Search for potential iron contamination zones in Burdwan district: an approach through fuzzy logic

摘要

Burdwan district is one of the iron affected areas in West Bengal. In the present research work, various controlling factors, i.e., geology, soil, rainfall, land use land cover are considered to delineate the potential iron contamination zones (PICZ) within Burdwan district. Assigning weights and ranks to various input factors and their sub-classes, respectively, are carried out in the ArcGIS 10.1 platform on the basis of Fuzzy logic technique. The PICZ, derived from modeling, is broadly classified into two classes, i.e., 'suitable' and 'unsuitable' where, 33% (2313 km~2) of the total area falls under unsuitable PICZ, indicating that the groundwater in these regions are unfit for drinking purpose due to iron contamination. The model output is further validated with three different kind of approaches, i.e., superimposition, success and prediction rate curve and ground truth verification. A total of 4727 reported points of iron contamination are used for superimposition technique where, an overall accuracy of 87% is obtained for the model prediction. Success rate curve (SRC) and prediction rate curve (PRC) shows an accuracy of 89% and 83%, respectively. Collection and analysis of water samples from the unsuitable PICZ for ground truth verification shows an accuracy of about 80%. The GIS technique has provided an efficient environment for analyses of handling large spatial data. The outcome of the research work will be helpful in better planning and management of groundwater resources.