Global and European temperature (CSI 012/CLIM 001) - Assessment published Mar 2009

Indicator definition

This indicator shows absolute changes and rates of change in average near-surface temperature for the globe and for a region covering Europe. Near-surface air temperature gives one of the clearest and most consistent signals of global and regional climate change, especially in recent decades. It has been measured for many decades or even centuries at some locations and a dense network of stations across the globe, and especially in Europe, provide regular monitoring of temperature, using standardised measurements, quality control and homogeneity procedures.

This indicator provides guidance for the following policy-relevant questions:

•  Will the global average temperature increase stay within the UNFCCC policy target of 2.0°C above pre-industrial levels?
•  Will the rate of global average temperature increase stay below the indicative proposed target of 0.2°C increase per decade?

Global average annual temperature deviations, ‘anomalies’, are discussed relative to a ‘pre-industrial’ period between 1850 and 1899 (beginning of instrumental temperature records).  During this time, anthropogenic greenhouse gases from the industrial revolution (between 1750 and 1850) are considered to have a relatively small influence on climate compared to natural influences. However it should be noted that owing to earlier changes in the climate due to internal and forced natural variability there was not one single pre-industrial climate and it is not clear that there is a rigorous scientific definition of the term ‘pre-industrial climate’.

Temperature changes also influence other aspects of the climate system which can impact on human activities, including sea level, intensity and frequency of floods and droughts, biota and food productivity and infectious diseases. In addition to the global average target, seasonal variations and spatial distributions of temperature change are important, for example to understand the risks that current climate poses to human and natural systems and to assess how these may be impacted by future climate change.

Units

Degrees Celsius and degrees Celsius per decade.

Temperature anomalies are defined as changes from pre-industrial averages

Policy context and targets

Context description

This indicator provides guidance for the following policy-relevant questions:

• Can the global average temperature increase stay within the UNFCCC policy target of 2.0°C above pre-industrial levels?

• Can the rate of global average temperature increase stay within the indicative proposed target of 0.2°C increase per decade?

The absolute change and rate of change in global average temperature are both important indications of the magnitude and variability of global climate change. They relate directly to global temperature policy targets. Temperature changes also influence other aspects of the climate system which can impact on human activities, including sea level, intensity and frequency of floods and droughts, biota and food productivity and infectious diseases.

In addition to the global average target, seasonal variations and spatial distributions of temperature change are important, for example to understand the risks that current climate poses to human and natural systems and to assess how these may be impacted by future climate change.

Targets

To avoid serious climate change impacts, the European Council proposed in its sixth environmental action programme (6EAP, 2002), reaffirmed by the Environment Council and the European Council of 22-23 March 2005 (Presidency Conclusions, section IV (46)), that the global average temperature increase should be limited to not more than 2 degrees C above pre-industrial levels (about 1.3 ⁰C above current global mean temperature). Furthermore the UNFCCC 15th conference of the parties (COP15) recognised, in the Copenhagen Accord (UNFCCC, Dec 2009), the scientific evidence for the need to keep global average temperature increase below 2 ⁰C above pre-industrial levels.  In addition, some studies have proposed a 'sustainable' target of limiting the rate of anthropogenic warming to 0.1 to 0.2 ⁰ C per decade.

The targets for both absolute temperature change (i.e. 2 ⁰C) and rate of change (i.e. 0.1 to 0.2 ⁰C per decade),(WBGU, 2003), were initially derived from the migration rates of selected plant species and the occurrence of past natural temperature changes. The EU target for global temperature increase (i.e. 2 ⁰C) has recently been confirmed as a suitable target from both a scientific and a political perspective.

The target for absolute temperature change (i.e. 2 ⁰C) was initially derived from the migration rates of selected plant species and past natural temperature changes. Further studies (IPCC, 2007; IARU, 2009; UNEP, 2009; Copenhagen Diagnosis, 2009) have indicated that climate change might still result in various impacts in vulnerable regions across the world, in particular in developing countries (including least developed countries, small developing island states and Africa), even below the 2 ⁰C target level. The UNFCCC Copenhagen Accord (Dec 2009) includes a review in 2015 of the scientific evidence (e.g. the IPCC fifth assessment report, due in 2014) for a possible lower global temperature target.

Mainstreaming climate change adaptation in EU policies is one of the pillars of the European Commission’s 2009 White Paper on adaptation and the EU Adaptation strategy. In the Europe 2020 strategy for smart, sustainable and inclusive growth , the following is stated on combating climate change: 'We must also strengthen our economies' resilience to climate risks, and our capacity for disaster prevention and response'.

Related policy documents

• EU Adaptation Strategy
  The European Commission today presents a package to advance action on adaptation to climate change: firstly, the EU strategy on adaptation to climate change sets out a framework and mechanisms for taking the EU's preparedness for current and future climate impacts to a new level; in a related measure, the Commission adopted a Green Paper on insurance in the context of natural and man-made disasters. This public consultation launches a wide debate on the adequacy and availability of existing insurance options.
• European Commission related policy documents
  European Commission related policy documents
• UNFCCC related policy documents
  UNFCCC related policy documents

Methodology

Methodology for indicator calculation

Various data sets on trends in global and European temperature have been used for this indicator:

• Global average seasonal and annual temperature: 3 datasets are used. The HadCRUT  is collaborative products of the Met Office Hadley Centre (sea surface temperature) and the Climatic Research Unit (land temperature) at the University of East Anglia.The global mean annual temperature deviations are in the original source in relation to the base period 1961-1990. The annual deviations shown in the chart have been adjusted to be relative to the period 1880 to 1899 in order to better monitor the EU objective not to exceed 2oC above pre-industrial values. Data source: EEA, based on HadCRUT and CRUTEM datasets.
• The GISS surface temperature is a product of the Goddard Institute for Space Studies under NASA. The original source anomalies are calculated in the relation to the 1951 to 1980 base period. Annual deviations shown on the chart are adjusted to the 1880 to 1899 period to better monitor the EU objective, of a maximum 2 oC global temperature increase above the pre-industrial values. The indicator has been calculated as a combination of land and sea temperature.
• The GHCN surface temperature is product of the National Climate Data Centre (NCDC) from National Oceanic and Atmospheric Administratioon (NOAA). Datasets are available as gridded product from 1880 onwards in monthly time step.  Dataset was created from station data using the Anomaly Method, a method that uses station averages during a specified base period from which the monthly/seasonal/annual departures can be calculated. Anomalies were calculated on a monthly basis for all adjusted stations having at least 20 years of data in the 1961–1990 base period. Station anomalies were then averaged within each 5° by 5° grid box to obtain the gridded anomalies. For those grid boxes without adjusted data, anomalies were calculated from the raw station data using the same technique.

• European average annual and monthly temperature: The sources of the data is the latest version of the gridded CRUTEM (land only). Europe is defined as the area between 35° to 70° Northern latitude, -25° to 30° Eastern longitude, plus Turkey (35° to 40° North, 30° to 45° East). The European anomalies are in the original source in relation to the base period 1961-1990. The annual deviations shown in the chart have been adjusted to be relative to the period 1850-1899. Data source: EEA, based on CRUTEM3 dataset

• Annual, winter (December, January, February) and summer (June, July, August) mean temperature deviations in Europe, 1860 to 2008 (^oC). The lines in the chart refer to 10-year moving average European land. Data source: EEA, based on CRUTEM datasets.

• Observed changes in warm spells and frost days indices in the period 1976 to 2009; (Changes in duration of warm spels  in summer, days per decade, and frequency of frost days in winter, days per decade). Warm spells are defined as a period of at least six consecutive days where the mean daily temperature exceeds the baseline temperature (average daily temperature the 1961 to 1990 period) by 5 ^oC. Frost days are defined as a day with an average temperature below 0 oC. Positive values indicate increase in frequency and negative values indicate a decreased frequency. Data source: http://eca.knmi.nl/ensembles

• Modelled number of heat index over Europe during summer. Heat index is defined as days with an apparent temperature above 40.7 ^oC. Apparent temperature is determined as a human-perceived equivalent temperature caused by the combined effects of air  temperature and relative humidity. (left chart: 1961 to 1990 average; middle: 2021 to 2050 average, right: 2071 to 2100 average). Data source: van der Linden P., and J.F.B. Mitchell (eds.) 2009: ENSEMBLES: Climate Change and its Impacts: Summary of research and results from the ENSEMBLES project. Met Office Hadley Centre, FitzRoy Road, Exeter EX1 3PB, UK. 160pp.

Global and European average time series for monthly temperature

In the original source the long-term annual and monthly mean HadCRUT3 global temperatures were calculated from 4349 stations for the entire period of the record. There is an irregular distribution in the time and space of available stations (i.e .denser coverage over the more populated parts of the world and increased coverage after 1950).  Maps/tables giving the density of coverage through time are given for land regions by Jones and Moberg (2003).  The gridding method was climate anomaly method (CAM), which means the station temperature data have been converted to the anomalies according to the WMO standards (baseline period 1961-1990 and at least 15 years of station data in the period) and grid-box values have been produced by simple averaging of the individual station anomaly values within each grid box.

GISS surface temperatures were calculated using around 7200 stations from Global Historical Climatology Network, United States Historical Climatology Network (USHCN) data, and SCAR (Scientific Committee on Antarctic Research) data from Antarctic stations. Additionally satellite SST has been included for the period after 1980. Temperatures were transformed into anomalies using station normalisation based on the 1951 to 1980 baseline period. Gridding has been done with reference station method using 1200 km influence circle (Hansen and Lebedeff 1987).

Surface temperature mean anomalies from Global Historical Climatology Network-Monthly (GHCN-M) has been produced at the NCDC from 2,592 gridded data points based on a 5° by 5° grids for the entire globe. The gridded anomalies were produced from GHCN-M bias corrected data. Gridded data for every month from January 1880 to the most recent month is available. The data are temperature anomalies in degrees Celsius (Jones, 2003).

Daily climate information

Although Europe  has a long history in collecting climate information, datasets containing daily climate information across the continent are scarce. Furthermore, accurate climate analysis requires long term time series without artificial breaks. The objective of the ECA project was to compile such a data set, consisting of homogeneous, long-term daily climate information. To ensure a uniform analysis method and data handling, data were centrally collected from about 200 meteorological stations in most countries of Europe and parts of middle East. Furthermore the data were, processed and analysed at one institute (i.e. KNMI) (Klok et.al. , 2008).

In order to ensure the quality of the ECA climate data set:

• Statistical homogeneity tests have been applied to detect breaks in the time series
• the meta-information accompanied with the data, has intensively been analysed, e.g. to check whether observed trends were not triggered by, for example, movements of stations.
• the final data set has been compared with other data sets, like the aforementioned data set of CRU.
• Findings of the different exercises have been discussed during workshops with representatives of countries.

Methodology for gap filling

Global and European average time series for monthly temperature

Grid values of HadCRUT, GISTEMP and GHCN data sets have been gridded using different interpolation techniques. Each grid-box value for the HadCRUT dataset is the mean of all available station anomaly values, except that station outliers in excess of five standard deviations are omitted (Morice et al, 2012;Brohan et al., 2005). GISTEMP temperature anomaly data are gridded into 8000 grid cells using reference station interpolation method with 1200 km influence circle (Hansen et al. 2006). GHCN monthly data consists of 2,592 gridded data points produced on a 5° by 5° basis for the entire globe (Jones, 2003).

Methodology references

• European Climate Assessment & Dataset project ECA&D ECA&D is initiated by the European Climate Support Network ECSN , and is supported by the Network of European Meteorological Services EUMETNET .
• Climate in Europe. Assessment of observed daily temperature and precipitation extremes Klein Tank, A.J. Wijngaard, and A. van Engelen (2002) European Climate Assessment, KNMI, the Bilt, the Netherlands
• KNMI Climate Explorer This web site gives the opportunity to explore monthly mean climate time series and the relationships between them.
• Updated and extended European dataset of daily climate observations Klok, E.J. and A.M.G. Klein Tank: Updated and extended European dataset of daily climate observations. Int. J. Climatol (2008)
• Uncertainty estimates in regional and global observed temperature changes: a new dataset from 1850 P. Brohan, J. J. Kennedy, I. Harris, S. F. B. Tett & P. D. Jones. Uncertainty estimates in regional and global observed temperature changes: a new dataset from 1850. J. Geophys. Res., 111, D12106 (2005)
• Global temperature change Hansen, J., Mki. Sato, R. Ruedy, K. Lo, D.W. Lea, and M. Medina-Elizade: Global temperature change. Proc. Natl. Acad. Sci., 103, 14288-14293, doi:10.1073/pnas.0606291103 (2006)
• WMO statement on the status of the global climate in 2011 WMO, 2010 WMO statement on the status of the global climate in 2009; World Meteorological Organization, Geneva, Switzerland.
• WMO statement on the status of the global climate in 2012 WMO statement on the status of the global climate in 2012. WMO, 2012 WMO statement on the status of the global climate in 2011; World Meteorological Organization, Geneva, Switzerland.
• WMO (2012) WMO (2012) Greenhouse Gas Bulletin: The State of Greenhouse Gases in the Atmosphere Based on Global Observations through 2011.
• Climate: observations, projections and impacts Understanding the potential impacts of climate change is essential for informing both adaptation strategies and actions to avoid dangerous levels of climate change.

Uncertainties

Methodology uncertainty

The observed increase in average air temperature, particularly during recent decades, is one of the clearest signals of global climate change.

Temperature has been measured over the centuries. There is a range of different methodologies which give similar results suggesting that uncertainty is relatively low. Three data sets have been presented here for the global temperature indicator. Global temperatures from HadCRUT3, GISTEMP, and GHCN have been homogenized to minimise the effects of changing measurement methodologies and location.

Data sets uncertainty

Each observation station follows international standards for taking observations set out by WMO. Each National Meteorological Service provides reports on how its data are collected and processed to ensure consistency. This includes recording information about the local environment around the observation station and any changes to that environment. This is important for ensuring the required data accuracy and performing homogeneity tests and adjustments. There are additional uncertainties because temperatures over large areas of the Earth are not observed as a matter of routine. These elements are taken into account by factoring the uncertainty into global average temperature calculations, thereby producing a temperature range rather than one uniquely definite figure (WMO, 2010). The uncertainty of temperature data has decreased over recent decades due to wider use of agreed methodologies and denser monitoring networks. Uncertainty of the temperature data comes from sampling error, temperature bias effect and from the effect of the limited observation coverage. Annual values of global and European temperature are approximately accurate to +/- 0.05 degrees C (two standard errors) for the period since 1951. They are about four times as uncertain during the 1850s, with the accuracy improving gradually between 1860 and 1950 except for temporary deteriorations during data-sparse, wartime intervals. Estimating accuracy is difficult as the individual grid-boxes are not independent of each other and the accuracy of each grid-box time series varies through time (although the variance adjustment has reduced this influence to a large extent). The issue is discussed extensively by Jones et al. (2003) Brohan et al. (2005), and Hansen et al. (2006).

Rationale uncertainty

According to the IPCC 4^th Assessment Report, there is very high confidence that the net effect of human activities since 1750 has been one of warming. Most of the observed increase in global average temperatures since the mid-20th century is very likely due to the observed increase in anthropogenic GHG concentrations.