Temperature and Precipitation Projections by Region

Chapter 11 of the IPCC latest report by Working Group 1 describes in detail model projections for a set of subcontinental regions that have been traditionally used by the climate change community since they were proposed by Giorgi and Francisco (2000). The chapter also analyzes the processes relevant to each region's climate and the ability of models to capture them, thus gauging the reliability of future projections. It also considers the consistency of future projections with changes already observed, when possible, and supplements GCM projections by regional modeling studies when available. Figures, tables and discussion provide a rich portrait of what scientific understanding, local expertise and modeling experiments suggest for the future at these regional scales.

Here we describe continent by continent the main findings summarized by the IPCC report (Christensen et al. 2007). We intend this as a quick reference, but we point at the report chapter, available online with its own supplementary material, for a more complete treatment of the subject. The scenario adopted throughout the description is A1B, considered close to "business as usual", in the way its rates of emissions remain similar to the current rates. Temperature and precipitation projections are based on 21 models that ran the A1B experiment. Future changes are computed as the difference between two 20-year averages within each simulation, 1980-1999 and 2080-2099. In the case of precipitation the change is expressed as a percentage of the 1980-1999 average.

Interestingly, it has been shown (Santer et al. 1990) that regional patterns of change of temperature and precipitation remain close to constant along the future simulations, and the "intensity" of the change is proportional to the global average temperature change signal. This result, known as pattern scaling, has been exploited for example in order to explore a large range of uncertainties by modifying

Table 3.1 Global average temperature change (with respect to the baseline 1981-2000) under three SRES scenarios for short (2030) medium (2050) and long term (2070 and 2090) projections

SRES scenario

2021-2040

2041-2060

2061-2080

2081-2100

B1

0.9

1.1

1.5

1.8

A1B

0.9

1.5

2.2

2.7

A2

0.9

1.5

2.2

3.2

the parameters of simpler models, cheaper to run under alternative - but equally plausible - settings. From the simpler models only the signal of global average temperature change is extracted. It is then applied to the normalized geographic patterns derived from GCMs to produce a large collection of regional projections (Murphy et al. 2004). This argument may be used to infer shorter term projections on the basis of the following end-of-the century changes. To a first degree of approximation one simply computes the ratio of global average temperature change at the end of the century under the A1B scenario and the same quantity at the shorter projection time. The projected changes in temperature and precipitation can be then rescaled by dividing them by this ratio. As reference, the global average temperature change (with respect to the baseline 1981-2000) under the three scenarios for short (2030) medium (2050) and long term (2070 and 2090) projections are listed in Table 3.1.

3.4.1.1 Africa

The African continent will very likely (with greater than 90% probability, according to a rigorous definition of the phrase in the IPCC report) experience warming in greater measure than the global average, and this is true for all seasons. The median warming projected by the ensemble is of over 3°C, with individual model projections ranging from close to 2°C for the cooler models to over 5°C for the warmer models. The drier subtropical regions will warm more than the moister tropics. Annual rainfall changes will vary across the different regions of the continent. Likely (with greater than 2/3 probability) there will be a decrease of precipitation amounts in much of Mediterranean Africa and the northern Sahara, in southern Africa in the winter rainfall region and western margins. On the contrary it is likely that East Africa will experience an increase in annual mean rainfall. Projections for the Sahel, the Guinean Coast and the southern Sahara are of contrasting sign.

3.4.1.2 Mediterranean and Europe

Annual mean temperatures in Europe are likely to increase more than the global mean with the largest warming affecting Northern Europe in the winter season and the Mediterranean basin in the summer season. The median annual warming for Northern Europe is projected to be more than 3°C, with a range from more than 2°C up to almost 5.5°C. For southern Europe the median is higher, 3.5°C (range: 2-5°C).

Precipitation changes across models show a larger agreement than for other regions of the world, suggesting increases in Northern Europe especially in the winter season and decreases in southern Europe, largest in the spring and summer seasons.

The Asian continent will warm more than the global average almost in its entirety, the exception being South East Asia. The models project a median warming of over 3.3°C, with a gradient increasing towards the northern latitudes. The range of projections goes from over 2°C to well over 5.5°C, with seasonal ranges touching 8.7°C for winter in Northern Asia. Precipitation in the boreal winter season is projected to increase over the entire continent, with larger confidence in the Northern regions and the Tibetan Plateau. Precipitation in summer is also likely to increase in northern Asia, East Asia, South Asia and most of Southeast Asia, while models tend to agree over a decrease of precipitation in central Asia.

3.4.1.4 North America

The annual mean warming is likely to be greater than the global mean warming for almost all regions of the continent, but especially so for winter in the high latitudes (where minimum temperatures show largest increases) and summer in the Southwest (where maximum temperatures do). The median temperature change across the ensemble is above 4°C for the higher latitudes (Alaska and Canada) and above 3°C for the continental US region. The individual model projections range from close to 3°C as their minimum and over 7°C as their maximum for the northern portion of the continent, and from just above 2°C and up to 5.8°C for the lower tier. Annual mean precipitation is very likely to increase in Canada and the Northeast USA, and likely to decrease in the Southwest.

3.4.1.5 Central and South America

The annual mean warming is going to be likely close to the global mean warming in the southernmost part of South America (median warming of 2.5°C, range between 1.7°C and 3.9°C) but larger than the global mean warming in the rest of the region (median warming of above 3°C, range between 1.8°C and over 5°C). Annual precipitation is likely to decrease in most of Central America and in the southern Andes, but there is less confidence in the models being able to simulate the regional variability in these mountainous regions. Winter precipitation in Tierra del Fuego and summer precipitation in south-eastern South America is likely to increase. The agreement of models over annual and seasonal mean rainfall change over northern South America, including the Amazon forest, is poor, and does not allow to draw conclusions in a direction or its opposite.

3 Climate Models and Their Projections of Future Changes 3.4.1.6 Australia and New Zealand

Warming is likely to be comparable to the global mean, with the southern areas warming less, especially in winter. Median projection is 2.6°C in Southern Australia, 3°C in the Northern part (ranges between 2°C and 4.5°C). Decreases in precipitation are consistently projected for South and Southwest Australia, especially in winter and spring. Precipitation is likely to increase in the west of the South Island of New Zealand. Changes in rainfall in northern and central Australia are uncertain.

3.4.2 Extremes

Indices of climate extremes have been devised to extract information from GCM simulations beyond the behavior of mean quantities. We would not trust GCMs to simulate the statistics of extremes that we observe at local scales: quantities simulated by GCMs are intended as averages over the grid boxes that divide up atmosphere, oceans, and land of the GCM domain. Still, within each model's scale and climatology, indices of tail behavior can be analyzed for changes under increased greenhouse gas forcings. In this case too, ensembles of GCMs are used to draw conclusions regarding the consistency of changes across simulations, i.e. the degree of inter-model agreement. It is also the case that changes in the behavior of simulated extremes can be considered in light of observed changes, and scientific understanding. In the latter case, changes in processes that we are already observing, or should be expected in the future, are explained and understood in the context of a system perturbed by increasing concentrations of CO2 in the atmosphere.

Many papers have recently addressed changes in extreme behavior. Here we briefly summarize some of our work that has specifically utilized GCM simulations. In Tebaldi et al. (2006) we analyzed five indices of extremes related to temperature:

• Frost Days, defined as the number of days in the year with minimum temperature below 0°C

• Growing Season Length, defined as the longest consecutive stretch of days in the year with mean temperature above 5°C

• Warm Nights, defined as the number of days in the year with minimum temperature (indicative of nighttime temperature) above the 90th percentile of climatology

• Heat Wave Duration, defined as the longest consecutive stretch of days in the year with maximum temperature exceeding climatological values by more than 5°C

• Extreme Temperature Range, defined as the difference between the warmest daily maximum temperature and the coolest daily minimum temperature in the year and five indices describing rainfall extreme behavior:

• Consecutive Dry Days, defined as the longest consecutive stretch of days in the year without precipitation

• Precipitation Intensity, defined as the annual average rain amount in wet days

• Number of Days with Rainfall Greater than 10 mm

• Percent of Total Precipitation Falling in Heavy Rain Days, defined as the percent of total yearly precipitation that fell in days whose rain amount exceeded the 95th percentile of wet-day climatology

• 5-Day Maximum Total Precipitation, defined as the largest amount falling in any consecutive 5 days during the year

Nine GCMs computed annual values of these indices from their gridded daily output of temperature (mean, min and max) and precipitation. The annual values were either averaged using the two traditional 20-year windows (present-day, 1980-1999 and future, 2080-2099) and the geographical patterns of the differences analyzed, or low-pass filtered time series (computed as 5-year running means) of global average values were considered.

The behavior of the five indices related to temperature extremes is consistent with what should be expected in a warming world. Heat Waves become longer, Frost Days diminish, Growing Season lengthens, Warm Nights become more numerous. The nine GCMs analyzed agree over the direction of the change, its significance and in large measure also over the geographical patterns of the changes. The analysis looked at three alternative SRES scenarios (high emission, A2, mid-emissions or business as usual, A1B and low-emissions, B1) and found significant differences in the intensification of the warming-related effects between lower and higher emission scenarios, especially in the second part of the twenty-first century. Interestingly however the geographical patterns of changes appear qualitatively similar across scenarios, in agreement with the pattern scaling arguments discussed above. These increases in temperature extremes are mainly the result of higher mean temperatures rather than increased interannual variability, as there is little model agreement on whether temperature variability will change (Meehl et al. 2007b).

Precipitation-related indices present a greater challenge in the quest for model agreement, at least in terms of spatial patterns. There are, however, some general messages that can be gathered from the analysis of the four indices related to intensification of rainfall: there is agreement across models that precipitation intensity will increase almost everywhere over land areas, in larger magnitude in the higher latitudes of the northern hemisphere. The level of inter-model agreement and statistical significance is less uniform over the globe than for temperature-related indices, with patches of regions where changes are not deemed significant by a majority of models. However, when averaged at the global scale all these indices show a significant increase, under all emission scenarios. Consecutive Dry Days is the index with larger inter-annual and inter-model variability. There are nonetheless large areas of the world where changes towards longest dry spells appear with a strong signal, like the Mediterranean basin, central Asia, South Africa, the Amazons and the West and Southwest of the United States.

Was this article helpful?

0 0
Renewable Energy 101

Renewable Energy 101

Renewable energy is energy that is generated from sunlight, rain, tides, geothermal heat and wind. These sources are naturally and constantly replenished, which is why they are deemed as renewable. The usage of renewable energy sources is very important when considering the sustainability of the existing energy usage of the world. While there is currently an abundance of non-renewable energy sources, such as nuclear fuels, these energy sources are depleting. In addition to being a non-renewable supply, the non-renewable energy sources release emissions into the air, which has an adverse effect on the environment.

Get My Free Ebook


Post a comment