Atmosphere, Год журнала: 2025, Номер 16(5), С. 616 - 616
Опубликована: Май 18, 2025
The aim of the homogenization climatic time series is to remove non-climatic biases from observed data, which are caused by technical or environmental changes during period observations. This bias removal generally more successful for long-term trends and annual means than monthly daily values. probability distribution (HPD) may improve data accuracy even when signal-to-noise ratio favors its application. HPD can be performed quantile matching spatial interpolations, but both them have drawbacks. study presents a new algorithm helps increase in all temporal scales. method similar matching, section mean values function (PDF) compared instead individual input dataset identical with results studied series. decides about statistical significance each break detected means, skips insignificant breaks. Correction terms removing inhomogeneity PDF calculated jointly Benova-like equation system, low pass filter used smoothing prime results, value between two consecutive breaks preserved such sections. initial version does not deal seasonal variations either other steps homogenization. has been tested connecting ACMANTv5.3, using overall 8 wind speed relative humidity datasets benchmark European project INDECIS. show 4 12 percent RMSE reduction scales, except extreme tails where part weaker.
Язык: Английский