Increases in the frequency and intensity of heat waves have serious impacts on human health, agriculture, energy and infrastructure. Here we use three simple metrics including the number of heat wave days, the length of heat wave season, and the annual hottest day temperature to characterize future changes in heat wave severity in China, based on large ensemble simulations conducted with the Canadian Earth System Model Version 2 (CanESM2) in the context of emergency preparedness. A heat wave day is defined as a day with daily maximum temperature reaching heat alert level (35°C). We find that global warming is associated with more severe heat waves including more heat wave days, longer heat wave season and higher hottest day temperature, and expansion of regions impacted by heat waves. While the increase in the magnitude of extremes in heat wave metrics with global warming level is close to linear, the increase in the frequency of extremes is much faster. For example, the historically hottest summer in 2013 in Eastern China, which occurs about one in 5 years in the 2013 climate, is projected to become more frequent than one in 2 years under 1.5°C global warming and almost every year would be worse than 2013 under 2°C warming. Additionally, the increase in the frequency of the extreme events is larger for rarer extremes. The frequencies for once-in-5-year, once-in-10-year, and once-in-50-year events increase by 2.5, 3.5, and 5.5 times under 1.5°C global warming, respectively.Plain Language Summary Heat waves have serious impacts on human health, agriculture, energy, and infrastructure. Though a few studies have investigated the future changes in heat waves in China, the heat waves defined in those studies are seldom of direct relevance to emergency preparedness. Here we study changes in heat wave days that are defined as daily maximum temperature above 35°C, a threshold for issuing a heat alert according to China's national Standards. We examine three simple metrics including the annual number of heat wave days, the length of heat wave season, and the annual hottest day temperature based on large ensemble simulations of CanESM2. We find that global warming is associated with more severe heat waves including more heat wave days, longer heat wave season, higher hottest day temperature, and expansion of regions impacted by heat waves. The increase in the heat wave metrics with global warming level is close to linear, while the increase in the frequency of extremes in these metrics is much faster. For example, the once-in-5-year event in the current climate, with a magnitude of the historically hottest summer in 2013 in Eastern China, is projected to become more frequent, to become a once-in-2-year event under 1.5°C global warming and every year event under 2°C global warming.