Angular Effects and Correction for Medium Resolution Sensors to Support Crop Monitoring

Abstract

Remote sensing imagery at medium spatial resolutions (20–60 m) such as Landsat, the advanced wide field sensor (AWiFS) and the disaster monitoring constellation (DMC) have been broadly used in mapping crop types and monitoring crop conditions. This paper examines the influence of viewing and illumination angular effects on surface reflectance of typical surface and crop types for both narrow swath (e.g., Landsat) and wide swath (e.g., AWiFS) sensors. Three types of angular effects: 1) view angle effect; 2) day of year effect; and 3) mean local time drift effect were analyzed based on both field and satellite bi-directional reflectance distribution function (BRDF) measurements. In order to correct these angular effects, a BRDF look-up map (LUM) for major cover types was built using the cropland data layer (CDL) and the Moderate-Resolution Imaging Spectroradiometer (MODIS) BRDF products. The BRDF LUM was applied to an AWiFS image to correct view angle effects in an agricultural area in central Illinois. The resulting nadir BRDF-adjusted reflectance (NBAR) provides a consistent data source for intra-annual crop condition monitoring and inter-annual time-series analysis.