Detecting unvegetated permafrost-thaw features in Arctic tundra using imaging spectroscopy

Details

• Journal Title:
  Environmental Research Ecology
• Personal Author:
  Tenorio, Francia ; Sousa, Daniel ; Hashemi, Josh ; Davidson, Scott J. ; Oechel, Walter C. ; Tweedie, Craig E. ; Thomas, Bryan D. ; Rauschenbach, Sonya ; Zona, Donatella
• NOAA Program & Office:
  OAR (Oceanic and Atmospheric Research) ; GML (Global Monitoring Laboratory) ; CESSRST (Cooperative Science Center for Earth System Sciences and Remote Sensing Technologies)
• Description:
  Freeze-thaw cycles, hydrology, soil mixing, and permafrost degradation shape landscape patterns in permafrost regions, forming unvegetated or sparsely vegetated landforms with varied topography. These features, which we refer to as unvegetated thermokarst polygons, result from permafrost thaw and may play a critical role in Arctic greenhouse gas (GHG) dynamics by unlocking vast amounts of carbon stored in permafrost regions. However, their spatial extent remains uncertain due to their variable size (∼0.5-15 m) and the challenges of detecting them in heterogeneous landscapes such as polygonal tundra. This study assesses the extent of unvegetated thermokarst polygons near Utqiaġvik, Alaska, using imaging spectroscopy and lidar data with a 1 m spatial resolution from the National Ecological Observing Network airborne observation platform on the joint-basis of a linear spectral mixture analysis and lidar-based relief. Based on a confusion matrix evaluating threshold models of our analysis against field-based results, the model that performed with the greatest F1 score (0.63), with balanced precision (0.71) and recall (0.57), estimated a 9.55% cover of unvegetated thermokarst polygons, or 380 000 m2 total area of unvegetated thermokarst polygons within the study extent. Ground truth data from 52 sites revealed that the spectral mixture model correctly mapped ∼81% of unvegetated thermokarst polygons. Our results highlight the challenges and potential of detecting unvegetated thermokarst polygons in complex Arctic landscapes with high-resolution remote sensing. Although these features currently cover a small area in the study site, their relevance may grow with increased landscape changes and GHG dynamics under warm conditions. Moreover, their extent could be greater in areas with more active permafrost degradation, emphasizing the need for detection methods to monitor these critical landforms.
• Keywords:
  • Imaging spectroscopy
  • Invegetated thermokarst polygons
  • Spectral unmixing
  • Hyperspectral
  • Permafrost degradation
  • Biogeochemical cycling
• Source:
  Environ. Res.: Ecol. 5 (2026) 015007
• DOI:
  https://doi.org/10.1088/2752-664X/ae487e
• Format:
  PDF
• Document Type:
  Journal Article
• Funding:
  Grant no. NA22SEC4810016
• Place as Subject:
  Arctic regions
• License:
  https://creativecommons.org/licenses/by/4.0/
• Rights Information:
  CC BY
• Compliance:
  Submitted
• Download URL:
  https://repository.library.noaa.gov/view/noaa/74449/noaa_74449_DS1.pdf
• File Type:
  [PDF - 6.28 MB]
• Collection(s):
  Office of Oceanic and Atmospheric Research (OAR) ; NOAA Cooperative Science Centers
• Main Document Checksum:
  urn:sha-512:8a46e7cef80217b3d5b15e1b2c1ed2b82f0971224d53d07f6567433f93962d9faba51466d99a7ed5acc9de2bd77e3da69b031ee56d52ab11ff3c88657c6bc959