Estimating the Leaf Area Index of agricultural crops using multi-temporal dual-polarimetric TerraSAR-X data: A case study on North East Germany

Leaf area index (LAI) is one of the most important ndicators of agricultural variables because of its relation to biophysical and biochemical properties of agricultural crops. Variations in LAI can be related to changes in leaf scattering properties, and these variations in leaf scattering properties can lead to changes in canopy backscattering behaviour. The objective of this study was to explore the potential of estimating LAI using multi-temporal dual polarimetric TerraSAR-X data in three different agricultural field crops, including winter wheat (Triticum aestivum L.), barley (Hordeum vulgare L.), and canola (Brassica napus L.). The relationship between LAI and the scattering coefficient (σ0) in TerraSAR-X was explored using three different approaches, including univariate regression, i.e., simple linear and nonlinear regression, multivariate regression, i.e., stepwise regression, and a semi-empirical water cloudmodel(WCM). The multivariate stepwise regression showed is capability to retrieve the LAI without any external input data, such as soil moisture, based solely on the polarization channels, i.e., HH or VV, and polarization variables, e.g. HH/VV or HH+VV. However, unlike the WCM, the stepwise method is not applicable with just one polarization channel. The results indicate that the leaf area index (LAI) was significantly and consistently correlated with σ0 throughout the growth stages using the stepwise regression and WCM approaches, whereas simple linear and nonlinear regression yielded relatively poor results except with barley