MapItAll
Illuminating the darkness with precision maps of neutral hydrogen across cosmic time
MapItAll is an ERC-funded project to apply high-dimensional Bayesian inference to the difficult problem of accurately characterising the uncertainties in the sky and instrumental models used to understand data from 21cm intensity mapping arrays.
About.
The problem being addressed by this project is to find a way of extracting a very weak radio signal that was emitted in the early stages of the Universe from in amongst all of the other, much brighter/louder sources of radio noise from Earth and our own Galaxy.
The radio signal we are interested in is called the "21cm line", and is emitted by hydrogen gas when it is in a neutral state (generally meaning that it is cold) at a wavelength of 21cm. About 400,000 years after the Big Bang, the Universe underwent a transition from being filled with hot ionised gas, to cooling down enough that neutral gas could form. After a few more million years, the first stars and galaxies switched on, "re-ionising" the Universe over a period of many millions of years. By making maps of the neutral hydrogen from this period, via its 21cm line emission, we can understand how and when the early stars and galaxies formed. After this period, the neutral hydrogen settles into dense clumps inside galaxies, and so we can also use it to map out the galaxy distribution, which in turn tells us things about how fast the Universe is expanding. This information can be used to study the mysterious "dark energy" component of our cosmos.
Unfortunately, the 21cm signal is very weak, being somewhere between 1,000 and 10,000 times fainter than the emission from our own Galaxy. We must therefore find ways of recovering it, while cleaning out the unwanted bright sources. This is made particularly difficult by the complexity of the radio telescopes we use to observe these signals; while they are very sensitive, they are also difficult to calibrate.
In this project, we are developing a novel statistical approach to doing this calibration that takes into account as many of the complicated features of the telescope, the unwanted signals, and the 21cm signal itself to make a very precise model of every measurement that we make, involving potentially millions of parameters that need to be fitted to the data. This requires some novel mathematical and computational techniques to be used to make the problem tractable to solve; using standard methods, it would be impossible to measure such a large number of parameters.
The MapItAll project team consists of Dr Phil Bull (PI), Dr Jacob Burba, Dr Mike Wilensky, and Katrine Glasscock. A number of collaborators are also involved in the project, including Dr Hugh Garsden (UoM), Dr Mel Irfan (UWC), and Fraser Kennedy (QMUL).
This project that has received funding from the European Research Council (ERC) under the European Union's Horizon 2020 research and innovation programme (Grant agreement No. 948764).
Publications.
This is a list of scientific articles published as part of the MapItAll project.
Code.
See below for links to the code(s) used as part of this project. Many of them are managed as part of an umbrella project on GitHub, HydraRadio.
Data access.
The data from the HERA and MeerKAT projects are subject to proprietary periods set by the collaborations and their principal funding agencies. For the data used in any given paper, please consult the data access statement within the paper, or email the lead author.