Prediction for drought strength in North China

Based on the analysis of spatial and temporal features of drought strength in North China, a new technique to predict the drought strength with an integration of both the droughty trend and the dynamic strong signal is proposed. Considered the prominent interdecadal and interannual variation of drought strength, which are separated by means of a nonlinear dynamic reconstruction,then the two models with different time scale to predict droughty trend are established respectively and integrated at last. A concept of dynamic strong signal is introduced into interannual varieties prediction model by use of signal to noise ratio method to search regions where the atmospheric and oceanic observations are prominently different from the multi-year mean. Therefore, possible influences of the lately anomaly of the atmosphere and ocean can be included in the prediction models. The extraseasonal hindcast from 1996 to 2002 show that the prediction model represents the droughty trend preferably and exhibits higher prediction skill.