As discussed in IPCC (1996, WG I, Section 6.6), output from transient runs of atmosphere-ocean general circulation models (hereafter referred to simply as GCMs) has become available that can be used as the basis for improved regional analysis of potential climate change. The main emphasis of current analyses is on the simulation of seasonally averaged surface air temperature and precipitation. Climate scenario information for North America is available from several GCMs. In IPCC (1990, WG I), one of the five regions identified for analysis of regional climate change simulation was central North America (35-50�N, 85-105�W). Output for this region from different coupled model runs with dynamic oceans was analyzed by Cubasch et al. (1994) and Kittel et al. (1998). Results for central North America, as well as the other identified regions, are depicted in Figure B-1 (Annex B), which shows differences between region-average values at the time of CO2 doubling and the control run, as well as differences between control run averages and observations (hereafter referred to as bias) for winter and summer surface air temperature and precipitation. These model results reflect increasing CO2 only and do not include the effects of sulfate aerosols. The biases in Figure B-1 (Annex B) are presented as a reference for interpretation of the scenarios because it can be generally expected that the better the match between control run and observed climate (i.e., the lower the biases), the higher the confidence in the simulated change scenarios. A summary of these transient model experiments is given in Table B-1 (Annex B). Most experiments use a rate of CO2 increase of 1%/year, yielding a doubling of CO2 after 70 years.
Scenarios produced for central North America by these transient experiments vary quite widely among models for temperature but less so for precipitation. GCM simulations also have been conducted that consider the effect of combined greenhouse gas- and direct sulfate aerosol-forcing on temperature, precipitation, and soil moisture (see Annex B). For central North America, the inclusion of sulfate aerosols results in a projected warming of 0-0.5�C in the summer and 1.4-3.4�C in the winter by the year 2100. In the case of precipitation, the inclusion of sulfate aerosol-forcing has little effect on the projections (see Annex B).
Using the Canadian Climate Centre (CCC) GCM (see Annex B), Lambert (1995) found a 4% decrease in cyclones in the Northern Hemisphere, though the frequency of intense cyclones increased. Lambert hypothesized that the latent heat effect is responsible for the greater number of intense storms. No change in storm tracks was evident. A few areas showed increased frequencies, such as off Cape Hatteras, over Hudson Bay, and west of Alaska. These results are similar to those of Rowntree (1993), who found a 40% increase in Atlantic gales, though fewer intense storms over eastern North America. Hall et al. (1994) and Carnell et al. (1996) found an intensification and northward shift of storm tracks.
Regarding sea-level rise scenarios, for IPCC Scenario IS92a, global mean sea level is projected to be about 50 cm higher by 2100 than today, with a range of uncertainty of 20-86 cm (IPCC 1996, WG I, Section 7.5). It is possible that for much of the North American coastline, future sea-level rise will be greater than the global average, given the higher historical rates of sea-level rise along the Gulf of Mexico and Atlantic coasts (see Section 8.2.2). By contrast, future sea-level rise along the Pacific coast may be less than the global average rise because of this region's generally lower historical rates. Even less sea-level rise might be expected in extreme northern North America, given the historical drop in sea levels at many locations (Titus and Narayanan, 1996).
Other reports in this collection