Hemispheric and global averages graph
(also available as a EPS and PDF)
More graphs below and here
|
Hemispheric and global averages graph (also available as a EPS and PDF) More graphs below and here |
HadCRUT is a global temperature dataset, providing gridded temperature anomalies across the world as well as averages for the hemispheres and the globe as a whole. CRUTEM and HadSST are temperature datasets for the land and ocean regions, respectively, and contribute to the global dataset.
New versions were published recently and updated in January 2021: HadCRUT5, CRUTEM5 and HadSST4 (see papers).
These are the recommended versions because of the improvements in data and processing methods over the previous versions. They are not yet being regularly updated but they currently have data up to December 2020. We are continuing to update the previous versions (HadCRUT4, CRUTEM4 and HadSST3) each month, and will do so until the new versions can be regularly updated. This will allow users to compare the two versions if they wish.
The new versions introduce a number of innovations and improvements that users need to be aware of, as explained in Morice et al. (2020), Osborn et al. (2020) and Kennedy et al. (2019). The compilation of land air temperature records now includes more stations, and biases in sea surface temperature observations have been reduced. Improved methods of analysis and gridding extend the spatial coverage and reduce coverage biases and reduce uncertainty, especially in the period since 1960.
For the land dataset, there are now two versions. CRUTEM5 uses the standard gridding method and is accompanied by an error model to provide estimates of the uncertainties in grid cell, hemispheric and global series. CRUTEM5alt uses a modified gridding method that removes the under-representation of high-latitude stations that occurs with the standard gridding method.
For the global land and ocean dataset, there are also now two versions. HadCRUT5 Non-Infilled uses similar gridding methods as HadCRUT4, i.e. temperature anomaly values are estimated only in grid cells close to where we have measurements. HadCRUT5 Analysis estimates temperature anomalies using the spatial connectedness of temperature anomaly patterns. This extends the the geographical coverage by estimating temperature anomalies further from the available measurements. This improves the representation of less well observed regions in our estimates of global and hemispheric temperature change.
For our best estimate of how global temperature has changed since 1850, we recommend you use the HadCRUT5 Analysis.
The latest version is HadCRUT5, formed used data from CRUTEM5 and HadSST4. It is currently "static", i.e. not yet being regularly updated. The previous version is HadCRUT4, formed from the combination of CRUTEM4 and HadSST3 and these versions are updated at roughly monthly intervals.
For all versions, hemispheric and global averages as monthly and annual values are available as separate files, alongside the gridded monthly data.
These datasets have been developed by the Climatic Research Unit (University of East Anglia and NCAS) jointly with the Hadley Centre (UK Met Office), apart from the sea surface temperature (SST) dataset which was developed solely by the Hadley Centre.
This webpage gives some brief information to users about the datasets including:
CRU gratefully acknowledges the long-term support for developing, improving and updating these datasets provided by the US Department of Energy (1984-2014) and by the UK National Centre for Atmospheric Science (NCAS) (2016-present), a NERC collaborative centre.
Additional support is also acknowledged: the 2016 update to CRUTEM4.5 and HadCRUT4.5, and the 2017 update to CRUTEM4.6 and HadCRUT4.6 were partially supported by NERC through the SMURPHS project (grant NE/N006348/1) and CRUTEM5.0 and HadCRUT5.0 were partially supported by NERC through both the SMURPHS (grant NE/N006348/1) and GloSAT (grant NE/S015582/1) projects.
The Met Office component of this work is supported by UK BEIS and Defra, through the Hadley Centre Climate Programme.
These major new versions will, in due course, be updated each month. Currently they are static versions for the period 1850-2020.
| Dataset | End month Updated |
Grid | Hemispheric & global means CRU format text files |
Station data | CEDA or Met Office For more data files including uncertainties |
|||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| HadCRUT5 Analysis | 2020-12 2021-01-14 | netCDF (21MB) |
|
Same as CRUTEM5 | Met Office: HadCRUT5 CEDA: HadCRUT5 |
|||||||
| Combined land [CRUTEM5] and marine [HadSST4] temperature anomalies on a 5° by 5° grid with greater geographical coverage via statistical infilling (Morice et al., 2020) | ||||||||||||
| HadCRUT5 Non-Infilled | 2020-12 2021-01-14 | netCDF (21MB) |
|
Same as CRUTEM5 | Met Office: HadCRUT5 CEDA: HadCRUT5 |
|||||||
| Combined land [CRUTEM5] and marine [HadSST4] temperature anomalies on a 5° by 5° grid with geographical coverage limited to grid cells close to where we have measurements (Morice et al., 2020) | ||||||||||||
| CRUTEM5 | 2020-12 2021-01-14 | netCDF (21MB) |
|
Text format (gzip) CRU format netCDF format (zip) |
Met Office: CRUTEM5 CEDA: CRUTEM5 |
|||||||
| Land air temperature anomalies on a 5° by 5° grid, not infilled (Osborn et al., 2020) | ||||||||||||
| CRUTEM5alt | 2020-12 2021-01-14 | netCDF (21MB) |
|
Same as CRUTEM5 | Met Office: CRUTEM5 | |||||||
| Land air temperature anomalies on a 5° by 5° grid, not infilled but with better representation of high-latitude stations (Osborn et al., 2020) | ||||||||||||
| HadSST4 | 2020-12 2021-01-14 | netCDF (21MB) |
|
Not applicable | Met Office: HadSST4 | |||||||
| Sea surface temperature anomalies on a 5° by 5° grid (Kennedy et al., 2019) | ||||||||||||
| Absolute 1961-1990 mean | Static | netCDF (<1MB) |
|
Not applicable | Not applicable | |||||||
| Absolute temperatures for the base period 1961-90 on a 5° by 5° grid (Jones et al., 1999). Note that in this file, latitudes run from South to North to match the HadCRUT5 gridded files (in contrast to the absolute temperature file supplied with HadCRUT4). | ||||||||||||
| Dataset | Full grid | End month Updated |
Hemispheric & global means | Hadley Centre | |||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| HadCRUT4 | NetCDF (21MB) | 2020-11 2020-12-21 |
| HadCRUT4 | |||||||
| Combined land [CRUTEM4] and marine [SST anomalies from HadSST3] temperature anomalies on a 5° by 5° grid (Morice et al., 2012) | |||||||||||
| CRUTEM4 | NetCDF (21MB) | 2020-11 2020-12-21 |
| CRUTEM4 | |||||||
| Land air temperature anomalies on a 5° by 5° grid (Jones et al., 2012) | |||||||||||
| CRUTEM4v | NetCDF (21MB) | 2020-11 2020-12-21 |
| ||||||||
| Variance adjusted version of CRUTEM4 | |||||||||||
| HadSST3 | NetCDF (21MB) | 2020-12 2021-01-05 |
| HadSST3 | |||||||
| Sea surface temperature anomalies on a 5° by 5° grid (Kennedy et al., 2011) | |||||||||||
| Absolute | NetCDF (1MB) | ||||||||||
| Absolute temperatures for the base period 1961-90 on a 5° by 5° grid (Jones et al., 1999). Note that in this file, latitudes run from North to South. | |||||||||||
Correction issued 30 March 2016. The HadSST3 NH and SH files have been replaced. The temperature anomalies were correct but the values for the percent coverage of the hemispheres were previously incorrect. The global-mean file was correct, as were all the HadCRUT4 and CRUTEM4 files. If you downloaded the HadSST3 NH or SH files before 30 March 2016, please download them again.
| Hemispheric/global average data file format |
|---|
for year = 1850 to endyear format(i5,13f7.3) year, 12 * monthly values, annual value format(i5,12i7) year, 12 * percentage coverage of hemisphere or globe |
|
Coverage of 0 means data not yet available
Download an R function to read this format |
Timeseries of global and hemispheric mean temperature anomalies as well as maps of the current year's data are available here.
Also see Tim Osborn's take on Ed Hawkins' famous temperature spiral.
For graphs (and data) of individual land grid cells or individual weather stations, use our CRUTEM Google Earth interface.
HadCRUT4 and CRUTEM4 datasets are available for further online analysis at the KNMI Climate Explorer.
Most years, we also add in updated data for stations that do not report in real time, by using
station data that we access from National Meteorological Services (NMSs) around the
world. These updates typically take place between May and September each year, as by then sufficient
NMSs should have made their monthly average data available for the preceding year. Where
available, we add in extra data from some NMSs when they make more homogeneous
data available. This includes routine updates from the USA, Canada, Russia, China,
Australia and a number of European countries.
How are the hemispheric and global anomaly series calculated?
Values for the hemisphere are the weighted average of all the non-missing, grid-box
anomalies in each hemisphere. The weights used are the cosines of the central latitudes of
each grid box. The global average for CRUTEM4 and CRUTEM4v is a weighted average of
the Northern Hemisphere (NH) and Southern Hemisphere (SH). The weights are two for the NH
and one for the SH. See Osborn and Jones (2014) for how this differs from previous versions of CRUTEM.
For HadCRUT4, the global average is the unweighted
average of the NH and SH values. In the timeseries files, the second row of integers is the
percentage of the surface area covered for each month from 1850. In the CRUTEM4 timeseries graphs above, we only
show the SH and global averages from 1856 onwards because the land data coverage in the SH is poor before 1856.
What are the basic raw data used?
For land regions of the world over 4800 monthly station temperature time series were used when CRUTEM4.0 was first
published. This increased through the quasi-annual improvements to the dataset, reaching over 7000 stations in CRUTEM4.6.
Coverage is denser over the more populated parts of the world, particularly, the United States,
southern Canada, Europe and Japan. Coverage is sparsest over the interior of the South
American and African continents and over Antarctica. The number of available stations
was small during the 1850s, but increases to over 4500 stations during the 1951-2010 period.
For marine regions, sea surface temperature (SST) measurements taken on board merchant
and naval vessels are used. As the majority come from the voluntary observing fleet,
coverage is reduced away from the main shipping lanes and over parts of the Southern
Oceans. Improvements in coverage occur after 1980 through the deployment of fixed and
drifting buoys. The development of the CRUTEM4 and HadSST3 datasets is extensively discussed in Jones et al. (2012)
and Kennedy et al. (2011). Both these sources also discuss the
consistency and homogeneity of the measurements through time and the steps that have been made
to remove non-climatic inhomogeneities.
Raw station data used to produce CRUTEM4 are available from the Met Office website
(CRUTEM4) and
the station data (and graphs) are also available
via our Google Earth interface.
Why are sea surface temperatures rather than air temperatures used over the oceans?
Over the ocean areas the most plentiful and most consistent measurements of temperature
have been taken of the sea surface. Marine air temperatures (MAT) are also measured and would,
ideally, be preferable when combining with land air temperatures, but they involve more
complex problems with homogeneity than SSTs (Kennedy et al., 2011). The problems are
reduced using only night marine air temperature (NMAT) but at the expense of discarding
approximately half the MAT data. Our use of SST anomalies implies that we are tacitly
assuming that the anomalies of SST are in agreement with those of MAT. Kennedy et al.
(2011) provide comparisons of hemispheric and large area averages of SST and NMAT anomalies.
Why are the temperatures expressed as anomalies from 1961-90?
Stations on land are at different elevations, and different countries calculate average monthly
temperatures using different methods and formulae. To avoid biases that could result from
these differences, monthly average temperatures are reduced to anomalies from the period with
best coverage (1961-90). For stations to be used, an estimate of the base period average must
be calculated. Because many stations do not have complete records for the 1961-90 period
several methods have been developed to estimate 1961-90 averages from neighbouring
records or using other sources of data (see Osborn and Jones, 2014; Jones
et al., 2012). Over the oceans, where observations are generally made from mobile platforms,
it is impossible to assemble long series of actual temperatures for fixed points. However it is
possible to interpolate historical data to create spatially complete reference climatologies
(averages for 1961-90) so that individual observations can be compared with a local normal
for the given day of the year (more discussion in Kennedy et al., 2011).
It is possible to obtain an absolute temperature series for any area selected, using data from the
absolute file, and then add this to a regional average of anomalies calculated from the gridded
data. If a regional average is required, users should calculate a regional average time series in
anomalies, then average the absolute file for the same region, and lastly add the average derived to
each of the values in the time series. Do NOT add the absolute values to every grid box in
each monthly field and then calculate large-scale averages.
Why do anomalies not average exactly zero over 1961-1990?
Not all regions
have complete data for the 1961-1990 period, so the anomaly data do not average exactly to
zero for this 30-year period. This applies to the global and hemispheric average series as
well as the individual grid-box series.
How are the land and marine data combined?
Both the component parts (land and marine) are separately averaged into the same 5°x5°
latitude/longitude grid boxes. The combined version (HadCRUT4 ) takes values from each
component and weights the grid boxes according to the area, ensuring that the land component
has a weight of at least 25% for any grid box containing some land data.
The weighting method is described in Morice et al. (2012).
How accurate are the hemispheric and global averages?
Uncertainty estimates are supplied with the same data given at the Met Office site:
CRUTEM4,
HadCRUT4.
Why can I not exactly reproduce the hemispheric and global averages for HadCRUT4 and HadSST3 that are given here?
Both these are ensemble datasets. This means that there are 100 realizations of each in order
to sample the possible assumptions involved in the structure of the various components of
the error (see Morice et al., 2012). All 100 realizations are available at the
above Hadley Centre site, but here we provide the ensemble median. For the gridded data
this is the ensemble median calculated separately for each grid box for each time step from
the 100 members. For the hemispheric and global averages this is again the median of the 100
realizations. The median of the gridded series does not produce the median of the hemispheric
and global averages, but the differences are small.
Why are values slightly different when I download an updated file a year later?
All the files on this page (except Absolute) are updated on a monthly basis to include the
latest month within about four weeks of its completion. Updating includes not just data for
the last month but the addition of any late reports for up to approximately the last two years.
Most years we also add in updated data for stations that do not report in real time, by using
station data that we access from NMSs around the
world. These additions typically take place between May and September, as by then sufficient
NMSs will have made their monthly average data available for the preceding year. Where
available, we add in extra data from some NMSs when they make more homogeneous
data available. The routine annual updates include data from the USA, Canada, Russia, China,
Australia and a number of European countries.
In addition to this the method of variance adjustment (used for CRUTEM4v)
works on the anomalous temperatures relative to the underlying trend on
an approximate 30-year timescale.
With the addition of
subsequent years, the underlying trend will alter slightly, changing the variance-adjusted
values. Effects will be greatest on the last year of the record, but an influence can be evident
for the last three to four years. Full details of the variance adjustment procedure are given in
Jones et al. (2001).
Previous versions
|
