Abstract: Weather radar serves as an extremely effective tool for monitoring and early warning of severe weather events. Based on precipitation data, Doppler weather radar base data, and rainstorm-flood disaster records in Hohhot, Inner Mongolia Autonomous Region from 2005 to 2022, characteristic parameters of radar echo, such as intensity, area, and gradient at the 1 km constant altitude plan position indicator (CAPPI) level, were selected. According to the scale differences of flood-triggering rainstorm radar echoes, a method for nowcasting such rainstorms using nested radar echo regions (combining "large area" and "small area") was developed. An indicator model for warning against flood-triggering rainstorms based on nested radar echo regions was established. The warning effectiveness of this model was objectively and automatically evaluated using the flood-triggering rainstorm event that occurred in Inner Mongolia Autonomous Region on August 8–9, 2024. The results indicate that: (1) The optimal "small area" in the nested radar echo region warning model is 289 km², with a corresponding side length of 17 km. Based on the characteristic values of radar parameters from historical cases of flood-inducing rainstorms, the thresholds for grading flood warning levels were determined. (2) Percentile method was applied to classify the intensity of convective cells identified within the optimal area, establishing the range of the area parameter S₄₀ (area with strong precipitation weight parameter greater than 40 dBz) for the "small area" in the nested region: blue warning (60 km²<S₄₀≤120 km²), yellow warning (120 km² < S₄₀≤180 km²), orange warning (180 km²< S₄₀≤210 km²), and red warning (S₄₀>210 km²). (3) Case validation shows that the nested radar echo region indicator model can effectively enhance the early warning capability for flood-triggering rainstorms. By using the "small area" radar indicators, this model can precisely locate the flood risk points and facilitate the issuance of progressive heavy rainfall warnings based on echo evolution. This contributes to improving the targeted issuance of timely, location-specific, and quantitative early warnings for severe precipitation and other disastrous weather events using weather radar.