Sensitivities, Impacts, and Adaptations
In Chapter 32 we wrote that the impact of a climate change on some activity is the integral during the change of the sensitivity times the rate of change of the climate. The hope, of course, is that adaptation can modify the sensitivity, ameliorating bad and increasing good impacts of a given climate change. In the sections that follow, the sensitivities, impacts, and adaptations of activities are examined. Because this is a U.S. report, much of the examination is of U.S. activities. The scenarios of change are generally within the ranges stated in our Assumptions, and they are given precisely in the cited publications.
Estimating the cost of impacts or adaptations is fraught with uncertainties. Uncertainties range from those about climate scenarios to ones about sensitivities and future technology. We do not know whether people will choose to adapt more or suffer more from harmful climate changes and benefit less from helpful climate changes. So, national let alone planetary estimates are difficult and may be misleading. Nevertheless, the scale or order of things must be judged. Accordingly, Table 34.1 gives some illustrative costs of impacts and adaptations.
The footnotes show that the cost estimates are drawn from diverse sources. Their accuracy ranges from the precision of the budget of the U.S. Weather Service to the imprecise multiplication of an assumed cost of a house by the number of houses that newspapers report that a storm destroyed. Few of the estimates, if anym include, for example, personal suffering, the advantages of a renewed home, or a construction boom after a flood. The accuracy of each cost can be judged from the cited sources.
These costs illustrate those of adapting and those that might be suffered more or less frequently if climate changed. For example, if hurricanes became more frequent and no one adapted, costs like the $5 billion for
TABLE 34.1 Illustrative Costs of Impacts and Adaptations in Current Dollars. An impact may help, as when a warmer climate reduces snow removal, or harm, as when a drier climate makes droughts more frequent. Adaptations may temper the harm or exploit the benefit of a new climate, as when a new and adapted wheat variety is created or forest planted. Some entries, like the U.S. gross national product (GNP) or the changing GNP per capita in the world, give a scale for judging the costs of impacts and adaptations. The numbers included for scale are in italics.
(continued on page 543)
(Table 34.1 continued from page 542)
(continued on page 544)
(Table 34.1 continued from page 543)
(continued on page 545)
(Table 34.1 continued from page 544)
(continued on page 546)
(Table 34.1 continued from page 545)
(continued on page 547)
(Table 34.1 continued from page 546)
eRiebsame et al. (1986).
fRiebsame et al. (1991).
gNational Hazards Research and Applications Information Center (NHRAIC), University of Colorado, Boulder. NHRAIC maintains an unreferenced data base on national hazards. Numbers referenced as NHRAIC are from their data base.
hThese tornados also caused 85 deaths. NHRAIC data base.
iThese floods also caused 47 deaths. NHRAIC data base.
jHurricane Hugo also caused 20 deaths. NHRAIC data base.
kDollar figures for average annual U.S. losses are estimates of total losses, including both private losses and government expenditures.
lRiebsame et al. (1986).
mPersonal communication from Office of Hydrology, National Weather Service, Silver Spring, Maryland, to W. Riebsame, NHRAIC, Boulder, Colorado, 1990.
nKessler and White (1983).
pRiebsame et al. (1986).
qThe actual expenditure in 1988 for the U.S. National Weather Service was $322,913,000. U.S. Office of Management and Budget (1989, p. I-F14).
sU.S. Department of Agriculture (1989b).
tU.S. Department of Agriculture (1989b).
vU.S. Bureau of the Census (1987, Table 670).
wForestry numbers are from Straka et al. (1989) unless otherwise noted.
xThe 1983 expenditures on about a half billion acres of State and private forest land was $0.50 per acre. The difference between this $0.50 and $1.36 times 736 million acres of total forest land is about a half billion dollars. U.S. Department of Agriculture (1986, Tables 661, 667, and 668).
yU.S. Bureau of the Census (1987, Table 670). Agriculture, etc., less farming.
zNational Plant Germplasm System, ARS, USDA operating costs only for regeneration, storage, and distribution. Personal communication from S. Eberhart, National Seed Storage Laboratory, Fort Collins, Colorado, to P. Waggoner, Connecticut Agricultural Experiment Station, May 13, 1991.
(continued on page 548)
(Table 34.1 continued from page 547)
aa$500 per year is the amount of the subsidy from the Center for Plant Conservation to member gardens for maintaining a sample. Personal communication from V. Heywood, Center for Plant Conservation, to P. Waggoner, Connecticut Agricultural Experiment Station, New Have, Connecticut, July 4, 1990.
bbRange is $50–$500 per acre for land far from cities; $300–$5,000 per acre for land near cities. Personal communication from J. Ball, Woodland Park Zoo, Seattle, Washington, to G. Orians, University of Washington, Seattle, Washington, April 1990.
ccCosts for food and labor only. Personal communication from J. Ball, Woodland Park Zoo, Seattle, Washington, to G. Orians, University of Washington, Seattle, Washington, April 1990.
ddCosts for food and labor only. Personal communication from J. Ball, Woodland Park Zoo, Seattle, Washington, to G. Orians, University of Washington, Seattle, Washington, April 1990.
eePersonal communication from J. Ball, Woodland Park Zoo, Seattle, Washington, to G. Orians, University of Washington, Seattle, Washington, April 1990.
ggU.S. Bureau of the Census (1987, Table 380).
hhU.S. Bureau of the Census (1988, Table 371).
iiCost for raw water from modifications to F. E. Walter Reservoir. Personal communication from R. Tratoriano, Delaware River Basin Commission, to D. Sheer, Water Resources Management, Columbia, Maryland, 1990.
jjNew Bureau of Reclamation, Central Valley Project. Cost for raw water at the plant. Does not include costs for delivery facilities to point of use. These figures are for construction costs of Auburn Dam allocated to water supply only23% of total construction costs. Other costs allocated to flood control, instream flow, hydropower, and recreation. Personal communication from J. Denny, U.S. Bureau of Reclamation, Sacramento, to D. Sheer, Water Resources Management, Columbia, Maryland, 1990.
kkIncludes cost of treatment and delivery facilities. R. Alpern, New York City Department of Environmental Conservation, First Intergovernmental Task Force Report.
llCosts for desalting run from $2,000–$5,000/acrefoot/yr capital costs, plus operating costs of $2,000–$4,000/acrefoot (mainly energy costs). This equates very approximately to $2,200–$5,400/acrefoot.
(continued on page 549)
(Table 34.1 continued from page 548)
mmNational average water rates for water delivered to the end user were on the order of $533 per acrefoot for small users, less for large users. Arthur Young Water and Wastewater Survey (1988).
nnMaximum of new contracts of U.S. Department of the Interior, Southern California. Personal communication from K. Frederick, U.S. Department of the Interior, to P. Waggoner, Connecticut Agricultural Experiment Station, February, 1991.
ooUse of 105 gallons per day (Solley et al., 1989) at $533 per acrefoot costs $63 per year.
ppAt $15 per acrefoot, the 3 ft evaporating in a year would cost $45 per acre.
qq27,000 acres in California produced 7,453 cwt of tomatoes valued at $18.30 per cwt. U.S. Department of Agriculture (1986).
rrNew York Times, December 19, 1989.
ssU.S. Bureau of the Census (1987, Table 670).
ttRaising an embankment from 12 to 15 ft high to 18 to 25 ft high to protect from major cyclones and to fortify them with concrete or boulders would cost about $25,000 per 100 ft (New York Times, May 12, 1991).
wwNational Research Council (1987).
yyNHRAIC data base.
aaaIllinois Department of Transportation (1986).
bbbIllinois Department of Transportation (1986).
cccFederal Insurance Administration (1984).
dddU.S. Bureau of the Census (1987, Table 670).
fffFrom Washington, D.C., to Oak Bluffs, Massachusetts. Personal communication from J. Ausubel, The Rockefeller University, to P. Waggoner, Connecticut Agricultural Experiment Station, May 10, 1991.
Hurricane Hugo would become more frequent. On the other hand, the cost of adaptation would include more frequent expenditures of $35 to $50 per person to evacuate or $30 billion to $90 billion to strengthen coastal buildings for stronger winds. In another example, a warmer and drier climate and no adaptation could raise the $800 million to $1,000 million per year for drought and cut the $3 billion for floods and $3 billion for winter storms and snows. Or, climate warming could raise the cost of floods by causing more rain and less snow in the spring. Adaptations would include costs for air conditioning and irrigation. They might include $1 million for an adapted wheat variety and some portion of the $33 million per year for the agricultural experiment station of a state in the Grain Belt. The residual impact would be the net of a new arrangement of production, comparative advantages, and prices.
Some entries in the table provide scale. For example, the U.S. gross national product (GNP) in 1986 of $4,235 billion is a standard for judging the $30 billion to $90 billion for strengthening coastal buildings for 100-mph winds. The projected change from a global average income of $3.0 thousand in 1985 to $7.1 to $35.6 thousand in 2100 suggests the future wealth for adaptation.
Again, these costs of impacts and adaptations are uncertain. Combining them with uncertain climate scenarios compounds the uncertainty. Nevertheless, the table illustrates the scale or order.
Before beginning these examinations of sensitivities, impacts and adaptations, we raise eight questions to keep in mind throughout the examination (Ausubel, 1991). They are familiar ones. Stating them at the outset makes our examination more exact. After the examination of activities, we will revisit these questions.
1. Is faster change worse than slow?
2. Will waiting to make policy and act drive up costs?
3. Are there only losers from climate change?
4. Will the most important impacts be on farming and from the rise of sea level?
5. Will changes in extreme climatic conditions be more important than changes in average conditions?
6. Are the changes unprecedented from the perspective of adaptation?
7. Will impacts be harder on less developed countries than on developed countries?
8. Are some hedges clearly economical?
Raising these questions at the outset provides a backdrop for our examinations. After examination of activities, we will see how these questions should be revised.
Primary Production of Organic Matter
Why this Subject
Investigation of sensitivity, impact, and adaptation to climate change begins with a paradox. The chief greenhouse gas, CO2, is feared for its effect on climate, but at the same time it is the key building material of all living things. Green plants are the eventual source of essentially all foods used by living organisms, whether plant or animal. They manufacture the food from CO2 and water in their green leaves, which are essentially all outdoors and hence subject to climate.
The vital role of plants for food, the peculiar effect of CO2 on them, and their exposure to climate cause us to examine farming, forestry, and the natural landscape early in this chapter. First, however, we examine commonalities among all three: photosynthesis, the pores that funnel CO2 in and water out of leaves, and the limits on experiments with systems of plants outdoors.
Using the energy from sunlight, plants convert CO2 from the air and water from the soil into food and oxygen. Since CO2 is the raw material for photosynthesis, one expects that enriching the air with CO2 will deliver more raw material and speed the formation of food. Although bottlenecks or limiting factors in the photosynthetic factory of a plant can restrict the speedup enabled by the delivery of more raw material, Figure 34.1 shows that the expected can happen. In a controlled atmosphere in a laboratory, raising CO2 from about 300 to 600 ppm speeds photosynthesis in corn by about 20 percent. In wheat it speeds photosynthesis more, by about 60 percent. Corn exemplifies plants called C4 whose photosynthesis is fast and yield is high today. Wheat typifies a more common sort of plant called C3 whose photosynthesis is slower than the other class today. Most plants in natural landscapes fall into the slower class.
CO2 arrives at the site of photosynthesis inside leaves through minute pores in the leaves. Since the interior of leaves is moist, water escapes through the pores. So much escapes that evaporation from an acre of foliage is about the same as from an acre of a lake. Not surprisingly, most plants have pores that close at night when photosynthesis stops. They also narrow when CO2 is abundant. The closing or narrowing saves water.
Some Initial Reasoning
Simple, direct argument from the physiological principles of the preceding paragraphs could lead to several broad conclusions about plants outdoors. At present concentrations of CO2 in the air, corn should yield more than wheat. Enriching the atmosphere to 600 ppm CO2 should increase all yields, especially plants with a photosynthesis like wheat that responds sharply to more CO2. It should close the gap between wheat and corn. It should especially favor the many plants in the natural landscape that have a responsive photosynthesis like wheat. Where ones with responsive photosynthesis compete with ones with unresponsive photosynthesis, the enrichment with CO2 should make the former more competitive. It should slow evaporation and save water. By speeding photosynthesis and slowing evaporation, CO2 should increase the efficiency of the use of water, the tons of yield per 1000 m3 of evaporation.
Experience supports the first conclusion: The average U.S. yield per acre of grain of corn is two to three times that of wheat. The other conclusions, however, cannot be tested in the same robust way. Instead, the behavior of an entire landscape must be argued or scaled up from laboratory or small controlled experiments. A logical question about scaling up is whether bottlenecks or limiting factors restrict the speedup of photosynthesis and slowing of evaporation.
About 1980 the debate was vigorous whether natural temperature, moisture, and nutrients outdoors would reduce the effects demonstrated in the laboratory. In the years since, experiments in chambers and laboratories have shown that enrichment with CO2 actually tempers the impacts of heat, drought, and salinity. More CO2 increases the growth of roots, shoots, and seed, more or less in step (Rogers et al., 1983; Cure, 1985; Oechel and Strain, 1985). Nitrogen deficiency does not bar the help of more CO2 to white oak seedlings (Norby et al., 1986) but may bar it for corn (Wong, 1979). Whether fertilizer deficiency bars the help of more CO2 remains ambiguous (Kimball, 1982; Goudriaan and de Ruiter, 1983; Pearcy and Bjorkman, 1983; Acock and Allen, 1985).
The argument about more CO2 narrowing pores and slowing evaporation shows the difficulties of scaling up. The narrowing of pores and slowing of evaporation can be shown in the laboratory. More CO2 may even increase photosynthesis of drought-stricken plants (Idso, 1988). The argument is whether the larger leaves encouraged by more CO2 plus the consequent warming of first the leaf and then the surrounding air will make the slowing minor (Jarvis and McNaughton, 1986). A comprehensive physical model of an unchanged canopy of a grassland, forest, or wheat field in a steady climate shows, however, that although 40 percent more stomatal resistance to evaporation would not cut evaporation 40 percent, it would cut it 12 to 17 percent (Rosenberg et al., 1990). In fact, when stomata were narrowed by a chemical spray, the evaporation from a barley field or a pine plantation slowed by 5 to 20 percent (Waggoner et al., 1964; Waggoner and Bravdo, 1967).
After all complications are tallied, a survey of many species, conditions, and experiments shows a ¼ to ½ percent change in growth for each percentage change in CO2 (Gates, 1985).1 In other terms, a rise in CO2 of half from 350 to 525 ppm would raise growth by 1/8 to ¼.
The generality of plants with a responsive photosynthesis gaining on others in rising CO2 has been demonstrated by crops and weeds. The responsive weed, velvet leaf, increased its growth more than the less responsive corn when CO2 was raised. Further, the weed, itch grass, which has the less responsive sort of photosynthesis, gained less than soybeans, which
have the more responsive sort of photosynthesis. Limited nutrients did not nullify the help of CO2, including greater height and leaf area that will affect competition for light (Patterson and Flint, 1980, 1982).
When four species grew together, the effects of moisture and CO2 amplified each other. One species with more responsive photosynthesis and that is adapted to wet soil gradually displaced one species with less responsive photosynthesis as CO2 and moisture were raised. The total weight of the four species rose with rising CO2, but the proportion produced by the two other species was low and remained low (Bazzaz and Carlson, 1984).
A final consideration is the long term. When the CO2 in the air above a tract of Alaskan tundra was raised, photosynthesis rose as expected but then declined to that of land exposed to normal air. In a continuing experiment in coastal wetlands near Chesapeake Bay, however, more CO2 in the air increased the carbon sequestered throughout the first 3 years and continues to increase it (Tissue and Oechel, 1987; Drake, 1989). So far, these contradictory results have not been reconciled.
Limitation on Experiments
Experiments with one or two factors controlled help us understand the effect of more CO2. Extrapolating or scaling up these brief experiments to the prolonged behavior of entire systems of plants is nonetheless full of uncertainty. In some experiments the light or ventilation in the chamber was unrealistic; the benefit of slower transpiration was not shown; the roots were restricted in pots, but the foliage was not crowded by neighbors; the treatments were not replicated or randomized; or the experiments were too brief to show adaptation, reproduction, or competition (Bazzaz et al., 1985; Eamus and Jarvis, 1989; Rose, 1989). Beyond these limitations of the experiments lie doubts whether unforeseen things like pests would appear and change the results of longer experiments or whether unforeseen interactions would arise in a whole system of plants unlike those in the simple one in a chamber.
Natural ecosystems as well as managed ones have herbivores eating plants and predators eating herbivores. The standing crop of plants on an area is a function not only of the primary production from plants but also of the amount of plant material removed by grazers and pests. Thus, even if increasing CO2 implies more plant growth, any change in the standing crop is uncertain. On farms plants will be protected by such techniques as integrated pest management and pest resistance in crops. In natural systems, on the other hand, CO2 fertilization might lower the nitrogen content of plants and thus change insect feeding or change populations of grazers and predators on them.
Similar doubts arise whenever reasoning cannot be tested because experiments
are impossible or because size and complication make them impractical. One wise course is to reason critically, rank the importance of the question against others, devote resources according to the rank, and then experiment as large and as long as resources permit. Another wise course is to analyze observations of the outdoors.
In an ideal world, all the living things on the land, in the soil, and in the ocean could be weighed annually and the change in carbon in them assessed just as the changing CO2 in the air can be seen in the record from Mauna Loa. This is, of course, impractical because things vary from place to place and because one large number would be subtracted from another, nearly equal, one.
The rings in long-lived trees have been examined for evidence of growth changing in step with the CO2 in the air. The trees were certainly growing faster, but one person interpreted it as evidence of the help of CO2 and another as the help of more precipitation (LaMarch et al., 1984; Gates, 1985).
Failing in a direct observation of changing growth of living things, one can turn to the CO2 record itself for evidence. While the average CO2 during the year is rising, the amplitude of the fluctuation from winter to summer increases about ½ percent per year. This is consistent with the CO2 enrichment of the air increasing growth worldwide. Although even this conclusion is clouded by uncertainties, it appears that the accumulation of biomass on the planet is either increasing or is steady, not decreasing (Revelle and Kohlmaier, 1986).
Because plants are sensitive to CO2 in the air, its concentration in the air is one of the things that will modify the change in the primary production of food as climate changes. In experiments more CO2 increases photosynthesis and slows evaporation. The question is whether these benefits will be large or small compared to the influences of climate and other changes.
Uncertainties attend scaling experiments up to the reality of a whole landscape. Nevertheless, the experiments do suggest thateven where factors like fertilizer or water are shortCO2 will speed photosynthesis, slow transpiration, and make plants with responsive photosynthesis more competitive. We know that any speeding, slowing, and competing will not be added to or subtracted from activities as they are today, but will modify the outcome of other changes as plants grow and interact in entire systems of plants.
Concentrate on Crops
The dimensions of U.S. agriculture can be seen in 1988 marketings in billions of dollars: livestock, 79; food and feed crops, 66; cotton lint and tobacco, 6; and lumber, 4. Although heat waves can kill chickens and blizzards can starve cattle, it will be argued later that animals are less sensitive to climate than crops. Of course, the economics of livestock production are tightly linked to the condition of range and pasture and with the supply of feed grains. A later section will be devoted to forests. So, the impact of climate change and adaptation of agriculture is examined here largely for crops, and this section is called Farming. Despite the riches of food in some places today, human population growth will swell demand during the decades of expected climate change.
Crop Sensitivity to Climate
It is not difficult to demonstrate that crops are directly sensitive to weather and climate. Frost kills citrus, drought shrivels wheat, and hail shreds soybean leaves. Frosts in Florida during the 1970s and 1980s contributed to the loss of much of that state's orange juice industry to Brazil (Miller and Glantz, 1988).
Another example is the 1988 drought in the United States. In North Dakota about the same area was planted in both 1987 and 1988, but in 1988 only 78 percent of the area was harvested and production was only 38 percent of the prior year. In Iowa about the same area was planted and harvested in both years, but only 69 percent as much was produced in dry 1988 as in 1987 (U.S. Department of Agriculture, 1989a).
There is a tendency in discussions of the potential effects of climate change on crops to consider only temperature and precipitation effects. However, changes in solar radiation (through altered cloudiness), humidity, and windiness are equally likely. They affect crops and must be borne in mind lest we think that prediction of climate change effects on crops will be less complex than it really is.
Weather also affects crops indirectly. When dry weather stops the sporulation of a fungus and its infection of grasshoppers, the hoppers swarm onto crops (Capinera and Horton, 1989). In wet weather, on the other hand, certain fungi blight crops (Waggoner, 1960).2
A neglected matter is how climate change would alter the organization and resources of soil, water, and the genetic basis for crops that undergirds farming. Muddy roads and washed-out bridges interrupt the supplies farmers need, and they stop farmers from carrying their produce to feed others.
But washed-out bridges can be rebuilt, whereas eroded soils are not so easily repaired and genes from extinct species can neither improve varieties nor make new crops.
So far, improvements in farming in many places have increased yields from the constant supply of natural resources as fast as human demands have grown or faster. Agricultural research and the farmers who apply its results have been highly successful. The clear question is: If climate changes during several decades, can these same people maintain the foundation of natural resources and raise more food for the escalating demands and numbers of people?
Soil is easily forgotten when new maps of vegetation zones are drawn for new climates. On the one hand, a new climate of wind or rain could erode an old productive soil. On the other hand, if the climate becomes favorable to crops where it was once arctic but the substratum is granite, centuries may pass before the soil becomes as fertile as the climate is favorable (Jenny, 1941; Joffe, 1949).
Moisture to sustain crops depends on amount and timeliness of precipitation and its balance against evaporation. The balance also fixes how much water is left to supply irrigation water by running off into streams and reservoirs and percolating into aquifers.
Only a few plant species among millions that grew at one time or another provided the genes for the crops that feed farm animals and us. Within a crop, genetic uniformity can cause vulnerability, as when Southern corn leaf blight arose and struck the U.S. corn crop in 1970. Fortunately, the genetic diversity of U.S. crops was broader in 1980 than in 1970 (Duvick, 1984). Nevertheless, we should be concerned that a climate change might extinguish wild races of major crops that might furnish valuable genes or even extinguish species that might become crops (Wilson, 1989). The large effort of preserving seeds in banks is described in the section ''The Natural Landscape," below.
At the outset of estimating impact, one wonders "of what?" If we want to know the impact of climate change on Iowa corn, we must ask: What will the change be in Des Moines? As related in our Assumptions, the future climate in any locality, especially the crucial factor of precipitation, will not soon be predicted confidently. So, we are left to think about sensitivities that can be multiplied by reasonable rates of climate change and integrated into the impact of a climate change.
One way of estimating the sensitivities is from the history of weather and yield. The regressions of Thompson (1988) exemplify the method. His regression coefficients for, say, corn yield on rain and temperature, month by month, are sensitivities. An early use of these sensitivities in assessing the impact of climate change produced such estimates as 11 percent less corn per 1°C warming in summer maximum temperature and 1.5 percent less for each 10 percent less summer rain (Bach, 1979). Later, a shift in the location of the Corn Belt (Blasing and Solomon, 1982; Newman, 1982), the help of CO2 enrichment, and even adaptation to the new climate were considered by Waggoner (1983).
Regression equations, while instructive, are not generally reliable for extrapolations beyond the range of the data used in their development. Also, they cannot be used to deal with CO2 enrichment effects. Process models or simulators that simulate plant growth, yield, and water use offer the best alternative to the regression model. Simulators, however, have their own limitations. For example, most simulate plant development by calculating growing degree days or some other index of heat accumulation. Given warmer temperatures, the simulators predict maturity earlier and so curtail the opportunity for production and accumulation of yield. This is reasonable. Plants do mature earlier in hot years. For this reason, however, simulated effects on summer crops may be too severe. Crops sown in the fall, on the other hand, benefit in simulation from the milder winters, break dormancy sooner, and mature before the hot, dry weather sets in. Hence, only moderate losses are often simulated due to even severe warming.
Thus far, most simulators have not dealt effectively with the episodic events that are most critical in determining yield, such as outbreaks of disease or the sterilization of pollen by extreme temperatures at critical times. One distinct advantage of the simulators is their ability to take direct account of CO2 fertilization. Another distinct advantage is that they can be used to consider adaptations such as different varieties, planting dates, and tillage practices. By submitting several years of weather records to the simulators, either actual or adjusted by some measure of climate change, the frequency distributions of annual yields can be calculated. Changes in frequency of bumper and disaster years may be as important an impact of climate change as changes in mean yields (Waggoner, 1983). A number of specific simulators that have been used in climate impact analysis are mentioned below.
Parry et al. (1988) summarized an ambitious study conducted through the International Institute of Applied Systems Analysis (IIASA) of impacts that concentrated on the cold and semiarid margins of agriculture. Various regressions and simulators were employed, as were other techniques of geography and climatology. Although various scenarios of climate change were considered, the results of one climate scenario from a global circulation
model were applied to all the regions. Existing sensitivities of agriculture to climate variability and potential sensitivities to climate change were identified. Preeminent effects were changes in length of the growing season and in growth rates changing the required growing season; changes in mean yield; changes in yield variability and in the certainty of expectable yields; changes in yield quantity; and changes in the sensitivity of plants to fertilizers, pesticides, and herbicides. Spatial shifts in comparative advantage and of crop potential could follow any or all of these. Climate change may also affect water balance and thus irrigation and flood control requirements as well as soil erosion, soil fertility, and pests.
Simulators were important in the Environment Protection Agency (EPA) assessment of the potential effects of climate change for U.S. agriculture (Smith and Tirpak, 1989). Regions studied included the Great Lakes (Ritchie et al., 1989), the Southeastern United States (Peart et al., 1989), the Great Plains (Rosenzweig, 1989), and California (Dudek, 1989). Various simulators were used, including CERES-maize (i.e., corn) (Peart et al., 1989; Ritchie et al., 1989; Rosenzweig, 1989), CERES-wheat (Rosenzweig, 1989), and SOYGRO (Peart et al., 1989; Ritchie et al., 1989). Fertilization effects of CO2 and opportunities for adaption were considered in these modeling exercises. The climate change scenarios produced by general circulation models (GCMs) for doubled greenhouse gases were applied to these models.
The major findings of these studies were that, without the direct effect of CO2, yields of wheat, soybeans, and corn declined in the Great Lakes, Southeast, and Great Plains regions, except in the northernmost latitudes, where frost-free season was lengthened. Decreases in yield stemmed mostly from the shortened life span of crops caused by higher temperatures. Differences among climate scenarios led to predicted yields ranging from mild gains to severe losses. Location within regions was important. Irrigation moderated yield losses as did higher CO2, least in the southeast and most in the north. However, significant increases in irrigation requirement were noted, especially with the more severe climate scenario. The adaptations tested, such as longer-season varieties of corn in Illinois, did not fully compensate for the loss of yield (Easterling, 1989).
More recently, the EPIC simulator has been used in a study sponsored by the Department of Energy to evaluate how climate change would affect agriculture in the Missouri-Iowa-Nebraska-Kansas (MINK) region (Easterling et al., 1991; Rosenberg and Crosson, 1991). Yields on some 50 representative farms were simulated by using the actual weather of the hot, dry 1930s as the scenario of climate change. Results varied between and within the four states because of differences in soil, crops, rotations, and other farming practices. Further, during the 1930s, the weather was not uniformly droughty in all portions of the region. In general, though, yields of summer crops
were reduced, on average, 18 to 25 percent. Wheat yields were reduced little. Irrigation requirement increased about 28 percent in Nebraska and Kansas. All of these effects were moderated by an increase in atmospheric CO2 content from 350 to 450 ppm. Such adaptations as altered planting dates, longer-season varieties, and tillage to conserve moisture moderated the yield losses.
Findings from the Parry et al. (1988) and the EPA (Smith and Tirpak, 1989) studies figure prominently in the recent report of the Intergovernmental Panel on Climate Change (IPCC) (1990b) in which estimated impacts are described as changes in productive potential against a baseline of present technology and management. Key findings were:
• It has not yet been demonstrated conclusively whether, on average, global agricultural potential will increase or decrease.
• Severe negative effects are possible in some regions, particularly those of high present-day vulnerability that are least able to adjust technically.
• Two broad sets of regions appear most vulnerable: (1) some semiarid tropical and subtropical regions (possibly western Arabia, the Maghreb, western Africa, Horn of Africa, southern Africa and eastern Brazil) and (2) some humid tropical and equatorial regions (possibly southeast Asia and central America).
• Changes causing water shortages in regions that are now exporters of grains (southern Europe, southern United States, parts of South America, and western Australia) may lessen their productive potential.
Whether one calculates the change in evaporation, the change in yield of crops in the southeastern United States or Great Plains, or the economic effect, the differing climate scenarios from reputable predictors produce farm impacts even more strikingly different than the scenarios themselves.
Difficult as it is to estimate yield sensitivities to the impacts of climate change, it may be even more difficult to estimate the economic impacts. Analyzing the impact of a climate change on the national income of the United States, Nordhaus (1991) estimated that the largest impact on a sector would be on farming. He estimated it would be +$12 billion to -$12 billion (see Table 34.6).
Adams et al. (1990), using a spatial equilibrium agricultural model of the United States, tabulated results of the EPA assessments mentioned above and found that for one scenario of climate change prices would fall 18 percent and the economic surplus would rise $9.9 billion, but for the other prices would rise 28 percent and the economic surplus would fall $10.5 billion. In the MINK study (Crosson et al., 1991; Rosenberg and Crosson, 1991), with no adjustments and no CO2 enrichment, a regional input-output
model shows that the decline in regional production of corn, sorghum, wheat, hay, and soybeans would be $4.4 billion (1982 dollars) or 1.4 percent of the total regional production to final demand, if all the decline were in exports. If the decline in the feedgrains, corn and sorghum, were to be only in exports, the loss would be $3.1 billion or 1.0 percent, but if the decline falls on animal producers the loss would be as high as $30 billion (or 10 percent of the regional production) because of impacts on meat packing, the largest manufacturing activity in the region. In all cases, higher CO2 reduces the losses.
When climate change is foreseen, the specter of worse pests usually rises. The relationships between climate and pests are clear; the northward movement of the overwintering range of insect pests, for example, can be calculated for scenarios of climate change on plant-pest interactions (Stinner et al., 1989).
There are, however, all sorts of pests, some favored by cool weather and some by warm, some by wet weather and some by dry. This diversity can be seen in maps of the prevalence of plant disease, such as the one that shows more apple scab in the humid western part of Washington State and less in the arid central part (Weltzien, 1978). The relationship between weather and weed pests led to the drawing of a "pestograph" with axes of temperature and moisture. Purple nutsedge flourishes in warm wet, weather; field bindweed in warm, dry weather; quackgrass in cool, wet weather; and Canada thistle in cool, dry weather (National Research Council, 1976).
Since the outlook is for warming, the opinion of an expert on tropical diseases of plants is relevant: "… it would seem that continued successful production of the majority of tropical crops, which are highly homogeneous genetically, belies both the contention of imminent danger from homogeneity as well as the oft-stated maxim that diseases are worse in the tropics because there is no winter. Neither seems to be a valid generality" (Buddenhagen, 1977). In short, the certain outcome of a climate change is not more nor less plant pests, but rather different ones.
Adaptation of Food Production
Adaptation also Makes Impacts Uncertain
Added to the uncertainty about climate in, say, 2050 is uncertainty about the sensitivity of farming then to any climate. The impact of a temperature change on, say, corn in Missouri is irrelevant if the Corn Belt moves somewhere else. If yield per acre and hence prices change, demand will change, and technology may change the sensitivity to weather.
The economic effects arrayed for the MINK study (summarized by Rosenberg and Crosson, 1991) were calculated on two unlikely assumptions: first, that
the climate changes immediately on the world as it is today and, second, that no adjustments to these changes are attempted. In both the MINK study and the earlier EPA and IIASA studies, some simple adaptations to the shortened crop-growing seasons are tested. The simulators show in general that longer-season varieties and earlier planting can help overcome yield losses. Wilks (1988) modeled the response of North American corn and wheat yields to a doubled CO2 climate change and reached a similar conclusion. He found that by selecting the most appropriate planting dates and cultivars for the changed climate, yield reductions could be minimized. Had he considered the effect of CO2 fertilization, his results likely would have been even more optimistic. Substitution of more resilient species, such as sorghum for corn, provides other tactical opportunities. In the MINK study, however, sorghum did not improve the farmer's balance sheet because of its lower price. On the other hand, simple changes in tillage to conserve water reduced loss of yield.
Can Farming Adapt?
If truly radical changes occur in climatefor example, a 5°C warming or severely curtailed or even greatly increased precipitationsimple adaptations will not suffice. It is more reasonable to believe that under such stringencies agricultural systems would be radically changed. Tropical savannah or tropical desert might exist where we now grow corn. Nonetheless, there are some indications that adaptations can be effected even to large changes in climate. The first is the adaptation of wheat to both colder and warmer climates (see Figure 34.2).
Hard red winter wheat has cultural and economic advantages over the competing spring wheats. Most important are that it is planted in the fall, which avoids waiting for the soil to thaw in the spring, and it is harvested before the heat and drought of summer can cut its yield. Its adaptation by breeding and other techniques is illustrated by the difference between the former northern boundary of Sidney, Nebraska (Location 1 in Figure 34.2), and the present one near Sidney, Montana (Location 2 in Figure 34.2). Compared to the old boundary, the new one has 20 percent less precipitation, an average temperature that is 4°C colder, a growing season that is 10 days shorter, and an annual amplitude of monthly means that are 6°C greater (Rosenberg, 1982).
Farmers adapt to a lack of rain by irrigating. From 1954 to 1984 the irrigated acreage in the United States rose from about 30 million to 45 million acres. An adaptation to a shortage of water is raising the efficiency of irrigation. Improved irrigation efficiencies during 1950 to 1980 reduced per-acre applications from the Ogallala Aquifer in the High Plains from Texas to Nebraska by a third. Among irrigation methods, efficiency rises
from 40 to 50 percent for furrow to 60 to 92 percent for trickle irrigation, but Peterson and Keller (1990) found that improving use of existing technologies was more economical than a change in technology.
An average increased demand for irrigation water of approximately 15 percent (for a mixture of alfalfa, corn, and winter wheat) was found by Allen and Gichuki (1989) for the Great Plains states from Texas to Nebraska in response to two GCM scenarios. Demands may be greater during peak periods, and growing seasons may be lengthened by various adaptations. Similarly, in the MINK study, irrigation demands increase by about 25 percent for corn and sorghum and 10 percent for wheat exposed to the climate of the 1930s. It is important to recognize that, in climate circumstances that create a greater demand for irrigation water, runoff to streams may also be reduced. And of course ground water supplies will not increase, but rather will decline more rapidly. Additionally, as Frederick and Kneese (1990) have shown, demand for limited water supplies for other usesmunicipal and industrial, fish and wildlife, recreation and navigationmay make water too expensive for agriculture. This trend is already occurring in the arid West and would likely be accentuated by any climatic change that decreases supplies or increases demand.
Can it Adapt Swiftly and Cheaply?
To adapt successfully to a climate change in a half century or so, adaptations must be quicker than the half-century and they must be economical. The average lifetime of a successful cultivated plant variety shows how fast varieties of a crop can be adapted. In 1981 plant breeders estimated that varieties lasted 7 years in corn, 8 in sorghum and cotton, and 9 in soybean and wheat. They opined that the life spans would grow shorter (Duvick, 1984). In 1990 a commercial seed company testified that a successful new variety of wheat has a useful life of about 6 years and costs $1 million to develop (Newlin, 1990).
Canola illustrates the speed of introduction of a wholly new crop. During World War II a small area of rapeseed was grown in Canada for marine lubricant. Figure 34.3 shows that during the next 30 years or so its production expanded rapidly to a level of more than 3 million tons per year. The expansion required breeding to make the oil edible, changing its name to canola, building processing plants, and organizing markets (Waggoner, 1990).
Irrigation adapts crops to dry climates but requires water and has potential long-term negative consequences that must be taken into account. During
recent decades irrigated area increased by less than 2 percent per year in the West, but by more than 10 percent per year in the East (Peterson and Keller, 1990). These normal life spans and costs of varieties, the introduction of a new crop in about 30 years, and the rates of change of irrigationall without the spur of changing climatesuggest that adaptation likely can keep ahead of the climate change, as most specialists see it in the next three to five decades, say. But rapid climate change, transient changes such as colling prior to warming, or major changes in the distribution (variability) of temperature, precipitation, storminess, and so forth, could, of course, challenge this essentially optimistic view of adaptation potentials.
Although predictions of future technologies for several decades have erred notoriously by missing large ones altogether, it is worthwhile to describe some strong possibilities. As for varieties, the survey of life spans found, for example, 60,000 corn cultivars and hybrids in preliminary trials in 1980 compared to 454 in commercial use. Future adaptation requires pursuit of such trials and underlying plant breeding by about 1,000 private and public workers, the seed banks, and the natural sources of diversity of crops (Wilkes, 1984). Expansion of irrigation in the West would be hard if the climate grew drier but easy in the East because only a fraction of the supply is consumed today (Peterson and Keller, 1990).
What of the developing world? We have presented a relatively optimistic view in this section of current and future agricultural capability to cope with at least the assumed climate changes. The implicit assumption has been that the research and the farmers in the developed world (more particularly in the United States) will be adequate to the task and that new tools such as biotechnology and computer-guided irrigation will make rapid adaptation easier in the future than it is today. But now we must address the question of means and resources. Can the developing world cope as easily?
There is no reason to believe that the developing countries as a group will be exposed to worse climatic changes than the developed countries. Not everywhere in the developing countries are the climates marginal. What is certain, however, is that their margins of survival are smaller and that the impacts of climatic change might be more immediate and profound where the infrastructure, including research capacity, is smaller. In fact, an Indian agriculturist, Jodha (1989), argues that farmers in developing countries use well-tried techniques in times of stress and that these are an arsenal from which to draw when the evidence of climate change becomes strong enough to convince them and their governments of the need. Jodha provides many examples of these responses, particularly from the Indian experience. In
climatically marginal areas, particularly in the semiarid tropics, even a slow, small change toward a worsening climate can, of course, accentuate climatic risks. It is vital to note, however, that the share of gross domestic product from agriculture in less developed countries decreased 50 percent between 1960 and 1980, so that the economic transformation of those countries may be lessening their sensitivity to climate (The South Commission, 1990).
Earlier we pointed out that, to date, impact assessments have tried to answer the question: What would be the impact of a future climate change on the world (nation, region, state) as it behaves today? Insights have surely been gained. However, the really relevant question is: How will a region behave at such time as the climate does change? Impact assessors are obliged to anticipate what agriculture (and all other sectors) will be like when climate changes are finally felt.
The MINK study described above attempts to do this. Starting from an understanding of the region as it is today, its agriculture, forests, water resources, and energy economy are projected to the year 2030. For example, absent climate change, its crops are projected to yield about 75 percent more than they do today. The imposition of the 1930s climate reduces yields by about 25 percent, but the CO2 enrichment offsets about half of that loss. Adaptationsboth autonomous in the sense of being easily accessible and relatively inexpensive to adopt and policy-driven adaptations prompted by the perception or knowledge of certain climate changebring yields back almost to the level of no climate change.
As in all such studies thus far, the results ought not to be taken as predictions but rather as illustrations of a method that should reveal the potential for adaptations to cope with future climate change.
If climate changes, crops will be the exposed and sensitive part of agriculture in both rich and poor countries. Their sensitivity is visible as when they prosper after a rain or wither during a drought. The speeding of their photosynthesis by rising CO2 or encouragement of one of their pests, on the other hand, are invisible. Their sensitivity also lies in the sensitivity of the soil and so forth on which they depend. The direct sensitivity of crops to weather is estimated both from history and experiments. When it comes to estimating the impact of climate change, however, the uncertainties of the scenarios of climate change and the invisible factors and the tempering of exchanges and adaptations cloud the crystal ball. Fortunately, actual experience shows that farmers do adapt promptly. To adapt promptly they need effective research outdoors that applies, among other things, ample biological diversity in breeding. They need incentives and freedom to adapt.
Forests and Grasslands
Concentrate on the Managed Trees
Although the production from the fifth of U.S. land in crops is valuable, forests and rangeland shape much of the landscape. In 1982 forest land was fully 29 percent and pasture of different kinds was another 26 percent of the total land area of the United States (see Figure 34.4a). The forests alone provide a quarter of the industrial raw materials, give millions of people recreation, shed much of the water we use, and shelter countless plants and animals (U.S. Office of Technology Assessment, 1983; Cordell, 1989; Flather and Hoekstra, 1989; Guldin, 1989).
This section will concentrate on the managed forests. The managed pastures may be considered part of farming, dealt with in the preceding section, ''Farming." The unmanaged forests and ranges are the subject of the next section, "The Natural Landscape." Figure 34.4b shows, of course, that forests managed for timber and unmanaged ones are not distinct classes. Here, however, the concentration is on managed ones.
Much forest is inaccessible, reserved, or unproductive (Figure 34.4b), removing it from practical management. Unproductive forests yield less
than 20 ft3/A/yr, making their management impractical. The 10 percentindustrial forest is probably the best land and the most managed. The public and nonindustrial private forests range from intensively managed to ignored. In sum something like half the forest land is probably managed today. Climate change that makes more or less land unproductive could change the estimate of "half."
Although planting forests to take in CO2 from the air is a prominent suggestion when climate change is discussed, the subject here is impact and adaptation to a change. The topic of reforestation is addressed by the Mitigation Panel in Part Three.
Because trees are long-lived organisms, they are adapted to withstand significant fluctuations in climate. Although a forester might want to factor climate change into the choice of the species of trees to be planted today, the uncertainty of climate scenarios, particularly at small-scale units that are most useful for forest management, renders prescriptive planning difficult. For example, storms like hurricanes profoundly affect tall and long-lived trees, but scenarios in such detail are too uncertain for profitable
reasoning. We consider only the climate changes considered in Chapter 32, the section "Assumptions."
Just as farming depends on natural resources, forestry does, too. Because forests and ranges are bigger and their yield of money per acre is less than from farms, they may depend more on the natural resources. In all events, soil and its moisture must be right for a new climate to continue to produce lumber and support cattle. And, adapted varieties must be found.
Sensitivities of Forests
As climate changed, the sensitivity of forests to climate changed them. Consider pines in Eastern North America during the 18,000 years since the Ice Age maximum. They started from a block in Georgia and Carolina, nearly disappeared 12,000 years ago, and then spread to occupy one zone from Louisiana to Carolina and another along the northern border of the Canadian prairie (Webb, 1986).
Changes in temperature, precipitation, and atmospheric concentrations of CO2 may have direct physiological effects on the growth of forest trees. For an individual tree, the impact of a climate change includes the direct effect of CO2 reported earlier. Although the longer growing seasons of a warmer climate would raise productivity, ill-adapted trees could suffer frost damage during their prolonged growth and others might not be chilled enough to germinate (Cannell and Smith, 1986; Cannell, 1987; Kimmins and Lavender, 1987). The sensitivity of individual trees to moisture is demonstrated by the difference in species from the north to south slopes of many ridges and from the dry, rocky summits to the wet marshes below.
The sensitivity of entire stands of trees has been estimated from tree rings, pollen deposits, and simulations. Although the evidence is equivocal, faster growth of tree rings has been interpreted as evidence of more CO2 in the air.
Pollen records provide evidence of the past range of various plants. Natural migration rates can be inferred from these data. The rate of migration of temperate forests, as inferred from pollen records, is about 100km per century (Davis, 1981). Isotherms on a map of growing season climate show a gradient of about 2°C per 500 km. So a climate change even in the low end of the assumed range of 1° to 5°C in a century or less would move the isotherms several times as fast as the pollen record shows forests have moved on their own.
More complicated computations by mathematical models generally confirm the rough estimation in the preceding paragraph (Solomon and Tharp, 1985; Solomon and Webb, 1985; Solomon, 1986b; Botkin and Nisbet, 1989; Urban and Shugart, 1989). Model results can be summarized as follows. The response of mature forests depends on the species composition. If the
mixture includes some species that are well suited to a warmer climate, the total stand biomass can increase despite the decline or death of some components. The increase may be rather rapid if the responsive species are dominants or can occur more slowly if the responsive species occupy lower canopy positions. If none of the species in the forest are adapted to the warmer climate, total biomass will decline, perhaps rapidly.
As on the farm, climate changewarming, changes in soil moisture, and changes in the composition of the atmosphereraises the specter of more pests in the forest. If the change itself weakens a tree, it will likely be attacked by some pest. In an analogue, trees weakened by defoliation or air pollution are then attacked by bark beetles or the fungus Armillariella (Smith, 1981). Also, a different climate encourages different pests. As discussed above, however, warmer climates do not today have greater outbreaks and epidemics than cooler onesjust different.
The calculated impact of doubled CO2 on forests as on farms depends on the climate scenario, region by region. Since that is uncertain, generalities drawn from Figure 34.5 must serve. It is the gist of complex mathematical models and relates growth as a percentage of the maximum for the forest type to a factor of climate, exemplified here by temperature (Botkin et al., 1989; Urban and Shugart, 1989). If the current climate is at the heavy vertical line, spruce fir dominates the forest. If the climate warms to I, both types grow faster and presumably, the growth of the whole forest is faster. If the climate warms on to II, the hardwoods have an advantage, although they may have to wait until the spruce fir die. If the climate warms all the way to III, both types decline.
After the decline, a new forest may or may not emerge. If the new climate were as hot and dry as western Texas, for example, none would emerge. If it were not so severe, the new forest would depend initially on the supply of adapted species on the spot. In a mixed forest these might well be on the spot. If the adapted species were absent, the new forest would depend on their arrival. Although they could be brought by people, their arrival by natural migration could be slow, as already mentioned.
In Figure 34.5 forests are exemplified by spruce fir and northern hardwoods and climate factors by temperature. The curves are set first by zero growth at the limit of the ranges of the two types of forest. Then the maximum of each curve is set at the same maximum at the climate optimum for each type, making the growth rates relative. While the general pattern is realistic, these models probably overstate the sensitivity of the forest to climate change. The limit of zero growth is set by the edge of the current range of the types, but competition more than inability of an
isolated tree to grow fixes the edge. So, the present forest could persist longer than predicted.
The natural pace of adaptation may be too slow to maintain a particular forest on a particular tract of land. Using climate scenarios like our assumptions, some researchers (Franklin et al., 1989; Urban and Shugart, 1989) estimate that the changes in forests we have discussed will appear in 40 to 70 years. To be effective, adaptive responses must operate within this time frame.
Move Lumbering or Use Less Wood
If climate becomes unfavorable to forests in one region and favorable in another, lumbering can move as it has before. Between the 1700s and
today, lumbering moved from colonial New England to the Great Lake states, from there to the South and West, and finally back to the South.
Another adaptation to a changed forest is simply using less timber. If timber becomes scarce, prices will rise and less will be demanded. History shows that actually happens. For example, since 1900 the price of softwood timber in constant dollars rose about 2.5 percent per year, and the consumption of softwood lumber fell from a peak of 530 board feet per capita in 1906 to only 190 today (Binkley, 1988; Binkley and Vincent, 1988; Fedkiw, 1989; Haynes, 1989). Economic adjustments compensate for impacts on the forest, they are felt far from the affected area, and climate change makes winners as well as losers.
Regenerate a Forest
Some forests in the United Statesperhaps half of the totalare tended, and on these lands managers can intervene to speed adaptation of forests. The costs of some interventions are given in Table 34.2. If a stand of trees dies, regenerating it with an adapted species is an obvious adaptation. At about $130 per acre based on 1988 costs in the South, it is costly.
To plant a site, foresters already choose trees genetically adapted to it. For example, loblolly pine grows from the wet coastal plains of Virginia and Carolina to dry Oklahoma and Texas. An island of loblolly in central Texas is aptly called the Lost Pines. In Arkansas fast-growing loblolly from North Carolina are planted on most sites, but drought-tolerant local ones are planted on dry sites (Farnum, 1990).
Although knowledge of the genetics of commercially important trees is advanced, the seed of most species of trees in the United States has not been collected and screened. On the other hand, since pines evolved from Mexico and a large number of species of this tree are found there, there is an active program to preserve this genetic material. Expecting warmer climates, foresters should look for candidates in places that are warmer today.
Regeneration is difficult, and a drier climate will make it more difficult. As Smith (1986) noted, "A very crucial race against time which the seedling must make is that of extending its roots downward faster than the loss of water through direct evaporation from the capillary fringe [of the soil] can overtake them." So, success may need first hardy seedlings of a variety adapted to the site and then appropriate control of shade and competitors.
Manage the Forest
Light, nutrients in the soil, and especially moisture govern the growth of a forest. Several ways of enhancing these are shown in Table 34.2. Where
TABLE 34.2 Cost of Forest Management Activities in the U.S. South, 1988
moisture is short, removing understory plants increases the productivity as much as 25 percent (Zahner, 1955). In terms of adaptation this could offset a 25 percent decrease in growth by a drier climate. Removing parts of the main canopy and thus the leaf area also saves moisture, prolongs summer growth, and saves some individuals. Cutting less adapted portions of a mixed forest has been estimated to increase productivity by as much as 25 percent (Larson et al., 1989). Although wet forests have been drained, the converseirrigationis too expensive to be practical outside nurseries or seed orchards.
If climate grows hotter and drier, controlling fires will be more important and costly, but if the climate grows wetter, it will be less costly. Fire control can be effective. In the early 1900s about 20 million acres burned each year in 40,000 incidents. By the 1960s the area burned declined to only 2 million to 3 million acres annually despite an increase in the number of incidents to 100,000 per year. The recent fires in Yellowstone reflect more a failure to use fire control and management measures than a failure of the measures themselves.
Just as the speed of replacement of capital stock was relevant in the discussion of innovation in Chapter 33, the section "The Tools of Innovation," a shorter rotation from planting to harvesting a forest raises its adaptability. Surprisingly, middle-aged forests are at most risk if climate changes;
young ones can be replaced at comparatively low cost and older ones are valuable to salvage.
The modern practice of replacing the mixture of hardwoods typical of late stages of succession with pure pine typical of the early stages has already shortened rotations. In plantations of southern pine the rotations are 15 to 35 years, and in those of Douglas fir on the Pacific coast they are 35 to 50 years. Although these uniform plantations may lack the diversity of natural forests, the speed of change creates flexibility.
Will the Adaptations Work?
Although adaptationsin the economic system or in the forestsseem technically feasible, will they work in time and on the scale needed? Are they flexible and robust enough for the uncertainty and prolonged wait of climate change? How costly in dollars and in carbon are they?
Because the future climate is uncertain, the most desirable adaptations would be robust, performing well in many climates. And they would be flexible, that is, quick and cheap. Regeneration is neither quick nor cheap. Planting a tree adapted to a climate commits it to that climate for decades. Table 34.2 shows it is not cheap, and eliminating the old stand and replacing it is not quick.
Some management measures rate better, as Table 34.3 shows. The table also shows that two of the measures already are supported by the federal government. Table 34.2 shows the cost of these measures. The costs in carbon must also be estimated lest the management practices we adopt actually worsen the production of greenhouse gases. Shortening rotations will reduce the amount of carbon stored in forests because little carbon is absorbed during regeneration and the small trees grown on short rotations
TABLE 34.3 Flexibility and Robustness of Management
produce few products, such as lumber, which store carbon for long periods. A plantation of loblolly pine takes in carbon fastest at age 30, about when it is normally harvested. On the other hand, spruce fir forests take it in fastest at age 95, older than these forests are usually harvested (Birdsey, 1990). The net carbon intake, including subtraction of the fuel used during management from the accumulation of wood in the trees, must be estimated to appraise the effect on greenhouse gases. This comprehensive analysis has not yet been done.
Forests and pastures each cover more than a quarter of the land in the United States. Because trees live long lives, they are naturally adapted to variations in climate. But their long lives also make them vulnerable to climate change. If plants adapted to the new climate are not on a particular tract of land, they await natural migration or being brought in. Natural migration is too slow to keep up with the anticipated climate change. The needed artificial regeneration of forests will be costly and difficult, and needs adapted species or varieties. The adaptation of valuable forests by management is practical if the methods are robust enough to work in many climates and flexible enough to change quickly and cheaply. Finding the will and way to apply them widely enough to make a difference is a challenge.
The Natural Landscape
Difference from Farming and Forestry
The natural landscape is made of unmanaged terrestrial ecosystems. It is a goodly portion of the 29 percent of U.S. land that is forest and 26 percent that is pasture. With respect to these natural ecosystems, the goals of adaptation are maintaining species richness, the major types of ecosystems, and the evolution that produces diverse living things adapted to circumstances.
Although they are unmanaged and therefore require no direct expense, the natural ecosystems help humanity in two ways. First, the species of animals, plants, and microorganisms are usefulpeople harvest them as game, fruit, drugs, and so forth. Also, the systems perform natural services. A growing forest, for example, absorbs CO2 and stores carbon as wood while emitting O2. A natural ecosystem can cleanse water while providing a place for exercise and aesthetic pleasure.
The value of these services is hard to quantify, but some attempts have been made. For systems, two values have been estimated. Building a system that would duplicate the treatment of waste water and the spawning
of fish in a hectare of Louisiana wetland would cost $205,000. The storage and purification of water, binding of soil, and fertilization by a hectare of streamside vegetation in Georgia is worth $2,000 per year. In carbon monoxide absorbed, the value of a hectare of pasture is 440 kg per year (Wharton, 1970; Gosselink et al., 1973).
No physiological evidence suggests that animals would be affected directly by the projected changes in concentrations of greenhouse gases. The chief effects on animals would be via climate changes induced by CO2 and by climatically induced changes in the plants that feed and shelter the animals.
Natural ecosystems are more vulnerable to climate change than are managed ones like farms or plantation forests, because, for example, natural ecosystems would not be irrigated nor would their components be replaced to adapt them to a climate change. Our earlier statement that climate was only one among many changes applies especially to natural ecosystems. The total change in an ecosystem depends not only on its sensitivity and the change in climate but also on the system's absolute sensitivity to a variety of other changes influencing soil and water chemistry (e.g., land use, water use, and pollutants) or habitat fragmentation as, say, clearing reduces the acreage of natural vegetation or modifies it.
Resemblances to Crops and Forests
Experiments showing the advantage of CO2 enrichment to plants with responsive photosynthesis versus those with less responsive photosynthesis were performed with young, rapidly growing annuals (Bazzaz and Carlson, 1984). Such experiments with a few plants are the basis for computing the sensitivity of entire stands of natural plants (Botkin and Simpson, 1989). The uncertainties of extrapolating or scaling them up to the behavior of entire systems of long-lived plants have already been mentioned, and they are great for the diverse systems of natural landscapes. Looking to the past for analogs of future change is fraught with difficulties because past climates are not known precisely and because what we do know suggests that they differed from those projected for the future. During climate changes since the Ice Age, both individual species and ecosystems appeared, moved, expanded, and disappeared (Spauling and Graumlich, 1986; Davis and Zabinski, 1990). Qualitatively similar changes are likely in the future.
Diversity is Accompanied by Both Sensitivity and Resistance
The rule that there is security in diversity is an axiom of ecology as well as finance. In a natural landscape many species take in carbon and make
food by photosynthesis. The loss of a few species would probably change the photosynthetic rate of the system little. In fact, diseases removed first the chestnut and then the elm from eastern forests, but their photosynthesis was quickly replaced by other species. So, climate change is liable to eliminate species from the natural landscape, but its diversity will protect those functions, such as photosynthesis, that are carried out by many species.
Some ecological processes are carried out by only a few species, however, and some species exert strong influences on the functioning of the entire system. For example, only a few species fix nitrogen, and the grazing of a single species of large mammal may alter landscapes. A single parasite infecting a pest like gypsy moth would alter forest dynamics considerably. Seastars and gulls are predators whose activities markedly change the structure of intertidal communities. If climate change removed one of these species or encouraged another, even a diverse landscape could be affected (Paine, 1980).
Some Computed Examples
The way natural landscapes might be affected and even transformed has already been discussed. On a tract, long-lived plants could persist, but not forever, even though their reproduction failed. Formerly suppressed but now better adapted plants could grow up, and new, adapted plants could slowly move in. Therefore, changes in an ecosystem are likely to involve both movement at the margins and reshuffling of the species that formerly defined the system.
Mathematical simulations of natural landscapes have been run that incorporate physiological processes of plants much as GCMs incorporate atmospheric processes. The simulators mimic the present zones of plants in North America. A simulator was fed the climate scenario produced by a GCM for a doubling of greenhouse gases. It produced modest changes in west central Ontario and east central Tennessee. Between the two, however, it resulted in a near disappearance of northern hardwood forest in northwest Michigan. A later quadrupling of greenhouse gases, however, caused a change from one sort of forest to another in Ontario. After four centuries, the mass of plants rose in Ontario, fell in Tennessee, and changed little in Michigan (Solomon, 1986a; Solomon and West, 1986). The simulation using doubled greenhouse gases supports the view that major changes
can happen on the margins of zones fed the simulator as well as the validity of the simulator itself.
Another simulation, which assumes summer drought grows more severe, predicts shrinking of forests in central North America eastward and northward as trees die and do not reproduce (Neilson et al., 1989). For crops some scenarios cause more yield and some cause less (Adams et al., 1990), and the same disparity is likely when forests are simulated with other scenarios from other GCMs. Anticipating impacts or designing adaptations is, therefore, difficult.
Changes in the plants that feed and shelter them would affect animals. For example, one simulation suggests that a warming of 2°C would eliminate the jack pine barrens of central Michigan in a century. Because these barrens are the only breeding ground for Kirtland's warbler, the simulation implies that it, too, would be eliminated unless the management activities currently employed by government agencies to enhance its breeding habitat could be adapted to offset these declines (Botkin et al., 1991).
Some Management of Natural Landscapes is Justified
Since natural landscapes are, by definition, unmanaged, we logically first examine the responses they make naturally. But a climate change caused by people may require human intervention to protect natural landscapes.
Although the changes described above were called impacts, they are also the natural responses of landscapes. The question is whether the responses are consistent with the goals we identifiednamely, maintaining species, ecosystems, and evolutionary processes.
The simulated landscapes resulting from doubled greenhouse gases support altered communities of plants rather than barrens. Nevertheless, achieving these goals for natural landscapes will be difficult. In the unmanaged systems of plants and animals that are much of our landscape, the ability to change may be much slower than climate change, making their future problematic. The slow pace derives from the long lives of some of their components, such as trees that live longer than the ones planted and cut for timber. It comes from the slow and chancy arrival of seed and migrant animals traveling on the wind, in currents, or along corridors. It comes from the
slow rates of succession and from even slower rates of evolution. For the sort of climate change we are assuming, timely adaptation of every species and conservation of the countless cooperators in the natural landscape are highly unlikely.
Intervening in Natural Landscapes
Intervening to maintain species diversity, ecosystem functioning, and evolution can take two broad forms. Components of an area can be saved off the site, which saves some diversity and may permit evolution. The means are gene and seed banks, libraries, gardens, and zoos. Or, the entire landscape can be saved in situ (where it is), which saves systems, diversity and evolution.
Banks or Gene Libraries
Banks for parts, such as tissue, are the most technological and furthest from natural ecosystems. The discovery that genes of a valuable species can be incorporated into a microbe opens the possibility of libraries of genes in test tubes. Going further, genes can be preserved in frozen or freeze-dried form. So far, libraries exist only for a few genes of medical or scientific interest. Although the technique is inexpensive, its extension to the countless species in nature has not begun.
Tissues, like embryos or protoplasts of both plants and animals, can be cultured in a laboratory. Specific cultural methods may be needed for each species. Further, regenerating whole organisms from the tissue may be impossible, or it may be hard or become harder the longer the tissue is cultured. Also, tissue cultures, like gene libraries, need skilled people and laboratories.
Nevertheless, tissue culture is used for plants that are propagated vegetatively or have short-lived seed. It is used for some animals. For example, six replications of 800 cultivars of grapes can be maintained as growing tips in two square meters by transferring them only yearly (Henshaw, 1975; Wilkins and Dodds, 1983).
From a packet of leftover seed saved in a cellar to thousands of collections in the National Seed Storage Laboratory in Fort Collins, seed is the classic way of banking a plant. Special attention is paid to safeguarding the genetic diversity in crops. For example, 60,000 accessions of rice and its relatives are stored at the International Rice Research Institute (International Rice Research Institute, 1983).
Because seed banks have been a reality for decades, we know their problems as well as their promise. Seed does die. Moreover, in the 2 million accessions in banks worldwide, information is lacking about sources in 65 percent, about useful characteristics in 80 percent, and about germination in
95 percent (Peeters and Williams, 1984). The value of seed banks to farming is nonetheless great. The Agricultural Research Service spends $26 million to $28 million annually on germ plasm work, and this does not cover many forms of gene storage (National Research Council, 1990a). Large banks for the diverse seed of natural ecosystems do not, however, exist.
Although generally invisible, microbes are part of the natural landscape. Like seed banks, collections of microbes have been maintained for some time, and about 40,000 species have been collected and are exchanged (U.S. Department of State, 1982). In standard culture, microbes are short lived, they do change, and they must be transferred to new media frequently. To avoid these difficulties, collections are kept cold or dry.
A far less technical way of preserving a species is in gardens. Although the managers of botanical gardens or arboreta are more self-conscious of their role, any gardener can help. The wide cultivation of Bougain villaea and the para rubber tree, for example, protects them from extinction in tropical America. By collecting and breeding novel plants, nurserymen preserve biological diversity. By selling novel plants, they diversify the places the plants are grown and so help preserve them.
The 1,500 botanic gardens and arboreta in the world annually help instill an appreciation of plants in 150 million visitors, and they also conserve biological diversity. A strategy recommended for botanical gardens is concentrating on saving wild species, especially the task of saving those of their locality. They have yet to organize themselves for this work (International Union for Conservation of Nature and Natural Resources, Botanic Gardens Conservation Secretariat, 1989).
Zoos are increasing their role in propagating and reintroducing rare and endangered animals. All the zoos in the world, however, now house only about a half million mammals, birds, reptiles, and amphibians. To maintain diversity within a species requires large populations. In the long run, large populations of less than 1,000 species of the more than 20,000 species of these animals can be cared for in zoos. So despite the value of zoos, capturing and reproducing species in them can do little to maintain the diversity of nature (Conway, 1986, 1989).
Maintaining the diversity and functioning of species and their systems is like maintaining a chain. It must be steadfast. Once the maintenance and
hence the chain or species fails, the prior efforts are lost along with the chain or species.
Unfortunately, such collections are often dismantled or simply deteriorate after the specialists who built them up are no longer active. …Although they are often of great value internationally, [the collections] may if they are not actively utilized come to be viewed as a drain upon … the institution where they are housed. Even [with money], it is difficult to provide … the meticulous and sustained care that is essential for their survival without the attention of a specialist who is deeply concerned with them. (Raven, 1981)
Because management of banks and gardens is difficult and costly, in situ preservation of species and systems must carry the heaviest burden in programs to adapt to climate change. Methods for in situ preservation include reserving land for samples of representative, rare and endangered ecosystems or habitats. They include increasing the biological diversity in disturbed sites by land management or introduction of species. Finally, helping systems of plants and animals migrate is a method particularly important if climate change moves climate zones faster than plants and animals can follow. All these methods are already being used to maintain biological diversity in the face of harvesting of plants and animals and destruction of their habitats. The possibility of climate change adds a reason and complication for their employment. These methods all require that land be set aside for these purposes. The costs are those of purchase and the other uses of the land that are thereby forfeited.
The importance of maintaining biological diversity is reviewed elsewhere (Wilson, 1988; Western and Pearl, 1989) and need not be examined here. Rather, we concentrate on maintaining it if the system must move. The first method is to pick up a species and move it. The cost of regenerating a simple plantation of trees has already been shown to be high, and moving a whole system of plants and animals would be even more costly. Heavy seeds that are not naturally dispersed far and small animals would most need assistance. Dispersing them in advance of a climate change is not likely to help, especially if we are unsure what new climate they must cope with. Fortunately, they can be carried after climate changed. Although we should watch for species endangered by climate change, we can wait.
It may be costly, in terms of both price and lost opportunities, to delay establishing migration corridors for species that are readily stopped by such barriers as freeways, open fields, and suburban sprawl. Determining the best locations for corridors requires examination of maps on which existing migration routes are plotted. This will reveal where the most important barriers lie and their types. With this information, key areas to be acquired
and the form of modification of land use that is needed can be identified. Actions may range from simply providing overpasses or underpasses across highways, maintaining native vegetation along railroad rights-of-way, acquiring easements on lands to foreclose uses that would impose barriers to migration, to outright purchase. Other uses of the land compatible with its serving as migration corridors also need to be determined so that total costs of establishing and maintaining corridors can be reduced.
Uncertainty about future climates in a region adds to the reasons for corridors that will allow migration from areas where climates are becoming unsuitable for the organisms living there to areas where climates are more suitable. Even though we do not know how climates may change, a system of corridors connecting natural ecosystems across latitudes and longitudes and elevationally is likely to be useful under many patterns of climate change. It is important to begin a program to establish corridors now because of the long lead times involved in assembling information on locations of existing protected areas, surveying lands, negotiating easements or purchases, developing management plans, and changing existing land use patterns where necessary. Also, land prices are likely to rise, especially in areas where potential corridors are vulnerable to urban development that may also render their future use as corridors impossible.
From the natural landscape we take valuable plants and animals, and as a system it gives us natural services. Its diversity, functioning as a system and place for evolution, must be maintained despite climate change. Some of its components have long lives. To maintain these systems and their components requires lots of land. Simulations of the impact of climate change show vegetation zones moving slowly, and, as they move, animals are affected. Components of the system and hence diversity and evolution can be helped by preserving them in banks and gardens. The systems, however, are so large and complex that they must surely be preserved outdoors. Adaptation of the natural landscape can be helped by moving species when they are in trouble. To reduce future troubles, corridors along which plants and animals can migrate need to be established.
Because growing numbers of people and their power inevitably press the natural landscape, we must remember that the impacts and adaptations to any climate change will be added to other enormous changes.
The Marine and Coastal Environment
This section examines the relationship between the postulated effects (global temperature rise is of primary relevance) of greenhouse gas accumulations
on the world's oceans and on the coastal eclogical zones of the United States.3
An increase in global surface temperature can increase the volume of water in the world's oceans. Two mechanisms are responsible. First, water itself expands slightly when heated, so heat conveyed from the air to the oceans will increase their volume. Second, higher surface air temperatures can raise the net melting rates of glaciers. There appears to have been considerable variation in sea level in the earth's history. During the last ice age, when global temperatures were some 5°C lower, sea levels are estimated to have been at least 100 m lower than today (National Research Council, 1990b). In the last interglacial period 100,000 years ago, they were over 5 m higher when temperatures were 1° to 2°C higher than today (Smith and Tirpak, 1989).
The Effects Panel (Part Two) estimates that sea level will rise an average of 0 to 60 cm (0 to 2 feet) toward the end of the next century due to warming caused by accumulations of greenhouse gases.4
An increase in the volume of seawater does not translate, however, directly to sea level rise as measured at the shoreline. Because some land areas are rising and others are subsiding relative to the earth's center, local changes in measured sea level can vary considerably. For example, the worldwide average sea level rise for the last century is 10 to 15 cm, but the level has risen about eight times that amount in Louisiana and generally about three times that global average on the Atlantic and Gulf coasts of the United States (Smith and Tirpak, 1989). The magnitude of local vulberability to sea level rise, of course, depends on the local topography, as rising seas will affect more land area in low deltas than where coastal elevation profiles are steep.
Ocean Circulation and Temperature
In addition to sea level, a second important (though not yet very well understood) physical consequence of greenhouse gas emissions is the potential effect on ocean temperatures and currents. Available computer models show that a warming of the lower atmosphere in the range expected for a doubling of CO2-equivalent emissions is associated with 0.2° to 2.5°C warming in sea surface temperatures (Intergovernmental Panel on Climate Change, 1990b). The models suggest that the maximum warming will occur in the arctic and antarctic regions in their respective winters. Because it would
reduce the meridional (north-south) gradient of sea surface temperatures, the relative warming toward the poles could diminish the intensity of ocean currents and trade winds (Mitchell, 1988). However, nearer to land masses, computer simulations show enhanced temperature differences between land and oceans, which could lead to stronger along-shore wind stresses. These types of changes could alter upwelling (Bakum, 1990; Intergovernmental Panel on Climate Change, 1990b). Because there is good correlation between areas of upwelling and very productive fisheries, changes in upwelling could have an impact on those ecosystems.
Some scientists have raised the possibility that a dramatic systematic change in global ocean circulations could be triggered as a result of greenhouse warming (Intergovernmental Panel on Climate Change, 1990b). Broecker and Denton (1990) speculate that altered patterns of rainfall and evaporation, and changes in seasonal intensity may cause the ocean and atmosphere to ''flip" into a very different mode of operation. With the flip, ocean circulation is changed, carrying heat around the world differently, resulting in glacial cycles. Although current climate models are insufficient to evaluate the probability of such events, their potential consequences could be immense (Broecker, 1987).
Sensitivities and Impacts for Coastal Habitats
The potential impact on coastal habitats would come mainly as a consequence of warming-induced rising sea level (except in those uncommon areas where there is compensating lift in coastal land masses), which could affect (1) the expanse and productivity of coastal wetlands, (2) shoreline habitats, and (3) barrier islands/reefs. Less well studied, but of growing concern, are (1) the effects of increased sea water temperature (other than thermal expansion); (2) disruption of ocean circulation, possibly resulting in alteration of global weather patterns, shifts in distributions of marine plants and animals, and disruption of upwelling sites; and (3) synergistic effects of rising sea level and increased water temperature.
Concern about coastal swamps and marshlands derives from their special ecological value and the fact that they are already under stress from human development, pollution, etc. Wetlands are among the most biologically productive of natural habitats (Intergovernmental Panel on Climate Change, 1990b), and they serve as nurseries for countless marine and terrestrial species. One measure of their commercial significance is that well over half (one estimate is 80 percent) of all fish caught spend part of their life cycles in coastal wetlands (Intergovernmental Panel on Climate Change,
1990b). In addition, a large proportion of migratory water and shore birds feeds in coastal wetlands as they migrate through the United States (Brown et al., 1990).
In the United States it is estimated that there are 18,000 km2 of coastal wetlands (U.S. Office of Technology Assessment, 1984). Most are located in the southeastern United States with about 40 percent in Louisiana (Brown et al., 1990). Where local sea levels rise, current wetlands face the risk of flooding, as periods of low tides no longer expose areas to the air. Areas where there are small tidal ranges are particularly sensitive.
Although wetlands have been maintained despite slowly changing sea levels in the past, there are two factors that may limit their responses to the potential effects of greenhouse gases. First, the average rate of change could be (at the upper range) several times faster than it has been in recent times. In the past, through a gradual process of sedimentation and peat buildup near the shore, the biologically productive area of the wetlands has expanded. However, if sea levels rise rapidly, the newly reached tidal areas will not nearly match the areas that have been flooded. Second, human development may restrict a wetland's inland expansion, if a wetland is bounded by a dike or bulkhead constructed to protect agriculture or manmade structures.
Initial attempts have been made to quantify the potential net loss of wetland areas as a function of sea level rise. The EPA estimates that 30 to 70 percent of current U.S. coastal wetlands would be lost with a 1-m rise in sea level, even disregarding the effects of man-made inland barriers. If such barriers are included, the loss is estimated at 50 to 80 percent (Smith and Tirpak, 1989).
The ultimate question is how the potential diminution of coastal wetlands would affect species survival and human activities. Given the complexity of ecological interactions and the incompleteness of our current understanding of the many qualitative and quantitative relationships involved, it is beyond the capacity of current science to provide specific predictions. Nonetheless, substantial reductions in available habitats are certain to reduce populations of many species and are likely to cause extinctions of some of them.
Rising sea levels would amplify the erosion of shorelines that is already occurring. The continental United States has approximately 51,000 km of shoreline, of which about one-half has been classified as erosional by the U.S. Army Corps of Engineers. Currently, only about 1,000 km or 2 percent is protected either by structural or "soft" means (e.g., beach nourishment) (Intergovernmental Panel on Climate Change, 1990b).
Small increases in sea levels can cause relatively large increments in beach erosion. A number of studies show "multipliers" ranging from 50 to many hundreds (Smith and Tirpak, 1989). Thus, beaches that are 50 m wide may well be severely eroded by a 30- or 60-cm rise in sea level. This could affect species (e.g., sea turtles, shorebirds) that nest on beaches or in adjacent dunes.
Coastal or shoreline ecosystems include not only coastal wetlands (such as the salt marshes and estuaries discussed above) but also marine intertidal and subtidal communities. Some of these ecosystems (e.g., kelp forests, coral reefs, and rocky intertidal areas) are among the most productive communities known (Valiela, 1984). Throughout history humans have relied directly and indirectly on these communities for food, medicines, fertilizers, tools, and supplies (Chapman and Chapman, 1980; Tseng, 1984; Santelices, 1989). Artisanal and commercial fisheries depend upon seaweeds, shellfish, and fish harvested directly from these habitats. In addition, commercial nearshore fisheries are often intimately linked to these shoreline ecosystems, either through the habitats or refuges provided to juvenile or adult stages or through dissolved and particulate organic matter transported from these ecosystems to nearshore waters. Because both developed and developing nations rely heavily on products from these shoreline ecosystems, the responses of these ecosystems to greenhouse warming are of great interest.
The often rich assemblages of marine plants and animals that inhabit intertidal and shallow subtidal regions of the world would undoubtedly be altered by the separate and combined effects of rising sea level, increased air temperatures, and warmer water temperatures, all of which may result from greenhouse warming. The consequences of other, less certain, changes (including alterations in the patterns of ocean currents, upwelling, and frequency and intensity of storms and El Niño/Southern Oscillation events) are potentially important but more difficult to evaluate.
The consequences of sea level rise to shoreline ecosystems would vary according to the type of organism and type of community. Shallow subtidal marine ecosystems, such as kelp beds in temperate regions and coral reefs in tropical waters, would probably not be affected much by changes in sea level. Intertidal coral reefs (those exposed to the air during low tides), on the other hand, could either be not affected or killed completely, depending on the rate of sea level rise relative to the rate of growth of the reef. Other intertidal organisms that are attached to the rock substratum (such as seaweeds, barnacles, and mussels) would likely be able to recruit higher on the shore and keep pace with the projected rates of sea level rise, as long as there is sufficient shoreline available.
Support for these predictions is based on evaluations of the consequences of abrupt uplift or depression of the shore following earthquakes. Although earthquake-caused shifts in shoreline height are known to severely disrupt local shoreline plants and animals for a few years (Haven, 1971; Castilla, 1990), recruitment, migration, and subsequent successional events should result in reestablishment of the communities that existed prior to the earthquake.
Increased air temperatures are hypothesized to result in compression of the intertidal zones throughout the world (Lubchenco et al., 1991). Most shores are characterized by horizontal bands or zones of plants and animals. The width of each zone is determined by the complex interaction of physical and biotic factors acting on each species. The upper limits of each zone are usually determined primarily by abiotic factors such as desiccationspecies do not live higher on the shore because they would dry out (Connell, 1972; Schoenbeck and Norton, 1978). The lower limits of most species appear to be determined by biotic factorsa superior competitor or a predator precludes a species from living lower on the shore (Lubchenco, 1980). These biotic interactions are generally more intense in the physically more benign lower intertidal and subtidal zones. Warmer air temperatures should result in a more intense desiccation regime on the shore. Species whose upper limits are determined primarily by desiccation would be unable to occupy the upper portions of their current ranges. This would result in compression of the zone occupied by each of those species. The extent of this compression would depend on several factors, including the absolute change in air temperature, the rate of change, and the variance in temperatures. Because each zone is composed of numerous species that interact in a complex fashion, it is not clear whether existing zones would all be compressed equally or whether some zones and/or species would be eliminated. There is insufficient information at present to predict the consequences of this hypothesized compression to the loss of biodiversity, habitat, or productivity of these ecosystems.
Rising seas could also affect the natural protection from the seas afforded by offshore barrier islands. Whether such barriers disintegrate as they are overwashed or are repositioned in such a way to continue to protect coastal habitats is uncertain and may depend crucially on local conditions and the actual rate of sea level rise (Intergovernmental Panel on Climate Change, 1990b). If sea level rises faster than coral reefs are able to grow, their role as a natural barrier in tropical and subtropical regions could be compromised.
If ocean temperatures rise, many marine organisms will be confronted with temperatures to which they are not usually exposed. In response to these changes, organisms may migrate, die, adapt, or exhibit no change. Individual organisms will thus respond directly to changes in temperature but will also be affected by changes in other species. For example, the overall response of a species may be strongly influenced by the availability of its prey and interactions with its competitors or with its symbionts. Recent events such as coral reef bleaching and mass mortality of marine populations suggest that even slight increases in temperature may result in severe disruption of ecosystems and loss of habitat diversity, species diversity, and genetic diversity. Particularly susceptible organisms include those currently living close to their thermal maxima, those dependent upon symbiotic interactions with other species, and long-lived species.
Four major widespread coral bleaching and mortality events were reported during the 1980s. These occurred in 1979–1980 and 1982–1983 in the Pacific Ocean and the Caribbean Sea; in 1986–1988 in the Indo-Pacific, Red Sea, and the Caribbean region (including the Flower Garden Banks in the Gulf of Mexico and Bermuda); and in 1989–1990 in the Caribbean (Glynn, 1991). The unprecedented geographic scale and frequency of these bleaching events have attracted considerable attention.
Bleaching results from the loss of symbiotic algae (zooxanthellae) that normally reside inside coral hosts and may supply up to 63 percent of the coral's nutrients (Muscatine et al., 1981). The expulsion of the algae can be triggered by a variety of conditions, including increased or decreased water temperature, increased visible and ultraviolet radiation, desiccation, decreased salinity, high sedimentation, and various pollutants (Glynn, 1991). Although all of these conditions are known to cause stress and result in the expulsion of the algae from corals, the available evidence strongly suggests increased water temperatures as the most likely cause of the recent bleaching events (Glynn, 1991).
The consequences of these bleaching events to coral reef ecosystems are not well known but are undoubtedly related to the intensity and spatial extent of the bleaching. During the unusually strong 1982 to 1983 El Niño/Souther Oscillation (ENSO)5 event, severe bleaching and mass mortality of corals were reported from Costa Rica (50 percent coral mortality), Panama (75 to 85 percent mortality), and the Galapagos Islands (97 percent mortality). The massive mortality in the Galapagos is being followed by severe bioerosion of the coral reefs. Sea urchins, other grazers, and internal boring animals are eroding the dead coral skeleton (Glynn, 1988a,b). If subsequent recruitment of new corals does not occur on a broad scale, coral reefs may disappear from these locations.
Another possible result of increased oceanic temperatures may be an increase in the incidence of diseases of marine organisms. Recent water temperature increases in the Caribbean were correlated with a massive dieoff of the black sea urchin. The mass mortality of black abalone in some California populations was also correlated with increased water temperatures during the 1982–1983 ENSO (Tissot, 1990). Although the causal relationships have not yet been established in these cases, the coincidence of these mass mortality events and increased water temperatures bears further investigation.
Valuable insight into effects of increased water temperatures may be gained by examining the effects of thermal discharge from nuclear power plants on nearby shoreline ecosystems (Lubchenco et al., 1991). Thermal discharge of 4° to 6°C above ambient in the Diablo Canyon Nuclear Power Plant resulted in a dramatic change in the species composition of nearby habitats. Although some of the species changes were predicted (e.g., increases in flora and fauna typical of warmer waters to the south), many others were surprises. The unexpected changes resulted from species influencing one another in addition to the direct responses of each species to the warmer water.
For example, an increase or decrease in the abundance of some species changed the competitive or predator-prey relationships of other species. Because of the complex interactions among species in an ecosystem, these biological interactions must be considered along with the direct effects of changes in abiotic conditions.
Highly mobile marine species would be expected to respond to increases in water temperatures by migrating to cooler waters. Such migration by coastal species could be blocked by natural or man-made barriers, but very little information is available to assess specific impacts on particular species.
Many marine plants and animals disperse in ocean currents. This dispersal may be to short or to very long distances, along the shore or across entire oceans. The boundaries of present-day biogeographic regions attest to the importance of oceanic currents in determining species' boundaries and ranges. Alterations in these currents would likely change patterns of species distribution and diversity, but exact predictions are beyond present understanding.
Sensitivities and Impacts for Ocean Habitats
Temperature, light, and nutrients are the suite of factors limiting productivity in the ocean. A modest rise in sea temperatures could, other things
being equal, have a positive effect for some species and could enhance photosynthesis and the fixing of CO2. However, many species of seaweeds achieve maximal production rates during the coolest months of the year (Mann, 1973). Thus, overall effects of increased ocean water temperature cannot be predicted. Moreover, large changes in water temperature could exceed the tolerance levels of tropical species (Smith and Tirpak, 1989).
More important than this is the possibility of changes in the patterns of ocean upwelling, which brings nutrients toward the surface and thus enhances primary organic production. As noted above, there may be compensating changes in ocean currents and windsintensifying upwellings that are near major land masses and moderating upwellings that are not. Existing information, the IPCC (Intergovernmental Panel on Climate Change, 1990b) judges, does not permit conclusions about which affect might dominate.
Again, the intricate relationships between any species and its physical and ecological setting make specific predictions impossible, as does the lack of knowledge of precise changes in temperature and circulation. Species in the open ocean can adjust to change by moving to more hospitable surroundings, a fact that led the IPCC (Intergovernmental Panel on Climate Change, 1990b) to conclude that the impact on overall biodiversity would be smaller than in coastal communities. However, such shifts in ranges of species could have large impacts (positive or negative) on countries that depend on commercially important marine species in nearby coastal waters.
Although specific predictions are not feasible, the scale and nature of the possible changes can be illustrated by historical experiences. For example, the slight ocean warming that was observed during the first half of this century was associated with the penetration of subtropical fish species into temperature latitudes, but ocean warming in the 1940s and 1950s coincided with both unusually high and low herring populations in two adjacent northern ocean regions (Intergovernmental Panel on Climate Change, 1990b).
During the 1982–1983 ENSO the anchoveta and mackerel catch off the coast of Peru declined, while the catch increased off the coast of Chile (Serra B., 1987).
The impacts of a possible change in the general patterns of ocean circulation cannot be gauged. However, because these patterns profoundly influence ocean habitats and, of course, much more, including large-scale weather and climate far inland, changes in ocean currents are certain to have major effects on distributions and abundances of marine species.
With regard to the possible loss of tidal wetlands, one adaptation to rising sea levels could be to prevent future coastal development activities
that would impede the inland expansion of intertidal ecological communities. Because development, once begun, is difficult to reverse, there is an understandable argument that near-term action is needed to preserve future options.6 The state of Maine has already taken such action (Smith and Tirpak, 1989). Indeed, coastal zone management is a very powerful way to both reduce likely property damage if sea levels rise and preserve intertidal habitats and the species that depend upon them.
With ocean (and surface air) warming, marine species will tend to migrate toward the poles. It may turn out that barriers to such migration exist, threatening some species. The "bridging" strategies specified above for terrestrial ecosystems, such as seed banks, artificial reserves, and corridors for migration, are not well suited to the marine environment. For example, seaweeds do not have seeds. While some marine species could be preserved in aquaria, large natural preserves could not be buffered against temperature change. As for providing corridors, present dispersal and migration routes are not known for most marine species.
Prospects for changing ocean habitatsfor any species or any region are so uncertain that it is difficult to contemplate adaptation measures. Some areas may lose their access to significant commercial species, and others may gain. About all that can be said is that the institutions that affect human use of the oceans' resources (e.g., fishery conventions) must consider future flexibility to be a high virtue as they make their rules and agreements. Much needs to be done to provide such flexibility for current variationflexibility that would have the bonus of helping adapt to global warming.
Marine and coastal ecosystems are potentially quite vulnerable to climate change. A sea level rise of 1 m could cause a loss of 30 to 70 percent of U.S. coastal wetlands. Compression of the intertidal zone that may also accompany rising sea level and increased air temperatures could result in a significant net loss of primary production to nearshore marine ecosystems. The potential effects on ocean temperature and currents are poorly understood at present, but increased incidence of coral bleaching could have disastrous consequences for tropical marine ecosystems. Alteration of ocean currents could result in altered global weather patterns and changes in upwelling intensity and location. Climate, fisheries, biodiversity, and shoreline would all probably be affected. Predictions of the exact consequences of these changes are difficult. At present, the potential for human intervention to ease adaptation in marine ecosystems seems quite limited.
Among the images that monopolize our thinking about climate change and prompt our decisions, pictures of dusty refugee camps in Africa today and cracked fields in Indiana in 1988 rank high. No question about climate change is more crucial than whether humanity will have reliable, potable, and cheap water. "Resources" are "available means," and so "water resources" are the available water or supply, the difference between precipitation and evaporation in the hydrologic cycle. Because the reservoirs, pumps, and pipes of water systems take a long time to put in place and have long lives, no question demands more foresight. Since cheap water must be collected upstream and run down hill, no question demands more local detail.
We warned that conventional wisdom may sometimes mislead, and the image of drought likely does that about water resources. A tabulation of climate changes affecting water resources shows regional averages for both soil moisture and stream runoff to undergo from a -50 to +50 percent change for an equivalent doubling of CO2 (Schneider et al., 1990). The regional differences among scenarios that make predictions of farm yields conflict also make opposing predictions of water resources. So, instead of an image of drought alone, we must be prepared for "some regions [to] benefit from changing precipitation patterns, while others…experience great losses" (Smith and Tirpak, 1989). And losses might be from excess as well as drought.
General warming provides an exception to the uncertainty engendered by differences among regional predictions. The predictions generally agree that all regions will warm. All else being equal, this will melt snow earlier, increase spring floods, and decrease summer flows and the reliability of storage (Gleick, 1987). The assumption that all else will be equal is, of course, unlikely to prove right.
Sensitivity of Water Resources to Climate
Climate affects first the income of precipitation and the expense of evaporation and then the net of runoff and water resources. From year to year precipitation is variable as the record for 1931–1976 in the Colorado watershed shows (Figure 34.6). For the 48 contiguous states, evaporation consumes two-thirds of the precipitation, and in the Colorado watershed it consumes more than eight-tenths. Runoff is generally a small difference between the larger quantities of precipitation and evaporation and hence fluctuates relatively more. Because the three are plotted on the same logarithmic scale in
Figure 34.6, the vertical swings of the lines are the relative, not absolute, variability. The record of runoff varies relatively more than the records of precipitation and evaporation.
Relative variability is a measure of sensitivity. In watersheds from Carolina and Florida to Kansas and Texas, the elasticity of runoff for a change in precipitation is computer to rise from 2 in the humid eastern part to 4 in the western. For a change in evaporation it is 1 to 2. The classic nomogram relating runoff to precipitation shows similar elasticities, although recent analyses show runoff is somewhat less sensitive to temperature changes than the nomogram showed (Langbein et al., 1949; Karl and Riebsame, 1989; Schaake, 1990).
The great variability of runoff evident in Figure 34.6 hinders the estimation of its sensitivity and obscures evidence of any trends. The practical consequences of such variability are illustrated by the Colorado River compact, which was made during a wet period and anticipated more water than the river can deliver over the long term (Frederick and Kneese, 1990). In an analysis of diverse streams across the United States, Matalas (1990) found little evidence of a trend amidst the year-to-year variability.
Beyond the sensitivity of runoff to climate, the impact of a climate change depends on whether supply, use and consumption of water are closely matched. The water withdrawn from streams and wells is said to be ''used," and part of the water withdrawn is evaporated, consumed by humans or livestock, or incorporated into products and said to be "consumed" (Solley et al., 1989). The averages for the 48 contiguous states could lull and mislead us because only 24 percent of the supply is withdrawn for use and only 7 percent of the supply is consumed.7
Individual water resource regions show the pinch between supply and use. The warning signals in Table 34.4 indicate where consumption, storage, variability, or use of groundwater or hydroelectricity suggest vulnerability. They actually show where supply and use are matched, so that a change would bring significant harm or benefit, depending on which way the climate changed. A climate change would make water resource matters considerably worse or better in the Great Basin, the Missouri, and the California water regions, especially.
The vulnerability of activities can also be appraised. These would surely show the impact of climate change on irrigation. It is most used where precipitation is light, evaporation is high, and the supply is small. It accounts for about half the withdrawals of fresh water and about four-fifths of the consumption. It waters only about an eighth of the crop acreage but produces more than a quarter of the market value. Still, it requires cheap water, and other users already take water from irrigation. So the impact, good or ill, of climate change on irrigation would surely be great. The computed effect on the Eastern states of a warmer and drier climate is an expansion of irrigation. In the Western states, however, a 3°C warming and 10 percent more precipitation would cut irrigated acreage by 15 percent. A 3°C warming and 10 percent less precipitation would cut it by 31 percent (Peterson and Keller, 1990). The impact of less irrigation in the West can be judged by the value of tomatoes from an average acre in California, which was $6,000 to $8,000 in 1989.
Before considering adaptations, one must remember that there are impacts on water supply other than climate change, such as those caused by human use. An example is the depletion of the Colorado River. The average annual depletions in the Upper Colorado River plus diversions and evaporation from reservoirs grew by about a third during each decade from 1952 to 1981, until they reached roughly a quarter of the undepleted flow in 1972–1981 (Kneese and Bonem, 1986). The impact of any climate change will be mixed with the impacts of other factors such as these depletions.
TABLE 34.4 Vulnerability of Water Resource Regions of the United States to Climate Change
How Water can be Managed
When rain regularly waters crops and the landscape, runs into streams, and fills wells, we need adapt and manage little. The variability illustrated in Figure 34.6 means humanity must usually manage water to get a reliable supply. As the population in a region grows and uses more water, the first signal that they must manage water may be during a drought. The signal
could, of course, come during a flood. The subject here, however, is water supply. The signal of a drought would be to begin managing water by controlling use of increasing supply.
If water on the surface is used, the sharing between users immediately becomes an issue. Users upstream can capture the water first. Unless everyone accepts a policy for allocating it, trouble may follow. In the humid Eastern United States, the riparian doctrine controls the use of water. Under it an owner of land adjoining a stream has the right to divert as much as he or she wants provided there is no "unreasonable" impact on others downstream. Although what is reasonable is decided by the courts, case by case, the number of cases has been small.
In the dry West the riparian doctrine breaks down because much more water is needed than is available. Determining "reasonability" case by case is unworkable. So most Western states have adopted a system of water rights called the "prior appropriation doctrine." Generally, it gives the right to prevent a newer user upstream from interfering with prior ''beneficial" uses. Beneficial is broadly defined.
Although these two allocation policies control the use of water, they have little to do with promoting economically efficient allocation, welfare, or environment. They allocate water without violence, promoting domestic tranquility.
To correct these shortcomings, people put forward new policies and even use some. Environmental regulations protect fish or stop pollution. Most do not balance benefits to the environment against other benefits but preempt water to maintain the environment.
Welfare is promoted by conservation measures that limit the right to waste water. Typically, they deal with toilets or showers. Or in an emergency they restrict sprinkling. Prices do affect the use of water (Frederick and Kneese, 1990), and conservation rates penalize using more than is needed for drinking, cooking, cleaning, and sanitation.
Economic efficiency is the goal of water markets. They are much talked about in the West. New users pay old ones to transfer their use to new ones. The transfers are usually limited to ones with no adverse effect on other existing users. When the transfers are between nearby and similar users, they go reasonably well. The lack of adverse effects of wider transfers, however, is often hard to prove, they find their way to court, and they become expensive and slow.
Forecasts of water supply make the control of use more efficient. From generating more hydropower and irrigating more acres to protection against flood, knowing how much water will be available increases its efficient use
(Glantz, 1982). Although all the policies can be used with forecasts to improve efficiency, the policies are usually imposed by different agencies and levels of government, and they are rarely coordinated.
The supply of water in a place might be made more reliable by long transport between basins, management of the watershed, or even cloud seeding. Generally, however, the supply is made more reliable by storing water in reservoirs, pumping water stored in aquifers, or reclaiming salt and waste water.
The reservoirs on the Colorado River can store water for years, but most reservoirs have only small volumes to smooth the supply for weeks or months. Many aquifers, on the other hand, are large enough to smooth the supply for longer times. The present supply can even be increased by mining ancient or fossil water.
Although salt and waste water can be reclaimed, it is costly in money and energy and only practical for valuable uses in dry places. The cost of desalinization may, of course, fall during a half century (Abelson, 1991). Some benchmark or exemplary costs are shown in Table 34.5. Note that the estimates of the cost of water from the Delaware, Platte, and Sacramento rivers do not include treatment and are only estimates for the present supply. The estimate from the Hudson River includes delivery.
Just as supply can be smoothed by storage that diversifies it over time, it can be smoothed by the diversity of joint operation of water supplies. This has been demonstrated for places as far apart as the Potomac and North Platte rivers and Houston, Texas. Despite its economy and promise, joint operation of independent or even federal systems is rare. The reasons include independent spirits and lack of credible evaluations of the benefits (Sheer, 1985).
How Water has been Managed
Water management has become a public, not private, concern because of the wide disruption of floods and droughts, the ubiquitous need for water for health and fighting fires, the need for political support and exercise of eminent domain to acquire strategic sites, and the economies of scale and natural monopoly. Generally, the demand for water grows annually by 1 to 3 percent, until a flood or drought impels governmental action, which increases supply by a leap that delays the need for additional action for years. Although a slowly growing need might have been known for years in advance, a flood or drought causes action. Often, a long-lived canal, dam, or levee is built.
TABLE 34.5 Costs of Present Water and for Securing Reliable Alternative Supplies
The safe yield of a reservoir is figured from the worst drought on record. Using it makes the optimistic assumption that the operator knows when a drought will end. When a new record is set by a worse drought, the safe yield is revised downward, which encourages action after extreme weather. All the forces described in the preceding paragraphs incline officials to view new facilities as insurance against disruptions and restrictions rather than as investments. Recently, growing environmental awareness seems to have made disruptions and restrictions more acceptable.
Climate change could change the timing of the leaps in strategy or means to adapt to the gradually changing demand for water. Action is generally precipitated by the extremes of flood and drought, and climate change would change the probabilities of these extremes and simply change the frequency of actions to adapt. The response of management will tend to lag behind the impacts of climate change. Unless, however, those changes are adverse and rapid relative to changes in the population-driven demands for water management, the overall impact of climate change is unlikely to be substantially more serious than that of the vagaries of the current climate.
Essentially, the chance of climate change introduces more uncertainty into the already uncertain realm of water management. The better prepared we are to deal with the extremes possible in the present climate, the easier it will be to deal with changes. Climate change makes the following more urgent:
• Improve management of present systems to deal with drought and flood.
• Develop and test methods to deal with more severe ones.
• When investing in new facilities, consider climate change by such a method as described in Chapter 33, the section "Making Decisions in an Uncertain World."
• Monitor to discover any trends.
• Learn the sensitivity and then the direction and size of the impacts of climate change.
Industry and Energy
Sensitivity and Income are not Congruent
Passing from the farms, forests, water, and natural land and sea scapes into the more man-made things broadly called industry, one encounters a paradox. The former outdoor things get the attention when climate change is considered, but the latter is where most of the money is.
In the national appraisal of effects by the EPA (Smith and Tirpak, 1989), the impacts on agriculture occupy 28 pages and the impacts on forests 21 pages. The only industrial subject, electricity, occupies 11 pages. The contributions to national income, however, of farming and forestry are about 3 percent of the total of all activities (see Table 34.6). The reason for the paradox is the sensitivity of the sectors. Nordhaus (1991) classified farms, forestry, and fisheries as "potentially severely impacted" but all others as "moderate potential impact" or "negligible effect" (see Table 34.6). His
TABLE 34.6 Economic Activities According to Their Sensitivities to Climate Change
estimate, in 1981 dollars, of the impact on farming8 of -$10.6 billion to $9.7 billion far exceeds the second-place estimate of -$2.8 billion for protection of open coasts. It also far exceeds the -$1.7 billion for electricity demand and the $1.2 billion for nonelectric space heat. Completeness requires examining the subjects called industry and electricity here because they make up about 97 percent of the U.S. national income; but we know they are less exposed to climate change.
Sensitivity of Electric Power Generation
According to Jäger (1985), "Relatively few studies have been made of the impact of climate change on energy demand." The Energy Department found that "the state of knowledge regarding the sensitivity of energy systems to climate change is very limited" (U.S. Department of Energy, 1989b). IPCC Working Group II began its summary of energy with threats to biomass and fuelwood rather than with electric power (Intergovernmental Panel on Climate Change, 1990a). Nevertheless, the most directly analyzable impact of warming is on the requirements for electricity for warming and cooling. The above IPCC report cites studies in six nations, and it states that results differ depending on how much energy is related to heating versus cooling.
An American study (Linderer, 1988) estimates the climate change impacts for aggregations of utility systems in the Great Lakes, the Southeast, the Southern Great Plains, California, and the United States as a whole. The most significant result is that for the United States as a whole new capacity requirements deriving from increased temperatures associated with climate change for the year 2066 range from an increase of 12 to 22 percent above the new capacity requirements that electric utilities would face arising from GNP growth and changes in energy intensity.
Whereas this additional capacity is a relatively modest fraction of the total growth in generating capacity that would be needed in any case, the absolute magnitudes involved are significant by many standards. The increase in new generating capacity from climate change would be between 200 and 400 GW and would require utilities to make capital investments of roughly an additional $200 billion to $400 billion in generating capacity from climate change alone. Not surprisingly, the largest increase in additional generating capacity from climate change occurs in the Southeast, where utilities already size their generating capacity based on summer cooling requirements. For the Southeast, utilities would have to add, between now and the year 2055, an additional 35 percent of total generating capacity above and beyond the incremental capacity that would be required because of other factors.
As the IPCC (1990b) Working Group II report states, "By changing
water resource availability, climate change may make some present hydroelectric power facilities obsolete and future energy planning more troublesome, although others may benefit from increased runoff." For the United States, Miller (1990) states that "electric power generation makes greater withdrawals of instream water than any other industry." However, "only a small portion of the water circulating through electric generation plants is actually consumed. Nearly all the currently operating steam power plants in the United States use a water-based cooling system. It is estimated that as of 1980, U.S. withdrawals of fresh water for thermo-electric cooling were equivalent to withdrawals for irrigated agriculture, but while 55 percent of irrigation withdrawals were consumptively used, the rate of the consumptive use of thermo-electric cooling was only 2 percent" (Miller, 1990). Electric utilities in the arid sections of the United States have adapted to water scarcity by adopting closed-cooling systems.
The mixture of fuel used for generating electricity would change sensitivity to climate change. First, restrictions on carbon emissions to mitigate the greenhouse effect would affect, especially, the burning of coal but not nuclear plants. Then, a climate change could help or hinder generation from water, wind, or solar energy. Finally, a change in the growth of plants would change the amount of biomass for burning.
The quantitative results support the judgment of modest sensitivity of electric power generation to climate change.
The large income but low sensitivity of industry tempers the statements that the greenhouse effect would disrupt national economies. Examples of such statements are, "This greenhouse effect may by early next century have increased average global temperatures enough to shift agricultural production areas, raise sea levels to flood coastal cities, and disrupt national economies.… A rise [in sea level] in the upper part of this range (25–140 cm) would inundate low lying coastal cities and agricultural areas, and many countries could expect their economic, social and political structures to be severely disrupted" (World Commission on Environment and Development, 1987).
The effects associated with a global cooling of roughly 5°C experienced during the last ice age 18,000 years ago are documented. They would have had major impact on the world's economy as it exists today. In particular, much of the industrial heartland of this country (the upper Midwest) would have been under an ice sheet. None of the effects in our base scenario (a doubling of CO2) would make large land areas uninhabitable, and thus the
impact would be fundamentally different (and much less severe, as indicated below).
The argument that the impact on the industrial sector would be relatively modest from the climate change in our scenario is put forward convincingly by Schelling (1983). Schelling says that "it is likely that most of the identifiable changes in welfare due to climate change would be, for most parts of the world, swamped by other uncertainties." Schelling goes on to say that one can argue, although not conclusively, that "if the change is slow, the adaptations and replacements, even the migrations, need not be traumatic or even especially noticeable against the ordinary trends of obsolescence, movement and change. The issue is suddenness and unexpectedness." He cautions that "it is wise to be concerned about any prospective change in some major index of climate."
Schelling says further "the only readily identified potential impact of significant magnitude on future living standards is on agriculture.… A fair guess seems to be that any likely rate of change of climate due to CO2 over the coming century would reduce per capita global Gross National Product by a few percentage points below what it would otherwise be. A curve of world per capita income plotted over time would be set back probably less than half a decade." Later, Schelling says "the pure temperature feedback on the use of energy, both as a cost saving (heating) and as an additional cost (cooling) and as a consequent damper or booster to fuel consumption, is of obvious relevance. Such estimates as there are do not indicate that any overall reduction or increase in energy use, due solely to temperature change, would be of major significance, whichever way the net effect goes." An estimate of change in "degree days of heating and cooling yield a result that is not an impressive fraction of the current cost of heating and cooling." Finally, Schelling says that "the most likely possibility emerging from the work done so far in relation to CO2 is that the impact of climate change on global income and production, and specifically the agriculture component of it, would not be of alarming magnitude."
Barbier (1989) states that the economic impact of any greenhouse effect will most likely be in terms of the rising cost of agricultural displacement and adaptation in the face of climatic instability. He places the impact on agriculture and water supply as a second-order impact and lists the economy (nonagriculture) as a third-order impact.
The one industrial sector that could possibly be an exception to this general conclusion is the forest products industry, since it depends directly on climate, as agriculture does. For this industry the rate of change in climate associated with the greenhouse effect could perhaps exceed the rate at which natural forests could evolve and migrate (U.S. Department of Energy, 1989b). However, analysis by Binkley and Dykstra (1987) indicates that the temperature increase associated with doubling of CO2 would increase
exploitable forest area and forest growth rates and yield a positive economic benefit (precipitation changes and CO2 fertilization were not addressed). This subject is discussed above in the section "Forests and Grasslands."
Even though there are references to economic disaster resulting from climate change, plausible arguments have been put forth that the impact will be relatively modest.
Relative Economic Power
The large income from industry makes this a logical place to examine the impact of climate change on relative national economic power. Both Schelling (1983) and Nordhaus (1991) conclude that the impact of climate change on global income and production would not be of alarming magnitude. Nordhaus specifically estimates that the upper limit on damage would be around 2 percent of total output (about 1 year's economic growth).
The panel searched for relative impacts on five major regions of the world: the United States, Western Europe, the former Soviet Union, China, and Japan. For agriculture a paper by Tobey et al. (1990) goes beyond change in yields to estimate changes in world prices for agricultural commodities and from that derives estimates of changes in economic value (producer and consumer surpluses) in the various regions of the world. Tobey et al. (1990) developed two cases to bound the likely impact on welfare from a climate change roughly equivalent to our scenario. In the optimistic case, none of the five key regions experience significant economic value changes. The largest change in the optimistic case, in fact, is an increase for the former Soviet Union of 0.3 percent of 1986 gross domestic product (GDP). In the pessimistic scenario, worldwide economic value from agricultural changes decreases by 0.5 percent of 1986 GDP. The changes in all the major regions of the world of concern are slightly less than the worldwide total, with the exception of China, which would experience a 5 percent decrease as a percent of 1986 GDP.
For natural resources the most obvious potential change is in access to Arctic oil reserves. The potentially recoverable reserves from the Arctic region are between 17 billion and 55 billion barrels for the United States and between 50 billion and 80 billion barrels for the former Soviet Union. If all these reserves were proven and recovered, they would extend the life of the oil reserves for the United States by roughly 10 years and the oil reserves for the former Soviet Union by roughly 20 years. Of course, these increases in reserves are small compared to the total proved oil resources in the world of roughly 700 billion barrels, of which about 500 billion are in the Middle East. Specifically, if the additional recoverable Arctic reserves from Canada and Norway are added to those of the United States and the
former Soviet Union, the increase in the proved reserves for the world would be only about 20 percent (Gleick, 1989).
Our review of the few reports and judgments strongly suggests that the climate change envisioned cannot affect relative economic power significantly.
While accepting the logical arguments about the envisioned climate change, the panel also accepts that surprises do happen. These could arise from nonlinear responses like the passage of a threshold. Two plausible ones are reorganization of the ocean-atmosphere system and changed runoff of water.
Significant reorganizations of the ocean-atmospheric system seem to be the key events that have triggered the advance and retreat of ice sheets for the previous ice ages (Broecker and Denton, 1990). Because our current geophysical fluid dynamics models cannot predict these kinds of nonlinearities (except when they already know what to look for), we need to determine on some scientific basis the confidence levels that current model predictions merit.
The second surprise could be changes in runoff following changes in precipitation. This was discussed in the section "Sensitivity of Water Resources to Climate," above. Even for present predictions from the models, regional variations could be strong for precipitation and the resulting water runoff. If the general distribution and flow of water in a region changed dramatically, the impact could be much more significant than we are predicting for agriculture, industrial distribution, and even the habitability of certain urban areas. Water is cheap, where it is generally available, but where water is not available, it is hard, if not impossible, to provide the quantities needed because of the high cost of its transportation.
In many places in this report the reader encounters the conclusion that if investments have shorter lives than climates adaptation is fairly easy. In Chapter 33, the section "The Tools of Innovation," we showed that many investments do have short lifetimes. The hard adaptations will come in those exceptional cases of long lifetimes.
One example of capability to adapt comes from electric power generation. According to Miller (1990), "It is estimated that by the end of the century, 30 years will be the average age of coal-fired power plants in operation in the United States, and it appears that the average useful service life of power plants may be lengthening as rising real construction costs
induce power companies to refurbish rather than abandon older plants.'' Thus, utilities face investment decisions concerning their cooling system technology with perhaps a 30- to 40-year lead time. As a result, they will have to factor into such decisions the uncertainty associated with the availability of water for cooling. Technical flexibility can be incorporated in the design of single power plants to hedge against uncertainty in the availability of water. Thus, to optimize decision making, utility executives need to have the best available information about potential changes in the distribution of water availability and the uncertainty in those estimates for a 30- to 40-year period. The adoption during a half century of an efficiency like superconductivity of transmission lines would, of course, sharply decrease demand for electricity, and such possibilities add to the uncertainty.
Another example of the ability of industry to adapt to climate warming, and that the adaptation required is relatively modest, was reported in The New York Times on December 20, 1989. Royal Dutch Shell spent a year assessing the impact of scenarios involving significant global warming as one of its major strategic planning initiatives. As a result of this work, Shell has reviewed its investment decisions that have time horizons that place the useful life of the investment well into the period of uncertainty about global climate change. Perhaps Shell's longest-lived investment is its gas platforms in the North Sea, where the reserves are so large that production from such platforms may go on for up to 70 years. Traditionally, Shell engineers have sized the platforms a standard 30 m above the water level, which is the height now thought necessary to stay above the waves that come in a once-in-a-century storm. As a result of reviewing possible sea level change from global climate change, Shell engineers have increased the height of the gas platforms to 31 or 32 m above the ocean. A 1-m increase will cost an additional $16 million and a 2-m increase, will be roughly double that. Even the higher cost, however, is only about 1 percent of the total of the platform, which costs a few billion dollars.
Another piece of evidence that sheds light on the adaptability of the industrial sector comes from a U.S. Department of Energy (1989a) report entitled A Compendium of Options for Government Policy to Encourage Private Sector Responses to Potential Climate Change. The report deals primarily with mitigation but does have some useful information for assessing adaptive capabilities. In particular, the report has some data on the lifetimes of industries' capital equipment. This report says that the electric utility sector has capital investments with lifetimes of 30 to 40 years. It goes on to say that there are other industries where large infrastructure investments are required, such as steel and paper, that have similar lifetimes of up to 40 years. A boiler, for instance, lasts about 40 years. Some papermaking machines are now 70 to 80 years old, although the technology in the United States is well behind that in Europe, and European machines
are shorter lived because of technological evolution. Such change is likely to come to the paper industry in the United States. In any case, evidence about the lifetime of capital investments in the industrial sector is critical for analyzing the ability of the industrial sector to adapt to climate change.
The evidence to date suggests that the industrial sector can adapt to the changes in global climate by making rather modest changes to the investment plans they would make without climate change. This conclusion seems to hold more clearly for adjustments to temperature change and sea level. As mentioned above, more careful work needs to be done on water runoff and how it may affect the availability of water for industrial use and distribution. The major conclusion, however, of a rather modest impact on the industrial sector does need to be confirmed through careful comparison, industry sector by industry sector, of the investment time horizons and those covered by the models for global climate change. Perhaps most important is a projection of ranges within which industrial managers could assume, with very high confidence, that they would have to operate.
Fortunately, industrial sectors, including electric power generation, produce much income but are only moderately or negligibly sensitive to climate change. Even though some writers have mentioned economic disasters that might follow from climate change, plausible arguments have been put forth that the impact will be relatively modest. Judgments suggest that the climate change envisioned will not affect relative economic power significantly. Generally, industry can adapt.
Tourism and Recreation
The grand tour that families take one summer to, for example, camp in Yellowstone and the Grand Tetons exemplifies the importance of recreation and how it depends on the climate. The one-third billion visits to national parks and two-thirds billion visits to state parks and recreation areas far exceed, for example, the 48 million attendance at major league baseball games or 36 million at college football.9 The dependence of tourism and recreation on water is exemplified by swimming and fishing. The section "Water Resources," above, addresses the relationship of water resources to climate.10 One-third of Americans swim and one-fifth fish. Expenditures on outdoor recreation are substantial. The $55 billion spent in 1985 on wildlife-associated recreation far exceeded the $14 billion spent for books and the $7 billion for new color TVs (U.S. Bureau of the Census, 1987).
Nevertheless, the issue of economic impact and adaptation for tourism and recreation is relatively insignificant since this sector represents about 1
percent of our total GNP. The part of this industry that is likely to be affected by climate change, is of course, the part that is closely associated with nature in one way or another. For this part the simple answer is that the industry will migrate to the new areas of nature that are attractive for tourism and recreation. The assets associated with this industry are primarily roads and lodging places. The useful lifetimes of these investments are on the order of 20 to 30 years. Thus, this industry can wait and see the impact of climate change and then migrate as the attractive areas move. Although such investments may not correspond exactly to the value tourists and sports enthusiasts may attach to particular opportunities, they do give us a first approximation of the likely economic impacts of greenhouse warming.
Skiing is exemplary in two ways. First, it is sensitive to climate. Second, just as popular taste has raised the ski industry during the past half century, a decline in popularity could ruin it in the decades to come, regardless of climate change. A study of recreation in the Great Lakes Basin used a scenario of climate change that shortened the skiing season by a third. Ontario suffered a $50 million loss. This loss would be countered, however, by a camping season about 40 days longer (Wall, 1989). Obviously, this has important consequences for the economies of specific locales. For the nation as a whole, however, such changes are likely to balance out. This probably applies to noneconomic aspects of tourism and recreation as well.
Settlements and Coastal Structures
Human settlements, including highly developed ones, exist in an amazing range of climatological conditions. Compared to this range, the potential climate change over the next century is small. Thus, it is unlikely that even when climate change exacerbates an already extreme situation (e.g., warming of the currently warmest spot on the planet) the resultant climate would render currently settled areas uninhabitable.
The pertinent questions, then, relate to the extent to which the costs of adapting existing settlements to the new conditions and of maintaining the settlements under the new conditions lead to abandonment or to important modification of the lifestyle or standard of living of the inhabitants.
Direct climate changes of importance to human settlements are primarily changes in the extremes and seasonal averages of temperature, and changes in the spatial and temporal distribution of rainfall. Though the impacts of these effects may be of importance in some cases, the secondary effects of
climate change on the level of water bodies are of much greater significance. Predicted sea level rise will result in a number of major impacts on coastal regions, while the opposite effectdropping of the level of the Great Lakes due to changes in rainfall patternswill have different impact on Great Lakes municipalities and facilities. Inland, increased variability may cause the most significant impacts.
Sea Level Rise
Major effects of sea level rise are inundation of some areas, shoreline erosion, storm damage, saline intrusion into groundwater and surface water, and groundwater table elevation. Inundation, erosion, and storms all lead directly to property damage, the extent of which depends upon the value of the property, its nearness to the shoreline, and the slope of the shoreline. Although the concentration of Americans along the coast makes these effects important to many, an estimate of the number affected by the wide range of a rise of 0 to 60 cm would lend a misleading sense of precision. Estimates of the inundation of considerable portions of, for example, Egypt and Bangladesh are for scenarios of a 1- to 5-m rise (Intergovernmental Panel on Climate Change, 1990b).
Harbor facilities, bridges, and certain recreational facilities, though suffering no physical damage from sea level rise, will nonetheless be impaired or rendered useless because of their dependence on design distances to the water level.
Sea level rise will cause other water-related difficulties in urban areas besides those that result from inundation, high water level, and erosion. The associated rise in the groundwater table may reduce the ability of the soil to assimilate rainfall and result in larger surface runoff. There will likely be increased seepage into basements, adverse effects on building foundations, and increased corrosion of buried pipelines.
In occasional instances, groundwater may intrude into landfills and waste disposal sites, resulting in leaching of hazardous materials.
A rise in sea level will also change tidal (Gleick and Maurer, 1990) and other currents, causing new patterns of scour and silt deposits in harbor facilities. Ocean outfalls of urban drainage systems and waste treatment facilities may no longer drain properly due to elevation and current changes.
The major impact of temperature rise is probably increased electrical demand for cooling in the warmer summers. Though total electric use may
not increase significantly (winter use may decrease), peak usage normally occurs in summer, and thus installed generating capacity will have to be increased to accommodate these greater peaks. Compared to the expected baseline no-climate-change required capacity increase, the added requirements are from 20 to 30 percent more in Miami, to 10 to 20 percent in New York City and San Francisco, and 10 percent in Cleveland and most of the Great Plains (Linder et al., 1987; Smith and Tirpak, 1989; Walker et al., 1989).
In some municipalities, warmer temperatures may exacerbate air pollution problems, particularly ozone concentrations.
Walker et al. (1989) imply that problems such as buckling of rails due to heat expansion will become much more common. Asphalt pavement softening may become more frequent in regions where it has been rare before. Some compensation would, of course, come in the form of fewer problems with frost.
Changes in Precipitation Patterns
Flooding difficulties are not limited to communities affected by sea level rise. Global circulation models indicate that in global warming in some communities the month with the highest rainfall could have 50 percent more rainfall than currently. If this additional rainfall is in the form of greater storms, drainage problems in a number of inland cities will require the same sorts of solutions as those in coastal cities (Titus et al., 1987). If, on the other hand, the additional rainfall is more evenly distributed in time, it may result in increased groundwater table levels with associated basement seepage, foundation damage, and subsidence (Walker et al., 1989).
Decreased rainfall, on the other hand, will result in a different set of impacts (in addition to water supply problems, treated elsewhere). Inland rivers used for navigation may no longer be navigable at all or may have reduced capacities. Water pollution problems may be aggravated by the combination of reduced flow and increased water temperature.
Great Lakes Impacts
Urban areas along the Great Lakes constitute a special case as a result of the prediction that levels in the lakes will decrease. Fluctuations in the level of the lakes is, of course, not new. As with sea level rise, this change will adversely affect recreational and harbor facilities that depend on a fixed range of distances from the water level. In addition, sewer and drainage outfalls may no longer have adequate submergence. Some wooden facilities that have been submerged will become vulnerable to dry rot.
Adaptation to Sea Level Rise
The choice facing coastal settlements is between abandoning areas subject to inundation and storm damage and protecting the land or infrastructure from the impacts or modifying it to withstand the impacts. Retreat, which may be economical in areas that are extremely difficult to protect or in low-valued regions, may require the greatest anticipation because it requires zoning measures to prevent growth and development of replacements for abandoned facilities, structures, and land (National Research Council, 1987). Protection by means of levees, dikes, and similar means, though costly, may usually be done in a matter of a few years. Other forms of protection, by modifying facilities to withstand or resist flooding and other impacts, are done more economically if planned in advance, so that the necessary modifications can be designed into new structures over a period of time (Titus and Barth, 1984; National Research Council, 1987).
Retreat and abandonment are not likely to be economical alternatives for major parts of urban areas. Some isolated facilities, particularly those that are obsolete and/or in disrepair, may be abandoned. But in most cases the replacement value of urban areas will make protection or adaptation of one sort of another, though expensive, the best alternative.
Protection against shoreline erosion can be achieved in two fundamental ways. Either the waves that cause the erosion must be intercepted before reaching the surf zone (breakwaters) or the shore profile must be armored or strengthened. Control may also be exerted in some cases, where sand transport along the shore is significant, by reducing the ability of waves to carry sand (groins and other structures) and by increasing the amount of sand available for transport (beach nourishment) (Sorensen et al., 1984).
Protection against inundation is provided by constructing a dike system or by filling in the area and surrounding the fill with a retaining structure. Many areas that already have dike and levee protection will be required to raise these structures, and new structures will be necessary to protect most coastal cities (Sorensen et al., 1984; National Research Council, 1987).
Cost estimates for these alternatives must be determined on a case-by-case basis because both the natural and the economic characteristics of regions are so widely different (National Research Council, 1983). A few estimates, however, can serve to provide a rough cost range. The Committee on Engineering Implications of Changes in Relative Sea Level estimated the annual costs of beach nourishment on the east coast of Florida in a sea level rise situation to be $13 to $82 per foot, amounting to 0.1 to 3.4 percent of the value of beachfront property (National Research Council,
1987). Gibbs (1986) estimated the cost of beach nourishment at Sullivan's Island, Charleston, South Carolina, to be $121 million. Removing $15 million from this estimate for the cost of stabilizing the backside of the island, the remaining cost is approximately $4,000 per foot. Using Gibbs's 3 percent discount rate for an indefinitely long period, this is equivalent to an annual cost of $120 per foot.
Goemans (1986) has estimated the cost in the Netherlands of adapting to a 1-m sea level rise to be $4.4 billion. Of this, 60 percent is for dikes and dunes. Over the 1,000 km of Dutch dikes, this amounts to about $1,000 per foot of existing dike. Gibbs (1986) suggested that the cost of seawall construction to protect Charleston, South Carolina, would be approximately $2,000 per foot of length, amounting to approximately 6 percent of the value of the protected property.
Titus and Greene (1989) synthesized the results of three different studies to estimate the nationwide costs of holding back the sea for various sea level rises. Using a set of plausible and believable assumptions, they calculated the cumulative costs of protecting barrier islands and developed mainland through the year 2100 to be $32 billion to $309 billion (1986 dollars) for sea level rises from 50 to 200 cm. For a 50-cm rise they provided a 95 percent confidence interval of costs from $32 billion to $43 billion.
Gleick and Maurer (1990) evaluated the costs of protecting San Francisco Bay from a 1-m sea level rise, which is more than our assumption. They estimated that building levees, seawalls, and other protective structures to protect the existing infrastructure would cost approximately $1 billion, plus an additional $100 million per year for maintenance. These costs excluded any costs of protecting natural ecosystems and wetlands.
Facilities that depend on fixed distances from water level will require modification. Docking facilities will need to be raised and/or protected against storm and high-tide damage. Some facilities, such as overhead cranes, will have reduced clearances that will diminish their usefulness (National Research Council, 1987). Such facilities may be raised or may simply not be utilized during periods of high tide.
Many bridges that cross bays, estuaries, and the downstream reaches of rivers draining into the sea will either need to be raised or will block marine navigation during high-tide periods (Walker et al., 1989). Other structures, not designed to permit navigation, may need to be raised or strengthened because they will be insufficiently high to protect them from storm damage.
Ocean outfalls of urban drainage systems and waste treatment facilities may, in some cases, need to be moved to permit discharge and/or otherwise protected to prevent damage (Smith and Tirpak, 1989).
Changes in currents within ports and harbors may increase the need for dredging, even though the mean sea level is higher (National Research Council, 1987).
Changes in groundwater levels combined with higher backwaters in river and drainage channels will require major improvements in urban sewage and drainage systems (Walker et al., 1989). Walker et al. (1989) suggest that in the city of Miami the cost of dikes and levees to protect from direct sea level rise will be relatively small, while the upgrading of storm sewers and drainage will be very expensive. In any city in which these problems occur, pipes and drainage channels will have to be increased in size either by replacement or by the addition of supplemental facilities. Systems that have depended upon gravity may need to shift to forced drainage or, in cases within tidal range, to combinations of locks and flap gates that permit drainage during low tides (Titus et al., 1987). Airports in coastal cities, often on landfill in estuaries or bays, may require special attention in this regard (Walker et al., 1989). Highway underpasses will need increased pumping capacity for drainage (National Research Council, 1987).
The elevated groundwater table can require additional adaptations. Increased seepage into basements may require expanded drainage facilities or improved waterproofing (National Research Council, 1987). Building foundations in some locations may be adversely affected and require stabilization to prevent subsidence. However, Walker et al. (1989) point out that in Dade County, Florida, building foundations will probably not be affected because large buildings already have foundations extending into the water table and homes are built on elevated lots. Streets, however, may need to be raised, resulting in additional drainage difficulties for buildings. Groundwater levels rising above buried pipelines will increase corrosion rates, resulting in more frequent leaks and replacement requirements (National Research Council, 1987).
Special problems may occur where the groundwater table rises into landfills and waste disposal sites (National Research Council, 1987). This could result in leaching of hazardous materials into groundwater if the landfill is not adequately sealed against such possibilities. The cost of remediation in such instances may be quite high.
Adaptations similar to many of those listed above will be necessary in inland cities where increased rainfall results in rises in groundwater levels.
Adaptation to Temperature Increase
In addition to increasing their electrical generation capacity, urban areas will need to make only minor adaptations to temperature change. These may include placing additional expansion joints in rails and increasing maintenance on asphalt roadways. However, northern municipalities may be able to build roads with reduced pavement thickness and may have reduced snow removal costs.
Because of the temperature effect on air pollution, even more stringent emission standards will be necessary for automotive exhausts and possibly for stationary combustion plants as well (Smith and Tirpak, 1989).
When to Adapt
Most of the anticipated impacts of global climate change on human settlements are on man-made facilities with relatively long lives (from 20 to 100 years). If these facilities are to be adapted rather than protected or abandoned, it will sometimes be economical to undertake the adaptation at the beginning of the life cycle rather than to retrofit later. One example of the difference in costs between initial design and retrofit is given by LaRoche and Webb (1987), who estimated the increased cost of an anticipated overhaul of a Charleston, South Carolina, drainage system if an 11-in. rise in sea level were included in the design and also the cost of accommodating an 11-in. rise by retrofit if the overhaul were first done for the current sea level. The added cost to accommodate now for the sea level rise was $260,000, 5 percent of the cost of the overhaul; the cost of the later retrofit was $2.4 million. However, anticipatory modification is not always cost effective. Waddell and Blaylock (1987) in a similar study of Ft. Walton Beach, Florida, concluded that the best way to combat sea level rise in this community was by floodproofing the houses. In this case, designing for future rise provided no significant savings over upgrading the floodproofing when the rise occurs.
As with adaptation, abandonment may be most economical when anticipated well in advance. This permits adequate time to provide economic facility replacement and prevents needless upkeep and modernization expenditures on facilities prior to the abandonment. On the other hand, the construction of new facilities, both protective (dikes, levees, etc.) and capacity-increasing (added power generation), can usually be done economically in a relatively short time and therefore can be delayed until needed.
As information increases, the probability and likely extent of climate change will become better understood. In the interim, as suggested by Titus et al. (1987), public officials must frequently consider whether to include allowances for climate change in current designs. They must view this decision as choosing between leaving future generations with a risk of significant damage or costly retrofits or leaving them with a certainty of added debt repayment for the anticipatory design. These issues are of principal importance when considering long-lived facilities that are difficult or expensive
to protect; in such cases, consideration of anticipatory modification of the facility design is warranted.
Humans have successfully adjusted to diverse climates. In the course of human evolution, people have advanced into every climatic zone between the equator and the poles and established permanent dwellings in most. Nevertheless, people have evolved with characteristics related to the climate where they have lived for a long time. For example, the short, heavy-set Eskimo is fitted to life in the Arctic, while the long-legged Masai is fitted to life in the savannah. The dense dark pigmentation of humans at low latitudes and its progressive decrease at high latitudes is another example of genetic adaptation to environment.
Birth rates are of the order of 20 per 1,000 population per year, and they are higher in many tropical climates than in some temperate ones. Death rates are of the order of 10 per 1,000 and are not correlated with climate (U.S. Bureau of the Census, 1987, Table 1379).
Besides long-term genetic adaptation, studies show humans are highly adaptable on shorter time scales as well. people routinely adapt to seasonal weather variations. Cold acclimatization provides more tolerance for low temperatures in the winter, while heat acclimatization provides more tolerance for hot summer temperatures. Some researchers suggest that people who move from a cool to a subtropical climate often adapt within 2 weeks (Rotton, 1983).
There are, however, limits to human adaptability. All permanent human settlements are in climates where the maximum temperature is 55°C and the minimum temperature is -60°C. These seem to be the limits of human tolerance. Over two-thirds of the world's settlements are in climates where the temperature ranges from a minimum of -5°C to -15°C to a maximum of 30° to 38°C. This is considered the mean comfort zone within which adaptation is easiest (Weihe, 1979).
The principal factor of climate that is biologically influential is temperature. Other factors mainly have importance with respect to temperature. Water vapor pressure gains in influence above 25°C and can be ignored below 10°C, while relative humidity is important at all temperatures. Precipitation is influential through cooling of the air and saturating it with humidity. Wind enforces convection, with a positive effect in the heat and a negative effect in the cold. Several attempts have been made to combine two or more meteorological factors as indices of physical comfort and predictors of habitability (Weihe, 1979).
Climate as a Direct Cause of Disease
Human health would be affected by any climate change because people are directly sensitive to climate and susceptible to diseases whose vectors are sensitive (Weihe, 1979; Kalkstein, 1989; Intergovernmental Panel on Climate Change, 1990a). For all direct impacts of climate, two factors are important: (1) the relative and absolute strength of the stimulus and (2) the exposure time. Injuries and maladjustments caused by climate are summarized in Table 34.7. Probably more important than injury by exposure in a warmer climate are maladjustments from insufficient adaptation to a change in climate. Although these generally occur among newcomers to a climate, they can affect indigenous populations. Although generally treatable, maladjustments such as heat stroke can kill if treatment is not immediate. Maladjustments such as congenital sweat gland deficiency and autoimmune hemolytic anemias render affected individuals permanently incapable of tolerating heat and cold, respectively. Because people can adapt to CO2 concentrations several times the present one, the projected rise in CO2 itself presents no particular hazard (Schaefer, 1958; Weihe, 1979).
The relationship between climate, particularly temperature, and mortality has been studied for over a century (Kalkstein, 1989). Generally, mortality peaks in winter, summer, or both. Weihe (1979) concludes that industrialized countries show winter peak mortality mainly from noncommunicable diseases (e.g., cerebrovascular disease). Developing countries, on the other hand, show peak summer mortality rates primarily from infectious diseases. Although mortality increases have been seen during extreme cold waves, several studies found the impact of cold weather on mortality to be less dramatic than that of hot weather (Kalkstein, 1989).
A major question about climate change and human health concerns acclimatizationthe ability of people to withstand stress with repeated exposure. It has been postulated that people who survive early heat waves become physiologically and behaviorally acclimatized and therefore deal more effectively with later heat waves (Marmor, 1975).
Kalkstein's (1989) estimation of the change in mortality from a 4° to 6°C warming in the summer underlines a paradox about the present: the cities with the highest weather-induced mortality are in the North rather than in the South. So, a scenario of warmer weather but no acclimatization caused the expected rise in mortality. If the population acclimatized, however, the predicted weather-induced mortality in the summer actually declined in about half the 15 cities examined. For the full sample of 15 cities, a moderate rise of mortality was predicted. On the other hand, Kalkstein noted, ''by the [estimation method], the comparatively modest rise in acclimatized mortality (as compared to unacclimatized predictions) parallels the response of people today who reside in southern cities with hot climates. Southern cities represent analogs of expected climate in northern cities, and these
TABLE 34.7 Health Disorders Caused Directly by Climate
warmer cities exhibit … fewer weather-induced deaths in summer." Fewer weather-related deaths in the winter were predicted, of course, if the climate warms. To keep the scale of this mortality in mind, remember that the 5 per 100,000 people whose deaths are attributed to summer weather in the surveyed cities is about a quarter the rate from motor vehicle accidents.
Communicable diseases are transmitted from one person to another by actual contact or contagion or by a vector. A person's body is colonized by parasites (e.g., viruses, bacteria, protozoa, and worms). Climate can play a part because parasites require certain climatic conditions for their existence.
Climate plays a minor role in contagious diseases transmitted directly from person to person (Weihe, 1979). Indirectly, weather causes the incidence of contagious diseases to rise in winter when people crowd into dwellings.
Climate affects the survival of viruses outside the host. For example, influenza virus survives longer in low humidity, while higher relative humidity favors the poliomyelitis virus. Viral diseases with seasonal morbidity are common in both temperate and tropical zones. Influenza A, for example, occurs primarily in the winter. Its morbidity is aggravated by sudden cold spells. Several enteroviruses, on the other hand, are most prevalent during the summer. It is not known whether the seasonality of these viral diseases is due to the impact of the climate on the virus, the host, or both. This makes it difficult to predict the impact of climate change on viral diseases. From the evidence about influenza, one would reason that a warmer climate could decrease the incidence of influenza, whereas increased climate variability could increase the incidence.
In these diseases, the pathogen is transmitted to a person by another agent, called the vector, such as a tick, flea, or mosquito. Climate can affect these diseases in several ways. It can affect both the infectious agent and the vector directly, or it can affect the vector by influencing the types of vegetation or intermediate hosts of the vector. Although the incidences of tick-borne Lyme disease and Rocky Mountain Spotted Fever are increasing, vector-borne diseases are relatively rare in the United States. Appraisals of the impact of climate change on vector-borne diseases in the United States have been attempted (Haile, 1989; Longstreth and Wiseman, 1989).
with varied results. However, they do not incorporate, understandably, such complications as changes in vegetation and habitats for intermediate hosts, less rain versus more puddles of irrigation water, or control measures. As a recent World Health Organization (1990) report noted, without regional climate projections "any predictions on future vector-borne disease risk can only be speculative."
Arteriosclerotic heart disease, nonrheumatic chronic endocarditis, and other myocardial degenerations kill fewer people in warm months than in cold months (Rose, 1966; Rosenwaike, 1966). Heat waves are another matter. Hospital admissions for cardiovascular effects seem to rise during heat waves (Sontaniemi et al., 1970; Gill et al., 1988).
Pneumonia, bronchitis, and influenza kill most in cold months (Weihe, 1979). Other respiratory diseases also are related to weather patterns. Currently, 3 to 6 percent of the U.S. population suffers from hay fever and asthma (Smith and Tirpak, 1989). So, changes in climate that change vegetation, and hence pollen, could affect these allergy sufferers.
Because a variety of skin infections are related to temperature and humidity, changes in climate could change them. Relative humidity and rainfall may play a more important role than temperature for some skin diseases. Longstreth and Wiseman (1989) report that in the U.S. Army skin diseases were the greatest single cause of visits by Army personnel to health care facilities in Vietnam. These skin disorders were directly affected by high rainfall and humidity.
Climate and Human Reproduction
Within a country, conception occurs less often during months with hot temperatures (Weihe, 1979), but across countries birth rates are higher in some tropical climates than in temperate ones. So, among the things that affect birth rates, a warmer climate will not likely be a large influence.
The time of maximum fertility affects infant mortality. For example, in the United States, statistically significant increases in preterm births and perinatal mortality during the summer months has been documented (Keller and Nugent, 1983; Cooperstock and Wolfe, 1986). So, a warmer climate could increase this mortality.
Indirect Effects of Climate Change on Human Health
Lack of safe drinking water and adequate sanitation are currently problems in many parts of the world. They bring with them a number of water-borne
diseases, such as typhoid and dysentery. If water supplies are changed by climate, as the section "Water Resources," above, discusses, health could change.
Similarly, climate change could change the food supply. So, climate change could affect famine and malnourishment and hence susceptibility to diseases. Warmer climates could cause more food spoilage, exacerbating enterovirus infections, which are already frequent in warm weather. Atmospheric hazards are the most deadly natural hazards in the United States (Riebsame et al., 1986). Floods, lightning, and tornados lead other atmospheric events in causing U.S. fatalities. Changes in the severity and frequency of storms could add to or subtract from these deaths.
The climatic changes anticipated are within the range now experienced among existing habitats and to which people who move from place to place have generally learned to adapt.
Older people may be at greatest risk of stress from heat waves; but adaptive means should be researched and found.
Arthropod-borne communicable disease may also increase in warmer, more humid climates if precautions are not taken.
Our adaptations are by and large within the present mainstream of medical research and international public health, but the consequences of neglect will be aggravated.
The pace of improvements in health from better technology and its application can and should greatly exceed the deterioration related to climate change. Failure to control global population growth, chemical pollution, and viral epidemics from presently ascertainable causes greatly outstrips the particular health burden from climate change.
How will climate change during the next 50 to 100 years alter where people live within the United States and over the whole planet? Will migrations, internal and international, take on new patterns and with what meaning for the United States? The implications of climate change for population distribution are scarcely treated in either the EPA (Smith and Tirpak, 1989) or IPCC (1990b) assessments. So, demographic and climate indicators in major countries were surveyed, historical responses to climate change were tabulated, and studies of climate and migration were examined.
Unlike subjects like farming or the natural landscape that may be affected
by climate change, migration is an impact if people flee a harmful climate and is an adaptation when they arrive in a congenial one. It may be an impact on the destination where they arrive. Nevertheless, as we did for the other subjects, we examine sensitivity, impact, and adaptation for migration.
In the section "Health," above, it was noted that humans live in environments from extreme heat to extreme cold, and from extreme aridity to extreme humidity. Roughly 77 percent of today's population of 5.3 billion people live in developing countries, most in warm climates. Fully 84 percent of the projected 8.2 billion people in 2020 will live in today's developing countries. Growth rates of population are fastest for tropical areas of the globe where temperatures average above 20°C annually, so regardless of climate change, more of the world's population will live in relatively warm climates in the next century (United Nations, 1989b).
Most migrations cover relatively short distances and follow established paths (Lee, 1966). Although international migrations are increasing, most migrations take place within nations and, if international, within regions. Restrictive policies of the industrialized countries are the major constraint on immigration, and, without those policy constraints, levels of international migration would be significantly higher worldwide given the economic and other disparities that exist among nations (Tabbarah, 1988). It is unlikely that international migration would increase significantly as a result of climate change.
Most migration is driven by people's desire to improve their economic and social well-being, but a significant and perhaps growing share is forced or refugee migration. The 1951 United Nations Convention Relating to the Status of Refugees established the definition of "refugee" that is now widely accepted by most countries, including the United States:
… any person who, owing to a well-founded fear of being persecuted for reasons of race, religion, nationality, membership of a particular social group or political opinion is outside the country of his nationality and is unable or, owing to such fear, is unwilling to avail himself of the protection of that country.…
By adopting a narrow definition of "refugee," the international community has been able to preclude overwhelming numbers of refugees. The definition leaves, however, a growing number of internationally displaced persons who may be victims of political confrontation although not persecuted themselves, or who may be victims of other forces (such as drought or famine), outside the protection of the United Nations High Commissioner for Refugees (Zolberg, 1989). Although the factors driving refugee migrations are complex, interrelated, and hard to measure, civil strife, war, and government structure seem to account for many and famine and state suppression
for fewer (Edmonston and Lee, 1990). Since most of the persons not qualifying as refugees are located in poor neighboring countries in Africa or Asia, the international community needs to consider how refugees and other migrants can be helped in those locations.
Within the United States
The U.S. population is among the most mobile in the world. About 17 percent of the population changes residence annually and 6 percent changes county of residence. Nonmovers are rareperhaps no more than 10 to 15 percent of adults have spent their entire lives in their county of birth. In addition to high levels of internal migration, about 1.3 million persons enter the United States annually from abroad. The majority of these are nationals of other countries coming as migrants or refugees to live and work in the United States and the remainder are U.S. citizens returning from residence abroad. No other country in the world accepts as many immigrants as the United States (Bogue, 1985).
Preference for warmer regions has been a key determinant of internal population shifts in the United States and other industrialized countries in the post-World War II era (Figure 34.7). Internal migrations toward the Sunbelt have been eased by science and technology developments that, for
example, cooled torrid summer air and controlled malaria in the South and along the Gulf coast. Government policies such as social security, differential state and local taxation policies, and siting of military and aerospace industries also contributed to the growth of the southern regions (Renas and Kumar, 1982; Pampel et al., 1984; Carlino and Mills, 1987; Voss et al., 1988).
Variability of climate in the same zone over time has also driven migration. For instance, in the 19th century and the first four decades of the 20th century in the Great Plains region, droughts fueled outmigration of thousands of people (Tannehill, 1947; Parry, 1978). Government policies and private acts can mitigate the effects of drought, as is demonstrated by reduced migration from the Great Plains during drought periods after the 1930s.
Migrations are sensitive even to slight temperature differentials. A study of migration flows within California, the state with the nation's largest elderly population, found a southward shift of population. Though most national studies of migration treat California in its entirety as a Sunbelt state, there exists a clear distinction between its northern and southern portions with respect to the direction migration takes (Ormrod, 1986).
The sensitivity of migration in other wealthy and industrialized societies resembles that in the United States. In nine such countries, for example, migration was triggered by retirement or widowhoodthe relatively young and affluent among the elderly tend to move to warmer and more amenable climate zones while the older elderly move to areas where they can obtain care and support from family and friends even if this involve a move back to a colder zone (Serow, 1987).
From Mycenaean drought more than a millennium before Christ to Sahelian droughts in our time, people have responded to climate events (see Table 34.8). However, drought is not the only climate event that has produced a response by people. Floods, cold weather, and even a plant pathogen newly arrived and encouraged by warm and moist seasons have increased outmigration or mortality, the response to dire conditions if the migration response is not available. Generally, places that were almost too cool, too dry, or too something became more so, and people were driven out.
Migration sensitivity to climate change appears to be correlated with three factors: the proportion of population living in regions prone to unstable political situations, the proportion of population that is poor or economically vulnerable, and the proportion living in coastal regions.
Political stability varies considerably across nations. Authoritarian regimes, war, civil strife, ethnic and religious conflicts, and state oppression
TABLE 34.8 Population Response to Climate Change in History
(continued on page 625)
(Table 34.8 continued from page 624)
have generated significant numbers of refugees in recent decades (Edmonston and Lee, 1990; U.S. Committee for Refugees, 1990). Most refugees have been produced by a few countries, but several other countries that have not can be considered ''prone" to produce them if political conditions worsen. Recognizing that politics and nations' administrative practices affect counts of refugees,11 one analysis (Edmonston and Lee, 1990) found that there were 8 million refugees in 1986, or slightly over 10 million taking into account 2.1 million Palestinians. By 1990 the generally accepted estimate of worldwide refugees is 15 million (U.S. Committee for Refugees, 1990). Populous countries with large land areas may generate internal migrants who flee one region for another without actually crossing an international border.
Since counts and definitions of refugees are based on political criteria and sensitivities, existing statistics do not identify how many might be considered environmental refugees or persons who moved because climate changes disrupted their livelihood. Environmental refugees are recognized officially only in Africa, where the Organization of African States has adopted a broad definition of refugees that includes persons who have left their homeland due to political, environmental, or other hardships (Zolberg, 1989).
The second source of sensitivity to climate change is the proportion of the population that is poor or economically vulnerable. Levels of economic development correlate strongly with the ability of governments or of people themselves to cope with or reduce the potentially negative effects of climate change on settled populations. Cases such as the stabilization of population in the Great Plains, the building of dikes in the Netherlands, and the Sunbelt shifts of population in the United States and other industrialized countries indicate that levels of economic and technological development mediate climatic effects on population settlements.
In developing countries the government and most of the people do not have the economic means to use improved technologies to adapt to climate change. As such, larger numbers of people are potentially sensitive to climate shifts, particularly if the latter erodes the sustenance base. Urbanization rates are high in developing countries, and by the year 2020 the United Nations projects that 57.7 percent of the world's population and over half of the developing countries' populations will live in urban areas (United Nations, 1989a). The effects of climate change on rural-to-urban migrations and on the process of urbanization in developing countries are unclear.
The third source of sensitivity to climate changethe proportion of population living at sea levelcorrelates closely with the other two sources. Developed countries have greater resources and capabilities to respond to sea level rises than do developing countries. Crude estimates suggest that about one quarter of the world's population resides within 100 km of a coast and would therefore be vulnerable to sea level rise. Priority needs to be given
to refining these estimates, taking into consideration altitude and development levels.
The climate change projected for the next 50 years or so might affect internal migration within the United States. For example, it might incline ongoing migration northward and increase migration to cooler mountains. Such a shift would be by many uncoordinated short moves, rather than a mass migration. The aging of the U.S. population and continued early retirement seem bound to increase the seasonal migrations south in winter and north in summer. Climate change could gradually shift the destinations of seasonal migrants, but flows are likely to continue northward or southward along existing migration corridors (e.g., East Coast toward Florida, Great Plains toward the Southwest).
If projections of sea level rise are correct, in the latter part of the 21st century, coastal populations may need to protect their homes or abandon them. Given existing investments in U.S. coastal settlements, people and governments are likely to respond by protecting those investments (see the section "Settlements and Coastal Structures," above). Thus, sea level rise is unlikely to stimulate migration from coastal areas in the United States, but it could in countries where development is too scanty to afford the adaptation of protection.
Large populations are a feature of many low coastal regions such as the Netherlands, Bengal, and Nile deltas. Concentrating on countries that would be hard pressed to build dikes, a 1-m rise in sea level, which is more than our assumption, would cover areas currently accounting for about 7 percent of habitable land and 5 percent of the population in Bangladesh and 12 percent of habitable land and 14 percent of the population in Egypt (Broadus, 1990). Thus, in those two countries alone, a 1-m rise in sea level could displace 14 million people.
Were climate change to displace settled populations in the world's most populous countries, China and India, significant numbers of refugees could be produced. Neither country currently produces a sizeable number of refugees, but even a small rate of refugee emigration from these populous countries would magnify the number of refugees. The impact of such increases would undoubtedly be greatest on neighboring countries in the Asian region.
If climate changes were greater than we have assumed, emigration from the United States could increase. This option would, of course, be limited to the small proportion of the population, largely the wealthy and educated, who would be able to locate jobs in other countries or develop the contacts that would allow such moves to take place (Kritz, 1991).
As pressures for entry increase, developed countries are likely to respond with further restrictions on immigration flows in order to keep the number of immigrants within manageable bounds. The discouragement of illegal migrations to the United States and greater assistance to help Third World refugees in the countries to which they fled rather than resettling them in the United States are also likely policy options. To the extent that climate change is likely to stimulate large inflows, the potential is greatest along existing paths. Examples are from Mexico, the Caribbean, and Central America to the United States, and from North and West Africa to Europe.
Developing countries have less control over their borders than developed countries and therefore are more vulnerable to illegal immigration. Adaptation to migration may thus be especially great at borders like that between Bangladesh and India were sea levels to rise substantially. Droughts in the Sahel region could also stimulate emigration toward coastal and central zones of Africa.
National and international governmental organizations should prepare for the likelihood that climatic change will be an added or complicating factor for migration. Within the United States, regulation of internal migration would be new. Indirect encouragement of migration, on the other hand, is not new and includes development of water supplies and regulation of construction on areas prone to flooding. The decrease in migration from the Dust Bowl and later droughts in the Great Plains illustrate how government policies can stabilize populations or reduce the likelihood of a migration response (Tannehill, 1947; Parry, 1978; Warrick, 1980).
In other countries, adaptations could range from ones that keep people home to others that allow them to settle in neighboring countries that share economic and political ties. Examples of the former include the dikes that make the Zuider Zee into a polder, the air conditioning that makes Texas and Florida attractive, or the climate and jobs that make mountainous regions prosper. Examples of the latter are the European Community, the Economic Commission for West African States, and the Trans-Tasman Agreement, all of which are regional treaties entered into by countries to further their common economic and political objectives and that allow some unrestricted migration of nationals of member states among the treaty signators.
Historically, people have adapted to a wide range of climates and climate changes. In the United States and other industrialized countries in recent decades, people have migrated toward warmer climates, indicating that they are not averse to hot weather. While migrants adapt by changing location, their impacts on receiving areas increase if they are fleeing refugees. Over
the centuries, refugees have fled drought, cold, and flood in addition to political persecution. Today, an array of factors, but especially civil strife, war, and government structure, drive refugees. These factors plus poverty and living at sea level make migration sensitive to climate change. Where nations have these sensitivities, migration could be a large impact of climate change.
Most people surely share the goal enshrined in the Preamble to the U.S. Constitution, "to insure domestic Tranquility." While some nations have the economic resources and political institutions that allow them to cope with failing harvests and flooded lands, other nations do not. Moreover, "in no country or city can the rich fortify themselves for long against the poor" (Tickell, 1990). The question is how climate will interact with such strains as terrorism, civil war, bankruptcy, and natural disasters to reduce or enhance the tranquility of nations and then the international condition.
The panel felt that it was important to point to the possibility that the precarious economic and political order in many countries around the world could be weakened further by climate change and to consider the implications for industrial countries. How might domestic and international tranquility be sensitive to climate change?
Social and technological development probably, on balance, lessens the vulnerability of societies to climate change (Kates et al., 1985), but sensitivity might be of three sorts: brittleness, accumulation of troubles, or overwhelming change.
Societies have developed increasingly elaborate technical and social systems to insulate themselves from recurrent weather fluctuations. These systems have allowed populations to grow in environments that might otherwise be considered "too hot or too cold" or "too wet or too dry" to sustain the livelihood of large numbers of people. Where such developments have occurred, a vulnerability develops if the technical or social systems fail. For example, if the cooling system fails in an extreme heat wave, mortality could increase among those susceptible to high temperatures. If farmers could not get access to credits to purchase fertilizers or equipment that allowed them to grow crops in poor soil, their vulnerability would be increased.
Accumulation of Troubles
The second sort of sensitivity stems from accumulation of troubles in countries already beset by economic and political difficulties. Here, a large or rapid change could trigger further stress. For instance, sea level rises could disrupt populations living in coastal zones and trigger migrations to other poor areas that will have difficulty feeding and housing them. To the extent that these migrations become international ones, political relationships between the sending and receiving countries could be strained.
Extreme climate change (e.g., a 5°C warming and a 1-m rise of sea level in 75 years) could require large-scale food relief and relocation of substantial populations from affected areas. Countries that are small territorially or that lack climatic variability will have difficulty adjusting. Domestic hardships could spill over into neighboring countries as refugees flee harsh conditions at home.
Variation in Sensitivity
Political sensitivity is likely greatest in countries already under population or environmental pressures. For the most part, they are also limited in administrative capacity, and natural disasters often bring a severe break-down. They are largely more agricultural, less industrialized, and have less technological capacity. Sensitivity is greater where many people live along the world's great rivers (whose flows might change), on low-lying coasts (that might eventually be flooded), or by agriculture that is finely tuned to current climate (e.g., monsoon rains).
Those societies that suffer the greatest economic losses from strategies to control emissions of greenhouse gases would also be sensitive. For example, petroleum or coal exporting nations would be susceptible if they could not find new sources of income, and nations that depend on high consumption of fossil fuels would suffer from draconian cuts.
On the other side of the coin, in parts of the world where climatic changes were economically beneficial, political stability could be enhanced.
The first impact could originate from claims that an economic or social goal has not been attained because of climatic change caused by human agency. The greenhouse issue has potential for exacerbating conflict between the poor, and therefore more vulnerable, countries and the industrialized ones that are less vulnerable.
If climate change is greater, the potential political and other effects associated with it will also increase. For instance, substantial reduction of water and buffer stocks of food often leads to increased malnutrition and mortality, especially among infants, and to population shifts internally or internationally. One can see an example in the horn of Africa. Until recently, the United Kingdom supported irrigation projects there. Now, the combination of drought and civil war has made it impossible to continue the projects. Therefore, the financing and experts have been pulled out, and refugees have been generated by both the environmental stress and civil strife (Tickell, 1990).
The paramount adaptation by migration will be within nations and depends on effective government. The most important transfers of money, resources and people will be within nations, but, as sensitivities increase, these could spill over to other countries. Spillover effects could be intensified if climate change alters the relative productive potential of countries.
A proper balance of initiative in the public and private sectors can facilitate adaptation. Governments are increasingly enacting environmental rules, as the example of the United States shows (Figure 34.8), both to prevent environmental damage and to adapt to changing conditions. Not all adaptive policies, however, call for direct government action. Some adaptations call for government to establish incentives and then let market forces signal adaptations. Others only require effective transmission of economic signals. The economic system should send accurate and rapid signals (usually prices) about scarcities of such goods as water. This is true for goods and services that are already exchanged in markets and for those in which mechanisms like markets can be established.
As related in the section "Migration," above, internal and international mobility of labor and other resources can be routes for adaptation. Where there is national diversity in climates and resources, effective governments can facilitate and ease the adaptation.
All adaptive strategies, and hence domestic tranquility, will be helped by effective government. The nations most vulnerable to climatic stress may be those with the weakest governments today. Attributes of effective administration include policy coordination at the highest level of government (heads of government need to pay attention, as they set priorities and allocate budgets); a structure that can turn policy into action; coordination among technical people at the working level; and links between government and nongovernmental organizations and between government and public
opinion (Tickell, 1986, 1990). The establishment of an effective government is likely to be as protracted as climate change.
Between nations, migration across frontiers, especially in large numbers, creates problems. Many in the receiving countries feel that the number of people and differences in culture that can be absorbed are limited. Unemployment in the receiving country makes it still harder. Resistance to immigration is popular politics, and more refugees only strengthen the resistance (Tickell, 1990).
Internationally, at least three kinds of actions may lessen the blows of climate change and hence serve as adaptations:
• Assistance for developing countries in building their own indigenous research and assessment. A government is more likely to understand and act if the country has national experts participating fully in the world scientific dialogue on climate change, its causes, its impacts, and adaptations to it.
• Assitance toward rapid, sustainable, balanced growth in per capita income and help in slowing population growth. In both the short and the long run, these are probably the best adaptive strategies. All-round knowledge plus skills and technologies for adapting to climate variation and change should also be provided where they are lacking. Ideally, climate change will fade out as a distinctive problem if climate-oriented policies become part of development policies in general (Meyer-Abich, 1980).
• Disaster reduction and relief. Strengthening both international and national ways to wider and prompter response to natural disasters helps adaptation. Care is needed lest short-term adjustments to climatic hazards do not slow long-term adaptation.
As the negotiations for the Law of the Sea well demonstrated, the process of creating new international law and new international institutions is usually painful and protracted. Furthermore, development assistance is rarely effective except as part of a long-term policy.
Concern about political tranquility stems from the fear that some governments may not be able to cope with the added stress of climate change. Many countries outside the industrial world may lack the government or resources to manage a continuous environmental crisis, which unfavorable climate change could bring. Difficulties of organizing coordinated, multi-lateral responses to problems such as hunger are already widely evident. Unfavorable climate change could aggravate present economic, political, and social problems and swamp national governments and international assistance.
Although it is impossible to assess in a scientific manner the contribution that climate change may make to global political tensions, the possibility that climate change will disrupt domestic tranquility cannot be ignored. An analytic base needs to be developed that would allow us to specify the conditions under which this outcome is likely to occur.
At the beginning of this chapter, we raised questions to keep in mind as we examined sensitivities, impacts, and adaptations. Drawing on our examination, we now comment on several key questions (Ausubel, 1991).
Is Faster Change Worse than Slow?
Some changes may be beneficial, such as the added rain predicted for some farming regions by some computations. So, the directions of change must be specified before the question is answered. For the answer to matter, the combination of climate change and sensitivity to it must be greater for an activity than other kinds of changes and sensitivities. Also, the usual
rate of change and adaptation of the activity must be slow or its forced rate of adaptation must be beyond our means for the answer to matter. Thus, the answer is not universally "yes," although a faster harmful change is clearly worse than a slow one in the same direction for, say, sensitive ecosystems or an impoverished coastal plain. For an activity like planting a new forest, a fast change that is anticipated may matter less than a slow one that is unforeseen. In general, rapid change will be more harmful than slow change, but an anticipated change, even if rapid, may be less harmful than an unforeseen slow one.
Will Waiting to Make Policy and to Act Drive Up Costs?
Sometimes we can calculate the cost of waiting versus acting. Acting now is justified if the cost of acting now relative to that of acting later is lower than a ratio that can be calculated from (1) the probability of the action being required, (2) the discount rate of money, and (3) how long the action can be postponed. The justified ratio is especially sensitive to the discount rate and the possible postponement. In practice these ratios can be computed only, if at all, to a very gross level of approximation. The cost of acting later will be affected by the growth in knowledge, technology, and wealth, which have grown in the past. As the world becomes more crowded, the reservation of sites for infrastructure, for the preservation of precious sites, or for the migration of ecosystems is likely to grow harder. Thus, they might be exempted from the rules of justification.
Are there Only Losers from Climate Change?
Under the assumptions we accepted of gradual warming and redistributed precipitation, there will be a complex and shifting set of winners and losers. Possible losses have frequently been predicted. Some wins may be absolute, for example, an increased crop yield. Some may be relative, as when yields fall less in one place than in another. The argument that there will be more losers than winners arises because we are currently adapted to present climate, and thus any change is likely to require at least some costly adaptation.
Will the Most Important Impacts be on Farming and from the Rise of Sea Level?
The adaptability of farming and even the forestry of managed trees is considerable and that of the natural unmanaged landscape or shore is less so. Cities and irrigated farms in dry regions show that water supply can be adapted, and there is room in many areas for further adaptation. Nevertheless, the impact of changes of climate on water supply is relatively large,
and the adaptation is likely to be costly. Despite the sensitivity of coasts to the rise of sea level, adaptation will likely preserve valuable sites like cities. So with reasonable adaptation by humanity, it is natural landscapes, shores, and water supply that will be left to suffer more.
Will Changes in Extreme Climatic Conditions be More Important than Changes in Average Conditions?
The foundation of this question is a belief that extremes will breach some threshold of sensitivity. Since warming is assumed, the obvious extreme and threshold of frost will be breached less and that of debilitating heat waves more frequently. Because a general increase in variability is not currently predicted, the cause of breaching more thresholds would be a gradual increase in the frequency of warming and in some places drying as their entire frequency distributions crept upward. The answer to the question lies in locating thresholds of sensitivity and predicting shifts in averages and then frequency distributions.
Are the Changes Unprecedented from the Perspective of Adaptation?
Although globally, on average, the assumed climate change for the next century exceeds any for tens of hundreds or thousands of years past, it is much smaller than that from day to night, from summer to winter, or between airports one might leave and reach in an hour. The ways humanity adapts to the present differences are impressive. It is important, therefore, to clarify how the projected climate change will differ from all the climatic variations that humans, the economy, and ecosystems are now accustomed to. Nevertheless, it seems that for impact and adaptation the unprecedented character matters primarily at margins of natural, forestry, or farming zones where the present climate is already close to a threshold of sensitivity.
Will Impacts be Harder on Less Developed Countries or on Developed Ones?
The impacts are likely to be larger for less developed countries because they derive a greater fraction of income from outdoors and hence sensitive activities like farming, forestry, or fisheries and because they lack wealth to afford adaptation and the infrastructure and technology to accomplish it. A sensitivity that might be forgotten is the effect of striving to mitigate carbon emissions upon a less developed country that depends on the extraction or processing of fossil fuel. In general, the consequences of climate change will likely be consistent with the findings of research about other health and environmental hazards that "richer is safer" (Wildavsky, 1980).
Are Some Hedges Clearly Economical?
Several hedges against gradual climate change are economical and are under way: testing crop varieties, replanting forests with adapted trees, coping with current variability in water supply, improving weather and climate information, clarifying policies for migration, and strengthening programs to reduce the effects of natural disasters. Also, the panel found other potential hedges such as better management of the demand for water and more prudent development of coastal zones. The question must be asked, however, why these apparently prudent actions are not already being pursued more actively. The panel noted the anticipatory investment in a higher offshore oil platform designed for many decades, and offered a simple method for including climate change in the calculus of risks, costs, and benefits of investments. Tools and applications exist for demonstrating more widely the economics of hedging.
1. The relative [(change of plant growth)/growth]/[(change of CO2)/CO2] is called beta or CO2 fertilization effect. In models relating the rise in CO2 in the air to its emission, a standard value of about 0.25 is used, and varying it changes predictions of future CO2 in the air greatly. See review in National Research Council (1983, p. 20) and Keeling and Bacastow (1977).
2. Damage to crops is figured from weather conditions.
3. This section concentrates on ecological effects and fisheries. Changes in the oceans could also affect tourism, recreation, and coastal structures, subjects covered in the sections ''Tourism and Recreation" and "Settlements and Coastal Structures."
4. In the recent past, concerns have been raised about the melting of the West Antarctic ice sheet, which rests on the sea floor. Such melting, it has been estimated, would add over 5 m to the sea level. Current estimates put such large-scale melting at least two centuries into the future.
5. El Niño/Southern Oscillation (ENSO) refers to an anomalous ocean/atmosphere interaction in which sea surface temperatures are warmer than usual in the central and eastern equatorial Pacific, and air pressure is high over the southeastern Pacific and low over the Indian Ocean. Changes in ocean upwelling, sea level, and rainfall accompany ENSO.
6. Note, however, that even without barriers, wetland areas may diminish significantly. Another adaptation strategy would be to develop measures to enhance critically the natural productivity of wetlands, perhaps by augmenting natural sedimentation rates.
7. Publications of the U.S. Geological Survey show 4,200 billion gallons/day (BGD) precipitation and 2,800 BGD evaporation, leaving 1,400 BGD runoff, in the 48 contiguous states. They also show 336 BGD withdrawn for use and 92 BGD consumed in 1985. Use of fresh water rose until 1980 and then declined 11 percent in 5 years because less was used in irrigation and industry.
8. Similar results were obtained by Adams et al. (1990).
9. The 1988 Statistical Abstract of the United States has the following relevant facts: 47 million fishermen and 17 million hunters (Table 379); 48 million people in attendance at major league baseball games and 36 million at college football games (Table 372); 345 million visits to national parks (Table 357) and 675 million visits to state parks and recreational areas (Table 362); $55,450 million (1985 dollars) for wildlife-associated recreation (Table 380); $14,072 million for books (Table 366); $7,250 million for purchases of color TVs (Table 371); 34 percent participation of people in swimming and 19 percent in freshwater fishing (Table 374).
10. For a discussion of recreation, wildlife, water, and climate change see Cooper (1990).
11. Some countries, for example, report certain aliens as refugees, but others classify similar migrants as economic or illegal migrants. It is often in the political interest of a government to understate its refugee population or in other cases to inflate the numbers. Some countries admit individuals under classifications other than refugee, and therefore they are not counted in the totals. Finally, asylum seekers, who are growing in number, are not reflected in refugee counts (U.S. Committee for Refugees, 1990).
Abelson, P. H. 1991. Desalination of brackish and marine waters. Science 251:1289.
Acock, B., and L. H. Allen, Jr. 1985. Crops responses to elevated carbon dioxide concentrations. In Direct Effects of Increasing Carbon Dioxide on Vegetation, B. R. Strain and J. D. Cure, eds. Report DOE/ER-0238. Washington, D.C.: U.S. Department of Energy.
Adams, R. M., C. Rosenzweig, R. M. Peart, J. T. Ritchie, B. A. McCarl, J. D. Glyer, R. B. Curry, J. W. Jones, K. J. Boote, and L. H. Allen, Jr. 1990. Global climate change and U.S. agriculture. Nature 345:219–224.
Allen, R. G., and F. N. Gichuki. 1989. Effects of project CO2-induced climatic changes on irrigation water requirements in the Great Plains states (Texas, Oklahoma, Kansas, and Nebraska). In the Potential Effects of Global Climate Change in the United States, Appendix C, Agriculture, J. B. Smith and D. A. Tirpak, eds. Washington, D.C.: U.S. Environmental Protection Agency.
Ausubel, J. H. 1991. A second look at the conventional wisdom about the impacts of global climatic change. American Scientist 79(3):210–221.
Bach, W. 1979. The impact of increasing atmospheric CO2 concentrations on the global climate: Potential consequences and corrective measures. Environment International 2:215–228.
Bakun, A. 1990. Global climate change and intensification of coastal ocean upwelling. Science 247:198.
Balzhiser, R. E. 1989. Meeting the near-term challenge for power plants. In Technology and Environment, J. H. Ausubel and H. E. Sladovich, eds. Washington, D.C.: National Academy Press.
Barbier, E. B. 1989. The global greenhouse effect, economic impacts and policy considerations. Natural Resources Forum (February):20–32.
Bazzaz, F. A., and R. W. Carlson. 1984. Response of plants to elevated CO2. I. Competition among an assemblage of annuals at different levels of soil moisture. Oecologia 62:196–198.
Bazzaz, F. A., K. Garbutt, and W. E. Williams. 1985. Effect of increased atmospheric carbon dioxide concentration on plant communities. In Direct Effects of Increasing Carbon Dioxide on Vegetation, B. R. Strain and J. D. Cure, eds. Report DOE/ER-0238. Washington, D.C.: U.S. Department of Energy.
Binkley, C. S. 1988. The impact of CO2-induced climate change on the world's forest sector. In The Impact of Climatic Variations on Agriculture, Volume 1, Assessments in Cool Temperature and Cold Regions, M. Parry, T. Carter, and N. Konijn, eds. Dordrecht, The Netherlands: Kluwer Academic Publishers.
Binkley, C. S. 1990. Climate change and forests. William P. Thompson Memorial Lecture given at Northern Arizona University, Flagstaff, June 5, 1990.
Binkley, C. S., and D. P. Dykstra. 1987. Timber supply. In The Global Forest Sector: An Analytical Perspective, M. Kallio, D. Dykstra, and C. S. Binkley, eds. New York: John Wiley & Sons.
Binkley, C. S., and J. Vincent. 1988. Timber prices in the U.S. South: Past trends and outlook for the future. Southern Journal of Applied Forestry 12:15–18.
Birdsey, R. A. 1990. Estimation of regional carbon yields for forest types in the United States. Unpublished manuscript. U.S. Forest Service, U.S. Department of Agriculture, Washington, D.C.
Blasing, T. J., and A. Solomon. 1982. Response of the North American corn belt to climate warming. Publication 2134. Oak Ridge, Tenn.: Environmental Sciences Division, Oak Ridge National Laboratory.
Bogue, D. J. 1985. The Population of the United States: Historical Trends and Future Projections. New York: The Free Press.
Botkin, D. B., and R. A. Nisbet. 1989. Projecting the effects of climate change on biological diversity in forests. In Consequences of the Greenhouse Effect for Biological Diversity, R. Peters, ed. New Haven, Conn.: Yale University Press.
Botkin, D. B., and L. B. Simpson. 1989. The global carbon cycle and land vegetation. Paper presented at the 198th National Meeting of the American Chemical Society, Miami Beach, Florida, September 10–15, 1989.
Botkin, D. B., R. A. Nisbet, and T. E. Reynales. 1989. Effects of climate change on the forests of the Great Lakes states. In The Potential Effects of Global Climate Change on the United States, Appendix D, Forests, J. B. Smith and D. A. Tirpak, eds. Washington, D.C.: U.S. Environmental Protection Agency.
Botkin, D. B., D. A. Woodby, and R. A. Nisbet. 1991. Kirtland's Warbler habitats: A possible early indicator of climatic warming. Biological Conservation 56: 63–78.
Broadus, J. 1990. Possible impacts of and adjustments to sea level rise: The cases of Bangladesh and Egypt. Contribution No. 7147. Woods Hole, Mass.: Woods Hole Oceanographic Institution.
Broecker, W. S. 1987. Unpleasant surprises in the greenhouse? Nature 328:123–126.
Broecker, W. S., and G. H. Denton. 1990. What drives glacial cycles? Scientific American (January):49–56.
Brown, L. R., A. Burning, C. Flavin, H. French, J. Jacobson, M. Lowe, S. Postel, M. Renner, L. Starke, and J. Young. 1990. State of the World 1990. New York: W. W. Norton.
Bryson, R. A., H. H. Lamb, and D. L. Donley. 1974. Drought and the decline of the Mycenae. Antiquity 48:46–50.
Buddenhagen, I. W. 1977. Resistance and vulnerability of tropical crops in relation to their evolution and breeding. Annals of the New York Academy of Sciences 287:309–326.
Cade, T. J., ed. 1988. Peregrine Falcon Populations, Their Management and Recovery. Boise, Idaho: Peregrine Fund.
Campbell, D. J. 1977. Strategies for Coping with Drought in the Sahel: A Study of Recent Population Movements in the Department of Maradi, Niger, Ph.D. dissertation. Clark University, Worcester, Mass.
Cannell, M. G. R. 1987. Climate warming and spring phenology of trees. In Woody Plant Growth in a Changing Chemical and Physical Environment, D. P. Lavender, ed. Proceedings of the International Union of Forestry Research Organizations Workshop, Vancouver, British Columbia, Canada, July 27–31, 1987. Vienna, Austria: International Union of Forestry Research Organizations.
Cannell, M. G. R., and R. I. Smith. 1986. Climate warming, spring budburst and frost damage on trees. Journal of Applied Ecology 23:177–191.
Capinera, J. L., and D. R. Horton. 1989. Geographic variation in effects of weather on grasshopper infestation. Environmental Entomology 18:8–14.
Carlino, G., and E. Mills. 1987. The determinants of country growth. Journal of Regional Science 27(1):39–54.
Castilla, J. C. 1990. Earthquake-caused coastal uplift and its effects on rocky intertidal kelp communities. Science 240:440–443.
Chapman, V. J., and D. J. Chapman. 1980. Seaweeds and Their Uses. New York: Chapman and Hall.
Connell, J. H. 1972. Community interactions on marine rocky intertidal shores. Annual Review of Ecology and systematics 3:169–192.
Conway, W. G. 1986. The practical difficulties and financial implications of endangered species breeding programs. International Zoo Yearbook 24/25:210–219.
Conway, W. G. 1989. The prospects for sustaining species and their evolution. In Conservation for the Twenty-first Century, D. Western and M. Pearl, eds. New York: Oxford University Press.
Cooper, C. F. 1990. Recreation and wildlife. In Climate Change and U.S. Water Resources, P. E. Waggoner, ed. New York: John Wiley & Sons.
Cooperstock, M., and R. A. Wolfe. 1986. Seasonality of preterm birth in the collaborative perinatal project: demographic factors. American Journal of Epidemiology 124:234–241.
Cordell, K. 1989. An analysis of outdoor recreation and wilderness situation in the United States, 1989–2040. Technical document supporting the 1989 Resources Planning Act Assessment. Washington, D.C.: Forest Service, U.S. Department of Agriculture.
Crosson, P.R., L. A. Katz, and J. Wingard. 1991. Agricultural production and
resource use in the MINK region without and with climate change. Working Paper IIA prepared for U.S. Department of Energy under agreement 041460-A-K1. Washington, D.C.: U.S. Department of Energy.
Cure, J. D. 1985. Carbon dioxide doubling responses: A crop survey. In Direct Effects of Increasing Carbon Dioxide on Vegetation, B. R. Strain and J. D. Cure, eds. Report DOE/ER-0238. Washington, D.C.: U.S. Department of Energy.
Davis, M. 1981. Quaternary history and the stability of forest communities. In Forest Succession: Concepts and Application, D. C. West, H. H. Shugart, and D. B. Botkin, eds. New York: Springer-Verlag.
Davis, M. B., and C. Zabinski. 1990. Changes in geographical range resulting from greenhouse warming: Effects on biodiversity in forests. In Consequences of the Greenhouse Effect for Biological Diversity. New Haven, Conn.: Yale University Press.
Drake, B. 1989. Elevated Atmospheric CO2 Concentration Increases Carbon Sequestering in Coastal Wetlands. Report DE-AC05-OR21400. Oak Ridge, Tenn.: Carbon Dioxide Information Analysis Center, Environmental Sciences Division, Oak Ridge National Laboratory.
Dudek, D. J. 1989. Climate change impacts upon agriculture and resources: A case study of California. In The Potential Effects of Global Climate Change in the United States, Appendix C, Agriculture, J. B. Smith and D. A. Tirpak, eds. Washington, D.C.: U.S. Environmental Protection Agency.
Duvick, D. N. 1984. Genetic diversity in major farm crops on the farm and in reserve. Economic Botany 38:161–178.
Eamus, D., and P. G. Jarvis. 1989. The direct effects of increase in the global atmospheric CO2 concentration on natural and commercial temperature trees and forest. Advances in Ecological Research 19:1–55.
Easterling, W. E. 1989. Farm-level adjustments by Illinois corn producers to climate change. In The Potential Effects of Global Climate Change in the United States, Appendix C, Agriculture, J. B. Smith and D. A. Tirpak, eds. Washington, D.C.: U.S. Environmental Protection Agency.
Easterling, W. E, M. McKenney, N. J. Rosenberg, and K. Lemon. 1991. A Farm-Level Simulation of the Effects of Climate Change on Crop Productivity in the MINK Region. Working Paper IIB prepared for U.S. Department of Energy under agreement 041460-A-K1. Washington, D.C.: U.S. Department of Energy.
Edmonston, B., and S. M. Lee. 1990. Factors affecting refugee emigration. Report OSC-DPS-UI-3. Washington, D.C.: The Urban Institute.
Farnum, P. 1990. Creating new stands: Adapting to the greenhouse effect through technological preparedness. Paper presented at the U.S. Environmental Protection Agency Conference, Forestry Responses to Climate Change, Washington, D.C., May 15–17, 1990.
Federal Insurance Administration. 1984. Retrofitting Flood-Prone Residential Structures. Washington, D.C.: Office of Loss Reduction, Federal Insurance Administration.
Fedkiw, J. 1989. The evolving use and management of the nation's forest, grasslands, croplands, and related resources. Technical document supporting the 1989 Resources Planning Act Assessment. Washington, D.C.: U.S. Forest Service, U.S. Department of Agriculture.
Flather, C. H., and T. W. Hoekstra. 1989. An analysis of the wildlife and fish
situation in the United States, 1989–2040. Technical document supporting the 1989 Resources Planning Act Assessment. Washington, D.C.: Forest Service, U.S. Department of Agriculture.
Franklin, J., V. Dale, D. Perry, F. Swanson, M. Harmon, T. Spies, A. McKee, and D. Larson. 1989. Effects of global climatic change on the forests of northwestern North America. In Consequences on the Greenhouse Effect for Biological Diversity. New Haven, Conn.: Yale University Press.
Frederick, K. D., and A. V. Kneese. 1990. Reallocations by markets and prices. In Climate Change and U.S. Water Resources, P. E. Waggoner, ed. New York: John Wiley & Sons.
Gates, D. M. 1985. Global biospheric response to increasing atmospheric carbon dioxide concentration. In Direct Effects of Increasing Carbon Dioxide on Vegetation, B. R. Strain and J. D. Cure, eds. Report DOE/ER-0238. Washington, D.C.: U.S. Department of Energy.
Gibbs, M. 1986. Planning for sea level rise under certainty: A case study of Charleston, South Carolina. In Effects of Changes in Stratospheric Ozone and Global Climate, Volume 4, Sea Level Rise, J. G. Titus, ed. Washington, D.C.: U.S. Environmental Protection Agency.
Gill, J. S., P. Davies, S. K. Gill, and D. G. Beevers. 1988. Wind-chill and the seasonal variation of cerebrovascular disease. Journal of Clinical Epidemiology 41:225–230.
Glantz, M. H. 1982. Consequences and responsibilities in drought forecasting: The case of Yakima, 1977. Water Resources Research 18:3–13.
Glantz, M. H., ed. 1988. Drought and Hunger in Africa. Cambridge: Cambridge University Press.
Gleick, P. H. 1987. Regional hydrologic consequences of increases in atmospheric CO2 and other trace gases. Climatic Change 10:137–160.
Gleick, P. H. 1989. The implications of climate change for international security. Climatic Change 15:309–325.
Gleick, P. H. 1990. Vulnerability of water systems. In Climate Change and U.S. Water Resources, P. E. Waggoner, eds. New York: John Wiley & Sons.
Gleick, P. H., and E. P. Maurer. 1990. Assessing the Costs of Adapting to Sea Level Rise: A Case Study of San Francisco Bay. Stockholm, Sweden: The Stockholm Environment Institute and Pacific Institute for Studies in Development, Environment, and Security.
Glynn, P. W. 1988a. El Niño-Southern Oscillation 1982–1983: Nearshore population, community, and ecosystem responses. Annual Review of Ecology and Systematics 19:309–345.
Glynn, P. W. 1988b. El Niño warming, coral mortality and reef framework destruction by echinoid bioerosion in the Eastern Pacific. Galaxea 7:129–160.
Glynn, P. W. 1991. Coral reef bleaching in the 1980s and possible connections with global warming. Trends in Ecology and Evolution 6(6):175–179.
Goemans, T. 1986. The sea also rises: The ongoing dialogue of the Dutch with the sea. In Effects of Changes in Stratospheric Ozone and Global Climate, Volume 4, Sea Level Rise, J. G. Titus, ed. Washington, D.C.: U.S. Environmental Protection Agency.
Gordon, P. 1982. Special Statistical Summary: Deaths, Injuries, and Property Loss
by Disaster, 1970–1980. Washington, D.C.: Federal Emergency Management Agency.
Gosselink, J. G., E. P. Odum, and R. M. Pope. 1973. The Value of the Tidal Marsh. Baton Rouge: Center for Wetlands Research, Louisiana State University.
Goudriaan, J., and H. E. de Ruiter. 1983. Plant growth in response to CO2 enrichment, at two levels of nitrogen and phosphorus supply: Dry matter, leaf area and development. Netherlands Journal of Agricultural Science 31:157–169.
Guldin, R. W. 1989. An analysis of the water situation in the United States, 1989–2040. Technical document supporting the 1989 Resources Planning Act Assessment. Washington, D.C.: Forest Service, U.S. Department of Agriculture.
Haile, D. G. 1989. Computer simulation of the effects of changes in weather patterns on vector-borne disease transmission. In The Potential Effects of Global Climate Change on the United States, Appendix G, Health, J. B. Smith and D. A. Tirpak, eds. Washington, D.C.: U.S. Environmental Protection Agency.
Haven, S. B. 1971. Effects of land-level changes on intertidal invertebrates, with discussion of postearthquake ecological succession. In The Great Alaska Earthquake of 1964: Biology, Washington, D.C.: National Academy of Sciences.
Haynes, R. W. 1989. An analysis of the timber situation in the United States, 1989–2040. Technical document supporting the 1989 Resources Planning Act Assessment. Washington, D.C.: Forest Service, U.S. Department of Agriculture.
Henshaw, G. G. 1975. Technical aspects of tissue culture storage for genetic conservation. In Crop Genetic Resources for Today and Tomorrow, O. H. Frankel and J. G. Hawkes, eds. New York: Cambridge University Press.
Idso, S. B. 1988. Three phases of plant responses to atmospheric CO2 enrichment. Plant Physiology 85:5–7.
Illinois Department of Transportation. 1986. Elevating or Relocating a House to Reduce Flood Damage. Chicago: Illinois Division of Water Resources, Illinois Department of Transportation.
Intergovernmental Panel on Climate Change. 1990a. IPCC: The First Assessment Report, Volume I, Overview and Policymakers' Summaries. Geneva: World Meteorological Organization.
Intergovernmental Panel on Climate Change. 1990b. Climate Change: The IPCC Impacts Assessment, W. J. McG. Tegart, G. W. Sheldon, and D. C. Griffiths, eds. Canberra: Australian Government Publishing Service. (Available from International Specialized Book Service, Portland, Oreg.)
International Rice Research Institute. 1983. Annual Report for 1981. Los Baños, Philippines: International Rice Research Institute.
International Union for Conservation of Nature and Natural Resources, Botanic Gardens Conservation Secretariat. 1989. The Botanic Gardens Conservation Strategy. Kew, Richmond, England: World Wide Fund for Nature and International Union for Conservation of Nature and Natural Resources Botanic Gardens Conservation Secretariat.
Jäger, J. 1985. Energy resources. In Climate Impact Assessment, R. W. Kates, J. H. Ausubel, and M. Berberian, eds. SCOPE 27. New York: John Wiley & Sons.
Jarvis, P. G., and K. B. McNaughton. 1986. Stomatal control of transpiration: Scaling up from leaf to region. Advances in Ecological Research 15:1–49.
Jenny, H. 1941. Factors of Soil Formation. New York: McGraw-Hill.
Jodha, N. S. 1989. Potential strategies for adapting to greenhouse warming: Perspectives from the developing world. In Greenhouse Warming: Abatement and Adaptation, N. J. Rosenberg, W. E. Easterling III, P. R. Crosson, and J. Darmstadter, eds. Washington, D.C.: Resources for the Future.
Joffe, J. S. 1949. Pedology. New Brunswick, N.J.: Pedology Publishing Company.
Kalkstein, L. S. 1989. The impact of CO2 and trace gas-induced climate changes upon human mortality. In The Potential Effects of Global Climate Change on the United States, Appendix G, Health, J. B. Smith and D. A. Tirpak, eds. Washington, D.C.: U.S. Environmental Protection Agency.
Karl, T. R., and W. E. Riebsame. 1989. The impact of decadal fluctuations in mean precipitation and temperature on runoff: A sensitivity study over the United States. Climatic Change 15(3):423.
Kates, R. W., J. H. Ausubel, and M. Berberian, eds. 1985. Climate Impact Assessment. SCOPE 27. New York: John Wiley & Sons.
Keeling, C. D., and R. B. Bacastow. 1977. Impact of industrial gases on climate. In Energy and Climate, National Research Council. Washington, D.C.: National Academy Press.
Keller, C. A., and R. P. Nugent. 1983. Seasonal patterns in perinatal mortality and preterm delivery. American Journal of Epidemiology 118:689–698.
Kessler, E., and G. F. White. 1983. The thunderstorm in human affairs. In Thunderstorms: A Social, Scientific, and Technological Documentary, E. Kessler, ed. Norman: University of Oklahoma Press.
Kimball, B. A. 1982. Carbon Dioxide and Agricultural Yield: An Assemblage of 430 Prior Observations. WCL Report II. Phoenix, Ariz.: U.S. Water Conservation Laboratory.
Kimmins, J. P., and D. P. Lavender. 1987. Implications of climate change for the distribution of biogeoclimatic zones in British Columbia and for the growth of temperate forest species. In Woody Plant Growth in a Changing Chemical and Physical Environment. Proceedings of the International Union of Forestry Research Organizations Workshop, Vancouver, British Columbia, Canada, July 27–31. Vienna, Austria: International Union of Forestry Research Organizations.
Kneese, A. V., and G. Bonem. 1986. Hypothetical shocks to water allocation institutions in the Colorado Basin. In New Courses for the Colorado River, G. Weatherford and F. L. Brown, eds. Albuquerque: University of New Mexico Press.
Kritz, M. M. 1990. Climate change and migration adaptations. 1990 Working Paper Series, No. 2.16. Ithaca, New York: Population and Development Program, Cornell University.
Kritz, M. M. 1991. International migration systems, processes and policies. In International Migration Systems: A Global Approach, M. M. Kritz, L. L. Lim, and H. Zlotnik, eds. London: Oxford University Press.
LaMarch, V. C., D. A. Greybill, D. A. Fritts, and M. R. Rose. 1984. Increasing atmospheric carbon dioxide: Tree-ring evidence for growth enhancement in natural vegetation. Science 225:1019–1021.
Lamb, H. H. 1981. An approach to the study of development of climate and its
impact on human affairs. In Climate and History: Studies in Past Climates and Their Impact on Man, T. M. L. Wigley, M. J. Ingram, and G. Farmer, eds. New York: Cambridge University Press.
Langbein, W. B., et al. 1949. Annual Runoff in the United States. U.S. Geological Survey Circular 5. Washington, D.C.: U.S. Geological Survey.
LaRoche, T. B., and M. K. Webb. 1987. Impact of accelerated sea level rise on drainage systems in Charleston, S.C. In Potential Impacts of Sea Level Rise on Coastal Drainage Systems. Washington, D.C.: U.S. Environmental Protection Agency.
Larson, B., C. Binkley, and S. Winnett. 1989. Simulated effects of climatic warming on the productivity of managed northern hardwood forests. Draft manuscript. School of Forestry and Environmental Studies, Yale University, New Haven, Conn.
Lashof, D. A., and D. A. Tirpak, eds. 1990. Policy Options for Stabilizing Global Climate. Washington, D.C.: U.S. Environmental Protection Agency.
Lee, E. S. 1966. A theory of migration. Demography 3(1):47–57.
Linder, K. P., M. J. Gibbs, and M. R. Inglis. 1987. Potential Impacts of Sea Level Rise on Coastal Drainage Systems. Report 88-2. Albany: New York State Energy Research and Development Authority.
Linderer, K. P. 1988. Regional and national effects of climate change on demands for electricity. In Second North American Conference on Preparing for Climate Change. Washington, D.C.: The Climate Institute.
Longstreth, J., and J. Wiseman. 1989. The potential impact of climate change on patterns of infectious disease in the United States. In The Potential Effects of Global Climate Change on the United States, Appendix G, Health, U.S. Environmental Protection Agency. Washington, D.C.: U.S. Environmental Protection Agency.
Lubchenco, J. 1980. Algal zonation in the New England rocky intertidal community: An experimental analysis. Ecology 61:333–344.
Lubchenco, J., S. A. Navarette, and B. N. Tissot. 1991. Consequences of global warming to shoreline ecosystems along the West Coast of North America. Unpublished manuscript. Department of Zoology, Oregon State University, Corvallis.
Mann, K.H. 1973. Seaweeds: Their productivity and strategy for growth. Science 182:975–981.
Marmor, M. 1975. Heat wave mortality in New York City, 1949 to 1970. Archives of Environmental Health 30:131–136.
Matalas, N. C. 1990. What statistics can tell us. In Climate Change and U.S. Water Resources, P. E. Waggoner, ed. New York: John Wiley & Sons.
McGovern, T. H. 1981. The economics of extinction in Norse Greenland. In Climate and History: Studies in Past Climates and Their Impact on Man, T. M. L. Wigley, M. J. Ingram, and G. Farmer, eds. New York: Cambridge University Press.
Meyer-Abich, K. M. 1980. Chalk on the white wall. In Climatic Constraints and Human Activity, J. H. Ausubel and A. K. Biswas, eds. London: Pergamon Press.
Miller, K. A. 1990. Water, electricity, and institutional innovation. In Climate
Change and U.S. Water Resources, P. E. Waggoner, ed. New York: John Wiley & Sons.
Miller, K. A., and M. H. Glantz. 1988. Climate and economic competitiveness: Florida freezes and the global citrus processing industry. Climatic Change 12:135–164.
Mitchell, J. F. B. 1988. Local effects of greenhouse gases. Nature 332:399–400.
Muscatine, L., L. R. McCloskey, and R. E. Marian. 1981. Estimating the daily contribution of carbon from zooxanthellae to coral animal respiration. Limnology and Oceanography 26:601–611.
National Research Council. 1976. Pest management. In Climate and Food. Washington, D.C.: National Academy of Sciences.
National Research Council. 1983. Changing Climate. Washington, D.C.: National Academy Press.
National Research Council. 1987. Responding to Changes in Sea Level: Engineering Implications. Washington, D.C.: National Academy Press.
National Research Council. 1990a. Managing Global Genetic Resources: The U.S. National Plant Germplasm System. Washington, D.C.: National Academy Press.
National Research Council. 1990b. Sea Level Change. Washington, D.C.: National Academy Press.
Natural Hazards Research and Applications Information Center (NHRAIC). 1990. NHRAIC Data Base. Natural Hazards Research and Applications Information Center, University of Colorado, Boulder.
Neilson, R. P., G. A. King, R. L. DeVelice, J. Lenihan, D. Marks, J. Dolph, B. Campbell, and G. Glick. 1989. Sensitivity of Ecological Landscapes and Regions to Global Climatic Change. Corvallis, Oreg.: U.S. Environmental Protection Agency.
Newlin, O. J. 1990. Testimony regarding amendments to strengthen the Plant Variety Protection Act. Presented to Department Operations, Research and Foreign Agriculture Subcommittee of the House Agriculture Committee, May 1, 1990.
Newman, J. E. 1982. Impacts of rising atmospheric carbon dioxide levels on agricultural growing seasons and crop water use efficiencies. In Environmental and Social Consequences of a Possible CO2-Induced Climate Change, volume II, Part 8. Washington, D.C.: Carbon Dioxide Research Division, U.S. Department of Energy.
Norby, R. J., J. Pastor, and J. M. Melillo. 1986. Carbon-nitrogen interactions in CO2 enriched white oak: Physiological and long-term perspectives. Tree Physiology 2:233–241.
Nordhaus, W. D. 1991. To slow or not to slow: The economics of the greenhouse effect, Economic Journal 101(407):920–937.
Oechel, W. C., and B. R. Strain. 1985. Native species responses to increased atmospheric carbon dioxide concentration. In Direct Effects of Increasing Carbon Dioxide on Vegetation, B. R. Strain and J. D. Cure, eds. Report DOE/ER-0238. Washington, D.C.: U.S. Department of Energy.
Ormrod, R. K. 1986. Evidence that California elderly are migrating southward. Sociology and Social Research 70(2):149–151.
Paine, R. T. 1980. Food webs: Linkages, interaction strength and community infrastructure. Journal of Animal Ecology 49:667–686.
Pampel, F. C., I. P. Levin, J. J. Louviere, R. J. Meyer, and G. Ruston. 1984. Retirement migration decision making: The integration of geographic, social and economic preferences. Research on Aging 6(2):139–162.
Parry, M. L. 1978. Climatic Change, Agriculture and Settlement. Folkestone, England: Dawson.
Parry, M. L., T. R. Carter, and N. T. Konijn, eds. 1988. The Impact of Climatic Variations on Agriculture. Dordrecht, The Netherlands: Kluwer Academic Publishers.
Patterson, D. T., and E. P. Flint. 1980. Potential effects of global atmospheric CO2 enrichment on the growth and competitiveness of C3 and C4 weed and crop plants. Weed Science 28:71–75.
Patterson, D. T., and E. P. Flint. 1982. Interacting effects of CO2 and nutrient concentration. Weed Science 30:389–394.
Pearcy, R. W., and O. Bjorkman. 1983. Physiological effects. In CO2 and Plants, E. R. Lemon, ed. American Association for the Advancement of Science Selected Symposium 84. Boulder, Colo.: Westview Press.
Peart, R. M., J. W. Jones, R. B. Curry, K. Boote, and L. H. Allen, Jr. 1989. Impact of climate change on crop yield in the southeastern USA: A simulation study. In The Potential Effects of Global Climate Change in the United States, Appendix C, Agriculture, J. B. Smith and D. A. Tirpak, eds. Washington, D.C.: U.S. Environmental Protection Agency.
Peeters, J. P., and J. T. Williams. 1984. Towards better use of gene-banks with special reference to information. Plant Genetic Resource Newsletter 60:22–32.
Peterson, D. F., and A. A. Keller. 1990. Irrigation. In Climate Change and U.S. Water Resources, P. E. Waggoner, ed. New York: John Wiley & Sons.
Raven, P. H. 1981. Research in botanical gardens. Botanische Jahrbuecher fuer Systematik Pflanzengeschichte und Pflanzengeographic 102:52–72.
Renas, S., and R. Kumar. 1982. Climatic conditions and migration: An econometric inquiry. Annals of Regional Science 17(1):69–78.
Revelle, R., and G. Kohlmaier. 1986. Increasing amplitudes of the seasonal CO2 cycle. In Climate-Vegetation Interactions, C. Rosenzweig and R. Dickinson, eds. Boulder, Colo.: University Center for Atmospheric Research.
Revelle, R. R., and P. E. Waggoner. 1983. Effects of a carbon dioxide-induced climatic change on water supplies in the western United States. In Changing Climate. Washington, D.C.: National Academy Press.
Riebsame, W. E., H. F. Diaz, T. Moses, and M. Price. 1986. The social burden of weather and climate hazards. Bulletin of the American Meteorological Society 67(11):1378–1388.
Riebsame, W. E., S. A. Changnon, and T. R. Karl. 1991. Drought and Natural Resources Management in the United States. Boulder, Colo.: Westview Press.
Ritchie, J. T., B. D. Baer, and T. Y. Chou. 1989. Effect of global climate change on agriculture: Great Lakes region. In The Potential Effects of Global Climate Change on the United States, Appendix C, Agriculture, J. B. Smith and D. A. Tirpak, eds. Washington, D.C.: U.S. Environmental Protection Agency.
Rogers, H. H., W. W. Heck, and A. S. Heagle. 1983. A field technique for the
study of plant responses to elevated carbon dioxide concentration. Journal of the Air Pollution Control Association 33:42–44.
Rose, E. 1989. Introduction. In The Potential Effect of Global Climate Change on the United States, Appendix C, Agriculture, J. B. Smith and D. A. Tirpak, eds. Washington, D.C.: U.S. Environmental Protection Agency.
Rose, G. 1966. Cold weather and ischaemic heart disease. Journal of Preventive and Social Medicine 20:97–100.
Rosenberg, N. J. 1982. The increasing CO2 concentration in the atmosphere and its implication on agricultural productivity. II. Effects through CO2-induced climate change. Climatic Change 4:239–254.
Rosenberg, N. J., and P. R. Crosson. 1991. Processes for identifying regional influences of and responses to increasing atmospheric CO2 and climate changeThe MINK Project: An Overview. Draft report prepared for U.S. Department of Energy under agreement 041460-A-K1. Washington, D.C.: U.S. Department of Energy.
Rosenberg, N. J., B. A. Kimball, P. Martin, and C. F. Cooper. 1990. From climate and CO2 enrichment to evapotranspiration. In Climate Change and U.S. Water Resources, P. E. Waggoner, ed. New York: John Wiley & Sons.
Rosenwaike, I. 1966. Seasonal variation of deaths in the United States, 1951–1960. Journal of the American Statistical Association 61:706–719.
Rosenzweig, C. 1989. Potential effects of climate change on agricultural production in the Great Plains: A simulation study. In The Potential Effects of Global Climate Change in the United States, Appendix C, Agriculture, J. B. Smith and D. A. Tirpak, eds. Washington, D.C.: U.S. Environmental Protection Agency.
Rotton, J. 1983. Angry, sad, happy? Blame the weather. U.S. News and World Report 95:53.
Santelices, B. 1989. Algas Marinas de Chile: Distribucion, Ecologia, Utilizacion y Diversidad. Santiago: Ediciones Universidad Catolica de Chile.
Schaake, J. C. 1990. From climate to flow. In Climate Change and U.S. Water Resources, P. E. Waggoner, ed. New York: John Wiley & Sons.
Schaefer, K. E., ed. 1958. Man's Dependence on the Earthly Atmosphere. New York: Macmillan.
Schelling, T. C. 1983. Climatic change: Implications for welfare and policy. In Changing Climate. Washington, D.C.: National Academy Press.
Schneider, S. H., P. H. Gleick, and L. O. Mearns. 1990. Prospects for climate change. In Climate Change and U.S. Water Resources, P. E. Waggoner, ed. New York: John Wiley & Sons.
Schoenbeck, M., and T. A. Norton. 1978. Factors controlling the upper limits of fucoid algae on the shore. Journal of Experimental Marine Biology and Ecology 31:303–313.
Serow, W. 1987. Why the elderly move: Cross-national comparisons. Research on Aging 9(4):582–597.
Serra B., R. 1987. Impact of the 1982–83 ENSO on southeastern Pacific fisheries, with an emphasis on Chilean fisheries. In Climate Crisis: The Societal Impacts Associated with the 1982–83 Worldwide Climate Anomalies, M. Glantz, R. Katz, and M. Krenz, eds. Boulder, Colo.: National Center for Atmospheric Research.
Sheer, D. P. 1985. Managing water supplies to increase water availability. U.S.G.S. Annual Water Survey 1985:101–112.
Smith, D. C., H. H. Borns, W. R. Baron, and A. E. Bridges. 1981. Climatic stress and Maine agriculture, 1785–1885. In Climate and History: Studies in Past Climates and Their Impact on Man, T. M. L. Wigley, M. J. Ingram, and G. Farmer, eds. New York: Cambridge University Press.
Smith, D. M. 1986. The Practice of Silviculture, 8th ed. New York: John Wiley & Sons.
Smith, J. B., and D. A. Tirpak, eds. 1989. The Potential Effects of Global Climate Change on the United States. Washington, D.C.: U.S. Environmental Protection Agency.
Smith, W. H. 1981. Air Pollution and Forests. New York: Springer-Verlag.
Solley, W. B., C. F. Merk, and R. R. Pierce. 1989. Estimated Use of Water in the United States in 1985. U.S. Geological Survey Circular 1004. Washington, D.C.: U.S. Geological Survey.
Solomon, A. M. 1986a. Linking GCM climate data with data from static and dynamic vegetation models. In Climate-Vegetation Interactions, C. Rosenzweig and R. Dickinson, eds. Report UCAR OIES-2. Boulder, Colo.: University Center for Atmospheric Research.
Solomon, A. M. 1986b. Transient response of forests to CO2-induced climate change: Simulation modeling experiments in eastern North America. Oecologia 68:567–579.
Solomon, A. M., and M. L. Tharp. 1985. Simulation experiments with late-Quaternary carbon storage in mid-latitude forest communities. Geophysical Monograph 32:235–250.
Solomon, A. M., and T. Webb. 1985. Computer-aided reconstruction of late-Quaternary landscape dynamics. Annual Review of Ecology and Systematics 16:63–84.
Solomon, A. M., and D. C. West. 1986. Simulating forest responses to expected climate change in eastern North America: Applications to decision-making in the forest industry. In Rising Carbon Dioxide and Changing Climate: Forest Risks and Opportunities, W. E. Shands and J. Hoffman, eds. Washington, D.C.: The Conservation Foundation.
Sontaniemi, E., U. Vuopala, E. Huhta, and J. Takkunem. 1970. Effect of temperature on hospital admissions for myocardial infarction in a subarctic area. British Medical Journal 4:150–151.
Sorensen, R. M., R. N. Weisman, and G. P. Lennon. 1984. Control of erosion, inundation, and salinity intrusion caused by sea level rise. In Greenhouse Effect and Sea Level Rise: A Challenge for This Generation, M. C. Barth and J. G. Titus, eds. New York: Van Nostrand Reinhold.
Spauling, W. G., and L. S. Graumlich. 1986. The last pluvial climatic episodes in the deserts of southwestern North America. Nature 320:441–444.
Stinner, B. R., R. A. J. Taylor, R. B. Hammond, F. F. Purrington, and D. A. MacCartney. 1989. Potential effects of climate change on plant-pest infestations. In The Potential Effects of Global Climate Change on the United States, Appendix C, Agriculture, J. B. Smith and D. A. Tirpak, eds. Washington, D.C.: U.S. Environmental Protection Agency.
Straka, T. J., W. F. Watson, and M. Dubois. 1989. Costs and cost trends for forestry practices in the South. Forest Farmer Manual 1989:8–14.
Tabbarah, R. 1988. Prospects for international migration. In International Migration Today, Volume 2, Emerging Issues, C. Stahl, ed. Paris: United Nations Educational, Scientific, and Cultural Organization.
Tannehill, I. R. 1947. Drought, Its Causes and Effects. Princeton, N.J.: Princeton University Press.
The South Commission. 1990. The Challenge to the South. Oxford, United Kingdom: Oxford University Press.
Thompson, L. M. 1988. Effects of changes in climate and weather variability on the yield of corn and soybeans. Journal of Production Agriculture 1:20–27.
Thorarinsson, S. 1944. Tefrokronologiska studier pa isaland. Geografiska Annaler: 1–127.
Tickell, C. 1986. Climatic Change and World Affairs. Lanham, Md.: University Press of America, Inc.
Tickell, C. 1990. Human effects of climate change. Lecture to the Royal Geographic Society, London, England, March 26, 1990.
Tissot, B. 1990. Geographic Variation and Mass Mortality in the Black Abalone; the Roles of Development and Ecology. Ph.D. dissertation. Oregon State University, Corvallis.
Tissue, D. T., and W. C. Oechel. 1987. Response of Eriophorum vaginatum to elevated CO2 in the Alaskan tussock tundra. Ecology 68:401–410.
Titus, J. G., and M. C. Barth. 1984. An overview of the causes and effects of sea level rise. In Greenhouse Effect and Sea Level Rise: A Challenge for This Generation, M. C. Barth and J. G. Titus, eds. New York: Van Nostrand Reinhold.
Titus, J. G., and M. S. Greene. 1989. An overview of the nationwide impacts of sea level rise. In The Potential Effects of Global Climate Change on the United States, Appendix B, Sea Level Rise, J. B. Smith and D. A. Tirpak, eds. Washington, D.C.: U.S. Environmental Protection Agency.
Titus, J. G., C. Y. Kuo, M. J. Gibbs, T. B. LaRoche, M. K. Webb, and J. O. Waddell. 1987. Greenhouse effect, sea level rise, and coastal drainage systems. Journal of Water Resources Planning and Management 113:2.
Tobey, J., J. Reilly, and S. Kane. 1990. An Empirical Study of the Welfare Effects of Climate Change on World Agriculture. Washington, D.C.: Research and Technology Division, Economic Research Service, U.S. Department of Agriculture.
Tseng, C. K., ed. 1984. Common Seaweeds of China. Beijing: Science Press (Kugler Publications bv, Amsterdam).
United Nations. 1989a. Prospects for World Urbanization 1988. New York: Department of International Economic and Social Affairs, United Nations.
United Nations. 1989b. World Population Prospects 1988. New York: Department of International Economic and Social Affairs, United Nations.
Unnewehr, D. L. 1989. Surviving the Storm. Boston, Mass.: All-Industry Research Advisory Council/National Committee on Property Insurance.
Urban, D. L., and H. H. Shugart. 1989. Forest response to climate change: A simulation study for southeastern forests. In The Potential Effects of Global
Climate Change on the United States, Appendix D, Forests, J. B. Smith and D. A. Tirpak, eds. Washington, D.C.: U.S. Environmental Protection Agency.
U.S. Bureau of the Census. 1987. Statistical Abstract of the United States: 1988. 108th ed. Washington, D.C.: Government Printing Office.
U.S. Bureau of the Census. 1987. Statistical Abstract of the United States: 1989. 109th ed. Washington, D.C.: Government Printing Office.
U.S. Committee for Refugees. 1990. World Refugee Survey: 1988 in Review. New York: U.S. Committee for Refugees, American Council for Nationalities Service.
U.S. Department of Agriculture. 1986. Agricultural Statistics, 1986. Washington, D.C.: U.S. Department of Agriculture.
U.S. Department of Agriculture. 1989a. Agricultural Statistics, 1989. Washington, D.C.: U.S. Department of Agriculture.
U.S. Department of Agriculture. 1989b. Inventory of Agricultural Research, Fiscal Year 1988. Washington, D.C.: U.S. Department of Agriculture.
U.S. Department of Commerce. 1975. Historical Statistics of the U.S., Colonial Times to the 1970s. Washington, D.C.: U.S. Department of Commerce.
U.S. Department of Energy. 1989a. A Compendium of Options for Government Policy to Encourage Private Sector Responses to Potential Climate Change. Washington, D.C.: U.S. Department of Energy.
U.S. Department of Energy. 1989b. Energy and Climate Change. Report of the DOE Multi-Laboratory Climate Change Committee, December. Washington, D.C.: U.S. Department of Energy.
U.S. Department of State. 1982. Proceedings of the U.S. Strategy Consensus Conference on Biological Diversity, November 16–18, 1981. Publication 9262, International Organization and Conference Series 300. Washington, D.C.: U.S. Department of State.
U.S. Office of Management and Budget. 1989. Budget of the United StatesFiscal Year 1990. Washington, D.C.: U.S. Government Printing Office.
U.S. Office of Technology Assessment. 1983. Wood Use: U.S. Competitiveness and Technology. Report OTA-ITE-210. Washington, D.C.: U.S. Office of Technology Assessment.
U.S. Office of Technology Assessment. 1984. Wetlands: Their Use and Regulation. Report OTA-0-206. Washington, D.C.: U.S. Office of Technology Assessment.
Utterstrom, G. 1955. Climatic fluctuations and populations problems in early history. Scandinavian Economic History Review III(1):1–47.
Valiela, I. 1984. Marine Ecological Processes. New York: Springer-Verlag.
Voss, P., R. Gunderson, and R. Manchin. 1988. Death taxes and elderly interstate migration. Research on Aging 10(3):420–450.
Waddell, J. O., and R. A. Blaylock. 1987. Impact of Sea Level Rise on Gap Creek Watershed in the Fort Walton Beach, Florida Area. Washington, D.C.: U.S. Environmental Protection Agency.
Waggoner, P. E. 1960. Forecasting epidemics. In Plant Pathology, An Advanced Treatise, Volume 3, J. G. Horsfall and A. E. Dimond, eds. New York: Academic Press.
Waggoner, P. E. 1983. Agriculture and a climate changed by more carbon dioxide. In Changing Climate. Washington, D.C.: National Academy Press.
Waggoner, P. E. 1990. Environmental change in food supply. Proceedings of 23rd Annual Convention of Canola Council of Canada, Ottawa, March 19–21, pp. 63–71.
Waggoner, P. E., and B. Bravdo. 1967. Stomata and the hydrologic cycle. Proceedings of the National Academy of Sciences 57:1096–1102.
Waggoner, P. E., J. L. Monteith, and G. Szeicz. 1964. Decreasing transpiration of field plants by chemical closure of stomata. Nature 201:97–98.
Walker, J. C., T. R. Miller, G. T. Kingsley, and W. A. Hyman. 1989. Impact of global climate change on urban infrastructure. In The Potential Effects of Global Climate Change on the United States, Appendix H, Infrastructure, J. B. Smith and D. A. Tirpak, eds. Washington, D.C.: U.S. Environmental Protection Agency.
Wall, J. 1989. Potential effects on tourism and recreation in Ontario. In First U.S. Canadian Symposium on Impacts of Climate Change on the Great Lakes Basin. Downsview, Ontario: Environment Canada.
Warrick, R. A. 1980. Drought in the Great Plains: A case study of research on climate and society in the USA. In Climatic Constraints and Human Activities, J. H. Ausubel and A. K. Biswas, eds. Oxford: Pergamon Press.
Webb, T. 1986. Vegetation change in Eastern North America from 18,000 to 500 Yr B.P. In Climate-Vegetation Interactions, C. Rosenzweig and R. Dickinson, eds. Report UCAR OIES-2. Boulder, Colo.: University Center for Atmospheric Research.
Weihe, W. H. 1979. Climate, health and disease. In Proceedings of the World Climate Conference. Geneva: World Meteorological Organization.
Weltzien, H. C. 1978. Geophytopathology. In Plant Disease, an Advanced Treatise, J. G. Horsfall and E. C. Cowling, eds. New York: Academic Press.
Western, D., and M. Pearl. 1989. Conservation for the Twenty-First Century. New York: Oxford University Press.
Wharton, C. H. 1970. The Southern River Swamp: A Multiple-Use Environment. Athens: School of Business Administration, Georgia State University.
Wildavsky, A. 1980. Richer is safer. Public Interest 60(Summer):23–39.
Wilhite, D. A. 1983. Government response to drought in the United States: With particular reference to the Great Plains. Journal of Climate and Applied Meteorology 22:40–50.
Wilkes, G. 1984. Germplasm conservation toward the year 2000: Potential for new crops and enhancement of present crops. In Plant Genetic Resources, C. W. Yeatman, D. Kafton, and G. Wilkes, eds. American Association for the Advancement of Science Selected Symposium 87. Washington, D.C.: American Association for the Advancement of Science.
Wilkins, C. P., and J. H. Dodds. 1983. The application of tissue culture techniques to plant genetic conservation. Scientific Progress 68:259–284.
Wilks, D. S. 1988. Estimating the consequences of CO2-induced climatic change on North American grain agriculture using general circulation model information. Climatic Change 13:19–42.
Wilson, E. O., ed. 1988. Biodiversity. Washington, D.C.: National Academy Press.
Wilson, E. O. 1989. Threats to biodiversity. In Managing Planet Earth, Special Issue. Scientific American 260:108–116.
Wong, S. C. 1979. Elevated atmospheric partial pressure of CO2 and plant growth. Oecologia 44:68–74.
World Commission on Environment and Development. 1987. Our Common Future. New York: Oxford University Press.
World Health Organization. 1990. Potential Health Effects of Climatic Change Geneva: World Health Organization.
Zahner, R. 1955. Soil water depletion by pine and hardwood stands during a dry season. Forestry Science 1:258–264.
Zolberg, A. 1989. Escape from Violence: Conflict and the Refugee Crisis in the Developing World. New York: Oxford University Press.