Integrating global and local data for flood adaptation in IDP camps near Dikwa, Nigeria

Northeastern Nigeria faces a compounding crisis driven by conflict-induced displacement and intensifying climate hazards. In Dikwa, Borno State, Internally Displaced Persons (IDPs) occupy flood-prone sites with inadequate infrastructure, exacerbating their vulnerability. Humanitarian operations in these data-scarce settings often lack the detailed flood risk information necessary for effective mitigation. This study presents an integrative flood modelling framework that couples global datasets with participatory local data to assess flood risks and evaluate adaptation strategies across 17 IDP camps. We developed a coupled hydrological-hydrodynamic model (Wflow and Delft3D FM) using global open-access data as a baseline. To address the limitations of global models, we integrated local meteorological records and participatory data collected via KoBoToolbox, including drainage characteristics and historical flood marks. Results indicate that relying solely on global datasets underestimated flood hazards and diverged from local observations. Integrating local data significantly improved model validity. We utilized the validated model to assess shelter-level exposure under various return periods (T2 to T100) and simulated the efficacy of a conceptual drainage network. The proposed interventions reduced the total population at risk by approximately 50% across all return periods. However, the analysis revealed trade-offs, where drainage diverted water effectively from major settlements but increased risk in specific localized areas. This research demonstrates that while global data enables initial assessments, local verification is essential for operational relevance. The findings provide a reproducible workflow for quantifying flood hazards and designing adaptation measures in complex humanitarian emergencies.