Resources
AGCI makes publicly accessible thousands of video presentations, research publications, and other resources from our workshops and projects. Use the search and filter options below to explore the resource library.



Accelerating Actionable Climate Information Through Machine Learning
To improve societal and environmental wellbeing, decision makers need climate information that is actionable. Actionable climate information needs to be timely; sufficiently detailed as to describe the region and processes of interest; trustworthy and truthful in acknowledging uncertainties; and policy- or decision-relevant. We have already begun to observe the transformative power of machine learning (ML) to produce actionable knowledge in weather forecasting (where we can conveniently check whether our predictions are correct). How could machine learning do the same, and more, in the context of climate change? This talk will introduce how we make climate predictions, and then describe ways in which machine learning can enable us to make more trustworthy and more detailed climate predictions, at a faster pace. We'll also explore how machine learning can provide a bridge between climate.

Causality and Interpretability
To improve societal and environmental wellbeing, decision makers need climate information that is actionable. Actionable climate information needs to be timely; sufficiently detailed as to describe the region and processes of interest; trustworthy and truthful in acknowledging uncertainties; and policy- or decision-relevant. We have already begun to observe the transformative power of machine learning (ML) to produce actionable knowledge in weather forecasting (where we can conveniently check whether our predictions are correct). How could machine learning do the same, and more, in the context of climate change? This talk will introduce how we make climate predictions, and then describe ways in which machine learning can enable us to make more trustworthy and more detailed climate predictions, at a faster pace. We'll also explore how machine learning can provide a bridge between climate.

Causal Inference and Discovery for Climate Science
To improve societal and environmental wellbeing, decision makers need climate information that is actionable. Actionable climate information needs to be timely; sufficiently detailed as to describe the region and processes of interest; trustworthy and truthful in acknowledging uncertainties; and policy- or decision-relevant. We have already begun to observe the transformative power of machine learning (ML) to produce actionable knowledge in weather forecasting (where we can conveniently check whether our predictions are correct). How could machine learning do the same, and more, in the context of climate change? This talk will introduce how we make climate predictions, and then describe ways in which machine learning can enable us to make more trustworthy and more detailed climate predictions, at a faster pace. We'll also explore how machine learning can provide a bridge between climate.