SCOPE 50 - Radioecology after Chernobyl

7.1 INTRODUCTION

The authorized disposal of radioactive wastes into the environment, or the accidental release of radionuclides from a nuclear facility have, as an inevitable consequence, the potential for increasing the radiation exposure not only of man, but also of domestic animals and native populations of wild organisms. For authorized disposals, controls are exercised to reduce the exposure of critical groups of humans to levels as low as reasonably achievable and, in any event, to maintain them (on average) within the appropriate recommended annual dose limit (at present 1 mSv a^-1 (ICRP, 1985)). The received wisdom is that such measures are likely to provide adequate protection for other species, although not necessarily for individual members of those species (ICRP, 1977, 1991). The ICRP have provided no evidence to support this belief, and no standards have been recommended for dose rate limits which would provide adequate protection for organisms other than man. This position clearly rests on a number of assumptions:

1. that humans occupy, or otherwise utilize the resources of, the environment contaminated by the waste to a similar extent as do populations of other organisms; 
2. that there are no mechanisms or pathways for the accumulation of radionuclides in some environmental compartment which lead to much-enhanced exposure of endemic populations, but which have no relevance to man; and
3. that other organisms are not markedly more radiosensitive than man.

The first two assumptions relate to the likelihood and extent of the radiation exposure, i.e. the problem of dosimetry, and the third concerns the radiobiological response, i.e. the assessment of the potential consequential effects.

In terms of making an assessment of the possible environmental impact of a disposal practice or a potential accident it is logical to deal with the problems in this order, i.e. radiation dosimetry and then radiobiological response. It should be recognized, however, that there is feedback between these two aspects of the overall problemthe assessment of the potentially damaging impact on the environmentwhich dictate the nature and relative importance of the information required in each case.

For example, for controlled radioactive waste disposals it is, in general, reasonable to conclude that the concomitant exposure of wild organisms will be of a similar magnitude to that experienced by humans, i.e. low dose rate and protracted in time. With the assumption that wild organisms are not more radiosensitive than humans, this means that lethal effects are unlikely to be a problem. This conclusion does, however, expose the existence of different perceptions of the risks resulting from the exposure of humans and those for all other organisms. For humans, the practice of radiological protection focuses on the individual for whom the risk of health effects arising from radiation exposure should be as low as reasonably achievable and, in any event, kept below some limiting value which is agreed to be acceptable. For the great majority of other species, it is the population (undefined at present, but see below) which would be considered the appropriate object for protection, with the fate of individuals usually being of much lesser concern; exceptions might be individuals of rare or endangered wild species or valuable domestic animals. This different perspective immediately alters the radiation responses which need to be considered for impact assessment and the provision of adequate protection. The survival of a population requires that its ability to maintain itself, through reproduction, within the normal range of variability and in the face of all the other vicissitudes of the natural environment, should not be affected. Notwithstanding these conclusions, it must be stressed that there can be no effects due to increased radiation exposure at the population level (or, indeed, at any higher community or ecosystem level) without the appearance of clear responses in individual organisms (Haux and Forlin, 1988). Thus it is the processes of gametogenesis in individuals and embryonic development, within the population, which are important and become the targets of dosimetric attention (IAEA, 1988, 1992b; NCRP, 1992).

Accident situations are likely to be different in a number of respects. It is probable that the release would be restricted in time and the magnitude of the resulting environmental dose rates, and their variation over time, could be much greater than would be the case for authorized disposals. For example, if, as was the case at the Chernobyl power plant (UNSCEAR, 1988b), the release were to consist of large quantities of short-lived radionuclides and, relatively, much smaller quantities of long-lived radionuclides then the dose rates in the immediate vicinity would be high initially but would fall fairly rapidly to much lower values, which would decline rather slowly due to both physical decay and the influence of environmental processes causing dispersion and dilution.

7.2 DOSIMETRY

Radiation dosimetry is the process of determining the energy absorbed in a specified target from a radiation field which, for waste disposal operations or accident situations, is generated by non-uniform, time-dependent distributions of radionuclides both in the external environment and within the organism of interest. For mobile animals, the incident radiation field from the external sources is also modified by their behaviour.

As indicated above, the relevant biological responses of wild organisms to radiation exposure suggest that the appropriate targets for dosimetry are likely to be the ovaries, stamens, germinating seeds and other propagative tissues in plants, and the gonads and the developing embryos in animals. These biological entities show enormous variation throughout the plant and animal kingdoms both in their geometrical form and their capacity to accumulate radionuclides from the external environment.

The incident radiation field consists of X- and -rays, (-particles and (-particles which have characteristic path lengths in tissue ranging from metres to approximately 0.1 mm. Thus it is clear that the spatial distributions of the radionuclides, relative to the defined targets, also need to be determined on these scales if dose (rate) estimates are to be made.

These factors concerning the spatial and temporal distributions of the radionuclides and the geometric variability of the targets mean that it is very rarely possible to determine the dose rate by a direct (or indirect) instrumental method; also, it is clearly impossible to adopt this approach for assessments either in the pre-operational phase of a waste-disposal practice or of the potential consequences of an accident. In these circumstances reliance must be placed upon dosimetric models, that is, computational methods which use information concerning both the distributions and behaviour of radionuclides in the environment and the absorption processes acting upon radiations in tissue, to estimate the radiation dose (rate) to the specified target. In view of the enormous number of species constituting the biosphere and the wide range of physical and chemical conditions prevailing in the niches occupied, it is inevitable that simplifications and generalizations would have to be made. In addition, the required full range of information on both the internal and external distributions of the radionuclides and their environmental behaviour is very rarely available even for a single species.

The processes by which energy is transferred from the radiation field and absorbed in the target tissue depend on the nature and energy of the radiation. A theoretical analysis of the energy deposition along the paths of - and (-particles and - and X-rays is possible but the resulting mathematical expressions are complex and, due to the stochastic nature of the processes involved, are not straightforward to apply for the purposes of dose estimation, particularly in an environmental context. As a consequence, simpler empirical formulae, with energy-dependent parameters, have been developed which describe the dose distributions around point sources (Berger, 1968, 1971; Loevinger et al., 1956; Harley and Pasternak, 1972; Woodhead, 1979). These can then be used to estimate the dose (rate) at a given point by integration of the expressions over an extended source distribution (Loevinger et al., 1956; Woodhead, 1979). The determination of the dose (rate) at different points within a target volume then allows an estimate to be made of the mean dose (rate) to the tissue or organ (Brownell et al., 1968; Ellett and Humes, 1971).

 These dosimeric approaches can be developed to deal with all the more or less complex situations which are to be found in the natural environment. The techniques and models introduced above provide estimates of the absorbed dose (rate), i.e. the radiation energy absorbed per unit mass of tissue. Radiobiological studies show, however, that different radiations can produce different degrees of effect for the same absorbed doses and that the variations are very largely dependent upon the ionization density produced along the particle tracks. This phenomenon is evaluated in terms of the relative biological effectiveness (RBE) of the radiation, defined as:

                         Weighted absorbed dose = QF absorbed dose.

Although the QF have, strictly, only been defined for use in human radiation protection applications, the values have been developed from RBEs obtained from experiments with a wide range of organisms and it is assumed that they can be applied in an environmental context. Such considerations are necessary due to the actual or potential contributions of (-radiation, for which a QF value of 20 is recommended, to the internal dose rate of many organisms.

For the purpose of dosimetry, the most difficult problem is to provide all the detailed information, concerning the distributions of the radionuclides in space and time, which is required for the application of the models. For an established waste disposal practice, relevant and specific information can be accumulated for the purpose of estimating the dose rates to particular organisms which might potentially be at risk (Woodhead, 1970, 1974, 1986; Pentreath et al., 1973, 1980); in exceptional circumstances it might be possible to confirm the results of the dose calculations by direct measurements (Woodhead, 1973). For pre-operational studies and the assessment of potential accidents a different approach is necessary. In both cases the radionuclide source must be defined both in terms of the spectrum of radionuclides present, and their emission rates over time. Appropriate models must then be applied to determine the development of the dispersed radionuclide distributions in the environment, including the accumulation by organisms. Available transport models are concerned entirely with producing a basis for estimates of dose 

Figure 7.1 Modelling framework

rates to human populations but, interpreted with care, can also provide information of use for environmental dosimetry. The assessment made of the potential impact of the dumping of packaged low-level radioactive waste in the deep northeast Atlantic Ocean may be quoted as an example (OECD/NEA, 1985). The modelling framework is indicated in Figure 7.1, from which it can be seen that the output of the transport model was a prediction of the radionuclide concentrations in water and sediment as a function of time. The highest concentrations are predicted to occur in the deep ocean in the immediate vicinity of the dumpsite and these were used as the basis for the estimates of the dose rates to the faunal types listed, together with their dimensions, in Table 7.1. The choice of organisms was based upon the following considerations:

1. Fish: radiobiological studies have indicated that these are probably the aquatic organisms most sensitive to radiation exposure (IAEA, 1976, 1988). Assessments for bathypelagic and benthic types allow the contribution from -emitters in the sediment to be highlighted.

2. Large crustaceans: these organisms generally have higher concentration factors than fish, thus increasing the importance of internal exposure relative to external sources. Again, assessments for bathypelagic and benthic types show the importance of both (- and -emitters in the sediment.
3. Molluscs: these organisms generally have higher concentration factors than the previous two groups and, with smaller size, show the effect of these factors on dose rate; and
4. Small crustaceans: assuming that the data available for surface-living zooplankton are applicable, these organisms would have the highest concentration factors for most elements. Also, being the smallest organisms, they show most clearly the effects of size on the relative contributions of internal and external sources.

 Table 7.1 Dimensions of the generalized geometrical models adopted to represent deepsea organisms

In the absence of more detailed information, the radionuclides were assumed to be uniformly distributed throughout the bodies of the organism. The volume-averaged absorbed fractions for -radiation, and the absorbed fractions at the centre of the volumes for -radiation, are shown in Figures 7.2a and 7.2b respectively, for the four organism geometries (see OECD/NEA, 1985 for more details). The results of calculations made for the natural radionuclides, past dumping and 5 years of expected future dumping (which did not take place) are given in Table 7.2. Clearly, these organisms, and their chosen geometries, cannot accurately represent all those present in the deep sea, but the dose rates estimated on the basis of these models do provide a realistic indication of the range of exposures likely to be experienced in the deep sea from the dumping of radioactive wastes.

The effect of a non-uniform distribution of the radionuclide body burden implied by a uniform whole-body concentration factor has been investigated for the cases where a small, centrally situated organ either preferentially accumulates, or discriminates against, -emitting isotopes. At low (-energies the dose rate scales directly with the radionuclide concentration, but at higher energies the dose rate falls below proportionality for preferential nuclide accumulation and increases above proportionality where there is discrimination; the size of the organ relative to the body size also influences the dose rate (Pentreath and Woodhead, 1988).

An alternative approach which has been adopted (IAEA, 1992b; NCRP, 1992) is to estimate the radionuclide concentrations in the environment which would, through reasonable human use of that environment for food production, leisure etc., result in the limiting dose equivalent of 1 mSv a^-1 being received by a critical group. These environmental concentrations may then be used as the basis for dose estimates to populations of wild organisms also inhabiting the area. The quality of the results thus obtained depends, amongst other things, upon the validity of both the assumptions made concerning the behaviour of the radionuclides in the environment and the models used to describe this behaviour and generate predictions of radionuclide distributions. Table 7.3 gives the estimates of the weighted absorbed dose rates to plants and animals which have been obtained for a controlled release to the atmosphere using this approach (IAEA, 1992b).

Figure 7.2 (a) Absorbed dose fraction for -radiation (b) Absorbed dose fraction for (-radiation.

Table 7.2 Estimates of the weighted absorbed dose rates (nGy h^-1) to deepsea organisms

Organism	Natural	Past	Past dumping plus 5 years
	background	dumping	expected future dumping
Fish
Bathypelagic	28460	52	73
Benthic	851400	73	100
Large crustaceans
Bathypelagic	9303600	360	460
Benthic	9904600	380	500
Molluscs	9303400	8000	10 000
Small crustaceans
Bathypelagic	2502900	1200	1600
Benthic	3304200	1300	1700

Table 7.3 Estimates of weighted absorbed dose rates to plants and animals for controlled atmospheric releases which deliver an effective dose equivalent of l mSv a^-1 to humans

Nuclide	Deposition	Equilibrium	Upper-bound	Equilibrium	Upper-bound
	rate	concentration	estimate of	concentration	estimate of
	(Bq m-2 d-1)	in plant	dose rate to	in animal	dose rate to
		tissue	plant tissue	tissue	animal tissue
		(Bq kg-1 dry)	(mGy d-1)	(Bq kg-1 wet)	(mGy d-1)
3H (as water)	6.6 x 105	1.8 x 106	0.14	1.8 x 106	0.14
14C		6.0 x 105	0.44	3.6 x 105	0.27
32P	5.4 x 101	3.2 x 103	0.049	6.8 x 101	0.001
95Zr	4.9 x 103	1.9 x 105	0.90	5.2 x 102	0.048
131I	1.0 X 103	5.8 x 103	0.029	1.7 x 102	0.0014
137Cs	2.8 x 101	1.9 x 103	0.13	3.9 x 102	0.074
239Pu	1.4 x 10-1	7.2	0.011	6.9 x 10-5	0.00026

It is clear from the foregoing that the major limiting factor in the determination of dose rates to wild organisms in contaminated environments is the availability of the necessary data on radionuclide distributions in the organisms and their environment. Where data are available, dosimetry models of sufficient sophistication have been developed to make complete use of the information provided on radionuclide distributions (Woodhead, 1979, 1986; Pentreath et al., 1973; Hoppenheit et al., 1980).

7.3 RADIATION EFFECTS

The effects of ionizing radiation have been studied in a wide range of plant and animal species at all scales of biological organization from biomolecules through cellular organelles, whole cells, tissues, organs, whole organisms, populations and, finally, at the community level. It is not possible to provide a comprehensive review of all the available information here (but see, for example, Bacq and Alexander, 1961, for a general introduction at the lower levels of organization; Whicker and Fraley, 1974 (plant communities) and Turner, 1975 (animal populations)). A general conclusion which can be drawn from these data, however, is that, because ionizing radiation acts by depositing energy and inducing change at the molecular level, there can be no effect at any higher level of organization without detectable change at successively lower levels (Haux and Forlin, 1988). In particular, radiation damage will not become apparent at the population level unless there is a clear response, in the individuals making up the population, in those characteristics which influence the maintenance of the population including individual morbidity, fertility (gametogenesis), fecundity (the production of viable offspring) and the gene pool. It follows that if the radiation dose (rate) either from a waste disposal practice, or arising from an accident, is sufficiently low that there can be no significant effects on these individual characteristics, then there can be no impact on the population, which, as already indicated, is generally accepted to be the object of protection.

1. what is the relevant population; and
2. for this population, what is the relevant dose (rate)?

In principle, the population could be defined as all the members of a particular species, but in practice this is rarely helpful. Ecologists have frequently found that the appropriate definition depends on the particular problem (and species) under study, and that it will be determined, among other factors, by the scale of interest in both space and time, thus often leading to a more limited concept (Bakker, 1971; Gross, 1986). A useful definition suggests that the population would be an essentially self-sustaining unit of a particular species, i.e. immigration into, emigration from, and interbreeding with, other populations of the species would be very minor factors in the overall maintenance of this population (Bakker, 1971). Thus it is clear, for example, that all the perch, brown trout or mussels in a freshwater lake could be considered as single populations, whereas ducks overwintering on the lake might be only a small proportion of a much larger interbreeding population. The more mobile the organism (including, for example, aerial dispersion of pollen and seeds, and water transport of gametes, eggs and larvae in the sea), the larger the space scale which needs to be considered in attempting to define the appropriate population.

Because the source of the radionuclides is likely to be restricted in space and time, it is probable that concentration gradients will exist in the environment and that the highest dose rates (all other considerations being equal) will arise in the immediate vicinity of the release. Thus it may be concluded that a range of dose rates will be experienced within a population, although of lesser extent, perhaps, for a mobile species as compared with a sedentary species, and that the increased radiation exposure could be delivered to either all the members of the population, e.g. many aquatic species inhabiting a pond or lake, or only a part of a population for many terrestrial and marine species. Given the large data requirements and obvious complexity involved in estimating average values, over the populations and through time, for the dose rates (or lifetime doses) it is usual to adopt a simpler approach. An assessment is made of the maximum dose rate likely to be experienced by individuals of a limited number of species which might be at risk in the contaminated environment, either because they belong to the most radiosensitive species or because their physiology and life-style are such that they will experience the greatest radiation exposures (see example above for the deep sea). It may then be reasonably assumed that, if the estimated dose rates are less than those which have been shown experimentally to produce no significant radiation effects, there can be no risk of damage to the population (or any higher level of organization).

7.3.1 MORBIDITY AND MORTALITY

For organisms other than man, morbidity (and mortality) are usually only considered as a consequence of acute (short term) high dose radiation exposure. Laboratory or controlled field studies of radiation-induced mortality usually determine the total dose required to kill 50 per cent of the organismsthe LD₅₀ within a specified period of time immediately post-exposure. Table 7.4 gives LD₅₀ data for a variety of organisms and provides a qualitative indication of the variations in radiosensitivity which exist between groups (Woodhead, 1971; Myers, 1989). Part of the wide variation in apparent radiosensitivity is due to the use of a 30 day time period for assessing the expression of the radiation-induced mortality (giving the LD_50/30) although this is strictly applicable only to small mammals where essentially all the deaths occur within this interval. For fish and other aquatic organisms it has been shown that a 5060 day period is necessary to encompass the acute responsea consequence of their poikilothermic metabolism (White and Angelovic, 1966)giving an increase in apparent radiosensitivity. Adjustment of the data for this factor reduces, but does not eliminate, the general tendency for radiosensitivity to increase with the increasing biological complexity of the organism. Another factor which influences the relative radio sensitivity between groups, e.g. insects and mammals, arises from the developmental progression through the life cycle. In the former, most of the cellular proliferation and differentiation occurs in the developing embryo, whereas for the latter, it continues in certain tissues throughout life. In general, developing embryos are more radiosensitive than fully formed adults and across groups of organisms this again tends to reduce the overall range of radiosensitivity; the developing vertebrate embryo does, however, retain its position of greatest radiosensitivity.

Table 7.4 Acute lethal doses of radiation

Class of organism	Lethal dose, Gy
	Adults	Developing embryos
Mammals	215	1 (mouse)
Birds	520	7 (chicken)
Amphibians	750
Fishes	7600	0.21 (salmon, trout and plaice)
Reptiles	1040
Crustaceans	15600	6 (production of young amphipods)
Molluscs	1001000
Echinoderms	390
Insects	203000	12 (weevil, fruit fly, wasp)
Higher plants (trees,
shrubs and herbs)	7800
Primitive plants
(mosses, lichens
and algae)	30>I2 000
Protozoa	1006000
Bacteria	501000
Viruses	20010 000

Such acute, high dose exposures are only likely to arise as a result of a large, accidental release of radionuclides. In this circumstance, the radiation exposure may range from acute (i.e. delivered in an interval which is short compared with the time of development of the biological response) to chronic (i.e. the exposure continues over a significant fraction of the lifetime of the organism) depending on the organism under consideration and the quantities and relative proportions of short-lived (~ days) and long-lived (~ years) radionuclides present in the release.

Protraction of radiation exposure generally increases the total dose necessary to cause mortality (or any other effect). This is a consequence of two complementary factors: at the cellular level, sub-lethal damage is reparable; and at the tissue level, lethally damaged cells can be replaced through cell proliferation to maintain a viable degree of organ (and organism) function. Where the exposure extends over a significant part of the lifetime of the organism, the total accumulated dose required to kill may be two to ten times greater than the acute LD₅₀ (Page, 1968; UNSCEAR, 1982a; Woodhead, 1984; IAEA, 1992b). The chronic, low-level dose rates which have been estimated to arise from controlled disposals of radionuclides to the environment (see Section 7.2) are very unlikely to lead to total accumulated doses to organisms which could result in significant mortality.

7.3.2 GAMETOGENESIS

Available reviews (Whicker and Fraley, 1974; IAEA, 1992b) do not provide any detailed discussion of the effects of radiation on gametogenesis in plants and there appear to be few data in the literature. A single reference (Woodwell and Rebuck, 1967) notes that flowering and seed set occurred in the herbaceous plant, Carex pensylvanica, in the dose rate range 0.1716 Gy d^-1. This plant species was, however, fairly resistant to the damaging effects of radiation in somatic tissue, so this rather low radiosensitivity is probably not typical of the plant kingdom.

For animals it is necessary to consider spermatogenesis and oogenesis separately because the developmental processes for the production of the mature gametes are generally different for the two sexes.

7.3.2.1 The testes

Exposure of the testes to ionizing radiation produces sterility which, depending on the dose rate and total dose, may be temporary or permanent. Acute doses which are a small fraction of the LD₅₀ may induce short periods of delayed sterility in mammals due to the damage to the most sensitive differentiating spermatogonial cell-type. Larger doses produce more immediate and longer-lasting periods of temporary sterility but recovery is possible due to the relatively radioresistant stem cell population. Permanent sterility is ultimately produced by doses greater than about half the LD₅₀ depending on the species (BEIR, 1980; UNSCEAR, 1982b).

Insects may be sterilized with acute doses which are much less than those required to kill them or otherwise affect their behaviour (Bacq and Alexander, 1961).

Similar conclusions have been drawn for aquatic organisms, and fish have been shown to be the most radiosensitive of the aquatic fauna for which data are available (IAEA, 1976; Woodhead, 1984).

Of more relevance to the majority of environmental situations is the potential influence of chronic, low-level exposure. The response of the testis is somewhat unusual in that protraction of a given total dose does not always lead to a reduction in the effect (UNSCEAR, 1982b). The most sensitive mammalian species appears to be the dog for which a life-time mean dose rate greater than 1.2 mGy d^-1 (50 µGy h^-1) has been determined to be the threshold for deleterious effects (Casarett, 1964; BEIR, 1980; UNSCEAR, 1982b). In fish also there appears to be considerable inter-species variability in the response of the testis to chronic irradiation with the most sensitive species examined showing minor effects at a mean dose rate of 1.1 mGy h^-1 (Rackham et al., 1992).

At dose rates above the threshold for effects, but below that at which a continuing decline into sterility occurs, there is evidence for a homeostatic response, through either mobilization of reserve proliferative capacity in the stem cell compartment or adjustment of the cell cycle kinetics, or a combination of both, leading to the maintenance of sperm production at equilibrium level below normal in both mammals (Hsu and Fabrikant, 1976; Erickson, 1978; UNSCEAR, 1982b) and fish (Hyodo-Taguchi and Egami, 1977).

The developing testis in the mammalian fetus also shows wide interspecies variations in radiosensivity, with the pig showing the greatest response at 2.5 mGy d^-1, the lowest dose rate employed, and being substantially more radiosensitive than the guinea pig, rat and mouse at 10 mGy d^-1 (UNSCEAR, 1986).

7.3.2.2 The ovary

The responses of the resting oocytes in adult mammals show wide variations between species with the acute dose required for sterilization being a small fraction of the LD₅₀ for the mouse and substantially greater than the LD₅₀ for the monkey (UNSCEAR, 1982b). Available information indicates less variation between fish species where the sterilizing dose is of the same order as the LD₅₀ (Woodhead, 1984). As might be expected, due to the cell division and differentiation in progress, the embryonic and neonatal ovaries are found to be more radiosensitive (UNSCEAR, 1982b; Woodhead, 1984).

For mammals, the available reviews of the effects of chronic irradiation on the adult ovary present the data in terms of offspring production (fecundity, see below), rather than as the numbers of oocytes at different stages of maturation present (UNSCEAR, 1982b). The developing ovary in the mammalian fetus is sensitive to the effects of continuous irradiation (although apparently less so than the developing testis) with a dose rate of 10 mGy d^-1 throughout gestation having a substantial effect in the pig, guinea pig, rat and mouse. The pig shows the greatest sensitivity with a significant response at the lower dose rate of 2.5 mGy d^-1 (UNSCEAR, 1986). Protraction of ovarian exposure in fish produces an apparent reduction in radiosensitivity, in contrast to the response of the testis. The most sensitive fish studied to date (Ameca splendens) has shown a substantial reduction in all stages of oocyte maturation at a dose rate of 1.1 mGy h^-1) over 290 days (Rackham et al., 1992); other species, however, have accumulated high total doses at higher dose rates over a greater proportion of the life with less damage becoming apparent (Woodhead, 1984). Exposure of chinook salmon embryos during the 80 day incubation period did not produce any effects on gonad development at dose rates below 100 mGy d^-1 (Woodhead, 1984).

The potential significance of life history attributes for the response to chronic irradiation has been noted by Turner (1975). Organisms with an extended time to sexual maturity might be at greater risk if, as seems possible, the cumulative dose is a more important determinant of ovarian damage than dose rate per seanimals with a shorter maturation time being able to reproduce before radiation-induced sterility intervenes (see also Myers, 1989).

7.3.3 FECUNDITY

If fecundity is taken to mean the number of viable offspring produced per female, it encompasses not only the effects of radiation in reducing fertility, but also both the induction, in otherwise apparently normal gametes, of mutations incompatible with embryonic development and neonatal survival, and the direct effects of radiation on embryonic development.

The total number of resting oocytes present in the ovary at sexual maturity generally far exceeds the number likely to mature and become potentially available for reproduction. A radiation-induced reduction in the number of immature oocytes does not, therefore, necessarily lead to a like reduction in the potential fecundity although it may result in a shortening of the span of reproductive life (BEIR, 1980; UNSCEAR, 1986; Myers, 1989).

In the mouse, for which the most data are available, dominant lethal mutation (as measured by the incidence of non-viable embryos conceived in subsequent matings with unirradiated females) are produced less efficiently by chronic than by acute exposure of the male gametes; also the post-meiotic cell stages are more sensitive than the premeiotic and spermatogonial cells. The induction rate for chronic exposure at 34-60 mGy d^-1) was estimated to be 1.4 x 10^-6 mGy^-1 per gamete for premeiotic cells and 5 x 10^-5 mGy^-1 per gamete for post-meiotic cells (UNSCEAR, 1982c). Available evidence indicates that the female mouse is much less sensitive, such that the effective rate of induction when both sexes are irradiated is half that for the males alone (BEIR, 1980). Data for other small mammals yield induction rates of similar magnitude (UNSCEAR, 1982c). It has also been concluded that aquatic organisms show a similar sensitivity for the radiation-induction of dominant lethal mutations as do the more intensively studied small mammals (Woodhead, 1984).

Experimental studies indicate that the developing embryo is more sensitive to the effects of acute irradiation, and sometime substantially so, compared with the adult organism. For insects the LD₅₀ may be reduced by a factor in the region of 10^-3 while for vertebrates the factor may be in the range 0.05 to 0.9largely reflecting the different basic biologies of the organisms (Myers, 1989). The many studies which have been made show that there is a generally declining trend of radiosensitivity with time post-fertilization but that, for the vertebrates in particular, there are phases in embryonic and fetal development, correlated with specific organogenic processes, which show increased radiosensitivity (UNSCEAR, 1977b, 1986; Woodhead, 1984). Again, protraction of a given total exposure, as expected, generally reduces the response (Myers, 1989). Lower doses and dose rates, while not producing immediately lethal effects, might produce teratological responses which may be incompatible with long-term survival in wild animals.

A number of experiments have shown that, despite chronic irradiation, the germ line can be maintained over many generations with minimal effects being apparent in the reproductive performance. Mice of four different strains have been exposed to an average of 20 mGy d^-1 over 10 generations (for a total accumulated dose to the germ line of 15 Gy) and have shown no effects on reproductive performance in the first litters (Stadler and Gowan, 1964). Similar results have been obtained for the rat (Brown et al., 1964).

The lifetime (> 950 days) offspring production by pairs of the fish, Poecilia reticulata, was reduced by 57 per cent at an average dose rate of 1.7 mGy h^-1) (the lowest used). It was concluded that this effect, which was progressive through life, resulted from both a reduction in fertility and the induction of dominant lethal mutations (Woodhead, 1977).

7.3.4 GENETIC EFFECTS

In wild populations, natural selection operates to adapt the gene pool transmitted from generation to generation to the local environment. This is not to say, however, that genetic uniformity exists because neither the environment nor the resulting selection pressure is stable over space and time. Thus there is a dynamic equilibrium between genetic variability and selection pressure. In the great majority of populations there is an enormous amount of genetic variability which is constantly being sorted into new combinations in each generation (Myers, 1989).

Irradiation of germ cells may produce both gene mutations and chromosome aberrations (changes in chromosome morphology or number). Such mutations are not different in kind from those occurring naturally but the proportions of the different types induced by irradiation may well differ (BEIR, 1980; UNSCEAR, 1977c). The induced mutations can have an impact which varies between lethality (as in the dominant lethals discussed above) and the apparently trivial, e.g. a change in eye colour (UNSCEAR, 1977c). Nevertheless, all are subject to selection pressure, and mutations are lost from the gene pool as a consequence of a reduction in fertility in, or by the pre-reproductive death of, individuals carrying the mutations.

Despite the variations in radio sensitivity, both of different gene loci within a species, and between species, the total dose required to double the spontaneous mutation rate appears to be constrained within the surprisingly narrow range of 0.33 Gy across a wide variety of organisms (Myers, 1989). Thus, for a given dose rate, the induction of genetic change will be much less significant for organisms with a short generation time than for long-lived, late-maturing species. Experimental studies of mice and fruit flies (Drosophila spp.), irradiated in each of many generations with doses equal to, or greater than, the doubling dose, have detected no effects on the health and general fitness of the populations (Myers, 1989).

7.4 CONCLUSIONS

Previous, more thorough, reviews of the available data, of which only a limited selection has been considered here, have concluded that there would be no significant effects in wild populations of:

1. terrestrial animals if the dose rate to the most exposed individuals were to be less than 1 mGy d^-1 (0.04 mGy h^-1); and
2. terrestrial plants, and aquatic plants and animals if the dose rates to the most exposed individuals were to be less than 10 mGy d^-1) (0.4 mGy h^-1) (IAEA, 1976, 1988, 1992b; NCRP, 1992).

Assessments of the dose rates actually (or potentially) arising from controlled waste disposal operations indicate that, with very few exceptions, these values have not been (or would not be) exceeded; often there is a large margin of safety (IAEA, 1976, 1992b; NCRP, 1992; OECD/NEA, 1985; Woodhead, 1970, 1973, 1974, 1984, 1986). The one major exception relates to dumping of packaged low-level radioactive waste into the deep ocean where, because of the remoteness of the release point from man and his food chains, and the time taken for radionuclide dispersion to take place, it is possible to envisage dumping rates, controlled on the basis of ultimate human exposure, at which very high dose rates could be delivered to benthic organisms at the dumpsite (IAEA, 1988).

The situation for a major nuclear accident is, as Chernobyl has shown, quite different. Effects were readily apparent in the pine trees close to the plant, and less dramatic responses have been detected in animals. Even here, however, it is debatable whether the long-term survival of the populations has been put at serious risk. Studies in progress should provide much useful information for the resolution of this concern.