This study proposes a method to estimate the timing of human-induced climate change (HICC) emergence from internal climate variability (ICV) for hydrological impact studies based on climate model ensembles. Specifically, ICV is defined as the inter-member difference in a multi-member ensemble of a climate model in which human-induced climate trends have been removed through a detrending method. HICC is defined as the mean of multiple climate models. The intersection between HICC and ICV curves is defined as the time of emergence (ToE) of HICC from ICV. A case study of the Hanjiang River watershed in China shows that the temperature change has already emerged from ICV during the last century. However, the precipitation change will be masked by ICV up to the middle of this century. With the joint contributions of temperature and precipitation, the ToE of streamflow occurs about one decade later than that of precipitation. This implies that consideration for water resource vulnerability to climate should be more concerned with adaptation to ICV in the near-term climate (present through mid-century), and with HICC in the long-term future, thus allowing for more robust adaptation strategies to water transfer projects in China.