- Location: Ireland
- Institution: Maynooth University, University College Dublin, University of Galway
- Status: Active
- Type: Independent
- Theme: Evidence Discovery & Integration
- Timeframe: 2024 - 2029
Share on social
Datasets on climate, biodiversity and water often come from very different sources and don’t connect easily. This means that when researchers try to combine them, the results can be incomplete or skewed, which can lead to misleading conclusions that don’t tell the full story.
Therefore, this project aims to use a form of artificial intelligence that doesn’t just identify patterns, but also tries to understand cause and effect. It can also explore what could have happened under different conditions. This makes it possible to work with messy, mismatched datasets in a more reliable way.
The outcome will be a better way to integrate different datasets and a clearer picture of what’s happening across our climate, ecosystems, and waterways for decision-makers.
Project Goals
-
Adapt and extend a machine learning model (Bayesian Causal Forests and Bayesian Regression Trees) so that it can work with climate data that varies across both location and time.
-
Gather and analyse a wide range of datasets related to climate, biodiversity, and water quality.
-
Build and test causal AI models using real-world data to ensure they perform accurately and reliably.
-
Use the models to explore alternative scenarios, examine the factors that influence environmental outcomes, and assess the potential impact of different decisions.
-
Apply the framework across other research areas within the Climate + Co-Centre, extending its value beyond this project.
-
Work closely with stakeholders to refine the approach and ensure it is practical and useful in real-world settings.
