Forest ecosystem on the edge: Mapping forest fragmentation susceptibility in Tuchola Forest, Poland

Abstract

Forest ecosystems, vital for maintaining global biodiversity and ecological balance, are increasingly threatened by fragmentation. This study addresses the critical issue in the Tuchola Forest of Poland, examining the effects of natural and human factors on forest fragmentation. Our objective was to identify the most suitable dataset for monitoring forest fragmentation from 2015 to 2020, ascertain the primary drivers of fragmentation, and map the areas at high risk. Utilizing the PALSAR (25 m resolution) and Dynamic World (10 m resolution) datasets, we discovered PALSAR's enhanced ability to detect changes in forest structure, particularly evident after a significant windstorm in 2017. This dataset proved crucial in highlighting the escalating trend of forest fragmentation, reinforcing its importance for environmental monitoring and policy formulation. Our analysis identified key factors influencing fragmentation, such as proximity to croplands, tree height and age, wind speed, and vegetation water content, with areas near croplands and having younger, shorter trees being most susceptible. Employing a Weight-of-Evidence (WOE) Bayesian modeling technique, we mapped forest fragmentation susceptibility, demonstrating our methodology's effectiveness through high accuracy validation (AUC of 0.82 and Kappa Index of 0.68). Our innovative approach in mapping susceptibility to fragmentation, especially after extreme weather events, marks a pioneering contribution in Poland. This research advances the understanding of forest fragmentation dynamics and offers a scalable model for global application, emphasizing the urgent need for targeted conservation strategies to preserve the integrity of forest ecosystems amidst climatic risk and anthropogenic pressures.