Allometric equations for estimating biomass and carbon stocks of on-farm bamboo species in the agricultural landscapes of Kenya

Abstract

The accurate estimation of above-ground biomass is critical for characterizing ecosystem function and accounting for carbon stocks. Cost-effective estimation of carbon stocks is required to understand the role of bamboo in climate change mitigation and its potential in carbon off-set projects. The information can help smallholder farmers access carbon market benefits. Despite massive planting of bamboo species in Kenya’s agricultural landscapes, allometric equations to estimate the potential biomass and carbon is not available. This study developed species-specific and multiple species allometric equations for estimating biomass and carbon for major bamboo species within the agricultural landscapes of Kenya. The study was done in selected farms covering different agro-ecological zones where bamboo has been planted widely. One hundred and thirteen bamboo culms were randomly selected for destructive sampling. The sampled culms were harvested and culm length, fresh weights of the stem, branches, and leaves measured. The sub-samples of each component were dried and the dry-to-green weight ratio used to estimate above-ground biomass (AGB) and carbon (AGC). We developed species-specific and pooled (multiple species) allometric equations by regressing DBH, height and wood density variables against AGB using five non-linear functions. We used 70 % and 30 % on development and validation of the models, respectively. Our findings indicated a combination of DBH and H achieved the best performance by having a high coefficient of determination (Adj. R²) and a low Akaike information criterion (AIC). The addition of wood density did not improve our models. The estimated AGB ranged from 55.2 ± 20.6 t ha⁻¹ to 79.9 ± 18.4 t ha⁻¹ while AGC was from 27.6 ± 10.3 t ha⁻¹ to 40.0 ± 9.2 t ha⁻¹. The developed species-specific and multiple species allometric equations will improve estimates of future carbon stocks.