Ecosystems: Although the area of potential distribution of temperate forests in Temperate Asia is, to a large extent, cleared and used for intensive agriculture, global climate change can be considered sufficient to trigger structural changes in the remaining temperate forests. The nature and magnitude of these changes, however, depend on associated changes in water availability, as well as water-use efficiency. Shifts in temperature and precipitation in temperate rangelands may result in altered growing seasons and boundary shifts between grasslands, forests and shrublands. Some model studies suggest that in a doubled CO2 climate there would be a large reduction in the area (up to 50 per cent) and productivity of boreal forests (primarily in the Russian Federation), accompanied by a significant expansion of grasslands and shrublands. There also would be a decrease in the area of the tundra zone of as much as 50 per cent,accompanied by the release of methane from deep peat deposits,and an increase (less than 25 per cent), in CO2 emissions.

Hydrology and Water Resources: Overall, most 2xCO 2 equilibrium scenario simulations show a decrease in water supply, except in a few river basins. Warmer winters may affect water balances because water demands are higher in spring and summer. Equilibrium climate conditions for doubled equivalent CO2 concentrations indicate that a decrease of as much as 25 per cent in mountain glacier mass is possible by 2050. Initially, runoff from glaciers in central Asia is projected to increase threefold by 2050, but by 2100 glacier runoff would taper to two-thirds of its present value. Model results suggest that runoff in the northern part of China is quite vulnerable to climate change, mainly as a consequence of changes in precipitation in spring, summer and autumn, especially during the flood season. To balance water supply with water demand, increasingly efficient water management is likely the best approach for Japan. In other parts of Temperate Asia, water-resource development will remain important; the central adaptation issue is how the design of new water-resource infrastructure should be adjusted to account for uncertainties resulting from climate change. The most critical uncertainties are the lack of credible projections of the effects of global change on the Asian monsoon or the ENSO phenomenon, which have great influence on river runoffs. Multiple-stress impact studies on water resources in international river basins are needed in the future.

Food and Fiber Production: Projected changes in crop yields using climate projections from different GCMs vary widely. In China, for example, across different scenarios and different sites, the changes for several crop yields by 2050 are projected to be: rice, -78 per cent to +15 per cent; wheat, -21 per cent to +55 per cent; and maize, -19 per cent to +5 per cent. An increase in productivity may occur if the positive effects of CO2 on crop growth are considered, but its magnitude remains uncertain. A northward shift of crop zones is expected to increase agricultural productivity in northern Siberia but to decrease (by about 25 per cent) grain production in southwestern Siberia because of a more arid climate. Aquaculture is particularly important to Temperate Asia. Greater cultivation of warm-water species could develop. Warming will require greater attention to possible oxygen depletion, fish diseases, and introduction of unwanted species, as well as to potential negative factors such as changes in established reproductive patterns, migration routes and ecosystem relationships.

Coastal Systems: An increase in sea level will exacerbate the current severe problems of tectonically and anthropogenically induced land subsidence in delta areas. Saltwater intrusion also would become more serious. A sea-level rise of 1 m would threaten certain coastal areas,for example, the Japanese coastal zone, on which 50 per cent of Japan's industrial pro-duction is located (e.g., Tokyo, Osaka and Nagoya); in addition, about 90 per cent of the remaining sandy beaches in Japan would be in danger of disappearing.

Human Health: Heat-stress mortality and illness (predominantly cardiorespiratory) are projected to more than double by 2050 resulting from an increase in the frequency or severity of heat waves under climate-change conditions projected by a transient GCM (GFDL X2, UKMO X6). Net climate change-related increases in the geographic distribution (elevation and latitude) of the vector organisms of infectious diseases (e.g., malarial mosquitoes, schistosome-spreading snails) and changes in the life-cycle dynamics of vectors and infective parasites would, in aggregate, increase the potential transmission of many vector-borne diseases. Increases in nonvector-borne infectious diseases, such as cholera, salmonellosis and other food- and water-related infections, also could occur because of climatic impacts on water distribution, temperature and micro-organism proliferation. Disease surveillance could be strengthened and integrated with other environmental moni-toring to design early warning systems; develop early, environmentally sound public health interventions; and develop anticipatory societal policies to reduce the risk of outbreaks and subsequent spread of epidemics.

Conclusions: The major impacts in Temperate Asia under global climate change are projected to be large shifts of the boreal forests, the disappearance of significant portions of mountain glaciers and water supply shortages. The most critical uncertainty in these estimates stems from the lack of credible projections of the hydrological cycle under global climate change scenarios. The effects of climate change on the Asian monsoon and the ENSO phenomenon are among the major uncertainties in the modeling of the hydrological cycle. Projections of agricultural crop yields are uncertain, not only because of the uncertainty in the hydrological cycle but also because of the potential positive effects of CO2 and production practices. Sea-level rise endangers sandy beaches in the coastal zones, but remains an anthropogenically induced problem in delta areas. Integrated impact studies considering multi-stress factors are needed.