As environmental governance undergoes a digital transformation, there is a decisive shift from static reporting toward dynamic ecosystems that facilitate real-time assessment and causal attribution. However, translating scientific insight into political action remains challenging, requiring raw data to be transformed into robust, legally defensible evidence. Recent advances in data science are expanding this capacity and offering more precise tools for impact assessment, explicit quantification of uncertainty and advances in explainability. At the same time, significant hurdles remain: The "black-box" nature of certain predictive models, the difficulty of communicating scientific uncertainty to non-experts, as well as the complexities of causal attribution in open systems must be addressed to ensure these methods are fit for purpose and lead to resilient governance.

This evolution enables a transition from reactive to proactive governance, providing the empirical rigor necessary to design and refine environmental interventions, such as water quality regulations and low-emission zones across spatio-temporal scales.

The session aims to provide a venue to present the latest progress in bridging the gap between environmental data science and policy implementation. We welcome abstracts focused on, but not limited to:

- Cross-sectoral case studies: Implementation of evidence-based governance in air quality, biodiversity, water management, or climate adaptation.
- Uncertainty quantification: Probabilistic modelling and frameworks for communicating scientific uncertainty to decision-makers.
- Interpretability and Explainability: Methods to ensure that data-driven models are transparent and legible for administrative and legal use.
- Causal inference: Causal impact assessment of policy interventions, counterfactual analysis, and the use of quasi-experimental methods.
- Evidence-based monitoring: Designing data collection networks and sensors specifically to support policy evaluation and KPIs.
- Adaptive governance: The use of real-time data streams for iterative policy adjustment and "course-correction" (feedback loops).
- Decision-support systems: The development of tools and metrics that translate scientific observations into actionable policy options.
- Institutional frameworks: Strategies for integrating data science workflows into government agency structures and legislative processes.