Estimation of aboveground biomass and carbon sequestration in a cocoa agroforestry system using UAV-LiDAR in northwestern Peru

Abstract

Accurate estimation of biomass and carbon in agroforestry systems is essential to assess their contribution to climate change mitigation and to improve their management. In this context, UAV-mounted LiDAR technology emerges as a fast, accurate, and non-destructive alternative for the structural characterization of cocoa agroforestry systems. This study aimed to estimate and analyze structural parameters, mainly tree height and diameter at breast height (DBH), as well as to calculate aboveground biomass and carbon sequestration in a cocoa agroforestry system, using LiDAR data obtained with a DJI Matrice 350 RTK UAV equipped with a Zenmuse L2 sensor, complemented by automatic tree segmentation in LiDAR360 and the application of species-specific allometric equations. The results showed a 93 % segmentation efficiency, with accuracies of 0.93 and 0.99 for DBH and height estimations, respectively. The evaluated plot, located at the Yanayacu Experimental Center (Jaén, Peru) and covering an area of 0.58 ha, had stored 15,492.5 kg of aboveground biomass and 7746.25 kg of aboveground carbon, with Mangifera indica and Cocos nucifera contributing more than 80 %. Consequently, this approach demonstrates the potential of UAV-based LiDAR to generate accurate and detailed information on system structure, enabling optimized management of high-biomass species and the development of more efficient and sustainable management strategies.