Mapping the Radio Sky: A New Approach

Building a map of the radio sky is super important, especially when we want to spot the 21 cm emission line signal from the time when the universe was very young, known as the Epoch Of Reionization (EoR). This map helps scientists in many areas of space research. To create this map, we used data from a radio survey called the LOFAR Two-meter Sky Survey (LoTSS) at 150 MHz. With this data, we figured out how bright different radio sources are, including various kinds of radio galaxies and active galactic nuclei.

We updated a computer program called the Tiered Radio Extragalactic Continuum Simulation (T-RECS) to make better mock radio source catalogs. The fake source counts from our updated work matched up better with what we actually see in the sky, especially for distant galaxies. We also saw that our model predicted a lower number of faint sources than T-RECS, which could be helpful for studying signals in the low-frequency band of the radio spectrum.

When it comes to the intensity maps of the 21 cm emission line at redshift around 6 and further out, they provide a unique way to look into what was happening in the universe during the Epoch of Reionization. However, a bit of a problem arises because the signals we want to detect get mixed up with unwanted background radio waves, coming from our own Milky Way and other sources that aren't quite what we’re looking for.

These unwanted sources can be classified into two main types: Active Galactic Nuclei (AGNs) and Star-Formation Galaxies (SFGs). Non-AGN sources create radio waves through things like supernova explosions or gas clouds. AGNs, on the other hand, are busy eating gas, and while doing so, they also blast out radio waves. AGNs can be loud or shy, and we call them radio-loud (RL) and radio-quiet (RQ) types.

The radio-loud AGNs produce most of their radio signals through powerful jets of particles. Depending on how they look and their behaviors, we can further divide them into High Excitation Radio Galaxies (HERGs) and Low Excitation Radio Galaxies (LERGs). For the radio-quiet AGNs, the origin of their radio signals is not very clear yet.

Simulating the radio sky is a handy tool for astronomers. For instance, it helps them figure out how complete their radio surveys are and predict what kinds of radio sources they might see in future surveys. In terms of the reionization research, it allows them to estimate the background noise that could obscure the 21 cm signal they want to study.

The T-RECS model is built to simulate the two main types of radio galaxies: AGNs and SFGs. It takes into account things like how bright the galaxies are at different distances and their clustering properties. However, since T-RECS gets its data mainly from higher frequencies, this can cause some gaps when trying to understand what is happening at lower frequencies.

In the last few years, there have been tons of new low-frequency radio surveys, which have given us more information than ever before. One of the most extensive surveys was done by LOFAR, which identified about 80,000 radio sources and categorized them into HERGs, LERGs, RQ-AGNs, and SFGs.

Compared to the T-RECS, the new catalog offers more variety by including the RQ-AGNs and extra sources at higher distances. This helps eliminate discrepancies we saw at those high distances.

The Square Kilometre Array (SKA) is on its way to being built and will provide even better observations in radio astronomy. This means we need to get our simulations right so we can prepare for what the SKA will detect and how to deal with the challenges in getting the EoR signal.

In this piece of work, we present a fresh outlook on how to classify radio sources and their models of evolution. We break it down into sections, starting with how we picked our data, followed by our models for AGNs and SFGs, and a comparison of our outcomes with what has already been observed.

Since radio waves don’t care about dust, using deep radio surveys gives us a clear view of galaxies and AGNs. Based on the deep 150 MHz LOFAR survey across three sky regions, which took a lot of telescope time, we created a catalog of 81,951 sources. We then classified these sources into different categories based on their brightness levels and characteristics.

For the sources with low radio waves, we figured out if they were SFGs or RQ-AGNs based on their radio spectra. High radio flux sources were split into LERGs and HERGs. We did this by comparing results from various codes designed to figure out how energy is spread across different wavelengths for each source. This helped us assign star formation rates and masses effectively.

The radio sources mainly consist of AGNs and SFGs, making our new catalog perfect for simulating the background sources for the 21 cm signal. To refine our selection, we looked out for sources within specific redshift ranges and a flux density limit that we could rely on. This helped us identify a total of sources while filtering out those that didn’t meet our criteria.

Next up, we describe our models for understanding the luminosity functions. These functions describe how bright different types of radio galaxies are and how this brightness changes over time. We fitted these functions based on what we learned from our catalog data, using a method known as Monte Carlo Markov Chain to get a range of probabilities for these measurements.

We set our luminosity functions for HERGs and LERGs based on how their characteristics change with redshift. This included updating values to represent how these functions evolved over time.

RQ-AGNs show signs of activity across various wavelengths, but they lack the strong radio jets typical of radio-loud types. The exact sources of their radio emissions are still up for debate. However, it appears that both star formation and a central core could play a role in producing radio emissions.

We also estimated the star formation rate for all sources. For RQ-AGNs and SFGs, we use a relationship between their radio luminosity and their star formation rate. For HERGs and LERGs, we pointed out an extra radio excess that we factored in when calculating the overall star formation rate.

Another aspect we looked at is the spectral index, which describes how the frequency of radio waves relates to each other. We found that the spectral index distributions for both AGNs and SFGs had consistent patterns, which we used to calculate the spectral indices for our sources.

In summarizing our model's findings, we rigorously compared our predictions to existing observations and other simulations to justify our work. We particularly focused on how closely our results matched the actual sources we expect to see in the sky.

In essence, our simulations and models have brought us closer to understanding the mix-up of radio sources in the universe. We discovered that our methods offer improved agreement with actual data, which might help in future studies of the radio sky.

We even made our modified T-RECS code available for others in the research community. This means that as we explore more of the low-frequency radio source sky, we're providing tools for others to join the hunt.

With a little bit of humor, running around in the universe while observing all these fascinating radio signals feels like trying to catch butterflies with a net made out of spaghetti- it’s challenging but oh-so-rewarding when you finally get a grasp of a few!