Using Geostatistics for Spatial Analysis of Soil Moisture Content, Electrical Conductivity, and pH at Paddy Fields

Abstract

Soil is dynamic due to various internal and external processes exerted on the soil, resulting in unique soil characteristics in space in short and long distances. Geostatistics (kriging) is the method of quantifying the spatial variation of soil properties. This research was mainly aimed at applying geostatistics to quantify and interpolate the spatial dependence and structure of three soil properties, namely pH, EC, and Soil Moisture Content (SMC) in a small area. This research was conducted on paddy fields in Mlandingan Kulon Village, Situbondo Regency. Sampling was conducted on an area of   9.2 ha with 31 sample points. Normal data distribution was found for pH and EC, whereas this was not the case for SMC. The results of the analysis showed that most of the pH values   were alkaline (>8), EC values were non-saline (<2 mm/cm), and SMC was in the low category (<20%). The results show that for three soil properties, weak dependencies were observed. The values of Root Mean Square Error (RMSE)  confirmed that kriging with exponential was better compared to the spherical model, resulting in the RMSE of 0.546 (pH), 0.041 (EC), and 1.512 (SMC).