Urban heat risk depends not only on thermal environment intensity but also on the spatiotemporal distribution and diurnal rhythm of population activity. Conventional assessments based on single-phase land surface temperature or static population data cannot capture intraday exposure variation or differences among functional spaces. Using a selected high-temperature day (13 June 2025) in Zhengzhou, this paper builds a Dynamic Thermal Environment Index (DTEI) fusing a static Landsat-derived thermal baseline with intraday variation from Himawari-9. It couples DTEI with a Population Activity Index (PAI) from hourly Baidu heat-map data to construct a 500 m grid-scale Dynamic Heat Exposure Index (DHEI). DHEI’s diurnal rhythm, spatial pattern, and functional-space differences are analyzed, and a two-factor decomposition identifies the hourly relative contributions of DTEI and PAI. Results show that: (1) DTEI, DHEI, and PAI peak at 13:00, 14:00, and 18:00, with DHEI lagging DTEI by 1 h and leading PAI by 4 h; (2) high-DHEI areas concentrate in the central built-up area with a stable spatial pattern, expanding outward and rising in level over the day; (3) exposure intensity follows commercial and residential > public service > transportation, green/recreational, and employment/industrial zones, with markedly different peak timings between groups; and (4) DHEI variation is driven not by a single factor but by alternating thermal environment and population activity contributions, with the peak formed by their joint effect in early afternoon. Dynamic urban heat exposure is thus a time-varying outcome of thermal environment and population activity coupling, offering a useful reference for time- and zone-specific heat-risk management.