Dr. Amr Hamada

Dr. Amr Hamada

Solar Physicist — NSO/NISP

Datasets

Below is a curated list of datasets that I have developed or contributed to, covering homogenized EUV solar observations, coronal hole detection, helioseismic far-side active region mapping, and machine-learning–ready resources.

1) Homogenized Extreme Ultraviolet (EUV) Synoptic Maps Dataset from SDO/AIA and SOHO/EIT

Authors: Hamada, A., et al.
Date of Release: 2020
Persistent Identifier / URL:
🔗 https://satdat.oulu.fi/solar_data/

A long-term homogenized EUV synoptic dataset combining SOHO/EIT and SDO/AIA observations, suitable for coronal structure studies, machine learning, and space weather research.

2) Coronal Hole (CH) Synoptic Maps Dataset

Authors: Hamada, A., et al.
Date of Release: 2020
Persistent Identifier / URL:
🔗 https://satdat.oulu.fi/solar_data/

A fully processed dataset of automatically identified coronal holes, including consistent boundaries, extracted features, and synoptic map representations.

3) ML-Ready Dataset for Far-Side Active Regions (Helioseismic + EUV)

Authors: Hamada, A., et al.
Date of Release: 2024
Persistent Identifier / URL:
🔗 https://dataverse.harvard.edu/dataset.xhtml?persistentId=doi:10.7910/DVN/8ZKY8Z
DOI: 10.7910/DVN/8ZKY8Z

A machine-learning–ready dataset combining far-side helioseismic phase-shift maps with EUV-derived active region masks, enabling research on far-side solar imaging and prediction pipelines.

4) FArSide Trained Active Region Recognition (FASTARR) Dataset

Authors: Hamada, A., et al.
Date of Release: 2025
Persistent Identifier / URL:
🔗 https://dataverse.harvard.edu/dataset.xhtml?persistentId=doi:10.7910/DVN/HSRJM4
DOI: 10.7910/DVN/HSRJM4

The complete FASTARR dataset used for training machine-learning models for far-side active region detection from helioseismic phase maps, including masks derived from GONG and STEREO/EUVI.

5) Deep Learning–Based Prediction of High-Speed Solar Wind Streams Dataset

Authors: Abraham-Alowonle, Joseph-Judah (Rights Holder), Hamada, Amr (Supervisor), et al.
Date of Release: 2025
Persistent Identifier / URL:
🔗 https://zenodo.org/records/14849304
DOI: 10.5281/zenodo.14849303

Dataset accompanying the deep learning model for predicting high-speed solar wind streams from coronal hole evolution, including time-series inputs, labels, and trained model outputs.