Urban heat islands (UHI) exacerbates the heat-related risk associated with global warming, increasing morbidity and mortality of urban residents. While the impacts of the spatial pattern of urban greenspace (UG) and its change on urban heat have been widely examined, there is less understanding of the aggregate effect of the change of UG- considering the loss and gain of UG simultaneously - on urban temperature. This study aims to fill this gap by using Beijing, China as a case study. Using a newly developed index - dynamic index of UG (UGDI) that simultaneously measures the loss and gain of UG in a certain unit of analysis, we investigated how changes in UG affect the daytime and nighttime land surface temperature (LST). We found: (1) A substantial proportion (49.90%) of grids with increased UG cover had increased LST during the daytime, with a magnitude ranging from 0.02 to 1.82 ℃, indicating that the increase in UG does not always result in reduction of LST. (2) UGDI had a significantly positive correlation with LST change, suggesting that increase in UG does not necessarily result in decrease of LST, which can be affected by the degree of dynamics of UG. (3) The evapotranspiration (ET) rate of vegetation for lost greenspace was higher than that of new greenspace, indicating that adding the same amount of UG might not able to provide the same amount of cooling effects provided by lost ones. Results can enhance our understanding on how (landscape) process affects ecological effect. Future research and practical management strategies shall move beyond net increase of UG and focus more on its change process. This finding provides new evidence for explaining the effect of the change of UG on LST, and offers new insights for planning and managing urban natural resource to enhance resilience of cities to climate warming.