The IPCC Data Distribution Centre The CRU Global Climate Dataset

A fundamental data requirement for climate change impacts research is an observed climatology that captures both the spatial and temporal characteristics of a range of surface climate variables. The climatology should be of sufficient spatial resolution to permit regional and country scale studies. In addition, accurate representation of variability in time demands careful attention to the temporal fidelity of the climatology.

Description
The CRU Global Climate Dataset available through the IPCC DDC consists of a multi-variate 0.5º latitude by 0.5º longitude resolution mean monthly climatology for global land areas, excluding Antarctica, strictly constrained to the period 1961-1990, together with monthly time series at the same resolution for the period 1901-1995. The mean 1961-1990 climatology comprises a suite of eleven surface variables: precipitation (PRE) and wet-day frequency (WET); mean, maximum and minimum temperature (TMP, TMX, TMN); diurnal temperature range (DTR); vapour pressure (VAP;) global radiation (RAD;) cloud cover (CLD); frost frequency (FRS); and wind speed (WND). The anomaly time series component comprises all variables except global radiation and wind speed. 

The mean 1961-90 climatology
The mean climate surfaces have been constructed from a new dataset of station 1961-1990 climatological normals, numbering between 19,800 (precipitation) and 3615 (windspeed). The station data were interpolated as a function of latitude, longitude and elevation using thin-plate splines. The accuracy of the interpolations are assessed using cross-validation and by comparison with other climatologies. A paper describing this climatology is due shortly for publication in the Journal of Climate as New,M., Hulme,M. and Jones,P.D. (1999) Representing twentieth century space-time climate variability. Part 1: development of a 1961-90 mean monthly terrestrial climatology J.Climate 12, 829-856.

The anomaly timeseries
The anomaly time series were constructed using historic anomalies derived from the monthly data holdings of the Climatic Research Unit and the Global Historic Climatology Network (GHCN). For the purposes of developing monthly gridded time series, the variables were classified as either primary or secondary. For the primary variables - PRE, TMP, DTR - sufficient data were available to enable interpolation directly from the station time series. In the case of secondary variables - CLD, VAP, FRS, WET - the available station time series were sparsely sampled in space and time. These variables had to derived indirectly from gridded time series of primary variables. Station data that were available for secondary variables were used to develop relationships to the primary variables, and to validate the derived gridded time series.

The full global climate dataset
To calculate monthly time series, grids of monthly anomalies relative to 1961-90 were calculated for each variable and applied to their respective 1961-90 climatology. The anomaly approach was adopted because the network of station normals was much more comprehensive than the network of station time series. The spatial variability in mean climate was best captured by the denser network of station normals, while the more sparse network of primary variable time series captured as much temporal variability as possible. A paper describing this full dataset is in preparation, but initial details can be found under the homepage of the lead developer, Mark New.

Viewing and availability
Selected fields and time series from this climatology can be viewed through the Data Visualisation pages of the DDC. Decade-mean and 30-year mean monthly fields can be downloaded from the Data Download pages. Access to the full year-by-year monthly dataset is achieved by lodging a request with the Climate Impacts LINK Project at the Climatic Research Unit (email: d.viner@uea.ac.uk, web site: http://www.cru.uea.ac.uk/link).