How to determine the leaf area index (LAI) of forests: A comparison of forest inventory versus satellite-driven estimates

Leaf area index (LAI) is a key measure of forest stand physiology and biomass production, and is essential within ecosystem modeling. There are two common approaches to obtaining LAI: (ⅰ) terrestrial forest inventory-based “bottom–up”, and (ⅱ) satellite-based “top–down” techniques. The purpose of this study is to compare terrestrial LAI from allometric functions applied to more than 30,000 trees of the Austrian National Forest Inventory (NFI) vs. satellite-based LAI estimates obtained from moderate resolution imaging spectroradiometer (MODIS) and Sentinel (Sentinel-3 TOC reflectance and PROBA-V) data across Austrian forests. We analyzed a satellite pixel-to-plot aggregation and obtained the full inventory data set for the LAI comparison. The results suggest that terrestrial vs. satellite (MODIS and Sentinel) driven LAI estimates are consistent, but (ⅰ) the variation of the terrestrial forest inventory LAI is larger vs. the pixel average LAI from satellite data, and (ⅱ) any satellite LAI estimation needs a forest stand density correction if the crown competition factor (CCF), a measure for stand density, is ​< ​250 to avoid an overestimation in LAI.