Applying LiDAR to Quantify the Plant Area Index Along a Successional Gradient in a Tropical Forest of Thailand

Abstract

Long-term monitoring of vegetation is critical for understanding the dynamics of forest ecosystems, especially in Southeast Asia’s tropical forests, which play a significant role in the global carbon cycle and have continually been converted into various stages of secondary forests. In Thailand, long-term monitoring of forest dynamics during the successional process is limited to plot scales assuming from the distinct structure of successional stages. Our study highlights the potential of coupling airborne light detection and ranging (LiDAR) technology and stand age data derived from Landsat time-series to track back forest succession, and infer patterns in the plant area index (PAI) recovery. Here, using LIDAR data, we estimated the PAI of the 510 sample plots of a seasonal evergreen forest dispersed over the study area in Khao Yai National Park, Thailand, capturing a successional gradient of tropical secondary forests. The sample plots age was derived from the available Landsat time-series dataset (1972–2017). We developed a PAI recovery model during the first 42 years of the succession process. We investigated the relationship between the model residuals and PAI values with topographic factors, such as elevation, slope, and topographic wetness index. The results show that the PAI increased non-linearly (pseudo-R2 of 0.56) during the first 42 years of forest succession, and all three topographic factors have less influence on PAI variability. These results provide valuable information of the spatio-temporal PAI patterns during the successional process and help understand the dynamics of tropical secondary forests in Khao Yai National Park, Thailand. Such information is essential for forest management and local, regional, and global PAI synthesis. Moreover, our results provide significant information for ground-based spatial sampling strategies to enable more accurate PAI measurements.