Spatial Enhancement of Modis Leaf Area Index Using Regression Analysis with Landsat Vegetation Index

Abstract

The Leaf Area Index (LAI) is an important indicator of vegetation development which can be used as an input parameter in hydrological and biochemical models (e.g. crop models for yield prediction and forecast) and is, thus, relevant information to monitor food production and to feed an early warning system for famine crisis. Satellite LAI data is available on a regular basis (high temporal resolution) with maps at regional or global scales (low spatial resolution). This study aimed at enhancing the spatial resolution of the MODIS LAI product to bring it to the Landsat resolution. The proposed method was applied in four sites with different climate and vegetation conditions. Regression analysis between MODIS EVI (Enhanced Vegetation Index) and LAI data was applied across time and the estimated regression equations were input in a downscaling model using Landsat EVI images and land cover maps. Comparison between the downscaled LAI values and LAI field measurements showed high correlation, with correlation coefficient values ranging from moderate (0.5–0.7 in two cases) to high (0.7–0.96 in five cases). The results show that it is possible to use this methodology to reliably estimate LAI at a 30m spatial resolution across various climates and ecosystems, thus supporting a food security early warning system.