PiggyCast - Improving Weather Prediction Accuracy through a Stacking-Based Ensemble AI Approach

Accurate weather prediction is fundamental to understanding and adapting to the impacts of climate change. As our climate shifts, the frequency and intensity of extreme weather events are changing, making reliable forecasts more critical than ever. This tutorial introduces PiggyCast, an ensemble machine learning model designed to improve weather prediction accuracy by stacking forecasts from various numerical, AI-based, and hybrid weather prediction models. We demonstrate how a combined approach, using gradient-boosted decision trees (XGBoost), can surpass the predictive performance of individual base models.

The main contributions of this tutorial are:

1. Developing and assessing an ensemble model:** We build and evaluate PiggyCast, a stacking-based ensemble model that leverages forecasts from state-of-the-art weather prediction models (IFS HRES, GraphCast, Pangu Weather, and NeuralGCM) and trains an XGBoost regressor on top of them to produce more accurate predictions of geopotential height at 500 hPa.
2. Investigating feature importance: We use SHAP (SHapley Additive exPlanations) values to analyze the contribution of each base model's forecast and geographic coordinates to PiggyCast's predictions, providing insights into the model's decision-making process.

Author(s):

• Josiah Kimani, African Institute for Mathematical Sciences (AIMS) - South Africa, josiah@aims.ac.za
• Oliver Angélil, Ishango.ai, oliver@ishango.ai
• Chris Toumping, Inshango.ai, chris@ishango.ai
• Steffen Knoblauch, University of Heidelberg, steffen.knoblauch@uni-heidelberg.de

Originally presented at the CCAI Tackling Climate Change with Machine Learning Workshop at NeurIPS 2025.

Access this tutorial

We recommend executing this notebook in a Colab environment to gain access to GPUs and to manage all necessary dependencies.

Estimated time to execute end-to-end: 30 minutes

Contribute to this tutorial

Please refer to these GitHub instructions to open a pull request via the "fork and pull request" workflow.

Pull requests will be reviewed by members of the Climate Change AI Tutorials team for relevance, accuracy, and conciseness.

Climate Change AI Tutorials

Check out the tutorials page on our website for a full list of tutorials demonstrating how AI can be used to tackle problems related to climate change.

License

Usage of this tutorial is subject to the MIT License.

Cite

Plain Text

Kimani, J., Angélil, O., Toumping, C., and Knoblauch, S. (2025). PiggyCast - Improving Weather Prediction Accuracy through a Stacking-Based Ensemble AI Approach [Tutorial]. In Conference on Neural Information Processing Systems. Climate Change AI.

BibTeX

@misc{jkimani2025piggycast,
  title={PiggyCast - Improving Weather Prediction Accuracy through a Stacking-Based Ensemble AI Approach
},
  author={Kimani, Josiah and Angélil, Oliver and Toumping, Chris and Knoblauch, Steffen},
  year={2025},
  organization={Climate Change AI},
  type={Tutorial},
  booktitle={Conference on Neural Information Processing Systems},
  howpublished={\url{https://github.com/climatechange-ai-tutorials/piggy-cast}}
}