The "source term" is a technical expression used to describe the accidental release of radioactive material from a nuclear facility to the environment. Not only are the levels of radioactivity released important, but also their distribution in time as well as their chemical and physical forms. The initial estimation of the Source Term was based on air sampling and the integration of the assessed ground deposition within the then Soviet Union. This was clear at the IAEA Post-Accident Review Meeting in August 1986 (IA86), when the Soviet scientists made their presentation, but during the discussions it was suggested that the total release estimate would be significantly higher if the deposition outside the Soviet Union territory were included. Subsequent assessments support this view, certainly for the caesium radionuclides (Wa87, Ca87, Gu89). The initial estimates were presented as a fraction of the core inventory for the important radionuclides and also as total activity released.
In the initial assessment of releases made by the Soviet scientists and presented at the IAEA Post-Accident Assessment Meeting in Vienna (IA86), it was estimated that 100% of the core inventory of the noble gases (xenon and krypton) was released, and between 10 and 20% of the more volatile elements of iodine, tellurium and caesium. The early estimate for fuel material released to the environment was 3 ± 1.5% (IA86). This estimate was later revised to 3.5 ± 0.5% (Be91). This corresponds to the emission of 6 t of fragmented fuel.
The IAEA International Nuclear Safety Advisory Group (INSAG) issued in 1986 its summary report (IA86a) based on the information presented by the Soviet scientists to the Post-Accident Review Meeting. At that time, it was estimated that 1 to 2 exabecquerels (EBq) were released. This did not include the noble gases, and had an estimated error of ±50%. These estimates of the source term were based solely on the estimated deposition of radionuclides on the territory of the Soviet Union, and could not take into account deposition in Europe and elsewhere, as the data were not then available.
However, more deposition data (Be90) were available when, in their 1988 Report (UN88), the United Nations Scientific Committee on the Effects of Atomic Radiation (UNSCEAR) gave release figures based not only on the Soviet data, but also on worldwide deposition. The total 137Cs release was estimated to be 70 petabecquerels (PBq) of which 31 PBq were deposited in the Soviet Union.
Later analyses carried out on the core debris and the deposited material within the reactor building have provided an independent assessment of the environmental release. These studies estimate that the release fraction of 137Cs was 20 to 40% (85 ± 26 PBq) based on an average release fraction from fuel of 47% with subsequent retention of the remainder within the reactor building (Be91). After an extensive review of the many reports (IA86, Bu93), this was confirmed. For 131I, the most accurate estimate was felt to be 50 to 60% of the core inventory of 3 200 PBq. The current estimate of the source term (De95) is summarised in Table 1.
From the radiological point of view, 131I and 137Cs are the most important radionuclides to consider, because they are responsible for most the radiation exposure received by the general population.
The release pattern over time is well illustrated in Figure 3 (Bu93). The initial large release was principally due to the mechanical fragmentation of the fuel during the explosion. It contained mainly the more volatile radionuclides such as noble gases, iodines and some caesium. The second large release between day 7 and day 10 was associated with the high temperatures reached in the core melt. The sharp drop in releases after ten days may have been due to a rapid cooling of the fuel as the core debris melted through the lower shield and interacted with other material in the reactor. Although further releases probably occurred after 6 May, these are not thought to have been large.
Table 1. Current estimate of radionuclide releases during the Chernobyl accident (modif. from 95De)
Core inventory on 26 April 1986 |
Total release during the accident |
|||
Nuclide |
Half-life |
Activity (PBq) |
Percent of inventory |
Activity (PBq) |
133Xe |
5.3 d |
6 500 |
100 |
6500 |
131I |
8.0 d |
3 200 |
50 - 60 |
~1760 |
134Cs |
2.0 y |
180 |
20 - 40 |
~54 |
137Cs |
30.0 y |
280 |
20 - 40 |
~85 |
132Te |
78.0 h |
2 700 |
25 - 60 |
~1150 |
89Sr |
52.0 d |
2 300 |
4 - 6 |
~115 |
90Sr |
28.0 y |
200 |
4 - 6 |
~10 |
140Ba |
12.8 d |
4 800 |
4 - 6 |
~240 |
95Zr |
1.4 h |
5 600 |
3.5 |
196 |
99Mo |
67.0 h |
4 800 |
>3.5 |
>168 |
103Ru |
39.6 d |
4 800 |
>3.5 |
>168 |
106Ru |
1.0 y |
2 100 |
>3.5 |
>73 |
141Ce |
33.0 d |
5 600 |
3.5 |
196 |
144Ce |
285.0 d |
3 300 |
3.5 |
~116 |
239Np |
2.4 d |
27 000 |
3.5 |
~95 |
238Pu |
86.0 y |
1 |
3.5 |
0.035 |
239Pu |
24 400.0 y |
0.85 |
3.5 |
0.03 |
240Pu |
6 580.0 y |
1.2 |
3.5 |
0.042 |
241Pu |
13.2 y |
170 |
3.5 |
~6 |
242Cm |
163.0 d |
26 |
3.5 |
~0.9 |
Fifteen years on, the estimation made in 1996 is still valid. However the results presented in Table 1 are incomplete with respect to the release of the short-lived radionuclides (132I and 135I). In the UNSCEAR 2000 report (UN00), the overall releases of short-lived radioiodines are presented on the basis of early and re-estimated informations (Ab86, Iz90); they are found to be substantially lower than those of 131I (1760 PBq), 1040 PBq, 910, 25 and 250 respectively for 132I, 133I, 134I and 135I, 132I is assumed to be in radioactive equilibrium with 132Te.
Figure 3. Daily release rate of radioactive substances into the atmosphere (modif. from IA86a)
The estimated daily releases of 131I during the accident is given in Table 2.
Table 2. Daily releases of 131I
Day of release
|
Daily releases (PBq)
|
26 April
|
704
|
27 April
|
204
|
28 April
|
150
|
29 April
|
102
|
30 April
|
69
|
1 May
|
62
|
2 May
|
102
|
3 May
|
107
|
4 May
|
130
|
5 May
|
130
|
Total
|
1760
|
Although the releases were considerably reduced on 5 and 6 May (days 9 and 10) after the accident), continuing low-level releases occurred in the following week and for up to 40 days after the accident, particularly on 15 and 16 may, attributable to continuing outbreaks of fires or to hot areas in the reactor. These later releases can be correlated with increased concentrations of radionuclides in air measured at Kiev and Vilnius.
The release of radioactive material to the atmosphere consisted of gases, aerosols and finely fragmented fuel. Gaseous elements, such as krypton and xenon escaped more or less completely from the fuel material. In addition to its gaseous and particulate form, organically bound iodine was also detected. The ratios between the various iodine compounds varied with time. As mentioned above, 50 to 60% of the core inventory of iodine was thought to have been released in one form or another. Other volatile elements and compounds, such as those of caesium and tellurium, attached to aerosols, were transported in the air separate from fuel particles. There were only a few measurements of the aero-dynamic size of the radioactive particle releases during the first days of the accident. The activity distribution of the particle size was found to be well represented as the superposition of two log-normal functions, one with an activity median aerodynamic diameter (AMAD) ranging from 0.3 to 1.5 µm and another with an AMAD of 10 µm. The larger particles contained about 80-90% of the activity of non-volatile radionuclides such as 95Zr, 95Nb, 140La, 144Ce and transuranium elements embedded in the uranium matrix of the fuel.
Unexpected features of the source term, due largely to the graphite fire, were the extensive releases of fuel material and the long duration of the release. Elements of low volatility, such as cerium, zirconium, the actinides and to a large extent barium, lanthanum and strontium also, were embedded in fuel particles. Larger fuel particles were deposited close to the accident site, whereas smaller particles were more widely dispersed. Other condensates from the vaporised fuel, such as radioactive ruthenium, formed metallic particles. These, as well as the small fuel particles, were often referred to as "hot particles", and were found at large distances from the accident site (De95). Typical activities per hot-particles are 0.1-1 kBq for fuel fragments and 0.5-10 kBq for ruthenium particles, the diameters being about 10 µm to be compared with sizes of 0.4-0.7 µm for the particles associated with the activities of 131I and 137Cs (De88, De91).
Radioactive contamination of the ground was found to some extend in practically every country of the northern hemisphere. European commission published on the basis of local measurements an atlas of contamination in Europe (De98).
Within the former Soviet Union
During the first 10 days of the accident when important releases of radioactivity occurred, meteorological conditions changed frequently, causing significant variations in release direction and dispersion parameters. Deposition patterns of radioactive particles depended highly on the dispersion parameters, the particle sizes, and the occurrence of rainfall. The largest particles, which were primarily fuel particles, were deposited essentially by sedimentation within 100 km of the reactor. Small particles were carried by the wind to large distances and were deposited primarily with rainfall. The radionuclide composition of the release and of the subsequent deposition on the ground also varied considerably during the accident due to variations in temperature and other parameters during the release. 137Cs was selected to characterise the magnitude of the ground deposition because (1) it is easily measurable, and (2) it was the main contributor to the radiation doses received by the population once the short-lived 131I had decayed. However, during the first weeks after the accident, most of the activity deposited on the ground consisted of short-lived radionuclides, of which 131I was the most important radiologically. All the maps established in the former Soviet Union were mainly based on the limited number of measurement of 131I, and they use 137Cs measurements as a guide. These maps must be regarded with caution, as the ratio of 131I to 137Cs deposition densities was found to vary over a large range in Belarus, 5 to 10, this ratio has been not seriously studied in many countries.
The three main spots of contamination resulting from the Chernobyl accident have been called the Central, Bryansk-Belarus, and Kaluga-Tula-Orel spots (Figure 4, pages 49-50). The Central spot was formed during the initial, active stage of the release predominantly to the West and North-west (Figure 5, pages 51-52). Ground depositions of 137Cs of over 40 kilobecquerels per square metre [kBq/m2] covered large areas of the Northern part of Ukraine and of the Southern part of Belarus. The most highly contaminated area was the 30-km zone surrounding the reactor, where 137Cs ground depositions generally exceeded 1 500 kBq/m2 (Ba93).
Areas of high contamination of 137Cs occurred thoughout the far zone, depending primarly on rainfall at the time the plume passed over. The Bryansk-Belarus spot, centered 200 km to the North-northeast of the reactor, was formed on 28-29 April as a result of rainfall on the interface of the Bryansk region of Russia and the Gomel and Mogilev regions of Belarus. The ground depositions of 137Cs in the most highly contaminated areas in this spot were comparable to the levels in the Central spot and reached 5 000 kBq/m2 in some villages (Ba93).
The Kaluga-Tula-Orel spot in Russia, centered approximately 500 km North-east of the reactor, was formed from the same radioactive cloud that produced the Bryansk-Belarus spot, as a result of rainfall on 28-29 April. However, the levels of deposition of 137Cs were lower, usually less than 600 kBq/m2 (Ba93).
In addition, outside the three main hot spots in the greater part of the European territory of the former Soviet Union, there were many areas of radioactive contamination with 137Cs levels in the range 40 to 200 kBq/m2. Overall, the territory of the former Soviet Union initially contained approximately 3 100 km2 contaminated by 137Cs with deposition levels exceeding 1 500 kBq/m2; 7 200 km2 with levels of 600 to 1 500 kBq/m2; and 103 000 km2 with levels of 40 to 200 kBq/m2 (US91).
Figure 4: Deposition of Caesium-137 in Belarus
Figure 5: Deposition of Caesium-137 in Ukraine
The principal physico-chemical form of the deposited radionuclides are: dispersed fuel particles, condensation-generated particles, and mixed-type particles. The distribution in the nearby contaminated zone (<100km) reflected the radionuclide composition of the fuel and differs from that in the far zone (>100km to 2 000 km). Large particles, deposited in the near zone, contained fuel (U, Pu) refractory elements (Zr, Mo, Ce and Np) and intermediate elements (Ru, Ba, Sr). The volatile elements (I, Te and Cs) in the form of condensation-generated particles, were more widely disperded in the far zone.
Deposition of 90Sr was mostly in the near zone of the accident as for 239Pu; the only area with plutonium exceeding 4 kBq m-2 was located within the 30-km zone, in the Gomel-Mogilev-Briansk area. (De98)
Outside the former Soviet Union
Radioactivity was first detected outside the Soviet Union at a Nuclear Power station in Sweden, where monitored workers were noted to be contaminated. It was at first believed that the contamination was from a Swedish reactor. When it became apparent that the Chernobyl reactor was the source, monitoring stations all over the world began intensive sampling programmes.
The radioactive plume was tracked as it moved over the European part of the Soviet Union and Europe (Figure 6). Initially the wind was blowing in a Northwesterly direction and was responsible for much of the deposition in Scan-di-navia, the Netherlands and Belgium and Great Britain. Later the plume shifted to the South and much of Central Europe, as well as the Northern Medi-terranean and the Balkans, received some deposition, the actual severity of which depended on the height of the plume, wind speed and direction, terrain features and the amount of rainfall that occurred during the passage of the plume.
Most countries in Europe experienced some deposition of radionuclides, mainly 137Cs and 134Cs, as the plume passed over the country. In Austria, Eastern and Southern Switzerland, parts of Southern Germany and Scandinavia, where the passage of the plume coincided with rainfall, the total deposition from the Chernobyl release was greater (exceeding 37 kBq m-2, with an extensive deposition in a 2-4 km2 area in Sweden within the commune of Gävle (exceeding 185 kBq m-2) (Ed91) than that experienced by most other countries, whereas Spain, France and Portugal experienced the least deposition. For example, the estimated average depositions of 137Cs in the provinces of Upper Austria, Salzburg and Carinthia in Austria were 59, 46 and 33 kBq/m2 respectively, whereas the average 137Cs deposition in Portugal was 0.02 kBq/m2 (Un88). It was reported that considerable secondary contamination occurred due to resuspension of material from contaminated forest. This was not confirmed by later studies.
Figure 6. Areas covered by the main body of the radioactive cloud on various days during the release
While the plume was detectable in the Northern hemisphere as far away as Japan and North America, countries outside Europe received very little deposition of radionuclides from the accident. No deposition was detected in the Southern hemisphere by the surveillance networks of environmental radiation (Un88).
Behaviour of deposited radionuclides
The environmental behaviour of deposited radionuclides depends on the physical and chemical characteristics of the radionuclides and on the type of fallout, dry or wet, the size and shape of particles and the environment. For example, particles produced by gas-to-particle conversion through chemical reactions, nucleation and condensation as well as coagulation have a large specific surface and are generally more soluble than explosion generated particles, such as large fuel particles particles generated by mechanical processes like explosion of fuel.
For short-lived radionuclides, such as iodine isotopes, the main pathway of exposure of humans is the transfer of the amount deposited on leafy vegetables that are consumed within a few days, or on pasture grass that is grazed on by cows or goats, giving rise to the contamination of milk. Long term behaviour is not relevant, because 131I has a physical half-life of only 8 days.
Radionuclides deposited on soil migrate downwards and reach the part of soil containing roots, and the time of residence in this area would partly determinate migration to vegetation. Observations strongly suggest that the migration profiles are established very early after contamination under the influence of the early conditions prevailing immediately after contamination, such as soil moisture and first rain events, which may be the paramount in determining the extent to which radionuclides will penetrate in depth (Br00). The vertical migration of 137Cs and 90Sr in soil of different type of meadows has been rather slow, and the greater fraction of radionuclides is still contained in the upper soil layers (0-10 cm). The effective half-time of clearance from root layer has been estimated to range from 10 to 25 years for 137Cs. Early after the accident the transfer coefficients of 137Cs to plant decreased by 1.5 to 7 times but later from the observed persisting mobility of radiocaesium, and particularly the increase in effective ecological half lives tending towards the physical decay rate of 137Cs, it now seems that the sorption-desorption process of radiocaesium in soils and sediments is tending towards a reversible steady-state and the practical consequences for plant contamination in the environment is that foodstuffs will remain contaminated for much longer than initially expected (Sm00).
The contribution of aquatic pathways to the dietary intake of 137Cs and 90Sr is usually quite small, However the relative importance, in comparison to terrestrial pathways, may be high in some lakes of Scandinavia and in Russia. In mountains we can observe by run-off some reconcentrations of the radioactivity in lower areas and for example in the South part of the French alps the 137Cs contamination was in 1992 about 20 000 Bq.m-3, corresponding to an activity of 1 760 Bq.kg-1 in soil samples. In some specific, small areas, (only a fraction of a square meter) hot spots have been measured at 55 800 Bq.kg-1 in 1992, 314 000 Bq.kg-1 in 1995, and500 000 Bq.m-2 in 2000. These hot spots are the consequences of the runoff of melting water coming from snow which fell after the 1986 contamination of the upper part of the mountain. These hot spots have been found in small basins lower in the forest or under larchs where snow accumulates. However these hot spots being of small surface (cm2 to m2) are offwalking tracks, pose little risk of irradiation for hikers. For example, it has been estimated that a hiker would receive about 0.001 mSv during a 4 hour rest in the vicinity of such a hot-spot. (Ma 97). These hot-spots will remain active for several decades, their decay following the physical half-life of 137Cs.
Drinking water in the affected areas is weakly contaminated, less than 1Bq of 137Cs or 90Sr per litre. The mean annual activity of 137Cs in the water of Pripiat river and in the Kiev reservoirs has now stabilised within a range of from 1 to 0.2 Bq.l-1 (Bq per litre), ten time higher than the values obtained before the 1986 accident. The 90Sr activity of the Pripiat river is sometimes higher than authorised levels for drinking water (2 Bq.l-1) due to meteorological conditions, rains and floods.
From 26 April to 6 March 1986, during the period of releases, the highest levels of radioactivity measured in water of the Pripiat river was of the order of 100 000 Bq/l, principally from 131I. The activity in the Pripiat declined to around a few thousand Bq/l by mid-May 1986, and to 200 Bq/l in June 1986. From the end of November 1986 to the beginning of 1987, the activity in the Pripiat was rarely measured above 40 Bq/l. From 1987 on, 137Cs and 90Sr were the only radionuclides measured in significant quantities. Since 1988, 90Sr is the radioelement of highest concentration in the waters of the Pripiat.
The chemical form of the 137Cs that was deposited is fairly insoluable, and is not quickly extracted from soil by surface runoff water. Most of the 137Cs transferred to the Pripiat river by runoff water came from the 30 km exclusion zone. As a result of this low solubility, only 1 to 5% of the initial 137Cs activity reached the Black Sea, the rest accumulating in various reservoirs of the Dniepr, of which more than half stayed in the Kiev reservoir.
The activity of 90Sr in Pripiat river water is a few times higher than the level authorised for human consumption, 2 Bq/l. During flooding in the fall of 1988, 90Sr activity reached 9.6 Bq/l. As a result of significant blockage of water during particularly high flooding, 90Sr concentrations reached 12.2 Bq/l in January 1991, and 5.9 Bq/l in February 1994.
In 1986, during the accident and the following months, the 137Cs activity released into the Dniepr was estimated to be 66 TBq. Subsequently, leaching from soils by surface water and floods resulted in a measurable increase of radionuclide concentrations in the Pripiat river. The following Table 3 indicates the respective influxes of 137Cs and 90Sr in the Pripiat between 1986 and 1998, as well as the resulting water concentrations.
The cities of Kiev, Kremenchug et Kahovsk are partly fed by Dniepr reservoirs (see Figure 7). The table shows the annual average levels of 137Cs and 90Sr in the Pripiat river from 1986 to 1998 (Po01), but it could be observed peaks of activity ten time higher during floods.
Graphs in Figure 7 show the evolution of 137Cs and 90Sr concentrations in these reservoirs from 1986 to 1998.
It has been shown that forests can deliver large radiation doses through the consumption of berries, mushrooms and game, but also through the industrial use of forest products. Radiological consequences result from energy production using radioactively contaminated biofuels from forests in the north of Europe and use of waste products or ashes and their recycling back to the forest as fertilizer.
On the forest podzolic soils, migration of 137Cs is pronounced, with increased amounts in the mineral layers ten years after aerial distribution. More than a decade after Chernobyl accident, the total inventory is still rising in pine trees of boreal forests. There is almost no 137Cs loss via runoff water from boreal forest ecosystems except from the wetter portions of bogs.
Figure 7. Possible groundwater flow directions in the Dniepr basin
137Cs and 90Sr concentrations in Bq/L-1, in the reservoirs of Kiev, Kremenchug and Kahkovka. Black area reresent activities entering in the reservoirs, white areas activities leaving reservoirs (From Po01)
Table 3: Evolution of average radioactivity in the Pripiat river since the accident in 1986 (From Poïkaprpov and Robeau, 2001)
Influx of 137Cs (TBq/a) |
Average spectrum activity of 137Cs in the Pripiat river (Bq/l) |
Influx of 90Sr (TBq/a) |
Average spectrum activity of 90Sr in the Pripiat river (Bq/l) |
|
1986 |
66,2 |
6,95 |
27,6 |
2,9 |
1987 |
12,8 |
1,65 |
10,4 |
1,34 |
1988 |
9,48 |
0,73 |
18,7 |
1,44 |
1989 |
6,44 |
0,521 |
8,97 |
0,725 |
1990 |
4,63 |
0,359 |
10,1 |
0.783 |
1991 |
2,9 |
0,208 |
14,4 |
1,033 |
1992 |
1,92 |
0,206 |
4,14 |
0,445 |
1993 |
3,48 |
0,208 |
14,2 |
0,838 |
1994 |
2,96 |
0,197 |
14,2 |
0,946 |
1995 |
1,15 |
0,11 |
3,4 |
0,326 |
1996 |
1,3 |
0,129 |
3,42 |
0,340 |
1997 |
1,7 |
0,158 |
2,68 |
0,25 |
1998 |
2,95 |
0,137 |
6,37 |
0,296 |
More than 16 years after the accident, only 2 to 3% of the deposited radioactivity still remains in the aerial part of the vegetation.
Since the accident, wood marketing has become regulated. Depending upon the intended use of harvested wood regulatory levels vary from 740 to 11 000 Bq 137Cs kg-1, which result in 30% of the Pine trees in the exclusion zone not being harvestable.
At this stage in time, the transfer of material by resuspension from more to less contaminated areas is not significant. The classical farming practices, mechanical treatment such as ploughing and mulching and the use of fertilisers are efficient countermeasures.
However, one year after the accident a storm resuspended deposited radioactivity in the exclusion zone, and the radioactivity of air in the Pripiat city increased by a factor of 1 000 and reached 300 Bq.m-3. Fires in forests have also led to increases of radioactivity. In 1992, in the vicinity of exclusion zone, radioactivity due to forest fires reached 20 Bq.m-3 for beta emitters and 70 mBq.m-3 from plutonium isotopes. Monitoring stations far from these zones registered some peaks of radioactivity.
It can be stated that there is now a fairly accurate estimate of the total radioactivity release, and the last years have strengthened previous evaluations. The duration of the release was unexpectedly long, lasting more than a week with two periods of intense release. Another peculiar feature was the significant emission (about 4%) of fuel material which also contained embedded radionuclides of low volatility such as cerium, zirconium and the actinides. The composition and characteristics of the radioactive material in the plume changed during its passage due to wet and dry deposition, decay, chemical trans-formations and alterations in particle size. The area affected was particularly large due to the high altitude and long duration of the release as well as changes in wind direction. However, the pattern of deposition was very irregular, and significant deposition of radionuclides occurred where the passage of the plume coincided with rainfall. Although all the Northern hemisphere was affected, only territories of the former Soviet Union and part of Europe experienced contamination to a significant degree. The environmental behaviour of deposited radionuclides is increasingly well known. More than sixteen years after the accident, radionuclides are still in the first layers of soils, maintaining a transfer to plants, particularly mushrooms, berries and forest products. Moreover the change of speciation of 137Cs in some soils has led to the fact that foodstuffs will remain contaminated for much longer than initially expected (Sm00). With the exception of some water tables, the contamination of environment is very well known. Contamination levels in soils decrease only slowly, mostly by transfer to plants. Most of the decrease in the coming years will be at only the rate of the physical half-life of 137Cs.