Distribution of radiocarbon in sediments of the cooling pond of RBMK type Ignalina Nuclear Power Plant in Lithuania

The vertical distribution of radiocarbon (14C) was examined in the bottom sediment core, taken from Lake Drūkšiai, which has served as a cooling pond since 1983 for the 26 years of the Ignalina Nuclear Power Plant (INPP) operation using two RBMK-1500 reactors (Russian acronym for”Channelized Large Power Reactor”). 14C specific activity was measured in alkali-soluble and -insoluble fractions of the sediment layers. Complementary measurements of the 210Pb and 137Cs activity of the samples provided the possibility to evaluate the date of every layer formation, covering the 1947–2013 period. In addition, 14C distribution was examined in the scales of pelagic fish caught between 1980 and 2012. Our measurements reveal that, during the period 1947–1999, the radiocarbon specific activity in both fractions exhibits a parallel course with a difference of 5 ± 1 pMC (percent of modern carbon) being higher in alkali-soluble fraction, although 14C specific activity in both fractions increased by 11.4–13.6 pMC during the first 15 years of plant operation. However, during the 2000–2009 period, other than previously seen, a dissolved inorganic carbon (DIC) → aquatic primary producers → sediments 14C incorporation pattern occurred, as the radiocarbon specific activity difference between alkali-soluble and -insoluble fractions reached 94, 25, and 20 pMC in 2000, 2006, and 2008, respectively. Measurements in different sediment fractions allowed us to identify the unexpected organic nature of 14C contained in liquid effluences from the INPP in 2000–2009. The discrepancy between 14C specific activity in fish scales samples and DIC after 2000 also confirmed the possibility of organic 14C contamination. Possible reasons for this phenomenon might be industrial processes introduced at the INPP, such as the start of operation of the cementation facility for spent ion exchange resins, decontamination procedures, and various maintenance activities of reactor aging systems and equipment.


Introduction
Lithuania near the borders with Belarus and Latvia and is the biggest lake in Lithuania (Fig 1). Lake Drūkšiai is a flow-through lake with 11 small streams flowing in and one stream flowing out which has a water-level regulating dam. The lake is characterized by a relatively slow (3-4 y) water exchange rate and a high areal diversity of bottom sediments. For 26 years, Lake Drūkšiai served as a cooling basin for the INPP. The nutrient load from the town of Visaginas (founded for workers of the INPP) and the increase in water surface temperature, altering vertical thermal stratification, have caused changes in the trophic state of the lake from (oligo) mesotrophic to almost eutrophic [30]. In order to recover the key environmental changes that occurred in the lake and its environs, the sediment sampling site (Sampling Station No. 1, 55 o 38'49" N; 26 o 35'07" E) of this study was selected at 29 m depth and was close to the deepest depression of Lake Drūkšiai (Fig 1). The 62 cm length sediment core was collected in December 2013 from ice surface using a Kajak gravity corer with acrylic sample tube. The core was sectioned in 1 cm slices in situ using piston rod immediately after sampling. Samples in containers were transported to the laboratory and stored at -40˚C until analysis. The characteristic for this lake depression are the highest recent sedimentation rate and finest sediments classified as muddy clay based on LOI and grain size analysis [31,32]. The clastic mineral material was dominating in these sediments with a concentration of carbonates and biogenic material ranging within 8.2-21.1% and within 19.6-32.6%, respectively [32].

Cs, 210 Pb, and 214 Pb measurements in sediment samples
Sediment chronology for the last 66 years was determined using the 210 Pb technique supported by anthropogenic 137 Cs as a chronostratigraphic marker. Samples were assessed by gamma-ray The sediment core site (station No 1) and water sampling site (station No 6) are marked by a circle and a triangle, respectively. The hydroelectric power plant (HEPP) and the dam on Drūkša river are marked by black bars 1 and 2, respectively. WWTP indicates location of the waste water treatment plant situated nearby a small lake connected to Lake Drūkšiai by Gulbinėlė rivulet. Industrial rain drainage system of Visaginas town via Smalva and Gulbinėlė streams is connected to Lake Drūkšiai. The short-dashed line means local railway route from regional line to INPP site.  Pb  and 214 Bi at 295, 352, and 609 keV), and 137 Cs (at 661 keV) isotopes were measured simultaneously using an HPGe well-type GWL-120-15-LBAWT detector (resolution 2.25 keV at 1.33 MeV) at the Nature Research Centre (Vilnius). Measurements were carried out in standard geometry, with 3 mL vials fitting to dimensions of the well-type detector. The mass of the samples was usually in the range of 2 to 4 g, but for a few top very fluffy samples it was approx. 1 g. The gamma spectrometric system was calibrated for counting efficiency using commercially available multi radionuclides standard sources and reference solutions for different densities and filing heights with the selected matrix as described in detail [31]. Uncertainties (±2 σ) of the radionuclide activity in the samples and the quality assurance procedures were evaluated by the GammaVision-32 software. The detection limit for the counting time of 200,000s was about 0.014 Bq for 137 Cs, 0.065 Bq for 210 Pb and 0.021 Bq for 214 Pb, while the measurement errors did not exceed 8%, 15%, and 20% for 137 Cs, 210 Pb, and 214 Pb, respectively.
The constant rate of the 210 Pb supply (CRS) model [32][33][34][35] with some modifications [9] was used. The initial data on the activity concentration of 210 Pb (total), 214 Pb, and 137 Cs in the sediment core versus depth with 1 cm resolution are given in S1 Fig. The specific activity of the total 210 Pb exponentially decreased from 360-460 Bq/kg in the 0-5 cm depth interval of the sediment core to 100-115 Bq/kg at a depth of 55-58 cm, with positive and negative 210 Pb deviations from exponential function. These small changes could be induced by slightly variable sediment mass accumulation rate (SMAR) and consequently more pronounced 210 Pb dilution for intervals with higher SMAR values and 210 Pb concentrating effect for intervals with lower SMAR values (S1 Fig). These deviations were accounted in calculation of partial sedimentation rates for certain depth (S2 Fig).
Below the depth of 58 cm, the specific activity of 210 Pb within the limits of uncertainties was close to the maximal level of 214 Pb in sediments, which corresponds to the supported 210 Pb level. Based on the conventional statistics ( 210 Pb average value in sediment core plus 3 standard deviations), the maximal level of 214 Pb in sediments was evaluated to be 85 Bq/kg.
The chronology based on 210 Pb for the second half of the 20th century was validated by using data on the artificial radionuclide 137 Cs fallouts from the atmosphere due to the Chernobyl NPP accident and nuclear weapons testing in the atmosphere, as independent chronostratigraphic markers. The depth profile of the 137 Cs activity in sediments was in good agreement with the 210 Pb CRS dates (S3 Fig). Two peaks of 137 Cs at the sediment depths of 49 and 33 cm corresponded to the 210 Pb dates of~1963.2 ± 2.2 and~1986.1 ± 1.0 AD. Below the 49 cm depth, 137 Cs activity concentration in sediments rapidly decreased. At the bottom of the core (59-60 cm depth slice), it was 7.4 ± 2.8 Bq/kg. Below the depth of 58 cm, which corresponds to the calendar date of 1950 ± 10 AD, the 210 Pb chronology is uncertain.
Details on 210 Pb and 137 Cs activity profiles as well as the age-depth model and sedimentation rate data are given in the supporting information (S1 Text, S1 and S2 Figs).

Sample pretreatment and measurements
Radiocarbon specific activity measurements were performed in two organic fractions of the lake sediment: alkali-soluble and organic matter that is not soluble in alkali. In agreement with Kleber and Lehmann's argument that alkaline extraction cannot separate humic and nonhumic substances [26], we decided not to use the terms of humus, humic acids, and other subcategories of humic substances.
Alkali-soluble and alkali-insoluble organic fractions in sediment samples were extracted using the acid-base-acid (ABA) method, including sequential washes with 1 M HCl (overnight), 0.2 M NaOH (for 1 h), and 1 M HCl (for 1 h) to yield the alkali-insoluble fraction [36]. The Coregonus albula scales (S1 Table), obtained from sets of 5 mature fishes (age of the third year (2+ y)) caught in summer in 1980-1999 and 2005-2012, were decalcified by immersing them into 1.2 N HCl for 2 min [37]. In doing research for this work, we did not work with living organisms. We measured 14 C specific activity in fish samples already collected (8-30 years ago) and stored at -40 o C until analysis. All samples were graphitized using Automated Graphitization Equipment AGE-3 (IonPlus AG) prior measurements with a 250 kV single stage accelerator mass spectrometer (SSAMS, NEC, USA) at the Center for Physical Sciences and Technology in Vilnius. The background of measurements was estimated to be 2.45 × 10 −3 f M (fraction of modern carbon) using phthalic acid (Merck, cat. #822298). The IAEA-C3 standard was used as a reference material (the percent of a modern carbon (pMC) value of 129.41). The 14 C/ 12 C ratio was measured with an accuracy better than 0.3%.
The measurements are reported in units of pMC [38,39]: where 14 a s is 14 C specific activity, A SN is the specific activity of the sample A S , normalized to δ 13 C = −25‰; A ON is the normalized specific activity of the standard. The freshwater reservoir effect (FRE) is defined as the difference between the radiocarbon isotope ratio ( 14 C/ 12 C) in the terrestrial and freshwater bodies. The radiocarbon reservoir age (RRA) was calculated according to the following equation [40]: where 14 a S T and 14 a S A is 14 C specific activity in the terrestrial and aquatic samples of the same period, respectively. As 14 a S T , for the period before 1957, we used atmospheric 14 C specific activity data from OxCal v4.2.4 [41,42]; for the period of 1957-2013, we used 14 C specific activity measurements in P. sylvestris from the background location and from the surroundings of INPP [17].
The total nitrogen (TN) (%) and total organic carbon (TOC) (%) content (for TOC/TN or so-called C/N ratio determination) in the sediments were measured using an elemental analyzer (Thermo Flash EA 1112). Standards sulphanilamide (Merck, cat. #111799) and nicotinamide (Sigma Aldrich, cat. #72345) were used for calibration. The long-term standard measurements were performed with a precision of <1.1% for TOC and <0.6% for TN. Before analysis, the sediment samples were acidified with 1 N HCl to remove carbonates, were then rinsed in de-ionized water to a neutral pH, and re-dried.

Radiocarbon specific activity in sediments
Corresponding results for both organic fractions are shown in Fig 2. 14 C specific activity in the alkali-insoluble organic fraction ranged from 78.8 to 109.5 pMC, whereas 14

Pre-operational period (1947-1983)
In 1953, the hydroelectric power plant (HEPP) was built on the Prorva River, the only outflow from Lake Drūkšiai. This resulted in an increase of the lake water level by 0.3 m. In the same year, the entire Apyvardė River flow was directed to Lake Drūkšiai by damming the Drūkša River downstream the Apyvardė River (Fig 1). As a result, the Lake Drūkšiai catchment area increased by 24% [43]. This event was one of the reasons for the increase in the sediment mass accumulation rate (S2 Fig). However, the increased water level did not result in significant changes in water and atmospheric CO 2 exchange rates, since 14 C specific activity was unchanged in both sediment fractions (in average 85.54 ± 0.58 pMC and 79.34 ± 0.81 pMC in alkali-soluble and -insoluble fractions, respectively) from 1947 until the beginning of the nuclear weapons tests. Nuclear weapons tests caused an increase in 14C specific activity in both sediment fractions by 19.9 pMC. Whereas in the atmosphere, 14 C specific activity increased by 95 pMC. This means that atmospheric carbon contributes only about 20% of the total carbon accumulated in the sediments. This also led to an increase in RRA in both fractions by 2850 ± 82 y. The RRA values before 1955 were 1092 ± 84 y and 1913 ± 82 y in alkalisoluble and -insoluble fractions, respectively.

Early operational period (1984-1999)
The INPP started its operational activity in December 1983. Accordingly, the radiocarbon specific activity in both fractions started to increase.
From 1947 to 1999, the radiocarbon specific activity in both fractions exhibits the parallel course, only 14 C concentration in the alkali-soluble fraction being 5 ± 1 pMC higher than in the alkali-insoluble fraction.
The  terrestrial origin organic matter contribution to the sediments did not exceed 16-22% (calculated assuming that the C/N ratio was 6.6 and 40 in phytoplankton and vascular plants, respectively) [44][45][46][47].
The DIC concentration and the isotope composition depend on various factors affecting lake water and atmospheric CO 2 exchange rates, such as organic and inorganic carbon transportation [48,49], the rates of organic carbon production, mineralization, and carbonate weathering and dissolution [50][51][52][53][54].
In 1982, the HEPP ceased electricity production, but the dam was still used for maintaining a high level of lake water required for cooling the INPP; i.e., the regime of watershed hydrology of Lake Drūkšiai basically remained unchanged. The mean water level fluctuation amplitude was kept at~0.2-0.6 m to maintain safety of water use and preservation regulations, stating that the annual lake water level fluctuation amplitude should not exceed 1.2 m in order to reduce the hazards of shore abrasion [43]. This means that changes in the 14 C specific activity in lake objects were not caused by the lake watershed hydrology affecting organic and inorganic matter transportation and water residence time [8,55,56]. 14 C specific activity measurements in DIC in Lake Drūkšiai were performed once or twice a year from 1980 to 2009 [29]. Until 2000, the trend of 14 C specific activity in DIC [29] coincided with the corresponding year radiocarbon measurement in tree rings, sampled from the P. sylvestris trees in the vicinity of the INPP (see Fig 3) [27].
Measurements showed that, until 2000, the values of 14 C specific activity in the C. albula scales within ±3 pMC corresponded to the values of 14 C specific activity in Lake Drūkšiai DIC (Fig 3A). It was proved in earlier investigations that the C. albula feeds exclusively on zooplankton [57,58]. Thus, the bioaccumulation of 14 C in higher trophic levels through a relatively short food chain (DIC ! aquatic primary producers ! zooplankton ! C. albula) is expected. The existence of such a semi-separate pelagic food chain was revealed by previous stable isotopes studies in Lake Drūkšiai [59]. The age of the caught C. albula was~2 y. This mean that the C. albula scales reflect averaged over~2 y 14 C specific activity values in the DIC of the lake water for this period.
Therefore, it can be stated that, until 2000, the main source of 14 C pollution was in DIC form. As can be seen in Fig 3A, both sediment fractions followed the 14 C specific activity change in the DIC and in fish scales. No changes in the distribution of radiocarbon in both fractions were also observed during this period.
The RRA in both fractions regained its former pre-bomb values at the beginning of the INPP operation in 1986-1987. From the start of the INPP operation up to 1995, RRA values in sediment fractions decreased by 537 ± 79 y. However, from 1995 to 1999, the RRA in both factions increased by 540 ± 63 y. In 1998-1999, an increase in 14 C specific activity (up to 10 pMC) was also observed in the tree samples from the vicinity of the INPP [27] (Fig 2). Enhanced maintenance activities are characteristic for the period 1998-2003. During this period, 245 fuel channel zirconium-niobium alloy tubes were extracted from the reactor cores and replaced by the new ones. During the replacement procedure the reactor cavity containing the graphite stack is opened and increased releases of 14 CO 2 are expected through the ventilation system of the plant.
Radiocarbon specific activity characteristic to Lake Drūkšiai with the assumption of no INPP impact was calculated for both sediment fractions (Eq 2) supposing that the RRA in both fractions had pre-bomb values (Fig 3B). Thus, during 1984-1999, excess of 14 C specific activity in both fractions reached 11.4-13.6 pMC. A similar increase in 14 C specific activity during this period should have been in DIC, as aquatic plants assimilating DIC are the main sedimentary organic source. No measurement data were collected for the period before 1980, so we can only determine the increase in 14 C specific activity in DIC after the INPP started to operate.
Characteristic DIC concentration value in the lake used to be~2 mM. A rough estimate showed that, during this period, the pollution from the INPP to Lake Drūkšiai used to be up to 0.35-0.41 GBq of 14 C per year. It includes both airborne and DIC discharges to the lake from the INPP. Assuming a 20% contribution of the airborne pathway for that period would only be a rough approximation due to changes in the CO 2 exchange rate between the air and lake water caused mainly by the industrial impact of the INPP. During the period of operation for both reactors (i. e. 1987-2004), the water surface temperature in some parts of the lake increased to 32-35 o C during the summer months because of the thermal load [60]. Due to the  14 C specific activity in the alkali-soluble and alkali-insoluble sediment fractions, in the scales of C. Albula and in DIC (data from [29]) during the period of INPP operation as well as radiocarbon specific activity that would have occurred without any INPP impact (data was obtained by Eq 2, assuming the constant RRA value until the bomb peak; 14 C specific activity data for terrestrial samples were taken measurements in P. sylvestris from the background location (BAGR)). (B) Evaluated INPP impact in terms of excess of 14 C specific activity (in pMC) in the alkali-soluble and -insoluble sedimentary organic fractions.
https://doi.org/10.1371/journal.pone.0237605.g003 elevated temperatures of the lake combined with excessive pollution by nitrogen and phosphorus delivered with the sewage water of the town of Visaginas, the trophic state of the lake has changed from (oligo) mesotrophic to almost eutrophic [30].

Late operational period (2000-2013)
The period from 2000 to 2013 is distinctive by a considerable increase of 14 C specific activity in alkali-soluble fraction. During the period from 2000 to 2002 (Figs 2 and 3A, S1 Table), a sharp increase of 14 C specific activity by~80 pMC was observed only in the alkali-soluble organic fraction, whereas a substantially smaller increase (by~4 pMC) of radiocarbon content can be seen in the alkali-insoluble fraction. During this period, radiocarbon specific activity values increased in DIC by 25 pMC. It must be taken into account that water samples were previously taken only once or twice a year, mostly at Sampling Station No. 6 (55˚34' 33" N, 26˚37' 15" E, Fig 1). The results of the measurements in DIC show 14  In fish scale samples taken in 2005-2012, 14 C specific activity corresponded well to radiocarbon change patterns in the alkali-soluble fraction for that period, but it was quite different from the values of 14 C specific activity in DIC. Samples from the 1985-1999 period showed ( Fig 3A) 14 C specific activity differences in fish samples (as well as in DIC) and alkali-soluble fractions up to 12.7 ± 2.1 pMC (and 10.6 ± 3.5 pMC if DIC is included when averaging). This indicates that, during the 2000-2009 period, there were active 14 C introduction processes other than DIC ! aquatic primary producers ! sediment fractions. In case the main source of 14 C pollution was DIC, both sediment fractions would follow the DIC 14 C specific activity trend.
The most plausible explanation for this phenomenon could be that, during this period, the INPP discharged elevated levels of 14 C incorporated into some specific water-soluble organic compounds used in the technological processes of NPP operation and maintenance. This additional 14 C may have been transmitted to the alkali-soluble sediment organic fraction. Earlier investigations showed that low molecular weight (<3.5 kDa) xenobiotics [62,63] and their metabolites [50] can be easily taken up and bioconcentrated by freshwater plants and organisms. Dietary uptake contributed little (up to a few percent) to overall uptake [64]. This may have been the reason for the same 14 (Figs 2 and 3A), the sources of 14 C were of a different origin and chemical nature. The unusual trends of 14 C specific activity in the bottom sediments reveal new paths of 14 C ingress into the lake ecosystem, which, compared to cases of routine operation, are much more likely to occur from the zones of accumulation of the technological chains of the INPP under decommissioning. It has to be noted that, in the late operational period, new processes of waste management, treatment, and deactivation have been introduced at the INPP.

Conclusions
Our study of radiocarbon distribution in organic alkali-soluble and alkali-insoluble sediment fractions of Lake Drūkšiai covers a period of 66 years . We analysed the new 14 C specific activity data obtained from the measurements of sediment core as well as in C. albula scale samples. We investigated how the radiocarbon distribution in this fish and DIC coincide due to expected DIC ! aquatic primary producers ! zooplankton ! C. albula 14 C accumulation chain. The data of radiocarbon specific activity in the tree rings of the INPP vicinity are also included in our analysis. This allowed establishing a full picture of 14 C specific activity changes over several decades in the main indicators of INPP environment. The slight increase in water level of Lake Drūkšiai caused by the hydroelectric power plant constructed in 1953 did not appear to have a significant effect on 14 C distribution in sediments. C/N measurements show that the main contribution (~78-84%) of organic fraction in sediments throughout the study period is of autochthonous origin. Thus, both before the commissioning of the INPP and during the first 15 years of its operation (until 2000) with minor maintenance activities, the radiocarbon specific activity in both fractions of bottom sediments exhibits the parallel course, and undisturbed chain DIC ! aquatic primary producers ! sediments governed migration of radiocarbon in Lake Drūkšiai sediments. During this period the 14 C specific activity values in C. albula scales and DIC samples were in a good agreement. 14 C redistribution observed in the last 11 years of operation of the INPP in both organic sediment fractions indicates that another 14 C accumulation pathway has taken place. This might be explained by the new processes of waste management and treatment implemented at the INPP in late operational period before the final shutdown of Unit 1 (end of 2004) and Unit 2 (end of 2009).
The gaseous discharges from the INPP with RBMK type reactors are mainly carbon dioxide containing 14 C. Although systematic monitoring of 14 C airborne releases started in 2008 only, it can be relatively well reproduced by 14 C measurements in tree rings near the INPP. Routine monitoring of radiocarbon in liquid releases was not performed, but it is expected that radiocarbon releases from RBMK type reactors is in dissolved inorganic carbon form. However, complex investigations of the radiocarbon specific activity in alkali-soluble and -insoluble fractions of sediment core layers indicate changes in 14 C redistributions between sediment fractions as well as discrepancies in radiocarbon specific activity values in the C. albula scales and DIC. This can retrospectively reveal facts of occurrence of elevated liquid radiocarbon releases, not only in DIC but also in dissolved organic compounds which might have been used in processes of maintenance and decontamination of technological circuits of the plant.  Table. 210 Pb, 214 Pb and 137 Cs activity data; age of the sediment layers, 14 C specific activity in sediment Alkali-Soluble (AS) and Alkali-Insoluble (AIS) sediment organic fractions as well as in the C. albula scales; total length of the C. albula (TL). Data is shown as mean ± standard deviation. (PDF) S1 Text. 210 Pb and 137 Cs sediment dating and sedimentation rate determination. (PDF)