Continued from previous page
The following findings from the models analysed in this chapter corroborate
results from the SAR (projections of regional climate change are given in Chapter
10) for all scenarios considered. We assign these to be virtually certain
to very likely (defined as agreement among most models, or, where only a small
number of models have been analysed and their results are physically plausible,
these have been assessed to characterise those from a larger number of models).
The more recent results are generally obtained from models with improved parametrizations
(e.g., better land-surface process schemes).
- The troposphere warms, stratosphere cools, and near surface temperature
warms.
- Generally, the land warms faster than the ocean, the land warms more than
the ocean after forcing stabilises, and there is greater relative warming
at high latitudes.
- The cooling effect of tropospheric aerosols moderates warming both globally
and locally, which mitigates the increase in SAT.
- The SAT increase is smaller in the North Atlantic and circumpolar Southern
Ocean regions relative to the global mean.
- As the climate warms, Northern Hemisphere snow cover and sea-ice extent
decrease.
- The globally averaged mean water vapour, evaporation and precipitation increase.
- Most tropical areas have increased mean precipitation, most of the sub-tropical
areas have decreased mean precipitation, and in the high latitudes the mean
precipitation increases.
- Intensity of rainfall events increases.
- There is a general drying of the mid-continental areas during summer (decreases
in soil moisture). This is ascribed to a combination of increased temperature
and potential evaporation that is not balanced by increases in precipitation.
- A majority of models show a mean El Niño-like response in the tropical
Pacific, with the central and eastern equatorial Pacific sea surface temperatures
warming more than the western equatorial Pacific, with a corresponding mean
eastward shift of precipitation.
- Available studies indicate enhanced interannual variability of northern
summer monsoon precipitation.
- With an increase in the mean surface air temperature, there are more frequent
extreme high maximum temperatures and less frequent extreme low minimum temperatures.
There is a decrease in diurnal temperature range in many areas, with night-time
lows increasing more than daytime highs. A number of models show a general
decrease in daily variability of surface air temperature in winter, and increased
daily variability in summer in the Northern Hemisphere land areas.
- The multi-model ensemble signal to noise ratio is greater for surface air
temperature than for precipitation.
- Most models show weakening of the Northern Hemisphere thermohaline circulation
(THC), which contributes to a reduction in the surface warming in the northern
North Atlantic. Even in models where the THC weakens, there is still a warming
over Europe due to increased greenhouse gases.
- The deep ocean has a very long thermodynamic response time to any changes
in radiative forcing; over the next century, heat anomalies penetrate to depth
mainly at high latitudes where mixing is greatest.
A second category of results assessed here are those that are new since the
SAR, and we ascribe these to be very likely (as defined above):
- The range of the TCR is limited by the compensation between the effective
climate sensitivity (ECS) and ocean heat uptake. For instance, a large ECS,
implying a large temperature change, is offset by a comparatively large heat
flux into the ocean.
- Including the direct effect of sulphate aerosols (IS92a or similar) reduces
global mean mid-21st century warming (though there are uncertainties involved
with sulphate aerosol forcing - see Chapter 6).
- Projections of climate for the next 100 years have a large range due both
to the differences of model responses and the range of emission scenarios.
Choice of model makes a difference comparable to choice of scenario considered
here.
- In experiments where the atmospheric greenhouse gas concentration is stabilised
at twice its present day value, the North Atlantic THC recovers from initial
weakening within one to several centuries.
- The increases in surface air temperature and surface absolute humidity result
in even larger increases in the heat index (a measure of the combined effects
of temperature and moisture). The increases in surface air temperature also
result in an increase in the annual cooling degree days and a decrease in
heating degree days.
Additional new results since the SAR; these are assessed to be likely due to
many (but not most) models showing a given result, or a small number of models
showing a physically plausible result.
- Areas of increased 20 year return values of daily maximum temperature events
are largest mainly in areas where soil moisture decreases; increases in return
values of daily minimum temperature especially occur over most land areas
and are generally larger where snow and sea ice retreat.
- Precipitation extremes increase more than does the mean and the return period
for extreme precipitation events decreases almost everywhere.
Another category includes results from a limited number of studies which are
new, less certain, or unresolved, and we assess these to have medium likelihood,
though they remain physically plausible:
- Although the North Atlantic THC weakens in most models, the relative roles
of surface heat and freshwater fluxes vary from model to model. Wind stress
changes appear to play only a minor role.
- It appears that a collapse in the THC by the year 2100 is less likely than
previously discussed in the SAR, based on the AOGCM results to date.
- Beyond 2100, the THC could completely shut-down, possibly irreversibly,
in either hemisphere if the rate of change of radiative forcing was large
enough and applied long enough. The implications of a complete shut-down of
the THC have not been fully explored.
- Although many models show an El Niño-like change in the mean state
of tropical Pacific SSTs, the cause is uncertain. It has been related to changes
in the cloud radiative forcing and/or evaporative damping of the east-west
SST gradient in some models.
- Future changes in El Niño-Southern Oscillation (ENSO) interannual
variability differ from model to model. In models that show increases, this
is related to an increase in thermocline intensity, but other models show
no significant change and there are considerable uncertainties due to model
limitations of simulating ENSO in the current generation of AOGCMs (Chapter
8).
- Several models produce less of the weak but more of the deeper mid-latitude
lows, meaning a reduced total number of storms. Techniques are being pioneered
to study the mechanisms of the changes and of variability, but general agreement
among models has not been reached.
- There is some evidence that shows only small changes in the frequency of
tropical cyclones derived from large-scale parameters related to tropical
cyclone genesis, though some measures of intensities show increases, and some
theoretical and modelling studies suggest that upper limit intensities could
increase (for further discussion see Chapter 10).
- There is no clear agreement concerning the changes in frequency or structure
of naturally occurring modes of variability such as the North Atlantic Oscillation
