A marked transition of Pb concentration (3.8 times increase) and isotopic composition (206Pb/207Pb from 1.21 to 1.18) was detected in all but the B5 core upon reviewing the core profiles (Fig. 3–4). Data collected shall be handled with care since only the upper part of the sediment (called phase-2) represents material deposited on a known time scale. This is the fraction that was deposited after the deep dredging of the sludge trap in 1979. Phase-1 sediment is much older than 40 years since it is known (Kutics 2021) that deep dredging affected 200 cm of sediment, leaving a bed that is expected to be older than 150–200 years. Many experts showed significantly altered Pb composition due to anthropogenic effects as early as 2000 years ago (Kamenov et al. 2020, Brannvall et al. 2001, Kylander 2005), but in the case of Balaton, it is expected that in the era preceding the industrial revolution, the contribution of lead from distant mining fields to the lake deposit was low, natural clastic or authigenic Pb deposition prevailed. Dolomites and other carbonate minerals dominant in the catchment area of Balaton, thus determining the natural sediment composition can contain amounts of Pb that are sufficient to explain Pb concentration of lower sediment phase. Lake Balaton has a large water capture area and a significant proportion of its deposit is derived from erosion through e.g. Zala River and Tapolca torrent. It is not ombrotrophic like many waters analyzed by Brannvall et al. (2005). Fluctuations in the proportion of inputs could also potentially explain minor shifts in isotopic composition of lead, but unlikely to cause the large differences presented in this paper. Increase of concentrations of Pb and other pollutants also point towards an anthropogenic source for the Pb profiles measured.
B5 sediment core, from the margin of the sludge trap, was distinct from the other cores as it did not show transition of Pb isotopic signatures (Fig. 4a), in contrast to that, it had a signature closer to deep core phases showing only a week trend from 1.206 towards lower values (1.198, R2 of fitting is 0.559). We concluded that B5 site did not accumulate significant amount of sediment, instead it was slightly eroded after 1979 deep dredging and affected by the 1999 thin layer (10-20cm layer) dredging, the surface of this site is believed to be between 50 and 100 years old. This suggests that the transition of Pb isotope ratios has happened in the early-mid 20th century, corresponding to exponential growth of industrialization at that era.
We compared Pb isotopic compositions with data gathered by Bölhoffer and Rosman (2000, 2001) and Hopper et al. (1991), and other papers (Fig. 7). on Pb isotopic composition of aerosols, fuel samples and ore samples in the scope of the prevailing wind direction that is from northwest. The results of the comparison are summarized in Fig. 7. The distribution of values of 206Pb/207Pb for many Central European countries is in the range 1.13–1.17. By supposing that Pb in the deep core is the natural background and the average concentration of Pb in phase-2 sample is the sum of the background and pollutant Pb, we calculated the expected isotope ratio of pollution to be 1.1775 ± 0.003 based on Eq. 1. This equation is the rearranged form of equation used by Kylander et al. (2005) for estimating the contribution of pollutant lead. This value is distinct from the data presented by Bölhoffer and Rosman (2001) for central Europe, but Hopper et al showed a value of 1.174 ± for 206Pb/207Pb in the 1988 leaded fuel. They have found leaded fuel 206Pb/207Pb ratios of 1.072 for Hungary that is closer to the Vienna leaded fuel 1.111 but also to the Broken Hill signature 1.039 (Richards, 1986, Reynolds, 1971) than to the expected pollutant lead isotopic composition. It must be mentioned that although the Hungarian leaded fuel has 206Pb/207Pb value distinct from the expected value for pollutant, and also 208Pb/207Pb values in the leaded fuel from Poland was 2.418 that is dissimilar compared to phase-2 samples 2.460 ± 0,003 and the expected value of pollutant lead 2.456 ± 0,004, two fact have to be pointed.
1: The values presented by Hopper (1990) for Hungary are measured in 1990 when Hungary was out of the scope of Warsaw Convention, becoming an open market for western companies, the samples from Poland on the other hand were collected in 1988 when it was part of Eastern Block.
2: Validity of Hopper’s data may be questionable based on the fact that they measured isotope ratio and concentration simultaneously, meaning no concentration matching before isotope ratio measurement was performed, also, they do not present the detector used, and whether or not they have performed dead time correction. 208Pb/207Pb values in a sample whose Pb concentration is multiple times higher than the external bracket standard can be underestimated if proper dead time calculation is omitted, but this effect will be totally absent for samples with concentration matching the standard solution, and this effect is also missing when 206Pb and 207Pb are measured since the abundance of these two is highly similar (our unpresented laboratory experience).
The differences between the literature data for leaded fuel samples and data presented here for phase-2 samples do not support the idea that leaded fuels are the main anthropogenic contributors to the Balaton sediment lead isotope signature. Data retrieved from the literature for different lead bearing ores show 206Pb/207Pb 1.179 (Russell and Farquahar, 1960) and 1.171 (Binczycki et al. 2014) with respective 208Pb/207Pb values 2.469 ± 0.003 and 2.456 ± 0.003 show great similarity to our data, suggesting that this signature is linked lead mining and subsequent utilization of Pb from these sources. Aerosol formed during mining and smelting of Pb does have the ability to travel long many thousand miles distances as seen in more than 2000 years old deposits in ombrotrophic peats and lakes in Spain and Sweden (Brannval et al. 2001 and Kylander et al. 2005). Lead was used for producing dyes and acid-lead batteries. Both of these industrial processes and the inadequate storage of waste containing these products may give rise to Pb emission. Lead is also an alloying material in steel manufacturing. Wear of machines produced from such alloy may also give rise to Pb emission. Leaded fuel Pb isotopic signatures in the literature suggest that ores from Harz Mountains and Southern Poland were not widely used for production of leaded fuel. Considering the geography and the prevailing wind directions in the Central European area encompassing Southern Poland, the Czech Republic, Austria and Hungary does support a model where lead from mines northwest to Krakow could reach Lake Balaton. In Second World War, there were heavy fights in Western Hungary. Considering lead containing projectiles, they can also account for lead burden significantly (Broomandi et al. 2020).
We recorded the change in Pb isotopic composition seeking for changes in the last decades. Pb isotopic composition is shifted between the upper new deposit and the bottom of the trench. In the newly formed deposit, we expected at least a gradual shift towards the uncontaminated bottom. To our surprise, such change was absent (Fig. 6). This is in contrast with the findings in USA regarding Great Lakes and lakes in Florida (Graney et al. 1995, Escobar et al. 2013). Where researchers suggested a shift in Pb isotopic composition towards the geologic Pb, this was accompanied by a decrease in Pb concentration due to slow resolution of pollution. They also showed changing Pb isotope ratio fingerprint for the pollutant tetraethyl lead due to a change in the lead ore used for the production of the fuel additive. Although levels of lead did not reach dangerous levels in Lake Balaton, the unchanging Pb isotopic composition in the phase-2 of cores suggests that Pb contamination in the catchment area can persist for extremely long times.
The most likely cause of such persistence is dust from urban area and agricultural soils (supported by our unpresented results showing that Hungarian soils of Pb pollution). Findings of Resongles et al. (2022) showed that Pb signatures in London urban area did not change significantly after drastic decrease of Pb emission, they predict that lead concentrations cannot be expected to drop any further in the near future. Persistence of Pb in agricultural soils are also shown by many other papers (Wang et al. 2021). Dust produced during ploughing may release Pb containing dust to atmosphere (Sharrat and Auvermann 2014) that can reach and contaminate Lake Balaton due to prevailing winds from northwest (Mezősi et al. 2015).
Our leaching experiment showed that extractability of Pb shows correlation with the extraction of Fe suggesting its association to Fe/Mn hydroxide particles, and not to carbonaceous fraction. 1M HCl leaching extract of phase-1 samples was characteristic of high concentration of Fe and Pb, while phase-2 showed low extraction yield for both Fe and Pb. Concentration of Pb in latter is 50 times lower than in the leachate of phase-1. This result is strikingly different from microwave digested (complete digestion) samples where phase-2 contained 4 times more Pb than phase-1.
The area having the highest amount of deposit 110-130cm (sediment depth of 206Pb/207Pb phase transition) was the western margin of the sediment trap, cores B11-B15. This area corresponds to the direction from which the predominant stream arrived to the sediment trap, there was a slightly elevated sediment thickness in the B1 and B6 (108-110cm, northern margin). The rest of the area showed somewhat smaller deposition rates, 80–96 cm. This shows that careful analysis of stream directions in the water body is of high importance in achieving the highest level of efficiency, while keeping the extent of the dredged area at a minimum level (Fig. 5b). If instead of a 100 ± 10 m trench perpendicular to the direction of dominant stream, two trenches of 40–50 m were created, the efficiency would grow by 10–15%, while the same area would be affected.
Transition of Pb isotopic composition in the phase limit is not always abrupt. There are peak like changes and/or smoother ramps in some of the cores (Fig. 4a). Peaks belong to occasional turbulences and sediment overturn, while smooher ramps to thorough mixing. We labelled the three main types of cores with types 1–3, where 1 means abrupt change in Pb isotopic composition, 2 means peak like changes or smooth transition or occasionally both, while 3 is B5 that was taken outside of the sediment trap. Type 1 is characteristic of the western margin of the trench and shows less disturbance. We conclude that the area covering the first few meters from the entry point of dominant water stream will have higher accumulation rate and can preserve pollution records better. The middle of the sludge trap shows characteristics of sediment inversion suggesting higher level of mixing and remobilisation (Fig. 5c).
Concentration of many elements that are traditionally associated with agricultural and heavy industrial activities showed correlation with the Pb isotopic signature, being elevated in the upper (phase-2) part (Fig. 7). Many authors (Nriagu and Pacyna 1988, Kemp et al. 2012) attribute Sb to the extent of emission of road vehicles, thus expecting higher concentration values for the areas where fuel additive derived Pb contamination is present. We found slightly elevated Sb in phase-2 compared to phase-1. Fe concentration was slightly elevated concomitantly with Pb. The major source of iron is probably dust associated with agricultural activities, but other industrial activities can also be significant contributors. Deposition of iron is sensitive to different chemical environments. Iron and manganese hydroxides that form during precipitation provide high adsorbent surface that can aid the deposition of other metal elements to a large extent (Boyle 1995).
The increase of Zn, Cu and Cd concentration in phase-2 compared to phase-1 was significant. Since a discontinuity exists in the cores due to dredging, the correlation of these elements with the Pb concentration needs to be regarded carefully, they do not represent a real causative connection, but it is worth noting that expansion of agricultural, urban and industrial areas is in most cases associated with elevated concentrations or, in some cases, toxic levels of these elements (Wang et al. 2021). It is important to note at this point that none of the examined elements reached hazardous level in any of the samples, but the changes were significant.
Prominent changes in 234U/235U composition was seen in the four sediment records examined for U isotopic composition, a peak like transient change is seen in the profilae roughly overlapping with the inflexion of Pb isotope ratio profile (Fig. 9). U in these peaks is not enriched or depleted, only the proportion of isotope 234U is varying considerably. Ratio of 235U and 238U is unaltered. The U and Mo peaks emerged at the time of deep dredging and slightly afterwards. This is supported by the fact that dredging is an invasive intervention leading to large amounts of suspended matter that can fall back to the lake bed (Zhang et al. 2010, Van der Berg et al. 2001).
U can be immobilized by bacteria by bio-sorption or reduction under anaerobic circumstances (Kalin et al. 2005, Anderson and Fleischer 1991, Mann and Fyfe 1985). Such circumstances can arise in the aftermath of algal and cyanobacterial blooms as a result of increased organic matter load provided by the mass of dead cells (Rastogi et al. 2015). From the 1970s until the early 1990s, overproduction of N fixing cyanobacteria Cylindrospermopsis raciboskii was significant, leading to water blooms (Pálffy 2013). There is evidence that molybdenum is accumulated in algal cells and can get trapped in the sediment during water blooming, therefore an increased biomass production in the lake may increase the level of Mo in the sediment. (Glass et al. 2008). U isotopic composition, as well as its concentration together with Mo might be a retrospective marker of water trophity and algal blooms. Dredged material is likely to have composed of such sludge containing high amounts of organic matter and having reductive nature. The fall out after dredging may be the source of the characteristic U and Mo peaks. The source of background U is likely to be lithogenic as carbonaceous materials may contain U (Nyachoti et al. 2017), karst water in the Transdanubian Mountains does indeed contain U, as shown in many karst waters in Hungary (Jobbágy et al. 2010).
Trends of Sr isotopic compositions in cores that have 2 separate phases, in terms of Pb, shows a gradual increase in the bottom core and a reversal at the top of the core (Fig. 10). Although data presented are not significant per se but considering the presence of the same trend in two cores suggests that the changes are real. Besides motor vehicle usage and industrial emission, there was another important anthropogenic factor at the presented time scale, namely the massive extraction of karst water from the reservoirs at Transdanubian Mountains, associated with mining activity that was gaining momentum in the 1950s which lead to a massive drop in karst water level, and ended in the early 1990s, followed by a slow but more or less constant replenishment (Csepregi 2007). It is highly probable that slight changes in the Sr isotopic composition are at least partially attributable to the fluctuations of karst water level and thus the carbonate input to the Lake Balaton from that particular source. The small change in Sr isotopic composition is strikingly dissimilar to the massive changes in Pb isotopic composition, this further strengthens the view that elevation of Pb cannot be merely the result of changing geological environment.