Neutral Hydrogen intensity mapping with MeerKAT

During the last two decades we have entered a "golden era" of cosmology. Using satellites and ground based telescopes we have gathered high quality data from the very early Universe, essentially from light emitted right after the Big Bang explosion, as well as from the late Universe, through the light emitted from stars and galaxies.

However, a big part of our Universe's history and volume remains unexplored. A way to attack this challenge is by observing the light emitted from the neutral hydrogen (HI) that filled the Universe for a long time after the Big Bang and before the first galaxies were formed. After that time HI resides within galaxies, so we can also use it as a novel way to study the late Universe. This is my main area of research; it is exciting because it opens a new observational window into the Universe and can push the boundaries of our understanding of astrophysics and cosmology.

In the next few years, HI surveys of exquisite sensitivity will be performed using radio telescopes, and part of the proposed research is working out new observational methods and techniques in order to maximise their science output. I have pioneered a new observational method that does not require the -difficult and expensive- detection of individual galaxies but maps the entire HI flux coming from many galaxies together in large 3D pixels (across the sky and along time). I aim to use this method to provide a 3D map of the Universe using HI intensity mapping data from the MeerKAT array. MeerKAT is a radio telescope located in Karoo, South Africa, and it is a pathfinder for the Square Kilometre Array, which is going to be the largest radio telescope in the world.

By using these data I want to answer fundamental questions about the Universe, for example what is the nature of dark energy - the mysterious substance responsible for the accelerated expansion of our Universe - and how galaxies evolve. But before doing that, much work is needed to improve the quality of the data and figuring out how to solve the many challenges that intensity mapping faces.

One of these challenges has to do with the huge emitted radiation (called "galactic synchrotron") from our own Galaxy, that severely contaminates our HI maps. In other words, when MeerKAT observes the night sky, it maps the HI cosmological signal, but it also maps the emitted synchrotron radiation, which is a thousand times larger than the signal! The cosmological signal we want to measure is therefore "buried" under a much larger astrophysical signal. We can attack this problem by exploiting the fact that foregrounds are smooth functions of frequency, while the HI signal has a spiky structure. So, in principle, foregrounds can be perfectly removed without losing any signal information. However in practice this is very challenging. I will use the available MeerKAT data sets to test the performance of various foreground cleaning and calibration techniques.

After the data has been processed and cleaned, I will move on to build a pipeline for the cosmological analysis of the HI intensity mapping signal. This pipeline will also account for the possibility of powerful synergies between HI and traditional optical galaxy surveys, by including cross-correlations data analysis tools. This is useful in order to obtain measurements that are free of systematic contaminations that often plague individual surveys (but drop out when combining them), and therefore more robust. Our developed tools will be used to analyse Science Verification data taken with MeerKAT, i.e. data that cover a larger part of the sky and observe for longer time, so they can be used to make precise cosmological measurements. We then hope to perform the first ever measurements of HI and cosmological parameters in the radio wavelength using the intensity mapping technique, exploit multi-wavelength synergies, and revolutionarise our understanding of galaxy evolution and dark energy.

The beneficiaries of this research include: (1) early career researchers (2) geophysics and medical physics (3) the Alan Turing Institute (4) South African economy and welfare (5) the public

(1) Early career researchers will benefit from this research in many ways. First of all, working on this project involves applying state-of-the-art and innovative data analysis techniques, as well as preparing the software tools and pipelines for the SKA. Hence, early career researchers will develop a variety of skills that will help them excel in any career path - in academia or industry- they choose to follow. These benefits are not only expected for the PDRA and the PhD student working on this project, but for a large number of researchers in the SKA, most notably students and PDRAs based in South Africa. This is going to be greatly enhanced by the participation of the PI in SA-DISCnet, an STFC-funded project for creating a collaborative data science training network across South Africa and the UK, part of SEPNET and AIMS (African Institute of Mathematical Science). The immediate beneficiaries will be astronomy and cosmology students and researchers, but the longer term use of our publicly available tools will benefit students and researchers from other branches of physics, engineering, and computer science.

(2) The PI's research project is crucial for the success of the innovative technique of HI intensity mapping. After the first generation of very large sky HI intensity mapping surveys has been commissioned, we will immediately start preparing for future experiments. The use of new technologies such as Phased Array Feeds (PAFs) can revolutionise cosmology with HI intensity mapping, and the success of this case is crucial to promote their use. Note that this technology has already been used for radio astronomy, for the first time, in the Australian Square Kilometre Array Pathfinder - ASKAP. The PI is in close contact with the Advanced Instrumentation team in ASTRON (Netherlands), and her research is instrumental for the science case promoting new technologies for the SKA. PAF technology has outstanding potential outside astronomy. Notable examples include geophysics and medical physics, that could greatly benefit from the very rapid imaging made possible by PAFs.

(3) The Alan Turing Institute is the national institute for data science, which QMUL is joining this year. Its mission is "to make great leaps in data science research in order to change the world for the better". The PI is going to collaborate closely with the Institute to develop collaborative programmes of research. SKA-related projects will benefit from the PI's research.

(4) The SA-DISCnet project will pilot an innovative course of training and internships for the next generation of data analysts, focusing on South Africa's economic development and welfare in the 21st century. Data intensive science is a major global growth area, and the PI will combine her participation in SA-DISCnet with her proposal's research products (for example the publicly available simulations and data analysis tools) to contribute further to this cause.

(5) Large sky radio and optical galaxy surveys are much about exploring the unknown and their findings will enrich humankind's understanding of Nature. The PI will use the motivation, the background, the developed tools and results of her research in public outreach activities (e.g. teaching in schools, open days, stargazing live) to generate public awareness of science.