Abstract: The precursor signal of a main mode affecting the summer precipitation in Heilongjiang province was analyzed using the summer precipitation data from 15 meteorological stations from 1981 to 2010,NCEP/NCAR (National Centers for Environmental Prediction/National Center for Atmospheric Research) reanalysis data and monthly average sea surface temperature from NOAA (National Oceanic and Atmospheric Administration) from 1980 to 2015.A statistical precipitation forecasting model for summer precipitation in Heilongjiang province was established based on the principal component regression method.The results show that the abnormal precursor signals affecting the main mode of summer precipitation in Heilongjiang province mainly include the height field in south regions of Novaya Zemlya and the Kara Sea last November,the inverse variation of the sea surface temperature between the central part of North Pacific and California vicinity last winter,and the anomaly sea surface temperature in the southern Indian Ocean.Through yearly cross hindcast and every four year independent sample prediction tests from 1981 to 2010,it indicates that the precipitation forecasting model established here performs very well.It is used in operation and gets a good prediction in the summer precipitation in Heilongjiang province in 2015.It suggests that this model may have the value for the summer prediction.

Key words: Summer precipitation, Precursor signal, Principal Component analysis, Stepwise-multiple linear regression, Test