Click for next page ( 120


The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 119
7. STUDYING THE EFFECTS OF ATMOSPHERIC DEPOSITION ON ECOSYSTEMS The widespread dispersal of trace pollutants--many of them toxic, carcinogenic, or mutagenic--has been discovered only recently, through tremendous advances in the technology of chemical analysis in the past two decades. These technical developments have increased the sensitivity of detection of most chemicals by orders of magnitude. The methods and programs available for assessing the environmental and biological effects of these trace atmospheric pollutants, however, have not reached a comparable degree of sensitivity or reliability. Substances are usually tested for their effects over short periods--rarely as long as one complete life cycle--and at relatively high dose rates, either singly or in very simple combinations with other pollutants. Often such tests are inadequate to assess long term, chronic problems. For example, it took thirty years of careful statistical studies of human populations, combined with experimental studies of animals, to document the linkage between smoking and lung cancer (U.S. DREW 1979~. There are many examples of unexplained mortalities that could well be caused at least partially by multiple pollutant stresses. Among them are the decline of leopard frog populations in Minnesota (McKinnell et al. 1979), the high cancer mortality in workers in the petrochemical industry (Thomas et al. 1980), and the high incidence of lip cancer in sucker fish in Lake Ontario. Epizootics of benign skin tumors (papillomas) were found in white sucker, Catastomas commersoni populations throughout the Great Lakes. However, elevated tumor frequencies (50.8%) were found in populations clustered around an industrial complex on Lake Ontario concerned with petrochemical refining. In this region there was an anatomical shift in the tumor location to the lips, an anatomical site which has near constant contact with bottom sediments (Sonstegard and Leatherland 1980~. Disentangling the chain of cause and effect responsible for such mortalities is an extremely difficult but extremely important task. Mortalities in animal or plant populations may serve--as did the coal miner's canary--to provide early warning of human hazard. Linkage s among pollutants are also of critical importance in determining how whole ecosystems are affected. Oxides of sulfur and 119

OCR for page 119
120 nitrogen are strongly linked in coal combustion, and both are linked to trace metals and trace organic molecules (e.g., polycyclic aromatic hydrocarbons) produced by combustion. In metal smelters there is less linkage between SOx and NOX and a much stronger link between SOX and metal emissions. In vehicle exhaust, NOx predominates strongly over SOx and is strongly linked to trace organic materials. The total effect of the linked emissions on an ecosystem depends on the combination of pathways and interactions of the individual emissions. Interactions along ecosystem pathways are especially diverse. For example, sulfate and nitrate deposited from the atmosphere to terrestrial ecosystems may travel together as rainwater percolates through soils, while hydrophobic and lipid-soluble organic molecules will follow other pathways. Because of greater biological demand and generally smaller supply, a greater proportion of nitrate than sulfate may pass into storage in biotic compartments of the ecosystem, leaving relatively more of the sulfate to pass through to streams and lakes. In aquatic or wetland habitats some sulfate may be reduced to sulfide in anaerobic environments and end up deposited as ferrous or other metallic sulfides in a sedimentary sink. Both nitrate and sulfate may be taken up by organisms and reach the sedimentary sink by a detrital pathway. Alternatively, some sulfate in anaerobic habitats may be reduced to volatile sulfides, and some nitrate to gaseous ammonia, which return to the atmosphere, where the sulfides may again be oxidized and interact with ammonia to form ammonium sulfate aerosols. Metals such as lead, which often accompany sulfate in particulate pollution, will follow yet another path, becoming strongly adsorbed to ion-exchange sites on soil clays or humic acids. Mercury, another metallic component of particulates from coal combustion, may become methylated by sedimentary microorganisms and thus be returned to the atmosphere as volatile dimethyl mercury; it can also volatilize as metallic mercury (Wood 1974~. Still other metals may be mobilized from soils by acidification due to atmospheric depositions; among these are nutrient elements such as calcium and potassium and toxic elements such as aluminum and zinc. Thus, not only are individual organisms affected by combinations of pollutants, but the ecosystem as a whole is affected by the combination of pollutants interacting with different parts of the system in various ways. Current knowledge of atmosphere/biosphere interactions suggests a number of broad-scale hypotheses concerning future ecological damage that could result from air pollutants. Of greatest global, long-term significance is the hypothesis that atmospheric pollutants such as carbon dioxide and oxides of nitrogen released in the course of energy generation may warm the planetary atmosphere (NRC 1977a, Kellog and Schware 19811. The consequences of such a "greenhouse effect" for the earth's terrestrial and aquatic biota are at present unpredictable, but major shifts in the global patterns of plant productivity and human habitation seem very likely (Kellogg 1978~. A second major hypothesis is that anthropogenic release of oxides of sulfur and nitrogen may alter appreciably the cycles of sulfur and nitrogen, again with consequences for the biota that are difficult to

OCR for page 119
121 predict. On the one hand, both elements are plant nutrients that are deficient in certain soils; on the other hand, both contribute to "acid rain" (more correctly acid deposition), which is already a serious regional problem in North America and Europe (Dochinger and Seliga 1976, Hutchinson and Havas 1980, Drablos and Tollan 1980~. In addition to directly affecting the biota, acid precipitation can alter substantially the cycling of both nutrients and toxicants in acid-sensitive ecosystems, particularly those on crystalline Precambrian rocks. A most recent and unexpected manifestation of acid-rain effects is the acidification of ground waters in Scandinavia tHultberg and Wenblad 1980), discussed in more detail in Chapter 8. A third, less well supported hypothesis is that energy-related air pollution is becoming a serious cause of biotic impoverishment. Increased acidification of fresh water streams and lakes by acid precipitation has destroyed sensitive populations of fishes and other aquatic organisms. Fallout of toxic heavy metals from the atmosphere and gaseous oxides of sulfur and nitrogen have been shown to have similar effects. Even if losses are only of local plant or animal varieties, a serious loss of genetic diversity may result. As pollution continues and increases, the effects will become more widespread and more severe. One of the most difficult problems to assess is the possibility that chronic low-level pollution of the atmosphere by teratogens, carcinogens, and mutagens is affecting the health, reproductive capacity, survival, and disease resistance of the biota, and thus indirectly the very structure and function of the biosphere that is our life-support system. By contaminating the environment and by causing the extinction of sensitive genotypes, air pollutants, acting singly or in combination, may change the biosphere in ways that are essentially irreversible. The possibility of major alterations in planetary ecology and biogeochemistry raises important questions concerning the kinds of programs that are necessary to predict, investigate, and cope with the problems. PREDICTING ANTHROPOGENIC EFFECTS ON ECOSYSTEMS Separating human-induced ecological changes from natural ones and measuring them is not a simple matter, particularly when the changes occur slowly. In describing such changes and in establishing causality, there is often an implicit assumption that in the absence of human intervention ecosystems are in some sort of distinguishable and predictable successional state. Ecological theory assumes that following small perturbations, an ecosystem will return to the original successional pattern. This is a useful theoretical concept but a difficult one to apply to particular ecosystems. Clearly distinguishable ecosystem successions are hard to predict even when human interference is absent. The maximum perturbation that an ecosystem can tolerate without irreversible change is not generally known. Moreover, there may be more than one possible successional pattern, even under natural conditions. On a long time scale,

OCR for page 119
122 climatic oscillations will influence the nature of the successional state. Superimposed on this oscillating system is the natural tendency of some species to become extinct while new ones develop. Human-induced changes must be considered against the background of a normal range of natural trends and fluctuations. The system with which we must deal is shown, greatly simplified, in Figure 7.1. Let us review, then, the types of information necessary for predicting the biological effects of atmospheric deposition. For predictive purposes, one would like to know which source characteristics and emission processes govern the various substances emitted, their forms and properties, and their rates of emission. Anthropogenic sources and their emission rates are readily measurable, but natural fluxes are difficult to estimate. Although the relationship between transport distance and residence time for particles of different size and type is reasonably well understood (Figure 5.1), much remains to be learned about pollutant transformations in the atmosphere. o The interfaces between atmosphere, lithosphere, hydrosphere, and biosphere (Figure 7.1a) are of critical importance in elucidating the ecological effects of pollutants. Transfers across them are controlled by the physical and chemical properties of the pollutants (or their alteration products), by the characteristics of the receptors, and by the nature of the transfer processes. Some aspects of transfer, such as wet deposition from the atmosphere, are relatively well understood; others, such as dry deposition, are not. Some important pollutant transfers and their effects are relatively simple to follow. For instance, SO2 transfer to plant leaves (A ~ O in Fig. 7.1a) causes a direct visible injury. However, SO2 may also exhibit a variety of indirect effects. By transfer to soils and oxidation to H2SO4 it may affect cycles of nutrients and toxicants and thus plant growth (A - S -- O in Fig. 7.1a). By transfer direct to waters, as H2SO4 in snowmelt percolating over frozen ground, it may result in fish kills (A - W - O in Fig. 7.la). Or by transfer to soils and then to waters as H2SO4, it may inhibit fish reproduction {A -- S -_ W _ O in Fig. 7.1a). It may even react through organisms back to the environment, as when SO2 fumigation kills the vegetation around a metal smelter, leading to severe soil erosion (A - O -- S in Fig. 7.1a). For each of the cases mentioned above, attention will have to be paid to all of the aspects of pollutant transfer shown in Fig. 7.lb. The transfer of a pollutant within and between the diverse terrestrial and aquatic compartments of ecosystems is extremely complex, depending upon the characteristics of both the pollutant and the numerous receptor sites and transfer processes within an ecosystem. Interactive effects upon the various species that dominate different trophic levels are especially difficult to predict, because each receptor species will respond differently (in greater or lesser degree) to a given pollutant and also to its effect upon other species.

OCR for page 119
123 (a) (b) (A) N/9 Atmosphere /\ (S) Soils ~ _ , .+ _ T ~ 1: 1 (O) I Organisms /' W! (W) 1~:' I Waters ~;;1 Em ission Concen - trat ion at Point Emission Amount Em itted Over Unit Time = Transport ~ Rate of Transport \/ diffusion 4;vironmental) |Rate of |Removal or |Accumula- Concentration at Target or Point in Environment Amount Reach i ng Target Over Unit Time / OCR for page 119
124 . . . The position of an element on the periodic table or its participation in known inorganic or organic chemical reactions offers useful clues to its behavior in the biosphere, because chemically similar elements are often treated similarly by organisms. Background concentrations of potentially toxic substances must be known, because even small anthropogenic releases of such substances may cause problems if the substance occurs naturally in near-toxic amounts. For example, the doubling of mercury emissions has apparently been enough to cause undesirable concentrations of the element in fish in a number of areas. The increases in atmospheric nitrous oxides and hydrocarbons from anthropogenic emissions can cause an increase in the photochemical formation of ozone to levels that are severely damaging to vegetation (NRC 1977b). Increased aluminum concentrations in surface water caused by acid deposition have also proved detrimental. Substances not normally found in near-toxic concentrations but released in large quantities by anthropogenic sources may also be problems. Examples are lead and oxides of sulfur and nitrogen. Synergisms and antagonisms among atmospheric pollutants must be considered. Effects of metallic toxicants (e.g., Hg, Al) are often more severe under acid conditions. Selenium may mitigate the effects of mercury (Rudd et al. 1980~; both are components of coal combustion products. It is also noteworthy that the NOX contribution to acid precipitation may supply both a limiting nutrient, nitrogen, and a potential toxicant, hydrogen ions, to receptor ecosystems, with the nutrient contribution likely to be more important in the short run and the acidifying component more significant over the long term (Tamm 1976, Abrahamsen 1980~. Residence times of substances in organisms, or in the substrates in contact with organisms, influence their toxic effects. Lead, strontium, and radium, which behave chemically like calcium, tend to be deposited in bone or calcareous shells, and their removal from organisms is therefore extremely slow. Lipid-soluble substances also tend to turn over very slowly. Organisms that deposit large amounts of fat will accumulate lipid-soluble pollutants most strongly; birds that accumulate great amounts of fats prior to migration often contain large quantities of chlorinated hydrocarbons and PCBs which can be released to the blood stream as the fats are used up. Hutchinson et al. (1979b) have shown that the degree of toxicity of a wide range of hydrocarbons, including chlorinated hydrocarbons, to freshwater green algae is predictable with some accuracy, based on a number of key physico-chemical parameters of the hydrocarbons. Aqueous solubility is one such parameter and the octanol-water partition coefficient is another, both simulate agueous-lipid partitioning. Infiltration into the

OCR for page 119
125 cellular membranes is also predictable, as detected by K and Mn leakage, based on these same molecular characteristics. Kenaga (1980) has shown that bioacccumulation in cattle and swine is highly correlated with bioconcentration factors in fish. He notes that bioconcentration factors in beef fat are negatively correlated with aqueous solubility and are positively correlated with 1-octanol-water partition coefficients. Radioisotopes such as radiocarbon and tritium, which are components of compounds subject to much greater rates of turnover in organisms, appear in general to be less harmful than substances that turn over slowly, at least in the short term--excluding genetic effects. Knowledge of food chains is important in tracing biomagnification. Top predators--i.e., organisms at the top of the trophic pyramid--will often concentrate most strongly substances that are affected by biomagnification. Predatory fish, for example, accumulate both chlorinated hydrocarbons and mercury. Information on distribution and abundance of sensitive receptors and distribution and size of pollutant sources is important for determining the effect of a particular pollutant upon a particular ecosystem. A maximum effect could be expected in a situation in which the receptor species is highly sensitive to a particular pollutant, the species is rare but ecologically important, and either the sources and receptors are located together or the pollutant has a long atmospheric residence time. Although we cannot provide foolproof prescriptions for the detection and evaluation of all pollutants, several general features characterize large numbers of undesirable pollutants, making it possible to categorize them to some degree. For instance, pollutants may be linked together at their source or share common transport or uptake mechanisms. They may also have the same targets in the biosphere, and exert similar effects upon them. A "comparative anatomy" constructed around similarities and differences among pollutants would reduce the number of surprise pollutants in the future, although there are always likely to be some pollutants with characteristics so unique as to defy prediction, or new categories of anthropogenic pollutants with properties vastly different from current ones. Table 7.1 is a detailed summary of the information required for an ecological assessment of a pollutant. This table reveals the characteristics that can make an atmospheric pollutant dangerous. First there must be significant emissions (A.1) of a substance with considerable effect upon organisms (E.1, E.3), ecosystems (F.), the environment (G.), or man-made materials (H.~. Long residence time in the atmosphere (B. 4) favors its spread but dilutes the amount deposited in any area. Resistance to transformation (Be 2) and detoxification by organisms (E.2) lengthen its persistence in the atmosphere and increase its resistance to degradation along ecosyste pathways (D.4~. However, some pollutants (e.g., inorganic Hg) are

OCR for page 119
126 TABLE 7.1 Major Factors to be Determined in an Assessment of the Ecological Effects of a Pollutant A. Emissions 1. Significant sources (coal, oil, natural gas, nonfossil fuels, natural processes) 2. Nature of source (stationary, mobile) 3. Nature of primary emissions (gases, particulates in different size fractions) B. Atmospheric transport 1. Significant modes (gases, particulates in different size fractions) 2. Physical and chemical transformation (type, degree) 3. Nature of transport (local, regional, global) 4. Residence times (up to days, weeks, months, longer) C. Deposition Dry (gaseous absorption, particle collision, settling) Wet (rainout, washout, as rain or snow) D. Ecosystem pathways 1. Readily transported (by water, or by volatilization back to atmosphere) 2. Likely to accumulate in sinks (e.g., soils, sediments) 3. Likely to accumulate in organisms (bioconcentration, biomagnification) 4. Readily degradable (abiotically, biotically) Biological consequences for microbes, plants, animals; in terrestrial, aquatic, wetland habitats 1. Toxicity (low to high, acute to chronic, affecting few to many organisms) 2. Detoxification and/or repair capability (low to high) 3. Type of effect (on health, development, growth, reproduction, phonology, behavior, heredity) Ecological effects 1. Energy flow (production, decomposition, storage) 2. Biogeochemical cycling (limiting nutrients, toxicants) 3. Structure (vegetation patterns, trophic levels, species composition, stability)

OCR for page 119
127 G. Effects on properties of the physico-chemical environment 1. Atmosphere (radiation balance, water balance, gaseous and particulate composition, visibility) 2. Hydrosphere (thermal properties, water balance, chemical properties) 3. Lithosphere (thermal properties, water balance, chemical properties) H. Effects on man-made materials 1. Economic (metals, other structural materials, fabrics, other organic materials) 2. Aesthetic (works of art, architecture)

OCR for page 119
128 readily altered to products (e.g., methylmercury) that are even more toxic. And in the case of SO2 and its alteration product H2SO4, the effects may be quite different, with the former affecting chiefly terrestrial ecosystems and the latter having a profound effect on sensitive streams and lakes {Glass and Loucks 1980~. Solubility in water (D.1) favors mobility in the ecosystem and the avoidance of sinks in soils and sediments (Dow. Lipid solubility, on the other hand, favors entry into organisms by transfer across cell membranes (D.5) and both bioconcentration and biomagnification along food chains (Dead. HIGHLY SENS ITIVE ORGANISMS AS EARLY INDICATORS Sensitive early indicators of pollutant stress may be useful in developing monitoring schemes to detect environmental stress in its initial stages. In most ecosystems, some species or groups prove much more sensitive than others. For instance, the bryophytes--i.e., mosses and liverworts--and lichens lack the waxy cuticle of the flowering plants, and thus their epidermal cell walls are directly exposed to the air. These cell walls have a highly charged cation exchange surface, and thus many deposited elements are retained by exchange. Such organisms are usually the most sensitive indicators of pollutant stress. Whole assemblages of lichens and bryophytes have become rare in urban areas (Holdgate 1979a, Johnsen 1980~. These organisms are so sensitive that their assemblages have been proposed as a general index of atmospheric pollution (Isecutant and Margot, discussed by Holdgate 1979a). Such organisms are also likely to accumulate trace substances. For example, trace metals have been shown to be concentrated from 1,000 to 50,000 times by bryophytes (Dvorak et al. 1978~. Ruhling and Tyler (1968 and 1970) have used the feather moss Hylocomium splendens for assessing lead loadings near highways in Sweden, and for examining north-south latitudinal gradients. Cesium residence times in mats of Cladonia lichens were found to range from 4 to 25 years (Liden and Gustafsson 1967~. Some aquatic plankton groups, such as the Chrysophyta or Crustacea, respond to minuscule quantities of pollutants (de Noyelles et al. 1980, Marshall and Mellinger 19801. Common responses include depression of photosynthesis and changes in species composition (de Noyelles et al. 1980~. Changes such as the above may in turn affect other members of the ecosystem that depend on the displaced species for food or protection. Among animals, Gammarus lacustris, a large benthic crustacean, and numerous molluscan species have proved more sensitive to acid precipitation than any other easily recognizable species (K.A. Okland 1980~. A North American animal with similar reactions, Mysis relicta, may prove even more useful. Like Gammarus, Mysis disappears when waters reach pH values of 5.8 to 6.0 or less (Schindler 1980a). The species is a glacial relict, and thus its distribution under earlier, unpolluted conditions is known quite exactly. Its absence in lakes

OCR for page 119
129 that are otherwise suitable may be an indicator of acidification. The mud minnow, Pimephales promelas, is equally sensitive to acidification (Schindler 1980a, McCormick et al. 1980~. In general, embryonic development is the most sensitive stage in the life cycle of an organism (McKim 1977), and the incidence of deformity or biochemical abnormality at that stage may serve as an early warning for more far-reaching effects occurring at later stages of the life cycle. While teratological and immunological studies are now commonly done with mammals, they have seldom been done on animal or plant populations in nature. The few studies that have been done on natural populations have proved to be very sensitive indicators of pollutant stress (e.g., Kennedy 1981, Daye and Garside 1980~. They may complement epidemiological studies of long-lived organisms to enhance our knowledge of the latent effects of long-term exposure to low concentrations of pollutants. Relatively tolerant accumulator organisms may also be useful in monitoring the dispersion and persistence of dangerous substances. For example, mosses and lichens readily accumulate radioactive fallout (Gorham 1958a, 1959) and heavy metals (Ruhling and Tyler 1971), and a global "mussel watch" is now monitoring the dispersal of dangerous contaminants (such as organochlorides, trace metals, petroleum hydrocarbons, and transuranic elements) in coastal marine waters (NRC 1980b). Prediction of sensitivity in organisms is sometimes difficult because a given organism can react in very different ways to different pollutants. For instance, lichens are exceptionally sensitive to SO2 pollution but are unusually tolerant of radiation stress (Hawksworth et al. 1973~. However, some reasonably useful generalizations can be suggested. In general, top carnivores are likely to be sensitive to persistent, fat-soluble pollutants capable of biomagnification (Carson 1962~. Species with small chromosome volumes are insensitive to irradiation by radioisotopes (Woodwell and Sparrow 1963, Woodwell 1967~. Opportunistic species that are colonists during early ecosystem succession also tend to be more resistant to pollutant stress than the usually longer-lived and more specialized organisms that inhabit relatively stable ecosystems. Presumably the opportunists are better able to tolerate diverse environmental conditions (Woodwell 1970) and also--because of high reproductive rates--to recover more readily from pollution damage. COMMON RES PONSES OF ECOSYSTEMS TO STRESS Ecosystems respond differently to different pollutants. For example, acid deposition affects chiefly aquatic ecosystems in watersheds with thin, coarse soils over crystalline igneous rocks. In this case, biotic effects are largely due to sensitivity of an abiotic compartment of the ecosystem. In contrast and as mentioned before, SO2, the gaseous precursor of acid rain, affects directly and primarily the organisms of terrestrial ecosystems--no matter what the geological substrate--because in water it is rapidly converted to

OCR for page 119
130 sulfate. The NOX that contribute to acid deposition, however, may be expected to have their greatest direct effect upon ecosystems whose vegetation is limited by nitrogen. In a more general way, ecosystems with a complex structure based on long-lived species are likely to be especially sensitive to pollutant stress (Woodwell 1970~. Woodwell (1970) pointed out the similarities in the responses of deciduous woodlands to two apparently quite different stresses, gamma radiation and SO2. He noted that in both cases large target trees close to a point source of either the radiation or the pollutant were badly damaged and that a zone of damage occurred. The most tolerant plants were sedges and grasses, both of which have their growing point--apical meristem--at or below the soil surface. This pattern appears to be fairly general. Species diversity decreases as a result of such stresses as SO2, radiation, fire, oil spills on land, extremes of cold (arctic and alpine), and high wind exposure or drought. The plants that seem best adapted to these stresses include a number of perrenial grasses and sedges. Annuals, which must complete a reproductive cycle each growing season, are very rare; large above-ground targets, such as trees and shrubs, are especially sensitive. Those species which have underground perennating organs such as bulbs and stolons or which are able to produce suckers from adventitious buds are likely to survive best. Vegetative reproduction is generally more successful under such stresses than sexual reproduction. Bryophytes and lichens are especially tolerant of a number of these stresses--e.g., cold, wind, and radiation--but, as we have seen, where the stress involves entry of a pollutant into cell walls, the absence of a cuticle is a crucial disadvantage. Lichens and mosses generally are susceptible to atmospheric pollutants, such as SO2, fluorine, and trace metals, which are absorbed on the exposed surfaces. They also may be susceptible to atmospheric hydrocarbons, as they are to direct application of petroleum. For example, of 31 species present initially at a tundra site in the Canadian arctic, 21 were bryophytes or lichens. But, one year after sprayed oil spills, a31 21 of these species were dead, while 7 of the 10 higher plant species had survived. After 8 years, 9 of these 10 were present, but only 1 bryophyte had reappeared (Hutchinson 1981~. It should be noted that these generalizations about stress-susceptible or resistant vegetation do not apply to the successional aftermath. When the effects of the pollution episode have diminished, pioneer species may be able to invade the previously stressed area de nova. Susceptibility of ecosystems to atmospheric pollutants seems to be reduced as annual precipitation is reduced, perhaps because toxicants can cause the greatest damage to organisms and populations that grow and metabolize rapidly, under conditions with optimum water and temperature. Emissions of SO2 from smelters in desert regions such as Arizona have much less effect on vegetation than equal SO2 emissions in temperate regions (Wood and Nash 19767. It therefore seems likely that rapidly growing tropical forests could sustain more severe damage from a given stress than temperate forests, and that

OCR for page 119
131 slow-growing arid or arctic ecosystems would be less sensitive. There is no dormant period in the tropics, while in both arid and arctic regions very extensive dormant periods occur, during which above-ground photosynthetic material is greatly reduced. Annual plants may be at special risk because they must successfully go through all the stages of sexual reproduction each year, and irregular episodes of exposure to harmful atmospheric pollutants could disrupt their reproductive cycle. One might predict that annuals in desert areas, where they are often abundant, could be locally eliminated while the perennials survive. Studies of the effect of smelters in plant populations in the Southwestern United States have confirmed this (Wood and Nash 19761. A newly emerging problem is the possibility that expanding use of diesel fuel in the U.S. auto fleet may increase emissions of suspended soot particles by more than an order of magnitude (Barth and Blacker 1978, Springer 19781. This problem is currently being investigated by the Diesel Impact Study Committee of the National Academy of Engineering (NRC, in press). Diesel particles are mostly in the submicron size range, and thus are very effective in reducing visibility. Of greater ecological importance, they carry a variety of adsorbed toxic, carcinogenic, and mutagenic compounds such as benzota~pyrene, and they are small enough to penetrate deep into animal lungs. The ecosystems most likely to be affected are those close to heavily traveled highways. Among primary producers, epiphytic lichens could well be sensitive, for reasons mentioned above. The micro- and meso-fauna that are exposed to concentrated through-fall from the canopy and that process leaf litter along food chains in the soil humus layer might also be sensitive. If biomagnification of lipid-soluble organic pollutants takes place, then carnivores at the top of the food chain ought to be regarded as potentially sensitive. In aquatic ecosystems adjacent to busy highways, the neuston (water surface) dwellers may be vulnerable, because many organic molecules concentrate to an exceptional degree in the surface film at the air-water interface. In the water beneath, filter-feeding zooplankton may well collect soot particles more or less indiscriminately along with their food, as may detritivores in the benthos. Cycle linkages may be of considerable importance in determining the total effect of air pollution upon a given organism and the ecosystem in which it occurs. For example, acid rain has both direct and indirect effects on fish; directly by toxicity of the hydrogen ion and indirectly by the mobilization of aluminum from soils. There is also a possibility, as suggested elsewhere in this report, that hydrogen ions may mobilize sufficient mercury from soils or sediments to become lethal as larger, older fish accumulate the element to high levels. While this report has emphasized the potential significance of atmospheric deposition in terrestrial and freshwater ecosystems, it is important to note that similar concepts apply to estuarine and coastal marine environments. Considerable evidence is available documenting an enhancement in concentrations of heavy metals and organic compounds in estuarine and coastal environments by atmospheric transport from

OCR for page 119
132 anthropogenic sources (Farmer et al. 1980, Bertine and Goldberg 1977, Seki and Paus 1979, Crecelius et al. 1975, Rodhe et al. 1980~. Understanding of the fate of anthropogenic pollutants delivered to the oceans by atmospheric transport is inadequate. Results from existing studies concerning the fate and ecological effects of anthropogenic pollutants show that the toxicity of trace metals to marine organisms is related to both concentration and chemical speciation (Steemann and Wium-Andersen 1970, Harriss et al. 1970, Allen et al. 1980~. Elevated concentrations of certain trace metals in seawater--some at concentration levels that can be found to occur in some marine ecosystems--have been shown to be toxic to estuarine and marine organisms in both laboratory culture and enclosed water-column experiments (Eisler 1973, Eisler and Wapner 1975, Grice and Menzel 1978~. A high priority should be given to research on analytical techniques, bioassay methodologies, and monitoring strategies for assessing the effects of both inorganic and organic pollutants in estuarine and coastal environments. Two recent workshop reports identify specific research needs (Goldberg 1919a,b). NATURAL RECORDS OF POLLUTION Records of the deposition of pollutants from the atmosphere in a given area may be found in the sedimentary record of lakes, reservoirs, marine deposits, and glaciers, and in annular deposits in trees and corals, if certain criteria are met. The processes that remove the pollutant from the atmosphere and the atmospheric residence time must have remained the same over the time interval of interest. The age assignments of the sedimentary strata must be accurate. And there must be no movement of the pollutant within the column following deposition. Anoxic sediments are especially attractive for assessing pollutant concentrations, especially if the sediments lie under anoxic waters. In such cases there is practically no disturbance by organisms to smear the record through movements of sediments. In many areas the solid nature of glacial sediments helps maintain the various strata as closed systems with respect to the migration of pollutants. The sedimentary record integrates the outputs of a variety of sources. For example, heavy metals may be introduced to the air by fossil fuel combustion, cement production, smelting, and many other activities. Although the relative contributions clearly are of great importance, knowledge of the overall atmospheric level and rate of increase is invaluable for regulatory activity. Sediments from southern Lake Michigan have been studied to determine the atmospheric burdens, sources, and rates of increase of several trace metals emanating from adjacent highly populated, highly industrialized and highly agricultural areas (Goldberg et al. 1981~. Sources were known to be primarily combustion processes involving coal, oil, and wood. Time horizons were determined for the Lake Michigan sedimentary column by the use of Pb-210 and Pu-239 + 240 chronologies. Primary energy sources in each time horizon were

OCR for page 119
133 identified by analyzing the characteristics of charcoals deposited in the strata (Griffin and Goldberg 1979~. This charcoal record shows that in the 19th century (1830 to 1900), the primary combustion processes involved natural and man-initiated burning of wood. This changed in the beginning of the 20th century, when coal became the important energy source. On the basis of the characteristics of associated charcoal deposits, oil is thought to have been responsible for pollutants in strata deposited after 1928. The charcoals indicated that in the period 1953 to 1978, about 76 percent of deposited charcoal had an origin in coal burning, 14 percent in oil burning, and 10 percent in wood burning. There is a steady increase of deposited carbon to about 1960 and then a distinct decrease to 1976 (Figure 7.21. Nine of the twelve metals assayed (Sn, Cr. Ni, Pb, Cu. Cd, Zn, Co, and Fe) had deposition profiles similar to that of the charcoal particles (Figure 7.3~. The recent decrease for these species is probably related to the improved retention of fly ash released from coal-and-oil burning facilities through the installation of control devices. Other studies have indicated that there were lower levels of atmospheric particulates in the atmosphere around Lake Michigan beginning in the late 1960s. Thus we can in principle derive a record of metals in the atmosphere above Lake Michigan for the time period spanning approximately the last century. With data on current atmospheric concentrations and present-day sedimentary concentrations, we can derive past atmospheric concentrations from sediment levels. Even though different quantities of metals may come from oil and coal burning, emissions from coke ovens, automobile exhausts, iron and steel plants, and cement production, their overall atmospheric concentrations are the important data for environmental regulators. By correlating changes in fossil remains with pollutant deposition records, insights into the ecological effects of pollutants can often be obtained. Because long-term records are obtainable, creeping changes can be separated from natural fluctuations and cycles. Changes in the relative abundance of species, extinction of sensitive forms, changes in preserved pigments, and changes in incidence of deformities have all been used as criteria for assessing effects. Paleolimnological techniques have been applied to the problems of eutrophication, acidification, and general environmental contamination (for example, Bradbury and Wadding ton 1973; Bradbury 1975; Gorham and Sanger 1976; Davis and Berge 1980; Davis et al. 1980; Norton and Hess 1980; Strand 1980; Warwick 1980a,b). The increased use of such techniques could partially overcome interpretive problems resulting from inadequate long-term baseline data. MONI TORI NG LARGE- SCALE POLLUT ION No area of the United States or indeed of our planet is free from the effects of air pollution. Acid precipitation is slowly eroding the environmental quality of remote wilderness areas in northeastern

OCR for page 119
134 o 20 - c) - I1J o Cal z 40 L`J 60 - 3$ $$ $333 _ D . ) - 1 1 1 1 0 0.10 0.20 % CARBON by WEIGHT >38 1970 1960 1950 1940 1930 1920 1910 1900 1890 1880 1870 1860 1850 1840 1830 FIGURE 7.2 Charcoal (elemental carbon) concentration as a function of depth in the Lake Michigan core. SOURCE: Goldberg et al. (1981). Reprinted with permission from Environmental Science and Technology. Copyright ~ 1981 by the American Chemical Society.

OCR for page 119
135 YEAR sso I ~ i ~ T j ' I 70 60 - _ 50 4O _ At/ 60 SO 40 3~0 20 10 0 DEPTH IN CORE (cm) YEAR ,~ ~ o 60 50 40 30 20 10 0 DEPTH IN CORE (cm) YEAR , 40 An _ 10 _ . , I YEAR 'sso soo 950 ' ' ' ' 1 ' ' I ' 1 ' ' ~20 ~ 1\ c l5 _ A) 1 60 50 40 30 20 10 0 DEPTH IN CORE (cm) _ 1500 _ ~ 1~ _ G _ 500 _ -~20 ~ ~10 _ 60 50 40 30 20 10 0 DEPTH IN CORE (cm) 40C Boo g _ 100 YEAR to soo /1 60 so 40 30 20 lo 0 DEPTH IN CORE (an) YEAR mo laso 1900 1950 1 loo , 100 cam ,,,,, 60 50 40 30 20 10 0 DEPTH IN CORE (cm) YEAR YEAR 950 18SO COO 1950 18SO 1900 19SO 100 60 | E ,,_ \ ~ ~ 1 ~60 50 40 30 20 10 0 DEPTH IN CORE (cm) , 0 x so so zo 10 0 DEPTH IN CORE (cm) llllll 60 50 40 30 20 10 C DEPTH IN CORE (em) YEAR 1 8S0 1900 , _ _ _ 1 ' ' , i I, I, . ~ ~ ~ ~ , . . . . . . YEAR 1850 1900 1950 YEAR 4 _ 950 ~50 ~900 ~950 0 ~ ~ ~ ~ I i ~ 1 6 0 50 40 30 20 10 0 DEPTH IN CORE tom) 100 _ G _ _ 50 _ _ 0 ~ 40 50 20 1 0 0 DEPTH IN CORE (cm) u O bO 40 30 20 10 0 DEPTH ~ CORE (cm) FIGURE 7.3 Metal concentrations (by dry weight) as a function of depth in the Lake Michigan core. SOURCE: Goldberg et al. (1981). Reprinted with permission from En- vironmental Science and Technology. Copyright 0 1981 by the American Chemical Society.

OCR for page 119
136 Canada, the United States, and Scandinavia; atmospheric deposition has also contributed chemicals of anthropogenic origin to the icecap of the Arctic. Sulfate, mercury, lead, and cadmium have polluted Greenland's glaciers in recent years (Boutron and Delmas 1980), and vanadium, a micro-constituent of fossil fuel, has been traced into remote areas of Alaska (Kerr 1979, Figure 7.4~. The development of techniques suitable for detecting slow degradation of the biosphere is hindered by a number of current scientific policies and attitudes. Granting agencies do not fund long-term studies; most grants are limited to 1, 2, or at most 5 years. But studies of the effects of low-level pollutants on ecosystems must be designed to go on for decades, so that slow, long-term trends can be separated from seasonal, annual, and multiyear fluctuations in both natural ecological phenomena and emissions of air pollution. For example, large seasonal fluctuations in the hydrogen ion concentration of Swedish rivers obscured the slow overall tendency for pH to decrease, until many years of data were combined (Figure 7.5). Because monitoring must now operate on the scale of decades rather than years, it is important that an appropriate coordination and funding organization be developed. Some funding for the initial phases of long-term studies has been made available recently by the National Science Foundation (Botkin 1977, 1978; Loucks 1979), but much remains to be done. The international system of biosphere reserves (Risser and Cornelison 1979) could provide suitable sites for baseline monitoring, and the proposed national network of experimental ecological reserves (TIE 1977) could provide sites for appropriate, long-term dose/response studies on the level of organisms, ecosystems, and whole watersheds. These ought to be combined with short-term laboratory experiments on both single-species and multi-species microcosms. Further development of theoretical models will also be needed, to test our predictive capabilities and to generate new insights into the behavior of species and ecosystems under pollution stress (West et al. 1980, Shugart et al. 1980~. A recent National Academy of Sciences report on ecotoxicology reviews the techniques available to test the effects of chemicals upon ecosystems (NRC 19811. Changes in scientific attitudes and educational systems will also be required. The Reemphasizing of taxonomic work in North American environmental science for the past few decades has left us with a dearth of first-rate taxonomists and few institutions that offer competent taxonomic training. Sound taxonomy is a key ingredient in any broad-scale, long-term analysis of an ecosystem, or in the paleoecological methods required to interpret past pollution records. Likewise, environmental monitoring has for the most part become the piecemeal application of techniques designed for other purposes. Assessment of environmental pollutants and their effects must be transformed into a science in its own right, dedicated to the development and application of methods for detecting long-term, chronic effects on organisms and ecosystems.

OCR for page 119
137 1 0.00 1.00 o cc z 111 z o O. 1 0 0.01 Barrow, Alaska 1 976-77 Excess Sulfate (micrograms per W l I l /\ / cubic meter) Excess Vanadium I\ (nanograms per cubic meter) \ i' Ny/ 1 1 1 1 1 1 1 1 1 1 1 1 O N D J F M A M J J A S FIGURE 7.4 Monthly mean concentrations at Barrow, Alaska, of atmospheric sulfate and vana- dium that could not be attributed to natural processes. Such excess vanadium is generally con- sidered to result from the burning of heavy indus- trial fuel oils. SOURCE: Kerr (1979). Reprinted with permission from Science 205: 290-293. Copy- right ~ 1979 by the American Association for the Advancement of Science. v'

OCR for page 119
138 8.0- 7.0 Ingvasta ~ ,~, , ~ \ .,, \~.\~ j.i\,~,,, :~/'II|~if~ ,$1\,,ff\,,, - ~ I- ~ ~ ~ ~ I 6.0- 85 i //\. 7.5 6.5- 8.5 7.5 6.5- i \. I \! I ai ,^. ./\ Al'' I, |_!\ '\l ~ I\ r ~ ~ ~ .~: me V! t 1 1 1 . 1 1 1 1 l - l ~ L L I l i\ - '` If.- !tl / \,-> I i4-l;|\l \\~'i\`f i:., Ail, 1 1 1 1 1 1963 1964 1965 1966 1967, 1968 1969 Kuggebro ~ 1 1~.~t 1 ~ 1 1 1 1 ! 'in I ~ I 1: I Ail . I ,, A\ I ~ I 1 1 1 FIGURE 7.5 Changes in pH values of a Scandinavian river. The data points, from three stations in the Savjann River, show the yearly periodic swings, and the arrow shows the average pH value. SOURCE: Oden and Ahl (1970).

OCR for page 119
139 SUMMARY The tremendous growth in the past few decades of our knowledge of atmospheric pollutants and their widespread dispersion has not been matched by an increase in our knowledge of their biological and ecological effects. Among the possible major effects envisaged are the serious perturbation of global sulfur and nitrogen cycles, the global dispersion of toxic trace metals, and the biotic impoverishment affecting the structure and function of major ecosystems. The study of atmospheric pollutants is very complex, because the substances must be followed from their an increase in our Knowledge ot tnelr the climatic Greenhouse effect, sources ot emission through atmospheric and ecosystem which major transformations can occur--to biotic and abiotic nathwaYs--a lone receptors, which fluctuate naturally in abundance and distribution, and which are affected in extremely diverse ways. Linkages of pollutants are important for predictive purposes, and may occur at all points from the sources of initial emissions to the ultimate receptors where their effects are manifested. Maximum pollution effect is expected where receptors are highly sensitive and ecologically important, and where sources and receptors are close together. (If the pollutant has a long atmospheric residence time such closeness is unnecessary.) For the prediction of effect it is important to know the distribution and abundance of sensitive receptors, the distribution and size of emission sources, and the properties of the pollutant that govern residence times in various ecosystem compartments. Sensitivity of organisms and ecosystems to pollutant stress may be caused by a wide variety of receptor properties, and may vary with the nature of the pollutant and the stage of development of the organism or ecosystem. Few generalizations are possible in this regard. Stratigraphic studies of pollutant accumulation and changes in fossils in sediments or long-term deposits of several types can provide a useful guide to the history of a pollution problem. Atmospheric pollution is now ubiquitous, but policies for investigating and dealing with it are not well developed. Evaluation of the long-term ecological effects of chronic environmental are vital, but appropriate organizations to accomplish such evaluations are lacking, as are suitably trained personnel. There an urgent need for new institutional arrangements and for training programs to establish ecotoxicology--the ecological assessment of environmental pollutants--as a new and independent scientific discipline. pollution