Details:

Journal Title:
Remote Sensing of Environment
Personal Author:
Liu, Lingling ; Zhang, Xiaoyang ; Yu, Yunyue ; Guo, Wei
Liu, Lingling ; Zhang, Xiaoyang ; Yu, Yunyue ; Guo, Wei Less -
NOAA Program & Office:
NESDIS (National Environmental Satellite, Data, and Information Service) ; JPSS (Joint Polar Satellite System Program Office) ; STAR (Center for Satellite Applications and Research)
NESDIS (National Environmental Satellite, Data, and Information Service) ; JPSS (Joint Polar Satellite System Program Office) ; STAR (Center for Satellite Applications and Research) Less -
Description:
Real-time prediction of vegetation phenology is critical for assisting crop monitoring, natural resource management, and land modeling in weather prediction systems. However, due to the lack of timely available satellite datasets and the inherent noise in time series, little attention has been paid to real-time and short-term predictions of vegetation phenology. The successful launch of the Visible Infrared Imaging Radiometer Suite (VIIRS) instrument onboard operational Suomi National Polar-orbiting Partnership (Suomi NPP) satellite makes this research possible because it can provide land surface observations in a timely fashion. This study introduces an operational system that provides real-time and short-term predictions of vegetation phenology. Specifically, the system integrates timely available VIIRS observations and the climatology (expectation and standard deviation) of vegetation phenology from long-term MODIS data to simulate a set of potential temporal trajectories of greenness development at a given time for each pixel. These potential trajectories are then applied to identify spring green leaf development in real time, predict the occurrence of greenup onset, mid-greenup phase and maturity onset, and analyze the uncertainty of the prediction across a variety of ecosystems in North America. The accuracy of real-time and short-term predictions was evaluated by comparing with standard VIIRS detections and near-surface PhenoCam observations in both 2014 and 2015 across North America. The results showed that the real-time prediction of spring phenological metrics from VIIRS were all significantly correlated with those derived from PhenoCam datasets (R2 > 0.96, P < 0.01) and closely comparable to the standard VIIRS detections with a mean absolute difference of < 10 days, 5 days and 5 days in greenup onset, mid-greenup phase and maturity onset, respectively. The mean absolute difference in the northern region for all three events was relatively smaller than that in the southern region. These findings demonstrate the capability of VIIRS observations to effectively predict temporal dynamics of vegetation phenology in real time at a continental scale.
Source:
Remote Sensing of Environment, 194: 89-99
DOI:
https://doi.org/10.1016/j.rse.2017.03.009
Document Type:
Journal Article
Funding:
NOAA:JPSS_PGRR2_14
Rights Information:
Accepted Manuscript
Compliance:
CHORUS
Main Document Checksum:
[+]
urn:sha256:9887264a821335378891f12e192875917904ae7f73faff7d12ac23e6824c991e
Download URL:
https://repository.library.noaa.gov/view/noaa/48458/noaa_48458_DS1.pdf
File Type:

Supporting Files

• No Additional Files

More +