CAM-NET: An AI Model for Whole Atmosphere with Thermosphere and Ionosphere Extension

TL;DR

CAM-NET is a high-accuracy, computationally efficient AI model that predicts the entire Earth's atmosphere from surface to ionosphere, enabling rapid simulations and flexible tracer modeling.

Contribution

It introduces a novel spherical Fourier neural operator-based architecture with a modular design for efficient atmospheric and tracer prediction, outperforming traditional models in speed.

Findings

Achieves over 1000x inference speedup compared to WACCM-X.

Maintains accuracy comparable to detailed climate models.

Successfully predicts key atmospheric parameters including winds and temperature.

Abstract

We present Compressible Atmospheric Model-Network (CAM-NET), an AI model designed to predict neutral atmospheric variables from the Earth's surface to the ionosphere with high accuracy and computational efficiency. Accurate modeling of the entire atmosphere is critical for understanding the upward propagation of gravity waves, which influence upper-atmospheric dynamics and coupling across atmospheric layers. CAM-NET leverages the Spherical Fourier Neural Operator (SFNO) to capture global-scale atmospheric dynamics while preserving the Earth's spherical structure. Trained on a decade of datasets from the Whole Atmosphere Community Climate Model with thermosphere and ionosphere eXtension (WACCM-X), CAM-NET demonstrates accuracy comparable to WACCM-X while achieving a speedup of over 1000x in inference time, can provide one year simulation within a few minutes once trained. The model effectively predicts key atmospheric parameters, including zonal and meridional winds, temperature, and time rate of pressure. Inspired by traditional modeling approaches that use external couplers to simulate tracer transport, CAM-NET introduces a modular architecture that explicitly separates tracer prediction from core dynamics. The core backbone of CAM-NET focuses on forecasting primary physical variables (e.g., temperature, wind velocity), while tracer variables are predicted through a lightweight, fine-tuned model. This design allows for efficient adaptation to specific tracer scenarios with minimal computational cost, avoiding the need to retrain the entire model. We have validated this approach on the $O^2$ tracer, demonstrating strong performance and generalization capabilities.