Ecological Forecasts: An Emerging Imperative James S. Clark, Steven R. Carpenter, Mary Barber, Scott Collins, Andy Dobson, Jonathan A. Foley, David M. Lodge, Mercedes Pascual, Roger Pielke Jr., William Pfizer, Cathy Pringle, Walter V. Reid, Kenneth A. Rose, Osvaldo Sala, William H. Schlesinger, Diana H. Wall, David Wear

Freshwater ecosystems

Humans have appropriated half of the accessible global freshwater runoff, and this could climb to 70% by the year 2025 (1). Nearly 2/3 of all rivers are regulated in some manner (2), causing fragmentation, deterioration, and losses of floodplains, wetlands, and riparian ecosystems (3). Irrigation has dramatically reduced water levels in major closed basins (e.g., Aral Sea, Lake Chad) and the discharge of river systems (e.g., the Colorado, Nile, Ganges, Amu Darya, and Syr Darya). In recent decades, more than 20% of the known 10,000 freshwater fish species have become threatened, endangered, or extinct (4). Forty percent of the human population occupies river basins that experience water scarcity; by 2050, this number will increase to 50%. With the exception of agencies in developed countries responsible for forecasting floods that might threaten life or property, there is no mechanism to warn of changes in freshwater ecosystems and how to respond to them. Providing ecological forecasts of the consequences of hydrologic change, pollution, and effects of exotic species on freshwater ecosystems represents a broad scientific challenge (3, 5-8).

Food Supply

By 2050, a burgeoning human population will depend on agricultural science to prevent widespread shortfalls in food supply. Past reliance on intensive land use, high-yielding crops, industrially produced fertilizers and pesticides, irrigation, and mechanization comes with environmental costs that can include damage to soil structure and contamination of water and food products (9, 10). Devastating consequences can result from failure to consider agricultural practice in the context of regional ecosystem function. During Hurricane Floyd, waste from factory-sized hog farms enriched streams and affected human health and the function of coastal estuaries.

Environmental and financial concerns are beginning to motivate less intensive agricultural systems (11-13) that require an understanding of interactions involving agriculture and basic ecosystem services, including supply of clean air and water, maintenance of soil fertility, nitrate leaching, and pollination. Forecasting the impacts of agriculture under developing agromanagement systems requires a broad examination of tillage practices, fertilizer use, crop rotations, and irrigation strategies in the context of local and regional ecosystem function. At present there is no integrated strategy for anticipating these interactions and their consequences for global food supply.

The Carbon Cycle

The continuing rise of CO₂ in Earth's atmosphere, and its potential to cause significant climate changes, demands two levels of ecological forecasting. The first level concerns the effects of higher CO₂ and temperature on plant growth, water use, and pest resistance, and how these responses will differ among species. Forecasts would have immediate relevance for farmers and foresters, who stand to lose economically if the wrong crops or trees are planted. Differential growth and competitive ability of species in response to rising CO₂ will determine how diversity, structure, and function may respond to changing CO₂ and climate. Public health officials could benefit from ecological forecasts of flowering phenology, pollen production, and severity of pollen allergens in the environment.

A second level of ecological forecast stems from the impact of plants on the rise of atmospheric CO₂, mediated by biosphere storage (or loss) of carbon, a key ecosystem service. Knowledge of biomass, net primary production, and soil carbon storage can be used to anticipate future ecosystem function and diversity in the face of climate change. These predictions have economic significance: for instance, the Kyoto protocol includes provisions for emissions trading. A country that sees net carbon storage in its vegetation and soils could sell that quantity as a "credit" to a country that is unable to curb its fossil fuel emissions. Accurate assessments of the carbon sequestration will be valuable to the emerging business of "emissions trading." The interactions involving CO₂, climate, nutrients, and plant physiology are not yet sufficiently understood to permit informative forecasts of carbon storage.

Living resources

Extinctions, invasions, and habitat loss impact ecosystem function (e.g., nutrient cycling, fire, primary productivity) and the capacity of ecosystems to supply critical goods and services. Diversity and habitat loss affect ecosystem variability and resilience to perturbations. Some of these effects can be immediate, whereas others are not apparent for decades. For example, salmon extinction has attenuated nutrient supply to Pacific Northwest rivers, with consequent, slow change in community composition and structure (7). Introduced cheatgrass in the western United States has altered composition, nutrient cycles, and disturbance regimes (14). The value of biodiversity extends beyond these rather direct goods and services. Nations around the world have invested in parks and protected areas, not only because there is economic value in tourism, but also because the public values biodiversity.

There is broad demand for biodiversity forecasts. Conservation biologists require predictions of extinction risk that are more accurate than simple species-area curves applied to habitat loss (15-21). For example, habitat loss does not lead to complete diversity loss outside of the remaining habitat island; some species persist, even flourish, in converted lands and at edges. Spatial aspects of extinction risk have conservation relevance (22, 23). There is need to anticipate spread and impact of nonindigenous species (NIS) on ecosystem function, food supplies, commerce, and recreation (24, 25). Introductions can be irreversible, and mitigation is difficult and expensive. Biodiversity prediction could have immediate impact on policy related to food supply, freshwater, and human health, it would publicize the biodiversity crisis of mass extinction, and it could inform preventative or mitigative actions against introduction and spread of NIS.

Disease

The recent outbreak of foot and mouth disease in the United Kingdom emphasizes the importance and potential of ecological forecasting. The disease appeared on a farm in northern England in February 2001. Within 2 weeks it was reported from at least 10 other locations in England. Over 3 million livestock have been slaughtered at a cost of $5.2 billion to Britain's farmers (26).

Scenarios for the course of the foot and mouth epidemic (27) proved remarkably accurate, but interactions involving ecological and socioeconomic factors typically make disease forecasting difficult (28, 29). Cholera dynamics depend on climate variability (30-32) and socioeconomic interactions (33). Measles epidemiology is most accurately predicted at the national level and within large cities. It is harder to predict at intermediate scales, where movements of infectious individuals depend on connectance of population centers.

Despite the complexity, forecasts and model scenarios have already provided invaluable guidance for prevention measures, the design of vaccination programs, and drug-use strategies. In the case of foot and mouth disease, scenarios for several potential interventions were the basis for the decision to escalate slaughter of infected herds. In general, preventive measures are critical when vaccines are not yet an effective option, especially in light of growing resistance to drugs and the breakdown of public health in large regions of the globe. Drug resistance is unlikely to be solved by the development of new drugs. Epidemiological models should continue to play a role with the evolution of new and resistant strains (34).

References

1. S. L. Postel, G. C. Daily, P. R. Ehrlich. Science 271, 785 (1996).
2. J. N. Abramovitz, Imperiled waters, Impoverished future: The decline of freshwater ecosystems. (Worldwatch Institute, Washington, D.C., 1996.)
3. C. M. Pringle, in press.
4. World Resources Institute (WRI), Pilot analysis of global ecosystems (PAGE): Freshwater Systems. Special Report of the World Resources Institute (2000). Available at: www.wri.org/wri/wr2000.
5. S. L. Postel, S. R. Carpenter, in Nature's services, G. Daily, Ed. (Island Press, Washington, D.C., 1997), pp. 195-214.
6. S. L. Postel, Pillar of Sand: Can the irrigation miracle last? (W. W. Norton and Company, New York, 1999).
7. R. J. Naiman, R.E. Bilby, P.A. Bisson. BioScience 50, 996 (2000).
8. National Academy of Sciences, Grand challenges in Environmental Sciences: Special report of the National Academy of Sciences (2000). Available at: www.nap.edu/openbook/0309072549/html.
9. H, Wagstaff, Agric. Ecosys. Env. 19, 1 (1987).
10. D. Tilman et al., Science 292, 281 (2000).
11. T. Dalgaard, N. Halberg, I. S. Kristensen. Nutr. Cycl. Agric. 52, 277 (1998).
12. N. Halberg, I. S. Kristensen. Biol. Agric. Hort. 14, 25 (1997).
13. DARCOF, Biannual report. 1996-1998. (Danish Research Centre for Organic Farming, Telje, Denmark 1999).
14. R. N. Mack, D. Simberloff, W. M. Lonsdale, H. Evans, M. Clout, F. A. Bazzaz, Ecol. Appl. 10, 689 (2000).
15. R. H. MacArthur, E. O. Wilson, The Theory of Island Biogeography. (Princeton Univ. Press, Princeton, NJ, 1967).
16. W. V. Reid, in Tropical Deforestation and Species Extinction, T. C. Whitmore, J. A. Sayer, Eds. (Chapman & Hall/IUCN, London, 1992), p. 55.
17. T. M. Brooks, S. L. Pimm, N. J. Collar, Conservation Biol. 11, 382 (1997).
18. S. L. Pimm, G. J. Russell, J. L. Gittleman, T. M. Brooks, Science 269, 347 (1995).
19. J. Harte, A. Kinzig, Oikos 80, 417 (1997).
20. J. Harte, A. Kinzig, J. Green, Science 284, 334 (1999).
21. E. Seabloom, A. P. Dobson, D. M. Stoms, in preparation.
22. D. F. Doak, P. C. Marino, P. M. Kareiva, Theor. Popul. Biol. 41, 315 (1992).
23. D. F. Doak, Conservation Biol. 9, 1370 (1995).
24. D. M. Lodge, Trends Ecol. Evol. 8, 133 (1993).
25. O. E. Sala et al., Science 287, 1770 (2000).
26. Economist, 5 May 2001, p 49.
27. R. M. Anderson, N. Ferguson, K. Donelly, Science, in press.
28. K. A. Schmidt, R. S. Ostfeld. Ecology 82, 609 (2001).
29. A. P. Dobson, J. Foufopoulos, Philos. Trans. Roy. Soc. London, in press.
30. A. A. Franco et al., Am. J. Epidemiol. 146, 1067 (1997).
31. R. Colwell, Science 274, 2025 (1996).
32. M. Pascual, X. Rodo, S. P. Ellner, R. Colwell, M. J. Bouma. Science 289, 1766 (2000).
33. E. Bonilla-Castro, P. Rodriguez Sehk, G. Carrasquilla, La Enfermedad de la Pobreza. El Colera en los Tiempos Modernos. Ediciones Uniandes Carrera.Santafe de Bogota (2000).
34. B. R. Levin, M. Lipsitch, S. Bonhoeffer, Science 283R. 806 (1999).