James Webb Space Telescope and its’ study of Dusty Star Forming Galaxies

This article is a scientific research paper that I wrote for my Astrophysics 320 class at Texas A&M.

1. Science Background

1.1 Dusty Star Forming Galaxies

Dusty Star Forming Galaxies (DSFGs) are galaxies that have highly intense star formation rates and are composed mainly of cold gas that blocks many wavelengths of light, especially UV, from getting through, and re-radiate it in the near IR wavelength. These galaxies provide insights into the early evolution and history of the universe. They represent the formation of galaxies during a phase of very rapid star formation right before quenching into much more quiet, non active galaxies (Casey et al. 2014). DSFGs provide astronomers with the opportunity to study and gain a deeper understanding of early Universe galaxy formation and interactions.

The evolution of DSFGs is an emerging field in astronomy, especially with the recent launch of JWST and its ability to gain high resolution imaging in the near-IR wavelength. DSFGs have many unique characteristics that will each be looked at more in depth in later sections of this report. These characteristics include lopsidedness, the formation of bulges, and outside-in quenching due to internal instabilities within the galaxy. These unique characteristics are currently being studied in the field to gain a better understanding of both the overall process and the cause of each one. Newer studies have begun to focus on the early stage quenching of DSFGs around cosmic noon in order to understand the overall evolution of these galaxies and their relationship to the evolution of the universe.

1.2 Cosmic Noon

Cosmic Noon refers to a time period when the Universe was roughly two to three billion years old and galaxies were undergoing peak star formation. While there is not an exact accepted definition, this time period has a redshift range somewhere between 1.5-3. Cosmic Noon followed the period of Cosmic Dawn, and during this period galaxies were growing 100s of times larger than those of their progenitors with star formation rates that had never been experienced before. Astronomers believe that this high growth rate was due to the abundant amounts of cold gas present during the Cosmic Noon era (Nielsen et al. 2020). This allowed galaxies to experience evolution like they never had before, and is also the time period that DSFGs became most prominent.

Cosmic Noon is a critical time period for astronomers to study and understand due to its uniquely high star formation rates and galaxy evolution. This time period also represents the overall shift of galaxies from the high formation rates to the quiescent and settled galaxies that we are more familiar with today. Through the study of Cosmic Noon and the galaxies present during this time period, scientists are able to have a better understanding of galaxy evolution and what causes galaxies to be shaped in the way that they are today.

1.3 How do DSFGs Form?

Dusty Star Forming Galaxies were first discovered with the data of the Cosmic Background Explorer (COBE) in the 1990s. The COBE was launched to observe light outside of the galaxy at IR and submillimeter wavelengths. Before this launch, all the galaxies that astronomers knew of in the early universe had been observed with visible telescopes. With the data collection made by COBE, it was revealed that infrared background light was nearly as intense as visible background light. It was also discovered that only around half of this infrared light could be attributed to known, observed galaxies (Long 2018). Because of this and the fact that dust absorbs most visible light, it was concluded that there must be a significant number of dusty galaxies that are too difficult to observe at optical wavelengths but exist in the far universe. Because of this and the launch of better and better infrared telescopes, astronomers have begun to understand these galaxies better.

Astronomers still do not have a clear explanation for how DSFGs formed in the early universe. The high mass, high star formation rates, and relatively low volume all being at the extreme end of the spectrum has been hard for astronomers to model. It is hard for astronomers to reconcile the formation and the sheer number of DSFGs that were present in the young universe. One idea astronomers are currently considering is that during this period in the Universe DSFGs underwent a high number of mergers with each other, causing them to have even more massive amounts of gas and dust present, allowing for extreme star formation rates (Bail et al. 2024).

However, what astronomers do know is that during the period when DSFGs were most populous, the Universe was primarily neutral, made up of large amounts of cold Hydrogen and Helium gas (Hoffmann 2021). This means that many of the galaxies that formed during this period were very large and gravitationally bound with the rich amount of gas present, ready to form stars from the collapse of the dust. The high amounts of dust was likely a major factor in the extreme star formation rates present, and it represents a process that was overall unsustainable for many of these galaxies in the long run.

1.4 DSFG Light Emission

As with a normal galaxy, dusty star forming galaxies emit light due to the rapid formation and sheer number of stars present within them. These stars undergo nuclear fusion which then emits photons of light. However, in the case with dusty galaxies, a significant fraction of the optical and UV light emitted by stars is absorbed by the dust. Therefore, it is difficult to observe DSFGs in the optical and UV wavelengths. The relative intensity of optical light, which is around 400-700 nm, and UV light, which is around 100-400 nm, of dusty galaxies is much lower (Casey et al. 2014).

To make up for the absorption of visible and ultraviolet light, dust re-radiates the energy of thermal emissions at longer wavelengths, mainly in the infrared region. The wavelength of this light is normally in the range of 3-500 μm—much longer than that of visible or ultraviolet light. This is what caused DSFGs to remain undetected for so long, as astronomers used to mainly look for galaxies in the visible spectrum and did not consider the infrared spectrum as discussed in section 1.3.

1.5 JWST observations

The telescope I am studying, the James Webb Space Telescope (JWST), is still relatively young and has only been observing for a little over two years. However, even with this fact, JWST has made significant observations on DSFGs that have contributed greatly to the understanding of their formation and evolution. More specifically, and as the paper discusses, JWST has done this through the Cosmic Evolution Early Release Science (CEERS) survey (Bail et al. 2024). This survey did not target DSFGs specifically, but it contains a significant amount of data that has enabled astronomers to study them intensively. The main observations made on DSFGs by JWST has been done by the NIRCam instrument, which observes in the near infrared range from 0.6-5.0 μm. As time passes, JWST will likely take closer looks at DSFGs. The specifics and involvement of JWST will be explored more in depth in the next section.

2. Telescope & Instrument

2.1 JWST background

Since the early 1990s, many astronomers have been calling for a space telescope to be developed that observes primarily in the infrared range. Construction for this telescope, which was formally named the James Webb Space Telescope (JWST) in 2002, began in 2004. Webb’s many pieces were constructed at a variety of locations throughout the world, and multiple countries contributed to the final project. Final assembly of all the parts took place starting in 2018 and ending in 2019, representing the conclusion of over 15 years of design and construction. Webb was finally launched on Christmas morning of 2021 where it started its 29 day journey to its final orbit around the second Earth Lagrangian point.

Webb was designed as the largest, most complex telescope ever sent to space. Webb’s main light collecting mirror is composed of 18 segmented hexagonal mirrors with a diameter of 6.6 meters. This design choice was made as it allows the ability to generate high resolution images and collect large amounts of light in the infrared range. Webb also consists of a sunshield that blocks bright visible and infrared light from the sun and other planets and allows it to capture light from distant, faint objects. This sunshield also allows the telescope to remain at a very low temperature, allowing the telescope to collect data better. Webb also has four scientific instruments which will be explored more in depth in a later section. Finally, Webb was designed to fold upon itself. This was done so JWST could fit inside of the launch vehicle, and then it subsequently unraveled itself while it travelled to its orbit location.

JWST was built in order to observe and study the early Universe in a way that had never been done before. Webb was designed to see the earliest stars and galaxies in the Universe in order to study their formation and evolution. Webb, being one of the most powerful and extensive telescopes ever built, has many other goals such as looking for extrasolar planets, studying the evolution of the solar system, and studying the evolution of galaxies from the earliest ones to the massive galaxies we observe today. These goals and design choices allow for the direct observations and study of DSFGs in the early Universe—especially those at cosmic noon. Through advancements in technology and more precise equipment, JWST is able to observe through the dust like never before.

2.2 Instruments of JWST

JWST has four main scientific instruments, each with a specific focus on a different scientific function. Three of the instruments—NIRSpec, NIRISS, and NIRCam—focus on the near infrared wavelength of light, while the final instrument, MIRI, focuses on the mid infrared wavelengths. These instruments are all in fixed positions on the telescope. Each instrument has a unique purpose, and it is often advantageous to use a combination of the four for observations. For this paper, I will be focusing mainly on the NIRCam instrument of JWST. The Near-Infrared Camera (NIRCam) has many capabilities that allow it to be the primary near infrared imager on Webb.

2.3 NIRCam specifics

The NIRCam, which stands for Near-Infrared Camera, instrument detects light in the same way that the other three instruments aboard JWST do. First, light runs into the 6.6 meter diameter primary mirror. From there, it is reflected on to the secondary mirror of Webb. Once reflecting off the secondary mirror, it is then focused into the Integrated Science Instrument Module (ISIM). From the ISIM, the light can be directed to any one of the four instruments, or any combination of the four. After the light is directed at the NIRCam instrument, it then applies any filters, blockers, or focusing required to record the light. As mentioned in the last section, the four instruments on Webb are in fixed positions, meaning they do not have to move or rotate to receive their light from the mirrors. This is the main purpose of the ISIM, which allows the instruments to remain fixed but also receive the light they need in order to make observations.

The NIRCam was designed to be the primary near infrared imager on the JWST. Because of this, it is able to provide high resolution imaging and spectroscopy for a variety of purposes and observations. It is also the only near infrared detector on JWST that has cornographic and time-series imaging capabilities, making it extremely important for many different applications. NIRCam has a large number of filters that it is able to activate or use for the purposes of observations and making specific wavelengths of light more apparent than others. Figure 1 shows the total throughput for each of the filters on NIRCam as well as their operating wavelengths.

In the chosen paper, red, green, and blue images from the F115W, F200W, and F444W were used in order to divide the observed galaxies into separate sub categories based on their colors (Bail et al. 2024). As the figure shows, two of the filters used in the paper were in the short wavelength channel, and the final filter used resides in the long wavelength channel. The F200W and F444W filters have relatively high throughput values, while the F115W filter has values lower than that of other filters likely due to the shorter wavelength of light it is responsible for observing.

2.4 Why NIRCam stands out

When compared to telescopes and instruments that came before it, NIRCam stands out above the rest due to its high number of capabilities and high resolution imaging that it provides. When compared to instruments on previous space telescopes like Hubble or Spitzer, NIRCam stands out for its ability to observe infrared wavelengths of light in a way that no other telescope has before. It does this by combining the high levels of sensitivity, resolution, and field of view, allowing it to stand out above its predecessors in every important category. When compared to Hubble instruments, NIRCam stands out due to its selection of observation wavelengths, which allow it to peer farther back into the early Universe and see more clearly through dusty galaxies. This allows NIRCam to study objects in the Universe that we were never able to before. When compared to Spitzer, which made observations in near infrared wavelengths, NIRCam stands out for its much higher resolution and field of view coverage. This allows NIRCam to study the same objects as Spitzer with much higher detail and coverage, allowing it to be more efficient.

NIRCam is the best instrument at its specific job that has ever been created by humans. From the high sensitivity, resolution, and field of view coverage that it provides, no previous telescope or instrument comes close to the capabilities that JWST provides with NIRCam, allowing us to study astronomy and DSFGs in a way that has never been done before.

3. Results from the Paper

3.1 Observations performed

The observations performed by the author in this paper are drawn from the JWST NIRCam imaging of CEERS. More specifically, the paper dives into data collected during the June 2022 pointings of CEERS, which are responsible for 40% of the area covered by NIRCam between June and December 2022 (Bail et al. 2024). This paper also used preexisting data from the Hubble Space Telescope in the same field. This allowed the opportunity to combine the JWST data with shorter wavelength data and give a more complete picture of the galaxies being looked at. This was mainly done to provide an advantage when doing SED Fitting, which will be discussed further in section 3.3.4.

This paper also employs the use of far infrared observations in order to make the initial galaxy selections. In doing so, they were able to narrow down the selections to only galaxies that were experiencing star formation, which was the main focus of the paper. By using this technique, 22 galaxies were chosen to look at more in depth. After selecting the galaxies that fell into the NIRCam/CEERS June 2022 region, further data analysis was done in order to come to conclusions about the DSFGs.

3.2 Reducing the Data

The paper I chose to write about did not do the data reductions within the study and instead focused more on the analysis of the data rather than the processing. However, it is made clear that they used a customized pipeline that had been previously developed by other researchers and astronomers. From the CEERS catalog, they used the background subtracted images in order to have more precise photometry later on in the analysis. This paper did however perform PSF matching on the data in order to correct the photometry measured by longer wavelength observations. This was done in order to ensure that observations made across different filters were analyzed at the same spatial resolution, allowing for cross referencing and comparison of images (Bail et al. 2024). This allows for a more precise analysis of the data, and eliminates errors that could arise when attempting to fit the SED.

3.3 Interpreting & Analyzing the Data

The paper provides insights into the different techniques used to identify unique characteristics of each galaxy analyzed. The galaxies were displayed as RGB images, where the red is represented by the F444W filter, the green by the F200W filter, and the blue by the F115W filter. From the identification of cores and bulges to clumps within the disk, each of the 22 galaxies was studied individually in order to ensure proper analysis, and the proper analysis could not have been done without the JWST NIRCam data.

3.3.1 Identifying Cores and Bulges

In order to identify the location of the cores and bulges of each galaxy, a visual inspection was made by locating the center of mass. This was done by analyzing the galaxies with the F444W NIRCam filter, which, depending on the redshift of the galaxy, probes the rest frame infrared wavelength between 1.1 and 1.8 μm for this study. This wavelength represents a good tracer of stellar mass, and therefore makes it clear for almost every case where the center of mass of the galaxy is in the RGB image (Bail et al. 2024). In the cases where the RGB image did not make it clear from visual inspection, the isolated NIRCam F444W filter was looked at, and it then became visually apparent where the center of mass of the galaxy was located from this filter.

For this study, a bulge and core both equate to the center of mass of the galaxy. However, a bulge represents a quiescent central concentration, while a core represents a star forming central concentration. The center of mass of most of the galaxies in this sample could not have been identified without the JWST NIRCam data, as the high dust content obscures the observations made by Hubble.

3.3.2 Identifying Lopsidedness

After defining the core/bulge of each galaxy, the rest of the galaxy was then considered to be the disk. In order to study and quantify the lopsidedness of the galaxies, two different properties were calculated for each galaxy. The first property was eccentricity. The eccentricity was calculated using the values of the center of the core, which was defined as the pixel with the highest flux density in the F444W filter (for the reasons mentioned about stellar mass above), and the center of the disk, which was defined as the center of mass when measured in the optical rest wavelength (F150W or F200W filter depending on redshift). The second property considered was asymmetry. This was once again calculated using the F444W filter as it represents the best stellar mass tracer for this study. The asymmetry was found by rotating each image by 180° and subtracting it from the original. For both of the properties, an equation was then applied to find a true value to represent them.

3.3.3 Identifying Clumps

The study also performs a visual analysis to identify clumps within the galaxy. A clump in a galaxy was considered as a visually identifiable concentration that had a different color or a higher magnitude in the RGB image. The clumps were mostly found in the shorter wavelength filters, F155W and F200W, as expected and predicted by astronomers. The number of clumps in galaxies ranged from 0 to 7 in this study, showing the wide range of characteristics of DSFGs.

3.3.4 SED Fitting

In order to extract physical properties of the galaxies such as star formation rates, stellar masses, and dust attenuation, spectral energy distributions (SEDs) were modelled for each galaxy. This was done using a tool called Code Investigating GALaxy Emission (CIGALE) which uses a Bayesian approach to generate probability distributions for each free parameter, of which there were 8 in this study. This allowed for a process that supplied statistically significant values. This was done under assumptions such as a single declining exponential model for the star formation history, fixed solar metallicity of Z = 0.02, and not including nebular emissions in the fit. All of these were justified by previous literature and provided accurate SED fitting. In order to obtain reliable data from the SED fitting, the 𝜒2 was set so that the CIGALE model had to return a value equal to or less than 1.67. If this condition was not met, a higher uncertainty was placed on the flux in each band, which did not impact the final derived values. This iterative process allowed the data and analysis to be more robust and accurate to observational data.

After obtaining the SED models for each galaxy, a two step process was then used in order to estimate the values of the star formation rate and stellar mass (Bail et al. 2024). The SEDs also provided information that allowed the galaxies to be classified into three different types, which will be discussed in the next section.

3.3.5 Classification

In this paper, galaxies were classified based on if the core/bulge, disk, or both were experiencing star formation. The star forming/quiescent characterization was based on three different factors. The first of these was the UVJ color-color diagram to determine if either the disk or core was in the quiescent region. The second factor was based on the position of each component relative to the galaxy main sequence (MS). The final indicator was based on the single exponential model used in the star formation history of the SED models. Each of these three factors were in agreement for the classification of each component in each galaxy except for in one component of one galaxy, where the UVJ color-color diagram did not agree with the other two factors.

After defining the process of determining whether a component was star forming or quiescent, three different types of galaxies emerged from the study. Type I was defined as having a star forming disk and a star forming core. Type II was defined as having a star forming core but a quenched disk. Finally, Type III was defined as having a star forming disk but a quenched core, which is similar to local spirals we observe (Bail et al. 2024). Figure 2 displays a plot of the type of each galaxy in this study and its corresponding redshift. This figure was generated using Python 3 and data provided in the chosen paper.

The figure above displays and confirms information that was previously assumed about DSFGs. The first piece of information it confirms is that, at higher redshifts in cosmic noon, galaxies experienced an extreme increase in star formation rates. This is evident as the figure shows that the galaxies with the highest redshifts are all Type I galaxies where both the core and disk are experiencing star formation. Then, the second piece of information this figure concludes is that galaxies began to slow down their relative formation rates as cosmic noon progressed and the Universe aged. This is shown as a majority of the galaxies at lower redshifts (and therefore towards the end of cosmic noon) are classified as either Type II or III galaxies, where one of the components (core or disk) is quenched while the other is star forming.

4. Conclusion

This report and project represents the connection between DSFGs and the NIRCam instrument aboard JWST. From an in depth look into each component individually to then seeing how JWST helps us learn more about DSFGs, this report helped me to gain a deeper understanding of the systems and processes that dictate both.