© Author(s) 2023. CC Atribution 4.0 License Hydrogeological characterization of karst springs of the white (Proteus anguinus anguinus) and black olm (Proteus anguinus parkelj) habitat in Bela krajina (SE Slovenia) Hidrogeološka karakterizacija kraških izvirov na območju habitata belega (Proteus anguinus anguinus) in črnega močerila ( Proteus anguinus parkelj) v Beli krajini (JV Slovenija) Katja KOREN 1 & Rok BRAJKOVIČ 1 Manca BAJUK 2 , Špela VRANIČAR 2 & Vesna FABJAN 2 1 Geological Survey of Slovenia, Dimičeva ulica 14, SI-1000, Ljubljana, Slovenia; e-mail: katja.koren@geo-zs.si, rok.brajkovic@geo-zs.si 2 High school Črnomelj, Gimnazija Črnomelj, Kidričeva ulica 18a, SI-8340, Črnomelj, Slovenia Prejeto / Received 19. 5. 2022; Sprejeto / Accepted 10. 7. 2023; Objavljeno na spletu / Published online 4. 8. 2023 Key words: hydrogeology, olm, ecology, nitrate, monitoring Ključne besede: hidrogeologija, močeril, ekologija, nitrat, monitoring Abstract The springs west of Črnomelj , in SE Slovenia, are the habitat of the black ( Proteus anguinus parkelj) and the white olm ( Proteus anguinus anguinus). Some of these springs are also the only known habitat in the world of endemic species of black olm. A steady decline in olm populations has been observed in this area over the past decades. Owing to the rapid runoff and groundwater flow high-resolution monitoring is essential in providing better insight into the hydrogeological characterization of the catchment area of springs. Specific factors and critical parameters of water behind said olm degradation have not yet been defined. Because the olm’s environment is largely aquatic, one potential critical parameter could be the higher water temperatures (>12 °C) or higher nitrate concentration (>9.2 mg/l). The six-month observation of the springs (July – December 2021) point to water temperature as a potential critical parameter since the water temperature of the springs exceeded 12 °C in months July and August. Nitrate concentrations could also be a second critical parameter in the degradation of the olm’s habitat. Maximum nitrate concentrations above 9.2 mg/l throughout much of the observation period (except for Dobličica spring). Due to less agricultural activity in December in the spring catchment area and a higher dilution rate due to reduced evapotranspiration and increased effective precipitation during this time of the year, the nitrate concentrations are decreased. The results of the measured parameters of groundwater could show the hydrogeological connection between the Otovski and Pački breg springs and between Šotor , Jamnice and Dobličica. The Obršec spring has an independent catchment area. A detailed estimation of the springs catchment area is possible due to a detailed geologic map. It is necessary to determine the origin of the nitrate (nitrate isotope analysis), to quantify the threshold values of the critical parameters, to define precisely all the causes of the olm deterioration, and to make proposals for appropriate measures to limit or even stop the decline of the olm population. Izvleček Izviri zahodno od Črnomlja , v JV Sloveniji so habitat črn ega ( Proteus anguinus parkelj) in belega močerila ( Proteus anguinus anguinus). Nekateri od teh izvirov so tudi edini znan habitat te endemične vrste črnega močerila. V zadnjih desetletjih je opazen upad populacije močerilov. Za boljši vpogled in ocenitev hidrogeoloških značilnosti prispevnega območja izvirov, je zaradi hitrega odtoka in toka podzemne vode pomembno pogosto spremljanje stanja. Potencialni vplivni dejavniki in parametri podzemne vode, ki bi lahko vplivali na slabšanje stanja ohranjenosti močerila še niso opredeljeni. Ker močeril večino časa živi v vodi, bi lahko potencialni kritični dejavnik bila višja temperatura vode (>12 °C) ali višja vsebnost nitrata v vodi (>9.2 mg/l). Izsledki šestmesečnega spremljanja kažejo, da bi potencialni kritični parameter za slabšanje stanja ohranjenosti močerila bila temperature vode nad 12 °C v mesecih julij in avgust v opazovanem obdobju. Vsebnost nitrata bi prav tako lahko bil kritični parameter oz. razlog za upad števila močerilov in slabšanje stanja tega habitata. Najvišje vsebnosti nitrata so mejno vrednost za močerila presegale skoraj čez celotno opazovalno obdobje (z izjemo izvira Dobličice), razen v mesecu decembru. Vzrok za to je zelo verjetno zmanjšana kmetijska dejavnost oz. višja stopnja razredčenja v tem delu leta zaradi zmanjšane evapotranspiracije in višjih količin efektivnih padavin. Rezultati izmerjenih parametrov podzemne vode kažejo, na verjetno hidrogeološko povezavo med izviri Otovski in Pački breg ter med izviri Šotor, Jamnice in Dobličica. Izvir Obršec ima samostojno prispevno območje. V prihodnje bo podrobnejša opredelitev prispevnega območja izvirov mogoča z detajlnim geološkim kartiranjem. Potrebno je ugotoviti izvor nitrata (izotopske analize nitrata), kvantificirati mejne vrednosti kritičnih parametrov, določiti vse mo žne vzroke za slabšanje stanja ohranjenosti populacije močerila in opredeliti predloge ukrepov za preprečevanje oz. ustavitev upada populacije močerilov. GEOLOGIJA 66/1, 151-166, Ljubljana 2023 https://doi.org/10.5474/geologija.2023.006 152 Katja KOREN, Rok BRAJKOVIČ, Manca BAJUK, Špela VRANIČAR & Vesna FABJAN Introduction Some springs and caves in the Bela krajina re - gion (SE Slovenia), in the area west of Črnomelj , are especially important and should be kept in good hydrogeological and geochemical condition (or work towards improvement), since they are the habitat of the black ( Proteus anguinus parkelj ) and white olm ( Proteus anguinus anguinus). The black olm is an endangered endemic subspecies known only from a few springs over less than 3 km 2 in the W part of Bela krajina. Based on Annex 6 (Red List of Amphibians) of the Habitats Directive (Council Directive 92/43/EEC), the white and black olm are classified as rare and vulnerable species. The clas - sification of the white and black olm as rare and vulnerable species was made based on a long-term scientific research of the distribution and decline in the olm population in the Bela krajina area. The problem of deterioration of the olm’s habitat has also been noted by locals, among them the stu - dents of the Črnomelj secondary school, who work to raise awareness among the wider local commu - nity and draw attention to the problem. Cave pol - lution and the consequent polluted groundwater affects these groundwater-dependant ecosystems (Mezga et al., 2016). In the long term, this could cause the decline of one of the most important symbols of subterranean biodiversity, the white olm, as well as the black olm in Bela krajina ( Sket, 1997; Aljančič et al., 2014; Ribeiro & Tičar, 2017). Olm lives in aquatic environments, in still and oxygen-rich waters with stable temperatures of 8 to 11 °C. Occasionally , enters the phreatic and epi- phreatic zones at high water levels (Aljančič et al., 2014; Mezga et al., 2015). Based on the conditions under which olm lives, water temperature above 12 °C could also be a potentially critical parameter for its degradation. Potential factors and related critical parameters affecting the preservation status of the olm’s hab - itat have not yet been properly defined. Past re - search (NLZOH, 2017) has determined the nitrate threshold value for olms, which consists of the pre - dicted no-effect concentration (PNEC), the nat - ural background concentration, and the expected variation of the natural background concentration. The main toxic effect of nitrate on aquatic animals appears to be the conversion of oxygen-carrying pigments (hemoglobin, hemocyanin) into forms incapable of transporting oxygen (methemoglobin, methemocyanin) (Jensen, 1996; Scott & Crunkil - ton, 2000). If the assessed critical parameters are found to have a significant inf luence, the next step is to find the causes behind certain excessive critical pa - rameter in groundwater and to limit or lower them using appropriate measures. The aims of this study are (I) to assess the ba - sic hydrogeological characteristics (water level, water temperature and electrical conductivity) of the observed springs as a response to water lev- els and water temperatures to precipitation (II) to determine whether water temperatures and ni - trate concentrations are in a range suitable for the olm, (III) to determine whether long-term national monitoring would provide a realistic assessment of the quality of spring water, and (IV) to determine whether there is a possible geological or hydrogeo- logical connection between the studied springs. The study area Geographical settings The study area lies in SE Slovenia, in Bela kra - jina, west of the town of Črnomelj ( Fig. 1 ), with a focus on six springs that are the habitat of white and black olm. The black olms were detected in the springs of Obršec, Šotor, Jamnice (also known as Jelševnik spring) and Dobličica, while only the white olm is known from Otovski breg and Pački breg springs (Gorički, 2017). In the catchment area of these six springs are village settlements, which have regulated water supplies but no sew- age system. On the slopes west of the springs (Do - blička gora, Stražni vrh, Rodine) there are homes with vineyards and permanently inhabited houses spread over a wider area of the catchment area of the studied springs. The potential sources of an - thropogenic impacts in this part of the karst area mainly consist of illegal landfills, the use of septic tanks in households, and the use of manure. Fur - thermore, in the immediate vicinity of the Obršec spring an illegal settlement with uncontrolled sewage disposal. Geological settings Bela krajina can be geotectonically divided into its NE part, which belongs to the transition area between the Internal and External Dinarides, and the remaining part, which belongs to the External Dinarides (Placer, 2008), where our study area is located. The lithostratigraphic succession of the study area is largely characterized by shallow ma - rine limestones and dolomites of Jurassic and Cre - taceous age (Fig. 2) (Bukovac et al., 1984a, 1984b; Vlahović et al., 2005). The studied area is char - acterised by outcrops of Jurassic and Cretaceous carbonates. The Upper Jurassic bedded limestones and bedded to massive dolomites are tectonically fractured and exhibit strong secondary porosity. 153 Hydrogeological characterization of karst springs of the white ( Proteus anguinus anguinus) and black olm ( Proteus anguinus parkelj)... Fig. 1. Geographical location with observation springs, inhabited area (GURS, 2016) and land use (GERK, 2023) in the study area. 154 Katja KOREN, Rok BRAJKOVIČ, Manca BAJUK, Špela VRANIČAR & Vesna FABJAN Fig. 2. The geological settings of the study area (modified after Bukovac et al., 1984a, 1984b). 155 Hydrogeological characterization of karst springs of the white ( Proteus anguinus anguinus) and black olm ( Proteus anguinus parkelj)... Lower Cretaceous limestones (Bukovac et al., 1984a, 1984b) in some parts also contain lens - es of dolomite and a breccia horizon. The entire Cretaceous succession exhibits strong and deep karstification, which is reflected in a large num - ber of karst dolines, vertical shafts, and caves. To the east, the study area is bounded by the tectonic contact with the Kanižarica coal basement, which formed in the Pliocene and was filled with fine- grained lake sediments and organic matter (coal) (Šinigoj et al., 2012). The Jurassic, Cretaceous and Neogene rocks and their contacts in the lower parts of the shallow karst are covered by clays of the Plio-Quaternary age, a thick (2–6 m) cover of residual and resedimented terra rossa and Quater - nary sediments of the Dobličica and Jelševniščica floodplains. Structurally, the wider area of Bela krajina is characterized by NW–SE trending Di - naric longitudinal structures (folds and faults). The fault planes are mostly characterized as re - verse faults with their dip towards the SW (Bu - kovac et al., 1984a, 1984b). The springs of the study area lie on the potential continuation of the reverse fault, defined on the basic geological map as the “Bosiljevo-Črnomelj” thrust (Bukovac et al., 1984a, 1984b; Habič et al., 1991b; Novak, 1996; Šinigoj et al., 2012). From the Dobličica spring towards the Šotor , Obršec, and Jamnice springs, this zone is covered by the Quaternary flood plain of the Dobličica River. N–S orientated fault zone outcrops only in some locations E of the village of Dobliče and W of village Otovec. Different struc - tural trends can be observed NW of the village Dobliče (Šinigoj et al., 2012). There the fault sys - tem shows NW–SE orientation. The structural relations between these two fault systems are not clear, as the possible fault intersection is covered by Quaternary sediments. These fault zones could be an important factor for groundwater conduc - tion (Čar, 2018). The springs studied are classified as karst springs. Dobličica and Otovski breg flow from Lower Cretaceous limestone and could not be directly connected to any of the known fault systems, while Obršec, Šotor , Jamnice and Pački breg springs flow from the Upper Jurassic lime - stone and dolomite and are probably located in a tectonic zone running in the NNE–SSW direction (Bukovac et al., 1984b; Habič et al., 1991b; Šinigoj et al., 2012). Observation springs Otovski breg is a spring in an unroofed cave from which the water flows to the surface through two syphons. Close by, another monitored spring is located (approximately 1 km SW from Otovski breg) called Pački breg. The white olm is present in these two (Pački breg & Otovski breg) observed springs. Both springs are located in the northern part of the studied area near the villages of Otovec a n d T u š e v d o l ( F i g . 1 ) . I n P a č k i b r e g t h e r e a r e three smaller springs that are a mere one meter apart and never run dry. The water flows to the surface in two horizontal syphon springs. In the third, the water springs vertically, which is obvi- ous when observed at high water level. The habitat of the black olm consists in the four observed springs located in the southern part of the studied area (Fig. 1) near the villages of Jelševnik and Dobliče. Šotor is a spring located on the Zupančič farm in Jelševnik, only 50 m away from the Jamnice spring. It is about 4.5 m wide, and the water comes to the surface in several sy - phons. The Šotor spring is covered with a tent to simulate a dark environment and a camera is in - stalled to observe the olms. The Jamnice spring is a funnel-shaped spring some about 2 m wide from where water outflows to the lake at the Zupančič farm. The Obršec spring is located 500 m south of the village of Jelševnik with two larger syphons 5 m apart. The southernmost and observably larg - est is the Dobličica spring. The spring Dobličica is located 2.4 km SE from the village Jelševnik (springs Šotor and Jamnice) and is a spring with a depth of more than 100 m (Novak, 1996). The spring is protected by a groundwater protection zone and is part of the public drinking water sup- ply system. Material and Methods Effective precipitation (P ef ), evapotranspiration (ETR) Data on hourly measurements of precipitation and daily potential evapotranspiration were ob- tained from the meteorological station Črnomelj- Dobliče ( S E A 2 0 2 2 a , 2 0 2 2 b ) . T h e e f f e c t i v e p r e - cipitation (P ef ) is the amount of total precipitation without runoff and evapotranspiration. Based on hourly measurements of precipitation and dai - ly evapotranspiration we calculated the highest amount of precipitation (P tot ) in one day, in a one- hour event, and the monthly volume. We also cal - culated the daily evapotranspiration (ETR) and effective precipitation. Based on the daily total precipitation (P tot ) and the daily potential evapo - transpiration (ETR) in the meteorological station Črnomelj – D o b l i č e , w e s i m p l i f i e d a n d a s s e s s e d the amount of daily effective precipitation (Eq. 1). [Eq. 1] 𝑃𝑃𝑃𝑃 𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒 = 𝑃𝑃𝑃𝑃 𝑡𝑡𝑡𝑡 𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡 − 𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸 𝐸𝐸𝐸𝐸 [ 𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚 ] 𝑊𝑊𝑊𝑊𝑊𝑊𝑊𝑊 𝐸𝐸𝐸𝐸 = 𝑊𝑊𝑊𝑊𝑊𝑊𝑊𝑊 𝑚𝑚𝑚𝑚 𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚 − 𝑊𝑊𝑊𝑊𝑊𝑊𝑊𝑊 𝑠𝑠𝑠𝑠 𝑡𝑡𝑡𝑡 𝑚𝑚𝑚𝑚 𝑠𝑠𝑠𝑠 𝑡𝑡𝑡𝑡 𝑡𝑡𝑡𝑡 𝑚𝑚𝑚𝑚 𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚 − 𝑡𝑡𝑡𝑡 𝑠𝑠𝑠𝑠 𝑡𝑡𝑡𝑡 𝑚𝑚𝑚𝑚 𝑠𝑠𝑠𝑠 𝑡𝑡𝑡𝑡 [ 𝑚𝑚𝑚𝑚 / ℎ] 156 Katja KOREN, Rok BRAJKOVIČ, Manca BAJUK, Špela VRANIČAR & Vesna FABJAN Field measurements Field measurements of water levels, tempera - ture, and electrical conductivity were carried out using the water level measurement logger Eltratec GSR 130NTG with sensors for water level, tem - perature, and electrical conductivity ( Table 1 ). The loggers were installed at the Obršec and Šotor springs from July to December 2021 and at the Pački breg spring from August to December 2021. The loggers recorded measurements at one-hour intervals. The water level, electrical conductivity and water temperature in Jamnice and Otovski breg were not monitored, we only measured ni - trate content. Data on water levels and temperature is record - ed every five minutes at the Dobličica spring and was provided by Komunala Črnomelj , a public util- ity company. Due to the wide measurement range and high measurement uncertainty of the elec - trical conductivity probe installed at Pački breg (± 400 μS/cm), the data obtained during the study for this spring was omitted. In this case, we mon - itored only the relative fluctuations in electrical conductivity. Response of water level (WLR) on rainfall event We defined the rain event as a maximum daily amount of precipitation (P tot ) of more than 25 mm. This rainfall amount (25 mm/day) was determined based on a significant simultaneous rise in water level in the observed springs, which occurred as a peak just a few hours after the rain event began. All rain events began with a rainfall rate greater than 0.2 mm/hour. The time of the beginning of the rainfall event (t start ) is the so-called beginning of the rainfall event. During this time, the water level does not change (WL start ). After some time during the rain event (t max ), the peak or maximum spring water level (WL max ) is measured. Based on the rainfall events and the water level rise in re- sponse to the rainfall event, we calculated the re- sponse rates or water level rise rate (WLR), which is a very simplified tool to roughly evaluate the re - sponse of the karst spring to rainfall events (Eq. 2). [Eq. 2] Water sampling Sampling for nitrate concentrations was car - ried out at weekly intervals between July and De- cember 2021 at the Obršec, Šotor , Jamnice, Otovs - ki breg, Pački breg and Dobličica springs ( Fig. 1 ). Sampling was performed in collaboration with students from Črnomelj H i g h s c h o o l ( G y m n a s i - um Črnomelj ), that were included into research as citizen science members. For sampling, we used 100 ml plastic bottles or two 50 ml plastic tubes for each sampling location and stored at 2–5 °C. Before collecting the water samples, the bot - tles were washed with water from the individual spring. Then the samples were taken to the Geo - logical Survey of Slovenia laboratory to a dark and cool place. Nitrate measurements Measurements of nitrate concentrations in wa - ter were carried out in the hydrogeological labo- ratory of the Geological Survey of Slovenia using a UV-VIS Spectro:IyserTM spectrometer. Based on the reflection of laser beams, the spectrum and nitrogen content of nitrate (NO 3 -N) are de - termined. Measurements were performed no later than 72 hours after sampling. The spectrometer is calibrated to the primary standards using known values. For a quality control check prior to sample measurement, the in-house standard (ultrapure water) was measured first, and the second in- house standard (tap water) was measured at the end of the measurement process. The same sam - ple from one bottle was measured three times. All measured values are corrected with the correction equation obtained using primary standards with known values and checked with in-house stan- dards. Results Effective precipitation (P ef ) and evapotranspiration (ETR) and rain events In the observation period July – December 2021, the highest monthly amount of precipita - tion (P tot ) came in July (141.99 mm) and the low - est in September (79.17 mm). The highest evapo - transpiration (ETR) coincided with higher air Table 1. Characteristics of sensors for water level, temperature and electrical conductivity measurements (Eltratec GSR 130NTG). Observation point Monitoring period Interval Water level Temperature Electrical conductivity Obršec July – December 2021 1 h 0 – 9.99 m 0 – 50 °C ± 0.3 °C 10 – 2000 µS/cm ±50 µS/cm (< 2 %) Šotor July – December 2021 1 h 0 – 9.99 m 0 – 50 °C ± 0.3 °C 10 – 2000 µS/cm ±50 µS/cm (< 2 %) Pački breg August – December 2021 1 h 0 – 99.99 m 0 – 50 °C ± 0.4 °C 0.1 – 10 mS/cm ± 400 µS/cm 𝑃𝑃𝑃𝑃 𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒 = 𝑃𝑃𝑃𝑃 𝑡𝑡𝑡𝑡 𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡 − 𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸 𝐸𝐸𝐸𝐸 [ 𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚 ] 𝑊𝑊𝑊𝑊𝑊𝑊𝑊𝑊 𝐸𝐸𝐸𝐸 = 𝑊𝑊𝑊𝑊𝑊𝑊𝑊𝑊 𝑚𝑚𝑚𝑚 𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚 − 𝑊𝑊𝑊𝑊𝑊𝑊𝑊𝑊 𝑠𝑠𝑠𝑠 𝑡𝑡𝑡𝑡 𝑚𝑚𝑚𝑚 𝑠𝑠𝑠𝑠 𝑡𝑡𝑡𝑡 𝑡𝑡𝑡𝑡 𝑚𝑚𝑚𝑚 𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚 − 𝑡𝑡𝑡𝑡 𝑠𝑠𝑠𝑠 𝑡𝑡𝑡𝑡 𝑚𝑚𝑚𝑚 𝑠𝑠𝑠𝑠 𝑡𝑡𝑡𝑡 [ 𝑚𝑚𝑚𝑚 / ℎ] 157 Hydrogeological characterization of karst springs of the white ( Proteus anguinus anguinus) and black olm ( Proteus anguinus parkelj)... t e m p e r a t u r e s a n d h i g h e r p l a n t t r a n s p i r a t i o n , i n J u l y (137.2 mm), with the lowest volumes in December (9.4 mm). The lowest monthly amount of effective precipitation (P ef ) was in August (2.25 mm), and the highest in November (117.37 mm) ( Table 2 ). In the period July – December 2021 we defined five rain events: at July 16 (72.3 mm/day), August 17 (29 mm/day), September 17 (42.1 mm/day), October 6 (27.1 mm/day) and December 2 (33.6 mm/day). Water level, temperature, and electrical conductivity in the observed springs In the period July – December 2021 we ob - served the hourly change in water level (WL), tem - perature (T), and electrical conductivity (EC) in three springs – Pački breg, Obršec, and Šotor . The measured values are presented in Figure 4 , 5, 6, and 7. The highest and lowest water level values, temperatures, and electrical conductivity of these four springs are presented in Table 3 . Table 2. Precipitation, evapotranspiration and effective precipitation in Črnomelj – Dobliče meteorological station (July – December 2021) (SEA 2022a, 2022b). Precipitation (P tot ) N=9169 Evapotranspiration) (ETR) N=185 Effective precipitation (P ef = P sum -ETR sum ) Month/Year Max [mm/h] Sum [mm/ month] Max [mm/ day] Sum [mm/month] Sum [mm/month] 7/21 50.9 141.99 6.8 137.2 4.79 8/21 23.5 108.25 5.3 106.0 2.25 9/21 34.4 79.17 3.5 73.7 5.47 10/21 27.0 105.83 2.6 31.2 74.63 11/21 23.6 131.77 1.4 14.4 117.37 12/21 33.9 97.57 1.2 9.4 88.17 Fig. 3. Total daily precipitation and effective precipitation in Črnomelj-Dobliče meteorological station (July – December 2021) (SEA 2022a, 2022b). Pački breg Obršec Šotor Dobličica WL [m] EC [µS/cm] T [°C] WL [m] EC [µS/cm] T [°C] WL [m] EC [µS/cm] T [°C] WL [m] T [°C] N=3430 N=3430 N=3409 N=3930 N=3709 N=3958 N=3926 N=3954 N=3954 N=41318 N=51780 Month/ Year Max Min Max Min Max Min Max Min Max Min Max Min Max Min Max Min Max Min Max Min Max Min 07/21 / / / / / / 1.91 0.76 452 424 13.9 11.5 1.47 0.48 535 357 17.3 10.5 1.43 0.37 25.6 10.7 08/21 0.53 0.33 930* 620* 11.8 11.5 1.44 0.56 451 420 14.7 11.5 0.71 0.40 442 403 11.3 10.5 0.42 0.37 14.1 11 09/21 0.59 0.25 970* 920* 12.0 11.7 1.63 0.29 474 431 14.1 11.4 1.03 0.41 435 418 10.7 10.5 0.83 0.43 13 10.9 10/21 0.76 0.27 990* 890* 11.9 11.4 1.63 0.3 507 430 12.8 11.2 1.9 0.54 549 418 11 10.4 1.82 0.39 12.4 9.7 11/21 0.85 0.41 970* 900* 11.7 10.9 1.82 1.19 489 452 11.6 11.3 2.09 0.81 625 435 11.8 10.2 1.96 0.49 10.7 9.0 12/21 0.86 0.42 960* 900* 11.5 10.9 1.8 1.36 461 450 11.5 11.4 2.08 0.91 695 442 11.9 10.1 1.91 0.49 10.7 8.3 *wide measuring range – deviations of ± 400 μS/cm Table 3. Highest and lowest water level, electrical conductivity water temperature in observed springs (July – December 2021). 158 Katja KOREN, Rok BRAJKOVIČ, Manca BAJUK, Špela VRANIČAR & Vesna FABJAN The highest water level in Pački breg spring (Fig. 4) was recorded in December (0.86 m) with the minimum evapotranspiration, and the lowest water level (0.25 m) in September, when the lowest total precipitation and highest water temperature were recorded. The highest water level in Obršec (1.91 m) ( Fig. 5) was measured in July, when evapotranspiration and effective precipitation also reached their maximum. The lowest water level in Obršec (0.29 m) was recorded in September, Fig. 4. Water level, temperature and electrical conductivity in Pački breg in comparison with effective precipitation. Fig. 5. Water level, temperature and electrical conductivity in Obršec in comparison with effective precipitation. 159 Hydrogeological characterization of karst springs of the white ( Proteus anguinus anguinus) and black olm ( Proteus anguinus parkelj)... like in Pački breg, when the lowest total precipi - tation was measured. The highest water level in Šotor (Fig. 6) was recorded in November (2.09 m) as well as in Dobličica spring (1.96 m) ( Fig. 7), when maximum effective precipitation was also recorded. The lowest water level in Šotor (0.4 m) and Dobličica (0.37 m) was observed in August, coinciding with the lowest effective precipitation. The maximum water temperature of the Šotor (Fig. 6) and Dobličica springs ( Fig. 7) was re - corded in July, and the minimum in December, along with the highest and lowest evapotrans- piration rates. The highest water temperature in Pački breg ( F i g . 4 ) was recorded in September (12.0 °C) in Šotor (Fig. 6) (17.3 °C) and Dobliči - ca (Fig. 7) (25.6 ° C - logger likely dry; other Fig. 7. Water level and temperature in Dobličica in comparison with effective precipitation. Fig. 6. Water level, temperature and electrical conductivity in Šotor in comparison with effective precipitation. 160 Katja KOREN, Rok BRAJKOVIČ, Manca BAJUK, Špela VRANIČAR & Vesna FABJAN maximum 12.4 °C) in July, and in Obršec in Au - gust (14.7 °C). The maximum water temperature exceeded the limit of 12 °C in Šotor spring in July (17. 3 °C), several time s in Obršec between July and October (14.7–12.8 °C) and in Dobličica also between July and October. Highest water tem - peratures were measured simultaneously with the highest evapotranspiration (Table 3). The highest electrical conductiv it y in Pački breg was recorded in October (990 µS/cm). The lowest electrical conductivity in Pački breg (620 µS/cm) was in July, during the period of highest evapo- transpiration. The highest electrical conductivi - ty in Obršec (507 µ S/cm) was, as in Pački breg, also recorded in October. In this spring the low - est electrical conductivity was detected in August (420 µS/cm), during the period of lowest effective precipitation. The highest electrical conductivity in Šotor was reco rded in Decem ber ( 6 95 µS/cm), when minimum water temperatures and evapo - transpiration were recorded. The lowest electrical conductivity in Šotor (357 µS/cm) was recorded in July, during the period of maximum evapotrans - piration. Response of water level to rainfall events Based on the rainfall events determined in the period July – December 2021 and water level rise in a spring after a short time at the beginning of the rain event we calculated the water level rise rate (WLR) ( Table 4 ). In average, among other springs in Obršec water level rise is the fastest (0.1 m/h) and in Pački breg the slowest (0.04). Concentration of nitrates in springs Nitrate concentrations measured weekly in collected water samples from the six monitored springs over a six-month period (July-December 2021) are shown in Figure 8. The basic statisti - cal analysis and the highest and lowest maximum concentrations in the springs are shown in Table 5. The highest concentration of nitrates in Pački breg (25.3 mg/l) and Otovski breg (29.2 mg/l) was recorded in September, when the lowest amount of total precipitation was recorded. The highest ni - trate concentration in Pački breg (20.8 mg/l) was in October. In Šotor the highest concentration of nitrates was recorded in July (15.1 mg/l), as in Jamnice (29.2 mg/l), during the period of maxi - mum evapotranspiration. Since spring 2010, Jamnice (also named Jelševnik), Otovski breg, Pački breg, and Dobličica are included in the national monitoring of the qual - itative status of groundwater. At the beginning, the national monitoring included sampling twice Table 4. Water level rise rate (WLR) in Pački breg, Obršec, Šotor and Dobličica spring (July – December 2021) as a response to a rainfall event. Pački breg Obršec Šotor Dobličica Month/ Year WL start WL max Δt [h] ΔWL [m] WLR [m/h] WL start WL max Δt [h] ΔWL [m] WLR [m/h] WL start WL max Δt [h] ΔWL [m] WLR [m/h] WL start WL max Δt [h] ΔWL [m] WLR [m/h] 07/21 / / / / / 16/7, 18:00 17/7, 11:00 17 0.99 0.06 16/7, 17:00 17/7, 10:00 17 0.96 0.06 16/7, 18:15 17/7, 14:15 23 1.04 0.05 08/21 17/8, 04:00 17/8, 10:00 6 0.16 0.03 17/8, 02:00 17/8, 10:00 8 0.88 0.11 17/8, 00:00 17/8, 08:00 8 0.26 0.03 / / / / 09/21 17/9, 15:00 17/9, 18:00 3 0.29 0.10 17/9, 14:00 17/9, 19:00 5 1.12 0.22 17/9, 07:00 17/9, 20:00 13 0.45 0.03 17/9, 15:00 17/9, 21:00 6 0.12 0.02 10/21 6/10, 20:00 8/10, 00:00 28 0.47 0.02 6/10, 19:00 7/10, 17:00 22 1.24 0.06 6/10, 18:00 7/10, 21:00 27 1.31 0.05 6/10, 21:00 8/10, 01:00 29 1.33 0.05 12/21 2/12, 14:00 3/12, 00:00 10 0.26 0.03 2/12, 12:00 2/122, 23:00 11 0.35 0.03 2/12, 14:00 2/12, 20:00 6 0.99 0.17 2/12, 14:00 3/12, 02:00 12 1.22 0.10 Average WLR 0.04 0.1 0.07 0.05 Δt = t max -t start [h] ΔWL= WL max -WL start [m] 161 Hydrogeological characterization of karst springs of the white ( Proteus anguinus anguinus) and black olm ( Proteus anguinus parkelj)... Fig. 8. Nitrate concentration in all observed springs with PNEC for olm. 162 Katja KOREN, Rok BRAJKOVIČ, Manca BAJUK, Špela VRANIČAR & Vesna FABJAN in a year. In the last few years sampling was per - formed just one time in a year. We compared, where possible, long-term national monitoring ni - trate concentrations (2010–2018) in these springs, with the results of our weekly observations and performed a basic statistical analysis of the data (Table 6 ). We calculated the median, mean, and maximum value of nitrate concentrations for all data obtained for the long-term (2010–2018) and the short-term (July – December 2021). The high - er values of nitrate concentration are represented in bold. Discussion Evapotranspiration, the sum of bare soil evap - oration, plant transpiration, and evaporation from precipitation intercepted by the canopy (Pollard & Thompson, 1995) and rainfall determine the spatial and temporal distribution of groundwater recharge (Jukić & Jukić, 2015). Land cover, like vegetation, changes the evapotranspiration and consequently has an influence on groundwater recharge (Kovačič et al., 2020). Due to vegetation cover, evapotranspiration in that area has a higher impact on groundwater recharge. Anthropogenic impact in this part of the karst area consists of some illegal landfill, the use of septic tanks in households and pouring manure on agricultural land. Decomposition under anaerobic conditions produces leachate saturated with or- ganic matter, which is characterized by a relatively high temperature, different from the temperature of the surrounding surface (Breg Valjavec & Zega, 2017). Additionally, in complex karst aquifers, significant temperature changes under a variety of hydrological conditions are a consequence of the inflow of water from different parts of the re - charge area (Petrič & Kogovšek, 2010). So water temperature can be considered as a natural tracer of groundwater flow (Saar, 2010). The electrical conductivity of water could also be used as a groundwater tracing tool. Electrical conductivity is determined by the dilution by pre - cipitation during rain events and can also be re- flected in higher concentrations of pollutants. The peaks in the electrical conductivity of monitored springs are likely controlled by the washing of pol- lutants from unsaturated zone during rain event (Kogovšek, 2011; Chang et al., 2021). Intensive transfer of contaminants occurred when the more permeable fissures were f lushed out, while some of the pollutants were retained in the less permeable part of the thick vadose zone (Kogovšek, 2011). The rough estimation of water level response (WLR) in Šotor a n d D o b l i č i c a s p r i n g s s h o w s u s the fast response in water level rise during a rain event in December, when the minimum evapo - transpiration was recorded, which was considered as an indicator of the high impact of evapotranspi - ration on the Šotor and Dobličica spring recharge Table 5. Highest and lowest monthly nitrate concentration in Pački breg, Otovški breg, Obršec, Šotor, Jamnice and Dobličica in the period of July – December 2021. Pački breg Otovški breg Obršec Šotor Jamnice Dobličica N=21 N=22 N=22 N=21 N=22 N=22 Month/Year Max Min Max Min Max Min Max Min Max Min Max Min 07/21 21.4 13.3 18.9 17.6 16.4 11.5 15.1 5.6 29.2 9.1 4.5 3.9 08/21 17 14.5 19.5 18.9 14.5 12.1 6.2 5.6 7.3 3.3 5.1 4.5 09/21 25.3 17 29.2 22.7 18.2 15.1 8.5 5.6 17.6 3.3 5.1 4.5 10/21 18.9 15.1 27.9 19.5 20.8 15.8 8.5 5.1 6.2 3.3 6.8 4.5 11/21 15.8 13.3 27.9 13.3 15.8 9.1 6.8 4.5 5.6 4.5 6.2 2.2 12/21 10.3 8.5 12.1 11.5 9.1 7.4 11.5 9.1 4.5 3.3 4.5 3.3 Table 6. Basic statistical analysis and comparison of nitrate concentration of short-term observations (July – December 2021) and nitrate concentrations of national monitoring (2010–2018). Higher nitrate concentrations are marked in bold). NO 3 - (mg/l) median mean max Jul–Dec 2021 (SEA 2022c) (2010–2018) Jul–Dec 2021 (SEA 2022c) (2010–2018) Jul–Dec 2021 (SEA 2022c) (2010–2018) Jamnice 4.76 3.20 (Jelševnik) 6.76 3.27 (Jelševnik) 29.22 4.25 (Jelševnik) Otovški breg 19.18 16.50 19.36 16.55 29.22 19.90 Pački breg 15.14 15.20 15.71 14.75 25.27 17.70 Dobličica 4.76 3.19 4.79 3.31 6.78 5.55 163 Hydrogeological characterization of karst springs of the white ( Proteus anguinus anguinus) and black olm ( Proteus anguinus parkelj)... dynamics. The highest (November) and lowest (July) effective monthly precipitation and water levels in Šotor and Dobličica spring were recorded in the same month. This basic assessment again demonstrates the important role of land cover in the Šotor and Dobličica recharg e area. The high - est (July) and lowest (September) amounts of to - tal precipitation and the highest and lowest water levels in the Obršec spring further indicate less impact from evapotranspiration and more direct infiltration, as well as the existence of a small in - dependent catchment area of the Obršec spring, as previously proposed by Habič (1991a) and Novak (1996). The nitrate ion (NO 3 – ), and consequently, ni - trate toxicity for aquatic animals, is due to nitrate ions (Camargo & Alonso, 2006). Based on these facts nitrate could be one of the potential critical parameters affecting the proposed threshold con - centration (PNEC), estimated at 9.2 mg/l NO 3 – (NLZOH, 2017). The nitrate ion occurs naturally in the nitrogen cycle and during nitrification but is also present in fertilizers in various forms. The most common anthropogenic sources of nitrate in groundwater are livestock and other agricultural production, wastewater, old landfills and illegal dumps, and fertilization with artificial fertilizers or digestate (NLZOH, 2017). The nitrate concen - tration above PNEC could be a major problem in Otovski breg, Pački breg, and Obršec, as con - centrations throughout most of the entire mon - itoring period (July – November) was exceeded, and occasionally also in Šotor and Jamnice (July 2021). There are no problems with high nitrate concentrations in Dobličica, as its catchment area is protected by the decree on water protection, wherein certain environmental interventions are not allowed (e.g. agriculture). The highest con - centration of nitrates was recorded in October in Dobličica, which did not exceed a concentration level of 7 mg/l (6.8 mg/l). Based on the geological and hydrogeological characteristics of Dobličica, low nitrate concentrations could be the result of the higher dilution rate seeing as it has the larg- est catchment area of all studied springs (Habič et al., 1991; Šinigoj et al., 2012). The main factor behind the high nitrate concentrations in the July – November period in Otovski breg and Pački breg could be agriculture, as in the case of Obršec, which includes an unregulated communal system (Habič et al., 1991a) in the catchment area. In most of the springs, the lowest maximum nitrate concentra - tions were recorded in December, which could be the result of little or no agricultural activity at that time of the year. A potential measure that could serve to ease nitrate concentrations would be to protect the springs with a decree limiting activi - ties that contribute to high nitrate concentrations in the catchment area of such springs. One of the proposed measures should be the regulation of wastewater drainage or the arrangement of a pub - lic sewage system. Also working with farmers on developing new fertilization techniques could con - tribute to a solution. We also compared the median, mean, and max - imum nitrate concentrations of long-term nation - al monitoring of the qualitative status of ground - water with low frequency of water sampling and short-term high-resolution sampling and mea - surements of nitrate concentrations. The compar - ison shows higher nitrate concentrations in the case of high-resolution sampling. In three of the compared springs – Jelševnik, Obršec and Otovs - ki breg – nitrate concentrations are higher in the case of high-resolution sampling, whereas the me - dian nitrate concentration in Pački breg is the ex - ception. Owing to their high solubility and mobil - ity, nitrates respond far more quickly and strongly to changes in hydrologic conditions and land use (Hem, 19985). So, in karst aquifers, low-resolu - tion monitoring of nitrates is unlikely to adequate - ly characterize the system, especially during rain - fall events (Pu et al., 2011). To assess the possible hydrogeological connec - tion between the studied springs, preliminary re - sults of detailed geological mapping at a scale of 1:5000 were used (Mušič et al., 2023). These data show some inconsistencies with previous geo - logical maps of the area (Bukovac et al., 1984a, 1984b; Šinigoj et al., 2012) in terms of stratig - raphy and structural relationships between fault zones. Therefore, only field verified (Mušič et al., 2023) fault zones were included in our interpre - tation. Thus, the majority of the connections cur - rently evaluated are based primarily on the hydro - geologic data collected. The basic hydrogeological characteristics of the Otovski breg are similar with that of Pački breg, as variations in nitrate concen - trations in these two springs are similar over the entire observation period. The connection between Otovski breg and Pački breg has already been con - firmed by previous tracer tests (Habič, 1991b). Al - though in Otovski breg and Jamnice only nitrate concentrations were monitored, a hydrogeological connection is also likely between the springs of Jamnice and Šotor . Nitrate concentrations fluctu - ate similarly in both springs and are low compared to the rest of the monitored springs. Comparisons of water levels, WLR, as well as nitrate concen - trations in Šotor and Dobličica springs also show 164 Katja KOREN, Rok BRAJKOVIČ, Manca BAJUK, Špela VRANIČAR & Vesna FABJAN similar spring dynamics. The Obršec spring has its own smaller catchment area and reacts quickly to precipitation, which drains in the NNE–SSW oriented fault zone (Bukovac et al., 1984a, 1984b; Šinigoj et al., 2012; Mušič et al., 2023). Conclusion The results of this study support the estab - lished knowledge of the dynamics that character - ize the karst springs in Bela Krajina, habitat of the black and white olm, and help to reveal the main problems that affect its conservation status. In or - der to try and solve the problem of the decline of the olm, it was first necessary to assess the basic hydrogeological characteristics of the six observed springs west of Črnomelj in Bela krajina – the hab - itat of the black and white olm – and to determine whether there were any possible geological or hy- drogeological connections between the observed springs. Due to their different hydrogeological charac- teristics, the springs react to weather phenomena differently, but some, like Pački and Otovski breg, have very similar dynamics, as do the Šotor and Dobličica springs. In the next step we evaluated the potential critical factors and water-related pa- rameters (nitrate concentration and temperature). The next step would require finding and specifi - cally defining the causes or critical water parame - ters using quantified threshold values and to take appropriate measures to slow or even halt entire - ly the decline of the olm. Nitrate concentrations throughout most of the entire monitoring period exceed the maximum threshold in Otovski breg, Pački breg and Obršec, and occasionally also in Šotor and Jamnice. During July – August (2021 ), the water temperature of the springs exceeded 12 °C in all four of the monitored springs. The comparison of high and low-resolution sampling indicates the importance of the high-res - olution monitoring in karst areas, where the runoff and groundwater flow are much faster compared to the flow in the intergranular aquifer. Further research is needed to constrain the hy- drogeological parameters over longer periods and to supplement our data using additional springs in the area. Said detailed hydrogeological data should also be further supplemented with a new detailed geological map of the area. It is necessary to de - fine the origin of nitrate (nitrate isotope analysis), quantify threshold values of the critical parame - ters, specifically define all the causes of olm dete - rioration, and make proposals for the appropriate measures to limit or even stop the olm population decrease. Authors contributions The authors' contribution is as follows: Katja Koren and Rok Brajkovič contributed to the con - ceptualization, analysis of data, writing, and re- view of the article. Authors Manca Bajuk, Špela Vraničar, and Vesna Fabjan, who participated in the present research as a Citizen Science Team, contributed through field observations, measure - ments, sampling, and participated in the discus - sion of the measurements. All authors read and agreed to the published version of the manuscript. Acknowledgments This study was financially supported by the Slove - nian Research Agency (research core funding Ground - waters and Geochemistry (P1-0020), Mineral Re - sources (P1-0025) and by the Slovenian Water Agency with the project Development of a support system for decision-making on the use of groundwater (2555-20- 470115) regarding groundwater-dependent ecosystems, including olms. For access to the sampling sites, we extend special thanks to the Zupančič farm and tour - ism Jelševnik and Komunala Črnomelj (Public Utility Company). Additionally, we would like to thank the Re - viewers for their valuable comments, which significant - ly contributed towards the improvement of the manu - script. References Aljančič, G., Gorički, Š., Năpăruş, M., Jeffery, W., R., Kuntner, M. & Lokovšek, T. 2014: A survey of distribution of Proteus Anguinus by envi- ronmental DNA sampling. Društvo za jamsko biologijo/ Society for Cave Biology. 10 p. Inter - net: https://www.cepf.net/sites/default/files/ edna_prezentacijacitluk.pdf (1. 2. 2022). Breg Valjavec, M. & Zega, M. 2017: Zasute vrtače – Potencialna nevarnost za vodne vire. In: Globevnik, L. & Širca, A. (eds.): Voda povezu - je, Drugi slovenski kongres o vodah, Podčetrek 19.–20. 4. 2017. Bukovac, J., Poljak, M., Šušnjar , M. & Čakalo, M. 1984a: Osnovna geološka karta SFRJ 1:100 000. Tolmač lista Črnomelj. Zvezni geološki zavod, Beograd: 63 p. Bukovac, J., Poljak, M., Šušnjar, M. & Čaka - lo, M. 1984b: Osnovna geološka karta SFRJ 1:100.000. List Črnomelj. Zvezni geološki za - vod, Beograd. Camargo, J.A. & Alonso, A. 2006: Ecological and toxicological effects of inorganic nitro - gen pollution in aquatic ecosystems: A glob - al assessment. Environment International , 165 Hydrogeological characterization of karst springs of the white ( Proteus anguinus anguinus) and black olm ( Proteus anguinus parkelj)... 32: 831–849. https://doi.org/10.1016/j.en - vint.2006.05.002 Cha ng , Y., Ha r t ma n n, A ., L iu, L ., Jia ng , G. H. & Wu, J. C. 2021: Identifying More Realistic Model Structures by Electrical Conductivity Obser - vations of the Karst Spring. Water Resources Research, 57. https://doi.org/10.31219/osf.io/ gw876 Council Directive 92/43/EEC of 21 May 1992 on the conservation of natural habitats and of wild fauna and flora (Habitats directive). Čar, J. 2018: Geostructural mapping of karstified limestones. Geologija, 61/2: 133–162. ht t p s:// doi.org/10.5474/geologija.2018.010 Gorički, Š., Stanković, D., Snoj, A., Kuntner, M., Jeffery, W.R., Trontelj, P., Pavićević, M., Gri - zelj, Z., Năpăruş-Aljančič, M. & Aljančič, G. 2017: Environmental DNA in subterranean biology: range extension and taxonomic impli - cations for Proteus. Sci. Rep. 7, 4505. ht t p s:// doi.org/10.1038/srep45054 GERK 2023: Javni pregledovalnik grafičnih podatkov MKGP. https://rkg.gov.si/GERK/ WebViewer/ (February 2023). GURS 2016: Inhabited area. Kataster stavb. https://eprostor.gov.si/imps/srv/slv/catalog. search#/metadata/5aa2b97b-2ea3-427f-a9fb- d7827c61c115 (February 2023). Habič, P., Kogovšek, J., Bricelj, M. & Zupan, M. 1991a: Izviri Dobličice in njihovo širše kraško zaledje. Acta carsologica, 17/19: 5–100. Habič, P., Kranjc, A., Kogovšek, J., Bricelj, M., Zupan, N., Drole, F., Drame, L. & Zadel, M. 1991b: Raziskovanje Dobličice in Krupe, 5. del, poročilo za leto 1990. Znanstvenoraziskovalni center SAZU, Postojna: 82 p. Hem, J.D. 1985: Study and interpretation of the chemical characteristics of natural water. US Geological Survey, Water Supply Paper, 2254: 1–263. Jensen, F.B. 1996: Uptake, elimination and ef - fects of nitrite and nitrate in freshwater cray- fish (Astacus astacus). Aquatic Toxicology , 34, 95–104. Juberthie, C., Durand, J. & Dupuy, M. 1996: Re - production of Proteus anguinus: results of 35 years of rearing in laboratory-caves of Moulis and aulignac (France). Mémoires de biospéolo - gie, 23: 53–56. Jukić, D. & Denić-Jukić, V. 2015: Investigating re - lationships between rainfall and karst-spring discharge by higher-order partial correlation functions. Journal of Hydrology, 530: 24–36. https://doi.org/10.1016/j.jhydrol.2015.09.045 Kogovšek, J. 2000: Ugotavljanje načina pretakanja in prenosa snovi s sledilnim poskusom v nara - vnih razmerah. Annales, 10/19: 133–142. Kogovšek, J. 2011: The role of the vadose zone in the transfer of pollution through karst aquifers to karst springs. In: Gabrovšek, F. (eds.): Kar - stology and development challenges on karst - Water, Karst Research Institute ZRC SAZU, Postojna: 21–37. Kovačič, G., Petrič, M. & Ravbar, N. 2020: Evalu - ation and Quantification of the Effects of Cli - mate and Vegetation Cover Change on Karst Water Sources: Case Studies of Two Springs in South-Western Slovenia. Water, 12/11: 3087. https://doi.org/10.3390/w12113087 Mezga, K., Janža, M., Prestor, J., Koren, K. & Šram, D. 2016: Groundwater dependent ecosystems – groundwater status indicators. Natura Sloveni - ae, 18: 35–42. Mezga, K., Janža, M., Šram, D. & Koren, K. 2015: Pregled ekosistemov odvisnih od stanja podzemnih vod. Priprava strokovnih podlag in strokovna podpora pri izvajanju vodne direk- tive za področje podzemnih voda (Direktiva 2000/60/EC); 2. UKREP DDU26: Analiza raz - položljivih zalog podzemne vode in površinske vode ter obstoječe in predvidene rabe vode za obdobje do 2021. Ljubljana: Geološki zavod Slovenije. Mušič, M., Dvorščak, L., Fariselli, S., Rome, T., Štern, M., Grum Verdinek, L., Brajkovič, R., Rožič, B. & Žvab Rožič, P. 2023: Student ge - ological mapping of the Black olm habitat and its catchment area in Bela krajina. In: Švara, A., Zupan Hajna, N. & Gabrovšek, F. (eds.): 30th International Karstological School "Clas - sical Karst", ZRC SAZU: 135-136. NLZOH, 2017: Ocena tveganja, ki ga predstav - lja nitrat za ekosisteme podzemne vode in za človeško ribico na projektnem območju LIFE Kočevsko. 25. Nacionalni laboratorij za zdrav - je, okolje in hrano, Center za okolje in zdravje Maribor: 20 p. Novak, D. 1996: Padavinsko zaledje izvira Jelševniščice. Naše jame, 38: 105–110. Petrič, M. & Kogovšek, T. 2010: Water tempera - ture as a natural tracer – a case study of the Malenščica karst spring (SW Slovenia). Ge - ologia Croatica, 63/2: 171–177. ht t p s://doi . org/10.4154/gc.2010.14 Petrič, M., Ravbar, N. & Kogovšek, T. 2011: Char - acteristics of karst aquifers, their vulnerability and endangerment. In: Gabrovšek, F. (eds.): Karstology and development challenges on 166 Katja KOREN, Rok BRAJKOVIČ, Manca BAJUK, Špela VRANIČAR & Vesna FABJAN karst - Water, Karst Research Institute ZRC SAZU, Postojna. 8–19. Placer, L. 2008: Principles of the tectonic subdi - vision of Slovenia. Geologija, 51/2: 205–217. https://doi.org/10.5474/geologija.2008.021 Pollard, D. & Thompson, S.L. 1995: Use of a land-surface-transfer scheme (LSX) in a global climate model: the response to doubling stoma - tal resistance. Global and Planetary Change, 10:129–161. https://doi.org/10.1016/0921- 8181(94)00023-7 Pu, J., Yuan, D., He, Q., Wang, Z., Hu, Z. & Gou, P. 2011: Highresolution monitoring of nitrate var - iations in a typical subterranean karst stream, Chongqing, China. Environtal Earth Science, 64: 1985–1993. https://doi.org/doi:10.1007/ s12665-011-1019-7 Ribeiro, D. & Tičar, J. 2017: The problematics of cave pollution in Bela krajina / »SOS Proteus« – Field Note. Natura Sloveniae, 19/1: 43–45. Saar, M.O. 2010: Review: Geothermal heat as a tracer of large-scale groundwater flow and as a means to determine permeability fields. Hy - drogeology Journal, 19: 31–52. ht t p s://doi . org/10.1007/s10040-010-0657-2 Scott, G. & Crunkilton, R.L. 2000: Acute and chronic toxicity of nitrate to fathead minnows (Pimephales promelas), ceriodaphnia dubia, and Daphnia magna. Environmental Toxicolo - gy and Chemistry, 19: 2918–2922. ht t p s://doi . org/10.1002/etc.5620191211 SEA 2022a: Črnomelj – Dobliče. Referenčna evapotranspiracija in padavine samodejnih postaj (dnevni podatki od leta 2017). Internet: https://meteo.arso.gov.si/met/sl/agromet/ data/arhiv_etp/ (10. 2. 2022). SEA 2022b: Meteorološki parametri Črnomelj – Dobliče. ARHIV – opazovani in merjeni me - teorološki podatki po Sloveniji. Internet: htt- ps:/ /meteo.arso.gov.si/met/sl/archive/ (10. 2. 2022). SEA 2022c: Kakovost podzemnih voda. Inter - net: https:/ /gis.arso.gov.si/portal/apps/ webappviewer/index.html?id=a16308bd - 37344559b1c5d5e515468f49 (10. 2. 2022). Sket, B. 1997: Distribution of Proteus (Amphib- ia: Urodela: Proteidae) and its possible ex - planation. Journal of Biogeography, 24/3: 263–280. https://doi.org/10.1046/j.1365- 2699.1997 .00103.x Šinigoj, J., Lapanje, A. & Poljak, M. 2012: Geologija območja zaledja Dobličice. Dolenjski kras, 6: 46–52. Vlahović, I., Tišljar, J., Velić, I. & Matičec, D. 2005: Evolution of the Adriatic Carbonate Platform: Palaeogeography, main events and deposi - tional dynamics. Palaeogeography, Palaeocli - matology, Palaeoecology, 220/3–4: 333–360. https://doi.org/10.1016/j.palaeo.2005.01.011 Voituron, Y., de Fraipont, M., Issartel, J., Guil - laume, O. & Clobert, J. 2011: Extreme lifespan of the human fish (Proteus anguinus): a chal - lenge for ageing mechanisms. Biology let - ters, 7/1: 105-107. https://doi.org/10.1098/ rsbl.2010.0539