Daily validation dataset
Daily surrogate data
The COST Action ES0601: Advances in homogenisation methods of climate series: an integrated approach (HOME) aims at improving homogenisation algorithms. Its working group 4 on daily data initiated a study to validate the correction methods for daily temperature series. A surrogate dataset with centenial daily temperature data has been produced based the statistical properties of shorter homogenized time series from Middle Europe. The power spectra of the stations (equivalent to the autocorrelation function) were reproduced, as well as the distribution of the series themselves and the difference times series. The variance of the difference time series was multiplied by a factor to increase of decrease the cross-correlations between the networks. The distributions are different for every month.
To this homogeneous surrogate data, break inhomogeneities were added every 20 years. These inhomogeneities were simulated by either switching the distribution to the one of a nearby station (to simulate a relocation of the station) or by perturbing the data with several functions to simulate a change in the measurements themselves. This dataset is available for download. Everyone is welcome to homogenize the data until the end of August. The results will be discussed at a meeting held during EMS2011 in Berlin. If you have any questions or would like to submit a homogenized dataset please contact the leader of WG4, Petr Stepanek.