Environmental Impacts of Natural Disasters
It is recognized that many significant nonmarket effects result from natural disasters, including environmental impacts. Though our committee had a keen interest in these topics, it became clear that these impacts—though often significant—did not fit easily with this study's main report and conclusions for the following reasons: (1) not all disasters result in significant ecosystem impacts (e.g., many earthquakes have but minor impacts on ecosystems); (2) some extreme events have positive impacts on ecosystems (e.g., floods can help rejuvenate floodplain vegetation and are important drivers of many ecological processes in floodplains); and, (3) these impacts are mainly nonmarket and are exceptionally difficult to quantify and/or monetize. Though there are emerging efforts in quantifying and monetizing ecosystem services (e.g., Costanza, 1997), they are in their infancy and are not yet widely accepted.
Nonetheless, the magnitude of the environmental impacts of many disasters compelled the committee to discuss them and we do so in this Appendix. Though no specific recommendations regarding how environmental costs should be incorporated in loss estimates we provide here, we encourage policymakers in the relevant executive branch agencies to devote more attention and perhaps research to these issues. It is important in assessing environmental impacts to distinguish between impacts of disasters on the natural environment from those on the human-made landscape environment. As mentioned, events that societies label as natural ''disasters" may also have beneficial ecological consequences. However, these benefits tend to only manifest themselves months or years after an extreme event (e.g., rejuvenation of a coniferous forest months and/or years after fires), or are often not readily apparent (e.g., recharging of groundwater stocks after a flood). These benefits to ecological systems are of course typically overshadowed by immediate, negative impacts on societies and structures; hence, the use of the term natural "disasters."1
The committee has recommended that the direct costs of damages to landscapes be included in loss estimates. See Table 2-2.
Three principles apply to the assessment of the costs and benefits of extreme geophysical events to the nation's ecological systems. First, although the more tangible, quantifiable damages of extreme events to infrastructure and economies may be difficult to calculate precisely, the costs to and benefits for natural ecosystems—even from such apparently straightforward impacts as numbers of fish killed or trees destroyed—are even less tangible and may be nearly impossible to quantify precisely. Moreover, even if the physical effects can be measured, the monetary values of those impacts cannot be stated with precision.
Second, existing ecological systems have already adapted in many respects to the forces created by extreme events, such as floods or droughts. This process is lengthy, extending over thousands of years and involving the evolution of species and complex physical systems. The effects of geophysical extremes often are not undesirable. For example, major natural disturbances, such as fires or floods, rejuvenate old forests. The critical factors are the frequency, intensity, and extent of natural disturbances. If disturbances occur too frequently and over large areas, then only pioneering, short-lived, and opportunistic species survive. If disturbances occur too infrequently, then slower-growing, superior competitors for light, water, and nutrients replace the pioneers. Maximum diversity is maintained by an intermediate level of disturbance, so that patches of pioneers and superior competitors alike occur within the landscape. All of this suggests that attempts to eliminate natural disturbances (rather than attempts to mitigate their adverse impacts) can be counterproductive and in some cases, as in the 1927 and 1993 floods on the Mississippi River and the Yellowstone fires in 1988, can make a disaster worse.2
Third, precisely because many disasters are indeed "natural," they often produce mixed outcomes for the environment: benefits to some parts of the natural system and losses to others. For example, some thinning of tree branches caused by high winds or ice accumulations from winter storms can allow for subsequent stronger tree development, and studies of the 1993 flood in the Midwest revealed major ecological benefits in the immersed floodplains. To the average human observer, floodplain forests appear to change scarcely at all from year to year, and therefore the death of trees during or after a major flood seems
catastrophic. However, the diversity of vegetation on the floodplain is a product of disturbances, such as major fires, droughts, and floods that occur very infrequently in terms of a human life span. Without droughts, floodplains would not get dry enough to burn, and fire-intolerant species would crowd out the wet prairies and trees. Even the most extreme geophysical events are thus not necessarily damaging to ecosystems, and in some circumstances can bring great benefits. Furthermore, the effects take months and years after the disturbance to assess, suggesting that immediate ecological or environmental accountings are prone to error.
Finally, as was outlined in the report, it is useful to assess the impacts of natural disasters by type of disasters. Because of their great spatial extent and longevity, major floods and droughts generally create the greatest environmental impacts, whereas earthquakes, hurricanes, thunderstorms, and winter storms cover less territory and their effects on the ecosystem are less pervasive and long-lasting. Below, we briefly review some case studies to illustrate the diverse environmental impacts of different categories of disasters, and the difficulties in precisely quantifying and monetizing these impacts.
Major floods create myriad effects on river-floodplain ecosystems. During periods of low flow, typically in midsummer, the rivers occupy channels. During rainy seasons, rivers spill into their floodplains, recharging the surrounding wetlands, forests, and lakes with fresh supplies of water, nutrients, and sediments. During great floods, floodplains do not merely store water, but become part of the flowing river itself, conveying water slowly downstream through the forests and marshes. Plant and animal species have adapted over time to exploit, tolerate, or escape seasonal floodpulses and exceptional great floods. The combination of the flood-adapted animals and plants, the seasonal flows and great floods, the river and its channels, and the complex patchwork of floodplain habitats constitute the dynamic and phenomenally productive river-floodplain ecosystem.
Large river-floodplain ecosystems provide valuable hydrological and ecological services and functions, such as flood storage and conveyance; the maintenance of biodiversity; retention, recycling, and conversion of potentially polluting nutrients into useful biomass; production of fish, wildlife, and forests; and the provisions of corridors for migratory fish and wildlife. Annual floodpulses help regulate and maintain these ecosystems by promoting exchanges of water, sediment, nutrients, and organisms between the rivers and their floodplains. Moreover, infrequent great floods and droughts help maintain habitat and species diversity (Sparks, 1996).
Flood of 1993. Though the record flood of 1993 in the Upper Midwest was an economic disaster, it was a boon to many plants and animals that lived in and along the Missouri and Upper Mississippi Rivers. Even the few species that appear to have been harmed by the flood, such as some trees, may benefit in the long term. Any harm that did occur may have been more the result of human factors rather than the flood itself, including failure of human-made levees, excessive loading of rivers and the Gulf of Mexico with herbicides and agricultural fertilizers, widespread dispersal of introduced pests, and the excessive drawdown of the Mississippi River after the flood.
It is not surprising that the flood of 1993 had both positive and negative effects on the river-floodplain ecosystems. Many mobile organisms have adapted to exploit such seasonal floods. For example, the flood benefited fish that spawned on the inundated floodplain, and wading birds in turn exploited the huge crop of young fish. In contrast, long-lived, stationary organisms, such as trees, were severely stressed or died as a result of the exceptionally long period of inundation. And yet the outcome for trees was not all bad. Many seedlings cannot germinate or grow in the shade of mature trees, so old forests were rejuvenated when mature trees died because of the 1993 flood.
Every component of the river-floodplain ecosystem, from the bottom to the top of the food chain, responded to the exceptional flood of 1993. At the shallow margins of the flood, nutrients were apparently released from newly flooded soils, stimulating phytoplankton. Aquatic insects likewise concentrated in the shallow water, perhaps consuming either the plankton or the remains of flooded terrestrial vegetation. Submergent aquatic plants grew in areas where the flood did not persist too long so they could reach sunlight. Where the flood rose higher and lasted longer, submersed aquatic plants virtually disappeared. About 52 species of fishes, representing 15 families, spawned on the floodplain during the flood (Maher, 1995). The abundant juvenile fish became food for larger fish and fish-eating birds, such as herons and egrets. The flood also took a heavy toll on trees, the longest-living organisms in the floodplain.
The 1993 flood caused a serious economic and environmental pest, the zebra mussel, to wash from the Upper Illinois River downstream into the lower Illinois and the Mississippi. In the process, zebra mussel larvae were carried far back into the floodplain and upstream into tributaries that were backed up by the mainstream rivers. Another potential pest was introduced when a fish farm on a tributary of the Mississippi flooded and Asian black carp escaped. The carp is able to consume endangered native mussels and clams and competes with the native fish and ducks that already consume zebra mussels.
The flood moved tremendous amounts of water to the Gulf of Mexico. Through erosion and flooding of agricultural soils throughout the Midwest, the floodwaters picked up vast quantities of various chemicals, including some from flooded industries along the rivers. Substantial quantities of these agricultural (and other) chemicals were transported into the streams and rivers, either as
dissolved matter or in suspension, and into the floodplains. This polluted water infiltrated floodplains and contaminated ground water aquifers.
There was an immense discharge of freshwater to the Gulf of Mexico during the summer of 1993. The delivery of this water and its dissolved and suspended materials affected the ecosystem of the Gulf of Mexico. Discharges of herbicides and nitrates to the Gulf of Mexico were substantially higher in 1993 than in prior years, stimulating plankton blooms. When the plankton died and sank, the decaying organic matter used up oxygen in the bottom layer of water, lowering oxygen levels over an area of 6,000 square miles (the so-called "dead zone") and threatening valuable fisheries. The total amount of atrazine delivered to the Gulf of Mexico by the Mississippi River from April to August 1993 was 1.2 million pounds, up about 25 percent from loads delivered during 1992. One million tons of nitrate-nitrogen were discharged to the Gulf of Mexico from April to August 1993, a value 37 percent larger than loads for 1992 (Goolsby et al., 1993).
In summary, the flood of 1993 exacerbated two preexisting environmental problems related to human activity. First, it facilitated the spread of a serious economic and environmental pest, the European zebra mussel, that accidentally had previously been introduced to the St. Lawrence-Great Lakes drainage by transoceanic ships (and facilitated other introduced pests, such as the Asian tiger mosquito). Second, nutrient loading of the Gulf of Mexico was substantially increased by the flood, and the summer "dead zone" in the Gulf consequently expanded, with potential detrimental impacts on the largest fishery in the United States. At the same time, the 1993 flood also vividly demonstrates the complexity and uncertainty over the range of positive and negative impacts upon floodplain ecosystems, as well as the overwhelming task of trying to distill precise figures for the full costs and benefits of an extreme geophysical event.
Unlike floods, droughts generally damage ecological systems and yield few offsetting benefits. In fact, the most subtle and enduring impacts of droughts occur in the environment. The cumulative stress on wetlands, wildlife, forests, ground water, and soils cannot be measured accurately, and many effects occur slowly and over a period of years, making them extremely difficult to quantify.
The problems generated by droughts begin with changes in the quantity and quality of water available in the hydrologic system. Drought damages both plant and animal species by depriving them of food and water, increasing their susceptibility to disease, and increasing their vulnerability to predation. As with floods, droughts produce a loss of biodiversity, and often increase erosion of dried soils when rain eventually comes. Droughts also degrade water quality, shifting salt concentration, pH levels and dissolved oxygen, while increasing
water temperatures. Even air quality is diminished because of increased dust and pollutants. Droughts also lead to more wildfires, while adversely changing salinity levels in coastal estuaries and reducing the flushing of pollutants.
Drought of 1988. The best documentation of environmental damages from a drought came from studies of the 1988 drought, which affected large portions of the United States. This event caused enormous reductions in streamflows in two major drought-affected regions. Plans to divert additional water from the Great Lakes to enhance the record low flows of the Upper Mississippi River system were halted by environmental concerns over the potential impacts of lowered water levels on the lakes (Changnon, 1989). Many streams were unable to handle industrial discharges and agricultural pollution, greatly limiting water quality and the use of water. Water supplies dropped to seriously low levels in the southeast United States, where many uses of river waters, including hydropower generation and navigation on major rivers, had to be curtailed. Saltwater intrusion up the Mississippi River beyond New Orleans was so severe that underwater sills were built to halt the intrusion.
The 1988 drought led to 68,000 wildfires that burned 5.1 million acres of federal forest land. Fire-fighting costs alone amounted to $300 million. The best-known fires were those in Yellowstone National Park, which captured national attention. The dry conditions in areas adjacent to the fires greatly reduced the number of tree seedlings, with mortality of 40 percent of the trees planted in the 10 years prior to 1988, including 150 million pine seedlings. The drought led to increased insect attacks on commercial forests, and 5.7 billion board feet of lumber were lost because of pine bark beetles. The total loss to U.S. forests was estimated at $5 billion (Riebsame et al., 1991).
The 1988 drought also caused sizable but unmeasured losses of fish, waterfowl, and wildlife. High water temperatures in bays along the East Coast caused an increase in oyster diseases, resulting in an 1988 harvest of 375 million bushels, the lowest on record for Chesapeake Bay (Avery, 1988; Changnon et al., 1996).
Finally, the high temperatures associated with the 1988 drought had profound effects on human health. Several thousands of deaths were attributed directly or indirectly related to the high temperatures. Many of these deaths occurred in the large urban areas of the central and eastern United States. Municipal governments responded by establishing cooling centers. Not surprisingly, a comprehensive study of the environmental impacts of the 1988 drought concluded that there were "no winners" in the ecosystems (Riebsame et al., 1991).
Hurricanes and Tropical Storms
Hurricanes and tropical storms create environmental damages within paths that vary from 50 to 150 miles in width. The environmental consequences largely consist of damages to trees and underbrush in the storm path. At the same time, the long-term ecosystem damages of these storms are uncertain. To be sure, during coastal storms in particular there is often significant erosion of shores and beaches. In the long run, however, nature generally has adapted to these events, so the extent of negative impacts of these events is not clear.
Severe Local Storms
Severe local thunderstorms—such as a major tornado striking Wichita or a thunderstorm producing large hailstorms in Dallas—are often labeled as natural disasters due to the attendant looses of life and economic losses, but in general these events are localized. They are not events that create serious, large-scale damages to the natural ecosystem. Nonetheless, it is possible that the cumulative environmental impacts of severe storms over a period as short as a year can be significant. Broad areas can suffer from numerous forest fires triggered by cloud-to-ground lightning. High winds and hail cause localized damages to plants and forests, although the total losses are considered to be relatively minor on a regional or national scale.
Heavy rains that lead to flash floods also can be environmentally damaging, at least locally. They increase soil erosion rates, and if they occur in mountainous areas the resulting flood can create massive damages to ecosystems in narrow mountain valleys.
Although the dominant losses from earthquakes are to structures and potentially to humans, these events can also result in adverse environmental consequences. Examples include flora and fauna damaged by the shocks and shifts in land surfaces, as well as alterations in local hydrologic systems. For example, the famed New Madrid earthquake in the central United States in the 19th century changed the course of the Mississippi River and created a cutoff lake. In the most affected areas, trees, shrubs, land cover, and habitats can also be destroyed. There are currently no estimates of the environmental or ecosystem losses from earthquakes (although the national, long-term impact is probably not great).
Strong and persistent synoptic scale (nonstorm) high winds can sweep over large areas and cause damage to trees and plants. High winds can also help promote large-scale fires, typically in dry western areas. Recent wind-driven fire catastrophes in California accounted for insured property losses of $1.5 billion in October 1992, rated as the third largest fire loss in the nation's history (III, 1993). Major brushfires enhanced by strong winds occurred in California in October 1993 and again in November 1993, together causing $815 million in insured property losses (III, 1995). These huge, wind-driven fires consumed all underbrush, ground cover, and trees over hundreds of square miles, but there is no known report documenting the value of these losses to the natural or landscape environments.
High winds and waves caused by severe extratropical cyclones damage beaches and shoreline ecosystems. This is a problem mainly along the East Coast when strong "Nor'easters" strike along shores ranging from 500 to 1,000 miles in length and in the Great Lakes, where winter storms create waves that severely erode beaches. However, these shoreline effects also can be viewed as an inherent part of nature to which coastal ecosystems have adapted.
Although natural disasters are by definition undesirable for humans, they often carry several ecological benefits. Floods are a prime example of the mixed economic and environmental effects. At the other extreme, droughts not only produce economic damage, but virtually all of their environmental impacts are also undesirable.
There are only limited quantitative data of the environmental costs of natural disasters. Relatively little effort by the private sector, academics and scientists, or the government has gone into this activity. Nonetheless, such studies as have been conducted reveal that numerous environmental problems caused by natural disasters often have significant consequences for ecosystems, and eventually people, societies, and economies. Thus, even though these environmental impacts may not readily translate into monetized losses (or gains) their importance strongly suggests they should be considered by governments, academia, and the private sector in the study and design of hazard mitigation and land use policies.
Avery, M. 1988. Heat Death Toll in 1988. Washington, D.C.: Center for Environmental Physiology.
Barry, J.M. 1997. Rising Tide. New York: Simon and Schuster.
Bhowmik, N. 1996. Physical effects: A changed landscape. Pp. 101–131 in The Great Flood of 1993. S. Changnon, ed. Boulder, Colo.: Westview Press.
Changnon, S.A. 1996. Lessons from the flood. Pp. 300–319 in The Great Flood of 1993. S. Changnon, ed. Boulder, Colo.: Westview Press.
Changnon, S.A. 1989. The 1988 drought, barges and diversion. Bulletin American Meteorological Society 70:1092–1104.
Changnon, S.A. 1978. How a severe winter impacts individuals. Bulletin American Meteorological Society 60:110–114.
Changnon, S.A. and D. Changnon, 1978. Winter storms and the record-breaking winter in Illinois. Weatherwise 31:218–225.
Changnon, S.A. and W.E. Easterling. 1989. Measuring drought impacts: The Illinois case. Water Resource Bulletin 25:27–42.
Changnon, S. A., K.E. Kunkel, and B. Reinke. 1996. Impacts and responses to the 1995 heat wave: a call to action. Bulletin American Meteorological Society 77:1497–1506.
Goolsby, D. A., W.P. Battaglin, and E. M. Thurman. 1993. Occurrence and Transport of Agricultural Chemicals in the Mississippi River Basin, July Through August 1993. Circular 1120–C, U.S. Geological Survey, Denver Colo.
Insurance Information Institute (III). 1995. The Fact Book 1995: Property/Casualty Insurance Facts. New York: III.
Insurance Information Institute (III). 1993. The Fact Book 1993: Property/Casualty Insurance Facts. New York: III.
Maher, R.J. 1995. Observations of fish community structure and reproductive success in flooded terrestrial habitats during an extreme flood event. Long Term Resource Monitoring Program. TRMP Technical Report 95-T, National Biological Survey, Onalaska, Wisc.
Pielke, R.A., Jr. 1995. Hurricane Andrew in South Florida: Mesoscale Weather and Societal Responses. Boulder, Colo.: National Center for Atmospherics Research.
Pilkey, O.H. and K.L. Dixon. 1996. The Corps and the Shore. Washington, D.C.: Island Press.
Riebsame, W.E., S.A. Changnon, and T.R. Karl. 1991. Drought and Natural Resources Management in the United States: Impacts and Implications of the 1987–89 Drought. Boulder, Colo.: Westview Press.
Sparks, R.E. 1996. Ecosystem effects: positive and negative outcomes. Pp. 132–162 in The Great Flood of 1993. S. Changnon, ed. Boulder, Colo.:Westview Press.