Estimation of Leaf Area Index of Hemiboreal Forests

ABSTRACT

ACRM, a two-layer homogeneous canopy reflectance model is applied for the estimation of leaf area index (LAI) of forests using Proba/CHRIS spectral signatures and model inversion. A regularity parameter is used in the model in order to account for the grouping of foliage into tree crowns. Results of LAI estimation at the Järvselja forested test site, in Estonia, are compared with ground truth readings available at the test site, and with LAI estimates by MODIS and MERIS/CYCLOPES LAI products for the test site. MODIS LAI estimates are systematically higher than MERIS estimates. The estimated LAI values from ACRM inversion using CHRIS data have wider range than MODIS and MERIS estimates for the test site because of smaller pixel size. In general, LAI values from CHRIS data are between MODIS and MERIS LAI values, however, systematically lower than allometric LAI of forest stands using forest inventory data and regressions suggested for the boreal zone in literature.

Keywords: