An Introduction to DASCH
This slide deck provides a high-level introduction to the DASCH project from an astronomer’s perspective.
Navigate using the buttons in the lower-right, the arrow keys on your keyboard, or swipe gestures.
What is DASCH?
DASCH is the project to digitize the Harvard College Observatory’s Astronomical Photographic Glass Plate Collection for scientific applications.
Plate A01271, showing the Orion Nebula in 1895.
The Harvard “plate stacks” collection includes more than 550,000 photographic glass plates, documenting the entire sky over a timespan of more than a century (~1880–1990).
The Harvard plates span the entire history of photographic astronomy, from the invention of the technique to the triumph of the CCD. It is the largest collection of its kind in the world.
This long time baseline is the scientific crux of the DASCH dataset.
Digital plate images can be calibrated to obtain accurate astrometry and photometry. DASCH makes more than a century of such measurements available for every location in the sky. Any particular sky position is covered by thousands of plates.
Williamina Fleming shown standing, right of center (Harvard University Archives)
The Harvard plates are a historical and cultural resource as well as a scientific one. Every plate in DASCH was once held by one of the hundreds of “women astronomical computers” who studied and curated the Harvard Plate Stacks while making discoveries of their own.
While some of these women now rank among the most famous names in astronomy (Cannon, Leavitt, Payne-Gaposchkin) their collective contributions to science were often overlooked — when not actively erased.
The new website StarGlass lets you explore a comprehensive database documenting all of the plates in the HCO collection, the handwritten notebooks describing their acquisition and analysis, and the people involved in that work.
While StarGlass is aimed at the general public, not just researchers, we hope that it will help even seasoned professionals better understand the HCO collection underlying DASCH.
DASCH digitized the Harvard plates using a one-of-a-kind high-speed scanner, visible in action here:
The scanner hardware and software were built by Bob Simcoe and Ed Los. It is documented in Simcoe et al (2006).
Out of the whole collection, ~430,000 plates are scanned: virtually every one that’s a candidate for photometric calibration.
DASCH’s primary data product is the collection of “mosaic” FITS files imaging each plate at 11 μm resolution. The average mosaic is 750 MB. The total image corpus is about 400 TB.
About 94% of DASCH mosaics have successful astrometric calibrations.
The initial solution is obtained with Astrometry.Net and then refined with distortion terms.
Plates without astrometric solutions have approximate sky locations based on written observing logbooks. You can query the DASCH/StarGlass databases by sky location to obtain a list of all overlapping plates and a wealth of associated metadata.
About 93% of astrometrically-solved mosaics have successful photometric calibrations.
DASCH photometric calibration is performed against different “reference catalogs” or “refcats”. The currently recommended refcat is APASS DR8, yielding B-band magnitudes with excellent long-term stability.
There are also data using ATLAS-REFCAT2 as the reference. The resulting photometry has substantial artifacts making it unsuitable for many purposes, but it is superior if you only care about astrometry.
Lightcurve of eclipsing binary RY Cnc after filtering. Low points are eclipses.
The DASCH data products include a lightcurve for every source in every refcat.
For each refcat, DASCH contains ~15 billion magnitudes for ~40 million sources, to depths of 14th-16th mag. Typical lightcurve RMSes are about 0.15 mag.
Raw DASCH lightcurves must be reviewed before being used scientifically. The same goes for other DASCH data products. There’s a reason we don’t use photographic plates anymore: they are prone to all sorts of defects!
While the DASCH pipeline aims to detect and handle as many issues as possible, there remain various known issues in the data that you must check for.
The recommended way to obtain and analyze DASCH data is using the Python package daschlab inside a JupyterLab environment. This video (linked in the other Getting Started materials) shows what it’s like.
The DASCH project has been a herculean effort spanning two decades — this introduction can hardly do justice to it. To learn more, explore the DASCH website:
If you’re interested in working with DASCH data for astronomy, we recommend these next steps:
- Watch the video on the previous slide
- Run the associated notebook yourself in our hands-on tutorial
- Check out other materials on the DRnext landing page
Thanks for your attention, and welcome to DASCH!