The role of soil spatial variability investigation in the management of the Chad Basin vertisols of N.E. Nigeria

Abstract

The Chad Basin of North-Eastern Nigeria is endowed with a vast expanse of vertisols that are not only rich in plant nutrient elements but also have a high water-holding capacity. These soils therefore are capable of supporting a large-scale production of food and forage crops alike. The ability of the soils to support large scale cultivation of forage crops is particularly significant in view of the fact that the North-East is the leading producer of livestock in Nigeria. Prediction of soil behaviour and crop productivity are conventionally based on soil test results. The reliability of soil test results on which management decisions are based cannot be greater than the reliability of the soil samples from which the results are obtained. The samples collected from the field must therefore be representative of that field. A procedure for obtaining representative soil samples, based on an adequate assessment of the soil spatial variability, is therefore presented. Surface soil samples (0–15 cm) were collected from 100 locations spaced at regular intervals of 2 m within a transect sited in the middle of a 2 ha field in the Chad Basin of North-Eastern Nigeria. The samples were analysed for exchangeable cations, saturation extract electrical conductivity (EC e ), particle size distribution, bulk density, organic carbon and pH. All the soil properties investigated were found to be normally distributed. Estimated sample autocorrelograms suggested that the observations of Exchangeable Na + , Exchangeable sodium percentage (ESP), EC e , Bulk density and silt content were spatially independent. Exchangeable Potassium and Magnesium, pH, percent sand and percent clay exhibited fairly strong spatial dependence while exchangeable calcium and organic carbon exhibited weak spatial dependence. The implications of the spatial dependence or independence of these soil properties in selecting appropriate soil sampling plans are discussed. Sample size calculations revealed that while it requires only one sample to estimate the mean values of clay content and pH within ±10% of the population means of these properties at the 95% confidence level, 696 samples would be required to estimate the mean value of exchangeable magnesium for the study area at the same precision level. The implications of these estimates of sample number requirements in obtaining reliable soil test results for improved management and enhanced productivity of these vertisols are discussed.