Background: Short-term heat exposure might induce stroke morbidity and mortality, and there were several studies explored the possible vulnerable populations. At present, the research on the modification effect of intra-urban landscape characteristics on the association between heat and stroke morbidity is limited, especially in China. Methods: We collected data on 22,424 first-ever strokes between 2010 and 2016 in Shenzhen, from June to August of each year. We adopted the case-only study combined with logistic regression models to examine the modification effects of 5 urban landscape characteristics. We studied the characteristics of relevant vulnerable populations through stratification analyses. Results: High values (refer to the median values) of nighttime land surface temperature (LST) and the proportion of impervious surface may aggravate the harmful effects of heat on stroke morbidity, with the OR values (95% CI) of 1.205 (1.053, 1.357) and 1.115 (1.010, 1.220); while, high values of NDVI and the proportion of water bodies may alleviate the harmful effects of heat, with the OR values (95% CI) of 0.772 (0.699, 0.845) and 0.821 (0.741, 0.901). The OR value of daytime LST was 1.004 (0.861, 1.147). Statistically significant modification effects were located in the population without Shenzhen's household registration; as for nighttime LST, statistically significant modification effects were located in females and the elderly. Conclusions: High values of nighttime LST and the proportion of impervious surface might aggravate the harmful effects of heat on stroke morbidity, while high green space and water cover might alleviate its effects. Immigrants were the related vulnerable populations. The government should take measures to cope with climate warming and pay attention to the health effects of heat on immigrants. (c) 2021 Published by Elsevier B.V.