A lightweight simulation suite for large-scale structure beyond GR

Hi-COLA is a package that runs fast, approximate N-body simulations of non-linear structure formation in reduced Horndeski gravity (Horndeski theories with luminal gravitational waves).

Hi-COLA is not hard-coded to solve specific Horndeski theories, but is designed to be generic with respect to the reduced Horndeski class. Given an input Lagrangian, Hi-COLA's frontend dynamically constructs the appropriate field equations and consistently solves for the cosmological background, linear growth, and screened fifth force of that theory. This is passed to the backend, an adaptation of the FML library, where a hybrid N-body simulation at significantly reduced computational and temporal cost compared to traditional N-body codes is run. By analysing the particle snapshots, one is able to study the formation of structure through statistics like the matter power spectrum.

Hi-COLA is publicly available on GitHub here.

See the Documentation folder on GitHub for guidance on how to install and use Hi-COLA. There you will find both a quickstart guide and a more detailed user manual.

Also see our paper, where we detail the work gone into creating the first version of Hi-COLA. We encourage the community to use Hi-COLA for their own research, and request that you cite us as detailed in the user manual (a handy bibtex is also provided there). Hi-COLA is provided under a CC BY 4.0 licence.

Hi-COLA remains under active development, with further extensions of the code and applications planned. You can contact the Hi-COLA team at [email protected] .


Come work with the Hi-COLA Team!

I'm currently advertising a postdoctoral position within the Hi-COLA team. The deadline for applications is 21st Jan 2024, and the position is due to start ~July-Sept. 2024. Here are some more details about the position:

Here are some more details on the research objectives and possible skill paths for this position. Probing the nature of dark energy and gravity is a major driver for galaxy surveys such as DESI, Euclid, LSST and others. Much of the new information from these surveys will come from the non linear regime of structure formation, which in general has to be studied numerically. However, many existing simulations focus on a small handful of gravity models, such as f(R) and DGP, thereby limiting the scope of new data to really influence the theory landscape.

My group at the ICG is working on a new simulation suite called Hi COLA, which aims to broaden the remit of these tests. We have constructed an extension of the FML code (H. Winther ) which can compute dark matter structure formation in any theory of gravity within the luminal Horndeski family of scalar tensor theories. Hi COLA consistently solves the background expansion history, modified forces and screening within this range of models. You can read more about Hi COLA here, and the code itself is on GitHub here .

We are looking for a postdoctoral researcher to join our team on this exciting effort. An ideal candidate would have experience with, or be willing to learn, numerical simulation codes such as COLA, Gadget, FML, etc. This would enable to candidate to work on development aspects of Hi COLA. There are numerous subtleties that arise when working with models beyond GR (e.g. stability), so experience in modified gravity models is also most welcome.

Experience is connecting simulation outputs to survey observables (e.g. clustering, weak lensing) is also warmly welcomed. Our group is actively involved in the Dark Energy Science Collaboration (DESC) of the Vera Rubin Telescope.

Applications for this position are to be submitted here. The position is also listed on the AAS jobs register here. Please note a PhD in cosmology, gravitational waves or other relevant topic is a requirement for this job.