Soil carbon pools based on physical, chemical and biological fractionation and their relationship to fertility of humic nitisols in the central kenya highlands

Abstract

Based on a survey conducted among 198 households in the Central Kenyan Highlands in which 33% of the farmers identified soil conditions as a leading constraint to crop productivity, 12 households in Kiambu District were randomly selected for more detailed investigation of soil management practices and analytical comparison of soils identified by farmers as being either productive or non-productive. Soil texture analysis was conducted and the parameter was used as a test of physical similarity, after which chemical analyses were conducted on soils with similar texture. The criteria used by fanners to distinguish productive soils versus non-productive soils were crop performance (100% of the fanners), soil tilth (50% of the farmers), soil moisture retention (50% of the farmers), soil colour (42% of the farmers) and presence of certain indicator animal and plant species (25% of the fanners). The fanners attributed the low fertility status of some patches of their farms to inadequate use of organic and inorganic fertilisers (100% of the fanners), removal of crop residues (100% of the farmers) continuous cropping (83% of the farmers), lack of crop rotation (66% of the farmers) and soil erosion (42~"O of the farmers). Sand and clay contents were not significantly different between the two soil categories but productive soils had significantly higher silt content than nonproductive soils (P < 0.001). Productive soils had significantly higher soil pH, ECEC. exchangeable cations, extractable phosphorus, total kjeldahl N, organic C and P (P<O.OOl) than non-productive soils. Soil microbial biomass C and Nand net N mineralisation (AMN) were significantly higher in productive soils than in non-productive soils (P < 0.01). Diurnal soil respiration (DSR) was significantly greater in productive soils than in non-productive soils (P < 0.05) and was positively correlated with net N mineralisation (r = 0.45) and microbial biomass nitrogen (r = 0.67). From these chemical analyses, productive soils had higher organic matter content and soil nutrients than non-productive soils. Nutrient depletion in the non-productive soils was attributed to soil organic matter decline. Productive soils had significantly greater particulate organic matter (53-200 urn) and light traction organic matter (> 1.37 g cm') C, Nand P contents than nonproductive soils (P < 0.05). A mild chemical oxidation procedure with potassium pennanganate (KMn04) proposed as a test of soil organic matter (SOM) quality was compared to more conventional SOM physical fractionation approaches. Potassium pennanganate oxidisable C for productive soils was also significantly greater than that for nonproductive soils (P < 0.001) and correlated highly with particulate organic matter (r = 0.69), light SOM fractions (r = 0.48 to 0.61), microbial biomass C (r = 0.46) and net N mineralisation (r = 0.50) showing that the KMn04 oxidation technique is a reliable and rapid technique and could be used for assessment of soil fertility status. Correlation of total soil C, Nand P with other soil quality indicators such as AMN and DSR were higher than correlation of SOM fractions with total soil nutrients, AMN and DSR, suggesting that detailed C fractionation procedures are not necessary for routine soil quality diagnosis of humic Nitisols. An index developed to measure the rate of change in the soil C dynamics of a system relative to a relatively more stable system, the Carbon Management Index (CMI) showed that productive soils had higher CMIs than non-productive soils (P < 0.001). Fanners' perceptions of soil quality were substantiated through soil chemical analyses and SOM fractions provided qualitative information on these differences. The study shows that fanners have knowledge of their soils, they are aware of the causes of the low soil fertility and in some cases, they had possible solutions to soil fertility depletion. However, application of these solutions is limited by social and economic constraints. Future research on soil fertility management should consider indigenous knowledge and fanners' practices as a starting point for any new interventions.