Solar Physicist • Scientific Programmer • Numerical Modeler • Educator
🌐 Live Website: https://sr-dash.github.io/
This repository hosts my personal academic website, developed to showcase research, publications, scientific software, teaching activities, and educational resources in solar physics and heliophysics.
The site serves as a central hub for my work involving solar observations, spectropolarimetry, coronal magnetic field diagnostics, numerical simulations, data analysis pipelines, and scientific visualization.
- Solar Physics
- Heliophysics
- Coronal Magnetism
- Spectropolarimetry
- DKIST & Cryo-NIRSP Observations
- Solar Eclipse Science
- Magnetohydrodynamic (MHD) Simulations
- Forward Modeling
- Inverse Problems
- Scientific Computing
- Data Visualization
- Space Weather
Peer-reviewed journal articles, conference proceedings, citation information, and links to published research.
Research projects spanning:
- Coronal magnetic field measurements
- Infrared spectropolarimetry
- Solar eclipse prediction studies
- Polarization synthesis
- Solar cycle investigations
- Numerical modeling of solar plasmas
Interactive tools and utilities developed for:
- Solar data analysis
- Visualization
- Spectral diagnostics
- Research workflows
- Scientific communication
Technical and educational articles covering:
- Solar physics concepts
- Scientific programming
- Numerical methods
- Data analysis techniques
- Research insights
Course material, lecture notes, tutorials, and educational resources related to astronomy, astrophysics, and computational science.
| Area | Description |
|---|---|
| Solar Corona | Structure, dynamics, and magnetic fields |
| Spectropolarimetry | Magnetic field diagnostics using polarized light |
| Cryo-NIRSP | Infrared coronal observations and analysis |
| Numerical Modeling | MHD simulations and forward modeling |
| Scientific Computing | High-performance scientific workflows |
| Data Visualization | Interactive and publication-quality visualizations |
| Solar Eclipses | Prediction, analysis, and educational resources |
Jekyll on GitHub Pages. Every page shares one layout, so the navigation, sidebar, footer, analytics snippet, and structured data live in a single place each rather than being copied into every file.
_config.yml site settings and plugins
_data/ content as data — publications, projects, courses,
resource catalogues, navigation, social links
_includes/ head, sidebar, footer, and the renderers that turn
_data into markup at build time
_layouts/ default.html (all pages) and post.html (articles)
_posts/ articles, in Markdown with YAML front matter
assets/css/ style.css, icons.css, plus per-section stylesheets
assets/js/ view controllers — no page fetches its own content
assets/vendor/ self-hosted MathJax, Prism and marked
scripts/ bib_to_data.py, which regenerates the publication data
Everything a visitor reads is in the HTML when the page arrives. No list on this site is assembled in the browser, so it all works without JavaScript and is visible to crawlers.
The site needs Ruby 3.3. macOS ships Ruby 2.6 at /usr/bin/ruby, which is too
old for Jekyll and should be left alone — it belongs to the OS. Install a
separate Ruby alongside it:
brew install ruby@3.3Then, once, in this directory:
bundle config set --local path vendor/bundle
bundle installAfter that:
bundle exec jekyll serveand the site is at http://localhost:4000. Gems land in vendor/bundle
inside the repo, which is gitignored and excluded from the build, so nothing
is installed globally and removing the folder undoes it.
Ruby 3.3 specifically: github-pages pins Jekyll 3.10, which does not work on
Ruby 4.x. Homebrew's unversioned ruby formula is 4.x, so install ruby@3.3
rather than ruby.
Normally you do not: .github/workflows/sync-publications.yml queries NASA ADS
on the 1st of each month, regenerates the bibliography and the data files, and
opens a pull request if anything changed — a new paper, or citation counts that
have moved. It needs a one-time secret:
- get a token at https://ui.adsabs.harvard.edu/user/settings/token
- add it under Settings → Secrets and variables → Actions as
ADS_TOKEN
Without the secret the job exits with a note rather than failing.
To run it by hand, or off-schedule:
export ADS_TOKEN=...
python3 scripts/sync_ads.py # add --dry-run to previewThe sync merges: it refreshes citation counts and appends new papers, and never removes an entry. That matters, because an ADS ORCID search only returns records where the ORCID is actually attached — for this profile, 12 of 22. Conference abstracts (AGU, AAS, SPD, SHINE, ASI) and papers where you are a middle author often have no ORCID claim, so treating ADS as authoritative would quietly delete real publications.
If ADS returns something you would rather not publish — an erratum, a
duplicate, a same-name mismatch — put its bibcode in scripts/ads_exclude.txt
and future syncs will leave it out.
Google Scholar indexes more than ADS does: preprints, theses, book chapters, and work published outside the astronomy journals. It has no API — no public interface, terms that prohibit scraping, and active blocking of automated clients — but you can export your own profile by hand and merge that:
- open your Scholar profile
- tick the box in the table header to select all articles (per page, if you have more than one)
- Export → BibTeX, save the file
- run:
python3 scripts/import_scholar.py ~/Downloads/citations.bibEntries already in the bibliography are skipped, matched by DOI first and then by normalised title, because Scholar's citation keys bear no relation to ADS bibcodes. New ones are appended under a marker so a later ADS sync leaves them alone. Nothing is removed, and re-running is a no-op.
Review the diff before committing: Scholar's metadata is looser than ADS's and it occasionally lists the same paper twice under slightly different titles.
The .bib remains the file everything derives from. If you edit it by hand,
regenerate the data with:
python3 scripts/bib_to_data.pyCI runs that same script with --check and fails if the two have drifted.
Drop a Markdown file in _posts/ named YYYY-MM-DD-slug.md:
---
layout: post
title: "..."
date: 2026-01-15
category: concepts # concepts | methods | research
tags: [Solar Physics, MHD]
excerpt: "One or two sentences; used on the hub, in the feed, and as the meta description."
---The articles hub, the category filters, the RSS feed and the sitemap all pick it up automatically.
Visit the website:
The website is automatically deployed through GitHub Pages.
- Google Scholar
- ORCID
- NASA ADS
- GitHub
(Links available on the website.)
solar physics, heliophysics, solar corona, coronal magnetism, spectropolarimetry, Cryo-NIRSP, DKIST, solar eclipse, MHD simulations, scientific computing, astrophysics, space weather, solar observations, data visualization, numerical modeling, astronomy education, scientific software
If you find any of the scientific resources, educational content, software, or visualizations useful in your research or teaching, please cite the associated publication whenever appropriate.
Soumyaranjan Dash
Solar Physicist | Researcher | Scientific Software Developer
🌐 Website: https://sr-dash.github.io/
📧 Contact information available through the website.
Content is CC BY 4.0, site code is MIT. See LICENSE for the split and NOTICE for bundled third-party components.
Figures reproduced from published papers remain subject to their publishers' terms; check the relevant publication before reuse.