Riccardi, TulliaD’Ammando, GiacomoWilson, LucyAttorre, FabioDavies, AndrewFarcomeni, AlessioMacFadyen, SandraStrydom, TerciaSmit, Izak P.J.Malatesta, Luca2026-03-032026-03-032025-11Riccardi, Tullia, Giacomo D’Ammando, Lucy Wilson, Fabio Attorre, Andrew Davies, Alessio Farcomeni, Sandra MacFadyen, Tercia Strydom, Izak P. J. Smit, and Luca Malatesta. 2025. “Investigating the Patterns of Tree Cover and Density in Relation to Abiotic and Biotic Factors in Kruger National Park.” Ecosphere 16(11): e70421. https://doi.org/10.1002/ecs2.70421.2150-8925 (online)10.5281/zenodo.1069117010.1002/ecs2.70421http://hdl.handle.net/2263/108707Across savanna ecosystems worldwide, the decline of large trees and the rapid expansion of shrubs present major conservation challenges. These trends are especially pronounced in South Africa's Kruger National Park (KNP), the country's largest protected area. To quantify their extent and identify their drivers, we conducted a spatial assessment of tree cover and density across KNP from 2011 until 2022. We then evaluated how these response variables are influenced by abiotic factors, including fire, climate, soil, and geology, and by biotic factors, such as the densities of African elephant adult male bulls and herds, including females and calves. We defined trees as land-cover elements that cast a distinct shadow and stand taller than 5 m. Using Collect Earth, an open-source software for augmented visual interpretation of high-resolution satellite imagery, we assessed tree cover and density on 4258 plots of 0.5 ha each. We recorded 27,918 trees, equivalent to an average density of 13 trees/ha. Counts in each plot were truncated to a maximum of 30 individuals. We validated our estimates of tree cover and height against independent, high-resolution airborne LiDAR measurements, which yielded an RMSE of 8.89% for trees taller than 3 m. The relative influence of selected predictors on tree cover and density was analyzed through logistic and survival regressions. Geology had the greatest influence on tree distribution, where both tree cover and density were higher on nutrient-poor granitic substrates than on nutrient-rich basalts. Tree cover and density were higher in areas with low fire frequency, close to main rivers, and with higher sand content in the soil. The mean annual rainfall showed a positive correlation with tree cover, while it had a negative correlation with the number of trees. Elephant bulls were found to be negatively correlated with both tree cover and density. In contrast, elephant herds exhibited a positive correlation with tree cover and density. This study highlights the importance of understanding the effects of multiple factors on tree distribution and aims to provide a baseline for assessing tree cover and density across KNP to support ongoing tree management strategies and contribute to future conservation priorities.en© 2025 The Author(s). This is an open access article under the terms of the Creative Commons Attribution License.African elephant (Loxodonta africana)Collect EarthDrylandsKruger National Park (KNP)Protected areasSavanna ecosystemSouth Africa (SA)Kruger National Park, South AfricaInvestigating the patterns of tree cover and density in relation to abiotic and biotic factors in Kruger National ParkArticle