Diffusion Modeling of the Solar Corona

Overview

Development stage

In development or acquisitionSource: a) Pre-deployment – The use case is in a development or acquisition status.

High-impact designation

Not high-impactSource: c) Not high-impact

Impact justification

Not reported

Start date

Not reported

Withheld from public reporting

NoSource: a) No

Mission

Topic area

Science

Operational functions

Image and video analysis, Forecasting and prediction Derived by keyword rules; see methodology

Problem the AI is intended to solve

Diffusion models such as DeepMind’s GenCast have demonstrated powerful performance in terrestrial weather forecasting, achieving results on-par and surpassing leading medium-range numerical weather simulations. We present our initial results to train a Denoising Diffusion Probabilistic Model (DDPM) of the solar corona, aimed as a first step at generating synthetic solar magnetic fields based on conditioning inputs. Conditioning inputs include multi-spectral imagery, magnetograms, and other measurements commonly used to frame coronal inverse problems. Our initial experiment targets the generation of synthetic global magnetic field configurations of the solar coronal basedon conditioning with the solar cycle phase. The model is trained on 10 years of WSA potential field source surface(PFSS) model runs, augmented with 15° to 345° azimuthal rotations to increase data diversity. Training is conducted inthe spherical harmonics domain, leveraging concepts from Fourier Neural Operators (FNOs) and Spherical Fourier Neural Operators (SFNOs). A physics-informed loss function, built on a differentiable spherical harmonic expansion, isused to maximize generation of realistic 3D magnetic potentials.

Expected benefits

First step towards probabilistic modeling of the solar corona

System outputs

3D Magnetic Field Structures in the Solar Corona

Technology

AI classification

Generative AISource: Generative AI: AI that generates new or synthetic content (e.g., images, videos, audio, text, code).

System name(s)

Not reported

Custom-developed code

Not reported

Public source code

Not reported

Sourcing

How it was built

Not reported

Vendor (as reported)

Not reported

Vendors (standardized)

None on the standard list

A vendor is the supplier named by the agency. It does not identify the underlying model or AI technology.

Data and privacy

Involves PII

Not reported

Privacy Impact Assessment

Not reported

Authorization to Operate

Not reported

Demographic features

Not reported

Training and evaluation data

Not reported

Federal Data Catalog

Not reported

Governance

0 of 8 minimum-practice questions answered. A blank answer means the agency reported nothing; it does not mean the practice is absent.

Potential impacts and how they were identified

Not reported