OPTIMIZING FORAGE-BASED COW-CALF OPERATIONS TO IMPROVE SUSTAINABILITY OF BEEF CATTLE AGRICULTURE AND WATER QUALITY PROTECTION AND MANAGEMENT
Title: Kriging analysis of soil properties: Implication to landscape management and productivity improvement
Submitted to: Journal of Soils and Sediments
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: April 5, 2008
Publication Date: June 2, 2008
Citation: Sigua, G.C., Hudnall, W.H. 2008. Kriging analysis of soil properties: Implication to landscape management and productivity improvement. Journal of Soils and Sediments. 8(3):193-202.
Interpretive Summary: One aim of soil science is to establish the cause and effect relationships between soil properties and soil behavior so that users of soil resources can predict the performance and behavior of soils. The prediction is made by matching the requirements of specific use to the characteristics of soils. The matching process however, is very complex because of the existing soil variability in the field. Soils with similar properties and environments are expected to behave similarly. Limiting the range of soil variables permits more accurate predictions of expected response to alternative soil management inputs and land use. The objectives of this study were to determine the spatial patterns and dependence of some selected physicochemical properties of soils and to assess the implication of dependence analysis to land management using 3-D kriging analysis in conjunction with geostatistical (GS+) model. Although the kriging analysis tool has been used by numerous scientists and engineers in mining and petroleum explorations, and environmental studies, few have applied this tool to the estimation of 3-D distribution of soil properties in the landscape which are affected by salt water. Results disclosed remarkable spatial dependence of some selected soil properties in the study area. The variability of individual properties as indicated by CV differed widely (p'0.05) which can be attributed to the interactive effect of local relief (micro-topography) and hydrologic patterns within the study area. Knowledge of the existing soil variability in the area can be utilized to develop a stochastic model that may describe the potential land use capabilities. The prediction is made by matching the requirements of specific land use to the characteristics of soils within the landscape. Results of an investigation of this type are of great interest to environmental scientist, water resource planners, regulators, decision makers, engineers, soil scientists, and resource managers.
Soil as a landscape entity contains wide ranges of physical, chemical, morphological, and mineralogical properties, both laterally and vertically. Soils with similar properties and environments are expected to behave similarly. Statement of land use potential depends in part on the precision and accuracy of the statements that can be made about the soils. This information has some practical applications in optimizing land management and productivity improvement. The spatial patterns and dependence of some selected physicochemical properties of brackish marsh and surrounding soils were investigated using a 3-D kriging analysis in conjunction with geostatistical (GS+) model. Surface and subsurface samples (profile sampling) were obtained across the 40 sampling sites. These samples were stored at 50C until the different physical (particle size) and chemical analyses were initiated. Soil chemical analyses included electrical conductivity (EC), Cl-1, pH, and water soluble Na, S, Ca, Mg, Fe, and Al by inductively coupled plasma spectroscopy. Soil extraction was performed via an automatic soil extractor. A 1:2 soil to extracting solution ratio was used for all the extractions. The concentration of Cl-1 in the soil sample was determined using a chloride electrode. Some selected physicochemical properties were interpolated by kriging. The procedures of univariate/multivariate analyses and kriging techniques were followed to evaluate the spatial dependence of some selected physicochemical properties of soils in the landscape. The kriging procedures included preliminary data analysis, structural data analysis, log kriging estimations, and image generations of spatial results. There was a significant spatial dependence and differences (p'0.01) for all soil parameters tested. The variability of individual soil property as indicated by coefficients of variation differed widely. Some soil properties in the lower horizons were more uniform than those at the surface because of less disturbances and effects of micro-topography. Soils in the area have developed markedly contrasting morphologies and properties that varied laterally (east-west) and vertically (north-south). Many of the differences and spatial dependence of soil properties in the study area that vary with topography are due to some combinations of microclimate, soil pedogenesis, geological surficial processes, and the sorting effects of water. Majority of the spatial variability of some soil properties within the study area were caused by the interactive effect of micro-topography and hydrologic pattern. Spatial variability affected soil performance. A uniform application of any soil amendments such as fertilizer or gypsum in the area that possessed spatially variable soil would result in over application in some parts of the area and under application in other areas.