The field-obtained information was entered into a Geographic Information System (GIS), developed specifically for the outcropping area of the GAS which allows the processing of information and the generation of thematic maps. The basic cartography (1:250,000) was developed in Arcview format and covered a total of 16 topographic sheets (1:50,000) of the Cartographic Plan of the Military Geographic Service (MGS) of the República Oriental del Uruguay (Sheets: Rivera, Paso de Ataques, Tranqueras, Cuñapirú, Minas de Corrales, Cuchilla del Ombú, Los Novillos, La Hilera, Arroyo de Clara, Masoller, La Palma, Paso del Cerro, Bañado de Rocha, Tacuarembó, Batoví, Curtina)15)(16)(17)(18)(19)(20)(21)(22)(23)(24)(25)(26)(27)(28)(29)(30. Digitization included hydrographic networks, major towns and cities, departmental and international borders, national routes, and secondary roads.
Wells, outcrops, and surveyed points were added to the mapping using Gauss Krüger projection. The fundamental geological survey for the analysis of vulnerability and risk of groundwater contamination was carried out by identifying and delimiting the main geological units, differentiating lithofacies, raising profiles and taking representative samples, in order to prepare the geological map of the region at a scale of 1:250,000, which is reproduced in Fig. 1. The well location map (Fig. 2) was generated from the census of wells and water points, fundamental for the preparation of the piezometric map and the water table depth map (Fig. 3 and 4). The well census consisted of obtaining hydraulic information, groundwater use, measurements of the total depth of the well, static level, dynamic level, suction level and flow rate, among others. Geographical coordinates were obtained with a Trimble data collector, with an error of up to 3 m, and the height above sea level (asl) was obtained from the topographic elevations, represented in the topographic sheets of the MGS.
Static levels were measured with a millimeter-graduated electrical probe, considering an error not greater than a centimeter. Based on the scale (1:250,000), the regional field survey comprised a total of 119 perforations, which provided a density of 1 well every 50 km2 (Fig. 2). A more detailed survey was carried out in the city of Rivera at a scale of 1:50,000, which included the OSE catchment wells. The survey included about 140 wells in total. The flow network, essential for understanding underground dynamics (identifying recharge and discharge zones, preferential flow directions and determining hydraulic gradients), was constructed from the values of the hydraulic dimensions (difference between the topographic level and the depth of the water level measured in the well).
The piezometric map was made using spatial data interpolation (ArcGIS) and then manually corrected, according to topography and surface drainage network. According to the scale of work and the distribution of the wells, an equidistance of 10 m was chosen, which allowed achieving an adequate definition of the piezometric morphology. The Water Table Depth Map of the outcropping GAS was elaborated to know the distribution of the water depth in the studied area, which is reproduced in Fig. 4.
The depth of the water table is an extremely important variable regarding the hydrogeological behavior of a region. It is the most important variable when determining aquifer vulnerability. The greater the depth, the thicker the unsaturated area and, consequently, the greater the storage capacity. On the contrary, if it is shallow, this capacity is reduced or canceled in areas where the water table emerges. In addition, in areas with no sewage services, shallower surfaces or groundwater outcroppings cause a high risk of contamination for the population. With regard to exploitation, if the water table is deep, more powerful pumping equipment, higher energy consumption and, therefore, higher extraction costs are needed. Fig. 4 reproduces the water table map, corresponding to shallow permeable levels of the Rivera-Tacuarembó aquifer unit, and shows that values lower than 5 m are located in the center-west of the area, coinciding with the discharge area. The depth increases towards the E and NE with depths greater than 20 m, and towards the SW, where depths greater than 25 m are observed. For the city of Rivera and its surroundings, the highest values are observed in the SE of the city, with more than 30 m, where the piezometric surface tends to a depression cone. Considering the totality of the studied area, the most frequent depth of the water table is between 10 and 20 m. The greater depths coincide with the division of surface and groundwater, with correspondence between high topographic and hydraulic values with greater water table depths, and low topographic and hydraulic with lower water table depths. This relationship is typical of areas where the water balance presents notorious excess, as happens in the studied area.
The water table depth is the most important variable to determine the vulnerability classification (Tables 3 and 4, and Figs. 10 and 11). As indicated by Foster and others36, it is misguided to believe that using more complex methods that consider a greater number of variables produces more reliable maps and closer to the system reality. In this specific case of the outcropping Guarani Aquifer, the use of other complex methods would not improve the vulnerability mapping, since the area does not present important changes in the topography, in the recharge of the system or the types of soil, being the depth of the water table and the vertical permeability the most important variables to study in detail and locally, according to the use of the soil and the potentially polluting load.