AGCI Session II: Characterizing and Communicating Scientific Uncertainty
Session Chairs: Dr. Richard H. Moss and Dr. Stephen H. Schneider
July 31 to August 8, 1996
Freshwater Ecosystems, Hydrology & Water Resources
John J. Magnuson
Center for Limnology, University of Wisconsin - Madison
Madison, Wisconsin
Magnuson reviewed some of the controversies that arose during the development of the water-related chapters of the 1995 IPCC WG II report as well as some of the major findings of the chapters related to fresh water. To begin, he points out that fresh water is a critical human resource and essential to the functioning of all natural ecosystems, and that freshwater systems are among the most sensitive responders to climate change (McKnight, Naiman).
Controversies in the IPCC chapter on fresh water ecology began early with the question of whether such a chapter should be included in the assessment at all. Prior to a rather late review of draft chapters, it was thought that the information would come through in chapters on hydrology, oceans and large lakes, non-tidal wetlands, and water supply; but it did not. The subject of fresh water ecology sometimes falls into the cracks between oceanography and terrestrial ecology. A second controversy had to do with water supply versus water quality issues. A third controversy arose because some authors, particularly in the water supply chapter, felt that the climate scenarios were so uncertain that perhaps they should not even be used.
In the end, Magnuson said he believed that issues dominated over personalities, and the 1995 IPCC Working Group II report usefully incorporated freshwater quality and ecology into a chapter on Hydrology and Freshwater Ecology. The review system was responsive and flexible in making a place for this information late in the writing and review process. The system could have failed but did not. Magnuson said that those involved learned that the lead authors should take a broad perspective and be inclusive rather than exclusive in the materials considered, that the limited number of pages allocated to a chapter not be used as a way to exclude points of view or significant new information, and that advocacy for a particular outcome must be recognized and dealt with.
Outcome advocacy
can interfere with improving the state of the science, increasing
understanding, and making appropriate responses to the findings.
The strongest advocacy in the water area concerned water resources. One extreme was an engineering perspective (we can fix this problem) while the other was an environmentalist perspective ("the sky is falling"), Magnuson says. Outcome advocacy can interfere with improving the state of the science, increasing understanding, and making appropriate responses to the findings.
Information on aquatic ecosystems is scattered through the IPCC Working Group II report in chapters on the cryosphere (Chapter 7), wetlands (Chapters 6 & 9), oceans (Chapter 8), hydrology and freshwater ecology (Chapter 10), water resources management (Chapter 14), and even fisheries (Chapter 16). While fresh waters are sensitive to climate change, it was not possible to model freshwater ecological impacts with the most recent models provided by the IPCC. The output of these models was available far too late to be incorporated. Also, the temperature and precipitation data provided were not sufficient to model many of the effects on aquatic ecosystems; data on humidity, cloud cover, and wind is needed, and extreme events are important.
Major results are water-related chapters indicate that as a result of climatic warming, the water supply per person and water for irrigation decline, wetlands and wetland areas decline, and cryosphere area and ice cover durations decline. At the same time, flow variability and floods increase, lake and stream temperatures increase, poleward dispersal of aquatic biota increases, and extirpations and extinctions increase. All of these were high confidence results with little uncertainty, provided that the climate changes as indicated in the climate scenarios.
Changes owing to climate warming in the per capita availability of water for direct human use is simulated to vary considerably from country to country. A key conclusion, however, is that water resources would become more critical with climate warming. Water levels of lakes and rivers are far more dynamic than those of the oceans and people have already had to adapt to such changes in the past, often at high costs.
As a result of
climatic warming, the water supply per person declines, wetlands
decline, and cryosphere area and ice cover durations decline. Flow
variability and floods increase, lake and stream temperatures
increase, and extinctions increase.
Focusing on the cryosphere, scenarios for 2050 in the Northern Hemisphere indicate a 6-20 percent decline in snow cover, a 25 percent decline in glaciers, 16 percent shrinkage in permafrost, and one month less lake and stream ice. In discussion, Tom Karl pointed out with regard to the projected 6-20 percent decline in snow cover that there has already been a 10 percent decline in snow cover observed in the last 20 years.
Magnuson says that more use of lake and stream ice data would be useful in future assessments. Ice phenologies for lakes and streams have several useful features. Many long records of "ice on" and "ice off" dates exist in northern temperate, boreal, and Arctic latitudes in North America, Europe and Asia. While direct observations are common, ice dates also can be estimated across large regions using satellite data. Ice on and off dates can be modeled well with physical process models using climatic data. Long-term changes in lake ice phenologies have common features globally, such as progressively shorter ice durations over several centuries; year-to-year variation reflects more local conditions and shorter-term phenomena such as El Niño. Recent papers pointing out some of these features are in a 1996 special issue of Limnology and Oceanography on climate change (McKnight et al., 1996).
Ecosystem effects of climate warming would be varied and touch on most physical, chemical, and biological aspects of lakes and streams. Interpretations in the 1995 assessment were from direct observations and from modeled changes forced by the GCM scenarios.
Lake and stream physics are responsive to climate change. Summarizing effects simulated under doubled CO2 conditions: stream temperatures track air temperatures, lake surface temperatures rise from 1 to 7°C, deep water temperatures range from -6 to +8°C, thermocline depth ranges from -4 to +4 meters, and the thermocline gradient becomes sharper. Some improvements can be made in the physical hydrodynamic models used in aquatic climate change analyses, but the uncertainty is less in these models than in the GCMs, Magnuson says.
In a warmer,
drier climate, less primary production takes place in lakes due to a
variety of changes. And clouds dramatically reduce primary production
in lakes.
Stream export of various compounds into a lake is strongly associated with an observed warm, dry period for a lake in northwestern Ontario monitored over 20 years, the Experimental Lakes Area (ELA). Dissolved organic carbon (DOC) declines because as water levels decline in the surrounding wetlands, there is less decomposition and hence less export of DOC via the streams to the lake. The water flowing into the lake becomes progressively clearer and clearer, thus the lake gets clearer and clearer, and light transmission increases. This changes the mixing depth and the depth of photosynthesis. Changes in runoff also caused changes in the export of phosphorous, silica, and other compounds. In a warmer, drier climate, less primary production takes place in the lake, owing to these changes. Many simulations of effects on lakes or streams place the boundary of the system at the edge of a water body and do not account for how climate change would affect land-water interactions and thus the export of materials from terrestrial to aquatic portions of the landscape.
Primary production, or the total amount of organic carbon fixed, would be responsive to changes in cloud cover. This is apparent from measurements at the North Temperate Lakes Long-term Ecological Research Site in Northern Wisconsin. Photosynthetically active radiation was monitored, lab incubations were done at different temperatures, and the carbon fixed was calculated on a daily basis for Crystal Lake, Wisconsin. A key expectation from the results is that clouds dramatically reduce primary production in lakes, underscoring the importance of the role of clouds and the need for a better understanding of how they can be expected to change if we are to simulate ecological effects. On the other hand, if primary production increases as has been simulated in some studies, dissolved oxygen would more likely be depleted in the deeper, colder waters of the lake and result in the loss of cold water fishes like lake trout.
The thermal niche of a fish species can be defined by lethal, controlling, and directive (thermal preference) criteria. The criterion with the broadest limits is the lethal criteria: the fish can survive within an upper and lower temperature limit. The optimum range, in which the fish's growth rate, swimming speed and other factors are greatest, is more narrow. The most narrow range is specified by behavioral thermoregulation of the fish; fish move to better temperature habitats in much the same way as we do when we move out of or into the sun or wind. These ranges are used to calculate how much more space there is in a lake for fish before and after climate warming and whether it increases or decreases on a annual basis. Simulations to date suggest that warming may benefit fishes in large deep lakes but be detrimental in streams, shallow lakes and in the shallow areas of large, deep lakes.
In conclusion, Magnuson reiterates that freshwaters are critical to the human condition, are sensitive to climate changes, interact with other influences, have heterogeneous spatial impacts, and have many uncertainties associated with them. A rich array of these potential effects of climate change are in the IPCC chapters.
Simulations to
date suggest that warming may benefit fishes in large deep lakes but
be detrimental in streams, shallow lakes and in the shallow areas of
large, deep lakes.
References
McKnight, D., D. F. Brakke, and P. J. Mulholland. eds., 1996. Freshwater Ecosystems and Climate Change in North America. Limnology and Oceanography, 41(5).
Naiman, R. J., J. J. Magnuson, D. M. McKnight, J. A. Stanford, 1995. The Freshwater Imperative. Island Press, Washington, D.C., USA. xvi + 165 pp.
Freshwaters are
critical to the human condition, are sensitive to climate changes,
interact with other influences, have heterogeneous spatial impacts,
and have many uncertainties associated with them.