Effect of Land Use Types and Soil Macrofauna on Soil Aggregate Stability and Carbon Storage in the Central Highlands of Kenya

Abstract

Changes from natural forests and bush lands to various land use types has been shown to have great influence on soil structure and carbon storage. However, little is still known about how these changes affect soil aggregate stability and C storage over the long-term, especially in Africa. A study was conducted in central highlands of Kenya to evaluate the influence of four land use types on soil aggregation and aggregates C content, as affected by the abundance of earthworms and termites. The four land use types included: (1) an undisturbed natural bush land dotted with Lantana camara L. shrubs and Croton macrostachyus Hochst. ex Delile. trees, (2) over 50 years old grazed pasture dominated by a mixture of Bermuda grass (Cynodon dactylon (L.) Pers.) and Kikuyu grass (Pennisetum clandestinum Hochst. ex Chiov.), (3) over 80 years old coffee plantation with Coffea arabica L. cv. SL 28, and (4) a maize field that had been cultivated continuously for the last 3 years. Sampling for soil aggregates and soil macrofauna was done using monolith method to a depth of 0.3 m. Separation of stable soil aggregates was done through wet sieving process, where seven aggregate fractions were obtained; large macroaggregates (LM), small macroaggregates (SM), microaggregates (m), silt and clay (s+c), coarse particulate organic matter (cPOM), microaggregates-within-macroaggregates (mM) and silt and clay within macroaggregates (s+cM). The aggregate C content was analyzed using wet oxidation method. Soil macrofauna were handpicked from the monolith soil and preserved in 75% ethanol and identified to genera or species where possible, using morpho-anatomical keys and comparison with reference collections. It was hypothesized that soil aggregates and aggregate C would decrease with increasing level of soil disturbance, with the magnitude of these effects being reduced by the abundance of earthworms and termites. Generalized linear models (GLM) were used to test the effects of land use types on aggregate fractions and aggregate-associated C using R statistical...................................................................................