SPATIAL PREDICTION OF HEAVY METAL POLLUTION FOR SOILS IN COIMBATORE, INDIA BASED ON ANN AND KRIGING MODEL

Abstract

The concentration of five soil heavy metals (Cr, Pb, and As) was measured in 121 sampling sites in Coimbatore, Tamil Nadu, India regions known as centres of pollution due to the chemical and metallurgical activities. The soil samples were collected from locations where the ground is not sliding and the probability of alluvial deposits is small. The concentration of heavy metals was measured by using Atomic Absorption spectrometer. Kriging and ANN techniques were used to develop the model to predict the constituents of the heavy metal in the soils. In some locations, the concentration for the investigated heavy metals exceeds the concentration admitted by the guideline. The highest concentration of lead (8.9 ppm) was found in Ukkadam Lake. The highest concentration of chromium was found in Ganapathi (3.6 ppm). The highest concentration of Arsenic (5.4 ppm) was found in Sidco Industrial Estate. The maximum admitted concentrations in the sensitive areas revealed to be exceed from five to twenty times.