Release notes for the CRU CY dataset (all versions) --- Introduction The CRU CY dataset consists of country averages at a monthly, seasonal and annual frequency, for ten variables. Spatial averages are calculated using area-weighted means. The CRU CY dataset is an update of the TYN CY dataset, which may be found on the CRU website: http://www.cru.uea.ac.uk/cru/data/hrg/ --- Relationship to Other Datasets CRU CY is derived directly from the CRU TS dataset, and version numbering is matched between the two datasets. Thus, the first official version of CRU CY is v3.21, as it is based on CRU TS v3.21. --- Format Each country file contains all the spatial means for a particular country and variable. The files are ASCII text, and should be readable with almost any application including a text editor (they import into Excel readily, start the import at line 4 or 5). Each file begins with three information lines: Line 1: Includes the date and time of produciton, and the date string uniquely identifying the CRU TS run on which the dataset is based. Line 2: Identifies the country, parameter (variable) and units (eg, mm). Line 3: Contains the time period, the missing value code, and the data format (as a Fortran-style string). There is then a data header, describing the following columns: YEAR JAN-DEC (12 columns, monthly means) MAM, JJA, SON, DJF (4 columns, seasonal means) ANN (annual mean) After these lines come the data, one line per year. For version 3.21, this covers 1901 to 2012. This data format is almost identical to that of the TYN CY dataset, although the reference to the CRU TS version and run (on line 1) is new. --- Notes on usage To understand this dataset, it is important to understand the construction and limitations of the underlying dataset, CRU TS. It is therefore recommended that all users read the relevant paper: Harris, I., Jones, P.D., Osborn, T.J., and Lister, D.H., 2013: Updated high-resolution grids of monthly climatic observations. In press, Int. J. Climatol., Doi: 10.1002/joc.3711. In brief: the CRU TS dataset prioritises completeness, and has no missing data over land. Where observations are unavailable, the 1961-90 monthly climatic mean is used as a substitute. In data sparse regions of the world, this can lead to repeated values, and this can show up in derived products such as CRU CY. Unlike the previous TYN CY dataset, CRU CY consists solely of country averages. Finally, note that the 'DJF' (winter) value for the last year will be missing; this is because the season straddles two years, and cannot be calculated without the January and February values.