Water and Environment Journal. Print ISSN 1747-6585

Hydrogeological characteristics of a groundwater-dependent ecosystem (La , ) Miguel Rodrıguez-Rodr´ ıguez´ 1, Francisco Moral1 & Jose Benavente2

1Physical, Chemical and Natural Systems, University Pablo de Olavide, , Spain and 2University of Granada, Granada, Spain

Keywords Abstract hydrogeochemistry; playa lakes; Southern By means of a simple water balance model, together with hydrogeochemical Spain. and morphological interpretation, the hydrogeological characteristics of a series

Correspondence of playa lakes forming an endorheic complex within the Guadalquivir river Miguel Rodrıguez-Rodr´ ıguez,´ Physical, basin in Southern Spain (La Lantejuela) have been evaluated. The lakes are Chemical and Natural Systems, University demonstrated to be groundwater-dependent ecosystems. The main source of Pablo de Olavide, C/ , km 1, Seville groundwater input to the lakes is from an unconfined detritic aquifer, the playa 41013, Spain. Email: [email protected] lakes being the natural discharge points from the aquifer within the endorheic complex. High rates of evaporation from the lakes induce a centripetal ground- doi:10.1111/j.1747-6593.2007.00092.x water flow pattern. This water body has been disturbed by a combination of extensive drainage works and intensive groundwater abstraction. There is a need for a sustainable water management strategy for the whole catchment area. It is hoped this will be an issue addressed within the Guadalquivir river basin management plan in accordance with the requirements of the European Water Framework Directive (WFD).

Introduction interest in the correct management of such ecosystems, which are very sensitive to hydrological alterations such as Throughout southern Spain, there exist numerous closed groundwater pumping or surface drainage. The Water Fra- basins hosting playa lakes, where small mountain ranges or mework Directive (WFD), which is applicable in all the hills separate semi-arid basins (Garc´ıa et al. 1997; Montes member states, encourages the investigation of ground- et al. 2004). Ephemeral playa lakes occur in arid to semi- water-dependent ecosystems (WFD-CISNo.2 2002, 2003). arid environments whenever internally drained basins are Finally, this type of methodology could help close the present formed by tectonic activity or dissolution processes, and the gap that exists between water managers and researchers floor of such basins intersects the water table (Duffy & (Borowski & Hare 2007), thanks to its simplicity. The main Al-Hassan 1988). Rainfall onto the related aquifers or the objectives to be addressed in this work are (i) to describe the catchment area of the playas enters the groundwater detritic aquifer -La Lantejuela (O-LL), about which through permeable materials overlying the impervious only a few hydrogeological studies have previously been substratum. Normally, the substratum is related to the published; (ii) to investigate the hydrological relations be- presence of Triassic sediments, which outcrop abundantly tween this aquifer and the playa lakes of La Lantejuela by in the Betic Cordillera (Vera 2004) and present very means of a simple water budget and (iii) to carry out a low hydraulic conductivity. The groundwater returns to preliminary hypothesis on the genesis of these systems. the surface in the form of springs, rivers or discharge to the playa lake system. However, a second type of circulation, which is unique to a closed basin, may also occur. Evapora- Main exposition tion concentrates the playa lake water, which turns into a brine. This brine, which is much denser, may flow into the Methods fresher groundwater. As long as brine is supplied, cellular flow circulation may take place below the playa lake Study site bottom until it reaches the impermeable substratum. This scenario is known as ‘free convective flow’ (Fan et al. 1997). The study zone is located near the town of Osuna in the The surface–groundwater relation between the playa southeast of the (, Spain) lakes and the unconfined aquifers in the area is of evident (Fig. 1).

Water and Environment Journal (2007) c 2007 The Authors. Journal compilation c 2007 CIWEM. 1 Hydrogeological characteristics of a groundwater-dependent ecosystem M. Rodrıguez-Rodr´ ıguez´ et al.

Guadalquivir river basin Seville province Spain

Andalusia

Palomarejo hill (309 m a.s.l.)

Playa lakes

La Lantejuela Blanco River Salado River 10 km Osuna Study zone Osuna weather station Corbones River ivir River Blanco River Seville Osuna

Guadalqu orre del Aguila reservoir Puebla de Cazalla

Fig. 1. Geographical sketch.

The topography is characterized by a plain that smoothly 4 m from the shore of the playa lakes. Groundwater descends towards the north, from an altitude of about samples were collected from the uppermost part of the 300 m above sea level (a.s.l.) in the proximity of Osuna to 47 wells that were included in the inventory. Field para- 150 m near La Lantejuela. Hills bordering the flood plain meters measured in surface and groundwater were elec- include Palomarejo (309 m a.s.l.), which is in the northern tric conductivity (EC25), water temperature and pH. part of the study zone (Fig. 1). Surface drainage is towards Major ions and nitrates were measured using ion chro- the west-located Corbones River and towards the east- matography and spectrophotometry techniques, at the located Blanco River (Fig. 2). The study area belongs to the hydrochemistry laboratory of Pablo de Olavide University Subbetic Units of the Betic Cordillera (External Zones) and (Seville). Geographic information system (GIS) was used to the sedimentary filling of the Guadalquivir river basin. to determine flooded surface and catchment area char- Nine water bodies formed the endorheic complex of La acteristics; several hydro-morphological surveys were Lantejuela, but the majority of these lakes are nowadays undertaken to check the results obtained. The Osuna threatened by drainage and filling and so only two of them weather station, located at an altitude of 255 m a.s.l., was are preserved and hydrologically functional (Calderon´ and selected to obtain meteorological data. A 29-year series of Ballestera playa lakes, see Fig. 2). Other anthropogenic mean temperatures and precipitation was used. Data on stresses such as pollution, eutrophication, catchment culti- daily evaporation were obtained using a ‘Class A’ eva- vation, road construction and groundwater pumping affect poration tank located at ‘Torre del Aguila’ reservoir these ecosystems. weather station (55 m a.s.l.). The equation of the water budget (Mansell & Rollet 2006) for average climatic conditions can be expressed as Field surveys, analytical procedures and water balance calculation DV ¼ P E þðSi þ GiÞSo Go; ð1Þ Sampling campaigns were carried out in 2004. Subse- quently, occasional morphological surveys were made in DV refers to the change in playa lake storage and is the 2005. Surface water samples were taken at a distance of result obtained from the water budget model (unknown

2 Water and Environment Journal (2007) c 2007 The Authors. Journal compilation c 2007 CIWEM. M. Rodrıguez-Rodr´ ıguez´ et al. Hydrogeological characteristics of a groundwater-dependent ecosystem

Ruiz Sanchez Blanco River

Consuegra Corbones Turquilla Gobierno Herrera Ballestera Calderon River La Lantejuela Verde de la Sal Calderon Grande Pedro Lopez Marchena

Estepa

Aguadulce

Osuna Pedrera Permeability and lithology River High. Sandy marls and conglomerates Medium. Sandy marls N Roads Town Low. Marls and clays 0 5 10 km Playa lake

Fig. 2. Permeability of the materials (Osuna-La Lantejuela aquifer).

variable). This component may not change on an inter- budget is a simple accounting scheme used to predict annual scale but important variations may be expected on soil–water storage, actual evapotranspiration (ET) and a monthly scale. The model can be validated by compar- water surplus by means of precipitation and potential ET ing its results (theoretical water level) with the time series at a given water-holding capacity (Thornthwaite & Math- data for the water level of the playa lake. er 1955; Jensen et al. 1990). Water surplus is the fraction P is direct precipitation onto the playa lake, estimated of precipitation that exceeds potential ET and is not stored by multiplying monthly P (mm) by the average flooded in the soil. The simple model used here does not distin- area at each time interval to obtain P (m3). guish between surface and subsurface run-off, and so E is evaporation from the flooded surface. This compo- water surplus includes both. nent of the water budget is estimated from Class A tank So is surface water outflow. As the playa lakes are evaporation data (mm), compared by correlation analyses located at the centre of a closed basin without any water- with other evaporation data from reservoirs of the Gua- discharging river or outlet, So is zero. 3 dalquivir river basin and scaled to obtain E (m ). Go is groundwater output (i.e. recharge or infiltration)

Si is surface water inflow (i.e. surface runoff) and Gi is from the lake to the aquifer. This term is not computed subsurface runoff. These two components of the playa in the equation because it is already included in the lake water balance are equal to 50% of the surplus of the soil–water budget (50% recharge). soil–water budget. The remaining 50% infiltrates to re- If the evolution of the water level in the playa lake is charge the aquifer. Previous studies (CHG-IGME 2001) similar to that observed in the field, then it is assumed have shown a relatively high transmissivity for the mate- that the water budget is correct. In that situation, the only 2 rials of the aquifer (10–100 m /day). Additionally, water water inputs to the lake are P and Si+Gi. If regional balances in other playa lakes of Andalusia have been groundwater discharge or hydrological alterations of the calculated using similar infiltration rates (Carrasco et al. catchment area such as drainage or over pumping take 2004). A soil–water budget in the catchment area was place, then the result will not fit the average evolution calculated to determine the water surplus. The soil–water of the water level observed in the playa lake. The

Water and Environment Journal (2007) c 2007 The Authors. Journal compilation c 2007 CIWEM. 3 Hydrogeological characteristics of a groundwater-dependent ecosystem M. Rodrıguez-Rodr´ ıguez´ et al. methodology of the water budget model presented in this to the north of the playa lake complex of La Lantejuela. work has been validated by successful previous applica- These materials display a medium level of hydraulic tions in other Andalusian wetlands (Benavente et al. conductivity (Fig. 2). 2006a, b). Autochthonous Formation (Upper Miocene and Plio-Quaternary): these appear discordant to the other Units described and are postorogenic. The main litholo- Discussion gies are conglomerates, gravels, sands, clays and evapor- ites that sometimes appear cemented. These materials Geological features are permeable, and have a high level of hydraulic Sedimentation in the Betic Cordillera took place during conductivity (Fig. 2). the Neogene in two tectonically different phases. From the Early to Middle Miocene, the Betic basin evolved with O-LL aquifer the main movements of orogenic structuring of the Cordillera, which were the collision between the Alboran´ The O-LL aquifer (Fig. 2) consists of a detritic unconfined Microplate (Internal Zones) and the South-Iberian Paleo- hydrogeological system made up by Plio-Quaternary allu- margin (in which deformed rocks of Mesozoic age made vial materials. The aquifer has an average thickness of up the External Zones). In this synorogenic phase, a 4–8 m in the northern part (near the playa lakes) and number of basins formed within the orogene as well as in about 40 m near Puebla de Cazalla and the Corbones River. the foreland basin outside it, such as the Proto-Guadal- Permeable outcrops occupy a surface area of 350 km2 and quivir basin. The sediments filling these basins were the aquifer materials display a heterogeneous hydrogeolo- deformed by the Betic orogeny. The second tectonic phase gical behaviour. The impermeable substrate of the aquifer took place during the Late Miocene to Quaternary. The contains clayey-evaporitic materials of Triassic age, which main features of the orogene were already determined are responsible for the relatively high salinity of the and this was the context in which the postorogenic basins groundwater (CHG-IGME 2001). Recharge takes place by formed, located in both the Internal and External Zones the infiltration of rainwater and occasional flooding of the and at the contact (Viseras et al. 2005). rivers. Discharge takes place by evaporation from the playa Simultaneously, diapiric movements associated with lakes and also by groundwater discharge to the main Mesozoic evaporitic materials occurred, leading to the streams and rivers of the area. Present-day discharge is also formation of elevations and karstification throughout the produced by the pumping activities carried out in numer- External Zones of the Cordillera. From the different ous irrigation wells throughout the area. regional geologic studies carried out in the study zone According to our measurements, the piezometric level (Cruz-Sanjulian´ 1974, 1986; Crespo 1977; Pignatelli & is close to the surface (3–4 m). Absolute altitudes a.s.l. Crespo 1977), the geologic materials outcropping in the range from 142 to 155 m a.s.l. The piezometric levels area can be grouped into three units: measured in the 47 wells of the inventory (Table 1) lead Alochthonous Formations (Subbetic and Olistostromic us to conclude that the directions of the groundwater flow Unit): multicoloured marls and clays of Triassic age that have an approximate component N3301E, from the proxi- include, in a chaotic mixture, blocks (olistolits) of varied mity of Osuna towards the playa lakes (Fig. 3). nature and ages between the Triassic and Miocene (dolo- Piezometric measurements indicate the existence of mites, volcanic rocks, limestones, marls, sandstones, etc.). a single groundwater system and that the base level of These outcrop to the east and the south of the study zone the playa lakes represents the phreatic level of the and constitute the impervious substratum above which aquifer. However, each lacustrine basin displays its own the youngest materials were deposited. The tectonics of separate characteristic hydrogeological behaviour as is the Alochthonous Units is determined by the gravita- discussed below. tional displacement of these materials, of plastic beha- The groundwater in the O-LL aquifer has nonhomoge- viour, that took place in a SE–NW direction towards the neous electrical conductivities, ranging from 1.2 mS/cm Miocene marine basin. These materials exhibit a low level in wells located in sandy conglomerates of Pliocene age to of hydraulic conductivity (Fig. 2). 16 mS/cm in wells located on the Quaternary clays and Para-autochthonous Formation: white marls and limes near the playa lakes. The salinity of the ground- marly limestones, locally with sandy bars, from Upper water increases towards the lacustrine basins. This hydro- Burdigalian to Messinian age. They show abundant chemical pattern seems to indicate that in a natural remnants of marine microfauna (sponge spicules, regime, transport processes take place towards the playa diatoms, etc.) and display a thickness of almost 100 m. lakes (the discharge zones of the O-LL aquifer) where The most extensive outcrop is that of the Palomarejo hill, evaporative brines are generated (Fig. 4).

4 Water and Environment Journal (2007) c 2007 The Authors. Journal compilation c 2007 CIWEM. M. Rodrıguez-Rodr´ ıguez´ et al. Hydrogeological characteristics of a groundwater-dependent ecosystem

Table 1 Location, salinity and water temperature of the wells inventoried

DWL WL EC25 Twater Code UTMX UTMY Z (m) (m a.s.l.) (mS/cm) ( 1C) Date recorded Observations O-LL-1 0312251 4132435 159 4.43 155 4.12 19.3 30 Sept. 2004 Aspirating pump O-LL-2 0312336 4132849 158 3.95 154 5.16 19.3 30 Sept. 2004 Aspirating pump O-LL-3 0312608 4133024 157 3.27 154 13.69 18.8 30 Sept. 2004 Not installed O-LL-4 0312435 4133252 157 3.00 154 7.20 19.3 30 Sept. 2004 Not installed O-LL-5 0312513 4133500 157 2.98 154 10.88 19.3 30 Sept. 2004 New well O-LL-6 0312899 4133844 161 9.03 152 7.09 20.3 30 Sept. 2004 Abandoned O-LL-7 0312702 4135107 155 3.50 152 4.18 20.2 30 Sept. 2004 Not installed O-LL-8 0312517 4135360 30 Sept. 2004 O-LL-9 0312762 4135760 158 3.40 155 2.15 21.5 30 Sept. 2004 Aspirating pump O-LL-10 0313037 4138336 3.10 11.90 21.7 30 Sept. 2004 New well O-LL-11 0312956 4138689 2.50 30 Sept. 2004 Submersed pump O-LL-12 0311313 4139277 150 2.50 148 2.85 22.3 30 Sept. 2004 Submersed pump O-LL-13 0309246 4138769 148 2.60 145 4.51 20.1 30 Sept. 2004 Submersed pump O-LL-14 0307461 4136901 151 3.00 148 13.8 22.8 30 Sept. 2004 New well O-LL-15 0307501 4136499 151 3.40 148 8.54 21.1 30 Sept. 2004 Aspirating pump O-LL-16 0307306 4136451 151 3.95 147 7.02 21.2 30 Sept. 2004 Aspirating pump O-LL-17 0307581 4136362 152 3.30 149 5.61 20.6 30 Sept. 2004 Not installed O-LL-18 0307616 4136056 156 4.10 152 4.81 20.2 30 Sept. 2004 Not installed O-LL-19 0308062 4135074 155 4.90 150 6.08 20.7 30 Sept. 2004 Submersed pump O-LL-20 0307967 4134991 155 4.85 150 4.74 20.4 30 Sept. 2004 Submersed pump O-LL-21 0308075 4134919 155 5.30 150 5.46 20.2 30 Sept. 2004 Aspirating pump O-LL-22 0308298 4134935 155 5.85 149 6.41 20.3 30 Sept. 2004 Aspirating pump O-LL-23 0308792 4135911 151 3.50 148 8.90 20.8 30 Sept. 2004 Aspirating pump O-LL-24 0308855 4138383 148 4.30 144 5.71 21.4 30 Sept. 2004 Aspirating pump O-LL-25 0311243 4134537 154 2.72 151 3.80 18.8 10 Oct. 2004 Not installed O-LL-26 0311580 4134552 154 3.00 151 10.22 19.6 10 Oct. 2004 Aspirating pump O-LL-27 0310888 4134208 156 3.93 152 5.76 19.9 10 Oct. 2004 Submersed pump O-LL-28 0310780 4134396 154 3.00 153 6.25 20.0 10Oct. 2004 Submersed pump O-LL-29 0310605 4134229 155 3.29 152 5.72 20.2 10 Oct. 2004 Submersed pump O-LL-30 0310167 4133858 157 4.26 153 6.93 20.1 10 Oct. 2004 Not installed O-LL-31 0309469 4134001 157 4.66 152 5.23 19.7 10 Oct. 2004 Aspirating pump O-LL-32 0310509 4134671 4.21 6.70 20.5 10 Oct. 2004 Not installed O-LL-33 0310131 4134869 154 3.50 151 6.19 20.8 10 Oct. 2004 Aspirating pump O-LL-34 0310245 4135161 153 3.13 150 8.35 20.8 10 Oct. 2004 Not installed O-LL-35 0309856 4136217 151 3.23 148 7.34 20.4 10 Oct. 2004 Not installed O-LL-36 0309824 4136346 151 3.30 148 8.34 20.2 10 Oct. 2004 Not installed O-LL-37 0309222 4136271 151 2.76 148 6.13 21.0 10 Oct. 2004 Not installed O-LL-38 0308944 4136809 150 2.96 147 4.86 21.0 10 Oct. 2004 Not installed O-LL-39 0308744 4137651 149 2.01 147 3.83 21.1 10 Oct. 2004 Aspirating pump O-LL-40 0307461 4139046 147 1.66 145 1.65 22.5 10 Oct. 2004 Aspirating pump O-LL-41 0307282 4139349 146 4.12 142 1.20 21.9 10 Oct. 2004 O-LL-42 0307809 4139618 145 1.53 143 2.42 21.9 10 Oct. 2004 Not installed O-LL-43 0307569 4139841 146 2.50 144 5.39 21.0 10 Oct. 2004 Aspirating pump O-LL-44 0307876 4139428 145 2.15 143 3.10 21.1 10 Oct. 2004 Aspirating pump O-LL-45 0308027 4139327 145 1.75 143 10.98 20.1 10 Oct. 2004 Not installed O-LL-46 0308022 4139249 144 2.18 142 9.97 20.0 10 Oct. 2004 Not installed O-LL-47 0308150 4139273 145 1.01 144 16.1 20.6 10 Oct. 2004 Not installed

Z, position (m a.s.l.) of the wells with an accuracy of 1 m (AME 2005); DWL, depth of the water level with an accuracy of 0.01 m; WL, water level with an accuracy of 1 m; O-LL, Osuna-La Lantejuela; a.s.l., above sea level.

This brine infiltrates through the sediments of the playa La Lantejuela playa lake endorheic complex lakes by diffusion and probably by advection processes, creating a zone of high-salinity groundwater below the This playa lake system consists of eight ephemeral lakes system (Figs 3 and 5). that have been considerably altered by human activities.

Water and Environment Journal (2007) c 2007 The Authors. Journal compilation c 2007 CIWEM. 5 Hydrogeological characteristics of a groundwater-dependent ecosystem M. Rodrıguez-Rodr´ ıguez´ et al.

(a) Consuegra 43 (b) 42 44 8 145 45 10 5 46 47 8 150 5 10 Verde de la Sal N 8 N Ballestera Calderon W E W E

S Cald Grande S

5

Pedro Lopez 150 5 Osuna Osuna

Fig. 3. Flow direction (a) and groundwater salinity (b) in the study area (September 2004).

18 4.5 Legend 16 4 47 R = 0.660 45 and 46 14 3.5 44 12 3 42 80 80 43 Playa lake 10 2.5 Scale of radii 60 60 8 2 Proportional to Cl (0–5200 mg/L) in the 40 EC (mS/cm) 6 1.5 upper diagram 40 4 1 20 20 R = 0.638 Depth of water table (m) 2 0.5 0 0 Mg SO 1000 1100 1200 1300 1400 1500 1600 1700 1800 Distance from playa lake (m) 80 80

Fig. 4. EC and depth of the water table in six wells (O-LL42 to O-LL47) 60 60 located in the studied area plotted against distance from Consuegra playa 40 40 lake (see Fig. 3). EC, electric conductivity; O-LL, Osuna-La Lantejuela. 20 20

EC (mS/cm) in well ‘O-LL46’ 80 60 40 20 20 40 60 80 7.38 7.4 7.42 7.44 7.46 7.48 7.5 7.52 7.54 7.56 7.58 Ca Na HCO Cl 2 Fig. 6. Piper diagram showing the hydrochemical facies in six wells 3 located in the studied area. The arrows indicate the geochemical evolution path of the groundwater: increasing concentrations of Cl 4 and Na+ from well 43 to the playa lake.

5 Depth (m) 6 the Salado River has been canalized to a tributary of the Corbones River, in accordance with the Agricultural Plan 7 for Osuna, established in 1967. The only playa lakes that have remained well preserved 8 are Ballestera and Calderon´ , declared Natural Reserves, Fig. 5. EC log in well No. 46 (O-LL 46, see Fig. 3). EC, electric conductiv- while the others are flooded only sporadically or have ity; O-LL, Osuna-La Lantejuela. been transformed into agricultural lands (Fig. 7). The playa lakes of the endorheic complex of La Lante- Formerly, it was a large endorheic area of more than juela exhibit high water salinity, although there is a large 300 km2 that extended around the town of Osuna. The degree of variation during each hydrological cycle. In the Salado River ended at this area, creating a continental fan Ballestera playa lake, the concentration of solids correlates delta. Nowadays, most of the playa lakes are drained and strongly with water volume and varies between 10 and

6 Water and Environment Journal (2007) c 2007 The Authors. Journal compilation c 2007 CIWEM. M. Rodrıguez-Rodr´ ıguez´ et al. Hydrogeological characteristics of a groundwater-dependent ecosystem

A B 200

175 Calderón (drained) Ballestera Gobierno 150 m a.s.l.

125

100 0246810km

A

Fig. 7. Cross section and hydrological function- ing in four of the studied playa lakes (Gobierno playa lake has been modified to host the water purifier of La Lantejuela). The downward haloki- netic movement refers to the depression formed up around the diapirs.

90 g/L. Similar salinity values have been measured in the playa lakes (discharge) or groundwater recharge from the Calderon´ playa lake. playa lakes to the aquifer must occur. With respect to hydrochemistry (Fig. 6), brine domi- The runoff (surplus) was estimated by applying the nated by Na and Cl develops in both systems, although precipitation and temperature records from the Osuna concentrations of sulphate and magnesium are also im- weather station (Table 2A) to the monthly water balance. portant. Playa lake brines are interpreted as the end- Because the catchment area of the playa lakes is located members of local groundwater flow lines moving slowly over relatively permeable materials, it is assumed that half towards the centre of each endorheic basin. of the runoff infiltrates, constituting a part of the recharge Table 2 shows the results for the water budgets in the of the O-LL aquifer. The other 50% produces direct playa lakes and the soil, together with meteorological surface runoff to the playa lakes. data. The surplus from the soil–water budget totals Table 2B shows the results of the monthly water 107.8 mm for a water-holding capacity of 100 mm (Table balance for the Calderon´ playa lake. The direct precipita- 2A), representing 20.5% of annual precipitation. tion (P) onto the surface area of 5.9 ha is 31.1 103 m3/ The monthly water balance of the Calderon´ and Balles- year. Excluding the flooded area, the catchment area has tera playa lakes was determined in accordance with the a surface of 75.1 ha. The surface runoff and subsurface methodology described above. We have assumed that, a groundwater recharge to the playa lake (Si+Gi) total priori, there is no water interchange between the playa 40.5 103 m3/year. The ET estimated in the model is lakes and the O-LL aquifer. If the results obtained from 71.6 103 m3/year. The model predicts a flooding period this theoretical balance for the average hydrological year with an average duration of almost 8 months, from (basically, the duration of the average flooding period of January to the end of August, and a maximum water the playa lake) are similar to the data known about the level of 65 cm. hydroperiod of the lagoon, then the resulting net volume According to the literature (Moreira & Montes 2005) of the water interchange between the O-LL aquifer and and field observations, the playa lake remains flooded for the playa lakes would be considered negligible. Other- o6 months each year. Thus, the hydroperiod observed is wise, a remarkable contribution of groundwater to the shorter than that predicted by the model. These differences

Water and Environment Journal (2007) c 2007 The Authors. Journal compilation c 2007 CIWEM. 7 Hydrogeological characteristics of a groundwater-dependent ecosystem M. Rodrıguez-Rodr´ ıguez´ et al.

Table 2 (A) Main characteristics of the hydro-meteorological station and soil water budget results (mm) (B) water balance results in Calderon playa lake (C) water balance results in Ballestera playa lake

Weather station Code Province Altitude Years rainf. Year init. Year end Year temp. Year init. Year end Osuna 5998A Seville 255 29 1967 1996 29 1967 1996

Month O N D J F M A M J J A S Total (A) Precipitation 48 69 78 76 62 52 53 36 16 4 16 17 527 Temperature 18.4 13.8 10.6 9.8 11.1 13.1 15.0 18.4 22.9 27.2 27.3 23.9 17.6 PET 65.9 33.9 20.1 17.9 22.2 37.2 50.9 82.4 125.1 175.3 166.2 115.0 912.1 Soil humidity 0.0 35.1 93.0 100.0 100.0 100.0 100.0 53.6 0.0 0.0 0.0 0.0 AET 48.0 33.9 20.1 17.9 22.2 37.2 50.9 82.4 69.6 4.0 16.0 17.0 419.2 Surplus 0.0 0.0 0.0 51.1 39.8 14.8 2.1 0.0 0.0 0.0 0.0 0.0 107.8

SW 81 ha SPL 5.9 ha

3 3 3 3 3 3 3 3 P (mm) Si+Gi (mm) E (mm) AET (mm) P ( 10 m ) Si+Gi ( 10 m )ET( 10 m )SH( 10 m ) D level (mm) Water level (mm) (B) Calderon October 48 0 90.8 48.0 2.8 0.0 2.8 0.0 0.0 0.0 November 69 0 58.6 34.0 4.1 0.0 2.0 2.1 0.0 0.0 December 78 0 38.1 20.0 4.6 0.0 1.2 5.5 0.0 0.0 January 76 26 40.5 17.9 4.5 19.2 2.4 5.9 354.2 354.2 February 62 20 54.9 22.2 3.7 14.9 3.2 5.9 260.4 614.6 March 52 7 83.6 37.2 3.1 5.6 4.9 5.9 62.6 677.2 April 53 1 89.9 50.9 3.1 0.8 5.3 5.9 23.5 653.7 May 36 0 115.4 82.4 2.1 0.0 6.8 5.9 79.4 574.3 June 16 0 178.4 69.6 0.9 0.0 10.5 5.9 162.4 411.9 July 4 0 218.6 4.0 0.2 0.0 12.9 5.9 214.6 197.3 August 16 0 201.6 16.0 0.9 0.0 11.9 5.9 185.6 11.7 September 17 0 151.0 17.0 1.0 0.0 7.6 0.0 0.0 0.0 Total 527 54 1321.4 419.2 31.1 40.5 71.6

SW 145 ha SPL 25.4 ha

3 3 3 3 3 3 3 3 P (mm) Si+Gi (mm) E (mm) AET (mm) P ( 10 m ) Si+Gi ( 10 m )ET( 10 m )SH( 10 m ) D level (mm) Water level (mm) (C) Ballestera October 48 0 90.8 48.0 12.2 0.0 12.2 0.0 0.0 0.0 November 69 0 58.6 34.0 17.5 0.0 8.6 8.9 0.0 0.0 December 78 0 38.1 20.0 19.8 0.0 5.1 23.6 0.0 0.0 January 76 26 40.5 17.9 19.3 30.6 10.3 25.4 148.8 148.8 February 62 20 54.9 22.2 15.7 23.8 13.9 25.4 100.8 249.6 March 52 7 83.6 37.2 13.2 8.9 21.2 25.4 3.2 252.9 April 53 1 89.9 50.9 13.5 1.3 22.8 25.4 32.0 220.9 May 36 0 115.4 82.4 9.1 0.0 29.3 25.4 79.4 141.5 June 16 0 178.4 69.6 4.1 0.0 45.3 20.1 141.5 0.0 July 4 0 218.6 4.0 1.0 0.0 21.1 0.0 0.0 0.0 August 16 0 201.6 16.0 4.1 0.0 4.1 0.0 0.0 0.0 September 17 0 151.0 17.0 4.3 0.0 4.3 0.0 0.0 0.0 Total 527 54 1321.4 419.2 133.9 64.5 198.3

All symbols are explained within the text.

ET, evapotranspiration; E, evaporation from the flooded surface; P, direct precipitation onto the playa lake; Si, surface water inflow; Gi, subsurface runoff; SW, surface of watershed; SPL, flooded surface of the lake; AET, actual evapotranspiration; SH, soil humidity; PET, potential evapotranspiration. could be explained by regional groundwater recharge from close to 150 m a.s.l. Therefore, this playa lake is perched the playa lake to the O-LL aquifer. Piezometric data (Figs 3, with respect to the saturated zone of the O-LL aquifer and 7 and 8) support this idea, because the Calderon´ playa lake can contribute to the recharge of the groundwater body. is situated at 157 m a.s.l. Piezometric measurements (Sep- The Ballestera playa lake has an average flooded surface tember 2004) indicate that the water level in this sector is area of 25.4 ha and a catchment area of 145 ha. The

8 Water and Environment Journal (2007) c 2007 The Authors. Journal compilation c 2007 CIWEM. M. Rodrıguez-Rodr´ ıguez´ et al. Hydrogeological characteristics of a groundwater-dependent ecosystem

C D

400 Palomarejo hill Osuna 300

Calderón 200

100 Diapiric movements

0 0 5 10 15 20 25 km

C

Local groundwater flow

A B

La Lantejuela

Regional groundwater flow Functional playa lake Drained playa lake Highway, roads Rivers D Fig. 8. Cross section of the Osuna-La Lantejuela Altitude (m a.s.l.) aquifer showing the direction of local and regio- Playa lake nal groundwater flow. Towns average precipitation onto the surface is 134 103 m3/ lake or in the aquifer depending on the hydrological year. Surface runoff from the catchment area is estimated season. In such a situation, the water of the playa lake to be 64.5 103 m3/year. The model predicts could recharge the aquifer in winter and vice versa. A 198.3 103 m3/year of ET (Table 2C). Under these condi- simple scheme of the hydrological regime in both playa tions, the hydroperiod would last from January to the end lakes and their relation with the O-LL aquifer is presented of June, and the water level would reach a maximum of in Fig. 7. Systematic piezometric information and detailed 25 cm in the months of February and March. A remark- geochemical investigation of this system are needed in able similarity exists between the forecasts of the model order to confirm this hypothesis. and the existing information about the actual hydroper- iod of the playa lake. The position of the playa lake Geomorphologic features: the fluvial network (145.5 m a.s.l.) with respect to the piezometric level of the O-LL aquifer in this sector (145 m a.s.l.) indicates that The main rivers of the study zone flow in a N–NW a hydraulic connection between the two systems may direction, from the southern highlands to the Guadalqui- exist. The hydraulic potential may be greater in the playa vir river basin. From west to east, these rivers are the

Water and Environment Journal (2007) c 2007 The Authors. Journal compilation c 2007 CIWEM. 9 Hydrogeological characteristics of a groundwater-dependent ecosystem M. Rodrıguez-Rodr´ ıguez´ et al.

Diapiric processes can be justified by the presence of Triassic evaporitic materials, as reported by several authors for nearby areas (Garc´ıa et al. 1991; Perez´ & Sanz 1994; Calaforra & Pulido-Bosch 1999).

Radial annular drainage network Conclusions (1) To the north of Osuna (southern Spain), Triassic Palomarejo hill evaporitic materials (Keuper) are abundant. In many places, Neogene and Quaternary materials are located above the Triassic clays, and in this area they constitute the O-LL aquifer. Blanco River (2) In the northern sector of the aquifer is the endorheic Salado River complex of La Lantejuela, comprising nine playa lakes. The analysis of hydrogeological measurements serves to detect the existence of local groundwater feeding to the Endorheic complex of La Lantejuela playa lakes. These are related to the O-LL aquifer, con- stituting local discharge zones and inducing a centripetal Peinado River groundwater flow into the playa lakes. Corbones (3) However, drainage works in these wetlands and intensive groundwater pumping have considerably mod- River ified the hydrological relations between the aquifer and the playa lakes. Advances in understanding these interac- Fluvial–lacustrine sediments tions and pressures are needed for a correct hydrological management of this type of groundwater-dependent eco- Fig. 9. The radial-annular drainage network around the Palomarejo hill system in the framework of the European WFD. can show the existence of a diapiric structure. The lakes of the endorheic (4) The geologic and geomorphologic characteristics – the complex of La Lantejuela are located around the peripheral depression. radial-annular fluvial network – of the area seem to indicate that a diapiric structure is responsible for the formation of the relief of the Palomarejo hill, of the basins Corbones, the Peinado, the Salado (of Osuna) and the where the playa lakes are located and of the O-LL Blanco. Around the Palomarejo hill, with a subcircular Plio-Quaternary detritic aquifer. shape and a diameter of more than 10 km, a remarkable distortion in the general disposition of the drainage net- work can be observed (Fig. 9). A fluvial network of radial- annular shape has developed in the hill, although this Acknowledgements network is very disorganized and presents numerous This work is a contribution to the projects ‘Definition of the endorheic areas, where the playa lake complex of La Hydrogeological Context of Andalusian Wetlands’ (Gua- Lantejuela is located. The playa lakes occupy a semicircu- dalquivir River Basin Authority) and IGCP-448 of UNES- lar area to the south of the Palomarejo hill (Fig. 9). CO and has been carried out by the Research Group RNM- The morphology of the relief and the distribution of the 126 of the Andalusian Autonomous Government. The fluvial network seem to indicate the existence of a diapiric critical comments of the anonymous reviewers and the structure. Accordingly, the Palomarejo hill would corre- Editor have helped us to improve the quality of this paper. spond to a diapiric dome and the lacustrine complex, the filling of the peripheral basin formed in the southern part of the diapir. The existence of active diapiric processes References would explain the modification of the fluvial network AME. (2005) Digital terrain model of Andalusia, Relief and pattern and the formation of the lacustrine basins as well topography, Regional Ministry of Andalusia, Andalusian as the detritic O-LL aquifer (Moral et al. 2006) due to Ministry of Environment (AME), Seville. fluvial–lacustrine sedimentation in the great endorheic Benavente, J., Moral, F., and Rodr´ıguez-Rodr´ıguez, M. (2006a) river basins formed (such as that of the Salado River Definicion´ del contexto hidrogeologico´ de humedales andaluces, continental fan delta) formed in the river trajectory before Lagunas de Cadiz,´ Guadalquivir River Basin Authority, the diapiric structure (Fig. 9). Seville, p. 182.

10 Water and Environment Journal (2007) c 2007 The Authors. Journal compilation c 2007 CIWEM. M. Rodrıguez-Rodr´ ıguez´ et al. Hydrogeological characteristics of a groundwater-dependent ecosystem

Benavente, J., Moral, F., and Rodr´ıguez-Rodr´ıguez, M. (2006b) Lake (Fuente de Piedra, Southern Spain). Hydrobiologia, 345, Definicion´ del contexto hidrogeologico´ de humedales andaluces, 131–141. Lagunas de Sevilla, Guadalquivir River Basin Authority, Jensen, M.E., Burman, R.D., and Allen, R.G. (1990) Seville, p. 175. Evapotranspiration and Irrigation Water Requirements. Borowski, I. and Hare, M. (2007) Exploring the Gap Between American Society of Civil Enginners, New York. Water Managers and Researchers: Difficulties of Model- Mansell, M. and Rollet, F. (2006) Water Balance and the Based Tools to Support Practical Water Management. Water Behaviour of Different Paving Surfaces. Water Environ. J., 20, Resour. Manage., 21 (7), 1049–1074. 7–10. Calaforra, J.M. and Pulido-Bosch, A. (1999) Gypsum Karst Montes, C., Gonzalez-Capitel, E., Molina, F., Moreira, J.M., Features as Evidence of Diapiric Processes in the Betic Rubio, J.C., Rodriguez, M., Gonzalez, C., and Rodriguez, I. Cordillera, Southern Spain. Geomorphology, 29, 251–264. (2004) Plan Andaluz de Humedales (Andalusian Wetland Plan). Carrasco, F., Andreo, B., Vadillo, I., Sanchez,´ D., Perez,´ I., V´ıas, Andalusian Ministry of Environment, Seville. J., and Cobos, A. (2004) Definicion´ del contexto hidrogeologico´ de Moral, F., Rodr´ıguez-Rodriguez, M., Benavente, J., and humedales andaluces, Lagunas de Malaga,´ Andalusian Cifuentes, V. (2006) Influence of diapiric processes in Ministry of Environment, Seville, p. 130. the genesis of the detritic aquifer Osuna-La Lantejuela CHG-IGME. (2001) Norma de Explotacion´ de la Unidad and the endorheic complex of La Lantejuela (Seville). Hidrogeologica´ 05.69 (Osuna-La Lantejuela). Spanish Groundwater in Mediterranean Countries. Spanish Geological Survey, Madrid, p. 21. Geological Survey, Seville, Spain, p. 123–128. Crespo, A. (1977) Hoja 986/14-40 (Fuentes de Andalucıa)´ . Spanish Moreira, J.M. and Montes, C. (2005) Caracterizacion´ ambiental Geological Survey (IGME), Madrid. de humedales en Andalucıa´ . Junta de Andaluc´ıa, Sevilla. Cruz-Sanjulian,´ J. (1974) Estudio geologico´ del sector Canete˜ La Perez,´ A. and Sanz, C. (1994) Tectonica´ de los Materiales Real-Teba-Osuna (cordillera betica,´ region´ occidental), triaSicos´ en el Sector Central de la Zona Subbetica´ (Cordil- Geodinamica,´ Granada, p. 431. lera Betica).´ Rev. Soc. Geol. Espana˜ , 7, 141–153. Cruz-Sanjulian,´ J. (1986) Hoja 1005/15-41 (Osuna). Spanish Pignatelli, R. and Crespo, A. (1977) Hoja 1004/14-41 (Marchena). Geological Survey (IGME), Madrid. Spanish Geological Survey (IGME), Madrid. Duffy, C.J. and Al-Hassan, S. (1988) Groundwater Circulation Thornthwaite, C.W. and Mather, J.R. (1955) The Water in Closed Desert Basins: Topographic Scaling and Climatic Balance. Drexel Inst. Technol. Publications Climatol., 8, 1–104. Forcing. Water Resour. Res., 24, 1675–1688. Vera, J.A. (2004) Geologıa´ de Espana˜ . Spanish Geological Fan, Y., Duffy, C.J., and Oliver, D.S. (1997) Density-Driven Society–Spanish Geological Survey, Madrid. Groundwater Flow in Closed Desert Basins: Field Viseras, C., Soria, J.M., Fernandez,´ J., and Garc´ıa-Garc´ıa, F. Investigations and Numerical Experiments. J. Hydrol., 196, (2005) The Neogene-Quaternary Basins of the Betic 139–184. Cordillera: An Overview. Geophys. Res. Abstr., 7, Garc´ıa, A., Gonzalez,´ J., Hernaiz, P., Zazo, C., and Goy, J. 11123–11125. (1991) Hoja 1069/12-46 (Chiclana). Spanish Geological WFD-CISNo.2 (2002) Identification of Water Bodies. Directorate Survey (IGME), Madrid. General Environment of the European Commission, Brussels. Garc´ıa, C.M., Garc´ıa-Ruiz, R., Rendon,´ M., Niell, F.X., and WFD-CISNo.12. (2003) The Role of Wetlands in the Water Lucena, J. (1997) Hydrological Cycle and Interannual Framework Directive. Directorate General Environment of Variability of the Aquatic Community in a Temporary Saline the European Commission, Brussels.

Water and Environment Journal (2007) c 2007 The Authors. Journal compilation c 2007 CIWEM. 11