|
What is the record of surface temperature fluctuations in the Northern Hemisphere over the last 1,000 years? How does the observed climate warming of the 20th century compare with this 1,000-year record? What do tree-rings tells us about the climate of the last 1,000 years relative to that of the 20th century? How much of the observed warming in the 20th century can be attributed to natural climate variability? How much of that warming is likely to be attributable to human activities? What has the biospheric response been to these changes, especially in the 20th century? What can be said about the rate of temperature change over the last 100 years or more? Was 1998 the warmest year in the last millennium? To what extent can the observed warming of 1998 be attributable to El Nino?
Without knowledge of natural climate variability at century and longer timescales, it is difficult to determine the significance of 20th century warming evident in the relatively short instrumental record of global surface temperature. To obtain a longer-term perspective on observed climate variability and change, one must resort to indirect measurements of climate variations derived from natural archives or "proxy" climate indicators such as tree rings, corals, and ice cores, supplemented with the few available long instrumental and historical climate records. Using such proxy data networks, research published in 1998 led to estimates of annual, global surface temperature patterns dating back to AD 1400. Averaging these reconstructions allowed the calculation of estimates of Northern Hemisphere mean temperatures back through AD 1400, as well as estimates of the uncertainties in these estimates. The warmth of the 1990s appeared to be unprecedented in this reconstruction, with three years during this decade (1990, 1995, and 1997) that were likely to be warmer than any other year since AD 1400. Reconstructions further back in time were not then possible with the available data networks.
Two significant events have occurred since that research was done that allow those original conclusions to be expanded upon. Based upon careful consideration of the sparse proxy data available for the years AD 1000 - AD 1400, estimates of yearly Northern Hemisphere mean temperatures have been made dating back to AD 1000, albeit with considerably larger uncertainties. Not withstanding these uncertainties, and taking into account the slightly warmer temperature estimates for the early part of the millennium, it would be difficult to argue that the 1990s were as anomalous when viewed in the context of the temperature history of the entire millennium, were it not for the record warmth of 1998. The year 1998 was observed to have been significantly warmer (by about 0.2 degrees C) than any other year in the instrumental record. In the context of the last 1,000 years, one can say with a high degree of confidence that 1998 was warmest year for the Northern Hemisphere.
Prior to AD 1400, surface temperature estimates depend upon certain key climate proxy data series, especially those in the higher elevations of the western U.S. which show a marked temperature sensitivity. These data also show non-climatic influences during the 19th and 20th century that may be related to CO2 increases or perhaps other factors. As such, these non-climatic influences must be subject to further research and analysis, before the data can be used in long-term climate reconstruction. Prior to about AD 1000, the sparseness of the available data preclude a meaningful estimate of hemispheric mean temperatures.
Based on the millennial hemispheric temperature estimates, conditions during the earlier centuries of the past millennium appear somewhat higher than those of the 15th-19th centuries. However, the data do not support the notion of the existence of a hemisphere-wide "Medieval Warm Period" relative to the late 20th century warming. Rather, the evidence suggests that the warmest decades of the Medieval era were comparable to early and mid 20th century temperatures, but not those of the late 20th century. Some evidence suggests that certain regions (e.g., the North Atlantic and Greenland) may have exhibited somewhat greater warmth, but at the hemispheric scale, the evidence does not support the notion of sustained periods of warmth during the past 1,000 years, comparable to the warmth of the late 20th century. Due to the sparseness of the proxy climate data for the years prior to AD 1400, and the difficulty in resolving temperature variations this far back in time, the uncertainties in hemispheric temperature estimates become considerably larger in the earlier centuries of the past millennium. Even with these expanded uncertainties, however, the 1990s, and 1998 in particular, appear to have exhibited hemispheric warmth that is unprecedented at least over the last 1,000 years.
In searching for a likely cause or causes that explain variations in the Earth's surface temperature changes over the last six centuries, a suite of plausible, candidate, climate-forcing influences such as changes in the brightness of the Sun, changes in the frequency and magnitude of volcanic eruptions, and human-caused increases in greenhouse gas concentrations, have been evaluated. This analysis suggested that only an increase in the concentration of greenhouse gases could explain the anomalous warmth of the late 20th century. With longer-term, millennial estimates of surface temperature change, other factors may yet prove to play a role. On timescales of tens to hundreds-of-thousands of years, the "astronomical theory" of climate change holds that changes in the geometry of the Earth's orbit relative to the Sun, bring about subtle changes in the distribution of solar radiation at the Earth's surface that may drive slow, but significant, long-term changes in climate. A number of recent climate modeling experiments suggest that such astronomical factors should have led to a slow cooling of the climate since about 6,000 years before present, at a rate of cooling of between 0.01 and 0.04 degrees C/century. Such theoretical considerations are in remarkable agreement with the observed cooling trend (about 0.02 degrees C/century) from AD 1000 through the mid 19th century, as observed in this temperature reconstruction. However, this long-term cooling trend undergoes a dramatic reversal over the course of the 20th century. Thus, the 20th century warming trend appears to be that much more anomalous when viewed in the context of the natural, long-term climate variability of the last millennium, and is therefore again, unlikely to be due to natural factors alone.
The measured record from thermometers, rain gauges and barometers, does not provide an adequate sample of the ways in which climate could vary under recent or present conditions, even if there were no human influence on climate. Planning that ignores this will be inadequate, whether its focus is resource use (e.g. energy or water) or mitigating the consequences of natural disturbances such as drought, floods and wildfires. This is because the instrumental record is often too short to represent the different ways climate can behave, and because this record was hardly started by the time human action had made measurable changes in the composition of our atmosphere. One therefore, cannot rely solely on the twentieth century instrumental record to assess the character of climate change.
Research using tree rings to derive estimates of climate variability provides many interesting insights. Examples include:
Work on using tree rings to detect the biosphere's response to climate variability and climate change is at a much earlier stage than their use as natural climate recorders, but some intriguing fragments of evidence have already emerged. In some regions as widely separated as the southern Rockies in the U.S., and Tasmania, high elevation trees have shown growth spurts in the last two to three decades that are unprecedented in at least the last thousand years. In the case of the southern Rockies, this seems to have been caused by an unusual combination of climatic conditions, rather than by any direct fertilization by increased carbon dioxide concentrations in the air. Trees growing at the highest elevations in the mountains in and around the Great Basin have been growing at an accelerated rate since the middle of the nineteenth century, and the link that did exist between their growth rate and local climate broke down at that time. There is no convincing climatic explanation for this, and alternatives, such as direct carbon dioxide fertilization have been proposed, but none has gained wide acceptance. An integrated program of research combining tree-ring records and vegetation remote sensing is needed to record and better assess the biosphere's response to climate change and variability.
The NASA/GISS (Goddard Institute for Space Studies) has developed a data set that provides estimates of global surface temperature change for 1880-1998, the period with significant global coverage of instrumental data. Urban influence on the record is substantial in certain locations, but is found to have only a small effect on the global estimates. The record shows global warming this century that is unambiguous and unusual. The five-year mean global temperature has increased about 0.7 degrees C since the late 1800s. The global surface temperature in 1998 was the warmest in the period of instrumental data. The rate of temperature change is higher in the past 25 years than at any previous time in the period of instrumental data. The warmth of 1998 is too large and pervasive to be fully accounted for by the recent El Nino. This analysis suggests that the global temperature may have moved to a higher level, analogous to the significant increase that occurred in the late 1970s.
The record of surface temperature change can be compared with satellite measurements of tropospheric temperature for the period since 1979. The satellite record is sometimes interpreted as being contradictory to the surface measurements. The GISS analysis indicates that the differences are actually small and within estimated measurement errors, and that the results are consistent with a long term warming trend at the surface and in the troposphere. This analysis further indicates that there has been a slight cooling in the United States in the past 50 years, particularly in the eastern half of the country. The latter observation raises questions regarding the likelihood of the observed temperature change in the U.S. catching up with the rest of the world, and the observational data on global climate forcings and the ocean necessary to answer the questions.