Carbonates and Evaporites https://doi.org/10.1007/s13146-018-0418-1 ORIGINAL ARTICLE Surface runoff and carbonates‑based definition of protection zones for Egirdir Lake in western Turkey Muhterem Demiroglu1 · Remzi Karaguzel1 · Mahmut Mutluturk2 · Cenk Yaltirak1 · Tolga Yalcin1 · Asli Donertas3 · Aysen Davraz2 · Zeynep Aktuna1 Accepted: 30 December 2017 © Springer-Verlag GmbH Germany, part of Springer Nature 2018 Abstract Freshwater of Isparta and Egirdir is supplied from the Egirdir Lake, which is the second largest freshwater lake of the Lakes District in Western Turkey. The Egirdir Lake has been studied within the framework of the Basin Protection Plan Special Provisions of the Egirdir Lake. The impact of runoff is taken into account in determining protection zones of the surface water reservoirs in Turkey. An approach that emphasizes the impact of groundwater flow in addition to the surface runoff has been adopted in this study. Water in Lake Egirdir is often classified as the Class II water according to terrestrial water resources quality criteria in Water Pollution Control Regulation of Turkey. The geological and hydrogeological studies reveal a significant amount of groundwater recharge into the Egirdir Lake through carbonate rocks and alluvial deposits outcropping in the basin, which is why Egirdir Lake still has a less contaminated water quality in spite of heavy pollutants. For this purpose, groundwater flow is prominently used in defining protection zones and surface runoff as well. The inner protection zone, which is defined as the 50-day travel time, and the outer protection zone, defined as the 400-day travel time, were estimated by infiltrometer and pumping tests in alluvium. Pumping tests results were used for the determination of hydraulic conductivities and groundwater levels for the determination of hydraulic gradients. Protection zones in karstic areas are based on the vulnerability map and large karstic springs. Keywords Egirdir Lake · Groundwater · Protection zone · Travel time Introduction Freshwater for the town of Isparta and the Egirdir district in Western Turkey, which amount of 30 hm3/year is supplied from the Egirdir Lake as well as 335 hm3/year for irrigation (IPDEU 2013). The Egirdir Lake with a water level of 917 m, 482 km2 surface area, 6–7 m average depth, is a tectonic lake which has about 4 billion m3 of water potential. The Egirdir Lake is being rapidly polluted due to a variety of activities carried out in the Egirdir Lake watershed. Point pollutants such as urban waste water, urban solid waste, * Muhterem Demiroglu demiroglum@itu.edu.tr 1 Department of Geological Engineering, Istanbul Technical University, Istanbul, Turkey 2 Department of Geological Engineering, Suleyman Demirel University, Isparta, Turkey 3 TUBITAK-MAM, Environment Institute (CE), Istanbul, Turkey industrial pollution loads and diffuse pollutants such as agricultural fertilizers, pesticides, animal waste and leakages are discharged into the rivers that recharge the lake. Domestic wastes of towns surrounding the lake are discharged into the rivers without proper treatment. Main leather processing and rose oil production are the most important and environmentally risky activities in the lake area. The use of pesticides and chemical fertilizers in agriculture create additional pollutant sources. As a result, the current operational status of the lake (drinking water supply, irrigation, fisheries production, and recreational use) is adversely affected (Beyhan and Kacikoc 2008; Sener 2010; Sener et al. 2013; Bulut et al. 2016). Up to now, maximum benefit has been achieved from the lake; however, measures taken have not been able to protect the lake from being contaminated. For this aim, protection zones of the lake should be defined properly. Given the growing population and inevitable demands for changing land use, protection zones should be as large as necessary and as small as possible (Alfoldi 1986; Kacaroglu 1999). 13 Vol.:(0123456789) Carbonates and Evaporites According to the “Water Pollution Control Regulation” in Turkey (SKKY 2008), drinking water reservoirs/lakes are protected by water protection zones in which certain activities are prohibited. They are (1) the absolute protection zone (0–300 m), (2) the inner protection zone (300–700 m), (3) the mid-range protection zone (700–1000 m) and (4) the outer protection zone (1000–3000 m). All activities are prohibited in the absolute protection zone (0–300 m) but this is not fully operational at present and only sensitive water bodies like Egirdir Lake that could become eutrophic in the near future are taken under control with assigned special provisions; i.e., it is crucial to protect the quality of the Egirdir Lake to ensure that water is completely safe to drink and to assure its sustainable use. For this purpose, the Egirdir Lake was studied within the framework of the “Basin Protection Plan Special Provisions of the Egirdir Lake and Assigning Project” (TUBITAK 2012) of the Ministry of Environment and Forestry. Lake water is often classified as the Class II water according to the terrestrial water resources quality criteria (SKKY 2008). The lake is directly recharged by significant amounts of groundwater (260 hm3/year), approximately 25% of total recharge, which explains why the Egirdir Lake is still less contaminated in spite of heavy pollutants. Quality of groundwater was determined almost as the Class I water according to terrestrial water resources quality criteria in Water Pollution Control Regulation of Turkey (SKKY 2008) and groundwater flow were predominently used, as well as surface runoff, in defining protection zones. The inner protection zone, which is defined as the 50-day travel time, and the outer protection zone, defined by 400-day travel time, were estimated by infiltrometer tests in alluvium. Pumping tests were used for the determination of hydraulic conductivities through Jacob and Cooper–Jacop time-drawdown plots for unsteady regime, and groundwater levels for the determination of hydraulic gradients. Data were used from Soyaslan (2004) and Seyman (2005). In this study, geological and hydrogeological structures reveal that approximately the 260 million m3/year of groundwater, mainly from the Kumdanli–Kasikara karstic system, recharge into the lake. In this sense, groundwater recharge areas were prominently used in defining protection zones in addition to surface protection zones. Karst aquifers exposed at the north and east of Hoyran recharge the lake directly with important springs (Tasevi, Asagitirtar, Kemerdamlari, Kayaagzi) assigned to the absolute protection area due to widespread developed karstic structures. No activity is allowed below an elevation of 919 m, which was determined as the maximum water level in the lake. No new construction is allowed in the lake green zone, which is the 30 m-width from the maximum water level within the absolute protection area. The tree and all plant species in this area are protected and bare areas are planted. Organic farming is to be resumed in agricultural areas located in the lake protection 13 area, which is determined as a 270 m-horizontal distance from the green zone in the absolute protection area. In the short-distance protection area determined as the 700 m-horizontal distance from the absolute protection area, good agricultural practices are implemented for the periods defined in the Watershed Conservation Plan Implementation Program. Industrial establishments that operate completely dry, do not produce hazardous waste, and do not produce industrial wastewater are allowed in the long-distance protection area. Geology and hydrogeology of study area The study area, the Egirdir Lake basin, located in the Lakes District in western Turkey is a rich region with regard to the freshwater potential. Egirdir Lake, is the second largest freshwater lake in western Turkey (Fig. 1.) Geological units of different ages and facies occur around the Egirdir Lake. The main units are composed of Paleozoic aged metamorphic rocks, Mesozoic aged carbonates, melanges, ophiolites and Tertiary aged gravelly, sandy clayey rocks. The main tectonic structure of the study area is the Isparta angle. The western edge of the Isparta angle forms the left lateral oblique, the Burdur-Fethiye Fault Zone (Senel et al. 1996). The Sultandagi thrust fault extending from the NW–SE and the normal Quaternary faults in front of it, forms the Sultandag mountains (Fig. 2). Egirdir lake and the nearby Beysehir lake take place inside the triangular shape of Isparta angle (Fig. 2). The Anamas Mountain, which extends in NW–SE direction, takes place in between the two lakes like a horst. The Egirdir and Hoyran lakes are two different tectonic basins. The Hoyran Basin underwent a 40° counterclockwise rotation relative to the simultaneously formed Egirdir-Kovada graben to their south (Karaman 2010). The Kovada Graben is a young tectonic structure with a linear feature extending from N to S about 2 km wide and 33 km long (Fig. 2). The NW–SE directional normal faults that form a fragmented structure, takes place at the NE edge of the Yalvac-Gelendost watershed. Miocene aged sediments cover these faults. The NE–SW trending normal fault forming the NW margin of the Yalvac-Gelendost watershed is overlain by the Miocene aged carbonate deposit (Figs. 1, 2). Mesozoic aged limestones and Paloecene aged olistostromes, situated on the impermeable ophiolite complex, show developed karst structures. The main karst structures are controlled by new tectonic deformation and layers rather than paleotectonic deformation. Springs drain into the Egirdir Lake and sinkholes located in the lake coast take place along the young fault plane (Fig. 3). The upper level of the Mesozoic aged deposits include the thin clayey limestone in lower Paleocene age. The Mesozoic aged series are overlain by Eocene aged flysches and Oligocene aged Carbonates and Evaporites Fig. 1 Location map of the Egirdir Lake basin conglomerates. Lower Miocene aged volcanic series on this unit have been overlain by Miocene aged terrestial-lacustrine series. Tertiary aged rocks, composed of claystone, sandstone and conglomerate are located in the east of Egirdir lake. Volcanic units are located in the west of the lake and overlie carbonates and flysch unconformably. Quaternary aged deposits are composed of materials such as clay, silt, sand, and gravel, unconformably covering all of the other lithological units (Fig. 3) (Sener 2010; Sener et al. 2013). The Barla Mountain, delineated by active normal faults in the southern boundary of the Uluborlu Senirkent Watershed, is located at the northern edge of the carbonate rocks with a thickness of more than 1500 m (Fig. 3). Kayaagzi spring discharges from karstic conduits extending to the main fault plane. The cave system has been developed on the normal fault plane outside the water line of the Hoyran Lake Basin. The springs discharge from the Mesozoic aged limestones overlain by the Kasikara Miocene aged sediments on the lakeside located at Kumdanli Watershed. Fossil caves were observed due to the upward movement of the basin by active fault systems in the south-west. The springs discharge into the Egirdir Lake and sinkholes that are located along the young fault planes (Fig. 3). A hydrogeological map has been prepared according to the physical properties of the outcrops in the Egirdir Lake Basin and hydraulic properties determined from pumping tests in these units (Fig. 3). Geological units, located around the Egirdir Lake, were primarily classified into two main groups of rock and granular environment, and then divided into sub-groups according to their aquifer parameters (with specific capacities > 2 l/s/m). Karstic carbonates Karstic carbonates are defined as limestones that are thickbedded massive geological units in contact with the Egirdir Lake. Limestones in the region belonging to different time periods (Paleozoic and Mesozoic) have mostly formed karst aquifers; discharge rates in karstic springs vary between 10 and 1000 l/s (Asagitirtar spring). These carbonates are classified as extensive, rich aquifers. Limestones with high topography in large areas are observed in the east, west and north of the Egirdir Lake basin. Both recharge and discharge may occur due to karstic structures. The main flow paths are based on conduit flows developed along the tectonic discontinuities. Many of the large springs occur at the contact 13 Carbonates and Evaporites Fig. 2 Main tectonics lines of Isparta angle and around (BFSZ Burdur-Fethiye Fault Zone, IA Isparta Angle, SDF Sultandagi thrust fault, KG Kovada graben, KL Kovada Lake, AM Anamas Mountains, BL Beysehir Lake, HL Hoyran Lake, EL Egirdir Lake, BUL Burdur Lake) (Hall et al. 2014a, b; Elitez et al. 2016) between highly impermeable and low permeable formations. Karstic aquifers mostly have high reservoir capacities, rapid transportation and easy response to contamination (Palmer 2010; Bakalowicz 2015; Parise et al. 2015; Gunay et al. 2015). For these reasons they are assessed with a particular attention. Springs mapped on the Egirdir Lake basin carbonates (Fig. 3). Groundwater flow are controlled by faults and fracture systems (Davraz et al. 2006, 2009; Karaguzel et al. 2006) for which the Tirtar and Ugullu springs (in Figs. 4, 5, respectively) can be given as examples. Large discharge rates were considered as clear evidences for flow paths which did not required further evidences such as tracer tests. Discharge is a function of the groundwater level in the hinterland on the area of principal recharge of the karst massif (Bonacci 2001). Groundwater is mainly recharged from the karstic environment at the north of the lake (Tasevi and from around Asagitirtar) and in the east (Kemerdamlari, Yenice, Yesilkoy coastal and underwater areas) (Fig. 3). Tirtar spring (with discharge rate of 1000 l/s) was showed on the cross-section (in Fig. 4). Wells in this region (Yalvac-Gelendost watershed) formed especially from limestone have high flow rates (Fig. 4). Hydraulic conductivity of the aquifer was determined between the range of 4 × 10−6–6 × 10−2 m/s, transmissibility changes within the range of 8.5 × 10−4–6.2 × 10−2 m2/s, and the storage of coefficient is between 0.005 and 0.38 (Soyaslan 2004). Limestone units are exposed in the north and south of the Uluborlu–Senirkent watershed. The Ugullu spring discharges 350 l/s of water from a tectonic contact between allochthonous Jurassic-Cretaceous aged limestones and Tertiary-Paleocene aged Uluborlu formation (Figs. 3, 5). Aquifer parameters in these wells have hydraulic conductivity of 1.4 × 10−5–2.2 × 10−4 m/s and transmissibility 2.1 × 10−3–3.4 × 10−2 m2/s (Seyman 2005). 13 Granular environment Granular environments are unconsolidated geological units exhibiting the characteristics of soil. Plio-Quaternary aged alluvial fans, sandy lake sediments, alluvial deposits, and Miocene aged conglomerates are classified as extensive rich aquifers in the Egirdir Lake Basin. Alluvial deposits and alluvial fans are the major granular aquifers in the Carbonates and Evaporites Fig. 3 Hydrogeological map and geological cross-sections of the Egirdir Lake basin Fig. 4 The Tirtar spring hydrogeological cross-section Egirdir Lake Basin. These units are exposed from the lake coast along the stream beds in an area of approximately 224.5 km2. Alluvial deposits show widespread distribution in the Uluborlu–Senirkent, Kumdanli and Gelendost–Yenice areas. These areas are important fruit growing areas of the basin where pesticides are used. Total PAH (Polycyclic aromatic hydrocarbons) and total pesticide in the groundwater are less than limit values that determine the class of water quality. Total pesticide was found to be less than 1 μg/l, this limit was not exceeded in six sampling stations. However, the limit value of the Class I water, 20 μg/l, was exceeded in 19 sampling stations in the dry period (EGKK, 2012). The highest concentration (48.02 μg/l) was recorded in Yenice-Afsar region where 13 Carbonates and Evaporites Fig. 5 The Ugullu spring hydrogeological cross-section agricultural activity and the use of arsenic-containing pesticides are intensive, suggesting that the pollution stems from pesticides. All of these alluvial areas are defined as permeable granular environments. The thickest and widespread sand and gravel levels forming a porous aquifer were cut in the Yalvac Plain. The thickness of this unit in the north and south of the Yalvac Plain is more than 200 m according to the geophysical survey of Topcam et al. (1977). The Gelendost granular aquifer reaches a maximum thickness in the central part of the plain. The thickness of the alluvium in the region was found to be 150–220 m according to geophysical measurements of Topcam et al. (1977). It is reduced to around 50–70 m thickness towards the north of the Gelendost plain. The alluvial unit is located in the sprawling plains throughout the Pupa creek within the Senirkent-Uluborlu graben and has about 140 km2 area. It is gravelly, sandy, clayey gravelly sand, sandy gravel and silt. The gravelly and sandy levels of the alluvium increases in the coasts of the Pupa Creek due to conglomerates coming as erosion materials. In the Senirkent-Uluborlu plain, Pliocene aged sediments consisting of polygenic coarse gravel, clay, silt and sand have similar hydrogeological characteristics with this alluvial deposit. Silt and conglomerates are mostly cut in wells in the northeast of Yalvac-Gelendost basin. Hydraulic conductivity of the alluvium varies between 3.6 × 10 −7 and 6.5 × 10 −6 m/s, transmissibility between 7.2 × 10 −5 and 2.5 × 10 −4 m 2/s, and storage coefficient between 0.003 and 0.008 in these wells, respectively (Soyaslan 2004). Hydraulic parameters of the alluvial aquifers in this region are reduced depending on the increase in the thickness of the clay level. Eocene aged thin-bedded marl and Miocene aged volcanic series are classified as fractured poor aquifers. Paleozoic aged schists which form the basis of the study area, Mesozoic aged ophiolite complex, Eocene aged flysch and Neogene aged clays are classified as very poor aquifers. 13 Material and method There are methods such as fixed distances (buffer zones), time of travel, and modelling studies for the delineation of the protection zones (US EPA 1997). A 50-day travel time is based on biological contaminant decay. It is designed to protect against the transmission of rapidly degrading toxic chemicals and some water-borne diseases. The 400-day travel time is based loosely on consideration of the minimum time required to provide delay, dilution and attenuation of slowly degrading pollutants (Carey et al. 2009). Darcy’s Law is only applicable under laminar flow conditions (Fetter 2001; White 2002). Groundwater flow occurs in large diameter conduits with high flow velocities (Palmer 2010; White 2002). As a result, Darcy’s Law is inapplicable within conduits. Through sinkholes and other epikarst features, contaminants can easily reach and spread rapidly over large distances in the conduit network (Figs. 4, 5). Natural attenuation processes related to filtration and retardation are often very low. Only dilution is the main process. Pollution is not treated in the karstic groundwater system. It is only transferred due to rapid flow, retention times for degradation processes are low and even particulate matter can be transmitted in turbulent flow. Because of the facts stated above, karst aquifers are extremely vulnerable to groundwater contamination (Ford and Williams 1989). In this manner, travel time method was used for granular environment in which flow type is laminar and Darcy Law is applicable. Travel times varies greatly even in the same area depending on the karst features in the Egirdir lake basin. Protection zones in karstic areas are based on the vulnerability map and persistence of developed karst structures in recharge areas with high discharge rates in discharge areas. The very high and, high vulnerable areas were prioritized. First, the vulnerability maps were prepared, taking into account the important parameters affecting the Carbonates and Evaporites Table 1 Ranges and weights of physical parameters Weight: 5 Class Weight: 2 Range Hydraulic conductivity (m/day) 0.001–0.1 1 0.1–1 2 1–2 4 2–4 6 4–10 8 10–30 9 + 30 10 Weight: 3 Weight: 5 Result Class Range Result Class Range Result 5 10 20 30 40 45 50 Soil type No soil Course texture Medium Fine Depth > 20 cm > 25 cm > 35 cm > 45 cm 10 9 8 3 7 6 5 4 20 18 16 6 14 12 10 8 Vadose zone Silt/clay Sand/gravel Volkanics Limestone Bedded lmt, sanstone Shist Alluvium Ophilitic comp 1 6 2 10 6 1 8 1 3 18 6 30 18 3 24 3 Weight: 2 Weight: 1 Weight: 5 Class Range Depth to groundwater (m) <1 10 1–2 9 2–4 8 4–10 7 10–30 6 30–65 5 > 65 3 Result 50 45 40 35 30 25 15 Weight: 3 Class Range Result Class Range Result Class Range Result Class Range Result Recharge (mm/year) > 400 400–300 300–200 200–100 < 100 5 4 3 2 1 10 8 6 4 2 Slope (%) 0–2 2–6 6–12 12–18 18 + 10 9 5 3 1 10 9 5 3 1 Distance to wells (m) 0–50 50–100 100–200 200–300 > 300 5 4 3 2 1 25 20 15 10 5 Aquifer type Karst Fructured Mixed Matrix Impermeable 10 6 7 7 1 30 18 21 21 3 Weight: 2 Weight: 5 Class Range Result Hydromorphology Hill Hillfront Plain 1 3 2 2 6 4 Weight: 3 Class Range Result Lineaments 0–200 200 + 5 1 25 5 Weight: 1 Class Range Result Class Range Result Land use Irrigated Dry land Green area Bare area 10 2 2 1 30 6 6 3 Drainage density 0.0–0.19 0.19–0.38 0.38–0.57 0.57–0.76 4 3 2 1 4 3 2 1 13 Carbonates and Evaporites determination of groundwater protection areas. These parameters are given in Table 1. The modified DRASTIC methodology was employed to define appropriate weights for the parameters (Aller et al. 1987; US EPA 1993; Demiroglu and Dowd 2013). The concept of groundwater vulnerability is that some land areas are more vulnerable to contamination than others and are defined as the zoning of different areas to protect groundwater against pollutants (Vrba and Zaporozec 1994). First vulnerability map was produced in France at 1:100,000 scale about 40 years ago. It has become more practical with the introduction of geographic information systems (GIS). Groundwater vulnerability assessment techniques have been developed to explain the effects on the environment during the transport of contaminants. There is no universal methodology for groundwater vulnerability assessment. In general, methods involving mathematical models are named with different symbols (Varol and Davraz 2010). The EPIK method was developed for karst aquifers (Doerfliger and Zwahlen 1995) while the AVI-method (Van Stempvoort et al. 1992), the PI Method (Goldscheider 2005), the PI method and the transit time method (Brosig et al. 2007) can be applied for all types of aquifers. One of the most widely used standard groundwater vulnerability methods is DRASTIC, developed by the United Fig. 6 The map of hydraulic conductivity distribution 13 States Environmental Protection Agency (USEPA) as a method for assessing groundwater pollution potential. This method uses seven parameters which are depth to water, net recharge, aquifer media, soil media, topography, impact of vadose zone media, and hydraulic conductivity of the aquifer in calculating a ‘vulnerability index’ (Aller et al. 1987). Also, in recent years, the DRASTIC method has been modified by using additional parameters and/or by ignoring the existing parameters according to the characteristics of the study area (Umar et al. 2009; Lee 2003; Simsek et al. 2006; Wang et al. 2007; Guo et al. 2007; Martinez-Bastida et al. 2010; Awawdeh and Jaradat 2010; Sener and Davraz 2013). Demiroglu and Dowd (2013) added lineatments, hydromorphology, land use and distance to wells into the DRASTIC method. This is the modified DRASTIC method used in this study. The aquifer type is a principal parameter in the vulnerability. The formations are classified on a scale of 1–10 depending on their features such as developed karstic structures, fractured, granular and impermeable units, and scaled to integrate with other maps. Protective cover or soil thickness of the upper unconsolidated zone, which includes both the soil and other geological overburden such as saprolite, epikarst and other quaternary deposits, is commonly regarded as one of the most important attributes in the Carbonates and Evaporites assessment of groundwater vulnerability (Doerfliger et al. 1999; Demiroglu and Dowd 2013). Protective cover is not a barrier to infiltrating contaminants, but it can supply time for the contaminants to degrade. Long infiltration times generally mean low aquifer vulnerability. The infiltration time depends on the thickness and hydraulic conductivity of the vadose zone. The soil type and the likely overburden thickness of the Egirdir Lake Basin area was defined using the non-commercial Google Earth satellite imagery at a ground pixel resolution of 10 m, borehole records, field observations and soil maps for this area. Soil maps data were obtained from State Hydraulic Works, the state owned organization responsible for water resources. Soil types are classified into eight classes depending on soil thickness and drainage capacity. Ranges and weights of soil parameters were given in Table 1. Groundwater depth and hydraulic conductivity (Fig. 6) overlays were prepared using an inverse distance weighted interpolation scheme. Hydraulic conductivity is a critical component in the vulnerability assessment. High hydraulic conductivity results in rapid contaminant movement. Additional parameters that affect vulnerability include hydromorphology, slope, drainage density, recharge, land use, lineaments and distance to wells. All lineaments that increased permeability and porosity were buffered by a distance of 200 m, and each buffer zone was assigned a value of five. Wells can act as a direct pathway for pollutants from the land surface into the water supply. Thus a major consideration in groundwater contamination is the position and condition of the well (De Silva and Hohne 2005). These features are assigned different weights depending on their importance with respect to vulnerability. Different classes are assigned weights ranging from one to five (Israil et al. 2006) depending on their influence on the groundwater vulnerability (Demiroglu and Dowd 2013). The topographical digital data were obtained from Mineral Research and Exploration General Directorate (MTA) of Turkey. The Digital Elevation Model (DEM) was prepared with ArcGIS 3D analyst, and the Triangular Irregular Network (TIN) and slope map were formed from the DEM. Slope gradation was estimated in degrees and reclassified Fig. 7 The groundwater vulnerability map of Egirdir Lake basin (TUBITAK 2012) 13 Carbonates and Evaporites Table 2 Infiltrometer test results (permeability of the soil in 2 h and 1 day) Infiltrometer experiment no. Average water intake rate (I0), (mm/h) T = 1440 min T = 120 min Average water intake rate (mm/h) Average water intake rate (mm/h) INF 1 INF 2 INF 3 INF 4 INF 5 INF 6 INF 7 INF 8 INF 9 INF 10 570 × T−0.37 0 65 × T−0.185 Canceled 48 × T−0.30 400 × T−0.15 30 × T−0.40 Canceled 120 × T−0.33 40 × T−0.30 38.7 (1.1 × 10−5 m/s) 0 (1 × 10−9 m/s) 17.0 (4.7 × 10−6 m/s) – 5.4 (1.5 × 10−6 m/s) 134.3 (3.7 × 10−5 m/s) 1.7 (4.5 × 10−7 m/s) – 10.9 (3.0 × 10−6 m/s) 4.5 (1.3 × 10−6 m/s) 97.0 (2.7 × 10−5 m/s) 0 (1 × 10−9 m/s) 26.8 (7.5 × 10−6 m/s) – 11.4 (3.2 × 10−6 m/s) 195.0 (5.4 × 10−5 m/s) 4.4 (1.2 × 10−6 m/s) – 24.7 (6.9 × 10−6 m/s) 9.5 (2.6 × 10−6 m/s) Fig. 8 N–S schematic illustration of alluvium in the Uluborlu–Senirkent (no scale) into four groups: 0–2, 2–6, 6–12, 12–18 and > 18°. Slope is important because it is linked to recharge rates. Low slopes allow longer time for infiltration. Very steep slopes, coupled with high drainage density, result in low infiltration and high runoff volumes. Therefore, high drainage density, which indicates a high runoff component was assigned a value of one (Demiroglu and Dowd 2013). Maps of each parameter were reclassified at each step to generate a composite map of the study area. These maps were assigned the appropriate weights, giving each a relative influence. The higher the resulting value, the greater the vulnerability, with the final map prepared by adding the scores of various parameters for each pixel (Fig. 7). Second, infiltrometer tests were carried out at selected locations in order to determine the permeability of the covered layer. There are three important alluvial areas in Uluborlu–Senirkent watershed, Kumdanli watershed, Yalvac–Gelendost watershed (Figs. 1, 3). These areas are places where orchard activities are made and intensive pesticides are used. Soil maps were also used to establish the test sites. Hand auger drilling was done at the test locations, 13 and samples were taken every 50 cm. Grain size distribution, consistency and classification of soils were defined following laboratory tests on the samples. The hydraulic conductivities of the soils have been determined by empirical methods based on the soil classifications of the field samples as well. Darcy equation is used for determining a water transit time of 50 days and time of 400 days. It is defined as: V = − Ki (1) where, V is velocity of groundwater flow, i is the hydraulic gradient of flow which is a negative value, and K is the hydraulic conductivity. The hydraulic gradient of flow is defined by (2) in which ΔH is the head difference and ΔL is the horizontal distance between the two points considered. Darcy equation can then be extended to i = ΔH∕ΔL V = −K ΔH nΔL (3) Clean sandy clay Good graded Gravel, silt, sand Clean clay Clean clay Clean clay Sandy clean clay Sandy silty clay Sandy silty clay Silty sand Silty sand and gravel Silty sand and gravel 9.3493E−09 0.00049708 8.1754E−09 8.3547E−09 6.802E−09 2.0842E−06 1.0155E−08 1.8998E−08 4.9318E−07 1.128E−06 5.078E−07 23 1 29 28 25 12 13 10 3 2 3 41 5 63 62 65 42 44 44 26 18 17 28 31 8 10 10 41 40 44 65 57 43 6 63 0 0 0 5 3 2 6 23 37 14 – 13 14 13 8 6 5 NP NP NP 16 – 20 20 19 19 20 20 NP NP NP 30 – 33 34 32 27 26 25 NP NP NP 2.70 2.74 2.68 2.69 2.69 2.53 2.52 2.50 2.55 2.58 2.59 0.00–0.50 1.00–1.50 0.00–0.50 0.50–1.00 1.00–1.50 0.00–0.50 0.50–1.00 1.00–1.50 0.00–0.50 0.50–1.00 1.00–1.30 PL (%) INF 1-1 INF 1-2 INF 2-1 INF 2-2 INF 2-3 INF 3-1 INF 3-2 INF 3-3 INF 4-1 INF 4-2 INF 4-3 14 – 19 18 17 20 22 24 15 14 13 Clay (%) Silt (%) Sand (%) Gravel (%) LL (%) PI (%) Grain size distribution Atterberg limits γs (g/cm)3 Depth (m) Wn (%) The groundwater vulnerability map of the Egirdir Lake basin is given in Fig. 7. Particular attention was paid to the “high” and “very high” vulnerable areas in defining the absolute protection areas around the lake. Karstic areas in the northern part of basin, the Uluborlu–Senirkent alluvial areas in the western part of the basin and some local alluvial areas in the eastern part of the basin are found most critical when the vulnerability map is analyzed in terms of groundwater pollution (Fig. 7). However, the local alluvial areas in the eastern part of the basin are excluded as they are shallow aquifers and far from recharging the lake. Irrigation, pesticide usage and other studies in agricultural activities were taken into account in defining the influence coefficients for the vulnerability map. Therefore, some areas, which have less risk of contamination, seems to be more critical. On the other hand, limestones have a greater risk of contamination than the agricultural activity areas; however, they were identified less critical to contamination due to the lack of settlement and agricultural activity. Uluborlu–Senirkent alluvial area covers the alluvium deposited by the Pupa river flowing in an east–west direction. The Pupa creek sediment which is occurred gravel-sand size at the source and deposited in the Egirdir Lake. It covers an area about 30 km-long. The grain size changes into silt–clay dimensions where approaches to the lake because of the velocity of the stream flow. The vast majority of the sediment forms carbonate pebbles rather than sand and clay. On the other hand, the areas close to the Egirdir Lake are covered by decomposed material from the Neogene units. It is very difficult to distinguish the alluvium-weathered Neogene boundary in this area. However, observations, sampling studies, and the physical properties of the samples clearly show that the Neogene clays are decomposition products. Infiltrometer tests at four points were carried out in order to determine leakage occurring in this area, adjacent to the lake water surface. The infiltration rates calculated from infiltrometer test results at different locations in the study area are presented in Table 2. The change of grain size distribution is more evident (Fig. 8). The accumulation of gravel and sand material is more intense at the southern region due to slope debris at the edge of the slope-induced fault, clayey in the middle sections, and sandy in the northern sections. As can be seen in Table 3, the hydraulic conductivity values are relatively Sample no. Table 3 Infiltrometer experiment results of samples taken by hand auger from Uluborlu–Senirkent alluvium to calculate groundwater velocity from one point with higher head to another with lower head, where, n is the effective porosity. Hydraulic conductivity (K in m/day) was calculated from pumping test data, the hydraulic gradient (i) was determined using the maps of groundwater level. CL GW–GM CL CL CL CL CL–ML CL–ML SM SM SM Class (USCS) K (m/s) Group (USCS) Carbonates and Evaporites Results and discussions 13 Carbonates and Evaporites Fig. 9 N–S schematic illustration of alluvium in the Kumdanli plain (no scale) high in the sandy and gravelly levels of the INF 1- and INF 4- experimental locations. Hydraulic conductivity is very low in the central part of the basin (INF 2). Hydraulic conductivity was calculated from the pumping tests as 5.99 × 10−6 and 1.99 × 10−5 m/s in the basin of Senirkent-Uluborlu area. The water travel distance was calculated as 31 m for 50-day and 253 m for 400-day in Uluborlu–Senirkent Plain. Kumdanli alluvium plain This area consists of alluviums of the Kumdanli and Kasikara streams. It is located in the northeast-southwest part of the lake watershed area. The alluvium is composed of sand, silt and clay. Gravelsized materials are very few. The areas approaching the lake are constantly under water and marsh because of the low slope. The infiltrometer experiments were conducted in the area adjacent to the lake to determine the infiltration from surface and subsurface at the three locations. Soil samples were taken at 50 cm intervals with a hand auger at the same locations. Classification and identification tests were carried out on the samples (Table 3). It is very difficult to distinguish the alluvium-weathered Neogene boundary in this area too. The experimental results reveal that the region is covered completely by clays (Fig. 9). Organic clays are clearly distinguished by color and odor in the identified central parts. Dense plant roots are observed on the surface in the organic peat areas. Therefore, relatively high infiltration values (K = 10−6 m/s) were found (Table 4). However, the underlying CH group soils infiltration values K are much too low (10−8 m/s) and shows the very low seepage from the surface in this area. The water travel distance was calculated as 0.5 m for 50-day and 4 m for 400-day in Kumdanlı Plain. Gelendost–Yenice alluvium plain This area is within the Neogene units in terms of geology. Previous studies have mapped these areas as alluvium. However, a large part of plains consists of weathered Neogen clays. It is difficult to distinguish the river-deposited alluvial materials within the Neogene units from the Neogene clays. The Gelendost–Yenice alluvium plain consists of two separate areas. The first is 13 a large part of the clay and sandy beach in the coastal region in which Afsar and Yenice are located and the north Afsar Creek passes through. The remaining alluvial Gelendost area the highway crosses over is consists of clays and sandy clay and marsh. The groundwater level is on the surface in the south. Soil samples were taken at 50 cm intervals with a hand auger at three locations (Fig. 3). Classification and identification tests were carried out on the samples (Table 5). A schematic cross-section of the underground structures containing both areas is shown in Fig. 10. The dense peat areas are seen particularly in the south. Hydraulic conductivity was calculated as 10−6–10−7 m/s using pumping tests in wells drilled in this region. This part of the plain is defined as the local permeable granular media. However, the classification of Yenice samples and the infiltrometer experiment results show the impervious environment (k = 10−9 m/s), but the region is classified as a local-permeable granular media from pumping tests. The water transit distance in the alluvial areas were calculated as 21 m for 50-day and 168 m for 400-day in Gelendost Plain, and it was calculated as 7 m for 50-day and 56 m for 400-day in Yenice surrounding. Bedre alluvial area There is a sandy beach in this area 25–30 m in length, with other areas covered by red gravelly sandy clays (terra rosa) derived from residual carbonates depending on the geological structure. Karababa alluvial area This area is quite short, and the rear of this area (a short band of sand, gravel, limestone) is completely covered with red gravelly sandy clays (terra rosa). Yesilkoy, Mahmatlar alluvial area This region appears between these two areas of deposition. The soils have relatively low clay content, and therefore higher permeability. However, the ophiolitic units associated with limestones are observed in this area. Ophiolitic units are generally defined as impervious even if limestones blocks in that are permeable. Sample no. INF 5-1 INF 5-2 INF 5-3 INF 9-1 INF 9-2 INF 9-3 INF 10-1 INF 10-2 INF 10-3 Depth (m) 0.00–0.50 0.50–1.00 1.00–1.50 0.00–0.50 0.50–1.00 1.00–1.50 0.00–0.50 0.50–1.00 1.00–1.50 Wn (%) 24 23 21 15 15 14 19 18 30 γs (g/cm)3 2.61 2.59 2.60 2.66 2.70 2.70 2.69 2.70 2.70 Atterberg limits Grain size distribution LL (%) PL (%) PI (%) Gravel (%) Sand (%) Silt (%) Clay (%) 60 54 57 63 54 53 82 69 48 22 24 23 24 24 23 28 28 23 38 30 34 39 30 30 54 41 25 1 1 1 0 0 0 0 0 0 21 42 25 3 5 7 4 2 9 41 27 37 38 50 50 41 46 59 37 30 37 59 45 43 55 52 32 Class (USCS) K (m/s) Group (USCS) CH CH CH CH CH CH CH CH CL 8.7631E−09 3.6505E−09 7.8548E−09 1.24E−09 1.0361E−08 2.9395E−08 1.1781E−08 1.15E−08 8.878E−09 Greasy clay and sand Sandy greasy clay Greasy clay and sand Greasy clay Greasy clay Greasy clay Greasy clay Greasy clay Clean clay Table 5 The test results of the samples taken by hand auger from Yenice-Gelendost alluvial area Sample no. 13 INF 6-1 INF 6-2 INF 6-3 INF 7-1 INF 7-2 INF 7-3 INF 7-4 INF 7-5 INF 8-1 INF 8-2 INF 8-3 Depth (m) 0.00–0.50 0.50–1.00 1.00–1.50 0.00–0.50 0.50–1.00 1.00–1.50 1.50–2.00 2.00–2.10 0.00–0.50 0.50–1.00 1.00–1.40 Wn (%) 27 12 17 11 12 13 15 19 18 18 16 γs (g/cm3) 2.65 2.58 2.62 2.70 2.73 2.71 2.73 2.73 2.71 2.72 2.70 Atterberg limits Grain size distribution LL (%) PL (%) PI (%) Gravel (%) Sand (%) Silt (%) Clay (%) 83 93 91 25 – – 26 26 36 49 60 38 32 31 18 – – 17 18 21 22 24 45 61 60 7 – – 9 8 15 27 36 0 0 0 3 3 3 2 0 0 0 0 4 2 1 44 50 57 55 44 36 5 1 48 28 29 43 34 28 32 39 43 57 41 48 70 70 10 13 12 11 17 21 38 58 Class (USCS) K (m/s) Group (USCS) CH CH CL–ML SM SM SC CL CL CL CH 10−8 10−8 1.8281E−08 10−5–10−4 10−5–10−4 6.67E−08 6.67E−08 4.9444E−09 1.01E−08 1.196E−08 Greasy clay Greasy clay Sandy silty clay Silty sand Silty sand Clayey sand Sandy clean clay Sandy clean clay Clean clay Clean clay Carbonates and Evaporites Table 4 Infiltrometer experiment results of samples taken by hand auger from Kumdanli alluvium Carbonates and Evaporites Fig. 10 N–S schematic illustration of alluvium in the Gelendost–Yenice (no scale) Fig. 11 Protection zones and special provisions of the Egirdir Lake basin (EGKK 2012) 13 Carbonates and Evaporites Barla alluvial area This is the delta of the river that comes from behind this alluvial area in the west. However, the surface is completely covered by red and brown gravelly and sandy clays. A band of gravel and sand is observed in a very narrow area of the coastal region. This area is impermeable. Therefore, surface runoff was prominently used in defining protection zones. To summarize for the whole basin, surface runoff instead of groundwater flow rate should be considered in determining protection areas, because the groundwater flow rate is very low in alluvial areas around the lake (Uluborlu–Senirkent locality, Kumdanli plain, Gelendost–Yenice plains, the Karababa, Yesilkoy, Mahmatlar, Barla, alluvial areas). Groundwater recharge areas were prominently used in defining protection zones in karstic systems, mainly the Kumdanli–Kasikara area where important springs recharge the lake (Tasevi, Asagitirtar, Kemerdamlari, Kayaagzi). However, there is not a significant groundwater recharge from carbonate rocks outcroppings along the western coast of the lakes. On the contrary, lake water escapes to other basins through these karstic structures which are clogged by State Hydraulic Works to keep water in the lake. Therefore, special provisions were not assigned for these areas (Fig. 11). The allocthonous limestones, exposed at the north and east of the Hoyran, important karstic springs (Tasevi Gencali, Asagitirtar, Kemerdamlari, Kayaagzi) recharge the lake directly, are completely covered by the absolute protection area due to the widespread developed karstic structures and approximately 125 × 106 m 3/year discharge rate to the Lake (Fig. 11). Conclusions Growing population and increasing demands for change in land use are key elements in defining protection zones that should be as large as necessary and as small as possible. In this sense, different criteria should be applied for different basins in establishing surface and groundwater protection areas. The 50-day groundwater transit length for the absolute protection area in the Egirdir Lake, Western Turkey, granular environments vary from a few meters to a few tens of meters. Accordingly, taking into account the 50-day groundwater travel distance, the absolute groundwater protection area is considered to be less than 300 m, the absolute protection zone width forced by Water Pollution Control Regulation in Turkey. On the contrary, when carbonates or geology based protection zones are concerned, the absolute protection zone covers wider areas. The vulnerability assessment performed in this study to define the lake protection areas could be validated by using hazard and risk maps together with tracer tests as a further step. Acknowledgements The authors would like to thank the TUBITAK Marmara Research Center for providing the valuable support. The manuscript has been edited in terms of its language by Darwin E. Fox, whom the authors deeply thank for the helpful comments improving the manuscript. Funding was provided by the scientific and technological research council of Turkey (TUBITAK). References Alfoldi L (1986) Groundwater protection in porous, fractured and karstic aquifer systems (general report)’. In: Symposium on groundwater protection areas (19th congress of IAH), proceedings, vol 19/2. Novinar Publishing House, Czechoslavakia, pp 89–103 Aller L, Bennett T, Lehr JH, Petty RJ, Hackett G (1987) DRASTIC: a standardized system for evaluating groundwater pollution potential using hydrogeologic settings. US-EPA 600 2-87-035, Washington DC, USA Awawdeh MM, Jaradat RA (2010) Evaluation of aquifers vulnerability to contamination in the Yarmouk River basin, Jordan, based on DRASTIC method. Arab J Geosci 3:273–282 Bakalowicz M (2015) Karst and karst groundwater resources in the Mediterranean. Environ Earth Sci 74:5–14 Beyhan M, Kacikoc M (2008) Management plans for conservation of water resources and Egirdir Lake case, water consumption, treatment, reuse: symposium book (in Turkish), pp 123–132 Bonacci O (2001) Analysis of the maximum discharge of karst springs. Hydrogeol J 9:328–338 Brosig K, Geyer T, Subah A, Sauter M (2007) Travel time based approach for the assessment of vulnerability of karst groundwater: the transit time method. Environ Geol 54(5):905–911 Bulut C, Kubilay A, Akcimen U, Yener O (2016) Seasonal variation of microbiological quality of Egirdir Lake. J Natl Appl Sci Suleyman Demirel Univer 20(1):50–53 Carey M, Hayes P, Renner A (2009) Groundwater source protection zones review of methods. Environment Agency, Rio House, Waterside Drive, Aztec West, Almondsbury, Bristol, pp 409. http://www.environment-agency.gov.uk Davraz A, Karaguzel R, Soyaslan I (2006) An approach to the underground water drainage between the Egirdir Lake and the Mediterranean Sea. In: Symposium of contemporary practices in engineering geology, proceeding book, Denizli, pp 293–302 Davraz A, Karaguzel R, Soyaslan I, Sener E, Seyman F, Sener S (2009) Hydrogeology of karst aquifer systems in SW turkey and an assessment of water quality and contamination problems. Environ Geol 58(5):973–988 De Silva CS, Hohne S (2005) Groundwater vulnerability assessment and protection measures—an example from Sweden. J Natl Sci Found Sri Lanka 34:43–51 Demiroglu M, Dowd J (2013) The utility of vulnerability maps and GIS in groundwater management: a case study. Turk J Earth Sci 23:80–89 Doerfliger N, Zwahlen F, (1995), Groundwater vulnerability mapping in karstic regions (EPIK), Practical Guide. Swiss Agency for the Environment. Forests and Landscape (SAEFL). Berne, pp 56 Doerfliger N, Jeannin P, Zwahlen F (1999) Water vulnerability assessment in karst environments: a new method of defining protection areas using a multi-attribute approach and GIS tools (EPIK method). Environ Geol 39(2):165–176 EGKK (2012), Basin Protection Plan Special Provisions of the Egirdir Lake (i), http://www.csb.gov.tr/gm/cygm/index .php?Sayfa=duyur udetay&Id=66 (visited on July 14, 2017) 13 Carbonates and Evaporites Elitez I, Yaltirak C, Aktug B (2016) Extensional and compressional regime driven left-lateral shear in southwestern Anatolia (eastern Mediterranean): the Burdur-Fethiye Shear Zone. Tectonophysics 688:26–35 Fetter CW (2001) Applied hydrogeology, 4th edn. Prentice Hall, Chicago, p 598 Ford DC, Williams PW (1989) Karst geomorphology and hydrology. Unwin Hyman Ltd, London, p 601 Goldscheider N (2005) Karst groundwater vulnerability mapping— application of a new method in the Swabian Alb, Germany. Hydrogeol J 13:555–564 Gunay G, Guner N, Tork K (2015) Turkish karst aquifers. Environ Earth Sci 74:217–226 Guo O, Wnag Y, Gao X, Ma T (2007) A new model (DRARCH) for assessing groundwater vulnerability to arsenic contamination at basin scale: a case study in Taiyuan basin, northern China. Environ Geol 52:923–932 Hall J, Aksu AE, Elitez I, Yaltirak C, Cifci G (2014a) The FethiyeBurdur fault zone: a component of upper plate extension of the subduction transform edge propagator fault linking Hellenic and Cyprus Arcs, eastern Mediterranean. Tectonophysics 635:80–99 Hall J, Aksu AE, King H, Gogacs A, Yaltirak C, Cifci G (2014b) Miocene-Recent evolution of the western Antalya Basin and its linkage with the Isparta Angle, eastern Mediterranean. Mar Geol 349:1–23 IPDEU (2013) Isparta Provincial Directorate of Environment and Urbanization, Isparta Provincial situation report, 2013 (in Turkish). http://www.csb.gov.tr/db/ced/editordosya/Isparta_icdr2012. pdf Israil M, Al-hadithi M, Singhal DC, Kumar B, Someshwar RM, Verma SK (2006) Groundwater resources evaluation in the Piedmont zone of Himalaya, India, using Isotope and GIS techniques. J Spatial Hydrol 6/1:105–119 Kacaroglu F (1999) Review of groundwater pollution and protection in karst areas. Water Air Soil Pollut 113:337–356 Karaguzel R, Davraz A, Soyaslan I, Seyman F, Sener S, Sener E (2006) Assessment of the Antalya water basins on pollution, Antalya drinking water and problems symposium proceedings, Antalya, pp 113–120 (in Turkish) Karaman M (2010) The tectonic evolution of Lake Egirdir, West Turkey. Geologos 16(4):223–234 Lee S (2003) Evaluation of waste disposal site using the DRASTIC system in southern Korea. Environ Geol 44:654–664 Martinez-Bastida JJ, Arauzo M, Valladolid M (2010) Intrinsic and specific vulnerability of groundwater in central Spain: the risk of nitrate pollution. Hydrogeol J 18:681–698 Palmer AN (2010) Understanding the hydrology of karst. Geol Croat 63:143–148 Parise M, Ravbar N, Živanovic V, Mikszewski A, Kresic N, MadlSzonyi J, Kukuric N (2015) Hazards in karst and managing water resources quality. In: Stevanović Z. (ed) Karst aquifers characterization and engineering. Professional practice in earth sciences. Springer, Cham, pp 601–687. https://doi.org/10.1007/978-3-31912850-4_17 Senel M, Dalkilic H, Gedik I, Sedaroglu M, Bilgin AZ, Oguz MF, Bolukbasi S, Korucu M, Ozgul N (1996) The stratigraphy of 13 autochthonous and allochthonous units in the east of Isparta Angle (West Taurus). Bull Miner Res Explor (in Turkish) 118:111–160 Sener S, (2010), Hydrogeochemical investigation of Egirdir Lake water and bottom sediments, PhD thesis. Suleyman Demirel University, Isparta, Turkey (in Turkish) Sener E, Davraz A (2013) Assessment of groundwater vulnerability based on a modified DRASTIC model, GIS and an analytic hierarchy process (AHP) method: the case of Egirdir Lake basin (Isparta, Turkey). Hydrogeol J 21:701–714 Sener S, Davraz D, Karaguzel R (2013) Evaluating the anthropogenic and geologic impacts on water quality of the Egirdir Lake, Turkey. Environ Earth Sci 70:2527–2544 Seyman F (2005) Hydrogeological study of the Senirkent-Uluborlu Basin, Suleyman Demirel University, Institute of Science and Technology, Master Thesis, Isparta (in Turkish) Simsek C, Kincal C, Gunduz O (2006) A solid waste disposal site selection procedure based on groundwater vulnerability mapping. Environ Geol 49:620–633 SKKY (2008) Ministry of forestry and water affairs, water pollution control regulation. Official Gazette, No. 26786, Ankara Soyaslan I (2004) Hydrogeological investigations of eastern Egirdir Lake and groundwater modeling. PhD thesis, Suleyman Demirel University, Isparta, Turkey Topcam A, Atigan A, Alkan N, Ozpinar B, Agacik O (1977) Hydrogeological studies of Hoyran-Gelendost and Yalvac plains. Geotechnical Services of General Directorate of State Hydraulic Works, State Hydraulic Works Printing House, Ankara (in Turkish) TUBITAK (Marmara Research Center, MRC) (2012) Basin Protection Plan Special Provisions of the Egirdir Lake, unpublished report (in Turkish) Umar R, Ahmed I, Alam F (2009) Mapping groundwater vulnerable zones using modified DRASTIC approach of an alluvial aquifer in parts of Central Ganga Plain, Western Uttar Pradesh. J Geol Soc India 73(2):193–201 US EPA (1993) Groundwater Resource Assessment. EPA 813-R93–003, USEPA Office of Groundwater and Drinking Water, Washington, DC US EPA (1997) State methods for delineating source water protection areas for surface water supplied sources of drinking water. EPA 816-R-97–008, USEPA Office of Water. https ://nepis .epa.gov. Accessed 31 Jan 2018 Van Stempvoort D, Ewert L, Wassenaar L (1992) Aquifer vulnerability index: a GIS-compatible method for groundwater vulnerability mapping. Can Water Resour J 18:25–37 Varol S, Davraz A (2010) Hydrogeological investigation of Sarkikaraagac Basin (Isparta, Turkey) and groundwater vulnerability. Water Int 35(2):177–194 Vrba J, Zaporozec A (1994) Guidebook on mapping groundwater vulnerability, IAH, vol 16. Heise, Hannover Wang Y, Merkel BJ, Li Y, Ye H, Fu S, Ihm D (2007) Vulnerability of groundwater in quaternary aquifers to organic contaminants: a case study in Wuhan City, China. Environ Geol 53:479–484 White WB (2002) Karst hydrology: recent developments and open questions. Eng Geol 65:85–105