Denis Erkal
banner
derkal.bsky.social
Denis Erkal
@derkal.bsky.social
Astrophysicist working on dark matter, near-field cosmology, and galaxy formation. All opinions are my own. He/him.
Here’s the Cygnus loop with a Seestar S50. It’s so big that one mosaic won’t fit it so I stitched two of them together. The total exposure is about 4 hours. It would be cool to go even wider and deeper. 🔭
August 22, 2025 at 7:56 PM
Yes, it’s super annoying since this used to be very useful. The issue is it thinks 3e19m is ~56 kpc instead of ~1 kpc but who knows what other units are/will be wrong. Maybe they’ve started using AI for their units?
August 4, 2025 at 7:19 AM
Anyone else notice any weird unit issues with Google lately? (3e19 m)/(1 kpc) gives 55.7 instead of ~1. Not good if you like using Google for quick unit conversions 🔭
July 24, 2025 at 8:18 PM
I tried to image the Palomar 5 globular cluster with a Seestar S50 for 3 hours. The cluster is the middle of this image and some of the faintest stars have Gaia G mags of ~17.5 and the right proper motions to be Pal 5 stars. 🔭
April 12, 2025 at 1:44 PM
Spent 4 hours on NGC 4631 (top galaxy, aka Whale galaxy) and NGC 4656 (bottom galaxy, aka Hockey stick galaxy) with a seestar last night. The hook in NGC 4656 is due to an interaction with other galaxies in the group. 🔭
March 31, 2025 at 8:15 AM
Today’s solar eclipse from the southern UK. It was taken with a seestar 50. I was very impressed how well it kept the sun centered in the timelapse! 🔭

youtube.com/shorts/f-wsa...
March 29, 2025 at 12:38 PM
The stars in the stream would also look different. Due to the rapid mass loss in SIDM, stars leave with a smaller dispersion at the previous pericenter, and thus create a much smaller spur in the stream. In CDM, the more gradual mass loss results in a significant spur, closer to what is observed 3/
March 21, 2025 at 10:34 AM
At the SIDM cross-sections considered in this work, the dark matter halo is almost completely stripped and there would be no dark matter remaining with the stars in Sagittarius. This figure shows the dark matter in CDM on the left and in SIDM on the right. 2/
March 21, 2025 at 10:34 AM
I also got this nice image of the Orion Nebula. The framing options are quite nice on the seestar and are great for large objects on the sky. 🔭
March 2, 2025 at 7:32 AM
I played a bit more with our department’s seestar and got this image of M3. The AIdenoise seems to pick out the stars resolved in the original image quite well. 🔭
March 2, 2025 at 7:29 AM
Had some fun tonight learning how to use one of our department’s seestar telescopes. They’re super easy to set up and use. I think our students will enjoy imaging galaxies with these! 🔭
January 30, 2025 at 11:51 PM
This final movie shows how NGC 1904 evolves in the Milky Way. The left panel shows a zoomed-out view centered on the Milky Way and the right panel shows a zoomed-in view centered on the globular cluster. The dumbbell morphology we see on the sky comes from the last stripping episode. 3/3 🔭
November 15, 2024 at 9:58 AM
We also see a clear distance and velocity gradient in the observed members that match our N-body simulation. This figure shows the CMD of the stream members and the stars on the top part of the dumbbell are fainter than those on the bottom. 2/3 🔭
November 15, 2024 at 9:58 AM
Paper day! Petra Awad led this nice paper (arxiv.org/abs/2411.08991) on the outskirts of NGC 1261 and NGC 1904. NGC 1904 has two jets of stars stripped at its most recent pericenter with the Milky Way. This pericenter was only ~0.12 kpc which led to this strong stripping event. 1/3 🔭
November 15, 2024 at 9:58 AM
Made this plot in ChatGPT. It shows the falloff of men’s & women’s running/swimming speed vs event duration. The y-axis shows the mean speed for each event normalized by the speed of the shortest event (100m for running, 50m for swim). It’s crazy how little the falloff is, 60% max speed for ~2 hours
May 19, 2024 at 8:21 PM
Aurora over Guildford in the Southern UK! 🔭
May 10, 2024 at 10:56 PM
Since we recovered all parameters of the mock impact, we can also look at where the subhalo would be today. Here we show the on-sky position of the mock impactor and it’s well recovered in all setups. This would be useful for follow up, eg Fermi, to search for a counterpart for indirect detection 5/
April 5, 2024 at 11:03 AM
We explored three different observational setups: one with no observational errors, one with present day errors (ie Gaia DR3 proper motions and AAT velocities), and one with future errors (ie Gaia DR5+4MOST). In all cases, we were able to recover the subhalo mass and scale radius quite well 4/
April 5, 2024 at 10:58 AM
We fit our mock data with our perturbed stream model and were able to recover all of the subhalo impact parameters, ie subhalo mass, size, velocity, impact time, impact parameter, and flyby velocity even with currently attainable observational errors 3/
April 5, 2024 at 10:54 AM
We based on the model on ATLAS-Aliqa Uma (AAU) which is a stream with a significant link in it. This figure shows the real observables for AAU along with our fiducial model of a stream perturbed with a 10^7 Msun subhalo. We then fit this mock data to see how well we could recover the subhalo. 2/
April 5, 2024 at 10:51 AM
Gave a fun public talk on How to weigh the Milky Way at our Surrey Astro evening. Lots of great questions from the audience!
January 24, 2024 at 10:16 PM
Another scene from DALL-E
November 9, 2023 at 10:02 PM
Exploring stellar streams with DALLE
November 8, 2023 at 1:57 PM
DALLE-3 is an impressive step up in what is possible. This is a one horse power cart.
November 2, 2023 at 11:28 PM
Did another in-class experiment for stat mech this week. This is a seesaw with one side held down by rubber bands and a laser pointer on the other side. If we heat up the rubber bands, will this lower or raise the laser pointer? 🧪
October 18, 2023 at 8:57 AM