Climate Change 2001: Working Group I: The Scientific Basis

	Other reports in this collection

Executive Summary

Overview
The best estimate of global surface temperature change is a 0.6°C increase since the late 19th century with a 95% confidence interval of 0.4 to 0.8°C. The increase in temperature of 0.15°C compared to that assessed in the IPCC WGI Second Assessment Report (IPCC, 1996) (hereafter SAR) is partly due to the additional data for the last five years, together with improved methods of analysis and the fact that the SAR decided not to update the value in the First Assessment Report, despite slight additional warming. It is likely that there have been real differences between the rate of warming in the troposphere and the surface over the last twenty years, which are not fully understood. New palaeoclimate analyses for the last 1,000 years over the Northern Hemisphere indicate that the magnitude of 20th century warming is likely to have been the largest of any century during this period. In addition, the 1990s are likely to have been the warmest decade of the millennium. New analyses indicate that the global ocean has warmed significantly since the late 1940s: more than half of the increase in heat content has occurred in the upper 300 m, mainly since the late 1950s. The warming is superimposed on strong global decadal variability. Night minimum temperatures are continuing to increase, lengthening the freeze-free season in many mid- and high latitude regions. There has been a reduction in the frequency of extreme low temperatures, without an equivalent increase in the frequency of extreme high temperatures. Over the last twenty-five years, it is likely that atmospheric water vapour has increased over the Northern Hemisphere in many regions. There has been quite a widespread reduction in daily and other sub-monthly time-scales of temperature variability during the 20th century. New evidence shows a decline in Arctic sea-ice extent, particularly in spring and summer. Consistent with this finding are analyses showing a near 40% decrease in the average thickness of summer Arctic sea ice over approximately the last thirty years, though uncertainties are difficult to estimate and the influence of multi-decadal variability cannot yet be assessed. Widespread increases are likely to have occurred in the proportion of total precipitation derived from heavy and extreme precipitation events over land in the mid- and high latitudes of the Northern Hemisphere.

Changes in Temperature and Related Variables

Changes in near-surface temperature from the instrumental record

• Average global surface temperature has increased by approximately 0.6°C since the late 19th century, with 95% confidence limits of close to 0.4 and 0.8°C. Most of this increase has occurred in two periods, from about 1910 to 1945 and since 1976, and the largest recent warming is in the winter extra-tropical Northern Hemisphere. The warming rate since 1976, 0.17°C/decade, has been slightly larger than the rate of warming during the 1910 to 1945 period (0.14°C/decade), although the total increase in temperature is larger for the 1910 to 1945 period. The most recent warming period also has a faster rate of warming over land compared with the oceans. The high global temperature associated with the 1997/98 El Niño event stands out in both surface and tropospheric temperatures as an extreme event, even after consideration of the recent rate of warming.
• Confidence in the magnitude of global warming since the late 19th century has increased since the SAR due to new analyses, including model simulations of land-surface temperature changes and new studies of the effect of urbanisation on global land temperature trends. There is a high level of consistency between changes in sea surface temperatures (SSTs) and near-surface land air temperatures across the land-ocean boundary over the 20th century, despite independent observing systems and independent bias correction factors for SSTs before 1942. The assessed warming is considerably larger than the total contributions of the plausible sources of error.
• Twentieth century temperature trends show a broad pattern of tropical warming, while extra-tropical trends have been more variable. Warming from 1910 to 1945 was initially concentrated in the North Atlantic and nearby regions. The Northern Hemisphere shows cooling during the period 1946 to 1975 while the Southern Hemisphere shows warming. The recent 1976 to 2000 warming was largely globally synchronous, but emphasised in the Northern Hemisphere continents during winter and spring, with year-round cooling in parts of the Southern Hemisphere oceans and Antarctica. North Atlantic cooling between about 1960 and 1985 has recently reversed. Overall, warming over the Southern Hemisphere has been more uniform during the instrumental record than that over the Northern Hemisphere.
  
• The patterns of global temperature change since the 1970s are related in part to the positive westerly phase of the North Atlantic/Arctic Oscillation and possibly to decadal to multi-decadal variability in the Pacific.
• A multi-decadal fluctuation of SST in the North Atlantic has been in a rising phase since about the mid-1980s. Warming in many regions of this ocean has accelerated over the last five years and is likely to have contributed to quite rapid parallel increases of near-surface air temperature in much of Europe.
• New analysis shows that the global ocean heat content has increased since the late 1950s. This increase is superimposed on substantial global decadal variability. More than half the heating is contained in the uppermost 300 m where it is equivalent to an average temperature increase of 0.037°C/decade.
• Analyses of mean daily maximum and minimum land surface air temperatures continue to support a reduction in the diurnal temperature range in many parts of the world, with, globally, minimum temperatures increasing at nearly twice the rate of maximum temperatures between about 1950 and 1993. The rate of temperature increase during this time has been 0.1°C and 0.2°C/decade for the maximum and minimum, respectively. This is more than twice the rate of temperature increase over the oceans during this time.

Changes in temperature-related variables

• Alpine and continental glaciers have extensively retreated in response to 20th century warming. Glaciers in a few maritime regions are advancing, mainly due to increases in precipitation related to atmospheric circulation changes, e.g., Norway, New Zealand.
• The duration of Northern Hemisphere lake-ice and river-ice cover over the past century, or more, shows widespread decreases averaging to about two fewer weeks of ice cover.
• There is a highly significant interannual (+0.6) and multi-decadal correlation between increases in the Northern Hemisphere spring land temperature and a reduction in the Northern Hemisphere spring snow cover since data have been available (1966). Snow cover extent has decreased by about 10% since 1966.
• A 10 to 15% reduction in sea-ice extent in the Arctic spring and summer since the 1950s is consistent with an increase in spring, and to a lesser extent, summer temperatures in the high latitudes. There is little indication of reduced Arctic sea-ice extent during winter when temperatures have increased in the surrounding region.
• New data from submarines indicate that there has been about a 40% decline in Arctic sea-ice thickness in summer or early autumn between the period 1958 to 1976 and the mid-1990s, an average of near 4 cm per year. Other independent observations show a much slower decrease in winter sea-ice thickness of about 1 cm per year. The influence of substantial interannual and inter-decadal variability on these changes cannot be assessed because of restricted sampling.
• By contrast, there is no readily apparent relationship between decadal changes in Antarctic temperatures and sea-ice extent since 1973. Satellite data indicate that after a possible initial decrease in the mid-1970s, Antarctic sea-ice extent has stayed almost stable or even increased since 1978.

Changes in temperature above the surface layer

• Analysis of global temperature trends since 1958 in the low to mid-troposphere from balloons shows a warming of about +0.1°C/decade, which is similar to the average rate of warming at the surface. Since the early 1960s no significant trends have been detected for the global mean temperature in the uppermost troposphere.
• Satellites have only been available since 1979. Between 1979 and 2000, based on satellites and balloons, the lower-tropospheric trend has been +0.04 ± 0.11°C/decade and 0.03 ± 0.10°C/decade, respectively. By contrast, surface temperature trends for 1979 to 2000 were greater, at 0.16 ± 0.06°C/decade. The trend in the difference of the surface and lower-tropospheric series of 0.13 ± 0.06°C/decade is clearly statistically significant. This is in contrast to near zero surface temperature trends over 1958 to 1978 when the global lower-tropospheric temperature warmed by 0.03°C/decade relative to the surface.
• It is very likely that these significant differences in trends between the surface and lower troposphere are real and not solely an artefact of measurement bias, though differences in spatial and temporal sampling are likely to contribute. The differences are particularly apparent in many parts of the tropics and sub-tropics where the surface has warmed faster than the lower troposphere. In some other regions, e.g., North America, Europe and Australia, lower-tropospheric and surface trends are very similar.
• Throughout the stratosphere, negative temperature trends have been observed since 1979, ranging from a decrease of 0.5 or 0.6°C/decade in the lower stratosphere to 2.5°C/decade in the upper stratosphere.