Paul Borang and Koppad A.G.
Abstract
The study was carried out at Yellapur Forest Division, Karnataka, with 86.97% forest area, exploring aboveground carbon (AGC) variations across different forest types and their correlation with LiDAR metrics. Ten one-hectare permanent plots were established, with 3 in dry and moist deciduous each, and 4 in semievergreen forests. All trees with >30 cm girth at breast height were inventoried. AGC estimates were 308.28 mg ha-1 in semi-evergreen, 207.77 mg ha-1 in moist deciduous, and 117.88 mg ha-1 in dry deciduous forests. LiDAR data were acquired using a Nextcore Lumos XM120 UAV LiDAR system. In dry deciduous and moist deciduous forests, the 95th height percentile correlated strongly with observed AGC, yielding predicted AGC values of 118.56 mg ha-1 and 208.86 mg ha-1, with R2 values of 0.91 and 0.83, respectively. Conversely, in semi-evergreen forests, the 50th height percentile outperformed other metrics, resulting in a predicted AGC of 310.90 mg ha-1 (R2=0.57).