Litterfall and litter nutrient dynamics under cocoa ecosystems in lowland humid Ghana

Abstract

There have been few studies quantifying litterfall, standing litterstock and gross litter decomposition following forest conversion to plantation crops such as cocoa. Additionally, an assessment of changing processes occurring in forest floor litter systems with plantation age is lacking. We investigated litterfall production, standing litter changes and litter decomposition along a chronosequence of shaded cocoa farm fields (secondary forest, 3, 15 and 30-year-old) in the moist semi-deciduous forest belt in the Ashanti Region of Ghana in West Africa over 24 months. Mean annual litterfall production differed significantly among study sites and ranged from 5.0 to 10.4 Mg DM ha−1. Similarly, standing litter differed significantly between land-use /plot ages. The results showed significant differences in quality between litter from forest and litter from cocoa plantations. Litterfall from forests had higher concentrations of nitrogen and lower concentration of soluble polyphenols and lignin compared to litter from cocoa systems. Monthly decomposition coefficients (k) estimated as \( k = {{\left( {{\text{A}} - \left( {{\text{L}}_1 - {\text{L}}_0 } \right)} \right)} \mathord{\left/ {\vphantom {{\left( {{\text{A}} - \left( {{\text{L}}_1 - {\text{L}}_0 } \right)} \right)} {\left( {{{\left( {{\text{L}}_1 + {\text{L}}_0 } \right)} \mathord{\left/ {\vphantom {{\left( {{\text{L}}_1 + {\text{L}}_0 } \right)} 2}} \right. } 2}} \right)}}} \right. } {\left( {{{\left( {{\text{L}}_1 + {\text{L}}_0 } \right)} \mathord{\left/ {\vphantom {{\left( {{\text{L}}_1 + {\text{L}}_0 } \right)} 2}} \right. } 2}} \right)}} \), where A is litterfall production during the month, L0 is the standing litterstock at the beginning of the month and L1 is the standing litterstock at the end of the month. Annual decomposition coefficients (k L ) were similar in cocoa systems (0.221–0.227) but higher under secondary forests (0.354). Correlations between litter quality parameters and the decomposition coefficient showed nitrogen and lignin concentrations as well as ratios that include nitrogen are the best predictors of decomposition for the litters studied. Our results confirm the hypothesis that decomposition decreases following forest conversion to shaded cocoa systems because of litter quality changes and that decomposition rates correlate to litter quality differences between forest and cocoa ecosystems. The study also showed that standing litter pools and litterfall production in recently converted cocoa plantations are low compared to secondary forests or mature cocoa systems. Management strategies involving the introduction of upper canopy species during plantation development with corresponding replacement of tree mortality with diverse fast growing species will provide high quality and quantity litter resources.