Using an expert-based model to develop a groundwater pollution vulnerability assessment framework for Zimbabwe

摘要

Predicting groundwater vulnerability to pollution in countries with sparse data is difficult. Using Zimbabwe as a case study, this study presents a simple expert-based GIS model that incorporates seven natural and anthropogenic factors for modelling groundwater vulnerability at a national scale. The predicted groundwater vulnerability was tested against three water quality parameters: pH, total nitrates and Total Dissolved Solids (TDS). The agreement between predicted vulnerability and measured groundwater quality was reasonably high for pH and total nitrates (R-2 >= 0.54, P < 0.05). For the dissolved solids, the relationship between vulnerability predicted by the model and measured TDS was weak (R-2 = 0.33). Model predictions indicate spatial variability of groundwater vulnerability to pollution, with high pollution in Mazowe and Manyame catchments dominated by high urban land use and commercial agriculture. Moderately high vulnerability of groundwater to pollution was predicted for Mzingwane and Runde catchments, whereas the lowest vulnerability of groundwater to pollution were predicted for the lower Sanyati sub-catchment and Gwayi catchment. The advantage of the expert-based GIS model presented here is that it requires only a limited and widely available input data set. Hence, it can be used in other countries with limited hydrogeological data to generate results showing spatial variability of groundwater vulnerability to pollution to guide sustainable groundwater management. We, however, recommend further research with case studies focusing on the most vulnerable catchments such as Mazowe, and more extensive validation of parameters, especially TDS, to improve the proposed model.