Statistical forecast of temperature dynamics within month on the basis of hydrodynamic model outputs

A method of a long-term forecast of smoothed temperature dynamics within month with a zero lead time for some Russian sites is considered. A procedure of a search for optimum predictors and a regression-analog prediction scheme are described. Hydrodynamic forecasts for the first ten days of the prognostic month are used as a predictor. Results of the prediction scheme verification on independent data with the use of hydrodynamic model outputs are discussed. Ways to improve the prediction scheme are suggested.