Climate Indicator Bulletins (CIBs) are user-driven climate information products which provide simple, effective and timely knowledge abstractions from the large amount of observation and reanalyses data available in EURO4M. The bulletins focus on user groups in sectors such as disaster prevention, health, energy, water resources, ecosystems, forestry, agriculture, transport, tourism and biodiversity at European, national and local levels.
This bulletin presents an overview of the temperature evolution in Europe based on a large number of measurements. Recent observations are placed in a historical context and special attention is given to long-term trends and record high/low values.
Last update: 25 January 2013.
Temperature evolution since 1950
Annual mean temperature
Surface air temperatures are measured at an extensive network of stations in Europe. These data are collected and agregated into European temperatures as described in Van der Schrier et al., 2012. The annual mean values in this E-OBS dataset are presented below for the land area between between 35˚-75˚N and 25˚W-45˚E (top panel, red and blue bars).
As in the global temperature record the warmest years in Europe cluster at the end of the series. Between 1950 and 2010 the Europe-average warming trend in the observations is 0.18 ˚C/decade. For comparison, the Northern Hemisphere (land-only) trend in the same period is 0.19˚C/decade and the global (land-only) trend 0.12˚C/decade (source: CRUTEM4v). The upward trend in Europe is most prominent since the mid-eighties.
The grey bars in all three panels indicate the estimated uncertainties which take into account the errors introduced by spatial interpolation over areas without observation stations, inhomogeneities in the temperature data that result from station relocations / changes in measurement instruments etc., and biases due to urbanisation (Van der Schrier et al., 2012). The uncertainties tell us that although we are not 100% certain about the ranking of individual years, the overall positive trend is clear.
The middle panel shows the difference between the European temperatures based on E-OBS and the European windows of the three best known global datasets: GISTEMP, CRUTEM4v, and GHCN-M. With a few exceptions all years in these other datasets fall within the estimated uncertainty range of E-OBS.
The bottom panel shows the difference between the European temperatures based on E-OBS and three datasets derived from re-analyses of past weather using numerical models: 20th Century Reanalysis, ERA-40, and ERA-Interim. EURO4M reanalysis datasets will be added when available. Large differences are found for the 20th Century Reanalysis which is much colder than E-OBS in recent years. Most likely this is due to a model deficiency, because the other two datasets fall within the estimated uncertainty range of E-OBS.
Frequency of warm and cold days
An analysis of daily temperatures in the E-OBS dataset indicates that the frequencies of warm and cold days are changing too. A warm (cold) day is defined as a day with an average temperature that belongs to the warmest (coldest) 10% in the period 1961-1990. Shown in the diagram below are the anomalies of the number of cold and warm days per year averaged over Europe. Recent decades saw more warm days and fewer cold days than before.
Monthly temperatures over the last two years
Although Europe is warming, relative cold months with negative deviations from the 1961-1990 climatological mean still occur, as depicted in the monthly averaged European temperature anomalies over the last two years.
February 2012 was the largest negative (cold) anomaly and December 2011 the largest positive (warm) anomaly for Europe as a whole. The spatial distribution of the temperatures in these months are shown below showing highest deviations over the eastern parts of the continent.
The most recent year: 2012
Europe-average temperature characteristics for 2012 are:
mean temperature 10.6˚C (= 0.9˚C warmer than the 1961-1990 average)
minimum (night-time) temperature 6.0˚C (= 0.9˚C warmer than the 1961-1990 average)
maximum (day-time) temperature 15.5˚C (= 0.9˚C warmer than the 1961-1990 average)
The map below shows that 2012 was relatively warm for nearly all European countries.
Seasonal values in 2012
Winter, spring, summer, and autumn in 2012 were all relatively warm (0.5, 1.3, 1.4 and 1.5˚C above the 1961-1990 average, respectively).
Over a time period of several decades all seasons are warming. In the top-10 of warmest seasons since 1950 there are only 2 springs, 1 summer, and 2 autumns, that are not from the most recent decade. Recent winters were less exceptional with 6 winters in the top-10 from before 2000, and with winter 2003 as the only recent season ranking in the top 10 coldest seasons. The ranking of all seasons and years since 1950 are given in the ranking list of European temperatures.
The most recent month: December 2012
Europe-average temperature characteristics for December 2012 are:
mean temperature 0.6˚C (= 0.5˚C colder than the 1961-1990 average)
minimum (night-time) temperature -2.7˚C (= 0.2˚C colder than the 1961-1990 average)
maximum (day-time) temperature 3.9˚C (= 0.8˚C colder than the 1961-1990 average)
The map below shows that December 2012 was relatively cold for Scandinavia, and Eastern Europe.
Higher resolution satellite and reanalysis data
Temperature data derived from satellite observations or reanalysis potentially provide more spatial detail than the maps based on E-OBS. An example for the maximum and minimum temperatures on 27 June 2011 at 3 km resolution is given below (top panels). These data are derived from the SEVIRI instrument on board METEOSAT. Comparison with E-OBS data for the same day (bottom panels) shows that there is good overall agreement.
It is too early to base European temperatures on satellites alone. The estimation of air temperature from satellite "skin" temperature observations is still under development. A downside of the use of satellites is the incomplete coverage owing to clouds obscuring the satellite's view of the earth's surface and the absence of satellites before the 1970s.
Another alternative to station observations are reanalyses. However, present-day reanalyses, such as the ones used in the bottom panel of the first diagram, are either very coarse (> 100 km resolution) or available for short time episodes only. The example for maximum temperature at 28 July 2010 below illustrates the recent progress made in the EURO4M project in regional reanalysis down to 12 km resolution. The available computing resources are currently insufficient to construct multi decadal time series at this resolution.
Note that E-OBS temperature maps for each individual day between 1950 and now can be found here
Indices for applications
From the basic climate data several indices are derived for specific applications. Below are two examples of indices based on daily temperature data. Other indices and trends in indices can be found here.
Heating Degree Days
This index is related to the energy consumption for heating of homes and offices. It is defined as the sum of the differences between the observed daily mean temperature and 17˚C for days where the average temperature drops below 17˚C (and heating is required). Maps are shown below for the long-term average Heating Degree Days over Europe (with highest values in Northern Europe and mountainous areas) and the anomalies over 2011 (with low values indicating reduced costs for heating almost everywhere as expected for a warm year).
Number of Tropical Nights
This index is a measure for high night-time temperatures which may have adverse effects on human health as illustrated during the Russian heatwave of Summer 2010. The index is a count of nights where the minimum temperature is greater than 20°C. Maps are shown below for the long-term average Number of Tropical Nights (with highest values in the Mediterranean) and for the anomalies over 2010 (the year of the Russian heatwave) and 2011 (the most recent year). In the area of Moscow, the number of tropical nights during July 2010 was 16, whereas people in this area are used to a value of only 0.5.
E-OBS version 6.0 with underlying station data from ECA&D. The time series plot shows the number of stations available in each year for minimum, mean and maximum temperature. The map shows the station locations.
We acknowledge the E-OBS dataset from the EU-FP6 project ENSEMBLES and the data providers in the ECA&D project. Haylock, M.R., N. Hofstra, A.M.G. Klein Tank, E.J. Klok, P.D. Jones, M. New. 2008: A European daily high-resolution gridded dataset of surface temperature and precipitation. J. Geophys. Res., 113, D20119, doi:10.1029/2008JD10201”
SEVIRI/METEOSAT satellite air temperatures were estimated by the UK Met Office from land surface "skin" temperatures provided by the EUMETSAT LSA-SAF using a regression-based approach.
Authors: Gé Verver, Else van den Besselaar, Albert Klein Tank, Gerard van der Schrier (KNMI, The Netherlands), Lizzie Good, Dale Barker (Met Office, UK), Phil Jones (Climatic Research Unit, University of East Anglia)
Please send your questions, remarks, suggestions to CIB-feedback.
The development of this Climate Indicator Bulletin is part of the EURO4M project. The research leading to these results has received funding from the European Union, Seventh Framework Programme (FP7/2007-2013) under grant agreement n° 242093.
This EURO4M Climate Indicator Bulletin is licensed under a Creative Commons Attribution-NonCommercial-NoDerivs 3.0 Unported License.