Estimation of planted forest leaf area index from TM imagery using the algorithm based on geometric-optical model

Global leaf area index (LAI) products such as MODIS LAI product et al. have relatively low spatial resolution (250m–7km) and can not meet the needs of high spatial resolution remote sensing applications. Therefore, it is necessary to explore the feasibility of the algorithm based on physical model for LAI retrieval using high spatial resolution remote sensing imagery. This study utilized the algorithm based on Four-scale model to derive LAI in planted forest from TM imagery. A set of land cover type dependent relationships between LAI and Simple Ratio (SR) are provided for various solar and view anlges. Bidirectional reflectance distribution function (BRDF) and clumping representation at canopy scale are both considered in the algorithm. The empirical model using NDVI as predicted variable is also considered for LAI estimation. A validation study was conducted with in-situ measurements of LAI in planted forest from Zhangye, Gansu province. Better accuracy in LAI prediction was observed from the inversion algorithm based on Four-scale model (R
²
=0.67, RMSE=0.50) than that from NDVI (R
²
=0.59, RMSE=0.67) compared with measured LAI, especially when LAI > 2.00. Moreover, the sensitivity analysis of inversed LAI to bands reflectance was carried out. LAI was more sensitive to reflectance at red band (ρ
red
) than that at near infrared band (ρ
nir
), with uncertainty value of reflectance range from −10% to −30%. This study prove the effectiveness of the algorithm based on Four-scale model in LAI estimation from TM imagery in planted forest and will be helpful in further developing physical models for high spatial resolution LAI retrieval.