Abstract: The florescence of the local peony(Paeonia suffruticosa Andr.) in the Heze City hardly meets the date of the Heze International Peony Fair,which greatly influences the local industrial development of the tree peony.Based on the analysis of the relationship between the florescence of the local peony and the soil temperature,a prediction model for florescence of tree peony based on the soil temperature was established by multivariate nonlinear regression analysis.The parameters included the days from January 20th to the first day with soil temperature above 4℃ and the average soil temperature from the first day to April 10th.The prediction model was validated by the florescence data in the last 28 years.The maximal error between the predicted date for the florescence of the tree peony and the factual florescence observed was with 4 days.The model could provide the prediction method to determine the date of the Heze International Peony Fair.

Key words: Prediction of florescence, Prediction of florescence, Soil temperature, Nonlinear multianalysis regressing analysis, Prediction model