Atmosphere

Theme commentary
Tom Beer, Michael Borgas, Willem Bouma, Paul Fraser, Paul Holper and Simon Torok
CSIRO Atmospheric Research
prepared for the 2006 Australian State of the Environment Committee, 2006

Major issue: climate change

As a result of human activities over the past 200 years, there have been increases in atmospheric greenhouse gas levels. Those due to human activities are known as anthropogenic. Continued increases in the levels of anthropogenic greenhouse gases, such as carbon dioxide, methane and nitrous oxide, are expected to lead to continued global warming and regional climate change.

Allowing for uncertainties in future greenhouse gas emissions and the response of the climate system, the CSIRO has estimated that by 2030 there will be a warming  of ‘0.4 to 2 °C over most of Australia, with slightly less warming in the north-west. By 2070, annual average temperatures will increase by 1 to 6 °C over most of Australia with spatial variations similar to those of 2030’ (Figure 4). 

In terms of rainfall  CSIRO has estimated that changes will tend to decrease in the south-west (-20% to +5% by 2030 and -60% to +10% by 2070), and in parts of the south-east and Queensland (-10% to +5% by 2030 and -35% to +10% by 2070) (Figure 5). Most other locations show changes which vary from -10% to +10% by 2030 and -35% to +35% by 2070. Decreases are more pronounced in winter and spring. .  

Subsequently, CSIRO and the Australian Greenhouse Office (2002) have estimated the likely changes in temperature and rainfall for 2030 and 2070. They have projected that temperatures will increase by around 1 degree by 2030 (± 0.8) and by between 3 and 3.5 degrees by 2070 (± 2.5). In terms of rainfall, most locations show a 4% decrease in rainfall by 2030 and a 12% decrease by 2070.


Sea level increases are discussed in Section 2.2 of the Coasts and Oceans Theme Commentary but the CSIRO has estimated that a rise in sea level of 9 to 88 cm by 2100 can be expected (CSIRO 2001).

Most sectors will be affected by global warming. Allen Consulting Group (2005, pp. 48–81) consider three major sectors: climate pressures on industry, natural environment, and on health and infrastructure. Some of the expected impacts include less water for cities such as Perth, an increased fire risk in south-eastern Australia and Tasmania, and the southward spread of mosquito-borne diseases in south-east Queensland (CSIRO and Australian Greenhouse Office 2002).

In 1992, Australia was a signatory to the UN Framework Convention on Climate Change and as a result established the Australian Greenhouse Office in 1998. One of the obligations under that convention was to develop, update and periodically publish national inventories of greenhouse gases, which take into account both sources and sinks. The rules to be used in developing these national inventories were worked out by the Intergovernmental Panel on Climate Change.

The Kyoto Protocol to the United Nations Framework Convention on Climate Change was negotiated in Kyoto, Japan in December 1997.The agreement came into force on February 16, 2005 following ratification by Russia on November 18, 2004. Australia signed the Kyoto Protocol but, along with the United States, refused to ratify it. Nevertheless, Australia is committed to meeting the Kyoto targets  that were established in 1997. Australia agreed to limit its greenhouse gas emissions by the first commitment period (2008 – 2012) to less than 108 per cent of its emissions in the benchmark year. As agreed under the UN Framework Convention on Climate Change, the benchmark year is 1990.

In Australia, national greenhouse gas emissions are compiled in the National Greenhouse Gas Inventory. Emissions are reported under the guidelines established by the UN Framework Convention on Climate Change as well as those established under the Kyoto Protocol.

From a global perspective, anthropogenic emissions represent the underlying forcing of climate change, and accurate national data are an essential step towards sound management and, ultimately, coordinated approaches to emission reductions.

Greenhouse gases

Carbon dioxide is the most important greenhouse gas that affects climate change. Its increase has contributed approximately 60 per cent of the additional heat that is trapped by the atmosphere. This additional heat is known as the enhanced radiative forcing and it has driven climate change over the past 250 years (Ramaswamy et al. 2001, p. 358, Table 6.1).

Carbon dioxide  data collected at Tasmania’s Cape Grim since 1976 have shown that concentrations have risen from 330 parts per million (ppm) in the mid-1970s to in excess of 375 ppm by the mid-2000s (Figure 6). The long-term growth rate, due to carbon dioxide release to the atmosphere from the burning of fossil fuels, has averaged about 1.5 ppm per year (Francey 2005).

Figure 6: Atmospheric carbon dioxide levels (blue, ppm) and annual growth rates (red, ppm) as observed at Cape Grim, Tasmania from 1976 to 2005

 Atmospheric carbon dioxide levels and annual growth rates as observed at Cape Grim, Tasmania from 1976 to 2005

Source: CSIRO Atmospheric Research and Cape Grim Baseline Air Pollution Station, Australian Bureau of Meteorology, January 2005.

Both the highs and lows in atmospheric carbon dioxide  growth rate are associated with El Niño events. The highest growth rate observed at Cape Grim was three ppm per year in 1998 due to release of carbon dioxide by El Niño-driven biomass burning events in the tropics; the lowest growth rate was about 0.5 ppm per year in the early 1980s, due to reduced release of carbon dioxide by the equatorial Pacific Ocean during El Niño events (when upwelling of carbon dioxide-rich waters is suppressed). During most major El Niño events, the enhanced terrestrial carbon dioxide releases due to drought-induced biomass burning are larger than the reduced ocean releases of carbon dioxide due to suppressed upwelling (Prentice 2001; Peylin et al. 2005).

In the early 1990s, there was a period of low global carbon dioxide growth rates, possibly associated with a large terrestrial carbon sink due to mid- to high-latitude cooling, following the eruption of the Mt Pinatubo volcano (Prentice 2001).

A comparison of Cape Grim data with measurements of air trapped in Antarctic ice shows that carbon dioxide levels have increased by 35 per cent since 1750 (Etheridge et al. 1996). Approximately half of the carbon dioxide released from the consumption of fossil fuels in the 1990s remained in the atmosphere, with the other half being removed by the oceans (27 per cent) and the biosphere (23 per cent) (Prentice 2001, Table 3.1).

The underlying, long-term global carbon dioxide growth rate has increased from less than 1.5 ppm per year in the mid-1970s to more than two ppm per year in response to international fossil fuel emissions, which have increased from between four and five billion tonnes of carbon per year in the mid-1970s to between six and seven billion tonnes of carbon today. This is despite global attempts to increase the efficiency of fossil fuel use. The major drivers of this increased demand for fossil fuels have been growing global population and the industrialisation of the developing world. Carbon dioxide from fossil fuels will continue to be the dominant factor in climate change due to greenhouse gases over the next 100 years.

Methane  is the second most important greenhouse gas that affects climate change. It has contributed approximately 20 per cent of the enhanced radiative forcing of all greenhouse gases over the past 250 years, an amount equal to one-third (33 per cent) of that due to carbon dioxide (Ramaswamy et al. 2001, Table 6.1). Methane also contributes indirectly to climate change through its role in the production of ozone, a powerful greenhouse gas, in the lower atmosphere. Recent calculations suggest that the direct and indirect climate forcing due to methane increases could be as much as 60 per cent of that due to carbon dioxide increases over the same period (Shindell et al. 2005).

Methane data , collected at Tasmania’s Cape Grim since 1976, show that concentrations rose from 1450 parts per billion (ppb) in the mid-1970s to 1700 ppb by the mid-2000s (Figure 7). Anthropogenic methane is released to the atmosphere from energy generation using fossil fuels (25 per cent), from ruminants (25 per cent), from rice agriculture (20 per cent), from landfills and waste treatment (15 per cent) and from biomass burning (ten per cent). In recent years the global contribution from rice agriculture has decreased, while that from energy, waste treatment and landfills has increased (Prather et al. 2001, pp. 248–251).

Figure 7: Atmospheric methane levels measured at Cape Grim, Tasmania (blue, ppb) and in the Cape Grim archive (green, ppb) and annual growth rates (red, ppb) from 1978 to 2005

 Atmospheric methane levels measured at Cape Grim, Tasmania (blue, ppb) and in the Cape Grim archive (green, ppb) and annual growth rates (red, ppb) from 1978 to 2005

Source: CSIRO Atmospheric Research and Cape Grim Baseline Air Pollution Station, Australian Bureau of Meteorology, January 2005.

The long-term growth rate of methane  has declined steadily over the past 20 years, from about 15 ppb per year in the mid-1980s to about zero over the past six years. The reason for the decline is uncertain, but possibly involves a reduction or stabilisation in methane release from the oil and gas industry, as methane is now considered a valuable resource; in the 1960s, unwanted methane was commonly either burnt (flaring) or released under high pressure (venting). A worldwide trend to minimise leaks from natural gas reticulation networks, particularly in the former Soviet Union, might also have contributed to the trend.

As well as showing a long-term decline, the methane growth rate  shows significant variability, the causes of which remain elusive. The growth rate was anomalously large in 1991 and 1998. The 1991 anomaly was followed by a drop in growth rate, which has been linked to the Pinatubo eruption in 1991, either by impacting on the chemical destruction of methane in the atmosphere or by suppressing methane emissions from wetlands. The 1998 anomaly was caused by tropical biomass burning.

Despite the observational evidence, atmospheric modellers continue to assume that methane is still growing in the background atmosphere and will continue to do so for the foreseeable future. The fact that methane levels have remained approximately constant over the past six years is at odds with the methane growth scenarios incorporated into the Intergovernmental Panel on Climate Report (IPCC 2001, pp. 266–267) and this will need to be addressed in future scenarios.

The third most important greenhouse gas is dichlorodifluoromethane (CFC-12). CFC-12, and all the Montreal Protocol gases combined, have contributed seven per cent and 13 per cent, respectively, of the enhanced radiative forcing over the past 250 years (Ramaswamy et al. 2001, Table 6.1). Under the Montreal Protocol, CFCs stopped growing in the global background atmosphere in the mid-1990s and are now declining by 0.3 per cent per year. In contrast, CFC-12 will continue to be the third most important greenhouse gas contributing to climate change since pre-industrial times, until it is overtaken by nitrous oxide in about the year 2010.

Nitrous oxide  is currently the fourth most important greenhouse gas impacting on climate change. Nitrous oxide levels have increased by about 17 pre cent and contributed six per cent of the enhanced radiative forcing over the past 250 years (Ramaswamy et al. 2001, Table 6.1). The major nitrous oxide sources are from soils (nitrifying and denitrifying bacteria, 40 per cent), agricultural activities employing nitrogen fertilisation (27 per cent), ocean biology (20 per cent), industrial processes (seven per cent, from nylon production, nitric acid production, fossil fuel fired power stations and vehicular emissions) and biomass burning (three per cent) (Prather et al. 2001, Table 4.2).

The enhanced nitrous oxide emissions from agricultural and natural ecosystems are believed to be caused by increasing soil nitrogen availability, which is driven by increased fertiliser use, increased molecular nitrogen fixation by soil bacteria, and increased nitrogen deposition (atmosphere to soils) (Nevison and Holland 1997).

The National Greenhouse Gas Inventory shows a slight overall decrease (–0.5 per cent) in Australian carbon dioxide emissions  from 1990 to 2003. Methane emissions are 4.4 per cent down from 1990. Other gases such as hydrochlorofluorocarbons (HFCs), PFCs and SF6 show a recent, large increase. Since 1990 there has been a steady and relatively large increase (40 per cent) in nitrous oxide emissions.

The overall trend in national greenhouse gas emissions  shows that reductions from 1990 levels in the years 1991 to 1997 (attributable primarily to carbon dioxide and methane) have not been maintained over 1998 to 2003, and that national emissions are now slightly above 1990 levels. By 2004 Australia’s net greenhouse gas emissions were 564.7 Mt CO2 -equivalents, which is a 2.3 per cent increase over 1990.

Australia now highlights estimates of emissions of greenhouse gases using the accounting procedures recommended in the Kyoto Protocol, which includes land clearing. Trees are sinks of carbon dioxide, so including land clearing in the inventory for 1990 decreases the available sinks and hence increases the actual emissions. In subsequent years, land clearing declined so that these subsequent years will show greater sinks compared with 1990. This change in accounting procedure abruptly increased estimated emissions for the benchmark year of 1990 (when a lot of land clearing took place).

Land clearing  has been (and still is) a significant factor for Australian greenhouse gas emissions. Land-clearing rates have varied over the data period (1990–2003), and, as discussed earlier, affect the calculated total carbon dioxide emissions (for more information and analysis, see AGO 2005).

Total net anthropogenic greenhouse gas emissions for Australia grew slightly over the period 1990–2004, albeit with some significant variations in total emissions  over time.

Australia’s efforts to curb greenhouse gas emissions have had an effect. Per capita emissions  declined from 32.2 in 1990 to 28.2 in 2004 (expressed in tonnes of CO2-equivalent) (Figure 8). In addition, greenhouse gas emissions per capita have fallen from 1.1 kilograms CO2-equivalent per dollar of gross domestic product  in 1990 to 0.7 in 2004 (Figure 9). This result represents a significant fall, and suggests that, at least in terms of using gross domestic product as an indicator, Australia has become more ‘greenhouse gas efficient’ or ‘carbon efficient’.

Figure 8: Australia’s greenhouse gas emissions per capita, 1990–2003

 Australia's greenhouse gas emissions per capita, 1990-2003

Source: adapted from AGO (2005, Tables A4.1 and A4.3)

In terms of the relative contribution of the various sectors  to Australia’s net CO2-equivalent emissions, stationary energy emissions, dominated by electricity production, accounted for 49.6 per cent of greenhouse gas emissions in 2004 (Figure 10). They have increased by 43 per cent from 1990 to 2004, with a particularly sharp rise from 1998 to 1999; most of the emissions  are carbon dioxide from coal, petroleum, natural gas and biomass. The largest increase was for coal, which now accounts for over 55 per cent of all stationary energy emissions. Emissions from petroleum also increased, but at a relatively lower rate. Natural gas emissions also increased significantly (45 per cent), even though as a proportion of all stationary energy sources, natural gas emissions are still small (14 per cent in 2003 and declining). The rather minor contributions from biomass used for energy generation also decreased.

Figure 10: Trends in carbon dioxide equivalent emissions and removals by sector,  1990–2004

 Trends in carbon dioxide equivalent emissions and removals by sector, 1990-2004

Source: DEH and AGO, 2006, National Greenhouse Gas Inventory 2004

Land use, land use change and forestry contributions  (6.3 per cent of greenhouse gas emissions in 2003) decreased very significantly in the period from 1990 to 2003, declining by 72.5 per cent, with a particularly sharp decline from 1990 to 1997. Agricultural emissions declined by about 1.6 per cent from 1990 to 1994, but started to rise until 2001, and then declined again. These fluctuations reflect climate variability and can be accounted for by changes in emissions from burning savanna and the number of livestock in the country. Thus, for example, the decrease in emissions from 2002 to 2003 was due to widespread drought over south-eastern Australia that led to declines in agricultural production.

The end result is that, as a proportion of total greenhouse gas emissions , stationary energy has increased from 36 to 50 per cent, transport emissions have increased from 11 to almost 14 per cent, agricultural emissions accounts for the same proportion at around 16.5 per cent, while land use, land-use change and forestry emissions have dropped from 23 to six per cent.

Hence, with only modest further reductions likely to be achievable in the land use area, and growing energy emissions, the trends in 1990–2004 data suggest that Australia’s overall greenhouse gas emissions (having risen only by 2.3 percent from 1990 to 2004) are likely to increase from 2004 onwards

Individual greenhouse sectors

  • Transport
  • Agriculture
  • Land use
  • -->

    Transport

    In 2004, the transport sector accounted for around 13.5 per cent of Australia’s net greenhouse emissions (up from 11 per cent in 1990) and it is a significant driver of net greenhouse gas emissions . Road transport emissions  increased by 29 per cent from 1990 to 2003; but because road transport emissions represent 90 per cent of the total transport emissions, they are also responsible for most of the increase. Domestic air transport emissions increased by 79 per cent, and are now about 6.6 per cent of total transport emissions.

    For road transport, passenger vehicles account for 55 per cent of emissions in 2003 (slightly down from 63 per cent in 1990), with trucks and light commercial vehicles accounting for the balance.

    Petrol continues to be the dominant contributor to emissions from the passenger vehicle class, although diesel has a dominant share in commercial vehicles. Within the passenger vehicle class, four-wheel-drive vehicles have had a rapid growth and this is expected to increase at the expense of other passenger vehicles. This translates into an increasing share of emissions that are attributed to four-wheel-drive passenger vehicles. Also, because four-wheel-drive vehicles are more fuel intensive, their carbon dioxide emissions increase more than proportionally (Wadud et al. 2003).

    Agriculture

    In 2004, the agricultural sector accounted for around 16.5 per cent of Australia’s nett greenhouse gas emissions  and, therefore, it is one of the major drivers in emissions.

    Carbon dioxide equivalent emissions from the agricultural sector  grew by two per cent from 1990 to 2004. The largest contributor is ‘enteric fermentation’ in livestock: this is the methane emitted by ruminants such as cows and sheep. Thus greenhouse gas emissions from agriculture depend on livestock numbers and, to a lesser extent, fertiliser application. Greenhouse gas emissions from agriculture decreased from 1991 to 1996, and then increased slowly from 1997 to 2001. Enteric fermentation in livestock now contributes 66 per cent of agricultural emissions (down from 74 per cent in 1990). Emissions associated with agricultural soils (for example, disturbance of land by cropping, improved pastures and the application of fertilisers and animal wastes) increased by 17 per cent from 1990 to 2004, with most of the increase occurring during 1997–2002. Agricultural soil emissions now represent 18 per cent of all agricultural emissions. The most significant increase came from prescribed burning of savannas, which grew by 67 per cent over the reporting period, and now represents 12 per cent of all agricultural emissions (compared with seven per cent in 1990). Minor sources (manure management, rice growing and field burning) are increasing, but they are too small to have a significant impact overall.

    Greenhouse gas emissions from agriculture strongly reflect climate variability. During drought years, livestock numbers and the area of fertilised crops are both lower, reducing greenhouse gas emissions from these two sources. These reductions are possibly offset by increases in savanna burning during dry years.

    Land use

    Emissions from land use, land use change and forestry  accounted for about six per cent of nett emissions by 2004. This sector was previously a much larger contributor, accounting for 23 per cent of net emissions in 1990. This decline, acting as a major offset for increases in other sectors, is the main reason that Australia’s overall nett greenhouse emissions have not risen significantly over the period 1990 to 2003; it effectively counteracts increases in other sectors.

    Discussion and conclusion: Climate variability and change

    The histogram of annual mean temperature anomalies  illustrates both climate variability and climate change (see Figure 2). Both the linear trend line and the graph showing the five-year mean have, since 1980, shown positive temperature anomalies. Although strongly indicative of global warming (within Australia, at least), the variability is also large (with a range of 1.8 °C), with the result that both effects need to be factored in to future assessments of climate trends. Australia State of the Environment 2001 stated that: ‘since 1910, Australian average surface temperature has increased by 0.76 °C, which is consistent with the global temperature increase of 0.6–0.7 °C’ (Manins, 2001, 3). The trend line indicates that in 2005 the Australian average surface temperature was 0.84 °C above that in 1910.

    For Australia, climate change models  indicate temperature increases of 1–6 °C by 2070. Rainfall change around the country will exhibit more spatial variability than temperature, but there is expected to be markedly less rainfall in Western Australia and significantly less rainfall in south-eastern Australia as global warming manifests itself. Some of the impacts that such changes in temperature and rainfall will have on the Australian environment and economy are covered in the Biodiversity , Coasts and oceans , Land , and Human settlements .

    In 2002, Australia changed the methodology used to estimate greenhouse gas emissions to follow the accounting procedures recommended in the Kyoto Protocol by including land clearing. On this basis, the increase between 1990 and 1998 was greatly reduced from the figure quoted in Australia State of the Environment 2001 (ASEC 2001). In 2006 it was estimated that the nett emissions in 2004 was 2.3 per cent above 1990 levels. Australia, though not formally bound by the Kyoto Protocol, continues to be committed to meeting a target of emissions being less than 108 per cent of 1990 levels by 2012.