Spatial Variability of Metal Elements in Soils of a Waste Disposal Site in Khulna: A Geostatistical Study

Abstract

The aim of this study was to identify the spatial variability of various metal elements in soils and to compare the variability of the outputs of different geostatistical methods using the concentration of heavy metals in the ground of the waste disposal site at Rajbandh, Khulna, Bangladesh, to find the best-fitted prediction model. In this study, sixty soil samples were collected from the waste disposal site at a depth of 0–30 cm from the existing ground surface. The concentration of metal components like Al, As, Ba, Ca, Cd, Cr, Co, Cu, Fe, Hg, K, Mn, Na, Ni, Pb, Sb, Sc, Sr, Ti, V, and Zn in soils were measured through standard technique. ArcGIS 10.5 was used to prepare the spatial distribution of these metals to determine their extents in the soil. Four geostatistical methods named inverse distance weighting (IDW), universal kriging (UK), ordinary kriging (OK), and empirical Bayesian kriging (EBK) were performed to generate the spatial distribution of these elements in the soil. The interpolation technique was used in this regard to reduce the possibility of error prediction of results. Prediction maps prepared using interpolation techniques showed that most of the contaminated spots were near the dumping point. There was seen a significant amount of low concentration of metals in some spots. Moreover, cross-validation was conducted to determine the best-fitted model. The results indicated that EBK was a better method with more accurate predictions based on MAPE, RMSPE, and SRMSPE values than that of other counterparts.