Irrigation activities have been recognized as one of the most potentially sustainable land management practices to tackle climate change. In this study, we investigated the impacts of present-day irrigation activities on regional temperature extremes changes in China using the state-of-the-art regional climate model RegCM-4.8.2, through two sets of long-term numerical experiments (i.e., with and without irrigation). The key findings are as follows. Firstly, by comparing the observations, RegCM4 has a good ability to reproduce both the spatial distribution and the volume of irrigation for provincial irrigation activities in China. In this context, by comparing the results of numerical experiments, we found that irrigation led to significant reductions (slightly changes) in the extreme indices associated with the warm (cold) tails of the maximum and minimum temperatures distributions, e.g., hot days, tropical nights (cold nights, frost days), particularly in the regions with intense irrigation (e.g., the North China Plain). Furthermore, these extreme indices were dominated by irrigation-induced shifts in mean temperatures, but the impacts of changes in temperature variability on extreme hot events were also non-negligible for the strongly irrigated regions. In addition, through attributing irrigation-induced changes in mean state temperatures, we revealed that the irrigation-induced repartitioning of the surface turbulent heat fluxes (i.e., sensible and latent heat) dominated the variations of temperatures and the associated extremes. However, region-specific land-atmosphere coupling features also led to differences in the irrigation-induced responses of surface energy components. We highlight the possible effects and mechanisms of irrigation on multiple types of extreme temperature indices, particularly the extreme hot events, which provide scientific clues for mitigating and adapting regional climate change.