Upload
others
View
2
Download
0
Embed Size (px)
Citation preview
UNDP Climate Change Country Profiles
South Africa
A. Karmalkar1, C. McSweeney
1, M. New
1,2 and
G. Lizcano1
1. School of Geography and Environment, University of Oxford.
2. Tyndall Centre for Climate Change Research
http://country-profiles.geog.ox.ac.uk
General Climate
“ラ┌デエ AaヴキI;が デエW ゲラ┌デエWヴミマラゲデ Iラ┌ミデヴ┞ キミ AaヴキI;が キゲ ヮ;ヴデ ラa デエW “ラ┌デエWヴミ HWマキゲヮエWヴWげゲ ゲ┌HデヴラヮキI;ノ zone. The climate of South Africa is largely influenced by variations in elevations and by the Indian
Ocean to the southeast and the Atlantic Ocean to the southwest, as well as the different rainfall
regimes (summer and winter) occurring over the country.
Temperatures can often exceed 32°C in the summer and can drop several degrees below freezing
point at higher elevations in winter. There is also a striking contrast between temperatures on the
country's east and west coasts, largely in response to the warm Agulhas ocean current that sweeps
southward along the Indian Ocean coastline in the east, and the cold Benguela current that sweeps
northward along the Atlantic Ocean coastline in the west. As a result, temperatures along the east
coast of the country are about 5°C warmer than temperatures at same latitudes along the west
coast. There is also very little difference in average temperatures from south to north, in part
because the inland plateau rises slightly in the northeast. The daily temperature ranges are,
however, markedly higher on the plateau where clear winter weather allows overnight freezing
temperatures to contrast with warm midday temperatures. South Africa experiences a rainy season
from convective type rainfall mainly in the summer months from November through March. The
southwest region, however, receives mostly mid-latitude frontal rainfall in winter months (May
through August). Rainfall also varies considerably from west to east, associated with the positioning
of the ocean currents and the general anticyclonic circulation over the subcontinent.. The north-
western region of the country receives less than 200 mm rainfall p.a. whereas the eastern parts
receive over 500 mm rainfall annually. The 250mm isohyet is generally accepted as being the
geographical boundary for rain にfed agriculture.
El Niño Southern Oscillation events and sea-surface temperature anomalies in the Indian and South
Atlantic Oceans significantly influence rainfall variability in South Africa.
South Africa
Recent Climate Trends
Temperature
Mean annual temperature averaged over South Africa エ;ゲ キミIヴW;ゲWS H┞ ;ヴラ┌ミS ヰくヶこC between 1960 and 2006が ;デ ;ミ ;┗Wヴ;ェW ヴ;デW ラa ヰくヱヴこC ヮWヴ SWI;SW (Table 1, Fig. 1).
Daily temperature observations show significantly increasing trends in daily temperature
extremes.
o TエW ;┗Wヴ;ェW ミ┌マHWヴ ラa けエラデげ1 days per year in South Africa has increased by 19 (an
additional 5% of days) between 1960 and 2003 (Table 2). The rate of increase is seen
most strongly in MAM.
o TエW ;┗Wヴ;ェW ミ┌マHWヴ ラa けエラデげ ミキェエデゲ ヮWヴ ┞W;ヴ キミ “ラ┌デエ AaヴキI; エ;ゲ キミIヴW;ゲWS H┞ ヱ6 (an
additional 4.3% of nights; Table 2). The rate of increase is seen most strongly in MAM.
Precipitation
Mean rainfall over South Africa has decreased by 1.5mm per month (3.7%) per decade since
1960 (Table 1, Fig. 3). The MAM season shows a statistically decreasing trend of 2.5mm per
month (6.6%) per decade in rainfall.
Identified trends in daily rainfall extremes are not statistically significant (Table 3).
GCM Projections of Future Climate
Temperature
The mean annual temperature is projected to increase by 1.1 to 2.4こC H┞ デエW 2060s, and 1.6
to 4.3こC by the 2090s (Table 1, Fig. 2). The range of projections by the 2090s under any one
emissions scenario is around 1-ヲこCく TエW ヮヴラテWIデWS ヴ;デW ラa ┘;ヴマキミェ キゲ ゲキマキノ;ヴ デエヴラ┌ェエラ┌デ デエW year. Warming is expected to be greater inland than in coastal regions (Fig. 2).
All projections indicate substantial increases in the frequency of days and nights that are
IラミゲキSWヴWS けエラデげ キミ I┌ヴヴWミデ Iノキマ;デWく o Annually, projections indicate that the frequency of けエラデげ S;┞ゲ will increase from
about 11% during the period 1970-1999 to 16-25% of days by the 2060s, and 20-39%
of days by the 2090s. Days that are hot for each season are projected to increase
most rapidly in DJF from 10.5% to 25-62% of days by the 2090s (Table 2, Fig 7, 8).
o Nights that are considereS けエラデげ aラヴ デエW ;ミミ┌;ノ Iノキマ;デW ラa ヱΓΑヰ-99 are projected to
increase from around 11% to 16-31% of nights by the 2060s and 19-44% of nights by
1 けHラデげ S;┞ ラヴ けエラデげ ミキェエデ キゲ SWaキミWS H┞ デエW デWマヮWヴ;デ┌ヴW W┝IWWSWS ラミ ヱヰХ ラa S;┞ゲ ラヴ ミキェエデゲ キミ I┌ヴヴWミデ Iノキマ;デW ラa デエ;デ ヴWェキラミ ;ミd season.
South Africa
the 2090s. Nights that are hot for each season are projected to increase most rapidly
in DJF, occurring on 27-78% of nights by the 2090s (Table 2, Fig. 9, 10).
o TエW ヮヴラテWIデWS キミIヴW;ゲW キミ デエW aヴWケ┌WミI┞ ラa けエラデげ S;┞ゲ ;ミS けエラデげ ミキェエデゲ キゲ エキェエWヴ キミ デエW north-eastern part of South Africa compared to the western part of the country
(Figs. 8, 10).
All projections indicate decreases in the frequency of days and nights that are considered
けIラノSげ2 in current climate (Table 2, Figs. 11-14).
Precipitation
Projections of mean annual rainfall from different models in the ensemble are broadly
consistent in indicating a small decrease in rainfall for South Africa (Table 1, Figs. 3-6).
Projected changes in precipitation, however, show wide variations across the country (Figs.
4, 6). Ensemble median values for all seasons are negative. The projections for the annual
mean rainfall anomaly for 2090s vary between -17 and +7%, while the median values vary
between -6 and -4%.
The proportion of total rainfall that falls in heavy rainfall3 events and maximum 5-day rainfall
amounts show very small changes in most model projections (Table 3, Figs. 15-20).
Additional Regional Climate Change Information
Model simulations show wide disagreements in projected changes in the amplitude and
frequency of future El Niño events, contributing to uncertainty in future climate variability in
projections for this region.
The projected increase in temperature is expected to exacerbate the possibility of drying, or
even counteract any possible increases in rainfall, through increased evapotranspiration,
which would detrimentally affect the agriculture and water resources of the country.
It is important to note that projections discussed in the report are based on climate models
with coarse spatial resolution, which hampers their ability to simulate regional-scale climate
features of the region.
For further information see Christensen et al. (2007) IPCC Wラヴニキミェ Gヴラ┌ヮ I ‘Wヮラヴデぎ けThe
Pエ┞ゲキI;ノ SIキWミIW B;ゲキゲげ, Chapter 11 (Regional Climate projections): Sections 11.2 (Africa).
2 けCラノSげ S;┞ ラヴ けIラノSげ ミキェエデ キゲ SWaキミWS ;ゲ デエW デWマヮWヴ;デ┌ヴW HWノラ┘ ┘エキIエ ヱヰХ ラa S;┞ゲ ラヴ nights are recorded in current climate of that region
or season. 3 A けHW;┗┞ ヴ;キミa;ノノげ W┗Wミデ キゲ SWaキミWS ;ゲ ; ラミW ┘キデエ ; S;キノ┞ ヴ;キミa;ノノ デラデ;ノ ┘エキIエ W┝IWWSゲ デエW デエヴWゲエラノS デエ;デ キゲ W┝IWWSWS ラミ ヵХ ラf rainy days in
current the climate of that region and season.
South Africa
Methods and Documents
This report provides basic analyses of observations and climate model data in the form of narrative,
S;デ; デ;HノWゲが ;ミS ェヴ;ヮエキIゲ ;ゲ ;ミ さラaa デエW ゲエWノaざ resource and can be consulted in investigations of
climate impacts, risk assessments, or adaptation options and for use in further research. It is
important to note that significant limitations and caveats are involved wherever climate model
projections are applied. Please refer to the following documents for additional information.
Rationale and Methodology: McSweeney, C., G. Lizcano, M. New, X. Lu, 2010: The UNDP Climate
Change Country Profiles. Bull. Amer. Meteor. Soc., 91, 157に166. doi: 10.1175/2009BAMS2826.1
http://journals.ametsoc.org/doi/pdf/10.1175/2009BAMS2826.1
Technical details: http://country-profiles.geog.ox.ac.uk/UNDPCCCP_documentation.pdf
South Africa
Data Summary
Table 1
Observed
Mean
1970-99
Observed
Trend
1960-2006
Projected changes by the
2030s
Projected changes by the
2060s
Projected changes by the
2090s
Min Median Max Min Median Max Min Median Max
Temperature
ふこCぶ
(change in
こC ヮWヴ decade)
Cエ;ミェW キミ こC Cエ;ミェW キミ こC Cエ;ミェW キミ こC
A2 0.6 1 1.3 1.5 2.1 2.4 2.8 3.9 4.3
Annual 17.6 0.14* A1B 0.5 1 1.3 1.5 2.2 2.4 2.3 3.2 4
B1 0.5 0.9 1.2 1.1 1.5 2 1.6 2.1 2.8
A2 0.7 1 1.2 1.4 2 2.5 2.6 3.6 4.6
DJF 22.6 0.10* A1B 0.6 0.9 1.5 1.3 2 2.6 2 3 4.4
B1 0.4 0.9 1.1 0.9 1.5 2 1.5 2.1 3
A2 0.6 0.8 1.4 1.5 2 2.6 2.6 3.6 4.6
MAM 17.7 0.15* A1B 0.5 1 1.3 1.4 2.1 2.7 2.2 3 4.2
B1 0.1 1 1.5 0.9 1.5 2.6 1.5 2 2.7
A2 0.4 1 1.4 1.5 2.1 2.4 2.8 4 4.3
JJA 11.7 0.13* A1B 0.4 1 1.4 1.6 2.1 2.6 2.2 3.1 3.8
B1 0.3 1 1.6 1.1 1.5 1.9 1.7 2 2.5
A2 0.5 1.1 1.5 1.4 2.2 2.5 2.9 3.8 4.4
SON 18.5 0.10* A1B 0.6 1.1 1.3 1.6 2.2 2.9 2.6 3.3 4.1
B1 0.6 0.9 1.2 1 1.6 1.8 1.6 2.1 3
Precipitation
(mm per
month)
(change in
mm per
decade)
Change in mm per month Change in mm per month Change in mm per month
A2 -4 -1 3 -9 -1 3 -10 -2 4
Annual 40 -1.5* A1B -5 0 3 -8 -1 4 -11 -3 2
B1 -5 -1 2 -6 0 2 -9 -2 5
A2 -5 -1 6 -16 0 7 -17 1 17
DJF 66.7 -1.9 A1B -6 -2 4 -9 -3 14 -18 -1 9
B1 -13 -2 6 -9 1 8 -14 -2 4
A2 -6 0 9 -14 -1 11 -10 0 10
MAM 37.4 -2.5* A1B -9 -1 11 -12 -2 4 -12 -1 13
B1 -9 -2 7 -10 -2 9 -15 -3 9
A2 -9 -1 1 -7 -2 6 -11 -5 2
JJA 15.6 -0.7 A1B -5 -1 3 -7 -3 -1 -9 -4 -1
B1 -8 0 2 -6 -1 1 -7 -3 0
A2 -8 -1 7 -10 -1 2 -19 -3 6
SON 40.3 -1.3 A1B -7 -1 9 -12 -2 5 -17 -4 8
B1 -6 0 5 -7 -2 5 -10 -1 10
Precipitation (%)
(mm per
month)
(change in %
per decade)
% Change % Change % Change
A2 -11 -2 8 -16 -2 5 -17 -6 7
Annual 40 -3.7* A1B -9 -2 4 -16 -4 10 -16 -5 4
B1 -9 -2 5 -13 -1 3 -15 -4 6
A2 -6 -1 13 -19 0 11 -19 2 18
DJF 66.7 -2.8 A1B -8 -2 6 -12 -2 24 -20 -2 15
B1 -17 -2 13 -11 1 14 -15 -3 5
A2 -13 0 18 -20 -4 12 -23 -1 19
MAM 37.4 -6.6* A1B -16 -2 21 -23 -4 11 -18 -2 15
B1 -16 -4 6 -24 -2 11 -21 -5 16
A2 -28 -6 5 -25 -12 25 -35 -22 5
JJA 15.6 -4.7 A1B -20 -4 13 -25 -10 -3 -35 -16 -3
B1 -24 -5 11 -24 -11 4 -25 -11 4
A2 -17 -2 8 -14 -2 6 -32 -7 12
SON 40.3 -3.1 A1B -14 -2 17 -21 -6 9 -28 -9 9
B1 -13 0 9 -16 -4 9 -23 -3 11
South Africa
Table 2
Observed
Mean
1970-99
Observed
Trend
1960-2003
Projected changes by the
2030s
Projected changes by the
2060s
Projected changes by the
2090s
Min Median Max Min Median Max Min Median Max
%
Frequency
Change in
frequency
per decade
Future % frequency Future % frequency
Frequency of Hot Days (TX90p)
A2 **** **** **** 19 24 25 26 32 39
Annual 11.6 1.17* A1B **** **** **** 19 23 25 23 28 39
B1 **** **** **** 16 18 22 20 22 29
A2 **** **** **** 25 32 39 37 48 62
DJF 10.5 0.72 A1B **** **** **** 22 30 38 33 39 59
B1 **** **** **** 18 24 29 25 30 42
A2 **** **** **** 20 27 33 28 44 52
MAM 12.7 2.52* A1B **** **** **** 21 28 29 29 37 50
B1 **** **** **** 17 22 25 20 27 33
A2 **** **** **** 21 28 33 34 48 55
JJA 11.7 1.55* A1B **** **** **** 23 28 33 29 37 49
B1 **** **** **** 19 21 25 22 26 32
A2 **** **** **** 22 29 33 31 41 51
SON 11.1 0.58 A1B **** **** **** 22 28 34 26 40 45
B1 **** **** **** 18 21 25 21 25 35
Frequency of Hot Nights (TN90p)
A2 **** **** **** 21 28 30 30 41 44
Annual 11.1 0.99* A1B **** **** **** 19 28 31 25 36 40
B1 **** **** **** 16 22 24 19 25 30
A2 **** **** **** 31 43 53 53 69 78
DJF 10.6 0.48 A1B **** **** **** 29 43 55 37 63 73
B1 **** **** **** 21 34 38 27 41 55
A2 **** **** **** 25 32 35 36 49 55
MAM 11.2 1.42* A1B **** **** **** 20 32 35 31 41 52
B1 **** **** **** 18 25 29 20 28 35
A2 **** **** **** 23 31 36 38 52 59
JJA 10.8 0.64 A1B **** **** **** 25 32 39 30 43 54
B1 **** **** **** 20 23 26 21 29 35
A2 **** **** **** 25 29 36 40 47 55
SON 11.4 1.05* A1B **** **** **** 24 31 35 33 39 47
B1 **** **** **** 19 24 27 22 28 35
Frequency of Cold Days (TX10p)
A2 **** **** **** 2 4 5 0 1 3
Annual 9.6 -0.67* A1B **** **** **** 2 4 5 1 2 3
B1 **** **** **** 4 5 6 3 4 5
A2 **** **** **** 2 3 4 0 1 4
DJF 10.7 -0.39 A1B **** **** **** 2 3 4 1 1 4
B1 **** **** **** 3 4 6 2 3 4
A2 **** **** **** 2 3 4 0 1 3
MAM 9.2 -1.31* A1B **** **** **** 2 3 5 1 2 3
B1 **** **** **** 3 4 6 2 3 6
A2 **** **** **** 1 2 4 0 0 1
JJA 9.5 -1.07* A1B **** **** **** 1 3 4 0 1 3
B1 **** **** **** 2 4 5 2 3 3
A2 **** **** **** 1 3 5 1 2 3
SON 9.8 -0.02 A1B **** **** **** 2 3 4 1 2 3
B1 **** **** **** 4 5 5 2 4 5
Frequency of Cold Nights (TN10p)
A2 **** **** **** 1 3 4 0 1 2
Annual 9.3 -1.12* A1B **** **** **** 2 3 4 0 2 3
B1 **** **** **** 3 4 5 2 3 5
A2 **** **** **** 1 2 3 0 0 1
DJF 9.3 -1.01* A1B **** **** **** 0 2 3 0 0 2
B1 **** **** **** 1 2 5 1 2 3
A2 **** **** **** 1 2 4 0 1 2
MAM 9.3 -1.36* A1B **** **** **** 1 3 5 0 1 4
B1 **** **** **** 3 4 6 2 3 5
A2 **** **** **** 0 2 3 0 0 1
JJA 9.3 -1.35* A1B **** **** **** 0 2 3 0 1 2
B1 **** **** **** 2 3 5 1 2 4
A2 **** **** **** 1 2 3 0 1 1
SON 9.3 -0.76 A1B **** **** **** 1 2 4 0 1 2
B1 **** **** **** 2 4 5 1 3 4
South Africa
Table 3
Observed
Mean
1970-99
Observed
Trend
1960-2006
Projected changes by the
2030s
Projected changes by the
2060s
Projected changes by the
2090s
Min Median Max Min Median Max Min Median Max
% total rainfall falling in Heavy Events (R95pct)
%
Change in %
per decade
Change in % Change in %
A2 **** **** **** -3 0 3 -2 1 6
Annual 31.6 -0.41 A1B **** **** **** -3 0 3 -3 1 6
B1 **** **** **** -2 0 4 -2 1 4
A2 **** **** **** -5 1 4 -4 2 6
DJF **** **** A1B **** **** **** -4 1 3 -4 2 6
B1 **** **** **** -4 1 5 -3 0 3
A2 **** **** **** -9 1 4 -7 1 7
MAM **** **** A1B **** **** **** -6 1 4 -7 1 6
B1 **** **** **** -6 0 4 -5 1 9
A2 **** **** **** -9 -3 4 -15 -6 3
JJA **** **** A1B **** **** **** -10 -3 2 -11 0 4
B1 **** **** **** -10 -1 4 -19 -1 2
A2 **** **** **** -12 -1 0 -18 -2 5
SON **** **** A1B **** **** **** -8 -1 0 -12 0 3
B1 **** **** **** -7 -1 2 -6 0 4
Maximum 1-day rainfall (RX1day)
mm
Change in
mm per
decade
Change in mm Change in mm
A2 **** **** **** -1 0 5 -2 1 7
Annual **** **** A1B **** **** **** -2 0 4 -1 1 5
B1 **** **** **** -1 0 6 -1 0 6
A2 **** **** **** -1 0 3 -2 1 4
DJF 27.2 1.01 A1B **** **** **** -1 0 2 0 0 4
B1 **** **** **** -1 0 2 -3 0 3
A2 **** **** **** -1 0 2 -4 1 5
MAM 18.4 -1.04 A1B **** **** **** -1 1 3 -2 1 5
B1 **** **** **** -3 0 3 -2 0 5
A2 **** **** **** -2 0 1 -3 -1 1
JJA 7.6 -0.68 A1B **** **** **** -3 0 1 -4 0 1
B1 **** **** **** -2 0 1 -4 0 1
A2 **** **** **** -3 0 1 -7 0 2
SON 20.2 0.3 A1B **** **** **** -3 0 0 -3 0 2
B1 **** **** **** -1 0 2 -2 0 2
Maximum 5-day Rainfall (RX5day)
mm
Change in
mm per
decade
Change in mm Change in mm
A2 **** **** **** -6 0 7 -4 1 12
Annual 101.5 0.99 A1B **** **** **** -6 1 5 -3 0 9
B1 **** **** **** -5 -1 8 -5 0 7
A2 **** **** **** -7 -1 5 -6 2 10
DJF 44.8 1.84 A1B **** **** **** -5 0 6 -5 0 10
B1 **** **** **** -5 0 5 -8 0 6
A2 **** **** **** -5 1 4 -7 3 10
MAM 31.2 -1.27 A1B **** **** **** -5 2 4 -5 1 8
B1 **** **** **** -5 0 7 -6 1 6
A2 **** **** **** -4 -1 2 -7 -3 1
JJA 12.2 -1.26* A1B **** **** **** -6 0 2 -7 -1 1
B1 **** **** **** -5 0 3 -7 -2 0
A2 **** **** **** -7 0 2 -14 -3 2
SON 27.5 -0.68 A1B **** **** **** -7 -2 0 -10 0 3
B1 **** **** **** -5 0 5 -6 -1 5
* indicates trend is statistically significant at 95% confidence
**** indicates data are not available
South Africa
Figure 1: Trends in annual and seasonal mean temperature for the recent past and projected future. All values shown are anomalies, relative to the 1970-1999 meanclimate. Black curves show the mean of observed data from 1960 to 2006, Brown curves show the median (solid line) and range (shading) of model simulations ofrecent climate across an ensemble of 15 models. Coloured lines from 2006 onwards show the median (solid line) and range (shading) of the ensemble projections ofclimate under three emissions scenarios. Coloured bars on the right-hand side of the projections summarise the range of mean 2090-2100 climates simulated by the15 models for each emissions scenario.
South Africa
Figure 2: Spatial patterns of projected change in mean annual and seasonal temperature for 10-year periods in the future under the SRES A2 scenario. All values areanomalies relative to the mean climate of 1970-1999. In each grid box, the central value gives the ensemble median and the values in the upper and lower cornersgive the ensemble maximum and minimum.
South Africa
South Africa
Figure 3: Trends in monthly precipitation for the recent past and projected future. All values shown are anomalies, relative to the 1970-1999 mean climate. SeeFigure 1 for details.
South Africa
Figure 4: Spatial patterns of projected change in monthly precipitation for 10-year periods in the future under the SRES A2 scenario. All values are anomalies relativeto the mean climate of 1970-1999.See Figure 2 for details.
South Africa
Figure 5: Trends in monthly precipitation for the recent past and projected future. All values shown are percentage anomalies, relative to the 1970-1999 mean climate.See Figure 1 for details.
South Africa
Figure 6: Spatial patterns of projected change in monthly precipitation for 10-year periods in the future under the SRES A2 scenario. All values are percentageanomalies relative to the mean climate of 1970-1999.See Figure 2 for details.
South Africa
Figure 7: Trends in hot-day frequency for the recent past and projected future. See Figure 1 for details.
South Africa
Figure 8: Spatial patterns of hot-day frequency for 10-year periods in the future under the SRES A2 scenario. See Figure 2 for details.
South Africa
Figure 9: Trends in hot-night frequency for the recent past and projected future. See Figure 1 for details.
South Africa
Figure 10: Spatial patterns of hot-night frequency for 10-year periods in the future under the SRES A2 scenario. See Figure 2 for details.
South Africa
Figure 11: Trends in cold-day frequency for the recent past and projected future. See Figure 1 for details.
South Africa
Figure 12: Spatial patterns of cold-day frequency for 10-year periods in the future under the SRES A2 scenario. See Figure 2 for details.
South Africa
Figure 13: Trends in cold-night frequency for the recent past and projected future. See Figure 1 for details.
South Africa
Figure 14: Spatial patterns of cold-night frequency for 10-year periods in the future under the SRES A2 scenario. See Figure 2 for details.
South Africa
Figure 15: Trends in the proportion of precipitation falling in ’heavy’ events for the recent past and projected future. All values shown are anomalies, relative to the1970-1999 mean climate. See Figure 1 for details.
South Africa
Figure 16: Spatial patterns of projected change in the proportion of precipitation falling in ’heavy’ events for 10-year periods in the future under the SRES A2 scenario.All values are anomalies relative to the mean climate of 1970-1999. See Figure 2 for details.
South Africa
Figure 17: Trends in maximum 1-day rainfall for the recent past and projected future. All values shown are anomalies, relative to the 1970-1999 mean climate. SeeFigure 1 for details.
South Africa
Figure 18: Spatial patterns of maximum 1-day rainfall for 10-year periods in the future under the SRES A2 scenario. All values are anomalies relative to the meanclimate of 1970-1999. See Figure 2 for details.
South Africa
Figure 19: Trends in maximum 5-day rainfall for the recent past and projected future. All values shown are anomalies, relative to the 1970-1999 mean climate. SeeFigure 1 for details.
South Africa
Figure 20: Spatial patterns of projected change in maximum 5-day rainfall for 10-year periods in the future under the SRES A2 scenario. All values are anomaliesrelative to the mean climate of 1970-1999. See Figure 2 for details.