Abstract: Flood propagation is pivotal to flood control security, affecting channel erosion-deposition processes, fluvial regime evolution, and river-lake interactions. Analyzing the composition and downstream variability of typical floods in the frequency domain has significant implications for uncovering the underlying mechanisms of flood propagation and formulating science-based flood mitigation strategies. Utilizing daily discharge data from hydrological stations between Yichang and Datong, this study applied signal processing-based time-frequency analysis to investigate the spectral characteristics of regional floods （2016, 2017） and basin floods （1998, 2020） to quantify the influence of key factors on flood signal intensity across river reaches. Key findings include:① Tributary/lake inflows, three-outlet bifurcation flows, and local catchment contributions predominantly modulated low-frequency （long-period） dominant signals, whereas short-term events, such as rainfall, primarily affected high-frequency （short-period） secondary signals. ② For both flood types, low-frequency signal intensity amplified downstream by 176.1%—240.3%, while high-frequency component attenuated to 45.5%—74.5%. ③ The Dongting Lake inflow exerted the strongest control on the dominant signal intensity, followed by Poyang Lake, Hanjiang River, and Qingjiang River, with its impact more pronounced during regional floods. This study provides novel insights into flood discharge dynamics and advances the mechanistic understanding of flood-wave propagation.