The relevant changes for this country data-set are as follows:

• The time-series for all available climate variables have been regridded and extended to 2000. The new observed time-series information entirely replaces the old information. See the individual variable files for each country in the country table.
• The mean climatology (1961-90) has not been altered in the grids; the mean climatology is still that of CRU CL 1.0 (see the Table). The climatic summary for each individual country is available through the country table. The consolidated file (all countries) for each variable is available through the variable table.
• In addition to the 30-year mean from 1961-90, for five variables we now also present the interannual standard deviation from that 30-year period. Please note that these data may include influences from the changes over time in the station networks contributing to the underlying grids (see discussion). Therefore, particularly for smaller countries and data-sparse regions, the 'standard deviation' presented here may not be entirely climatic, but also a function of the number of observations. The consolidated file (all countries) for each variable is available through the variable table.

Guidance
This data set is intended for use in trans-boundary research, where it is necessary to average climatic behaviour over a wide area into statistics that are representative of the whole area. The user should note that any single country may contain substantial spatial variations of climate within it, and that climate changes over time.

This data set should not be used to represent climate at a point, or for sub-regions of the countries and territories supplied here. If climate data is required for points or sub-regions, enquiries should be made to the national meteorological agency of the country concerned.

This data-set should only be used for time-series analysis (e.g. trends or regression) with the greatest of care. This data-set is based on the CRU TS 2.0 grids, and the discussion of the limitations on the use of that data-set is relevant here.

Please note that the standard deviation data may include influences from the changes over time in the station networks contributing to the underlying grids (see discussion). Therefore, particularly for smaller countries and data-sparse regions, the 'standard deviation' presented here may not be entirely climatic, but also a function of the number of observations.

The credibility of the data set rests upon observed grids previously constructed (CRU TS 2.0 - see details of the original data). The work of this author was to aggregate the grid box values into averages representing countries and territories (see details of the data processing).

For many countries, the accuracy is better for more recent decades than for earlier decades of the 20th century (although data is provided wherever possible). In addition, the useful precision of the data may be less in earlier decades of the 20th century (although the recorded precision is constant throughout). For some variables and countries there is insufficient data to make a reasonable estimate of year-to-year variations in the early part of the 20th century; in such cases, the values recorded are 'relaxed' towards the average from 1961-1990, and therefore are not representative of the real-world year-to-year variations.

One climatic season (December-February) overlaps from one calendar year to the next. The value we supply begins in the December of the year recorded, and ends in the February following the year recorded. Therefore the last year recorded always contains the missing value (-999.0) in the DJF column.

Further information may be obtained from the research paper published in Area. An electronic version is available.

Data Structure
The average climate is available as:

• Web pages, each with all the variables in a single country ('average' column in Table); this contains the 30-year mean only.
• ASCII files, each with all the countries for a single variable (Table); this contains the 30-year mean and the inter-annual standard deviation for the same 30-year period.

• ASCII files, each with one variable in one country (right-hand columns in Table).

Disclaimer
The author bears no responsibility for the accuracy of the data set. The countries listed below include a variety of sovereign states, dependent territories, and disputed territories. No political statement is being made by the inclusion or exclusion of a particular territory, or by the labelling of a particular territory by a particular name.

Data Processing
The original data took the form of a value for each month and each box on a 0.5 degree latitude / longitude grid (CRU TS 2.0). We assigned each box to a single country. For each country we calculated the weighted mean of the values from its constituent grid boxes for each month in turn. Each grid box was weighted by surface area, using the cosine of the latitude. The seasonal and annual values are the means of their constituent months.

Original Data
Station observations were first collected by national meteorological, hydrological and related services, and were acquired through the free and unrestricted exchange of meteorological and related data. These observations were gridded. The gridded data-set is publicly available, and is being published in a peer-reviewed scientific journal:
Mitchell, T.D., Carter, T.R., Jones, P.D., Hulme,M., New, M., 2003: A comprehensive set of high-resolution grids of monthly climate for Europe and the globe: the observed record (1901-2000) and 16 scenarios (2001-2100). Journal of Climate: submitted.