Introduction

Water resources available around the Eastern Mediterranean Region, are suffering severely from the increases in temperature and severe droughts. This resulted in an increase in the demand for water resources, mainly on islands like Cyprus. The water resources of Cyprus are increasingly stressed by the rapidly growing tourism sector, anthropogenic activities and agricultural investment (Tsiourtis 2004; Endreny and Gokcekus 2008; Türker et al. 2009). A reliable source of groundwater supplies for domestic and irrigation use is less affected by droughts and other climatic changes. Groundwater in Cyprus has traditionally been viewed as a personal source for domestic and agricultural purposes, which results in overexploitation that endangers the aquatic ecosystem (Ergil 2000; Milnes and Renard 2004; Milnes 2011).

In recent years, to protect the groundwater resources requires understanding of the hydrogeological characteristics of the available aquifers. The wide spectra of hydrogeological studies provide a useful framework to manage groundwater resources. Simulation and optimization models (Uddameri and Kuchanur 2007; Feng et al. 2011); remote sensing and GIS approaches (Pandey and Kazama 2011; Lee et al. 2006); numerical models (Garg and Ali 2000; Liu et al. 2006); and safe yield and groundwater balance analysis (Mollema et al. 2012; Ersoy and Gültekin 2008) are some of those valuable contributions. The safe yield and balance method can suitably assess the aquifer management activities in Yeşilköy aquifer, as it can objectively assess the rate at which groundwater can be abstracted from an aquifer without causing an undesirable adverse effect (Heath and Spruill 2003) and helps to model a regression relation for predicting the future behavior of the aquifer in terms of water balance.

Accordingly, the objective of this study is to evaluate groundwater resources of the Yeşilköy aquifer system involving the quantification of recharge, storage, discharge and the sustainable safe yield based on groundwater budget. The goal is to investigate the past performance of the aquifer for developing regression analysis to put forth the present disposition of the water table and to predict the future behavior of the aquifer system.

Groundwater resources of northern part of Cyprus

Groundwater is a major source of fresh water supply in northern part of Cyprus. It supplies 92 % of water to meet the domestic and irrigation requirements. Low rainfall over the past few years with more frequent periods of drought revealed the need for asystematic and comprehensive approach for the proper and effective management of water resources in Cyprus. Price et al. (1999) analyzed long-term temperature data of two different meteorological stations in Cyprus and found increasing trends of approximately 1 °C/100 years in the annual mean temperatures. Following this, Giorgi and Lionello (2008) studied the climate change projections for the Mediterranean region and determined a large influence of climate change on the water budget of Mediterranean countries. The above facts reinforce the need for a long-term perspective to optimize the water demand and prevent uncontrolled and uncounted activities on groundwater resources. The general mismanagement of groundwater resources in the area has already caused negative effects such as seawater intrusion in coastal areas, quality and quantity deterioration, and environmental problems on agricultural lands (Gökçekuş et al. 2002; Gökmenoğlu et al. 2004).

Neglecting the size of the water bearing formations, there are 12 major aquifers listed in Table 1, distributed around the northern part of Cyprus. Among the 12 groundwater resources, the three main domestic and irrigation water supplier aquifers are distributed on three different basins. Güzelyurt Coastal Aquifer is on Mesarya Plain; Girne Range Aquifer is in Girne Limestone Mountains; and Yeşilköy aquifer is at the inlands of Karpaz Peninsula as shown in Fig. 1.

Table 1 The aquifers at northern part of Cyprus
Full size table
Fig. 1
figure 1

The location of three main aquifers located at northern part of Cyprus

Full size image

The Güzelyurt coastal aquifer has always been the focus of engineers and scientists as it is the largest aquifer in Cyprus. Electro-Watt Company (1973) carried out a feasibility study on irrigation methods at Güzelyurt region to optimize the water consumption from the aquifer. The Turkish State Water Works (DSI 1975, 1976) has established a comprehensive study about the Güzelyurt Aquifer. They concentrate on the water balance of the aquifer taking into consideration the salt water intrusion problems while proposing dams for recharging activities at the region. Gökçekuş and Doyuran (1996) proved that in addition to salt water intrusion, the quality degradation of the aquifer can be linked to the bedrock contamination. A study has been done by Ergil (2000) who estimated the available freshwater volume within the aquifer and roughly predict the remaining lifetime of the aquifer by the help of the volumetric approach. Later, Gökmenoğlu et al. (2004) had a detailed study and indicated that due to over-pumping, the Güzelyurt Aquifer has been deteriorated by seawater intrusion.

The Girne Range Aquifer is the second largest aquifer of northern part of Cyprus. It used to be one of the most qualified aquifer of Cyprus. The geologic formation of Girne Range Aquifer allows the spring discharges in which more than 30 naturally flowing springs were active at the area. These springs were used to discharge 0.5–4 MCM per year. However, extensive over-pumping during the last 20 years has resulted in a dramatic depletion of the water levels such that most of the springs are dried up. The total area of the aquifer is 85 km2, the average rainfall is 400–450 mm and the thickness of the aquifer is 400 m. Dixey (1972) has estimated that 167 mm direct seepage from the precipitation replenishes the aquifer by 11.5 MCM per year. Gökmenoğlu et al. (2004), based on 22 drilled monitoring wells and geophysical studies, worked out the hydrodynamic structure of the Girne Range Aquifer. During the last decade, water levels within the aquifer have exhibited a depleting trend, due to reduced recharge and increased abstraction.

Yeşilköy is the third main aquifer that occupies 5.46 km2 surface area. The aquifer is used for agricultural and domestic water supply purposes and has provided a supply of water since the late 1960s. Yeşilköy aquifer has never been studied to evaluate its hydrogeology or water balance. The Yeşilköy basin is 100 km northeast of Nicosia, the capital. It can be divided into two distinct geographic environments: the high plain to the southeast and the lower plain to the southwest. The high plain has elevations of 100–150 m, moderate relief and dendritic drainage. The gradient is 3.5 % towards the southwest. The southwest drainage is identified by a spring which was once discharging towards the outlet of the basin. The rural settlements in the region motivated agricultural activities as the major economic factor.

The region has sustained competitive agricultural activities on Colocasia esculenta farming since the late 1970s. On the other hand, the yields of Colocasia esculenta declined in response to the increased depletion of the groundwater sources and drop in level. As a result, the main crop pattern at the area has changed to Solanum tuberosum (potatoes) which consumes less water and ripens with high quality due to the structure of the surrounding soil. Unfortunately, the change in the crop pattern did not change the inevitable deterioration of the aquifer.

The Yeşilköy aquifer

Hydrogeology of the region

The Karpaz Peninsula in which Yeşilköy aquifer is located, has mainly impervious rocks of the Kythrea formation overlain by generally thin Pliocene–Pleistocene sedimentary rocks, which are in part pervious. Only at Yeşilköy and Dipkarpaz/Rizokarpasia regions, these sedimentary rock layers are thick enough to form small groundwater units. The individual small compartments of limestone can be found in patches in this area with rather small storage capacity (UNDP 1970).

The shape of the Yeşilköy aquifer is an elliptic surface and mainly consists of the Quaternary calcarenite and forms a permeable aquifer above relatively impermeable marl, which acts as an aquitard.

There are two ephemeral streams tangentially bypassing the Yeşilköy aquifer boundary. These are the Göz and the Değirmen streams. Due to impermeable Kythrea formation, these ephemeral streams mainly flow to the sea without recharging into the aquifer.

The unconfined groundwater in sandstone, sandy marls and calcarenite is the description of the composition of the Yeşilköy aquifer. The smooth and uniform variation in hydraulic head of the water table and the responses of aquifer to recharging and discharging is the strong indication that the aquifer is homogenous and isotropic (Gökmenoğlu et al. 2004). The aquifer has been in used since the 1960s. The only pumping test in the area was conducted in April 2007 which is explained in detail by McKinney and Brown (2007).

The well-known Theis (1935) model is later incorporated with the pumping test results of McKinney and Brown (2007) and the hydraulic properties of Yeşilköy aquifer are identified. The result of the pumping test using the Theis method (Fig. 2) shows that the transmissivity T of the aquifer was 1,782-m2/day, and the storage coefficient S was on the order of 0.0012. The results satisfied the hydraulic properties of sandstone and calcarenite subsurface of the region.

Fig. 2
figure 2

The type curve and curve matching by Theis (1935) graphical method

Full size image

Since 2000, local authorities initiated a monitoring program for monitoring the water level and water quality of the Yeşilköy aquifer. The recorded database consists of extensive collection of static and dynamic water levels and water quality data. The sampling frequency was twice a year. Recharge to the aquifer system occurs mainly as downward percolation of precipitation. Although the water table tends to rise during the rainy season (mid-November to mid-March), recharging of the aquifer is not successfully managed to balance the water consumption of the aquifer. Discharge from the aquifer system occurs primarily by abstraction from the wells used for irrigation and domestic water supply of the whole Karpaz Peninsula, including 15 villages with approximately 10,000 residents.

The average depth of the Yeşilköy wells is divided into two different groups. At the northern lowland part of the basin, the static water levels in the wells range between 6–10 m below ground surface and the depth of these wells are around 12–15 m. This basin is underlain by hydrogeologic units with small well yields. The pumping rates from these wells is around 4–5 m3/h, in which the well is then run out of water and can be reusable after 1 or 2 days. The main aquifer is located at the southern upland region. The depths of the wells in this region are approximately 75–100 m with the capacity of 15–35 m3/h discharging from 30 cm diameter wells. According to the above-mentioned hydraulic characteristics of the region, the total water bearing units are divided into two sub-basins working independently of each other. The location and the distribution of wells at the main aquifer are given in Fig. 3.

Fig. 3
figure 3

Distribution of wells at the Yeşilköy main aquifer

Full size image

The chemistry of groundwater from the Yeşilköy aquifer has been studied to classify the type of water. The main sampling parameters were total dissolved solids, major cations (Na+, Mg2+, Ca2+, and K+) and anions (Cl−, SO4 2−, CO3 2− and HCO3 −). The pH values and specific conductance are also measured. Figure 4 shows the graphical illustration based on Piper (1953) trilinear diagram of the chemical analysis of twelve water samples collected from different wells. The major dominant cations are sodium and potassium, increasing the possibility of pollution of aquifer due to fertilizers used for the agricultural activities like growing Solanum tuberosum and other crop types. The dominant anions, on the other hand, are chloride and carbonate.

Fig. 4
figure 4

The Piper diagram representation of twelve different wells at Yeşilköy aquifer

Full size image

It can be concluded that the water chemistry of the Yeşilköy aquifer is NaCl, NaHCO3 and KCl. The absence of Calcium and Magnesium indicates that the precipitation base recharging of the aquifer is not dominant since the saturation of geological formation (e.g. calcarenite) with rainwater is not observed. On the other hand, the availability of KCl increases the possibility of pollution caused by fertilizers used for agricultural purposes. The NaHCO3 content is also the indication of basic water characteristics for the region. The Sodium adsorption ratio (SAR) of the aquifer ranged between 2.95 and 8.23, showing that the groundwater of the Yeşilköy aquifer is suitable for agricultural purposes.

The meteorology of the region

The climate of the study area can be classified as semi-arid and is characterized by warm, relatively hot summers and mild winters. Temperatures commonly exceed 35 °C during May, June, July and August and fall below 4 °C for a few days in the winter months. The average annual temperature at Yeşilköy basin is about 20 °C. The long-term (1975–2005) average annual precipitation at the region is 486.6 mm. During the winter, the rainfall is associated with cold fronts and is more uniformly distributed than the rest of the year.

Evaporation is the most significant water loss from the catchment area of the region. Dense uniform vegetal cover minimizes the evaporation rates, relative to the other parts of Cyprus, however, became significant when evapotranspiration is considered. Depending on the type of cultivated crop and vegetal cover, evapotranspiration rates vary in space and time. It remains as an unknown parameter to be quantified in safe yield analysis of the Yeşilköy aquifer in the following sections.

Data analysis and water balances

The drawdown rates and the depletion characteristics of the aquifer

In Yeşilköy aquifer, almost 42 wells have been monitored since 2000. The monitoring program was followed generally twice a year for each well. The monitored static water levels of 30 wells are given in Fig. 5. The figure shows that at the end of 2009 the water level of the aquifer varied between 69 and 87 m from the mean sea level. The declining trend of the water level during the monitoring period, with respect to time, can be formulated by a power relationship,

$$ {\text{swl}}_{\text{avg}} = 10^{6} t^{ - 0.9} $$
(1)

where t is the time and swlavg is the average static water level of the aquifer. During the monitored time interval annual average drop in water level elevation was observed to be around 0.6 m (an increase of 1.8 m during winter and a decrease of 2.4 min summer). In mid-March 2008, two data loggers (Solinist level logger model 3001) were installed into two observation wells YK-109 and YK-21. The objective was to provide a continuous monitoring program for the two wells. The monitoring program also provided data on aquifer response to recharging by precipitation. The calcarenite property of the aquifer was sensitive to the hydraulic flow within the aquifer. The change in water level as a result of continuous monitoring of the wells YK-21 and YK-109 is given in Fig. 6.

Fig. 5
figure 5

The monitored static water levels of the Yeşilköy aquifer

Full size image
Fig. 6
figure 6

The drop in water levels on YK109 and YK21 until the end of June 2008

Full size image

It is apparent that the static water level in each well follows a depleting trend with respect to time. Seldom discontinuities in the depleting trend are observed due to rapid infiltration of precipitation in the region. The rapid influence of precipitation on water table levels are confirmed with the precipitation data at gages located at Ziyamet/Leonarisso and Yenierenköy meteorological stations (Table 2). The daily response of aquifer due to precipitation is clear as the precipitation date and discontinuities on depleting curve are corresponding with each other.

Table 2 The precipitation rates and dates at Ziyamet and Yenierenkoy meteorological stations
Full size table

The agricultural and domestic water consumption analyses

A field survey was conducted for all irrigated lands in the Yeşilköy basin, which helped to collect the agricultural based information that was used for the definition of the crop pattern, farm sizes, and irrigation techniques. Field survey data also provided the necessary information for water consumption estimates at the region. The location of survey sites was mapped by using a GPS receiver using the geodetic datum WGS84. All the field data of Yeşilköy obtained by the help of GPS were superimposed on a satellite image using GIS Arcview (ESRI) software. The findings were useful to categorize the type of crop pattern, mainly given as Solanum tuberosum and Hordeum (Figs. 1 and 2, ESM only). The total planting area and its crop distribution were also important outcomes of field survey.

About 150 farmers are active and Solanum tuberosum is the most common crop in the area, covering 83.8 ha out of 170 ha. Hordeum is the second leading crop, covering 38 % of the planted areas. The total area of Hordeum is estimated to be 75.7 ha. It is important to note that the Hordeum is only experiencing precipitation based agriculture. The other crops such as citrus and other fruits cover only 7 % of the total area. The cultivated land was 23.5 ha, covering 12 % of the total area. Solanum tuberosum is planted twice a year, in February and in August. Each year the government announces that the yield during August is half the amount planted in February (İnce et al. 2005, 2006, 2007, 2008, 2009). The agricultural water consumption from the wells is uncounted, and thus the annual water consumption for irrigation is unknown.

A well survey has been carried out to estimate the water abstraction from each well. The results of survey studies listed 177 private and 6 governmental wells. Among the 177 private wells, 85 wells were either abandoned or drilled as stand-by or observation wells, and the remaining 92 wells were used as production wells. Well locations were recorded and imported into the Yeşilköy geographical information mapping system based on the latitude and longitude information provided by a Magellan 600 GPS. These locations were electronically overlain on satellite maps to evaluate the relationship between each well and corresponding land use for irrigation purposes. Table 3 provides a breakdown of the survey results for the 183 wells. The results of the well survey show that there are 75 wells used for agricultural purpose. Among those 75 wells merely 62 wells are operating efficiently which has a considerable effect on groundwater budget estimation. The rate of pumping ranges between 15 and 35 m3/h. The annual consumption of each well is given in Fig. 3 (ESM only). Some of these wells pump for 16 h a week, whereas some operate for 48 h a week. The diameter of each of the wells in the area is 30 cm.

Table 3 Summary of well survey results
Full size table

The irrigation water consumption from each irrigation well of 62 wells add up to 0.906 MCM which is far above the water consumption of water treatment plant and household uses (Table 3). The livestock and gardening water consumption can be described as the water consumption for small gardening works and animal breeding like hens and sheep.

Six water supply wells, supplying domestic water demands of the entire Karpaz peninsula, are discharging 0.66 MCM per year. The details of consumptions of each well are given in Table 4.

Table 4 The annual consumption of water supply wells
Full size table

Safe yield analysis

A successful groundwater management study is required at Yeşilköy region to preserve the sustainability of groundwater resources and agricultural activities. The general water balance equation of the Yeşilköy aquifer at any time interval can be expressed as:

$$ \pm \Updelta S = \sum {I(t) }- \sum {O(t)} $$
(2)

where ±ΔS is the change in groundwater storage defined as negative when the groundwater overexploitation is dominant and positive when the rate of abstraction is less than the rate of recharge. ∑I(t) is the total volume of inflow, which consists of recharge from precipitation (I ppt) and recharge from irrigation returns (I irr). ∑O(t), on the other hand, is the total volume of outflow which consists of abstraction for irrigation (O irr), abstraction for domestic use (O dom), abstraction for livestock and gardening (O lsg) and spring outflow (O spr). Hence, the annual average inflow to the Yeşilköy aquifer is:

$$ \sum {I(t) = I_{\text{ppt}} + I_{\text{irr}} } $$
(3)

and the average annual abstraction from the aquifer can be estimated as:

$$ \sum {O(t) = O_{\text{irr}} + O_{\text{dom}} + O_{\text{lsg}} + O_{\text{spr}} } $$
(4)

The closed and spherical basin shape of the Yeşilköy region forces the surface runoff to drain towards the drainage area of the aquifer. Therefore, except for evapotranspiration there is no other loss from the precipitated water within the drainage area. Therefore, the remaining water from precipitation is directly infiltrating into the aquifer. On the other hand, due to high temperature the mean daily evapotranspiration values are generally higher than the precipitation rates except in winter months. According to the analysis of Cropwat (Smith 1992) that is based on the FAO Penman–Monteith method, the mean daily evapotranspiration values in the study area are summarized in Table 5.

Table 5 Evapotranspiration and Precipitation rates at Yeşilköy from November to April
Full size table

The FAO Penman–Monteith method calculates the evapotranspiration values (ETo) based on the following equation:

$$ {\text{ET}}_{\text{o}} = \frac{{0.408\Updelta (R_{\text{n}} - G) + \gamma (900/(T + 273))U_{2} (e_{\text{s}} - e_{\text{a}} )}}{{\Updelta + \gamma (1 + 0.34U_{2} )}} $$
(5)

where ETo is the reference evapotranspiration (mm/day), R n is the net radiation at the crop surface (MJ/m2day), G is the soil heat flux density (MJ/m2day), T is the mean daily air temperature at 2-m height (°C), U 2 is the wind speed at 2-m height (m/sec), e s is the saturation vapor pressure (kPa), e a is the actual vapor pressure (kPa), Δ is the slope vapor pressure curve (kPa/°C), and γ is the psychometric constant (kPa/°C).

The crop pattern evaluated at the end of the field survey was used as input data of evapotranspiration calculations. The result of the evapotranspiration calculations has shown that the aquifer is recharging only during the winter period and the total volume of recharge is 1.01 MCM (drainage area × depth of annual recharge). This value has been achieved considering that the drainage area of the aquifer is 6.39 km2.

The results of continuous monitoring in wells YK109 and YK21 and the results of water quality analysis have shown that the irrigation return to the aquifer is considerably low. Due to the high temperature and the high evapotranspiration rates during the irrigation seasons irrigation returns are estimated to be as high as the 10 % of the irrigation water. Thus, the total annual inflow into the aquifer is:

$$ \sum {I(t) = 1.01 \times 10^{6} + (0.1 \times 0.906) \times 10^{6} } = 1.1 \times 10^{6} \;{\text{m}}^{3} /{\text{year}} $$
(6)

As was calculated in the previous section, the annual domestic water supply from the aquifer is 0.66 MCM, and the groundwater abstracted from the aquifer to cover irrigation needs is 0.906 MCM. The groundwater abstraction for the livestock and gardening and water treatment purposes is considerably negligible when compared with the domestic and irrigation purposes. As the head of the springs of the region is higher than the groundwater level, there is no spring flow at the region. Thus, the average annual abstraction from the aquifer can be estimated as:

$$\sum {O(t) = 0.66 \times 10^{6} + 0.906 \times 10^{6} + 0.03 \times 10^{6} + 0 = 1.596 \times 10^{6}\, {\text{m}}^{3} /{\text{year}}}$$
(7)

From the estimates made above, the mean annual change in groundwater storage from the Yeşilköy aquifer is 0.496 MCM. The budget calculations prove that the current abstraction is not sustainable, and overexploitation of the aquifer exists. The impacts of the overexploitation can be terminated by reducing the rate of abstraction within the limits of the safe yield of the aquifer. Safe yield of the groundwater is that amount of groundwater that can be abstracted without adverse and destructive effects on the groundwater system. In this study, the safe yield is estimated by developing a relationship based on the studies performed previously by Goes (1999), Naik and Awasthi (2003) and Voudouris (2006). Their findings were helpful to define the specific yield S y and safe yield based on the water-level fluctuation method. The method uses the water level rise in wet and water level decline in dry seasons. The modified form of the specific yield equation can be given as:

$$ S_{\text{y}} = \frac{{\left( {\sum {I(t)} -\sum O (t)} \right)_{\text{WET}} }}{{A\Updelta {\text{WT}}_{\text{WET}} }} $$
(8)

in which subscript WET denotes the wet season of the region (November to April), A represents the drainage area of the aquifer basin and ΔWT means the change in water level elevation. The summary and the evaluation of the parameters used for the results of the Eq. (8) are illustrated in Table 6. Using these parameters, the specific yield for the Yeşilköy aquifer is estimated to be 0.030 or 3.0 %. The safe yield can be deducted from the following equation in which the ratio of the average water level decline in the dry period to the average water level increase in the wet period becomes significant:

$$ {\text{SafeYield}} = \frac{{\Updelta {\text{WT}}_{\text{DRY}} }}{{\Updelta {\text{WT}}_{\text{WET}} }}(\Updelta S)_{\text{WET}} + (O_{\text{dom}} )_{\text{WET}} + I_{\text{irr}} $$
(9)

where subscript DRY denotes the dry season of the region (May to October), and ΔS, O dom, ΔWT and I irr are defined earlier. Incorporating the data given in Table 6 and the Eq. (9) estimates the safe yield for the Yeşilköy aquifer as 0.84 MCM annually. Under these circumstances, it is obvious that the sustainability of the aquifer is not foreseen and the safe yield exceeds the natural recharge and the aquifer is overexploited. The safe yield can be attained by decreasing the water abstraction from the aquifer, which can be achieved by lowering the water volume for irrigation use or maintaining the domestic water supply from another source like seawater desalination.

Table 6 Explanation of the parameters used for specific yield and safe yield calculations
Full size table

Abstraction effects on water level fluctuations

The water surface elevation ranges from 69 to 87 m above mean sea level. Water level tends to fluctuate in response to recharge and withdrawal rates. Accordingly, the groundwater levels at the center and southwest of the Yeşilköy aquifer generally are higher than the northeast part because of over abstraction. The lower groundwater level elevations at the northeastern part are due to continued pumping in that region. The governmental wells, which are used for domestic water supply, are in this region. Therefore, instead of static water levels, most of the readings refer to the dynamic water levels at that region. The water level surface of the aquifer, based on well survey studies, is given in Fig. 7.

Fig. 7
figure 7

The water table surface of the Yeşilköy aquifer, 2010 (m)

Full size image

On the basis of the previous studies (Theis 1935; Wang et al. 2004), it is apparent that the elevation of the head in any well can be defined as a function of storage coefficient S, transmissivity T, radius of interference r, discharge Q, and the time since the pumping starts t:

$$ h(t) = f(r,S,T,t,Q,\Updelta t) $$
(10)

The application of the dimensional analysis on the representation of dependent and independent parameters results in the relation between two dimensionless parameters described as:

$$ \frac{Th(t)}{Q}\alpha \frac{\Updelta t}{t} $$
(11)

in which h(t) is the static water level; Δt is the time difference between the two consecutive monitoring periods and t is the total monitoring period. The dimensionless term “h(t)(T/Q)” is the head parameter and “Δt/t” is the period parameter. The relationship between the head parameter and the period parameter can be further evaluated using the data collected from the wells in Yeşilköy aquifer. The outcome will help to understand and empirically formulate the fluctuations and the spatial distribution of water level. The preliminary studies have shown that the relationship between the two parameters is a sixth order polynomial with a negative trend:

$$ \frac{Th(t)}{Q} = a\left( {\frac{\Updelta t}{t}} \right)^{n} - \,b\left( {\frac{\Updelta t}{t}} \right)^{n - 1} +\, \cdots e\left( {\frac{\Updelta t}{t}} \right)^{n - (n - 1)} +\, f $$
(12)

The regression analysis is used to generate a best fit equation for the collected data for Yeşilköy aquifer. This will help to make estimates and predictions about the behavior of the aquifer in future predictions. Figure 8 describes the relationship between the period parameter and the head parameter for Yeşilköy aquifer. It is important to note that during the selection criteria of monitored well data used in regression analysis, the wells which have missing data within the last 10 years were taken as outliers. The resultant empirical equation for defining the relationship between the period parameter and the head parameter is given in Eq. (13). The empirical equation is derived by considering that the aquifer is homogenous and isotropic:

$$ \frac{Th(t)}{Q} = 9\left( {\frac{\Updelta t}{t}} \right)^{6} - \left( {\frac{\Updelta t}{t}} \right)^{5} +\, 8\left( {\frac{\Updelta t}{t}} \right)^{4} - \,2\left( {\frac{\Updelta t}{t}} \right)^{3} +\, 0.26\left( {\frac{\Updelta t}{t}} \right)^{2} - \,0.015\left( {\frac{\Updelta t}{t}} \right) + \,0.0235 $$
(13)
Fig. 8
figure 8

The relation between ∆t/t and (T/Q)h(t)

Full size image

The coefficient of determination, R 2 of the equation provides strong relationship with the monitored data and the predicted equation. Equation (13) invokes the two important assumptions: (1) the discharge from the aquifer is constant; (2) the equation is valid for limited time period, representing the relationship of the aquifer parameters for the last 10 years. A clear deficit is expected for the wells pumping at the northeast part of the aquifer. These wells are mainly the governmental wells supplying domestic water for the surrounding villages. The wells which are close to the upper (north) boundary of the aquifer are also alarming for future use due to the fault of north boundary. The safer yields can be obtained from the western part of the aquifer. The groundwater level data shows that hydraulic head of the aquifer drops towards the east due to the heavy pumping from the central and eastern part of the aquifer. After all these observations it can be concluded that further deterioration of the aquifer is inevitable unless the pumping rates are optimized and irrigation methodology is changed.

Conclusion

Groundwater resources have been evaluated for the Yeşilköy aquifer, which is subjected to numerous pressures by ongoing intensive agriculture and expected tourism development at Karpaz Peninsula. The Yeşilköy aquifer is the main water source for local use at Karpaz Peninsula. A well survey, field survey and pumping test have been performed in this region to evaluate the physical and hydrogeological conditions of the basin. The detailed information about the wells was collected during well survey studies, whereas the field survey was successful to define in detail the crop pattern of the area. The hydraulic characteristics of the aquifer were determined by pumping tests performed in 2007 by McKinney and Brown. The transmissivity and storage coefficient of the aquifer were calculated as 1,782 m2/day and 0.0012. The annual groundwater abstractions for agricultural and domestic purposes from the aquifer have been estimated as 0.906 and 0.66 MCM, respectively. Evapotranspiration estimates were carried out using field survey results, in which the crop pattern of the basin was mapped and precipitation rates were gathered from meteorological stations close to the study area. Recharge of the aquifer has been estimated on the basis of the difference between the monthly evapotranspiration and precipitation values in which the drainage area of the basin is expected to have no surface runoff from the infiltration area. The resultant annual groundwater recharge due to precipitation and irrigation return has been estimated as 1.01 and 0.09 MCM, respectively. On the basis of the water balance study, the annual deficit from the aquifer is estimated to be 0.496 MCM.

The specific yield and the safe yield analyses of the aquifer were carried out on the basis of the water level fluctuations, which have been monitored for 10 years. The regional specific yield has been estimated as 3 % (0.03) for which safe yield analysis has been calculated as 0.84 MCM annually.

Time dependent water level fluctuations of the aquifer have been defined in terms of dimensionless parameters named as head parameter and period parameter and an empirical sixth order equation is obtained to describe the water level fluctuations of the aquifer. The equation was used to predict the short-term behavior of the water level if the present hydrogeological conditions prevail.

It is concluded that currently the aquifer is overexploited as the abstraction rates exceed both recharge and safe yield of the aquifer. It is recommended to optimize the irrigation water abstraction from the aquifer, which can be achieved by decreasing the water consumption with the application of water-saving techniques like drip irrigation or adapting full-cost pricing of water use with the implementation of metering to the irrigation wells. Abandoning the domestic water supply wells and replacing them by other sources like seawater desalination can be helpful to minimize the abstractions from the aquifer.