A new generic sub-model for radionuclide fixation in large catchments from continuous and single-pulse fallouts, as used in a river model

This paper presents a new general sub-model for fixation in catchment areas to be used within the framework of a river model for substances such as radionuclides and metals from continuous and single-pulse fallouts. The model has been critically tested using data from 27 European river sites covering a very wide geographical area and contaminated by radiocesium and radiostrontium from the Chernobyl accident and from the nuclear weapons tests (NWT fallout). This modelling approach gives radionuclide concentrations in water (total, dissolved and particulate phases) at defined sites on a monthly basis. The overall river model is based on processes in the upstream river stretch and in the catchment area. The catchment area is differentiated into inflow (approximately dry land) areas and outflow (approximately wetland) areas. The model has a general structure, which can be used for all radionuclides or substances. It is simple to apply in practice since all driving variables may be readily accessed from maps and standard monitoring programs. The driving variables are: latitude, altitude, catchment area, mean annual precipitation and fallout. Note that for large catchments, this model does not require data on the characteristic soil type or the percentage of outflow areas (wet lands) in the catchment, as in most previous models, since in practice it is very difficult to obtain reliable data on characteristic soil type or percentage of outflow areas, especially in large and topographically complex catchments. Modelled values have been compared to empirical data from rivers sites covering a wide domain (catchment areas from 3000 to 3,000,000 km2, precipitation from 400 to 1700 mm/year; fallouts from 1600 to 280,000 Bq/m2; altitudes from 0 to 1000 m.a.s.l. and latitudes from 41 degrees to 72 degrees N). The river model with its sub-model for fixation predicts close to the uncertainty factors given by the empirical data, which have been shown to be about a factor of 1.6 for 137Cs and a factor of 2.2 for 90Sr in river water. The obtained characteristic uncertainty factors for 137Cs from the Chernobyl fallout is 2.4, for 137Cs from the NWT fallout it is 1.3 and for the 90Sr results from the NWT fallout it is 3 using the new model.     

A new general dynamic model predicting radionuclide concentrations and fluxes in coastal areas from readily accessible driving variables

This paper presents a general, process-based dynamic model for coastal areas for radionuclides (metals, organics and nutrients) from both single pulse fallout and continuous deposition. The model gives radionuclide concentrations in water (total, dissolved and particulate phases and concentrations in sediments and fish) for entire defined coastal areas. The model gives monthly variations. It accounts for inflow from tributaries, direct fallout to the coastal area, internal fluxes (sedimentation, resuspension, diffusion, burial, mixing and biouptake and retention in fish) and fluxes to and from the sea outside the defined coastal area and/or adjacent coastal areas. The fluxes of water and substances between the sea and the coastal area are differentiated into three categories of coast types: (i) areas where the water exchange is regulated by tidal effects; (ii) open coastal areas where the water exchange is regulated by coastal currents; and (iii) semi-enclosed archipelago coasts. The coastal model gives the fluxes to and from the following four abiotic compartments: surface water, deep water, ET areas (i.e., areas where fine sediment erosion and transport processes dominate the bottom dynamic conditions and resuspension appears) and A-areas (i.e., areas of continuous fine sediment accumulation). Criteria to define the boundaries for the given coastal area towards the sea, and to define whether a coastal area is open or closed are given in operational terms. The model is simple to apply since all driving variables may be readily accessed from maps and standard monitoring programs. The driving variables are: latitude, catchment area, mean annual precipitation, fallout and month of fallout and parameters expressing coastal size and form as determined from, e.g., digitized bathymetric maps using a GIS program. Selected results: the predictions of radionuclide concentrations in water and fish largely depend on two factors, the concentration in the sea outside the given coastal area and/or adjacent coastal areas and the ecological half-life of the radionuclide in the sea. Uncertainties in these factors generally dominate all other uncertainties, e.g., concerning the surface water retention time, the settling velocity of the particulate fraction, the distribution coefficient regulating the fluxes in dissolved and particulate phases, the catchment area influences and the factors regulating biouptake and excretion of the radionuclide in fish. This means that the conditions in the sea are of paramount importance for the conditions in the coastal area, even for relatively enclosed coastal areas. This coastal model may be regarded as a tool for testing working hypotheses on the relative roles of different processes in different coastal areas. Such information is essential for getting realistic expectations of various remedial measures, such as coastal dredging discussed in this work.     

Modelling the transport of radionuclides from land to water

This paper addresses some fundamental problems related to the structure and function of catchment areas in general and for seasonal (weekly, monthly) mass balance calculations of radionuclides (and metals, organics and nutrients) in particular. A new catchment area model has been developed and critically tested. This modelling approach is based on mechanistic principles. The catchment area is differentiated into inflow ( approximately dry land) areas and outflow (=wetland) areas. The model also accounts for time-dependent fixation of substances in the catchment related to single-pulse fallouts. The model has a general structure. It is simple to use since there are only two soil type classes and three categories for the outflow areas. In critical tests, the model was put within a framework where it is intended, i.e., within a more comprehensive lake model. Radiocesium was used as a tracer in these tests. Modelled values were compared to empirical data from 23 lakes (351 data on (137)Cs in water, as well as in sediments, small fish and on suspended particles) covering a very wide limnological domain (latitudes from 42 to 61 degrees V, altitudes from 0 to 1090 m ASL, catchment areas from 0.17 to 114,700 km(2), precipitation from 430 to 1840 mm/year, lake areas from 0.042 to 1147 km(2), mean depths from 1.1 to 90 m, pH from 5.1 to 9 units, potassium concentrations from 0.23 to 27.5 mg/l, total P concentrations from 8.3 to 100 microg/l and theoretical water retention times from 0.02 to 137 years). When modelled values were compared to empirical data, the slope was almost perfect (0.99) as well as the coefficient of determination (r2 = 0.96).     

A blind test of the MOIRA lake model for radiocesium for Lake Uruskul, Russia, contaminated by fallout from the Kyshtym accident in 1957

This paper presents results of a model-test carried out within the framework of the COMETES project (EU). The tested model is a new lake model for radiocesium to be used within the MOIRA decision support system (DSS; MOIRA and COMETES are acronyms for EU-projects). This model has previously been validated against independent data from many lakes covering a wide domain of lake characteristics and been demonstrated to yield excellent predictive power (see H?kanson, Modelling Radiocesium in Lakes and Coastal Areas. Kluwer, Dordrecht, 2000, 215 pp). However, the model has not been tested before for cases other than those related to the Chernobyl fallout in 1986, nor for lakes from this part of the world (Southern Urals) and nor for situations with such heavy fallout as this. The aims of this work were: (1) to carry out a blind test of the model for the case of continental Lake Uruskul, heavily contaminated with 90Sr and 137Cs as a result of the Kyshtym radiation accident (29 September 1957) in the Southern Urals, Russia, and (2) if these tests gave satisfactory results to reconstruct the radiocesium dynamics for fish, water and sediments in the lake. Can the model provide meaningful predictions in a situation such as this? The answer is yes, although there are reservations due to the scarcity of reliable empirical data. From the modelling calculations, it may be noted that the maximum levels of 137Cs in fish (here 400 g ww goldfish), water and sediments were about 100,000 Bq/kg ww, 600 Bq/l and 30,000 Bq/kg dw, respectively. The values in fish are comparable to or higher than the levels in fish in the cooling pond of the Chernobyl NPP. The model also predicts an interesting seasonal pattern in 137Cs levels in sediments. There is also a characteristic "three phase" development for the 137Cs levels in fish: first an initial stage when the 137Cs concentrations in fish approach a maximum value, then a phase with relatively short ecological half-lives followed by a final phase with long ecological half-lives more or less corresponding to the physical decay of radiocesium.     

Evaluating land cover change and its impact on hydrological regime in Upper Shire river catchment, Malawi

A study was conducted to investigate hydrological impacts of land cover changes in the degradation of the hydrological on flow regimes of the Upper Shire river, Malawi. Remote sensing techniques were used to inventory temporal changes of land cover changes in the catchment. Hydrological data were analyzed to reveal the alterations and trends for two periods; 1989 and 2002. The study revealed significant changes in magnitude and direction that have occurred in the catchment between 1989 and 2002, mainly in areas of human habitation. Trends in land cover change in the Upper Shire river catchment depict land cover transition from woodlands to mostly cultivated/grazing and built-up areas. The land cover mapping showed that 23% of the land was covered by agricultural land in 1989. Subsistence agricultural area has increased by 18%, occupying 41% of the study area in 2002. The effects of the derived land cover changes on river flow in the Upper Shire river were investigated using the semi distributed soil and water assessment tool (SWAT) model. River flows were found to be highly variable and sensitive to land cover changes. Simulation results show that 2002 land cover data produces higher flow peaks and faster travel times compared to the 1989 land cover data. The changes detected indicate the effects of land use pressure in the catchment. The study highlights the importance of considering effects of land use and land cover changes on ecosystems, and water resources for an informed decision on proper catchment planning and management.     

基于BP神经网络的鄱阳湖水位模拟

考虑到鄱阳湖水位受流域五河与长江来水等多因素的共同作用而表现出高度非线性响应,采用典型的三层BPNN神经网络模型来模拟鄱阳湖水位与其主控因子之间的响应关系。分别将湖口、星子、都昌、棠荫和康山水位作为目标变量进行BPNN模型构建和适用性评估。结果显示：综合考虑流域五河及长江来水（汉口或九江）的BPNN水位模型,空间站点水位模拟精度（R2和Ens）可达090以上,各站点的均方根误差（RMSE）变化范围约050～10 m,若忽略长江来水的影响作用,仅将流域五河来水作为湖泊水位的主控影响因子,模型训练期与测试期的纳希效率系数（Ens）和确定性系数（R2）显著降低,且低于050,均方根误差（RMSE）也明显增大（124～288 m）,意味着综合考虑流域五河与长江来水是获取结构合理、精度保证的鄱阳湖水位模型的重要前提。同时建议针对鄱阳湖湖盆变化对水位的影响,尽可能选择一致性较好的长序列数据集来训练和测试BPNN模型。所构建的BPNN神经网络模型可进一步结合流域水文模型,用来预测气候变化与人类活动下流域径流变化对湖泊水位的潜在影响,也可作为一种有效的模型工具来回答当前鄱阳湖一些备受关注的热点问题,如定量区分流域五河与长江来水对湖泊洪枯水位的贡献分量,为湖泊洪涝灾害的防治和对策制定提供科学依据     

Discharge of 137Cs and 90Sr by Finnish rivers to the Baltic Sea in 1986-1996

The total amounts of 137Cs and 90Sr transported from Finland by rivers into the Gulf of Finland, Gulf of Bothnia and Archipelago Sea since 1986 were estimated. The estimates were based on long-term monitoring of 137Cs and 90Sr in river and other surface waters and on the statistics of water discharges from Finnish rivers to the above sub-areas of the Baltic Sea. The total amounts of 137Cs and 90Sr removed from Finland into the Baltic Sea during 1986-1996 were estimated to be 65 and 10 TBq, respectively. The results show that, although the deposition of 137Cs was much higher than that of 90Sr after the Chernobyl accident, the amount of 137Cs removed from Finland is only six times as high as that of 90Sr. This emphasizes the importance of 90Sr while considering radiation doses from surface waters and 137Cs while estimating doses via pathways from catchment soil, lake sediments and biota after a fallout situation.     

Distribution of U-Th nuclides in the riverine and coastal environments of the tropical southwest coast of India

A reconnaissance study has been made on the distribution of 238U, 234U, 232Th and 230Th in soils, water, suspended particulate matter (SPM) and bottom sediments in the Kali river basin around Kaiga, its estuarine region and the adjacent Arabian Sea to obtain the baseline data of U-Th series nuclides in view of the commissioning of nuclear power reactors at Kaiga, near Karwar, on the southwest coast of India. Drainage basin soils developed over greywackes (the dominant litho-unit upstream) are lower in 238U/Al and 232Th/Al ratios by factors of 3-5 in comparison with those developed over tonalitic gneisses (the dominant litho-unit downstream). The dominance of the former type of soils is reflected in the composition of river-bottom sediments derived from the upstream drainage basin during the monsoon. The 232Th in bottom sediments tends to increase towards the estuarine and coastal areas, presumably due to deposition of heavy minerals and onshore transport of coastal sediments into the estuary. The dissolved U in the Kali river is low (0.001-0.02 microg/l) when compared to the major Indian rivers as the Kali river flows through U-poor greywackes. Thus, the input of dissolved U to the Kali estuary is dominated by sea water. Although there is some evidence for the removal of dissolved U at low salinity during estuarine mixing, its behaviour is conservative in the lower estuary (at higher salinities). The removal rate of dissolved U from the Kali river basin is similar to that reported from other tropical river basins. The U flux from all the west-flowing rivers of Peninsular India is estimated at 26.3 x 10(6) g/yr to the Arabian Sea which is about 2% of the flux from the Himalayan rivers to the Bay of Bengal.     

"Blind" testing of models for predicting the 90Sr activity concentration in river systems using post-Chernobyl monitoring data

Two different models for predicting the time-dependent mobility of (90)Sr in river systems have been evaluated using post-Chernobyl monitoring data for five large Belarusian rivers (Dnieper, Pripyat, Sozh, Besed and Iput) in the period between 1990 and 2004. The results of model predictions are shown to be in good agreement (within a factor of 5) with the measurements of (90)Sr activity concentration in river waters over a long period of time after the accident. This verifies the relatively good accuracy of the generalised input parameters of these models which were derived primarily from measurements of (90)Sr deposited after atmospheric nuclear weapons testing (NWT). For the cases studied here, the simpler AQUASCOPE model performed just as well as the more complex "Global" model which used GIS-based catchment data as an input. The reasons for this are discussed. Exponential decay equations were also curve-fitted to the data for each river to help assess the uncertainties in the predictive models.     

苏北里下河典型区河网水系演变特征研究

在自然地理气候变化和人类活动胁迫的双重作用下，平原河网地区水系格局发生了很大的变化。以江淮下游里下河典型区--盐都区为例，基于1970s、1980s和2010s年的3期水系数据，利用GIS和遥感解译等技术，选取河网水系数量、结构和连通指标，对河网水系时空演变特征及其影响因素进行分析研究。结果表明：（1）研究区河网密度和河网水面率均呈不断减少趋势，两者分别下降21.3%和10.2%。研究区河流曲度较小，河网连通性较好，河网曲度和连通度总体上变化较弱。（2）研究区河网水系变化存在时空差异性。河网水系在缓慢城市化时期（1970s～1980s）的衰减率高于快速城市化时期（1980s～2010s），特别在城镇化发展缓慢区域表现显著。城镇化发展快速区域，其河网水系持续减少且变化剧烈。（3）研究区河道变化存在着等级差异性。城镇化等人类活动对河网水系的影响，呈现出先低等级（Ⅳ级）河道、后高等级（Ⅰ~Ⅲ级）河道的特征规律。其中，低等级河道数量持续减少且变化强度剧烈，高等级河道数量变化不大且其面积和长度均有所增加。 关键词： 平原河网；水系变化；演变特征；里下河地区     

Contamination of fish by organochlorine pesticide residues in the Ouémé River catchment in the Republic of Bénin

In the Republic of Bénin, aquatic ecosystems are subject to poisoning risks due to the inappropriate use of pesticides, such as washing of empty bottles in rivers and using pesticides to catch fish. In some areas, cotton fields are located near riverbanks, increasing the probability of pesticide emission to the river. To assess contamination levels in the Ouémé River catchment area, different fish species were collected from different geographical areas along the river. DDT, its metabolites and isomers were the most frequently identified pesticides in fish flesh, alpha-endosulfan, beta-endosulfan, dieldrin, telodrin, lindane and octachlorostyrene were also detected. Concentrations of pesticide residues in fish ranged from 0 to 1364 ng/g lipid. A preliminary risk assessment indicated that the daily intake of chlorinated pesticides by people consuming fish from the Ouémé River still is rather low and does not present an immediate risk.     

鄱阳湖水流运动与污染物迁移路径的粒子示踪研究

湖泊水流运动和污染物的迁移路径对湖泊水质起着关键作用。采用水动力和粒子示踪耦合模型,并结合野外粒子示踪实验来调查鄱阳湖洪水季节空间水流运动和污染物迁移路径。模拟结果显示,流域五河点源入湖污染物和湖区空间面源分布污染物均会沿着不同方向迅速进入湖泊主河道,在快速水流推动下逐渐向北部湖区迁移,但部分污染物因东北部湖湾区存在顺时针方向环流而发生长时间滞留和富集。野外粒子示踪实验同样表明,东北部湖湾区污染物迁移路径随着该湖区水流运动表现得复杂多变,除了可以明显观察到顺时针方向的污染物运动,还可以发现污染物向湖岸边界和湖汊迁移,最终滞留在东北部湖湾区,而康山湖区污染物在快速水流推动下主要沿主河道向西北主湖区迁移,未发现污染物进入东北部湖湾区的迹象。鄱阳湖洪水季节水流运动和污染物迁移路径具有枯水期水流-污染物传输特性。模拟得出鄱阳湖洪水期的平均换水周期约89 d,表明其可能需要近3个月时间才能彻底完成一次换水。研究成果可为大型通江湖泊"江湖关系变化"研究提供参考,对保护"一湖清水"、保障长江中下游水环境安全具有重要现实意义。     

Association of plutonium with sediments from the Ob and Yenisey Rivers and Estuaries

The present study applied sequential extraction techniques to investigate the binding and mobility of plutonium (Pu) in sediments from the rivers and estuaries of the Ob and Yenisey. As a study site, the Ob and Yenisey are particularly interesting as both rivers have weapons-grade Pu sources in their catchment areas, including the Russian Pu production and reprocessing plants at Mayak, Tomsk-7 and Krashnoyarsk, and the Semipalantinsk nuclear weapons testing site in Kazakhstan. Plutonium activity and ²⁴⁰Pu/²³⁹Pu ratios were determined using accelerator mass spectrometry (AMS). Sequential extractions showed that between 47 and 80% of the Pu in Yenisey River sediments and 35–53% of the Pu in soils around the Techa River are mobilized with weak oxidising agents, which can indicate that Pu is bound to organic material. In contrast, Pu in Ob and Yenisey Estuarine sediments was more strongly bound, with 60–100% being found in the HNO₃-extractable fraction. This change in speciation could reflect either that Pu bound to organic material in the Techa and Yenisey River sediments becomes more fixed to the sediments with time, or that organic-bound Pu is mobilized and released to the water when the sediments encounter the more saline water of the Ob and Yenisey estuaries. In general, ²⁴⁰Pu/²³⁹Pu ratios were relatively consistent between different extraction fractions, although, in whole sediments, an increase in ratio was observed with distance from the source. This reflects the increased influence of weapon fallout from catchment runoff within the river systems, as compared to the weapons-grade sources close to the production and reprocessing plants. Knowledge of Pu speciation in the Ob and Yenisey Rivers, and the processes controlling its behaviour in estuarine systems, can improve predictions of its transfer and subsequent environmental impact to Arctic Seas.     

Plutonium and americium inventories in atmospheric fallout and sediment cores from Blelham Tarn, Cumbria (UK)

The objective of this paper is to report on the results of a study of 238Pu, 239 + 240Pu and 241Am inventories onto Blelham Tarn in Cumbria (UK). The atmospheric fallout inventory was obtained by analysing soil cores and the results are in good agreement with the literature: 101 Bq m(-2) for 239 + 240Pu; 4.5 Bq m(-2) for 238Pu and 37 Bq m(-2) for 241Am. The sediment core inventory for the whole lake is compared to the atmospheric fallout inventory. The sediment activity is 60-80% higher than the estimated fallout activity, showing a catchment area contribution and in particular the stream input.     

黄浦江上游地区水环境质量演变趋势

为全面反映黄浦江上游地区水质情况，在黄浦江上游地区共选取包括黄浦江干流及上游主要来水水系的7个监测站点的15年监测数据，作为本次黄漓江上游地区水质状况演变分析的主要依据。黄漓江上游水源地处东部平原感潮河网地区．水源地内河道密布、纵横交错，水系结构极为复杂，水源承接上游江苏、浙江来水，其水源保护具有鲜明的区位特殊性和区域综合性。长三角地区河流有机污染严重，普遍出现水体黑臭现象，尤其是近10年来，这一问题日趋突出，许多中小河流的水体溶解氧几乎为零。出现季节性的或终年的水体黑臭。如果不能保证上游江浙来水水质，即使能有效控制水源地点源、非点源污染，而水源水质也很难保证。因此．除了加强对黄浦江上游点源、非点源污染的控制力度，江浙来水亦将成为决定水源水质的重要因子。     

Reference conditions for rivers of the German Baltic Sea catchment: reconstructing nutrient regimes using the model MONERIS

We introduce an approach for establishing reference conditions (RC) for rivers of the German Baltic Sea catchment, based on predictive modelling. An extensive data set of statistics from the year 1880 was coupled with literature data, providing a comprehensive basis for the calculation of regional historical nutrient emissions into rivers, river nutrient concentrations, and nutrient loading into the sea. Four different scenarios were calculated: scenario 1 assumed RC following previously established criteria for lakes (Poikane et al. in Environ Manag 45(6):1286–1298, 2010), scenario 4 applied the nutrient emission conditions of 1880 (considering non-intensive land usage, the presence of tile drainage systems, sewer systems, and the human population from 1880), and scenarios 2 and 3 considered intermediate conditions. Our results showed that nutrient emissions from scenario 1 accounted for approximately one-tenth of the currently observed total nitrogen (TN) and total phosphorus (TP) emissions. The nutrient emissions calculated for 1880 (scenario 4) were found to be 47 % (TN) and 55 % (TP) higher than those calculated considering reference lake conditions (scenario 1). Our results suggest that RC nutrient concentrations in rivers in the German Baltic Sea catchment were clearly below 0.05 mg/l for TP and 1 mg/l for TN. This range is in accordance with historical and calculated pristine nutrient concentrations from other rivers in similar catchments.     

Distribution of environmental Cesium-137 in tree rings

Environmental ¹³⁷Cs concentrations in tree rings grown since 1914 and 1929 were measured and compared with the record of ¹³⁷Cs deposition from weapons fallout. ¹³⁷Cs was found in tree rings formed prior to 1945 when the first nuclear weapon test was conducted. The radiocesium record in wood is not correlated with that of fallout deposition.     

Plutonium isotopes in the lower reaches of the River Rhône over the period 1945-2000: fluxes towards the Mediterranean Sea and sedimentary inventories

Plutonium isotopes in the Rh?ne River originate from both the weathering of the catchment basin contaminated by global atmospheric fallout, and the liquid effluents released from the Marcoule reprocessing plant since 1961. Due to a new treatment process applied to the liquid effluents, a decrease of two orders of magnitude in the industrial plutonium discharged into the River Rh?ne has been registered from 1991. Today, 238Pu industrial inputs to the River Rh?ne are still about 10 times higher than those derived from global fallout, while 239+240Pu inputs from industrial and global fallout sources are of similar importance, i.e. 1 GBq y(-1). Our results indicate that the river sedimentary compartment act either as a sink or a delayed-source term of plutonium for the freshwaters depending on the hydraulic regime and flood events. This compartment may then represent an important industrial delayed-source term for the River Rh?ne freshwaters in the coming years as the Marcoule reprocessing plant is being dismantled. These results were obtained from samples collected from the lower course of the River Rh?ne over the 1987-1998 period and analysed for 238Pu and 239+240Pu activities. Both river sedimentary inventories of plutonium isotopes and effective outputs from the River Rh?ne towards the Gulf of Lions have been estimated for each year over the 1945-2000 period. Regarding 239+240Pu, the sedimentary inventory accumulated since 1945 is estimated to be 172+/-35 GBq. If mobile, this amount represents a significant delayed-source term of plutonium on the scale of the Rh?ne watershed.     

A general approach to transform a lake model for one radionuclide (radiocesium) to another (radiostrontium) and critical model tests using data for four Ural lakes contaminated by the fallout from the Kyshtym accident in 1957

This paper presents results of a model test carried out within the framework of the COMETES project (EU). The aim of the work was to change the structure of the MOIRA lake model for radiocesium so that it can be applied more generally for, in principle, all types of radionuclides and heavy metals. This general lake model is used within the MOIRA decision support system (DSS; MOIRA and COMETES are acronyms for EU-projects). The model is based on a set of differential equations and a specific modelling structure. It incorporates all important fluxes to, from and within lakes in a general manner. Yet the model is driven by a minimum of variables accessible from standard maps and monitoring programs. The model can be separated into two parts, a general part with equations applicable for all types of water pollutants and a substance-specific part. This model has previously been validated for 137Cs from many lakes covering a wide domain and yielded excellent predictive power. The alterations discussed in this work are meant to be general and radiostrontium is used as a typical element. Radiostrontium is known to be more mobile than radiocesium and all abiotic parts of the model handling fixation and mobility have been altered. The new model for 90Sr has been critically tested using data from four lakes heavily contaminated with 90Sr from the Kyshtym accident in the Southern Urals, Russia, using empirical data from a period from 1958 to 1995 for 90Sr in fish (here goldfish), water and sediments.