Join or visit
Students and visitors are welcome to write. These are directions where the group has open code and public data, and where a new person could add something. Nothing here is a vacancy.
Open code and data
4 directions, each with a public repository or record.
- 01
Magnetic field from velocity gradients
VGT.jl turns H I and CO spectral-line cubes into plane-of-sky field orientation maps. The method gave the foreground direction in the LMC paper (Alina et al. 2023).
Where you could contribute: Testing the method on more cubes, and checking its angles against dust and starlight polarization.
- 02
Cold neutral gas in the LMC
ROHSA-LMC runs the ROHSA Gaussian decomposition (Marchal et al. 2019) on LMC H I cubes, with velocity-layer cubes and disk de-rotation.
Where you could contribute: Reading the decomposed components, and separating the cold neutral medium from the rest of the atomic gas.
- 03
Supergiant shells and the field
LMC-shells compares the rims of optical supergiant shells (DeMCELS Hα and [S II]) with the field directions from velocity gradients and from Planck, with synthetic controls.
Where you could contribute: Extending the rim-tangency test and the controls to more shells.
- 04
Filament angles with a neural network
A U-Net trained on synthetic images returns the position angle of filamentary structures without per-map tuning (Umetaliev et al. 2025). The model is open on Zenodo.
Where you could contribute: Applying it to new Planck, Herschel and H I maps, and measuring filament orientation against the field in larger samples.
In progress
Lines the group is working on now. Details follow once the group has agreed to show them.
- In progress
Field from GASKAP-HI velocity gradients
The same velocity-gradient method applied to GASKAP-HI 21 cm data, at 30 arcsecond resolution instead of the 16 arcminutes of GASS.
- In progress
Cold neutral medium, separated
Gaussian decomposition of every H I spectrum (ROHSA) separates the cold neutral medium from the rest of the atomic gas.
Before you write
Positions, funding and computing are not advertised on this page. If one of the directions above interests you, tell us which one and what you have worked with, and we can discuss what is possible. Contact details are on the contact page, which also shows why a group at 51° N works with data from the south.