Governed by the Clausius-Clapeyron (CC) equation, daily mean temperature (T_m) and precipitation extremes would be theoretically linked by atmospheric moisture. However, precipitation extremes cannot systematically follow the CC rate of 7% per warming degree, due to moisture limitations. In this study, the observational tri-pole relation among T_m, atmospheric moisture and precipitation intensity over China have been investigated. The results indicate that atmospheric moisture (specific humidity and dew-point temperature) is positively correlated with T_m (precipitation) across four seasons at the interannual timescale. Particularly, the increase in precipitation extremes is accompanied by high atmospheric moisture, but is different in four seasons. In comparison, the year-to-year relation between T_m and atmospheric moisture is stronger than that between atmospheric moisture and precipitation. The atmospheric moisture has acted as a bridge linking T_m and precipitation extremes. T_m is highly correlated with precipitation extremes, while the relation also shows seasonal difference. The difference may be attributed to the negative scaling of daily precipitation extremes and precipitation efficiency (defined as the percentage of moisture in the air converting into precipitation) with T_m, when T_m exceeds ~25 °C.