Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Scalable Machine Learning Analysis of Parker Solar Probe Solar Wind Data

Daniela Martin, Connor O'Brien, Valmir P Moraes Filho, Jinsu Hong, Jasmine R. Kobayashi, Evangelia Samara, Joseph Gallego

TL;DR

The paper tackles scalable analysis of Parker Solar Probe solar wind data, which exceeds 150 GB, by introducing a distributed processing framework using Dask and a quantum-inspired Kernel Density Matrices (KDM) method for univariate and joint density estimation and anomaly detection. The approach enables sampling and inference on large, high-volume time series data and provides a pathway toward reproducible analyses. Applied to SWEAP measurements across 0–1 AU, the method reveals physically meaningful trends such as solar wind speed increasing with distance and proton density decreasing, along with an inverse speed–density relationship tied to solar wind structures. Hyperparameter choices, preprocessing steps, and open-source tooling are provided to support reproducibility and adaptability to other large-scale in situ datasets and space weather forecasting.

Abstract

We present a scalable machine learning framework for analyzing Parker Solar Probe (PSP) solar wind data using distributed processing and the quantum-inspired Kernel Density Matrices (KDM) method. The PSP dataset (2018--2024) exceeds 150 GB, challenging conventional analysis approaches. Our framework leverages Dask for large-scale statistical computations and KDM to estimate univariate and bivariate distributions of key solar wind parameters, including solar wind speed, proton density, and proton thermal speed, as well as anomaly thresholds for each parameter. We reveal characteristic trends in the inner heliosphere, including increasing solar wind speed with distance from the Sun, decreasing proton density, and the inverse relationship between speed and density. Solar wind structures play a critical role in enhancing and mediating extreme space weather phenomena and can trigger geomagnetic storms; our analyses provide quantitative insights into these processes. This approach offers a tractable, interpretable, and distributed methodology for exploring complex physical datasets and facilitates reproducible analysis of large-scale in situ measurements. Processed data products and analysis tools are made publicly available to advance future studies of solar wind dynamics and space weather forecasting. The code and configuration files used in this study are publicly available to support reproducibility.

Scalable Machine Learning Analysis of Parker Solar Probe Solar Wind Data

TL;DR

The paper tackles scalable analysis of Parker Solar Probe solar wind data, which exceeds 150 GB, by introducing a distributed processing framework using Dask and a quantum-inspired Kernel Density Matrices (KDM) method for univariate and joint density estimation and anomaly detection. The approach enables sampling and inference on large, high-volume time series data and provides a pathway toward reproducible analyses. Applied to SWEAP measurements across 0–1 AU, the method reveals physically meaningful trends such as solar wind speed increasing with distance and proton density decreasing, along with an inverse speed–density relationship tied to solar wind structures. Hyperparameter choices, preprocessing steps, and open-source tooling are provided to support reproducibility and adaptability to other large-scale in situ datasets and space weather forecasting.

Abstract

We present a scalable machine learning framework for analyzing Parker Solar Probe (PSP) solar wind data using distributed processing and the quantum-inspired Kernel Density Matrices (KDM) method. The PSP dataset (2018--2024) exceeds 150 GB, challenging conventional analysis approaches. Our framework leverages Dask for large-scale statistical computations and KDM to estimate univariate and bivariate distributions of key solar wind parameters, including solar wind speed, proton density, and proton thermal speed, as well as anomaly thresholds for each parameter. We reveal characteristic trends in the inner heliosphere, including increasing solar wind speed with distance from the Sun, decreasing proton density, and the inverse relationship between speed and density. Solar wind structures play a critical role in enhancing and mediating extreme space weather phenomena and can trigger geomagnetic storms; our analyses provide quantitative insights into these processes. This approach offers a tractable, interpretable, and distributed methodology for exploring complex physical datasets and facilitates reproducible analysis of large-scale in situ measurements. Processed data products and analysis tools are made publicly available to advance future studies of solar wind dynamics and space weather forecasting. The code and configuration files used in this study are publicly available to support reproducibility.
Paper Structure (5 sections, 2 equations, 3 figures, 1 table)

This paper contains 5 sections, 2 equations, 3 figures, 1 table.

Table of Contents

  1. Introduction
  2. Background
  3. Experimental Setup
  4. Results and Discussion
  5. Conclusion

Figures (3)

  • Figure 1: Boxplots of three PSP solar wind parameters from 2018 to 2024: (left) solar wind speed [km/s], (middle) proton density [cm-3], and (right) proton thermal speed [eV].
  • Figure 2: Univariate distributions of PSP's solar wind speed. Left: Cumulative Distribution Function (CDF). Right: Probability Density Function (PDF). Each curve corresponds to a different heliocentric distance (0-0.6 AU), as indicated in the legend. Distances beyond 0.6 AU are omitted due to sparse measurements and increased noise.
  • Figure 3: Bivariate distributions of PSP's solar wind speed versus proton density at different heliocentric distances (AU). Left panel: 0.3-0.4 AU. Right panel: 0.4-0.5 AU.