Search for a command to run...
Abstract Flood forecasting is essential for disaster preparedness and risk reduction, particularly in data-scarce regions where purely data-driven models are often infeasible. This scoping review synthesised 94 peer-reviewed studies published between 1994 and 2024, applying the PRISMA framework to trace the evolution of deterministic hydrologic–hydraulic model coupling for flood forecasting. The review shows that HEC-HMS is the most widely used hydrologic model (33%), while HEC-RAS dominates hydraulic applications (45%). Their combination (HEC-HMS + HEC-RAS) appeared in 23 studies and consistently achieved strong predictive performance with coefficients of determination above 0.98 and inundation mapping accuracies of 80–94%. Other effective couplings, such as SWAT–HEC-RAS and CREST–ANUGA, were less frequently applied but demonstrated context-specific advantages. Regional disparities were pronounced, where 78% of studies were concentrated in high-income nations across North America, Europe, and Asia, with Africa contributing fewer than 5%. This imbalance reflects structural barriers such as sparse hydrometric networks, limited access to high-resolution data, and inadequate computational and technical capacity. Case studies from Southern Africa, including the application of HEC-HMS and HEC-RAS in flood-prone catchments, illustrated that deterministic coupling methods can be adapted to resource-limited environments and still deliver actionable forecasts. The review highlights that the value of model coupling lies not only in predictive accuracy but also in its feasibility and transferability across diverse hydrological contexts. Lessons drawn from global and regional applications suggest that deterministic model coupling remains a practical pathway for enhancing flood early warning systems in data-scarce regions if calibration, validation, and local constraints are systematically addressed.