The objective of this work is to estimate regional rainfall thresholds obtained from a combination of high-resolution gridded rainfall data and debris flow events. The debris flow events were associated with rainfall data, determining triggering and non-triggering rainfall events. The method for determining the thresholds is based on an empirical–statistical approach, and the predictive performance of the thresholds is evaluated whit “true skill statistics”. The validation of the thresholds was carried out by selecting one year of the debris flow dataset to focus on the operability evaluation of thresholds in early warning systems in Peru. In addition, another validation method based on a random selection of the events was used to compare the validation procedure. The thresholds were determined for 11 rainfall regions in Peru. The best predictive performance is the mean daily intensity-duration I-D threshold curve, followed by accumulated rainfall E. This study is highly important to Peru because is one of the first approximations of rainfall thresholds at a regional scale in Peru.