Implications Of Conservation Agroforestry On Carbon Sequestration, Soil Fertility And Costs In The Drylands Of Machakos County, Kenya

ABSTRACT

Kenya’s drylands constitutes about 80% of the country’s size. These drylands are prone to relatively high vegetation and general environmental degradation; translating to changes in soil physical and chemical properties. Conservation agriculture with trees presents an opportunity to reduce vegetation and soil degradation thereby enhancing soil characteristics and carbon sequestration, which is an important component of redressing dryland problems associated with greenhouse gases and climate change. This study was set to assess the potential of conservation agriculture with trees (CAWT) in terms of carbon sequestration under dryland agroecosystem context. The study was carried out as part of ongoing experimentation established in short rain(SR) season of 2012 by the World Agroforestry Centre in a trial site at the Agricultural Training Centre (ATC) in Machakos county, Kenya. Specifically, the study assessed (i) the soil physical and chemical properties under conservation and conventional agriculture systems with and without trees, (ii) the amount carbon sequestered under conservation and conventional agriculture systems both with and without trees for both above and below ground biomass, and (iii) the cost effectiveness of conservation agriculture with trees and the potential thereof to earn carbon credits. The trials adopted a randomized block design with two farming systems (conventional and conservation agriculture) as the main blocks, 10 treatments and there replicates, summing to a total of 60 plots. In the fields, two shrub species (Calliandra calothyrsus Meissn. and Gliricidia sepium Jacq.) and Cajanus cajan were planted in three different spacing (1.5x1 m, 3x1 m, 4.5x1 m) for maize-legume intercrops. Soil samples were taken from 0-30 cm depth and analyzed for selected physical and chemical characteristics using standard laboratory methods. As a measure of biomass determination, trees were harvested and sampled for wood and leaves. The resulting biomass was converted to carbon content. Analysis of variance was carried out using Genstat version 14 and means separated using LSD at p