Akira Hiratsuka

GIS Based Fuzzy Optimization Method to Groundwater Vulnerability Evaluation

Recognition of groundwater vulnerability to pollution will help in protection groundwater sources and managing groundwater quality conflicts. The DRASTIC model has been used as a valuable tool for assessing groundwater vulnerability over the large geographical areas involving a variety of hydrogeological settings. DRASTIC index uses linear model to calculate the final vulnerability index and the factor that influence the groundwater vulnerability must be divided into ranges and then be given ratings. This system is unable to describe a continuous transition from the easiest to be polluted to the most difficult to be polluted that is fuzzy nature of the groundwater vulnerability to pollution. In this paper, GIS based fuzzy optimization model has been presented to evaluate the groundwater vulnerability to pollution by taking the fuzziness into consideration. Finally, the result is compared with DRASTIC method in a case study of Kathmandu Valley basin, Nepal. It is shown that the fuzzy optimization method incorporates more efficiently the fuzzy nature of the groundwater vulnerability to pollution and is more applicable in reality.

Current status and trend of nitrate-N and factors affecting its concentration in shallow groundwater systems of Kathmandu

Nitrate-N is commonly examined as indicator to trace the impact of anthropogenic activities on groundwater quality. The aim of this paper is to describe the current status and trend of the nitrate-N contamination in different sources of shallow groundwater systems of Kathmandu, Nepal which is vital for managing risk associated with nitrate-N contamination. Special emphasis is given to examine the various factors affecting nitrate-N concentration. Water samples were collected and analyzed for nitrate-N and other parameters from ninety sources, including shallow tube wells, dug wells and stone spouts (locally called Dhunge Dharas). Although, mean annual nitrate-N concentrations in shallow groundwater of Kathmandu has slightly decreasing trend over the period from 1999 to 2008, it ranged from 0 to 26.4 mg/L indicating its vulnerability to contamination. The results confirmed 16% of the sampled wells exceeded USEPA guidelines of 10 mg/L of nitrate-N; however, another 33% of the wells had impacted levels of nitrate-N between 2 and 10 mg/L. This result can be attributed to different anthropogenic activities at shallow depth such as leaking septic and sewer systems and disposal of human waste. Moreover, this study revealed mutually exclusive relationship and negative correlation between nitrate-N and iron that suggest nitrate is being consumed by natural denitrification in some extent.

Combining Rule-Based Fuzzy Model with GIS to Predict Groundwater Vulnerability to Contamination

The groundwater is the major natural resources for drinking and other various purposes; however, it is under threat of degradation of its quality from concentrated and/or dispersed sources of contamination in urban area of developing countries. Therefore, contamination of groundwater has become a major concern of decision makers and water managers involved with the management of water quantity and quality and their relationship to human health in recent years. Since, groundwater monitoring and mathematical modeling is very costly in regional scale vulnerability assessment, one frequently used water quality protection strategy is to develop contamination potential maps of groundwater, and then prioritize those wells located in the potentially highly contaminated areas for testing of contaminants. However, generation of contamination potential maps is not an easy task due to lack of sufficient hydrogeologic database and inherent uncertainty. This procedure is usually carried out using GIS based overlay index method and process based model. DRASTIC model, a popular index method is used to assess groundwater vulnerability over the large geographical areas on the basis of general hydrogeological knowledge of the contaminant transport in aquifer media. DRASTIC index uses linear model to calculate the vulnerability index and the factors that influence the groundwater vulnerability must be divided into ranges and then be given ratings. This system is unable to describe a continuous transition from the easiest to be polluted to the most difficult to be polluted that is fuzzy nature of the groundwater vulnerability to contamination. Therefore, this paper aims to estimate the risk of an aquifer to be polluted from concentrated and/or dispersed sources which applies the ability of the fuzzy ruled-based system to cope with the modeling of a nonlinear system and presentation of the output of the fuzzy system in the framework of GIS. This may be possible to describe this concept with a set of linguistics rules in the form of IF-THEN. This paper presents the fuzzy inference system benefiting from fuzzy engine and expert knowledge-based DRASTIC parameters for nonlinear mapping of the intrinsic groundwater vulnerability concept. The developed fuzzy model is applied to a shallow aquifer of Kathmandu Valley, Nepal and the result is compared with that of the DRASTIC model. This methodology has potential for facilitating groundwater vulnerability modeling at a regional scale and can be used for other regions, especially in the case of data scarcity, however, would require incorporation of appropriate input parameters and rule base suitable for the region.