Temperature record of the past 1000 years

From Wikipedia, the free encyclopedia - View original article

Jump to: navigation, search
For information on the description of the Medieval Warm Period and Little Ice Age in various IPCC reports see MWP and LIA in IPCC reports
Reconstructions of Northern Hemisphere temperatures for the 2nd millennium according to various older articles (bluish lines), newer articles (reddish lines), and instrumental record (black line)

The temperature record of the 2nd millennium describes the reconstruction of temperatures since 1000 CE on the Northern Hemisphere, later extended back to 1 CE and also to cover the southern hemisphere. A reconstruction is needed because a reliable surface temperature record exists only since about 1850. Studying past climate is of interest for scientists in order to improve the understanding of current climate variability and, relatedly, providing a better basis for future climate projections. In particular, if the nature and magnitude of natural climate variability can be established, scientists will be able to better detect and attribute anthropogenic global warming. Note, however, that although temperature reconstructions from proxy data help us understand the character of natural climate variability, attribution of recent climate change relies on a broad range of methodologies of which the proxy reconstructions are only a small part.[1]

According to all major temperature reconstructions published in peer-reviewed journals (see graph), the increase in temperature in the 20th century and the temperature in the late 20th century is the highest in the record. Attention has tended to focus on the early work of Mann, Bradley, and Hughes (1998, 1999) whose "hockey stick" graph was featured in the 2001 United Nations Intergovernmental Panel on Climate Change report. The methodology and data sets used in creating the qversion of the hockey stick graph were disputed by Stephen McIntyre and Ross McKitrick, but the graph is overall acknowledged by the scientific community.

Instrumental Temperature record of the last 150 years.


General techniques and accuracy

By far the best observed period is from 1850 to the present day, with coverage improving over time. Over this period the recent instrumental record, mainly based on direct thermometer readings, has approximately global coverage. It shows a general warming in global temperatures.

Before this time various proxies must be used. These proxies are less accurate than direct thermometer measurements, have lower temporal resolution, and have less spatial coverage. Their only advantage is that they enable a longer record to be reconstructed. Since the direct temperature record is more accurate than the proxies (indeed, it is needed to calibrate them) it is used when available: i.e., from 1850 onwards.

Quantitative methods using proxy data

As there are few instrumental records before 1850, temperatures before then must be reconstructed based on proxy methods. One such method, based on principles of dendroclimatology, uses the width and other characteristics of tree rings to infer temperature. The isotopic composition of snow, corals, and stalactites can also be used to infer temperature. Other techniques which have been used include examining records of the time of crop harvests, the treeline in various locations, and other historical records to make inferences about the temperature. These proxy reconstructions are indirect inferences of temperature and thus tend to have greater uncertainty than instrumental data.

In general, the recent history of the proxy records is calibrated against local temperature records to estimate the relationship between temperature and the proxy. The longer history of the proxy is then used to reconstruct temperature from earlier periods. Proxy records must be averaged in some fashion if a global or hemispheric record is desired. Considerable care must be taken in the averaging process; for example, if a certain region has a large number of tree ring records, a simple average of all the data would strongly over-weight that region. Hence data-reduction techniques such as principal components analysis are used to combine some of these regional records before they are globally combined. An important distinction is between so-called 'multi-proxy' reconstructions, which attempt to obtain a global temperature reconstructions by using multiple proxy records distributed over the globe and more regional reconstructions. Usually, the various proxy records are combined arithmetically, in some weighted average. More recently, Osborn and Briffa used a simpler technique, counting the proportion of records that are positive, negative or neutral in any time period.[2] This produces a result in general agreement with the conventional multi-proxy studies.

The Mann, Bradley & Hughes (1998) reconstruction has been described as "the "first systematic, statistically based synthesis of multiple climate proxies", and it introduced explicit statistical error bars.[3] Mann, Bradley & Hughes (1999) extended the proxy temperature record back 1000 years, and the term "Hockey Stick" graph was coined by Jerry Mahlman, director of the Geophysical Fluid Dynamics Laboratory, when he saw this graph.

Several reconstructions suggest there was minimal variability in temperatures prior to the 20th century.[4] More recently, Mann & Jones (2003) have extended their reconstructions to cover the 1st and 2nd millennia. The work was reproduced by Wahl & Ammann (2007).

The two Mann, Bradley & Hughes studies, together with Jones et al. 1998, Briffa 2000 and others[5] formed a major part of the IPCC Third Assessment Report's conclusion that "the rate and magnitude of global or hemispheric surface 20th century warming is likely to have been the largest of the millennium, with the 1990s and 1998 likely to have been the warmest decade and year".[6] The 2007 IPCC Fourth Assessment Report cited 14 reconstructions, 10 of which covered 1,000 years or longer, to support its conclusion that "Average Northern Hemisphere temperatures during the second half of the 20th century were very likely higher than during any other 50-year period in the last 500 years and likely the highest in at least the past 1,300 years".[7]

Qualitative reconstruction using historical records

It is also possible to use historical data such as times of grape harvests, sea-ice-free periods in harbours and diary entries of frost or heatwaves to produce indications of when it was warm or cold in particular regions. These records are harder to calibrate, are often only available sparsely through time, may be available only from developed regions, and are unlikely to come with good error estimates. These historical observations of the same time period show periods of both warming and cooling.

Astrophysicist Sallie Baliunas notes that these temperature variations correlate with solar variation[8] and asserts that the number of observed sunspots give us a rough measure of how bright the sun is. Balunias and others have suggested that periods of decreased solar radiation are partially responsible for historically recorded periods of cooling such as the Maunder Minimum and the Little Ice Age. The same argument would imply that periods of increased solar radiation contributed to the Medieval Warm Period, when Greenland's icy coastal areas thawed enough to permit farming and colonisation.


The apparent differences between the quantitative and qualitative approaches are not fully reconciled. The reconstructions mentioned above rely on various assumptions to generate their results. If these assumptions do not hold, the reconstructions would be unreliable. For quantitative reconstructions, the most fundamental assumptions are that proxy records vary with temperature and that non-temperature factors do not confound the results. In the historical records temperature fluctuations may be regional rather than hemispheric in scale.

In a letter to Nature Bradley, Hughes & Mann (2006) pointed at the original title of their 1998 article: Northern Hemisphere temperatures during the past millennium: inferences, uncertainties, and limitations[9] and pointed out more widespread high-resolution data are needed before more confident conclusions can be reached and that the uncertainties were the point of the article.

The hockey stick controversy

There is an ongoing debate about the details of the temperature record and the means of its reconstruction, centered on the Mann, Bradley & Hughes (1998) "hockey stick" graph. Stephen McIntyre and Ross McKitrick[10] claimed various errors in the methodology of Mann, Bradley & Hughes (1998) and that the method of Mann, Bradley, and Hughes "when tested on persistent red noise, nearly always produces a hockey stick shaped first principal component". In turn, Michael E. Mann (supported by Tim Osborn, Keith Briffa and Phil Jones of the Climatic Research Unit) have disputed the claims made by McIntyre and McKitrick.</ref>[11] The IPCC Fourth Assessment Report says that McIntyre and McKitrick "may have some theoretical foundation, but Wahl & Ammann (2007) also show that the impact on the amplitude of the final reconstruction is very small (~0.05°C)."[12]

See also



External links