Modelling Above Ground Biomass Using Sentinel 2 and Planet Scope Data in Dense Tropical Montane Forests of Tanzania

Abstract

Forest biomass estimation using field -based inventories at a large scale is challenging and generally entails large uncertainty in tropical regions. In this study, we investigated the performance of Sentinel 2 and Planet Scope data for above ground biomass (AGB) modelling, in the tropical rainforest of Tanzania. A total of 296 field inventory plots were measured across the west Usambara mountain forests. The results showed that, Sentinel 2-based model fitted using GLMs had better performance (cvRMSEr = 67.00 % and pseudo-R²= 20%) as compared to Planet Scope-based models (cvRMSEr = 72.1 % and pseudo-R²= 5.2%). Overall GLMs resulted into models with less prediction errors in contrast to random forest when using Sentinel 2 data. However, for the Planet Scope, there was marginal improvement when using random forest (cvRMSEr = 72.0%). Models that incorporated texture variables produced better prediction accuracy as compared to those with band values and indices only. The study has shown that, Sentinel 2 and Planet Scope remotely sensed data can be used to develop cost-effective method for AGB estimation in tropical rainforests of Tanzania.