Resource information
Kakamega National Forest Reserve is a tropical forest ecosystem at high risk of irreplaceable biodiversity loss due to persistent human-induced pressures. The aim of this paper is to assess the effect of fragmentation and forest cover loss on forest ecosystems in Kakamega National Forest Reserve, with the objectives: (1) to quantify the forest cover loss and analyse fragmentation in the Kakamega forest ecosystem and (2) to analyse the effect of forest cover loss on the spatial distribution of the Kakamega forest ecosystem at different timescales. Hansen global forest change data was used as an input training dataset on the Google Earth Engine platform (GEE) to estimate the area of forest cover loss by aggregating the sum of pixel values, and to provide a time series visualization of forest change by the extent of cover loss using Sentinel-2 and Landsat 7 false colour composites (RBG) in QGIS software. Fragmentation analysis was performed using reclassified forest loss and distribution data from the Hansen product as binary raster input in Guidos software. Total forest cover loss over 20 years was estimated at 826.60 ha. The first decade (2000–2010) accounted for 146.31 ha of forest cover loss, and the second decade (2010–2020) accounted for 680.29 ha of forest cover loss. Forest area density (FAD) analysis depicted an increase in the dominant layer by 8.5% and a 2.5% decrease in the interior layer. Morphological spatial pattern analysis (MSPA) illustrated a change in the core layer of 96% and a 14% increase in the openings class layer. Therefore, this study demonstrates that forest cover loss and landscape pattern alteration changed the dynamics of species interaction within ecological communities. Fragmented habitats adversely affected the ecosystem’s ability to recover the loss of endemic species, which are at risk of extinction in the backdrop of climate change. Anthropogenic drivers i.e., the clearing of natural forest and conversion of forest land for non-forest use, have contributed significantly to the loss of forest cover in the study area.
