Author
 |
CHOPPING, MARK |
|
Submitted to: Remote Sensing of Environment
Publication Type: Peer Reviewed Journal Publication Acceptance Date: 3/31/2000 Publication Date: 6/15/2000 Citation: N/A Interpretive Summary: In this study a number of linear semi-empirical kernel-driven (LiSK) bi-directional reflectance distribution function (BRDF) models are adjusted against an extensive Advanced Very High Resolution Radiometer (AVHRR) dataset collected over a variety of semi-arid cover types in the southern part of New Mexico and parts of Chihuahua, Mexico. The data were acquired as part of the May 1997 Prototype Validation Exercise (PROVE) campaign, an activity of the NASA Earth Observing System Terra validation program. The aim is to investigate the behavior of these BRDF models under conditions of sparse angular sampling such as those provided by medium-resolution satellite-borne sensors such as the AVHRR. The processing methods used are fully described, and important science issues are addressed including the angular sample provided by the two AVHRRs, calibration of digital counts to spectral radiance, retrieval of surface reflectance via atmospheric correction, contamination screening and inversion protocols. Large-scale BRDF retrieval via inversions of these LiSK models with data from the two AVHRRs provides an important test of viability of future regional and global operational retrievals with the Moderate Resolution Imaging Spectro-Radiometer (MODIS) and Multi-angle Imaging Spectro-Radiometer (MISR) sensors on NASA's Terra satellite, while characterizing the directional reflectance behavior of a wide range of southwestern semi-arid and arid surface types from spaceborne measurements and allowing corrections for BRDF artifacts in multi-angular optical remote sensing data. Technical Abstract: Linear semi-empirical kernel-driven (LiSK) bi-directional reflectance distribution function (BRDF) models of the type to be used in the MODIS/MISR BRDF/albedo product (MOD43) are adjusted against multi-angular reflectance data derived from the Advanced Very High Resolution Radiometer (AVHRR) flown on two National Oceanic and Atmospheric Administration Polar Orbiting Environmental satellites. These models are appropriate for use with data covering large areas since they are analytical in formulation and can be adjusted very rapidly using matrix inversion techniques. The results show that these models are able to describe BRDF for a wide variety of southwestern desert surfaces and provide both a means for correcting for directional phenomena in off-nadir satellite data and for extracting structural information from multi-angular reflectance datasets. While model fits to observations are generally very good (the average root-mean-square-deviation for the 235,680 locations is around 3 percent reflectance in both visible and near-infrared channels), interpretation of the retrieved structural information is not straightforward. This may be owing in part to simplifying assumptions made in model derivations which are too severe and in part to problems of calibration and of accurately retrieving surface bi-directional reflectance at high solar and viewing zenith angles. |
