Spatial variability analysis and mapping of soil physical and chemical attributes in a salt-affected soil

Abstract

Knowledge of inherent spatial variability of soil physical and chemical properties is needed for more accurate site-specific management of soil nutrients. In this study we investigated the spatial variability of a wide range of soil physical and chemical properties including soil texture fractions (percentages of sand, silt, and clay denoted as Sand, Silt and Clay, respectively), soil water content (WC), bulk density (BD), gypsum, organic carbon (OC), electrical conductivity (EC), pH, Ca, Mg, Na, exchangeable sodium percentage (ESP), sodium absorption ratio (SAR), available phosphorous (AP), and available potassium (AK) in a saline-alkaline soil catena in Sistan Plain, southeast of Iran. Soil samples were collected from two depths (0–15 and 15–30 cm) on a nearly regular grid at 113 sites over an 85-ha agricultural field. Statistical analysis of soil properties showed that Na, Mg, Ca, WC, EC, ESP, and SAR have a large coefficient of variation (CV) (more than 50%) and BD and pH have a low CV (less than 15%) for both layers. The correlation among soil properties varies for two layers; while Silt, WC, EC, ESP, Na, and gypsum are statistically (p < 0.01 and p < 0.05) correlated with most of physical and chemical properties in topsoil, Sand, EC, and OC are the most dominant properties in subsoil. Geostatistical autocorrelation analysis of soil properties were examined based on the “range of spatial continuity” and “nugget to sill” ratio. Accordingly, AP and subsoil ESP have the lowest spatial correlation while texture fractions are the most auto-correlated variables in space. The spatial structure of soil properties followed either a spherical or an exponential model with a minimum correlation distance of 70 m for AP to almost 800 m for soil fractions. The results indicated that spatial continuity generally increases and decreases with depth for soil physical and chemical properties, respectively. The difference in spatial variability of soil properties could be attributed to internal factors (e.g., the forming processes of soil) as well as external factors (e.g., human activities). The maps of soil physical and quality parameters were generated using either kriging or inverse distance weighting methods depending on cross-validation results. In general, topsoil layer has a greater amount of EC, ESP, SAR, pH, Na, Ca, Mg, and OC than subsoil while Silt, WC, and gypsum were often higher in subsoil. OC maps showed that the whole area is relatively low in organic carbon, mainly due to hot and dry climate and windy conditions in Sistan. The maps of soil nutrients provide useful information for adapting an efficient and precision agricultural production management.