Forest Variable Estimations Using TanDEM-X Data in Hyrcanian Forests

Abstract

The objective of this study was to estimate forest variables using TanDEM-X interferometric synthetic aperture radar (InSAR) data acquired over the Shastkalate forest of Gorgan in northern Iran. Inventory variables, including diameter at breast height, tree height (Lorey’s mean tree height), basal area and volume, were collected from 112 circular sample plots with a size of 0.1 ha. Interferometric phase height and coherence were computed from TanDEM-X data. Stepwise multiple linear regression was used to develop models describing the relationship between field-derived forest variables and the InSAR based statistical metrics. The validation was carried out using the leave-one-out cross-validation method. The estimation accuracy results in terms of relative RMSE for Lorey’s mean tree height, basal area and volume were 11.0, 34.1, and 37.8%, respectively, whereas the corresponding figures for R2adj were 32.0, 10.0, and 16.0%, respectively. The results were good enough for estimating forest variables and it was concluded that TanDEM-X could be used in Hyrcanian hardwood forests.