Introduction

Although there are carbonate aquifers in southern Spain functioning in an unaltered regime or with a low exploitation, many of them are nowadays affected by an intensive use of its groundwater resources, as defined by Llamas and Custodio (2003). They are normally groundwater bodies near lowland—campiña—towns with a high water demand for agriculture and for population supply. Increasing groundwater withdrawal has led to lowering of the water table in some aquifers and even to the desiccation of springs and groundwater-dependent ecosystems (e.g., wetlands). As a result, this type of effect on the springs associated with carbonate aquifers in the south and southeast of Spain is not uncommon. The reduced size of these aquifers (usually <100 km2), the high permeability due to the strong karstification in the zone of dynamic storage and the strong variability of the recharge make the exploitations unsustainable even if the pumping rates are less than the mean recharge. Declines in the piezometric surface below the level of spring drainage are usually generated, which are even more evident during the frequent drought periods.

As for quality, there is a trend toward an increase in total salinity, as well as a risk of contamination by nitrate, which should be controlled. Likewise, at the borders of the aquifer that coincide with the Triassic materials containing gypsum and halite, a significant increase in extraction entails a risk of contamination by chloride and sodium (Martos-Rosillo et al. 2009).

As many authors have stated, sustainability of groundwater bodies is a necessary goal for the long-term welfare of both humans and the environment (Henriksen et al. 2008; Sophocleous 1997, 2000). In this sense, abundant documentation has reported on the development of indicators to quantify the degree of exploitation to which a groundwater body is subjected. In fact, groundwater sustainability indicators, based on monitoring programs, maintain the sustainable management of groundwater resources, help to analyze the extent of natural processes and human impacts on groundwater systems in space and time and facilitate communication and public participation in resource planning and policy (Vrba and Lipponen 2007). The objective of such indicators is to define, by means of diverse variables and synthetic expressions, specific aspects of the quantitative and qualitative state of groundwater and to permit comparisons over different aquifers as well as a proper planning and management of the available water resources. Such indicators are not dimensional and are obtained from data that are relatively easy to gather, providing information about the state of an aquifer and the possible tendencies and/or impacts taking place in it.

Nonetheless, given the great number of variables and uncertainty associated with the terms employed, previous to their application, it is crucial to have sufficient knowledge of the workings of the aquifer under study (Lavapuro et al. 2008; Pernía et al. 2005; Pernía and Lambán 2007). The approximation and definition of such indicators should always be approached with a consideration of the hydrodynamics and specific characteristics of each aquifer, avoiding exclusively statistical approximations. The adequate definition of these indicators is likewise essential, given that they are oriented toward a correct management and planning of groundwater resources, taking into account social aspects (availability and use), economic factors (development, catchments and protection) and environmental concerns (drops in levels and effects in ecosystems or contamination).

The objective of the present study is, firstly, to apply a series of groundwater sustainability indicators—both quantitative and qualitative—to the use of groundwater resources, in accordance with the European Water Framework Directive (WFD) and the Groundwater Daughter Directive (GDD) (EU 2000, 2006) and, secondly, to characterize a clear example of extensive groundwater exploitation in a carbonate aquifer—Sierra de Becerrero—by means of these indicators. Finally, some recommendations of sustainable management in this type of groundwater bodies are proposed.

Study site

The Sierra de Becerrero is situated in southern Spain, in the central area of Andalucía (within the province of Sevilla) and in the basin of the Guadalquivir River (Fig. 1). It covers a total surface of some 51 km2 and has topographic limits ranging between 460 and the 844 m a.s.l. of the Cerro de Becerrero.

Fig. 1
figure 1

Hydrogeological map of the Sierra de Becerrero aquifer

Full size image

The uses of the groundwater are for urban supply and irrigation. Groundwater urban supply cover the demand of 24,171 inhabitants (3,599,849 m3/year) with 149 m3 per person per year, while there are 250 ha of olive groves irrigated with groundwater from the aquifer that totalized a volume of 2,000 m3/ha per year (CAP 2000).

It lies in a zone with a semi-arid to humid climate (mean precipitation, 500 mm/year), characterized by a very irregular inter-annual distribution of precipitation: short periods of very intense rainfall alternating with long dry periods (drought) that are less intense (Martos-Rosillo 2008).

From a geological standpoint, the materials are aged from the Triassic to the lower Miocene (Cruz-Sanjulián and Divar 1986). The Triassic materials are represented mainly by multicolored clays and marls with gypsum and halite and layers of sandstone and black dolomite. The lower and middle Jurassic rocks form the main aquifer, which is represented by 400 m of limestone, overlain by a thick (700 m) sequence of marls with an age ranging from the late Cretaceous to the Burdigalian. The structure of the Sierra de Becerrero corresponds to an anticlinal dome, with a northward vergence. The most representative geomorphological features are the presence of exokarstic forms, especially karren fields and the absence of known endokarstic forms.

The aquifer of the Sierra de Becerrero is confined at its borders and unconfined in the central zone, where there are 26.2 km2 of permeable outcrops. Geological maps (Cruz-Sanjulián and Divar 1986), different geophysical prospection campaigns (IGME 2006) and the analysis of core drills (Martos-Rosillo 2008) have indicated that the boundary of the aquifer is closed to groundwater flow. The impermeable substrate is made up of clays and marls of Triassic age. Infiltration of rainwater recharges the aquifer and discharges through springs, now regulated by catchment works. These works and the groundwater withdrawal associated to them have evidenced an unremitting decrease in both groundwater levels and water quality. The piezometric level is currently between 446 and 453 m a.s.l. which implies that depth varies between 18.1 and 116.7 m from the surface, and groundwater flow in S direction. A piezometric control network is available, with groundwater quality and hydrometric data, plus historic data obtained from the 1970s until the present. The control points correspond mostly to wells and piezometers. At present, there are 12 points of monthly piezometric control and biannual quality control. The gauging stations comprise 2 points with continuous recordings since 2001.

In Fig. 2 it can be observed that the intensive exploitation of the aquifer lowers the piezometric level below the position of the main springs (drainage level) which, in turn, cause their desiccation. Comparison of piezometric level with the precipitation shows a good correlation, as well as a rapid recovery of the piezometric level and the spring flows in rainy years.

Fig. 2
figure 2

Evolution of the piezometric levels, pumping rates and comparison with precipitation in the Sierra de Becerrero

Full size image

The waters of the aquifer have a major ion content of calcium and bicarbonate, with an electrical conductivity (EC) average of 480 µS/cm—between 269 and 813 µS/cm. In some areas of the aquifer and after periods of drought, magnesium and sulfate contribution to the water ionic content can also be important. This temporal variability can be observed in Fig. 3 (well #24). The figure shows the time series of the main physicochemical parameters for the period 1977–2006, together with the evolution of the piezometric level in well #24 and a hydrograph of the spring located next to the well (so-called Gilena spring). The periods in which the spring is inactive coincide with the piezometric level in #24 is less than 467 m a.s.l. The water of this spring is calcium bicarbonate and calcium–magnesium bicarbonate, with an average EC of 502 µS/cm—between 370 and 565 µS/cm. During the drought of the early eighties of last century, water obtained from #24 exhibited on several occasions, calcium sulfate magnesium bicarbonate chemical facies. During dry periods, the decline in the piezometric level is always followed by an increase in water salinity. Increasing salinity is due to the increased concentration of chloride, sodium, and magnesium sulfate, and a decreased concentration of bicarbonate and calcium. Nitrate concentration in the groundwater has experienced a slight increase, which could probably be associated with urban pollution, given the proximity to the town of Gilena (Martos-Rosillo 2008).

Fig. 3
figure 3

Evolution of the piezometric level in the well #24 and of the main physical–chemical parameters of the groundwater, compared to daily recharge and the drainage flow from the Gilena spring (Martos-Rosillo 2008)

Full size image

Materials and methods

Development of groundwater sustainability indicators (henceforth, GWSIs) has been taken up by UNESCO under the Sixth Phase of the International Hydrological Programme (IHP), Theme 2: Integrated Watershed and Aquifer Dynamics. This effort also draws in the expertise and support of the International Atomic Energy Agency (IAEA) and the International Association of Hydrogeologists (IAH). These organizations together also utilize the support of a group of select professionals and have formed a groundwater indicators working group (WG) composed of UNESCO, IAEA and IAH experts. The indicators proposed for this group although simple, are both scientifically based and policy-relevant (Vrba and Lipponen 2007).

A selection of the indicators proposed by this group was applied in Sierra de Becerrero (Spain), taking into account the intensive use of its groundwater resources and the specific characteristics of this aquifer. Groundwater indicators selected are: (a) total groundwater abstraction/groundwater recharge, (b) total groundwater abstraction/exploitable groundwater resources, (c) groundwater depletion and groundwater quality. The last two indicators have been enlarged and developed for Geological Survey of Spain (IGME) experts (Pernía et al. 2005; Pernía and Lambán 2007) in regards to the previously established by the WG. In this work, we selected four main GWSIs based on their importance in the field of groundwater sustainability and because they proved to be the most reliable, based on the data collection and methodology utilized. The GWSIs selected are described below.

Total groundwater abstraction with respect to groundwater recharge: (abstraction/recharge) × 100

The main sources of recharge are infiltration of rainwater, infiltration from surface water bodies, irrigation losses, and losses from urban zones (GIWG 2004). The estimation and/or calculation of recharge involves the combined application of diverse methods (soil water balance, numeric simulation, chloride balance, Darcy’s law, and others) selected according to the climatic and hydrogeological conditions and available data. In any case, considering both the possible natural variability and the high number of variables and parameters involved, uncertainty in the estimation of the recharge is always a possibility. It is also important to indicate the time scale employed in evaluating the recharge. In semi-arid regions a short but intense rain event may be more significant than a regular precipitation of greater duration. Therefore, the use of mean annual values for precipitation to estimate recharge should be carefully considered. The total output of groundwater encompasses the total water extracted from the aquifer by means of wells, springs or lateral discharges, for urban supply, agriculture, industry or other uses (GIWG 2004). In many countries, some of these outputs are already known, as it is obligatory to register groundwater withdrawals in order to obtain the corresponding extraction permission. For hand-dug wells with a domestic use, the data is generally obtained by means of qualitative estimates. Information for natural outputs (e.g., springs or groundwater discharge) may be calculated when sufficient data are available.

It is recommended to work with numeric ranges, in order to take into account both the variability and the uncertainty associated with the terms used.

Finally, the concept of induced recharge may be defined as such: when the cone of depression of a pumping well lowers the adjacent water table below a stream, a hydrological process known as induced recharge can be produced. As a result, the stream begins to lose water to the adjacent groundwater aquifer near the well.

For the indicator total groundwater abstraction with respect to groundwater recharge, two principal scenarios may be distinguished: (1) sustainable use of the groundwater with respect to quantity (total output ≤ recharge + induced recharge) and (2) non-sustainable use of the groundwater (groundwater mining) with respect to the amount (total output > recharge + induced recharge). This indicates, in a preliminary manner, the degree of sustainability with regard to the use of groundwater, although aspects related to quality and/or affections to groundwater-fed ecosystems are not considered.

Total groundwater abstraction with respect to groundwater resources: (abstraction/exploitable resources) × 100

The term corresponding to the total groundwater abstraction is the same as that used in the indicator described above. The groundwater resources are defined as the amount of water that can be extracted from an aquifer in light of socioeconomic needs at the time, while maintaining ecological conditions (GIWG 2004). Bearing this definition in mind, as well as the variations existing in the natural recharge, it is proposed that we consider this term as a percentage of the volume of water situated above the drainage level, i.e., the volume of water that may be extracted without damaging the aquifer dynamics. To establish this percentage, we will take into account both the hydrogeological and climatic conditions (recharge) and the pace of exploitation. Therefore, from a quantitative standpoint (IGME 2004):

$$ R_{d} /t = \left( {R_{v} /t_{r} } \right) - Q_{e} , $$

where:

R d /t :

Exploitable resources

R v :

Volume of water between the piezometric level and the minimum surge level

t r :

Time estimated for extraction of reserves in view of recharge

Q e :

Flow necessary to maintain ecological or environmental conditions.

Groundwater depletion

In order to determine both the actual state and the temporal trends of the groundwater storage in the aquifer under study, the following methodology is proposed (IGME 2004):

  1. (a)

    The representation of the temporal series of groundwater withdrawal, i.e. pumping flows, piezometric levels and precipitation in the area allows us to obtain information about the hydrodynamics of the aquifer and of the exploitation trends.

  2. (b)

    The gathering of graphical and numerical indexes for both the groundwater extraction flows and the piezometric levels.

With respect to the pumping flows, we propose the representation of data on a daily scale with the aim to evaluate and quantify volumes, using different methods for the decomposition of the hydrograph. With regard to the piezometric levels, we propose obtaining trends for the median (numeric value separating the higher half of a frequency distribution, from the lower half), quartile evolution and accumulated deviation of monthly precipitation with respect to the mean piezometric level. Then, we can obtain a so-called “piezometric evolution index”. Moreover, it would be advisable to obtain a frequency diagram for the last campaign carried out in the aquifer with respect to the mean value of the control period (EEA 1998) and to compare the values measured in the last campaign with the mean, median, minimum, maximum and standard deviation corresponding to the control period. Then, we can obtain a so-called “piezometric situation index” for that campaign. This piezometric situation index is proposed in order to obtain straightforward information that could be easily interpreted by non-specialized personnel.

Groundwater quality

Groundwater quality aspects must be observed, because groundwater quality changes due to various human activities may affect the overall groundwater exploitability. In order to adequately define the quality of groundwater, it is necessary to differentiate between the quality of natural origin and the modifications of this quality as a consequence of exploitation and/or contamination of groundwater due to human activity. We define “natural origin” as the range of concentrations in the water of a certain element, species or substance present and deriving from natural geological, biological or atmospheric sources (EESD 2003). Notwithstanding, the “natural origin” is different for every aquifer and, for instance, depends on the particular combination of its geological and hydrodynamic characteristics. The definition calls for historic data series as well as knowledge—which is generally lacking—of the hydrodynamic functioning of the aquifer.

The methodology and quality GWSIs proposed here is as follows (Pernía and Lambán 2007): the definition of the present quality and risk of contamination of the groundwater may be described on the basis of the last sampling campaign(s). After eliminating erroneous data and/or inadequate analyses, the predominating hydrogeochemical facies are defined (regional hydrogeochemical characterization), and the water is classified, taking into account both its potability (according to Directive 98/93/CE and Spanish Water Law) and its risk of alkalinization and salinization from the soil (according to the norms established by Thorne and Peterson 1954).

Aside from the descriptive parameters (temperature, pH, BOD and EC) (EESD 2003) and the major ions (Ca, Mg, Na, K, HCO3, Cl and SO4), the following parameters will be analyzed as possible indicators of contamination: NO3, NO2, NH4, PO4, (EEA 1998), dissolved organic carbon (DOC) and total organic carbon or TOC (EESD 2003). The inorganic substances that most frequently lead to problems in water quality are Fe, Mn, F, Cl, SO4 and trace elements such as As, Ba and Ni (of natural origin) as well as NO3, K, Na, Cl, SO4, NH4 and PO4 (anthropic origin) (EESD 2003).

Depending on the availability of data and pressures or tensions to which the groundwater is subjected, the following may also be considered: heavy metals, organic substances, pesticides and others. The pressure exerted upon groundwater will depend mainly on the density of the population, the use of the soil, the focal points of contamination, and/or the use of fertilizers. The parameters indicating contamination or a risk thereof will be selected in agreement with the classification of the chemical state proposed in Table 1 (modified from Costa 2002 in IGME 2004), as well as indicators of quality in agreement with the specific characteristics of a given aquifer.

Table 1 Risk of contamination and chemical state of the groundwater according to its use
Full size table

Such a classification is based on the concept of “contamination” (according to the Spanish Water Law), which does not consider possible problems associated with the natural quality of groundwater. The term V N corresponds to a range of concentrations resulting from the spatial and temporal variability of the chemical composition of the aquifer water (Fig. 4), with certain values possibly surpassing the limit value (V L ) in view of the water’s use. The risk of contamination depends on the use of the water, for which reason the limit values (V L ) are to be established in light of that use. In order to evaluate the present state of groundwater quality, it is proposed that one obtains, for the parameters indicative of contamination, a diagram of accumulated frequencies derived from the values registered in the last sampling campaign with respect to the mean value during the control periods. Moreover, it is proposed that, for these parameters, the so-called quality index of the chemical state be obtained; this is defined as the quotient between the concentration of the parameter of reference at the present time and the concentration assigned to that parameter in order that the water body may be designated for a particular water use (or that assigned as the threshold for a good chemical state).

$$ {\text{IC}}_{\text{p}} = {\frac{[P]}{{V_{L} }}} $$
ICp :

the index of quality of the chemical state of the parameter “p”

[P]:

current concentrations of the parameter “p” (last sampling campaign)

[V L ]:

value limit depending on use or threshold value for the chemical state of parameter “p” (according to WFD and GDD).

given the following weighting of values:

Fig. 4
figure 4

Schematic representation of the accumulated frequency with respect to the logarithm of the concentration of each indicator, showing different types of distributions and processes (Manzano et al. 2004)

Full size image

  • ICp > 1—Insufficient quality or poor chemical state

  • 0 ≤ ICp ≤ 0.5—Quality between a concentration higher than the limit of analytical detection and the mean concentrations of required limit and/or threshold

  • 0.5 < ICp ≤ 1—Quality between a concentration of required limit and/or threshold and the concentration of the required limit value or threshold.

Secondly, the temporal hydrogeochemical characterization and determination of tendencies are completed. A joint treatment of all the chemical analyses at all the points available throughout the control period is undertaken in order to obtain the most significant statistical parameters. For each control parameter, diagrams of accumulated frequencies are plotted. The mean value will be used as the level of regional reference for the comparison of different elements among themselves. The observed populations do not often present a normal or lognormal distribution due to the coexistence of more than one process (Fig. 4). For the determination of tendencies, the temporal evolution of the previously selected parameters at the control points is displayed, and the annual evolution of the mean, median, standard deviation, minimum, maximum and quartiles (EEA 1998) are calculated and graphically represented. Whenever possible, instead of the annual evolution, the evolution between the different sampling campaigns should be considered. Likewise, by means of linear regression, the curve of the best fit corresponding to the temporal evolution of the median for each one of the selected parameters is calculated. Insofar as it is possible, we distinguish between: (a) tendencies of natural origin (variations on the aquifer scale or, on a smaller scale, due to the heterogeneity of the medium), (b) tendencies caused by exploitation of the aquifer (acceleration of natural tendencies due to the increase in the flow speed), and (c) tendencies related to contamination (point or diffuse). We propose that an index of the evolution of the chemical state be used in order to establish a criterion for assessing the evolution of the chemical state at a control point or for the groundwater body. This index is the quotient between the difference of the present concentration of the parameter and the value of the natural supply for that parameter (median of the range of concentrations) and the difference between the maximum concentration of that parameter during the control period and the value or natural supply for that parameter (median of the range of concentrations),

$$ {\text{IE}}_{\text{P}} = {\frac{{\left[ P \right] - \left[ {V_{N} } \right]}}{{\left[ {P_{ \max } } \right] - \left[ {V_{N} } \right]\,}}} $$
IEp :

Index of evolution of the chemical state of the parameter p

[P]:

Current concentration of parameter “p” (last sampling campaign)

[P max]:

Maximum concentration measured during the control period

[V N ]:

Concentration of natural supply, median of the range of concentrations (if not known, to be substituted by the minimum concentration of parameter p during the control period).

with the following weighting:

IEp = 0:

Concentration equal to the natural supply (median range of concentrations) or the minimum concentration of the control period

0 ≤ IEp < 0.5:

Concentration between the natural value (median range of concentrations) or the minimum and average of the control period

0.5 < IEp < 1:

Concentration between the average value and maximum of the control period

IEp = 1:

Concentration equal to the maximum value during the control period.

To avoid the influence of analogous maximum values (e.g., sources of focal contamination, analytical errors), this index of evolution is proposed, in addition to a consideration of the median concentration corresponding to the control period. The quality index describes the risk of contamination and type of contamination with regard to the use of the water or the chemical state of a specific point or water body. The index of evolution allows for an evaluation of the current state in comparison with the control period.

Results

Total groundwater abstraction with respect to groundwater recharge: (abstraction/recharge) × 100

We evaluated the recharge in the aquifer by two different methods: the daily balance of water in the soil for the period 1977–2006 (Milly 1994a, b) and the chloride balance (Eriksson and Khunakasem 1969), giving an average recharge of 200 ± 140 mm/year (i.e., 5 ± 4 hm3/year). Water outputs from this aquifer takes place mainly by means of the exploitation of extraction wells although, during wet years, the springs draining the aquifer may be temporally active. For the supply of the 24,200 inhabitants that depend on this aquifer, 3.6 hm3/year are extracted, to which we must add another 0.4 hm3/year that is directly extracted from the aquifer for industrial use (limestone quarries). Meanwhile, groundwater proceeding from the aquifer is used to irrigate olive groves with an annual withdrawal of approximately 2,000 m3/ha per year, which amounts to a consumption of 0.5 hm3/year (IGME 2006). Therefore, the total extraction comes to some 4.5 hm3/year. If we take into account the variability of the recharge resulting from the hydrometeorological balances and consider as the total output the sum of the extractions (4.5 hm3/year) together with a minimum of 0.5 hm3/year to be drained by the main springs, we arrive at the range of values for this indicator presented in Table 2. The results shown in Table 2 confirm that values for this GWSI ranging between 48 and 417% demonstrate that non-sustainable use is being made of the groundwater during dry periods, as well as in years with medium precipitation, the use is very close to surpassing the “threshold of non-sustainability”.

Table 2 Range of values obtained for the indicator “Total outputs of groundwater with respect to Recharge” in terms of the period and the average recharge produced throughout the period analyzed (1977–2006), in the wet years (mean precipitation + 1 standard deviation) and in the dry years (mean precipitation − 1 standard deviation)
Full size table

Total groundwater abstraction with respect to groundwater resources (abstraction/exploitable resources) × 100

As previously stated, the total output (extractions plus natural release) was estimated at 5 hm3/year (IGME 2006). The exploitable resources are given by

$$ R_{d} /t \, = \left( {R_{v} /t_{r} } \right) - Q_{e} $$

where:

R v :

Volume of water between the piezometric level after a wet hydrological year (1996, 1997, 2003) and the level of the minimum groundwater level (460 m). The average volume of rock obtained is about 256 hm3, so if an effective porosity of 0.03 is considered (Martos-Rosillo 2008), the resulting volume of water is 7.7 hm3. This porosity was obtained from the analysis of the porosity of the matrix and from the study of the variations of groundwater storage in the aquifer

t r :

Time in which the reserves are to be extracted, in view of recharge

Q e :

Natural discharge considered to be minimal, according to specific environmental conditions: 0.5 hm3/year (Vázquez-Mora et al. 2001).

The range of values resulting from the application of this indicator is presented in Table 3. In one average hydrological year—with an approximate recharge of 5.2 hm3/year—the time of extraction of R v should be equal to or greater than 1.4 years, if the use of the groundwater is to be sustainable in relation to the aspects of quantity. For shorter time periods, the development is not sustainable.

Table 3 Range of values obtained for the indicator “Total extractions with respect to available groundwater resources” in terms of time of extraction of reserves
Full size table

Groundwater depletion

In the Sierra de Becerrero there are no surface watercourses, nor any evidences of previously existing natural springs. This situation allowed us to estimate the evolution of the groundwater flows. For this reason, we represent only those points with periodic data of the piezometric level, as available (Fig. 2). In Fig. 5 we can observe the groundwater level evolution during the study period and the evolution of the median value (Fig. 5a), the evolution of the quartiles (Fig. 5b), and the accumulated deviation of monthly precipitation with respect to the median piezometric level (Fig. 5c). It is important to obtain both the median and the average in order to best compare the asymmetry of the frequency distribution. We present the indexes of the piezometric state applied to the Sierra de Becerrero in Fig. 5. The frequency diagram obtained from the difference between the values measured in 2006 with respect to the mean value for 1997–2006 (Fig. 6b) exhibit a generalized piezometric dropdown trend. It can be observed that the piezometric level is 8 m below the median level (Fig. 6a). Meanwhile, the comparison with the rest of the statistic parameters corresponding to the period 1997–2006 (Fig. 6b) points to a situation of proximity between the values obtained in 2006 and the 10th percentile of the analyzed period.

Fig. 5
figure 5

Indexes of piezometric evolution: a curve of the median, b evolution of the 25th, 50th and 75th percentiles and c accumulated deviation of monthly precipitation with respect to the median of the piezometric level (1997–2006)

Full size image
Fig. 6
figure 6

Indexes of the piezometric situation :a frequency diagram for the year 2006 with respect to the mean value of the period 1997–2006 and b comparison of the monthly values in the year 2006 with respect to the median, 10th and 90th percentiles and standard deviation during 1997–2006

Full size image

The analysis of the piezometric evolution showed a particular hydrological behavior. First of all, there is an evident decaying tendency for the piezometric level during the whole study period, and secondly, the recovery of the piezometric level in every piezometer is very fast after rainy periods. Both aspects will be discussed in detail later on in this paper.

Groundwater quality

Quality during the campaign of December 2006 and risk of contamination

If we apply the abovementioned methodology to the aquifer of the Sierra de Becerrero, we observe that, in December 2006, the groundwater had a calcium bicarbonate composition with significant chloride and sodium contents (Fig. 7). This water is good enough for human consumption and has a low risk of soil alkalinization and salinization (norms by Thorne and Peterson). Nitrate concentrations ranged from 10 to 34 mg/l, with a mean of 23.5 mg/l (47% V L ), implying a low risk of NO3 contamination. When considering the maximum values (68% V L ), the contamination risk is moderate. Similarly, when a total of 17 analyses of NO2, NH4 and PO4 were performed at the main piezometers (2006), all resulted in concentrations below the detection limit. The dissolved organic carbon (DOC) was analyzed and found to be between 0.1 and 0.7 mg/l of O2. The parameters for contamination selected are NO3, Cl and Na, and those for quality are EC, HCO3 and Ca.

Fig. 7
figure 7

Current chemical composition of the groundwater of Sierra de Becerrero (December 2006)

Full size image

Hydrochemical characterization and trend determination

We carried out a joint treatment of the 322 chemical analyses available (1976–2006), resulting in the diagrams of accumulated frequencies presented in Fig. 8. We confirmed that the water from the aquifer of the Sierra de Becerrero is calcium bicarbonate and that a significant variability in chloride and sodium exists (greater standard deviation). Moreover, we observed a normal distribution for bicarbonate and calcium whereas chloride and sodium showed greater asymmetry, indicating a process of mixing with salt water. At the same time, the greater variability of the nitrate concentrations can be attributed to a process of contamination. The results obtained allowed for the maintenance of the indicators of contamination and quality previously established. The indicator of quality for nitrate (0 < 0.47 < 0.5) showed a moderate risk of contamination. The indexes of quality for chloride and sodium (0 < 0.15 and 0.11 < 0.5) indicate slight effects. For the determination of tendencies, the annual evolution for the mean, median, standard deviation, minimum, maximum and quartiles (EEA 1998) of NO3, Cl and Na (GWSI of contamination) during the period 1991–2006 were evaluated (Table 4). Figure 9 displays the annual evolution of the percentiles corresponding to 25, 50 and 75% of the nitrate during the period 1991–2006. Additionally, we calculated the slope of the curve fitting the values of the median for each of these parameters. The slopes obtained were 0.50 (nitrate), −0.55 (chloride) and −0.36 (sodium). The index of evolution for nitrate equals 0.65, so, in the last sampling campaign (December 2006), we obtained concentrations ranging between the natural source and the maximum for the control period, registered in 2003. The indexes of evolution of chloride and sodium are, respectively, 0.32 and 0.37. A summary of all of the indexes related to quality is presented in Table 5.

Fig. 8
figure 8

Frequency diagram (log concentration–accumulated frequency) during 1976–2006

Full size image
Table 4 Number of sampling points, analyses performed, and annual evolution of the statistical parameters corresponding to NO3 during 1991–2006 (IGME database)
Full size table
Fig. 9
figure 9

Evolution of the percentiles and linear regression of the median for the case of nitrate (1991–2006)

Full size image
Table 5 Summary of the indexes related with quality in the Sierra de Becerro
Full size table

Discussion

Some of the GWSIs proposed in the present study have recently been applied in different countries (Girman 2007; Hirata et al. 2007; Lavapuro et al. 2008; Pernía and Lambán 2007; Traore 2008) in order to contrast aspects of the management and planning of groundwater resources at different scales, ranging from national to regional and, finally, to the aquifer scale (Vrba and Lipponen 2007). From the results of these papers, the authors concluded that the best scale for applying such type of indicators is the aquifer scale. Many of the GWSIs applied are useful for detecting negative trends in water quality, but correct application requires long-term, representative data series (Melloul and Goldenberg 1998; Vrba and Lipponen 2007). The indexes of piezometric evolution obtained for the Sierra de Becerrero underline the existence of a descending trend for the piezometric level during the studied period (1997–2006) (m = −0.0022; Fig. 5a), as well as a fast recovery of such levels with respect to precipitation (Figs. 5a, c).

The analysis of the different indicators of variation of the storage allowed us to observe a descending trend in the evolution of the piezometric levels, which is in accordance with the increasing groundwater withdrawal taking place in this groundwater body (Bredehoeft et al. 1982; Custodio 2003; Margat 1992). In the long term, it is probable that the piezometric surface will stabilize under the current pumping regime and the new hydrodynamic situation. However, the adaptation to this new equilibrium is causing the exhaustion of the springs that drain the aquifer in a natural regime. This situation implies that the present degree of exploitation is not compatible with the maintenance of the springs. In order to avoid exhaustion of the springs during the periodic drought periods, which last 4 years, the total amount of groundwater withdrawal should not exceed 1.4 hm3/year.

Regarding the qualitative indexes, the quality of the water in this groundwater body appears to be deteriorating. A certain ascending trend is observed with respect to the nitrate content in the groundwater over the past 4 years (although not reaching 50 mg/l), along with a stabilized behavior regarding the contents of chloride and sodium. Anyhow, the content of such ions in the groundwater are slightly higher for the period 2000–2006 and lower for the period 1998–2000.

Conclusions

The current degree of exploitation of the carbonate aquifer of the Sierra de Becerrero is generating both a drop in the quantity and the quality of the groundwater. The deterioration in the quality is mainly related to a nitrate increase in groundwater and the piezometric dropdown clearly affects the natural discharge of the aquifer—the springs—which were formerly principal discharge zones and are nowadays temporarily inactive. In this sense, and in view of the results obtained in this research, we affirm that the current exploitation of this aquifer is unsustainable. This is quite clearly reflected by the GWSIs put forth in the present study. The descending evolution of the piezometric levels in this groundwater body is due to a continuous and increased exploitation and to the adaptation of the aquifer to a new hydrodynamic situation. If the current degree of exploitation—representing less than the mean recharge—is stabilized, we can expect that the piezometric surface will be stabilized below the drainage level of the springs, so the natural discharge would become sporadic, even after a series of wet years. On the other hand, the present intensive use of the groundwater in aquifer generates a greater regulatory capacity and gives the inhabitants of this region a greater degree of guarantee of supply. Groundwater supply is, for the time being and in anticipation of the arrival of infrastructure for water transport connected with two major reservoirs, the only source for water supply. The maximum volume of pumping needed to maintain a minimal discharge of drainage in the main springs—estimated at 0.5 hm3/year—after the drought periods was calculated. This maximum volume of pumping should be less than 1.4 hm3/year, considerably lower than the current demand of supply to the population and the average annual recharge. To a considerable extent, both the decline in piezometric levels and the alteration of the groundwater quality could be minimized by moving the main exploitation wells away from the discharge zones, which are near the triassic rocks and the agricultural areas. If pumping wells are moved closer to the zone of aquifer recharge, where there are fewer potential sources of contamination, both the risk of contamination and the descending trend of the groundwater level could be minimized. This would obviously entail an additional economic cost (pumping the water from greater depths, as the pumps would be at a higher level), which calls for an evaluation by means of cost/benefit analysis, including the environmental costs, in accordance with the directions of the WFD.