|
|
DOWNLOAD DATA
Stored files in the database are in NetCDF format (Network Common Data Form).
The database is organised in a way that each variable for each experiment (scenario) and dataset (corrected/uncorrected) is stored in a separate file. For the period 1951-2000 daily files have 18263 records (one record is written for each day) and 31 MB, while monthly files have 600 records (one record for each month) and 1.1 MB. For the period 2001-2100 daily files have 36524 records and 62 MB, and monthly files have 1200 records and 2.1 MB.
Files name convention is given as: var_yst-ynd_sc_cr.nc, where:
var is the name of the variable and can be one of the following: tg (mean temperature), tx (maximum temperature), tn (minimum temperature), rr (precipitation)
yst is the starting year of the experiment: 1951 or 2001
ynd is the ending year of the experiment: 2000 or 2100
sc is the label for a scenario and can be: 20c3m, A1B or A2
cr is a label for bias corrected or uncorrected data: corr (for corrected data) or uncorr (for uncorrected data)
monthly data have the same names as daily, except they have an additional letter "m" in front of the variable name (for an example, mtg_2001-2100_A1B_corr.nc is a file with monthly data, while file tg_2001-2100_A1B_corr.nc contains daily data)
RAW (UNCORRECTED) MODEL DATA
DAILY DATA (link)
MONTHLY DATA (link)
BIAS CORRECTED MODEL DATA
BIAS CORRECTED DAILY DATA (link)
BIAS CORRECTED MONTHLY DATA (link)
Recommendation:
For impact studies it is strongly recommended to use bias corrected dataset.
|