IN31A-3706:
Improvements and Extension to a Global Earth System Data Record of Daily Landscape Freeze-Thaw Status Determined from Satellite Microwave Remote Sensing

Wednesday, 17 December 2014
Youngwook Kim, Univ Montana, NTSG, Missoula, MT, United States; The University of Montana, FLBS, Polson, MT, United States, John S Kimball, The University of Montana, Flathead Lake Biological Station, Polson, MT, United States, Jinyang Du, University of Montana, Numerical Terradynamic Simulation Group, Missoula, MT, United States and Joseph M Glassy, Lupine Logic Inc., Missoula, MT, United States
Abstract:
A global satellite microwave Earth System Data Record of daily landscape freeze-thaw status (FT-ESDR) has been commonly used to quantify cold temperature impacts on productivity, phenology, evapotranspiration and the terrestrial carbon cycle. Overlapping 37 GHz, vertically polarized brightness temperature (Tb) measurements from the Scanning Multichannel Microwave Radiometer (SMMR) and Special Sensor Microwave Imager (SSM/I) were integrated to produce a temporally consistent and continuous global daily FT data record from 1979 to 2012 and derived at 25-km pixel resolution. In this study, we develop and evaluate FT-ESDR enhancements, including expanded record length and spatial coverage, alternate algorithm calibrations, and a finer scale FT classification. A larger global domain is evaluated that encompasses all land areas affected by seasonally frozen temperatures, including urban, snow-ice dominant, barren, and permafrost landscapes. The FT retrieval is obtained using a seasonal threshold algorithm (STA) that classifies daily Tb changes in relation to frozen and non-frozen Tb reference states on a per-pixel basis. STA sensitivity to FT reference states is evaluated and alternative ancillary data are applied for defining Tb reference conditions, including surface temperatures from global reanalysis and MODIS land surface temperature (LST) seasonal climatology. The resulting FT record shows mean annual spatial classification accuracies of 92 and 86 percent for PM and AM overpass retrievals relative to in-situ temperature measurements. Despite the larger domain and longer record, the new FT-ESDR showed a 1-3 percent spatial classification accuracy improvement over previous FT-ESDR versions. Areas with enhanced accuracy include the Central USA, Central Asia, and North and Central Europe. Sub-grid land surface spatial heterogeneity effects on the aggregate FT retrievals are also assessed to refine FT-ESDR data quality metrics. The results of this study are being applied for continuing FT-ESDR production and utility enhancements, and to inform development of similar FT algorithms and products from the NASA SMAP mission. This study was funded under the NASA MEaSUREs program.