Journal of Hydrology: Regional Studies 24 (2019) 100611

Contents lists available at ScienceDirect

Journal of Hydrology: Regional Studies

journal homepage: www.elsevier.com/locate/ejrh

Tracing groundwater recharge conditions based on environmental

T
isotopes and noble gases, Lom depression, Bulgaria
Marianna Túria,c, , Mihály Molnára, Tatiana Orehovab, Aglaida Totevab,
⁎

Vladimir Hristovb, Aleksey Benderevb, Anikó Horvátha, László Palcsua

a
Isotope Climatology and Environmental Research Centre (ICER), Institute for Nuclear Research, Hungarian Academy of Sciences, Bem tér 18/C, H-
4026, Debrecen, Hungary
b
Department of Hydrogeology, Geological Institute, Bulgarian Academy of Sciences, 1113, Sofia, Bulgaria
c
Department of Mineralogy and Geology, University of Debrecen, Egyetem tér 1, H-4010 Debrecen, Hungary

ARTICLE INFO ABSTRACT

Keywords: Study region: Lom depression, Bulgaria.

Environmental isotopes Study focus: A multi-tracer investigation was applied to identify the recharge conditions and
Noble gases isotope hydrological character of four aquifers in the Lom depression (Northwest Bulgaria) using
Groundwater environmental isotopes (δ18O, δ2H, δ13C, 3H, 14C) and noble gases. The radiocarbon age model of
Residence time
Ingerson and Pearson was used to estimate the mean residence time of groundwater samples from
Lom depression
four aquifers (Dacian, Pontian, Meotian, Sarmatian). Our study focuses on the study of recharge
Bulgaria
conditions and provides an additional information to hydrodynamic understanding of the four
aquifers.
New hydrological insights for the region: The mean residence time of groundwater samples re-
present the last twelve thousand years. In addition to a recently recharged groundwater sample,
some samples represent the early Holocene and samples closely correspond to the late Pleistocene
or the transition time between the early Holocene – late Pleistocene. The lower noble-gas re-
charge temperature values may indicate that the recharge occurred during the Late
Glacial–Holocene climatic transition. These data can also be used to verify the existence of pa-
leowater from the Pontian and Sarmatian aquifers on a trend of the mean residence time of older
water heading north away from the study area. Besides the paleoclimatological investigation, it
was determined that the Pontian aquifer has no hydraulic connection with the Danube River. The
sandy layers and lenses forming the Dacian aquifer are apparently hydraulically isolated and
therefore have high degree of heterogeneity.

1. Introduction

In the past several decades, there has been increasing concern for global water resources as influenced by climate change. This
concern has motivated studies that examine the effects of climate change on the water cycle and their specific impacts on aquifers at
regional and local scales (Aeschbach-Hertig and Gleeson, 2012). Groundwater is critical to water supply, irrigation and industry.
Further assessment of its sustainability in response to future changes requires the better knowledge about hydrological features and

⁎
Corresponding author at: Isotope Climatology and Environmental Research Centre (ICER), Institute for Nuclear Research, Hungarian Academy of
Sciences, Bem tér 18/C, H-4026, Debrecen, Hungary.
E-mail address: turi.marianna@atomki.mta.hu (M. Túri).

https://doi.org/10.1016/j.ejrh.2019.100611
Received 10 December 2018; Received in revised form 30 April 2019; Accepted 15 June 2019
Available online 24 June 2019
2214-5818/ © 2019 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY license
(http://creativecommons.org/licenses/BY/4.0/).
M. Túri, et al. Journal of Hydrology: Regional Studies 24 (2019) 100611

Fig. 1. Schematic cross section of the studied area is plotted from A to Á. The pentagon marks symbolize the locations of sampling points and their
colors sign the aquifers. Lines on the topographic map show the extensions of the aquifers and the assumed depth of the aquifers in the cross section.

processes as well as climate-driven changes in specific aquifers.

Environmental isotope hydrology is a powerful tool in discovering the origin of water and the characterization of aquifer con-
ditions. Using environmental tracers (isotopes, noble gases, etc.), groundwater recharge conditions such as mean residence time
(MRT) and climatic influences at the time of recharge can be determined (Fontes, 1980; Clark and Fritz, 1997; Sarkar et al., 2017;
Joshi et al., 2018). MRT values are usually estimated by the radiocarbon content of the water (Stute and Deak, 1989; Stute and
Schlosser, 1993; Aeschbach-Hertig et al., 1999; Klump et al., 2007; Varsányi et al., 2011). A direct quantitative method using the
temperature dependence of noble gas solubility in water allows the reconstruction of past mean annual temperatures.
In this study, a multi-tracer investigation was carried out for a multilayered aquifer system in the Lom depression in Bulgaria,
which combines hydrochemical data, environmental isotopes (δ18O, δ2H, δ13C, 3H, 14C) and noble gases in order to reconstruct the
recharge conditions and identify the isotopic character of this system. Although such isotopic hydrological studies exist for Bulgaria,
the data coverage is poor and general knowledge is still scarce (Rank et al., 1985; Pulido-Bosch et al., 1999; Magro et al., 2010;
Machkova et al., 2014, as well as unpublished data).
The aim of this first sampling campaign was to examine the aquifers of the Lom depression from a paleoclimatological point of
view, and to use this information for late Pleistocene and Holocene paleoclimate reconstruction. Our study focuses on the study of
recharge conditions and aiming to provide an additional information to hydrodynamic understanding of the four aquifers.

2
M. Túri, et al. Journal of Hydrology: Regional Studies 24 (2019) 100611

2. Study area

The study area is located in Northwest Bulgaria within the Danubian Plain and covers about 2500 km2 (Fig. 1). From a geological
point of view, this is the Lom depression (the most outstanding tectonic feature in Northwest Bulgaria), and from a hydrogeological
point of view, this is the Lom artesian basin (part of the North Bulgarian artesian basin).
The climate of the study area is a fully-humid warm temperate zone with hot summers according to the Köppen-Geiger climate
classification (Cfa). The recharge area belongs to the fully-humid warm temperate zone with warm summers (Cfb) (Peel et al., 2007).
According to observations, the mean annual rainfall varies from 546 to 677 mm. The average air temperature on the Danubian Plain
is 11.4 °C.
The study area is bounded on the North by the Danube River and its tributaries – the Lom River and the Tsibritsa River flow from
the south within the study area. The relief of the interfluvial areas is flat, with altitudes up to about 150–200 m above sea level. In
general, the relief is inclined slightly to the north and northeast towards the Danube River, with the lowest altitudes at about 30 m.
The main drainage artery of the region is the Danube River.

2.1. Geology

From a geological point of view, the Lom depression is located in the western part of the Moesian plate and is mainly filled by
Neogene sediments (Archar Formation, the Meotian – Lower Pontian sediments, Brusartsi Formation) and Quaternary deposits
(Kojumdgieva, Popov, 1988). The Archar Formation (N1p) mainly consists of well-sorted sand. The Meotian – Lower Pontian sediments
that occur in the central part of the Lom depression are mainly built of clay. The Brusartsi Formation (N2) mostly consists of clay with
sandy lenses. All Neogene layers dip toward the northeast. Quaternary sediments are widespread in the region and were deposited
under a variety of sedimentary environments, including: alluvial, fluvial, and aeolian (loess) origins.

2.2. Hydrogeological characterization

Hydrogeological conditions in the study area are mostly determined by geological factors. The main hydrogeological structure
here is the Lom artesian basin (Yovchev, Ryzhova, 1962), coinciding with the tectonic unit of the Lom depression. This basin is an
upper part of the North Bulgarian artesian region - the largest hydrogeological structure in Bulgaria, with well-expressed hydro-
dynamic, hydrochemical and hydrogeothermal zonality both in vertical and in horizontal directions (Antonov, Danchev, 1980).There
is a multilayered aquifer system in these formations, in which five hydrogeological units can be distinguished (Table 1, Fig. 1).
The uppermost Quaternary aquifer is associated with alluvial sediments (in the Danube lowlands and river valleys), Early
Pleistocene fluvial and fan deposits, and loess deposits. The most permeable Quaternary sediments are alluvial deposits (gravels and
sands, with intercalations of clays), with discordant layering relations in comparing to the older formations, both aquiferous and low
permeable.
The object of the study is groundwater formed in Neozoic strata. They sink from the western, southern and eastern periphery to
the central northern part of the study area. Based on lithostratigraphic features of the sediments, the most important hydrogeological
units (HGU) can be distinguished (Table 1).
The Dacian-Romanian complex (DRA, Dacian aquifer), which is fully contained within the Brusartsi Formation (Table 1), is located at
the upper part of Neozoic sediments. From a hydrogeological point of view, it is rather heterogeneous. The formation mainly consists
of clay interspersed with sandy lenses with thickness from 0.5 to 20 m. Most of the lenses are completely isolated, and some of them
are interconnected. The sandy lenses at the top of the Brusartsi Formation receive recharge from fluvial Early Pleistocene deposits
(boulders, gravel and coarse-grained sands). The local groundwater discharge occurs through springs on the slopes of the Danube
River and its tributaries. Some artesian wells have been established, providing a continuous discharge to the surface. Regionally, the
Dacian-Romanian complex is an aquiclude, with coal layers at the bottom. The hydraulic conductivity of individual sand lenses is in
the range of 0.64 to 2.1 m/d (Benderev et al., 2010; Stoyanov, 2019).
The Upper-Pontian aquifer (UPA, Pontian aquifer) covers an area of 1700 km2 and it is the most essential aquifer in the study area.
It is important for water supply to settlements (by a large number of wells) and therefore it is quite well investigated (Benderev et al.,
2010; Stoyanov, 2019). The aquifer is related to sediments of the Archar Formation (Table 1) with total thickness of about 80–100 m.
It is built of different grained sands, locally with intercalations of clayey sands.
The hydraulic conductivity of the aquifer is relatively high but varies in the fraction of sand, related to the paleogeographic
environment during sedimentation. In the western parts of the study area, near the town of Lom, the Upper-Pontian deposits are
rather coarse with transmissivity from a few hundred to over 2000 m2/d (Benderev et al., 2010). To the east, finer-grained sands are
deposited, leading to a reduction of transmissivity values (from 40 to155 m2/d). At the eastern edge, the transmissivity is about
20 m2/d. Locally, a highly permeable zone is identified near the village of Valchedram, with transmissivity around 750 m2/d.
Outcropping zones of the aquiferous Archar Formation, as well as the areas where it is covered only by the loess deposits, are
located on the peripheral parts of the Lom depression (Fig. 2). It is in these peripheral parts that the main recharge of the Upper-
Pontian aquifer occurs. Due to flat relief and widespread loess cover, the surface outflow is relatively slow and the rainwater in-
filtrates into the loess complex as groundwater recharge into the aquifers, which generates baseflow of the rivers or outflow through
springs.
The aquiferous sandy layers dip to the central parts of the depression, where they are covered by sediments of the Brusartsi
Formation. Most clayey deposits in the lower part of the last formation play the role of a regional aquiclude separating the two

3
 M. Túri, et al.

Table 1
Lithostratigraphic and hydrogeological units in the study area (after Kojumdgieva and Popov, 1988; Nakov et al., 2001; Ivanov et al., 2002; Yaneva and Ognjanova-Rumenova, 2012; Yovchev and
Ryzhova, 1962; Benderev et al., 2010).

Formation Age Thickness Lithological description Hydrogeological unit

Brusartsi Late Pontian-Romanian from 40-50 to 120-140 m locally gray pure and sandy clay with intercalations of sand and locally strata of lignite coal Dacian - Romanian complex (DRC)
in the area of Lom
Archar Late Pontian from 20-30 to 100-120 m sand with intercalations of clay Upper-Pontian aquifer (UPA)

4
Smirnenski Maeotian-Pontian up to 500-550 m in the area of the Lom alternation of predominantly claystone and sandstone. The sandstone content Maeotian – Lower Pontian aquitard
depression decreases to the north
Florentin Chersonian 70-80 m; the top of the formation is usually varved calcareous claystone with rare layers of dolomitic limestone and limestone
eroded
Furen Late Bessarabian/ Early 15-550 m limestone with local reef development Sarmatian aquifer
Chersonian
Krivodol Volhynian/ Bessarabian from 20 to 150-200 m and more gray-blue calcareous claystone with intercalations of clayey limestone Sarmatian aquitard
Deleina Badenian 50-200 m limy clays
Journal of Hydrology: Regional Studies 24 (2019) 100611
M. Túri, et al. Journal of Hydrology: Regional Studies 24 (2019) 100611

Fig. 2. Zoning of the Pontian aquifer: 1 - zone of outcrops of the aquifer; 2 - zone where the aquifer is covered by the Loess Formation; 3 – zone of
aquifer recharge through alluvial sediments; 4 - discharge zone of the aquifer into the Archar-Orsoya lowland; 5 - zone where the aquifer is covered
by low permeable sediments and the aquifer is confined; 6 - groundwater table elevation m asl, 7 - top of the Archar Formation.

hydrogeological units: DRA and UPA. The general direction of the groundwater flow is to the north. The aquifer is confined in its
inner part with a hydraulic pressure head of between 40 and 80 m. The hydrodynamic features of UPA presented in Fig. 2 are derived
both from observations and groundwater modeling results (Benderev et al., 2010; Stoyanov, 2019).
The drainage of UPA occurs into alluvial deposits of the Danube River at the western and eastern edges of the aquifer. One part of
the groundwater probably flows north towards Romania, where the sandy fraction in the Dacian-Romanian sediments prevails
(Enciu, 2009).
The Upper Pontian aquifer overlays the Meotian – Lower-Pontian aquiclude (MLPA). The latter is mainly made up of the clayey
sediments of the Smirnenski and Florentin Formations (Table 1), with intercalations of aquiferous sandstones and dolomitic lime-
stones in the first and the second formations respectively.
Outside the areas covered by the Archar Formation, these clayey formations are covered by loess deposits and outcrop only in river
valleys. Data on the spatial position of the MLPA is mainly based on boreholes for oil and gas exploration. There are only a few
drillings with available hydrogeological information and these are mainly in areas where the aquifer is relatively shallow. The
aquiferous sandy and carbonate layers within MLPA are confined with difficult water exchange, except for of its outcropping zones.
The Meotian – Lower-Pontian sediments are underlain by those of Sarmatian age (Fig. 1) that refer to various lithostratigraphic
units with complex spatial relationships due to Facies changes (Cheshitev and Filipov, 1989a, b). These units have been studied in
more details outside the study area. Within the scope of the Lom depression, there is information only on their lithological com-
position based on deep boreholes. The most permeable is limestone of the Furen Formation (Table 1), forming the Sarmatian aquifer
(SA). In outcropping zones, it is karstified with transmissivity up to 220 m2/d. Sarmatian sediments of the Krivodol and Delein
Formations consist of clays with enhanced carbonate content form the Sarmatian aquiclude (Table 1).
As a whole, in the study area the Sarmatian aquifer is confined, with delayed water exchange. In the present study, attention is
mainly focused on the northern zones, where the Sarmatian sediments dip downwards (Filipov, 1995; Haydutov, 1995), with
transition from unconfined to confined conditions.

3. Methods

Groundwater sampling was conducted in October, 2016. Four water samples were taken from the Dacian aquifer and seven from
the Pontian aquifer, as well as two from the Sarmatian aquifer and one groundwater sample from the Meotian aquifer (see Table 1).
The sampling points are shown on the map and the cross-section (Fig. 1). For artesian wells and pumping station, the groundwater
samples were collected directly at the well heads after the boreholes/wells had been primed. Three types of wells were identified and
sampled: artesian wells, water supply wells and observation wells (Table 2). The artesian wells flow continuously. The samples from
the water supply wells were taken during exploitation periods, and the two observation wells were primed to remove more than three
volumes of the water before sampling.
The groundwater temperature, pH, electrical conductivity, alkalinity, as well as redox potential were measured in situ in the field.

5
 M. Túri, et al.

Table 2
Chemical components of the groundwater samples collected during the field campaign held in October, 2016; (ORP: oxidation-reduction redox potential; COD: chemical oxygen demand).

well no. well name well type depth (m) HGU T (°C) pH ORP (mV) EC (μS/cm) Na+ (mg/l) K+ (mg/l) Ca2+ (mg/l) Mg2+ (mg/l)

1 S. Mahala artesian well 64 DRA 14.5 6.7 64 484 19.9 1.6 69.2 12.9
2 Fishfarm water supply well 60 DRA 16.4 7.3 −14 1031 53.3 3 139.5 31.2
3 Komoshtitsa artesian well 120 DRA 15.4 7 −65 731 48.7 0.9 84.8 21.5
pumped
4 Valchedram water supply well 80 DRA 15.9 7.2 −11 632 41.5 1 81.4 14.4
5 Vasilovtsi water supply well 75 UPA 13 7.5 177 510 11.8 1.5 67.6 13.3
6 Agroinvest observation well 106 UPA 15.4 7.6 73 812 50.5 1.9 113.4 25.9
7 RUA water supply well 114 UPA 15.8 7.8 −18 788 56.7 1.9 81.4 16.3
8 Hospital water supply well 180 UPA 14.8 7.5 124 844 44.6 1.7 109.8 31.2
9 Komoshtitsa deep observation well 220 UPA 13.7 7.6 203 840 55.1 1.1 89.2 33.3
10 Dolno Linevo water supply well 200 UPA 16.5 7.1 −27 832 38 1.5 98.2 29.3
11 Dolni Tsibar artesian well 150 UPA 14.6 6.3 −10 1135 103.5 2.4 93.2 22.8
12 Bukovets artesian well 40 SA 12.9 6.9 23 800 28.7 1.4 122.9 20.9
13 Smirnenski water supply well 120 SA 15.3 7 43 632 33.8 2 78.3 20.8
14 Cherni Vrah artesian well 819 MLPA 13.9 6.8 232 722 43.6 1.3 106.3 21.6

6
well no. NH4+ (mg/l) Cl− (mg/l) HCO3− (mg/ SO42−(mg/l) NO3−(mg/l) NO2− (mg/l) PO43− (mg/l) F− (mg/l) COD TOC (mg/l) Ca/Mg molar SO4+HCO3/ SO4/Ca
l) ratio Ca + Mg molar ratio
(meq/meq)
(mg/l)

1 0.1 5.1 301.9 19 0.3 0.1 < 0.5 0.3 1 0.4 3.2 1.2 0.1
2 1.7 22 550.5 109.6 0.3 0.1 < 0.5 0.2 4.2 3.6 2.7 1.2 0.3
3 0.1 10 438 33 0.3 0.1 < 0.5 0.2 0.8 0.4 2.4 1.3 0.2
4 0.2 8.7 378.8 27 0.3 0.1 < 0.5 0.1 0.7 0.4 3.4 1.3 0.1
5 < 0.05 9.6 224.9 35 31 < 0.05 < 0.5 0.3 < 0.5 0.6 3.1 1 0.2
6 1.1 24 420.2 62 8.6 0.2 0.7 0.1 1.2 1.3 2.7 1 0.2
7 2.6 63 372.9 0.9 0.3 0.1 < 0.5 0.2 1.5 0.5 3 1.1 0
8 1.3 25 544.9 19 0.4 0.1 < 0.5 0.2 0.6 0.6 2.1 1.2 0.1
9 < 0.05 4.4 541.6 14 17 < 0.05 < 0.5 0.5 < 0.5 0.4 1.6 1.3 0.1
10 1.6 19 494.2 16 0.4 0.1 < 0.5 0.2 1.2 0.8 2 1.2 0.1
11 5.5 172.7 396.6 4.1 0.3 0.2 < 0.5 0.3 2.4 0.5 2.5 1 0
12 < 0.05 16 461.7 57 0.4 0.1 < 0.5 0.5 < 0.5 0.6 3.6 1.1 0.2
13 0.2 9.1 369.9 41 0.3 < 0.05 < 0.5 0.3 0.5 0.5 2.3 1.2 0.2
14 < 0.05 3.5 476.5 9.8 6 0.1 < 0.5 0.3 0.6 0.3 3 1.1 0
Journal of Hydrology: Regional Studies 24 (2019) 100611
M. Túri, et al. Journal of Hydrology: Regional Studies 24 (2019) 100611

The HCO3−, Cl−, NO3−, NO2−, F−, PO43-, SO42-, Ca2+, K+, Na+, Mg2+, and NH4+ analyses were conducted at the ICER laboratory,
Debrecen, Hungary. The anions were measured by ion chromatography and trace elements by MP-AES (microwave plasma atom
emission spectrometer), additionally NH4+ was determined by spectrophotometry. The COD (chemical oxygen demand) of the water
samples was determined by titration, while the TOC (total organic carbon) by high temperature catalytic oxidation with infrared
detection.
Stable isotopic analyses of water and dissolved inorganic carbon (DIC) from the water samples were carried out with an auto-
mated GASBENCH II sample preparation device attached to a Thermo Finnigan DeltaPLUS XP mass spectrometer (Vodila et al., 2011).
Hydrogen and oxygen isotopes of the groundwater, as well as carbon and oxygen isotopes of carbonate samples were analyzed and
then expressed as δ2H, δ18O, and δ13C values relative to V-PDB for δ13C and V-SMOW for δ2H and δ18O, following the equation: δ =
(Rsample/Rstandard-1)*1000, where R is the 2H/H, 18O/16O or 13C/12C ratio in the sample or in the international standard. The precision
of the measurements is better than ± 0.15 ‰ for δ18O and ± 2 ‰ for δ2H, and ± 0.1 ‰ for δ13C.
The 14C content in the dissolved inorganic carbon (DIC) from the groundwater samples were measured using accelerator mass
spectrometry (EnvironMICADAS AMS). No preservatives were used, but samples were stored in dark and cold conditions. DIC content
of the groundwater sample was degassed as CO2 with addition of phosphoric acid (85%) in vacuum sealed gas reaction vessels at
75 °C for 2 h (Molnár et al., 2013a). The carbon-dioxide samples were cleaned in a cryogenic gas extraction line and converted into
graphite for AMS analyses. The 14C results were corrected for decay of the standard and for isotope fractionation using on-line 13C
measurements on the AMS (Molnár et al., 2013b).
Noble gas and tritium measurements were performed by a VG5400 noble gas mass spectrometer (Fisons Instruments). The
groundwater samples for dissolved noble gases were stored in copper tubes sealed by stainless-steel pinch-off clamps. The mass
spectrometer has an automated extraction and purification system for degassing noble gases from water samples and after the sample
passes through this system the mass spectrometer measures the noble gas concentrations as well as 3He/4He isotope ratios (Papp
et al., 2012). The tritium concentrations were measured using the 3He ingrowth method with a detection limit of 0.012 tritium units
(TU) or better, and the accuracy is 2.4% for TU between 1 and 20 (Palcsu et al., 2010).

4. Results and discussion

4.1. Hydrogeochemistry

Chemical parameters of the 14 samples are summarized in Table 2. Shallow and deep wells were sampled: there are water samples
from the Dacian aquifer with depths between 60–120 m and from the Pontian aquifer from depths between 75–225 m, and one
sample from the Meotian aquifer at a depth of 819 m. The temperature of the groundwater from the Dacian aquifer was between 14.5
and 16.4 °C and the water samples from the Pontian aquifer was in the range 13.0–16.5 °C. The aquifers are generally characterized
by neutral pH values (6.27–7.84).
In terms of dissolved anions, HCO3− dominates in the waters followed by SO42- and then Cl−. In case of dissolved cations the
dominant ion is Ca2+ followed by Mg2+. The Ca-HCO3 water type is dominant in all aquifers except for the RUA well which is of Ca-
Na-HCO3 type and for Dolni Tsibar, which has Ca-Na-HCO3-Cl chemistry.
It can be seen that moving away from the recharge zone, the dissolved solids become slightly higher along the prevailing flow
lines (Fig. 1, Table 2). A study of possible water-rock interaction processes which control the groundwater chemistry is necessary to
understand the basic geochemical characteristics and processes of an aquifer. Possible carbonate dissolution processes can be de-
termined based on the Ca/Mg molar ratio. In this case, the Ca/Mg ratio of the samples is in the range from 1.6 to 3.6 (Table 2), where
the ratio between 1 and 2 is attributed to calcite dissolution and a ratio higher than 2 is typical for the contribution of calcite rich
minerals (Katz et al., 1997). Probable ion-exchange processes can be characterized by an equivalent ratio of Ca + Mg versus
SO4+HCO3 of groundwater samples, which values are between 1.0 and 1.3 (Table 2) and this indicates that ion exchange can occur
in the area and suggests dissolution of carbonate and gypsum (Fisher and Mullican, 1997). This molar ratio is appropriate for
estimating the contribution of gypsum, calcite and dolomite to the water composition, although the SO4/Ca molar ratio of
groundwater samples ranges between 0 and 0.3 (Table 2) may confirm the lack of gypsum dissolution. For a better illustration of the
chemical character of the groundwater, ions are plotted in a Piper diagram (Fig. 3).

4.2. The stable oxygen and hydrogen isotope composition of samples

The stable isotopic values of groundwater samples from all the studied aquifers vary between -9.82 ‰ and -14.24 ‰ for δ18O, and
between -68.7 ‰ and -103.1 ‰ for δ2H (Fig. 4). The samples from the Dacian aquifer display less variability between -10.44 ‰ and
-12.44 ‰ for δ18O, and -72.4 ‰ to -89.6 ‰ for δ2H. The GMWL (Global Meteoric Water Line) shows the average relationship
between the hydrogen and oxygen isotope ratios in precipitation, which is expressed as a worldwide average δ2H = 8*δ18O+10
(Craig, 1961). Although there is no measured stable isotope composition of precipitation in the area, the recent precipitation at
Ramnicu Valcea, Romania (GNIP station of IAEA) located 150 km northeast of the research area can be considered representative for
the Danubian Plain. Monthly δ18O values in regional precipitation are -2.5 ‰ to -15.0 ‰ and -23.6 to -111.0 for δ18O and δ2H,
respectively. The prevailing winds for the study area are from west and northwest and seasonally from south. Generally, west-east or
south-north transport of air masses over Bulgaria prevails. Zonal circulation has a larger influence on precipitation in Bulgaria
compared to meridional circulation (Nojarov, 2017). Some prominent local geographical features including the Black Sea in the east,
and Carpathian Mountains and Balkan Mountains to the north and west, contribute to the evolution of air masses (Knight et al.,

7
M. Túri, et al. Journal of Hydrology: Regional Studies 24 (2019) 100611

Fig. 3. Piper diagram of the groundwater samples.

Fig. 4. δ2H versus δ18O values of groundwater investigated in this study as compared to the GMWL and LMWL.

2004). According to the stable isotope data for Ramnicu Valcea precipitation, the LMWL (Local Meteoric Water Line) can be cal-
culated and gives a regional average δ2H = 7.03*δ18O-1.44 and the hydrogen and oxygen isotope distributions are correlated for
meteoric waters, thus these data reflect a meteoric origin for groundwater samples.

4.3. Water mean residence time

To better understand water transport in a multilayered aquifer system, the residence time of groundwater can be a useful tool.
Tritium, radiocarbon, noble gases and stable isotopes can be used both for age dating as tracers in groundwater systems. These
environmental isotopic tracers can be used to estimate residence time, and to determine environmental conditions upon recharge.

4.3.1. Tritium
As tritium is a radioactive isotope with a short half-life of 12.32 years (Lucas and Unterweger, 2000), it can be used to identify
modern recharge. Natural tritium is produced in the atmosphere due to nuclear reactions between galactic cosmic rays and atmo-
spheric gases. The tritium concentration in precipitation is representative of the atmosphere and is about 10 TU in Central Europe.
The average value in Debrecen, Hungary is 11.3 TU (Palcsu et al., 2018) and 11.66 TU in Râmnicu Vâlcea, Romania (Varlam et al.,
2013). Our tritium results are shown in Table 3. Most of the groundwater samples showed very low tritium concentrations (lower
than 0.02 for RUA, Valchedram, Komoshtitsa pumped, Cherni Vrah, Dolno Linevo, Hospital, S. Mahala, Komoshtitsa deep, Dolni

8
 M. Túri, et al.

Table 3
Isotope data of sampled points with modelled mean residence time of waters using Ingerson and Pearson (1964) model.

14 3
well no. well name C pMC H (TU) δ13C (‰, VPDB) δ2H (‰, VSMOW) δ18O (‰, VSMOW) pH ε13CHCO3-CO2(g) δ13Csoil A0 Ingerson and Pearson (yr BP) aquifer

1 S. Mahala 63.18 ( ± 0.22) 0.014 ( ± 0.003) −14.75 −72.4 −10.11 6.67 −9.11 −15.89 91.49 3061 Dacian
2 Fishfarm 53.11 ( ± 0.19) 1.232 ( ± 0.025) −13.11 −87.5 −12.14 7.34 −8.89 −16.11 80.21 3409 Dacian
3 Komoshtitsa pumped 29.15 ( ± 0.19) 0.003 ( ± 0.005) −13.42 −89.6 −12.2 7 −9.00 −16.00 82.73 8622 Dacian
4 Valchedram 23.67 ( ± 0.14) 0 ( ± 0.005) −13.32 −89.4 −12.44 7.23 −8.95 −16.05 81.78 10249 Dacian
5 Vasilovtsi 98.6 ( ± 0.28) 6.818 ( ± 0.057) −12.86 −68.7 −9.82 7.5 −9.28 −15.72 80.64 −1663 Pontian

9
6 Agroinvest 23.85 ( ± 0.13) 0.150 ( ± 0.005) −11.9 −103.1 −14.24 7.61 −9.00 −16.00 73.36 9288 Pontian
7 RUA 33.61 ( ± 0.16) 0 ( ± 0.020) −14.51 −88.6 −12.14 7.84 −8.96 −16.04 89.16 8065 Pontian
8 Hospital 46.15 ( ± 0.18) 0.009 ( ± 0.007) −13.65 −77.9 −11.08 7.46 −9.07 −15.93 84.53 5004 Pontian
9 Komoshtitsa deep 63.68 ( ± 0.20) 0.015 ( ± 0.005) −10.6 −75.7 −10.71 7.6 −9.20 −15.80 66.17 316 Pontian
10 Dolno Linevo 39.83 ( ± 0.17) 0.008 ( ± 0.005) −15 −79 −11.32 7.13 −8.88 −16.12 91.74 6897 Pontian
11 Dolni Tsibar 20.83 ( ± 0.12) 0.016 ( ± 0.004) −13.93 −86.7 −12.07 6.27 −9.10 −15.90 86.34 11756 Pontian
12 Bukovets 46.95 ( ± 0.18) 0.986 ( ± 0.020) −12.21 −70.2 −9.87 6.91 −9.29 −15.71 76.67 4055 Sarmatian
13 Smirnenski 15.94 ( ± 0.11) 0.107 ( ± 0.007) −10.92 −85.1 −11.77 6.96 −9.01 −15.99 67.36 11917 Sarmatian
14 Cherni Vrah 60.37 ( ± 0.20) 0.006 ( ± 0.004) −14.59 −73.1 −10.1 6.84 −9.18 −15.82 90.91 3384 Meotian
Journal of Hydrology: Regional Studies 24 (2019) 100611
M. Túri, et al. Journal of Hydrology: Regional Studies 24 (2019) 100611

Fig. 5. The yearly 3H concentrations in precipitation from 1961 to 2009 in Vienna, Austria (IAEA/WMO (2009)) with the calculated initial tritium
content of Vasilovtsi by the time of recharge.

Tsibar). Groundwater from the Vasilovtsi well had the highest tritium content of 6.8 TU (indicating modern recharge), followed by
Fishfarm with a value of 1.23 TU and Bukovets with a value of 0.99 TU. There are two wells with a value in the range of 0.1-0.15 TU
(Smirnenski, Agroinvest). The measureable tritium in Smirnenski well along with the lowest radiocarbon content (15.94 pMC)
attracts attention and suggests mixing with recent water. This mixing process could have occurred due to a malfunction of the well or
from a well with multiple screen depths, or from leakage. Assuming this sample represents aquifer water, we infer that the Sarmatian
aquifer is affected by mixing between recent and old groundwater. A similar observation was made by Joshi et al., 2018 with regards
to the isotopic signature of the groundwater from aquifers of NW India. This is a reasonable interpretation as the recharge zone is
located nearby.
Since groundwater from the Vasilovtsi well has the highest tritium concentration, it is worth calculating the 3H/3He age of
groundwater. Although dissolved 3He in groundwater may have several origins, the 3H/3He age of Vasilovtsi can be calculated and
more precise groundwater residence time can be estimated. The tritiogenic 3He concentration is calculated by subtracting from the
measured helium content that amount due to atmospheric solubility, excess air and radiogenic production (Schlosser et al., 1989).
Based on these calculations, the tritiogenic 3He is 60.1 TU and the 3H/3He age is calculated to be 40.6 ± 0.7 years. According to the
calculations and tritium recovery curve of precipitation in Vienna (Fig. 5), the initial tritium concentration at the time of recharge
was 66.9 ± 2.9 TU, and the decay from the time of the mid 1960s and mid-1970s "bomb" peak this peak provides a reference point
for determining groundwater residence time. The samples were taken in 2016 and a 3H/3He age of ˜40 years implies the recharge
events could have occurred around 1976 (Fig. 5).

4.3.2. Modelled radiocarbon ages

In groundwater, the DIC (dissolved inorganic carbon) results from CO2(aq)+HCO3−+CO32- which are dissolved in the recharge
water, and these contain atmospheric 14C. This is a useful tool for dating groundwater, and is measured in dissolved inorganic/
organic carbon. For estimation of groundwater residence time, we use the radiocarbon model of Ingerson and Pearson (1964). This
method produces similar ages to a variety of other methods (Ingerson and Pearson, 1964; Fontes and Garnier, 1979; Mook, 1980;
Salem et al., 1980; Plummer and Glynn, 2013). To calculate the apparent age of the groundwater (in years), the following equation
can be used:

t=8267 ln (A0/At)

To estimate the residence time of groundwater, the initial 14C content (A0) of DIC in groundwater at the recharge zone must be
evaluated. Here the Vasilovtsi well can be used as representative for initial 14C content, based on the 3H/3He age and the measured
14
C content which is 98.6 pMC. This is somewhat higher based on the groundwater recharge that can have occurred during the last 50
years and is therefore likely to contain a component of bomb derived 14C. The results of the Ingerson and Pearson model (1964) are
given in Table 3. In this model, a dilution factor (q) is used to estimate A0, according to the equation: q=(δ13CDIC- δ13Ccarb)/(δ13Csoil-
δ13Ccarb). Here, δ13CDIC is the measured δ13C in groundwater, and δ13Ccarb is the δ13C value of the dissolved carbonate which is
usually 0 ‰. The δ13Csoil term is the δ13C value of the CO2 in the soil, which depends on the plants growing there (in temperate zone:
C3), C3 plants are about -25 ‰ (Table 3). An enrichment factor was used to calculate q because the soil can be enriched in 13C based
on equilibrium exchange processes between soil CO2 and DIC at increasing pH. The dilution factor (q) can be calculated based on the
equation: q=(δ13CDIC- δ13Ccarb)/ ((δ13Csoil+ Ɛ13CDIC-CO2(soil))- δ13Ccarb) where the Ɛ13CDIC-CO2(soil) depends on pH following the δ13C

10
M. Túri, et al. Journal of Hydrology: Regional Studies 24 (2019) 100611

Fig. 6. Terrigenous helium concentration in groundwater concerning different aquifers in respect to groundwater age (yr BP).

mixing model of Pearson (1965) and Pearson and Hanshaw (1970), see δ13C correction (Table 3). The error in the corrected ages was
derived by propagating the errors of the independent elements in the dilution factor equation, including errors for measured
radiocarbon content and δ 13C values of DIC, pH, and alkalinity are the measurement errors. The corrected ages are reported with one
sigma error.
The 14C content of the groundwater samples range from 15.9 pMC to 98.6 pMC (Table 3, Fig.1). These data suggest that three
aquifers of four might contain waters recharged during the early Holocene - late Pleistocene (Pontian, Dacian, Sarmatian). There are
two wells in the Pontian aquifer (Dolni Tsibar, 20.83 ± 0.12 pMC; Agroinvest, 23.85 ± 0.13 pMC) with low radiocarbon content.
The wells near the area of Agroinvest are characterized by higher radiocarbon content, moving away from the well (in any direction)
the radiocarbon content is increasing. In the Dacian aquifer, there is only one well, the Valchedram, which might show late Pleis-
tocene recharge given the level of 23.7 pMC. In the Sarmatian aquifer, the oldest water was obtained from all samples with 15.94
pMC for the Smirnenski well. This is of interest, because the distance between Bukovets and Smirnenki is less than 5 km although the
screen depth of Smirnenski well is 100 m deeper, so this difference may be related to another flow path.
The age of the groundwater is the mean water residence time in the aquifer of each individual groundwater sample. This was
calculated from the average of ages derived from the various radiocarbon models. The uncertainty associated with residence times is
obtained as the standard deviation of ages from the different models, and then this average age is consistently used in the following
interpretations. The 3H/3He age determination showed the most recently-recharged sample was Vasilovtsi from the Pontian aquifer
(Fig. 5). A number of samples appear to be recharged during the late Holocene (Komoshtitsa deep, S. Mahala, CherniVrah, Fishfarm,
Bukovets) with ages between 3000 and 4000 calibrated years before present (yr BP) (Fig. 6). In addition, there are samples with ages
between 5000–6800 yr BP (Hospital, Dolno Linevo) from the Pontian aquifer. Furthermore, there are groundwater samples with
residence time between 8000-9300 yr, which represents the early Holocene (RUA, Komoshtitsa pumped, Agroinvest) (Fig. 6).
Samples with ages between 10 000–12 000 yr BP (Valchedram, DolniTsibar, Smirnenski) closely correspond to the late Pleistocene or
the early Holocene – late Pleistocene transition.

4.3.3. Terrigenous helium accumulation

Terrestrial 4He (4Heterr) can be used to classify groundwaters with longer mean residence times (Solomon, 2000). It is used
especially when characterizing the old fractions in groundwater that have experienced mixing. Modelled water mean residence times
plotted versus 4Heterr content identify groundwater samples with relatively long residence times that contain excess 4Heterr (Fig. 6),
likely incurred during the early Holocene – late Pleistocene climatic transition. We observe a tendency of increased terrigenic helium
contents in older groundwater from the Pontian aquifer and in the Sminenski sample (Fig. 6).

4.3.4. Hydrogeological aspects

The correct interpretation of the results requires the exact location of the boreholes sampled. Some of the boreholes correspond to
the deeper confined zone of the aquifer, and others to the outcrops of the aquiferous formations. Considering the groundwater
properties of the aquifers and corresponding boreholes, the following hydrogeological settings can be defined.

4.3.4.1. Sarmatian. One of the boreholes (Bukovets) is located within the outcrops of the aquifer, and the second one (Smirnenski) is
below the local erosion base (Fig. 1). The covered (confined) part of the aquifer is generally characterized by very slow water
exchange. This is mainly observed at the village of Smirenenski. The groundwater residence time for the borehole in the village of
Smirnenski shows weak water exchange and the highest groundwater age, possibly due to the influence of deeper water.

11
M. Túri, et al. Journal of Hydrology: Regional Studies 24 (2019) 100611

4.3.4.2. Meotian. Relatively younger groundwater compared to other waters indicates a probable connection between the
Smirnenski Formation sandy layers with the outcropping sands of the Pontian aquifer (Fig. 1).

4.3.4.3. Dacian. A specific feature of this aquifer complex are sandy lenses and layers that vary in thickness, location in depth, and
hydraulic conductivity values (depending of the sandy content). The wide range of groundwater age in two neighboring boreholes
close to the village of Komoshtitsa (300 yr BP and 8600 yr BP) indicates poor hydraulic connection between different lenses and
layers as well as different recharge and discharge conditions. This significant difference also suggests the absence of a hydraulic
connection in this area between the waters of Dacian and underlying Pontian aquifers.

4.3.4.4. Pontian. The numerous samples of the Pontian aquifer (7 samples) demonstrate groundwater ages that range from about 300
to 11 700 yr BP. One exception is the water sample from the borehole in the village of Vasilovtsi, where water can be considered to be
modern. In this area, the Pontian aquifer is relatively shallow and could have a direct connection with the river passing near the
village at lower elevation. The highest values (from about 8000 to 11 700 yr BP) are common in the central part of the aquifer, with a
reduced age near the town of Lom (to 5000 yr BP). This phenomenon could be explained by significantly higher hydraulic
conductivity values in this area. A second possibility could be intensive groundwater exploitation near to the town of Lom.
The various samples taken from the Pontian aquifer were analyzed for possible interrelations between parameters determined
from individual groundwater samples. The most indicative parameter is thought to be the groundwater age. Increasing groundwater
age is associated with increasing concentrations of Na+, K+, Cl− concentrations and increasing electrical conductivity of water.
Moreover, an increase nitrite NO2− and ammonium NH4+ contents is registered along with increasing of COD (chemical oxygen
demand) in these water samples. Conversely, the redox potential and nitrate NO3− show evident decrease with the increasing
groundwater age.

4.4. Noble gas temperature and paleoclimate reconstruction

The concentrations of noble gases are suitable indicators to estimate the groundwater recharge temperature. The noble gas
concentrations of Ne, Ar, Kr and Xe are derived from the atmosphere and after their dissolution into the water their concentrations do
not change along the flow path since they take part neither in chemical nor biological processes. The interpretation of noble gas
concentrations in groundwater requires inverse modelling, which can determine reliable noble gas solubility or recharge temperature
(NGT). This requires a correction for excess air and a model proposed to determine its formation. The closed system equilibration
model of gas partitioning was used in this study (Aeschbach-Hertig et al., 1999). Noble gas results are shown in Table 4 and the
uncertainty of the noble gas temperature determination is about ± 0.6 °C.
We already noted that the average air temperature in the Danubian Plain is 11.4 °C. Water temperature near in the recharge area
(Vasilovtsi) is 13 °C, as measured in the field, which may reflect ground temperature. The noble gas temperatures measured vary
between 8.1 °C and 13.4 °C (Table 4). NGTs of the Dacian aquifer vary between 9.95 °C and 11.17 °C although the NGTs from the
Pontian aquifer shows higher variability 8.1 °C to 13.4 °C.
Somewhat lower NGT values in the Smirnenski, Dolni Tsibar and Valchedram wells presumably indicate recharge during the
glacial-Holocene transition (Fig. 7). A likely admixture of recent groundwater to glacial water is conceivable due to the low
radiocarbon content with measureable tritium in a water sample from the Smirnenski well, and a minimal tritium content in the Dolni
Tsibar well.
The waters from the Sarmatian and Pontian aquifers are presumed to have recharged during late glacial times, or to have
incorporated glacial water that result in NGT somewhat lower than groundwater from the mid and late Holocene (Fig. 7). These
groundwater samples are characterized by NGTs of about 8.3 °C, whereas the late Holocene waters have NGTs around 12 °C, which is

Table 4
Noble gas results and calculated noble gas temperature (°C) data of groundwater samples and their fit to the CE-model.

well no. well name He Ne Ar Kr Xe R/Ra NGT ± ΔT X2 p (%) 4

Heterr
(ccSTP/g) (ccSTP/g) (ccSTP/g) (ccSTP/g) (ccSTP/g) (°C) (°C) (ccSTP/g)

1 S. Mahala 7.58E-08 2.69E-07 4.33E-04 9.55E-08 1.33E-08 0.84 10.57 0.69 0.03 86.8 (1.19 ± 0.15)E-08
2 Fishfarm 7.23E-08 2.44E-07 3.96E-04 8.85E-08 1.27E-08 0.78 11.17 0.57 0.63 42.6 (1.31 ± 0.14)E-08
3 Komoshtitsa pumped 1.47E-07 4.93E-07 5.37E-04 1.08E-07 1.44E-08 0.77 9.95 0.61 0.04 84.6 (1.82 ± 0.30)E-08
4 Valchedram 9.36E-08 2.79E-07 4.32E-04 9.53E-08 1.35E-08 0.69 9.97 0.59 0.18 67.1 (2.62 ± 0.16)E-08
5 Vasilovtsi 6.45E-08 2.74E-07 4.17E-04 9.25E-08 1.29E-08 2.70 10.98 0.59 0 94.5 (0.00 ± 0.15)E+00
6 Agroinvest 1.02E-07 3.16E-07 4.67E-04 9.97E-08 1.37E-08 0.68 10.37 0.68 0.17 68.3 (2.58 ± 0.18)E-08
7 RUA 2.12E-07 2.59E-07 4.22E-04 9.57E-08 1.40E-08 0.28 8.38 0.52 0.93 32.7 (1.49 ± 0.25)E-07
8 Hospital 1.61E-07 2.59E-07 4.24E-04 9.72E-08 1.41E-08 0.37 8.13 0.51 0.08 76.9 (9.81 ± 0.20)E-08
9 Komoshtitsa deep 7.48E-08 3.67E-07 4.20E-04 8.91E-08 1.21E-08 0.94 13.42 0.54 0.05 82.6 0.00( ± 1.84E-09)
10 Dolno Linevo 1.81E-07 2.63E-07 4.34E-04 9.85E-08 1.41E-08 0.33 8.25 0.56 0.05 82.4 (1.19 ± 0.02)E-07
11 Dolni Tsibar 5.57E-07 2.61E-07 4.39E-04 9.82E-08 1.41E-08 0.13 8.51 0.61 0.57 45.2 (4.95 ± 0.06)E-07
12 Bukovets 5.29E-08 2.40E-07 3.85E-04 8.70E-08 1.22E-08 1.00 11.27 0.57 0.18 87.4 0.00( ± 1.29E-09)
13 Smirnenski 1.33E-07 2.62E-07 4.30E-04 9.76E-08 1.38E-08 0.45 8.23 0.58 0.01 91.4 (7.08 ± 0.18)E-08
14 Cherni Vrah 8.04E-08 2.23E-07 4.21E-04 8.91E-08 1.21E-08 0.89 13.37 0.55 0.04 84.9 0.00( ± 1.89E-09)

12
M. Túri, et al. Journal of Hydrology: Regional Studies 24 (2019) 100611

Fig. 7. Noble gas temperatures vs. modelled water mean residence time (year) with errors (standard deviation of modelled ages).

similar to the mean annual air temperature of 11.4 °C so the largest temperature change observed is about 3.7 °C. These recharge
temperature differences might represent the transition time between these two periods. This transition period can be also observed in
the stable isotope data, where the groundwater with larger residence times has more negative δ2H and δ18O values, suggesting the
groundwater recharge occurred during the Late Glacial–Holocene climatic transition period (Table 3, Fig. 8).
Paleoclimate reconstruction studies in Europe based on dissolved noble gases in groundwater have shown higher temperature
differences between the Holocene and glacial periods, for example 5–7 °C in the Bohemian Cretaceous Basin (Corcho Alvarado et al.,
2013) and at least 5 °C in the Glatt Valley, Switzerland (Beyerle et al., 1998). In the Pannonian Basin, the warming after the last
glacial maximum (LGM) reached a temperature of 9.1 °C (Varsányi et al., 2011) and this temperature in the Ledo-Paniselian Aquifer is
9.5 °C (Blaser et al., 2010). Comparing these NGT values in Europe, the temperature differences obtained here appear to be too low.
Comparable observations were made for a Jurassic limestone aquifer in Poland, where the NGT was 2–3.5 °C lower than recent
recharge groundwater (Osenbrück et al., 1993). Assuming some mixing with recent waters, these noble gas temperatures could reflect
either mixing, or a transition phase between the two climatic periods. The lower NGT difference between the colder and warmer
recharge waters can be explained as that the mean residence times do not reflect groundwater recharged at the LGM, i.e. the
groundwater at the oldest part of the aquifer does not reflect glacial temperatures.

5. Conclusion

This study presents a hydrogeological and palaeoclimatic interpretation for the Lom depression, Northwest Bulgaria. A multi-
tracer approach is presented to reconstruct the recharge temperature and identify isotopic hydrological character of the aquifers
using the environmental isotopes: δ18O, δ2H, δ13C, 3H, 14C, and noble gases. The stable isotope results show that the recharge sources
of the groundwater studied are local meteoric water. For a groundwater sample from the Vasilovtsi well, the 3H/3He groundwater age
of 40.6 ± 0.7 years was confirmed. Tritium content in deep groundwater with low radiocarbon content was determined for three

Fig. 8. Modelled water mean residence time (year) (standard deviation of modelled ages) against stable oxygen isotopes ( ± 0.2 ‰).

13
M. Túri, et al. Journal of Hydrology: Regional Studies 24 (2019) 100611

wells. The presence of tritium can be attributed to mixing process between old and recent waters due either to overexploitation or
leakage in the well. The Ingerson and Pearson radiocarbon model was used to estimate mean residence times taking into account
chemical and isotope hydrological parameters to achieve precise groundwater ages from the 14C content of dissolved inorganic
carbon. Modelled groundwater residence times fall within the last twelve thousand years, with samples from the early Holocene with
ages between 8000–9300 yr BP (RUA, Komoshtitsa pumped, Agroinvest) and older waters that date to between 10 000–12 000 yr BP
(Valchedram, Dolni Tsibar, Smirnenski), from the late Pleistocene or the transition time between the Holocene and Pleistocene.
The noble gas results provide quantitative information on the temperature of recharge and qualitative information on the time of
recharge. They make it conceivable to estimate the climatic conditions under which the recharge took place and to confirm the time
of recharge. Values of 3–4 °C lower NGT for Smirnenski, Dolni Tsibar and Valchedram presumably indicate that the recharge occurred
during the glacial-Holocene transition. An admixture of recent Holocene groundwater to glacial water is likely due to the low
radiocarbon content in presence of some tritium in water sample from Smirnenski well, and a low tritium content for the Dolni Tsibar
well.
The results of this study significantly change our previous knowledge about the hydrodynamics of the Lom Basin. In particular, we
have shown that the Pontian aquifer has no hydraulic connection to the Danube River.The sandy layers and lenses forming the Dacian
aquifer are apparently hydraulically isolated and therefore have high degree of heterogeneity.

Acknowledgement

We are grateful for the help of Prof. A.J. Timothy Jull for the English revision and for his constructive comments. The research
was supported by the European Union and the State of Hungary, co-financed by the European Regional Development Fund in the
project of GINOP-2.3.2-15-2016-00009 ‘ICER’.

Appendix A. Supplementary data

Supplementary material related to this article can be found, in the online version, at doi:https://doi.org/10.1016/j.ejrh.2019.
100611.

References

Aeschbach-Hertig, W., Gleeson, T., 2012. Regional strategies for the accelerating global problem of groundwater depletion. Nat. Geosci. 5 (12), 853.
Aeschbach-Hertig, W., Peeters, F., Beyerle, U., Kipfer, K., 1999. Interpretation of dissolved atmospheric noble gases in natural waters. Water Resour. Res. 35,
2779–2792.
Antonov, H., Danchev, D., 1980. Groundwater in the Republic of Bulgaria. . “Technika’’ Publisher, Sofia, pp. 360 p (in Bulgarian).
Benderev, A., Spasov, V., Mihaylova, B., Vasileva, T., 2010. Hydrogeological settings of the Lom coal basin and problems relevant to the future mining development.
Eng. Geol. Hydrogeol. 25, 123–142.
Beyerle, U., Purtschert, R., Aeschbach-Hertig, W., Imboden, D.M., Loosli, H.H., Wieler, R., Kipfer, R., 1998. Climate and groundwater recharge during the last
glaciation in an ice-covered region. Science 282 (Iss. 5389), 731.
Blaser, P.C., Kipfer, R., Loosli, H.H., Walraevens, K., Van Camp, M., Aeschbach-Hertig, W., 2010. A 40 ka record of temperature and permafrost conditions in
northwestern Europe from noble gases in the Ledo‐Paniselian Aquifer (Belgium). J. Quat. Sci. 25 (6), 1038–1044.
Cheshitev, G., Filipov, L. (Eds.), 1989. Geological Map of Bulgaria in Scale 1:100,000. Kozloduy Map Sheet. Committee on Geology. Company for Geophysical surveys
and Geological mapping, Sofia.
Cheshitev, G., Filipov, L. (Eds.), 1989. Geological Map of Bulgaria in Scale 1:100,000. Lom Map Sheet. Committee on Geology. Company for Geophysical surveys and
Geological mapping, Sofia.
Clark, I.D., Fritz, P., 1997. Environmental Isotopes in Hydrogeology. Lewis Publishers ISBN1-56670-249-6.
Craig, H., 1961. Isotopic variations in meteoric waters. Science 133, 1702–1703.
Corcho Alvarado, J.A., Pačes, T., Purtschert, R., 2013. Dating groundwater in the Bohemian Cretaceous Basin: understanding tracer variations in the subsurface. Appl.
Geochem. 29 (February 2013), 189–198.
Enciu, P., 2009. Pliocene and Quaternary of the Western Part of the Dacian Basin. Academiei Romane, Bucuresti, pp. 251 p.
Filipov, L. (Ed.), 1995. Geological Map of Bulgaria in Scale 1:100,000. Montana Map Sheet. Committee on Geology. Company for Geophysical surveys and Geological
mapping, Sofia.
Fisher, R.S., Mullican III, W.F., 1997. Hydrochemical evolution of sodium-sulfate and sodium-chloride groundwater beneath the northern Chihuahuan Desert, Trans-
Pecos, Texas, USA. Hydrogeol. J. 5 (2), 4–16.
Fontes, J.C., 1980. Environmental isotopes in groundwater hydrology. Handbook of Environmental Isotope Geochemistry, vol. 1.
Fontes, J.C., Garnier, J.M., 1979. Determination of the initial 14C activity of the total dissolved carbon-a review of the existing models and a new approach. Water
Resour. Res. 15, 399–413.
Haydutov, I. (Ed.), 1995. Geological Map of Bulgaria in Scale 1:100,000. Belogradchik Map Sheet. Committee on Geology. Company for Geophysical surveys and
Geological mapping, Sofia.
IAEA/WMO, 2009. Global Network of Isotopes in Precipitation. The GNIP Database. Accessible at:. http://www.iaea.org/water.
Ingerson, C.W., Pearson, F.J., 1964. . Estimation of Age and Rate of Motion of Ground-water by the 14C Method. Recent Researches in the Field of Hydrosphere,
Atmosphere, and Nuclear Geochemistry, Sugawara Festival Volume. Maruzen Company, Tokyo, Japan, pp. 263–283.
Ivanov, D., Ashraf, A.R., Mosbrugger, V., Palamarev, E., 2002. Palynological evidence for miocene climate change in the Forecarpathian Basin (central paratethys, NW
Bulgaria). Palaeogeogr. Palaeoclimatol. Palaeoecol. 178 (1), 19–37.
Joshi, S.K., Rai, S.P., Sinha, R., Gupta, S., Densmore, A.L., Rawat, Y.S., Shekhar, S., 2018. Tracing groundwater recharge sources in the northwestern Indian alluvial
aquifer using water isotopes (δ18O, δ2H and 3H). J. Hydrol. 559, 835–847.
Katz, B.G., Coplen, T.B., Bullen, T.D., Davis, J.H., 1997. Use of chemical and isotopic tracers to characterize the interactions between ground water and surface water in
mantled karst. Groundwater 35 (6), 1014–1028.
Klump, S., Tomonaga, Y., Kienzler, P., Kinzelbach, W., Baumann, T., Imboden, D.M., Kipfer, R., 2007. Field experiments yield new insights into gas exchange and
excess air formation in natural porous media. Geochim. Cosmochim. Acta 71 (6), 1385–1397.
Knight, C.G., Raev, I., Staneva, M.P., 2004. Drought in Bulgaria: A Contemporary Analog for Climate Change. Ashgate Publishing Limited, Hants, England.
Kojumdgieva, E., Popov, N., 1988. Lithostratigraphy of the neogene sediments in Northwestern Bulgaria. Palaeontol. Stratigr. Litol. 25, 3–26 (in Bulgarian with English

14
M. Túri, et al. Journal of Hydrology: Regional Studies 24 (2019) 100611

abstract).
Lucas, L.L., Unterweger, M.P., 2000. Comprehensive review and critical evaluation of the half-life of Tritium. J. Res. Inst. Stand. Technol. 105 (4), 541.
Machkova, M., Pavlova, V., Benderev, A., Pulido Bosch, A., Lichev, A., 2014. Stable isotopes in the main karst aquifers in Dobrudzha region (NE Bulgaria). Proc. XX
Congress of Carpathian Balkan Geological Association, 24 to 26 September 2014 388–391.
Magro, G., Gherardi, F., Bellani, S., 2010. Noble gases in karstic and thermal waters of Strimon basin (Greece-Bulgaria). In: Birkle, Torres-Alvarado (Eds.), Water-Rock
Interaction XIII. Taylor & Francis Group, London, pp. 345–348.
Molnár, M., Janovics, R., Major, I., Orsovszki, J., Gönczi, R., Veres, M., Leonard, A.G., Castle, S.M., Lange, T.E., Wacker, L., Hajdas, I., Jull, A.J.T., 2013a. Status report
of the new AMS 14C sample preparation lab of the Hertelendi Laboratory of Environmental Studies (Debrecen, Hungary). Radiocarbon 55 (Nr 2–3), 665–676.
Molnár, M., Rinyu, L., Veres, M., Seiler, M., Wacker, L., Synal, H.-A., 2013b. EnvironMICADAS: a mini 14C AMS with enhanced Gas Ion Source Interface in the
Hertelendi Laboratory of Environmental Studies (HEKAL), Hungary. Radiocarbon 55 (Nr 2–3), 338–344.
Mook, W.G., 1980. Carbon-14 in hydrogeological studies. In: In: Fritz, P., Fontes, J.C. (Eds.), Handbook of Environmental Isotope Geochemistry, vol. 1. Elsevier,
Amsterdam, pp. 49–74.
Nakov, R., Burchfiel, B.C., Tzankov, T., Royden, L.H., 2001. Late Miocene to Recent Sedimentary Basins of Bulgaria. – Geological Society of America Map and Chart
Series MCH 088, Scale 1/200,000. pp. 28 p.
Nojarov, P., 2017. Circulation factors affecting precipitation over Bulgaria. Theor. Appl. Climatol. 127 (1-2), 87–101.
Osenbrück, K., Weise, S.M., Zuber, A., Grabczak, J., Ciezkowski, W., 1993. Noble Gas Temperatures and Ages of Some Glacial and Buried Brine Waters in Poland.
Isotope Techniques in the Study of Past and Current Environmental Changes in the Hydrosphere and the Atmosphere. IAEA, Vienna, pp. 319–336.
Palcsu, L., Major, Z., Köllő, Z., Papp, L., 2010. Using an ultrapure 4He spike in tritium measurements of environmental water samples by the 3He-ingrowth method.
Rapid Commun. Mass Spectrom. 24 (5), 698–704.
Palcsu, L., Morgenstern, U., Sültenfuss, J., Koltai, G., László, E., Temovski, M., Major, Z., Nagy, J.T., Papp, L., Varlam, C., Faurescu, I., Czuppon, G., Bottyán, E., Túri,
M., Rinyu, L., Jull, T., 2018. Modulation of cosmogenic tritium in meteoric precipitation by the 11-year cycle of solar magnetic field activity. Sci. Rep. 2018, 8.
Papp, L., Palcsu, L., Major, Z., Rinyu, L., Tóth, I., 2012. A mass spectrometric line for tritium analysis of water and noble gas measurements from different water
amounts in the range of microlitres and millilitres. Isotopes Environ. Health Stud. 48 (1), 494–511.
Pearson Jr, F.J., 1965. Use of C-13/C-12 Ratios to Correct Radiocarbon Ages of Material Initially Diluted by Limestone, Paper Presented at Sixth International
Conference on Radiocarbon. US At. Energy Comm., Pullman, Wash.
Pearson Jr., F.J., Hanshaw, B.B., 1970. Sources of Dissolved Carbonate Species in Ground Water and Their Effects on carbon-14 Dating.
Peel, M.C., Finlayson, B.L., McMahon, T.A., 2007. Updated world map of the Köppen-Geiger climate classification. Hydrol. Earth Syst. Sci. Discuss. 4 (2), 439–473.
Plummer, L.N., Glynn, P.D., 2013. Radiocarbon dating in groundwater systems. Isotope Methods for Dating Old Groundwater. pp. 33–89.
Pulido-Bosch, A., Lopez-Chicano, M., Calaforra, J.M., Cal Vache, M.L., Machkova, M., Dimitrov, D., Velikov, B., Pentchev, P., 1999. Groundwater problems in the
karstic aquifers of the Dobrich region, northeastern Bulgaria. Hydrol. Sci. J. 44 (6), 913–927.
Salem, O., Visser, J.H., Dray, M., Gonfiantini, R., 1980. Environmental isotopes used in hydrogeological study of north-eastern Brazil. Arid-Zone Hydrology:
Investigations With Isotope Techniques, Panel Proceedings Series (1980) 134.
Sarkar, A., Shekhar, S., Rai, S.P., 2017. Assessment of the spatial and temporal hydrochemical facies variation in the flood plains of North-West Delhi using integrated
approach. Environ. Earth Sci. 76 (19), 665.
Schlosser, P., Stute, M., Sonntag, C., Münnich, K.O., 1989. Tritiogenic 3He in shallow groundwater. Earth Planet. Sci. Lett. 94 (3-4), 245–256.
Solomon, D.K., 2000. 4He in groundwater. Environmental Tracers in Subsurface Hydrology. Springer US, pp. 425–439.
Stoyanov, N., 2019. Mathematical Modeling in Hydrogeology. Numerical 3D Models Using Finite Difference Method. Publishing house, “V. Nedkov”, Sofia, pp. 246 p
(in Bulgarian).
Stute, M., Deak, J., 1989. Environmental isotope study (14 C, 13 C, 18 O, D, noble gases) on deep groundwater circulation systems in Hungary with reference to
paleoclimate. Radiocarbon 31 (3), 902–918.
Stute, M., Schlosser, P., 1993. Principles and applications of the noble gas paleothermometer. AGU monograph on climate change in continental isotopic, records.
Geophys. Monogr. 78, 89–100.
Varlam, C., Stefanescu, I., Duliu, O.G., Faurescu, I., Bogdan, D., Soare, A., 2013. Tritium level in Romanian precipitation. Isotopes in hydrology. Marine Ecosystems
and Climate Change Studies. Vol. I. Proceedings of an International Symposium. pp. 139–146.
Varsányi, I., Palcsu, L., Kovács, L.O., 2011. Groundwater flow system as an archive of palaeotemperature: noble gas, radiocarbon, stable isotope and geochemical study
in the Pannonian Basin, Hungary. Appl. Geochem. 26 (2011), 91–104.
Yaneva, M., Ognjanova-Rumenova, N., 2012. Momin brod and Rasovo members - new lithostratigraphic units of brusartsi formation (late miocene-early pliocene), lom
coal basin, NW Bulgaria. Comptes Rendus de L’Académie Bulgare des Science 65 (9), 1239–1248.
Yovchev, R., Ryzhova, E., 1962. Groundwater in Northern Bulgaria. Main Directorate for Geology and Subsoil Protection, Sofia, Bulgaria (in Russian).

15