The climate system - IPCC findings

The Intergovernmental Panel on Climate Change made its last Assessment of climate change science in 2007.1 The results are published in several books (IPCC 2007a-d; several thousand pages of text). The key findings concerning the underpinning science include what has been observed to change over the past century or so, and what is anticipated to happen through this century.

What has happened to the climate?

We are all aware that the climate where we live is variable from day to day and from year to year. Some of this variability is well understood in terms of its causes and some of it is less so. However, if we average climatic conditions - temperature or rainfall for example - over longer time periods and/or over larger areas of the Earth, then the degree of variability is decreased. For example, the average temperature of the Earth as a whole varies from year to year by no more than a few tenths of the degree Celsius, whereas diurnally and from day to day we can experience changes of 10°C or more; that is, variations a 100 times larger. Thus, in trying to identify secular changes in the climate, we will always have greater difficulties the more regional or short term our measurements.

Modern records of the climate of the Earth, with a few exceptions, commenced around 150 years ago. So it is only for this period of time that we can observe with a high degree of precision what has happened to the global climate system. In summary, the IPCC Fourth Assessment the Report concluded that:

Gases

• Currently concentrations of the greenhouse gases carbon dioxide, nitrous oxide and methane concentrations far exceed those of last 650,000 years

• The increases of these gases are primarily due to fossil fuel use, in the case of carbon dioxide, and agriculture and land-use changes for the other gases.

Temperature

• The warming is unequivocal, evident in air and ocean temperatures, melting of snow and ice and rising sea-levels (Figure 7)
  • Global mean temperatures have risen over past 100 years by 0.74 +/-0.18
  • Rate of warming in last 50 years is approximately double that of the last 100 (0.13oC +/- 0.03 per decade)
  • Warmest years 1998, 2005, 2002, 2004
  • Eleven of last 12 years rank amongst the 12 warmest years on record
  • Land warming faster than over the oceans
  • Snow cover has decreased in most regions, especially in spring and summer
  • Freeze-up and break-up dates for river and lake ice (variable). For the northern hemisphere:
    • Freeze-up occurring later a by 5.8 +/-12.6 days per century
    • Break-up occurring earlier at a rate of 6.5 +/-1.2 days per century
  • Arctic sea-ice extent decline of 2.7 +/-0.6 per decade
• The warming is very likely (>90% certainty) mainly an effect of human activities and likely to be due to natural variability with the impact of human activities being at least five times greater than that due to variations of the sun.

Figure 7: Change to the global climate system through the past 150 years. (a): General warming of the surface of the Earth: (b): Rise of sea level caused primarily by expansion of oceanic water as it warmed, and (c): The loss of snow cover in the Northern Hemisphere (Source: IPCC 2007d).

During the past 100 years, the earth has warmed by about 0.74°C. Warming of about the same amount has also occurred in Australia. This warming is observed at the surface of the Earth and in observations of the deep ocean temperatures where the warming is smaller but consistent in all major ocean basins of the world. The reader can look at the patterns of rainfall and temperature change over Australia by going to the Bureau of Meteorology website (see attached Table of web sites).

The Earth's systems have responded to this warming. These are responses that physically are very closely connected to the Earth’s temperature. The first of these is global sea level. Sea levels on average around the Earth have increased by about 18 centimetres over the past hundred years. This increase is largely due to the direct effects of warmer temperatures on the expansion of oceanic water, with small additional effects related to melting of ice and changes to land-water storage.

The second of these is glaciation - the melting of major snow and ice regions of the Earth again, a process where the physical connection between temperature is strong. Observations show clearly that, for example, the extent of Arctic sea-ice is now about 8%, reduced from what it was 25 years ago and its thickness decreased by about 40% (see Figure 8). Glaciers around the Earth are receding, and the length of the snow-cover season in northern parts of Eurasia and North America has been reduced by a month or more.

Other changes to the global climate system

Changes to the global climate system, that themselves are a response to the general warming, include wind patterns, precipitation, ocean salinity, sea ice, ice sheets and aspects of extreme weather. The IPCC (2007a) lists the following changes:

• Precipitation has generally increased over land north of 30oN from 1900-2005 and decreased in the tropics since 1970s
• There have been substantial increases in heavy precipitation events
• The have been more common droughts, especially in tropics and subtropics since 1970
• Tropospheric water vapour concentrations have been increasing
• “Global dimming” is neither global in extent nor has it continued after 1990
• Cloud changes dominated by ENSO appear to be opposite over land and ocean
• Changes of large-scale atmospheric circulation are apparent
• Mid latitude westerly winds have generally increased in both hemispheres
• Tropical cyclones have likely (>66%) increased in intensity since the 1970s.

The Intergovernmental Panel on Climate Change report on impacts, adaptation and vulnerability7 concluded that:

• Many natural systems (continental and oceanic) being affected by regional climate change (>90% confidence)
  • Natural systems are vulnerable to climate change and some will be irreversibly damaged.
• Global influences
  • Biological: Earlier time of leaf-unfolding and seasonal greening, bird migration, egg-laying, Pole-ward shift of ranges of plants and animals.
  • Physical: Glacial lakes, ground instability, enhanced runoff, heating of lakes.
• Human influences
  • Many human systems are sensitive to climate change and some are vulnerable
  • Projected changes in climate extremes could have major consequences for Indigenous livelihood
  • Some human systems have been affected by recent increases in floods and droughts
  • Human health-heat-related mortality, infectious disease vectors, allergenic pollen
  • Reducing greenhouse-gas emissions to stabilise their atmospheric concentrations would delay and reduce damages caused by climate change
  • Adaptation will be necessary to complement efforts to reduce net greenhouse gas emissions.
• Regional influences
  • Some coastal erosion impacts already
  • Agricultural and forestry management.

Many other changes to the climate system have been reported. They include a generally moister global atmosphere with more rainfall in some regions and less in others (more about this later). It also includes observed changes to cloudiness, to the intensity of storms and lengthening of growing season. Changing intensity of storms is of particular significance.

Figure 8: Observed seasonal Arctic sea ice extent (1900-2003). Decline is attributable to general warming (from the National Snow and Ice Data Center, Boulder Colorado; http://nsidc.org/).

Around the world, the insurance industry has become convinced that, at least in part, rising insurance liabilities have resulted from more intense storms than existed in the past. There are, of course, other factors that could be responsible for these trends, and these are considered by the insurance industry. But it is also true that in the past few years we have had more reports of severe tropical storms, tropical storm Katrina (Figure 9; August 25, 2005) that hit Florida being perhaps the most publicised example.

Figure 9: Satellite picture of tropical storm Katrina, August 25, 2005.

Theory suggests that a warmer surface with deep warm surface ocean waters provided greater energy for the development of such storms. An analysis of all storms identified around the globe from satellite imagery available for the last 30 years shows that while the frequency of tropical storms has not changed, the frequency of storms categorised as level four or five storms has increased by 100% (Figure 10; Webster et al. 2005).

Figure 10: Changes in the global number of tropical storms of three magnitudes Category 1, 2 and 3, and 4 and 5 (most extreme), showing that the overall frequency of storms has not changed but there has been a marked increase of more extreme storms. Based on Webster et al. (2005).

For Australia, observed changes include:

• Max temperature +0.54°C over last 90 years
• Min temperature +1.08°C over last 90 years
• Increase in hot days (>35°C) 5 days over past 50 years
• Increase in hot nights (>20°C) 9 nights over past 50 years
• Decrease in cold days (<15°C) 7 days over past 50 years
• Decrease in cold nights (<5°C) 8 nights over past 50 years
• Southern Ocean temperature +0.2°C over last 30 years
• Wetter north-western Australia
• Drier southern and eastern Australia
• Hotter droughts
• Increase proportion total rainfall from extreme rain-days in eastern Australia, decrease in southwest Australia
• Tropical cyclones decreased frequency, increased intensity
• Sea-level rise of 10 cm over past 80 years
• Glacial retreat at Heard Island.

There is a substantial published literature in which authors have observed changes to biological systems, the cause of which they attribute to changes in regional climate. Evidence for such changes linked to climate change in Australia/New Zealand include: a range of ecosystems - semi-arid woodlands, Eucalypt savannas, rain forest/woodland, subalpine, mangroves, coral reefs; many specific species or genera including birds, Antarctic beech, mammals, insects (including genetic changes), sea urchins, marine mammals, fish, invasive species; and behavioural changes such as flowering phenology, earlier migration, egg laying and seed production.

There is, perhaps, a little less certainty as to what extent these are climatically forced changes, as other factors may have been involved in at least some cases. We will consider this a little further in the next section.

Attributing cause

The climate of the Earth has changed significantly over geological timescales. In many cases the cause of such changes is understood. For example, it is understood that around 65 million years ago the Earth was struck by a 10 km diameter asteroid, which caused an immediate and substantial change to the climate that resulted from changes in atmospheric composition and persisted for some time. Associated with this were major changes to the distribution of plants and animals around the world.

It is now established that during the past million years, the Earth’s climate has oscillated between interglacial conditions, similar to those of today, and glacial conditions where much of the Earth’s surface was covered with ice and snow. These variations have been driven by the obliquity of the Earth on its axis and the precession of the equinoxes that resulted in slight changes to the amount of sunlight arriving at the Earth’s surface and to its distribution between the two hemispheres. Using the climate models described above, scientists have shown that a significant amount of the cooling and warming fluctuations over this period can be described by this “wobbling” of the Earth on its axis. Interestingly enough, the degree of cooling during the glacial periods can only be described quantitatively when, in addition to the solar radiation effects, the changing water vapour, carbon dioxide, methane and nitrous oxide levels in the atmosphere are also included. In this case, temperature change was followed by carbon dioxide change that then enhanced the temperature effects - this unlike current global warming, where carbon dioxide concentrations lead temperature changes.

Periodically, the Earth is subjected to major volcanic eruptions, singly or in groups, which are large enough to inject into the upper atmosphere volcanic dust that reflects solar radiation and lowers the amount of energy available at the Earth’s surface. Following such eruptions, the Earth cools by a few tenths of a degree for a period of about two years that corresponds to the time it takes for the dust to be slowly removed from the atmosphere. Such events are not predictable but very likely will continue in the future.

Modern observations also show us that the Sun is not always emitting exactly the same amount of energy. Over several years or even a decade or so, changes of a few tenths of a degree Celsius in the Earth’s mean temperature can result from these fluctuations. Again, these variations are not predictable at present, but are likely to continue. The volcanic and solar radiation variations together add to the variability of the Earth’s temperature on decadal timescales, by overlaying the general warming that is occurring due to greenhouse gases.

There is some evidence that the clearing of land also affects at least regional climate. This might be expected as such clearing lowers the surface resistance to air flowing over it and changes the hydrologic balance of the region and also the reflectivity of the surface. Again, these are variations that are generally small, regional and overlay the general warming trend.

Mainstream climatic science is now convinced (with greater than 90% certainty) that at least most of the general warming that occurred over the past 50 years was due to greenhouse gases and not to the other potential causes of climate variability or change. This conclusion has been reached after rigorous examination of each of the known causes of historical/geological climate variations to see if any one or number of these could have caused the observed change. But this is not to deny that uncertainty remains with respect to attributing changes of specific components of the climate system to global warming due to greenhouse gases (See Figure 11).

Changes that are closely connected physically to the general warming process, processes such as the expansion of sea water and thus sea-level rise, and the melting of ice, and thus de-glaciation, are relatively well understood and simple. Thus the observed changes in global sea level and the loss of land, sea and coastal ice can be confidently attributed to the warming
process. However, the more complex the physical connections are, such as where there are opportunities for interactions between competing effects, and positive or negative feedbacks between components of the more complex system, the more difficult it is to draw conclusions about what is cause and effect.

Figure 11: The attribution of observed climatic change (and their impact on human/natural ecosystems) on higher concentrations of greenhouse gases depends on the complexity of the connections (physical and dynamical) and the number of confounding extraneous factors.

Further, the more regional and the more short-term specific observations are, the less confident scientists will be that those changes can be related directly to the general warming of the planet. The more remote an observed change is both physically or dynamically from the general warming of the climate system, the lower the level of confidence in cause and effect will be. Indeed, when one is looking at the response of biological/human systems to the climate then both the level of complexity rises and the possibility of other factors forcing change becomes higher. Again, scientists will be less confident in their attribution of changes to the underlying warming of the planet.

Finally, the system is not under experimental conditions, where all other things are held constant. Observed changes can be caused by other human-induced impacts such as land clearing, air pollution and land-use practices. In complex systems, where the climate has impacted on the regional environments together with other human effects, confident attribution of change to the original climate change remains difficult, if not impossible.

But it is important to also realise that confidence will not be expressed in a cause-and-effect relationship by climate scientists unless in the testing of a hypothesis they are 99% sure that that connection exists. This is the basis of the scientific approach and culture. However, in the real world, in assessing one’s exposure or risk associated with changes, the probability required in order to determine action is a totally different probability. It is highly dependent on the magnitude of the change that is anticipated. For example, no one would get on aircraft if they were only 99% sure that the aircraft was not going to crash. This simply reflects the fact that the magnitude of the outcome is sufficiently severe that a much lower probability of a crash is required. Similarly, if the future of agriculture, or the Great Barrier Reef, or Melbourne’s water supply, is likely to be severely impacted by climate change, then policy development will not and should not wait until scientists are 99% sure that it is going to happen.

Figure 12: Rarely are decisions made with the benefit of perfect and comprehensive knowledge. Thus, they are best handled in a risk management framework where risk is dependent on both the magnitude of the impact of an event, if it happens, and its probability of occurrence (after Allen 2005).

Most decisions that are made by governments, companies or individuals are made without perfect information and in a risk management framework (Figure 12). Here, the risk is assessed as the product of both the probability of an event occurring, and the magnitude of the impact if it does occur. This has to be weighed against the capacity of natural or human systems to adapt to the magnitude and rate of the change. In the climate change issue, where such risks threaten long-term resilience, then intervention may be necessary through deliberate and advanced adaptation, and/or mitigation of the effect via the strategic modification of the long-term characteristics of the global energy system.

What are the projections for the future?

The climate of the future will remain variable. This variability will be superimposed on a gradual warming, with associated changes to the circulation of the oceans and atmosphere that will lead to regional trends in the climate. How much the global average climate will change depends primarily on how much human activities further change the concentration of greenhouse gases, particularly carbon dioxide, in the atmosphere. As carbon dioxide is released mainly into the atmosphere from the combustion of fossil fuels, then the future depends very much on the nature of the global use of energy. This has two components. First, the demand for energy derives from the requirement/necessity for the growth in use to meet economic and social needs, particularly in the developing world. Second, the introduction of improved efficiencies in the transference of energy to useful outcomes (transport, heating, cooling, industrial processes) and the application of technologies that avoid the emission of carbon dioxide in the atmosphere.

In this and the coming decade, the Australian Bureau of Agriculture and Resource Economics projects that Australians will require an annual growth in energy use of 2.1%. Other countries will be able to satisfy their needs with a reduction in energy use although most will require a growth rate similar to that in Australia or larger for developing countries.

Assessing exactly how humans will meet these energy requirements and at the same time meet the challenges of changing climate is extremely difficult. The IPCC has chosen to develop emission scenarios, each representing a different view of how the future may unfold (IPCC 2000; see examples in Table 4). These are not predictions of the future, but alternative futures that can be examined in terms of the climatological consequences, should they turn out to be what happens.

Table 4: Selected IPCC scenarios for which temperature rises through this century are projected in Figure 13. For more details see IPCC (2000) and http://www.ipcc.ch/pub/srese.
pdf.

All of these scenarios lead to a growth in carbon dioxide in the atmosphere through this century. They all therefore anticipated further warming of the Earth and concomitant changes to the climate. The range of anticipated warming through this century, reflecting these alternative futures, and the uncertainty that still exists in the sensitivity of the planetary temperature is a concentration of between 1.1°C and 6.2°C (see Figure 13).

Figure 13: Projected rise in temperatures (°C) from 1990 through 2100 for a range of possible emission scenarios. The key refers to specific IPCC scenarios (see Table 3 for more information; IPCC 2000). From IPCC (2007d).

In summary the projected global climatic changes are as follows:

• Mean temperatures
  • 2025 0.6-0.7°C Higher over land/high latitude
  • 2095 1.7-4.0°C
• Extreme temperatures
  • Rise in frequency of extremes of maximum; decline in extreme minimum
  • More frequent, intense Longer lived heat waves
  • Decrease in frost days Mid to high latitudes
  • Increased growing season Mid to high latitudes
• Mean precipitation
  • Increase in high latitudes
  • Decrease sub-tropics/mid latitudes
• Extreme precipitation
  • Intensity of events to increase
  • Longer periods between events (sub-tropics/mid latitudes
• Tropical cyclones (hurricanes, typhoons)
  • Increased peak wind and precipitation
  • Overall less frequent; geographic shifts uncertain
• Mid latitude storms
  • Fewer with a pole-ward shift (several degrees); increased wind speed and wave heights
• Snow and ice
  • Snow cover and sea-ice extent decrease
  • Glaciers and ice caps lose mass
  • Loss of Arctic sea ice as early as mid 21st century
• Carbon cycle
  • Loss of CO2 absorption efficiency
  • Greater atmospheric accumulation of carbon dioxide
• Sea level
  • By end of century from end of 20th century, 0.19-0.58 metres
  • Limited knowledge of potential additional increase due to melting of ice flows
• Ocean acidification
  • 0.14-0.35 pH units in 21st century
  • Southern Ocean exhibits under-saturation with consequences for marine organisms.

Recently the CSIRO (Australia) and the Australian Bureau of Meteorology analysed the combined projections for all 23 IPCC models under the full range of possible emissions scenarios.7 With some assumptions about the independence of these, an attempt was made to specify the probability of certain levels of warming and other climatic changes. The following are the estimates made in this study of changes in Australia where the mean value is the 50th percentile (most probable projections and the range represent the 10th to 90th percentile).

Temperature by 2030 (Figure 14)

• Mean:
  • 1.0°C (0.6-1.5) Slightly less on coast/ more inland
• Extremes:
  • Increase in diurnal range, hot days and warm nights, modest increase in frost frequency

Precipitation by 2030 (Figure 15)

• Mean:
  • Decrease almost everywhere
• Intensity:
  • Increased daily precipitation intensity, dry days.

Snow

• Decrease in depth and season length
• Earlier maximum depth

Solar radiation

• Slight increase in south

Relative humidity

• Small decrease by 2030 (1.07±2.0-1.5% over most of Australia

Potential evaporation (Figure 16)

• Increase over Australia by 2030; largest in the south and east plus 2% (0-6%)

Drought

• Increase occurrence over most of Australia by 2030
• Greatest in southwest

Wind

• Tendency for small increase winds speeds over most coastal areas of 2-5% by 2030
• Decrease at 30oS in winter

Fire risk

• Substantial increase in fire risk in southeast
• Risk to be determined elsewhere

Sea level

• Global increase of 18-59 cm by 2100
• Some regional differences
• Possibly greater by 10-20 cm due to de-glaciation

Ocean acidification

• Increase, most at high latitudes
• Aragonite saturation at high latitudes by 2050

Severe weather

• Tropical cyclones: Likely increase in more intense categories
• Possible decrease in numbers overall

Thunderstorms

• Hail risk increase in southeast

Climatic modes

• El Nino events drier; frequency not necessarily changed
• Southern Annual Mode: Trend toward positive phase (weaker westerlies over southern Australia).

Figure 14: Best estimate (50th percentile)8 of change of seasonally averaged surface temperature (oC) relative to 1990 by 2030 for the A1B emissions scenario (moderate emissions growth). Source CSIRO/BoM (2007).

Figure 14 suggests that by 2030 Australian temperatures will warm by approximately the same amount as through the past 100 years. This will be slightly seasonally dependent and greater at night and in inland regions. Precipitation is, however, more problematic, but the indication from the combined projections of 23 IPCC climate models is that there will be a loss of rainfall almost everywhere across the country (Figure 15). This is a very confident projection particularly south of 30oS and less certain in the north where there remain uncertainties concerning monsoonal activities.

Available water, that is, that which remains in the soil or runs off into rivers and dams for future use, is the net difference between the amount of precipitation and the evaporation. Both of these are relatively large numbers with the difference being often quite small. It is for this reason, at least, that when rainfall changes occur, we often experience large percentage changes in available water. But when both rainfall decreases and evaporation increases, there is a potential for very large changes in available water. Figure 16 shows that the evaporation from open water surfaces, soils and plants (evapo-transpiration) is anticipated, by 14 of the IPCC models, to increase across the country. When coupled with rainfall decreases this suggests that Australia’s most probable prognosis is a far drier continent.

These changes, together suggest the potential for substantial habitat loss for a wide range of ecosystems such as rainforests, wetlands, coral, coast lines, alpine areas, and consequently, the reduction of ecosystem services relevant to water supply, biodiversity, tourism, fishing, etc. Some of the most vulnerable ecosystems identified by the IPCC Fourth Assessment Report (IPCC 2007b) are outlined in Table 5.

Figure 15: Best estimate (50th percentile) of change of rainfall by 2030 (% of 1961-1990) (Emissions scenario A1B; weighted results of 23 models). Source CSIRO/BoM (2007).

Figure 16: Best estimate (50th percentile) of change of evapo-transpiration by 2030 (% relative to 1990; Emissions scenario A1B; weighted results of 14 models). Source CSIRO/BoM (2007).

Coastal settlements are and will be increasingly exposed to sea-level rise, intense storm surges/cyclones, increased erosion, degraded beaches and infrastructure. Bush fire frequency and intensity, urban drainage, human health (see WHO 2003), invasive species, refugees of vulnerable/disadvantaged peoples of the world (Dupont and Pearman, 2006).

Together, these changes are likely to be reflected in increased social and economic trauma, issues of cultural significance, change insurability and insurance costs. Such changes will require adaptive responses as, irrespective of the actions taken to mitigate against further warming, the planet is already committed to 0.5-1.0°C further warming.

Table 5: Potential change to some key and vulnerable ecosystems in Australia with the wider qualitative implications to the economy (based on IPCC 2007b).