Spatial Variability of Soil Fertility and Microbial Properties in Kenyan Natural Pastures
Caroline Chepkoech *
Department of Crop and Soil Science, School of Agriculture and Biotechnology, University of Eldoret, P.O Box 1125-30100, Eldoret, Kenya.
Nicholas Rop
Department of Crop and Soil Science, School of Agriculture and Biotechnology, University of Eldoret, P.O Box 1125-30100, Eldoret, Kenya.
*Author to whom correspondence should be addressed.
Abstract
Soil fertility decline and spatial heterogeneity remain major constraints to sustainable pasture productivity in smallholder agroecosystems, particularly in sub-Saharan Africa, where management practices and agroecological variability strongly influence soil health outcomes. Despite the ecological importance of natural pasture systems in supporting livestock production, limited information exists on how soil physicochemical and microbial properties vary across agroecological gradients and farmer management typologies. This study therefore assessed spatial patterns of soil fertility in natural pasture ecosystems across four agroecological zones (UM3, UM4, LH1, and LH2) in Kenya, comparing pioneer (P; profit-making) and control (C; non-profit-making) farms. Soil samples were collected from 58 smallholder households (29 pioneer and 29 control farms) and analysed for key physicochemical properties, including pH, texture, total carbon, total nitrogen, phosphorus, nitrate, ammonium, cation exchange capacity, and soil moisture, alongside microbial biomass carbon and nitrogen. Heatmap visualisation and hierarchical clustering were used to evaluate spatial variability and relationships among soil properties. The results revealed pronounced spatial heterogeneity in soil fertility indicators, with clear clustering patterns among both sampling sites and soil variables. Pioneer-managed farms in UM4 exhibited higher total carbon and nitrogen concentrations, indicating enhanced organic matter accumulation under improved pasture management. However, these soils showed reduced microbial biomass carbon, microbial biomass nitrogen, and available phosphorus, likely associated with sandy textures and increased nutrient leaching. In contrast, UM4 control farms displayed higher microbial biomass and nitrate levels, reflecting intensified microbial activity and nutrient cycling under favourable moisture conditions. In the UM3 zone, pioneer farms were characterised by improved soil structural and chemical fertility, including higher clay content and optimal pH, while control farms exhibited elevated cation exchange capacity and ammonium concentrations, indicating enhanced nutrient retention. In the lower highlands, LH2 pioneer farms showed higher phosphorus availability, whereas LH1 control farms recorded increased nitrate and phosphorus concentrations, suggesting active mineralisation and nitrification processes. Hierarchical clustering grouped soil properties into biological, chemical, and textural fertility clusters, highlighting strong functional linkages among soil indicators. The study demonstrates that agroecological gradients and farmer management systems jointly regulate soil fertility dynamics in natural pasture ecosystems by influencing biological activity, nutrient availability, and soil structural properties.
Keywords: Soil fertility, spatial variability, natural pastures, agroecological zones, microbial biomass carbon, microbial biomass nitrogen, soil physicochemical properties, heatmap analysis, hierarchical clustering, pasture management