TTS 9906-543 

Radioactive Contamination of the Techa River and its Effects

Dmitriy Burmistrov*#Mira Kossenko *) and Richard Wilson+)

*) Research Center for Radiation Medicine, Chelyabinsk +) Harvard University, # now at Menzie-Cura and Associates Inc

Department of Physics

Harvard University

17 Oxford St.

Cambridge, MA 02138

Phone: 617-495-3387

Fax: 617-495-0416

E-mail: wilson@huhepl.harvard.edu


Mira Kossenko and Richard Wilson

Introduction

Soviet nuclear science and technology made great achievements before the Second World War. But War (1941 - 1945) completely stopped all Soviet studies in this field. However during this period scientists in the USA and England made revolutionary progress in investigation of the applications of nuclear fission. Unfortunately this progress was initially used for military purposes. The end of Second World War was marked by use of nuclear weapons against Japan. The mere possession of nuclear weapons enabled the USA to exert enormous international pressure and led the government of the USSR to try to "balance the power". A special committee was established by the State Committee for Defense (Resolution number 9887 dated 20th August 1945)  for the production of nuclear weapons, with an unprecedented sense of urgency: the "Uranium Project" (1,2). 

Fig 1. Location of the first Soviet plutonium production plant: "Mayak" Production Association

The Uranium Project in the USSR was implemented under great economic difficulties, with a dearth of skilled scientific and technical specialists and material resources. But the results were achieved in a short time. Two experimental design facilities were established in Kirovís Factory in Leningrad ( now St.Petersburg). The first equipment was designed for uranium isotope enrichment by gaseous diffusion, the second was a laboratory for the development of heavy-water reactors fuelled by natural uranium; and the construction of two plants for the uranium enrichment and plutonium production respectively were started. The second of these plants was the first Soviet plant for weapon-grade plutonium production and was located in the Southern Urals (Fig.1). Igor Kurchatov was appointed Scientific Leader of this facility. A site in the Ural Mountains was chosen as the best place for the construction because of the convenient and safe position in the middle of the country far from the borders of the USSR (and at the time inaccessible for raiding aircraft from beyond the borders), yet close to necessary resources and communications. In the summer of 1945 the site selection survey was made, and in autumn of 1945 the government commission decided to construct the first of seven military reactors on the southern shore of Lake Kyzyltash. Soon, a number of auxiliary enterprises were set up, and  they were connected by network of roads, power and water supply communications. Houses for the employees of the enterprises were built nearby. It was the industrial complex, which is known now as "Mayak" Production Association (1,2). Originally a closed city called only by its postal code (Chelyabinsk-40, Chelyabinsk-65), the city was given the name of Ozersk in 1994. 

The main purpose of the "Uranium Project" was the creation of Soviet nuclear weapons as fast as possible. By the end of 1947 the first military reactor (A-plant) was ready to operate. The reactor was started on 8th of June 1948 and by 19th June 1948 it had reached the designed power (100 MW). It had been operating for plutonium production for 39 y when it was shut down on 16 June 1987. Three additional reactors were put into operation in the period 1950-1952 have also been shut down. All these four reactors were graphite-moderated reactors with direct water cooling loops, using cooling water from the lake Kyzyltash. Only two reactors of the seven built during the operation of the Mayak are still working for civilian needs (1). The radiochemical plant (plant B) for extraction of 239Pu from uranium irradiated in the reactor of the A-plant was put into operation in December of 1948. And in February of 1949 the first plutonium concentrate from the plant-B was converted to high-purity metallic plutonium for the first Soviet atomic bomb at the metallurgical plant (V-plant).

In the 1950s nuclear technology was not yet well developed and there was not enough knowledge in the world and still less in the USSR about the fate of radioactive wastes in natural ecosystems and the effects of radiation on humans. Therefore, efforts to prevent discharges of radioactivity into the environment were insufficient, and territories in the vicinity of Mayak were severely contaminated by radionuclides, including long lived 137Cs and 90Sr .

In the first years of the Mayak operation there were three accidents accompanied by large releases of radioactivity in the environment (1-4): 

1. Techa River contamination (1949-1956, ~ 100 PBq); 

2.Kyshtym accident (1957, ~ 70 PBq)

3.An incident with dispersion of radioactive dust (1967, ~ 20 TBq).

Soon after the start of the Mayak plant, the problem of storage of radioactive wastes of the radiochemical plant arose. The usual way that radioactive wastes in the first years of the Mayak operation were managed was to dump low-contaminated wastes directly to the Techa River system (some of them were passed through the absorbers). This was the practice in the period 1949-1956. Highly contaminated wastes were directed to the tank farms of a special storage facility (complex-C). In January of 1950 a special facility for the decontamination of high-level radioactive wastes was built, and the construction of additional tank farms at complex-C was stopped. Unfortunately, it was found soon that the decontaminating facility did not operate as intended, and this failure resulted in an increase of radioactive releasers in the Techa River. In addition the tanks of complex-C had to be cooled constantly to prevent self-overheating of wastes that contained high-levels of chemicals and radioactive materials. Eventually some leakage occurred through imperfections in the cooling system. Since the water from the Techa River was used as the cooling agent in this system, large amounts of radionuclides (with the daily mean of 0.16 PBq) were eventually discharged into the river in the period 1950-1951. As a result appreciable quantities of radioactivity accumulated in the water and bottom sediments of the river, especially in the upper reaches. In October of 1951 the discharge of radioactive materials into the Techa was reduced and the wastes were collected in an enclosed artificial reservoir known now as the Lake Karachay. Later, arrangements were made to close the Upper Techa and prevent future dissipation of radionuclides deposited there. 

The Techa River contamination was the most serious "accident" but was not the last one. In 1957 the cooling system of one of the storage tanks containing highly radioactive liquid wastes failed to operate properly and the tank overheated and exploded (Kyshtym explosion; 1,4). This tank contained 70,000-80,000 kg of wastes mainly in the form of nitrate compounds. Energy released by radioactive decay caused a daily temperature increase of 50C to 60C. The temperature was controlled using cooling water, which was completely replaced every 12 h. Apparently, the cooling process and temperature control system for the exploded tank failed, the temperature of the waste increased to 330-3500C and the cooling water completely evaporated. On the 29 September 1957 the tank exploded. About 90% of the radioactive material contained in the tank (700 PBq) was deposited close to the site of the explosion, but the rest (about 70 PBq) formed a radioactive cloud, which reached a height of about 1 km, and drifted with the wind to the north-north-east direction, forming the new contaminated territory. This contamination site known now as East Urals Radioactive Trace (EURT). It covers about 15,000-20,000 km2 territory, with approximately 5% (about 1,000 km2) with initial radionuclide densities greater than 70 GBq/km2 (1).

Finally in 1967 there was little snow and the spring and summer were hot and dry. Evaporation reduced the level of water in the Karachay Lake dramatically, and the radioactive dust from the bare banks was spread by wind over surrounding territories. About 1800 km2 were contaminated at the levels 1-10 GBq/km2, and the contamination reached 50-70 km from the Mayak site. By 1967 the residents of territories near the Karachay lake had been resettled.

In spite of military secrecy, the fact that a severe accident had occurred was known in the west as early as 1961. The general outlines of the Kyshtym explosion was deduced by American scientists about 1975 (4) but no details were known. In 1990 the scientific research related to these accidents were declassified, and the first public articles appeared in Priroda (5) The number of scientific publications about radioactive contamination in Southern Urals, and particularly about the Techa River contamination, is rapidly growing . The purpose of this review is to collect the information about the Techa River accident and related investigations and make it more accessible to a wider range of readers. The Techa River accident was selected by two reasons: 1) it was the most significant accident in the region with the largest number of people exposed at levels of radiation of more than 0.05 Sv; 2) the current state of this case study is more advanced than the study of the other accidents, and 3) a statistically significant radiation impact on the population was detected.

The Techa River contamination

Location and hydrological characteristics

The Techa River is the right tributary of the Iset River. It belongs to the basin of the Kara Sea (Fig. 1). The sources of the river are the Irtyash and Kyzyl-Tash Lakes. It falls into the Iset River, which in turn falls into the Tobol River. The length of the river is about 240 km. At the time of the radioactive pollution there were 40 villages on the river with a resident population about 28,000 (Fig. 2). These were mostly small agricultural villages. Brodokalmak village was the center of the administrative region

Fig. 2. Map of the Techa River and of the villages located on its banks before contamination occurred. The number of residents, their ethnicity and current status (evacuated or not) are also shown. Total population in 1950 was 27454 (reproduction from (37)).

The riverbed includes layers of turf, silt and clay. There were flood swamps which measured 300 m to 2 km in width along the river shoreline; the most swampy areas were located between the villages of Nadyrov Most and Muslyumovo. The flood soils were composed of turf-bog soils that give way to meadow-turf ones along the boundaries of the swamps. The thickness of the turf layer ranges from 10 cm to 3 m, the turf contains a considerable amount of mineral inclusions and increased percentage of ash. Clay and sandy loam, less frequently sand, compose the underlying layer of turf. 

Downstream of the village of Muslyumovo the river has a well-formed bed (Fig.3), its bottom consists of layers of sand and slime, in some places of sand and gravel. The mean width and depth of the river during the summer time are 22 m and 0.5 m to 1 m, respectively. The stretches of the river from Kyzyl-Tash Likes to the village of Muslyumovo were for the most part swampy with a poorly marked winding bed overgrown with water plants. The width of the riverbed varied from 3m to 15m and its depth ranged from 0.5m to 2 m. 

The Techa River receives its supply of water from melting snow and intensive spring floods. The main source of water supply to the Techa during the summer months is discharge from groundwater formed by atmospheric precipitation. In the 1950s the flow rate varied from 2 m3/s to 10m3/s. The Techa river is rather small and shallow. That was one of the reasons that there was appreciable contamination of the river environs. 

Fig. 3. Techa river near the Nizhnepetropavlovskoye village (148km downstream the site of releases) with the houses in the background

As shown in fig. 4, in the upper reaches of the Techa there is now a cascade of hydraulic-engineering constructions (6). 

 
a) b)
 
c) d)
Fig. 4. Schematic maps of the upper reaches of the Techa: 

a)At the beginning of "Mayak" PA operation; b) After creation of the Koksharov Pond; c) After creation of reservoir No 10; d) After the upper reaches are made a "closed system". The main sources of contamination were liquid radioactive wastes from the radiochemical plant and reactor cooling water dumped into the Kyzyl-Tash Lake. Two other radiation accidents gave a minor contribution to the Techa River contamination also (Kyshtym accident producing the EURT and Dust Transfer from the banks of Karachay Lake) (reproduced from (6)).

Only the Metlinsky pond existed when Mayak began operation. Later, in 1951 the Koksharov Pond was created to prevent direct flow of radionuclides to the Metlino village (Fig 5). It was the first village downstream 7 km from the point of the releases, and the residents accumulated the highest radiation doses there. In 1956 arrangements were started to restrict the influence and spread of radioactive contamination: residents of upper reaches of the river were evacuated and a series of additional dams were built (Fig.4).

Fig. 5. Metlino village was situated 7km downstream of the site of releases of liquid radioactive wastes by "Mayak" plant, before the evacuation of residents in 1951. Later almost all buildings were destroyed, although the buildings seen at this picture (the church and the mill) remain. They are used now for collection of brick samples for thermoluminescent dosimetry measurements (18-20). The water reservoir at the foreground is the reservoir N10 created in 1956. Now the buildings are still decaying by natural processes. Metlinsky pond is situated behind the mill buildings. (The picture by E.I.Belova is presented from personal archives of Dr. Nelli Safrnova with her permission)

Mayak as a source of Techa contamination

The production of weapons-grade plutonium at Mayak required a number of stages. At the first stage the natural uranium was irradiated in the uranium-graphite reactors with thermal neutrons to generate 239Pu in the fuel at the reactor complex-A. After irradiation the fuel was treated at the plant-B to separate of the 239Pu from natural uranium. Then the product enriched by 239Pu was directed to the metallurgical plant-V to produce high-purity metallic plutonium (1,6). 

There were a number of sources in this chain that contaminated the Techa River. The reactors in the complex A contained only one cooling loop: the water from Kyzyltash Lake circulated directly through the reactor core and released back to the lake. As a result cooling water contained radionuclides with short half-decay times generated by irradiation of material in the active zone of reactor. The water from the contaminated lake was one, but not the main, source of Techa River pollution. There were various kinds of waste products after extraction of uranium and plutonium from the irradiated fuel at the radiochemical plant B. Liquid radioactive wastes with intermediate and low-level radioactivity were dumped directly to the Techa River. In 1949 high-level wastes were routed to the tank farms in the complex-C. However in order to reduce the volume of materials going to the tanks, a process for decontamination of high-level wastes was introduced in 1950 with a portion of the radioactivity directed to the tanks and a portion released to the river (6). In July 1951 it was discovered that this process did not work as intended, and during this period large quantities of radionuclides had been released into the river. It was also noted that sometimes  cooling water from the tanks of Complex C was discharged into the Techa. 

Leaks in the tank- cooling lines caused some of these discharges to be highly contaminated. These "wild releases" were unmonitored and unnoticed until 1951. Over this period, about half of the total release to the Techa River resulted from routine releases and about half from wild releases (6). 

The nature and quantity of the released radionuclides in 1950-1952 were estimated by Ilyin in his Doctoral Thesis (7,6), the head of the Mayak Central Laboratory. These estimates were made on the basis of all available measurements and with knowledge of technological processes at Mayak. Later (1953-1956) similar estimates were made by Marey (8) in his doctoral thesis using the results of measurements by special expeditions under the supervision of Moscow Institute of Biophysics. The sources of information used by Ilyin and Marey as well as their original reports as of 1999 are still classified. They estimated that about 100 PBq of beta-emitters were released into the Techa river, and approximately 98% of this total activity was dumped in the period March 1950 to November 1951. The total amount of gamma-emitters was as much as 24% of the released beta activity, and the released activity of alpha-emitters was less than 2 TBq. About 70% of released activity were incorporated in solid particles of sodium nitrate and acetate. The estimated average daily releases and their approximate radionuclide composition are presented in table 1. 

Table 1. Estimates of nature and quantity of radionuclide releases into the Techa River in earlier period of operation of the Mayak plant. Based upon Ilyin (7) and Marey (8).
 

Year

Release, TBq/d 90Sr+89Sr

%

137Cs

%

Rare Earths 103,106Ru

%

95Zr+95Nb

%

1949 2.6 5 10 - 55 30
1950until March 32.2
 
15 15 6 45 9
March 1950 to Sept 1951 161 21 12 27 26 14
1951 since October 3.74) 26-58* 4-15* 10-61* - 8-25*
1952 0.9 26-58* 4-15* 10-61* - 8-25*
1953 0.74 26-58* 4-15* 10-61* - 8-25*
1954 0.56 44 10 45 1

*Only range estimation is available.

Note: the data were originally presented with excessive precision, but their sum sometimes is not equal to 100% . Therefore they are rounded in this table, to emphasize that the estimation is approximate. 

As shown in table 1, in the period of 1952-1955, the releases of the radiochemical plant were much reduced.

Another significant source of contamination was the reactor cooling water that flowed into the Techa River from Kyzyl-Tash Lake. For example, during seven months in 1953, the activity released from the reactor was five times the release from the radiochemical plant (6). The water of Kyzyl-Tash Lake entering the Techa was also contaminated by the activation radionuclides 32P, 35S, and 45Ca. The ultimate fate of radionuclides in the river system was determined by a number of parameters including radioactive decay; transport and dilution by water flow; sedimentation of particulate fraction of the releases; and absorption and desorption processes in soluble fraction of releases with participation of particles of natural admixtures in the water with consequent sedimentation of these particles (9-10). 

The nature of the releases, the resettlement arrangements in the Upper Reaches, the specific transport processes and the transport features of specific radionuclides determined the following pattern of the contamination (1,3,6,9,10,11,12). The radioactivity decreased fast with calendar year since 1951, and with distance from Mayak. The short-lived radionuclides and the radionuclides in particulate form decayed or settled almost totally in the Koksharov and Metlinsky ponds. In the rest of the river the main contaminants were 137Cs and 90Sr. 

Since 1964 the river was effectively in a self-cleaning regime, accompanied by the slow decay of 137Cs and 90Sr. The radioactivity was accumulated predominantly in river bottom sediments: the ratio of concentrations in bottom sediments to water concentration varies in the 100-1000 fold depending on the nature of the sediments. Flood plain soils were contaminated during spring floods, especially during the outstanding flood of spring 1951. Houses and land were contaminated due to the activity of people and agricultural animals. 

Regular environmental measurements were started in the summer of 1951 when the mistakes in the treatment of radioactive waste became evident (3,6). The most representative measurements (almost daily at a representative set of locations) were made for the total beta activity of river water. Eventually samples of river sediments and flood plain soils were collected at different locations along the river and a number of radiation dose-rate measurements were made in air near the river edge, in the streets of the villages and in houses. One of main purpose of these measurements was the determination of critical groups of the population - those who accumulated large doses - in order to make decisions about their evacuation from contaminated territories. That goal did not require systematic and precise monitoring and attention was centered on high-level values. This approach has added difficulties in the reconstruction of radiation doses of the exposed people. 

The radionuclide concentrations decreased 10 fold in water and radioactive concentrations in the bottom sediment decreased 100-1000 fold from the upper reaches to the mouth (Fig 6). 

Fig. 6. The results of measurements of total radionuclide concentration in river water and bottom sediments in 1951 and 1953. The radioactivity of the water decreased approximately 10 fold along the river. The units of the y axis are different for each line and are shown in the inset describing each line.   Data are taken from Vorobia et al. (6)).

The dilution by clean water of the Iset River caused an additional 10-fold decrease. And there was further 100-1000 fold dilution in the Tobol River (3). Thus the contamination of Iset and Tobol Rivers can be considered to be comparatively insignificant. The most significant radiation conditions were in the upper and middle Techa. 

In 1951 the level of pollution of water in the Metlinsky Pond (Fig. 4a, 5) allowable levels for 90Sr exceeded by 2,000 to 3, 000 and the 137Cs level exceeded 100 times (3). The radiation exposure rate reached in some places the levels at which in only one hour a person could have accumulated a radiation dose comparable with the maximum annual allowable dose for radiation workers. Since late 1951, when the massive releases ceased, there was about 100-fold decrease in exposure rates (3). However in the subsequent period the exposure decreased slowly due to slow decay of cesium in the coastal strip of the river.

Population Protection Measures

The elevated radiation levels in the Upper Reaches of the Techa in 1950-1952 were of concern to the health of local inhabitants, particularly because they used the river as the basic and often sole water supply resource. The situation required a set of measures to protect people's health (3). 

Effective from the autumn of 1951 it was officially forbidden to use the river for drinking, swimming, domestic and agricultural needs. Then the migration of residents of the village of Metlino began. Despite the reduction in the releases the concentrations of radionuclides still exceeded allowable levels. For this reason a new dam was built on the swampy area downstream of the Metlino village (Fig. 4c), which was constructed in 1956. However a significant reduction was not achieved. By that time the main source of pollution of water was releases of radionuclides from bottom sediments. Therefore, it was decided to construct additional dams and canals, which totally isolated the hydraulic engineering objects of Mayak and the polluted marshes in Upper Techa River from the rest of the Techa in 1963 (Fig. 4d). 

During this period a set of additional administrative measures were taken about the territories along the Techa River. A number of riverside fields covering about 80 km2 were forbidden for agricultural uses (3), the riverside territories were fenced by barbed wire and warning signs were set up. Special security guards were organized to enforce these measures. Water pipes and wells were constructed for maintenance of drinking and economic water supply. The use of the water of the Iset River for drinking and fishing was forbidden (3). These measures decreased the levels of irradiation of the population. Meanwhile the resettling of the residents was extended, and by 1960 about 7500 residents had been resettled (3) (Fig.2).

Radiation Doses of the Population

The reconstruction of doses to the population that were exposed is always a difficult task. Because of its importance the way in which dose to one individual was estimated is described in some detail. At the Techa River people were irradiated both internally and externally. The reconstruction of the internal dose (mainly due to 90Sr) is more important for the largest part of the population, and it is more reliable than the estimation of the external dose. 

The external gamma-irradiation was predominant only for those who lived in the Upper Reaches of the Techa in 1950-1952 and consumed relatively small amount of water from the river. The main cause of external irradiation was gamma-irradiation by 137Ba , the decay product of 137Cs. The 137Cs was mainly accumulated in bottom sediments. The important factors determining the level of external irradiation were the dose rate in air near the river where the inhabitants spent daily. Some additional irradiation occurred in streets of the villages, which were contaminated via human activity, and inside houses due to usage of river water for domestic activity. 

Except for the residents of upper Techa the main pathway of dose accumulation for riverside individuals was internal irradiation by beta particles of ingested 90Sr and 137Cs (11,12). Cesium did not give a large contribution to internal irradiation for two reasons: it accumulates in organ tissues uniformly, and it has a comparatively short retention time in an organism (about 100 d for an adult). The contribution of 90Sr (including its decay product 90Y) to the internal dose was more significant because it is concentrated in human bone tissues for a relatively long time. The largest doses of internal irradiation were accumulated by those who were teen-agers in the 1950's: their skeletons were growing rapidly requiring the largest amounts of calcium and hence, the 90Sr (11,12). The main source of ingested strontium in 1950-1951 was drinking water from the river. Later, after the wells and pipes construction and prohibition of usage of the river water, the main source of radionuclides intake was the milk from cows pastured in the contaminated territories. 

These estimations for external irradiation are based on: 1) average values of the dose rate measured near the river bed, on the streets of the villages and in the houses; and 2) a survey by M. Saurov (Moscow Biophysics Institute), who estimated the mean times spent at the river by residents of the different age-cohorts in the 1950's (11, 12) 

The data available do not provide information on the variations in individual dose levels between the residents of a village. Instead the average value for specified age groups and specified settlement is assigned to each resident. In this approach the personal variability of the external dose is described only by the variability in personal residence histories. 

A possible method for dose reconstruction consists of measurement of the number of Electron Paramagnetic (Spectral) Resonance centers (ESR) in natural crystals contained in tooth enamel and cement and in other bone tissues. This number is proportional to the dose of external irradiation accumulated by a person (13-17). However, these measurements are difficult and expensive and not many of them have yet been made for the Techa residents. Additional verification of the levels of external exposure can be made now using Thermolumeniscent Dosimetry (TLD), using the ability of some natural minerals. A number of such measurements were made using brick samples taken from buildings of Metlino and Muslyumovo villages to validate estimations of external exposure in the 1950's. They are in reasonable agreement with calculations as well (18-20). 

In the late 1950s special branch N4 of Moscow Biophysics Institute (now called Ural Research Center for Radiation Medicine: URCRM) was organized in Chelyabinsk city (Fig. 1). Its purpose was to control the health status of the population irradiated as the result of Techa River contamination and (later) the Kyshtym accident of 1957. A 90Sr whole-body counter (WBC SICH-9.1) was designed there in 1974 (21). Because neither the 90Sr nor its daughter 90Y emit gamma rays WBC SICH-9.1 was designed to measure bremsstrahlung resulting from beta emissions. WBC SICH-9.1 could measure the content of the 90Sr and 137Cs separately using specific features of their photon spectra (21). 

The WBC SICH-9.1 is a big metal tube (shielding room), with four photon detectors connected to a computerized analyzer. The detectors were fixed in the central vertical plane of the shielding room. On the bed frame a fabric is placed which stretches under the weight of the body in such a way that the medium plane of the body is at equal distances from each detector. During measurement the person lying on the bed is moved through the detector array. The motion is controlled by signals from the analyzerís real-time clock. The measurements are made during controlled stops of the bed (21). Calibration of the WBC SICH-9.1 was carried out in 1974 using phantoms designed in two different laboratories (21). The difference in the calibration coefficients from different laboratories was about 10%. 

The WBC SICH-9.1 measurement provides the radioactivity content of the body at the time of measurement and does not give the desired accumulated radiation dose. To calculate the total dose absorbed by tissues during the whole period of irradiation, one should know the dose-rate at each moment of the irradiation time for integration over time. Unfortunately, there were no WBC SICH-9.1 measurements before 1974, and when they were started the period of high radionuclide body burdens for the exposed peoples was in the past. Thus the dose must be reconstructed using a mathematical model, which describes the radionuclide content in human tissues depending on time (22-26). An example of such a calculation is shown in Fig.7. Moreover the WBC SICH 9.1 measurements are available only for part of the exposed population. For the rest of the population a Reference Model is used, which represents the average 90Sr content for the group of people exposed at given age at a fixed location. If individual measurements are present we can adjust the Reference Model, multiplying it by a factor, which provides the best fit of the model predictions for the individual body burden (Fig.7). This adjustment is based on the assumption that the temporal pattern of radionuclide intake for the given person was the same, as for the "reference person", but the value of intake was greater (or less) with the factor, described above, due to specific personal behavior or residence location. The main reason for variability of radionuclide intakes between persons was the presence or absence of wells with clean water. 

Fig. 7. Example of the reconstruction of strontium-90 body burden. The lower dashed line represents the Reference Model describing the average body burden for the residents of the Techa of the same age and residence history as for the person with individual code 65737. The upper curve represents the model, adjusted to represent the individual Whole Body Counter (WBC) measurements for this person: the adjustment factor is equal 5.97 in this case (WBC data are reproduced with permission by E.Tolstikh, private communications). 

The Reference Model is used for persons with absence of individual measurements. 1kBq corresponds to 1000 radioactive decays per second in the skeleton. Each of these decays gives its contribution to accumulated dose. Thus reference and individual accumulated doses are proportional to the square of the areas under the corresponding solid lines. In the shown extreme case the individual dose is greater than the reference dose, but the opposite situation can take place for other persons, and in the average value of individual correction factor is equal to 1.

An additional difficulty was that the mathematical model required information about the dynamics of radionuclide intake for the given person or for a group of persons. 

It is possible to determine the intake of 90Sr using measurements on each person's teeth (23). The teeth enamel is formed only in the first years of life Therefore, the measured radioactive decay is almost entirely from the 90Sr that entered the enamel in childhood. The 90Sr content in the enamel can be determined using conventional beta counting using one and the same procedure. 

Absolute values of the intake could be obtained using the more precise WBC SICH-9.1 measurements. Given the time-pattern of the intake for 90Sr, and using estimated ratios of concentration of other radionuclides in food relative to 90Sr, it was possible to estimate the intake of other radionuclides. At present, it was assumed that these ratios were the same as the corresponding radionuclide concentration ratios in the river water for given time period and location. Given the intakes, and using model curves analogous to those shown in Fig.7 but for other radionuclides, it was possible to estimate the contribution of all spectra of ingested radionuclides (Fig. 8). 

Fig. 8. Structure and dynamics of individual Red Bone Marrow dose for the person 65737. 

Fig.9. The radiation dose accumulated in the Red Bone Marrow for permanent residents of the Techa riverside. The "Reference Model" is used for all members of the Techa River Cohort. Effective dose equivalent limit for continuous exposure for public during 50 years is equal to 0.05 Sv (5 rem) (26).

It is clear from the above considerations that it is impossible to base the dose assessment for the population using only direct measurements. They must be accomplished by calculations with mathematical models fitted to data. In order to use the mathematical model for individuals based on their residence histories the Techa River Dosimetry System (TRDS) code was created. For a better understanding of the scheme of the dose calculation with TRDS the calculation process is described for the red bone marrow dose for a particular person (code number 65737). 

A database MAN was created in the URCRM to support the follow-up of exposed peoples. In particular, it contains information about person's residence history, which is used for individual dose calculations. Each exposed person has a "system number" used as an identification code. The person with the system number 65737 was born in 1928 in the Ibragimovo village (Fig.2). In 1953 he moved to the Muslyumovo village where he died in 1995. The selected person was exposed in two locations during the period of 1950 to1955, when external exposure and radionuclide intakes were significant. The total period of dose accumulation was 45 y.

The radiation dose is calculated in the TRDS according to the following formula

(1)

where

Do = cumulative dose to organ o;

m = month of exposure;

L(m) = place of residence for the mth month (river location identifier); 

r = radionuclide identifier; 

Age(m) = the age of the person in the mth month;

RISr-90 (Age(m),L(m))= reference intake of 90Sr for given time and location 

fr(m,L(m)) = fraction of r-th radionuclide in river water relative to 90Sr;

DFro(Age(m)) = dose factor for radionuclide r in organ o;

p = place of exposure identificator (indoor, outdoor, at the river bank)

DRp(month,L(month)) = dose rate in air 

Ao(Age(m)) = air-to-organ correction, independent of energy (Sv/Gy)

Tp = time spent at the given place of exposure

The reference intake of 90Sr in 1950 in the Ibragimovo village, obtained from teeth beta-counting in (21) as described above, was about 2 MBq. The person 65737 resided in Ibragimovo throughout the whole of 1950. There are two options: 1) use the "Reference Model" for the intakes for the whole population; and 2) use the individual model, based on WBC measurements. The first approach is appropriate if we want to know the doses for the largest number of people, who can have no WBC SICH-9.1 measurements, the second one can be applied for selected sub-cohort of peoples with known WBC SICH-9.1 measurements. These two approaches should not be mixed for biasing reasons: persons with the highest doses have the higher probability to have WBC SICH-9.1 measurements which are available for selected persons. They are shown in Fig.7. The ratio of the reference and the adjusted 90Sr content for this person is about 6. If the individual approach is used, the reference intake of 2 MBq must be multiplied by 6 resulting in 12 MBq. The reference intakes of other radionuclides, are calculated from 90Sr reference intake RISr-90 and radionuclide fraction in river water fr (Eq.1), using the model of radionuclide transport in the river (9) resulting in 1500 kBq of 89Sr, 1000 kBq of 137Cs, 300 kBq of 103Ru, 1500 kBq of 106Ru , 700 kBq of 95Zr, 23 kBq of 95Nb, 0.350 kBq of 141Ce and 87 kBq of 144Ce. These values were calculated using model of radionuclide transport, the ratios of concentration of each radionuclide and the 90Sr concentration in river water for given time period and location. It was assumed that these ratios are the same, as the analogous ratios of radionuclide concentrations in resident's diet for this time and location. To use the individual model each of the listed intake values should be multiplied by 6. 

Given the values of the intakes, the accumulated dose of internal irradiation for the person 65737 can be calculated using tabulated dose coefficients DFro values from instant intake of 1 Bq of each radionuclide in 1950, when the person was 22 y old for subsequent 45 y of exposure. These tables of "dose-coefficients" are available for the majority of radionuclides in the recommendations of the National Council on Radiation Protection and Measurements (27,28). We use the tables calculated on the basis of these recommendations by V.Berkovsky (29). But for 90Sr and 137Cs the model (22) is used, which is based on a large number of WBC measurements and therefore takes account of the specifics of the irradiation on the Techa river better. The product of the intake values (multiplied by 6) in 1950 and dose coefficients give the following contributions to the accumulated dose: 0.05 Sv(89Sr), 3 Sv(90Sr), 0.08 Sv(137Cs), 0.0003 Sv(103Ru), 0.0014 Sv(106Ru), 0.002 Sv(95Zr), 0.00003 Sv(95Nb), 0.00000005 Sv(141Ce), 0.0001 Sv(144Ce).

The dose from external irradiation is calculated as follows. According to the survey  in 1950, an adult person spent about 150 h at the shoreline (river bank), 1410 h in the streets of the village, about 3960 h inside his house, and the rest of the time on clean territories far from the river. Using available measurements it can be shown, that the average ratio of the dose-rate in air in the residence area to the dose-rate near the shoreline was about 0.035, and the mean ratio of dose rates in the houses to dose-rate in the street is about 0.5. The dose rate in air near the riverbank in 1950 at the Ibragimovo village was about 2.4 Sv/y. This value was obtained by interpolation of actual measurements carried out in the 1950's (11). To calculate the external irradiation dose in the red bone marrow, the shielding effect of other tissues of the organism must be considered. These coefficients of conversion of dose-rate in air to dose-rate in a target-organ Ao are available (30). Using these age-dependent conversion coefficients the fraction of the dose-rate in the red bone marrow of the person 65737 was about 0.73 of the dose-rate of external exposure. The corresponding contributions of exposure at the river, in the residence area and inside the house were 0.04 Sv, 0.013 Sv and 0.019 Sv respectively. Thus the contribution of the first year of exposure for the person with the number code 65737 was about 3 Sv. The contribution of internal irradiation to the total organ dose was about 98%. The main contaminants, 90Sr and 137Cs, give about 98% of the dose by internal radiation. The above conclusion applies only to the bone marrow. If one calculates the dose in target-organs other than red bone marrow, the contribution of 90Sr should be much smaller. 

The contributions of other years to the exposure are calculated in the same way, but using values of the intakes, dose coefficients, exposure rates and behavior specific for the given location, calendar year and current age of the person. The difference is only in account of the contribution of the year 1953, when the person moved to the new location. There is no information on the exact date of this migration for this selected person. Therefore, it was assumed, that the migration took place in the middle of the year. Actually the contribution of the year 1953 is the sum of half-year contribution of residence in the Ibragimovo village, and half-year contribution of residence in the Muslyumovo village. The contributions of the 1954 and 1955 years are calculated using intakes and the exposure rates corresponding to the Muslyumovo for these calendar years.

The dynamics of accumulated red bone marrow dose for the person 65737, calculated in the described way, and the structure of this dose is shown in fig. 8. The distribution of accumulated doses in the exposed population of the Techa river residents, calculated without taking account of individual WBC measurements, but using the Reference Model only, is shown in Fig.9. It can be seen that the selected person has an extremely large dose as compared to the majority of the exposed population. 

It is easy to understand that there are many shortcomings, uncertain and variable factors in the above calculations. It is possible to estimate their probabilistic distributions, and calculate the distribution of accumulated dose for the person 65737 (or any other person) instead of the average dose. It allows as at least attributing the error range for each calculated individual dose. This approach is more informative and reliable, but it is only in progress now for the Techa River Dosimetry System.

Radiation Effects in the Exposed Population

There are at least two important reasons for studying the effects of the radiation exposure on the health of the exposed population. The first radiation measurements were begun during the summer of 1951- 2 years after releases of radioactive material into the Techa had started. In the upper reaches of the river, at the shores of the Metlinski pond 7 km from the discharge point, gamma background levels were 5 R/h (0.05 Gy/h) at some sites. These levels were high enough to initiate a study of radiation pathology among the residents of the riverside villages. Visiting teams of physicians of Medical-Sanitary Department Number 71 (Chelyabinsk 40) and the Biophysics Institute, USSRís Health Ministry (Moscow), conducted medical examinations of the population. Visiting examinations at that time, consisted of interviewing the patients, an assessment of the patientís health status by pediatrician or internal medicine physician, gynecological examinations of women, and peripheral blood studies including counts of all morphological elements. In the summer of 1951 only a few of the Metlino residents were examined. The task of the next visiting team, which arrived in the area one year later (1952), was to conduct dynamic examinations of both members of the Metlino community and residents of villages located on the Techa downstream of Metlino. The first reports showed the presence of patients with complaints of easy fatigue, general weakness, sleeplessness, headaches and dizziness, nausea, reduced memory, pains in bones, stomach and intestine. Objective symptoms were mainly represented by impairments of hemopoiesis manifested by leucopenia and trombocytopenia. It was at that time that the occurrence of chronic radiation sickness was proposed among the exposed population. Later on the studies allow the observation of such effects of irradiation as increased leukemia and solid cancer cases rate among exposed peoples and possible genetic effects (late effects). 

Chronic Radiation Sickness

The concept of chronic radiation sickness (CRS) was introduced by A.K. Guskova, G.D. Baisogolov, et al., who were faced with the necessity to designate the disease developed by several hundred workers of the Mayak facility early in the 1950s (32-34). Accordingly, chronic radiation sickness is a complex clearly outlined syndrome which develops as a result of a protracted exposure of the organism to radiation with single or total doses exceed systematically the permissible limits of occupational exposure.

It is well known that high radiation exposure (> 1 Sv) over a short period of time (less than a day) leads to the so-called "Acute Radiation Sickness". The Lethal 50% dose (at which 50% of the exposed persons die) varies from 3.5 Sv without intervention to 5 Sv with medical intervention. In most countries of the world anyone who is exposed in an accident to 0.50 Sv or more is taken off duties involving radiation exposure or removed from residence in the accident area. But around "Mayak" from 1949 to 1955 this was not done and many people including workers at the reactor or plutonium extraction plant and villagers along the Techa river received high doses for several years in succession. The result was a disease, unique to Mayak and Techa, hereto unknown in the worldís official disease nomenclatures. Most members of the teams visiting the Techa villages had already gained some clinical expertise in diagnosing and treating CRS among the Mayak workers. Nonetheless the diagnosis of chronic radiation injury in exposed residents of the Techa River was a difficult problem. There was no information on the levels of exposure, and health status of the population prior to the exposure. It was difficult to make a differential diagnosis between general somatic and radiation pathology. It was noted that verification of the diagnoses in children was more difficult than in adults because as a rule, children are unable to clearly formulate their complaints. 

The reports prepared in 1952-1962 initially identified 1159 cases of CRS. The disease was diagnosed in 65% of the total adult population and in 63% of all children examined in the village of Metlino. However, a considerable percentage (6%) of cases of CRS were diagnosed among the residents of the lower reaches and in villages located on the Isset River. With improved dose estimates and increased expertise in the follow-up of exposed people, the diagnoses of CRS were becoming better substantiated, and a proportion of diagnoses made earlier were revised. A special commission revised the CRS diagnoses during 1959-1964 mainly based on new medical and dosimetric information in the period 1959-1964 taking into account the information contained in all available medical and dosimetric records. The commission reported that "given the low body content of radionuclides and in view of the fact that CRS, especially at the first stage, has no specific symptoms, such high disease incidence seems questionable". The diagnoses of 940 cases were considered reliable and a dynamic follow-up was begun. 

However, for reasons of military secrecy, no mention of CRS was made in the patients' medical records and it was forbidden to inform patients that they had CRS. A codified name was used - ABC (astheno-vegetative syndrome) or the term "special disease" or "specific injury". Sometimes only the stage of the disease was indicated (stage 1 - first stage of the CRS). All of these terms were well understood by the medical personnel of the URCRM clinic. 

The main symptoms of CRS are the changes in peripheral blood composition. It was observed that the hemoglobin level was below 120 g/L among a considerable proportion of patients in the early period after diagnosis of CRS as shown in fig. 10a and 10b. 

 
a) Men b) Women
Fig.10. Hemoglobin levels for exposed peoples.

The hemoglobin values which could be interpreted as anemia (below 100 g/L for women, below 126 g/L for men) were recorded in 26% of women and 30% of men during the period of the development of the disease. But hemoglobin values were influenced by the inadequate life conditions during the post-war period, imbalanced nutrition, deficient in microelements and vitamins so that using a control group was essential. In the period 1951-1955 patients with CRS manifested considerably reduced thrombocyte counts as compared to controls. The distribution of thrombocyte counts is shown in figure 11a. Leukopenia, defined as a leukocyte count 4.1x109/L and lower, was noted during the period of the development of CRS and was most pronounced in patients with highest doses (over 1 Sv). 

In some patients leukocyte counts were in the range from 2.1x109/L to3.5x109/L during the period of development of CRS, and is shown in figure 11b. 

 
a) b)
Fig.11. Levels of thrombocytes (a) and leucocytes (b) among patients with CRS cases, exposed and control group of population.

It was also attempted to attribute to CRS other less significant changes in the blood parameters, nervous system, bone tissue structure, immunological resistance, miocardial functions and gastric secretions. 

The clinical symptoms described above were mostly observed in the period characterized by external radiation exposure, radionuclide incorporation and significant dose rates. This was when the diagnoses of CRS were made. Later on, when the exposure of the population to the unconfined source of radiation (river) had ceased, external exposure stopped and internal dose rates decreased considerably. During this latter period clinical picture was characterized by regression of pathological symptoms and processes of gradual repair. Most patients had recovered by the late 1960s. 

None of the symptoms observed in the patients are unique to CRS and all can be encountered in other diseases. Therefore, a well-substantiated diagnosis of CRS had to fulfil two conditions:

(1.) Availability of clear information on the nature of exposure and exposure dose received by the patient;

(2.) Establishment of a differential diagnosis between radiation injury and general somatic diseases. 

Data derived from medical records show that at the time of CRS diagnoses a large proportion of patients were suffering from various general somatic diseases which may have manifested symptoms imitating those of CRS. Most commonly infectious and parasitic diseases were diagnosed, brucellosis being the most common one (149 cases). Rural districts of Chelyabinsk region have been endemic for brucellosis since early 1930s. Differentiation between brucellosis and radiation pathology was always the focus of attention: the time of the occurrence and dynamics of clinical symptoms were carefully studied, immunological methods of diagnosis and skin tests using brucellosis vaccine were applied. The relatively small doses received by a number of patients with CRS, and the presence of general somatic diseases at the time of diagnosis of radiation pathology suggests that in a number of cases the diagnosis of CRS was incorrect. At the same time, both the verification of the individual doses and verification of the diagnostic procedures were sufficient to conclude that at least 66 cases of CRS diagnoses were well substantiated. These cases satisfied the main conditions for CRS verification: the rates of exposure doses accumulated in the red bone marrow was close to 1 Sv (100 rem) for one of the exposure years, and the period of CRS was free of other diseases with symptoms similar to those of CRS. Note that the presence of general somatic diseases did not preclude the development of clinical manifestations resulting from radiation exposure, moreover, it contributed to the development of radiation injury. Based on our experience the likelihood of the development of CRS was higher in those patients who had developed endocrine diseases, chronic infectious processes, avitaminosis, or reduced functions of bone marrow. Such concurrent conditions resulted in a decrease of the organism=s reserves, and determined, to a significant degree, the individual radiosensitivity.

The dose distribution for CRS patients is shown in figure 12 with doses re-estimated using the best available information.. 

Fig.12. The radiation dose accumulated in the red bone marrow for observed cases of chronic radiation sickness (CRS). 

These data contradict somewhat statements by the International Commission on Radiological Protection (ICRP) on chronic radiation exposure (35). The ICRP publication suggests a higher threshold dose value for similar changes in blood caused by radiation: 0.4 Sv/y for homopiesis inhibition, 1 Sv/y for lethal bone aplasia. However, this apparent disagreement may be based on the following: 

1. The estimates of individual doses may be and is likely to be inaccurate; 2. The proportion of CRS diagnosis may be erroneous; 3. The threshold dose value for occurrence of non-stochastic effects in a heterogenic population may be lower than that listed in ICRP-41 because in accidental situations the population may be exposed to a wide range of hazardous materials and conditions, along with radiation exposure. 

Some of the residents exposed on the Techa did not develop CRS, but in the early years of exposure they manifested isolated radiation reactions most commonly represented by hematological changes demonstrated by peripheral blood studies. 

Cancer Mortality

An increased incidence of leukemia and other malignant neoplasms is a well-known effect of high radiation exposure. In 1967 cohorts of exposed residents of the Techa River were idetified by the staff of Branch No 4 of the Moscow Biophysics Institute (now URCRM). The cohorts were defined on the basis of records from the earlier examinations and addresses from bureau and taxation records. The Techa River Cohort (TRC) includes 26500 people who were living in villages on the Techa riverside (36, 37) throughout the period of the maximal releases from 1949 through 1952 (Table 2). Their ages at the beginning of the radionuclide releases ranged from 0 to 96 with a mean age of 29 y. About 58% of the cohort members are women. The cohort includes two ethnic groups: Slavs (mainly Russians) and Tatar/Bashkir. During the last decades the personal data and residence history data for members of TRC have been updated and corrected in accordance with newly received archival documents and personal interviews. Later all these data were arranged in searchable form in the MAN computer database (11,12). 

The area of the follow-up of the cohort TRC was defined to include all the territories of the five rural districts in the Chelyabinsk and Kurgan oblasts through which the Techa River flows. These districts include all of the contaminated villages and all the evacuated villages. Between 1949 and 1989 almost 25% of the cohort members are known to have left this area. Roughly 80% of the cohort members who left the area moved to nearby cities and towns in the Chelyabinsk, Yekaterinburg (formerly Sverdlovsk) and Kurgan oblasts. The dates and places of migration have been obtained from address bureaus in the relevant oblast and were included in the MAN database. 

To follow-up the mortality structure of the members of the TRC cohort the death certificates were matched manually with the cohort roster using age, name and residence at death. Additional information on vital status and the migration have been obtained anecdotally during the course of clinic visits or other contacts with family members of the cohort subjects. At the end of 1989 the vital status was unknown for about 36% of non-migrant members of the TRC. Table 3 contains a detailed summary of the nature of the mortality follow-up data as of the end of 1989 (37). 

Table 2 Techa River Population 
 
Exposed in 1950-1952 Exposed in utero and progeny of exposed parents Exposed since 1952 Total
Number 26,500 29,700 7,800 64,000

Table 3 Mortality Follow-up 
 
TRC

1950-1989

Non-migrants Migrants
Number % Number %
DEAD 9,307 46 926 15
CAUSE KNOWN 6,483 70 185 20
CAUSE UNKNOWN 2,824 30 741 80
ALIVE 3.715 18 2,315 37
STATUS UNKNOWN 7,175 36 3,011 48
TOTAL 20,197 100 6,288 100
TOTAL TRC 26,485

Tables 4 and 5 summarize cases of leukemia and solid cancers for the Techa River Cohort. In these tables the "Background" is the prediction of the dose-response model for unexposed peoples (extrapolation to zero dose).   There were 40% excess leukemia cases. 50 were observed compared to the 29 background cases. At doses to the red bone marrow of over 0.50 Sv the 18 observed leukemia cases were three times the 6 background cases. Accordingly, a statistically significant dose-dependent increase of the leukemia mortality was observed. The number of solid cancers was 3% greater than background with an excess of 30 cases (out of 969). Similarly an increased number of additional cases of cancer was mostly observed at doses of over 0.5 Sv (50 rem). Excess cancer cases accounted for about 2% at doses below 0.5 Sv, and about 15% (14 cases excess over the background of 79) at doses of over 0.5 Sv.

Table 4 Mortality from Leukemia among Non Migrants 
 
Red Bone Marrow Dose, Sv Person-years of Observation Background Observed
0.005-0.10 103,031 4 3
0.10-0.20 194,858 9 13
0.20-0.50 200,144 10 16
0.50-1.00 93,873 4 9
>1.00 49,398 2 9
TOTAL 641,304 29 50

Table 5 Solid Cancer mortality among Non Migrants 
 
Soft-Tissue Dose, Sv Person-years of Observation Background Observed
0.005-0.10 459,576 711 716
0.1-0.2 96,297 125 126
0.20-0.50 19,582 24 34
0.50-1.00 32,204 46 52
>1.00 33,645 33 41
TOTAL 641,304 939 969

A value for the radiation-related carcinogenic risk was obtained in earlier studies encompassing a slightly smaller (33 y) period of follow up (1950-1982) (36).  These values are listed in Table 6    The data at the end of 1982 (31 y follow up) were analyzed to give a risk value for  for leukemia mortality: 0.48 - 1.1 cases per 104 person-years per Sv based upon an absolute risk model. For cancers other than leukemia (solid cancers) the risk is based upon a relative risk model. 

Table 6 Estimate of the Risk 
 
Risk estimates Techa River Cohort Atomic Bomb Survivors (LSS) 
Leukemia Excess absolute risk per 10,000PYSv, Linear model 0.85 (0.24-1.45) 2.94 (2.43-3.49) 
All cancer (except leukemia) Excess relative risk per Sv 0.65 (0.27-1.03) 0.41 (0.32-0.52) 

Footnote: * - 90% confidence intervals given in parentheses 

The linear-quadratic model for the dependence of excess leukemia cases on accumulated dose does not fit the leukemia data from the Techa river better than the linear model. The absolute value of the leukemia risk for exposed members of the Techa cohort is somewhat lower than the respective value for A-bomb survivors in Hiroshima and Nagasaki: Life Span Study (LSS) (2.94 per 10 000 person-years per Sv). The data on leukemia are consistent with the traditional Dose Rate Reduction Factor (DRRF) of two to three that is used for estimating radiation risks for continuous radiation exposure to multiply the risk deduced from LSS. The absolute risk for solid tumors is comparable with that for A-bomb survivers (38) but the data are not sufficient to tell whether or not the DRRF is different from unity for these tumors also. 

Thus, the calculations of excess cancer cases based upon the data from the 33-39 y follow-up (since onset of exposure), and the values obtained give us reason to suggest that an extension of follow up is unlikely to result in the number of cancer cases exceeding 200. 

Possible genetic effects

Since genetic effects following radiation exposure might be anticipated, a study was conducted with the aim of assessing the health status of the offspring of people exposed in the Techa riverside villages. In particular, the unfavorable outcomes of pregnancies, birth rate, and the mortality rate among progeny of exposed parents are studied. No reliable effects of radiation were established. 

Unfavorable outcomes of pregnancies were studied during the period from 1956 through 1973 based on labor histories retained at the maternity homes of the village of Mouslyumovo located on the Techa, and of the town of Kunashak. Both exposed and unexposed women were admitted to those maternity homes. In all, the outcomes of 2,460 pregnancies were studied (39). The following unfavorable outcomes of pregnancies were assessed: 1)incidence of spontaneous abortions, 2) incidence of stillbirths, 3) percentage of deaths in the early neonatal period (during the first week of life), and 4)incidence of fatal developmental defects. 

The most of the unfavorable outcomes of pregnancies were spontaneous abortions (from 2.7 to 9.8% in different groups: controls, with one parent exposed, with both parents exposed). The incidence of spontaneous abortions in the control group constituted from 2.1% to 4.2%, in when one or both parents were exposed it was estimated to be from 3.6% to 6.1%. The rate of stillbirth was 26.2 cases per 1000 (exactly 63 cases per 2405 neonates; twin births were included in the calculations) which is essentially higher than the respective value for the USSR in 1980 of 9.11 cases per 1000 (40). The rates of stillbirth were actually the same for both cases with exposed parents and for controls. The coefficient of total early neonatal mortality for all labors studied was about 2.99 cases per 1000 live birth: the respective value for children born to exposed parents was about 4.18, and for controls it was about 1.75 cases per 1000 live births.  According to the available information there were only 3 children with congenital developmental defects. These data do not allow to draw a definite conclusion. 

When both parents wereexposed the rate of unfavorable outcomes of pregnancies was slightly lower (6.1%) than in cases with one parent exposed (9.1%-11.7%). The rate of unfavorable outcomes of pregnancies for controls was about 5.8%. If we assume that a proportion of perinatal deaths can be attributed to radiation exposure and make a comparison with the controls of the group with one parent exposed, the doubling gonadal dose will be equal to 0.02 Sv If the doubling gonadal dose is estimated on the basis of loss of offspring in the groups with both parents exposed, its value will be much higher:0.42 Sv. Thus, the range of the doubling dose is large, and the value correspondingly uncertain. There is insufficient statistical power to draw reliable conclusions. 

 
a) b) 
Fig.13.Birth Rate per 1,000 for control group and exposed on the Techa River of Russian (a) and Tatar-Bashkir (b) population in Chelyabinsk region. 

An analysis of the birth rate does not provide directly the effect of radiation exposure because of changes in family planning practice. But birth rates do provide whether or not the process of population reproduction has been affected in an indirect way. They characterize the reproductive function in the exposed population. Birth coefficients were calculated based on the registry of exposed people and their offspring born in the period from 1950 through 1974 (41). Later than 1974 the number of cohort members of reproductive age had decreased considerably. By the time the presented study was completed the total offspring included 20,278 persons representing the first-generation and 3,411 persons representing the second-generation, i.e., in all, the progeny of parents and grandparents exposed to radiation on the Techa River. The unexposed population of two districts in Chelyabinsk Region (Krasnoarmeysky and Kunashaksky) which numbered about 78,000 in the 1950s was identified as a control group. Birth rate coefficients for controls were estimated based on data on birth of 43,000 people in the period 1950-1974. 

As shown in figures 13a and 13b, the highest birth rate was observed in the 1950s and then decreased  throughout the follow-up period up to its end (1974). This dependence was traced for both unexposed controls and for residents exposed on the Techa. The negative birth trend was typical of the entire population of Russia. It results primarily from the removal of the prohibition against abortion and the widening scope of family planning practice. Birth rate dynamics, shown in this study, are governed by the dependencies common to Russia as a whole, and it indicates the correctness of these calculations. On one hand, even if radiation affects the birth rate, it does not change essentially the general dynamic dependencies. Also figure 13b shows that the birth rate for the group composed of the ethnic Tatar and Bashkir population  is 1.5 times higher than that for the ethnic Russian population shown in figure 13a. This can be attributed to different national and religious traditions. Birth coefficients for the Tatar and Bashkir group reached 43 per 1000 in the period 1950-1959.  This increase is characteristic of families without family planning traditions, and of communities where no contraceptives are used by women, and abortions are regarded as a serious violation of moral principles. On the whole, the results of the study testify to the lack of a decrease in birth rate and, accordingly, to normalcy of reproductive function in persons whose doses to gonads ranged from 0.03 Sv (3 rem) to 1.3 Sv (130 rem). 

An analysis of the death rate and of causes of death for persons assigned to the offspring of exposed parents is an informative parameter used in assessment of potential genetic effects of radiation exposure. Mortality has been followed-up for 35 y since the onset of exposure and encompassed the period from 1950 through 1984 (42). Among 1,661 death cases of offspring of exposed parents born in 1950 and later, 1,330 cases are confirmed by death certificates retrieved from the regional ZAGS (Civil Registrar=s Office) archives. The analysis was made only for these cases.

The main cause of death in these cases was disease of the respiratory organs. Infectious diseases ranked second, and trauma third. However, the prevalence of one or other cause of death is to a great extent dependent on age. Thus, for infectious diseases and diseases of respiratory organs account for 62.4% of deaths among children under 1 y, congenital anomalies, neonatal diseases and ill-defined conditions in total account for 23.4%, and traumas for only 1.5%. For subjects aged 20 y and older, trauma ranked first among causes of death (72.6%). The most informative parameter is infant mortality, i.e. death rate among children under 1 y. There were 972 deaths among children under 1 y which is 58.8 % of the deaths among offspring. The most clear-cut dependence is represented by a decrease in infant death rate, which can be traced over the period from 1950 through the end of the study. By comparison, in Kunashaksky and Krasnoarmeysky districts of Chelyabinsk region among the offspring of both exposed and controls one child in ten died in the period 1950-1952 while in 1970-1974 the death rate decreased to 15-30 death cases per 1,000. This negative temporal trend was observed for both controls and children representing offspring of exposed parents. 

There was, however, a slight increase in the number of deaths from endogenous causes and ill-defined conditions among the progeny of exposed parents compared to controls. But this was against an overall increase of the incidence of the three classes of diseases (congenital abnormalities, diseases of neonates and ill-defined conditions). The death rate from these causes for the offspring of exposed parents (total groups of Slavs and Tartar-Bashkir population) made up 11.8 (5% CI:9.96-13.87), whereas for controls the respective value was 8.22 (95% CI: 7.92-9.77), P<0.05. 

Discussion and Conclusion

Mistakes in putting into operation the weapon-grade plutonium production plant "Mayak" caused contamination of surrounding territories. The most serious radiation accidents involved the release of liquid radioactive waste of the radiochemical plant directly into the river. As a result the river ecosystem was highly contaminated by radionuclides. It was the first tragic accident with environmental pollution by radionuclides in human history. Experience gained in fighting its dangerous consequences was of great help in later accidents in Kyshtym (1957) and Chernobyl (1986). 

About 28,000 residents of the riverside villages accumulated  doses of an average of 0.3 Sv. which significantly exceeds allowable dose-limit for the public of 0.05 Sv. A number of villages were evacuated. The villages on the contaminated territories, which are still inhabited, now have additional social problems. People whose exposure was estimated to be especially high receive social benefits. But this system of benefits is imperfect, and many exposed people feel it leads to injustice. This problem is now very difficult to solve, because the economic situation in Russia is unstable. 

The medical study of the exposed population started soon after the moment of irradiation by visiting teams of physicians. Systematic medical treatment, dose assessment and epidemiological studies began later when a special branch of the Moscow Biophysical Institute was organized in Chelyabinsk, which is now URCRM. Many unique materials and findings were obtained during more than forty years of study in the URCRM and declassified in 1990. A statistically significant  excess of 21 leukemia cases (out of 50) and a statistically insignificant excess of 30 solid cancers (out of 969) were seen in a cohort, about half the total population, of those permanent residents who were exposed to radiation on the banks of the Techa River. 

Data from the Techa River can address a number of important questions, that have not been measured in humans. These include the influence of exposure duration (dose rate) and the role of ethnic factors. The study of the peoples exposed as a result of Techa river contamination can provide unique contributions to the knowledge of the impact of radiation on humans and hence on radiation protection and safety.

Acknowledgements: This review, as with all reviews, would not have been possible without the extensive work and the approval of many people. Dmitry Burmistrov would like to thank the Richard Lounsbury Foundation and Harvard University for generous support. He is grateful to Dr. Marina Degteva, Dr. Nelli Safronova and Dr. Eugeniya Tolstykh for printed materials related to this review, and Dr. Elena Shihkina for her assistance. Mira Kossenko wishes to thank Dr. Angelina Guskova for extremely useful discussions on diagnostic of Chronic Radiation Sickness, and Dr. Dale Preston for his help in analysis of stochastic effects in the Techa River Cohort. Richard Wilson would like to thank the UCRM for their hospitality on many occasions. Many of the reviewed studies were supported by the U.S. - Russian Joint Coordinating Committee on Radiation Effects.

References

1Joint Norwegian-Russian Expert Group for Radioactive Contamination in the Northern Areas, Sources Contributing to Radioactive Contamination of the Techa River and Areas Surrounding the Mayak Production Association, Urals, Russia (October 1997). 

2V.N. Novikov, V.S. Tolstikov, The Mistery of "Sorokovka" (Ekhaterinburg, Izd. "Uralskiy rabochiy", 1995, in Russian). 

3Radiation Contamination of the Territory of Chelybinsk Oblast (Chelyabinsk,Yuzhno-uralskoe Knizhnoe Izdatelstvo,1992, in Russian). 

4J.R.Trabalka,L.D.Eyman,F.L.Packer et al "Another Perspective on the 1958 Soviet Nuclear Accident", Nuclear Safety, 20: (no. 2): 206-210 (1979) 

5Nikipelov,B.V., Lizlov, A.F. and Koshurnikova N.A. "Experience with the First Soviet Nuclear Installation: Irradiation Doses and Personnel Health" (in Russian) Priroda (Nature) February 1990

6M.I. Vorobiova, M.O. Degteva, D.S. Burmistrov, N.G. Safronova, V.P. Kozheurov, L.R. Anspaugh, B.A. Napier, "Review of Historical Monitoring Data on Techa River Contamination", Health Physics 76(6) : 605-618 (1999)

7D.I. Ilyin, Migration of Radioactive Materials from Open Reservoirs, Doctoral thesis, (Moscow, 1956, in Russian). 

8A.N. Marey, Sanitary Consequences of the Releases to Surface Waters of Radioactive Wastes of Atomic Industry Enterprises, Doctoral thesis, (Moscow, 1959, in Russian). 

9M.I. Vorobiova, M.O. Degteva "Simple Model of Radionuclide Transport in the Techa River Contaminated by the Mayak Releases", Health Physics, 77 (2): 142-149 (1999)

10D.S. Burmistrov, M.I. Vorobiova, M.O. Degteva, I. Linkov, R. Wilson, "Radioactive Contamination of the Techa River: Environmental Records and Multimedia Modelling", Conference Proceedings Vol. 59 "Nuclear Data for Science and Technology" (part II) : 1376-1380 (Bologna: Societa Italiana di Fisica, 1997). 

11M.O. Degteva, V.P. Kozheurov, M.I. Vorobiova, "General Approach to Dose Reconstruction in the Population Exposed as a Result of the Release of Radioactive Wastes into the Techa River", Sci. Total Environ, 142: 49-62 (1994). 

12M.O. Degteva, V.P. Kozheurov, D.S. Burmistrov, M.I. Vorobiova, V.V. Valchuk, N.G. Bougrov, E.A. Shishkina "An Approach to Dose Reconstruction for the Urals Population", Health Physics, 71(1): 71-76 (1996). 

13A.A. Romanyukha, E.A. Ignatiev, M.O. Degteva, V.P. Kozheurov, A. Wieser, P. Jacob, "Radiation Doses from Ural Region," Nature 381: 199-200 (1996). 

14A.A. Romanyukha, M.O. Degteva, V.P. Kozheurov, A. Wieser, P. Jacob, E.A. Ignatiev, M.I. Vorobiova, "Pilot Study of the Urals Population by Tooth Electron Paramagnetic Resonance Dosimetry", Radiat. Environ. Biophys., 35: 305-310 (1996). 

15A. Wieser, A.A. Romanyukha, M.O. Degteva, V.P. Kozheurov, G. Petzold, "Tooth Enamel as a Natural Beta Dosemeter for Bone Seeking Radionuclides", Rad. Prot. Dosimetry, 65, (Nos. 1-4): 413-416 (1996). 

16A.A. Romanyukha, M.O. Degteva, V.P. Kozheurov, A. Wieser, P. Jacob, M.I. Vorobiova, E.A. Ignatiev, E.A. Shishkina, A.A. Koshta, "Retrospective Evaluation of External Component of Individual Doses for Techa Riverside", IRPA9: 1996 International Congress on Radiation Protection. Vienna, Austria, April 14-19, vol.3: 111-113 (1996). 

17A. Wieser, A.A. Romanyukha, K. Bunzl, W. Kracke, M.O. Degteva, V.P. Kozheurov, "The Absorbed Dose in Tooth Enamel by Sr-90 Body Burden", IRPA9: 1996 International Congress on Radiation Protection. Vienna, Austria, April 14-19, vol.1: 425-435 (1996). 

18N.G. Bougrov, V.K. Vlasov, O.V. Kiryuhin, N.L. Fatculbayanova, "Thermoluminescence Measurements of Ceramic Samples from Accidentally Polluted Territory of Southern Urals", Radiation Measurements, 24, (no.4): 493-498 (1995). 

19H.Y. Goksu, L.M. Heide, N.G. Bougrov, A.R. Dalheimer, R. Meckbach, P. Jacob, "Depth-dose Distribution in Bricks Determined by Thermoluminescence and by Monte-Carlo Calculation for External -Dose Reconstruction", Appl. Radiat. Isot., 47,( no.4): 433-440 (1996). 

20N.G. Bougrov, H.Y. Goksu, E. Haskell, M.O. Degteva, R. Meckbach, P. Jacob, "Issues in the Reconstruction of Environmental Doses on the Basis of Thermoluminescence Measurements in the Techa Riverside", Health Physics, 75: 1-10 (1998). 

21V.P. Kozheurov, "SICH-9.1- A Unique Whole-body Counting System for Measuring 90Sr via Bremsstrahlung: the Main Results from a Long-term Investigation of the Techa River Population", Sci. Total Environ, 142(1-2): 37-48 (1994). 

22M.O. Degteva, V.P. Kozheurov, "Age-dependent Model for Strontium Retention in Human Bone", Rad. Prot. Dosimetry, 53: 229-234 (1994). 

23V.P. Kozheurov, M.O. Degteva "Dietary Intake Evaluation and Dosimetric Modelling for the Techa River Residents Based on in vivo Measurements of Strontium-90 in Teeth and Skeleton", Sci. Total Environ., 142: 63-72 (1994). 

24E.I. Tolstykh, V.P. Kozheurov, O.V. Vyushkova, M.O. Degteva, "Analysis of Strontium Metabolism in Humans on the Basis of the Techa River Data", Radiat. Environ. Biophys., 36: 25-29 (1997). 

25M.O. Degteva, V.P. Kozheurov, E.I. Tolstykh, "Retrospective Dosimetry Related to Chronic Environmental Exposure", Rad. Prot. Dosimetry, 79: 155B160 (1998). 

26E.I Tostykh, M.O. Degteva, V.P. Kozheurov, D.S. Burmistrov, "Strontium Transfer from Maternal Skeleton to the Fetus Estimated on the Basis of the Techa River Data",. Rad. Prot. Dosimetry, 79: 307-310 (1998).

27NCRP (National Council on Radiation Protection and Measurements). "Recommendations on limits for exposure to ionizing radiation," NCRP Report, No 91 Bethesda, MD, NCRP(1987) 

28ICRP (International Commission on Radiological Protection).Age-dependent Doses to Members of the Public from Intake of Radionuclides:Part 2: Ingestion Dose Coefficients, Annals of the ICRP, v. 23, no. 3/4, ICRP Publication 67, Oxford, Pergamon Press (1993) 

29Berkovsky, V "Computer System for Reconstruction and Prediction of Human Internal Doses", Kiyev, Radiation Protection Institute, Doctoral Thesis (1992) (in russian) 

30ICRP (International Commission on Radiological Protection)"Conversion Coefficients for use in Radiological Protection against External Irradiation", Annals of the ICRP, v. 26, no. 3/4, ICRP Publication 74, Oxford, Pergamon Press (1995). 

31K.F.Eckerman, J.C.Ryman "External exposure to radionuclides in air, water and soil",Federal Guidance Report No 12, EPA 402-R-93-081,Oak Ridge, TN, Oak Ridge National Laboratory (1993) 

32A.K.Guskova, G.D.Baysogolov. "Radiation diseases in human". Moscow, 1971, 387p..(in Russian) 

33N.D.Okladnikova., V.S.Pesternikova , M.V.Sumina, N.Ya.Kabasheva, T.V. Azizova. "Human Chronic Radiation Sickness caused by External Gamma-irradiation, long term stage" 

Bulletin of Academy of Medical Sciences (1991), 2:22-26 (in Russian). 

34N.D.Okladnikova, V.S.Pesternikova, M.V.Sumina, V.N.Doshchenko. "Occupational Diseases from Radiation Exposure at the first Nuclear Plant in the USSR". The Science of the Total Environment 142 9-17 1994 (in Russian) 

35ICRP (International Commission on Radiological Protection). Nonstochastic Effects of Ionizing Radiation. ICRP Publication 41, Oxford, Pergamon Press (1984). 

36M.M.Kossenko, M.O.Degteva. "Cancer mortality and risk evaluation for the Techa river population." Sci Total Environ. 142:73-89 (1994). 

37M.M. Kossenko, M.O. Degteva, O.V. Vyushkova, D.L. Preston, K. Mabuchi, V.P. Kozheurov, "Issues in the Comparison of Risk Estimates for the Population in the Techa River Region and Atomic Bomb Survivors", Rad. Res., 148: 54B80 (1997). 

38Y.Shimizu, H.Kato, W.J.Schull, D.L.Preston, Sh.Fujita,D.A.Pierce. "Studies of the mortality of A-bomb survivors 9. Mortality, Part I: Comparison of risk coefficient for site specific cancer mortality based on the DS86 and T65DR shielded kerma and organ doses." Rad Res 118:502-524(1989). 

39M.M.Kossenko, P.I.Izhevsky, M.O.Degteva, A.V.Akleyev, O.V.Vyushkova. "Pregnancy outcome and early health status of children born to the Techa River population." SciTotal Environ. 142:91-100(1994). 

40Population of the USSR in 1987. Statistical data overview. Moscow (1988)(in Russian). 

41M.M.Kossenko. "Birth Rate for Population Exposed to Radiation in the Techa Riverside Villages. Meditsinskaya Radiologiya i Radiatsionnaya Bezopastnost, 40(6):25-29(1996)(in Russian). 

42M.M.Kossenko. "Studies of Death Rate among Offspring of Exposed Parents." Meditsinskaya Radiologiya i Radiatsionnaya Bezopasnost, 41(3): 4-10 (1996) (in Russian)