This project is currently under development.
Research
Help us determine if tracings of galaxy spiral arms are clustered correctly
The goal of this project is to untangle tracings of spiral galaxy arms produced by Spiral Graph volunteers. Our algorithm groups, or clusters, tracings together to indicate individual arms, but it's not always right. We need your help to judge if the algorithm produced correct results.
Your classifications will help train an A.I. that will select the appropriate parameters for clustering tracings into individual arms. Your contributions will make our A.I. smarter!
Why is this important?
Galaxies are not always symmetric and their individual spiral arms may differ in shape from each other. The three example galaxies below may appear symmetric at first glance, but their arms are actually different from each other. Some are longer, some are bifurcated, and some have a different pitch angle.
NGC 1300 | M 101 | PGC 1212361 |
|---|---|---|
 |  |  |
Our pitch angle measurement software, P2DFFT, assumes the input image is perfectly symmetric. Measuring the pitch angle of each arm individually lets us bypass that assumption.
How you can help
We need you to judge whether or not our clustering algorithm correctly grouped tracings together into individual arms.
The Task
- First, you will be shown a set of images.
A. One image will be the aggregated set of tracings of the galaxy's arms, obtained from multiple different volunteers in Spiral Graph.
B. The second image will be the result of our algorithm's attempt at grouping the tracings into individual arms, while ignoring any outliers. Each color represents a different cluster of tracings.
C. The final image is an overlay of the second image on top of the first image.
A. Aggregated Tracing Image | B. Clustered Image | C. Overlay Image |
|---|---|---|
 |  |  |
- Then you will be asked "Do the tracings appear to be clustered correctly into individual arms?" You simply have to use your best judgement and select yes or no!
The Impact
We will use your judgement classifications to train an A.I. algorithm we are developing. The A.I. will help us select the correct parameters used by our clustering algorithm based on the aggregate tracing image. This will help us more accurately cluster tracings into individual arms and allow us to measure the pitch angle of each separately. We hypothesize that this will allow us to more accurately determine the "global" pitch angle of non-symmetric galaxies.
Why is measuring pitch angle important?
The pitch angle of spiral arms relates to other parameters of the host galaxy that are more difficult and time consuming to measure. These parameters include the mass of the black hole found in the nucleus[1], bulge stellar mass[2], galaxy total stellar mass[3], maximum rotational velocity, and dark matter content of the galaxy[4]. By measuring the spiral arm pitch angle we can quickly and easily estimate these parameters and identify interesting galaxies for more detailed, follow-up telescope observations.
For example, we can use pitch angles to search for black holes in the centers of large galaxies with masses that range from hundreds to hundreds of thousands of times the mass of our Sun (intermediate mass black holes). Most large galaxies have central black holes with masses that range from millions to billions of times the mass of our Sun (supermassive black holes).
This project is supported by NASA's Citizen Science Seed Funding Program[5].
Treuthardt, P., Tezbasaran, A., & Hewitt, I. (2024). Spiral Graph: Cluster Buster - A Participatory Science Project to Improve the Identification of Spiral Arms from All-Sky Survey Galaxy Images. Zenodo. https://doi.org/10.5281/zenodo.15882378 ↩︎