Main Article Content

Relationship between carbon stocks and tree species diversity in a humid Guinean savanna landscape in northern Sierra Leone


Edward Amara
Janne Heiskanen
Ermias Aynekulu
Petri KE Pellikka

Abstract

Global sustainable development goals include reducing greenhouse gas emissions from land-use change and maintaining biodiversity. Many studies have examined carbon stocks and tree species diversity, but few have studied the humid Guinean savanna ecosystem. This study focuses on a humid savanna landscape in northern Sierra Leone, aiming to assess carbon stocks and tree species diversity and compare their relationships in different vegetation types. We surveyed 160 sample plots (0.1 ha) in the field for tree species, aboveground carbon (AGC) and soil organic carbon (SOC). In total, 90 tree species were identified in the field. Gmelina arborea, an exotic tree species common in the foothills of the Kuru Hills Forest Reserve, and Combretum glutinosum, Pterocarpus erinaceous and Terminaria glaucescens, which are typical savanna trees, were the most common species. At landscape level, the mean AGC stock was 29.4 Mg C ha−1 (SD 21.3) and mean topsoil (0–20 cm depth) SOC stock was 42.2 Mg C ha−1 (SD 20.6). Mean tree species richness and Shannon index per plot were 7 (SD 4) and 1.6 (SD 0.6), respectively. Forests and woodlands had significantly higher mean AGC and tree species richness than bushland, wooded grassland or cropland (p < 0.05). In the forest and bushland, a small number of large diameter trees covered a large portion of the total AGC stocks. Furthermore, a moderate linear correlation was observed between AGC and tree species richness (r = 0.475, p < 0.001) and AGC and Shannon index (r = 0.375, p < 0.05). The correlation between AGC and SOC was weak (r = 0.17, p < 0.05). The results emphasise the role of forests and woodlands and large diameter trees in retaining AGC stocks and tree species diversity in the savanna ecosystem.

Keywords: aboveground carbon, soil organic carbon, tree species


Journal Identifiers


eISSN: 2070-2639
print ISSN: 2070-2620