Objective: Grounded in ecological system theory , this study considers the longitudinal association between interparental conflict (IPC) and risk-taking behavior among adolescents and investigates the underlying mediating and moderating mechanisms. Background: IPC has been shown to have a detrimental impact on adolescent behavior development. However, little is known about the processing mechanisms underlying the association of IPC and risk-taking behavior from an ecosystem perspective. Method: This study conducted a longitudinal design (3 time points, 3 months apart) with the sample comprising 550 middle school students in southeastern China (52.91% males; mean age at Time 1 = 15.37). The performed measurements encompassed IPC (T1), deviant peer affiliation (T2), school climate (T3), risk-taking behavior (T1/T2/T3), and demographic information. Results: The moderated mediation model revealed that after controlling for T1/T2 risk-taking behavior, T1 IPC was longitudinally and positively correlated with T3 risk-taking behavior through T2 deviant peer affiliation. Furthermore, the indirect effect of T2 deviant peer affiliation was significantly stronger under a low-level T3 school climate. Conclusion: Adolescent risk-taking behavior is the joint effect of risk factors from multiple microsystems. Specifically , IPC is associated with higher levels of deviant peer affiliation, which, in turn, are associated with more risk-taking behavior. Moreover, the negative school climate serves as a risk factor to enhance the adverse impact of deviant peer affiliation on adolescents' risk-taking behavior.