The HADRT2.2 data are global monthly fields of radiosonde temperature anomalies at standard pressure levels on a 10 degree latitude by 20 degree longitude grid from 1958 to 2000. Anomalies are calculated with respect to 1971-1990 climatology. Anomalies are available for 9 standard levels (850, 700, 500, 300, 200, 150, 100, 50, 30hPa) as well as tropospheric (850 - 300hPa) and stratospheric (150 - 30hPa) averages. The data are degree Celsius anomalies from 1970-1990 means. Anomalies are calculated for each of about 200 sonde stations worldwide and grid values derived from these. HADRT2.2 is an eigenvector reconstructed grid data set from 1958 - 2000, on a 10 degree latitude by 20 degree longitude grid, created from HadRT2.1. The eigenvector reconstruction was used to infill missing seasons or years in boxes with 70% of seasonal or annual data available. This dataset has been superseded by the HadAT dataset also available from CEDA.

The HADRT2.2u data are global monthly fields of radiosonde temperature anomalies at standard pressure levels on a 10 degree latitude by 20 degree longitude grid from 1958 to 2000. Anomalies are calculated with respect to 1971-1990 climatology. Anomalies are available for 9 standard levels (850, 700, 500, 300, 200, 150, 100, 50, 30hPa) as well as tropospheric (850 - 300hPa) and stratospheric (150 - 30hPa) averages. The data are degree Celsius anomalies from 1970-1990 means. Anomalies are calculated for each of about 200 sonde stations worldwide and grid values derived from these. HADRT2.2u are as HADRT2.2 but created from HADRT2.0. This dataset has been superseded by the HadAT dataset also available from CEDA.

The HadAT2 data are global radiosonde gridded temperature anomalies at standard levels (850, 700, 500, 300, 200, 150, 100, 50, and 30hPa) in the troposphere and in the lower stratosphere from 1958 to December 2012. This monthly timeseries are available on a 10 degree longitude by 5 degree latitude basis. This dataset supersedes the HadRT dataset. All values are anomalies relative to the monthly 1966-95 climatology.

The Met Office Hadley Centre produced the HadSLP1 dataset which replaces the Global Mean Sea Level Pressure (GMSLP) data sets, and is a unique combination of monthly globally-complete fields of land and sea pressure observations a 5 degree latitude-longitude grid from 1871 to 1998. The advantages of HadSLP1 over GMSLP2 are an improved land station data base, new interpolation scheme and the incorporation of local detail while safeguarding against random errors. Like GMSLP2, HadSLP1 was developed by Tracy Basnett and David Parker, in collaboration with R.J. Allan (previously at CSIRO and now at the Met Office) and M.J. Salinger (NIWA). Marine observations were taken from the Met Office Marine Data Bank (MDB) and from the NOAA Comprehensive Ocean-Atmosphere Data Set (COADS) (Woodruff et al, 1987). The MDB data were given priority in the blend, which is described in Basnett and Parker (1997). Land observations are a combination of data obtained from CSIRO (Australia), NIWA (New Zealand), CRU (University of East Anglia), GHCN, and operationally-received "CLIMAT" messages. The latter were used mainly for updating station series and for the Arctic. HadSLP1 pressures are reconstructed using a reduced-space optimal interpolation procedure, followed by superposition of quality-improved gridded observations onto the reconstructions to restore local detail.

The HADRT2.3 data are global monthly fields of radiosonde temperature anomalies at standard pressure levels on a 5 degree latitude by 10 degree longitude grid from 1958 to 2000. Anomalies are calculated with respect to 1971-1990 climatology. Anomalies are available for 9 standard levels (850, 700, 500, 300, 200, 150, 100, 50, 30hPa) as well as tropospheric (850 - 300hPa) and stratospheric (150 - 30hPa) averages. The data are degree Celsius anomalies from 1970-1990 means. Anomalies are calculated for each of about 200 sonde stations worldwide and grid values derived from these. HADRT2.3 is a globally complete dataset based on HadRT2.1 1958-2000, but with gaps filled in by reference to the second derivative of the corresponding NCEP reanalysis temperature fields, using the Laplacian technique of Reynolds, 1988: A Real-Time Global Sea Surface Temperature Analysis. J. Climate (see docs for a link to this publication) This dataset has been superseded by the HadAT dataset also available from CEDA.

The HADRT2.0 (unadjusted) data are global monthly fields of radiosonde temperature anomalies at standard pressure levels on a 5 degree latitude by 10 degree longitude grid from 1958 to July 2004. Anomalies are calculated with respect to 1971-1990 climatology. Anomalies are available for 9 standard levels (850, 700, 500, 300, 200, 150, 100, 50, 30hPa) as well as tropospheric (850 - 300hPa) and stratospheric (150 - 30hPa) averages. The data are degree Celsius anomalies from 1970-1990 means. Anomalies are calculated for each of about 200 sonde stations worldwide and grid values derived from these. This dataset has been superseded by the HadAT dataset also available from CEDA.

The HadAT1 data are global radiosonde gridded temperature anomalies at standard levels (850, 700, 500, 300, 200, 150, 100, 50, and 30hPa) in the troposphere and in the lower stratosphere from 1958 to December 2002. This monthly timeseries are available on a 10 degree longitude by 5 degree latitude basis. This dataset supersedes the HadRT dataset. All values are anomalies relative to the monthly 1966-95 climatology.

The HadAT data are global radiosonde gridded temperature anomalies at standard levels (850, 700, 500, 300, 200, 150, 100, 50, and 30hPa) in the troposphere and in the lower stratosphere from 1958 to December 2012. This monthly timeseries are available on a 10 degree longitude by 5 degree latitude basis. This dataset supersedes the HadRT dataset. All values are anomalies relative to the monthly 1966-95 climatology. The gridded product is derived from 676 individual radiosonde stations with long-term records. Because of the criteria of data longevity the resulting dataset is limited to land areas and primarily Northern Hemisphere locations. Radiosondes are single launch instruments and there have been many changes in instruments and observing practices with time. HadAT has used a neighbour-based approach to attempt to adjust for these effects and produce a homogeneous product suitable for climate applications. Zonal averages were created by averaging all available gridbox values in each 5 degree latitude band. Large scale mean timeseries for the globe and the tropics (defined here as 20N to 20S) are simply cos(lat) weighted zonal mean field values. This reduces the spatial sampling bias towards Northern Hemisphere mid-latitudes. Other plausible large-scale averaging techniques would impact the timeseries values calculated. Cautionary note from the Met Office Hadley Centre It is important to note that significant uncertainty exists in radiosonde datasets reflecting the large number of choices available to researchers in their construction and the many heterogeneities in the data. To this end we strongly recommend that users consider, in addition to HadAT, the use of one or more of the following products to ensure their research results are robust. Currently, other radiosonde products of climate quality available from other centres for bona fide research purposes are: - Radiosonde Atmospheric Temperature Products for Assessing Climate (RATPAC) - RAdiosonde OBservation COrrection using REanalyses and Radiosonde Innovation Composite Homogenization ) - IUK (Iterative Universal Kriging) Radiosonde Analysis Project Also see a comparison with Microwave Sounding Unit (MSU) data by Remote Sensing Systems (see links under Docs tab below) .

MOHSST, (Met Office Historical Sea-Surface Temperature) is a gridded dataset of sea-surface temperature anomalies covering the period 1856-2006. MOHSST has now been superceeded by HadSST2. We now recommend use of HadSST2 instead of MOHSST for all purposes. MOHSST is only still available in case it is needed for direct comparison with earlier work where MOHSST was used. MOHSST is produced by taking in-situ measurements of SST from ships and buoys, rejecting measurements which which fail quality checks, converting the measurements to anomalies by subtracting climatological values from the measurements, and averaging the resulting anomalies on a 5 by 5 degree monthly grid. Up to 1996 the measurements used are those in the U.K. Marine Data Bank; more recent years use data coming in through the GTS. After gridding the anomalies, bias corrections are applied to remove spurious trends caused by changes in SST measuring practices, and the data are smoothed to reduce noise. The data were provided by the Met Office Hadley Centre.

This dataset holds Northern Hemisphere (north of 15 deg. N) daily and monthly series of 1000-500hPa Geopotential Height Thickness Daily and Monthly Series. The data is gridded on a 5x10 degree grid. The data is available for the period 1945 to 2005. The geopotential thickness between pressure levels — difference of the 1000 hPa and 500 hPa geopotential heights for example — is proportional to mean virtual temperature in that layer. The data is supplied by the Met Office, Hadley Centre.