Assessing interactions between groundwater and surface water in a Pampean stream

Although water law and water policy often consider groundwater and surface water as separate resources, both water bodies are functionally interdependent. Surface-water bodies such as rivers, lakes and wetlands can receive groundwater inflow, recharge the groundwater or do both actions in different sectors or seasons. This interaction between water systems leads to the mixing of their water qualities (Winter 1999; Sophocleous 2002; Hancock et al. 2005; Goldscheider et al. 2006). High quantities of nutrients or other dissolved chemicals derived from natural processes and human activities in surface water can be transferred to the connected groundwater system and vice versa. Thus, the interaction of ground water and surface water has been shown to be a significant concern in issues related to water supply, water quality and degradation of aquatic environments (Winter 1999). The interactions of streams, lakes, and wetlands with groundwater are governed by the positions of the water bodies with respect to groundwater flow systems, geologic characteristics of their beds, and their climatic settings (Tóth 1999; Winter 1999).

Despite the fact that groundwater and surface water interactions are controlled by their hydraulic connection and may be considered essentially one resource, the interactions are difficult to observe and measure (Winter 1999; Sophocleous 2002; Custodio 2005). All methods have shown significant uncertainty, as usually happens when humans try to quantify nature.

The regional geological and hydrological features were identified in the field and hydrogeological data were obtained from 25 wells, most of them penetrating the upper 20 m of the unconfined sedimentary aquifer. Climatic and land use features were analyzed for the study area. The historical precipitation data were collected from ‘La Perdiz’ station (33 °54′52.54″S – 64 °24′16.39″W). In this first stage of the study, the hydrological survey was carried out in summer and the stream flow was measured in this season in different gaging sections. Twenty-five groundwater and surface water samples were obtained and field parameters were measured in situ: pH (Electrode Orion 9104), electrical conductivity (EC) (Hanna Instruments, HI 9033) and temperature (T) (YSI Model 95 Handheld Temperature System).

The samples were collected in 1 L plastic bottles and were analyzed within 24 h of collection, in the laboratory of the National University of Rio Cuarto. The following parameters were analyzed: Na⁺, K⁺, HCO₃⁻, Cl⁻, Mg²⁺, Ca²⁺, SO₄²⁻ (American Public Health Association (APHA) et al. 2005 ).

Although this work was carried out in the medium-low basin as a result of a requirement of a water company, some samples were collected in the upper and middle basin to improve hydrological interpretations. Moreover, and even though this research was carried out in summer, an expeditious survey was carried out in winter to measure EC and isotope values. The numerical geochemical models were developed using the PHREEQC software (Parkhurst and Appelo 1999) taking into consideration the major ions (HCO₃⁻, SO₄²⁻, Cl⁻, Na⁺, K⁺, Ca²⁺, Mg²⁺) and the stable isotopes (δ¹⁸O/δ²H). The activity of ²²²Rn in surface water and groundwater was measured in situ in five samples using the RAD7 radon detector (Durridge Co.) belonging to INGEIS.

Climatic changes which occurred in South America during the Quaternary provoked deflation and deposition of large masses of silt and fine sand which formed loess and loessoid units of regional extension in several areas of the continent. The winds derived from the Patagonian ice field during the Last Glacial Maximum transported to the Pampean plain silts and fine sands formed by frost action in the Andean Mountains Range. In humid periods, these loessical and sand plains were crossed by different fluvial systems while, in the interfluvial areas, soils were developed (Iriondo 1997).

The Ají stream starts at the confluence of the Cortaderas and Zelegüa streams with some supply from the Suco lagoon. From the middle basin, it is a permanent stream as the result of groundwater contribution. The flow varies between summer (0.8 m³/s) and winter (in the order of 0.40 m³/s) with an average water velocity of 0.33 m/s (Table 1).

The groundwater from the unconfined aquifer exhibits variable EC from 650 μS/cm (fresh water) linked to recharge in dunes (Bécher Quinodóz 2014), up to 14,000 μS/cm (salty water) in the hydrohalomorphic areas (Figure 3). Groundwater shows a geochemical evolution in the flow direction from fresh to salty water, changing from sodium bicarbonate to sodium sulfate and, finally, to sodium chloride geochemical type (Table 4, Figure 5).

The analyses of relationships show that surface water exhibits marked homogeneity, with freshwater from the beginning to the end of the studied area (Figure 5), despite having traveled tens of kilometres and maintaining, in most sections, hydraulic connection with an aquifer that provides brackish to salty groundwater. This situation allows the interpretation that the aquifer has little influence on the mixture.

In some samples, the similarity between aquifer and stream isotopic features are closely linked to induced recharge from the stream freshwater when groundwater is extracted in wells located at the stream side (Figure 4(b), Section A₂). This singular situation holds an environmental positive connotation, because the induced recharge improves the groundwater condition. More studies are necessary to evaluate the winter behavior, but preliminary δ²H and δ¹⁸O isotope results showed more fractionation in winter stream samples as a result of evaporation. The surface water will also be more enriched in winter as a result of the aquifer contribution.

Taking into account the conceptual model, it was considered that A_Z2 (upper basin) is a mixture between A_Z1 (surface water from mountain area) and SL (surface water from Suco lagoon). Table 5 shows chemical and isotope data of samples used in the numerical models.

The results (Table 5) showed that to obtain the sodium bicarbonate water type of Zelegüa stream in AZ₂ (final solution), the initial solutions must be mixed in a ratio 92/8 (AZ₁/SL) and cation exchange and salt dissolution processes (CaSO₄:2H₂O) must be produced.

The A₁ sample was considered a mixture (Table 6) of A_Z2 (surface water, Zelegüa stream), A_C1 (surface water, Cortaderas stream) and M₈₃ (groundwater from unconfined aquifer). The results (Table 6) showed that to obtain the sodium bicarbonate water type of Ají stream in the A₁ section (final solution), the initial solutions must be mixed in a ratio 19/72/9 (A_Z2/A_C1/M₈₃) and cation exchange and salt dissolution processes (CaCO₃;CaSO₄:2H₂O) must be produced.

The A₃ sample was considered a mixture (Table 7) of A₁ (surface water of Ají stream, which is identical to A₂) and M₁₃₉ (groundwater from unconfined aquifer).

The results (Table 7) showed that to obtain the sodium bicarbonate water type of Ají stream in the A₃ section (final solution), the initial solutions must be mixed in a ratio 98/2 (A₁/M₁₃₉), and cation exchange and salt dissolution processes must be produced (CaCO₃;CaSO₄:2H₂O).

The ²²²Rn regional background value in surface water was tentatively established at 1 pCi/L, on the basis that generally there is no ²²²Rn in surface water (Ellins et al. 1990). This result indicates that, at the sampling moment, surface flow predominated and groundwater contribution was approximately only in the order of 2% (Table 9, Figure 8).

The explained surface water-groundwater relationships showed the control exerted by the relief, the lithology and the climate. The chemical composition of both hydrological systems is essentially the result of different water velocity in each body. This situation leads to mixing processes between surface water and local, intermediate or regional groundwater giving rise to the observed geochemical pattern. The aquifer-stream relationships in the studied stream reach, as was observed from the hydraulic data, showed effluent/influent varying spatially as a result of the fluvial geomorphology and human actions, although a natural gaining stream type prevails.

The isotopic behavior (δ²H – δ¹⁸O) of both hydrological systems also indicates water mixtures. This point is validated by ²²²Rn and the numerical geochemical modeling which indicates also a small contribution from the aquifer (∼2%) in summer.

Taking into account the isotopic and modeling results and although fresh surface water flows in areas with high groundwater salinity (up to 14 mS/cm in the low basin), it may be established that there is strong predominance of surface water in the summer mixing process. As a result of surface water high velocity and the incoming of such small groundwater quantities, salts that come from the aquifer are diluted.

Furthermore, the stream water samples and those extracted from adjacent wells exhibit the same low salinity (in the order of 0.4 g/L) and geochemical type (sodium bicarbonate) due to induced recharge from stream to aquifer during pumping. This is a promising situation for conjunctive use of surface water and groundwater for different purposes. In this regard, the first measurements in the winter season showed that the stream water has low electrical conductivity and it is still influenced by the fresh water coming from the mountains. Thus, it was considered a vital water supply offer and some wells located beside the stream have already been built to generate induced recharge of surface fresh water towards the aquifer. More studies are required, especially extensive and thorough sampling in the two contrasting seasons, if a more comprehensive description of the entire basin is to be attempted.

The research was supported by FONCYT (Fondo para la Investigación Científica y Tecnológica) and MINCYT (Ministerio de Ciencia, Tecnología e Innovación Productiva) (PID 35/08 Prestamo BID) National University of Río Cuarto and the National Council of Scientific Technical Research (CONICET) and IAEA/RC/17385.