Conversion of Natural Forest to Cocoa Agroforest in Lowland Humid Ghana: Impact on Plant Biomass Production, Organic Carbon and Nutrient Dynamics

Thumbnail Image
Journal Title
Journal ISSN
Volume Title
This study was conducted to assess the effects of forest conversion to shaded-cocoa system on plant biomass, nutrient fluxes and soil physico-chemical properties along a chronosequence (forest, 3, 15 and 30-year-old cocoa farms) in the Moist Semi-deciduous Forest Zone of the Ashanti Region, Ghana. It also explored farmer indigenous knowledge and perceptions of soils and soil fertility dynamic processes. Plant biomass and above-ground organic carbon and nutrient pools significantly declined following changes in land-use compared to soil pools. Tree biomass constituted the largest pool ranging from 12.7 ±1.6 Mg ha-1 for the 3-year-old cocoa system to 209.3±33.3 Mg ha-1 in the forest. Soil Organic Carbon (SOC) in 0-60 cm soil depth did not change significantly over a 30-year period and ranged from 49.0±2.3 to 67.4±1.1 Mg C ha-1 in 3 year-old shaded cocoa system and forest respectively. SOC significantly declined only in the top (0-10 cm) soil at 3 years after conversion but recovered at 15 years. Thirty-year-old shaded-cocoa systems yielded up to 151 Mg C ha-1 primarily stored in established trees (both cocoa and shade trees) and soil pools. Total N declined only in the 10-20 cm soil depth in 3 and 15 year-old treatments but remained stable in all other soil depths across the chronosequence while available P stocks declined significantly. Soil exchangeable Ca, K and Mg stocks remained relatively stable with a tendency to improve, and cation exchange capacity (CEC) and base saturation increased more or less along the chronosequence. Soil bulk density (gm cm-3) increased significantly with increasing age of plantation only for the top 0-10 cm soil layer but did not differ among sites for similar depths. Despite the apparent stability of soil C stocks and nutrients (0-60 cm) along the chronosequence, soil quality declined under cocoa land-use at 3 years. Microbial biomass demonstrated a strong seasonal variation. However, conversion of forest did not result in a significant decline in microbial biomass. Mean annual litterfall and stand litterstocks differed significantly among land-uses. Litterfall ranged from 5.0 Mg ha-1 in 3-year-old cocoa to 10.4 Mg ha-1 forest systems while stand litterstocks were from 3.6 to 5.9 Mg ha-1 in 3 and 15-year-old farms respectively. Annual decomposition coefficients (kL) were similar in cocoa systems (0.221-0.227) but greater under forests (0.354). Estimated nutrient inputs from litterfall was 4 to 165 kg ha-1yr-1 of P and Ca respectively in 15-year-old and forest plots respectively. Turnover of fine roots was 3,591, 1,427, 2,466 and 4,066 kg ha-1yr-1 for forest, 3, 15 and 30-year-old plots respectively. Nutrient inputs through turnover of fine roots were estimated to be 16-31 kg N ha–1year–1, 2 -5 kg P ha–1year–1, 9-36 kg K ha–1 year–1, 18-47 kg Ca ha–1year–1 and 3-25 kg Mg ha–1year–1 across the chronosequence. There were significant differences in incident rainfall, throughfall and stemflow chemistry. Mean annual inputs of nutrients fluxes in incident rainfall were 5.7 kg N, 0.14 kg P, 13.6 kg K, 9.43 kg Ca and 5.6 kg Mg ha-1 yr-1. Rainfall loading or net canopy exchange was negative for total N at all sites while concentrations of P and the basic cations increased in throughfall relative to incident rainfall. Throughfall on average constituted about 95% of the total solute inputs of rainfall origin to forest floor. The mean N and P input-output balances were negative showing the system’s ‘no external input’ character. Farmers in the study had a well-developed knowledge system of their soils and related fertility processes. They derived their knowledge from observable plant and soil characteristics namely; soil color, crop yield, water retention capacity, difficulty to work soil, type and abundance of indicator weeds, leaf color or deficiency symptoms observed on crops and presence and abundance of soil macro-fauna. The qualitative perceptions of farmers matched scientific assessment of fertile or infertile soils. The results suggest the integration of local and scientific knowledge to facilitate the processes for formulating policies and development plans for agriculture truly participatory, gender sensitive and collaborative approaches. Enhancement farmers’ capability to adopt improved farm management and land preparation methods is required to conserve the soil and sustain long-term productivity.
A Thesis submitted to the Department of Agroforestry, Faculty of Renewable Natural Resources, Kwame Nkrumah University of Science and Technology in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY IN AGROFORESTRY
Litterfall, stand litterstocks, forest conversion, litter quality, nutrient fluxes, indigenous soil knowledge