Does topographic normalization of landsat images improve fractional tree cover mapping in tropical mountains?

Abstract

ABSTRACT:Fractional tree cover (Fcover) is an important biophysical variable for measuring forest degradation and characterizing land cover. Recently, atmospherically corrected Landsat data have become available, providing opportunities for high-resolution mapping of forest attributes at global-scale. However, topographic correction is a pre-processing step that remains to be addressed. While several methods have been introduced for topographic correction, it is uncertain whether Fcover models based on vegetation indices are sensitive to topographic effects. Our objective was to assess the effect of topographic correction on the accuracy of Fcover modelling. The study area was located in the Eastern Arc Mountains of Kenya. We used C-correction as a digital elevation model (DEM) based correction method. We examined if predictive models based on normalized difference vegetation index (NDVI), reduced simple ratio (RSR) and tasseled cap indices (Brightness, Greenness and Wetness) are improved if using topographically corrected data. Furthermore, we evaluated how the results depend on the DEM by correcting images using available global DEM (ASTER GDEM, SRTM) and a regional DEM. Reference Fcover was obtained from wall-to-wall airborne LiDAR data. Landsat images corresponding to minimum and maximum sun elevation were analyzed. We observed that topographic correction could only improve models based on Brightness and had very small effect on the other models. Cosine of the solar incidence angle (cos i) derived from SRTM DEM showed stronger relationship with spectral bands than other DEMs. In conclusion, our results suggest that, in tropical mountains, predictive models based on common vegetation indices are not sensitive to topographic effects.

Topographic Correction of Landsat TM-5 and Landsat OLI-8 Imagery to Improve the Performance of Forest Classification in the Mountainous Terrain of Northeast Thailand

“…However, it should be noted that the ability of radar (C-band) to penetrate clouds and dense vegetation gives the SRTM DEM an advantage over optical observations. The superior performance of the SRTM DEM in comparison with the ASTER GDEM is in line with previous studies [8,21,32,33,[65][66][67].…”

“…Teillet et al [19] proposed the C-correction method, which considers the difference between bands under diffuse irradiation. As such, the C-correction method is a band-specific regression coefficient topographic correction method, which incorporates a modified cosine correction parameter, C. Based on the linear relationship between IL and reflectance data, the empirical constant (C) can be automatically calculated for each band of Landsat TM-5 and OLI-8 data (see Equation (4)) [8,26,46].…”

“…Additionally, the extent to which forest classifications could be improved if topographic correction is applied to individual source imagery (Landsat TM-5 and OLI-8) and combined with widely used advanced machine learning classifiers such as random forest (RF), is not yet known. Machine learning classifiers are of growing interest to many researchers because of their non-parametric nature, whilst also providing a way of estimating the importance of individual variables in the classification [4,8].…”

The Added Value of Stratified Topographic Correction of Multispectral Images

“…In order to overcome this limitation, some authors proposed automated image-classification approaches previous to the topographic correction [19,20] while some others decided to stratify by thresholding vegetation indices, such as the Normalized Difference Vegetation Index (NDVI) [16][17][18]27]. Land cover based STOC applications mostly used Minnaert based TOC methods (i.e., MIN or EMIN [17,22,24,30]), but some other methods (e.g., SE, SCS + C and CC methods) were also used, although less frequently [16,20].…”

Spectral features based tea garden extraction from digital orthophoto maps

“…A number of methods employed various indices for extraction of vegetation information from remote sensing data (Pena-Barragana et al 2004). Traditionally, NDVI has widely been used to distinguish vegetation and nonvegetation areas (Adhikari et al 2015). In addition, other vegetation indices have been proposed such as excessive green index (Qin 2014), greenness and brightness (Adhikari et al 2015), ratios between spectral bands (Pena-Barragana et al 2004) etc.…”