Understanding arsenic and uranium sources and mobility in groundwater flow systems from two areas in San Luis Potosi, Mexico: Cerritos and Villa de Reyes Basins
- Verständnis von Arsen- und Uranquellen und -mobilität in Grundwasserströmungssystemen von zwei Gebieten in San Luis Potosi, Mexiko: Cerritos- und Villa de Reyes-Becken
Cauich Kau, Dario del Angel; Rüde, Thomas R. (Thesis advisor); Cardona Benavides, Antonio (Thesis advisor); Castro Larragoitia, Guillermo Javier (Thesis advisor)
Aachen : RWTH Aachen University (2021)
Dissertation / PhD Thesis
Dissertation, Rheinisch-Westfälische Technische Hochschule Aachen, 2021
Arsenic (As) and Uranium (U) are trace elements present in (ground) waters. Worldwide As and U have been identified in groundwaters with elevated concentrations above the reference levels set by the World Health Organization (WHO) threatening the human health especially in arid to semi-arid areas where the groundwater is the main source not only for drinking water for the population but also for the industry. In addition, water in the vadose zone and shallow groundwater are influenced by evaporation, in particular in these regions. This study deals with geogenic, anthropogenic, and the evaporation effect on As and U concentrations in two regions in the Mexican Altiplano, Cerritos Basin (CB) and the Graben of Villa de Reyes (GVR) in which elevated concentrations of As and U in groundwater have been detected. Even though they are located in different physiographic entities with different, geological, calcareous rocks affected by a mineralized granitic intrusive body characterize the CB and igneous rocks characterize the GVR, and hydrogeological characteristics they have in common elevated As and U concentrations in groundwater. Therefore, the main objective of this study is to understand the processes and dynamics controlling As and U mobility in groundwater flows systems as well as the identification of the primary and secondary sources and the interaction between the geological media and the groundwater. The chemical characterization of the water samples gave that in both areas the pH is in circum-neutral conditions. At the CB 73% of the samples are within the As Mexican threshold for water for human consumption; however, for the (WHO) guideline just 55 % are within. In the case of U, currently the Mexican drinking standard does not include the U in the mandatory elements; however, 96% of the samples in the area are within the WHO guideline. At the GVR all the samples are within the As Mexican threshold and 36% are within the WHO guidelines. For U 97% of the samples are within the WHO critical value. The evaluation of the water types in the CB showed the influence of the in the calcareous units being the Ca-HCO3 the main water type and at the south the wells tapping the gypsum units the main water type was the Ca-SO4. On the other hand, at the GVR the main water type was the Na-HCO3 with temperatures up to 42.4 °C and is distributed along the GVR tapping the basin fill sediments in which the highest As concentrations were found. One of the processes identified by Nicolli et al. (2010) for As and U enrichment in shallow groundwater is the evaporation process. Therefore, in order to evaluate the evaporation effect in the As and U concentrations on the shallow aquifer at the CB environmental stable isotopes δ2H and δ18O were used. The results of the deuterium excess reflected the fractioning of the water vapor. These allowed to classify the output in three groups each one in function of the deuterium excess and the As and U concentrations. These groups are from dug wells located basin fill, which showed the higher evaporation effect. The interpretation indicates that with a lower deuterium excess As and U concentrations increase demonstrating that the evaporation of shallow groundwater contributes to an increase of As and U concentrations in the arid climate of the region. Additionally a geochemical modeling evaluation was done in order to evaluate evaporation process at the CB. The modeling showed that the evaporation effect has influence in the As concentrations behavior in which adsorption/desorption processes play an important role in the increase or decrease of the As concentrations. In contrast, at the GVR no evaporation effect was observed due to depth of the wells. The statistical evaluation using the orthogonally rotated solution explained a 61% of the total variance in both areas, at the CB with four factors and the GVR with five factors. In both areas, each factor was defined using the compound, element, or elements with the highest scores. The factors obtained at the CB are the following. The sulfate factor had samples from the deep wells and dug wells tapping the Gypsum-Anhydrite formation in the south. The sodium-chloride factor was linked to deep groundwater. The arsenic-uranium factor gathered samples from deep wells and dug wells tapping the basin fill. The nitrate factor is related to diffuse sources associated with intensive farming activities. Likewise, a group of samples did not have high scores on any of these factors as was indicated by the 39% of unexplained variance. At the GVR five factors were defined. The uranium factor grouped samples from wells located on the western flank of the GVR where the San Luis - Tepehuanes major fault system crosses the graben. The lithium factor is also spatially related to the San Luis – Tepehuanes fault system. The potassium factor gathered several samples that are distributed over the whole study area. The arsenic-sodium factor is spatially related to the center of the GVR and the eastern section where the old groundwaters, high temperatures, and arsenic values are located pointing to a deep groundwater with a maximum temperature measured at the surface of 42.4 °C. The chloride-sulfate factor showed samples located in the same areas as factor 1 and 2. Finally, the samples of no factor did not have high scores on any of the four or five factors as was indicated by the 39% of unexplained variance. Some of these samples located at the borders of the study area imply that these sites could have local conditions. In both areas the As and U concentrations found in several solid samples exceed the global background values. At the CB, the samples with the highest As concentration were one rock sampled from the granitic intrusive body, one sediment sample taken in a riverbed downstream of the intrusive body, and one tailing from the old mining works associated to the intrusive body. The results allowed identify the intrusive body and associated mineralization as the primary As and U to the CB. The secondary source are the sediments and soils derived from the intrusive body where adsorption/desorption play an important role. On the other hand, at the GVR the volcanic glass has the highest As concentration and is considered the primary source of As to groundwater, especially in high altered rocks with rhyolitic composition. Additionally, a possible source could be the small mineralizing processes associated to magmatic activity in the west margin of the GVR. The long-term column test (CT) experiments using artificial rainwater, with the chemical characteristics of the area, were performed in order to evaluate the contribution of the mining tailings and in two soils influenced by the intrusive body and tailing material on the As and U concentrations at the CB. The mineralogical characterization using XRD method was applied in order to identify the main mineralogical phases present in the samples in both study areas. The main phases presented in the rocks, soils, and sediments were silicates such as quartz and feldspar and carbonates like calcite and dolomite. For the mining tailing, the main phase was the secondary oxide minerals goethite and hematite. A Qemscan evaluation applied to a tailing and a soil sample provided morphological information and the surface compositional distribution of the particles. In the CT experiments As and U showed similar behavior in the different samples. The As and U started with low concentrations and after a few volume changes their concentrations started to rise until the maximum concentration. However, after reaching the maximum concentration a depletion of the concentrations started. The As kept a “constant” concentration, in contrast the U concentrations were very low indicating a possible depletion or the pH conditions constrained its mobilization. The source of the As concentrations is associated with the Fe-hydroxides and additional source from sulfides oxidations is not evident because the sulfate concentrations were very low along the duration of the experiment. Therefore, the results showed that the mining tailings and the soils influenced by the intrusive body constitute a long-term source for As in the area, the first as a primary source and the soils as secondary source. In addition, the Qemscan showed that in the soil and in the tailing the As are co-precipitated/adsorbed to secondary minerals like Fe-hydroxides. According to statistical analysis, a group of samples showed elevated Li and Ge concentrations and temperatures. The Li geochemistry is useful as an indicator for regional groundwater flow and as a proxy to relative residence time in the aquifer, the higher the concentration the older the relative residence time. Therefore, in order to identify the groundwater mean residence time at the GVR five samples with the highest temperatures and Li concentration located across the regional fault system and tapping the latite Portezuelo formation were evaluated. The evaluation included radiocarbon isotopes, a calcite calibration collected manually from the debris of two drilled wells tapping the latite Portezuelo, and the methodologies from different authors were applied. The calculated corrected Libby ages results were ages from 12,300 to 17,500 B.P. indicating groundwater of Pleistocene age, slightly older than the ages calculated by Carrillo-Rivera et al. (1992). The youngest ages, located one at center and the second to the eastern flank of the GVR, can result from mixing with younger waters of higher levels. The oldest ages, located at the middle of the GVR where the San Luis - Tepehuanes major fault system crosses the valley, give a minimum age of groundwater that also they can be mixed up with younger shallower groundwater and can be correlated with the Last Glacial Maximum. Even though in the GVR the groundwaters are Pleistocene age they do not have elevated As and U concentrations that according to the rock analysis the altered volcanic glass adsorb the As concentrations, being this a natural control in the area. Summarizing all the information produced and interpreted in both areas a hydrogeological section for each area was proposed. At the CB, the sediments derived from the intrusive body are transported downstream and are deposited in a pond south of the Guadalcazar Polje. There, the surface water in contact with the soils and sediments with high As concentrations and seeps into a karst system below. Additionally, this water enriched with As gets in contact, in the karst system, with the waste water of the nearby town (As 32 µg/L). The water seeping into the karst system continues its path to the Cerritos valley as an intermediate flow system that probably gets in contact with shallow water contributing to its As concentration. In these shallow groundwaters the evaporation effect tends to increase the As concentration. The increase of the As in the these waters is controlled by the desorption/adsorption processes according to the interpretation of the hydrochemical model, the XRD, and the Qemscan results the Fe-hydroxides are the main responsible for these processes. At the Graben of Villa de Reyes the samples with the highest As concentrations have a thermal component, are Na-HCO3 water type, and are located in the center of the study area where the San Luis – Tepehuanes major fault system crosses the GVR that can be a preferential flow path. For natural conditions previously to the intensive pumping rate, the preferential groundwater flow direction was toward to the NE of the GVR. Actual piezometric information shows two drawdown cones as a result of heavy pumping rate at the center, where the oldest groundwater are located, and at the NE of the GVR. According to the groundwater mean residence time, the current pumped groundwater suggest an increase in the ascendant flow indicating abstraction from deeper and older zones. Even though the groundwater is from Pleistocene age, high temperature, and deep sited the As concentrations are in an average of 11.4 µg/L. Due to the As concentrations are higher in the devitrified matrix this can be associated with adsorption/desorption process. The As is absorbed in non-devitrified suggest natural control in the area. According to the statistical analysis, the uranium factor gathered samples from wells located on the western flank of the GVR where the major fault system crosses the graben. This implies that uranium is related to mineralized waters of alkaline-earth bicarbonate type linked to regional fault system. The bicarbonate type facilitates uranium mobility as uranyl carbonate complexes. The main source of As in the GVR is the volcanic glass, especially from rocks with high alteration grade and rhyolitic composition and a possible source could be the small mineralizing processes associated to magmatic activity in the west margin of the GVR. These findings are in line with the results that Banning et al. (2012) found in the area. Besides, Alarcón-Herrera et al. (2020) described several areas within Mexico with similar geological characteristics and As concentrations as in the GVR. To close, at the GVR the As concentrations with an average of 11.4 µg/L in high temperature, deep sited, old groundwater are due to natural control in the glassy matrix. In contrast, in the CB the As and U concentrations in groundwater are controlled by the Fe-hydroxides present in the sediments derived from the intrusive body and the associated mineralization. In addition, the evaporation effect has an important role in the enrichment of these elements in the shallow aquifer. Finally, with a geological approach the control mechanisms of As and U mobility in both study areas cannot be compared due to they have different characteristics, in the CB the groundwater flows through the calcareous units from the Valles-San Luis Platform. In contrast, at the GVR the groundwater flow through the volcanic rocks overlying the calcareous units of the Mesozoic basin. As final remark, the approach used during this research could be used not only in Mexico, but also worldwide in sites with similar geological conditions.