Mapping of seasonal freeze-thaw transitions across the pan-Arctic land and sea ice domains with satellite radar

• Published in: Journal of Geophysical Research Vol. 117; no. C8; pp. C08004 - n/a
• Main Authors: Mortin, J., Schrøder, T. M., Walløe Hansen, A., Holt, B., McDonald, K. C.
• Format: Journal Article
• Language: English
• Published: Washington, DC Blackwell Publishing Ltd 01.08.2012; American Geophysical Union
• Subjects: active microwave; Air temperature; atmosfärvetenskap och oceanografi; Atmospheric Sciences and Oceanography; Climatology; Cryosphere; Earth sciences; Earth, ocean, space; Exact sciences and technology; freeze-thaw; Geophysics; High performance computing; Ice; Marine; Oceans; pan-Arctic; Remote sensing; Sea ice; Snow depth; Surface temperature; Temperature measurement; terrestrial; polar regions; Arctic region; algorithms; Climatology; Buoy; outliers; networks; Outlier; Meteorological station; Atmospheric temperature; Ground surface; sea ice; spatial resolution; Snow cover; thawing; detection; Time series; Satellite observation; cartography; Layer thickness; Freeze thaw cycle; Temperature measurement; Radar observation; QuikSCAT satellite; Timing; Pan; radar methods
• ISSN: 0148-0227; 2169-9275; 2156-2202; 2156-2202; 2169-9291
• DOI: 10.1029/2012JC008001

To monitor the pan‐Arctic seasonal freeze‐thaw transitions of the land surface and sea ice, we analyze daily backscatter data from satellite scatterometry to examine the time series on an annual basis by applying an optimal edge detection scheme, and iterate against an internal median climatology to mitigate unreasonable outliers. By applying this novel algorithm to resolution‐enhanced QuikSCAT data from 1999 to 2009, we have mapped a decade of seasonal freeze‐thaw transitions across the landmass and sea ice north of 60°N at a spatial resolution better than 5 km. The data set has been validated against surface air temperature measurements and snow depth obtained from a distributed network of weather stations and drift buoys. Most retrieved timings from surface and QuikSCAT measurements agree to less than a week at thaw transition for both land and sea ice and at freeze transition for sea ice, indicating successful retrieval over a range of surface covers. While the spatial pattern of freeze‐thaw transition changes substantially from year to year, the interannual variability of the mean transition timing over a particular surface is small. Key Points First map of pan‐Arctic seasonal freeze‐thaw transitions over land and sea ice Validation against freeze‐thaw transition retrieved from surface measurements Spatiotemporal patterns of pan‐Arctic seasonal freeze‐thaw transition