Abstract: Based on the geopotential height field products from 1983 to 2015 provided by the DERF (Dynamic Extended Range Forecast) operational system version 2.0,NCEP/NCAR (National Centers for Environmental Prediction/National Center for Atmospheric Research) reanalysis data,and monthly mean temperature data in Northeast China,a scheme for monthly-temperature prediction was established by using the downscaling interpretation and application of the DERF v2.0 products.A temperature prediction model in November in Northeast China was established based on the above methods.The results show that this model has a good forecasting performance,with a high Ps score and can be used in the climate forecasting operation.Meanwhile,the model's stability for forecasting the temperature in November is discussed,and some solutions to improve this model are provided.