Quantifying micro-environmental variation in tropical rainforest understory at landscape scale by combining airborne LiDAR scanning and a sensor network
Key message : We combined aerial LiDAR and ground sensors to map the spatial variation in micro-environmental variables of the tropical forest understory. We show that these metrics depend on forest type and proximity to canopy gaps. Our study has implications for the study of natural forest regeneration. Context: Light impacts seedling dynamics and animals, either directly or through their effect on air temperature and relative humidity. However, the micro-environment of tropical forest understories is heterogeneous. Aims: We explored whether aerial laser scanning (LiDAR) can describe short-scale micro-environmental variables. We also studied the determinants of their spatial and intra-annual variation. Methods: We used a small-footprint LiDAR coverage combined with data obtained from 47 environmental sensors monitoring continuously understory light, moisture and temperature during 1 year over the area. We developed and tested two models relating micro-environmental conditions to LiDAR metrics. Results: We found that a volume-based model predicts empirical light fluxes better than a model based on the proportion of the LiDAR signal reaching the ground. Understory field sensors measured an average daily light flux between 2.9 and 4.7% of full sunlight. Relative seasonal variation was comparable in the understory and in clearings. In canopy gaps, light flux was 4.3 times higher, maximal temperature 15% higher and minimal relative humidity 25% lower than in the forest understory. We found consistent micro-environmental differences among forest types. Conclusions: LiDAR coverage improves the fine-scale description of micro-environmental variables of tropical forest understories. This opens avenues for modelling the distribution and dynamics of animal and plant populations.