Skip to main content
ARS Home » Research » Publications at this Location » Publication #282272
Title: Generating 30-m land surface albedo by integrating landsat and MODIS data for understanding the disturbance

Author
item SHUAI, YANMIN - National Aeronautics And Space Administration (NASA)
item MASEK, JEFFREY - National Aeronautics And Space Administration (NASA)
item Gao, Feng
item SCHAAF, CRYSTAL - Boston University
item WILLIAMS, CHRISTOPHER - Clark University

Submitted to: Meeting Abstract
Publication Type: Abstract Only
Publication Acceptance Date: 5/1/2012
Publication Date: 5/9/2012
Citation: Shuai, Y., Masek, J., Gao, F.N., Schaaf, C., Williams, C. 2012. Generating 30-m land surface albedo by integrating landsat and MODIS data for understanding the disturbance. Meeting Abstract. 2012 CDROM.

Interpretive Summary:

Technical Abstract: Land cover change affects climate through both biogeochemical (carbon-cycle) impacts and biogeophysical processes such as changes in surface albedo, temperature, evapotranspiration, atmospheric water vapor, and cloud cover. Previous studies have highlighted that forest loss in high latitudes could cool the Earth through increased land surface albedo, while deforestation in the tropics could warm the Earth through decreased evaportranspiration rates and increased carbon emissions. However, these studies have not used data at the spatial resolution of human land dynamics (e.g. 30m). There is a need for improved estimates of land surface albedo at high resolution to fully understand the role of land cover change in climate forcing. Our initial “concurrent” approach required simultaneous acquisitions from Landsat and MODIS. BRDF information from MODIS products was retrieved for Landsat-based spectral clusters, and the albedo was calculated by scaling the per-cluster BRDF by the 30m Landsat reflectance value. However, this approach can only be applied for the post-2000 MODIS era. Here we present a “pre-MODIS era” approach to generate 30-meter albedos using Landsat surface reflectance (1970s-2000) and Look-Up-Tables (LUT) of anisotropy information extracted from MODIS BRDF data. A class-based LUT based on NLCD (National Land Cover Data) for non-disturbed and disturbed categories is assembled from high quality MODIS BRDF parameters (MCD43A1 product) and disturbance datasets (Monitoring Trends in Burn Severity dataset and the North American Forest Dynamics dataset). Each entry in the LUT reflects a unique combination of land cover type, disturbance age and type, season/month, and sensor wavelength-band. An initial version of the NLCD-type snow-free BRDF LUT has been generated for the Pacific Northwest of the United States. As a first evaluation, the BRDF evolution of evergreen forests in the NW US (MODIS tile h09v04) resulting from a non-fire disturbance (mostly harvest) during 1985-2008 is marked by a gradual increase in the NIR band and a reduction in the red & SWIR reflectance, and a different recovery trajectory was found for fire-disturbed forest. Once the classification-based LUT is acquired, the albedo-to-nadir-ratio method was applied to calculate the spectral albedos, followed by a narrow-to-broadband conversion to generate the broad-band shortwave albedo. Our preliminary comparison over the study region shows that the LUT-based land surface albedo, especially in the evergreen forest region, has a good agreement with retrievals from the “concurrent approach” with fine resolution albedo information, and wider dynamic albedo range than MODIS 8-day 500-m albedo product. Future work will expand the study area to include the full range of North American vegetation regimes. We intend to evaluate the net change in both winter and summer albedo across the continent due to land cover change processes occurring since the 1970’s. This information will also feed into a global analysis of albedo changes occurring since the 1700’s parameterized using the IPCC Global Land Use harmonization data set.