An Improved Physics-Based Model for Topographic Correction of Landsat TM Images

Abstract

Abstract:Optical remotely sensed images in mountainous areas are subject to radiometric distortions induced by topographic effects, which need to be corrected before quantitative applications. Based on Li model and Sandmeier model, this paper proposed an improved physics-based model for the topographic correction of Landsat Thematic Mapper (TM) images. The model employed Normalized Difference Vegetation Index (NDVI) thresholds to approximately divide land targets into eleven groups, due to NDVI's lower sensitivity to topography and its significant role in indicating land cover type. Within each group of terrestrial targets, corresponding MODIS BRDF (Bidirectional Reflectance Distribution Function) products were used to account for land surface's BRDF effect, and topographic effects are corrected without Lambertian assumption. The methodology was tested with two TM scenes of severely rugged mountain areas acquired under different sun elevation angles. Results demonstrated that reflectance of sun-averted slopes was evidently enhanced, and the overall quality of images was improved with topographic effect being effectively suppressed. Correlation coefficients between Near Infra-Red band reflectance and illumination condition reduced almost to zero, and coefficients of variance also showed some reduction. By comparison with the other two physics-based models (Sandmeier model and Li model), the proposed model showed favorable results on two tested Landsat scenes. With the almost half-century accumulation of Landsat data and the successive launch and operation of OPEN ACCESS Remote Sens. 2015, 7 6297Landsat 8, the improved model in this paper can be potentially helpful for the topographic correction of Landsat and Landsat-like data.

The Impact of Hurricane Maria on the Vegetation of Dominica and Puerto Rico Using Multispectral Remote Sensing

“…Optical remotely-sensed images in mountainous areas are subject to radiometric distortions induced by topographic effects, which need to be corrected before quantitative applications [37]. As shown in Table 2 and DEM data, the sun elevation angle is higher than most of the terrain slopes, and does not vary greatly over the study period.…”

The effect of topographic normalization on fractional tree cover mapping in tropical mountains: An assessment based on seasonal Landsat time series

“…Furthermore, land cover stratification based on NDVI has been successfully used to consider the land cover dependency of c (Hantson and Chuvieco, 2011;Li et al, 2015;Reese and Olsson, 2011), and such stratification was also applied in this study. Each image was divided into three NDVI classes: <0.4, 0.4-0.6, >0.6 (Fig.…”

Mapping Irrigated Areas of Northeast China in Comparison to Natural Vegetation

“…For example, there are large estimation errors over the eastern regions of Jilin province because of the mountainous areas. Optical remotely sensed images in mountainous areas are subject to radiometric distortions induced by topographic effects, which need to be corrected before quantitative applications [47]. In addition, the moisture condition substantially changes with the topography and slope aspect, and precipitation is not the only factor which determines the moisture condition of croplands and the adjacent forests [48].…”