Snowmelt detection from QuikSCAT and ASCAT satellite radar scatterometer data across the Alaskan North Slope

Autor: Emily J. Sturdivant, Karen E. Frey, Frank E. Urban
Jazyk: angličtina
Rok vydání: 2019
Předmět:
Zdroj: GIScience & Remote Sensing, Vol 56, Iss 1, Pp 87-108 (2019)
Druh dokumentu: article
ISSN: 1548-1603
1943-7226
15481603
DOI: 10.1080/15481603.2018.1493045
Popis: The timing of seasonal snowmelt in high-latitude tundra has implications ranging from local biological productivity to global atmospheric circulation, yet remains difficult to quantify, particularly at large spatial scales. Snowmelt detection in such remote polar environments is possible using satellite-based microwave scatterometers, such as NASA’s QuikSCAT. QuikSCAT measured scattering in Ku-band, which is sensitive to snowmelt signals, from 1999 until the antenna failed in 2009. The Advanced Scatterometer (ASCAT) (2006–2021 (projected) operational), which operates at C-band, may be able to extend the QuikSCAT record, but existing techniques fail to adequately monitor tundra environments. Here, we designed a departure threshold algorithm to produce a consistent 15-year time series of melt onset for the tundra of the Alaskan North Slope, using the overlap period for the enhanced resolution datasets to calibrate the ASCAT melt detection record against QuikSCAT. We produced a time series of day of year of melt onset for 4.45 km x 4.45 km grid cells on the Alaskan North Slope from 2000–2014. Time series validation with in situ mean daily air temperature produced mean R2 values of 0.75 (QuikSCAT) and 0.72 (ASCAT). We qualitatively observed a difference between early-season melt, which occurred rapidly and was driven by strong wind events, and more typical melt, which occurred gradually along a latitudinal gradient. We speculate that future melt timing will have greater frequency of early-season onset as climate change destabilizes the high-latitude atmosphere.
Databáze: Directory of Open Access Journals
Externí odkaz: