Document KJwVdEZYX9847n0Ra8Zv0Vm2w

ENVIRONMENTAL RESEARCH 36, 144-159(1985) 6650 Code 01 02 43 60 Ambient Air Concentrations of Asbestos Fibers Near the Town of Asbestos, Quebec 74 81 ^vt ,V ^ # Bhawan Singh and Jean-Pierre Thouez Dipartement de Geographic, Universal de Montreal, Montrial, Quebec H3C 3J7, Canada Accepted March 5, 1984 Ambient air concentrations of asbestos fibers were measured during the period June 20 to August 12, 1980, at three locations; Danville, Asbestos, and WottenviDe in the eastern townships of Quebec, Measurements were done with low-volume samplers and measure ment periods extended from 3 to 13 days. Fiber counts were done by means of electron microscopy. Results indicate that overall fiber concentrations are related to atmospheric stability and to the direction of the prevailing wind with respect to the source of emis sion. C 19(5 Academic Pres*, loe. INTRODUCTION The commercial mining of asbestos, whether of the serpentine or the amphibole variety, leads to several types of hazards and inconveniences, one of the major ~ of these being the pollution of the ambient air. Emanations into the ambient air normally ocetr during mining of the mineral, transportation of the raw fibers, and the various manufacturing processes utilizing asbestos as a raw material. The dangers to health, especially the incidence of asbestosis, a type of pneumonoconisis or disease of the lung, pleural and peritorial mesothelioma, and even lung cancer and other diseases arising from these airborne fibers have been already well documented (National Institute for Environmental Health, 1973; McDonald and McDonald, 1975; Buchanan, 1979; Jones, 1979; Capel, 1979; Newhouse, 1979; Levine et al., 1979; Selikoff and Hammond, 1979). This risk to human health arises mainly from the inhalation of asbestos fibers that are contained in the ambient air. In the eastern townships of Quebec, the dominant fiber is of the chrysotile variety belonging to the serpentine mineral group. Chrysotile fibers are typically curved in appearance and occur in.open bundles which tend to split into fibers and fibrils (Chariebois, 1978). The dimensions of these fibers and fibrils tend to vary in both length (ranging from 1000 A to tenths of millimeters (Nicholson et aL, 1970)) and in diameter (ranging from 20 to 40 nm (Chariebois, 1978)). In addition, chrysotile fibers, which are generally white in appearance, have a specific gravity of 2.4-2.6 at-room temperature, a tensile strength of562579 mg/m2, and have a relatively low resistance to add breakdown. At the town of Asbestos where there exists a single open-pit mine whose area is about 3 km2, total annual production in 1976 was 600,000 metric tons (Brulotte, 1976). In our study we monitored ambient air concentrations of asbestos fibers at three locations during the summer of 1980; at Asbestos where the mine is located, at Danville which lies to the west, and at WottenvQle which lies to the east. Fiber counts were performed by means of electron microscopy. Variations in fiber con- OOI3-935I/85 53.00 Copyright C 19(5 by Academic Pres*. Inc. AO rights of reproduction in any fora reserved. 144 C35794 1293 PLAINTIFF'S EXHIBIT AL-1198 I ASBESTOS FIBERS IN AMBIENT AIR 145 centrations in time and space were examined with respect to prevailing meteo rological conditions, mainly atmospheric stability and turbulence and horizontal wind speed and direction. ATMOSPHERIC. DIFFUSION Ambient air concentrations of asbestos fibers represent the normal background concentrations for the region in question (Sdbastien el at., 1979) plus local pol lution which, under specified conditions, can exceed background concentrations by several orders of magnitude (Bruckman and Rubino, 1977). Local pollution is mainly the result of the difficulty in reducing thf emissions of fine particles of asbestos during excavation and factory operations'fHarwood el at., 1975), emis sions that occur during transport of the asbestos mineral from mines to factories and thence to waste disposal sites (Harwood, 1971), and emissions from open pits and disposal sites (Ase et a!., 1976). In addition, one can mention the emis sions associated with workers involved in the various mining and manufacturing activities who transport and liberate fibers from their clothes and body. In this study, the focus is on local pollution, that is to say on fibers emanating from a local point source. Two classes of factors generally determine the level of air pollution at a partic ular location. First, there is the nature of the relevant emissions and of the source regions. Such distinguishing characteristics as the physical, chemical and geo metric properties of asbestos fibers, as have been described above, are critical. The actual release of these fibers into the ambient air is essentially accomplished through mining operations such as blasting, factory operations such as crushing, and transport of such materials as crushed serpentine rocks and tailings in open bodied trucks. We find it appropriate to describe the source region as an extended or large point source whose boundaries are delineated by the extent of the mining pits and adjoining milling plants and of the disposal sites (see Fig. 1). Second, there is the state of the atmosphere. Wind can accelerate and even initiate the release of fibers into the air. After release into the ambient air, how ever, further dispersion of the fibers is strictly controlled by atmospheric turbu lence and wind speed. The intensity of turbulent diffusion that is mechanically generated depends upon the horizontal wind speed as it influences internal shearing between moving air layers and upon surface roughness as it controls frictional shearing between the ground and the wind. The turbulent eddies gen erated in this manner are characteristically smaller. Thermal diffusion on the other hand, whose turbulent eddies are typically larger, is essentially a function of surface beating, mainly through solar radiation receipt and absorption. Generally if the size of the turbulent eddies is smaller than the cloud of fibers, they will diffuse it; if they are larger, they will tranpsort it (Shaw and Munn, 1971). - The wind field is critical with respect to horizontal dispersion. The wind speed determines both the distance of downwind transport and the amount of pollutant dilution, whereas the wind direction controls the path across which the pollutants are diffused and transported. When wind is blowing, asbestos fibers are dispersed both in the along-wind direction and by turbulent eddy diffusion in the aciosswind and vertical directions. Atmospheric stability however, as characterized C35?94 1294 Fid. I, General she of reseurch. 146 SINGH AND THOUEZ C35794 1295 ,, $` ASBESTOS FIBERS IN AMBIENT AIR 147 by the ambient air thermal stratification not only controls the intensity of tur bulence and wind speed but also the depth of the surface mixed layer. This latter factor is critical in determining the extent of the vertical dispersion of fibers. METHODS AND ANALYSES Ambient air concentrations of asbestos fibers were measured during the period 20 June to 12 August 1980 at three locations in and around the open-pit Jeffrey mine located near the town of Asbestos, Quebec (Fig. 1). The area is fairly uniform topographically with elevations ranging from about 210 to about 375 meters (Fig. I). The first site, at Danville, was about 3 km west of the mine and the tailing disposal site. The sampling equipment was-located in a fairly extensive grass-covered open site adjoining the town's municipal building. The second site was at Asbestos, a town situated on the eastern extremity of the mining site. The sampling equipment was installed on the roof of the town's municipal building, an elevation of about 10 meters. Wind speed and direction were also measured at this location. The third site was located at Wottenville, a rural town situated about 11 km east-south-east of the mining site. The sampling equipment was located in an open site among cultivated fields. The monitoring of ambient air concentration of asbestos fibers was performed by means of low-volume air samplers designed by Sdbastien et al. (1979) and modified by the Institute of Occupational Health and Safety, McGill University. The face velocity of this sampler, using a Gast Rotary Carbon Vane pump, is 8.4 cm s'1. This gives a flow rate of about 5 liter min'1. The effective filter area is roughly 962 mm2, with a diameter of about 47 mm. Upon this base filter was placed a metrical membrane filter of pore size 0.8 pm and diameter 47 mm for the collection of fibers. The height above ground or roof top, of samplers, was about 1.5 m. Samples were collected at intervals that varied from 2 to 6 days according to the sequence of weather conditions, except for one instance where the sampling period lasted 13 days because of the absence of an operator. Un fortunately, the sampler at the Asbestos site broke down after 2 weeks of oper ation and as a result, data for this site is missing after 4 July. Upon retrieval, the filters, with fibers attached, were enclosed in sealed containers, kept as immobile as possible and stored at room temperature. The preparation ofgrids for counting of fibers via electron microscopy followed very closely the method suggested by Sibastien et al. (1978). A portion (I cm2) of each filter, on the surface of which were deposited the asbestos fibers, was removed. These were then inserted into conical Pyrex tubes which- were then placed in a low-temperature asher (LTE 507) and subjected to activated oxygen for a period of about 6 hours. When ashing was complete, the residue was re suspended in 100 ml of specially filtered distilled water. To facilitate the counting of fibers, the residue was then ultrasonified for IS min in an attempt to separate bundles into single fibers. The liquid mixture containing the chrysotile fibers was then filtered through a polycarbonate nuclepore membrane (pore size, 0.1 pm; effective diameter, 35 mm). After drying, a portion of the filter was coated with a carbon film that was deposited via evaporation in an enclosed vacuum. Portions of the coated filter C35794 1296 148 SINGH AND THOUEZ were placed directly on 200-mesh electron microscope grids. Two grids were prepared for each sample. A transmission electron microscope .was then used for counting the fibers on each of the two grids. Fibers were identified according to their geometric prop erties: fibril appearance and hollowed cylindrical ends. Magnification ranged from 2000 to 10,000 times by increments of 2000 and from 10,000 to 50,000 times by increments of 10,000. For each grid, five fields, each measuring 80 x 80 jun, were chosen for mea surement of the number and length of fibers according to the criteria described previously. Fiber counts, expressed as a mass concentration, were derived in the manner described below (Gibbs and Rowlands, 1980; Gibbs and Hwang, 1975). First the number of fibers per liter of air sampled (FV) was derived as (1) where FV = number of fibers per liter (liter-1), IT = total fiber count for the 5 fields chosen, EFA = the effective filter area (962 mm2), ST. = total surface area of the 5 fields chosen (32,000 pm2), FR = total volume of air (liters). The factor EFA represents the total effective area of the original 35-mm filter and the factor 9.62 represents a dilution factor since samples of the original filter (100 mm2) were redeposited upon a larger filter area (962 mm2). In order to arrive at the final value of fiber concentration, the following rela tionship was exploited (Gibbs and Rowlands, 1980) FC = (FV x 103) x(25 x 10-3) x ((Da)2 x x LX) (ng m-3) (2) where FC = fiber concentration in ng m-3, D = mean diameter of each fiber, here chosen to be 0.08 pm, IJL *= mean length of fibers (pm). In the above equation (2), the first term converts liters (liters-1) to cubic meters (m-3); the second term represents the density of chrysotile fibers (ng m-3); and the last term represents the total volume of fibers (pm3). During the period of data collection, wind speed and direction and general synoptic conditions relating to sky condition, air temperature, and humidity were recorded at the Asbestos site. In addition, the surface synoptic charts and the corresponding tephigrams were analyzed so as to derive estimates of atmospheric stability. Wind speed and direction and air temperature and its corresponding dewpoint, together with the degree of cloud cover, were extracted from the syn optic charts. The vertical temperature lapse rate and the mixing height were C35794 1297 ASBESTOS FIBERS IN AMBIENT AIR 149 calculated from the change of vertical temperature gradient which was obtained from one of the surrounding stations where this data is available, i.e., Maniwaki, Quebec, Portland, Maine, or Albany, New York. The choice of station depended on the direction of air movement shown on the synoptic charts. Two indices of stability, chosen on the basis of their relativeness to the study and of the availability of data, were then calculated. The first stability parameter has been proposed by Shaw and Munn (1971). They state that over relatively uniform surfaces, as are found in our study area, an appropriate stability param eter is the dimensionless stability index (SI) which incorporates both buoyancy and wind effects and which is expressed as where g = acceleration due to gravity (m sec 2) T = temperature of ambient air (K) near the ground (Z m) 3[ = az the vertical temperature lapse rate ("K m-1) T = adiabatic lapse rate ("K m "') Z? = mean horizontal wind speed (m sec"1) z = height of u (m). Furthermore Shaw and Munn (1971) suggested the use of another index of atmospheric diffusion, namely the ventilation coefficient VC (m2 sec-1) where VC = MH x u ! (4) MH = the mixing height of the turbulent layer (m). Because the wind speed does not change much with height in the turbulent layer, the parameter VC represents the rate of horizontal diffusion of fibers. Following the lead ofIslitzer (1965) and ofPasquill and Smith (1971), the indices of stability, as calculated by equation (4), are then related to PasquflTs (1961) stability classes so as to arrive at a more descriptive, yet objective, characteriza tion of atmospheric stability (see Table 4). DISCUSSION OF RESULTS Tables 1, 2, and 3 summarize the relevant data and the final calculation of the concentration of asbestos fibers for the three stations under study. Due to instru ment malfunction, data are missing for Asbestos from the 4th ofJuly onward and for Wottenvflle from the 27th of June to the 4th of July. However, it is evident from the available data that generally the highest concentrations occurred at As bestos and the lowest at Danville (see Table 5) due to downwind distance from the mining area. 150 SINGH AND THOUEZ C35794 1299 TABLE I Summary op Relevant Data for Calculating A mbient A ir Concentration op Fibers: Da n v ille ASBESTOS FIBERS IN AMBIENT AIR TABLE 2 Summary of Relevant Data for Calculating A mbient A ir Concentration of Fibers: A sbestos 151 C35794 1300 152 SINGH AND THOUEZ C357941301 TABLE 3 Summary of Rbi.bvant Data for Calculating A mbient A ir Concentrations of Fibers: Wottenville  ASBESTOS FIBERS IN AMBIENT AIR TABLE 4 Descriptive Stability Classes Derived from the Stability Index (SI) Range of SI values Stability class -1.0 to -0.7 -0.5 to -0.4 -0.4 to -0.18 -0.17 to -0.13 -0.12 to -0.01 0 0.01 to 0.03 0.03 to 0.05 0.05 to 0.11 >0.11 Extremely unstable Moderately unstable Moderately to slightly unstable Slightly unstable Slightly unstable to neutral Neutral Neutral to slightly stable Slightly -stable Moderately stable Very stable 153 When fiber concentrations are related to atmospheric stability and windiness, several interesting relationships become evident. For the station at Danville, which lies to the west-north-west of the mining area, wind direction seems to be critical in determining ambient air fiber concentrations (Fig. 2): winds from the east are associated with highest fiber concentrations, whereas westerly and south erly winds are associated with lowest concentrations. Atmospheric stability also seems to be a factor since lowest concentrations occur under relatively unstable conditions, when mixing is enhanced, and highest concentrations occur under more stable neutral conditions, when vertical mixing is subdued. Examining the values of the ventilation coefficients, their relationship to fiber concentration is less clear (Fig. 3) although it appears that higher fiber concentrations are generally related to higher ventilation coefficients, the latter being the consequence of in creased windiness and hence turbidity, which tends to increase the diffusion of fibers to nearby stations. Again, the influence of wind direction seems to be dominant. Fig. 2. Fiber concentration, stability index, and wind direction, Danville. C35794 1302 154 SINGH AND THOUEZ FiC. 3. Fiber concentration, ventilation coefficient, and wind direction. Danville. For the Asbestos station, which lies to the immediate east of the mining area, it is very evident that wind direction is the dominant factor, with westerly winds being associated with highest fiber concentrations and easterly winds, the reverse (Fig. 4). The influence of atmospheric stability is also evident with highest con centrations occurring under near neutral conditions when diffusion is restricted. These relationships are also evident when considering the ventilation coefficients: VC (Fig. 5). In Fig. 5, the influence of wind direction however seems to overrate that of ventilation, although higher ventilation coefficients seem to give rise to C357941303 ASBESTOS FIBERS IN AMBIENT AIR 20- W *40- N-W 155 120- ,Xo o- N-W 40- E OH-----1------ j----- i------ i---- i-------1---- ------ 1-----1------ r 0 2 4 10 VC 10 ) FlC. 5. Fiber concentration, ventilation coefficient, and wind direction. Asbestos. higher fiber concentrations, this again being most likely due to increased windi ness. For the town of Wottenville, which lies to the east-south-east of the mining area, highest fiber concentrations are not entirely explainable in terms of wind direction. It seems that highest concentrations occur when winds are south easterly, that is coming from a source region other than the mining area (Fig. 6). The relatively unstable atmospheric condition on this occasion (Fig. 6), however. C3579* 1304 156 SINGH AND THOOE2 Fig. 7. Fiber concentration, ventilation coefficient, and wind direction, Woctenvdic. indicate that diffusion of the fibers is intensified so that wind direction may be a poor indicator of fiber concentration distribution. For this station then, it appears that atmospheric stability (SI) is more criticaTthan wind direction in determining the ambient air concentration of fibers. When examining the influence of the ventilation coefficient VC (Fig. 7), its relationship to fiber concentration is not well defined although again, generally, it appears that higher fiber concentrations occur when VC is moderately high signifying that increased windiness and mixing allows for a greater diffusion of the fibers away from the mining areas. Also, for the same wind direction, higher fiber concentrations are related to higher VC values, this emphasizing the influence of windiness and turbidity. Putting these various variables and parameters together in Table 5, it is evident, for the time periods when* data are available, that highest fiber concentrations occur at Asbestos, this being mainly due to wind direction with respect to the source region and to proximity to the source region. Even these very local con centrations however, as shown previously, seem to be sometimes related to at mospheric diffusion processes. Fiber concentrations at the other stations, espe cially WottenvQle, however seem to be more closely related to atmospheric dif fusion than to wind direction. Furthermore, for each station, there exist significant temporal variations in fiber concentration, these being due very likely to the sequence of synoptic events. CONCLUSION Because of the paucity of data points, we were unable to carry out exhaustive statistical tests of our hypotheses. However, several conclusions can be drawn from the results of this study. First, it seems that there are large temporal vari ations in ambient air fiber concentrations in the study regions regardless of site, and that these variations are essentially related to synoptic conditions, mainly windiness and turbulence. Furthermore, for stations located close to the source ASBESTOS FIBERS IN AMBIENT AIR 157 .ziz ^(OU Cl] I co <o co oo > cococo T3 C. 5 60 *> DO 09 03 m <C V rV. ,| inn <N ^C O' rj (N *-i IN - ^N O" ?O) rJ *> (J if * |S Qu uQ uC QOUflUDi Ua , UI Oi Ua u c o' s> b. "S>. .*1_ &-P Cu QoQu ^ ^ r- . -- CN ~- | OnS r IcN** ^ -- mmm s.u: *S v 2 S-- 3 -0S0 S oS III r oo ^*r C? O*n C*0\ ^s rI- toI --I --~I *%n> mtri ^-- -- Ov n m f| o] 09 (N 03 09 --Nn--N>O^--*OnOW>0\ II S3! T2?i J*ofo II 1 ass P* NT E s3i VV Sz "* <Q I? Ij 2 4) . wa >s 11 3 III g >* --g5jonillc--v < c ~ Jf = od** t C35794 1306