# Optimizing Geoengineering Interventions Using Differentiable Climate Models

arXiv: https://arxiv.org/abs/2609.12528 · HTML: https://mlsanity.com/paper/2609.12528/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 10 of its 43 references that are in the map.
- First posted on arXiv 11 Sep 2026.

## Builds on (most cited first, 10 of 10)

- [Neural General Circulation Models for Weather and Climate](https://mlsanity.com/paper/2311.07222/) (2023)
- [Spherical Fourier Neural Operators: Learning Stable Dynamics on the Sphere](https://mlsanity.com/paper/2306.03838/) (2023)
- [Gradients are Not All You Need](https://mlsanity.com/paper/2111.05803/) (2021)
- [LUCIE: A Lightweight Uncoupled ClImate Emulator with long-term stability and physical consistency for O(1000)-member ensembles](https://mlsanity.com/paper/2405.16297/) (2024)
- [Differentiable Programming for Earth System Modeling](https://mlsanity.com/paper/2208.13825/) (2022)
- [Stratospheric Aerosol Injection as a Deep Reinforcement Learning Problem](https://mlsanity.com/paper/1905.07366/) (2019)
- [Maximizing simulated tropical cyclone intensity with action minimization](https://mlsanity.com/paper/1905.00515/) (2019)
- [Constructing Extreme Heatwave Storylines with Differentiable Climate Models](https://mlsanity.com/paper/2506.10660/) (2025)
- [LUCIE-3D: A three-dimensional climate emulator for forced responses](https://mlsanity.com/paper/2509.02061/) (2025)
- [Optimal scenario design for climate emulation](https://mlsanity.com/paper/2606.19302/) (2026)

## Abstract (arXiv)

The deployment of a geoengineering program to cool Earth's climate may be imminent. It is crucial that tools be developed to ensure that such a program would achieve its objectives while minimizing disruption. Here we exploit recently developed differentiable atmospheric models to demonstrate a novel geoengineering control strategy. In the differentiable primitive-equation atmospheric model JAX-GCM we impose a uniform +4,K ocean warming and ask what pattern of sea-surface temperature cooling -- in five ocean-masked zonal bands of prescribed SST forcings whose amplitudes are free -- returns land near-surface air temperature closest to the model's own unwarmed climatology. This idealized set-up represents a cooling pattern that could be delivered physically either by marine cloud brightening or stratospheric aerosol injection. Gradients through chaotic dynamics decorrelate from the true sensitivity beyond the Lyapunov horizon, so we optimize greedily over segments of 8 to 14 days, following receding-horizon control. The learned strategy removes 92.3 ± 0.4% of the realized land warming across a ten-member ensemble of two-year rollouts, and a three-year run sustains it. If we use the spatial pattern of land temperature as the optimization objective, the distributions of precipitation, evaporation, and specific humidity over land are restored as well, even though they are not included in the objective function. The learned strategy from JAX-GCM replayed in the AI emulators LUCIE and NeuralGCM without re-optimization is successful, suggesting robustness. These promising results demonstrate a strategy for designing optimal climate interventions that can be applied broadly for geoengineering scenarios under consideration.