Abstract:

A comparative test of two vulnerability mapping methods (COP and PaPRIKa) specifically dedicated to for karst aquifers was carried out on two Mediterranean carbonate aquifers. The vulnerability maps obtained for each aquifer present important differences. To identify and determine the origin of these differences, the results were statistically analyzed using sensitivity analysis, coefficients of determination and scatter graphs. In addition, the global vulnerability (Gv) parameter was used to measure the general vulnerability of the aquifer and to compare the results obtained. This statistical analysis led us to conclude that the main cause of differences between these two methods used to assess aquifer vulnerability lie in the relative importance of the parameters employed in calculating the vulnerability index. For the PaPRIKa method, the variable related to infiltration (slope and karst features) has the most influence, with less weight being assigned to the protective capacity of layers overlying the aquifer. For the COP method, the most influent variable is defined by the layers overlying the aquifer, together with infiltration characteristics, determined by the relative importance of different forms of infiltration in each aquifer. The vulnerability mappings performed using the COP method present greater coherence with the known hydrogeological behavior of the study areas, especially the Spanish aquifers. Nevertheless, further hydrogeological investigations are needed, such as ones to validate the obtained vulnerability maps.