Research

Data-Intensive Astrophysics

Astronomy has ushered in the big data era. In the past two decades, multi-wavelength large surveys have collected tremendous amount of data across the sky, transforming astronomical research from data-limited to data-rich science. Such large volume datasets enable a new data-intensive approach of performing astronomical research, which focuses on addressing astrophysical questions by designing innovative methods for extracting novel information from big datasets.

My research leverages this approach to explore and understand various components of the Universe, including gas, molecules, dust, stars, black holes, magnetic fields, and dark matter halos and the interplay between these components through cosmic time. I have utilized and combined massive datasets from several sky surveys, including GALEX (UV), SDSS, HSC, DESI Legacy Surveys (Optical), WISE (Infrared), Planck , LAB 21cm, FIRST, NVSS, LOFAR (Radio), to extract rich astrophysical information, to reveal novel astronomical phenomena, and to explore the physical mechanisms driving those newly discovered phenomena. In addition, I am keen to use the massive spectroscopic datasets from two ongoing/upcoming surveys, DESI and PFS, in which I have been participating. In the following, I will describe my research projects with this data-intensive approach.

Survey and telescope datasets that I use to conduct data-intensive research.

Large Sky Surveys

To perform data-intensive research, big data with rich information is needed. I have been participating in the development of large sky surveys to acquire such massive datasets. Currently, I am participating in the development of the DESI and PFS surveys which will obtain tens of millions of spectroscopic measurements of stars, galaxies, and supermassive black holes. I can not wait to use these datasets to not only explore the astrophysics of galaxy evolution but also reveal new fundamental physics that drives the accelerating expansion of the Universe.

I have been leading the visual inspection efforts of all types of galaxies observed by DESI and utilizing the VI catalogs to inform the target selection criteria and quantify the expected performance of the DESI main survey. The results are summarized in Lan et al. (2023) and here.

Top: SDSS / Bottom: DESI

A diagram of the circumgalactic medium from Tumlinson, Peeples, and Werk (2017).

The Circumgalactic Medium

Gas around galaxies, the circumgalactic medium (CGM), contains signatures of gas flow processes which have been considered as important mechanisms driving galaxy evolution.

Observations: By utilizing massive datasets, I have been investigating the properties of the CGM and its connection to galaxies. My research has revealed several new findings, including

New correlations between the properties of galaxies and their CGM (Lan et al. 2014, Lan & Mo 2018)
Small-scale structure of the CGM (Lan & Fukugita 2017)
The co-evolution of galaxies and their CGM detected for the first time (Lan 2020)
Novel constraints on the magnetic fields of the CGM (Lan & Prochaska 2020)

These observational results have challenged the current models of galaxy evolution and thereby motivated new theoretical investigations (see Nelson et al. 2020 as an example). Moreover, the science and the developed techniques are the main drivers for the next generation sky surveys, including the DESI Legacy Imaging Surveys and the Prime Focus Spectrograph Survey, in which I have participated in their development.

Theory: Besides observational studies, I have developed semi-analytic models to investigate the physical mechanisms that give rise to the observed properties of the CGM (Lan & Mo 2019). In addition, I am also working on the formation and evolution of the circumgalactic dust (Hirashita & Lan 2021).

The CGM is one of the active and important research topics suggested by the Decadal Survey 2020!

Galaxy Evolution in Dark Matter Halos

Galaxies form and evolve in dark matter halos. Studying the properties of galaxies in dark matter halos will therefore help us to better understand the complex processes involved in galaxy evolution. I have been investigating the properties of galaxies in dark matter halos with statistical methods and obtaining unprecedented measurements which provide novel insights on the formation mechanisms of galaxies. For example, in Lan et al. (2016), I characterized galaxy population in dark matter halos with a mass range over three orders of magnitude and used this measurements to constrain models of galaxy evolution (Lim, Mo, Lan, Ménard 2017).

I plan to push such measurements to fainter galaxies and at higher redshifts with the survey datasets, including HSC, PFS, and DESI. This will reveal how galaxies and their dark matter halos co-evolve through time and the corresponding physical mechanisms.

An image of a galaxy cluster from HSC

A giant radio galaxy observed by LOFAR (Lan & Prochaska 2021).

Supermassive Black Holes

Over the past two decades, astrophysicists have realized that the supermassive black holes in the center of galaxies play a crucial role in driving galaxy evolution. The supermassive black holes are expected to eject substantial amount of energy and mass out of galaxies. This so-called feedback mechanisms remove gas from galaxies and maintain the heat content of the CGM of galaxies. These processes quench the star-formation activity of galaxies and transform star-forming galaxies into passive ones.

Radio jets and lobes around galaxies are one of the key observational signatures of black hole feedback. By studying the properties of radio jets, one can shed new lights on the feedback mechanisms. To this end, I have been investigating how these radio jets influence the properties of galaxies and why radio jets have different sizes with a new radio survey dataset (LOFAR). In Lan & Prochaska (2021), I measured the environment of an extreme population of radio galaxies, called giant radio galaxies, and showed that environments do not play a role on the sizes of radio jets. This new measurement places a strict constraint on the models of the origin(s) of the giant radio structures.

Mysterious Molecules in the Universe

DIB map (interactive map) from Lan, Menard, and Zhu (2015) and Zasowski, Menard et al. (2015)

About 100 years ago, Mary Heger discovered some absorption lines, called diffuse interstellar bands (DIBs), induced by unidentified interstellar materials in our Milky Way (read the history of DIBs). Since then, the origins of DIBs have been a puzzle. Until today, more than 600 DIBs have been discovered.

In 2015, a group of chemists confirmed that C60 is the carrier of 2 DIBs. This result shows that those DIBs are induced by complex large molecules in the space and suggests that there is a reservoir of complex molecules like C60 between stars which may play an essential role in the formation of stars, planets, and human beings.

I have been interested in these puzzling DIBs. To study their nature, in Lan et al. (2015), I developed a statistical technique and applied it to the SDSS spectra of stars, galaxies, and quasars. By doing so, I produced the largest map of the distributions of about 20 DIBs across the sky and used this map to explore the properties of DIBs. The results revealed that the carriers of DIBs tend to live in different environments and informed us about the formation and destruction mechanisms of those mysterious molecules.

I plan to continue to explore DIBs by using the upcoming DESI spectra which will enable me to produce even large DIB maps and to reveal the 3D structure of such mysterious molecules across our Milky Way!

Machine learning applications in Astrophysics

In order to extract astrophysical information from huge-volume astronomical datasets, I have been using machine learning methods to largely increase the efficiency of the data analysis process. For example, recently, I am using deep learning methods, such as convolutional neural networks, to explore the CGM absorption line features. I am interested in developing and incorporating novel machine learning techniques in my explorations of the Universe.