Event-dependent flash-flood regulation by small reservoirs in the Qianhe River Basin: A BTOP-based scenario analysis

Abstract Spatially dispersed small reservoirs can alter flash-flood routing, but their event-scale effects are difficult to constrain without operational records. We coupled tributary storage-release nodes to BTOP routing in the Qianhe River Basin, representing 15 small reservoirs as nine equivalent nodes. The underlying BTOP model was calibrated with 12 hourly storm-flood events and independently evaluated against six further events. Validation Nash-Sutcliffe efficiency values ranged from 0.74 to 0.87, with a mean of 0.81. Five prescribed reservoir-state scenarios varied initial storage and the spill threshold. Under the reference scenario, simulated outlet peak attenuation ranged from 0 to 2.04% and averaged 1.27%; event-volume reduction averaged 0.14%, and peak timing changed by at most 1 h. Mean event-level peak-attenuation ranges were 0.91 and 0.96 percentage points over the tested initial-storage and spill-threshold intervals, respectively. These ranges describe conditional scenario responses and do not rank intrinsic parameter importance. The contribution is a tributary-based implementation for examining small regulation signals under limited operating data. The results are restricted to the six events and prescribed states; they neither reconstruct historical reservoir operation nor establish an advantage over an outlet-aggregated reservoir representation.