Bridging Idealized and Operational Models: An Explainable AI Framework for Earth System Emulators

This repository contains the implementation and experiments for the paper "Bridging Idealized and Operational Models: An Explainable AI Framework for Earth System Emulators".
The notebook provides end-to-end code for building and training an LSTM-based framework that connects idealized and operational model data for Earth system emulation.

Contents

-Bridging_model_main.ipynb`: Main Jupyter notebook with all experiments and analyses.

Requirements

The notebook uses Python 3 with the following key dependencies:

numpy
pandas
matplotlib
pytorch
scikit-learn

You can install them with:

pip install -r requirements.txt

Usage

1. Clone this repository:

   git clone https://github.com/pbehnoud/bridging_model.git
   cd bridging_model

2. Open the notebook in Jupyter:

   jupyter notebook Bridging_model_main.ipynb

3. Run all cells to reproduce the results.

Data

The notebook assumes access to observational and operational model data as described in the paper.
Processed datasets and code are available at https://github.com/pbehnoud/bridging_model.

Citation

If you use this code or data, please cite:

<!-- @article{behnoud2025bridging,
  title={Bridging Idealized and Operational Models: An Explainable AI Framework for Earth System Emulators},
  author={},
  journal={Nature Communications},
  year={2025}
} -->