Synoptic Drought Monitoring: A Multivariable Framework for Water Cycle and Impact Integration

Objective

Drought is a complex and globally significant hazard that generates cascading impacts across ecosystems, infrastructure, and society. Early-warning forecasting systems can substantially mitigate these impacts. However, most research conceptualizes drought as a set of isolated events, without considering its interconnectedness within the water cycle or its associated impacts. Advancing monitoring and forecasting requires a systems-based framework that quantifies and visualizes all water-cycle components using a unified indicator set. The SyndroDry project addresses this need by developing a synoptic drought monitoring and forecasting framework based on systems theory, establishing a new theoretical model for drought assessment. The primary objective is to provide a visualized, multivariate framework for operational monitoring and forecasting that improves early warning capabilities. The SyndroDry approach incorporates stakeholder knowledge and scientific evidence to define impact-based thresholds, embedding these thresholds within a multivariate representation of the water cycle. The project will evaluate advanced predictive methods, including Long Short-Term Memory (LSTM), GLM-Boost, Tree-based, and hybrid machine-learning models, to identify the most robust strategies for synoptic drought forecasting. Ultimately, the project will deliver an operational tool that provides actionable early warnings to water-resource managers, supporting timely decision-making and reducing economic and ecosystem losses from drought.