Abstract:
The fragmentation of forests into smaller isolated patches is a threat to biodiversity conservation. Fragmentation reduces the forest area and increases forest edge effect. Reduction in forested area may lead to increased local species extinctions, while effects of increased forest edge alter species interactions by increasing the degree of interaction among edge and forest interior species. Intensification of land uses surrounding forest patches resulting from population increase exacerbate the edge effects thereby having significant impact on tree species diversity and the potential of the forest fragments to sequester carbon. Comparative studies were undertaken in five main forest fragments of varying sizes in Taita Hills, Kenya to examine the effects of forest fragmentation on forest cover change, tree species diversity and above-ground carbon stock. Belt transects proportional to the forest size were established along forest edge-interior gradient, and square plots (10x10m) established systematically for assessment of the forest structure, tree species richness and diversity, edaphic factors and above-ground carbon stock. Analysis of Landsat images of 1973, 1987, 2001, 2012 and 2016 provided information on forest cover change of the fragments between 1973 and 2016. Data collected from the field was analysed using GenStat version 16.0 and CANOCO for Windows version 4.5 while Landsat images were analysed using ArcGIS version 10.2. Statistical testing for significance was performed at 95% confidence interval. The results showed that the annual rate of deforestation in Taita Hills was 0.5% and the forest fragments lost 23.2% of their cover between 1973 and 2016. The forest area lost was higher in Vuria (43%) and Chawia (32.7%) which are non-gazetted forests managed by the County government of Taita Taveta but lower in Fururu (3.2%), Mbololo (13.7%) and Ngangao (16.8%) which are gazetted forests managed by Kenya Forest Service. There was no significant edge effect on tree species diversity in Chawia (p=0.564)), Fururu (p=0.689), Mbololo (p=0.369), Ngangao (p=0.576) and Vuria (p=0.338), although the species diversity was higher in the forest interior than in the intermediate forest and on the forest edge in all the five forest fragments. Carbon stocks were not significantly affected by the edge effect in Chawia (p=0.704), Fururu (p=0.908), Mbololo (p=0.544), Ngangao (p=0.991) and Vuria (p=0.590). There was a significant edge effect on the abundance and distribution of key dominant and adaptable species i.e. Macaranga conglomerata, Albizia gummifera, Syzygium guineense, Xymalos monospora, Tabernaemontana stapfiana and Maesa lanceolata (p=0.012). There was no significant edge effect on species richness in Chawia (p=0.490), Fururu (p=0.507), Mbololo (p=0.303), Ngangao (p=0.886) and Vuria (p=0.090). Large forest fragments had significantly higher species richness and diversity in all the habitats along the forest edge-interior gradient than small forest fragments (p=0.001) and (p=0.001) respectively. There was no significant effect of fragment size on carbon stock (p=0.634). There was significant edge effect on soil pH in Mbololo (p=0.001), Ngangao (p=0.001) and Vuria (p=0.042) but not in Chawia (p=0.700) and Fururu (p=0.667). There was significant edge effect on soil electrical conductivity in Fururu (p=0.048) but not in Chawia (p=0.896), Mbololo (p=0.367), Ngangao (p=0.555) and Vuria (p=0.861). The edge effect significantly affected soil nitrogen in Fururu (p=0.038) and potassium level in the soil in Mbololo (p=0.002) and Ngangao (p=0.035). The results show that forest fragmentation contributed to substantial loss in forest cover and hence reducing the capacity of the forest fragments to sequester carbon. Besides, the edge effect created by fragmentation contributed to reduction in floral species richness and diversity, and affected edaphic conditions. These results provide robust evidence for developing land use policies and appropriate management strategies for effective conservation of fragmented forest ecosystems.