Pan-tropical Analysis of the Temporal Relationship of Forest Disturbances and Precipitation

By Angeliki Chrysanthou

Abstract
Analysis of the relationship between forest disturbance patterns and precipitation patterns is important for understanding the dynamics of the pan-tropical forest under changing climate conditions and different kinds of human activities. In tropical regions, where persistent cloud coverage often limits the use of optical remote sensing products, making use of radar-based forest disturbance datasets enables a more efficient way to examine the intra-annual relationships between forest disturbance and precipitation with spatial and temporal precision. This study conducted cross-correlation analysis utilizing monthly RAdar for Detecting Deforestation (RADD) (Sentinel-1) and Climate Hazards Group InfraRed Precipitation with Station data (CHIRPS) time series from 2020 to 2023 across the tropical regions in Central America, South America, Africa, and Southeast Asia, using a 0.5° resolution. To access the intra-annual relationship between precipitation and forest disturbances (PDR), the strength of the correlation and the time lag was used. Furthermore, accessibility proxies were further used to study their influence in the studied relationship. Results revealed that the majority of the studied areas were dominated from significant negative correlation, ranging between -0.289 > cc ⩽ -0.7. The significant negative correlation appeared to be on average closer to cities for all regions with overall smaller variations in travel time to cities in comparison with the significant positive and non-significant grid cells. The magnitude of the negative correlation shows a uniform pattern in relation to travel time to settlements. This pattern is crucial for understanding how travel time impacts access to resources, particularly in rural areas. Few areas showed significant positive correlation, which was mainly a result of natural causes, such as flooding in case of Colombia in South America. The findings provide new insights in the influence between forest disturbance, precipitation, and accessibility, underscoring the necessity to support forest conservation and management during climate change that directly affects the pan-tropics.

Keywords: Forest Disturbance; Precipitation; Accessibility; Pan-tropics; Cross-correlation; Sentinel-1