A comprehensive assessment of grain supply, demand, and ecosystem service flows is essential for identifying grain movement pathways, ensuring regional grain security, and guiding sustainable management strategies. However, current studies primarily focus on short-term grain provision services while neglecting the spatiotemporal variations in grain flows across different scales. This gap limits the identification of dynamic matching relationships and the formulation of optimization strategies for balancing grain flows. This study examined the spatiotemporal evolution of grain supply and demand in the Beijing-Tianjin-Hebei (BTH) region from 1980 to 2020. Using the Enhanced Two-Step Floating Catchment Area method, the grain provision ecosystem service flows were quantified, the changes in supply–demand matching under different grain flow scenarios were analyzed and the optimal distance threshold for grain flows was investigated. The results revealed that grain production follows a spatial distribution pattern characterized by high levels in the southeast and low levels in the northwest. A significant mismatch exists between supply and demand, and it shows a scale effect. Deficit areas are mainly concentrated in the northwest, while surplus areas are mainly located in the central and southern regions. As the spatial scale increases, the ecosystem service supply–demand ratio (SDR) classification becomes more clustered, while it exhibits greater spatial SDR heterogeneity at smaller scales. This study examined two distinct scenarios of grain provision ecosystem service flow dynamics based on 100 and 200 km distance thresholds. The flow increased significantly, from 2.17 to 11.81 million tons in the first scenario and from 2.41 to 12.37 million tons in the second scenario over nearly 40 years, forming a spatial movement pattern from the central and southern regions to the surrounding areas. Large flows were mainly concentrated in the interior of urban centers, with significant outflows between cities such as Baoding, Shijiazhuang, Xingtai, and Hengshui. At the county scale, supply–demand matching patterns remained consistent between the grain flows in the two scenarios. Notably, incorporating grain flow dynamics significantly reduced the number of grain-deficit areas compared to scenarios without grain flow. In 2020, grain-deficit counties decreased by 28.79 and 37.88%, and cities by 12.50 and 25.0% under the two scenarios, respectively. Furthermore, the distance threshold for achieving optimal supply and demand matching at the county scale was longer than at the city scale in both grain flow scenarios. This study provides valuable insights into the dynamic relationships and heterogeneous patterns of grain matching, and expands the research perspective on grain and ecosystem service flows across various spatiotemporal scales.

land-use type  /  soil aggregate  /  density fractions  /  carbon transfer  /  Tibetan Plateau