During the past two centuries, many arid and semiarid areas worldwide have experienced important changes in land use as humans continue to occupy and diversify the land, thus invariably stressing the availability of water and often causing water tables to decrease (Jackson et al. 2001; WHO 2005; Kafri and Yechieli 2010; Pascual-Ferrer et al. 2014). Within arid areas, a common occurrence are closed (endorheic) basins whose surface water and groundwater are enriched with salts (Grünberger 2005; Elmore et al. 2008; Kafri and Yechieli 2010; Rodríguez-Rodríguez et al. 2012).

Groundwater mined in closed basins around the world, e.g., the United States, Mexico, China, Argentina, and Chile, report high concentrations of arsenic (As) and fluoride (F), often greater then established limits for human health (Luo et al. 1997; Welch et al. 2000; Smedley and Kinniburgh 2002; Smedley et al. 2002; Espósito et al. 2011; González Dávila 2011). However, there remain many closed basins that have not been fully characterized and where the hydrological processes are little known. Numerous closed basins in northern Mexico and the United States Southwest belong to this category (Rosen 1994; Janeau et al. 2005).

The limits for As established by the (WHO 2005) and the United States Environmental Protection Agency (USEPA) are 0.010 mg L−1 for As (USEPA 2013), whereas the Mexican norm stands at 0.025 mg L−1 (NOM-127-SSAI- 1994). For fluorine, both the WHO and Mexican norms recommend 1.5 mg L−1 and the USEPA 4.0 mg L−1, although the USEPA also has a secondary drinking water regulation of 2.0 mg L−1 for F, mainly to prevent unwanted cosmetic effects (skin and tooth discoloration).

In arid areas, drought intensity is expected to increase in the near future as a result of global climate change (Strzepek et al. 2010) which, coupled with the low productivity of salt-affected soils of closed basins, will likely encourage changes in land use and a greater dependence on groundwater resources (Rosen 1994; Jackson et al. 2001). In other arid regions around the world facing this predicament, land-use data are commonly added to contaminant data to search for possible sources and impact of these contaminants (Robinson and Ayotte 2006; Farooqi et al. 2007; Rodríguez-Rodríguez et al. 2012). Gaining knowledge of the natural behavior of the aquifer systems and monitoring their levels and water quality is therefore vital to the attainment of effective and sustainable use of the resources.

In northern Mexico, the health effects caused by the prolonged ingestion of groundwater that is high in As and F have been addressed in numerous studies, some of which have focused on dental fluorosis (Alarcón-Herrera et al. 2001b) and/or chronic arsenicosis (Del Razo et al. 1993; Ochoa-Reyes et al. 2009). In arid and semiarid regions of Mexico and nearby areas, these potentially toxic elements have been attributed mainly to its natural (geogenic) occurrence (Planer-Friedrich et al. 2001; Gutiérrez et al. 2009; Ortega-Guerrero 2009; Scanlon et al. 2009; Reyes-Gómez et al. 2013) and, in some cases, to environmental degradation generated by human activities, such as land-use change, presence of mine tailings, and presence of chemical and agricultural wastes (Alarcón-Herrera et al. 2001a; Razo et al. 2004; Armienta and Segovia 2008; Gutiérrez et al. 2012).

Local studies focusing on groundwater in arid and semiarid parts of the state of Chihuahua report As and F concentrations spanning over a wide range of values from 0.001 to 0.11 mg L−1 As and 0.3 to 9.7 mg L−1 F (Piñón-Miramontes et al. 2003; Ruiz-Payan et al. 2005; Mahlknecht et al. 2008; Espino-Valdés et al. 2009). Prolonged droughts accentuate the shortage of the availability of water, which in this particular region seems to trigger an increase of the concentrations of As and F in groundwater (Núñez-López et al. 2007).

Long-term monitoring of the quantity and quality of drinking water in Mexico is performed sporadically by federal agencies [e.g., Comisión Nacional del Agua (CONAGUA)] and by state and municipal offices (e.g., Junta Estatal y Municipal de Agua y Saneamiento). The different purposes driven by these agencies to gather the data with a lack of periodicity cause the records of groundwater levels and water quality to be incomplete and difficult to collect (Arreguin-Cortéz et al. 2010). To fill in some of the gaps, wells within the Laguna El Cuervo closed basin located in Chihuahua, Mexico (Fig. 1) were sampled by the investigators in 2005 and then again in 2007, 2010, and 2011. Not all of the wells were sampled each year due to a variety of impediments ranging from man power to obtaining permission from well owners. This monitoring effort will continue on a medium- to long-term basis to document the evolution of drinking water quality and potentiometric levels.

Fig. 1
figure 1

Map of the study area showing wells and sediments sampling locations. The dashed line marks the boundary of the Laguna El Cuervo closed basin. The town of Aldama is located at its southernmost part

Full size image

In addition to documenting the fluctuation of As and F content in groundwater, the data were analyzed to confirm if these two contaminants occur together and to identify possible causes of the high As and F occurrences. These findings are more relevant after a portion of the basin of approximately 228 km2 (3.5 % of study area), which had historically been used for grazing, changed to agricultural land use in 2010 and is now being irrigated with groundwater for the most part. In 2011, cotton, alfalfa, and wheat were growing on these plots and were being irrigated with groundwater.

To aid in the identification of possible sources and distribution of As and F contaminants, the geochemistry of the sediments (alluvium) was included in this study. A previous study reported that both As and F were present in the rhyolites and other volcanic rocks that are main contributors to the sediment fill of this basin (Reyes-Gómez et al. 2013). The spatial and temporal distribution of As and F were analyzed for both groundwater and sediments by means of ArcGISRM (ESRI, Redlands, CA) spatial tools software.

In summary, this study aims to document the contents of As and F in the aquifer within the Laguna El Cuervo closed basin, their change with time, and their degree of co-occurrence. In addition, the identification of specific parameters that may cause these variations is sought based on a 7 year-long monitoring of groundwater and using four monitoring wells.

Materials and Methods

Area of Study

The study area comprises the endorheic basin known as Laguna El Cuervo. This basin has a surface area of 6,479 km2 and belongs to a series of closed basins known as Cuencas Cerradas del Norte with Laguna El Cuervo occupying 7.3 % of the combined area (Grünberger 2005; Janeau et al. 2005) as shown in Fig. 1. The geology consists of sedimentary and volcanic rocks that outcrop at the edges of the basin and a thick layer of alluvium ≤250 m deep (Dobson et al. 2008). The outcrops on the southern part of the study area are mainly limestone and include the sierras El Torreño, El Morrión, and, to the east, La Tasajeña. To the east and west, the rocks are predominately igneous with an abundance of rhyolites and tuffs.

The topography is typical of the basin and range with slight surface elevation differences inside the basin from 1,226 m around the northern edge of the basin to 1,185 m in the center. The mountains rise approximately 300 m above the basin floor. The sampled wells were all located in the basin with the exception of one well (well no. 27), which was located at a greater elevation. Sediment fill is composed of Quaternary alluvium of lacustrine origin (INEGI 1984) deposited in layers with larger size particles at the edges and finer sediments toward the center. Layers with high clay content have developed perched aquifers and a few springs. Geomorphologically sodic–saline soils make up the central part of the basin (Grünberger 2005; Janeau et al. 2005). The most abundant soil types are Orthic Solonchak and Chromic Vertisol. In the north and eastern parts of the basin, Euthric Regosol soils are also present (INEGI 1984).

The aquifer underlying this basin is the alluvial aquifer known as Hormigas, which extends over a surface of approximately 6,014 km2; its average water extraction has increased from 1.5 (INEGI 1999) to 12.8 Mm3 (CONAGUA 2008). The aquifer has a transmissivity of 0.006 m2 s−1 and an average recharge rate of 25.6 Mm3 year−1. The water is of medium drinking quality with total dissolved solids (TDS) between 400 y 800 mg L−1 except in the vicinity of the town of Aldama (population 18,000) where TDS values ≤2,500 mg L−1 have been reported (CONAGUA 2008). Cones of depression reported for the Aldama area indicate that this is an intensively used aquifer. The aquifer adjoins at its southwest part with the aquifers Tabalaopa-Aldama and Aldama-San Diego the transmissivities of which vary between 0.0046 and 0.12 m2 s−1 and of which the TDS of the water ranges from 400 to 2,000 mg L−1.

The climate in this region is arid to semiarid with an annual average temperature of 12 to 18 °C and has extreme temperature variations (García 2003). The climates of the north- and south-central parts of the basin have been classified as desert, and the west and southwest parts as steppe (Janeau et al. 2005).

The annual average precipitation is 335 mm with monsoonal rains (July to September) contributing approximately 75 % of the water and sporadic winter rains (November–January) contributing 20 %. The evaporation rate has been reported as >2,000 mm year−1 (INEGI 1999).

Analytical Determination of As and F in Groundwater

Water samples were collected from 27 wells as shown in Fig. 1. Two private hand-dug wells (nos. 25 and 26) were also included in this study. Eight additional wells located within a few kilometers of the basin were also sampled and were used solely for comparison purposes. Water was collected in plastic bottles of 1-L capacity and kept cool during their transport to the laboratory where they were analyzed according to standard methodology; this is described in more detail in Reyes-Gómez et al. (2013). As was determined by atomic absorption spectrophotometry (Sigma GBC AAS AVANT) according to the USEPA’s 815-R-00-010 method after acid digestion in a microwave oven (MARSX-CEM). F content was determined by an ion-selective electrode (Orion 1260), which had been previously calibrated with standard solutions.

The temperature (T) and electrical conductivity (EC) in water samples were determined with a portable meter Corning CHECKMATE I, the pH with a portable potentiometer team (3STAR; Orion); the depth to the water level (D) was determined in situ with a piezometric probe. TDS values were derived from EC using a factor of de 0.65 mg L−1/µScm−1 (Miyamoto et al. 1995), and salinity was estimated according to the equation by Bodelón et al. (1994): salinity = 0.1175 + 1.0053 log10(EC) in g L−1.

As in Sediments

Thirty sediment samples (sample locations are included in Fig. 1) were collected from the top 20 cm of the soil, stored in plastic bags, and transported to the laboratory where they were air-dried, disaggregated, and analyzed using an inductively coupled plasma spectrophotometer according to standard procedure (method USEPA3050B). To these data sets, 110 data reported by the Servicio Geologico Mexicano were added because they were analyzed using the above-mentioned methodology, thus resulting in a total of 140 data points.

The relationship between As content in sediments and groundwater was investigated using a technique from (Zeng and Wu 2013), which consists of using the minimum distance between the source of contaminants and water distribution through runoff to determine the possible impact. In our case, the source consisted of the four sediments with the highest content of As—S1, S7, S11, and S17—located in the south-central and northeastern part of the aquifer. Next we determined the distance to the nearest wells to test if the sediments would influence the As concentration in the wells (locations shown in Fig. 1). The direction of groundwater has been reported by Secretaría de Recursos Hidráulicos (SARH 1981) and (CONAGUA 2008) as flowing from the edge of the basin toward the center with some lateral and ascending flows in the deeper parts of the basin fill.

Water Levels and Precipitation

The standardized precipitation index (SPI) was calculated to evaluate the effect of dry and wet periods on the dilution and concentration of F and As in wells. The SPI incorporates the precipitation recorded a few months before the target date, which can be as long as 48 months, and transforms the deficit or excess to a standardized value through an incomplete Gamma function (McKee et al. 1995; Núñez-López et al. 2007). For this study, we calculated SPI-3, which included 3 months of precipitation before the date of sampling. For this calculation, it was assumed that 3 months was a long-enough time for infiltrated water to come in contact with the water table. Based on local reports (SARH 1981; CONAGUA 2008) it was found that after 24 days of a rain event, some localities showed an increase in water table of ≤4 m. In addition, the depths of water table measurements were compared with concentration data to test the hypothesis that the levels of As and F are greater in deeper parts of the aquifer.

Data Processing and Spatial Analysis

Concentrations of As and F in wells were added to the attribute table of the geographical coverage of data including location of wells and water quality. Box-plot statistical analysis was applied to the As content in wells and sediments and F in wells to identify zones of anomalous concentrations. Statistical principal component analysis (PCA) with varimax rotation was performed for each of two settings: (1) As and F concentrations and other water parameters for all wells sampled in 2011 and (2) concentrations among wells. The program XLSTAT 8.0 was used to conduct the statistical analyses. The 2011 data are shown in detail because it was the most complete data set from the sampled years. Data from other years were not included for simplification purposes.

The produced geographic information system (GIS) system documented the evolution of As and F concentrations through time once data collected in 2011 were added to previously obtained data reported in Reyes-Gómez et al. (2013). In addition, their co-occurrence and their relation to other variables, such as land use, could be substantiated with these additional data. Co-occurrence was evaluated using regression analysis after determining the significance of correlation and the efficiency of the model (Mayer and Butler 1993).

An index of differences in concentration for the two elements of interest was established using the Mexican guidelines, which are 0.025 mg L−1 for As and 1.5 mg L−1 for F. The baseline values for a trend analysis were set as those first measured (2005 for As and 2007 for F). The proposed index was named the “index of evolution and complement of As and F contamination” and is mathematically expressed as follows:

$${\text{IEC}}_{{{\text{As}},F}} = \left( {\frac{{C{\text{As}}_{i} ,F_{i} \times 100}}{{C{\text{As}}_{\text{LM}} F_{\text{LM}} }}} \right) - 100,$$

where IECAs,F represents the % difference (excess or deficiency) of the actual As or F concentration with respect to the established limit, CAsi,F i is the concentration of As or F (mg L−1) for the year of measurement, and CAsLM,F LM is the recommended limit of As or F for the zone where this index will be used. The two values of 100 represent the conversion factor to percent. The results obtained for IECAs,F are % values that deviate from the norm (IECAs,F  = 0) to positive values (concentrations exceeding the norm) or negative values (concentrations lower than the norm).

Results

Water Quality and As Species Present

As concentrations in groundwater varied between 4 and 117 and F concentrations between 1.1 and 6.8 mg L−1. Additional water parameters (e.g., pH, T, TDS, EC) were measured to characterize the water and to determine their possible influences to As and F concentrations. Water quality of samples collected in 2005, 2007, and 2010 are reported in Reyes-Gómez et al. (2013). Results for the 35 samples collected in 2011 are listed in Table 1. The pH of the water was neutral to slightly alkaline (pH 6.5–8.5) in the majority of sampled wells with wells no. 23 through 25, in the central part of the basin, having pH values >8.5. EC varied over a wide range of values from 482 to 3,900 mS cm−1 (TDS 235–2,535 mg L−1), and seven wells contained brackish water (TDS > 1,000 mg L−1). Table 1 also shows the depth to the water table. The depth to the water table increases toward the center of the basin with two notable exceptions: the shallow depth observed in wells no. 25 and 26, which is explained by these wells being hand-dug wells over a spring, and the wells around the city of Aldama, which exhibit large variations from one well to the next likely as a result of varying groundwater extractions used to irrigate the crops and also influenced by water from the Rio Chuviscar diverted to irrigate parcels. The Rio Chuviscar flows through the southern part of the city just outside the study area.

Table 1 Water-quality parameters for groundwater sampled in 2011
Full size table

Principal component analysis (PCA) applied to the water-quality parameters showed that three groups of variables—GI, GII, and GIII—combined were able to explain 74.23 % of the variance (Fig. 2). GI grouped As and F concentrations, T, and pH; GII comprised Eh; and GIII combined EC, TDS, and salinity. Because the data set is relatively small and PCA is best applied to large data sets and highly correlated variables, we also applied Pearson correlation to the data (see “Spatial Distribution of As and F in Groundwater” section) to verify the co-occurrence of As and F.

Fig. 2
figure 2

PC results for water-quality parameters using varimax rotation. Of the variance, 76.7 % was explained with groups G1, GII, and GIII. Sal salinity, TH total hardness, Alk alkalinity, D depth to water table

Full size image

Spatial Distribution of As and F in Groundwater

Spatial distribution of As concentrations for groundwater collected in 2011 is shown in Fig. 3. As concentrations exceeded the Mexican norm of 0.025 mg L−1 with the greatest concentrations occurring in two wells located in the central part of the basin (nos. 23 and 24). Similar to the pattern obtained for As distribution, the distribution map of F concentrations (Fig. 4) shows concentrations exceeding the Mexican norm of 1.5 mg L−1 over a large portion of the study area with the highest concentration near the center of the basin. One well (no. 24) contained the second highest content of both As and F. From 27 wells within the basin, only four at the south and easternmost parts of the study area were lower than the F norm. The distribution of As and F concentrations for years 2007 and 2010 showed approximately the same patterns.

Fig. 3
figure 3

Spatial distribution of As concentrations (μg L−1) of samples collected in 2011

Full size image
Fig. 4
figure 4

Spatial distribution of F concentrations (mg L−1) of samples collected in 2011

Full size image

Pearson correlation was applied to the 15 water levels recorded in 2011. A significant correlation of As concentration with depth was found for As (r = −0.429, p < 0.05), but no significant correlation was found for F. For the four water levels recorded in 4 years of measurements, with all data combined (Table 2), the trends indicated that greater contents of As and F were associated with more shallow depths (r = −0.38, p < 0.02 and r = −0.30, p < 0.07, respectively).

Table 2 Correlation coefficients obtained for the water-quality values in Table 1
Full size table

Co-occurrence and Evolution of As–F

The Pearson correlation analysis (Table 2) showed a strong association between As and F and of these elements with pH. The co-occurrence was confirmed for each of the 3 years of measurements after correlations for 2007, 2010, and 2011 yielded r = 0.70, 0.84, and 0.55, respectively. These values validated previously reported results of high correlation between As and F in this aquifer (e.g., Reyes-Gómez et al. 2013).

To document the evolution of As and F content in groundwater for this aquifer, four wells were used as monitoring wells and were sampled in 2005, 2007, 2009, and 2010 (Figs. 5, 6). Depth to water table was recorded in all but one of these wells. From the four monitoring wells, one well consistently surpassed the Mexican norm with respect to As (positive IECAs index) and three with respect to F, whereas one well, near the town of Aldama (well no. 12), showed significantly lower concentrations of As and F in every sampled year. The pattern followed by most wells was an increase in IECAs,F (As and F) from 2007 to 2010 followed by a slight decrease in 2011.

Fig. 5
figure 5

IECAs variations with time for monitoring wells sampled in 2005–2011. Negative values of SPI-3 represent dry periods and positive values wet periods. IEC index accumulated concentration of As

Full size image
Fig. 6
figure 6

IEC F variations with time for samples collected in 2007–2011 for four monitoring wells. Negative values of SPI-3 represent dry periods and positive values wet periods. IEC index accumulated concentration of Fluoride

Full size image

To evaluate the relationship of the abundance (or scarcity) of precipitation to As and F concentrations in groundwater, SPI-3 was plotted (bars) on the same graph with IECAs,F in Figs. 5 and 6. Figure 5 shows that the As in wells no. 7 through 9 increased slightly during the initial phase of the dry period 2010 to 2011 and then decreased to the original levels. Well no. 12, with lesser As content, remained relatively unchanged with time (SPI-3 > 1.0; r = 0.10–0.45 between As and SPI-3). With respect to F, the relations were found to be statistically not significant. Well no. 12 remained with low F content that was relatively unchanged with time as it did for As. After comparing IECAs,F and SPI-3 patterns, a relationship between IECAs,F and SPI-3 was far from evident showing that weather alone was not able to explain the variations in As and F concentrations. When comparing the pattern of IECAs,F with the decrease of water table depths, however, the wells that experienced a greater decrease in depth (wells no. 7 and 12) in 2010 and 2011 were the wells that remained near or within the recommended limits for both As and F. This possible relationship of IECAs,F index and an increase in water extraction must be further tested because the number of data was too small to draw any conclusion of statistical significance. The influence of a recent establishment of a settlement of farmers (year 2010, Fig. 1) and subsequent increase in water demand for drinking and irrigation purposes was thus inconclusive.

It is important to note that monitoring wells were clustered in the southernmost part of the study area, near the town of Aldama, and for this reason these results are not representative of the entire study area; well no. 12 was the northernmost location of the monitoring wells.

As in Sediments

As content in surface sediments within the basin varied between 1.8 and 28.3 mg kg−1 (Fig. 7). A box-plot analysis was used to identify anomalous concentrations (>14.3 mg kg−1) and plotted in a map: Anomalous concentrations are shown with large crosses and are concentrated in the south and south-central parts of the basin. To test if sediments with greater As concentrations affected nearby wells (nos. 10 through 14 and 19 through 22, Table 3), a correlation of As concentration with distance to its closest high As sediment sample was performed. The results showed a greater concentration of As in wells that are farthest from the points of reference (S1, S7, and S11), i.e., toward the center of the basin (r > 0.91, p < 0.001). Wells in the central zone showed a trend of less concentration of As with increasing distance between wells no. 19 through 23, 30, 32, and 35 and reference point S17 (r = 0.81, p < 0.017). In both cases, the highest concentration of As in well water was observed toward the center of the basin, which is expected from the point of view of preferential direction of groundwater flow (SARH 1981).

Fig. 7
figure 7

Spatial distribution of As in sediments (mg kg−1)

Full size image
Table 3 Parameters used in determining the relation between As in wells and As in sediments
Full size table

Discussion

The concentrations of As and F in the aquifer varied greatly within the basin and became less pronounced with time. For As, 55 % of the samples were >0.025 mg L−1, and 83 % were greater the F limit of 1.5 mg L−1 (Table 2 and Figs. 3, 4). The co-occurrence of As and F was verified for water samples collected in 2011 (r = 0.55, p < 0.05), and once these were added to previous data (2007, 2010) the results of these years combined showed an even greater degree of co-occurrence (r = 0.59, p < 0.05).

Co-occurrence can be explained as a result of the chemical weathering of rocks rich in As and F present in and around the basin. The rocks exposed in the basin include tuffs, rhyolites, shales, and carbonates, samples of which were analyzed for As and F content by Reyes-Gómez et al. (2013). A high content of As was found associated with volcanic glass present in rhyolites and tuffs, whereas small amounts of fluorapatite occurred in rhyolites and in veins associated with the emplacement of these volcanic rocks as well as in shale, thus asserting that these rocks can be the source of As and F in groundwater.

A high correlation was also obtained for F concentrations to pH (Table 2) agreeing with reported increase in dissolution of carbonate rocks and release of As from iron oxides occurring in aquifers under reducing conditions (Thomas et al. 2008; Sposito et al. 2011; Kim et al. 2012; Karro and Uppin 2013). Although reducing conditions may prevail in the deeper parts of the aquifer, at the groundwater extraction depths, however, our measurements of Eh indicated oxidizing conditions, which agrees with conditions found in endorheic basins of arid areas elsewhere (Smedley et al. 2002).

Both volcanic glass and fluorapatite are present in the rhyolites that abound in the southwest part of the study area (Reyes-Gómez et al. 2013). Furthermore, As adsorbed to iron hydrates may mobilize under the reducing conditions prevailing in deeper parts of the aquifer (Thomas et al. 2008; Kim et al. 2012). As a result of groundwater extraction and/or vertical flows within the aquifer, solubilized As and F may be pushed up to more shallow parts (Wang et al. 2009) where they remain in solution after coming into contact with oxidizing conditions (CONAGUA 2008; Brito-Castillo et al. 2010; Scanlon et al. 2009). In our study, spatial analysis of surface sediments showed greater As content near the sources of As-rich rocks in the south-central part of the area (Fig. 7) and then again toward the center of the Laguna El Cuervo watershed, which agrees with the direction of groundwater flow and in support of the processes mentioned previously and with the geohydrological model proposed for this basin (Reyes-Gómez et al. 2013).

As and F also fluctuated with time. The fluctuation, however, did not correlate in a significant manner with either the depth to water table or with droughts. In addition to the relatively short period that this basin has been monitored, other factors that may have hindered a clear view of the relationship between water quality and land-use change are the location of monitoring wells (covering only one corner of the study area) and slow groundwater flow.

Conclusion

The distribution of the potentially toxic elements As and F in groundwater of the aquifer showed a large area where As concentrations surpassed the Mexican reference limit for drinking water (25 µg/L−1) with concentrations >100 µg L−1 in the central part of the basin. For F, concentrations were > 1.5 mg L−1 in most of the aquifer with values >5 mg L−1 in the same locations where As were detected at the highest levels. One well (no. 12) with consistently low content of As and F points out to a different source of groundwater, possibly leakage from the Rio Chuviscar, a river flowing directly south of the study area.

Co-occurrence of these two contaminants tested positive by PCA and Pearson correlation analysis. Because both As and F are at or near contamination levels, co-occurrence must be considered while planning for remediation actions.

The As and F concentrations were found to vary with time based on four monitoring wells sampled in 2005, 2007, 2010, and 2011. The As and F concentrations observed in the aquifer could not be fully explained by depth to the water table, amount of precipitation, or association to surface sediments. This suggests that these variations may be a combination of the above-mentioned factors together with contributions from other processes—including adsorption/desorption and underground recharge of the aquifer by vertical, horizontal, and back flows—as well as a definite tendency to increase toward the center of the basin for both As and F.

The impact of droughts and land use on As and F concentrations yielded inconclusive results, possibly because the 7-year monitoring interval was insufficient to show a trend. Continued monitoring of these wells and adding monitoring wells to cover other locations within this aquifer are necessary to fully understand the conditions that cause As and F to be present in high concentrations.