The objectives of the study were to validate the predictive models of winter mortality and population dynamics of bean leaf beetle, Cerotoma trifurcata (Förster), (Coleoptera: Chrysomelidae) developed by Lam and Pedigo (2000b) and Lam et al. (2001) and apply the models in developing new strategies for the beetle management. The study was conducted at Iowa State University Johnson Farm near Ames, Iowa from 1999 through 2002. The experimental design and procedures for sampling the adult populations were following the method used by Lam et al. (2001). Validation of the models was conducted by substituting the data from 1999 through 2002 into the models and comparing the predicted results with the sampled beetle populations. The results showed that the models could predict the trend of the beetle populations accurately in the following season. Regression analysis indicated that the first-generation beetle density was highly correlated to the predicted overwintered population. Additionally, the results demonstrated that winter temperature is the most important limiting factor and mid-summer precipitation is the second limiting factor on the bean leaf beetle population dynamics. The strategies of applying the predictive models in a management program for bean leaf beetle are discussed.