Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Climate Change in the Great Lakes Region WALTRAUD A. R. BRINKMANN University of Wisconsin-Madison The Great Lakes region is small enough so that temperature anomalies are generally of the same sign basin-wide. The long-term trends in temperature that have occurred in the Great Lakes region during this century have been very similar to the trends in the average surface air temperature for the Northern Hemisphere. The early part of this century was relatively cold; this was followed by a period of above average temperatures from the 1930s to the mid to late 1950s, followed by a return to lower temperatures. There is some indication of a swing back toward higher temperatures in recent years. Precipitation changes in the Great Lakes basin have been more complex because precipitation is generally much more variable over space as well as over time as compared to temperature. Specifically, there is a climatologicalâa precipitationâboundary that divides the region into a northwestern and a southeastern portion; the boundary is a result of shifts in the importance of the two major storm tracks across the basin. This has led to precipitation trends over the two portions that have often been in opposition and sometimes been similar. Climate changes determine natural changes in that portion of the total water supply to a lake that is generated within the lake basin itself (the sum of lake precipitation plus runoff minus lake evaporation)âreferred to as "net basin supplies." Changes in pre- cipitation onto the basin will result in changes in lake precipitation 115
116 GREAT LAKES WATER LEVELS as well as in the amount of runoff. Changes in temperature will also have an effect on net basin supplies since temperature is in a complex way related to lake evaporation. However, evaporation removes less than half of the natural water input (from lake precipitation and runoff) into the Great Lakes. It is therefore precipitation that drives the system. There is, consequently, a poor and sometimes opposite association between the net basin supplies to Lake Superior and sup- plies to the lower lakes (particularly for moderate supplies); and it is this that makes Lake Superior useful as a reservoirâone of the bases of current lake level management. Extreme net basin supply events have, however, been basin-wide at times, and these have led to the low waters of the 1930s and 1960s, and the high waters of the 1950s, 1970s, and 1980s. What about the future? We are still not very good at making forecasts for decades in advance. Perhaps the recent trend of alter- nating extremes in water supplies will continue; but we do not really know. At a longer time scale, the effect of increasing atmospheric carbon dioxide concentrations will become important. But here, too, there is uncertainty. The uncertainty is not with the trend in carbon dioxide but with the effect of this on the climate over the Great Lakes region. All general circulation models predict an increase in temper- ature over the region with a doubling of carbon dioxide, although the magnitude of the predicted increase varies somewhat from model to model. An increase in temperature could lead to an increase in evaporation. Concerning precipitationâthe climate parameter that drives the systemâthe models do not even agree on the sign: some predict an overall increase, and others predict an increase for one season and a decrease for another. The predicted changes in precipitation are to some degree re- lated to predicted shifts in circulation patterns. Therefore, another important question to consider is how circulation shifts will affect the lack of association or sometimes out-of-phase association between the upper and lower lakes that has been a characteristic of the net basin supplies to the lakes.
Climate Change: The Knowns and Unknowns STANLEY A. CHANGNON, JR. Chief Emeritus Illinois State Water Survey I have considered herein the climate change issue within the context of the Great Lakes. Hence, I have focused my comments more on regional than global scale change and I have reviewed some major knowns and unknowns from my perspective. I am an applied climatologist and have principally studied the climate of the past 100 years in the Midwest and North America; I am not a global climate change expert nor modeler. As an applied scientist for the past 35 years, I have interacted with a wide variety of private and public sector representatives at the local, state, regional, national, and international levels. Thus, my comments about climate change are based on my own scientific research and on observations about what current knowledge of climate change means to those who function in the region. Obviously, we know that the earth without human influences has experienced some extremely wide shifts in climate; here where Chicago stands there have been past epochs with tropical forests and others with glaciers nearly a mile thick; hence, people and their waste products are not needed to change the climate dramatically. Natural forces, as yet poorly understood, have made and are making the climate change. I see four major uncertainties relating to the issue of global climate change that have specific reference to the Great Lakes (or any other region). These are: 117
118 GREAT LAKES WATER LEVELS 1. In what way, in the near climatic future (50 to 100 years), will various climate conditions depart from their 50- to 100-year current averages, and more importantly, what will be the change (if any) in interannual and interdecadal variability? 2. How rapidly can or will the climate conditions shift, either in their averages and/or in their extremes? 3. Can man somehow affect the natural climatic processes and will natural forces tend to counteract or amplify the signal? 4. Can our society adjust its most weather-sensitive activities (agriculture and water resources) to the changes? Most experts in impacted sectors believe that if the future shifts are moderately slow (multidecadal) and the future variability does not become too great, satisfactory adjustments can be made. ... Is it true? The three most common questions posed to me by those in weather-sensitive activities are: Will the climate change that some atmospheric scientists are predicting occur? Why is there no speci- ficity at the regional (action) scale over what weather conditions will change, the magnitude of change, and when? Third, will the changes really bring on severe impacts and require such adjustments that society cannot adapt to the change in a satisfactory manner? What do we really know? Based on the past 140 years of wa- ter level records on the Great Lakes, we know that the climate has fluctuated considerably, sufficient to produce a 3-ft range in lake levels around the long-term average. We also know that the record extremes have occurred quickly, within a period of 20 years, and we also know that shorelines and human activities have been consider- ably impacted by these shifts. Furthermore, we know that climate conditions in the Great Lakes Basin prior to the past 100 years have been more extreme than anything we have sampled. In fact, during the last 2,000 years climate conditions have been both much wetter/cooler and warmer/drier than anything we have experienced since 1860. Interestingly, if one uses Larson's historical lake level reconstructions, one can project that the Great Lakes Basin is in a period that will be wetter and cooler over the next 100 to 200 years. This possibility for natural conditions is in opposition to a future drier climate that the carbon dioxide models seem to predict for the area over the next 100 years. If both tendencies alone are correct, how will the two interact? This is an interesting scientific issue and one that is unresolved. We further know that the climate in the Great Lakes Basin over the past 100 years reveals a distinct trend, particularly emphasized
CLIMATE CHANGE: THE KNOWNS AND UNKNOWNS 119 during the last 50 years, to a cloudier, cooler, and wetter regime. The past 15 years have been the wettest period in the modern history of the Great Lakes; as a result, two periods of record high levels have been experienced (1970s and 1980s). Given this evidence, it is not surprising that it is difficult for those impacted by weather in the region to get too concerned over a predicted change in climate of uncertain magnitude and timing. Most climatologists agree that current science lacks the capa- bility to make climate predictions for periods beyond a few months ahead, and that without this capability, the past is the best predictor of the future. The past at this time is not indicating a shift in climate in the direction that many climate modelers claim will occur. Thus, predictions of an indeterminant change in regional climate based on models limited by the data included and the assumptions used ap- pear out of place and without much skill. But, we know that there have been marked regional and large-scale changes in climate. What do we know more specifically about inadvertent (man- made) climate change? Major urban areas in the basin such as Chicago and Detroit, modify every aspect of their climate and change clouds and precipitation 100 km beyond them. We also know that large industries, cooling facilities, and jet contrails increase clouds, but if other man-made climate changes exist, they are lost in the noise of natural variability of climate. I believe that those who predict a man-induced climate change of severe proportions over the next 25 to 75 years must provide more compelling evidence than exists now and must become much more definitive about the spatial and temporal features of change before climate change will become a major regional issue leading to action. When will the change occur? Will it be warmer or cooler, wetter or drier, over all or portions of the basin? Will there be more or less variability, and greater extremes of wetness and dryness than those of the past 100 years? Convincing answers to such questions must be provided before major decisions and plans are made.
Economic Research, The Greenhouse Effect, and Fluctuating Great Lakes Water Levels RICHARD F. KOSOBUD University of Illinois at Chicago Economic research into long-run climate change induced by at- mospheric trace gas accumulations has sought to develop frameworks for appraising the contributions of economic activity to these accu- mulations and the feedback impacts of changing climate on economic activity and welfare. The ultimate aim is to evaluate policy choices. Three frameworks, or models, have been developed that continue to be actively studied and that have potential application to a study of changing climate implications for hydrologic systems such as the Great Lakes. My object is to describe, briefly, these models and their potential applications. I must state right off that, in my view, the results to date of research just begun do not lead to deterministic pre- dictions of catastrosphic difficulties nor to policy prescriptions such as heavy taxes on fossil fuels; rather, these frameworks at present pro- vide an interesting way to think about the long-run future, shrouded in many uncertainties as it is and smudged with our present state of ignorance. These frameworks can also direct our attention to critical areas for our research effort. Economists have so far applied mathematical programming (MP), computable general equilibrium (CGE), and midlevel sectoral (MS) models to possible greenhouse effects and their consequences. Each has strengths and weaknesses that I can hint at by describing the essential stages of the problem that the models must investigate. First, I will introduce some economic jargon because here it is useful: I will define an environmental quality function to be a relationship 120
ECONOMIC RESEARCH 121 showing how economic activity affects the states of nature or hu- man affairs beyond the market or central planner calculations. The economic models must be extended to take these externalities into account. I will define an environmental damage function to be a relationship showing how we could, in principle, attach benefits or costs to these external impacts, including impacts on Great Lakes variables. All three models contain detailed specifications showing how, over the long run, global and regional economic activity consumes fossil fuels that emit carbon dioxide and perhaps other trace gases into the atmosphere. To projections of economic activity of these economic models must be added environmental quality submodels indicating how trace gases accumulate, how this leads to changing climate, and how changing climate affects the Great Lakes, among other systems. Finally, an environmenta1 damage submodel must be added to permit benefit-cost calcuations and policy appraisal. All this is no small order! The MP model maintained at the University of Illinois contains less detail on the economy but more information on energy technol- gies and changing least-cost use of these resources over time as prices rise, e.g., for depleted oil. The MS model maintained at the Oak Ridge National Laboratory contains great detail on energy sectors and regions, but must be solved for future time paths in stages. The CGE model maintained at Vanderbilt University permits appraisal of impacts on different income groups, but is limited as to size. All models permit projections of long-run economic growth paths with attendant demands for energy consumption and trace gas emis- sion. Aspect of the models can be varied to reflect the range of expert opinion so that the range of projections that result enable us to measure uncertainty, a vital task in this uncharted area. Granting these uncertainties, how do we specify the environment quality and damage functions? Concentrations of trace gases in the atmosphere depend on com- plex interactions among the atmosphere, oceans, and biota. Most of the models, for example, assume a carbon cycle in which slightly less than half of the carbon dioxide emitted remains in the atmo- sphere. Trace gas accumulations give rise to the well-known green- house effect and hence to changing climate. Many uncertainties re- main about how the stochastic processes of climate will be affected, and what weather manifestation changes, such as observable pre- cipitation, temperature, cloudiness, and wind patterns, will result. Much research effort is being put into general circulation modeling of changing climate both globally and regionally, and economists look
122 GREAT LAKES WATER LEVELS to further results with great interest because adding this submodel is crucial. For our purposes, what is important is the impact of these changes on climate-sensitive sectors that are likely to include agricul- ture and hydrological systems, among others. What we can add to our framework models are those hydrological models that allow for climate variable impacts on water levels, nm-offs, and other charac- teristics. Continuing to focus on impacts on the hydrologic system, we may append to our frameworks a specification of the water supply dependence on such variables as precipitation and evaporation (the latter in turn depending on temperature). This relationship can be intricate, and its parameters can be highly uncertain. Many observ- able manifestations of climate besides precipitation and temperature may be important, for example, cloud cover and wind patterns. Our framework permits experimentation with alternate specifications so that we can study the consequences of changing one feature, and then another in order to project long-run implications. This ability to simulate alternatives is the great merit of our approach, given the enormous gaps in knowledge. My own survey of hydrological research indicates a number of models under development that we can add to and test within our frameworks. A particularly difficult hydrological issue is the variabil- ity question of changing climate and its relationship to fluctuating water levels. The variance rather than the level of these variables may be important, and it will be challenging to estimate changes of variances within our projection framework. However, we can exam- ine the historical record and we can by careful study "transplant" to the Great Lakes area other regional climates that may be closer to future long-run patterns and in this way "think about" fluctuating water levels. Our final stage is to estimate benefits and costs of these altered hydrologic, and other sectoral, patterns and hence provide a guide to appraisal of public policies. Most research into the greenhouse effect has gone into earlier stages such as the carbon cycle and the issue of changing climate. I argue that what is needed now is an increase in the share of effort devoted to studying the later stage long-run impacts of changing climate on systems such as the Great Lakes and the consequences for human welfare and for policy. Preliminary work has begun. I suggest that sustained research is called for. The economic frameworks I have outlined provide, in my view, a promising approach.
Climate Change and Great Lakes Levels MARIE E. SANDERSON University of Windsor The Great Lakes Institute of the University of Windsor has com- pleted two research contracts on topics related to climate change and Great Lakes levels. The first was for the Atmospheric Environment Service (Environment Canada) on possible climate change and the impact on lake levels, navigation, and hydroelectric power gener- ation on the Great Lakes (summary appeared in Climate Change Digest #3, 1987). The second (just completed) was for the Don- ner Canadian Foundation on future lake levels and the hydrologic, environmental, and political impacts. In the first study, the Canadian Climate Centre of Environment Canada provided to the researchers a projection of climate conditions in the Great Lakes Basin with an atmospheric carbon dioxide (CO2) concentration twice that of preindustrial times (2 x CC^). The data on monthly temperature and precipitation represented the modified output of the Goddard Institute of Space Studies' (GISS) General Circulation Climate Model. The average annual warming in the Great Lakes Basin projected by this model is approximately 4.5°C, slightly more in winter and less in summer. Annual precipitation is projected to increase approximately 8 percent for points in the central and western basin, but to decrease by 3 to 6 percent in the eastern basin. There are, of course, many assumptions and uncertainties in such large-scale models, and the outputs are very tentative estimates of future climate for any specific area. An additional problem in 123
124 GREAT LAKES WATER LEVELS applying the output of the model to the Great Lakes Basin is the fact that there are only 10 data points in or near the basin. Our researchers were also provided with three sets of data from a Great Lakes hydrologic model from the Inland Waters Directorate, Environment Canada. The years 1900-1976 were used as the basis- of-comparison period (BOC). The BOC data showed average annual lake levels and flows that would have occurred during the period 1900- 1976 under current diversions, regulation practices, and physical conditions of the lakes and connecting channels (diversions include 142 m3s~1 into Lake.Superior via Long Lac and the Ogoki River, 91 m3s~1 out of Lake Michigan at Chicago and 198 m3s~1 from Lake Ontario through the Welland Canal). Thus, with these assumptions, the variability seen under BOC conditions is due to changes in climate only. During the BOC period, the variability of average annual levels ranged from 0.99 m for Lake Superior to 2.14 m for Lake Ontario. A second set of data gave the levels and flows (for the period 1900- 1976) that would occur under the GISS 2 x CO2 climate scenario. The average level of Lake Superior was seen to decrease by 20 cm, of Michigan-Huron by 60 cm, Erie by 44 cm, and Ontario by 85 cm. A third set of data gave lake level and flow data for 2 x CO2 climate plus the impact of increased consumptive use of Great Lakes water as projected by the International Joint Commission in 1981 for the year 2035. In this scenario, average lake levels decreased by an additional 10 to 20 cm. The graph (Figure 1) shows the levels that would have occurred on Lake Erie during the period 1900-1976 under these three scenarios. It is seen that the frequency of occurrence of extreme low levels as in the 1930s and 1960s could increase to 75 percent of the time. For the Donner study, we continued and expanded our modeling work on climate change and lake levels. We used the Geophysical Fluid Dynamics Laboratory (GFDL) as well as the GISS model output and refined the net basin supply models. In the Great Lakes Basin, the projected average GFDL monthly temperature change is 1.5°C less than in the GISS model. Our refined runoff model gives monthly as well as the annual runoff amounts previously obtained for the GISS model. For Lake Erie we found that runoff during the BOC period averaged 81 cm depth on the lake surface, whereas under GFDL conditions it was 68 cm, and under GISS condition, 61 cm. New estimates of over-lake evaporation under the two climatic change scenarios were also determined, and increases over BOC conditions were seen, especially in the high-evaporation period
CLIMATE CHANGE AND GREAT LAKES LEVELS 125 174.50 174.25 174.00 173.75 173.50 173.25 173.00 172.75 172.50 172.25 172.00 Basis of Comparison ' Climatic Change Climatic Chang* plus Consumptive Use 1900 1910 1920 1930 1940 FIGURE 1 Lake Erie levels, 1900-1976. 1950 1960 1970 1980 JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC FIGURE 2 Lake Erie net basin supply. Monthly averages, 1958-1983.
126 GREAT LAKES WATER LEVELS September-December. Overall, net basin supplies were found to de- crease (as in Figure 2) with GFDL showing less decrease than GISS projections. In addition, we examined the potential effects on Great Lakes levels, under climate change scenarios, of large-scale diversions into the basin, such as the Grand Canal scheme. We found that a diversion of 1560 m3s 1 into Lake Huron from James Bay would increase Lake Erie levels and compensate for the lowering of the levels under the GISS scenario. Under the GFDL scenario, Lake Erie levels would be raised 45 cm above historic levels, thus introducing the possibility of the export of water southward. It is seen from the above results that our Great Lakes hydrologic response model permits many different future scenarios of Great Lakes levels to be explored.
Preliminary Results From EPA Study of Impacts of Global Warming on the Great Lakes Basin JOEL B. SMITH U.S. Environmental Protection Agency In 1986, Congress asked the Environmental Protection Agency to conduct two studies on global warming. The first study would examine options to limit emissions of greenhouse gases. The second study, which will be referred to as the Effects Report, ". . . should examine the health and environmental effects of climate change. This study should include, but not be limited to, the potential impacts on agriculture, forest, wetlands, human health, rivers, lakes and estuaries as well as other ecosystems and societal impacts." My presentation to the Colloquium on Great Lakes Water Levels will review some of the preliminary results from the Effects Report relevant to the Great Lakes Basin. Since the results have not been peer reviewed, they will not be available for citation or quotation. GOALS FOB THE EFFECTS BEPORT The goal of the Effects Report is to try to give a sense of the possible direction of changes from a global warming as well as the magnitude. We are examining some of the following issues: ⢠The range of effects under different warming scenarios; ⢠Sensitivities of systems to changes in climate; ⢠Regional differences among effects; ⢠Interactions among effects on a regional level; 127
128 GREAT LAKES WATER LEVELS Regional Case Studies Core Analytic Areas Outputs Southeast Great Lakes Great Plains California Report to Congress Climate Change Scenarios Water Resources Agriculture Forests Sea Level Rise Biodiversity Health Infrastructure Air Pollution Policy Research Plan National Studies Models/ Data bases Agriculture Sea Level Rise Health Energy Demand Policy Others Risk Communication Workshops FIGURE 1 Elements of Effects Report. ⢠Uncertainties; and ⢠Policy implications. ELEMENTS OF THE EFFECTS REPORT The elements of the Effects Report are displayed in Figure 1. We will use climate change scenarios (described below) to examine potential changes in core analytic areas on a regional and, in some cases, national level. We are studying impacts in these analytic areas in the Great Lakes, California, the Southeast, and the Great Plains. In addition, we are conducting national studies on agriculture, sea level rise, energy demand, human health, and other issues. The Environmental Protection Agency (EPA) plans to sum- marize and discuss the implications of these studies in a report to Congress. We intend to deliver the report before the end of 1988. The EPA will continue research into the impacts of climate change after this report is completed. Our Office of Research and Development is preparing a research plan to help guide our efforts. METHODOLOGY AND SCENARIOS These studies are being conducted by leading researchers in academia and government. They will generally use "off-the-shelP models of the relationship between climate and their analytic area. For example, the Great Lakes Environmental Research Lab (GLERL)
IMPACTS OF GLOBAL WARMING 129 is using its hydrologic model of the Great Lakes to study possible impacts of climate change on lake levels. In some cases, we have engaged the services of experts to conduct literature reviews and workshops on specific issues. We developed a consistent set of scenarios to be used in the analysis of the potential impacts of climate change. The researchers are using these scenarios as inputs to their models. The scenarios combine outputs from General Circulation Models (GCMs) with his- toric climate data.1 Specifically, the scenarios use average monthly outputs from the Goddard Institute for Space Studies (GISS), the Geophysical Fluid Dynamic Laboratory (GFDL), and the Oregon State University (OSU) GCMs, which have doubled concentrations of greenhouse gases. Those results are combined with actual meteo- rologic data from 1951 to 1980. This assumes that average temper- ature, precipitation, winds, and other factors change, but that daily, interannual, and spatial climate variability remain the same. Goddard Institute for Space Studies also has run a transient experiment in which trace gases are gradually increased from the 1950s until the middle of the next century. This transient run shows how climate may change in the near future. We also combine this run with historic data to create a transient scenario. In some cases we will also use the 1930s as an analog of short-term warming. GREAT LAKES CASE STUDY The Great Lakes are the largest body of freshwater in the world. The lakes are a source of water for consumption, transportation, hydropower, and recreation. Recent high lake levels demonstrate the sensitivity of the lakes to changes in climate. The lower lakes, especially Lake Erie, have had serious pollution problems, but have shown improvement in recent years. The focus of the Great Lakes case study is on the lakes them- selves, but we also examine other impacts of climate change on the region. Changes will be estimated for the lake levels for all lakes and for ice cover on Lakes Superior and Erie. The potential effects on the thermal structure of Lakes Erie and Michigan will be examined and results will be used to study possible impacts on Great Lakes fish. We will look at the impacts of recent high and low lake levels on Great Lakes shorelines, and the potential impacts of climate change on hydropower in the region and on shipping on the lakes. An agriculture study will give crop yield estimates for the region.
ISO GREAT LAKES WATER LEVELS In addition, we are looking at potential response to climate change by farmers, such as extending the growing season or adding irrigation. Another study examines changes in nonpoint source runoff from farms. Possible changes in forests in the region will be studied through the use of stand simulation models to demonstrate long-term equi- librium changes in species composition, and through the analysis of paleovegetational data to estimate the response of forests to past climatic change. Finally, we will attempt to identify the institutions and policies affected by all of these changes. PROJECTS The specific projects in support of this case study are as follows: Title Researcher Institution Changes in Lake Levels Croley Changes in Ice Cover Assel Thermal Structure of McCormick Lake Michigan Thermal Structure of Blumberg Lake Erie Great Lakes Fisheries Magnuson Regier Impacts on Shorelines Changnon Impacts on Shipping Keith Crop Yields Ritchie Farm Level Response Easterling Stand Simulation Botkin Modeling Pollen Response Surfaces Overpeck Seedling Distribution Davis Policy Implications Brah Great Lakes Environmental Research Lab Great Lakes Environmental Research Lab Great Lakes Environmental Research Lab Hydroqual, Inc. University of Wisconsin University of Toronto Illnois State Water Survey Engineering Computer Optecnomics Michigan State University Illnois State Water Survey University of California, Santa Barbara Lamont-Doherty Geological Observatory University of Minnesota Center for the Great Lakes
IMPACTS OF GLOBAL WARMING 131 NOTE 1. The GCMs simulate the physics and dynamics of the global atmosphere. They can be used to simulate current climate and to simulate climates with different atmospheric constituents, such as increased concentrations of greenhouse gases. These models will estimate climate on a regional scale, although each grid box may be several hundred miles wide.
Panel Discussion: State Coastal Erosion Management Programs
