Groundwater Vulnerability Assessment of Lahore City based on Modified DRASTIC Model in GIS

Abstract

Water quality is considered as a major issue in the mega cities of developing countries. The city of Lahore has over 10 million population with the highest population density in the Punjab Province, Pakistan. The objectives of the study were to a) assess the temporal variations in physicochemical groundwater parameters and arsenic in the study area and b) create groundwater pollution vulnerability map of the study area using a modified DRASTIC model. The water samples (n = 73) were collected from tubewells and tested for arsenic and physicochemical parameters (alkalinity, calcium, pH, total dissolved solids, turbidity, hardness, and chlorides). Similar physicochemical water quality parameter data were also obtained from Water and Sanitation Agency (WASA) for the year 2012. The data exploration was performed using SPSS software. ArcGIS spatial analyst, geostatistical analyst, Getis-Ord Gi* statistics and model builder were used for mapping the water quality and analysis of the modified DRASTIC model. Various interpolation methods (radial basis functions, inverse distance weighting and kriging) were used for mapping the water quality to come up with the best groundwater quality map. Comparison of different interpolation methods based on the lowest root mean square error (RMSE) and mean absolute error (MAE) closer to zero showed that kriging produced better results than the other interpolation methods. The spatial distribution map of groundwater arsenic of 2015 showed that the arsenic concentrations above WHO guideline (10 μg/L) exist almost in the entire study area, whereas, the concentrations are above the national standards for drinking water quality of Pakistan (50 μg/L) in some areas of Ravi Town (Farrukhabad/Shahdra). Getis-Ord Gi* statistics revealed a significant (p < 0.05) hotspot of arsenic in the vicinity of River Ravi. While in the eastern parts of the study area, the arsenic concentrations had a decreasing trend. viii The modified DRASTIC model revealed that 51.38%, 36.77%, 9.48% and 2.37% area could be categorized in terms of vulnerability as low, moderate, high and very high, respectively. The ‘high’ and ‘very high’ vulnerability classes were found along the River Ravi and nearby areas, especially in Ravi Town (Farrukhabad/Shahdra). It is recommended that the arsenic removal plants should be installed at all the tubewells with higher arsenic levels (arsenic concentration > 50 ppb) to provide safe drinking water to the citizens of Lahore. Moreover, the spatio-temporal techniques used for analysis of water quality should be applied in other mega cities for better monitoring and management of drinking water quality.