Abstract: Temperature and temperature changes are important environmental risk factors that affect human health, especially mortality. This study investigates impacts of temperature and temperature changes on mortality of residents in Qinhuangdao based on daily meteorological data and resident mortality data for all districts and counties in Qinhuangdao from 2014 to 2020. The generalized additive model (GAM) and the distributed lag nonlinear model (DLNM) are used to explore the impacts of temperature and temperature changes on the number of non-accidental mortality and cardiovascular and respiratory mortality from three perspectives of temperature, diurnal temperature change (DTR) and temperature change between neighboring days (TCN). The modeling study is stratified by sex and age, and the relative risk (RR) is used to quantify the mortality risk of exposure to specific temperature changes. In addition, a non-parametric bivariate response surface model is used to explore the interaction between temperature and temperature change. The results are as follows: (1) The annual peak of non-accidental mortality, cardiovascular mortality and respiratory mortality in Qinhuangdao occurs in January, the coldest month of the year. The impact of temperature on the three types of deaths is mainly dominated by lagged cold effects, while high temperature has immediate effects. (2) The overall exposure response curve of DTR and non-accidental and cardiovascular mortality shows a U-shaped distribution, and high DTR has significant risk effects on the above two types of deaths. Among them, cardiovascular mortality is most affected, and the cumulative 3 d relative risk of extremely large DTR (19℃) is 1.27 (95%CI:1.15—1.4), while the risk effect of respiratory mortality cannot pass the significance test. (3) The exposure curve of the overall effect of TCN on non-accidental mortality and circulatory mortality shows a nonlinear increasing trend, and positive TCN has significant risk effects. (4) In terms of sex and age grouping, famale are more sensitive to temperature changes, while temperature changes have a lagged effect on male, and the elderly are more susceptible to temperature changes. (5) The synergistic effect of low temperature and temperature change exacerbates the mortality risk. The effect of temperature and its changes on the mortality of the three types is mainly the cold effect in Qinhuangdao. Particularly, the combination of low temperature background in winter and large temperature changes has the greatest impact on mortality risk of local elderly residents, and they should be protected in a timely manner.