Home Research Teaching Publications CV Contact

21-cm signal from the Cosmic dawn

Cosmic dawn refers to an era in the history of the Universe when the first stars and galaxies were assembled from primordial gas. Observational studies of this era are one of the frontiers of contemporary astronomy. So far we know that sometime between 100 and 200 million years after the Big Bang (it is now 13.7 billion years after the BB), the first stars and galaxies formed. The intense ultraviolet and X-ray radiation they emanated heated and ionized (i.e. separated the electrons and protons that form Hydrogen gas) the diffuse intergalactic gas around them.

The most promising probe of the then prevailing conditions is the 21-cm line of hydrogen. Hydrogen atoms preferentially interact with radio-waves with a wavelength of 21-cm or a frequency of 1.4 GHz (our WiFi routers operate at about twice this frequency) and act as a thermometer telling us about the evolution of the temperature of intergalactic gas with time. This evolution tells us precisely when the first stars turned on and how much their radiation heated the gas around them. In addition, the spatial fluctuations or "graininess" of the 21-cm signal also encodes information about the nature of the first stars and protogalaxies; for instance, did a few massive objects do all the work or was it numerous smaller galaxies?

These observations are challenging because the signal we are after is extremely faint in comparison to everything else around us (up to a million times!), such as radiations from our own Galaxy, all the other galaxies in the Universe, and human generated radio waves. This not only means we need massive telescopes that can pick up the faint whispers from the early Universe, but also advanced data processing techniques that can "get rid" of all the bright unwanted sources of radiation. I am a core member of one such Herculean effort being undertaken with the Low Frequency Array (LOFAR). My efforts within the project are primarily focussed on developing novel observational techniques and to understand and mitigate the corrupting influence of the Earth's atmosphere on radio waves.


Astrophysical Lensing is a gift of nature, as it is basically a magnifying glass of cosmic proportions. Most people know of gravitational lensing (as predicted by Einstein's theory), which is caused by the warping of light by gravity. Lensing is not restricted to the gravitational type; anything that warps or bends light can lens, e.g. a magnifying glass. The plasma that pervades the interstellar medium (space between stars) also can warp radio waves and give rise to plasma lensing. This nice part is that the mathematics of lensing is basically the same for gravitational of plasma lenses. I am interested in both, especvially on small scales that we have not yet probed.

Extreme scattering events

I am intrigued by a type of plasma lensing phenomenon called extreme scattering events (ESE). ESE are time-symmetric "U-shaped" dips in the brightness of radio sources. They were discovered more than 30 years ago but have eluded a convincing explanation so far. As far as we can tell, ESEs are caused by dense plasma globules in the Galaxy. But their inferred densities (and hence pressures) are so high that they should simply explode and disperse into the interstellar medium. One work-around is to make them self gravitating (like stars and planets); the force of gravity then keeps them intact. But doing so would imply that they constitute a significant fraction of Galactic dark matter: an exotic if not unpalatable conclusion.

To make progress with the ESE mystery, one of my PhD mentors Ger de Bruyn (RIP), and I decided to try a different tack. Ger had discovered the most extreme example of interstellar scattering from a source called J1819+3845 back in the days. We knew that the plasma cloud responsible had to be extremely dense (more than 100 particles/cc; the background density is more like 0.3 particles/cc) and that relatively close by (about the distance to Alpha-Centauri system). We had a specimen in our backyard for careful study. We then tried to detect a complementary plasma phenomenon called Faraday rotation that is associated with magnetised plasma in the interstellar medium. We used the old Westerbork Radio Telescope and LOFAR telescope to make a Faraday map of the sky towards J1819+3845. We found a dense elliptical globule about 1 x 2 deg in size with J1819+3845 on the turbulent edge of the globule which may have caused the scintillation. This is the first time we were able to make an "image" of the structure causing extreme scintillation and taught us that relatively large elliptical globules floating around in interstellar space may be causing all the anomalous scintillations when they begin to eclipse a radio source. We are following up on this discovery to learn the precise nature of these sources.

Gravitational milli-lensing

I was searching for ESEs in the data from the 40-m but ended up discovering a new phenomenon called Symmetric Achromatic Variability (SAV). Basically, we found the tell-tale U-shaped signature we were looking for in one source called PKS 1413+135, but this signature persisted at all radio frequencies (hence achromatic), something that is very hard to explain in terms of plasma lensing. We conjectured that gravitational lensing might be the culprit because gravity affects radio waves of all frequencies identically. It turns out that the lens should be much more massive than individual stars but much less massive that a typical galaxy. This intermediate mass range (thousand to a million times the mass of the Sun to be specific), is a fertile ground to test different dark matter models; an exciting prospect I am following up on.

Fast Radio Bursts

Fast Radio Bursts (FRB) are intense radio bursts that only last about one-thousandths of a second. Discovered in 2007, we do not know a whole lot about them other than (a) they are fairly common (thousands of them go off each day), (b) they come to us from cosmologically large distances and (c) some of them repeat. I developed an interest in FRBs during my time at Caltech, mostly thinks to fellow-postdoc Vikram Ravi. I have done work related to modelling the FRB population and statistical considerations relating to associating FRBs with slower transient events (lasting hours to days after the FRB) called "afterglows". I was also involved in getting a novel 10-dish FRB localization machine going at the Owens Valley Radio Observatory, Finally, I have also worked on the propagation of FRBs though the circum-galactic medium of intervening galaxies and what we can learn about the internal small-scale structure of the media from future FRB surveys.