SDO/EVE Solar Extreme Ultraviolet Irradiance Spectra for Machine Learning
- Provider
- Shah Mohammad Bahauddin (LASP, University of Colorado Boulder); NASA-IBM AI4Science
- License
- CC-BY-4.0 (source)
- Size
- 15 files
- Format
- NetCDF, CSV (in sampled files)
- Stored at
- Hugging Face (https://huggingface.co/datasets/nasa-ibm-ai4science/euv-spectra)
- Access
- Free and public. No account with the data provider is needed.
- DOI
- 10.57967/hf/7280
- Category
- Astronomy & Space
- Source
- https://huggingface.co/datasets/nasa-ibm-ai4science/euv-spectra
About SDO/EVE Solar Extreme Ultraviolet Irradiance Spectra for Machine Learning
Time-aligned extreme ultraviolet (EUV) irradiance spectra from the Extreme Ultraviolet Variability Experiment (EVE) on NASA's Solar Dynamics Observatory (SDO), prepared for image-to-spectrum regression and forecasting. Spectra come from the EVE Level 2b MEGS-A product, adjusted to 1 AU and downsampled from 60-second integrations to a 12-minute cadence. Each record is a 1,343-bin spectrum spanning about 6.5 to 33.3 nm at 0.02 nm resolution, in W/m^2/nm, covering May 2010 to May 2014 in Solar Cycle 24, including quiet-Sun and flare periods; 189,344 instances in total. Uses include predicting EUV spectra from SDO/AIA coronal images, differential emission measure inversion and EUV proxies for ionosphere and thermosphere models. Listing the repository (measured 2026-11-13) shows train.csv (2.99 GB), validation.csv, test.csv, leaky_validation.csv and euv_wavelengths.csv at the root, plus archive/ with 8 files (8.6 GB) including AIA_EVE_dataset_combined.nc and AIA_EVE_temporally_stratified.nc. Splits are by date (train Feb 15 to Dec 31 of each year; validation and test from late January of different years). Open the CSVs with pandas and the netCDF files with xarray. Released under CC-BY-4.0; cite Woods et al. 2012 for the EVE instrument.
How to load this dataset in Python
import pandas as pd
df = pd.read_csv("hf://datasets/nasa-ibm-ai4science/euv-spectra/archive/test_eve_spectra.csv")
df.head()Data structure
File types in sampled files: NetCDF, CSV
Files
Example file: archive/test_eve_spectra.csv
Analyze this dataset without downloading it
On Scigantic, SDO/EVE Solar Extreme Ultraviolet Irradiance Spectra for Machine Learning opens in a hosted JupyterLab notebook with the data already mounted, so you can read files where they are stored instead of copying them first. Each notebook includes an AI assistant that can see the mounted files and write and run analysis code. Open the dataset page to launch a notebook, or browse the full catalog.
This page is also available as Markdown and as JSON.
Related Astronomy & Space datasets
- FMI Kilpisjarvi All-Sky Camera: 557.7 nm Auroral Keograms, Finnish Meteorological Institute, 1997-2001 (CDAWeb on HelioCloud)
- New Horizons SWAP Solar Wind Around Pluto: Outer-Heliosphere Solar Wind Protons and Interstellar Pickup Ions, 2008-2024 (CDAWeb on HelioCloud)
- TRACERS Tandem Reconnection and Cusp Electrodynamics Reconnaissance Satellites: Cusp Ion, Electron and Search-Coil Data (CDAWeb on HelioCloud)
- LOFAR ELAIS-N1 cycle 2 observations on AWS
- Mars-Bench S5Mars: Curiosity Mastcam Terrain Segmentation (9 classes, 1200 x 1200)
- Polar Spacecraft Magnetosphere and Aurora Data: MFE, EFI, HYDRA, TIMAS, TIDE, PWI, VIS and UVI, 1996-2008 (CDAWeb on HelioCloud)
- Mars-Bench ConeQuest Segmentation: Martian Cones in MRO CTX (512 x 512)
- Unblurred Coadds of the Wide-field Infrared Survey Explorer (unWISE)