{"Bibliographic":{"Title":"Mauna Loa Observatory : a 20th anniversary report","Authors":"","Publication date":"1978","Publisher":""},"Administrative":{"Date created":"08-17-2023","Language":"English","Rights":"CC 0","Size":"0000400585"},"Pages":["QC\n875\n.U52\nM3\n1978\nOF\nCOMMUNITY\nMauna Loa Observatory\nwinted STATES OF Austral\na 20th Anniversary Report\nJohn Miller, Editor\nSeptember 1978\nU.S. DEPARTMENT OF COMMERCE\nNational Oceanic and Atmospheric Administration\nEnvironmental Research Laboratories","QC\nNOAA Special Report\n875\nAND ATMOSPHERIC\nU52 M3\nNOAA SECURITY\nMAUNA LOA OBSERVATORY\n1978\n\"A 11 20th Anniversary Report\nSS SOCUATIMENT OF\nJohn Miller, Editor\nCENTRAL\nAir Resources Laboratories\nLIBRARY\nSilver Spring, Maryland\nFEB 9 1979\nSeptember 1978\nU.S. Department of Commerce\nJuanita Kreps, Secretary\nNational Oceanic and Atmospheric Administration\nRichard A. Frank, Administrator\nEnvironmental Research Laboratories\nBoulder, Colorado\nWilmot N. Hess, Director\n79\n0343","Hilo, with Mauna Loa in the back-\nground.\nii","To the memory of Harry\nWexler, a scientist with re-\nmarkable perception, breadth,\nand foresight; a man whose\nwarmth, sincerity, and\ngenerosity set an enviable goal\nfor those who were fortunate\nenough to know him; a re-\nsearch director with outstand-\ning ability who had much to do\nwith the establishment of\nMauna Loa Observatory.","CONTENTS\nDedication\nForeword\nL. Machta\n-\nLetters\nD. Pack, J. Steiner, R. Pueschel\nAcknowledgments\n1\nIntroduction\nK. Hanson\nHistory of Mauna Loa\nObservatory\nThe United States Exploring\nExpedition on Mauna Loa\n3\nC.S. Ramage\nEarly Days of the Mauna Loa\nObservatory\n10\nR.H. Simpson\nThe Construction of the Present\nObservatory\n13\nR. Stair\nThe First Twenty Years; An\nUnscientific Remembrance\n17\nB.G. Mendonca\nThe Observatory\n24\nH. T. Ellis\nMauna Loa Observatory-1968\n26\nL.H. Ruhnke\nA single, rare silver-sword grows on\nthe north slopes of Mauna Loa, pro-\ntected by a fence and periodically vis-\nited and inspected.\niv","Geology of Mauna Loa\nIce Nucleation Properties of Soils From\n86\nHawaii and the Continent\nG. Langer, D. Goto\nThe Volcanic Environment of Mauna\nLoa Observatory-History and\nAerosol Concentrations and Fallout\nProspects\n28\nat Mauna Loa\n89\nJ. P. Lockwood\nH.L Volchok\nIsotopic Composition of Lead Aerosols\nGeophysical Measurements\n92\nat Mauna Loa Observatory\nat Mauna Loa\nH. J. Simpson, E. A. Catanzaro\nSolar Measurements\nGas Measurements\nZenith Skylight Characteristics in the\nThe Influence of Mauna Loa Observatory\nSunrise Period at Mauna Loa\n105\non the Development of Atmospheric\nK.L. Coulson\n36\nCO2 Research\nC.D. Keeling\nObservations of the Solar Corona\nat Mauna Loa\n115\nAtmospheric Tritium Sampling on\nC. Garcia\n55\nMauna Loa\nH.G. Ostlund, A.S. Mason\nMeteorological Measurements\nAerosol Measurements\nThe Complexity of the Wind Patterns\nat Mauna Loa\n122\nLidar Measurements at Mauna Loa\nJ. M. Miller\nObservatory\n59\nR.W. Fegley, E.W. Barrett, H. T. Ellis\nClimate and Water Balance on the\nIsland of Hawaii\n129\nA Possible Effect of Local Volcanic\nJ. O. Juvik, D.C. Singleton, G. .G. Clarke\nActivity at Mauna Loa Observatory\n66\nB.A. Bodhaine\nIce Nuclei at Mauna Loa\n72\nMauna Loa, A Pictorial\nE. K. Bigg\nHistory\n141\nVolcanoes and Ice Nucleus Monitoring\nat Mauna Loa Observatory\n77\nBibliography\n155\nC. . M. Fullerton, C.J. Garcia\nNatural lava formations like \"the\nhead of Charles de Gaulle\" and \"a\nformer keeper of the mountain\" are\nthe scenic features on the road to\nMauna Loa Observatory.\nV","FOREWORD\nMan has always been an observer of his environment. Fragmentary\nobservations and records were made even by prehistoric peoples as a matter of\nsurvival. With the advent of the scientific method the need for systematic\nobservation of natural phenomena became apparent, and astronomical and\ngeophysical observatories for that purpose were eventually established\nthroughout the world.\nMauna Loa Observatory (MLO) started as a small meteorological station\nin the early 1950's. It is young in comparison with more established\ninstitutions such as Pic du Midi or Davos. But however young, MLO is\ndistinctly set apart from all others by its location and purpose. Its location is\nunique-high on a mountain thousands of kilometers away from any\ncontinental landmass. The environment is one of the cleanest in the world,\nand thus is ideal for MLO's purpose, which is to monitor constituents in the\natmosphere that could cause climatic change.\nIn the short time since its establishment, MLO has become a world-\nrenowned institution, because of its measurements of trace gases and solar\nradiation, and because it is the only facility in the world that has been making\ncontinuous CO2 measurements over the last two decades. MLO is now intended\nto become a model baseline station within the Global Monitoring for Climatic\nChange (GMCC) network, of which it is a part.\nTo mark the observatory's 20-year existence, this volume describes the\nmilestones of its history, highlights work done in the past, and outlines its\npresent purposes and goals.\nLester Machta\nDirector, Air Resources Laboratories\nSilver Spring, Maryland\nMore than 18 miles ahead on the\nroad, the observatory is barely visible\n(center) on the slope of snow-capped\nMauna Loa.\nvi","As one of the people who originally climbed around Mauna Loa\nwith Bob Simpson and Charles Woffinden, I am pleased to say hello on\nthis twentieth anniversary celebration.\nHow well I recall the weekly torture of driving to the summit site in\nthe early fifties. As I recall, the distance to the summit from Hilo was\nabout 50 miles. It took 4 hours for the drive: 2 hours to the 11,000-ft\nlevel, and then 2 hours for the last 10 miles to the summit. First it was in\na reclaimed WWII ambulance-type vehicle, then in a new Dodge power\nwagon. The impetus for the purchase of the new wagon was undoubt-\nedly the fact that the old vehicle gave out near the summit when I was\nthere with the director of what was then the Hawaii Aeronautics\nCommission. We walked for some seven hours - until about eight in the\nevening - before we were \"rescued\" by prisoners from Kulani prison at\nabout the 6,000-ft level of the road.\nThe earliest efforts at data collection were simply to profile the tem-\nperature, humidity, pressure, and wind along the slope. Because of the\narduous trip all the way to the summit, attempts were made to modify\nstandard instruments for 30-day operation. Several Rube Goldberg\ncontraptions were designed by me and my staff in Hilo, but none was\neffective We returned to standard operation and made weekly trips.\nEven then there were problems. The diurnal lifting of the inversion layer\nand the influx of marine air caused the ink to blot, caused the paper to\nstretch, and in general, was a thorn in the side. I believe the first project\nat the slope observatory was conducted by C. C. Kiess and C. Corliss\nfor the National Bureau of Standards and the National Geographic\nSociety during the summer of 1956. Dr. Nakaya and a group from\nHokkaido University were there when I left the Islands in January 1957.\nA reclaimed World W\nMy wife and I visited the observatory last June, just 2 weeks to the\nhicle provided a ride to\nday short of the anniversary of dedication. I found some aspects of the\nearly fifties.\nobservatory unchanged, but others quite different. The instrumentation\nand the vast array of projects are a far cry from our early beginnings.\nBut the awesome beauty of the mountains remains the same.\nI want you to know how I look forward to reading articles on\nGMCC and Mauna Loa's part in it. I swell with pride knowing I was\nthere at the beginning. My congratulations to the observatory on this\nanniversary. I know there will be many more. After a 20-year marriage,\nMadam Pele has accepted you and will do you no harm.\nJames W. Steiner\nvii","Best wishes on your twentieth anniversary. Your success has been\nThe two years that I spent working at the Mauna Loa Observatory\ndue not just to instruments, buildings, and budgets. The most important\n(MLO) amount to a mere 10% of its existence. Although my time there\nfactor in the observatory's history has been the people who have worked\nwas short it represents the most significant part of my professional\nthere. Their dedication, skill, and persistence have made this Pacific\ncareer.\nisland mountain top unique.\nMy involvement with the United Nations Environment Program\nMauna Loa's founding marked both a beginning and an end: In 1956\nover the past three years has brought me to many countries in Europe,\nthe observatory's initiation was the beginning of organized surveillance\nAfrica, and Central and South America. In talking to colleagues I find\nof atmospheric chemical behavior. It also represented the end of an era\nrepeatedly not only that MLO is NOAA's most widely known field\nduring which mankind had behaved as though it could assault the air-\nstation but also that my work there is better known than all my work in\nocean-land environment and leave no trace. Now, 20 years later, the\nGermany, California, Washington, and Colorado. It is sometimes dif-\ninformation collected at Mauna Loa, by its staff and by scientists from\nficult to explain why I am no longer involved with the observatory's\nall over the world, has shown that man is leaving marks on his\noperation.\nenvironment. Carbon dioxide, chlorofluorocarbons, tritium, and carbon\nThe record shows that about 25% of my publications are related to\nmonoxide data all reflect human activity, even on this remote and still\nwork that was done at MLO during only 15% of my professional life.\nrelatively clean atmosphere.\nThe paper on solar radiation trends at Mauna Loa that I had the pleasure\nThe observatory records also document the role of natural events.\nto write with Howard Ellis is the one most frequently cited in the scien-\nIts solar radiation data are a clear and quantitative indication of the\ntific literature in spite of the fact that solar radiation is outside my line of\neffect of massive volcanic eruptions, such as the 1963 Mount Agung\nexpertise. This is proof of the scientific creativity that MLO stimulates.\nemissions.\nA relatively high scientific output can be explained partly by the\nMauna Loa has played a vital role in the still incomplete\ngeography and meteorology that surround Mauna Loa; these have often\nunderstanding of the interlocking roles of trace gases and photo-\nbeen described and praised and are very conducive to scientific work.\nchemistry. The interrelationships of ozone, nitrous oxide, methane, and\nEven more important for the success and satisfaction that I enjoyed at\ncarbon monoxide, documented over time, will assist in determining the\nthe observatory was the help from people. The constant flow of visitors\nextent to which human activity affects the atmosphere. The observatory's\nwho stopped in Hawaii on their travels between Asia or Australia and\ndominant role in assessing the growth of carbon dioxide needs no elab-\nAmerica made life very interesting. Visiting scientists such as Earl\noration.\nBarrett, Gerhard Langer, and Gote Ostlund provided lots of ideas,\nFortunately, the observatory no longer stands alone. The three addi-\nincentive, and cooperation.\ntional observatories in the NOAA Geophysical Monitoring for Climatic\nIt is a particular pleasure to thank Helmut Weickmann for having\nChange program now support and expand your role. Other nations have\ncreated and Lester Machta for having extended the opportunity for me to\njoined in similar programs in a truly global effort to measure important\nwork at MLO. I am most grateful to John Chin, Howard Ellis, Bernard\natmospheric constituents and document their changes over time.\nMendonca, Judy Pereira, Al Shibata, and Alan Yoshinaga for the\nLet me congratulate you on completing your second decade, and\nloyalty, enthusiasm, and diligence with which they performed their tasks.\nthank you for the Mauna Loa contributions - past, present, and future\n\"Happy Birthday,\" MLO, and may your excellence long endure.\n- to an understanding and appreciation of our environment.\nDonald H. Pack\nRudolf F. Pueschel\nviii","INTRODUCTION\nDuring MLO's first twenty years scientists have been attempting to gain\nonly the simplest comprehension of the processes and mechanisms of climatic\nchange. This book records the history and drama of MLO's contribution to that\neffort.\nOur present knowledge is meager indeed. On the one hand, it is rather\nACKNOWLEDGMENTS\nsimple to list the various factors that are probably involved with climate\nThe editor is especially grateful\nchange - some within the climate system itself, and some outside the system,\nto all those who made pictures avail-\nwhich are not affected by climatic changes they produce. On the other hand,\nable, including Rudy Pueschel, Earl\nwe are desperately lacking in an understanding of how these factors operate -\nBarrett, Barry Bodhaine, Bernard\nthat is, what are the processes and mechanisms of climate change?\nMendonca, and Robert Pyle. Thanks\nNature stores information on climate change in various ways. There are\nare also due to Judy Pereira for re-\nrecords in tree rings, in the layering of continental ice sheets, and in sediments\ntyping many of the papers, to the edi-\nof bogs and ocean bottoms. Interpretation of this stored information is a form\ntor's wife, Sylvia, for compilation\nof \"monitoring,\" one that can evaluate in a short time changes in climate that\nand pre-editing of the manuscipt, and\nhave occurred over long periods of time. In addition, it may now be possible\nto the staff of ERL Publication Serv-\nfor man to copy nature's methods by \"banking\" air, water, soil, or biological\nices, Boulder, for beyond-the-ordi-\nsamples in suitable containers and environments to provide information on\nnary help in designing and preparing\nlong-lived constituents through laboratory analysis of the samples in future\nthis report.\ndecades and centuries. This, too, is a form of \"monitoring.\" However, many\nof the factors causing climate change cannot be captured in these ways but\nmust be observed directly if their essence is to be monitored. It is in direct ob-\nservation that MLO has made an important contribution.\nMany people deserve thanks for their foresight and perseverence, which\nproduced a valuable 20-year record of trace materials in the atmosphere.\nMany thanks are also due to John Miller, who conceived of this publication,\nand to those who gave valuable time to record their recollections of MLO.\nKirby J. Hanson\nDirector, Geophysical Monitoring\nfor Climatic Change\nBoulder, Colorado\nFinal steep approach to MLO, hard\nsurfaced here, to minimize dust, but\nlittle more than one car wide.\n1","HISTORY OF MAUNA LOA\nScientific research on the slopes of\nMauna Loa predated the present observa-\ntory by many years. Colin Ramage's\naccount in this volume of Wilkes's expedi-\ntion describes some of the early observa-\ntions made on the mountain. Not only\nmeteorologists but also geologists,\nbotanists, and biologists claim to have\nfound the slopes of Mauna Loa a fascinat-\ning place for scientific study, both in the\npast and in the present. Probably most\ninteresting, however, is the effect of the\nmountain on the people who work there.\nThe following articles present some of the\npersonal remembrances of the people\ninvolved in establishing and maintaining\nthe Mauna Loa Observatory.\nAerial view of MLO, looking to the\nnortheast, in the early '60s before the\nHAO installation. The observatory\nwas still using power from a genera-\ntor (shed, lower left). The four CO2\nintakes are visible at the corners of\nthe site.\n2","THE UNITED STATES EXPLORING EXPEDITION\n155° 04 W\nON MAUNA LOA\nHAWAII\nC. S. Ramage\nDepartment of Meteorology\nUniversity of Hawaii, Honolulu\n2\nLate on the afternoon of December 9, 1840, a dilatory sea breeze nudged\n12\nthe flagship of the first United States Exploring Expedition into Hilo harbor on\nthe northeast coast of the island of Hawaii. Lieutenant Charles Wilkes, USN\nHilo\n19° 43'\n8\n(1845, 1851), commanding from the sloop of war U.S.S. Vincennes, was\nN\npersonally responsible for the expedition's hydrographic and meteorological\n6\n14\nobservations. His instructions from the United States Congress to \"extend the\nDec.\nbounds of science and promote the acquisition of knowledge\" had led him to\n15\n8\nMauna\n1840\nprepare a scientific foray to the top of the great volcano, Mauna Loa, which\n10\nLoa\n22\nbulged 13,680 ft (4174 m) above sea level 40 mi (64 km) to the west-southwest\n12\n16\n(Fig. 1). On the authority of the King of Hawaii, Kamehameha III, 200 porters\n21\n20\n22 Dec.\nassembled, and early on December 14 they set out with a party of 16 from the\n18\n13 Jan.\nship. A distinguished member was Dr. Gerrit P. Judd, a medical missionary\n19\nKilauea\nfrom Honolulu, who later became Prime Minister of Hawaii.\nSunday\nLt. Wilkes had with him three mercurial barometers, thermometers, a\nCamp\nDaniell silver cup hygrometer (Middleton, 1969), and a Pouillet capsule hy-\nN\ngrometer. (The expedition volume on physics, which was to have included\ndetails of the instruments, was never published. A search of the original\npapers failed to unearth the information [Reichelderfer, 1940].) With them he\nmeasured \"shade,\" temperature, and dewpoints but did not specify the\n0\n10\n20 miles\nmethod of exposure.\nContour Interval is 2000 ft.\nFigure 1. The island of Hawaii\nshowing height contours (feet) and\nthe route, dates, and camp sites of\nthe expedition to Mauna Loa.\n3","After the first day of the 9-day trip to the summit, the route lay pathless\nover rugged lava and was usually extremely difficult. Shoes rapidly wore out\nand were sometimes replaced by improvised sandals made from ti leaves. In\nthe typically moderate trade wind weather, little rain fell. The small army\nreached Kilauea volcano, 4,000 ft above sea level, on the 16th, (Wilkes, 1845,\n1851) and\nMauna Loa burst upon us in all its grandeur. The day was extremely\nfine, the atmosphere pure and clear, except a few flying clouds, and this\nimmense dome rose before us from a plain some twenty miles in breadth\n[Figure 2]. I had not, until then, formed any adequate idea of its magni-\ntude and height. The whole dome appeared of a bronze colour, and its\nuninterrupted smooth outline was relieved against the deep blue of a\ntropical sky. Masses of clouds were floating around it, throwing their\nshadows distinctly on its sides.\nI now, for the first time, felt the\nmagnitude of the task I had undertaken.\nAfter spending Dec. 17 inspecting the vigorous eruption taking place in\nHalemaumau fire pit, the entourage ascended about 1,000 ft on the 18th.\nWilkes noted that \"this part appeared to have suffered much from drought; for\nin passing along we came to several narrow and dry water-courses, but met\nwith no water.\"\nAt 2:00 p.m. near 5,100 ft, the upper limit of the forest was reached, and\n\"as the clouds began to pass over, and obscure the path, we determined to\nhalt and encamp\nWe were now for a long time enveloped in mist for we\nhad reached the region of cloud [Fig. 3].\" One dewpoint and three temperature\nobservations were made at this camp (Fig. 4); during the night a cold westerly\n\"mountain breeze\" lowered the temperature to 43°F (6.1 C).\nOn the 19th, another 1,000 ft was climbed, during which \"we lost all\nsigns of trees, and were surrounded by low scraggy bushes.\" Soon after being\nenveloped in clouds between 2:00 and 3:00 p.m. the party made camp at\nabout 6,100 ft, \"at which we found ourselves above the region of clouds, and\ncould look down upon them.\" That night, Wilkes first remarked what was to\nprove a commonplace phenomenon above this level:\non pulling off my\nclothes, I noticed the quantity of electrical fluid elicited, which continued for\nJ.\n4","some time to affect the objects about me, particularly a large guanaco-robe I\nhad to sleep in.\"\nThe group remained at \"Sunday\" Camp until early on the 21st. Wilkes\nmade one dewpoint and seven temperature observations (Fig. 4). \"Mountain-\nsickness\" appeared, and \"we all began to experience great soreness about the\neyes, and a dryness of the skin.\" At noon on the 20th, \"I found it impossible\nto obtain the dewpoint with one of Pouillet's hygrometers, but after the clouds\nreached us in the afternoon it was found at 10°.\" Wilkes now realized that\n\"the want of water would prove the greatest difficulty I should have to\nencounter\" and sent a party ahead to determine if snow was lying on the\nhigher slopes.\nIt was a most beautiful day: the atmosphere was mild, and the sun\nshone brightly on all above us. We enjoyed a clear and well-defined\nhorizon, the clouds all floating below us in huge white masses (Fig. 5), of\nevery variety of form, covering an area of a hundred or more miles;\npassing around as they entered the different currents, where some\nacquired a rotary motion that I had never before observed. The steam-\ncloud above the volcano was conspicuous, not only from its silvery hue,\nbut by its standing firm, like an immense rock, while all around and be-\nneath it were in motion. The vault overhead was of the most cerulean\nblue, extending to and blending with the greenish tint of the horizon;\nwhile beneath the clouds, the foreground and distant view of the island\nwas of a dark green. The whole scene reminded me of the icy fields of\nthe Southern Ocean; indeed the resemblance was SO strong, that it\nseemed only to require the clouds to have angular instead of cumular\nshapes, to have made the similarity complete. It was perceived, that as\nmasses of clouds met they appeared to rebound, and I seldom saw them\nintermingle; they would lie together with their forms somewhat com-\npressed, and their outlines almost as well preserved as when separated\nand alone. After three o'clock, when the sun was retiring, the clouds\nadvanced up the mountain-side, and finally we became immersed in\nthem. This happened on both days at nearly the same hour.\nAfter\ndark the mist cleared off, when we saw the majestic cloud of the volcano\nhanging as though illuminated in its position.","From the 14th through the 21st the Hilo observations reflected a steady\ntrade-wind regime. By sheer chance, the 5,100-ft camp lay just below the trade\n70°\nwind inversion and Sunday Camp just above. Wilkes made all the observa-\ntions necessary to delineate the trade wind inversion-mist, cold, high dew-\npoint, and scattered vegetation at the lower camp; extreme dryness (\"around\n60°\nnothing but a dreary waste\") and warmth at Sunday Camp, below which only\nclimbing\nthe hot, moist venting from the volcano broke a flat cloud top. More than\n50°\nthis, he described the diurnal variation of the inversion on the mountain-rising\nin mist\nas the afternoon upslope winds lifted moisture and clouds to Sunday Camp,\nSunday\n°F\n5100 ft.\n6100 ft.\ndew\nsinking as the nocturnal downslope winds took over. At the lower camp, the\n40°\nCamp\npoint\ntemperature fell between 6:00 p.m. and 9:00 p.m. but at Sunday Camp, as the\nclouds once more retreated below, it rose. Wilkes was puzzled by his observa-\n30°\ntions but unaccountably advanced no hypothesis to explain them. This\nbrilliant man had come within a whisker of discovering the trade wind\n20°\ninversion more than 15 years before Piazzi-Smyth (1858). (In August 1856,\ndew\nwet/dry bulb thermometers in an instrument shelter strapped to the back of a\npoint\nmule were read every few hundred feet of an ascent and descent of Pico de\n10°\nTeide in the Canary Islands. Piazzi-Smyth identified a sharp temperature\ninversion at 2,000 ft which coincided with a strong lapse in humidity and the\n18\n00\n06\n12\n18\n00\n06\n12\n18\n00\n06\ntop of a dense layer of clouds (see also Riehl, 1954). Ironically, Wilkes had\n18 th\n19th\n20th December 1840\nalready observed effects of the trade wind inversion not far from where Piazzi-\nFigure 4. Temperatures, dew points,\nSmyth made his discovery. \"Whilst at Madeira [September 17-24, 1838]\nand weather at 5100- and 6100-ft\nwe found the height of the vapour plain to be about 4,000 ft above the level\nlevels.\nof the sea\nwhich corresponds to the highest point of cultivation\nThere is little doubt that the vapour plain must have considerable influence\nupon the climate of Madeira.\")\nThe group left Sunday Camp early on the 21st. \"The ascent now became\nmuch steeper\nfor the whole face of the mountain consisted of one mass of\nlava, that had apparently flowed over in all directions from the summit. The\nsun shone brightly, and his rays seemed to fall with increased power on the\nblack lava\nmany suffered from nausea and headache, and the desire for\nwater redoubled\nCamp was made at about 9,700 ft, and although snow had been found\nfarther up, water was still very short.\nFigure 5. \"We enjoyed a clear and\nwell-defined horizon, the clouds all\nfloating below us in huge white\n\" (Photo by C. Garcia.)\nmasses.\n6","7","Nearly 300 persons had occupied Sunday Camp, but by 3:00 p.m. on the\narrived when the sea-breeze set in from the different sides of the island:\na\n22nd the last of the porters could go no farther, and at 13,200 ft Wilkes found\nmotion was then seen in the clouds at the opposite extremities, both of\nhimself \"with the guide and nine men, with nothing for a covering but the\nwhich seemed apparently moving towards the same centre, in undula-\nsmall tent used for the instrument.\nA southwest gale was blowing, snow\ntions, until they became quite compact, and so contracted in space as to\nbegan to fall, and the temperature dropped to 15°F (-9.4°C). A wall of\nenable us to see a well-defined horizon; at the same time there was a\nclinkers helped keep the worst of the elements at bay, although at 4:00 a.m.\nwind from the mountain, at right angles, that was affecting the mass,\non the 23rd the snow broke through the canvas roof. The storm continued,\nand driving it asunder in the opposite direction. The play of these masses\nbut even so, by 1:00 p.m. the last climb, through a foot of snow, was fin-\nwas at times in circular orbits, as they became influenced alternately by\nished, and the small party stood on the edge of the summit crater.\nthe different forces, until the whole was passing to and from the centre\nin\nAs happens during winter in Hawaii, the trade winds had been disrupted\nevery direction, assuming every variety of form, shape, and motion.\nby a \"Kona storm,\" a middle tropospheric cyclone or a large-amplitude trough\nOn other days clouds would approach us from the southwest, when\nin the polar westerlies in the eastern half of which southerly or southwesterly\nwe had a strong northeast trade-wind blowing, coming up with their\nwinds and rising motion prevailed throughout the troposphere. The trade wind\ncumulous front, reaching the height of about eight thousand feet,\ninversion disappeared, and deep, precipitating clouds extended far above the\nspreading horizontally and then dissipating. At times they would be seen\ntop of Mauna Loa. Wilkes subsequently learned that Hilo was on the pro-\nlying over the island in large horizontal sheets, as white as the purest\ntected lee side of the island with respect to the storm, but they had exper-\nsnow, with a sky above of the deepest azure blue that fancy can depict.\nienced at Honolulu, on the nights of the 23rd and 24th, \"a very heavy storm\nfrom the southwest, simultaneously with the one that annoyed us on the\nBetween January 8th and 10th, another storm passed over, repeating the\nmountain.\" A greater degree of cold was experienced there than they had had\nsequence of southwesterly gales with snow, a shift to west with clearing, and\nfor years.\nthen a slow decrease in force.\nA wall of clinkers was built around living and instrument tents, and the\nparty managed to survive a recurrence of violent southwest winds on the\nThese gales reminded me strongly of those we experienced among the\nnights of the 25th and 26th.\nice on the Antarctic cruise. I regretted I had no anemometer, to ascertain\nThen the summit wind veered to the west, the trade wind regime was re-\nthe direction, changes and force of the wind. It is remarkable that these\nestablished, and from December 28 to January 7, weather remained fine. Tem-\nsevere gales all occurred during the night, beginning in the evening and\nperatures ranged from 15° to 20°F and reached 45° to 55°F in the afternoons.\ncontinuing until the next morning. I attempted to ascertain the velocity\nWilkes was occupied making gravity and magnetic measurements but failed to\nof the clouds by the rate of progress of their shadow across the crater,\ncoax the hygrometers into detecting atmospheric moisture.\nmarking the time of the passage; and the greatest velocity in many trials\nof those from the southwest was about forty-seven miles an hour. It was,\nDuring our stay on the summit, we took much pleasure and interest in\nhowever, observed, in these experiments, that the swiftness of the clouds\nwatching the various movements of the clouds; this day in particular\nseemed to increase in passing over the apex of the cone or crater.\nthey attracted our attention; the whole island beneath us was covered\nWhether this was the effect of being able to compare their movements\nwith a dense white mass, in the centre of which was the cloud of the\nmore nearly with fixed objects, I am not prepared to say; but I am\nvolcano rising like an immense dome. All was motionless, until the hour\ninclined to believe that in some cases, as they touched the mountain-side,\n8","e forced upwards and over the summit, with a much greater\nfor the first half of the crater than the last. The shortness of the\nt elapsed in passing the diameter of the crater, little more than a\ncludes the supposition that they had changed their form suffi-\no alter the figure of their shadow. The wind was blowing what\ne termed a strong gale, when the experiments were made.\nquit the summit on the 13th and reached Sunday Camp at 5:00\ney were \"soon enveloped in mist.\" The great observer remained\nACKNOWLEDGMENTS\nnear miss. \"The vapour plain, or the height at which the\nremained, was about 5,000 feet above the sea\n[They] were\nI am grateful to my Hilo colleagues\nove the height of 8,000 feet except during the stormy weather.\"\nfor taking the photographs used to\nillustrate this article.\nREFERENCES\nMiddleton, W. E. K., 1969: Invention of the\nMeteorological Instruments. Johns Hopkins\nPress, Baltimore, 362 pp.\nPiazzi-Smyth, C., 1858: Astronomical\nexperiment on the Peak of Tenerife. Phil.\nTrans. Roy. Soc. London, 148:465-533.\nReichelderfer, F. W., 1940: The contribution of\nWilkes to terrestrial magnetism, gravity and\nmeteorology. Proc. Amer. Phil. Soc.,\n82:583-600\nRiehl, H., 1954: Tropical Meteorology,\nMcGraw-Hill, New York, 392 pp.\nWilkes, C., 1845: Narrative of the United\nStates Exploring Expedition During the Years\n1838, 1839, 1840, 1841, 1842 (in five\nvolumes). Lea and Blanchard, Philadelphia.\nWilkes, C., 1851: United States Exploring Expe-\ndition During the Years 1838, 1839, 1840,\n1841, 1842. Vol. XI Meteorology.\nC. Sherman, Philadelphia, 726 pp.\n9","Shortly after I arrived in Honolulu in the spring of 1948 and had bene-\nEARLY DAYS OF THE MAUNA LOA\nfited from an indoctrination tour of the Hawaiian Islands by Luna Leopold of\nOBSERVATORY\nthe Pineapple Research Institute, I wrote an article published in the Honolulu\nAdvertiser about the opportunities for research and in particular regarding\nMauna Loa as an ideal natural laboratory for geophysical and atmospheric\nR. H. Simpson\nresearch. The potential benefits of establishing a summit weather observatory\nUniversity of Virginia\nwere stressed.\nCharlottesville, Virginia\nMuch to my surprise several days later I received a call from Tom Vance,\nthen Director of Institutions for the Territory of Hawaii, who offered to join\nme in seeking means of locating an observatory near the summit. His interest\nhad been kindled because of the possibility of co-locating a ski lodge near the\nsite which could be largely operated by and used as a means of progressively\nrehabilitating prisoners at the Kulani camp, a somewhat experimental facility\non the slopes of Mauna Loa.\nIn the few weeks that followed, a plan of action was evolved for carrying\nout this joint venture. After a quick trip to Washington to obtain the endorse-\nments of Francis Reichelderfer, then Chief of the U.S. Weather Bureau, and his\nDirector of Research, Harry Wexler, Tom Vance and I set about the task of\nimplementing the program, with the unfunded blessings of the Weather Bureau\nand with a preliminary endorsement of the Department of the Interior to enter\nNational Park grounds for this purpose.\nThe first task was to determine a specific site for the structure and to\nsurvey the route of an access road that would minimize the risk to personnel\nat the site and provide egress in case of volcanic eruption. The aid of Gordon\nMacdonald, then Director of the Volcano Observatory at Halemaumau was\nenlisted for this task, and four survey trips were made to the summit before a\nsatisfactory route was determined and a site for the structure pinpointed at the\n13,453-ft (4100 m) level just below the upper rim of Mokuaweoweo crater.\nI was able to obtain the loan of two giant road-building machines from\nthe Navy at Pearl Harbor, which also supplied water transportation to Hilo\nand the necessary spare parts. The funds for fuel to operate these machines\nand to procure the materials for construction of a small building to house the\nThe site survey party on March 9,\nautographic meteorological instruments were reprogrammed from the Weather\n1951, included (left to right) Charles\nBureau Pacific Projects budget. The labor for road construction and\nM. Woffinden, U.S. Weather Bureau;\nW. P. Mordy, Head, Meteorology\nDiv., Pineapple Research Co.; Tom\nVance, County of Hawaii; Leon Sher-\nman, UCLA; James Kealoha, Chair-\nman, County of Hawaii; Mr. You-\nman, U.S. Weather Bureau; James\nSteiner, U.S. Weather Bureau, Hilo\noffice.\n10","supporting engineering services and for construction of the building was\nsupplied by the Department of Institutions.\nMonth after month went by with discouragingly slow progress, in part\nbecause labor was available only sporadically and in part because of diffi-\nculties with the road routing. On a number of occasions one half mile of road\nbed would be roughed in only to encounter a giant lava tube that could not be\nbypassed or bridged with the funds available. Nevertheless, a road was\ncompleted to the summit, traversible by four-wheel-drive vehicles, and the\nframe observatory structure, built in sections at Kulani and trucked to the\nsummit, was assembled at the selected site. While cinder cone materials were\nbeing spread to improve the road bed, meteorological instruments were\ninstalled at the site, and the observatory was equipped with the necessary fa-\ncilities for overnight lodging and cooking by Weather Bureau personnel.\nLittle more than a year after construction began, the Mauna Loa summit\nobservatory was officially opened and dedicated by Oren Long, Governor of\nthe Territory of Hawaii, in the presence of more than 25 invited guests: distin-\nguished scientists, administrators, politicians, and educators, from Wash-\nington, D.C., and half a dozen states as well as from the Territory. The open-\nair ceremony took place in a 20-knot wind, with light intermittent snow and a\ntemperature in the upper twenties (°F). A barbequed beef dinner was served\naround a roaring bonfire following the ceremony. The excursion in four-\nwheel-drive vehicles to the summit, punctuated by frequent stops for rests, for\naccommodation to the thin air, and for a few deep breaths from the \"walk-\naround\" oxygen bottles borrowed from military supplies, was completed\nwithout incident.\nUnfortunately, after the road-building equipment was returned to the\nNavy, the road began to deteriorate rapidly, and getting to the summit site to\nservice equipment and collect records became progressively more arduous.\nFailure to obtain funds to improve the road SO that a standard passenger car\ncould reach the summit prevented Tom Vance from establishing a ski lodge.\nNovember 16, 1952, expedition to\nFinally, in 1953 it was necessary to discontinue trips to the summit observa-\nevaluate site locations.\ntory, and observations started in 1949 were discontinued.\nMeanwhile, the studies of cold lows migrating westward across the\nPacific, and questions raised by Clarence Palmer as to their origin, as well as\n11","studies of their influence on the weather of Hawaii and the role they might\nplay in the formation of typhoons as they reached the western Pacific, led me\nto propose that local measurements of total-path ozone in Hawaii might\nprovide useful inferences concerning the character of circulations in the upper\ntroposphere and lower stratosphere and shed light on the evolution and devel-\nopment of these upper lows. However, no interest could be aroused or oppor-\ntunities found to initiate such observations until 1955, when by a most re-\nmarkable coincidence while vacationing in the canyon country of the western\nUnited States I met a Bureau of Standards scientist, Ralph Stair, who was\nworking on solar flare observations at Sacramento Peak Observatory. He\ncomplained of the protracted periods in which atmospheric conditions limited\nor completely curtailed his observations. I asked why he didn't arrange to\nmake observations from a more ideal spot like Mauna Loa. This led to a\ndiscussion of our abortive venture at the summit site, my frustration in seeing\nobservations there discontinued, and my new interest in having ozone\nmeasurements from a suitable Hawaii site.\nThe outcome of this encounter was a tentative agreement that we would\nwork in our respective agencies to reestablish a Mauna Loa Observatory\nthrough a cooperative program in which the Bureau of Standards would do\nthe engineering, let the contract, and supervise the construction, and the\nWeather Bureau would fund the construction, drawing upon funds recently\nappropriated for hurricane research at the newly established National Hurri-\ncane Research Project, to which I had just been assigned as the first director.\nThe understanding was that a new site would be sought at which personnel\ncould remain indefinitely and that the program would include not only\nmeteorological observations but also facilities for solar coronagraph studies, a\nDobson spectrophotometer for measuring total path ozone, and suitable in-\nstrumentation for measuring atmospheric particulates and carbon dioxide.\nWhen I returned to Washington and presented the proposal to Harry\nWexler with an indication of my willingness to divert hurricane research\nmoney to help reestablish the Mauna Loa Observatory, he was immediately\nenthusiastic and set about the task of assigning project leaders to coordinate\nthe program. Ralph Stair was similarly successful, and the seeds were effec-\ntively sown from which ultimately the fine observatory at the 11,150-ft (3400\nm) level was established.\nThe Mauna Loa summit observatory\nstructure prior to its opening at the\n13,453-ft level. R. H. Simpson (left)\nand Joachim Kuettner (right).\n12","The Mauna Loa Observatory had been a dream of many for several\ndecades. Although the U.S. Weather Bureau had from time to time set up tem-\nporary observation posts at a number of places along the trail leading to the\nsummit of the mountain, the origin of the observatory came about in a rather\npeculiar manner differing greatly from any plans made or contemplated by\nTHE CONSTRUCTION OF THE PRESENT\nthose most interested in having a permanent weather station on the north\nOBSERVATORY\nslope of Mauna Loa.\nThe National Bureau of Standards (NBS) had been interested in the\nintensity of the ultraviolet solar radiation since about 1928. This interest grew\nwith time and expanded in scope to include not only the total solar ultraviolet\nRalph Stair\nbut also the spectral intensity through a wide region of the spectrum; the total\nNational Bureau of Standards\namount and vertical distribution of ozone; and some information on water\nvapor, dust, and other pollutants in the atmosphere. In connection with this\nwork, much effort was expended on the development of new instrumentation\nand standards of radiant energy. Much of this work was financed through\nnon-Bureau sources, including the Army, Navy, Air Force, National Academy\nof Sciences, and Air Pollution Foundation, with assistance from others, such\nas the High Altitude Observatory, Upper Air Research Observatory, U.S.\nWeather Bureau, Lowell Observatory, and California Institute of Technology.\nAs the instrumentation and standards were improved through the years,\nthe lack of atmospheric clearness remained the chief obstacle to obtaining high\naccuracy in the measurements of spectral solar energy. Measurements were\nmade in various localities: Washington, D.C.; Mount Evans, Colorado;\nThe observatory building in June\nClimax, Colorado; San Juan, Puerto Rico; Flagstaff, Arizona; Pasadena,\n1956.\nCalifornia; White Sands, New Mexico; and Sunspot, New Mexico.\n13","result of the shielding effect of Mauna Kea the atmosphere would be least\nIt was at Sunspot, in June 1955, that we had the good fortune to\nturbulent and most free of clouds. Furthermore, a mountain cinder road\nreceive a prominent visitor from the Florida Hurricane Center - none\nwas already in existence in this area. Accordingly, an arrangement was\nother than Dr. Robert H. Simpson himself. As there had been a dust\nmade for a conference (in December 1955) with the Honorable Samuel\nstorm of considerable magnitude over the adjacent New Mexico desert,\nWilder King, then Governor of Hawaii, for the purpose of transferring\nthe sky was very bright with the sun coming through weakly. Dr.\ntitle to a suitable area on the north slope of Mauna Loa (within the\nSimpson was very sympathetic to our need for a clear sky with direct\nMauna Loa forest and game reserve) from the Territory of Hawaii to the\nsunlight coming through at high intensity and for seeing Venus in the\nDepartment of Commerce of the federal government. Mr. King agreed to\ndaytime and unobscured by clouds. (Venus was clearly visible in the\nthe transfer, and it was accomplished through the proper Hawaiian\ndaytime at Climax, Colorado, except when there were clouds.) We\nchannels, with survey by C. L. and D. J. Murray for an area of 4.05\nwanted a better location but knew not where to turn, since we had\nacres (10 hectares) in the shape of a square 420 by 420 ft (tax map 4-4-\nconsidered all possible locations within the continental United States and\n16; C.S.F. No. 12333). The exact area was chosen previous to the survey\nfound no place superior to Sunspot.\nby Ralph Stair, Roy L. Fox, and James W. Steiner, of NBS; Pacific Area,\nDr. Simpson had a suggestion that we agreed to consider. It was to\nU.S. Weather Bureau; and Hilo office, U.S. Weather Bureau, re-\nset up our laboratory on one of the mountains in the Hawaiian Islands.\nUpon our return to Washington, a number of conferences at the NBS and\nspectively.\nIn picking the site for the observatory a number of considerations\nat the Washington office of the U.S. Weather Bureau resulted in an\nwere taken into account. The site chosen was near the mountain cinder\nagreement for NBS to use some $25,000, which would be transferred from\nroad at the upper terminus of its better condition, that is, at the highest\nthe U.S. Weather Bureau to the Radiometry Laboratory of NBS, in the es-\nelevation possible to be reached with two-wheel-drive vehicles. The\ntablishment of a laboratory for the joint use of the Weather Bureau and\nelevation, approximately 3400 m, was considered near the limit for\nNBS. The amount of money available was insufficient for the construction\nof a permanent building through the normal construction channels, but\nextended living and working for most individuals. The area was on a\nfor use in establishing a durable \"shelter\" for practical use in all kinds of\nrecent volcanic lava flow which, because of its elevation above adjacent\nareas to the right and left, offered promise of some protection from\nweather expected on one of the Hawaiian mountains it was considered\nadequate if strict economies were followed throughout.\nfuture flows on this side of the mountain.\nThe 4.05-acre area was marked at the northeast corner by a spike set\nFollowing preliminary discussions and evaluations in Washington, it\nin concrete which was labeled \"Stair, 1955.\" The other three corners and\nwas decided that the best location for a solar radiation laboratory in the\nthe midpoints of the north and south sides were defined by 3/4-in iron\nHawaiian Islands would be on the north slope of Mauna Loa, where as a\nDedication ceremonies on June 28,\n1956, marked the official beginning of\nthe observatory.\n14","pipes driven into the lava.\nhandled by NBS, with consultations and interviews with U.S. Weather\nThe transfer of this parcel of Mauna Loa to the Department of\nBureau personnel. This work was performed primarily on regular Bureau\nCommerce (of which NBS and the U.S. Weather Bureau were parts)\nschedule with no cost to the construction fund. As a matter of fact, to\nhaving been arranged, numerous interviews were conducted with other\nkeep all costs to a minimum no drawings or blueprints were prepared, all\ngovernment and business people within the Hilo area concerning the best\nwork being performed from detailed specifications.\nprocedures to follow in getting the most desirable building on the site at\nWith the building specifications in order, the requests for bids on\nthe least cost. A first thought was to construct the building of lava\nconstruction were handled through the offices of NBS in the usual official\nblock, of which there was an abundance within the area. However,\nmanner.\ninvestigation showed this to be more expensive than the use of cast\nDuring construction, in order to keep costs down, all inspections\ncinder block manufactured in Hilo. (The possibility of a lava block\nand consultations with the contractor were handled by Hilo U.S.\nbuilding was left open, however, for a time when more money might be\nWeather Bureau personnel through occasional visits to the site followed\navailable, through the incorporation of a concrete ledge on the building\nby simple reports to NBS.\nfoundation.) Other interviews regarding available material resulted in the\nBy June 1956 the building was essentially complete as described and\nlocation of local overstocks of certain items and sizes (for example,\nillustrated in the September 1956 issue of Weather Bureau Topics, the\naluminum sash and roofing sheets). The final specifications for the\nOctober 1956 issue of National Bureau of Standards Technical News\nbuilding included only materials available locally SO as to permit\nBulletin, and the November 1956 issue of Discovery.\nconstruction without delay or extra cost. All possible contractors within\nThe construction of the observatory proceeded smoothly through-\nthe area were interviewed and made acquainted with the local supplies in\nout. There were helping hands in all quarters with no dissidents any-\nstock.\nwhere. Donations came from several organizations. One, in the form of\nThe Hawaiian Kulani prison authorities were extremely helpful in\na radio transceiver, was supplied by the Radio Section of NBS. A small\nthe construction of the new observatory. They not only continued to\nlibrary was furnished by the Smithsonian Institution and the Department\nkeep the roadway in repair but volunteered to \"level off\" the 4-acre area\nof Agriculture. Only minor problems developed, and those were\nahead of the contractor and to deliver to him the water required during\nprimarily the result of \"corner cutting\" to get the job done at low cost.\nconstruction. The latter was a most important item as it relieved the\nFor example, the placing of the diesel electric generator near the building\ncontractor of all expenses that might be involved in obtaining and\nunder the water tank required moving it later to a more distant location\noperating water tanks.\nbecause of vibrations and noise that interfered with experiments inside or\nAll developments in the specifications for the new building were\nnear the building. Although the construction of the observatory building\n15","was carefully specified to contain steel rods to hold the structure securely\nattached to the foundation in case of a hurricane, no such forethought\nwas included in the case of the \"outhouse\"; it went the ways of the winds\nupon the arrival of the first 100-mph hurricane.\nIn getting the \"most possible for the least cost,\" one can run into\nLava formation at the 7000-ft level on\nreal trouble. This happened in the case of the NBS project leader, who\nMauna Loa in December 1966.\nfound it necessary to become plumber and janitor for a day to get the\nwater running and the floors clean for the first scientific observing team\nto use the observatory (Dr. C. C. Kiess and associates of NBS for the Na-\ntional Geographic Society).\nBy dedication time, the building was completely furnished (down to\nbed sheets, kitchen utensils, dishes, and flatware) for comfortable living\nfor six observers. Government inspection (by Ralph Stair) confirmed ful-\nfillment of the contractor's obligation, although the observatory faced\n\"magnetic south\" rather than geographic south, which shows that one\ncan never be too careful in writing a specification - the contractor used\na compass. The building was duly accepted by NBS, the contractor paid,\nand the Mauna Loa Observatory dedicated on June 28, 1956, the intro-\nductory remarks being made by the Honorable Samuel Wilder King,\nGovernor of Hawaii.\n16","THE FIRST TWENTY YEARS; AN UNSCIENTIFIC REMEMBRANCE\nBernard G. Mendonca\nGeophysical Monitoring for Climatic Change\nBoulder, Colorado\nI am sure that enough will be said in this 20-year commemorative about\nand John Chin. These people broke ground for those who followed. Their\nthe scientific work done at Mauna Loa Observatory (MLO) and the uniqueness\nwork was long, arduous, and often frustrating. Conveniences were minimal,\nand suitability of its setting for atmospheric monitoring. Certainly, the\nand the daily travel over rough roads from Hilo to the observatory and back\nobservatory has served as a fountainhead for scientific studies and subsequent\nwore on the will of all.\nscientific publications on the atmospheric sciences. Without doubt it should\nPeople like \"Doc\" Foss, the first electrical technician and instrument man\nalso continue to do so for the next 20 years or more.\nat MLO, made life interesting in the first trying years. His belief in 1960 in\nSo instead of dwelling on the science at MLO I would like to recall briefly,\nsolar power, his pattern of a solar hot water heater, and his formation of a\nfor old times' sake, some of the people and events associated with the\ncompany to produce it in Hilo attested to his inventiveness and vision of\nobservatory that will never make the scientific publications and in the dimness\nthings to come. He was a master of improvisation and modification of the\nof time will be forgotten. What is related is not complete, all-comprehensive,\nscientific instrumentation at MLO and saved the day many a time by being able\nor plucked from an accurate chronology, but comes from memory with all its\nto adapt state-of-the-art instrumentation to the needs of the monitoring\nfailings.\nprograms at MLO. Some of his other vocations were orchestra leader and\nMy first contact with MLO was in August of 1958 as a part-time\ndruggist and drugstore owner, whence the name \"Doc.\"\ngovernment employee. To support myself while attending the University of\nBut even before the \"official\" opening of MLO in 1957-1958 during the\nHawaii, I was working under Saul Price in the U.S. Weather Bureau Pacific\nIGY, preliminary meteorological measurements at test sites had been done at\nRegion Office in Honolulu. Saul Price, together with Jack Pales (the first\nMauna Loa. In 1956, Dr. Ralph Stair (solar physicist from the National\ndirector of MLO), was responsible for the operation of the observatory. At that\nBureau of Standards) with the help of other people got the funding to build\ntime the observatory had a Honolulu branch, which consisted of Saul Price,\nthe first permanent building at the present site of the observatory for his solar\nRichard Sasaki, and myself. The observatory was officially 1 year old by then.\nand planetary studies (Mars).\nData flowed from Mauna Loa to the Honolulu branch for processing analysis\nAs early as 1951 the first meteorological measurements on a routine basis\nand eventual publication in scientific journals. The International Geophysical\nwere started at the summit of Mauna Loa. A wooden building was erected and\nYear (IGY) was winding down, but because of the high quality of\nmaintained by the U.S. Weather Bureau rawinsonde crew at the Hilo airport\nmeasurements made at MLO during the IGY the observatory operation was\nunder the official in charge, Ray Busniewski. Weekly trips were made by\ncontinued in the succeeding years. Those were prosperous times. The staff\nHoward Tatum, Roy Sodetani, George Nii, and others over a nearly\nexpanded to a maximum of about 13 people (counting the Honolulu branch),\nimpossible jeep trail to collect meteorological data at the summit. The\nand many came and went. A few names remain in mind: Jack Pales, director;\nbeginnings of benchmark monitoring on Mauna Loa had its humble origin\nCliff Kutaka; Harry Arashiro; Colby \"Doc\" Foss; Bill Cobb; Howard Ellis;\nwith this sturdy crew. They showed that it could be done routinely.\n17","In 1959 the observatory proposed and obtained a 4-year contract\nchart paper for the recorders, with a few dollars extra for the light bill. Under\nsupported by the Department of Defense to do spectral solar radiation\nHoward Ellis, the director at the time, the basic programs were kept going,\nmeasurements, with an emphasis on measurements of the upper tropospheric\nand we waited. For 2 years, nothing much changed, and the observatory hung\naerosols and their relation to the Bowen hypothesis of worldwide precipitation\nin limbo.\nanomalies and meteor showers. The CO2 and atmospheric ozone programs\nBy 1966, things began to pick up and continued for several years. Funding\nwere also coming into their own, and SO were benchmark measurements of\nonce again became available; a new director, Dr. Lother Ruhnke, obtained\natmospheric electricity parameters during this interval. Then, at the end of\ncommercial power and a telephone to the observatory. Previously, diesel\n1963, an austerity wave washed through all government agencies, the ripples\ngenerators had provided power and a multitude of headaches to the staff who\nextending to the MLO programs. There was serious talk and maneuvering to\nhad to maintain them. An office in Hilo on the University of Hawaii campus\nshut down the observatory permanently. Funds suddenly were not available,\nwas established with secretarial help and several new staff members. Soon the\ninterest in long-term monitoring had lost its champions, and the expense and\nfirst part-time university student help were on the payroll, sharing in the\neffort of running an observatory thousands of miles distant on a volcano in\nprogram-an innovation that has lasted to the present. Things started to hum\nthe middle of the Pacific was hard to justify. The staff dropped from a high of\nagain!\nabout 13 to 3. Personnel transferred to other jobs, and programs terminated\nThe scientific community had expanded, and two other agencies joined\none by one. In February 1964, Howard Ellis, Mike Keyes, and I were the only\nthe observatory on the mountain. The High Altitude Observatory (HAO), a\nones who remained on the official staff, mainly because we had pleaded\nsection of NCAR in Boulder, had set up a solar coronascope observatory\ndesperately (with the help of others, all of whom I do not know) for the\nadjacent to the observatory in 1965. The Atomic Energy Commission (AEC)\ncontinuation of the observatory. I do know that Jack Pales; Saul Price; Nels\nhad set up a classified program on the mountain in 1963 for 2 years, which\nJohnson, director of the Pacific Supervisory Office; and Dr. C. D. Keeling\nafterwards was continued intermittently for several years. The AEC program\nwere involved. Later, Dr. Helmut Weickmann, director of the Atmospheric\nwas classified, that is, it was classified until the newspaper, to everyone's\nPhysics and Chemistry Laboratory in ERL, took us under his wing after the\ndismay, flashed the headlines \"Mauna Loa Station to Monitor 'N' Tests\" on\nreorganization of the U.S. Weather Bureau into the Environmental Science\nFriday the thirteenth, some months after the AEC had arrived. The HAO\nServices Administration. With his help the observatory survived the lean\ncoronascope has stayed to the present.\nBudgeting once again became hard, and there was a slight recession. Help\nyears.\nfrom other agencies lessened the slack, and the observatory endured. The next\nAs a result of the cutback during this austere time the Honolulu branch\nbig push came in 1971. Under the directorship of Rudolf Pueschel several\nwas closed, the staff was reduced to 3, the Hilo office was closed, and a\nsignificant papers were published, the Mauna Loa data showing the worthiness\nstarvation budget was given for salaries, gasoline to get to the mountain, and\n18","of benchmark monitoring for climatic change. MLO was incorporated as the\nprincipal geophysical monitoring station of a new section in the Air Resources\nLaboratories (ARL) of the Environmental Research Laboratories in NOAA. This\nsection was designated Geophysical Monitoring for Climatic Change (GMCC)\nand started under the directorship of Donald H. Pack and Lester Machta,\nDirector of ARL. For the first time, MLO was part of an agency set up to do the\nkind of atmospheric research that the observatory had been involved in for the\npast 13 years. It has remained so to the present. Now, three other\nobservatories have been added, and a central office in Boulder, Colorado,\ncoordinates the network.\nHowever, in the evolution of such an observatory there are many\nadventures, milestones, and passing interests that are as integral a part of the\nhistory of the observatory as the scientific work that it was set up for. To\nattempt to comment on them all is of course impossible, but I'd like to\nmention a few:\nROAD AND TRAVEL\nA major nemesis to the operation of the observatory over the years has\nbeen the roads used for access to the observatory. The standing joke for those\nwho have worked there is that the access road is the only road on the island\nthat no one will lay claim to for fear of being obligated for its maintenance.\nThe first road was laid out (by some sweet-talking and creative accounting\nand financing of several government agencies) with prison labor from the\nKulani honor camp. As it is jokingly retold, an inmate was set on a bulldozer\nand told to cut a road to the summit of Mauna Loa. When funds ran out, the\nobservatory was built at the end of the road.\ne 1966.\n19","The first route used for access began at the Kulani honor camp turnoff on\ntaken a heavy toll on vehicles. With the AEC, HAO, NOAA, and the county\nthe belt highway. This preexisting road was called the Stainback Highway and\ngovernment all contributing, another major upgrading was begun, and this\nvery soon was corrupted to \"strainback\" highway by those who made the\ntime most of it was oiled to a smooth hard surface. Again the road repair was\nweekly trips to the observatory. It terminated at the Kulani honor camp\ndone by prison inmates.\n(elevation 7,000 ft). The daily search and shakedowns of people and vehicles\nDuring this period the HAO staff championed the upkeep of the road.\nby the guards at the camp were an unbelievable experience for first-time\nDick Hansen and Charles Garcia were instrumental in organizing the work\ntravelers to the observatory. Later the guards got to know the staff and waved\ndone on the road. After each major project the staff members of HAO and MLO\nthem through the camp unless something was amiss. Above the camp the road\nwould prepare a Hawaiian luau for the inmates, and feasting was the order of\nran along the northeast ridge and required four-wheel-drive vehicles. A good\nthe day.\ntrip with no breakdowns required 2 1/2 hours for the 40 miles. It took a hard\nDuring the periods when the road fell into disrepair, the vehicles used for\nhead, resilient behind, and cast-iron stomach to make the journey and keep\ntravel to the observatory suffered drastically. Punctured brake lines, broken\none's breakfast. Maintenance of the road was virtually nonexistent, and the\nsprings, busted shocks, sheared-off wheels, and ruptured oil pans were not\ngravel surface deteriorated rapidly with the heavy tropical rains. It was not\nunusual but routine monthly maintenance. It didn't take long before the word\nunusual to come across wild boar rooting among wild orchids on the shoulder\nwas out to the local garages in Hilo to avoid the MLO GSA-vehicle contract\nof the road in the early morning and later to see wild goats eating ohelo\nlike the plague. Car rental agencies in Hilo started to attach clauses to their\nberries along the road above the camp.\nstandard contracts that forbade trips to MLO.\nIn 1963, with the establishment of the AEC project at Mauna Loa, funds\nWhen the observatory was at its wit's end in finding suitable repair\nbecame available for upgrading the road. Howard Ellis, who was director at\ngarages, an agreement was made with a garage in upper Kaumana, the last\nthe time, diverted funds and equipment and convinced the road crew to cut\none before the saddle road. The proprietor, Noburo Yamanouchi, who quietly\ntwo sections of the new road north from the 8,300-ft elevation to connect with\nran the garage in the local community, accepted the maintenance of our\nan old isolated hunters' road and the saddle road. The event was hailed as the\nvehicles as a personal challenge. He was a master mechanic and through his\nopening of the highest road in the Pacific by the press, who also noted that\npersonal efforts kept our vehicles on the road through the most adverse\nonly the state of Colorado had a higher road. To the observatory staff it\nconditions from 1964 to 1972, when he retired. His dedication would often\nmeant welcome relief that driving on the rugged road through the Kulani\nresult in midnight repairs to get a car ready for the morning trip. He was so\nhonor camp was over and that the new saddle route was now paved for half\nfamiliar with our vehicles that he could identify the arrival of the observatory\nthe trip to the observatory. Everyone was happy.\ncars approaching his garage by their characteristic rattles before he saw them.\nBy 1965 the road to the observatory had again fallen into disrepair and\nHe delighted in being able to tell the current customers to their surprise that\n20","the observatory boys were approaching the garage before they were seen. Mr.\nSnow time at MLO.\nYamanouchi would certainly have qualified as an honorary member of the\nMLO staff.\nThe road and daily travel to the observatory are still part of the rugged,\nrustic appeal of the observatory's setting.\nVISITORS\nTo anyone who has spent time at remote observatories, visitors are a way\nof life. Over the years, MLO has had more than its share. Visitors have been a\nmixed blessing: the visiting scientists generally are a welcome and refreshing\ncontact with the outside world, a source of fresh ideas, and an update of\nevents occurring in one's field of interest. The casual visitor, on the other\nhand, could make the observatory staff practice their virtues to the limit. The\nsign-in guest book maintained at the observatory contains a virtual who's who\nof visitors over the last 20 years. The visitors can arbitrarily be classified into\nseveral categories: (1) the visiting scientist; (2) the visiting VIP; (3) the\noutdoorsperson (hiker); and (4) the casual inquisitive walk-on, the prearranged\ntours and the snow people.\nThe snow people were a phenomenon that no one anticipated at the\nobservatory. With the opening of the saddle route access road to the\nobservatory, a typical family passenger car with some strain was able to\nascend the mountain to the 3.4-km altitude of the observatory for the first\ntime in the island's history. The significance of the elevation was that in the\nwinter the snow line of the mountain often descended to about 2.8-km\nelevation. The stage was set to permit the people of the island to drive up to\nfreshly fallen snow after a winter cold front passage and experience for the\n21","first time in their lives snowman building, snowball throwing, sledding, and\nall sorts of winter sports. Once the word got out that this could be done, the\nresult was mayhem on the slopes of the mountain after each snowfall in the\nwinter. A good weekend would bring more than 500 cars a day packed with\npeople, hot Thermoses, picnic lunches, makeshift sleds, and every possible\nadaptation of Hawaiian wear into winter fashions. The upper reaches of the\nmountain turned into an instant winter carnival land. The effect on the\nobservatory staff driving to work was bewilderment. The net effect was that\nthe local community was accepting the establishment of a scientific\nobservatory on Mauna Loa. To this day, to a lesser degree, the arrival of the\nsnow people every winter is a Mauna Loa phenomenon.\nTHE WHITE DOG\nNo reminiscence of MLO would be complete without mentioning the white\ndog of Mauna Loa. Much has been told about the mysterious phantom dog\nthat would appear on the mountain to forewarn of a volcanic eruption.\nHawaiian legend relates a tale of Pele, who is the fiery goddess of the\nvolcanoes on Mauna Loa, and her companion dog, whom she would send as a\nmessenger to alert the people whenever an eruption was imminent.\nThe white dog was first noticed by the observatory staff during the latter\npart of 1959. At that time the staff were living on site for up to a week at a\ntime on rotating shifts. Because of this housekeeping, a rubbish dump was\nsoon developed to the west of the observatory. The contention of the staff was\nthat a stray white dog had discovered the dump and foraged it for food.\nAttempts by the staff to befriend it and later to capture it, no matter how\npersistent or devious, failed. The dog for some reason would have nothing to\n22","the staff members were hesitant to talk about it to anyone they did not know.\ndo with the observatory staff. Soon the dog disappeared and was presumed to\nTo this day the mystery of the white dog is just that-a mystery.\nhave found its way back to the populated regions of the island. In December\n1959, Kilauea Iki erupted.\nTo the amazement of the staff the dog reappeared at the observatory\nIn any remembrance such as this there is an irresistible urge to list names\nseveral months later and again was spotted intermittently for a month or so\nand events to recall the life and scientific history at the observatory over the\nand then disappeared. This pattern of appearances and disappearances\nyears. One consequence of such listings of course is to omit inadvertently\ncontinued until 1966. Since then, to my knowledge, no one has seen the dog.\nsomeone deserving or some significant event from the litany. Therefore I will\nIts appearances or disappearances were never regular, and at times it was seen\nabandon my strongest desires to list and leave it to someone else. However, I\nat the summit as well as farther down the access road to the observatory. It\nwould be remiss if I did not mention two names. The first is Harry Wexler.\nwould never have anything to do with anyone and whenever pursued would\nDr. Wexler's belief in MLO showed a vision in ground-based baseline\nalways easily outdistance its pursuers over the rough lava and run to the top\nmonitoring that was far ahead of its time, and his support for the observatory\nof the mountain. The staff could never determine where it obtained food when\nwas indispensible for its establishment. The other is Charles D. Keeling, whose\nit was not at the dump (months at a time) in the desolate environment of the\nCO2 monitoring at MLO is the textbook case of proper monitoring for climatic\nmountain nor why, if it did descend the mountain when it was not seen, it did\nchange. Through his tenacious effort and strictest attention to measurement\nreturn to roam the mountain top for months at a time. This was especially\ntechnique the atmospheric CO2 measurements at MLO constitute the best\npuzzling in view of the fact that the staff sometimes discovered lost hunters'\ncontinuous CO2 record in the scientific community.\ndogs wandering close to the observatory, always in the most pitiful condition.\nI am fortunate to have had even a small part in the program at MLO over\nIn every case, starvation and exposure to the elements had just about done in\nthe past 19 years. Knowing the people and being associated with the program\nthese hunters' dogs.\nat the observatory during its growing years, contributing to the work and\nConcerning the belief that the white dog was a messenger of pending\nseeing it mature into a worthwhile endeavor have been gratifying. On this\neruptions, it is true that it was sighted sometimes before an eruption, but it\ntwentieth anniversary I'd like to congratulate the long line of people who have\nwas also sighted many other times when no eruption occurred. The dog did\nparticipated in MLO's history and the many more who will be involved in its\ncreate a problem for the staff in that when a staff member would describe the\nfuture.\nappearances of the dog to visiting scientists or to the public the response\nwould invariably be looks of worry and discomfort or of concern and a fear\nthat this staff member had finally gone stark crazy. The story of the dog was\ndefinitely out of place among scientific endeavors at the observatory, and soon\n23","THE OBSERVATORY\nHoward T. Ellis\nMauna Loa Observatory\nHilo, Hawaii\nAs I remember, I first heard about Mauna Loa Observatory (MLO), where\nroad to the saddle road; when completed it made the drive to MLO much\nI\ncame to spend a large part of my career, in September 1957 through an\neasier and shorter.\narticle in Topics, an official monthly publication for U.S. Weather Bureau\nIn November 1963, word reached us from the U.S. Weather Bureau that\nemployees. Even though I had no intention of going there, I kept hearing\nthe observatory would probably have to be closed down because of lack of\nabout MLO from Jack Pales, who shortly before I joined the Physical Science\nfunds and that everybody would be transferred. Despite this grim outlook\nI\nLaboratory in Washington, D.C., had left that institution to become the first\nwas determined to continue anyway and stated that I would find my own\nphysicist-in-charge at MLO, then an IGY station under the administration of Dr.\nfunds for running the place if it came to that extreme. In February 1964, Jack\nH. Wexler. Jack's frequent calls and letters asking for advice and equipment\nPales left for a new job in Las Vegas, and the administration of the observa-\nprocurement kept us informed about the place.\ntory was turned over to me. To save money, I immediately closed the admin-\nIn 1960 I received an urgent call from Charles Lee, of the U.S. Weather\nistrative office in Hilo and personally hauled all the equipment to the\nBureau personnel office, asking me to take the position of principal assistant at\nmountain. The staff had been reduced to three including myself, and I tried to\nMLO. At that time I was also attending graduate school at American\naccomplish as much as I could by frequently staying on the mountain over-\nUniversity, and naturally the decision was a difficult one. Eventually, I\nnight and working 7-day weeks. On the whole I remember these 2 years as a\naccepted the job and left for Hawaii in February 1961 with Dr. Wexler's\npeak of personal accomplishments and good relations with the Hilo\ninstructions to concentrate on the interpretation of ozone and carbon dioxide\ncommunity.\nmeasurements.\nIn March 1966, Lothar Ruhnke took over the leadership of the observa-\nMy first days on the mountain were quite a shock, since my expectations\ntory until 1968, when another austerity period set in with his departure. Again\nof job conditions had been quite different. I was strongly tempted to ask for a\nI was in charge of the facility until 1970, when new funds became available\nreassignment on the mainland, but after longer deliberations I decided to stay\nthrough the increased interest in air pollution and environmental problems.\nand do my best at the observatory and also to adjust completely to the local\nFortunately, since then, MLO has not had to face any serious financial\nculture, an easy task thanks to the friendliness of Hilo people and their\nproblems, and I hope that this situation will continue many more years.\napparent fondness for newcomers.\nWork routine on the mountain in those days consisted of manning the\nfacility 24 hours a day. Generally, five people, but never fewer than two,\nstayed at the observatory on a staggered schedule. We were all working a 6-\nday week, since observations were also made at a site at the Hilo airport 12\nhours a day. The mountain instrumentation, with the exception of a\ncontinuous carbon dioxide analyzer, was duplicated at the Hilo site. Data\nreduction was done on a day-by-day basis, thus insuring good-quality data.\nThere was little change concerning the work on the mountain until\nJanuary 1963, when Jack Pales, then director, suffered a heart attack, and I\ntook on more responsibilities. My first project was to improve access to the\nobservatory. Until that time we had to use the Kulani road, which was in very\npoor condition. I decided to have a 4-mile road built that linked the Kulani\n24","Hikers ascend Mauna Loa, starting at\nHigher on the trail, a hiker rests by\nthe trail marker, yellow paint on the\nthe observatory.\nlava. The observatory is visible (cen-\nter right).\nFrom even higher, the observatory is\nstill visible (right).\n25","MAUNA LOA OBSERVATORY - 1968\nLothar H. Ruhnke\nNaval Research Laboratory\nWashington, D.C.\nIt was an exciting period. The observatory had struggled through 10 years\nprison camp, and the county provided vehicles and machines free of charge.\nof uncertainties in regard to being recognized as a permanent research station\nEven private donations (e.g., $5.00 from Mr. Pete Beamer) helped. For the 10-\ndespite a well-established record of fine observations unique to the high-\nyear anniversary celebration we could invite the public to drive over a stretch\naltitude tropical environment. But the national awareness of the importance of\nof newly paved road to MLO for an open house. More than 300 guests viewed\nwatching and maintaining our natural atmospheric environment had just\nthe facilities, heard lectures on the problem of man-made contamination of our\nbegun. Slowly but definitely the idea began to form that there was really a na-\nnatural environment, and saw NCAR's coronascope and a movie on the sun's\ntional need for the Mauna Loa Observatory (MLO), that its existence should\neclipse.\nnot depend on occasional support by short expeditions or sporadic investiga-\nBut there were also more subtle changes at the observatory, not easily\ntions of the mountain climate. We started to believe that there always would\nvisible to the public. Usually observatory personnel were busy taking scientific\nbe a Mauna Loa Observatory, and we began to work with confidence on the\nreadings and repairing research gear, whereas data evaluation and data inter-\nfuture of this installation.\npretations were done by other research agencies and universities. The change\nTen years ago, diesel generators were replaced by permanent commercial\ncame when observatory personnel discovered that they themselves were most\npower, telephone service was installed, jeeps and trucks were replaced by\nable to process their own data and publish their own research results without\nsedans, and the long road to the observatory was covered with blacktop to\noutside help. I remember the joy and surprise we all had when Bernard\ndecrease dust contamination of our aerosol and pollution measurements. And\nMendonca's first paper on Mauna Loa's wind circulation was accepted by the\nwhat a job it was just to organize such work. Whose job was it anyway to im-\nJournal of Applied Meteorology. Where did we find the time for this addi-\nprove the road? Federal agencies supported the observatory but felt no respon-\ntional research and still maintain all routine observations? We established a\nsibility for the road. The state judged priorities in road construction in terms\nbase office in Hilo at the university campus and limited visits to the mountain\nof economic returns on investment and asked how many more federal\nobservatory to essential observatory duties. This gave us time at sea level to\nemployees would be hired if better roads were available and when would their\ndiscuss our measurement results.\nstate taxes pay for the construction (I estimated 723 years but kept the answer\nThe remaining problem was how to eliminate administrative work. Fortu-\nconfidential). However, there was also the county, with little funds but a big\nnately, permission was granted to hire a secretary. The first applicant was\nheart to help, and last but not least we had the Hilo Chamber of Commerce,\nJudy Bright and I hired her. And what luck we had with Judy. No more letter\nwhich helped with publicity and an open eye on all developments at the ob-\nwriting while driving up the mountain in an old jeep; no more trouble with\nservatory. And then it happened somehow. Research projects paid for\nour private filing systems; no more missed appointments; and finally, someone\nmaterials, the Federal Highway Department took over construction and super-\nto answer the telephone with a genuine aloha spirit.\nvision, the State of Hawaii provided free labor through the Kulani honor\nThe Hilo office gave us also other advantages. Because of its location at\n26","observatory. A local circulation transports and accumulates insects at the\nthe Cloud Physics Observatory, we were induced to use this research environ-\nmountain top, which inhibit accurate sky observations through foreward-\nment to our benefit. We formed a common research association with regular\nseminars to discuss our research and to have a forum for visiting scientists to\nscattered sunlight.\nI wish MLO continued success in the coming decades and I am very sorry\ngive lectures. Besides that we shared a scientific library and some laboratory\nthat I cannot personally thank all my friends in Hawaii for the wonderful time\nand shop space. It became a period of learning that helped all of us to appre-\nciate the unique research opportunities at MLO. While it is customary for\nI had with them 10 years ago.\nresearchers in the environmental sciences to organize expeditions and field\nexperiments far from home, we practically lived inside a scientific gold mine.\nThe tropical inversion below the observatory level keeps surface-produced\ncontaminants well away from the observatory, so that it is possible to study\nlarge-scale tropospheric characteristics. The tropical weather regime of Hawaii\nand the local circulation are predictable with a high degree of accuracy, and\ncertain weather situations repeat themselves with great regularity. Solar radia-\ntion and nocturnal radiational cooling are very pronounced and lead to easy\nexperiments on the atmospheric radiation budget. I wish the national energy\ncrisis had occurred 10 years ago; then I would have urged Howard Ellis to\nstart research on solar energy. Conditions are ideal at MLO for research on\nphotovoltaics and thermoelectric power conversion. Most solar energy\nconversion systems increase in efficiency with increasing temperature differen-\ntials. High insolation coupled with very low radiational sky temperatures is\nvery favorable for such research.\nBut even 10 years ago, research subjects were plentiful, and many tem-\nporary as well as permanent researchers were attracted to MLO. I remember\nwell a research expedition to the top of Mauna Loa not only to celebrate our\ntenth anniversary but also to celebrate the first anniversary of NCAR's solar\ncoronascope. This tenth anniversary expedition showed us clearly that the\n13,000-ft-high top of the mountain was not necessarily the best place for an\n27","THE VOLCANIC ENVIRONMENT OF MAUNA LOA\nOBSERVATORY, HAWAII\nHISTORY AND PROSPECTS\nJohn P. Lockwood*\nU.S. Geological Survey\nHawaiian Volcano Observatory, Hawaii\nINTRODUCTION\nMauna Loa Observatory (MLO) is located near the summit of one of the\nlargest and historically most active volcanoes in the world. During 118 years\nof written history, between 1832 and 1950, Mauna Loa volcano erupted on an\naverage of once every 31/2 years. Then, after the large eruption of June 1950,\nthe volcano began a period of inactivity that lasted 24 years. During this\nanomalously long period of inactivity, the U.S. Geological Survey's Hawaiian\nVolcano Observatory (HVO) installed a seismometer at the volcano's summit to\nmonitor small earthquakes (1964) and established a small geodetic network to\nmonitor deformation at the mountain's top (1965). However, no significant\nactivity was noted until 1974 when a sharp increase in microseismic activity\nwas accompanied by marked swelling of the summit region. This summit\nexpansion has continued to the present and, coupled with the summit eruption\nof July 1975, has suggested that a large eruption may occur on Mauna Loa\nwithin a few years (Lockwood et al., 1976).\nThe renewed activity necessitated a dramatic increase in HVO's monitoring\ncapability, and observatory crews began to make frequent trips to the volcano\nsummit. Throughout this build-up of activity and in the years before, HVO\n*Publication approved by the Director, U.S. Geological Survey\nAerial view of MLO in the mid- '70s.\nFaintly visible above the observatory\nsite is the signal cable to the seismo-\ngraph in the caldera.\n28","20°N\nscientists and technicians stayed hundreds of person-nights at MLO to prepare\n156°W\nfor predawn treks to Mauna Loa's summit. Thus nearly every HVO staff\n155°\nmember since MLO's founding in 1956 has been a recipient of MLO hospitality.\nMAUNA KEA\nA feeling of closeness exists between MLO and HVO, partly because of mutual\ncourtesies extended over the years and partly because of the realization at HVO\nHUALALAI\nthat the MLO staff is at the \"cutting edge\" of Pele's sword while working at\nMLO, that the MLO staff is vulnerable to certain types of Mauna Loa eruptions,\nand that HVO has a responsibility to MLO to offer the best possible advice on\nRift Zone\n?\nthe eruptive threat to Mauna Loa Observatory. For all these reasons I am\nMAUNA\nhappy to contribute a few words to this volume celebrating MLO's twentieth\nCaldero\nHawaiian Volcano Observatory\nLOA\nanniversary.\nKILAUEA\nGEOLOGIC BACKGROUND\no\n25Km\nMauna Loa can be divided into two parts: an eruptively active north and\nnorthwest side, and an eruptively inactive southeast side (Fig. 1). The two\nEruptively active NW flank\nparts are separated by the southwest and northeast rift zones. At Mauna Loa's\nEruptively passive SE flank\n19°\napex is Mokuaweoweo caldera, the geologic heart of the volcano.\nEruptions on Mauna Loa are of two different types: summit and flank.\n1.\nSummit eruptions are here defined as those originating within Mokuaweoweo\ncaldera, whereas flank eruptions occur in the rift zones or north and northwest\n20°N\nflanks of the volcano. Summit eruptions typically do not occur below about\n3,700 m; flank eruptions typically occur below this elevation. Although more\nnumerous than flank eruptions, summit eruptions produce considerably less\nMAUNA KEA\nlava and tend to be of shorter duration.\nFlank eruptions are themselves separable into two categories: rift and\nHUALALAI\nradial eruptions. Rift eruptions constitute more than 90% of the flank\noutbreaks and are limited to Mauna Loa's two rift zones. Radial eruptions\nB\nC\noccur only on the northwest side of Mauna Loa, on eruptive fissures radial to\nZone\nMokuaweoweo caldera, or inclined at high angles to the rift zones.\nMAUNA\nLOA\nThese different types of eruptions pose potential threats to different parts\nof Mauna Loa, and the volcano can be divided into four areas to indicate\nKILAUEA\nrelative activity (Fig. 2). MLO is situated on the area exposed to the threat\nof\nA\nsummit eruptions or radial flank eruptions. It is not, however, directly\nthreatened by the high-volume rift-type flank eruptions.\nC\nType of Eruptive Threat\nA\nNo appreciable threat\nB\nRadial eruptions only\nFigure 1. Map showing principal\nFigure 2. Map showing approximate\n19°\nRift eruptions only\n19°\nareas of Mauna Loa susceptible to\nstructural features of Mauna Loa\nSummit eruptions\nD\n(plus radial on NW side)\n2\nvolcano.\ndamage by different types of\n156°W\n155°\nMoung Loa Observatory\neruptions.\n29","155°\nVOLCANIC GEOLOGY OF AREA AROUND MLO\nIncomplete\nMLO is located directly downslope from Mokuaweoweo caldera and is\nLava Flow Contact\n1942\ncompletely surrounded by lavas that were erupted from vents either within or\nEruptive Vent\n19° 30'\nradial to the caldera (Fig. 3). On Fig. 3 these lavas have been divided into nine\nRoad\nage groups: four historic flows, four groups of latest prehistorio age (including\n1975\n1975\nLova of July 6, 1975\nfive flows), and an undivided group of older prehistoric lavas that includes\nA\nmany separate flows. The division of the prehistoric lavas into \"latest\n1942\nLava of April 26, 1942\nprehistoric\" and \"older prehistoric\" age groups at 1,000 years is a best guess\n1933?\nLava of December, 1933?\nbased on the amount of weathering compared with weathering of historic\n1942\nIncomplete\n1852\nLava of February 17, 1852?\nflows. However, the division point probably does fall somewhere between 600\nand 2,000 years. Lavas of the younger eight groups consist predominantly of\n1933\nthe aa form (blocky lava rubble); lavas of the older group consists\nLava Flows of Latest Prehistoric Age\npredominantly of pahoehoe (smooth-surfaced lava).\n(<1000yrs b.p.?)\nWithin the area shown in Fig. 3, six eruptive fissures radial to Mokuaweo-\nweo caldera have been identified (Fig. 4), all of them prehistoric. One of these\n1933?\neruptive vents cuts across the east end of the MLO compound, at the entrance\nUndivided Lava Flows of Older Late\n5\nPrehistoric Age\n(1000-5000 yrs b.p.?)\nhairpin curve. The age of this vent is estimated to be 200 to 400 years on the\nbasis of its lavas compared with historic lavas in the area.\no\nIKM\n1852?\nORIGIN OF THE FLOW UNDERLYING MLO\nThe MLO buildings are constructed on a very fresh blocky aa flow,\ntypically with the following modal composition: 22% clinopyroxene, 12%\nplagioclase, 5% olivine, and 61% very fine grained matrix of pyroxene,\nplagioclase, and olivine crystallites set in dark brown glass clouded by opaque\nminerals. This lava was erupted from a 1.2-km-long vent located 3.5 km\nupslope from MLO. The vent consists of an acruate fissure (A on Fig. 3)\noriented perpendicular to the northeast rift zone. A spatter and reticulite\n(pumice) rampart, 3-8 m high, along the fissure attests to a \"curtain of fire\"\n5\nlava fountain that probably reached heights of 20-40 m. Lava was erupted as\n19° 32.5'\nMauna Loa\nfrothy pahoehoe at the vent but changed to aa as it moved rapidly downslope.\n1852\nObservatory\nFarther than 1 km below the vent, only aa is found.\nFigure 3. Generalized geologic map of\nthe area surrounding MLO.\n30","The age of this flow is not known. It is included with lavas of late\nprehistoric age on the reconnaissance geologic map of Hawaii (Stearns and\nA\nMacdonald, 1946, Plate 1) and yet shows no more weathering than known\nnineteenth century lava flows in the same climatic area on Mauna Loa.\nSpecifically, it shows no more weathering than the 1843 aa flow that crosses\nsimilar climatic zones 7 km to the east. In mapping this flow the question\narose whether this lava could have formed during historic time but not have\nbeen recognized owing to its then inaccessible location.\nWhen the written accounts of witnesses to nineteenth century Mauna Loa\neruptions are reviewed, one eruption is notable because its site has not been\npreviously recognized. At 3:30 a.m. on February 17, 1852, citizens of Hilo\nwere awakened by a Mauna Loa summit eruption. Reverend Titus Coan,\nHilo's precise chronicler of Hawaii's volcanic eruptions from 1840 to 1881,\nwrote (Coan, 1852, pp. 219-220) that\nimmense columns of burning lava shot up heavenward to the height of\n300 or 400 feet, flooding the summit of the mountain with light, and\ngilding the firmament with its radiance. Streams of light came pouring\ndown the mountain, flashing through our windows, and lighting up our\napartments so that we could see to read large print.\nIn two hours the\nmolten stream had rolled, as we judged, about fifteen miles down the\nside of the mountain. The eruption was one of terrible activity and\nsurpassing splendor. But it was short. In about twenty-four hours all\ntraces of it seemed to be extinguished.\nHowever, in his autobiography (Coan, 1884, pp. 270-280) Coan stated\nthat the eruption lasted 40 hours and flowed at the apparent rate of 15-20\nmph.\nThree days later a much larger and well-documented flank eruption\nfollowed at the 2,600-m level on the northeast rift zone (Coan, 1852, pp.\n220-224). The earlier eruption of February 17 was generally forgotten, and the\nonly information as to its location are the words of Coan (1852, p. 219) that\n\"it was from the same point, and it flowed in the same line, as the great\neruption which I visited in March, 1843.\"\nFigure 4. Aerial photograph of MLO\nshowing relation to Mokuaweoweo\ncaldera and young lava flows. The\nlines mark locations of eruptive\nvents. The lava underlying MLO was\nerupted from the vent labeled A.\n31","There are two 1843 eruptive vents, but only the upper one could have\nbeen seen from Hilo. The upper 1843 vent had not previously been designated\non any map, but from Coan's lucid description (Coan, 1844, p. 47) the vent\nwas identified during current studies on the northeast rift zone as hill 11,427\n(lat. 19 31.6 N., long. 55 31.8 W.). A subsequent ground visit confirmed\nAZIMUTH\nthat this was indeed the upper 1843 vent known to Coan. (It was later\n245°\n246°\n247°\ndiscovered that Howard Powers had reached the same conclusion many years\nbefore [T. L. Wright, 1976, personal communication].) To reconstruct\n04° 00'\nReverend Coan's 1852 view, it was then necessary to find where he had lived.\nHis house, since demolished, had been located in Hilo on Haili Street just\nSpatter Rampart \"A\" Fig. 3\nbelow the present Haili Church (O. Lyman, 1977, personal communication).\n(1852 Vent?)\nLines of sight were projected from the site of Coan's home to the upper 1843\nvent and to the source vent for the flow on which MLO is constructed (vent A,\nFigs. 3 and 4). From the perspective of Coan's home the MLO vent was indeed\nnearly coincident with the 1843 cone (Fig. 5), even though the cone is located\n6 km to the east and 430 m lower in elevation than vent A. The vertical scale\nis exaggerated four times in Fig. 5, SO that the visual coincidence of the two\n03°45'\n1843 Cone\nvents is even closer than this figure indicates.\nThe MLO flow is only 5 km long, however, SO that if the reconstruction is\ncorrect, Coan's estimate of a \"fifteen mile\" length is in error. There are no\nknown young flows 15 miles long on the northeast rift that could fit the 1852\ndescription. Furthermore, a 15-mile-long flow would indicate an anomalously\nhigh 12-km/hr flow rate if Coan's 2-hour flow duration is correct. Assuming a\n5-km length gives a 2.5-km/hr flow rate-much more reasonable in view of\nhistorically observed flow rates in this area.\nIt thus seems likely, all considered, that the aa lavas underlying MLO may\n03° 30'\nbe the products of the eruption viewed by Reverend Coan on the early\nmorning of February 17, 1852. For those interested, a dramatic summary\ndescription of the 1852 eruptions was published in the Hilo Tribune Herald\n(Allan, 1929).\nFigure 5. Diagram of possible view\nfrom downtown Hilo at 4:00 a.m. on\nFebruary 17, 1852 (azimuths and\nvertical angles are from the site of the\nhome of Reverend Titus Coan).\n32","THE PROSPECTIVE VOLCANIC HAZARD AT MLO\nMLO is located in an area that is exposed to a serious volcanic threat, as\nshown by Figs. 3 and 4. Mullineaux and Peterson (1974, Plate I) placed the\narea encompassing MLO in their zone of highest volcanic risk, zone F. They\ndivided the volcanic hazards contributing directly to this risk as lava flows,\nfalling rock fragments, gases, and particle and gas clouds (Mullineaux and\nPeterson, 1974, pp. 29-40). The risk related to the last three phenomena is\nconsidered negligible at MLO, and lava flows are considered the only hazard\nthat seriously threatens the MLO facilities. Some perspective can be placed on\nthe threat of lava flows by studying the area's history, as recorded by flows\nnear the observatory.\nThe area shown in Fig. 3 encompasses approximately 18 km-. Within this\nmapped area, 21% of the surface rocks are of historic age (on the assumption\nthat the year of the MLO flow was 1852), 23% were erupted in later prehistoric\ntime, and 56% were erupted in earlier prehistoric time. Thus, 21% of the\nsurface rocks were erupted within the 140-year period of written observation.\nThe 140-year historic record (since about 1837) is very short and may well not\nbe typical of the future. Several lines of evidence now suggest that Mauna Loa\nmay have been more active during this period than in the late prehistoric past,\nbut even so, the conclusion is inescapable that MLO is located in an area of\nhigh volcanic risk. Firm statistical limitations cannot be assigned to this risk\nbecause of the lack of knowledge about the prehistoric record and because of\npotential variability in the future.\nThis then is the unfortunate state of our ability to quantify the volcanic\nrisk to MLO. Researchers at MLO can, however, take comfort in two facts:\n(1) the risk to life at MLO is very slight, and (2) MLO is situated on relatively\nhigh ground provided by the 1852(?) lava. Future flows will tend to flow\nMokuaweoweo,\nde, at an eleva-\naround the observatory.","REFERENCES\nAllan, V.M., 1929: Eruption of 1852 one of\nsplendor: Hilo Tribune Herald, October 6,\n1929.\nCoan, Titus, 1844: Untitled letter about 1843\neruption: The Missionary Herald,\n40(2):44-47.\nCoan, Titus, 1852: On the eruption of Mauna\nLoa, Hawaii, February, 1852: Amer. J.\nScience, 2nd ser., 14(40-52):219-224.\nCoan, Titus, 1884: Life in Hawaii: Anson, D.\nF. Randolf & Co., New York, 340 p.\nLockwood, J. P., R. Y. Koyanagi, R. T. Till-\ning, and D. W. Peterson, 1976: Mauna Loa\nthreatening: Geotimes, 21(6):12-15.\nMullineaux, D. R., and D. W. Peterson, 1974:\nVolcanic hazards on the island of Hawaii:\nU.S. Geol. Survey Open-File Rept. 74-239,\n61 p.\nStearns, H. T., and G. A. Macdonald, 1946:\nGeology and groundwater resources of the\nisland of Hawaii: Hawaii Div. Hydrography\nBull. 9, 363 p.\n34","Gas measurements\nCO2 instruments (left) in the new ob-\nservatory building 1978, with meteor-\nological and solar radiation strip\ncharts (right).\nJohn Chin with ozonesonde at the\nNational Weather Service airport sta-\ntion, Hilo, in March 1970. The MLO\nballoon-supported instrument used\nthe standard Weather Bureau radio-\nsonde transmitter and receiver at the\nairport.\n35","fossil fuel was too slight to expect a rise in atmospheric CO2 to be detectable.\nTHE INFLUENCE OF MAUNA LOA OBSERVATORY\nThe idea was again convincingly expressed by Callendar (1938, 1940) but still\nON THE DEVELOPMENT OF ATMOSPHERIC CO, 2\nwithout solid evidence of a rise in CO2.\nRESEARCH\nThe first unmistakable evidence of atmospheric CO2 increase was\nfurnished by continuous measurements made at MLO and by measurements of\nflask samples collected periodically at the South Pole. These data, obtained in\nCharles D. Keeling\nconnection with the International Geophysical Year (IGY), were precise enough\nScripps Institution of Oceanography\nto indicate a rise in concentration in 1959 when compared with the results of\nUniversity of California at San Diego\nthe previous year (Keeling, 1960). Further measurements have shown a persist-\nent year-to-year increase.\nAlong with new observations have come increasingly refined calculations\nof the heating effect of increased atmospheric CO2. One of the most widely\nINTRODUCTION\naccepted climate models emerging from this effort indicates that the earth's\nsurface would warm by 4°C above the present average global temperature for\nThe increasing amount of CO2 in the atmosphere from the burning of\na fourfold increase in CO2, by 6°C for an eightfold increase (Geophysics\nfossil fuels has become a serious environmental concern. Central to this\nStudy Committee, 1977). A rise in CO2 as great as eightfold before coal\nconcern is the question whether a rise in CO2 constitutes a peril to man by\nreserves are exhausted has been predicted using a geochemical model cali-\nraising world temperatures, as many scientists now claim. That a rise in CO2\nbrated by the Mauna Loa and South Pole trends (Keeling and Bacastow,\nis occurring is unquestionable, however. Mauna Loa Observatory (MLO) data\nare providing dramatic evidence of that: they show amounts more than 10%\n1977).\nSuch a high average global temperature has probably not occurred for\nover amounts recorded before the Industrial Revolution, and a rise of 6% in\ntens of millions of years. Accompanying such warming may be shifts in\nthe last 19 years alone.\nrainfall patterns and in agricultural zones. Polar ice may melt or break up and\nNinety-seven percent of the energy demand of the industrial world is met\nlead to coastal flooding (Geophysics Study Committee, 1977). These problems,\ntoday by burning fossil fuels. Even if the industrialized world were to decide\nonce upon us, will not be easily overcome. Once high CO2 levels are reached,\nto shift to other energy sources as rapidly as possible, the annual consumption\nthey will probably decrease only slowly as deep ocean water gradually absorbs\nof fossil fuels would double before the shift was complete. Without such a\nthe excess CO2. Concentrations well above preindustrial levels are likely to\nshift, a peak annual rate ten or even twenty times today's rate may occur\npersist for at least 1,000 years, along with attending climatic problems (Keeling\nbefore fuel reserves, especially coal reserves, are exhausted. Thus a large addi-\nand Bacastow, 1977).\ntional increase in atmospheric CO2 is likely in the next few decades. As\nWhether or not a large CO2 increase will occur and persist depends on the\nRevelle and Suess (1957) wrote, \"Through his worldwide industrialized civil-\nnatural carbon cycle, about which we still know too little. How much CO2\nization, man is unwittingly conducting a vast geophysical experiment. Within\nfrom fossil fuel will remain in the air during the next centuries? How much\na few generations he is burning the fossil fuels that slowly accumulated in the\nwill be taken up by the oceans and by vegetation on land? These questions\nearth over the past 500 million years.\"\ncannot be answered from present knowledge. Sustained monitoring of CO2 at\nThe idea that CO2 from fossil fuel burning might accumulate in air and\nsites such as MLO is an indispensable aid to validate predictions stemming from\ncause a warming of the lower atmosphere was speculated upon as early as the\ncalculations of the behavior of the carbon cycle.\nlatter half of the nineteenth century (Arrhenius, 1903). At that time the use of\n36","Viewed in this context, the reasons for measuring atmospheric CO2 at\nMauna Loa seem compelling. A few of us remember, however, that the\noriginal decision to study CO2 at this remote site was not easily made. Because\nthe story is closely involved with MLO being established in the first place, it\nseems appropriate to recount here some of the human aspects of this story and\nits scientific perspective.\n358\nHISTORY\n310\n310\nThe IGY, which began in 1957, offered scientists for the first time an or-\n330\nganizational setting for study of atmospheric CO2 on a global scale. In view of\n340\nthe importance of knowing whether airborne CO2 was rising worldwide, such\na study was long overdue. The data published before the IGY led to a general\nbelief that CO2 concentrations depended greatly on location with no clear time\n350\ntrend (Bray, 1959). Observations varied from under 200 parts per million\n328\n(ppm) near the North Pole to over 350 ppm in continental air and air near the\n355\nequator (Buch, 1948). Owing to this apparent spatial variability, a whole\n325\nnetwork of stations was deemed necessary to detect any significant global\nH\n352\ntrend.\n341\nIn the early 1950's, Carl G. Rossby suggested that Stockholm University's\n363\nMeteorological Institute, which he directed, should participate in an extensive\n331\n336\ninvestigation of trace chemicals in the atmosphere as a prelude to the IGY. At a\nconference held on the subject in 1954, participants decided to plan for a\n330\nworldwide network of CO2 monitoring stations, possibly including a site in\nthe Hawaiian islands (Eriksson, 1954). Responsibility for setting up stations in\n333\n340\nthe Pacific region fell to Wendel Mordy, a conference member and chief\n329\n350\nmeteorologist of the Pineapple Research Institute in Honolulu.\nWhen I learned that Mordy was interested in measuring atmospheric CO2\n361\nin the Pacific region, I informed him of CO2 studies I had begun in 1955 while\nat the California Institute of Technology. In contrast to previous studies, I had\n350\n331\nfound practically constant atmospheric CO2 in turbulent air near midday.\nMeanwhile, CO2 monitoring had just begun in Scandinavia under the\ngeneral direction of Kurt Buch of Finland. The Scandinavian data (Fig. 1) re-\n360\nsembled past work, with greatly varying CO2 concentrations - even though\nFigure 1. Concentrations of\natmospheric CO2 over Scandinavia\n(ppm) on February 20, 1955. The ap-\nproximate pattern is shown by\ncontour lines in a manner similar to\nBischof's (1960, Fig. 6). Concentra-\ntions were determined by absorption\nof CO2 in barium hydroxide solution.\n37","recording CO2 gas analyzer on Mauna Loa since it would be possible to live\nspecial care was being taken to sample in open areas away from local in-\nthere and tend the analyzer as necessary. As far as I knew, no one had ever\nfluences (Fonselius et al., 1955). My daytime CO2 results were close to the\nbefore suggested measuring atmospheric CO2 continuously. Wexler asked a\nScandinavian means, but the variability was far less - even though I had\nnumber of questions in rapid-fire, covering both the scientific and the prac-\ntaken special care to sample in densely vegetated areas where local influences\ntical. He was especially interested in costs. We went SO far as to discuss setting\nwould predominate. Specifically, I had found that everywhere I went the air a\nup a second continuous CO2 analyzer in Antarctica. Then the interview was\nfew tens of meters from the plants on sunny days tended to reach a nearly\nover. Altogether it took almost exactly 15 minutes, as scheduled. Wexler had\nconstant CO2 level of about 315 ppm (Keeling, 1958). In an attempt to under-\nmade up his mind to press for CO2 measurements at Mauna Loa using monies\nstand why, I took measurements in some exposed windy areas away from\nwhich he hoped would be made available by the participation of the United\nplants: at high elevation in the White Mountains (Fig. 2) and Sierra Nevada of\nCalifornia, on ocean beaches, and over ocean water near the equator (Keeling,\nStates in the IGY.\nDuring this same spring of 1956 the oceanographic community was\n1961). All these data were also near 315 ppm. I concluded that the CO2 in air\nmaking plans to participate in the IGY. Roger Revelle, as director of the\nhad a characteristic background concentration, at least near the west coast of\nScripps Institution of Oceanography, was a leader in this effort. Revelle had\nthe United States and Central America where I had sampled. Evidently, on\nan intimate knowledge of the natural CO2 cycle going back to his student\nsunny days this background level prevailed even near plants.\ndays, and he wanted to make sure that man's \"vast geophysical experiment\"\nThus I became concerned that the proposed measurements in Hawaii and\nwould be properly monitored and its results analyzed. Revelle believed that a\nelsewhere might not be accurate enough to establish this background CO2\nCO2 program should include ocean water studies as well as atmospheric\nlevel. Although Mordy soon decided not to participate in CO2 studies, my\nmeasurements. With this in mind and with Wexler's concurrence, he arranged\nconcerns reached the attention of Harry Wexler, Director of Research for the\nfunding for a laboratory for CO2 measurements at Scripps, and I was invited\nU.S. Weather Bureau. Wexler was a friend of Rossby and an ardent supporter\nto run it. Although it had not been decided precisely what kind of CO2\nof broadly based meteorological studies. He invited me to Washington early in\nprogram should be implemented as part of the United States IGY effort, I\n1956.\naccepted his offer.\nThe Weather Bureau already had a small wood frame hut near the\nWexler's support of continuous measurements of atmospheric CO2 at MLO\nsummit of Mauna Loa where some simple automatic instruments were housed.\nwas a bold decision not widely accepted at the time. Wexler knew that I had\nIn 1955, at Wexler's urging, plans were underway to construct a larger, more\nlocated a manufacturer of nondispersive infrared CO2 gas analyzers, but he\npermanent structure where people would live and tend more complicated in-\nalso knew that I had not yet been able to test such an analyzer. Even the firm\nstruments. During my interview with Wexler, which I recall began promptly at\nitself did not claim that its infrared analyzer was accurate enough for the task.\n8:00 a.m., I talked to him about the possibility of setting up a continuous\n320\nInyo Mountains, Calif.\n315\n9\n10\n11\n12\n13\n14\nMarch 1956\n38","It had been designed principally for industrial uses which did not demand high\ntalked about plans for an ambitious instrument-based United States program.\naccuracy. I was relying on the judgment of one of the firm's engineers that the\nIronically, I had SO far obtained CO2 data using quite inexpensive devices\ndevice was inherently very sensitive and stable. The firm couldn't even lend\n- glass sampling flasks, a liquid nitrogen cooled freeze-out trap, a mercury\nme one to test. The basic instrument was expensive and required costly addi-\ncolumn manometer. But my manometric method could not be used for a large\ntional equipment to operate as an air monitor at a remote field station. Refer-\nprogram because a single sample took over an hour to analyze. The infrared\nence gases to calibrate the instrument did not exist.\ngas analyzer was needed to speed up the work without sacrificing high ac-\nTo most of the IGY planners who heard about the CO2 infrared analyzer\ncuracy.\nLate in the summer of 1956 I arrived at Scripps to begin implementing the\nscheme in 1956, such expensive and complicated equipment seemed unneces-\nsary. Both the earlier published data and the new Scandinavian data, appear-\nnew U.S. atmospheric CO2 program. In all, four gas analyzers were pur-\nchased. One was hastily outfitted for Antarctic field work. Shipment to Little\ning in print every 3 months, proved that atmospheric CO2 variations were SO\nAmerica couldn't be delayed. This first venture turned out, in fact, to be too\nlarge that traditional methods of chemical analysis would always remain ade-\nquate. I distrusted these variable data, but my distrust was based on no more\nhasty. No useful data were obtained at Little America until the second Ant-\nthan a few hints from my own data. The most important of these was the near\narctic field year in 1958.\nconstancy of CO2 over five days for samples taken at 3,500 m in the White\nAs soon as the Antarctic shipment was off - on the same vessel that was\nMountains. Wexler had been especially impressed by the White Mountains\nto have carried Admiral Byrd to Antarctica, had he been able to go - I began\nrecord (reproduced in Fig. 2). He felt that if this record was typical of back-\nsystematically to test the new analyzers. In March 1957, continuous measure-\nground air, high measurement accuracy at a site on Mauna Loa just might pay\nments of air began at Scripps. Soon afterward I assembled another apparatus\nfor Mauna Loa. But there were numerous delays and problems with the air-\noff in the IGY program.\nRevelle soon agreed to the new infrared analyzer method, but he preferred\ncraft and shipboard programs. These delays were especially bothersome\na network of measuring locations in which such analyzers would be used to\nbecause the IGY had already begun. Soon it would be over, and ships and air-\nanalyze air collected in flasks, from ships and aircraft for example.\ncraft would not be available.\nRossby remained dubious. I had a chance to meet him just once at an IGY\nAs it turned out, when the equipment for Mauna Loa was ready, I\nplanning meeting at Scripps during 1956. Someone pointed me out to him\ncouldn't install it. Revelle insisted that I give first attention to aircraft and\nacross a grass lawn during a recess. As he walked up to greet me, he remarked\nshipboard sampling, and the aircraft program was not yet underway. He rein-\nfor the benefit of some nearby acquaintances, \"Ah\nforced his view of the matter by refraining from signing my travel orders to\nza yong man wiz za\nmachine.\" He seemed upset at this abrupt new American plan to buy\nvisit Mauna Loa. As the IGY approached its July 1958 ending date, Wexler\nbecame very anxious about Mauna Loa. At length he took action himself and\nexpensive gadgetry to measure CO2. His skepticism became obvious as we\nFigure 2. Variation in atmospheric\nCO2 over barren ground near White\nMountain Research Station in the\nInyo Mountains of California during\nMarch 1956 (adapted from Fig. 2 of\nKeeling, 1961). Concentrations were\ndetermined manometrically from\nliquid nitrogen temperature\ncondensates. The arrow identifies the\nminimum concentration plotted in\nFig. 4, accepted as representative of\nwest coast U.S. air.\n39","sent to me Ben Harlin, the meteorologist who had operated the CO2\nwinter temperature inversions. The maximum at Mauna Loa occurred in May\nequipment at Little America in 1957. With help from Jack Pales, the first\njust before temperate and boreal plants add new leaves. The seasonal pattern\ndirector of MLO, Harlin installed the analyzer at MLO in March 1958 without\nwas highly regular and almost exactly repeated itself during the second year of\nmy assistance. To our great surprise, on the first day of operation it delivered\nmeasurements at Mauna Loa. Thus there was no need to wait for statistical\nwithin 1 ppm the CO2 concentration that I had told Harlin to expect on the\nstudies to prove the reality of the oscillation as would have been required had\nbasis of my earlier manometric data and preliminary test data obtained at\nless exact chemical methods been used. I soon reviewed my 1955-1956\nScripps.\nmanometric data and discovered that they showed a similar seasonal variation\nOf course this agreement was an accident. The mean of the daytime\n(Bolin and Keeling, 1963).\nmanometric and Scripps data just happened to be close to the value typical for\nNo one had expected to determine the long-term rate of rise in CO2\nthe month of March. Indeed, the next month's data did not agree - the con-\nduring the IGY even though establishing the rise was the principal purpose of\ncentration rose by over one ppm. The following month's mean concentration\nthe program. Revelle and others had expected that the IGY program at best\nwas still higher. Electrical power failures then shut down the equipment for\nwould furnish a reliable \"baseline\" CO2 level which could be checked 10 or 20\nseveral weeks. When measuring resumed in July, the concentration had fallen\nyears later, after the rise in CO2 was large enough to stand out against local\nbelow the March value. I became anxious that the concentration was going to\nvariability. But because of the regularity of the seasonal variation at Mauna\nbe hopelessly erratic, especially when the computed concentration fell again in\nLoa a rough estimate of the long-term rise was possible after only two years\nlate August. Then there were more power shutdowns.\n(Bolin and Keeling, 1963).\nFinally, after my first visit to Mauna Loa in November, the concentration\nFortunately, funding for CO2 measurements at MLO was continued after\nstarted to climb steadily month by month. Gradually a regular seasonal\nthe IGY. By early 1962 it was possible to deduce that approximately half of the\npattern began to emerge: we were witnessing for the first time nature's\nCO2 from fossil fuel was accumulating in the air and that a sink must be\nborrowing of CO2 for plant growth during the summer and returning the loan\ncarrying a substantial fraction away (Keeling, 1960). Revelle and Suess (1957)\neach succeeding winter. Earlier published data for Europe also showed a\nhad predicted that much of the CO2 from fossil fuel would be absorbed by the\nseasonal trend of sorts (Bray, 1959), but the maximum concentration, arrived\noceans. The earlier published CO2 data had argued against their view,\nat statistically from a highly irregular pattern, was in January, a time of year\nhowever, because the rise in CO2 seemed to be close to that predicted if all of\nwhen CO2 from burning is likely to accumulate near the ground because of\nthe CO2 from fossil fuel accumulated in the air. This latter conclusion was\nreinforced in 1958 after several years of the Scandinavian network data\n336\nbecame available (Callendar, 1958). But after four years of measurement\nat\nMauna Loa the question was settled in favor of the Revelle-Suess prediction.\n332\n328\n324\n320\nFigure 3. Monthly average\nconcentrations of atmospheric CO2 at\nMLO since the beginning of\n316\nmonitoring in 1958. Concentrations\nwere determined with a nondispersive\n312\ninfrared gas analyzer as described by\n1960\n1965\n1970\n1975\nKeeling et al. (1976a), p. 539.\nyear\n40","sea breezes. The CO2 record was twice interrupted for several months when\nAs the Mauna Loa record has been further extended, additional interesting\nfeatures of the long-term trend have revealed themselves. These include pertur-\noceanographic work was in progress, but a nearly unbroken continuous record\nbations that appear to correlate with the trade winds and with sea surface tem- exists from April 1958 to June 1960. Since the Mauna Loa analyzer was op-\nperature (Bacastow, 1976; Machta et al., 1976; Newell and Weare, 1977). The\nerating during this period, these data, and a few more in 1962, are useful in\nseasonal pattern has also been scrutinized to see if variations in amplitude\nadjusting the La Jolla record to a common basis with Mauna Loa.\nfrom year to year are meaningful. So far the pattern is too regular to reveal\nMost of the 1955-1956 manometric data reflect local CO2 emanating from\nsignificant variations (Hall et al., 1975). Now after nearly 20 years of measure-\nplants and soil. The minimum values for each location, occurring typically\nments, the Mauna Loa record (Fig. 3) appears as a natural yearly cycle gradu-\nnear midday, as already noted, may not have been markedly influenced by\nally being dwarfed by a long-term rise - a dramatic example of inadvertent\nplant activity, however. A plausible reason for this is that the sampling loca-\ninfluence by man on his environment.\ntions I had chosen were in wild areas which had never been disturbed very\nmuch by humans. In wild areas the photosynthetic withdrawal of atmospheric\nCO2 by the plants and the release of CO2 by plant respiration and decomposi-\nTHE WEST COAST DATA\ntion should not differ greatly. The net change in the CO2 concentration of the\nEven though the manometric CO2 data obtained shortly before the IGY\nlocal air should therefore be relatively small, especially if air turbulence,\nplayed a prominent role in deciding the strategy of the United States CO2\ntypically maximal at midday, further diminishes the net effect.\nprogram, they had never been compared with the infrared CO2 data for\nAt several control sites on ocean beaches and barren mountains, where I\nMauna Loa. Until a pressure broadening correction was recently applied to the\nalso sampled during 1955 and 1956, the CO2 concentrations usually agreed\nlatter data (Keeling et al., 1976a), a precise comparison was not possible. It\nwith the minimum values found near plants. For example, in Yosemite\nseems worthwhile now to review these earlier measurements and to recon-\nNational Park in June 1955, the lowest value found for forest air was 316.2\nstruct, as closely as possible, the global concentrations of CO2 back to 1955.\nppm; a few miles away over barren terrain near Lake Tenaya, I found 315.9\nThis reconstruction is greatly aided by additional infrared measurements\nppm (Eriksson, 1954).\nof CO2 obtained between 1957 and 1962 at La Jolla, California. Although\nThe minimum CO2 concentrations for all CO2 sites in the western United\nthese data were obtained as a by-product of instrument testing, they are never-\nStates are listed in Table 1 and plotted in Fig. 4, except that data have been\ntheless a useful record of air from the same general geographic area as the\nomitted if the humidity was not measured, since for these data it is impossible\nearlier manometric data. Except for a few days when air was sampled from a\nto determine the CO2 concentration versus dry air. Most of the measurements\nlaboratory window, all measurements were made near the end of a 1,000-foot\nwere obtained in California, but a few were obtained farther north in the state\nocean pier where the air was often free of local disturbances, at least during\nof Washington and several from Arizona.\n41","Table 1. Minimum atmospheric carbon dioxide concentrations (relative\nto dry air) by direct manometric analysis, for various sites near the\nwest coast of the United States and Central America\nLocation\nElevation\nDate\nLocal\nMinimum CO2 Type of Site\n(above sea level)\nTime\nConcentration\n(ppm)\n1955\nBig Sur State Park\n70 m\nMay 18\n12:15\n319.3\nforest\n(36 °N., 122 °W.)\nYosemite National Park\n2500 m\nJune 2\n12:30\n316.2\nforest\n(38°N., 119 W.)\n3000 m\nJune 3\n10:00\n315.9\nbarren ground\nOlympic National Park\n170 m\nSept. 7\n13:30\n312.6\nforest\n(48°N., 124 °W.)\n313.8\nocean beach\n0 m\nSept. 7\n15:15\nGulf of Tehuantepec\n314.4\nover ocean\n(9°N., 89 °W.)\n10 m\nDec. 1\n5:30\n314.9*\n1956\nBorrego Valley, California\nbarren desert\n340 m\nFeb. 1\n10:30\n314.1\n(33°N., 116 °W.)\nbarren snow field\nInyo Mountains, California\n3800 m\nMar. 10\n20:00\n316.2\n(37°N., 118 °W.)\nOrgan Pipe Cactus National Monument\n550 m\nApr. 22\n0:00\n316.1\ndesert brush\n(32°N., 113°W.)\nMay 16\nHoward Pocket, Arizona\n2100 m\n15:30\n317.4\nforest\n(35°N., 112°W.)\nMay 18\nTelephone Hill, Arizona\n2600 m\n15:00\n320.0\nforest\n(37 °N., 112 °W.)\nBig Sur State Park\n70 m\nJune 6\n12:00\n318.4\nforest\n(36°N., 122°W.)\nforest\nYosemite National Park\n2500 m\nJune 11\n12:00\n316.4\n(38°N., 119°W.)\n* *Adjusted to 33 °N.\n42","as a e e A A\nthe minimum CO2 concentration from a suite of samples collected aboard ship\noff the coast of Nicaragua near 9°N, in 1955. This minimum has been adjusted\nupward by 0.5 ppm on the basis of the average latitudinal gradient found by\nBolin and Keeling (1963) between 9° and 33 °N for the appropriate month of\nsampling.\nThe continuous measurements obtained at La Jolla from 1957 to 1962 are\nhighly contaminated by local and regional urban sources of CO2. Even the\ndaily minima, which usually occurred during sea breezes, vary considerably\ndepending on the history of the air. Highest values typically occurred when\nthe air had previously passed near the city of Los Angeles to the northwest.\n320\nTo reduce further the influence of contamination, the daily minima were ar-\nTH\nranged into calendar weeks, and weekly minima were identified. As noted\nalready in 1960 (Keeling, 1961), these weekly minima scatter much less than\nBS\nBS\nthe dailies. Also, unlike the dailies their monthly means show a consistent\ntrend suggestive of uncontaminated air.\nHP\nThese monthly means are listed in Table 2 and plotted in Fig. 5. One\nIM\nentry, for June 1958, is omitted from further consideration because only one\nYF\nYF\nweekly minimum was obtained that month. Also, as indicated in the table, a\nOF\nfew obviously contaminated minima were omitted in assembling the monthly\nYB\n315\nmeans. The means for April 1958 through March 1960 have been published\nGT\n(Keeling, 1961). These, and previously unpublished data for 1957, 1960, and\n1962, are here reported according to the 1974 manometric CO2 mole fraction\nBV\nOB\nscale, using formulas for conversion from an adjusted index scale (Keeling et\nal., 1976a).\nThe manometric and infrared data (Figs. 4 and 5) display a seasonal\nOF\nvariation similar to but of greater amplitude than that for Mauna Loa. The\n310\n1955\n1956\nBS, Big Sur; YF, Yosemite forest; YB,\nFigure 4. Minimum concentrations of\natmospheric CO2 at various sites near\nYosemite barren ground; OF,\nthe west coast of the United States\nOlympic forest; OB, Olympic beach;\nduring 1955 and 1956. Concentrations\nGT, Gulf of Tehuantepec; BV,\nwere determined manometrically from\nBorrego Valley; IM, Inyo Mountains;\nliquid nitrogen temperature conden-\nOP, Organ Pipe; HP, Howard\nsates. Sites are identified as follows:\nPocket; TH, Telephone Hill.\n43","Average date\nNo. of weekly\nAverage observed\nminima included\nCO2 concentration\n(ppm)\nTable 2. Mean of weekly minimum concentration of atmospheric carbo\n1957-Mar. 24\ndioxide (relative to dry air) by infrared gas analysis, for Scripps pier,\nApr. 17\n4\n315.91\nLa Jolla, California, at 33°N, 117°W, elevation 8 m\nMay 16\n4\n315.92\nJune 16\n4\n315.43\nSept. 12\n5\n310.16\nOct. 4\n2\n311.15\n1958-Apr. 21\n3*\n316.72\nMay 16\n3\n317.71\nJune 22\n1\n319.08*\nJuly 17\n5\n313.52\nAug. 18\n4\n310.83\nSept. 13\n4\n311.08\nOct. 19\n3*\n313.17\nFigure 5. Monthly averages of\nNov. 20\n5\n315.64\nweekly minimum atmospheric\nDec. 20\n4\n316.76\nconcentration at La Jolla, Calif\n1959-Jan. 19\n3+\n317.38\nConcentrations were determine\nFeb. 14\n4\n316.89\na nondispersive infrared gas an\nMar. 18\n5\n317.89\nApr. 18\n4\n317.52\nMay 17\n4\n317.52\nJune 14\n4\n317.65\nJuly 14\n5\n313.95\nAug. 16\n4\n310.52\n320\nSept. 12\n4\n311.14\nOct. 13\n5\n314.58\nNov. 8\n2*\n316.19\nDec. 16\n5\n315.95\n1960-Jan. 16\n3*\n317.19\nFeb. 14\n4\n317.94\nMar. 17\n5\n317.95\nApr. 16\n4\n319.82\n315\nMay 12\n3\n320.57\nJune 12\n2\n318.58\n1962-Mar. 25\n2\n320.43\nApr. 13\n4\n320.06\nMay 6\n2\n321.07\n* One weekly minimum omitted from average.\n310\n1957\n1958\n1959\n1960\n1961\n19\n44","seasonal variation, however, is clearly evident only for the La Jolla data\nbecause the 1955-1956 manometric data involve so many missing months and\nextend over less than two years.\nSeveral of the manometric data appear to be inconsistent with the\nseasonal trend. That the two CO2 minima for Big Sur State Park may be too\nTable 3. Adjusted manometric data, 1955-1956, and infrared gas analyzer\nhigh, both in 1955 and 1956, is not surprising because sampling was done in a\ndata, 1957\npublic campground where daytime automobile traffic may have produced\nseveral ppm of contamination. Also, the CO2 minimum for Telephone Hill,\nConcentration (ppm)\nArizona, seems too high relative to Howard Pocket, but there is no obvious\nLocation\nMonth\nDeparture\nAdjusted\nAdjusted\nand\nfrom linear\nreason, since the site was in a remote forest north of the Grand Canyon.\nto 15th\nfor seasonal\nYear\ntrend\nFinally, the pair of CO2 minima for the Olympic National Park agree with\nof the month\nvariation\neach other but are both considerably higher than had been expected for the\nBig Sur State Park\nMay 1955\n319.36*\n316.40\n3.46\nmonth of sampling on the basis of the La Jolla data, again for no obvious\nYosemite National Park\nat 3000 m\nJune\n315.05\n313.72\n72\nreason.\nBefore deciding on the disposition of these possibly contaminated values,\nOlympic National Park\nan adjustment of the data was made to the 15th of the month of sampling in\nat 170 m\nSept.\n312.81*\n317.16\n3.98\nGulf of Tehuantepec\nDec.\n315.36\n313.94\n0.59\norder to reduce scatter resulting from uneven spacing in time. The adjustments\nBorrego Valley\nFeb. 1956\n314.06\n312.92\n-0.55\nwere made following a procedure described previously (Keeling et al., 1976b).\nInyo Mountains\nMar.\n316.35\n314.75\n1.23\nSpecifically, the individual monthly concentrations X(t), in ppm, where t de-\nOrgan Pipe Cactus\nnotes the time in years after January 1, 1955, were fit by the method of least\nNational Monument\nApr.\n315.91\n313.23\n-0.35\nsquares to an oscillatory-linear trend function:\nHoward Pocket\nMay\n317.42\n314.46\n0.82\nTelephone Hill\nMay\n320.06*\n317.10\n3.46\nX(t) = Q1 sin 2nt + Q2 cos 2nt + Q3 sin 4ut +\n(1)\nBig Sur State Park\nJune\n317.74*\n316.41\n2.71\nQ4 cos 4xt +Qs+Q6t\nYosemite National Park\nThe four possibly contaminated data mentioned above were tentatively\nat 2500 m\nJune\n316.08\n314.75\n1.05\nomitted from the computation. The parameters of best fit were found to have\nLa Jolla\nMar. 1957\n315.46\n313.86\n-0.36\nthe values:\nLa Jolla\nApr.\n315.85\n313.17\n-1.11\nLa Jolla\nMay\n315.94\n312.98\n-1.35\nQ1 =\n2.86883 ppm\n= 6.64806 ppm\nLa Jolla\nJune\n315.51\n314.18\n-0.21\nQ2\n(2)\n0.879716 ppm\n= 312.684 ppm\n=\nLa Jolla\nSept.\n310.26\n314.61\n0.16\nQ3 = -1.51123 ppm\nQ6 = 0.6954 ppm yr\nLa Jolla\nOct.\n312.03\n314.47\n-0.04\nOn the basis of equations (1) and (2), the data, including the tentatively\nJudged to be contaminated.\nrejected values, were adjusted to the 15th of the month as listed in Tables 3\nand 4. Next, the data were seasonally adjusted using the first four terms of\n45","equation (1), and the resulting trend data were plotted as shown in Fig. 6.\nFrom this plot it becomes clear that the questionable values, shown as crosses,\nshould be rejected. A statistical computation bears this out: the four values\ndiffer by factors of 3.4 to 4.9 times the root mean square departure of the\nremaining 13 data points for 1955-1957 with respect to equations (1) and (2).\nTable 4. Comparison of atmospheric carbon dioxide concentrations at\nThe next step was to establish from overlapping data the difference in\nLa Jolla with the long-term trend in concentration at MLO\nseasonal variation and long-term trend for Mauna Loa and La Jolla. First,\nfrom the entire Mauna Loa record of monthly averages from March 1958\nMonth\nConcentration\nTrend at\nDifference\nthrough December 1976, the average seasonal variation and seasonally ad-\nat La Jolla*\nMauna Loa*\njusted trend for that station were established.\nApr. 1958\n316.54\n315.34\n1.20\nSeveral methods have been used previously to separate the long-term\nMay\n317.73\n315.40\n2.33\ntrend at Mauna Loa from the associated seasonal variation (Bacastow, 1977).\nJune\n315.47\n-\n-\nHere I have chosen to express the trend by a cubic spline function (Reinsch,\nJuly\n313.72\n315.53\n-1.81\n1967) and the seasonal variation as an average of the monthly mean\nAug.\n310.96\n315.59\n-4.63\nSept.\n311.15\n315.64\nconcentrations after subtracting the trend. Since the two features are not\n-4.49\nOct.\n312.82\n315.69\n-2.87\nuniquely separable, an iterative procedure was used. First, an estimate of the\nNov.\n315.30\n315.74\n-0.44\nlong-term trend was found assuming a linear increase with time, and a prelimi-\nDec.\n316.67\n315.78\n0.89\nnary estimate of the seasonal variation was obtained. Then consistent with this\nJan. 1959\n317.42\n315.83\n1.59\nseasonal variation, the original monthly values were seasonally adjusted, and\nFeb.\n316.89\n315.88\n1.01\na cubic spline function was passed through the adjusted data points. Further\nMar.\n317.79\n315.93\n1.86\niterations were carried out until the adjusted values approached constancy.\nApr.\n317.43\n315.99\n1.44\nThis convergence was rapid, and because of the high regularity of the seasonal\nMay\n317.56\n316.05\n1.51\nvariation, the seasonal variation found was similar to that found by using a\nJune\n317.57\n316.13\n1.44\nleast squares fit based on equation (1).\nJuly\n313.85\n316.20\n-2.35\nNext, as shown in Table 4, the long-term trend for Mauna Loa, expressed\nAug.\n310.57\n316.29\n-5.72\nSept.\nas a spline function, was compared with the La Jolla data adjusted to the 15th\n311.24\n316.38\n-5.14\nOct.\n314.75\n316.46\n-1.71\nof each month. For the relatively short period of the comparison it seems\nNov.\n316.72\n316.55\n0.17\nreasonable to assume that the long-term trends for Mauna Loa and La Jolla\nDec.\n315.93\n316.63\n-0.70\ndiffer by only a constant. On the basis of the monthly differences between the\nJan. 1960\n317.20\n316.70\n0.50\nMauna Loa trend and the La Jolla data (last column of Table 4), mean differ-\nFeb.\n317.94\n316.77\n1.17\nences between stations were determined for each month. The sum of these dif-\nMar.\n317.88\n316.84\n1.04\nferences is -0.42 ppm; that is, the La Jolla weekly minima, on average, are\nApr.\n319.79\n316.90\n2.89\nlower by that amount than the Mauna Loa trend. Since the expected latitu-\nMay\n320.52\n316.96\n3.56\ndinal difference between stations according to aircraft and shipboard data\nJune\n318.34\n317.01\n1.33\nanalyzed by Bolin and Keeling (1963) is -0.20 ppm, the weekly minima agree\nMar. 1962\n320.08\n318.37\n1.71\nclosely with expectations in spite of the high degree of selection involved in\nApr.\n320.13\n318.43\n1.70\nobtaining them. Evidently, the large irregular variations in the original La Jolla\nMay\n321.02\n318.49\n2.53\nrecord are almost solely owing to high values, probably produced by urban\nAdjusted to the 15th of the month.\nDetermined for the 15th of the month from a spline fit of the seasonally adjusted monthly\nsources.\nmeans for 1958-1976, inclusive.\nNext, from the west coast data, 1955-1962, a long-term trend and an\naverage seasonal variation were found in the same manner as that just\n46","described for the Mauna Loa record. Because of the considerable gaps in the\ndata, the trend in all iterations was assumed to be a straight line. The final\ntrend, shown in Fig. 7, obeys the relation\nX(t) seasonally adjusted = Q5 + Q6t\n(3)\nwhere\nQ5 = 312.592, Q6 = 0.7167 ppm yr`\n(4)\n320\nand, again, t = 0 for January 1, 1955.\nThe corresponding seasonal variation, shown in the third column of Table 5,\nagrees closely with that obtained (see the second column) by comparing the La\nJolla data for 1958-1962 with the Mauna Loa spline function trend. The only\nmonth where the agreement is possibly unsatisfactory is December which\nincludes the data point from 9°N. This discrepancy does not appear to be\nsignificant, however, in view of the scatter of the other 1955-1956 data.\nEvidently it makes little difference which seasonal variation is used in\n315\nfurther analysis. Since the seasonal variation based on the entire data set from\n1955 to 1962 results in slightly lower scatter, I chose this representation.\nTo express the comparison of the pre-1958 U.S. data with the Mauna Loa\nrecord, I have devised what I call \"proxy\" data. My goal is to produce, with\nthe least interpretive adjustment, a set of monthly values valid for Mauna Loa\nfor 1955 through 1957. On the basis of the difference between the seasonal\nvariations for Mauna Loa and the west coast U.S. data for 1955-1962, with\ndue regard for the average difference of 0.42 ppm between locations (Table 6\n310\nand Fig. 8), the west coast U.S. data were converted to equivalent Mauna Loa\n1955\n1956\n1957\n1958\n1959\n1960\n1961\n1962\nmonthly means. In this way the scatter is included, and no judgment of the\nlong-term trend is placed on these early measurements. The results are listed in\nTable 7 and plotted in Fig. 9. Finally, a long-term trend line was established\nFigure 6. Long-term trend in the\nfor the seasonally adjusted Mauna Loa record including these proxy data\nminimum concentration of atmos-\n(Table 8 and Fig. 10). The previous iterative method was again used to\npheric CO2 near the west coast of the\nseparate the trend from the seasonal variation. Since the Mauna Loa record\nUnited States based on the data of\nalready includes 19 years of direct data, the new data have a negligible effect\nFigs. 4 and 5 after adjustment of each\npoint to the 15th of the month of\non the computed seasonal variation for Mauna Loa, shown in the third\nobservation. The seasonal variation,\ncolumn of Table 6. Also, since the spline function at any part of the record is\nexpressed by the first four terms of\nequation (1) as a harmonic function\nwith 6- and 12-month terms, was sub-\ntracted to obtain seasonally adjusted\nconcentrations shown as dots.\nPossibly contaminated data are indi-\ncated by crosses. The straight line is a\nleast-squares best fit through the\nplotted points.\n47","Table 5. Seasonal variation in atmospheric carbon dioxide near U.S.\nsensitive only to nearby data, the inclusion of the early data affects the trend\nwest coast determined by summing monthly concentrations with the\nline only near its former beginning in 1958.\nlong-term trend removed\nThe small difference of 0.42 ppm between the La Jolla and Mauna Loa\ntrends where they overlap suggests that the La Jolla weekly minima are not\nbiased, but actually one need not make this assumption in accepting the proxy\nMonth\nLa Jolla\nWest Coast\nDifference\ndata, provided that the original west coast minima for 1955-1957 have the\n1958-1962*\n1955-1962\nsame bias as those for 1958-1962. This appears reasonable for 1957 because\nJan.\n1.47\n1.46\n.01\nthe data are for the same location as the 1958-1962 data and were selected in\nFeb.\n1.51\n1.22\n.29\nthe same way. Indeed, as can be seen from Fig. 10, the seasonally adjusted\nMar.\n1.96\n2.05\n-.09\nproxy data for 1957 appear to be consistent with the direct data (1958 and\nApr.\n2.23\n2.18\n.05\nlater) both as to scatter and trend. Thus one is encouraged to accept the 1957\nMay\n2.90\n2.87\n.03\nproxy data as reliable.\nJune\n1.81\n1.87\n-.06\nJuly\nOne is less confident that the 1955-1956 proxy data are unbiased. Their\n-1.66\n-1.70\n.04\nAug.\n-4.76\n-4.79\n.03\nscatter is greater, and a backward extrapolation of the relatively steep trend\nSept.\n-4.40\n-4.37\n-.03\nline for 1958 suggests that they could be too high by as much as 1.0 ppm. On\nOct.\n-1.87\n-2.11\n.24\nthe other hand, the rise and fall in trend indicated by the spline function for\nNov.\n0.28\n0.28\n.00\n1955-1956 (Fig. 10) is similar to abrupt changes in trend that have occurred\nDec.\n0.52\n1.04\n-.52\nmore recently, for example in 1973. Thus one cannot easily decide that the\no = .20\nproxy data for 1955-1956 are wrong.\nWe are probably expecting too much to consider that these early data\n*Monthly means of the fourth column entries of Table 4 normalized by adding 0.42 ppm to\nmight tell us something about a change in the long-term trend. These data are\neach value.\nBased on comparison with the linear trend for the west coast of the United States, ex-\nbetter regarded as the kind of \"baseline\" data which Revelle had in mind to\npressed by equation (3).\nobtain during the IGY. At least they add evidence that no very unusual\n320\ncircumstances influenced the atmospheric CO2 record immediately before sys-\ntematic data collecting began during the IGY.\nFigure 7. Long-term trend in atmos-\npheric CO2 for the west coast of the\nUnited States, as in Fig. 6 except that\nthe seasonal variation was determined\n315\nas an average of the monthly average\nconcentrations after subtracting a\nlinear estimate of the trend. Data\nidentified in Fig. 6 as possibly\ncontaminated are not shown.\n310\n1955\n1956\n1957\n1958\n1959\n1960\n1961\n1962\n48","Difference\nWest Coast*\nMauna Loa\nMonth\n-1.17\n-0.15\nJan.\n1.02\nTable 6. Seasonal variation in atmospheric carbon dioxide - compari-\n-0.27\nFeb.\n0.80\n0.53\nson of west coast United States, 1955-1962, with MLO, 1958-1975\n-0.42\n1.21\nMar.\n1.63\n0.51\n2.27\nApr.\n1.76\n0.29\n2.74\nMay\n2.45\n0.79\nJune\n1.45\n2.24\n2.99\n0.87\nJuly\n-2.12\n4.09\nAug.\n-5.12\n-1.12\n2.11\nSept.\n-4.79\n-2.68\n-2.99\n-0.46\nOct.\n-2.53\n-1.82\n-1.96\nNov.\n-0.14\n-1.59\n-0.97\nDec.\n0.62\n*Third column of Table 5 reduced by 0.42 ppm.\n6\nFigure 8. Atmospheric CO2 as a\nfunction of the month of the year\n0\ndetermined as a departure of the\nmonthly mean concentration from the\nlong-term trend for Mauna Loa. Data\nare shown for MLO by dots, and for\nthe west coast of the United States by\ncrosses. Months 1 to 6 (January\nthrough June) are plotted twice to\nreveal the seasonal patterns more\nfully.\n+\n-6\n6\n6\n12\nMonth\n49","Figure 9. Trend in atmospheric CO2\n335\nconcentrations at MLO. The dots\nindicate the monthly average concen-\ntration. Data in 1955, 1956, and 1957\n330\nare proxy data based on observations\nfor the west coast of the United\nStates. The oscillatory curve is a\n325\nspline fit of the sum of the long-term\ntrend and the average seasonal varia-\ntion determined as in Fig. 7.\n320\n315\nTable 7. Monthly average concentration of atmospheric carbon dioxide\n310\n(ppm) at MLO expressed according to the 1974 manometric mole\n1954\n1956\n1958\n1960\n1962\n1964\n1966\n1968\n1970\n1972\n1974\n1976\n1978\nfraction scale\nYear\nJan.\nFeb.\nMar.\nApr.\nMay\nJune\nJuly\nAug.\nSept.\nOct.\nNov.\nDec.\n1955\n315.84*\n313.77*\n1956\n313.79*\n315.93*\n316.42*\n317.71*\n316.87*\n1957\n315.04*\n316.36*\n316.23*\n316.30*\n312.37*\n311.57*\n1958\n316.33\n317.59\n317.93\n317.71\n315.92\n315.15\n314.02\n312.83\n313.64\n314.71\n1959\n315.62\n316.59\n316,94\n317.77\n318.29\n318.24\n316.67\n314.96\n314.12\n313.58\n315.14\n315.77\n1960\n316.62\n317.16\n317.90\n319.21\n320.02\n319.74\n318.15\n316.00\n314.23\n314.07\n315.04\n316.19\n1961\n316.97\n317.74\n318.63\n319.43\n320.47\n319.71\n318.78\n316.84\n315.16\n315.56\n316.14\n317.13\n1962\n318.06\n318.59\n319.74\n320.63\n321.21\n320.83\n319.55\n317.75\n316.27\n315.62\n316.84\n317.70\n1963\n318.80\n319.08\n320.15\n321.49\n322.25\n321.50\n319.67\n317.61\n316.25\n316.17\n317.01\n318.36\n1964\n319.37\n322.19\n320.49\n318.48\n317.13\n317.02\n317.84\n318.78\n1965\n319.55\n320.65\n321.15\n322.31\n322.35\n322.19\n321.53\n319.13\n317.99\n317.70\n319.15\n319.27\n1966\n320.22\n321.23\n322.13\n323.30\n323.57\n323.29\n322.36\n319.71\n317.89\n317.54\n319.36\n320.51\n1967\n321.60\n322.03\n322.50\n324.00\n324.46\n323.46\n322.19\n320.57\n318.91\n318.81\n320.24\n321.59\n1968\n322.15\n322.73\n323.50\n324.52\n325.11\n325.06\n323.62\n321.55\n319.89\n319.80\n320.73\n322.25\n1969\n323.73\n324.53\n325.62\n326.58\n327.24\n326.53\n325.63\n323.28\n322.21\n321.67\n322.61\n324.07\n1970\n324.91\n325.81\n326.85\n328.07\n327.97\n327.77\n326.44\n324.92\n323.49\n323.50\n324.34\n325.39\n1971\n326.46\n326.93\n327.56\n328.23\n329.51\n329.04\n327.87\n326.00\n324.06\n324.20\n325.48\n326.62\n1972\n327.30\n328.20\n328.50\n330.22\n330.58\n329.48\n328.56\n326.77\n325.39\n325.72\n326.97\n328.09\n1973\n329.16\n330.02\n330.95\n331.95\n332.85\n332.58\n331.30\n329,64\n328.12\n327.67\n328.69\n329.05\n1974\n329.84\n331.13\n331.93\n333.16\n333.53\n332.73\n331.77\n329.63\n327.87\n327.84\n328.77\n330.12\n1975\n330.64\n331.20\n331.89\n333.14\n333.78\n333.75\n332.06\n330.25\n328.85\n328.58\n329.61\n330.82\n1976\n331.75\n332.81\n333.55\n334.62\n335.01\n334.58\n333.22\n331.24\n329.48\n329.19\n330.35\n331.72\nProxy data.\n50","335\nFigure 10. Long-term trend in atmos-\npheric CO2 concentration at MLO.\n330\nThe plot is the same as Fig. 9 except\nthat the seasonal variation has been\nsubtracted out.\n325\n320\n315\n310\nTable 8. Seasonally adjusted concentration of atmospheric carbon\n1970\n1972\n1974\n1976\n197\n1954\n1956\n1958\n1960\n1962\n1964\n1966\n1968\ndioxide (ppm) at MLO for the 15th of each month expressed according\nYear\nto the 1974 manometric mole fraction scale*\nAug.\nSept.\nOct.\nNov.\nDec.\nMay\nJune\nJuly\nMar.\nApr.\nJan.\nFeb.\n314.10\n314.16\n314.21\n313.81\n313.88\n313.96\n314.03\n1955\n314.13\n314.31\n314.28\n314.24\n314.21\n314.17\n314.33\n314.34\n314.33\n1956\n314.25\n314.28\n314.31\n314.50\n314.60\n314.71\n314.83\n314.22\n314.30\n314.39\n314.12\n314.16\n1957\n314.11\n314.09\n314.09\n315.77\n315.79\n315.46\n315.55\n315.63\n315.69\n315.74\n315.15\n315.26\n315.36\n1958\n314.94\n315.05\n316.26\n316.37\n316.48\n316.58\n316.67\n315.99\n316.07\n316.16\n315.88\n315.93\n1959\n315.82\n315.85\n317.08\n317.11\n317.13\n317.15\n317.18\n317.22\n317.01\n317.05\n316.83\n316.90\n316.96\n1960\n316.76\n318.18\n317.75\n317.84\n317.94\n318.02\n318.10\n317.33\n317.40\n317.48\n317.56\n317.66\n1961\n317.27\n318.73\n318.76\n318.80\n318.84\n318.57\n318.63\n318.68\n318.45\n318.51\n1962\n318.25\n318.32\n318.39\n319.08\n319.12\n319.16\n319.21\n319.27\n319.03\n319.05\n1963\n318.87\n318.91\n318.94\n318.98\n319.01\n319.74\n319.78\n319.82\n319.86\n319.61\n319.66\n319.70\n319.44\n319.50\n319.56\n1964\n319.33\n319.39\n320.36\n320.43\n320.50\n320.57\n320.63\n320.20\n320.28\n1965\n319.90\n319.95\n320.00\n320.06\n320.12\n321.03\n321.08\n321.13\n321.19\n321.25\n320.90\n320.94\n320.98\n320.80\n320.85\n1966\n320.69\n320.75\n321.89\n321.97\n322.06\n321.65\n321.73\n321.80\n321.54\n321.59\n1967\n321.32\n321.38\n321.43\n321.48\n322.93\n323.08\n323.24\n323.42\n322.68\n322.80\n322.22\n322.30\n322.39\n322.47\n322.57\n1968\n322.14\n324.62\n324.73\n324.83\n324.94\n325.05\n324.38\n324.50\n324.10\n324.24\n1969\n323.60\n323.78\n323.94\n326.30\n325.93\n326.03\n326.13\n326.22\n325.71\n325.82\n1970\n325.16\n325.28\n325.38\n325.49\n325.60\n327.04\n327.13\n327.22\n327.30\n326.77\n326.86\n326.95\n326.60\n326.68\n1971\n326.38\n326.45\n326.52\n328.20\n328.39\n328.59\n328.81\n329.02\n327.89\n328.03\n1972\n327.39\n327.47\n327.56\n327.65\n327.76\n330.37\n330.45\n330.50\n330.54\n330.57\n329.99\n330.14\n330.27\n329.63\n329.82\n1973\n329.23\n329.45\n330.75\n330.78\n330.81\n330.70\n330.72\n330.73\n330.68\n330.69\n330.62\n330.64\n330.66\n1974\n330.59\n331.50\n331.61\n331.72\n331.82\n331.29\n331.40\n330.90\n330.96\n331.03\n331.11\n331.20\n1975\n330.85\n332.40\n332.47\n332.37\n332.44\n332.27\n332.31\n332.34\n332.17\n332.23\n1976\n331.93\n332.02\n332.10\nEntries before March 1958 are based on proxy data.\n51","EPILOGUE\nSince these proxy data for Mauna Loa were originally obtained from\npresented at the International Union of Geodesy and Geophysics meeting in\nsampling sites presumed to be disturbed locally, it seems paradoxical that truly\nHelsinki in 1960. But reform was on the way. Walter Bischof in 1959 had\nreliable data were not obtained by investigators who deliberately sought\nassumed responsibility for Swedish measurements. He soon became suspicious\nundisturbed locations to obtain baseline CO2 data. As Bray (1959) noted,\nof their variability on the basis of discrepancies between ground-level and\nseveral nineteenth-century investigators, who claimed analytical analyses\naircraft sampling (Bischof, 1960). Also, he had begun to use an infrared gas\naccurate to 1.0 ppm, made serious attempts to obtain data representative of\nanalyzer. With this abandonment of the traditional chemical method of\nlocally undisturbed air. I conclude that these scientists, perhaps from an inade-\nanalysis, the Swedish CO2 data ceased to include unreasonably low CO2\nquate knowledge of meteorology and atmospheric motion, underestimated the\nvalues. Then in 1960 Bischof turned to investigating suspiciously high values\ndifficulty in finding truly uncontaminated sites. When their analytical and\nusing aircraft to verify ground-level data. Probably within a year or two,\nsampling methods failed to give them the high reproducibility that they\nconsiderably more accurate systematic data would have begun to appear from\nthought they had attained, they ascribed the scatter to the atmosphere itself\nthe Scandinavian program.\nand not to weaknesses in their methods.\nBut it is far from certain that a Scandinavian site as reliable as MLO would\nIn the first half of this century declining interest in atmospheric CO2 was\nhave soon been established. The Scandinavian investigators lacked the funds\nkept alive by only a few investigators. The most notable was Kurt Buch of\nto embark on an ambitious continuous sampling program at a remote station.\nFinland, who concluded after many years of study that the CO2 concentration\nMany years might have passed before data of the quality of the Mauna Loa\nvaried systematically with air mass. His claims (Keeling and Bacastow, 1977)\nrecord would have been forthcoming. Indeed, high costs almost caused MLO to\nthat high arctic air had concentrations in the range of 150 to 230 ppm, north\nclose down in 1964 in spite of its obvious value as a CO2 sampling site.\nand middle Atlantic air, 310 to 345 ppm, and tropical air, 320 to 370 ppm,\nDisruptions under that threat of closure account for a serious gap in the CO2\nstrongly influenced preparations for the IGY CO2 program, especially the Scan-\nrecord during the early part of 1964. Problems of cost also contributed to the\ndinavian program, which he initially supervised. When from inadequate\ndecision to shut down the South Pole continuous-analyzer program at the end\nchemical and sampling techniques the Scandinavian pre-IGY program produced\nof 1963. If these two remarkable sites had not already been established and\nCO2 concentrations in the same range as previous data, these new data were\nyielded high-quality data before 1964, it is likely that the stimulus to start\nreadily justified as resulting from different properties of the air masses passing\nwork at such remote sites would not have occurred for at least several more\nover the sampling sites (Fonselius et al., 1956).\nyears because of financial impediments. Thus it was a fortunate circumstance\nHow long would the findings of the Scandinavian CO2 network have been\nthat Wexler and Revelle in 1956 saw the value of using the IGY organization to\naccepted if new manometric and infrared studies had not been begun? The\ncheck out the possibility of near constancy in atmospheric CO2 by inaugurat-\nScandinavian data continued to appear in the back pages of Tellus until after\ning a precise sampling program. We all recognize now that such a program is\nthe infrared analyzer results for Mauna Loa and other locations had been\nessential if we are to document adequately the rise in atmospheric CO2.\n52","REFERENCES\nACKNOWLEDGMENTS\nMany people who could not be included in the historical discussion con-\nArrhenius, S. A., 1903: Lehrbuch der\nKosmischen Physik, V. 2, Hirzel, Leipzig,\ntributed to the planning of atmospheric carbon dioxide measurements at\n477-481\nMauna Loa. I am particularly indebted to Oliver Wulf, U.S. Weather Bureau\nBacastow, R. B., 1976: Modulation of atmos-\nscientist stationed at the California Institute of Technology in 1955 and 1956.\npheric carbon dioxide by the southern oscil-\nWulf first brought my manometric work to the attention of Dr. Wexler. I am\nlation. Nature, 261:116-118.\nalso indebted to Paul Humphrey, Dr. Wexler's assistant, who coached me on\nBacastow, R. B., 1977: Influence of the\nsouthern oscillation on atmospheric carbon\nmaking the best use of the short time I would have to talk with Wexler.\ndioxide. In Fate of fossil fuel CO2 in the\nHumphrey later devoted many hours to coordinating funding and logistics\noceans. Edited by N. R. Andersen and A.\ninvolved in setting up CO2 research at Mauna Loa.\nMalahoff, Plenum, N.Y., 33-43.\nIn addition, I am indebted to Kenyon George, engineer of the Applied\nBischof, W., 1960: Periodical variations of the\nPhysics Corporation, Pasadena, California. George patiently replied to my\natmospheric CO2 content in Scandinavia.\nTellus, 12:216-226.\ndetailed questions during 1956 about the performance characteristics of his\nBolin, B., and C. D. Keeling, 1963: Large-scale\nfirm's nondispersive infrared gas analyzer. He was not himself convinced that\natmospheric mixing as deduced from the\natmospheric CO2 could be determined by infrared analysis to the accuracy I\nseasonal and meridional variations of carbon\nsought, but his frank answers and total lack of bias provided sound arguments\ndioxide. J. Geophys. Res., 68:3899-3920.\nin favor of trying out the infrared method during the IGY.\nBray, J. R., 1959: An analysis of the possible\nrecent change in atmospheric carbon dioxide\nI also owe thanks to John Miller, present director of MLO, who suggested\nconcentration. Tellus, 11:220-230.\nthis article and allowed me time to complete it, and to Robert Bacastow, who\nBuch, K., 1948: Der Kohlendioxydgehalt der\ndevised the computer programs that executed many of the computations of\nLuft als Indikator der meteorologischen Luft-\nthis paper and who offered valuable criticisms. Financial support for the work\nqualitat. Geophysica (Helsinki), 3:63-79.\ndescribed was by the Climate Dynamics Program of the U.S. National Science\nCallendar, G. S., 1938: The artificial\nproduction of carbon dioxide and its\nFoundation under grants ATM76-23053 and ATM77-25141.\ninfluence on temperature. Q. J. R. Meteorol.\nSoc. (London), 64:223-240.\nCallendar, G. S., 1940: Variations of the\namount of carbon dioxide in different air\ncurrents. Q. J. R. Meteorol. Soc. (London),\n66:395-400.\n53","Callendar, G. S., 1958: On the amount of\nKeeling, C. D., R. B. Bacastow, A. E. Bain-\ncarbon dioxide in the atmosphere. Tellus,\nbridge, C. A. Ekdahl, P. R. Guenther, L. S.\n10:243-248.\nWaterman, and J. F. S. Chin, 1976a: Atmos-\nEriksson, E., 1954: Report on an informal\npheric carbon dioxide variations at Mauna\nconference in atmospheric chemistry held at\nLoa Observatory, Hawaii. Tellus,\nthe Meteorological Institute, University of\n28:538-551.\nStockholm, May 24-26, 1954. Tellus,\nKeeling, C. D., J. A. Adams, C. A. Ekdahl,\n6:302-307.\nand P. R. Guenther, 1976b: Atmospheric\nFonselius, S., F. Koroleff, and K. Buch, 1955:\ncarbon dioxide variations at the South Pole.\nMicrodetermination of CO2 in the air, with\nTellus, 28:552-564.\ncurrent data for Scandinavia. Tellus,\nKeeling, C. D., and R. B. Bacastow, 1977:\n7:258-265.\nImpact of industrial gases on climate. In\nFonselius, S., F. Koroleff, and K. E. Wärme,\nEnergy and climate, National Academy of\n1956: Carbon dioxide variations in the at-\nSciences, Washington, D.C., 72-95.\nmosphere. Tellus, 8:176-183.\nMachta, L., K. Hanson, and C. D. Keeling,\nGeophysics Study Committee, 1977: Overview\n1976: Atmospheric carbon dioxide and some\nand recommendations. In Energy and\ninterpretations. In Fate of fossil fuel CO2 in\nclimate, National Academy of Sciences,\nthe oceans, Edited by N. R. Andersen and A.\nWashington, D.C., 1-31.\nMalahoff, Plenum, N.Y., 131-144.\nHall, C. A. S., C. A. Ekdahl, and D.E.\nNewell, R. E., and B. C. Weare, 1977: A rela-\nWartenberg, 1975: A fifteen-year record of\ntionship between atmospheric carbon dioxide\nbiotic metabolism in the Northern\nand Pacific sea surface temperature. Geo-\nHemisphere. Nature, 255:136-138.\nphys. Res. Lett., 4:1-2.\nKeeling, C. D., 1958: The concentration and\nReinsch, C. H., 1967: Smoothing by spline\nisotopic abundances of atmospheric carbon\nfunctions. Numerische Mathematik,\ndioxide in rural areas. Geochim. Cos-\n10:177-183.\nmochim. Acta, 13:322-334.\nRevelle, R., and H. E. Suess, 1957: Carbon\nKeeling, C. D., 1960: The concentration and\ndioxide exchange between atmosphere and\nisotopic abundances of carbon dioxide in the\nocean, and the question of an increase of at-\natmosphere. Tellus, 12:200-203.\nmospheric CO2 during the past decades.\nKeeling, C. D., 1961: The concentration and\nTellus, 9:18-27.\nisotopic abundances of carbon dioxide in\nrural and marine air. Geochim. Cosmochim.\nActa, 24:277-298.\n54","ATMOSPHERIC TRITIUM SAMPLING\nON MAUNA LOA\nH. Gote Ostlund and Allen S. Mason\nUniversity of Miami\nMiami, Florida\nOUTLINE OF SCIENTIFIC EFFORT\nThe data obtained from Mauna Loa have confirmed the value of remote\nThe Tritium Laboratory of the University of Miami developed a technique\nsampling stations. Since 1971, only one incident of apparently local contami-\nfor simultaneous sampling of tritiated water vapor (HTO) and tritium gas\nnation has occurred. This event, between February and March 1972, was an\n(HT) in 1968. The technique has been described fully elsewhere (Ostlund and\nincrease in the local HT mixing ratio by a factor approaching 2. It has never\nMason, 1974) and will only be summarized here. A 1- to 2-liter/min air flow\nbeen explained. In late 1973 Mauna Loa data permitted the estimation of the\nfrom a diaphragm pump is mixed with 1% of tritium-free hydrogen. The\nHT seepage to the atmosphere from a series of three large underground ther-\nmixture is passed through a trap of 300 g of molecular sieve which absorbs all\nmonuclear explosions in the Arctic. A 15% HT mixing ratio increase appeared\nwater species (HO, HTO, etc.) and then into a trap of 100 g of palladium-\nand persisted for several months following the detonations (Mason and\ncoated molecular sieve. This trap oxidizes the carrier hydrogen plus the am-\nOstlund, 1974). The long time series of data have enabled us to detect trends\nbient species (H, HT, etc.) and adsorbs the resulting water in situ. The traps\nin the global HT budget (Ostlund and Mason, 1974).\nare sent to Miami, where the samples are recovered by baking the traps under\nA new sampler has been installed with the capability of separating\nvacuum. The water samples are reduced to hydrogen gas and analyzed for\nsamples from the characteristic upslope and downslope wind regimes. It is ac-\ntritium by low-level proportional gas counting. The mixing ratios of HT and\ntuated by a controller provided by the Health and Safety Laboratory of the\nHTO are calculated from the volume of air sampled, the amounts of water\nU.S. Energy Research and Development Agency (ERDA).\nsamples recovered, and the specific activity of the samples. The specific\nactivity of the water vapor is also calculated, but that of the atmospheric hy-\ndrogen is not obtainable because of the dilution by carrier gas.\nAt the inception of the program it was recognized that data from remote\nsites would be helpful in establishing the global backgrounds of HT and HTO.\nFairbanks, Alaska, was chosen as the first such site in 1970, in 1971, Mauna\nLoa, Hawaii, was added, and in 1975, Baring Head, New Zealand.\n55","FIELD OPERATIONS\nThe first tritium sampler at Mauna Loa Observatory (MLO) was a\nprototype, assembled on site by the first author with the capable assistance of\nHoward Ellis of MLO. It was located in a trailer adjacent to the main building.\nThe visit was memorable to Östlund both for the experience of visiting Mauna\nLoa for the first time and also for the experience of seeing the areas around\nHilo and finding the unique white wild strawberries that grow in the national\npark near Kilauea, just as tasty as the red ones in the author's old homeland,\nsubarctic Sweden.\nThe sampler was relocated in the main building in 1972 and inspected\nthere by the second author in the fall of that year. This was also a first visit\nand was memorable to Mason for the observing of lidar operations at night.\nThe drive up at night, in the company of Ron Fegley and Howard Ellis, was\nmost impressive.\nThe new sampler was sent to MLO in late 1975 and placed into manual\nmode operation by Howard Ellis. In April 1976 the second author met Herbert\nVolchok and Fred Wilson, both of the Health and Safety Laboratory, ERDA,\nfor the installation of the upslope-downslope controller. Working in the wind\nand rain and helping to push the rented station wagon up the last slope to the\nREFERENCES\nobservatory form Mason's chief memories of that visit.\nÖstlund, H. G., and A. S. Mason, 1974: At-\nOur association with MLO has been both fruitful and enjoyable. The\nmospheric HT and HTO I. Experimental\nunfailing efforts of the directors and staff, with particular recognition to\nprocedures and tropospheric data 1968-72,\nHoward Ellis, are the mainstay of our operations at MLO.\nTellus, 26(1-2), 91-102.\nMason, A. S., and H. G. Östlund, 1974: At-\nmospheric HT and HTO: Major HT injec-\ntions into the atmosphere 1973, Geophys.\nRes. Lett., 1(6), 247-248.\n56","Barry Bodhaine with flame photo-\nmeter measuring concentration of\naerosol particles at MLO, March\n1972. An NRC Resident Research As-\nsociate at the time, Barry later\nbecame a staff member.\nAerosol measurements\nMeasuring condensation nuclei with\nGardner counters at Kilauea volcano\nduring an eruption in 1969 (Bernard\nMendonca and Janice Sias).\nThe \"NCAR\" acoustic ice nucleus\ncounters.\nRecording equipment for the ice nu-\ncleus counters, and the G.E. counter.\nThe buildings and trailers housing ice\nnucleus experiments in 1967. The\ntrailers have since been removed. The\nbuilding now contains a kitchen and\nliving quarters for staff and visitors.\n57","Mauna Loa Observatory in March\n1978. Structures, left to right: Camera\nbuilding, CO2 sampling intake on 80-\nft tower, instrument shelter, \"clean\nrest rooms\", precipitation sampler,\nAEC sampling platform, \"Butler\nbuilding\" for storage, Dobson dome,\nmain building, new solar radiation\nplatform, lidar dome, \"AEC build-\ning\".\n58","LIDAR MEASUREMENTS AT MAUNA LOA\nOBSERVATORY\nUnpacking the lidar at MLO in 1971.\nRonald W. Fegley\nGeophysical Monitoring for Climatic Change, NOAA,\nBoulder, Colorado\nEarl W. Barrett\nAtmospheric Physics and Chemistry Laboratory, NOAA,\nBoulder, Colorado\nHoward T. Ellis\nMauna Loa Observatory, NOAA, Hilo, Hawaii\nThe Mauna Loa lidar was the first in the world to be used in the field for\nroutine continuous observations. Its main purpose is to provide detailed infor-\nmation on stratospheric events in the northern tropics (Fegley and Ellis, 1975\na,b). An accurate time-history of optical backscatter in the lower stratosphere\nabove Hawaii now exists for 1973-1978. From this parameter, simple optical\nmodels can be used to calculate other important parameters such as aerosol\nmass loading, number densities, and stratospheric extinction coefficient. Even-\ntually, stratospheric aerosol size distribution information may also be obtain-\nable using two-color lidar. These data from Mauna Loa Observatory (MLO), a\nprototype station, will eventually be supplemented with those from other\nstations in the GMCC network to provide valuable information for climate\nstudies (Oliver, 1976). A secondary use of the lidar is to provide profiles of\nbackscatter in the lower atmosphere to improve our understanding of aerosol\nand cloud dynamics in the tropical atmosphere.\nRichard Proulx beside the target used\nfor calibration and alignment of the\nlidar.\n59","ference filter. The detector is a 10 stage PM with S-20 response. No blanking\nDESCRIPTION OF THE MLO LIDAR SYSTEM\npulse is applied to this tube; some overload does occur when full laser power\nThe transmitter for the system is a ruby laser with nominal maximum\nis used. The output of this PM is fed to the digitizer through a low-pass filter\npulse energy of 3 joules at 694.3-nm wavelength. Pulses are formed by\nof the same type as in the long-range receiver, but has a cutoff frequency of 9\nPockels-cell Q-switching; nominal pulse length is 30 ns. Beam divergence is\nless than 7 mrad; maximum pulse repetition rate is 4 per minute. An interfer-\nMHz so that finer details of the return are preserved. Both PM housings are\nequipped with slots for neutral-density filters, so that the returns from the cali-\nence filter centered on the ruby wavelength, a converging lens, and a solid-\nbration and alignment target can be attenuated, or so the larger receiver can\nstate photodiode are located behind the rear laser reflector; the radiation\nbe used in the daytime. Control circuitry is provided to steer the syn-\nsampled by the lidar system provides a measure of relative pulse energy as\nchronizing trigger pulses from the laser system and the power-measuring diode\nwell as synchronizing pulses for the data display and processing components.\nto appropriate points on the digital voltmeter and digitizer.\nThe diode output is fed to an integrator and sample-and-hold circuit; the\nIn the interest of simplicity, and to avoid laser output power losses in\noutput, which is proportional to pulse energy, is measured by a digital volt-\nprisms, no beam-steering mechanisms are used. The laser and the two\nmeter. The BCD information produced by the DVM is also picked up by the\nreceiving systems are attached rigidly to a heavy I-beam which is moved in\nautomatic data processing (ADP) system at read time and used in evaluation\nazimuth and elevation by gearing. The entire steerable system rests on a heavy\nof the data.\nsteel pedestal, which is bolted to the concrete floor of the observatory-type\nBecause of the tremendous dynamic range of lidar signals (100 dB or\ndome.\nmore), the system is equipped with two receivers. The long-range collector is a\nThe transient recorder, or digitizer, stores 2024 digitized sequential\n40.6-cm diameter (16-in) catadioptric Cassegrain instrument. The collected light\nsamples of the output of either receiver. Maximum sampling rate is 108\nis collimated by an achromat lens of 50-mm diameter and 160-mm focal\nsamples per second (one per 10 ns, or one per 150 cm of range), and the reso-\nlength. The collimated beam passes through an interference filter with 1-nm\nlution is 8 bits (one part in 256). Sampling is initiated by the sync pulse from\nbandwidth before impinging on the cathode of a 14-stage photomultiplier tube\nthe monitor photodiode at lasing time. After sampling, the digitized data are\nwith S-20 response. Output of the photodetector is coupled by 50-ohm coaxial\ndisplayed repetitively at low rate on any convenient oscilloscope. The operator\ncable to a low-pass filter with a 4-MHz cutoff frequency. The output from the\ncan thereby assure himself that the data from a particular shot are of\nfilter is fed through a very short length of cable to the input port of the\nacceptable quality and that the laser has not malfunctioned. If all is well, the\ntransient recorder. (This system was specially designed by one of the authors,\noperator switches to the dump, or data transfer, mode. Contents of memory\nEarl W. Barrett, hereafter referred to as E.W.B.)\nare transmitted sequentially (8 bits in parallel) over a 77-m cable to the mini-\nTo prevent overload of the PM tube by the near-field return, alternate\ncomputer. The two-way multiplexed data-transmission system between the\ndynodes of the tube were supplied with a blanking pulse which began at laser\ndigitizer and the computer was designed by Dennis Wellman of APCL. In order\nflash lamp firing time and ended at a time that could be preset by the\nthat the number of digitized shots sent to the computer for averaging (to\noperator. A time corresponding to a range of 7.5 km was approximately\nimprove signal-to-noise ratio) shall agree with the preset value supplied by the\ncorrect to avoid PM tube overload and also excessive signal clipping at the\noperator to the computer, a digital LED display at the operator's position in\ninput of the digitizer.\nthe dome is updated each time a shot is received by the computer. The relative\nThe second receiver, intended for short-range and daytime use out to 7.5\nlaser energy as displayed on the digital voltmeter is also transmitted to the\nkm, uses a 10.2-cm-diameter (4-in) telescope of the same type, a 25-mm-\ncomputer via this data exchange link.\ndiameter, 60-mm focal length collimating lens, and a 1-nm bandwidth inter-\n60","The data processing is controlled by a BASIC main program, with calls to\nassembly language subroutines that control data and flag transfers between the\nconsole and the computer. The main program can be altered to suit different\ncircumstances. In the original setup the operator supplied the computer with\nthe sampling rate (i.e., maximum range) and the number of shots to be\naveraged. After receipt of the data from the last shot, the computer printed\nout on the teletype the time mean relative backscatter (average of all shots)\ncorrected for range and Rayleigh scattering extinction and also averaged in\nrange over 10 sample intervals. The theoretical Rayleigh backscatter coefficient\nfor each of these 100 significant levels in a tropical standard atmosphere was\nalso printed out for use in interpreting the results. Program changes have been\nmade in the 3 years since the system was placed in operation; some of these\nchanges will be discussed in a following section.\nTo provide a target with stable scattering properties for absolute calibra-\ntion of the system, and to aid in boresighting the system (making the axes of\nthe laser beam and the receiver telescopes parallel), a billboard 3 m (10 ft) on\na side was erected at a point on the mountain some 300 m (1000 ft) south of\nthe dome. The billboard can be lowered when not in use. The surface is fiber\nglass sheeting painted with flat white paint (85% diffuse reflectance). Three\nblack spots, spaced the same distances apart as the axes of the laser and tele-\nscopes, are provided for assistance in the boresighting operation. Unfortu-\nnately, the board was damaged by wind and repaired with glue and therefore\nno longer presents a homogeneous surface. It therefore cannot be used for\nabsolute calibrations, but is still useful for alignment.\nThe physical arrangement of the system in the dome as of June 3, 1974, is\nshown in the figures. Fig. 1 shows the transmitting and receiving optical hard-\nware in position for making a vertical sounding. The laser with its plastic\nFigure 1. Laser transmitter and\nreceiving optics mounted on steerable\nI-beam support in observatory dome.\n61","stray-light baffle is at the left, bolted to one flat face of the I-beam. The short-\nrange receivers, scope, and PM tube housing are in the center, attached to the\ninner surface of the beam. The long-range receiver is at the right, attached to\nthe other flat face of the beam. The PM tube housing is concealed by the\nheavy steel trunnion; it is located at the lower right. The handwheel at the far\nright is the elevation control.\nFig. 2 shows the operating position as seen from above. From left to right,\nthe various components are laser power supply and control circuits; the\nsystem master control box and sync pulse distributor; the digital voltmeter for\nthe power monitor; and the rack containing the PM power supplies and\nblanking-pulse generator, the digitizer, the data transmit-receive link to the\ncomputer, and the monitor oscilloscope. Cabling to the PM tubes passes\nthrough the rigid conduit at the right.\nFigs. 3 and 4 show the operating position from eye level. The laser power\nsupply, master control box, and digital voltmeter appear in Fig. 3; the remain-\ning hardware is displayed in Fig. 4. The miniboxes connected to the digitizer\ncontain the special low-pass filters. The two PM power supplies are at the\nbottom of the rack with the digitizer just above them. The blanking-pulse\ngenerator occupies the very narrow panel above the digitizer; the next panel\nabove is the data transmitter-receiver with LED displays of bit status and shot\ncount. At the top is the monitor oscilloscope (a laboratory scope served as\nFigure 2. Lidar operating position\ntemporary substitute for the regular unit which was being repaired).\nviewed from lidar pedestal.\nDESIGN AND INSTALLATION OF THE MLO LIDAR\nSYSTEM\nThe desirability of having a high-powered lidar system at MLO was the\nsubject of several discussions between E.W.B. and Helmur Weickmann,\nDirector of the Atmospheric Physics and Chemistry Laboratory (APCL), as\nearly as 1968. The influence of the dust from the eruptions of Agung volcano\nin 1963 on the solar radiation received at Mauna Loa (Ellis and Pueschel,\n1971) and the concern over pollutants from aircraft operating in the strato-\nsphere were indications of the value of long-term monitoring of stratospheric\naerosol levels. While solar radiation measurements can be used to estimate\nFigure 3. Center of operating position\nviewed from eye level.\nFigure 4. Right side of operating\nposition viewed from eye level.\n62","total particulate loadings, lidar has the additional capability of ranging to\ndesigned, installed, and debugged. It took four years, from the preliminary\npermit the study of the vertical distribution of the aerosols.\ndesign work of summer 1970 to the first computerized soundings taken on\nFunding for a lidar system was included in the APCL budget for fiscal\nMay 30, 1974. The entire system was turned over to Ronald Fegley on June 3,\n1971. At that time the observatory was a part of APCL under the direction of\n1974.\nRudolf Pueschel.\nOPERATIONAL EXPERIENCE\nThe lidar system was designed to be as versatile as possible, so that it\ncould be used not only for stratospheric aerosol monitoring but also for low-\nIn August 1972, Ronald Fegley became Director of MLO. Because of his\nlevel aerosol and cirrus cloud studies. Preliminary design work took place in\nprior experience with atmospheric aerosols and lidar systems, he was\nthe summer and fall of 1970, and component procurement began in December\ninterested in starting a long-term atmospheric monitoring program to detect\nof that year. By the early summer of 1971 the major elements of the system\natmospheric phenomena that would escape short-term observational programs.\nhad been obtained, and the mechanical design work (by David Eyre, APCL\nStratospheric soundings began during the fall of 1972 and were recorded\nmechanical engineer) was well advanced.\nphotographically. These photos were laboriously hand-reduced and a\nAt about that time a major administrative reorganization took place, in\nprocessing technique was developed by Howard Ellis. The manual technique\nwhich MLO was transferred from the jurisdiction of APCL to that of the Air\nwas designed to allow a \"clean air\" calibration to be made at some upper\nResources Laboratory (ARL).\ntropospheric level.\nThe lidar system began to take shape on the mountain in October 1971,\nIn March 1973 problems had been resolved to the point that the authors\nwhen David Eyre, with an engineer from a private contractor and a crew of\nwere confident of their results. Therefore that date is given as the beginning of\nNOAA workmen, erected the observatory-type dome near the east end of the\nthe Mauna Loa stratospheric data record.\nparking area and set up the heavy pedestal that supports the laser and\nAfter the first computer was installed, refinement of the analysis program\nreceiving optics. At the same time E.W.B., assisted by Bruce Uhlenhopp,\ncontinued over the years. Routines were developed to allow real-time\nelectronic engineer, and Charles Johnson, electronic technician, of APCL,\ngraphical display of the scattering ratio profile on a cathode-ray tube display.\ncompleted the design and construction of the signal-handling and control\nAutomatic calibration of the lidar returns at the clean air level and at\ncircuitry. In February 1972 E.W.B. and Richard Proulx, meteorological\nmaximum range was implemented. Data were averaged and put onto paper\ntechnician of APCL, proceeded to MLO to put the system together and check it\ntape for semi-permanent storage. Corrections were made to the program to\nallow analysis at various zenith angles.\nout.\nThe installation and initial testing of the lidar system was characterized by\nProcedures were finally well developed when the intense dust cloud from\nproblems like those that accompany the development of any complex scientific\nFuego volcano in Guatemala appeared in late 1974. This event confirmed\nsystem: failure in the laser power supply, overloading on the receiver-photo-\ninterest in stratospheric monitoring and led to discussions concerning\nmultiplier tube, radio frequency interference problems, and a high receiver\nexpansion of the lidar network to other GMCC observatories.\nnoise level due to fluorescent afterglow from the laser being scattered off the\nIn fall 1976 a second computer dedicated to lidar operation was installed\ndome into the receiving optics. The automatic data-processing system also\nin the dome and SO total operation was possible at one site. The lidar has\noffered difficulties which meant a year's delay before it became operational.\noperated reliably, considering the complexity of the electro-optical system. The\nHardware and software problems arose with the minicomputer. Yet despite\npersistent problems have been due to the poor room-temperature regulation in\ndelays and frustrations all the necessary components of the system were\nthe dome. Future GMCC lidars will be housed in a temperature-regulated room\n63","and will fire through a glass skylight. Future development will also emphasize\nfurther automation to decrease the difficulties of operation in the cold at high\n25\nelevations.\nSUMMARY OF RESULTS\n20\nThe first regular lidar data were taken in March 1973. The returns were\nrecorded on photographic film, and reduced manually. Data were taken ap-\nTropopause\nproximately once every 2 weeks. Profiles were generally the average of a few\nshots.\n15\nThis technique continued until about May 1974, when an electronic\ntransient digitizer was incorporated to simplify the analysis. Data were then\ntaken approximately once every week with additional series made in the case\nof significant geophysical events such as volcanic eruptions. From August 1976\nto the present, data have been taken approximately once every 2 weeks.\nFig 5\nA stratospheric profile was included in each of the above data sets.\nProfiles of lower regions in the atmosphere were also included in many of the\n5\ndata sets. For example, Fig. 5 shows a profile taken on January 26, 1977. A\n10-10\n-8\n-6\n10-6\n10\nsmooth layer covering the entire sky was observed visually, and the lidar\nNon - Rayleigh Backscatter Coef.\nshowed it to be at 16.3 km msl. The maximum backscatter was two orders of\nmagnitude greater than that of adjacent layers. This layer reappeared\nsporadically several times, but dissipated rather rapidly compared with the\n1974 volcanic event (Fegley and Ellis, 1975a). Within error, air trajectory\nanalysis placed the volcanic source near Nyiragongo volcano in the country of\nZaire, Africa. By the conventional definition of tropopause height, the layer\nwas slightly below the tropopause, which may explain the short lifetime of this\nveil.\nFuego\nFig. 6 summarizes the Mauna Loa stratospheric data. From the short data\nvolcano\nrecord, it is hard to say whether there is a steady background level at about\nNyiragongo\n0.3 X 10-4/sr or a slowly declining value since 1973. Other observers have\n2.0\n5.0\nreported a stratospheric injection early in 1973, a month or two before the\nvolcano\nonset of our measurements (Hofmann et al., 1976).\nFigure 5. Profile of lidar backscatter\nFigure 6. Aerosol backscatter\nintegrated through 16- to 24-km layer\nfor January 26, 1977, at Mauna Loa.\nFig. 6\nabove MLO. (Data are preliminary.)\nNote intense veil at 16.3 km msl.\nWavelength is 0.6943 um. Volcanic\nassignments are tentative. Aerosol\n1973\n1975\n1977\ncolumn density is calculated using\nsimple optical model.\n64","REFERENCES\nEllis, H. T., and R. F. Pueschel, 1971: Solar\nradiation: Absence of air pollution trends at\nMauna Loa. Science, 172:845.\nFegley, R. W., and H. T. Ellis, 1975a: Lidar\nobservations of a stratospheric dust cloud\nlayer in the tropics. Geophys. Res. Lett.,\n2(4):139-141.\nFegley, R. W., and H. T. Ellis, 1975b: Optical\neffects of the 1974 stratospheric dust cloud.\nAppl. Optics, 14(8):1751-1752.\nHofmann, D. J., J. M. Rosen, J. M. Kierman,\nand J. Laby, 1976: Stratospheric aerosol\nmeasurements IV: Global time variations of\nthe aerosol burden and source considera-\ntions. J. Atmos. Sci. 3(9): 1782-1788.\nOliver, R. C., 1976: On the response of\nhemisphere mean temperature to strato-\nspheric dust; an empirical approach. J. Appl.\nMeteorol., 15(9):933.\n65","A POSSIBLE EFFECT OF LOCAL VOLCANIC\nACTIVITY AT MAUNA LOA OBSERVATORY\nB. A. Bodhaine\nGeophysical Monitoring for Climatic Change\nNOAA, Boulder, Colorado\nINTRODUCTION\nThe Geophysical Monitoring for Climatic Change (GMCC) program of the\nDuring July 1975 an experiment was performed in cooperation with Keith\nNational Oceanic and Atmospheric Administration (NOAA) is currently\nBigg in an attempt to obtain a better understanding of the character of surface\nmonitoring background atmospheric properties at four baseline sites: Barrow,\naerosols at MLO. His techniques (Bigg, 1977) are especially applicable at MLO\nAlaska; Mauna Loa, Hawaii; American Samoa; and South Pole. All stations\nbecause it is likely that Mauna Loa, owing to its high elevation, is often\nare now operational and are collecting continuous data for many gases, parti-\nexposed to a background upper tropospheric aerosol composed primarily of\nculates, and solar radiation (NOAA, 1975). Mauna Loa Observatory (MLO),\nsulfates. The chance occurrence of an eruption of Mauna Loa (5 km from the\nlocated at an altitude of 3.4 km on the island of Hawaii, has been in operation\nsite) during this experiment provided a unique opportunity for a comparison\nof the instruments measuring background, island, and locally generated\nsince 1956.\nAerosol monitoring is an important part of GMCC because of the potential\nvolcanic aerosol.\neffects of aerosols on climate. Aerosols may interact directly with solar or\nThe local meteorology of Mauna Loa has been discussed by Mendonca\ninfrared radiation and could produce a general warming or cooling depending\nand Iwaoka (1969) and Mendonca (1969) and has been summarized (NOAA,\non the physical properties of the aerosol and its location in the atmosphere.\n1974a, 1974b, 1975). The sources of aerosols on the island of Hawaii have\nFurthermore, aerosols may have an indirect effect by acting as cloud or ice\nbeen studied by Bodhaine and Pueschel (1972), Pueschel and Mendonca\nnuclei which may affect the albedo, weather, or climate of the earth. It is con-\n(1972), and Pueschel et al. (1973). Also, Bodhaine and Mendonca (1974) have\nceivable that mankind may have the capability of producing enough aerosols\ncompared normal background aerosol monitoring conditions with a brief\nto compete with those occurring naturally, although this is open to question at\nperiod of volcanic contamination at Mauna Loa.\npresent.\nCurrently, the GMCC aerosol program is monitoring Aitken nuclei and\ntotal light scattering. Activities planned include the monitoring of cloud con-\nINSTRUMENTATION\ndensation nuclei (CCN), ice nuclei (IN), and the chemical properties of\naerosols, although it is not yet clear what methods are most suitable for back-\nAerosol monitoring equipment used for this study included a Pollak CN\ncounter, a G.E. automatic CN counter, and a four-wavelength integrating\nground locations.\n66","nephelometer. The basic design and operation of a four-wavelength nephe-\nchanges in the chemistry of the aerosol may occur, which could also have im-\nlometer have been described in detail by Ahlquist and Charlson (1969) and\nportant implications for both direct and indirect effects on climate but which\nCharlson (1972), whereas the Mauna Loa instrument was described by\nwould not show up as changes in either concentration or size. For this reason\nBodhaine and Mendonca (1974).\nit is important to perform regular intensive experiments, such as this one with\nBriefly, the Pollak counter is an expansion instrument operated in the\nBigg, in conjunction with background monitoring.\noverpressure mode at a supersaturation of about 300%. It is calibrated in\nIn general, instruments that measure Aitken nuclei respond to aerosols in\nterms of the attenuation of a light beam passing through the cloud produced\nthe size range 0.001 um < r < 0.1 m, and the nephelometer responds in the\nby the expansion and is sensitive to less than 10 nuclei/cm3. The G.E. nucleus\nsize range 0.1 um < r < 1.0 um, SO that it is not unusual to find little or no\ncounter is an expansion instrument operating in the underpressure mode, also\ncorrelation between the two types of instruments if the aerosol size distribu-\nat a supersaturation of about 300%. Detection of the cloud is accomplished at\ntion is shifted towards one end of the size spectrum. However, under stable\nlow forward scattering angles. The instrument performs five expansions per\nbackground conditions the two types of instruments generally show high cor-\nsecond and is sensitive, in its present modified form, to less than 10\nrelation. With the addition of Bigg's technique it is possible to span the size\nnuclei/cm3.\ndistributions of both instruments to provide a better understanding of the size\nThe Mauna Loa nephelometer measures the total integrated light\nrange in which they overlap, in addition to obtaining information on the\nscattering due to aerosols (bsp) simultaneously at wavelengths 450, 550, 700,\nchemistry of the aerosol.\nand 850 nm and is capable of measuring bsp of the order of 10-7 m-1, which is\nFig. 1 gives a composite presentation of wind direction and speed at 3-\nhour intervals, relative humidity, surface ozone, Aitken nuclei measured by\nabout 1% of the molecular scattering of air.\nboth Pollak and G.E. counters, and 550-nm light scattering for July 1975 at\nTo gain a better understanding of the aerosol data it is necessary to assess\nother atmospheric variables such as relative humidity, wind direction, and\nMLO. While, literally, a \"mountain\" of information is contained in this illustra-\nspeed. As a further aid in determining the origin of air masses at Mauna Loa,\ntion, it is useful to point out a number of features before delving into a\nit is useful to examine the surface ozone record, produced by an instrument\ndetailed analysis of relationships among the various parameters.\nbased on the electrochemical cell method (NOAA, 1974a). In general, the\nThe volcanic eruption occurred at the summit of Mauna Loa (altitude of\nhighest ozone values are associated with air of upper tropospheric or,\n3.6 km) directly upslope (due south) of the observatory at approximately 2330\npossibly, stratospheric origin.\nhours (LST) on July 5, 1975. At about 0800 hours the next morning, the\nThe equipment supplied by Bigg consisted of an impactor and an electro-\npower lines feeding the observatory were threatened by a lava flow, and the\nstatic precipitator which captured particles directly upon electron microscope\npower was turned off. This was unfortunate, since data lost during this 24-\nscreens. These screens were then subjected to further chemical treatment and\nhour outage would have been valuable indeed. Other than this one outage, all\ninvestigated by means of electron microscopy as discussed more fully by Bigg\ninstruments performed properly throughout July except for the G.E. counter,\nwhich was inoperative during the last two weeks.\n(1977).\nThe classic diurnal upslope-downslope wind flow is apparent for much\nof July, showing a southerly component at night which brings dry upper\nDISCUSSION\ntropospheric air to the site and a northerly component during the day which\nIt is likely that global trends in the atmospheric aerosol may show up as\nbrings humid island or marine air from below the temperature inversion to the\nchanges in concentration or size distribution. However, it is possible that\nsite. This diurnal effect is especially apparent in the humidity trace but also\n67","shows up to a certain degree in the records of all quantities monitored at\nperiod, good correspondence between Aitken nuclei and light scattering is\nseen. After the eruption, high levels of Aitken nuclei were observed for about\nMauna Loa. Occasionally, this diurnal pattern is overpowered by strong\na week, after which all quantities returned to approximately pre-eruption\nsoutherly winds, usually from the southeast, and probably associated with\nlevels. For the week following the eruption, Aitken nuclei show a general\nlarge-scale high-pressure regions over the Pacific. Experience has shown that\ndecrease, and light scattering shows a general increase with no obvious rela-\nthese strong southeasterlies are usually very dry (RH < 20%) and provide the\ntionship between shorter-term variations. Apparently, the island was contami-\nbest conditions for monitoring upper tropospheric air. Examples of these\nnated with a small-sized volcanic aerosol which gradually aged to a larger-\nstrong southeasterlies are seen on July 13-15, 24-26, and 30.\nsized aerosol before being cleansed from the local atmosphere. It is interesting\nPrevious to the eruption, all quantities showed fairly typical behavior\nthat Aitken nuclei show an inverted diurnal behavior subsequent to the\nwith some evidence of the diurnal cycle. Ozone data show high nighttime\neruption (i.e., the downslope wind gives higher values than the upslope wind,\nvalues, whereas Aitken nuclei data show background levels of about 150/cm3,\nwhereas light scattering shows no obvious short-term relationships, even\nand bsp (550 nm) data approach a low of about 10-7m- Also, during this\nthough upslope effects are apparent on the ninth and on the tenth).\nTypical diurnal oscillations in all quantities appear July 18-23, with a\nFigure 1. Hourly averages of wind\ngradual return to background levels during the early morning hours of July 23\ndirection and speed plotted at 3-hour\nand 24. Although fairly typical background aerosol conditions occur July\nintervals; hourly averages of relative\n24-26, the period July 28-30 is more typical of high ozone values associated\nhumidity; hourly averages of surface\nwith these strong southeasterlies.\nozone; hourly averages of conden-\nIt is interesting to compare the Bigg (1977) data with the data presented\nsation nuclei from a G.E. counter\nhere. The concentrations of large particles given in Bigg's Figure 8 follow very\nalong with occasional observations\nclosely the trends in light scattering data for the entire month, whereas there is\nfrom a Pollak counter; hourly\naverages of total light scattering at\n550 nm.\nWind\nWind\n100\n100\n0\n0\n60\n60\n40\n40\ngu\n20\n20\n0\n0\n5\n5\ncm\nEruption\n4\n4\nN\n3\n3\n60\n2\n2.\n-5\n-5\nPower Out\nE\n-6\n-6\n-7\n7\n-8\n-8\n1\n2\n3\n4\n5\n6\n7\n8\n9\n10\n11\n12\n13\n14\n15\nJuly 1975\na\n68","no correspondence between those data and the Aitken nucleus data for July\nand are given in Fig. 2 of this paper (along with Bigg's size distribution\n8-15. This further supports the contention that the volcano contributed a\ncurves). It is found that the upslope flow (solid curve a) shows a tendency\nsmaller-sized aerosol, probably derived from gas-to-particle conversion\ntowards larger particles with slope B III 3.3 and slightly concave upward.\nprocesses, whereas the nephelometer was seeing a larger haze-type aerosol.\nHowever, the downslope flow (solid curve b) shows a tendency towards\nDuring the last two weeks of July, conditions returned to the usual Mauna Loa\nsmaller particles with a steeper size distribution at larger particle sizes and\nbackground.\nslightly concave downward. Although the nephelometer cannot resolve a\nJuly 12 shows unusually high light scattering values in agreement with the\nbimodal size distribution such as that appearing in Bigg's Figure 7, there is a\nhigh concentrations of large particles in Bigg's Figure 8. Richard Hansen, of\ndefinite indication of a change of slope over its range. Note, however, that the\nthe NCAR High Altitude Observatory (personal communication), observed an\nvertical scale is arbitrary and that actual concentration must be provided by a\nunusual \"incredibly dramatic sky brightness pattern, mottled, milky white,\nsupplementary method. Furthermore, Heintzenberg (1976) investigated the\ndistinct haze, etc.,\" above Mauna Loa on that day. Although the source is\nMauna Loa aerosol in September 1975 with a Royco-225 particle counter. By\nunknown, it was most likely from the island itself, and it is significant that the\napplying a numerical inversion technique to three channels of Royco data and\nsurface aerosol measurements were strongly correlated with the remote effects\nfour channels of nephelometer data he found size distributions quite similar to\nof the upper troposphere.\nthose given by Bigg with a minimum for downslope flow occurring at about r\nSince the four-wavelength nephelometer essentially measures light\n= 0.4 um.\nscattering as a function of wavelength, a three-segment approximation of the\nAngstrom exponent may be derived, and information on the aerosol size\ndistribution may be inferred (see Butcher and Charlson, 1972, for a brief\ndiscussion). Calculations of the slope of the aerosol size distribution were\nperformed on the nephelometer data for the periods shown in Bigg's Figure 7\nWind\nWind\n100\n100\n0\n0\n60\n60\n40\n40\ngu\n20\non\n20\n0\n0\n5\n5\n4\n4\n3\n3\n2\n2\n-5\n-5\n-6\n-6\n-7\n7\n-8\n16\n17\n18\n-8\n19\n20\n21\n22\n23\n24\n25\n26\n27\n28\n29\n30\n31\nJuly 1975\n69","ACKNOWLEDGMENTS\nCONCLUSIONS\nI thank Dr. Bigg for the loan of his sampler and for his effort in reducing\nIt is suggested that Bigg's techniques are clearly applicable to the GMCC\nthe large amount of data from this instrument. I also thank Ed Lundin and\naerosol monitoring program and in fact provide useful information not\nCris Maeda for their assistance in analyzing the Mauna Loa aerosol data.\notherwise easily obtained. Furthermore, his size distribution data rather con-\nclusively support a bimodal size distribution with a minimum in the vicinity of\nr = 0.3 m.\nIn light of Bigg's results concerning the relative appearance of sulfuric acid\nand ammonium sulfate in the background aerosol at MLO, further investiga-\nREFERENCES\ntions of the influence of Mauna Loa volcano are currently underway.\nMendonca, B. G., and W. T. Iwaoka, 1969:\nAhlquist, N. C., and R. J. Charlson, 1969:\nThe trade wind inversion at the slopes of\nMeasurement of the wavelength dependence\nMauna Loa, Hawaii. J. Appl. Meteorol.,\nof atmospheric extinction due to scatter.\n8:213-219.\nAtmos. Environ., 3:551.\nNOAA, 1974a: Geophysical Monitoring for\nBigg, E. K., 1977: Some properties of the\nClimatic Change No. 1; Summary Report\naerosol at Mauna Loa Observatory. J. Appl.\n1972. Environmental Research Laboratories,\nMeteorol., 16:262-267.\nBoulder, Colorado.\nBodhaine, B. A., and R. F. Pueschel, 1972:\n2\n10th\nNOAA, 1974b: Geophysical Monitoring for\nFlame photometric analysis of the transport\nClimatic Change No. 2; Summary Report\nof sea salt particles. J. Geophys. Res.\n1973. Environmental Research Laboratories,\n77:5106-5115.\nBoulder, Colorado.\nBodhaine, B. A., and B. G. Mendonca, 1974:\nNOAA, 1975: Geophysical Monitoring for\nPreliminary four-wavelength nephelometer\nClimatic Change No. 3; Summary Report\nmeasurements at Mauna Loa Observatory.\n1974. Environmental Research Laboratories,\nGeophys. Res. Lett., 3:119-122.\n10-2\nBoulder, Colorado.\nButcher, S. S., and R. J. Charlson, 1972: An\nPueschel, R. F., and B. G. Menconca, 1972:\nIntroduction to Air Chemistry. Academic,\nSources of atmospheric particulate matter on\nNew York.\nHawaii. Tellus, 14:139-149.\nCharlson, R. J., 1972: Multiwavelength nephe-\nPueschel, R. F., B. A. Bodhaine, and B. G.\nlometer measurements in Los Angeles smog\nMenconca, 1973: The proportion of volatile\naerosol. J. Colloid Interface Sci., 39:240-265.\naerosols on the island of Hawaii. J. Appl.\nHeintzenberg, J., 1976: Determination in situ of\nO\n10-\nMeteorol., 12:308-315.\n0.5\nthe size distribution of high tropospheric\n0.2\n0.3\n1.0\n0.1\naerosol particles. Report prepared for NOAA-\nRadius ( um)\nGMCC.\nMendonca, B. G., 1969: Local wind circulation\non the slopes of Mauna Loa. J. Appl.\nMeteorol., 8:533-541.\nFigure 2. Estimate of the size distribu-\ntion of the Mauna Loa aerosol.\n(a) Upslope winds, July 21-22.\n(b) Downslope winds, July 21-23. Solid\nlines are three-segment approxi-\nmations derived from nephelometer\ndata with arbitrary vertical scale.\nDashed lines are from Bigg (1977).\n70","Equipment for measuring atmospheric\nelectricity was installed on MLO's 4-\nIn 1959, Bill Cobb and Byron Phillips\nacre site in 1960. The ground screen\nwere part of the atmospheric electri-\nis at the far left: The screen is now\ncity program at MLO. The instru-\ngone, and the building has been\nment shown measured electric current\nmoved. The building at the lower\nfrom the ionosphere to the ground.\nright housed the generator.\nThe current-measuring instrument is\nin the center of the ground screen.\nAtmospheric electricity instruments\nare housed in the building. The chim-\nney is the intake for air, which is\nsampled to determine electrical prop-\nerties of the atmosphere.\n71","in principle. With a container of manageable size, a slow expansion to match\nnormal rates of cooling of the air in clouds causes most of the available water\nICE NUCLEI AT MAUNA LOA\nto be deposited on the walls instead of on the cloud drops. There it promptly\nfreezes, shedding into the air ice crystals that are indistinguishable from, but\ncompletely independent of, those formed on nuclei. In addition the ice crystals\nE. Keith Bigg\nare difficult to detect.\nDivision of Cloud Physics, CSIRO\nMany of the early measurements probably include spurious ice crystals in\nthe count, and therefore are completely worthless. It is impossible, of course,\nSydney, Australia\nto know now whether precautions to avoid counting spurious crystals were\nadequate. In the absence of nearby special sources, such as silver iodide\ngenerators, blast furnaces, and so on, any concentrations exceeding 10/liter at\n-20°C or warmer must be suspect. The methods used to overcome the defects\nBACKGROUND TO THE MEASUREMENTS AT\nof a simple system have varied widely, but all have had to compromise\nMAUNA LOA OBSERVATORY\nbetween practicality and the provision of cloudlike conditions.\nThe \"spurious ice crystal\" problems were overcome in the early 1950s by\nBergeron's (1935) proposal that the formation of ice crystals is essential to\ncoating the container walls with glycerine or glycol. These substances do not\nthe initiation of precipitation in many clouds led to interest in the concentra-\nseem to hamper the measurements if used carefully in normal circumstances;\ntion of ice crystals in clouds and in the particles, or ice nuclei, on which they\nin warm humid situations, however, they rapidly become diluted, allowing ice\nformed. Pioneer work by Findeisen and Schulz (1944) shows that such\ncrystal production from the walls, whereas in very dry situations they become\nparticles are relatively rare and are present in quite variable concentrations.\ntoo concentrated and reduce cloud lifetime. Detection of ice crystals in early\nWhen Schaefer (1946) demonstrated that spectacular changes occurred in\nexperiments was accomplished usually by observing their scintillations in\nclouds seeded with dry ice, the prospect of beneficial weather modification\na\nbeam of light, but many natural particles scintillate, giving false counts.\ncreated considerably more interest in the nuclei. Clearly, the ability to predict\nSupercooled solutions of sugar or other soluble materials turned out to be very\noccasions when concentrations of natural ice nuclei would be too low for\nuseful for crystal counting, since they grow ice crystals only when an ice\nefficient production of precipitation would be helpful in selecting occasions\ncrystal falls into the solution, provided that the supercooling is kept to a\nsuitable for cloud seeding, and a knowledge of the sources of the ice nuclei\nminimum. Two of the most popular early methods of counting ice nuclei have\nand meteorological conditions should be sufficient to allow such a prediction if\nused either static \"mixing\" chambers, where the air is introduced into a cold\nthe sources were well defined and not too numerous.\ncontainer coated with glycerine and having a supercooled sugar solution at the\nIn principle, there could scarcely be a simpler measurement than that of\nbottom, or \"rapid expansion\" chambers. In the latter method the air is cooled\nfinding the concentration of ice nuclei. All that is necessary is to cool a sample\nto the temperature of the container walls without more than transient conden-\nof humid air to the temperature of interest and to count the ice crystals that\nsation; it is then cooled to some lower temperature by a sudden expansion. Al-\nfall from the resulting cloud. The only stipulations should be that cooling\nthough the results of measurements using these two methods may agree at one\nrates, concentrations, and sizes of cloud drops are comparable with those in\ntemperature, they usually disagree considerably at others, owing probably to\nnatural clouds. Unfortunately, measurements are rarely as simple as they seem\n72","the effects of different temperature-humidity histories or of cloud drop prop-\nThis theory stated that meteor dust reaching the earth's atmosphere on certain\nerties or lifetimes. Neither method really provides a good approximation to\ncalendar dates reached the lower troposphere 30 days later, where it seeded\ncloud conditions, though the maximum supersaturation and cloud drop prop-\nclouds and produced rain. Critics had claimed that this was entirely improb-\nerties in a mixing chamber are probably a better representation of those in\nable, but Bowen's answer was that measurements of ice nuclei would demon-\nnatural clouds than occurs in an expansion chamber.\nstrate the correctness of his hypothesis. Indeed, there was a certain measure of\nMore recent counting methods include the filter technique and several\nsuccess in early measurements which were concentrated on the month of\ncontinuous-flow cloud chambers, of which the \"acoustic counter\" is the most\nJanuary; Bowen claimed (though others disputed) well-defined anomalies in\nimportant. In the filter method all the particles in a sample of air are collected\nrainfall about January 12, 22, and 31. About a dozen rapid-expansion counters\non a filter; the filter is then cooled in a thermal diffusion chamber to the tem-\nwere made in Australia and shipped to various sites, including MLO. Later the\nperature of interest, and the humidity is raised; after a suitable interval the ice\nU.S. Weather Bureau constructed many similar instruments, and these were\ncrystals that form are counted. The temperature-humidity history of the air\nused both at Hilo and at MLO.\nsample could be made to match that in a cloud exactly if it were not for\nThe early results showed that, when the observatory was isolated from\nvapour depletion by hygroscopic particles and the filter itself; however, inter-\nlow-level air, concentrations of ice nuclei were always less than 0.1/liter even\naction between cloud drops and the nuclei is difficult, perhaps impossible, to\nat -24°C (The early work appears not to have been published in detail, al-\nmodel.\nthough I have a sample of such early data.) In other sites at this temperature,\nIn the acoustic counter, air is drawn into a cold chamber and sucked out\nnuclei were present to the extent of 10/liter or more. This result, of course,\nthrough a special orifice in which particles larger than a certain size make\nargued strongly against the meteor dust hypothesis, unless the unusual condi-\naudible clicks that can be counted. The early models of the acoustic counter\ntions at the observatory had caused some instrumental malfunction.\nwere designed more for plume tracking from silver iodide burners than for\nI made my first visit to MLO in April 1959 and was able to compare\ncounting natural ice nuclei, and a few spurious counts did not matter.\nresults from the operation of the U.S. Weather Bureau instrument used as an\nHowever, the attractions of a real-time printout soon led to the use of these\nexpansion chamber with those obtained when it was used as a mixing chamber\ncounters for counting natural nuclei under conditions where the inherent diffi-\nwith additional moisture supplied. The results were totally different: as a\nculties of ensuring appropriate humidification and ice-free operation rendered\nmixing chamber the instrument gáve concentrations of nuclei comparable with\nmany of the results useless. After many refinements the acoustic counter is ap-\nthose from other sites.\nparently capable of giving reproducible results when it is used with care.\nIn 1960 the U.S. Weather Bureau commenced continent-wide sampling of\nHowever, since, in common with all other methods in use, there is a possi-\nnuclei with expansion chambers, and that at Mauna Loa again showed un-\nbility of consistent bias due to failure to model natural clouds exactly, its\nusually low counts, except during periods of upslope flow (Price and Pales,\nresults cannot be considered absolute.\n1964). To verify my 1959 impressions, I spent a longer time at the observatory\nin 1961 (Bigg, 1964) and again found concentrations to be much the same in\nupslope and downslope flows. The difficulties that I experienced in dry condi-\nINITIATION OF PROGRAMS OF COUNTING ICE NUCLEI\ntions were an increase in concentration of the glycerine on the walls, leading\nAT MAUNA LOA OBSERVATORY\nto reduced cloud lifetime and droplet size. There was also an increase in sugar\nAt the time Mauna Loa Observatory (MLO) commenced operations,\nconcentration of the surface layer of the sugar solution, leading to failure to\nBowen's (1953) \"meteor dust\" hypothesis had created a major controversy.\ndetect ice crystals that fell into it. When diluted glycerine and sugar were used,\n73","the results were similar to those obtained elsewhere.\nchamber were collected on a moving paper roll and allowed to evaporate. Ice\nIn 1964, Droessler and Heffernan (1965) attempted to settle the question\nnuclei on the paper were subsequently revealed by humidifying the filter paper\nof whether humidity was the main variable by using the recently developed\nat a temperature of -15°C and pouring a supercooled solution over it to reveal\n\"millipore filter\" technique for counting the nuclei. Since the filters were not\nthe ice crystals. Any environmental effects at the observing site should there-\nprocessed on the site, the answer should have been independent of ambient\nfore have been avoided, while frost from the walls of the cold chamber should\nconditions. They attempted to allow for reductions in the count by hygro-\ncontain no ice nuclei and therefore fail to be counted.\nscopic particles (more numerous in the upslope air) by simultaneously\nDetailed measurements from the acoustic counter at Mauna Loa were not\nsampling filters at two volumes at three sites: the observatory, Hilo at sea\npublished, but the Japanese technique showed precisely what Droessler and\nlevel, and a midlevel site (Kulani prison). Their conclusion was that there was\nHeffernan's (1965) filters had shown earlier: no significant diurnal variation.\nno significant difference in ice nucleus content of the air above and below the\nFortunately, the 3-hourly measurements of the acoustic counter at Blue Glacier\ntemperature inversion. This result might have settled the question if the filter\nwere reported in detail in a University of Washington publication and could be\nmethod had not had some obvious defects and had not failed to model natural\ncompared with the 6-hourly measurements of the Isono et al. (1971)\nclouds in allowing for capture of nuclei by water drops. Another problem was\ninstrument. The fluctuations were found to be exactly out of phase. From this\nthat of \"background nuclei,\" i.e., those that come with the filters.\nI concluded, because of my doubts about the acoustic counter's performance,\nThe advent of an \"acoustic counter\" for ice nuclei (see Langer et al., 1967)\nthat it was responding to an ambient condition negatively correlated with ice\nprovided an opportunity to reexamine concentrations of ice nuclei on a con-\nnucleus concentrations.\nPossibly as a result of this disagreement - though it was never discussed\ntinuous and automatic basis. The first reported measurements with the counter\nin the literature - two acoustic counters were operated simultaneously at the\nat MLO were made from January to March 1967 (Nagamoto et al., 1967); these\nobservatory through 1971-1972 with some improvements in construction and\nshowed much the same sort of variation with upslope and downslope winds\nprocedure. The possibility that frost on the walls or erratic flow patterns\nthat had been found earlier by Kline (1963) with expansion chambers. I had\nwithin the counters could contribute to variations in apparent ice nucleus con-\nseen several of the early commercial counters in action and was unimpressed:\nthey clearly failed to work in low humidities and were prone to frost troubles\ncentrations was reduced by considering only those occasions when the reading\nof the two counters agreed. The results of a long series of observations have\nand changes in counting rate which were due to changes in internal air flow\nbeen reported and discussed by several authors (Mendonca and Langer, 1973;\npatterns. Langer (1971) conceded these points but claimed that modifications,\nMendonca and Pueschel, 1973; Fullerton et al., 1975). It was unequivocally\nwhich he has subsequently introduced, would avert these difficulties. At the\nshown once again that air coming from below the trade wind inversion\nWorkshop on Ice Nuclei held at Ft. Collins, Colorado, in 1971, two modified\ngenerally had a higher ice nucleus content than the air above the inversion.\ncommercial counters were tested which were plainly unreliable, SO that I still\nAt the same time, further filters were exposed, and some of these were\nfelt doubtful on these points. On the other hand, Langer himself operated a\nsent to me for analysis. The mean concentrations of ice nuclei are given below\nmore sophisticated counter which appeared to be very good.\n(in numbers per cubic metre active at -15°C):\nIn 1968 a most important experiment was organized at MLO, at Blue\nGlacier in Washington, and at other sites, to compare two independent tech-\nDate\nMorning\nAfternoon\nniques. The first used the acoustic counter, the results being analysed by\nHobbs et al. (1971a) and Hobbs et al. (1971b). The second used a novel instru-\n2/23-3/4/72\n2.7(16 filters)\n2.2 (11 filters)\nment described by Isono et al. (1971) in which ice crystals falling from a cold\n5/11-5/19/72\n4.7 19 filters)\n3.7 (20 filters)\n74","transient supersaturations on admitting the air sample, to ensure frost-free\nThe situation therefore is that the acoustic counter, the reliability of\noperation and proper detection of ice crystals. The tedious and time-\nwhich seems to have been adequately demonstrated, gives an entirely contrary\nconsuming nature of the operation of the instrument would preclude its use\nconclusion to the filter method, which itself has been found to agree\nexcept on an intermittent basis at times selected according to the origins of the\nreasonably well with other techniques in other localities.\nair. The fact that it gives a better representation than most other techniques of\nHow is this conflict to be resolved? At present the only simple solution\nthe processes in a cloud and is inherently simple outweighs its disadvantages.\nseems to be to invoke the demonstrated difference between nuclei activated\nIf the results from the Ohtake (1971) chamber should agree with the\nfrom the vapour phase (known as \"deposition nuclei\") and those which require\nfindings of the acoustic counter, there could be little doubt that failure to\ncontact with a water drop to become activated (known as \"contact nuclei\").\ndetect contact nuclei when filter methods were used was the reason for the\nAs I have used it, the filter method detects only the former, whereas the\nconflicting results. All the work that has gone into the measurements would\nacoustic counter could detect both. If we assume that nuclei from below the\nnot then have been wasted, for a valuable distinction between the nature of\ntrade wind inversion were exclusively \"contact\" nuclei, the results from the\nsea level and high-altitude nuclei would have been demonstrated.\ntwo methods would no longer be in disagreement.\nThere is a difficulty in this simple solution: neither the mixing chamber\nmeasurements that I had made using diluted glycerine and sugar nor the\nexpansion chamber measurements made by Droessler and Heffernan (1965)\nyielded unusually low concentrations of nuclei in air that appeared to have\noriginated from above the trade wind inversion. Langer (1971) suggested that\nbecause tests with clean air were not made, the chambers we used may have\nbeen contaminated, producing their own nuclei. Since this proposal would\nremove what appears to be an otherwise insuperable difficulty, I am now\nmore ready to accept it than I would have been in 1971.\nCONCLUSION\nFurther experiments should be carried out at MLO to verify that the\ndifferences are indeed due to the response of different techniques to different\ntypes of nuclei. Quite clearly they should be carried out by using methods\ncapable of detecting both deposition and contact nuclei. I would recommend\nthe use of Ohtake's (1971) \"cloud settling chamber\" for this purpose. The\nprinciple with this instrument is to generate a cloud in a region above the tem-\nperature inversion of a static cold chamber and allow the cloud drops to fall\nthrough the precooled air sample. With care it is possible to avoid more than\n75","REFERENCES\nBergeron, T., 1935: On the physics of clouds\nLanger, G., J. Rosinski, and C. P. Edwards,\nand precipitation. Proc. 5th Assembly UGGI,\n1967: A continuous ice nucleus counter and\nLisbon, Vol. 2, p. 156.\nits application to tracking in the troposphere.\nBigg, E. K., 1964: Geographical differences in\nJ. Appl. Meteorol. 6:114-125.\nconcentrations of ice nuclei. Mon. Weather\nMendonca, B. G., and G. Langer, 1973: Ice\nRev. 92:355-356.\nnucleus counts in varying ambient humidities\nBowen, E. G., 1953: The influence of\nusing an NCAR ice nucleus counter. J.\nmeteoritic dust on rainfall. Aust. J. Phys.,\nAtmos. Sci., 30:1452-1454.\nMendonca, B. G. and R. F. Pueschel, 1973: Ice\n6:490-497.\nDroessler, E. G., and K. J. Heffernan, 1965: Ice\nnuclei, total aerosol climatology of Mauna\nnucleus measurements in Hawaii. J. Appl.\nLoa, Hawaii. J. Appl. Meteorol. 12:156-160.\nNagamoto, C. T., J. Rosinski, and G. Langer,\nMeteorol., 4:442-445.\nFindeisen, W., and G. Schulz, 1944: Experi-\n1967: Ice nuclei concentration in Hawaii\nmentelle Untersuchungen uber die atmos-\nduring the period 7 January to 10 March\npharische Eisteilchenbildung. Forschungs-und\n1967. J. Appl. Meteorol. 6:1123-1125.\nOhtake, T., 1971: Cloud settling chamber for\nErfahrungsberichte des Reichswetterdienstes,\nice nuclei count. Preprints, Int. Conf.\nSer. A, No. 27.\nFullerton, C. M., C. Garcia, and G. Langer,\nWeather Modification, Canberra, Australia,\n1975: Nine months of ice nucleus monitoring\nAm. Meteorol. Soc., 38-41.\nat Mauna Loa, Hawaii. Meteorol.\nPrice, S., and C. Pales, 1964: Ice nucleus\nRundschau, 28:178-190.\ncounts and variations at 3.4 km and near sea\nHobbs, P. V., and G. C. Bluhm, and T.\nlevel in Hawaii. Mon. Weather Rev.\nOhtake, 1971a: Transport of ice nuclei over\n92:207-221.\nthe North Pacific Ocean. Tellus, 23:238-239.\nSchaefer, V. J., 1946: The production of ice\nHobbs, P. V., C. M. Fullerton, and G.C.\ncrystals in a cloud of supercooled water\nBluhm, 1971b: Ice nucleus storms in Hawaii.\ndroplets. Science, 104:457.\nNature, 220:90-91.\nIsono, K., M. Komabayashi, T. Takeda,\nT. Tanaka, K. Iwai, and M. Fujiwara, 1971:\nConcentrations and nature of ice nuclei in\nthe rim of the North Pacific Ocean. Tellus,\n23:40-59.\nKline, D. B., 1963: Evidence of geographical\ndifferences in ice nuclei concentrations. Mon.\nWeather Rev., 91:681-686.\nLanger, G., 1971: Comments regarding\noperation of the NCAR ice nucleus counter. J.\nAtmos. Sci., 28:1074-1076.\n76","VOLCANOES AND ICE NUCLEUS MONITORING\nAT MAUNA LOA OBSERVATORY\nC. M. Fullerton\nCloud Physics Observatory, Dept. of Meteorology\nUniversity of Hawaii, Hilo, Hawaii\nC. J. Garcia\nHigh Altitude Observatory, National Center for\nAtmospheric Research², Hilo, Hawaii\nINTRODUCTION\nMeasurements of atmospheric ice-forming nuclei (hereafter referred to as\nIN) have been carried out at Mauna Loa Observatory (MLO), Hawaii, almost\nfrom the time it was established in 1956. The location, altitude, and physical\nsite of the observatory, combined with a relatively well defined diurnal\ncirculation, make MLO a particularly attractive and potentially valuable site\nfor IN measurements.\nIt is somewhat disappointing, therefore, that IN measurements at MLO\nhave been rather sporadic, that most programs were carried on for only a few\ndays to a few weeks, and that the longest period of continuous IN\nmeasurements lasted less than 1 year and ended 5 years ago.\nIdeally, a regular program of observations should have been established\nand maintained years ago, along the lines of the outstanding carbon dioxide\n-\nContribution No. 77-9 of the Department of\nMeteorology, University of Hawaii.\nThe National Center for Atmospheric Research is\nsponsored by the National Science Foundation.\n77","The assumption has been suspect, however, since Price and Pales (1963)\nand solar radiation monitoring projects long identified with MLO. A number of\nreported a decrease in IN concentration during local eruptions. Moreover,\nreasons may be suggested for the lack of a long-term IN monitoring program:\nwhen Pueschel and Langer (1973) sampled IN concentration near potential\n1. A fundamental skepticism exists concerning the importance of IN as\nsources in Hawaii, they found the lowest concentration in the effluents of\nvariable constituents of the atmosphere. After years of worldwide study the\nMauna Ulu, an active vent of the Kilauea volcano. When an NCAR counter\nbasic nature, structure, composition, and origin of such nuclei remain largely\nand membrane filters, both at -20°C, were used, measurements directly in the\nunknown. Under these circumstances it is difficult to assess precisely what role\nfume showed IN concentrations of about 0.5 IN/liter. Thus it appears that local\nthese elusive particles may play in atmospheric processes.\nvolcanic activity is not a direct source of IN. How, then, may the observations\n2. Although a number of methods and devices have been developed to\nof Hobbs et al. (1971b) be explained? In them, clear evidence of pronounced\ndetect and measure IN, no single device has become established as the\npeaks in IN concentration, as measured at MLO, followed volcanic eruptions in\nrecognized (standard) instrument. Workshops conducted to compare\ninstruments and technical approaches to the problem, including the activation\n1969.\nPerhaps the simplest way of explaining the variety of results reported by\nprocess employed, have not resolved the differences. This is probably to be\nearlier investigators is to consider local volcanic activity as an indirect cause of\nexpected, in view of the fundamental uncertainties mentioned above.\nIN activity through initiation of burning. Thus lava may set surrounding scrub\n3. Several investigators have made IN measurements at MLO, but none has\nvegetation and trees afire and in this way generate IN. Hobbs and Locatelli\ncarried out a carefully monitored continuous program of sufficient duration to\n(1969) observed IN from a natural forest fire. Pueschel and Langer (1973)\nestablish even seasonal trends at the site. As investigators come and go, and\nidentified the burning of leaves during sugarcane harvesting as the most\ninstrument types and methods change, the basic continuity that such a\nprolific source of IN on Hawaii, a finding verified by Fullerton et al. (1975)\nprogram requires has never become established.\nduring a 9-month IN monitoring program at MLO. Laboratory studies by\n4. Certain questions of long-standing interest remain unresolved. From our\nLanger et al. (1974) indicate that smoke from burning ohia-lehua branches is\nperspective one of the most fascinating is the possible relationship between\nlocal volcanic activity and the IN concentration measured at MLO. This\nan active IN source at -20°C.\nOur thesis is that some of the peaks in IN concentration observed at MLO\nproblem has been the subject of several research papers and a recent brief\nduring local volcanic eruptions arise from vegetation burned as a secondary\nreview which concluded that \"the correlation between ice nuclei and volcanic\neffect of lava flows. Not all increases in IN concentration, however, are so\nactivity on Hawaii remains largely inconclusive\" (Langer et al., 1974).\nrelated to volcanic activity. We know that sugarcane fires are the dominant\nlocal IN source and that major \"IN storms\" of distant origin occasionally arrive\nVOLCANIC ACTIVITY AND ICE NUCLEI\nat MLO.\nData on the type and amount of material burned during eruptions are\nWe believe that confusion has arisen in the past because of the assump-\ndifficult to obtain for past events. We may, however, estimate the potential\ntion that IN are present, or at least potential, in fresh volcanic effluent. This\nfor burning on the basis of these considerations:\nassumption derives from the report of Isono et al. (1959) that volcanic erup-\n1. Eruptions within the summit caldera of Kilauea volcano, especially those\ntions in Japan (principally, Mount Asama in November 1958) are a source of\nin Halemaumau, should be accompanied by little or no burning, simply\nIN. If the assumption is valid for Hawaii, then local eruptions should produce\nbecause frequent volcanic activity prevents vegetation from becoming\nIN, which under favorable wind conditions should increase the measured IN\nestablished.\nconcentration at MLO.\n78","2. Eruptions confined to a given location may initially set the peripheral\nwith the first downslope flow following both phases 3 and 4 of the eruption\nvegetation afire, but such burning would not long continue unless lava flowed\n(Hobbs et al., 1971a), and volcanic smog was noted at MLO between 12 and 24\nfrom the source region into forested areas.\nhours following the outbreak of volcanic activity.\n3. Rift eruptions into areas long free of lava inundation should set afire the\nThus, depending on local wind conditions, rainfall, and the strength of\nexisting vegetation and continue to burn new areas as the lava moves.\nthe trade wind inversion, volcanic effluent may be detected at MLO within a\n4. Rainfall before an eruption may decrease the material burned; rainfall\nfew hours to a few days following an eruption. It is possible, of course, that\nafter an eruption may extinguish fires that might otherwise continue.\nstrong northerly winds and/or heavy rainfall would completely prevent transit\nIf burning vegetation does produce IN or release IN to the atmosphere,\nof volatile material to MLO. In the absence of high winds and rainfall, once\nexisting wind patterns will determine the trajectory and transit time to MLO.\nvolcanic smoke reaches MLO, the trade wind inversion and local circulation\nRainfall during transit may reduce the IN concentration by \"washout and\nwill tend to retain the effluent over the island for several days unless synoptic\nrainout\" of aerosol particles. Kilauea volcano is located at an elevation of 1.2\nconditions develop to purge the atmosphere of contaminants.\nkm on the southeast flank of Mauna Loa, whereas MLO lies on the northern\nIn the past ten years there have been four periods of local volcanic\nslope of the mountain at an elevation of 3.4 km. Thus the volcano is not\nactivity during IN monitoring at MLO. Each of these periods will be examined\nfavorably situated with respect to upslope winds advecting nuclei directly to\nfor evidence of increases in IN concentration that may be associated with\nMLO.\nburning vegetation set afire by lava. Such cases usually appear as modulations\nPrice and Pales (1963) argue that volcanic effluent probably reaches MLO\nof the normal IN variation due to other local sources (mainly sugarcane\nby either or both of two primary trajectories. The first, illustrated by the\nburning) and to \"IN storms\" of remote (off-island) origin.\nKilauea Iki eruption in 1959, is \"southwestward along the slope and over the\nlower portions of the ridge of Mauna Loa by the prevailing trade wind\nERUPTIONS AND BURNING VEGETATION\nnorthward along the western (leeward) side of the island, and thence into the\nBefore considering the eruptive episodes in detail we will outline our\nMauna Loa-Mauna Kea saddle through its western entrance. Smoke of\nexpectations of how each may have affected the IN concentrations measured at\nvolcanic origin first reached MLO from the south-southwest on November 21,\nMLO. Two types of eruptions must be considered: those in the summit area of\nseven days after the eruption began.\"\nKilauea volcano, and those occurring along the volcano's rift zones. Three\nThe other, more direct trajectory was observed in the Kapoho rift\nsamples of each type of eruption are available.\neruption (1960) where effluent moved \"directly westward into the Mauna\nSummit eruptions\nLoa-Mauna Kea saddle, whence it ascended to the observatory within the\ndaytime upslope wind, appearing there first less than thirty hours after the\nDuring the Kilauea Iki eruption in 1959 and the eruption of Halemaumau\nonset of volcanic activity.\" In these two cases, effluent arrived at MLO during\nin 1971, only limited areas of forest, on the periphery of lava fountaining,\nupslope winds within 1 to 7 days after initiation of volcanic activity.\nwere set afire. We would expect, under the most favorable wind conditions,\nOther trajectories exist, however, depending on the presence and strength\nonly a slight enhancement in the IN background for a few days after the initial\nof the trade wind inversion. If the inversion is absent or weak, a vigorously\neruption.\nexpelled effluent may rise vertically and later be brought down to MLO by the\nThe eruption of 1967-1968 took place entirely within the summit caldera,\ndownslope wind flow. This trajectory apparently was followed in the\nand there is no record of any vegetation burning. We would expect, therefore,\nAloe-Alae craters eruption (1969) when IN concentrations increased abruptly\nno IN activity as a result of this eruption.\n79","Price and Pales found \"the median count with effluent actually present was\nRift zone eruptions\nonly slightly greater than the background median\nthe\nvalues\nbeing,\nThe Kapoho eruption in 1960 broke out in a sugarcane field, and con-\nrespectively, 4.0 and 3.4 IN/10 liters at -24°C. However, the median count in\nsiderable vegetation burned during the eruption. We would expect rather\nthe period before the eruption was 3.3 IN/10 liters, whereas after the eruption\nmajor increases in IN count, especially in the early stages of the eruption.\nit was 4.2 IN/10 liters.\nSimilarly, the third and fourth phases of the Mauna Ulu eruption (June 1969),\nThe results of Price and Pales appear to be precisely what might be\nin flowing to the sea, crossed and set afire major forested areas. Certainly,\nexpected from a summit eruption in which the lava flow was confined within\nthese two phases should be evident as major increases in IN concentration at\nthe walls of the crater, and very little vegetation was burned. Only a slight en-\nMLO. Indeed, they are the clearest examples of isolated burning, caused by\nhancement of the IN count would be expected, and that is what was observed.\nvolcanic activity, in the events described in this report. The Mauna Ulu series\nWe view the data presentation of Price and Pales as indicating a general\nof eruptions in February-May 1972 is more complicated, but forest was\nincrease in IN activity at MLO between about November 25 and December 28,\nburned, and IN concentration should have been enhanced during this period.\n1959, although the absolute count remained only slightly above background\nlevels.\nKILAUEA IKI ERUPTION (1959)\nVariations in IN concentration during January and February 1960 appear\nThe eruption began at 2008 on November 14, 1959, on the southeast wall\nto be different from those measured during the Kilauea Iki eruption. While the\nof Kilauea Iki crater, which lies on the eastern edge of the Kilauea caldera.\noverall background count is lower, several peaks appear, especially in\nFissures quickly lengthened both eastward and westward, and there soon was\nmeasurements made during downslope wind conditions.\na line nearly 0.8 km long of lava fountains about 15 m high. Most of the\nfountaining lasted for only a few hours, and by the next morning only two\nERUPTION (1960)\nsmall fountains remained active.\nAt 1930 on January 13, 1960, lava broke out in a sugarcane field a few\nActivity began to increase on November 17, and by the afternoon of\nhundred meters northwest of the village of Kapoho, on the east rift of Kilauea\nNovember 18 a single fountain reached a height of 230 m; on November 19 it\nvolcano. The erupting fissure opened gradually eastward until the line of lava\noccasionally reached 300 m. On November 21 the height of the fountain\nfountains was almost 1 km long. By midnight the three westernmost fountains\nvaried between 50 and 380 m until early evening, when fountaining suddenly\nhad ceased activity, but farther east was a row of 15 to 20 fountains, ranging\nceased at 2000. For the next four days there was no activity.\nin height from 10 to 100 m. Steam blasts accompanied the fountains sporadi-\nJust after midnight on the night of November 25-26, the lava fountain\ncally throughout the eruption.\nsuddenly resumed activity, beginning a series of brief eruptive episodes\nBy the morning of January 14 the number of lava fountains was reduced\nranging in length from 2 to 32 hours, separated by brief quiescent periods. On\nto seven, and through most of the eruption there were only one to three active\nNovember 28 the lava fountain reached a height of at least 460 m, possibly\nfountains. The main fountain occasionally reached heights of nearly 300 m,\neven 520 height unprecedented in the records of Hawaiian volcanoes.\nwith ash clouds to 600 m.\nCycles of eruption and withdrawal were repeated at least 14 times until the\nLava entered the ocean on the morning of January 15 with a spectacular\neruption ended on December 19, 1959.\nburst of rolling clouds of white steam. Gradually, over the next few days, lava\nPrice and Pales (1963) show a plot of IN counts at MLO from October 15,\nengulfed nearly the entire village of Kapoho. On January 28 an ash-laden\n1959, to March 15, 1960. Measurements were made with an expansion\nsteam cloud rose from the 1959 vent in Kilauea Iki crater.\nchamber twice daily, at 0800 and 1400. In the case of the Kilauea Iki eruption,\n80","By the time the Kapoho eruption ended on February 9, 1960, a total\nHALEMAUMAU ERUPTION (1967-1968)\nvolume of about 115 X 106 m³ of lava had poured out (three times that\nAt about 0230 on November 5, 1967, lava fountains broke out along a\ncontained in the Kilauea Iki lava lake), and the shoreline had been pushed\nline extending nearly north-south across the floor of Halemaumau. Lava\nseaward nearly 0.8 km, adding approximately 2 km2 of new land to the island.\npoured into the crater at a rate of more than 1.2 X 106 m³/h. Fountains\nIn the 10-day period (January 6-15, 1960) just prior to the Kapoho\ngradually increased in height from 15 to 60 m. By midnight the strength of the\neruption, the Hilo Airport Weather Station received a total of about 60.7 cm\nactivity was decreasing, and an hour later, fountaining stopped.\nof rain. Daily accumulations ranged from 1.9 to 18.0 cm. Rainfall was\nThe eruption resumed on the morning of November 9 and gradually\nespecially heavy during January 13-15, when 38 cm were recorded in the 3-\nincreased in strength over the next several days, with occasional lava jets as\nday period. Volcanic haze was first observed in Hilo on January 18.\nhigh as 15 to 20 m. This phase of the eruption ended on November 19 at 1945.\nThe weather then changed markedly, with no measurable rainfall in Hilo\nThis alternation of periods of fountaining and lava lake activity with\nbetween January 19 and February 10. Rainfall again became prevalent from\nperiods of drainback and inactivity continued for 3 months, through a total of\nFebruary 11 through 17, when 39 cm fell in daily amounts ranging from 1.8 to\n28 active periods. Phase 29, which began on February 27, 1968, signalled the\n14.8 cm. This rainfall seems to have effectively washed out the volcanic\nbeginning of continuous lava lake activity, which lasted until the end of the\neffluent present at MLO.\neruption on July 8, 1968.\nThese heavy rainfall periods resulted from synoptic changes accompanied\nFrom late January to March 1968, simultaneous IN measurements were\nby advection of nucleus-laden air to Hawaii. We would therefore interpret the\ncarried out at four widely separated locations around the rim of the North\nhigh IN counts recorded by Price and Pales (1963) just before the onset\nPacific Ocean. Observations were taken at MLO and in Alaska, Washington\n(January 10-15) and just after cessation (February 16-22) of the Kapoho\nState, and Japan. Two instruments were used in Hawaii. Throughout the\neruption as \"IN storms\" of remote origin, similar to those discussed by Hobbs\nmeasurement period (January 27 to March 31), IN active at -21°C were\net al. (1971a).\nmonitored with the NCAR continuous IN counter (Langer et al., 1967). From\nDuring the drought period (January 19-February 10), several IN peaks\nabout February 1 to March 1, observations also were taken with an IN\nappeared in the measurements taken at MLO. The increases in IN concentration\ncollector (at -15°C), and samples were examined by electron microscope\nthat were recorded about January 21 and 30 and February 2 and 5-7 clearly\n(Isono et al., 1971). Bigg (1973) has noted some ambiguity in the published\nexceeded any IN peak recorded during the Kilauea eruption. Although most\nresults from the two instruments during measurements made in Washington\npeaks occurred in the morning (0800) measurements, under apparently clear\nState as a part of this program. He asserts that a detailed comparison of the\nair downslope wind conditions, they may still indicate IN generated by\ntwo data sets shows fluctuations in IN concentration that were exactly out of\nvegetation burned during the eruption. Similar IN peaks during the 1969 rift\nphase.\neruption also appeared during downslope flow at MLO (Hobbs et al., 1971b).\nDetailed variations in IN activity cannot be extracted from published\nIt should be noted that the nightly downslope flow at MLO normally\nrecords of the Hawaii measurements. Unfortunately, Hobbs et al. (1971a)\nchanges to a daytime upslope wind at just about 0800. Measurements made\npresented the -21°C IN counts at MLO in the form of 1-week running means,\nduring this transitional period may not be truly representative of either\nand Isono et al. (1971) showed the -15°C data as 28-hour running means.\nupslope flow (conditions below MLO) or downslope flow (air descending from\nHaving been involved with the NCAR counter measurements at MLO\nhigher elevations).\nduring this period, we have some doubt about the quality of the results.\nThrough most of the period we were essentially learning how to operate the\n81","With the exception of two small IN peaks on June 7 and 9, again during\ninstrument. Later measurements, in the summer of 1969 and in 1971-1972,\ndownslope wind conditions, the IN count remained virtually zero from noon\nwere taken with much greater care and appreciation of potential sources of\non June 4 to noon on June 12. Island-wide rainfall occurred on June 11-12, 1\nerror.\nBoth the -21°C and the -15°C data are consistent in showing that the\ncm of rain falling in the volcano area on June 12 up to 0900.\nThe third phase of the eruption began at 1330 on June 12 and lasted for\nlowest IN activity measured at the four observation sites occurred at MLO. The\n21.5 hours. Lava fountains up to 150 m again shot from the fissure between\nwinter of 1967-1968 was unusual in Hawaii. Trade winds were almost totally\nAloi and Alae craters, and a voluminous and rapid lava flow poured\nabsent from early December to early March. Trade wind conditions returned\nsouthward down and over the cliffs of the Hilina fault system to form a broad\nabout the second week in March and continued through the month. Wide-\npool on flat land about 0.6 km from the coastline. Volcano observatory\nspread, heavy rainfall occurred in the last week of March. The major increase\npersonnel reported that considerable forest burned during this lava flow.\nin IN concentration in late March is attributed to an \"IN storm,\" similar to\nLate on the afternoon of June 12 the IN concentration at MLO began to\nthose that occasionally appeared in Alaska and Washington State during the\nincrease. During the downslope wind flow on June 12-13 the IN count peaked\nsimultaneous monitoring program. We do not associate this with local IN\nat approximately the same concentration as that measured on June 1-2. MLO\nsources.\npersonnel reported \"heavy volcanic smog\" at the observatory early on the\nIn the absence of detailed information on the variation in IN count at\nmorning of June 13. The IN count returned to background levels about noon\nMLO, we can only conclude that the concentration was very low (less than\non June 13 and remained practically zero until June 26. Hobbs et al. (1971b)\n0.1 IN/liter) until the last few days of March. It is probable that only minor\nargue convincingly that the June 12-13 IN peak was related to the third phase\nfluctuations about the background level occurred during the measurement\nof the eruption.\nperiod.\nThe fourth phase of the eruption began at 2145 on June 25 and ceased\nabout 9 hours later. Fountains 125 m high in the same east rift area again sent\nALOI-ALAE CRATER ERUPTION (1969)\nvoluminous lava flows down the Hilina fault scarps. A narrow tongue of lava\nentered the ocean just east of Apua Point.\nOn May 24, 1969, lava fountains broke out at three locations along the\nVolcanic smog was first noted at MLO on the morning of June 26, but the\neast rift zone from just southeast of Aloi crater to Kane Nui o Homo. Lava\nIN count remained low until downslope winds began about 2100. The IN con-\npoured into Alae crater and covered part of the Chain of Craters Road.\ncentration then increased rapidly to values similar to those recorded on the\nActivity ceased on May 25 but recommenced on May 27 and continued until\nnights of June 1-2 and 12-13, with a clear peak during downslope flow on\nMay 29.\nJune 26-27. By noon on June 27 the IN concentration returned to background\nIN measurements began at MLO on June 1 using two NCAR counters, one\nlevels and remained low until the afternoon of June 29. Undoubtedly, the June\noperating at -21°C and the other at -15°C. Pronounced IN peaks were\n26-27 IN peak was related to the fourth phase of the eruption.\nrecorded during downslope wind flow on June 1 and 2, and minor IN peaks\nSynoptic conditions then changed dramatically, beginning on June 28-29.\nduring downslope flow the succeeding two mornings. Afternoon values,\nCirrus clouds were common, and the winds at MLO were constantly from the\nduring upslope flow, were low but clearly higher than the background IN\nsoutheast at speeds exceeding 10 m/s. Island-wide rainfall occurred from June\nconcentration. We suggest that the enhanced IN activity on June 1-4 may be\n29 to July 11, with several days of moderate rainfall in the volcano area.\nrelated to the second phase of the eruption (May 27-29) through localized\nThese conditions suggest that the \"IN storm,\" measured at MLO between June\nburning of the forest around the eruption site.\n82","30 and July 10, was not of island origin but rather was due to advection of IN\nclosely in time to the annual hiatus in sugarcane field burning. Under these\ninto the air above Hawaii by upper level synoptic disturbances (Hobbs et al.,\nconditions, very low IN concentrations would be expected at MLO, and this is\n1971b).\nwhat was observed (Fullerton et al., 1975).\nHALEMAUMAU ERUPTION (1971)\nMAUNA ULU ERUPTION (1972)\nOn August 14, 1971, there was a 10-hour eruption (0900-1900) in\nMauna Ulu began fountaining again on the afternoon of February 6, 1972.\nHalemaumau, with fountaining up to 75 m. Forest on the edge of the crater\nConsiderable quantities of new lava were deposited from Alae crater on\nwas set afire, and some 26 ha were burned before the fire was brought under\nFebruary 12-16. Activity continued, but at much less magnitude, until\ncontrol on August 15 at about 2000.\nFebruary 25, when lava overflowed the Mauna Ulu summit crater. From\nFortunately, the day before the eruption, IN monitoring began at MLO\nFebruary 25 to March 4-5, lava flowed out to form a tongue about 400 m\nusing two modified NCAR ice nucleus counters, operating in parallel. Detailed\nwide and about 4 km long. Fountaining averaged about 40 m, with bursts to\nIN plots from the two instruments for the period August 13-23, 1971, are\n75 m. Low-level activity continued over the next two weeks.\ngiven by Langer et al. (1974). These plots indicate enhanced IN activity at MLO\nOn March 18, new vents opened west and southeast of Mauna Ulu and\nfollowing the eruption, in particular, pronounced peaks in the daytime upslope\nwere vigorously active until March 23-24. At the same time, the eruptions in\nflow. Additional details on the monitoring program and island-wide rainfall\nMauna Ulu crater remained strong. Throughout April and early May the\nare given by Fullerton et al. (1975).\neruption continued at a fairly steady rate, lava flowing from one or more\nAnother eruption began about 1920 on September 24, 1971. The initial\nvents essentially without interruption.\noutbreak occurred on the floor of Kilauea caldera, with lava flows cascading\nFullerton et al. (1975) show plots of IN concentration at MLO from\ndown into Halemaumau and fountains rising to about 50 m. Sporadic and\nFebruary 14 to May 15, 1972. Numerous high IN peaks appear. Those on\nirregular activity continued until the evening of September 29. Fullerton et al.\nFebruary 26-28 and March 22-27 may well reflect vegetation set afire by\n(1975) show plots of IN activity from August 13 to September 28. IN concen-\nmigrating lava flow. High IN concentrations on several days in April and May\ntrations, very low on September 22-23, increased progressively from\nmay similarly be related to volcanic activity. Unfortunately, the chronology of\nSeptember 24 to September 27. IN peaks appeared daily in the upslope flow.\nthe eruption is not sufficiently detailed to positively correlate IN peaks with\nThe interpretation of IN events in August and September 1971 is difficult.\nvegetation ignited by lava. Many of the days exhibiting low IN concentrations\nIN peaks in mid-September may be related to sugarcane fires, which were\noccurred during periods of heavy and widespread rainfall, for example, on the\ncommon because of dry weather island-wide. Furthermore, at the same time\nafternoon of February 18 and 23 and on March 4-5.\nthese two brief eruptions took place in the summit area, activity continued on\nthe east rift, the Mauna Ulu vents emitting copious fume. During the\nSeptember 24-29 Halemaumau eruption, fountaining also occurred at Mauna\nUlu, and as late as mid-October, sporadic sounds of fountaining and splashing\nlava could be heard from the area. After this time, however, seismic evidence\nindicated that the 1971 phase of the Mauna Ulu eruption had ended.\nThere was no volcanic activity on the island from mid-October 1971\nthrough early February 1972. This lull in volcanic activity corresponded\n83","Ice Nucleus (IN) Monitoring and Volcanic (VOL) Activity on the Island of Hawaii\nIN Activity\nIN Monitoring\nexpected/observed\nVolcanic Activity\nReferences\nPeriod\nIN Instrument\nSlight EB*/Slight EB*\nKilauea Iki (summit)\n10-liter expansion\n15 Oct 59 to\nIN:\n14 Nov 59 to 19 Dec 59\nPrice and Pales (1963)\nchamber\n15 Mar 60\nKapoho (E rift)\nEB* and isolated IN\n(modified Bigg-Warner),\nVOL:\n13 Jan 60 to 19 Feb 60\npeaks/\n-24 °C daily counts at\nMacdonald and Abbott\nIsolated IN peaks\nLava entered sea on\nat 0800 and 1400\n(1970)\n15 Jan at 1930 hrs\n(\"IN storm\")\nHalemaumau (summit)\nNone/Cannot extract\nNCAR counter at -21 °C;\n25 Jan 68 to\nIN:\n5 Nov 67 continued\ndetailed IN activity\nHobbs et al. (1971a)\n31 Mar 68\ncontinuous\nfrom published data;\nIN collector at -15°C\nIsono et al. (1971)\nthrough 28 phases\n27 Feb 68: Phase 29\nvery low IN count at\n(Isono)\nVOL:\nMLO (\"IN storm\")\nbegins: continuous\nMacdonald and Abbott\nlava lake to 8 July\n(1970)\nMajor IN peaks from\nAloi-Alae (E rift)\nTwo NCAR counters\n4 June 69 to\nIN:\nphases 2, 3 and 4/\nHobbs et al. (1971 a,b)\nPhase 1: 24-25 May\ncontinuous; one at\n10 July 69\nExpected peaks ob-\n2: 27-29 May\n-21 °C, the other at\nVOL:\nserved (\"IN storm\")\nMacdonald and Abbott\n3: 12-13 June\n-15°C\n4: 25-26 June\n(1970)\nWright et al. (1975)\nLava entered sea on\n26 June at 0835\nHalemaumau (summit)\nEB* and isolated IN\nTwo NCAR counters in\n14 Aug 71 to\nIN:\npeaks/Expected EB*\n14 Aug; 24-29 Sept\nparallel at - °C but\nMendonca and\n14 May 72\nobserved along with\nMauna Ulu (E rift)\nPueschel (1973)\nwith different super-\noccasional IN peaks; low\nLanger et al. (1974)\n1969 activity con-\nsaturations\nIN activity in absence\ntinues to mid-Oct\nFullerton et al. (1975)\nof volcanic activity\n6 Feb 72 to mid-May\nVOL:\nHawaiian Volcano Ob-\nand continuing\nservatory field notes\nenhancement of IN\nbackground count","CONCLUSIONS\nA reasonable case appears to have been made relating volcanic activity on\nthe island of Hawaii to increased IN concentration at MLO through vegetation\nset afire by lava flows. With the exception of the 1968 Halemaumau eruption,\nwhere adequate documentation of the IN measurements is not available,\nexisting data indicate the following:\n1. Kilauea summit eruptions, characterized by minimum burning, produce\nonly slight enhancement of the IN concentration at MLO.\n2. Rift zone eruptions through forested areas may, under optimum condi-\ntions, give major IN peaks.\nIN concentrations at MLO are modulated by prevailing winds and clearly\naffected by widespread rainfall. Sugarcane field burning is probably the major\nsource of IN on the island and thus the determinant of the IN background\nREFERENCES\nlevel. \"IN storms\" occasionally move into Hawaii to further complicate the\nBigg, E. K., 1973: Ice nucleus concentrations in\nLanger, G., C. Garcia, B. Mendonca, R.\nproblem.\nremote areas. J. Atmos. Sci., 30:1153-1157.\nPueschel, and C. Fullerton, 1974: Hawaiian\nThe accompanying table summarizes volcanic activity and IN monitoring\nFullerton, C.. C. Garcia, and G. Langer, 1975:\nvolcanoes-a source of ice nuclei? J.\nprograms at MLO from 1959 to 1972, along with our interpretation of how\nNine months of ice nucleus monitoring at\nGeophys. Res., 79:873-875.\nlava-induced burning of forested areas may have affected the IN concentration.\nMauna Loa, Hawaii. Meteorol. Rundsch.\nMacdonald, G., and A. Abbott, 1970:\nWe urge that a program of continuous IN measurements be established\n28:178-190.\nVolcanoes in the Sea. University of Hawaii\nHobbs, P., and J. Locatelli, 1969: Ice nuclei\nPress, Honolulu.\nand maintained at MLO. If this is done, future measurements may confirm that\nfrom a natural forest fire. J. Appl.\nMendonca, B., and R. Pueschel, 1973: Ice\nforest burning is the connection between eruptions and enhanced IN concentra-\nMeteorol. 8:833-834.\nnuclei, total aerosol, and climatology at\ntions. The type and amount of material burned should be determined.\nHobbs, P., G. Bluhm, and T. Ohtake, 1971a:\nMauna Loa, Hawaii. J. Appl. Meteorol.\nTrajectory and transit times should be assessed and information on the\nTransport of ice nuclei over the North\n12:156-160.\nPacific Ocean. Tellus, 23:28-39.\nstrength of the trade wind inversion obtained. A detailed chronology of\nPueschel, R., and G. Langer, 1973: Sugar cane\nHobbs, P., C. Fullerton, and G. Bluhm, 1971b:\nfires as a source of ice nuclei in Hawaii, J.\neruptic activity should be maintained. With these data it should be possible to\nIce nucleus storms in Hawaii. Nature Phys.\nAppl. Meteorol. 12:549-551.\nestablish clearly the link between eruptions and the ice nucleus concentrations\nSci., 230:90-91.\nPrice, S., and J. C. Pales, 1963: Local volcanic\nmeasured at MLO.\nIsono, K., M. Komabayasi, and A. Ono, 1959:\nactivity and ice nuclei concentrations on\nVolcanoes as a source of atmospheric ice\nHawaii. Arch. Meteorol. Geophys.\nnuclei. Nature, 183:317-318.\nBioklimatol. Ser. A., A13:398-407.\nIsono, K., M. Komabayasi, T. Takeda, T.\nWright, T., D. Swanson, and W. Duffield,\nTanaka, K. Iwai, and M. Fujiwara, 1971:\n1975: Chemical compositions of Kilauea east-\nConcentration and nature of ice nuclei in rim\nrift lava, 1968-1971. J. Petrol. 16:110-133.\nof the North Pacific Ocean. Tellus, 23:40-59.\nLanger, G., J. Rosinski, and C. Edwards, 1967:\nA continuous ice nucleus counter and its\napplication to tracking in the atmosphere. J.\nAppl. Meteorol. 6:114-125.\n85","ICE NUCLEATION PROPERTIES OF SOILS FROM\nHAWAII AND THE CONTINENT\nGerhard Langer and David Goto\nNHRE, National Center for Atmospheric Research\nBoulder, Colorado\nINTRODUCTION\nOur interest in the nucleation properties of soils arose from a study\ncarried out at the Mauna Loa Observatory, Hawaii, to determine the extent to\nwhich local sources form ice nuclei. A major source of nuclei was found to be\nthe smoke or, more specifically, the ash particles from cane field fires\n(Pueschel and Langer, 1973) - the sugarcane leaves are set on fire just before\nharvest. A subsequent observation that lava dust nucleates effectively (Langer\net al., 1974) led us to question whether the soil from the sugarcane plantations\nhas enhanced nucleation properties because it contains lava and trace elements\nfrom fertilizer and agricultural sprays. An alternative explanation is that the\nplant growth process leads to the synthesis of a nucleating material, and\ntherefore all the activity originates with plant ashes. In this paper we discuss\nsoil nucleation activity and the possibility that dust generated by vehicles and\ncultivation is an erratic source of nuclei. Soil samples from the continental\nUnited States were also examined to provide a basis for comparison with\nHawaiian soils, since to our knowledge no such work had previously been\ncarried out.\nOn the island's eastern coast, the\nsmoke plume from burning sugar\ncane (lower right) rises into cumulus\nclouds.\n86","EXPERIMENTAL PROCEDURE\nThe soil samples are collected right at the surface, dried at 100 °C, and\nsieved through a 44-um screen. The fluidized bed disperser shown in Fig. 1 is\nused to produce an aerosol from about 0.1 g of soil sample; the dust is\nIce\ngradually released from the bed of glass beads, and particles larger than 10 um\nNucleus\nCounter\nin diameter are separated out by a cyclone. The size distribution and the\nCyclone To Remove\nconcentration of the resulting aerosol are determined with a light-scattering\nExcess\nParticles Over 10u\ncounter. Another stream of the aerosol is diverted to two NCAR ice nucleus\nRoyco\ncounters (Langer, 1973), one operating at -16°C and the other at -20°C. The\nParticle\nCounter\ndata are thus obtained in terms of the fraction of the particles nucleating out\nSecondary Venturi,\nof the total aerosol. We assume that the particles of less than 0.5-um diameter,\nDisperser And Diluter\nwhich are not counted by the light-scattering device, are not involved in the\nnucleating process. That assumption certainly merits verification; however, the\nFlowmeter\nnecessary resources were not available for our study.\nIn the case of the Hawaiian soils, some additional work was done to\n15 lpm\nnarrow down the particle size range in which most of the nucleus activity\nClean Compressed Air\nexists. By adjusting the airflows in the aerosol generator we reduced the\n50 lpm\nfraction of particles leaving the system by a factor of 10 in the 3- to 9-um\nPinch Clamp\nrange. As can be seen from Table 1, there is essentially no activity in the 0.5-\nH\nFilter\nto 3-um-size range. By making use of this fact and recalculating the nucleus\nFilter\n1 cm\nactivity on the basis of the particles in the 3- to 9-um range for the Hawaiian\nPinch Clamp\nsoil, we obtained the results shown in parentheses in Table 1. Those results\nDip Tube For Operation Above\nshow a high activity for the large particles; they also indicate that cane ashes\nOr Below Top Of Bed\nare no more active than soil.\nFluidized Bed (300-400 Dia.\nIn Hawaii cane fires do represent an important source of ice nuclei in\nGlass Beads) Plus Soil Sample\n7 cm\ncomparison with windblown soil particles because the soil is usually too moist\nto release much dust. After a fire the residual ashes of the dry leaves become\nPorous Glass Disc\nairborne when the harvesting machines pick up the stalks. The smoke itself,\nFlowmeter\nmostly oily organic material, does not nucleate until the burned leaves are\nmechanically disturbed and ash particles are airborne. In contrast to the\n30 lpm Clean Compressed Air\nsituation in Hawaii, windblown soil could be an important source of ice nuclei\nin the continental United States, especially during dust storms in the\nsouthwest.\nFigure 1. Fluidized bed disperser to\ngenerate soil aerosols.\n87","Table 1. Soil Ice Nucleus Activity\nSource of soil\n% of particles active as ice nuclei\n0.5- to 9-um range 0.5- to 3-um range\nAt-16°C At -20 °C At-16°C\nAt-20°C\nMiles City, Montana, 15 dif-\nferent locations, includes\ncultivated fields.\n0.01-0.06\n0.08-0.9\n-\n-\nNE. Colorado, 2 locations,\nprairie soil.\n0.01-0.07\n0.04-0.2\n-\n-\nTexas, 4 locations, includes\nREFERENCES\nwindblown dust.\n0.01-0.1\n0.09-0.7\n-\n-\nNebraska, 4 locations, culti-\nPueschel, R. F., and G. Langer, 1973: Sugar-\nvated fields and prairie soil.\n0.01-0.02\n0.02-0.03\n-\n-\ncane fires as a source of ice nuclei in Hawaii.\nHawaii:\nJ. Appl. Meteorol. 12:549-551.\nPapaya field, barren lava\nLanger, G., 1973: Evaluation of NCAR ice\n0.06 (3)*\nnot detectable\nsand until 1974.\n0.01 (.5)*\nnucleus counter, Part I: Basic Operation. J.\nabove background\nAppl. Meteorol. 12:1000-1011.\nLanger, G., C. J. Garcia, B. G. Mendonca,\nSugarcane field, freshly\nR. F. Pueschel, and C. M. Fullerton, 1974:\nplanted, in use 15 years.\n.008 (.4)*\n0.07 (4)*\nHawaiian volcanoes - a source of ice\nSugarcane field, mature\nnuclei? J. Geophys. Res., 79:873-875.\nplants, in use 20 years.\n.008 (.5)*\n0.05 (3)*\nSugarcane field, recently\nharvested, in use 20\n.01 (.6)*\n0.08 (5)*\nyears.\nAshes from cane leaves,\n11\nright after fire.\n.009 (.4)*\n0.04 (2)*\nThe Montana soil samples were obtained through the courtesy of Dr. A. B. Super,\nBureau of Reclamation; Texas and Nebraska soil samples from Dr. D. Gillette, NCAR;\nHawaii soil samples from Dr. J. Miller, Mauna Loa Observatory, NOAA.\nRecalculated on the assumption that all of the activity is in the 3- to 9-um size range.\n88","AEROSOL CONCENTRATIONS AND FALLOUT AT\nThe isotope of major interest in the HASL fallout studies has been Sr90\nbecause of its potential toxicity. Generally following calcium's geochemical\nMAUNA LOA\npathways (e.g., United Nations, 1972; Bennett 1976a) after being deposited on\nthe earth's surface (primarily in precipitation), Sr90 enters the food chain by\nH. L. Volchok\nbeing taken up in vegetation and ultimately appears in measurable quantities\nHealth and Safety Laboratory, ERDA\nin many foods, cow's milk being perhaps of most concern (Bennett, 1976b).\nFrom time to time a shorter-lived isotope, Sr89 was also measured.\nNew York, N. Y.\nAll of the fallout deposition data are published quarterly (Hardy, 1977)\nand are reproduced here as Table 1. The data are also used in computing the\nannual fallout and deposit on the earth's surface (Feely, 1976) as part of HASL's\nThe Health and Safety Laboratory (HASL) participated in two\nunique responsibility for maintaining a complete inventory of Sr90 from\nenvironmental programs at the Mauna Loa Observatory (MLO): (1) collections\nnuclear explosions.\nof monthly fallout deposition, and (2) continuous high-volume aerosol\nIn April 1976, additional collectors were installed at MLO to obtain\nsampling.\nseparate samples of wet and dry deposition. The wet and dry samples are\nbeing studied specifically with the goal of obtaining \"global baselines\" for\nmajor cations, anions, and other trace substances in precipitation.\nFALLOUT DEPOSITION COLLECTION\nFallout deposition collection at Mauna Loa was started in February 1959\nAEROSOL SAMPLING\nas part of an expanding research program concerning the global distribution of\nThe HASL Surface Air Sampling Program is a direct outgrowth of a\ndebris from nuclear weapons tests (Hardy and Klein, 1959; Harley, 1976). At\nprogram initiated by the U.S. Naval Research Laboratory (NRL) in 1957 and\nthat time this and 21 other new sites were chosen and set up at U.S. Weather\ncontinued through 1962 (Lockhart et al., 1964). The NRL program at Mauna\nBureau stations, through the cooperation of Dr. Lester Machta.\nLoa was established in 1960. Direction was transferred to HASL in 1963. The\nThe sampling device that permitted this rapid expansion of our fallout\nobjective of the program is to study temporal and spatial distributions of\nstudies is a plastic funnel-ion exchange column combination. The substances of\nspecific natural and man-made radionuclides as well as other trace substances\ninterest (originally fission product cations) are retained in the column while the\nbulk of the precipitation passes through. Dry fallout is also captured and\nin surface air.\nThe present network of 21 stations extends from about lat. 76°N. to lat.\nretained in the top of the column. Only the actual column needs to be\n90°S. Air is drawn through filter paper continuously at the rate of about\nreturned for analysis. Hence the system is practically accident-free and can be\n1 m³/min and, for the most part, analyses are carried out on monthly\noperated by relatively untrained personnel. The record at Mauna Loa is\ncomposites. We have routinely analyzed the filters for 7 radionuclides and\nvirtually unmatched in our entire 150-station network. Through more than 17\nstable lead; however, as many as 13 different isotopes have been measured for\nyears of continuous monthly sampling (206 samples), only two samples failed\nspecial purposes. Additional stable trace metal analyses are being added to the\nto arrive at HASL.\n89","ram. Since portions of many of the samples dating from 1968 have been\nthe record for successful delivery of samples from the site to the laboratory\nI, these and other analyses may be carried out at selective sites\nhas been very close to perfect.\nspectively.\nWe acknowledge with great respect the on-site personnel at MLO, who\nAll of the surface air data from this program are also published quarterly\nhave SO successfully maintained our sampling programs year in and year out:\ny et al., 1977), and the quality of the analytical results is summarized\nWe are well aware of the functional difficulties in performing complex\neach year (Toonkel et al., 1977). Tables 2 and 3 are included as examples\ntechnical manipulations at this altitude and therefore take this opportunity to\ne available data. Table 2 lists the monthly Sr90 results for the entire\ncommend all of the people associated with the Geophysical Monitoring for\nd of the program. Table 3 summarizes the concentrations of the natural,\nClimatic Change program (and its predecessor programs at MLO) for this\nic ray-produced nuclide, Be7. These analyses were started in 1970. Again,\nextraordinary cooperation.\nTable 1. Monthly Sr90 Fallout at MLO, Hawaii\nLat. 19°32'N, Long. 155°35'W, Elevation 3,401 m (Column)\n(Millicuries per square kilometer)\nJan.\nFeb.\nMar.\nApr.\nMay\nJune\nJuly\nAug.\nSep.\nOct.\nNov.\nDec.\nTotal\n1959\n0.49\n0.01\n0.02\n0.00\n0.03\n0.01\n-\n*\n0.04\n0.60\n-\n1960\n0.06\n0.07\n0.07\n0.05\n0.06C\n0.13C\n*\n*\n0.00C\n0.00C\n*\n*\n0.44\n1961\n0.01C\n0.05C\n0.05C\n0.06C\n0.00C\n0.01C\n*\n0.02\n0.02\n0.03\n0.25\n-\n-\n1962\n0.16\n0.02\n0.41\n0.41\n0.57\n0.07\n0.00\n0.02\n0.05\n0.00\n2.31\n0.02\n4.04\n1963\n0.32\n0.15\n0.84\n0.31\n1.42\n0.76\n0.39\n0.18\n0.17\n0.03\n0.02\n0.23\n4.82\n1964\n0.12\n0.10\n0.31\n0.23\n0.23\n0.01\n0.20\n0.06\n0.03\n0.05\n0.06\n0.09\n1.49\n1965\n0.16\n0.06\n0.09\n0.18\n0.27\n0.08\n0.07\n0.07\n0.04\n0.02\n0.03\n1.07\n1966\n*\n0.25\n0.04\n0.04\n*\n0.01\n0.02\n*\n0.01\n0.06\n0.02\n0.04\n0.49\n1967\n0.03\n0.01\n0.06\n0.05\n0.04\n0.01\n0.03\n0.01\n0.01\n*\n*\n*\n0.25\n1968\n0.01\n*\n0.05\n0.11\n0.02\n0.05\n0.02\n0.04\n0.04\n0.01\n0.02\n*\n0.37\n1969\n0.02\n0.02\n0.04\n0.02\n0.02\n*\n*\n0.02\n0.01\n0.01\n0.16\n1970\n*\n0.01\n*\n0.11\n0.04\n*\n0.01\n0.01\n*\n*\n0.02\n0.20\n1971\n0.02\n0.01\n0.01\n0.14\n0.05\n0.04\n0.02\n*\n0.01\n*\n0.30\n1972\n0.10\n0.02\n0.02\n0.02\n0.02\n0.01\n0.03\n*\n*\n*\n*\n0.04\n0.26\n1973\n*\n*\n*\n*\n*\n*\n*\n*\n*\n*\n0.00\n1974\n*\n0.04\n0.05C\n0.05C\n0.05\n0.02\n*\n*\n*\n0.21\n1975\n0.03\n0.04\n*\n*\n*\n*\n*\n0.01\n*\n*\n0.08\n1976\n*\n0.00\nData not available\nZero or trace\nC: Proportioned from originally consolidated data\n90","Table 2. Surface Air Sr90 Concentration at MLO, Hawaii\n(Femtocuries/cubic meter)\nSep.\nOct.\nNov.\nDec.\nJune\nJuly\nAug.\nApr.\nMay\nJan.\nFeb.\nMar.\n4.14\n5.86\n6.31\n1957\n-\n23.00\n22.10\n22.50\n27.90\n1958\n8.11\n9.45\n27.90\n-\n1.84\n1.73\n5.59\n6.25\n1960\n4.19\n-\n-\n-\n-\n-\n2.70\n1.45\n1.45\n7.61\n7.61\n4.95\n4.95\n2.70\n4.44\n4.44\n1961\n1.73\n1.73\n- No data\n19.50\n19.50\n9.19\n9.19\n17.60\n17.60\n39.30\n36.70\n36.70\nErrors are less than 20% except:\n1962\n22.00\n22.00\n39.30\n19.50\n81.60\n39.00\n17.30\n22.00\n20.80\n89.70\n70.90\n87.30\nA-error between 20% and 100%\n1963\n51.30\n78.80\n49.90\n12.60\n8.96\n8.26\n9.10\n68.50\n84.80\n42.90\n36.10\n15.40\nB-error greater than 100%\n1964\n42.60\n96.20\n49.90\n4.19\n6.95\n16.50\n18.40\n13.90\n7.87\n4.64\n6.34\n1965\n20.30\n40.80\n27.60\n22.80\n9.86\n6.84\n2.58\n1.94\n1.10\n1.97\n3.60\n16.50\n14.00\n14.10\n1966\n8.07\n8.40\n1.40\n0.79\n0.68\n3.82A\n0.98\n5.63\n2.75\n1.80\n1967\n5.38\n5.59\n3.24\n-\n5.80\n3.08\n2.66\n2.03\n1.07\n0.66\n1.80\n6.80\n5.18\n1968\n2.57\n1.38\n3.70\n4.34\n2.46\n1.80\n2.17\n1.14\n1.08\n4.54\n5.76\n4.88\n1969\n2.28\n1.79\n2.73\n2.27\n2.62\n0.87\n1.05\n0.00\n8.16\n5.71\n4.86\n1970\n1.58\n4.08\n7.74\n8.25\n2.29\n1.45\n1.01\n0.89\n1.22\n4.15\n2.70\n7.18\n10.40\n5.11\n1971\n2.67\n2.44\n0.41\n0.57\n1.96\n1.66\n0.66\n0.58\n0.47\n1.60\n2.12\n2.32\n1972\n2.05\n2.09\n0.42\n0.16\n0.14\n0.16\n0.80\n0.50\n0.21\n1.52\n1.21\n1973\n0.52\n1.18\n0.68\n1.24\n1.76\n0.99\n0.66\n0.71\n0.87\n1974\n0.67\n2.47\n4.28\n3.47\n3.89\n3.18\n0.12\n0.20\n0.23\n1.07\n0.43\n0.30\n2.49\n2.81\n1.45\n1975\n1.30\n2.06\n1.52\n0.47\n0.40\n1976\n0.43\n-\n-\n-\nTable 3. Surface Air Be7 Concentration at MLO, Hawaii\n(Femtocuries/cubic meter)\nNov.\nDec.\nAug.\nSep.\nOct.\nMay\nJune\nJuly\nFeb.\nMar.\nApr.\nJan.\n166.00\n247.00\n197.00\n298.00\n181.00\n141.00\n246.00\n253.00\n1970\n178.00\n221.00\n260.00\n209.00\n215.00\n164.00\n229.00\n205.00\n1971\n236.00\n250.00\n242.00\n167.00\n172.00\n136.00\n255.00\n168.00\n176.00\n149.00\n168.00\n304.00\n215.00\n1972\n241.00\n206.00\n132.00\n176.00\n165.00\n173.00\n151.00\n118.00\n255.00\n196.00\n216.00\n1973\n189.00\n197.00\n194.00\n213.00\n161.00\n186.00\n218.00\n194.00\n224.00\n247.00\n213.00\n201.00\n1974\n136.00\n191.00\n247.00\n188.00\n226.00\n246.00\n231.00\n259.00\n276.00\n225.00\n307.00\n249.00\n184.00\n231.00\n1975\n199.00\n227.00\n191.00\n1976\n262.00\n-\n-\n-No data\nErrors are less than 20% except:\nA-error between 20% and 100%\nB-error greater than 100%\n91","REFERENCES\nBennett, B. G., 1976a: Sr-90 in human bone-\nISOTOPIC COMPOSITION OF LEAD AEROSOLS AT\n1975; Results for New York City and San\nFrancisco. USERDA Report HASL-308.\nMAUNA LOA OBSERVATORY\nBennett, B. G., 1976b: Sr-90 in the Diet-\nResults through 1975. USERDA Report\nHASL-306.\nFeely, H. W., 1976: Worldwide deposition of\nH. J. Simpson and E. A. Catanzaro\nSr-90 through 1975. USERDA Report\nHASL-308.\nLamont-Doherty Geological Observatory\nFeely, H. W., L. Toonkel, and M. Schonberg,\nPalisades, New York\n1977: Radionuclides and lead in surface air.\nUSERDA Report HASL-315, Appendix.\nHardy, E. P., 1977: Appendix to Health and\nSafety Laboratory's Environmental\nQuarterly. USERDA Report HASL-315\nINTRODUCTION\nHardy, E. P., and S. Klein, 1959: Strontium\nprogram. Quarterly Summary Report,\nIndustrial societies have begun to appreciate, especially during the last few\nHASL-69, USAEC TID-4500.\ndecades, the importance of understanding the effects of their activities on the\nHarley, J. H., 1976: A brief history of long-\natmosphere. Degradation of air quality in urban areas is one of the most\nrange fallout. USERDA Report HASL-306.\nobvious results of modern civilization; there have been other, more subtle\nLockhart, L. B., Jr., R. L. Patterson, Jr., A. W.\nSanders, Jr., and R. W. Black, 1964:\nchanges on a global scale. Nuclear weapons testing in the atmosphere,\nSummary report, Fission product\nespecially during the early 1960s, produced radioactive debris that spread\nradioactivity in the air along the 80th\nthroughout the troposphere and lower stratosphere. Combustion of fossil fuels\nmeridian (west) 1957-1962. NRL Report\n-\nhas produced a readily measurable and continuing increase in the CO2 content\n6104.\nof the global atmosphere (Pales and Keeling, 1965; Keeling et al., 1976).\nToonkel, L., M. Schonberg, and H. W. Feely,\n1977: HASL surface air sampling program,\nthe quality of analysis-1975. USERDA\nReport HASL-315.\nUnited Nations, 1972: Ionizing radiation;\nLevels and effects, Vol. 1: Levels, a report of\nthe United Nations Scientific Committee on\nthe effects of atomic radiation. U.N.\nPublication Sales No. .72.IX.17/1972.\n92","chemical differences can potentially be exploited to help define the anthro-\nAlthough the effect of humans on hemispheric scale aerosol patterns is less\npogenic portion of background aerosols. Heavy metal concentrations of\nclear, measurements of electrical conductivity of marine air from surface ships\naerosols from remote areas have been determined by a number of\nindicate that the aerosol burden of the air over the North Atlantic probably\ninvestigators. During 1970-1971, aerosols collected at Mauna Loa Observatory\nhas substantially increased over the last half century (Cobb and Wells, 1970).\n(MLO) were analyzed to determine which heavy metals would be the most\nThe potential global climatic implications of significant changes in atmospheric\npromising as indicators of that portion of the aerosols derived from distant\ngases and aerosols are still intensively debated, but there is no question about\n(greater than 1,000 km) urban sources (Simpson, 1972a). On the basis of the\naltered urban climates and measurable health effects on a local scale during air\nanalytical techniques used in that study, observed concentrations of metals in\npollution episodes. Considerable research is now in progress on a number of\nsamples from MLO and other sites on the island of Hawaii, and relative\npotential interactions of man with large-scale climatic trends.\nstrengths of natural and anthropogenic sources of several metals (Fe, Mn, Zn,\nIt is important to establish the extent and consequences of changes in the\nCu, Ni, Pb), it was concluded that lead was the best indicator of man's\ncharacteristics of the atmosphere far removed from local pollution sources,\ncontribution to aerosols in remote regions. Duce et al. (1975) have discussed\nparticularly with respect to aerosols. Aerosol concentrations in the upper tro-\nthe enrichment relative to average crustal ratios of a number of trace elements\nposphere and some remote areas of the lower troposphere are very small in\nin aerosols from remote areas. This enrichment may be due primarily to\ncomparison with most continental areas, and especially with urban areas. The\nnatural processes which supply volatile trace elements to the atmosphere or to\nterm background aerosols has often been used for aerosols in remote areas of\npollution sources. Thus defining processes that are most important to the\nthe lower troposphere and much of the upper troposphere, but the relative\naerosol budget of lead in remote areas could provide better understanding of\nimportance of the sources and many of the chemical and physical properties of\nthe atmospheric transport of a large number of trace elements.\nthese aerosols are still largely undefined (Junge, 1968). Background aerosols\nA number of other studies of atmospheric lead concentrations in remote\nare probably a mixture derived from both natural and anthropogenic sources\nareas have been made. Chow et al. (1969) used isotope dilution mass spec-\nin continental areas and by natural processes at the sea surface. Since the rela-\ntrometry to analyze air filter samples collected from shipboard in the Pacific.\ntive importance of the various sources is not established, we cannot accurately\nThey estimated lead concentrations in central Pacific aerosols to be on the\nassess the potential alteration by humans of the balance of the background\norder of 1 ng/m3 of air (1 ng = 10-9 g), with significant uncertainty introduced\naerosol burden.\nby shipboard contamination problems. The results of several studies of back-\nAerosols in urban regions are chemically substantially different from\nground aerosol lead concentrations were summarized by Chow et al. (1972),\naerosols in remote areas of continents, the most obvious differences being the\nand isotope dilution mass spectrometry data for remote mountain sampling\nhigh concentrations of sulfur and heavy metal in urban aerosols. These\n93","sites in California were reported, indicating an annual average baseline for the\ncontinental United States of ng/m3. Published Indian Ocean values range\nfrom 1 to 4 ng/m3 (Egorov et al., 1970); values for arctic areas have been\nreported to be 0.2 (Egorov et al., 1970) to 0.5 ng/m3 (Murozumi et al., 1969).\nHoffman et al. (1972), using atomic absorption spectrometry, reported an\nSTABLE ISOTOPE COMPOSITION AS A TRACER FOR THE\naverage of 3 ng/m3 on the windward side of Oahu, while Simpson (1972a),\nSOURCE OF AEROSOL LEAD\nalso using atomic absorption spectrometry, suggested that the concentration of\nIn studies of human contribution of metals to the overall composition of\nlead in aerosols collected above the trade wind inversion on the island of\naerosols, lead has several advantages relative to other atmospheric heavy\nHawaii averaged about 1-2 ng/m3. Volchok (1973) has summarized lead con-\nmetals introduced in large amounts from urban areas. One is the relatively\ncentration data for the five most remote of the twenty surface-air monitoring\nhigh proportion of total aerosol lead that can be directly attributed to a single\nstations maintained by the Health and Safety Laboratory of ERDA primarily\nsource: automobile exhaust. In addition, there are large variations in the\nfor study of radioactive debris from nuclear weapons tests. These stations, two\nisotopic composition of anthropogenic aerosol lead. There are four nonradio-\nin the Northern Hemisphere and three in the Southern Hemisphere, all have\nactive (stable) isotopes of lead: 204 Pb, 206 Pb, Pb, and 208 Pb. All of these\nmedian values below 10 ng/m³ of aerosol lead, the lowest values at all the\nisotopes were present in the initial material that accreted to form the solar\nsites being in the range of 0.2 to 2 ng/m3 found by other investigators. All of\nsystem, and the amounts of the latter three isotopes have increased with time\nthe published values for the concentrations of aerosol lead in the central\nbecause they are the stable daughter products of the radioactive decay of U,\nPacific are quite low, averaging about 3 orders of magnitude less than the\n235 U, and 232 Th (Patterson et al., 1955). The isotopic composition of lead\nvalues found in urban continental air.\naveraged over the bulk of the crust of the earth has evolved in a reasonably\nUnfortunately, there are no measurements of aerosol lead in remote areas\ncoherent fashion to a composition similar to that represented by pelagic\nthat predate the large-scale use of lead alkyl additives to gasoline, so there is\nsediments in the world ocean today (Chow, 1958; Chow and Patterson,\nno direct evidence that aerosol lead burdens in remote regions have increased\n1962), although there is significant regional variation. Lead ores, on the other\nappreciably because of leaded gasoline combustion. However, on the basis of\nhand, have considerable variation in stable isotopic composition depending\nmass balance calculations a number of investigators believe that most of the\nupon a number of complicated processes including the geologic period in\npresent total standing crop of aerosol lead is probably anthropogenic\nwhich the concentration of the lead into ore bodies occurred (Brown, 1962).\n(Patterson, 1965). Estimates of the natural level of aerosol lead in remote areas\nThe isotopic composition of the lead used for gasoline additives is variable\nvary greatly but in general are substantially lower than the 1 to 2 ng/m3\nbecause different ore leads are used by different tetraethyl lead producers, and\ncurrently found (Patterson, 1965). The most dramatic evidence suggesting that\nso it is difficult to predict the isotopic composition of anthropogenic aerosol\ncurrent levels of background aerosol lead are far above natural levels is the\nlead of any given region. However, a few investigators, as discussed below,\nincrease of lead content found in polar snows from Greenland and Antarctica\nhave been able to determine a great deal about urban regional variations in\n(Murozumi et al., 1969). These measurements, made by isotope dilution mass\nisotopic composition of aerosol lead and to use these variations to deduce the\nspectrometry on large samples with good stratigraphic control, indicate that\nsource of the lead in environmental samples. These studies of isotopic compo-\ncurrent Greenland snow contains lead in amounts hundreds of times larger\nsition have generally focused on areas of relatively high aerosol lead content\nthan snows of a few thousand years ago.\nand have not been used to explore the isotopic composition of aerosols from\nremote regions.\n94","because the local lead ore in Missouri had an anomalously high radiogenic\nChow and Johnstone (1965) demonstrated that aerosol leads from Los\nisotopic composition and could be readily distinguished from nearly all gaso-\nAngeles, snow lead from a rural area in California, and lead from gasolines\nline-derived leads. A similar study in California (Rabinowitz and Wetherill,\nmarketed in the Los Angeles area all had almost identical isotopic\n1972), this time near a smelter emitting leads with a very nonradiogenic\ncompositions, which could be readily distinguished from average crustal lead\n\"signature,\" gave comparable results. It was shown that horses from near the\nas represented by lead found in Pacific sediments. The dominance of automo-\nsmelter site that had died of lead poisoning had derived approximately equal\nbile exhaust lead in the isotopic composition of environmental lead samples\namounts of lead from the smelter and from local automobile exhaust. Chow\nnear major roads has been clearly shown in several subsequent studies (Chow,\nand Earl (1972) measured the lead isotopic composition of coals and found it\n1970; Ault et al., 1970). Chow (1971) later demonstrated that aerosols and\nto be substantially different from that of most gasoline leads, thus indicating\nsoils from a number of cities had lead isotopic compositions identical with the\nthe possibility of distinguishing aerosol lead derived by coal burning from\ngasoline marketed in those cities. There can be little question that aerosol lead\nautomobile exhaust lead. In general, lead smelters and coal combustion appear\nin urban areas is derived almost entirely from local automobile exhausts. One\nto be relatively small sources of aerosol lead in comparison with automobile\nof the most interesting findings of the latter study was the large spread in\nexhaust except near point sources, and do not significantly affect the global\nisotopic composition among the urban lead aerosols from region to region:\nscale budget of aerosol lead. Studies of lead smelter and coal lead isotopic\naerosol leads from Southeast Asia were very primitive (low in radiogenic\ncompositions do indicate, however, the potential for distinguishing two\nlead), approaching the composition of some of the most ancient lead ore\nsources of aerosol lead if the isotopic compositions are significantly different.\ndeposits, while aerosol leads from the United States and Europe were much\nmore radiogenic. The isotopic composition of urban environmental leads was\nextremely large, approaching the total range found for economically important\nlead ores (Brown, 1962; Chow, 1971) except for the most anomalous lead ores\nwhich have received very large radiogenic lead contributions. The observed\nLEAD ISOTOPE MEASUREMENTS ON AEROSOL\nrange of variation for urban environmental leads is extremely large in\nSAMPLES FROM MLO\ncomparison with the precision of the measurements and with the degree of\nA study of aerosol chemistry on the island of Hawaii was made during\nagreement between the gasoline leads and aerosol leads within each urban\n1970-71, much of the work being based on samples collected at MLO\narea. Thus aerosol leads from different urban regions have characteristic\n(Simpson, 1972a). In general, it was shown that aerosols could be collected at\nisotopic compositions which potentially could be exploited to help determine\nMLO that were not contaminated by lead from towns along the coast of the\nthe fate of lead aerosols traveling far from the relatively discrete urban source\nisland of Hawaii and that lead concentrations above the trade wind inversion\nareas.\nwere in the range found by Chow et al. (1969) for central Pacific samples col-\nWhen lead is supplied to environmental samples from two sources of\nlected on shipboard. The chemistry of sodium, potassium, calcium, and mag-\ncomparable magnitude, it is frequently possible to delineate the magnitude of\nnesium in aerosols collected above the trade wind inversion at MLO was shown\neach source with sufficient sampling and to determine the relative importance\nto be substantially different from fresh marine aerosols and also from \"frac-\nof the contribution of each to an individual sample. Rabinowitz and Wetherill\ntionated\" marine aerosols that could periodically be collected at the observa-\n(1972) used isotopic measurements to define the extent of local lead contami-\ntory site (Simpson, 1972b). The aerosol cation chemistry at MLO resembles\nnation from a smelter and milling operation in Missouri, in the presence of a\ngreatly that of precipitation from continental areas (Junge, 1963), suggesting\nlarge diffuse source of aerosol lead from automobiles. This was possible\n95","also analyzed for lead isotopic composition (Table 1). Of the 30 samples from\nTable 1. The Isotopic Composition of Aerosol Lead from Hawaii\nMLO for which lead aerosol concentrations were reported (Simpson, 1972a) all\nbut 6 had concentrations of - 3 ng/m3 or less, 1-2 ng/m3 being typical. Most\nof the samples with higher values (maximum of - 23 ng/m3 were from\nSample\nSample\nCollection\nRatios\nperiods in which local contamination of the atmosphere at the observatory site\nNumber\nCode*\nPeriod\nPb/204Pb\nPb/207\nPb\n206\nPb/208\nPb\nng/m³\nby automobile exhaust lead was dominant. During a few well-documented\n(1970)\nperiods, large numbers of visitor automobiles had come to the observatory site\n1\nP 4\n11/9-11/16\n18.90\n1.210\n0.4926\n1.1\nwhile aerosol collection was underway. These sampling periods had lead con-\n2\n9\n12/2-12/16\n18.52\n1.189\n0.4978\n1.0\ncentrations 5 to 10 times greater than those found at other periods. As an\n3\nP 10\n12/2-12/16\n18.10\n1.164\n0.4809\n1.1\nextreme example, during the last week of December 1970 approximately 500\n4\nP 13\n12/23-12/30\n18.65\n1.195\n0.4896\n5.5\nautomobiles visited MLO because of the coincidence of school vacation and a\n5\nP 14\n12/23-12/30\n18.60\n1.193\n0.4896\n7.3\n6\nsubstantial quantity of snow down to the elevation of the observatory. We\nH 13\n12/2-12/9\n18.08\n1.163\n0.4800\n1.6\n7\nH 14\ndetermined the isotopic composition of samples from such periods plus several\n12/9-12/16\n17.65\n1.138\n0.4731\n4.9\n8\nH 16\n12/23-12/30\nfrom periods showing lead concentrations more typical of the whole collection\n17.77\n1.144\n0.4755\n22.5\n9\nC 1\n11/19-11/20\n18.29\nperiod of 9 months, when the chance of site contamination was extremely\n1.176\n0.4832\n104.0\n*P-Samples collected in remote area of MLO site, - 5 m above ground.\nunlikely. Thus the data in Table 1 should be reasonably representative of the\nH-Samples collected from HASL surface air network filter - 1 m above ground near\ntotal range of lead isotopic composition of aerosols at MLO during late 1970,\ncenter of MLO site.\nincluding the few periods of significant local contamination as well as those in\nC-Sample collected near the Cloud Physics Observatory at the edge of Hilo.\nwhich background aerosol lead is dominant. The samples have a large range\nThe precision of the measurement of the isotopic ratio of 206 Pb to 207 Pb was generally\n(Fig. 1) in isotopic composition, the least radiogenic values being found for the\nbetter than + 0.2%\nfew periods of highest lead concentration in which local contamination was\ndominant. It appears that local lead contamination has an isotopic composi-\ntion quite different from that of background aerosol lead usually present at\nthat aerosols collected above the trade wind inversion are comparable to\nMLO. The distribution of values is linear, falling along a trend close to that\n\"background\" aerosols found in the upper troposphere over continents and\nfound by Chow (1971) for lead ores and urban aerosol leads.\nrepresent material injected into the atmosphere far from the local Hawaiian en-\nEach aerosol sample that we analyzed generally represents 1 to 2 weeks of\nvironment (Simpson, 1972b). These aerosols could thus represent predomi-\ncollection with a high volume pump (1-2 m³ min). Samples indicated with the\nnantly continental material derived from a mixture of distant source regions,\ncode letter H in Table 1 were from the Health and Safety Laboratory (HASL)\ndespite their collection in an area surrounded by ocean for thousands of\nsurface air monitoring site at MLO, which was located ~ 1 m above the ground\nkilometers. Such a source is consistent with that suggested much earlier by\nnear the middle of the MLO site during the period of our sample collection. We\nJunge (1957) on the basis of the ammonia and sulfate composition of aerosols\nanalyzed quarter sections of each of the circular (20-cm diameter) polystyrene\nabove the trade wind inversion in Hawaii.\nfilters used in the HASL network, each used for 1 week of continuous sampling.\nSome of the aerosol samples from Hawaii that were analyzed for Na, K,\nThe total volume of air sampled by each of the HASL filters was 7-8 X 103 m³.\nCa, Mg, Fe, Mn, Zn, Cu, Ni, and Pb by atomic absorption spectrometry were\nSamples indicated with the code letter P in Table 1 were collected from a more\n96","20\nremote corner of the observatory site, relative to the observatory parking lot,\nusing 20 X 25 cm polystyrene filters from the same stock supply as that of the\nHASL filters. Aerosol samples indicated with an odd number following the code\nletter P were composites of 1-2 weeks of \"day\" filtering (0600-2100). Even-\nnumber samples were 1- to 2-week composites of \"night\" filtering (2100-0600).\nAll of these letter samples were collected ~5 m above the ground surface, and\nA\neach represents a bulk composite of 10-25 X 103 m³ of air. One of the\n19\nsamples (C1) in Table 1 was collected over ~24 hours near the edge of Hilo,\nthe largest town on the island of Hawaii.\nP\nFilter samples were ashed at 425 °C in a covered beaker for - 6 hours, and\nthe residue was dissolved in 5 ml of HCI and 2 ml of HNO (high purity) and\ndiluted to a total volume of 25 ml. The analytical scheme followed to this\npoint is essentially that of Hoffman et al. (1972) and was originally intended\nhere to be used only for flame atomic absorption analysis of a number of\n18\nelements on the same sample. All of the sample preparation to this stage was\ndone in a laboratory in Hilo (Cloud Physics Observatory), and filter leading\nand handling was done at MLO. Blanks measured by atomic absorption spec-\ntrometry of unexposed filters (5 samples), as well as the back half of exposed\nfilters (9 samples), averaged 15%-20% of the total sample values for aerosol\nconcentrations of 1-2 ng/m3 and appeared to result primarily from the filter\nmaterial rather than from filter handling procedures or reagents. We were not\nable to obtain a very good estimate of the lead isotopic composition of the\n17\noverall blank.\n1.0\n1.1\n1.2\nThe chemical procedures described below were carried out in a \"clean lab\"\n06Pb/207P\nat Lamont-Doherty Geological Observatory by using reagents purified to\nreduce the possibility of reagent Pb contamination. An aliquot containing ~ 1\nFigure 1. Aerosol lead determined in\nug of Pb was withdrawn (by pipette) from the sample bottle, placed in a 30-\nthis study (Table 1) for the island of\nor 50-ml Teflon-brand resin-coated beaker, and evaporated to a single drop.\nHawaii during 1970. All of the data\nThe drop was then diluted to 150 ml with triple distilled water, all of which\npoints except one are for samples\nwas poured into a 250-ml Teflon-coated beaker (final pH - 3). One milliliter\nfrom MLO. Average values for Pacific\nof a purified copper (0.5 mg/ml) solution was added to the sample, and a\n(P) and Atlantic (A) ocean pelagic\nTeflon-covered magnetic stirring rod put into the beaker. Two 50-ml platinum\nsediments are also included (Chow,\nwire electrodes were suspended in the sample, forming an electrolytic cell\n1958; Chow and Patterson, 1962).\nacross which a 1.85-V potential was impressed, with continuous stirring.\n97","Under these conditions, PbO plates out on the anode, and Cu (and other\n20\nmetals) on the cathode. The electroplating was generally allowed to proceed\novernight, although 6 to 8 hours were usually sufficient. The PbO was then\nstripped from the platinum wire by immersing the wire in 1 ml of a 99:1 2%\nHNO3: H2O solution. This solution was evaporated to a large drop, and then\ntwo drops of a 0.375 molar H3PO4 solution and two drops of a concentrated\nsilica gel solution were added, and the solution was reevaporated to a single\nA\ndrop. The lead isotope measurements were made on a 30-cm radius of curva-\n19\nture, 60° sector field, solid source mass spectrometer. A single-filament\nP\nrhenium ribbon (30 mm wide) source was used. The sample drop was pipetted\nonto the filament and dried. The mass spectrometric analysis was performed\nX\nby resistance heating at a filament temperature of 1250 + 100 C. Pb ion\nX\ncurrents were on the order of 10-11 A and were measured with a vibrating reed\nelectrometer connected to an expanded scale recorder. The precision of the\nL.A.\nisotopic ratio measurement was generally better than 0.2%.\n18\nS.F.\nFigure 2. Pollutant lead reported by\nChow (1971) for urban aerosols and\n17\nsoils from a number of cities\n1.2\n1.0\n1.1\nthroughout the world. SF is San Fran-\n206Ph/207Pb\ncisco; LA is Los Angeles; P and A are\npelagic sediments from the Pacific\nand Atlantic oceans (Chow, 1958;\nChow and Patterson, 1962). The lead\nisotopic composition of coastal sea-\nwater off southern California, which\nis believed to be contaminated with\nsewage and other sources of pollutant\nlead, is indicated with x's (Patterson\net al., 1976). Data for the average of\ntwo samples of Hudson River water\n(Catanzaro, unpublished data) are\nindicated with a +.\n98","DISCUSSION\nsediment lead as might be assumed on the basis of the urban aerosol data in\nFig. 2 (Chow, 1971). Recent data (1972-1974) for lead from coastal seawater\nThe range of observed data in Fig. 1 is large, indicating at least two\noff Los Angeles and San Diego, California (Patterson et al., 1976), indicate\nsources of aerosol lead with significantly different isotopic compositions. In\nisotopic compositions considerably more radiogenic than those for aerosol lead\ngeneral, the samples from periods of local site contamination at MLO have less\nfrom Los Angeles (Chow, 1971). Representative isotopic compositions for\nradiogenic lead and have an isotopic composition quite similar to that found\ncoastal sea-water samples believed to be substantially contaminated with\nfor aerosols from Los Angeles (L.A.) and San Francisco (S.F.) by Chow (1971)\nsewage lead and other pollutant lead are shown in Fig. 2 (Patterson et al.,\n(see Fig. 2). Since gasoline is supplied to Hawaii from California, this\n1976). We have also recently analyzed the isotopic composition (Fig. 2) of lead\nobservation appears relatively easy to explain.\nfrom two samples of water from the Hudson River. The isotopic compositions\nThe presence of an end-member with more radiogenic lead isotopic\nof lead in water samples from both of these environments, which can be\ncomposition is very interesting. In general, samples with less likelihood of\nexpected to be substantially affected by pollutant lead, are similar to the\nlocal site contamination are found toward the upper end of the data in Fig. 1.\nisotopic composition of Pacific and Atlantic pelagic sediment lead (Chow,\nThe lead in these samples cannot be supplied as a contaminant of lead from\n1958; Chow and Patterson, 1962). The relationship of the isotopic composition\nthe local Hawaiian lava dust because the concentration of lead in the lava is\nof lead from water samples collected near large urban areas during the last\ntoo low, by more than 2 orders of magnitude, to supply the lead collected in\nseveral years to the published aerosol lead isotopic compositions from urban\nthese aerosols. Aerosol mass calculated from the residual ash of exposed filters\nareas during the late 1960s and early 1970s is not simple to interpret, but the\ncollected 5 m above the ground was 0.25 /m3. In addition, the isotopic\nobserved differences do make conclusions about the source of aerosol lead in\ncomposition of lead from Hawaiian lavas is less radiogenic than that of nearly\nbackground aerosols in Hawaii considerably more difficult to reach.\nall the samples near the upper end of the curve. Samples discussed here were\nFinally, it was recently brought to our attention by Clair Patterson that\nnot collected during periods of local contamination of the MLO atmosphere by\nthe isotopic composition of industrial lead in the United States underwent a\nemissions from the active volcano Kilauea.\nsubstantial change during the period 1964-1974. The Pb/207P ratio\nThe isotopic composition of pelagic sediments from the North Pacific (P)\nincreased from about 1.14 to about 1.20 as a consequence of more use of lead\nand North Atlantic (A) oceans (Chow, 1958; Chow and Patterson, 1962) is\nfrom southern Missouri (Chow et al., 1973; Chow et al., 1975). The data that\nindicated on Figs. 1 and 2. These data are probably the best indications avail-\nwe used here to characterize urban aerosol lead from California were collected\nable of the lead isotopic composition of aerosol lead prior to man's industrial\nduring the mid- to late-1960s and thus probably are not a good indicator of\ncontribution of lead. The trend of isotopic composition for aerosol lead at\nthe isotopic composition of urban aerosol lead in the United States during the\nMLO has one end-member very similar to that of Pacific pelagic sediments.\nperiod of late 1970, when our samples at MLO were collected. If automobile\nFrom published data (Chow, 1971) the weighted average isotopic composition\nexhaust leads averaged over the Northern Hemisphere now have a 206Pb/207Pb\nof automobile exhaust lead appears to be considerably less radiogenic than\nratio of approximately 1.20 (as appears not to have been the case 10 years\nthat from Pacific pelagic sediments. Thus if the aerosol lead composition at\nago), the isotopic composition of this pollutant lead does not differ ap-\nMLO were dominated by automobile exhaust lead from distant urban sources,\npreciably from pelagic sediment lead, making it much more difficult to resolve\nthe expected isotopic composition would be considerably less radiogenic than\nthe source of remote area aerosol lead on the basis of the lead isotopic\nthat of pelagic sediment lead. There are indications, however, that the isotopic\ncomposition.\ncomposition of pollutant lead may not differ as much from Pacific pelagic\n99","The data we report for aerosol lead from MLO are consistent with the\ninterpretation that a significant portion of background aerosol lead in remote\nareas today has an isotopic composition similar to that which probably existed\nbefore man's large contribution of aerosol lead. It is thus possible that back-\nground aerosol lead today in the central Pacific may be largely natural. This\ninterpretation conflicts substantially with the large increase of lead in polar\nsnows from recent years compared with snow from preindustrial periods\n(Murozumi et al., 1969) and with aerosol lead flux estimates (Patterson, 1965),\nboth of which suggest that the anthropogenic source dominates background\naerosols by more than an order of magnitude. The large change in the isotopic\ncomposition of industrial lead in the United States over the last decade does\nintroduce a substantial complication. The similarity of the isotopic\ncomposition of industrial lead in the United States during the early 1970s to\nthat in aerosol lead at MLO during late 1970 is also consistent with the\ndominance of pollutant lead in aerosols of remote areas in the Northern\nHemisphere. The data from MLO are sufficiently interesting, however, to\nsuggest that lead isotopic measurements of aerosols from remote areas could,\nif sufficient data were collected, provide key information for understanding the\ncurrent global cycle of aerosol lead.\nACKNOWLEDGMENTS\nThis study was initiated under an Environmental Science Services Admin-\nistration (now the National Oceanic and Atmospheric Administration -\nNOAA) postdoctoral research program sponsored by the National Research\nCouncil. The scientific advisor, Dr. Helmut Weickmann, of NOAA in Boulder,\nColorado, provided generous support and guidance. R. Pueschel, J. Pereira,\nB. Mendonca, H. Ellis, and J. Chin, of MLO helped throughout sample\ncollection. J. Bowen, J. Naughton, and V. Lewis, of the University of Rhode\nIsland, and A. Lazrus and G. Gendrud, of NCAR, provided advice on atomic\nabsorption analytical techniques. H. Volchok, of the Health and Safety Lab-\noratory of the Energy Research and Development Administration, supplied the\nfilter materials and extensive logistic support. C. Patterson provided valuable\ncriticism during preparation of the paper. K. Antlitz assisted in preparation of\nthe manuscript.\n100","REFERENCES\nJunge, C. E., 1968: Airborne dust at Barbados\nAult, W. U., R. G. Senechal, and W. E\nSimpson, H. J., 1972b: Aerosol cations at\nChow, T. J., K. W. Bruland, K. Bertine, A.\nand its relation to global tropospheric\nErlebach, 1970: Isotopic composition as a\nSoutar, M. Koide, and E. D. Goldberg, 1973:\nMauna Loa Observatory. J. Geophys. Res.,\naerosols. Geochim. Cosmochim. Acta,\nnatural tracer of lead in the environment.\nLead pollution: Records in Southern Cali-\n77:5266-5277.\n32:1219-1222.\nEnviron. Sci. Technol. 4:305.\nfornia coastal sediments. Science,\nVolchok, H. L., 1973: The tropospheric\nKeeling, C. D., R. B. Bacastow, A. E. Bain-\nBrown, J. S., 1962: Ore leads and isotopes.\nbaseline concentration for lead. USAEC\n181:551-552.\nbridge, C. A. Ekdahl, Jr., P. R. Guenther,\nEconomic Geology, 57:673-720.\nChow, T. J., C. B. Snyder, and J. L. Earl,\nReport HASL-273, 79-87.\nL. S. Waterman, and J. F. S. Chin, 1976: At-\nChow, T. J., 1958: Lead isotopes in sea water\n1975: Proceedings of FAO/IAEA\nmospheric carbon dioxide variations at\nand marine sediments. J. Marine Res.,\nSymposium, Vienna, Austria.\nMauna Loa Observatory, Hawaii. Tellus,\nCobb, W. E., and H. J. Wells, 1970: The elec-\n17:120-127.\nChow, T. J., 1970: Lead accumulation in road-\n28:538-551.\ntrical conductivity of ocean air and its corre-\nMurozumi, M., T. J. Chow, and C. Patterson,\nside soil and grass. Nature, 225:295-296.\nlation to global atmospheric pollution. J.\n1969: Chemical concentrations of pollutant\nChow, T. J., 1971: Isotopic identification of in-\nAtmos. Science, 27:814.\nlead aerosols, terrestrial and sea salts in\ndustrial pollutant lead. Proceedings of the\nDuce, R. A., G. L. Hoffman, and W. H.\nGreenland and Antarctic snow strata.\nSecond International Clean Air Congress,\nZoller, 1975: Atmospheric trace metals at\nGeochim. Cosmochim. Acta, 33:1247.\nAcademic, New York.\nremote northern and southern hemisphere\nPales, J. C., and C. D. Keeling, 1965: The con-\nChow, T. J., and C. C. Patterson, 1962: The\nsites: Pollution or natural. Science,\ncentration of atmospheric carbon dioxide in\noccurrence and significance of lead isotopes\n187:59-61.\nHawaii. J. Geophys. Res., 70:6053.\nin pelagic sediments. Geochim. Cosmochim.\nEgorov, V. V., N. Zhigalovskaya, and S. G.\nPatterson, C. C., 1965: Contaminated and\nMalakhov, 1970: Microelement content of\nActa., 26: 263-308.\nnatural lead environments of man. Arch.\nChow, T. J., and M. S. Johnstone, 1965: Lead\nsurface air above the continent and the\nEnviron. Health, 11:344-360.\nisotopes in gasoline and aerosols of Los\nocean. J. Geophys. Res., 75:3650.\nPatterson, C. C., G. Tilton, and M. Inghram,\nAngeles Basin, California. Science,\nHoffman, G. L., R. A. Duce, and E. J.\n1955: Age of the Earth. Science, 121:69-75.\n147:502-503.\nHoffman, 1972: Trace metals in the\nPatterson, C. C., D. Settle, and B. Glover,\nChow, T. J., J. L. Earl, and C. F. Bennett,\nHawaiian marine atmosphere. J. Geophys.\n1976: Analysis of lead in polluted coastal\n1969: Lead aerosols in marine atmosphere.\nRes., 77:5322.\nseawater. Marine Chem. 4:305-319\nEnviron. Sci. Technol.. 3:737-740.\nJunge, C. E., 1957: Chemical analysis of\nRabinowitz, M. B., and G. W. Wetherill, 1972:\nChow, T. J., and J. L. Earl, and C. B. Snyder,\naerosol particles and of gas traces on the\nIdentifying sources of lead contamination by\n1972: Lead aerosol baseline: Concentration at\nIsland of Hawaii. Tellus, 9:528-537.\nstable isotope techniques. Environ. Sci.\nWhite Mountain and Laguna Mountain,\nJunge, C. E., 1963: Air chemistry and radioac-\nTechnol. 6:705-709.\nCalifornia. Science, 178:401-402.\ntivity. Academic, New York, 328 pp.\nSimpson, H. J., 1972a: Aerosol and\nChow, T. J., J. L. Earl, 1972: Lead isotopes in\nprecipitation chemistry at Mauna Loa Obser-\nNorth American coals. Science, 176:510-511.\nvatory. NOAA Tech. Report, ERL\n101","Ralph Stair making solar radiation\nmeasurements with a scanning photo-\nmeter, August 29, 1966.\nSolar radiation instruments at MLO,\nIn the early '70s, Mauna Loa Obser-\nvatory had one of the first automated\nleft to right: normal incidence pyr-\nheliometer, infrared hygrometer,\nsolar radiation data recording sys-\nglobal pyranometer.\nitems.\nBernard Mendonca measuring ozone\nWalter Komhyr and Al Shibata modi-\nBernard Mendonca and Rudy\nwith a Dobson spectrophotometer,\nPueschel check out solar radiation\nfying a total-ozone Dobson spectro-\nphotometer at MLO in February\nMarch 1972.\ninstruments, 1970.\n1972.\n102","On June 28, 1966 Dick Hansen of\nHAO and Lothar Ruhnke of MLO\n(and others) installed a radiometer to\nmonitor solar activity and cloud\ncover at the summit of Mauna Loa.\nSolar measurements\n103","","relatively rapid buildup, whereas in the South Pacific area the available\nmeasurements indicate the aerosol content to be stable. The longest record of\nreliable measurements on which an assessment of aerosol loading of the atmos-\nphere can be made is that taken at Mauna Loa Observatory (MLO), Hawaii.\nThe measurements of direct solar radiation taken at the observatory over a 19-\nyear period show a strong but temporary buildup of particulate matter\nfollowing the eruption of Mt. Agung in 1963, but no trend, either upward or\ndownward, is evident in the data available at the present time (Machta, 1972;\nZENITH SKYLIGHT CHARACTERISTICS IN THE\nEllis, 1977, private communication). If there is a trend, its magnitude is less\nSUNRISE PERIOD AT MAUNA LOA\nthan the probable error of the measurements.\nIn view of the importance of the aerosol problem it is desirable to get as\nmany independent measurements of aerosol loading as possible. A sensitive\nKinsell L. Coulson\nmethod of detecting the presence of aerosols and obtaining a rough estimate of\nUniversity of California, Davis, California\ntheir properties is through their light scattering effects. The scattering of\nsunlight by aerosol particles is different from that by the oxygen, nitrogen,\nand other gaseous constituents of the atmosphere. The total amount of energy\nscattered, its angular distribution, and its state of polarization depend strongly\non the numbers; physical properties of light from the sunlit sky are sensitive to\nthe existence of aerosols in the atmosphere, and measurements of these quan-\ntities should provide a simple and useful method of characterizing atmospheric\nINTRODUCTION\nturbidity on a long-term basis.\nThere is well known controversy about the effects of accumulations of\nMLO is an advantageous site for making skylight measurements for\nsmoke, dust, smog, and other types of particulate matter in the atmosphere. If\naerosol monitoring. Since the observatory is located in an oceanic setting far\nwe are putting aerosols into the atmosphere faster than the atmosphere is\nfrom major landmasses, the natural aerosol loading is minimal, and pollution\ncleansing itself, it is not illogical to expect some long-term climatic effects to\nfrom local sources is generally confined to the atmospheric layer below the\nresult from the buildup of the worldwide burden of aerosols. Unfortunately, it\ntemperature inversion associated with the trade wind regime. In fact, Mauna\nis not clear just yet whether the overall effect of such an increase of aerosols\nLoa is thought to have some of the cleanest atmospheric conditions to be\nwould be a cooling of the planet, as some scientists are predicting, or a\nfound in the world. In addition, a large complement of mutually supporting\nwarming of the planet. The most acceptable available analysis of the problem\nmeasurements, such as incident solar radiation at different wavelengths,\nindicates that the trend could go either way, depending on the characteristics\nparticle counts at the station, and lidar probes of the atmosphere, are made on\nand location of the aerosols themselves.\na routine basis at the observatory.\nHowever, whether there is a progressive buildup of atmospheric aerosols\nThe present investigation is designed to measure in detail the intensity and\non a worldwide basis is also a question for which the answer is not completely\nstate of polarization of the light from the sunlit sky and to determine, if\nknown. The evidence on the point is conflicting. In some locations, such as\npossible, the best method of using skylight data for characterizing atmospheric\nover the North Atlantic and in the Caucasus, there is strong evidence of a\nturbidity at the observatory.\n105","100\nTHEORETICAL EXPECTATIONS\nSince the atmosphere at Mauna Loa is very clear, the Rayleigh approxi-\n=0\nmation, in which all particles are assumed to be small in comparison with the\n80\nA o\nwavelength of the radiation, should be a good theoretical model with which to\nA = 0.25\nevaluate the measurements. In preparation for this work an extensive set of\nFigure 1. Polarization as a function of\nA = 0.50\ncomputations of the intensity I(T; 0, Oo, 0-00; A), degree of polarization P(T; 0,\n60\n-A\n0.25\n=\nzenith angle in the plane of the sun's\nA = 0.75\nOo, 0-00; A), and angle X(T; 0, Oo, 0-00; A) of the plane of polarization with\nvertical for a plane parallel Rayleigh\nrespect to the vertical direction was carried out. Here T = T(A, h) is normal\n0.50\natmosphere at a wavelength of 0.365\n40\noptical thickness of the atmosphere at the wavelengths 1 and altitude h of the\num and the altitude of MLO for four\n0.75\nobservatory; (A, o) and (Ao, 00) are the zenith angle and azimuth of the\ndifferent values of surface albedo and\nobserved direction and the sun, respectively; and A is albedo of the surface.\nfor solar zenith distances of 78.5°\n20\nThe values of the different parameters used in the computations are listed in\n(solid curves) and 23.1° (dashed\nTable 1. The various combinations of the parameters resulted in a total of\ncurves).\nSun\n159,936 individual values from the computations.\no\nIn the interest of brevity, emphasis here is confined to theoretical results\nfor two wavelengths. The first, l = 0.365 um, is characterized by a relatively\n80\n40\n0\n40\n80\nlarge optical thickness (T = 0.333) with much multiple scattering, whereas the\nZenith Angle in Degrees\nsecond, 1 = 0.80 um, corresponds to a small optical thickness (T = 0.015) and\na minor significance of multiple scattering effects in the atmosphere. Because\nof this change of scattering efficiency with wavelength for the Rayleigh atmos-\nphere, it is to be expected that radiation is morensitive to aerosol effects at the\nlonger wavelengths than it is at the shorter wavelengths. As will be seen\nbelow, this expectation is borne out by the measurements.\nThe degree of polarization as a function of zenith angle in the plane of the\n100\nsun's vertical (plane defined by the sun and the zenith direction) for 1 = 0.365\nA o\nFigure 2. Polarization as a function of\num is shown for four values of surface albedo (A = 0, 0.25, 0.50, and 0.75)\nA 0.25\nzenith angle in the plane of the sun's\nand two zenith angles of the sun (Oo = 23.1 and 78.5) in Fig. 1. Multiple\n80\nA 0.50\nvertical for a plane parallel Rayleigh\nscattering is responsible for the decrease of the maximum polarization from\nA 0.75\natmosphere at a wavelength of 0.80\n100% to about 80% and for the existence of the negative branches of the\n60\num and the altitude of MLO for four\ncurves, with the attendant neutral points (points of zero polarization). Of\n0.25\ndifferent values of surface albedo and\nparticular interest are the depolarizing effects of surface reflection, the reflected\nfor solar zenith distances of 78.5°\nradiation being assumed to be unpolarized and isotopic in the outward\n40\n0.50\n(solid curves) and 23.1 (dashed\nhemisphere. In all cases, increasing surface reflection decreases the magnitude\n0.75\ncurves).\nof the polarization maximum, the decrease being most pronounced for the\n20\nSun\nSun\no\n80\n40\no\n40\n80\nZenith Angle in Degrees\n106","higher sun elevation. This increase of surface influence with increasing sun\nthe extreme case when the sun is at the horizon, it is necessary to extrapolate\nelevation is due to the efficient transmission of the direct solar beam through\nthe data beyond their region of theoretical validity. Although this is admit-\nthe atmosphere, and the consequent strong illumination of the surface, at the\ntedly a questionable procedure, it does give some insight as to the probable\nhigher sun elevations. At the same time the efficient transmission causes less\neffects of surface reflection in the measurements.\nmultiple scattering at the higher sun elevations, a feature which explains the\nFrom curves such as those shown in Figs. 1 and 2 it is possible to deter-\nincrease of the polarization maximum with increasing sun elevation for the\nmine the degree of polarization at the maximum as a function of the various\ncase of no surface reflection (A = 0).\nparameters of the computations. In Fig. 3 the magnitude of the polarization\nPolarization curves for 1 = 0.80 um are shown in Fig. 2. The very high\nmaximum, Pmax is shown as a function of sun elevation for 1 = 0.365 um\nmaxima here attest to the paucity of multiple scattering at this long wave-\n(dashed curves) and 1 = 0.80 um (solid curves) and for four different values\nlength, whereas the decrease of polarization with increasing surface reflection\nof surface albedo. The extrapolation from a sun elevation a = 5.7° to a =\n0°\nis evidence of the efficiency with which the direct solar beam is transmitted\nis shown by the dotted segments of the curves. It is evident that at l =\ndownward through the atmosphere to illuminate, and be reflected from, the\n0.365um, surface reflection has a significant effect on Pmax max at all sun\nelevations, including 1 = 0°, if the extrapolation is valid in any sense.\nsurface.\nSince the present investigation is emphasizing the polarization and\nIntuitive reasoning would indicate that even when there is no appreciable\nintensity of skylight in the zenith direction for the case when the sun is very\nradiation incident on the surface from the direct solar beam, strong scattering\nnear the horizon, either above or below, the change of polarization maximum\nby the optically thick atmosphere would cause a significant amount of diffuse\nwith sun elevation is of particular interest. In fact, the theoretical computa-\nillumination of the surface, and reflection of the diffuse radiation would\ntions, being based on the assumption that the curvature of the atmospheric\nproduce the depolarizing effects shown in the diagram. At 1 = 0.80 um, on\nlayers is negligible (a plane parallel atmosphere), do not apply at very low sun\nthe other hand, there is little diffuse radiation to illuminate the surface, SO at\nelevations. To obtain a qualitative estimate of the features to be expected in\nvery low sun elevations, for which the energy in the direct beam is minimal,\nthe effect of surface reflection must also be minimal. In the diagram, the\ncurves for A # 0 have been extrapolated to closely approximate that for A =\n0 for a = 0°. This is undoubtedly an exaggeration, but the trend of the curves\n100\nA=0\nat a > 5.7° indicates that surface effects at the longer wavelengths must be\nvery small at the time the sun is at or below the horizon. This fact will be\nused in the interpretation of the skylight measurements.\n90\nFigure 3. Maximum polarization\n80\nA o\nversus solar zenith angle for a plane\nparallel Rayleigh atmosphere at\nwavelengths of 0.365 um (dashed\ncurves) and 0.80 um (solid curves)\n70\nand the altitude of MLO for four\ndifferent values of surface albedo.\nThe dotted extensions of the curves\n60\nA = 0,25\nrepresent an extrapolation to a sun\nelevation of 0°.\n0.75\nA 0.50\n50\no\n20\n40\n60\n80\nSun Elevation in Degrees\n107","horizon. Obviously, this would correspond to the zenith polarization\nRESULTS OF MEASUREMENTS\nmaximum occurring at the time the sun is on or slightly above the horizon.\nMode of Operation\nThe fact that the zenith polarization reaches its maximum when the sun is 2.5°\nThe normal mode of operation for measurements in the period from first\nto 3.5° below the horizon, or at scattering angles of 92.5° to 93.5°, must be\nlight in the morning until the sun was a few degrees above the horizon was to\nbrought about by the geometry of the sun-atmosphere combination in the\norient the polarimeter in the direction of the zenith and step continuously\nsunrise period.\nthrough the four sets of color filters. A measurement of intensity, degree of\nFrom intuitive reasoning one can obtain an idea as to the possible cause\npolarization, and orientation of the plane of polarization in each of the eight\nof the apparently anomalous behavior of the short-wavelength curves of Fig.\nwavelength ranges was obtained at approximately 50-s intervals throughout\n4. As the first direct rays of the sun illuminate the high atmosphere at solar\nthe period. The uncertainty in each measurement depends somewhat on wave-\ndepression angles of 6° to 4°, primary scattering by the tenuous upper atmos-\nlength, the sensitivity of the photomultiplier tubes decreasing strongly near the\nphere is predominant, and the angular dependence of the polarization is about\nshortwave and longwave limits. Thus measurements at 0.32 and 0.90 um have\nas would be expected. However, as the solar depression angle decreases, the\na probable uncertainty of perhaps 0.5% during the dawn period, whereas\ndirect rays descend lower into the atmosphere, and the rapidly increasing at-\nthose at the other wavelengths have a probable uncertainty of approximately\nmospheric density causes a rapid transition from primary scattering to multiple\n+0.2%. In the post-sunrise period the uncertainties were somewhat less than\nscattering as the predominant feature of the radiation field. As the transition\nthese values.\ndevelops, the polarizing effects of multiple scattering overcompensate the\nBecause of the large number of measurements made in this series, it is\nnormal increase of polarization to be expected as the scattering angle\nnecessary for purposes of illustration to select cases that are typical of different\napproaches 90°, thereby causing the indicated decrease of P. As the sun\ntypes of atmospheric conditions. Cases of low atmospheric turbidity will be\ncontinues to rise to the horizon, the atmospheric pathlength of the incident\ndiscussed first, and the data from these cases will be used for comparison with\nrays through the spherical atmosphere tends to decrease, with the result that\nthe measurements made under more turbid or anomalous atmospheric\nthe depolarizing effects of multiple scattering combine with the regular increase\nconditions.\nof polarization to create the plateau shown by the curves. After sunrise the\noptical pathlength of incident radiation continues to decrease; multiple\nLow-Turbidity Cases\nscattering is alleviated sufficiently to counteract the angular decrease of polar-\nization as the sun moves to less than 90° from the zenith to cause the plateau\nAll of the measurements made under conditions of low turbidity\nof the curves to persist after sunrise for 1 = 0.365 and 0.40 um. At 1 = 0.32\ngive results similar to those shown by the curves of Figs. 4 and 5,\num the optical pathlength is SO large that a small change with sun elevation\nin which degree of polarization P in the zenith direction is plotted as a\ndoes not appreciably change the relative importance of multiple scattering, and\nfunction of sun elevation. In the range of wavelengths from 0.32 to 0.50 um,\nshown in Fig. 4, the polarization increases monotonically with increasing\nthe normal change of P with changing scattering angle proceeds about as\nwavelength, and the behavior of P appears to be anomalous in some respects.\nexpected.\nIt will be remembered that for Rayleigh scattering the maximum polarization is\nThere are obviously some features of the curves of Fig. 4 that these simple\nat 90°, or slightly less, from the direction of the sun, and many measurements,\nconcepts do not explain. First, the change of position of the maximum with\nboth in the past and in the present series, show this expectation to be closely\nwavelength appears to be the reverse of what would be expected. Since\napproximated in the actual atmosphere when the sun is well above the\nmultiple scattering increases with decreasing wavelength, one would expect the\n108","depolarizing effects to show up at larger solar depression angles for short\n90\nwavelengths than for longer wavelengths, but the data do not bear out this\nFigure 4. Polarization in the zenith\nexpectation. Astronomical refraction would tend to cause a shift in the right\n0.50 pm\ndirection as a function of sun\ndirection to explain the observations, but the magnitude is probably too small.\n80\nelevation for four different wave-\nSecond, the curve for l = 0.50 um appears not to follow the simple explana-\nlengths in the shortwave part of the\n0.40 um\ntion proposed above, but in this case the effects of non-Rayleigh-type particles\nsolar spectrum. The data were taken\nmodify the behavior from what it would be for a pure Rayleigh atmosphere.\nin very clear atmospheric conditions\nThis may indeed be the case for the shorter wavelengths as well, but, as\n0.365\n70\num\non February 19, 1977.\ndiscussed below, such aerosol effects are particularly evident as wavelengths 1\n0.32\num\n0.60 um.\nThe curves of zenith polarization versus sun elevation for the longer\n60\nwavelength ranges are shown in Fig. 5. Here the scale of the ordinate is\nstaggered so as to separate the curves. Two features of the diagram indicate\n85\nthat primary scattering dominates the radiation field at these wavelengths.\n0.90\n80\nFirst, the maximum is located at approximately 90° from the sun, the shift of\n50\n-5\n0\n5\n10\nits position, which was ascribed to multiple scattering effects at shorter wave-\n75\nSun Elevation in Degrees\nlengths, being absent from these curves. Second, the magnitude of the\n85\nmaximum (about 85.5%) is essentially independent of wavelength in this\n0.80\n80\nrange, a fact which indicates that multiple scattering effects must be very\nsmall. The high value of the maximum shows that atmospheric turbidity must\n75\n90\nhave been very small, since scattering at these wavelengths is strongly\nFigure 5. Polarization in the zenith\ndominated by the non-Rayleigh particles in the atmosphere.\n85\ndirection as a function of sun\nAnother interesting feature of Fig. 5 is the behavior of the zenith polariza-\n0.70\nelevation for four different wave-\n80\ntion at solar depression angles of 3° to 6°. The pattern shown by these curves\nlengths in the longwave region of the\nis typical of that on almost all days for which the early morning measurements\n75\nsolar spectrum. The data were taken\nhave been made at Mauna Loa, although, of course, the details vary. The\nin very clear atmospheric conditions\n70\nmagnitude of the secondary maximum, the magnitude of the minimum, and\n90\non February 19, 1977.\nthe position of the minimum all show day-to-day variations that seem to be\n85\nassociated with high-level, either stratospheric or upper tropospheric, layers of\n0.60\num\naerosols. If this is proved, by auxiliary measurements such as lidar probes, to\n80\nbe the case, then the measurement of zenith polarization should be a simple\n75\nand valuable means of characterizing and monitoring turbidity of the upper\n-5\n0\n5\natmosphere.\nSun Elevation in Degrees\n109","Under most conditions the minimum of a curve occurs at a solar\nfeature is well defined in the average. Similarly, the secondary maximum is\ndepression angle of approximately 4°, and the secondary maximum occurs\nreasonably pronounced for the average. If the existence of these features is to\nwhen the sun is 5.0° to 5.2° below the horizon. The solar ray tangent to the\nbe ascribed to an aerosol layer, as was postulated above, then the top of the\nearth's surface is at heights of about 25 km and 16 km for solar depression\nlayer would be at an average altitude of about 26 km and reach its maximum\nangles of 5° and 4°, respectively. The polarization curves would thus indicate\ndensity at 16 to 17 km MSL in average low-turbidity atmospheric conditions at\nthat an aerosol layer with a top at about 25 km and a maximum density at 16\nMauna Loa. The primary maximum for the average at 0.80 um occurs at the\nkm exists over Mauna Loa a large part of the time. In isolated cases the\ntime when the sun is just on the horizon, as it would for predominantly\naerosol becomes relatively diffuse, its altitude varies somewhat from day to\nprimary scattering in a clear atmosphere.\nday, and occasionally it largely, but not completely, disappears. Some data to\nThe average low-turbidity curve for 1 = 0.365 um shows the primary\nsubstantiate these statements are given below.\npolarization maximum in the zenith direction to occur at a solar depression\nangle of about 2.9°, with a very minor minimum at about sunrise. This curve\nis much more difficult to interpret than that for the longer wavelength, and\nAverage Low-Turbidity Curves\nthere is a greater scatter in the data from which the average was determined.\nIn order to have some standard curves for comparison purposes, 10 days\nAs was pointed out previously, the short-wavelength radiation is strongly\non which the maximum polarization at 0.80 um was at least 84.8% were\ninfluenced by the depolarizing effects of multiple scattering, and cloud\nselected as low-turbidity days, and average low-turbidity curves for l = 0.365\nreflection plays a very important role in determining the shortwave radiation\nand 0.80 um were computed from the ten sets of data. The curves resulting\nfield, even for very low sun elevations.\nfrom this process are shown in Fig. 6. Although slight variations of the\nposition of the presunrise minimum at 1 = 0.80 um tend to broaden the\nZenith Polarization on Days with Moderate to\noutline of the average curve from its configuration on individual days, the\nHigh Turbidity\n90\nWe are now in a position to characterize conditions of moderate to high\natmospheric turbidity by comparison of zenith polarization measured in such\ncases with the average low-turbidity curves of Fig. 6. On February 17 there\nwas evidence of dust, perhaps of volcanic origin, in the troposphere or lower\n0.80 um\n80\nstratosphere over the region of Hawaii. A layered structure was visible above\nthe horizon in all quadrants. The yellow color was more pronounced than\nusual in the early morning sky, and a definite aureole was visible in the region\nof the sun. Unfortunately, the lidar system at the observatory was inoperable\n70\nat the time, but measurements by the solar corona research project at the site\nindicated the skylight intensity at one solar diameter from the solar disc to be\n0.365 um\n3 to 4 times the average value during the morning of February 17. Such an\nFigure 6. Polarization at the zenith as\nintensity increase is associated with increased forward scattering of sunlight by\n60\n-8\n-6\n-4\n-2\no\n2\n4\n6\n8\na function of sun elevation for\nparticulate matter in the atmosphere.\nSun Elevation in Degrees\naverage low-turbidity conditions at\nMauna Loa. The curves were\nobtained by averaging the\nmeasurements taken on 10 days\nselected for atmospheric clarity\n(February 10, 15, 19, 21, 23, 26;\nMarch 4, 9, 10, 11, 1977).\n110","The degree of polarization, as measured at the zenith during the sunrise\nminimum being shifted to a solar depression angle of 3.5° and the magnitude\nperiod at a wavelength of 1 = 0.80 um on February 17, is compared with that\nof the maximum being decreased by about 3% from its low-turbidity value.\nfor average low-turbidity conditions in Fig. 7a. The measurements indicate the\nAn even more anomalous case occurred on March 1, as shown by the\nhigh-altitude conditions to be near average, but the main aerosol layer at high\ncurve of Fig. 7c. On that morning, however, there were wisps of cirrus clouds\naltitudes appeared to be thinner than average. The flattening of the curve in\nin various parts of the sky and considerable upslope motion carrying air from\nthe range of solar depression angles of -4.2° to 0° indicates that the turbidity\nlower levels to the mountain top. Care was taken to make sure that there were\ncomponent has extended well below its average sunrise altitude in the\nno visible clouds near the zenith as the measurements were made, but there\natmosphere. The shift of the maximum from its average sunrise position to a\nmay well have been some depolarization effects introduced by reflection from\nsolar depression of about 1.3°, together with a decrease of polarization of over\nhigher-level clouds or by effects of particles that had not yet formed visible\n5% after sunrise, would likewise indicate increased depolarization effects of\nclouds. The curve shows obvious optical instability which may have been\naerosol scattering well down into the troposphere.\nproduced by either actual or nascent clouds. Whether or not such particulates\nA second case of obvious atmospheric turbidity occurred on March 25.\nconstitute atmospheric turbidity depends on the definition of the term, but\nCloudiness was very minor throughout the morning and forenoon of that day,\nthey certainly do have a major effect on the field of skylight polarization.\nbut there was a considerable amount of dust or heavy haze in the atmosphere.\nThe period from March 31 to April 7 was characterized by strong vertical\nDust layers extending to several degrees above the horizon were visible, the\nmotions in the atmosphere as a weather system moved over the area. Large\nsunrise sky was unusually yellow, and a definite solar aureole persisted\ncumulus clouds with occasional thunderstorms were prevalent, and upslope\nthroughout the forenoon. The zenith polarization measurements for March 25\nmotion on Mauna Loa brought clouds over the observatory by midforenoon\nare compared with the average low-turbidity curve in Fig. 7b. The behavior of\non most days. The data points for this period are much more scattered than\nthe polarization field on that day is obviously anomalous, the position of the\nusual, an indication of optical instability in the atmosphere. In Fig. 8, polari-\nzation curves derived from zenith measurements at 1 = 0.80 um are shown for\nthree days of the period (April 2, 3, and 4). As in the previous diagram, the\n85\ndashed curves represent the low-turbidity average for the same wavelength.\nThe curves of Fig. 8a show the degree of polarization to be up to 4.5% lower\n80\nAverage\nthan the low-turbidity average. By April 3 (Fig. 8b), however, the minimum\nwas very pronounced and was shifted to occur at a solar depression angle of\n75\nabout 2.9°. These features would seem to indicate an increase of tropospheric\n2/17/77\na\nturbidity between April 2 and 3, although the polarization maximum was\n70\n85\nslightly higher on April 3 than on the previous day. The most anomalous case\nFigure 7. Polarization at the zenith as\nof the period occurred on April 4, as shown in Fig. 8c. Here the polarization\na function of sun elevation as\n80\nAverage\nmaximum was about 8.5% lower than the average for low-turbidity\nmeasured at a wavelength of 0.80 um\nconditions, and the early morning minimum bears little resemblance to that of\non days with moderate to high\n3/25/77\n75\nmore normal atmospheric conditions. Measurements not shown here indicate\natmospheric turbidity (solid curves)\nb\nthat by April 7 the atmosphere had returned to a more nearly normal\ncompared with that for average low-\n70\n85\nturbidity situation.\nturbidity conditions (dashed curves).\n80\nAverage\n3/1/77\n75\nC\n70\n-8\n-6\n-4\n-2\no\n2\n4\n6\n8\nSun Elevation in Degrees\n111","DISCUSSION\n85\nThe possibility of using measurements of skylight polarization in the\nzenith direction during sunrise or sunset periods to characterize turbidity\n80\nAverage\nconditions in the upper atmosphere has long been recognized. Linke (1951)\n4/2/77\nmade a long series of observations of zenith skylight measurements by both\n75\nvisual and photoelectric means. Fesenkov (1961) has studied the problem by\na\nmeans of some theoretical approximations applied to twilight phenomena but\n70\n85\nconcluded that zenith twilight phenomena are unsuitable for studying the\nAverage\noptical properties of the atmosphere at high altitudes. Dietze (1963), on the\n80\nother hand, has used zenith polarization measurements to study the penetra-\n4/3/77\ntion of cosmic dust into the upper atmosphere and attributed a decrease of\n75\npolarization, observed to occur at solar depression angles of 6° to 12°, to\nb\nconcentrations of cosmic dust in the 80- to 100-km region of the atmosphere.\n70\n85\nA discussion of the entire twilight problem has been given by Rozenberg\nFigure 8. Polarization at the zenith as\n(1966).\na function of sun elevation as\n80\nAverage\nAlthough the concept of using zenith skylight measurements for\nmeasured at a wavelength of 0.80 um\ncharacterizing atmospheric turbidity is not new, there are certain aspects and\non days with moderate to high\n75\nresults of the present series of measurements that do appear to be unique to\natmospheric turbidity (solid curves)\n4/4/77\nC\nthe series. First, the measurements made here extend to wavelength ranges\ncompared with that for average low-\n70\n-8\n-6\n-4\n-2\no\n2\n4\n6\n8\nlonger than those of previous investigations. For instance, the measurements\nturbidity conditions (dashed curves).\nSun Elevation in Degrees\nand theory on which Fesenkov based the opinion of zenith twilight phenomena\nbeing unsuitable for studying atmospheric properties were for radiation at\nvisible wavelengths, in which case multiple scattering of the primary twilight\ncomponent introduces extreme complexity into the problem. In the infrared\nregion, however, the multiple scattered component is essentially negligible in\nTable 1. Values of Parameters Used in Computations for the Rayleigh\ncomparison with that arising from primary scattering, and SO the problem is\nModel\ncomparatively much simpler at the longer wavelengths. A second feature of\nthe present series is the high altitude of the observatory at which the measure-\nA\n1\n0\nOo\n0-00\nT\nments were made. The two aspects of high altitude and long wavelengths\n(degrees)\n(degrees)\n(um)\n(degrees)\ncombine to make a very small normal optical thickness of the atmosphere\n0.32\n0.625\n90.0\n68.9\n43.9\n0\n0\n0\nabove the site. For instance, at a wavelength of 0.80 um and an altitude of\n0.365\n0.333\n88.3\n67.0\n41.4\n23.1\n0.30\n0.25\n3460 m the normal optical thickness of the Rayleigh atmosphere is only 0.015.\n0.40\n0.246\n86.6\n65.2\n38.7\n36.9\n60\n0.50\nThe optical thickness discussed by Fesenkov (1961) was over 13 times this\n0.50\n0.098\n84.8\n63.3\n35.9\n53.1\n90\n0.75\n0.60\n0.047\n83.1\n61.3\n32.9\n66.4\n120\n0.70\n0.025\n81.4\n59.3\n29.5\n78.5\n150\n0.80\n0.015\n79.6\n57.3\n25.8\n84.3\n180\n0.90\n0.009\n77.9\n55.2\n21.6\n76.1\n53.1\n16.3\n74.3\n50.9\n8.1\n72.5\n48.7\n70.7\n46.4\n112","(0.20), and so, of course, multiple scattering effects are much more important\nREFERENCES\nin his results.\nDietze, G., 1963: Polarization of skylight\nAlthough further work, both experimental and theoretical, needs to be\nduring twilight as an indicator of cosmic\ndone to arrive at a more complete physical understanding of the field of\ndust. In Proceedings of the All-Union Scien-\nskylight polarization when the sun is very near the horizon, the measurements\ntific Meteorological Congress, vol. 6, Series\nActinometry and Atmospheric Optics,\nmade here indicate that a useful index of atmospheric turbidity can be\nHydromet. Pub. House, Leningrad,\nobtained very simply by routine monitoring of the degree of polarization in\npp. 92-101 (in Russian).\nthe direction of the zenith during the early morning period. Although similar\nFesenkov, V. G., 1961: The twilight method in\nmeasurements near sunset should yield similar results, local turbulence brought\nthe study of the optical properties of the\nabout by daytime heating would tend to introduce pollution from local\natmosphere. In Scattering and Polarization of\nLight in the Atmosphere, translated from\nsources and thereby render the polarization field during the period near sunset\nRussian, Israel Program for Scientific\nless representative of overall atmospheric turbidity than that which occurs in\nTranslations, Jerusalem, 1965, pp. 196-214.\nthe sunrise period.\nLinke, F., 1951: Meteorologisches\nTaschenbuch, Giest und Portig, Leipzig.\nMachta, L., 1972: Mauna Loa and global\ntrends in air quality, Bull. Amer. Meteorol.\nSoc., 53:402-420.\nRozenberg, G. C., 1966: Twilight: A Study in\nACKNOWLEDGMENTS\nAtmospheric Optics. Translated from\nThe support for this investigation, which was provided jointly by the\nRussian by A. E. Stubbs, Plenum Press, N.Y.\nResident Research Associateship Program of the National Academy of Science,\nthe University of California, and MLO, is gratefully acknowledged. Special\nthanks go to the director, Dr. John Miller, and staff of the MLO for their\nexcellent cooperation and support of the skylight measurements, as well as for\ntheir personal kindness and hospitality. Computations of skylight parameters\nfor the Rayleigh atmosphere were performed by Bruce Fitch. The work was\ncarried out while the author was on a 6-month sabbatical leave from the\nUniversity of California, Davis.\n113","","OBSERVATIONS OF THE SOLAR CORONA AT\nMAUNA LOA\nCharles Garcia\nHigh Altitude Observatory, National Center for\nAtmospheric Research\nHilo, Hawaii\nThe sun's corona is best observed at times of total eclipse, when the\nmoon, appearing almost identical to the sun in diameter, blots out the\noverwhelming brilliance of the solar disk (a million times that of the corona),\nand when the earth's atmosphere along the eclipse path is also in shadow and\ninterference from scattered sunlight is much reduced. But however\nbreathtaking and however useful to science they may be, total eclipses are SO\ninfrequent and of such short duration that they cannot really give us an\naccurate picture of the long-term behavior of the corona.\nInvention of the coronagraph in 1930 by the French astronomer Bernard\nLyot made day-to-day observations of the corona possible, and this instrument\nhas been used to document many aspects of coronal activity that earlier were\nunknown or only suspected. Even in the thin clear air of high altitudes,\natmospherically scattered light degrades coronagraph observations; neverthe-\nless, they have been invaluable in portraying the corona's behavior in a\ncontinuous rather than piecemeal fashion.\n115","SOLAR CYCLE CHANGES IN THE CORONA\nBasic changes occur in the appearance of the corona during the 11-year\nsolar sunspot cycle. Near solar maximum, when 100 or more sunspots may be\nvisible at one time, the corona appears to be almost circular, with brilliant\nstreamers radiating from all latitudes. At solar minimum, when few sunspots\ndevelop, equatorial streamers predominate, polar features are much reduced,\nand large rifts often appear near the poles. At this time, total coronal radiance\nis less than half as great as it is during solar maximum.\nSolar cycle changes in the corona were originally discovered by piecing\ntogether data from drawings and written descriptions of total eclipses. The\nadvent of eclipse photography in 1851 introduced a more reliable method of\nrecording eclipses and revealed also the difference in total radiance between\nsolar maximum and minimum. Solar cycle changes have now been\ndocumented photographically and spectroscopically on a continuing basis with\ncoronagraphs at high-altitude sites in France, Switzerland, Germany, Czecho-\nslovakia, the USSR, Japan, and the United States.\nMAUNA LOA OBSERVATORY STUDIES\nRichard and Shirley Hansen, Charles Garcia, and Eric Yasukawa, working\nat the High Altitude Observatory (HAO) site on Mauna Loa, Hawaii, are\nFigure 1. Mauna Loa K-coronameter\nstudying features of the inner corona with two highly specialized coronagraphs,\nobservations superposed on HAO's\nthe K-coronameter and the Coronal Activity Monitor. These instruments,\n1966 eclipse photograph. The lines\ndeveloped by HAO engineers, are especially adapted to map coronal radiance\nare intensity plots and correspond\nphotoelectrically. They measure the polarized K-component of the corona,\nwell with the principal coronal\ni.e., light scattered by free electrons in the sun's atmosphere. Both the K-\nfeatures (streamers, condensations,\ncoronameter and the Coronal Activity Monitor measure this radiation by\nholes, etc.). The K-corona is\nN\ncomposed of white light scattered by\nfree electrons in the sun's atmosphere,\nE\nW\nFigure 2. K-coronameter July monthly averages for 1964 (dot), 1966\nand its radiance is a measure of\n(dash), and 1967 (solid) depict solar cycle changes in corona. Radial\nelectron density in the corona.\ndistance from the edge of the circle is proportional to radiance at 0.125\nsolar radii from the limb of the sun. Note that these are radiance plots\nand not diagrams of the shape of the corona.\n116","making concentric small-aperture scans around the sun at predetermined\nfilamentary, or quiescent, prominence typically divides the regions of opposite\ndistances above the edge, or limb, of the sun while the solar disk is occulted as\npolarity. As the sun's rotation brings active centers near the limb of the sun,\nin other coronagraphs (Fig. 1).\nthe association between active centers and coronal features becomes evident.\nMauna Loa data verify that pronounced changes occur in the distribution\nWhen seen at the edge of the sun, these features are broad helmet-shaped\nof radiance in the corona during the solar cycle. For example, during the\nstructures; at greater heights above the sun's surface they taper to narrow\nascending phase (from 1964 to 1967) of the recent solar cycle the total radiance\nstreamers.\nof the lower corona doubled, and there was a progressive migration of zones\nUltimately, the trailing poleward end of the system appears to break its\nof enhanced activity toward the poles, as shown in Fig. 2. During July of the\nties with the equatorward portion and to line up magnetically with the polar\nsolar minimum year 1964 the general level of radiance was low and day-to-\nfield. A long magnetic tunnel arches over the filamentary prominence that\nday changes were slight. In 1966, zones of increased activity appeared in the\nseparates the trailing field from the polar field, and a helmet streamer extends\nNorthern Hemisphere. The general level of radiance was two to three times as\noutward as part of the polar corona. The equatorward end of the original\ngreat as that during solar minimum. In 1967, radiance had increased in the\nactive region, now magnetically unbalanced, may form a transequatorial arch\nsouthern quadrants also, although a relative void remained over the south\nif a similar region of opposite polarity exists at nearby longitudes in the\npole.\nopposite hemisphere.\nEVOLUTIONARY CHANGES\nSOLAR ROTATION\nFrom the wealth of synoptic observations being made at Mauna Loa, the\nThe sun's rotation causes systematic and recognizable changes in the ap-\nevolution of individual coronal features such as the streamers seen in eclipse\npearance of coronal features as they move into and out of the plane of the\nphotographs can be related to other solar phenomena. The corona appears to\nsky, projecting for several days above the solar limb. Occasionally, very-high-\nincrease in radiance with the emergence of an active center, a localized region\nlatitude features may be followed continuously as they rotate around the pole.\ncontaining sunspots, and other manifestations of strong magnetic fields. Such\nan active center, which develops at a low latitude, gradually disperses, and a\nCORONAL TRANSIENTS\nportion of its magnetic field migrates poleward, trailing behind the equator-\nward portion because of the differential rotation of the sun. For a time,\nFrom 1970 to 1973, more than a dozen major abrupt depletions of the\nmagnetic field lines connect the two regions, but as the trailing field spreads\ninner corona were observed at Mauna Loa. This coronal material, expelled\npoleward and eastward, field lines also join this region to the polar field. A\nfrom the sun, was almost invariably associated with eruptive hydrogen alpha\n117","prominences and correlated in both position and time with outward moving\ntype-1V radio sources as seen with a coronagraph aboard the OSO-7 satellite,\noperated by the Naval Research Laboratory. These coronal transients are\n1\nabrupt and short lived, and occur in localized regions of the corona (Fig. 3).\n1\nSeveral coronal transients observed at Mauna Loa have also been\n2\nrecorded by the radioheliograph at Culgoora, Australia, operated by the Com-\nmonwealth Scientific and Industrial Organization (CSIRO). On March 21, 1970,\nfor instance, the radioheliograph recorded three distinct radio sources, one of\nwhich moved to a distance of six solar radii from the limb of the sun. Mauna\n1\nLoa data suggest that then, as well as during several other transient events, the\nN\nloss of material blown out into space depleted part of the corona temporarily,\n1\nfor radiance was markedly reduced in the immediate area of the eruption.\nE\nW\n22\nRecovery was rapid, however, and by the next observing day the coronal\nstructure had returned to its preflare shape and brightness (Fig. 4).\nS\nHigh\nEruptive\n2\nProminence\nNo.1 August 1970 @ 2002 UT\nNo.2 August 1970 @ 0255 UT\n1\nFigure 3. Example of coronal\ndepletions at 0.75 solar radii from the\nlimb of the sun, obtained from K-\ncoronameter records. Brightness of\nthe corona is represented by radial\ndistance from the circle. During the\ninterval between these measurements\n(about seven hours), a large promi-\nnence erupted from the SE limb with\nattendant depletion of the coronal\nstreamer (cross hatching) and also in\na separate event a flare occurred near\nthe west limb with a somewhat\nsmaller depletion of coronal ray.\n118","NEW INSTRUMENTATION\nThe 11-year cycle of solar activity will reach a maximum in 1979. A large\neffort will be made by HAO to study events occurring on the active sun. HAO\nwill have a solar coronagraph on the Solar Maximum Mission satellite due to\nbe launched into circular orbit in late 1979. To support and complement the\nSMM coronagraph, a new K-coronameter and a prominence monitor are being\nbuilt by HAO in Boulder, Colorado, for use at Mauna Loa, Hawaii.\nTime resolution will be improved by a factor of several hundred over that\nof the telescope now used at Mauna Loa. All scan heights will be measured\nsimultaneously at a number of scanned position angles. Computers will\ncontrol much of the telescope operation and digitize the data on magnetic\ntape. Together these instruments will measure and interpret coronal electron\ndensity and magnetic field structures on a rapid time scale to provide a better\nunderstanding of the dynamics of coronal transients.\nFigure 4. Eruptive prominence seen\nwith the Mauna Loa Ha coronagraph\non October 28, 1972. This eruption\ncaused a depletion in the solar corona\nas coronal material was carried\noutward from the sun during the\neruptive phase of the prominence.\n119","Saul Price and Jack Pales used a teth-\nered balloon for meteorological\nsounding of the atmosphere (1958).\nWeather-releated equipment at MLO\nincludes (left, from top) microbaro-\ngraph, wind-speed and -gust and\ndirection recorder, and operations re-\ncorder, and (right, from top) tem-\nperature, and solar radiation records.\nKulani Mauka, MLO's climatological\nIn the mid-'70s wind was measured at\nstation at 8300-ft elevation.\na site below the observatory by the\n\"Hilo-Kona\" road.\n120","Meteorological measurements\n121","The Complexity of the Wind Patterns at Mauna Loa\nObservatory\nMauna Kea\nJohn M. Miller\nHilo\nMauna Loa Observatory, NOAA\nHilo, Hawaii 96720\nObservatory\nKona\nMauna Loa\nProfile\nINTRODUCTION\nThe purpose of the U.S. baseline stations (Mauna Loa, Samoa, Barrow,\nand South Pole) is to monitor trace materials in the atmosphere that may\ncause climatic change. In order to interpret these data, meteorological param-\neters such as surface measurements, vertical soundings, and air trajectories\nhave been employed (Keeling et al., 1976; Pack et al., 1977; and others). A\nparameter that has been used in interpreting the Mauna Loa Observatory\n(MLO) data is the surface wind direction and speed.\n30 Kilometers\n0\n10\n20\nThe wind regime at MLO consists of two major components-local and\nN\nsynoptic. The local upslope (northerly wind)-downslope (southerly wind)\nsystem at the observatory has frequently been described (Lavoie, 1967;\nNW\nNE\nMendonca, 1969; and others). The daytime upslope flow is not, however, a\nsimple anabatic flow, because the strength of either the low-level trades on the\nE\nW\neast side of the island or the sea breeze on the west side determines whether\n4200\nthe upslope has a westerly or easterly component (Fig. 1).\n3600\nSW\nSE\nThe second wind system, i.e., the large-scale synoptic flow, which reflects\nS\nglobal circulation and usually has an easterly or westerly component, occurs\n2400\nalso at the site. The difficulty of determining which system prevails arises\n1200\nwhen the winds come from the SE., SW., NE., or NW. To distinguish between\nthe local wind systems and the synoptic flow is one aspect of establishing\n0\nProfile\nwhether a given measurement is representative of a global or local value.\nFigure 1. Topography and profile of\nthe island of Hawaii.\n122","MEASUREMENTS\nAs the local winds accentuate the north and south components of the\nFor approximately 20 years, wind measurements at MLO have been made\nwinds, the synoptic flow strengthens the easterly and westerly directions.\nwith a cup anemometer and recorder using a digital system to eight points of\nDuring the winter period the west winds are increased. The easterly\nthe compass. For an evaluation of this record, a 2-year period (February 1,\ncomponent is strongest during the summer period but may appear throughout\n1975 to January 31, 1977) was chosen, and the hourly averaged wind direction\nthe year depending on the position of the intertropical convergence zone.\nand speed were summarized.\nDuring the transition period between summer and winter seasons, the best\nupslope-downslope pattern exists. A summary of the local flow patterns and\nthe regional synoptic patterns is given in Table 1.\nWIND DIRECTION\nThe average hourly wind direction over the 2-year period is plotted in\nWIND SPEED\nFig. 2. The most obvious feature of this figure is the regularity of the upslope\ndirection, or northerly component, of the wind. This anabatic wind, which\nOver the 20 years of wind measurements, moderate to strong winds at\nMLO (speeds > 10 mph (4.5 m/s)) have been observed to indicate a domi-\nprevails only during the daytime period (0800-1900 LST), may, however, be\ninfluenced by the northeasterly trades progressing up the Saddle area from the\nnance of the synoptic pattern over the local wind system. Even during these\nHilo side or by the westerly sea breeze from the west Kona coast of the island.\nstrong wind periods, the upslope has enough influence to swing the westerly\nThese two regimes fight a daily battle to impose on the upslope winds an\nflow from south to north during a given diurnal period. A similar effect is\nseen in the shift from south to north during the strong easterly flow, although\neasterly or a westerly component. In addition, a synoptic flow may sometimes\nbe imposed on local wind systems and thus complicate the picture.\nthis effect is less pronounced (Fig. 2).\nThe southerly, or downslope, component is present during the evening\nA typical case of how the wind speed may help to differentiate between\nsynoptic and local winds is shown in Fig. 3. The wind direction data for\nand early morning. However, even though winds with a southerly component\nmay prevail throughout the day during some months, i.e., September 1975,\nJanuary 1976 are separated into two classes; all cases above 10 mph show the\n1976, and May 1977 (Fig. 1), the synoptic pattern interferes with the katabatic\npredominance of the southeasterly flow throughout the whole day and indicate\nwinds in many cases. The diversity of the wind patterns at the same time of\na strong synoptic influence on the wind pattern. This synoptic situation occurs\nthe year can be seen by comparing the January patterns.\ntypically in the winter with a west-east polar front north of the islands but a\nThe large-scale synoptic systems leave a significant imprint on the MLO\nsoutheasterly flow over Hawaii.\nsurface winds. Meteorologists have divided the synoptic patterns that affect\nthe Hawaiian Islands into two simple categories-summer and winter\nSUMMARY\n(Worthley, 1967). During the winter period (October-April) the intertropical\nThe Mauna Loa wind system is a complex interaction of both the local\nconvergence zone moves farther south allowing the more westerly flow\nflow systems and the regional synoptic patterns. To determine whether the\nassociated with extra-tropical disturbances to prevail. The summer period\nMLO trace material measurements represent global or local values, an analysis\n(June-August) is characterized by the semipermanent high-pressure situation to\nof the wind speed and direction is a useful method.\nthe north and east of the Hawaiian Islands. May and September are\nconsidered periods of transition.\n123","April\nJune\nJuly\nFebruary\nMay\nMarch\n1975\n1975\n1975\n1975\n1975\n1975\nSW\nW\nNW\nN\nNE\nE\nSE\nS\n1976\n1976\n1976\n1976\n1976\n1976\nSW\nW\nNW\nN\nNE\nE\nSE\nS\n01\n12\n24 01\n12\n24\n01\n12\n24\n01\n12\n24\n12\n24\n01\n12\n24\n01\nLocal Time\nLocal Time\nFigure 2. Average hourly wind\ndirections at Mauna Loa. Averages\nare compared for two years.\n124","August\nSeptember\nOctober\nNovember\nDecember\nJanuary\n1975\n1975\n1975\n1975\n1975\n1976\nSW\nW\nNW\nN\nNE\nE\nSE\nS\n1976\n1976\n1976\n1976\n1976\n1977\nSW\nW\nNW\nN\nNE\nE\nSE\nS\n01\n12\n24\n01\n12\n24\n01\n12\n24\n01\n12\n24\n01\n12\n12\n24\n01\n24\nLocal Time\nLocal Time\n125","January 1976\nJanuary 1976\nWind Speed <10 mph (4.5 m/s)\nWind Speed >10 mph (4.5 m/s)\nSW\nTable 1. Interpretation of Surface Wind Directions at\nNW\nMauna Loa Observatory\nTime\nPrevailing\nN\nDirection\n0800-1900 LST\n1900-0800 LST\nSeason\nSouthwest\nOccurs only with\nCombination of\nWinter\nstrong synoptic\ndownslope and\nNE\ncomponent\nsynoptic\ncomponent\nWest\nInsignificant,\nInsignificant\nNone\nE\nrecorded during\nwind shifts\nNorthwest\nMajor occurrence\nInsignificant\nNone\nduring this period,\nSE\neffect of local sea-\nbreeze plus some\nsupport from synoptic\nNorth\nLocal upslope flow,\nInsignificant\nNone\nno synoptic\ncomponent\n01\n12\n24\n01\n12\n24\nLocal Time\nLocal Time\nNortheast\nLocal upslope\nInsignificant\nSlightly more\nflow plus a synop-\noften in summer\ntic component\nFigure 3. Comparison of hourly\nEast\nMainly synoptic\nInsignificant\nSlightly more\naverage winds per wind direction\nplus some local\noften in summer\nas separated into winds below and\neffects\nabove 10 mph.\nSoutheast\nDuring strong\nLocal effect\nAll year long\nsynoptic flow from\nfortified by\nwith slightly\neast\nsynoptic\nmore cases\nin summer\nSouth\nInsignificant\nLocal downslope\nAll year\nflow with little\nsynoptic influence\n126","REFERENCES\nKeeling, C. D., R. B. Bacastow, A. E. Bain-\nbridge, C. A. Ekdahl, P. R. Guenther, L. S.\nWaterman, and J. F. S. Chin, 1976: Atmos-\npheric carbon dioxide variations at Mauna\nLoa Observatory, Hawaii. Tellus, 28:538-\n551.\nLavoie, R. L., 1967: Air motion over the wind-\nward coast of the island of Hawaii. Tellus,\n19:354-358.\nMendonca, B. G., 1969: Local wind circulation\non the slopes of Mauna Loa. J. of Appl.\nMeteorol. 8:533-541.\nPack, D. H., J. E. Lovelock, G. Cotton, and C.\nCarthorp, 1977: Halocarbon behavior from\na long time series. Atmos. Environ. 11:329-\n344.\nWorthley, L. G., 1967: Weather phenomena in\nHawaii; Part one, synoptic climatology of\nHawaii. HIG 67-09, Hawaii Institute of\nGeophysics, University of Hawaii, pp. 1-40.\n127","Snow on Mauna Loa\n128","CLIMATE AND WATER BALANCE ON THE ISLAND\nOahu\nMolokai\nOF HAWAII\nMaui\nPacific Ocean\nHawaii\n20°N\nJames O. Juvik, D. C. Singleton, and G. G. Clarke\nDepartment of Geography, University of Hawaii\nHawaiian Islands\nHilo, Hawaii\n156°W\n1000\n1500\n500\n2500\n250\n2500\nINTRODUCTION\nThe island of Hawaii, with a surface area of only 10,455 km², exhibits a\nspectacular range of climatic diversity comparable with that found on large\n250\ncontinents. Three major factors contribute to this climatic diversity:\nMauna\n1. Topographic relief. The volcanic mountains of Mauna Kea and Mauna\n500\nKea\nHilo\nLoa reach summit elevations of 4,205 m and 4,168 m, respectively. The\naltitudinal range provides for a diversity of temperatures, and the mountains\n500\nthemselves are barriers that induce orographic precipitation.\n2500\n2. Large-scale synoptic wind field. The strong and persistent northeast trade\n1000\n3000m\nwinds interact with the island topography to produce distinctive windward\nMauna\nand leeward climates. The associated upper-level trade wind inversion exerts a\n1500\nLoa\nparticularly strong control on mountain precipitation gradients.\n3. Local circulation. Differential heating and cooling of the land, water,\nmountain, and lowland areas on Hawaii give rise to localized wind regimes\nwhich add to the island's climatic diversity.\n1500\n1000\nFigure 1. Distribution contours of\n1500\nmean annual rainfall (mm), super-\n1000\nimposed on topographic map of the\n0\n10 20km\n500\n500\nisland of Hawaii. (Redrawn from\nTaliaferro, 1959; State of Hawaii,\n1970.)","A HUMID TROPICAL CLIMATES\n(All monthly mean temp 18°C)>\nTROPICAL CONTINUOUSLY WET\nAs\nAf\n(Each month mean rainfall 6 cm)\nTROPICAL WINTER-DRY\nAw\nHAWAII\n(At least one winter month\nmean rainfall <6 cm)\nTROPICAL SUMMER-DRY\nAs\nAm\n(At least one summer month\nWaimea\nmean rainfall < 6 cm)\nTROPICAL MONSOON\nAm\n(High annual rainfall with\nBWh\nshort dry season)\nBSh\nB ARID AND SEMI-ARID CLIMATES\n(Annual evaporation exceeds\nAs\nprecipitation)\nBSh HOT SEMI-DESERT\nHilo\nCsb\n(Moderate rainfall, mean\nannual temp. > 18°C)\nAf\nBWh HOT DESERT\nCfb\n(Low rainfall, mean annual\nCfb\ntemp. > 18°C)\nKailua\nC TEMPERATE CLIMATES\n(Mean temp. of coldest month between\nObservatory\n- 3° and + 18°C)\nCfb CONTINUOUSLY WET WARM\nVolcano\nTEMPERATE\nET\n(At least four months mean\ntemp. > 10°C < 22°C)\nCsb SUMMER-DRY WARM\nCsc\nTEMPERATE\n(Temp. as above, 70% of\nAw\nannual rainfall in six winter\nAs\nmonths, dryest summer\nmonth rainfall < 3 cm)\nN\nPahala\nCsc SUMMER-DRY COOL\nTEMPERATE\n(Precipitation as above, less\nthan four months mean\nAf\ntemp. > 10°))\nD COLD CONTINENTAL CLIMATES\n0\n10\n20km\n(None in Hawaii)\nE ICE CLIMATES\nFigure 2. Distribution of Köppen cli-\nPERIGLACIAL CLIMATE\nET\nmate types on the island of Hawaii.\n(Warmest mean monthly\ntemp. between 0° and 10°C)\n130","KOPPEN CLIMATIC ZONES\nsummer-dry climates are not common anywhere in the world, since for most\ntropical locations rainfall is at a maximum in the summer, the result of\nIntegrating the altitudinal temperature gradients with the annual,\nincreased convective instability in the high-sun period. Outside of Hawaii the\nseasonal, and spatially variable rainfall regimes results in a diverse\nAs climate type occurs only in southern Madras (India) and adjacent northern\ncombination of climatic environments. The Köppen climate classification uses\nSri Lanka.\nmonthly temperature and precipitation characteristics in a descriptive system\nThe leeward or Kona coast of Hawaii contains the only extensive area of\nthat distinguishes broad regional and global climatic zones. The system has\nsummer maximum rainfall in the Hawaiian archipelago (Aw, winter-dry\nbeen often criticized for its empirical approach and lack of emphasis on\nclimate). Isolated from the prevailing trade wind flow by intervening high\n\"dynamic processes\" (e.g., Carter and Mather, 1966); however, as a \"first\nmountains, the Kona coast's dominant circulation pattern is formed by a\napproximation\" the Köppen classification offers useful insights into regional\nlocalized land-sea breeze regime. Increased land surface temperatures in\nclimatic patterns.\nsummer strengthen the daily sea breeze regime and increase convective\nFour broad Köppen climatic zones are distinguished on the island of\ninstability, leading to a high frequency of afternoon thundershowers. The\nHawaii. They are organized primarily as concentric altitudinal bands on the\nvertical structure necessary for thundershower development is further assured\nmountain slopes. Fig. 2 illustrates the spatial distribution of these Köppen\nby the high mountains, which exclude the trade wind aloft and limit the\nclimatic types on the island. The map was constructed on the basis of\npotential for strong vertical wind shear. The presence of a strong shearing\ntemperature (absolute or extrapolated) and precipitation data from 55 island\nforce would otherwise tend to destabilize these leeward convectional cells.\nstations. Discussions of the zones follow.\nAlthough there is generally a summer rainfall maximum throughout Kona, the\nAw climate gives way to Af at elevations above 400 m, where, by virtue of\ngeneral orographic position, there is adequate precipitation in all months.\nHumid Tropical Zone (A climates).\nCharacterized by warm temperatures throughout the year and relatively\nhigh annual rainfall, humid tropical climates occupy the lower slopes of the\nArid and Semi-Arid Zones (B climates).\nisland from sea level to about 450 m (slightly higher in warmer areas of\nleeward Kona). This tropical zone may be further differentiated on the basis of\nA classic rain shadow desert exists on the leeward side of the Kohala\nrainfall seasonality. Large-scale synoptic disturbances in winter (mid-latitude\nmountains. Smaller and lower (maximum elevation 1,670 m) than Mauna Kea\ncyclonic storms) produce substantial rainfall that is to some extent independent\nand Mauna Loa, the Kohala mountains are incapable of blocking out trade\nof slope aspect or elevation, and as a result most locations on the island\nwind flow to leeward: Having become depleted of moisture during windward\nexhibit an absolute winter maximum in rainfall. However, windward areas of\nascent, the trades warm adiabatically to leeward, promoting a hot, arid zone.\nHawaii also receive substantial orographic rainfall throughout the year, with\nWith only 190 mm of annual rainfall, Kawaihae on the leeward Kohala coast\nthe result that there is no distinct dry season (Af climate). Lowland areas on\nis the driest location in the Hawaiian archipelago. The Köppen system\nthe island that are transitional in location between windward and leeward\ndistinguishes two climatic subtypes, the true desert (BWh climate) and the\nreceive less orographic rainfall (since they are not oriented normal to trade\nsemidesert (BSh climate) on the basis of relative aridity. In leeward Kohala the\nwind flow) and exhibit a distinctive summer dry season (As climate). Humid\ntrue desert gives way to semi-desert at higher elevations.\n131","Temperate Zone (C climates)\nthe treeless arctic tundra. The upper tree line of Mauna Kea (3,000 m)\ncorresponds fairly closely to the C/E boundary mapped in Fig. 2. On Mauna\nAverage air temperature in Hawaii decreases with altitude at the rate of\nLoa the tree line is much lower for edaphic reasons (recent lava).\nabout 0.55°C/100 m (Price, 1973). When the criteria of the Köppen\nclassification are used, at elevations above 400-500 m on the mountain slopes,\ntropical climates grade to temperate as a result of decreasing average\nWATER BALANCE\ntemperatures. As a result of the moderating influence of altitude, almost two-\nThe preceding discussion of Köppen climatic zones on Hawaii provides a\nthirds of the \"tropical\" island of Hawaii possesses a temperate upland climate.\ngeneral overview of the dramatic range in regional climatic diversity found on\nThe majority of this zone is characterized by warm summers and adequate\nthe island. However, this descriptive approach says little about the direct\nprecipitation in all months (Cfb climate). Except for the absence of a stronger\nlinkage of climate to physical and biological processes at the earth/atmos-\nseason variability the upland Hawaii climates are analogous to those of similar\nKöppen designation in Pacific coastal areas of North America. Ascending\npheric interface.\nAn integration of seasonal moisture supply (precipitation) with the evap-\norographic clouds compressed between the rising mountain slope and an\noration and transpiration demands of the environment (determined primarily\nupper-air temperature inversion produce frequent ground level mountain fog,\nby solar energy inputs) provides an index of moisture surplus or deficit. Such\nan important moisture source for upland vegetation (Juvik and Perreira, 1974).\nindices can illuminate direct process/response relationships between climate\nAt still higher elevations on Mauna Kea and Mauna Loa (above 2,000 m) there\nis a tendency toward summer drought. The increased strength and frequency\nand the terrestrial ecosystem.\nIn an initial survey of water balance climatology on the island of Hawaii,\nof the trade wind inversion in summer (modal elevation 1,800 m) inhibits the\nMueller-Dombois (1966) constructed a series of climate diagrams 21 stations\nvertical penetration of orographic clouds and precipitation to the higher\n(see Fig. 3). This type of diagram, popularized by Walter and Lieth (1960),\nslopes. This summer-dry zone (Csb climate) also occurs at a lower elevation in\nportrays seasonal curves of mean monthly temperature and precipitation.\nleeward Kohala, Mauna Kea, and Mauna Loa, where summer orographic\nAccording to Muller-Dombois (1976) an index of precipitation efficiency is\nprecipitation is largely absent. Above 2,500 m on both Mauna Loa and Mauna\nbuilt into the diagrams by making one degree of temperature (Celsius) equal to\nKea the summer-dry regime changes from warm to cool (Csc climate).\ntwo millimeters of precipitation in the scaling of the two ordinates. This is\nbased on the assumption that monthly potential evapotranspiration (in milli-\nmeters) is roughly equal to twice the mean monthly temperature (Gaussen,\nAlpine (periglacial) Zone (E climates).\n1954). Wherever the precipitation curve drops below the temperature curve, a\nAbove 3,200-m level on Mauna Kea and Mauna Loa all months have a\ndrought season is indicated. Thus the graph is transformed into a water\nmean temperature below 10°C, and the climates are classified as periglacial\nbalance diagram with the temperature curve interpreted as an index of\n(ET). Nighttime freezing is common throughout the year. Although it exhibits\npotential evapotranspiration.\na winter maximum, annual rainfall is very low (200-400 mm) and variable.\nA serious problem inherent in this graphing technique is the tendency to\nAbove the 3,500-m level, winter snowfall accounts for a substantial portion of\napproximate evapotranspiration with a simple linear function of air tempera-\nthe seasonal precipitation. Köppen used the 10 °C (warmest month) boundary\nture (i.e., the 2:1 ratio). Chang (1959, 1968) has reviewed the problems of\nto separate the C and E climates on the basis that trees will not normally grow\ntemperature-based estimations of potential evapotranspiration and points out\nwhere mean temperatures fall below this level. Hence E climates characterized\nThe Temperate (C) and Polar (E) climates as originally proposed by Köppen were not\napplied in high-altitude tropical environments, which, because of their orographic\ncomplexity, lack of meteorological data, and absence of strong seasonality, were simply\ndesignated as highland climates (H). In more recent global and regional climatic maps of\nthe tropics, highland C and E climatic areas are frequently portrayed in order to show\napproximate altitudinal analogs for these broad latitudinal climatic zones.\n132","that solar radiation rather than temperature is the primary forcing function in\n°C\nthe evaporative process. Temperature-based estimates of evaporation implicitly\n°C\nmm\nmm\nassume a strong correlation between temperature and solar radiation. In\nHolualoa\nHalemaumau\n500\n500\nHawaii, as a result of advection and the buffering effect of the surrounding\nBeach\nmarine environment, there is generally poor correlation between solar\nradiation and temperature. In Hilo, for example, the range in mean monthly\n50\n100\n50\n100\nair temperature is only 1.4°C between June (24.2°C) and December (22.8°C).\nBy contrast, the receipt of solar radiation in June (563 Ly; see solar radiation\ndata for 1965 from Löf et al, 1966) is more than twice that in December (263\nLy).\nIt is obvious from the above comparison that temperature-based estimates\nof evapotranspiration cannot be expected to portray realistically the seasonal\nfluctuations implied in the radiation data. However, in the absence of a dense\nnetwork of solar radiation monitoring stations on the island, upon which more\n0\n0\n0\n0\nsophisticated spatial modeling of evapotranspiration might be based, it is\nJan\nDec\nJan\nDec\nnecessary to revert to some form of temperature-derived estimation in a \"first\napproximation\" of water balance regimes.\nThornthwaite (1948) has developed perhaps the most widely adopted\n°C\n°C\nmethod of estimating potential evapotranspiration. His empirical formula is\nmm\nmm\nbased essentially on air temperature:\nHawaii Nat.\nMauna Loa\n500\n500\nPark Hq.\nObservatory\nE = 1.6 (10T/I)a.\n(1)\n50\n100\n50\n100\nPotential evapotranspiration E is computed from mean monthly temperature T\nand an empiric \"heat index\" I, which itself is an exponential function of tem-\nperature; a is a constant.\nTo obtain mean monthly evapotranspiration, the values derived from\neq. (1) are corrected for mean daylength and number of days in the month. The\nThornthwaite equation, although subject to the general limitations of all tem-\n0\n0\n0\n0\nperature-based methods, might be expected to give better results than the\nJan\nDec\nJan\nDec\nWalter method in Hawaii, since potential evapotranspiration is expressed as an\nexponential rather than linear function of temperature.\nFigure 3. \"Walter\" climate diagrams\nfor four Hawaii island stations (From\nMueller-Dombois, 1966).\n133","In Fig. 4, monthly values of potential evapotranspiration derived by both\nthe Walter and the Thornthwaite methods have been plotted along with class\n\"A\" pan-evaporation data for Hilo and Pahala. It is evident that Walter\nFigure 4. Monthly estimated potential\ngrossly underestimates pan evaporation (here assumed to be equal to potential\nevapotranspiration and measured pan\nevapotranspiration) and also fails to detect the seasonal rhythm apparent in\nevaporation for Hilo and Pahala\nthe pan data. Thornthwaite also underestimates pan evaporation but does so\ndata.\nin a fairly consistent manner and achieves a strong covariation with the pan\ndata in seasonal rhythm. This suggests that the Thornthwaite method might be\n200\nuseful in Hawaii if a correction factor could be derived to compensate for the\nHilo\nPahala\nconsistent underestimation exhibited in Fig. 4.\nClass \"A\" Pan\nIn Fig. 5, monthly values of the Thornthwaite potential\n150\nevapotranspiration estimate are plotted against pan evaporation for Hilo and\nThornthwaite\nPahala. With a regression coefficient of 0.844, approximately 71% of the\nobserved variation in pan evaporation can be explained by variation in the\n100\nThornthwaite estimate. (The regression coefficient is significant at 0.01 level.)\nPotential evapotranspiration Y can thus be reasonably predicted from the\nWalter\n50\nThornthwaite values X by the linear regression equation\n0\nY = 42.8 + 1.016 (X)\n(2)\nJ\nS\no\nN\nD\nM\nA\nM\nJ\nJ\nA\nJ\nF\nM\nA\nM\nJ\nJ\nA\nS\no\nN\nF\nBefore eq. (2) can be applied as a general (island-wide) correction factor for\nthe Thornthwaite potential evapotranspiration estimate, it must be verified\nFigure 5. Relationship between\nthat the relationship established in Fig. 5 (for two lowland locations) is equally\nmonthly class \"A\" pan evaporation\nvalid for mid- and high-altitude areas of the island\nand estimated monthly potential\nThere are no class \"A\" pan evaporation data for inland mountain areas of\nevapotranspiration (Thornthwaite) for\nHawaii with which the lowland-derived correction factor might be compared.\nHilo and Pahala data.\n175\nHowever, Juvik and Clarke (1976) have accumulated limited experimental data\nr = 0.844\non mountain evaporation gradients in Hawaii Volcanoes National Park on the\nY = 42.5 + 1.016(X)\neast flank of Mauna Loa. These data were obtained by using four constant-\nlevel pan evaporimeters (Fig. 6) situated along an altitudinal transect between\n150\nsea level and 2,000 m.\nIn Fig. 7, measured mean daily evaporative rates (averages from 133 days\nof simultaneous readings taken from September 1974 through May 1975) are\n125\nHilo\nPahala\n100\n50\n75\n100\n125\nPotential Evapotranspiration (mm)\n134","plotted against elevation. There is a clear linear decrease in evaporation over\nthe altitudinal range surveyed (approximately 0.72 mm/day/1,000 m). Fig. 7\nalso shows the corrected (eq. 2) Thornthwaite potential evapotranspiration\nvalues (mean of 9 months, September to May) derive I from temperature-\nrecording stations that occur near the evaporimeter transect. There is good\nagreement between the Thornthwaite and the evaporimeter values (differences\nb.\nrange from 1% to 12%), largely because air temperature also decreases\nlinearly with elevation. On the basis of the close agreement in Fig. 7, the\ncorrected Thornthwaite estimate was considered acceptable to use for all areas\nof the island in the derivation of monthly and annual potential evapotran-\nspiration from temperature data.\nCorrected Thornthwaite estimates of monthly and annual potential evapo-\ntranspiration were computed from standard tables (Thornthwaite and Mather,\n1975) and eq. (2), for 30 stations on the island of Hawaii. The\nevapotranspiration data were then integrated with monthly precipitation\nvalues to produce seasonal water balance diagrams (Fig. 8). Because some of\nthe water surplus received in the wet season is stored as soil moisture for\nutilization during dry periods, the computation of seasonal water balance must\nincorporate a parameter describing the moisture storage capacity of the soil.\nOn the geologically youthful island of Hawaii, soils are not generally well\ndeveloped except for limited areas where ash deposits occur in high-rainfall\nFigure 6. Constant-level pan\nzones. Recent lava flows exhibiting little or no soil development cover\nevaporimeter. a) field installation\nsubstantial portions of the island. The average depth to bedrock for 75\nwith inner tube reservoir; b)\ndifferent Hawaii island soil types and subtypes has been calculated at 0.89 m\nevaporimeter detail. Note foam\nwith only moderate variation (Sato et al., 1973), lending quantitative credence\ninsulation around evaporation pan.\nto this stated geological youthfulness of the island. Soil moisture storage\ncapacity has not been well studied for most Hawaiian soil types. A value of\n125 mm/m is the average moisture capacity for ten different soil types for\n5\nFigure 7. Relationship between\nwhich data are available. If this is assumed to be representative, then the\nmeasured and predicted evapo-\nPredicted\naverage soil moisture storage capacity for all Hawaii island locations would be\n(Thornthwaite & Equation 2)\ntranspiration along an altitudinal\n111.2 mm (i.e., 0.89 X 125). In the construction of water balance diagrams\n4\ntransect in Hawaii Volcanoes\nfor all island stations this value was rounded off to 100 mm so that standard\nNational Park on the east flank of\nmoisture depletion tables could be employed in water balance calculations\nMauna Loa (pan data from Juvik and\n(Thornthwaite and Mather, 1957).\n3\nMeasured\nClarke, 1976).\n(Constant Level Pan)\n2\n0\n1000\n2000\nElevation on Mauna Loa (m)\n135","mm\n500\n6\n4\n5\nLaupahoehoe\nPaauhau\n2\n3\nWaimea\nNiulii\nKahua (Beach)\n1\nPuu Lalau\n10m\n566m\n837m\n3m\n1157m\n23m\n200\n100\n3556\n2098\n1466\n3062\n179\nmm\nmm\nmm\nmm\nmm\n1634\n1615\n1484\n1022\n1946\n1296\nmm\n1362\n500\n18\n17\n16\n15\n14\nAina Hou Makai\nMauna lu\n13\nKapoho\nVolcano\nKulani Mauka\nKulani\n1370m\n34m\n838m\n2527m\n1574m\n1209m\n200\n100\n1346\n2524\n1955\n2522 1205 mm\n2505\nmm\nmm\nmm\nmm\n1255\n1039\n1654\n1518\n1256\nmm\n1116\n500\n27\n26\n23\n24\n25\nMiddle Holualoa\nOld Kona Airport\n22\nKainaliu\nPapaloa\nOpihihale\nHonaunau\n145m\n5m\n457m\n1560m\n387m\n332m\n200\n100\n1064\n1228\n1803\n1110\n567\n630\nmm\nmm\nmm\nmm\nmm\nmm\n1640\n1511\n1496\n1492\n1743\n1251\n0\nJ FMAMJJA S OND J FMAMJJA SOND J F M A M J AS OND JFMAMJJ ASOND JFMAMJ J A S o N D J F MAM J J A S\nSOIL MOISTURE\nSOIL MOISTURE\nWATER\nWATER\nACTUAL\nPOTENTIAL\nPRECIPITATION\nRECHARGE\nUTILIZATION\nEVAPOTRANSPIRATION\nSURPLUS\nDEFICIENCY\nEVAPOTRANSPIRATION\n136","mm\n500\n7\n8\n9\n10\n11\n12\nKeanakolu\nHale Pohaku\nPuu Huluhulu\nSaddle Road\nHilo\nMauna Loa\nObservatory\n1609m\n2810m\n2043m\n712m\n12m\n200\n3397m\n100\n475\nmm\n689\n1110\n1948\n640\n4771\n3526\nmm\nmm\nmm\nmm\nmm\n1139\n1210\n1065\n1394\n1669\n500\nJFMAMJJASONDJFMAMJ J A SOND JFMAMJJ A SOND\n19\n20\n21\nPahala\nKiolakaa\nManuka\n265m\n320m\n533m\n200\nSurplus\n0\n3\n-1000\n0-1000 mm\n900/\nin\n2\nKohala 5\n100\n20°N\n4\n6\nDeficit\n30\n< 1000 mm\n1513\n1157\n1369\n180000\n7\nmm\nmm\n500\nmm\n30000\n1670\n1612\n1571\n# 4205\nMauna Kea\n500\nA\nDeficit\n28\n29\n30\nHualalai\n8\nHolualoa\nHonuaula\nKeamoku\n26\n0-1000 mm\nA\n10\n11\n980m\n1930m\n952m\n200\n9\n29\n28\n13\n27\n12\nSurplus\nMauna Loa\nA\nA\n14\n1000 mm\n24\n4\n4168 m\n25\nA\n18\n16\n15\n23\n100\n3000\n17\n22\n2400\n2065\n19\n1012\n601\nDeficit\n1800\nmm\nA\nmm\nmm\n1409\n1126\n1360\n0-1000 mm\n1200\n0\n10\n20 km\n0\nJ F M A M J . J AS OND J F M A M J J A S o N D J F M A M J J A S OND\n21\n600\nScale in Kilometers\nA\n20\nFigure 8. Annual water balance on\nthe island of Hawaii. Ratios are mean\nannual precipitation over mean\nannual potential evapotranspiration\nin millimeters.\n137","SUMMARY\nThe 30 water balance diagrams constructed for the island depict both\nsteep gradients and pronounced regional differences in seasonal moisture\nThe Thornthwaite water balance diagrams and map demonstrate\nsurplus and deficit. In Fig. 8 the difference between annual precipitation and\ngraphically the tremendous climatic diversity on the island of Hawaii.\npotential evapotranspiration has been mapped in four zones:\nAlthough the Köppen map (Fig. 2) shows only a relatively small portion of the\n1. Annual surplus exceeding 1,000 mm. This zone comprises 20% (2,100\nisland to be arid or semi-arid (Bwh and BSh), from the water balance analysis\nkm² of the island area and is restricted to the high-rainfall regions of\nit is evident that nearly 60% (zones 3 and 4 above) of the island experiences\nwindward Mauna Kea, Mauna Loa, and the summit area of Kohala. The\nan annual moisture deficit.\nannual moisture surplus in this zone ranges as high as 3,377 mm (station 10) at\nmiddle elevations. All stations within this zone (stations 2, 6, 10, 11, 14, and\n15) exhibit an absolute winter maximum in precipitation, and a secondary\nsummer maximum also occurs at elevated stations where summer orographic\nprecipitation is exaggerated (e.g., stations 2 and 14).\n2. Annual surplus between 0 and 1,000 mm. This zone comprises 21%\n(2,200 km ² of the island area and extends from middle to high elevations on\nthe windward slopes down to sea level in those areas where slope aspect is not\noriented perpendicular to prevailing trade wind flow, and thus the orographic\nrainfall component is diminished. For the windward stations there is typically\na moderate summer drought (stations 5, 7, 9, 13, and 17) from 2 to 5 months\nlong. The increased strength of the trade wind inversion in particular limits\nsummer rainfall at higher elevations. The localized core area of high\nconvectional rainfall in Kona also falls within this moisture zone. However,\nhere the deficit period occurs in winter (stations 24, 28, and 29) and is not\nsevere.\n3. Annual deficit between 0 and 1,000 mm. This zone comprises 54% (5,600\nkm² of the island area and occupies a predominantly leeward location on\nKohala, Mauna Kea and Mauna Loa. The drought period may be concentrated\nin either the summer (on the windward side) or the winter (on the Kona side)\nand is typically 6 to 12 months long.\n4. Annual deficit exceeding 1,000 mm. This zone comprising 5% (550 km²\nof the island area is restricted to leeward Kohala and north Kona. The annual\nmoisture deficit may exceed 1,900 mm (station 1).\n138","KNOWLEDGMENTS\nThis research was supported in part by grants from the Hawaii Natural\nory Association and the U.S. Department of Interior, Office of Water\nources Research.\nREFERENCES\nCarter, D. B., and J. R. Mather, 1966: Climatic Mueller-Dombois, D., 1966: Climate. Chap. IV State of Hawaii, 1970: An inventory of basic\nclassification and environmental biology.\nin Atlas for Bioecology Studies in Hawaii\nwater resources data-island of Hawaii. Rep.\nPubl. Climatol. 19(4):305-390.\nVolcanoes National Park, edited by Maxwell\nR34, Dept. of Land and Natural Resources,\nChang, J. -H., 1959: An evaluation of the 1948\nS. Doty and D. Mueller-Dombois, U.S.\nHonolulu, Hawaii, 188 pp.\nThornthwaite classification. Ann. Assoc.\nNational Park Service, 507 pp. (Republished\nTaliaferro, W. J., 1959: Rainfall of the\nAmer. Geogr., 49(1):24-30.\nas Hawaii Agr. Exp. Sta. Bull. 89, 1970.)\nHawaiian Islands. Hawaii Water Authority,\nChang, J.-H., 1968: Climate and Agriculture\nMueller-Dombois, D., 1976: The major\n394 pp.\nAldine Publishing Co., Chicago, 304 pp.\nvegetation types and ecological zones in\nThornthwaite, C. W., 1948: An approach\nJuvik, J. O., and G. G. Clarke, 1976:\nHawaii Volcanoes National Park and their\ntoward a rational classification of climate.\nTopoclimatic gradients in Hawaii Volcanoes\napplication in park management and\nGeogr. Rev., 38:55-94.\nNational Park (abstract). In Proceedings of\nresearch. In Proceedings of the First\nThornthwaite, C. W., and J. R. Mather, 1957:\nthe First Conference in Natural Sciences,\nConference in Natural Sciences, Hawaii\nInstructions and tables for computing\nHawaii Volcanoes National Park, edited by\nVolcanoes National Park, edited by S. W.\npotential evapotranspiration and the water\nS. W. Smith, Department of Botany,\nSmith, Department of Botany, University of\nbalance. Publ. Climatol. 10(3):1-311.\nUniversity of Hawaii, Honolulu, p. 113.\nHawaii, Honolulu, 149-161.\nWalter, H., and H. Lieth, 1960:\nJuvik, J. O., and D. J. Perreira, 1974: Fog\nPrice, S., 1973: Climate. In Atlas of Hawaii,\nKlimadiagram-Weltatlas, Jena.\ninterception on Mauna Loa, Hawaii. Proc.\nedited by R. W. Armstrong, University of\nAssoc. Amer. Geogr., 6:22-25.\nHawaii Press, Honolulu, 53-60.\nLöf, G. O. C., J. A. Duffie, and C. O. Smith,\nSato, H. H., et al., 1973: Soil Survey of Island\n1966: World distribution of solar radiation.\nof Hawaii, State of Hawaii. U.S.D.A. Soil\nRep. 21, Solar Energy Laboratory, Univ. of\nConservation Service, 115 pp. (plus maps).\nWis., 59 pp. (plus maps).\n139","Howard Tatum (now Meteorologist-\nin-Charge at Hilo), Bob Williams,\nHarry Wexler, Jack Pales, and \"Doc\"\nColby Foss at the Hilo airport, 1959.\nThis pyrheliometer, attended in 1959\nby Jack Pales, is still in use. To the\nright are early radiometers, no longer\nused. The generator building is visible\nbehind Pales (Official U.S. Navy\nPhotograph).\nThe observatory building in June\n1956.\n140","A pictorial history\nof Mauna Loa Observatory\nSaul Price at the dedication of the\nsummit weather station in 1954.\n141","Mauna Loa and Weather Bureau per-\nsonnel in November 1959. Standing:\nBill Cobb (MLO), Ray Busniewski\n(WB), Nels Johnson (WB), Cliff Ku-\ntaka (MLO), (a Weather Bureau\nvisitor?), and Jack Pales (Director,\nMLO): kneeling: Tom Tyrrel (MLO).\nThe kitchen and two bunk rooms at\nMLO (1973) have been displaced as\nwalls have been relocated and new\nequipment has been installed.\nCoke\nCoke\n142","The Dobson shelter (with sliding\nroof) in April 1958.\nThe CO2 2 instrumentation at MLO in\nSeptember 1959.\nCliff Kutaka taking a Dobson spec-\ntrophotometer measurement at the\nobservatory in the late 1950s.\nagoc,\n143","Solar radiation instruments at the Hi-\nlo Airport, 1960. MLO's Hilo offices\nwere in the Air Cargo Terminal\nbuilding.\nMLO interior in the early 1960s. The\nhand calculator is still part of the\nequipment. The microphone and\ntransceiver were used to communicate\nwith those traveling between the ob-\nservatory and Hilo, in case of trouble\non the road.\n144","Lothar Ruhnke, Director, MLO, and\nWayne Iwaoka, summer student em-\nployee, take measurements with a\nsmall particle counter, June 28, 1966.\nIn the main building of MLO, August\nAt MLO in June 1967 (left to right):\n1966: Bill Waters, Ralph Stair,\nWayne Iwaoka, Sharon Nagata, Ber-\nMrs. Stair, and Howard Ellis.\nnard Mendonca, Helmut Weickmann\n(Director, APCL), Howard Ellis, Judy\nBright, Lothar Ruhnke, and Josef\nMueller.","MLO/HAO Open House, December\n4, 1966. Dick Hansen (in baseball\ncap) conducts a tour.\n146","Jim Simpson (NRC Resident Research\nAssociate) at lava flow, Kalapana Na-\ntional Park, September 1970. MLO\nfollowed the flow of lava for a lab-\noratory experiment and measured the\ngaseous effluent on site as the lava\nentered the sea.\nMLO staff during visit of Deputy Dir-\nector of ERL, in front of the Cloud\nPhysics Observatory in Hilo. Left to\nright: John Chin, Rudolf Pueschel\n(Director, MLO), Bernard Mendonca,\nJudy Bright Pereira, Robert Knecht\n(Dep. Director, ERL), Howard Ellis.\nMauna Loa Observatory entertains\nvisitors on its 15th anniversary.\n147","1\nDon Pack, Rudy Pueschel, and Larry\nMehau (Island Land-Use Committee\nmember) at MLO's 15th Anniversary\nopen house, January 9, 1972.\nMLO staff in 1972. Standing: Howard\nEllis, John Chin, Marge Kealanahele,\nAlan Yoshinaga, Ronald Fegley,\nKneeling: Mark Goldman, Al Shi-\nbata.\n148","Student Aides Charmaigne Makanui\nand Leona Ferreira, August 1976.\nMauna Loa Observatory staff mem-\nbers and visitors in February 1972.\nBack row: Visitor (?), Charles Garcia,\nBernard Mendonca, Dick Proulx, Vis-\nitor (?), and Earl Barrett. Front row:\nVisitor (?), Eric Yasukawa, Barry\nBodhaine, Walter Komhyr, John\nChin, Bob Grass, and Al Shibata.\nThe front dome houses the Dobson\nspectrophotometer. The NCAR coro-\nnagraph is in the rear dome.\n149","Judy Bright Pereira\nBy April 1975 facilities and equip-\nment had proliferated to include tow-\ners, trailers, diesel fuel tanks, domes,\nantennas, and signs of a multitude of\nindividual experiments (photo by\nMargie Kealanahele\nRobert K. McGill).\n150","The new solar radiation array, 1977.\nMauna Kea is in the background.\nKin Coulsen, U. of California at Da-\nMauna Loa Observatory during re-\nvis, was Senior Scientist at Mauna\nconstruction in 1977.\nLoa Observatory for six months in\n1977.\n151","XIX XXXX\nXXX\n8\n6\n4\n1\n2\n7\n9\n3\n5\nMauna Loa staff in March 1978:\n1. Howard Ellis 2. Mamaru (Al) Shi-\nbata 3. Judy Pereira 4. Alan Yash-\ninaga 5. Cynthia Aki 6. John\nChin 7. Dick Cram 8. Duane Hard-\ning 9. John Miller. Missing: Sandra\nIreland, Lily Kam, and Dan Naka-\nmura.\nMauna Loa Observatory in March\n1978 had a new tower for measuring\nsolar radiation, and a new 40-ft air\nsampling intake stack, looking here\nalmost as tall as the 80-ft tower in the\nbackground.\nin the newly modernized\nvilding Alan Yoshinaga is","s\nThe Rev. Edward Kealanahele blessed\nthe observatory, staff members, and\nvisitors participating in the celebra-\nMauna Loa Observatory 20th anni-\ntion.\nversary celebration, January 28, 1978.\n153","The Honorable Daniel Akaka, U.S.\nRepresentative from Hawaii, and Dr.\nLester Machta, Director, Air Re-\nsources Laboratories, addressed the\ngathering.\nA cake to celebrate the anniversary.\nFROM. C.P.O.\n154","BIBLIOGRAPHY\nAngell, J. K., and J. Korshover, 1976: Global analysis of recent total ozone\nFegley, R. W., and H. T. Ellis, 1975: Lidar observations of a stratospheric dust cloud\nfluctuations. Mon. Wea. Rev., 104(1):63-75.\nlayer in the tropics. Geophys. Res. Lett., 24:139-141.\nBarrett, E. W., R. F. Pueschel, P. M. Kuhn, and H. K. Weickmann, 1970: Inadvertent\nFegley, R. W., and H. T. Ellis, 1975: Monitoring of a stratospheric dust cloud using\nmodification of weather and climate by atmospheric pollutants. ESSA Tech. Report\nlidar. Presented at the Optical Society of America in Anaheim, Calif., Mar. 17-21,\nERL 185-APCL 15, 110 p.\n1975. Available from the Optical Society of America.\nFegley, R. W., and H. T. Ellis, 1975: Optical effects of the 1974 stratospheric dust\nBigg, E. K., 1968: Ice nucleus concentrations in Hawaii. J. Appl. Meteorol., 7:951-952.\ncloud. Applied Optics, 14:1751-1752.\nBodhaine, B. A., 1972: The effects of ammonia on the electrification of freezing and\nFox, R. L., 1956: The Mauna Loa Observatory, Weatherwise, 9(7):147-150.\nsplashing water drops. Tellus, 24:473-479.\nFox, R. L., 1956: New Mauna Loa Observatory unit. Nature, 178(4545):1272.\nBodhaine, B. A., and R. F. Pueschel, 1972: Flame photometric analysis of the transport\nFullerton, C. M., C. J. Garcia, and G. Langer, 1975: Nine months of ice nuclei\nof sea salt particles. J. Geophys. Res., 77(27):5106-5115.\nmonitoring at Mauna Loa, Hawaii. Meteorol. Rundschau, 28:178-190.\nBodhaine, B. A., and R. F. Pueschel, 1974: Source of seasonal variations in solar\nGarratt, J. R., and G. I. Pearman, 1972: Atmospheric carbon dioxide. In Proceedings\nradiation at Mauna Loa. J. Atmos. Sci., 31:840-845.\nof the International Clean Air Conference, Melbourne, Australia. May 15-18, 1972,\nBodhaine, B. A., and B. G. Mondonca, 1974: Preliminary four-wavelength nephe-\nClean Air Society of Australia and New Zealand, CONF 72-05-41.\nlometer measurements at Mauna Loa Observatory. Geophys. Res. Lett., 1:119-122.\nCharlson, R. J., W. M. Parch, A. P. Waggoner, and N. C. Ahlquist, 1974:\nGoldberg, B., and W. H. Klein, 1977: Variations in the spectral distribution of\nBackground aerosol light scattering characteristic: nephelometric observations at\ndaylight at various geographical locations on the Earth's surface. Solar Energy,\nMauna Loa Observatory compared with results at other remote locations. Tellus,\n19:3-13.\n26:345-360.\nGoldman, M. A., 1974: Carbon dioxide measurements and the local wind patterns at\nChin, J. F. S., H. T. Ellis, B. G. Mendonca, R. F. Pueschel, and H. J. Simpson, 1971:\nMauna Loa Observatory, Hawaii. J. Geophys. Res., 79(30):4550-4554.\nGeophysical monitoring at Mauna Loa Observatory. NOAA Tech. Memo. ERL\nGrass, R. D., 1973: Observations of total ozone with Dobson spectrometers. Observa-\nAPCL-13, 36 p.\ntions and Measurement of Atmospheric Pollution, Special Environmental Report No.\nCobb, W. E., and B. B. Phillips, 1962: Atmospheric electric measurement results at\n3, WMO No. 368, 413-422.\nMauna Loa Observatory. Tech. Paper No. 46, U.S. Weather Bureau, 252 p.\nHansen, R. T., S. F. Hansen, and S. Price, 1965: An example of meteorological consid-\nCoulson, K. L., R. L. Walraven, and S. B. Soohoo, 1974: Polarization of skylight at\nerations in selecting an observatory site in Hawaii. Publications of the Astronomical\nan altitude of 3416 m (11200 ft) on Mauna Loa, Hawaii. Contributions in Atmos-\nSociety of the Pacific, 78(460):14-29.\npheric Science No, 9, 49 p.\nHerbert, G. A., D. H. Pack, R. Fegley, D. Hoyt, W. Komhyr, J. Miller, and\nCouncil on Environmental Quality, Federal Council for Science and Technology, 1975:\nC. Turner, 1973: Geophysical monitoring for climatic change-the NOAA program.\nFluorocarbons and the environment. A report of the Federal Task Force on Inad-\nObservation and Measurement of Atmospheric Pollution, Special Environmental\nvertent Modification of the Stratosphere (IMOS), 36-39.\nReport No. 3, WMO No. 368, 334-345.\nDroessler, E. G., and K. J. Hefferman, 1965: Ice nucleus measurements in Hawaii. J.\nHill, W. J., P. N. Sheldon, and J. J. Tiede, 1977: Analyzing worldwide total ozone for\nAppl. Meteorol., 4:442-445.\ntrends. Geophys. Res. Lett., 4(1):21-24.\nEllis, H. T., and R. F. Pueschel, 1971: Solar radiation: Absence of air pollution trends\nHobbs, P. V., C. M. Fullerton, and G.C. Bluhm, 1971: Ice nucleus storms in Hawaii.\non Mauna Loa. Science, 172:845-846.\nNature, 230:90-91.\nEllsaesser, H. W., R. F. Pueschel, and H. T. Ellis, 1972: Turbidity of the atmosphere:\nsource of its background variation with the season. Science, 176:814-815.\n155","Hoffman, G. L., and R. A. Duce, 1971: Copper contamination of atmospheric particu-\nKruger, P., 1969: P210 in surface air along the slopes of Mauna Loa volcano, Hawaii.\nlate samples collected with Gelman hurricane air samplers. Environ. Sci. & Tech.,\nFinal rep. to U.S. AEC, SU-326-PA-16-3, 24 p.\n5:1134-1136.\nKruger, P., and A. Miller, 1966: Transport of radioactivity in rain and water across\nHogan, A. W., 1975: Continuing survey of maritime aerosols. Final report, NOAA\nthe trade wind inversion at Hawaii. J. Geophys. Res., 71:4243-4256.\nGrant No. 04-4-022-11, Atmos. Sci. Research Center, State U. of N.Y. at Albany,\nLanger, G., C. J. Garcia, B. G. Mendonca, R. F. Pueschel, and C. M. Fullerton, 1974:\n77 p.\nHawaiian volcanoes: a source of ice nuclei? J. Geophys. Res., 79:873-875.\nHoyt, D. V., and G. A. Herbert, 1974: Ground based measurements of solar radiation\nLockhart, L. B., Jr., R. L. Patterson, Jr., A. W. Sanders, Jr., and R. W. Black, 1964:\nby geophysical monitoring for climatic change. Observations and Measurement of\nSummary report, Fission product radioactivity in the air along the 80th meridian\nAtmospheric Pollution, Special Environmental Report No. 3, WMO No. 368,\n(west), 1957-1962. NRL Report 6104.\n506-515.\nMachta, L., 1971: The role of the oceans and biosphere in the carbon dioxide cycle.\nHoyt, D. V., 1974: A review of presently available solar radiation instruments. Pro-\nPresented at the Nobel Symposium 20, \"Changing Chemistry of the Oceans,\"\nceedings of the Solar Energy Workshop, NOAA/ARL, 37-41.\nGothenberg, Sweden, Aug., 1971.\nKeeling, C. D., R. B. Bacastow, A. E. Bainbridge, C. A. Ekdahl, Jr., P. R. Guenther,\nMachta, L., 1972: Mauna Loa and global trends in air quality. Bull. Amer. Meteorol.\nL. S. Waterman, and J. S. Chin, 1976: Atmospheric carbon dioxide variations at\nSoc., 53(5):402-420.\nMauna Loa Observatory, Hawaii. Tellus, 28:538.\nMachta, L., G. Cotton, W. Hass, and W. Komhyr, 1975: Erythemal ultraviolet solar\nKeeling, C. D., and J. C. Pales, 1965: The concentration of atmospheric carbon\nradiation and environmental factors. Proc. 4th Conference on the Climatic Impact\ndioxide in Hawaii. J. Geophys. Res., 70:6053-6076.\nAssessment Program, Cambridge, Mass., Feb. 4-7, 1975, U.S. Dept. of Trans-\nKiess, C. C., C. H. Corliss, H. K. Kiess, and E. L. R. Corliss, 1957: High-dispersion\nportation Report DOT-TSC-OST-75-38 405-411.\nspectra of Mars. Astrophys. J., 126(3):579-584.\nMachta, L., G. Cotton, W. Hass, and W. D. Komhyr, 1977: CIAP measurements of\nKline, D. B., 1963: Evidence of geographical differences in ice nuclei concentrations.\nerythemal solar ultraviolet radiation. Proc. Joint Symposium on atmospheric ozone,\nMon. Wea. Rev., 91:681-686.\nVol. III, K. H. Grasnick (ed.), 87-104.\nKline, D. B., 1972: Measurement of ice nuclei and associated chloride particle concen-\nMachta, L., K. Hanson, and C. D. Keeling, 1976: Atmospheric carbon dioxide and\ntrations at Mauna Loa Observatory. J. Appl. Meteorol., 11:684-687.\nsome interpretations. Fate of fossil fuel CO2 in the oceans, N. R. Andersen and\nKomhyr, W. D., and T. B. Harris, 1977: Measurements of atmospheric CO2 at the\nA. Malahoff, Eds., Plenum Press, N.Y., 131-144.\nU.S. GMCC baseline stations. Air Pollution Measurement Techniques, Special\nMason, A. S., and H. G. Östlund, 1974: Atmospheric HT and HTO: Major HT\nEnvironmental Report No. 11, Report and Proceedings of the WMO Air Pollution\ninjected into the atmosphere, 1973. Geophys. Res. Lett., 1(6):247-248.\nMeasurement Techniques Conference (APOMET), Gothenberg, Sweden, 11-15\nMendonca, B. G., 1969: Local wind circulation on the slopes of Mauna Loa. J. Appl.\nOctober 1976, WMO No. 460, 9-19.\nMeteorol., 8(4):533-541.\nKomhyr, W. D., and T. M. Thompson, 1977: Fluorocarbon-11 measurements at the\nMendonca, B. G., and W. T. Iwaoka, 1969: The trade wind inversion at the slopes of\nU.S. GMCC baseline stations. Air Pollution Measurement Techniques, Special\nMauna Loa, Hawaii. J. Appl. Meteorol., 8(2):213-219.\nEnvironmental Report No. 11, Report and Proceedings of the WMO Air Pollution\nMendonca, B. G., and G. Langer, 1973: Ice nucleus counts in varying ambient humidi-\nMeasurement Techniques Conference (APOMET), Gothenberg, Sweden, 11-15\nties using an NCAR ice nucleus counter. J. Appl. Meteorol., 30(7):1452-1454.\nOctober 1976, WMO No. 460, 208-215.\nMendonca, B. G., and R. F. Pueschel, 1973: Ice nuclei, total aerosol and climatology\nKruger, P., 1967: Transport of radioactive aerosols across the trade wind inversion at\nat Mauna Loa, Hawaii. J. Appl. Meteorol., 12(1):156-160.\nHawaii. Tellus, 19:381-391.","Miller, J. M. (ed.), 1973: Geophysical Monitoring for Climatic Change, No. 1,\nPrice, S., 1957: Notes on the climate of Mauna Loa. Proc. Ninth Pacific Science\nSummary Report-1972. U.S. Dept. of Commerce, NOAA/ERL, 79 p.\nCongress, Bangkok, 1957, Vol. 13, Meteorology, p. 17.\nMiller, J. M. (ed.), 1974: Geophysical Monitoring for Climatic Change, No. 2,\nPrice, S., and J. C. Pales, 1959: Mauna Loa High Altitude Observatory. Mon. Wea.\nSummary Report-1973. U.S. Dept. of Commerce, NOAA/ERL, 104 p.\nRev., 87:114.\nMiller, J. M. (ed.), 1975: Geophysical Monitoring for Climatic Change, No. 3,\nPrice, S., and J. C. Pales, 1960: Some observations of ozone at Mauna Loa Obser-\nSummary Report-1974. U.S. Dept. of Commerce, NOAA/ERL, 107 p.\nvatory, Hawaii. Symposium on Atmospheric Ozone, Monograph No. 3, Inter-\nMiller, A. J., J. M. Miller, and R. M. Rotty, 1975: Two case studies correlating the\nnational Union of Geodesy and Geophysics, p. 37.\nbaseline CO2 record at Mauna Loa with meteorological and oceanic parameter's.\nPrice, S., and J. C. Pales, 1963: Local volcanic activity and the ice nuclei concen-\nNOAA Tech. Memo. ARL-49, 9 p.\ntrations on Hawaii. Archiv für Meteorologie, Geophysik und Bioklimatologie, Ser.\nMoore, H. E., S. E. Poet, L. A. Martell, and M. H. Wilkening, 1974: Origin of 222Rn\nA, 13(3-4):398-407.\nand its long-lived daughters in air over Hawaii. J. Geophys. Res., 79(33):5019-5024.\nPrice, S., and J. C. Pales, 1963: Mauna Loa Observatory: The first five years. Mon.\nNakaya, U., J. Sugaya, and M. Shoda, 1957: Report of the Mauna Loa expedition in\nWea. Rev., 91:665-680.\nthe winter of 1956-1957. J. Fac. Sci., Hokkaido University, Ser. II, 5(1):1-36.\nPrice, S., and J. C. Pales, 1964: Ice nucleus counts and variations at 3.4 km and near\n(Reprinted by Munitalp Foundation, N.Y., as Occasional Paper No. 0028.)\nsea level in Hawaii. Mon. Wea. Rev., 92:207-221.\nOltmans, S. J., 1973: Surface ozone monitors. Observations and Measurement of\nPueschel, R. F., B. A. Bodhaine, and B. G. Mondonca, 1973: The proportions of\nAtmospheric Pollution, Special Environmental Report No. 3, WMO No. 368,\nvolatile aerosols on the island of Hawaii. J. Appl. Meteorol., 12(2):308-315.\n394-403.\nPueschel, R. F., and H. T. Ellis, 1972: \"Reply\": Turbidity of the atmosphere: Source of\nOstlund, H. G., A. S. Mason, and A. Ydfalk, 1972: Atmospheric HT and HTO,\nits background variation with the season. Science, 176:815.\n1968-71. Tritium Laboratory Data Report No. 2, Rosenstiel School of Marine and\nPueschel, R. F., and G. Langer, 1973: Sugar cane fires as a source of ice nuclei in\nAtmospheric Sciences, University of Miami, Miami, Florida, pp. 10-11, 28-29.\nHawaii. J. Appl. Meteorol., 12(3):549-551.\nÖstlund, H. G., and A. S. Mason, 1974: Atmospheric HT and HTO: Experimental\nPueschel, R. F., C. J. Garcia, and R. T. Hansen, 1974: Effects of atmospheric water\nprocedures and tropospheric data 1968-1972. Proceedings of the International\nvapor and volcanic aerosols. J. Appl. Meteorol., 13:397-401.\nSymposium on Atmospheric Trace Gases, Mainz, West Germany, Tellus, 26(1-\nPueschel, R. F., L. Machta, G. E. Cotton, E. C. Flowers, and J. T. Peterson, 1972:\n2):91-102.\nNormal incidence radiation trends on Mauna Loa, Hawaii. Nature, 240:545-547.\nPack, D. H., R. Fegley, G. Herbert, D. Hoyt, W. Komhyr, J. Miller, and C. Turner,\nPueschel, R. F., and B. G. Mendonca, 1972: Sources of atmospheric particulate matter\n1973: Geophysical monitoring for climatic change; the NOAA program. In Observa-\nin Hawaii. Tellus, 24(2):139-149.\ntions and the Measurement of Atmospheric Pollution, Special Environmental Report\nRobinson, G. D., 1974: Scattering and absorption properties of atmospheric particles\nNo. 3, WMO No. 368, 334-345.\ndeduced from routine records of solar radiation. CEM Report No. 4149-507, The\nPales, C. J., and C. D. Keeling, 1965: The concentration of atmospheric carbon\nCenter for the Environment and Man, Inc., Hartford, Conn., 37-41.\ndioxide in Hawaii. J. Geophys. Res., 70(24):6053-6076.\nRobinson, G. D., 1976: Examination of some solar radiation records from Mauna Loa\nPearman, G. I., and J. R. Garratt, 1972: Global aspects of carbon dioxide. Search,\nObservatory. CEM Report No. 4181-543, The Center for the Environment and Man,\n3(3):67-73.\nInc., Hartford, Conn., 33 p.\nPearman, G. I., 1977: Further studies of the comparability of baseline atmospheric\ncarbon dioxide measurements. Tellus, 29:171-181.","Rotty, R. M., 1973: Global production of CO2 from fossil fuels and possible changes\nVolchok, H. L., L. Toonkel, and M. Shonberg, 1975: Radionuclides and lead in surface\nin the world's climate. Presented at the ASME-IEEE Joint Power Generation Confer-\nair. U.S. ERDA Report HASL-298, Appendix, p. B-1-B-140.\nence, New Orleans, La., Sept. 16-19, 1973, ASME Paper 73-Pwr-11, 12 p.\nWall, A. C., 1959: Measurements of the vertical distribution of ozone at Mauna Loa\nRuhnke, Lothar H., 1969: Area averaging of atmospheric electric currents. J. Geomag.\nObservatory, Hawaii. Thesis, University of Hawaii, Honolulu, 60 p.\nGeoelec., 21:453-462.\nWatkins, J. A. (ed.), 1976: Geophysical Monitoring for Climatic Change, No. 4, Sum-\nRuhnke, L. H., and J. T. Dennett, 1967: Mauna Loa Observatory: High altitude\nmary Report-1975. U.S. Dept. of Commerce, NOAA/ERL, 131 p.\nscience in a tropical maritime environment. ESSA Tech. Memo. ERLTM-APCL 10,\nWilkening, M. H., 1971: Atmospheric radon-222 and lead-210 in Hawaii. Interim\nReport, Atmospheric Sciences Section, National Science Foundation, Washington,\n34 p.\nSemonin, R. G., 1972: Comparative chloride concentrations between Mauna Loa\nD.C.\nObservatory and Hilo, Hawaii. J. Appl. Meteorol., 11:688-690.\nWilkening, M. H., 1974: Radon-222 from the island of Hawaii. Science, 183:413-415.\nSeto, Y. B., R. A. Duce, and A. H. Woodcock, 1969: Sodium-to-chlorine ratio in\nHawaiian rains as a function of distance inland and elevation. J. Geophys. Res.,\n74:1101-1103.\nShaw, G. E., 1976: Properties of the background global aerosol and their effects on\nclimate. Science, 192:1334-1336.\nSimpson, H. J., 1972: Aerosol and precipitation chemistry at Mauna Loa Observatory.\nNOAA Tech. Report ERL-248-APCL 24, 56 p.\nSimpson, H. J., 1972: Aerosol cations at Mauna Loa Observatory, J. Geophys. Res.,\n77:5266-5277.\nStair, R., and H. T. Ellis, 1968: The solar constant based on new spectral irradiance\ndata from 310 to 530 nanometers. J. Appl. Meteorol. 7(4):635-644.\nStair, R., and R. G. Johnston, 1956: Some studies of atmospheric transmittance on\nMauna Loa. J. Res. Natl. Bur. Stand., 61:419-425.\nTelegadas, K., 1972: Atmospheric radioactivity along the HASL ground-level sampling\nnetwork, 1968 to mid-1970, as an indicator of tropospheric and stratospheric\nsources. J. Geophys. Res., 77(6):1004-1011.\nVolchok, H. L., 1975: Worldwide deposition of SR-90 through 1974. U.S. ERDA\nReport HASL-297, p. I-1-I-56."]}