The approach of FutureWater and Galayr is designed to be both scientifically rigorous and contextually grounded, ensuring that the developed drought model is locally relevant, sustainable, and fully integrated into existing national systems such as those of SODMA and NADFOR. The model will merge top-down (data-driven, machine learning-enabled) and bottom-up (stakeholder-informed) approaches, combining satellite data, climate indices, and indigenous knowledge to co-develop impact-based forecasts and consensus-based triggers for anticipatory action.
For the development of the drought forecasting model and the knowledge transfer we will focus on the following pillars:
- A phased work plan that spans institutional capacity assessments, model development, validation, stakeholder consultations, and hands-on capacity building.
- Application of state-of-the-art forecasting models ranging from ARIMA and regression to more advanced machine learning techniques, while maintaining focus on usability and institutional adoption.
- A strong emphasis on knowledge transfer, including training programs and the establishment of a collaborative knowledge-sharing platform using the FutureWater Academy platform
- A robust risk management plan, including mitigation strategies for data scarcity, stakeholder disengagement, and institutional turnover.