Boreal forest leaf area index from optical satellite images: model simulations and empirical analyses using data from central Finland

A key variable controlling the global carbon, water and energy cycles is the leaf area index (LAI), which thus has an important role in current biosphere-atmosphere research. Several international networks promote the development of remote sensing techniques for global assessment of LAI. We present results from the VALERI (VAlidation of Land European Remote sensing Instruments) network’s test site at Hirsikangas in central Finland which represents one of the very few boreal forests in the global networks. In this study, three years of LAI data from Hirsikangas together with medium-resolution satellite images (SPOT HRVIR, Landsat 5 TM and IRS) over the site were used (1) to study the relationships between LAI and different spectral vegetation indices (SVIs), and (2) to test the performance of a physically-based forest reflectance model (PARAS). Simulations with PARAS showed good agreement with empirical data. Results indicated that inclusion of the SWIR band improves the performance of SVIs used for LAI estimation in the boreal forests.