Archive | July 2020

Welcome to Disk Detective v2.0!

Hi, all:

As you may have seen over the past month, or from today’s email blast from Zooniverse, Disk Detective v2.0 is LIVE!!!!!

Here’s a rundown of some of the things you may notice are different now:

  • We’ve moved onto the Zooniverse Project Builder. This means that it’s much easier for us to change out image sets as necessary (e.g. adding on objects that receive Gaia parallaxes), and it’s easier for us to update the project otherwise.
  • Instead of using SDSS and DSS2, we’re using PanSTARRS and Skymapper for our images at visible wavelengths. These surveys have higher spatial resolution and probe deeper than SDSS and DSS2 do, so we’ll be able to see fainter background objects and multiple objects that are very close together.
  • Instead of the AllWISE images, we’re using image coadds from unWISE. These images (which are also used to assemble the data for Backyard Worlds: Planet Nine) have better spatial resolution than the AllWISE atlas, which will let us detect background objects more easily in the WISE data. However, this data set comes with its own unique set of issues, which I’ll discuss below.
  • We’ve added a new circle in the images. Much like the red circle in DDv1.0 (and the outer circle in DDv2.0) correspond to the W4 point spread function (PSF; how light spreads out on the camera chip), this circle corresponds to the 2MASS PSF. This lets us know if our 2MASS data has multiple objects blended together, which could indicate interesting phenomena like a binary system!
  • We’ve added a new option for classification: “multiple objects in the inner circle.” This is to indicate that the 2MASS and WISE photometry data will have two stars blended within it, which could affect our detection of a disk.

As part of this, today we deployed a new subject set. This set has the same objects as the previous set, but with improved image processing of the unWISE images to correct for saturation of the chip. We will be using classifications from the previous data set, but thought it was important to get the improved processing uploaded as soon as possible.

Please try out DDv2.0 and tell us what you think! If you have feedback on things that need clarification, feel free to reach out to us on Talk. And keep following us here, and on Twitter and Facebook for updates on the project!