Shortwave broadband black-sky surface albedo estimation for Arctic sea ice using passive microwave radiometer data

Abstract

[1] A new remote sensing method to estimate the optical broadband black-sky surface albedo using passive microwave radiometer data has been developed. In this research, the Advanced Very High Resolution Radiometer (AVHRR) based shortwave broadband black-sky albedos processed using the algorithms of the new surface broadband albedo CM-SAF product SAL, and the Advanced Microwave Scanning Radiometer (AMSR-E) microwave data from the National Snow and Ice Data Center, Boulder, Colorado, have been employed. To analyze the correspondence between AVHRR-based optical albedo and microwave brightness temperature, AMSR-E frequencies of 6.9, 18.7, and 36.5 GHz have been tested by fitting a third-degree polynomial curve to the SAL albedo and the AMSR-E data points. The best correlation occurs in 6.9 GHz vertical polarization brightness temperature (R2 = 0.92). To illustrate and compare the spatial variabilities of the SAL and AMSR-E albedos of the sea ice during the melting and refreezing period, maps of 13 weeks have been prepared of both albedos from June to August 2007. The albedo time series from AMSR-E and AVHRR data are calculated for the combined sea ice and open water cover for the Northern Hemisphere as a whole, and for six specific sea ice regions: the Arctic Ocean, the Kara and Barents Seas, the Greenland Sea, the Labrador Sea, Hudson Bay, and the Canadian Archipelago. The standard errors between the optical and passive microwave estimates varied from about 0.001 (Greenland Sea) to 0.04 (Canadian Archipelago) being 0.013 on the average in absolute units. The relative standard errors are then smaller than 5% in most of the regions.