IPCC Fourth Assessment Report: Climate Change 2007
Climate Change 2007: Working Group I: The Physical Science Basis

Executive Summary

Radiative forcing (RF)[1] is a concept used for quantitative comparisons of the strength of different human and natural agents in causing climate change. Climate model studies since the Working Group I Third Assessment Report (TAR; IPCC, 2001) give medium confidence that the equilibrium global mean temperature response to a given RF is approximately the same (to within 25%) for most drivers of climate change.

For the first time, the combined RF for all anthropogenic agents is derived. Estimates are also made for the first time of the separate RF components associated with the emissions of each agent.

The combined anthropogenic RF is estimated to be +1.6 [–1.0, +0.8][2] W m–2, indicating that, since 1750, it is extremely likely[3] that humans have exerted a substantial warming influence on climate. This RF estimate is likely to be at least five times greater than that due to solar irradiance changes. For the period 1950 to 2005, it is exceptionally unlikely that the combined natural RF (solar irradiance plus volcanic aerosol) has had a warming influence comparable to that of the combined anthropogenic RF.

Increasing concentrations of the long-lived greenhouse gases (carbon dioxide (CO2), methane (CH4), nitrous oxide (N2O), halocarbons and sulphur hexafluoride (SF6); hereinafter LLGHGs) have led to a combined RF of +2.63 [±0.26] W m–2. Their RF has a high level of scientific understanding.[4] The 9% increase in this RF since the TAR is the result of concentration changes since 1998.

— The global mean concentration of CO2 in 2005 was 379 ppm, leading to an RF of +1.66 [±0.17] W m–2. Past emissions of fossil fuels and cement production have likely contributed about three-quarters of the current RF, with the remainder caused by land use changes. For the 1995 to 2005 decade, the growth rate of CO2 in the atmosphere was 1.9 ppm yr–1 and the CO2 RF increased by 20%: this is the largest change observed or inferred for any decade in at least the last 200 years. From 1999 to 2005, global emissions from fossil fuel and cement production increased at a rate of roughly 3% yr–1.

— The global mean concentration of CH4 in 2005 was 1,774 ppb, contributing an RF of +0.48 [±0.05] W m–2. Over the past two decades, CH4 growth rates in the atmosphere have generally decreased. The cause of this is not well understood. However, this decrease and the negligible long-term change in its main sink (the hydroxyl radical OH) imply that total CH4 emissions are not increasing.

— The Montreal Protocol gases (chlorofluorocarbons (CFCs), hydrochlorofluorocarbons (HCFCs), and chlorocarbons) as a group contributed +0.32 [±0.03] W m–2 to the RF in 2005. Their RF peaked in 2003 and is now beginning to decline.

— Nitrous oxide continues to rise approximately linearly (0.26% yr–1) and reached a concentration of 319 ppb in 2005, contributing an RF of +0.16 [±0.02] W m–2. Recent studies reinforce the large role of emissions from tropical regions in influencing the observed spatial concentration gradients.

— Concentrations of many of the fluorine-containing Kyoto Protocol gases (hydrofluorocarbons (HFCs), perfluorocarbons, SF6) have increased by large factors (between 4.3 and 1.3) between 1998 and 2005. Their total RF in 2005 was +0.017 [±0.002] W m–2 and is rapidly increasing by roughly 10% yr–1.

— The reactive gas, OH, is a key chemical species that influences the lifetimes and thus RF values of CH4, HFCs, HCFCs and ozone; it also plays an important role in the formation of sulphate, nitrate and some organic aerosol species. Estimates of the global average OH concentration have shown no detectable net change between 1979 and 2004.

Based on newer and better chemical transport models than were available for the TAR, the RF from increases in tropospheric ozone is estimated to be +0.35 [–0.1, +0.3] W m–2, with a medium level of scientific understanding. There are indications of significant upward trends at low latitudes.

The trend of greater and greater depletion of global stratospheric ozone observed during the 1980s and 1990s is no longer occurring; however, it is not yet clear whether these recent changes are indicative of ozone recovery. The RF is largely due to the destruction of stratospheric ozone by the Montreal Protocol gases and it is re-evaluated to be –0.05 [±0.10] W m–2, with a medium level of scientific understanding.

Based on chemical transport model studies, the RF from the increase in stratospheric water vapour due to oxidation of CH4 is estimated to be +0.07 [± 0.05] W m–2, with a low level of scientific understanding. Other potential human causes of water vapour increase that could contribute an RF are poorly understood.

The total direct aerosol RF as derived from models and observations is estimated to be –0.5 [±0.4] W m–2, with a medium-low level of scientific understanding. The RF due to the cloud albedo effect (also referred to as first indirect or Twomey effect), in the context of liquid water clouds, is estimated to be –0.7 [–1.1, +0.4] W m–2, with a low level of scientific understanding.

— Atmospheric models have improved and many now represent all aerosol components of significance. Improved in situ, satellite and surface-based measurements have enabled verification of global aerosol models. The best estimate and uncertainty range of the total direct aerosol RF are based on a combination of modelling studies and observations.

— The direct RF of the individual aerosol species is less certain than the total direct aerosol RF. The estimates are: sulphate, –0.4 [±0.2] W m–2; fossil fuel organic carbon, –0.05 [±0.05] W m–2; fossil fuel black carbon, +0.2 [±0.15] W m–2; biomass burning, +0.03 [±0.12] W m–2; nitrate, –0.1 [±0.1] W m–2; and mineral dust, –0.1 [±0.2] W m–2. For biomass burning, the estimate is strongly influenced by aerosol overlying clouds. For the first time best estimates are given for nitrate and mineral dust aerosols.

— Incorporation of more aerosol species and improved treatment of aerosol-cloud interactions allow a best estimate of the cloud albedo effect. However, the uncertainty remains large. Model studies including more aerosol species or constrained by satellite observations tend to yield a relatively weaker RF. Other aspects of aerosol-cloud interactions (e.g., cloud lifetime, semi-direct effect) are not considered to be an RF (see Chapter 7).

Land cover changes, largely due to net deforestation, have increased the surface albedo giving an RF of –0.2 [±0.2] W m–2, with a medium-low level of scientific understanding. Black carbon aerosol deposited on snow has reduced the surface albedo, producing an associated RF of +0.1 [±0.1] W m–2, with a low level of scientific understanding. Other surface property changes can affect climate through processes that cannot be quantified by RF; these have a very low level of scientific understanding.

Persistent linear contrails from aviation contribute an RF of +0.01 [–0.007, +0.02] W m–2, with a low level of scientific understanding; the best estimate is smaller than in the TAR. No best estimates are available for the net forcing from spreading contrails and their effects on cirrus cloudiness.

The direct RF due to increases in solar irradiance since 1750 is estimated to be +0.12 [–0.06, +0.18] W m–2, with a low level of scientific understanding. This RF is less than half of the TAR estimate.

— The smaller RF is due to a re-evaluation of the long-term change in solar irradiance, namely a smaller increase from the Maunder Minimum to the present. However, uncertainties in the RF remain large. The total solar irradiance, monitored from space for the last three decades, reveals a well-established cycle of 0.08% (cycle minimum to maximum) with no significant trend at cycle minima.

— Changes (order of a few percent) in globally averaged column ozone forced by the solar ultraviolet irradiance 11-year cycle are now better understood, but ozone profile changes are less certain. Empirical associations between solar-modulated cosmic ray ionization of the atmosphere and globally averaged low-level cloud cover remain ambiguous.

The global stratospheric aerosol concentrations in 2005 were at their lowest values since satellite measurements began in about 1980. This can be attributed to the absence of significant explosive volcanic eruptions since Mt. Pinatubo in 1991. Aerosols from such episodic volcanic events exert a transitory negative RF; however, there is limited knowledge of the RF associated with eruptions prior to Mt. Pinatubo.

The spatial patterns of RFs for non-LLGHGs (ozone, aerosol direct and cloud albedo effects, and land use changes) have considerable uncertainties, in contrast to the relatively high confidence in that of the LLGHGs. The Southern Hemisphere net positive RF very likely exceeds that in Northern Hemisphere because of smaller aerosol contributions in the Southern Hemisphere. The RF spatial pattern is not indicative of the pattern of climate response.

The total global mean surface forcing[5] is very likely negative. By reducing the shortwave radiative flux at the surface, increases in stratospheric and tropospheric aerosols are principally responsible for the negative surface forcing. This is in contrast to LLGHG increases, which are the principal contributors to the total positive anthropogenic RF.

  1. ^  The RF represents the stratospherically adjusted radiative flux change evaluated at the tropopause, as defined in the TAR. Positive RFs lead to a global mean surface warming and negative RFs to a global mean surface cooling. Radiative forcing, however, is not designed as an indicator of the detailed aspects of climate response. Unless otherwise mentioned, RF here refers to global mean RF. Radiative forcings are calculated in various ways depending on the agent: from changes in emissions and/or changes in concentrations, and from observations and other knowledge of climate change drivers. In this report, the RF value for each agent is reported as the difference in RF, unless otherwise mentioned, between the present day (approximately 2005) and the beginning of the industrial era (approximately 1750), and is given in units of W m–2.
  2. ^  90% confidence ranges are given in square brackets. Where the 90% confidence range is asymmetric about a best estimate, it is given in the form A [–X, +Y] where the lower limit of the range is (A – X) and the upper limit is (A + Y).
  3. ^  The use of ‘extremely likely’ is an example of the calibrated language used in this document, it represents a 95% confidence level or higher; ‘likely’ (66%) is another example (See Box TS.1).
  4. ^  Estimates of RF are accompanied by both an uncertainty range (value uncertainty) and a level of scientific understanding (structural uncertainty). The value uncertainties represent the 5 to 95% (90%) confidence range, and are based on available published studies; the level of scientific understanding is a subjective measure of structural uncertainty and represents how well understood the underlying processes are. Climate change agents with a high level of scientific understanding are expected to have an RF that falls within their respective uncertainty ranges (See Section 2.9.1 and Box TS.1 for more information).
  5. ^  Surface forcing is the instantaneous radiative flux change at the surface; it is a useful diagnostic tool for understanding changes in the heat and moisture surface budgets. However, unlike RF, it cannot be used for quantitative comparisons of the effects of different agents on the equilibrium global mean surface temperature change.