SCOPE 50 - Radioecology after Chernobyl

6.1 INTRODUCTION

About 70 per cent of the population of the current developed world lives in urban areas and it is therefore essential that an understanding of the pathways and behaviour of pollutants, including artificial radionuclides, in towns and cities is obtained. For various reasons, however, the urban environment has largely been excluded from terrestrial studies until recent years. Perhaps one of the greatest catalysts for research within this field was provided by the Chernobyl reactor accident. The environmental processes operating within an urban system may be similar to those observed in a rural environment. Therefore the emphasis of this chapter will not be on describing these general processes, but rather the aim is to highlight how pathways of artificial radionuclides are modified by features of the urban environment. Finally, it is shown how a knowledge of the behaviour of artificial radionuclides in the urban environment may be used to develop a strategy for decontamination.

6.2 CONTAMINATION PROCESSES 

6.2.1 DRY DEPOSITION

Underwood (1987) has critically reviewed the applicability of a single deposition velocity to model plume depletion over urban complexes and found that careful interpretation is required. Since urban complexes generally have a spatial extent of many kilometres the concept of an `averaged bulk deposition velocity' is feasible. However, in reality the roughness elements of the urban canopy are never homogeneous, thus only local deposition velocities can be determined and these will not be strictly applicable to the whole city. These fundamental problems, combined with the technical difficulty in deducing the downward flux above the top of the roughness elements (buildings, etc.), has resulted in a limited application of the concentration gradient approach.

An alternative approach is to measure the accumulated flux on both outdoor and indoor sections of the urban canopy and relate that to integrated air concentrations. Initially, work concentrated on ¹³⁷Cs particles derived from weapons fallout, using surfaces protected from wet deposition and weathering. However, the Chernobyl accident enabled the determination of deposition velocities to numerous surfaces. Attempting to apply the deposition velocities derived by this method to the whole of an urban canopy is even more complex than using a bulk deposition velocity. Macro scale considerations include the influence of buildings on wind direction and speed, leading to greater turbulence. On the meso and micro scales the composition and texture of the individual surfaces will govern deposition. In addition, enhancement may occur at projections (e.g. window sills), discontinuities (e.g. mortar joints) or edges, negating any attempt to scale up deposition velocities from a portion of a surface, say 10 bricks, to the whole wall.

Roed (1987a) and Nicholson (1987) studied the dry deposition of Chernobyl radionuclides to various building materials and a few of their results are summarized in Table 6.1. The deposition velocities for particular radionuclides and surfaces were reasonably similar, whether measured in open spaces, near edges or in densely built-up areas. However, the deposition velocities of the majority of radionuclides to roof material (e.g. 0.0270.09 cm s^-1 for ¹³⁴Cs) were higher than those to roads (e.g. 0.00450.011 cm s^-1 for ¹³⁴Cs). This was believed to be due to an increase in wind speeds and turbulence at roof level. There also appeared to be an order of magnitude difference between the deposition velocities to walls (e.g. 0.0010.002 cm s^-1 for ¹³⁷Cs) and roads for radiocaesium, although this was not observed in the case of ¹³¹I. Deposition velocities measured for Chernobyl radioiodine were generally higher than those for radiocaesium, due to the presence of different radioiodine species in the fallout (particulate, gaseous and methyl iodide). Measurements of ¹³⁴Cs deposition velocities to surburban trees and bushes (0.030.13 cm s^-1 were higher on average than those to roof surfaces, while deposition velocities to grass were nearly proportional to the mass of grass per unit area, resulting in a wide range of values (0.0150.099 cm s^-1 for ¹³⁴Cs) (Roed, 1987a). 

Deposition velocities within houses have also been measured for radionuclides in Chernobyl fallout and ranged from 0.006 to 0.058 cm s^-1 (Roed and Cannell, 1987). In addition, results from this study were used to determine the ratio of indoor to outdoor aerosol concentrations (0.25 for ¹³⁷Cs and 0.4 for ¹³¹I). The reduction in concentration was considered to be due to the filtering effects of cracks, crevices and pores in the building structure, with the higher ratio for ¹³¹I being due to a smaller particle size.

Table 6.1 Deposition velocities for ¹³⁴Cs, ¹³⁷Cs and ¹³¹I to various urban surfaces

		Deposition velocities (cm s-1)
Reference	Radionuclide	Roof	Wall	Road
Roed (1987a)	137Cs	0.0280.035	0.0010.002
	134Cs	0.0270.029		0.00450.011
	131I	0.020.32	0.0270.03	0.030.033
Nicholson (1987)	134Cs	0.030.09	<0.004<0.0005

It can be seen that each surface in the urban environment will have a unique deposition velocity resulting from the nature of the surface and its location within the urban windfield. The effects of inhomogeneities appear to be minimal and therefore deposition velocities will apply to the whole surface (Roed, 1987a). A summary of urban surface deposition velocities is provided in Figure 6.1. This allows the deposition velocity to the whole urban canopy to be expressed by the following equation (Underwood, 1987):

where A_i is the total surface area of the type i per plan area of city and V_gⁱ is the corresponding deposition velocity. Recent work by Roed and Jacob (1990) on Chernobyl radiocaesium produced a V_g (bulk urban) estimate of 0.03 cm s^-1 for 0.4 µm particles using this method. Despite its limitations, this `naive' estimate of the urban deposition velocity may result in reasonable values for certain applications.

Figure 6.1 Dry deposition velocities of Chernobyl radiocaesium to various urban surfaces (cm s^-1). From Nicholson, 1987; Roed, 1987a, Roed and Cannell, 1987; Roed and Goddard, 1990; Roed and Jacob, 1990.

6.2.2 WET DEPOSITION

It has been well established that major conurbations (> 1 million people) cause enhancement of both precipitation, occult deposition and the frequency of thunderstorms (e.g. Changnon, 1980). In addition, scavenging ratios have been observed to decrease from upwind of a city to downwind (Gatz, 1977), but this may be due to increased air concentrations at the downwind sites. Thus, it was argued that ground level pollutants had not been efficiently scavenged by rainfall and only when the city-derived species reached cloud level, some distance downwind, would they actually be efficiently scavenged. Conversely, pollutants derived far enough upwind of the city will be subject to efficient in-cloud scavenging above an urban complex. In this situation, the apparent enhanced levels of precipitation suggest that wet deposition of particles and gases will be greater in urban areas. However, it should be realized that the observed zone of rainfall enhancement in major conurbations did not occur over the cities themselves, but developed downwind over rural regions. Therefore, since one cause of enhanced precipitation is the additional particulate matter suspended in the atmosphere, the relative position of industrial plants is of prime importance to the location of this zone. Furthermore, precipitation enhancement has been observed only for cities > 1 million people; thus, any effect that the majority of cities may have on rainfall is probably small.

Thunderstorms may have a tendency towards low wash-out efficiency due to their short duration, but their importance lies in the fact that they process large volumes of air and thus may lead to localized hot spots of deposition.

Fog frequency in central London has been shown to be greater than in the surrounding rural area (Hall, 1984), which, combined with the generally high concentration of pollutant species found in fogwater (Waldman et al., 1983) suggests that this deposition pathway may be important.

Although wet deposition is more efficient at removing pollutants (particularly particles) from the atmosphere, dry deposition is pervasive through time and space. Hence, it is likely that for continuous pollutant releases both processes are of similar importance. However, for instantaneous releases of radionuclides into the atmosphere, it is of great significance whether the resulting radio-active cloud is intercepted by rainfall. This was demonstrated quite dramatically by the Chernobyl reactor accident, where large regions of north-west Britain received much higher levels of fallout due to the coincidental heavy rainfall (Clark and Smith, 1988). Hence it may be tentatively suggested that the increased frequency of rainfall in major urban complexes could alter the balance of fallout mode in favour of wet deposition, especially where particulate contaminants are concerned. For these reasons, an assessment of accident consequences needs to take into account the relative position of nuclear installations and major cities in relation to predominant wind directions.

6.2.3 MECHANICAL TRANSPORT PROCESSES

The mechanical transport of dust and pollutants in the outdoor environment is generally of greater significance in urban rather than rural areas, due to the predominance of vehicles. However, mechanical transport probably has greater impact within indoor environments, where humans provide the motive force. Consequently, the extent of outdoor contamination will determine the importance of this pathway.

Two possible sources of radionuclides to the outdoor urban environment are from nuclear fuel fabrication and reprocessing plants. Contaminated dust may be physically picked up by the tyres of heavy plant vehicles and subsequently deposited on local road surfaces. In addition, material transported by lorry may also accidentally become incorporated into street dusts. Evidence for the existence of these pathways was found by Al-Khayat (1989) with elevated levels of ²³⁴Th present in dust at the entrances to the Springfield nuclear fuel processing plant (UK) and also on the roads used by the plant vehicles.

Radionuclide contamination of houses in the coastal village of Ravenglass, near Sellafield, was attributed to the mechanical transport (on the bottom of shoes) of finegrained sediments from the Esk estuary (DOE, 1987). This conclusion was based on the significant increase in the plutonium content of house dust, from occupants who never visited the foreshore, to those who made frequent visits. Harrison (1979) observed that the particle size distributions of house dust and soil were similar and deduced that a major pathway for dust into homes was by mechanical transport of soil on shoes. This has been confirmed using a fluorescent tracer (Cannell et al., 1987). 

6.3 REMOVAL PATHWAYS

6.3.1 INTRODUCTION

Once deposition has occurred, the majority of pollutant species rapidly lose their individual identity, through becoming sorbed to the host soil particles. Surface deposits of urban dusts and pollutants are subject to a variety of environmental and anthropogenic removal processes, the principal transport pathways being resuspension, mechanical cleaning or hydrologically based.

6.3.2 RESUSPENSION

6.3.2.1 Resuspension in the outdoor environment

Attempts to directly assess the modifications imposed on resuspension by urban complexes have been virtually non-existent. Not only is it necessary to resolve wind-driven resuspension from a variety of hard and soft surfaces, but the influence of mechanical resuspension has to be considered. Added to this the resuspension rate is virtually indeterminate, due to the complex windfields, whilst the resuspension factor becomes even more location-specific. Thus, it is necessary to infer resuspension behaviour from a combination of separate studies.

Sehmel (1973b) has provided much of the data on vehicle-induced resuspension from ZnS tracer studies on an asphalt road. A resuspended fraction of 10^-510^-2 was determined for a single vehicle pass through, or alongside, the freshly deposited tracer. The resuspended fraction was observed to increase with the size of the vehicle and also the square of the car speed. This was explained by an increase in turbulence, although tyre stresses may have been important. Weathering of the tracer was found to decrease the resuspended fraction, but the amount of tracer removed between experiments by wind-induced resuspension was not known and therefore the decrease represents a lower limit. Nevertheless, the observation is not unreasonable since particles may become attached to the road surface or lodged in crevices where they will be less easily resuspended.

Since the rate of vehicles driving over a contaminated surface may reach 1 vehicle s^-1 on a major trunk road, the resuspension rate will range from 10^-5 to 10^-2 s^-1, which is relatively high when compared to wind-driven resuspension. However, the spatial extent of vehicle-induced resuspension is much more limited than the surfaces available to wind-driven resuspension.

A simple resuspension model developed by Hamilton et al. (1985) produced a resuspension rate of 1.7 x 10^-5 vehicle^-1, consistent with the experimental results described above. Further predictions of the model suggest that vehicle-induced resuspension will reach equilibrium with the deposition flux after several dry days on a major trunk road, but this would never be the case for a side street with low traffic density.

Nicholson (1988a) reviewed some results on resuspension due to pedestrian movement, with a reported range of resuspended fractions from 1 x 10^-5 to 7 x 10^-4 per pass. This mode of resuspension is unlikely to contribute dramatically to the urban aerosol, although it may be of great significance to an individual's exposure.

Wind-induced resuspension from vegetated regions within cities will presumably differ little from rural areas, except with regard to the different wind speeds experienced. Resuspension from hard surfaces is less easy to predict: the fast drying nature of these surfaces will certainly ensure that dust and contaminants will be available to resuspension processes for longer periods of time. However, deposited species on asphalt roads become less erodable with weathering, as well as being progressively moved to sheltered positions and actually removed from the surface via runoff and cleaning. Thus, resuspension rates and factors for hard urban surfaces will have a time dependence attributable to all these processes.

Some estimates of resuspension factors have been made for urban areas and range from 5 x 10^-5 up to 10^-3 m^-1 (Linsley, 1978). These relate to the surface contamination on paved areas, whilst J.A. Garland (personal communication, 1990) has suggested that a value of 10^-6 m^-1, in relation to the initial deposition, may not be inappropriate. Unfortunately the former assumes no influence from regional resuspension, whilst the latter neglects removal of contaminants from paved surfaces via runoff.

6.3.2.2 Resuspension in the indoor environment

A number of studies were carried out in the 1960s to establish resuspension factors for indoor environments, in particular nuclear installation changing rooms. Fish et al. (1967) determined resuspension factors for an asphalt tile floor and found values ranging from 9.4 x 10^-6 m^-1 for `light' work up to 1.9 x 10^-4 m^-1 for vigorous activity. Resuspension from different floor coverings was studied by Jones and Pond (1967). An increase in movement produced a corresponding increase in the resuspension factor; a value of 5 x 10^-5 m^-1 was proposed as a reasonable resuspension factor. A marked correlation between the number of personnel in a changing room and airborne contamination levels was noted by Brunskill (1967). He determined resuspension factors ranging from 10^-3 to 2 x 10^-4 m^-1 for loose contaminants on a concrete floor. However, the resuspension factor for waxed linoleum was found to be smaller.

Resuspension factors have also been measured for carpeted surfaces and ranged from 1 x 10^-5 m^-1 for resting conditions to 2 x 10^-5 m^-1 during activity conditions. Unfortunately, no measurements were made during cleaning operations, when a large amount of resuspension would seem likely.

6.3.3 HYDROLOGICALLY INDUCED TRANSPORT

The removal of pollutants via the movement of water may occur by two distinctive mechanisms: migration due to the infiltration of water and wash-off resulting from runoff. Migration is generally considered to be a solution or colloidal phase transport process and only in certain situations, such as bioturbation, is this not the case. However, direct removal from surfaces may occur in either the solid or solution phases.

Within rural areas, infiltration is considered to be the dominant removal pathway for water, with runoff being of little significance for the well-vegetated regions of northern Europe. During urbanization, however, there is expected to be an increase in the proportion of impervious area compared to the previous rural region. This has a significant impact on the hydrological regime, with runoff increasingly dominating infiltration (Hall, 1984). Consequently, larger volumes of water are discharged into receiving streams, lag times are shortened and peak flow rates increased. Nevertheless, runoff from so-called impervious urban surfaces rarely exceeds 90 per cent and typically ranges from 34 to 83 per cent of the total rainfall (Ellis et al., 1987). This deficit is not only a reflection of the road storage capacity, but is believed to be due to seepage losses at road and pavement joints, spray from vehicles and direct infiltration through the surface itself (Ellis et al., 1987). Thus, in the urban environment the following processes will dominate:

1. Migration of soluble pollutants through pervious surfaces, ultimately to the groundwater.
2. Wash-off of particles and soluble pollutants from impervious surfaces into road side drains (gully pots).

3. Desorption of pollutants from particles and surfaces, followed by wash-off into gully pots or migration through surface and substrate to the groundwater. 

The magnitude of these different pathways for urban pollutants will depend on the physico-chemical form of the species and initially whether the contaminants were dry or wet deposited. It has been suggested that after about a day or so dry deposited species will behave as though they had been wet deposited, presumably through moisture appearing on the surface from light rain, mist or dew (Crick and Brown, 1990). Prior to this, rainfall would be expected to remove a reasonable proportion of the dry deposited species (Section 6.4.3). The wash-off of previously deposited pollutants from road surfaces is an important removal pathway very characteristic of the urban environment and has been the subject of a great number of studies in relation to non-radioactive contaminants (e.g. Ellis, 1986). The relatively fine grain sizes and steep slopes associated with roof wash-off ensure that these surfaces do not act as a sink, although it is possible for guttering to act as a temporary trap. The remaining processes, in particular migration of soluble species through pervious surfaces and the desorption of pollutants from impervious surfaces, will be discussed under retention behaviour (Section 6.4).

Radionuclides removed from urban surfaces by hydrological processes will ultimately be discharged into natural watercourses or bodies. The reduced exposure to radionuclides in the urban environment may therefore be offset by a potential increase in exposure from aquatic pathways such as drinking water or the food chain.

6.3.4 CLEANING

The removal of dust and consequently of pollutants by cleaning is a practice widely conducted in urban environments. Studies on the effectiveness of cleaning have been confined mainly to external dusts, although one study concerned with the decontamination of internal surfaces has been reported.

It is widely accepted that street sweeping has a beneficial effect on litter control and thus provides aesthetic improvements to urban environments. However, Sartor and Gaboury (1984) addressed the question of whether street cleaning has a significant effect on the water quality of urban runoff. The removal efficiency of conventional rotary and vacuumized street sweepers was found to be dependent on the particle size of the street dust: although highly efficient at removing the large particles (> 2 mm), including litter (79 per cent), their efficiency at removing the finer, more contaminated, grain size fractions was observed to be low (15 per cent of the < 43 µm fraction).

Revitt and Ellis (1980) showed that the consequence of this preferential removal of the finer grain size fraction was to cause a shift towards the finer particle end. Hydroflushing was performed after the mechanical rotary sweeping and appeared to have little effect on the coarse particle sizes (> 250 µm), but removed an additional portion of the 30250 µm particles. The resultant reduction in dust and total lead loadings were only 30 and 27 per cent respectively for the rotary sweeping, although greater efficiencies of 45 and 55 per cent were observed for hydroflushing. 

Malmquist (1978) also concluded that street sweeping was relatively inefficient at removing the highly contaminated fine dust, but observed higher removal efficiencies of 57 per cent for suspended solids, 65 per cent Pb, 31 per cent Zn and 61 per cent Cu. Similar total solid removal efficiencies were reported by Bender and Terstriep (1984) ranging from 14 to 56 per cent. However, they pointed out that the < 250 µm fraction was relatively unaffected by street cleaning. The reduction in mean lead concentration for subsequent wash-off ranged from 5 to 50 per cent. Although there appears to be some debate over the removal efficiency of street cleaning after a single pass (565 per cent for Pb), Sartor and Gaboury (1984) have demonstrated that the governing factor for the overall efficiency is the relative frequency of cleaning and storm events. The manner in which street cleaning is practised in the UK, with a typical cleaning interval of 2 months, prevents it from being an effective technique for removing road surface pollutants. However, in some cases it may increase the efficiency of wash-off by breaking up large particles and redistributing them throughout the street (Sartor and Garboury, 1984).

Much less attention has been paid to the efficiency of indoor cleaning techniques. Cannell et al. (1987) conducted experiments on the decontamination of carpet tiles impregnated with a 0.6 µm fluorescent tracer, using a light industrial/commercial dry vacuum cleaner. They found that for carpets vacuumed for just 20 s m^-2, 61 per cent of the tracer was removed from carpets which had not been trodden on, while only 37 per cent could be removed from those which were traversed 60 times. Further more intensive decontamination (5 minutes) removed totals of 72 and 52 per cent for the two carpeted surfaces.

6.4 RETENTION BEHAVIOUR 

6.4.1 INTRODUCTION

The retention behaviour of urban pollutants will naturally be a function of the input and removal rates of the processes previously described. There have been a number of studies, particularly since Chernobyl, on the retention behaviour of radionuclides adsorbed to impervious urban surfaces. In addition, the migration behaviour of pollutants through the vadose zone is generally well understood (e.g. Yaron et al., 1984), but little is known about the retention of particulate bound contaminants on impervious urban surfaces.

6.4.2 PERVIOUS URBAN SURFACES

The migration of radionuclides through soil is dependent primarily on the adsorption equilibria with the components present in a soil column (e.g. clay minerals, zeolites, hydroxides of iron and humic substances) and the porewater velocity. Molecular diffusion, plant root uptake and volatilization may also have significant influences. Migration rates are therefore quite variable, e.g. 0.617.9 cm y^-1 for ^239+240Pu (Coughtrey et al., 1984).

Long term migration rates, derived from weapons fallout ¹³⁷Cs, range from 0.07 to 1 cm y^-1 for the various soil horizons in a forest, compared to 0.20.4 cm y^-1 for grassland (Schimmack et al., 1989). Initial migration rates determined for Chernobyl-derived ¹³⁷Cs were several orders of magnitude higher than these values (forest, 2600 cm y^-1; grassland, 1800 cm y^-1). This was attributed to a high porewater velocity and the non-attainment of sorption equilibrium. Residence times of 2.5 y cm^-1, 2.1 y cm^-1 and 3.3 y cm^-1 have been calculated for ¹³⁷Cs ⁹⁰Sr and iodine respectively (Coughtrey and Thorne, 1982). This results in mean residence times of 75, 63 and 99 y for the top 30 cm of soil.

The reduction in dose due to weathering above a soil plot contaminated with ¹³⁷Cs was investigated by Gale et al. (1964). Migration of the ¹³⁷Cs resulted in an exponential decrease of the dose rate, due to increased absorption of the gamma rays by overlying soil layers. There was a high migration rate for the first four years, followed by a much longer term retention, which may even have been the radioactive decay of ¹³⁷Cs.

6.4.3 IMPERVIOUS URBAN SURFACES

Warming (1982), Sandalls et al. (1986) and Gjørup et al. (1986) undertook pre-Chernobyl experiments on both natural and artificial weathering of radionuclides from a variety of urban surfaces. These studies were followed up with an assessment of decontamination techniques. The dose rate above areas of asphalt and concrete contaminated with ⁸⁶Rb (applied in solution) exhibited an exponential decrease with time (Warming, 1982; Gjørup et al., 1986). Rain was found to be the most effective agent at removing ⁸⁶Rb and losses were tentatively attributed to wash-off rather than migration through the surface. The age of the surface appeared to have some effect, with contaminant removal occurring more readily from young surfaces. About 30 per cent of the ⁸⁶Rb on asphalt underwent removal, with a half-life independent of radioactive decay, ranging from 10 to 200 d, while for concrete surfaces 50 per cent had half-lives ranging from 80 to 200 d. The remainder was assumed to be removed over the much longer timescale reported by Gale et al. (1964) (92 y half-life). Three decontamination techniques were attempted: fire hosing, addition of potassium fertilizer and vacuuming. Only fire hosing appeared to be reasonably successful, with a 50 per cent removal efficiency if attempted within 14 days of the initial contamination. However this was reduced to < 10 per cent if left for 40 days. Jacob et al. (1987, 1990) measured the reduction in dose above paved areas in Munich, Germany following the Chernobyl accident, and fitted a regression model describing the exponential decay in activity.

The percentage of weapons fallout ¹³⁷Cs and cosmogenic ⁷Be retained by roofing material was investigated by Gjørup et al. (1986). They showed that the slope of the roof and type of roofing material were both important parameters. The retention by different roofing material was in the order: red tile > eternite > cement tile > silicone-treated eternite.

Table 6.2 Percentage of wet deposited Chernobyl radionuclides intercepted by various roof surfaces (Roed, 1987b)

The validity of these results was established by measurements made soon after the Chernobyl accident (Roed, 1987b). Table 6.2 shows that for 45° slopes, silicone-treated eternite again had the least capacity for retaining ¹³⁷Cs (22 per cent of deposition), while untreated eternite had the greatest (87 per cent deposition). The other Chernobyl radionuclides showed similar behaviour except for ¹³¹I which was only retained by the red tile.

Studies on the subsequent weathering of the adsorbed radionuclides showed virtually no removal of ¹³⁷Cs, with < 5 per cent washed off from August to December 1986. Sandalls et al. (1986) and Wilkins (1987) found slightly higher removal rates from a variety of building materials. The removal of activity was found to be a function of the total rainfall and therefore the retention behaviour was expressed in terms of the integrated rainfall (Figure 6.2). Wilkins (1987) conducted laboratory experiments on caesium deposited as a dry aerosol and then subjected to simulated rainfall. There appeared to be an appreciable removal of the activity by the first irrigation, but thereafter the retention behaviour followed that of wet deposition (Figure 6.2). Further experiments showed that no difference in behaviour could be discerned between old and new building materials or inland and coastal samples. Examination of the retention of Chernobyl radiocaesium by roofing materials collected from intact buildings supported these results (Nicholson, 1988c).

The overall efficiencies with which building materials intercept and retain depositing radionuclides may be expressed as the interception/retention factor (IRF) (Sandalls, 1987a):

Figure 6.2 Detection behaviour of both wet and dry deposited ¹³⁴Cs on building materials. After Wilkins, 1987.

Sandalls (1987a) determined these IRF values for urban surfaces in Cumbria, 20 weeks after Chernobyl. He concluded that the overall retention ranged from 5 per cent for glazed roof tiles and up to 60 per cent for unglazed, while housebricks ranged from 18 to 34 per cent. Studies on decontamination of building materials by Sandalls (1987b) provide an insight into the retention behaviour of radionuclides. Non-porous materials such as glass showed little affinity for caesium and thus efficient decontamination could be achieved with pure water. Porous materials, however, exhibited a greater retention capacity for caesium. Since decontamination by ammonium nitrate appears particularly successful for concrete (> 90 per cent removal in 20 h), but less so for clay tiles (40 per cent in 20 h), it may be concluded that the availability and selectivity of ion exchange sites governs the interception and retention behaviour of radionuclides.

Allott et al. (1990) have derived the environmental half-lives and mean residence times of Chernobyl radiocaesium-contaminated dust in the UK town of Barrow-in-Furness. The radiocaesium levels in dust were observed to follow an exponential decrease with time. The environmental half-lives describing this behaviour ranged from 190 to 370 d for a variety of internal and external sites. Mean residence times for the dust, derived from these half-lives, ranged from 150 to 250 d. The broadly similar half-lives observed for the quite different sites within this town emphasized the pervasive nature of dust and its associated contaminants and the universal nature of the environmental processes controlling its behaviour in the urban environment.

6.5 DECONTAMINATION STRATEGIES 

6.5.1 INTRODUCTION

Since urban environments may potentially become contaminated from accidental releases of radioactivity and the artificial radionuclides released may have a long-term retention behaviour (Section 6.4), various decontamination procedures have been considered. The aim of these procedures is to reduce the level of contamination and consequently shorten the period of time for which interdiction is necessary.

6.5.2 DECONTAMINATION TECHNIQUES

From a knowledge of the retention behaviour of artificial radionuclides in the urban environment, a range of decontamination procedures may be identified. Certain techniques are simply the rigorous application of normal practices, such as road sweeping, digging gardens or vacuuming indoors. Others are more comprehensive and even destructive, including removal of the road surface (road planing), ploughing of parks and roof replacement.

6.5.3 COST AND EFFECTIVENESS OF TECHNIQUES

In order to assess the overall cost of applying a decontamination technique to a city, compared to the dose reduction achievable for the population, it is necessary to consider the following factors:

1. Distribution of the deposited material with respect to different urban surfaces.
2. The contribution of the deposited artificial radionuclides on these surfaces to the dose rate.
3. The decontamination or dose reduction achievable on the individual surfaces, using appropriate methods.
4. The cost required to treat different urban surfaces.
5. The relative proportion of the urban surface types within a city.

The relative cost and decontamination achievable for various techniques has been determined for ¹³⁷Cs in relation to the unit area of the surface treated (Roed, 1988; Tawil et al., 1985). These data have been used to calculate the overall percentage dose reductions which may be achieved by the decontamination techniques (Roed and Andersson, 1991; Robinson et al., 1990). The models employed took account of the different interception and retention abilities of the urban surfaces, the shielding properties of the buildings and population habits (e.g. where they spend their time). The overall relative costs of treating unit area of city were derived from the individual surface data and the proportion of the surface within the city.

Figure 6.3 Relative cost and effectiveness of different decontamination techniques for the urban environment.

Comparing predicted levels of dose reduction with the city costs (Figure 6.3), the most cost-effective techniques appear at the bottom of the graph. The two techniques which are apparently the most cost-effective of those considered are digging gardens and road sweeping. Replacing roofs would be least cost-effective.

6.5.4 STRATEGY PLAN

In defining a strategy which optimizes decontamination in the urban environment, the cost and effectiveness of the procedures must be considered along with other factors, such as practicability and time required to complete the treatment. These have been incorporated into the schemes presented in Table 6.3 (Roed and Andersson, 1991). The relatively inexpensive procedures of digging the garden and defoliating trees may reduce the total dose to about 30 per cent and should therefore be of first priority. Decontamination of walls and internal surfaces, however, would have little effect on dose and should be given the lowest priority.

Table 6.3 Decontamination strategies for the urban environment

6.6 SUMMARY

Dusts and pollutants may enter an urban environment by either wet deposition, mechanical transport or erosion of building material. The major contamination processes in the outdoor environment appear to be wet and dry deposition, with the former leading to much greater deposition per unit time. Dry deposition is highly variable due to the complex variety of surfaces and wind domains, whilst wet deposition may possibly be enhanced for cities with populations greater than one million, because of their influential increase on precipitation events. Contamination of the indoor environment appears to occur mainly via the mechanical transport of dust on the bottom of shoes.

A major difference between urban and rural areas results from the different interception abilities of pervious and impervious surfaces. In open rural areas the wet deposition of pollutants usually results in fairly uniform contamination. However, wash-off from impervious urban surfaces causes the surface contamination within cities to be highly variable.

The retention behaviour of pollutants deposited on urban surfaces will be dependent on the nature and rate of removal processes. Weathering in the outdoor environment is probably governed by wash-off, whilst cleaning constitutes the major removal process in houses. Resuspension in urban environments may be greater due to vehicular disturbances; however, this has to be balanced against the possibility of a reduction in wind-induced resuspension, resulting from lower windspeeds. Contaminants deposited on pervious surfaces will be subject to long-term weathering processes with the effective half-life of ¹³⁷Cs after 4 years being little different from its radioactive half-life. A faster weathering rate has been observed for the dose due to radionuclides on road surfaces, whilst half-lives of 190-370 d have been reported for dust contaminated with artificial radionuclides.

These studies form the basis of modelling artificial radionuclide behaviour in the urban environment and may therefore be used to predict the dose reduction achievable for different decontamination techniques. When the cost of the techniques is considered, it is apparent that digging gardens and defoliating trees should be of first consideration in a decontamination programme.

Artificial radionuclides have been used successfully in the study of individual urban environment pathways, with particular attention being paid to dry deposition and mechanical transport. Future studies on wet deposition, resuspension and wash-off would benefit from the use of these tracers. However, the greatest value and future scope of artificial radionuclides as tracers in the urban environment is probably in the study of the retention behaviour of pollutants, so that urban models may be improved.

ACKNOWLEDGEMENTS

We are indebted to Professor J. B. Ellis (Middlesex Polytechnic), Dr K. W. Nicholson, Dr F. J. Sandalls and Mr J. A. Garland (AEA Technology, Harwell) for their constructive criticisms and extremely helpful comments on this chapter. The financial support of the National Environment Research Council to R.W.A. is gratefully acknowledged.