The ecological environment of resource-based cities is threatened by multiple ecological risks. The main objective of this study is to investigate the dynamics of land use and landscape ecological risk in resource-based cities. Huaibei, a typical resource-based city in China, was taken as a case study. By using the Markov model and landscape metrics, this study first analysed the land use changes in 1989, 2002 and 2016. Then an assessment model was developed to examine the ecological risk caused by the land use changes. Additionally, a spatial autocorrelation analysis was adopted to reflect the spatial heterogeneity of the landscape ecological risk. The results showed that urbanization has a greater impact on land use than the mining activity in Huaibei. The level of landscape ecological risk showed a tendency to initially increase and then decrease. In 1989 and 2002, the extremely high and high-risk regions were concentrated in the northern mining area, but these regions were reduced and scattered throughout the northern and middle areas of Huaibei in 2016. The area of medium-risk regions continuously increased throughout the study period. The spatial autocorrelation of the landscape ecological risk was positive and was weakened in 2016. The results suggest that the pattern of land use and the landscape ecological risk in resource-based cities showed an obvious spatiotemporal variation. Understanding the phase characteristics is necessary for ecological restoration.