Some of the galaxies/halos in the Caterpillar suite of simulations. We identify and investigate the ultra-faint dwarfs in each of these halos. Image from Griffen et al (2016).


Kinematic clustering of ultra-faint dwarf galaxies

 Summer 2021

 Throughout its evolution in cosmic time, the Milky Way has accreted (or “devoured”) many smaller dwarf galaxies. Since these accretions happened millions to billions of years ago and at really long timescales, we cannot observe these accretions in real time. So how can we possibly reconstruct the Milky Way’s accretion history? Astrophysicists do this by observing the stars of now-accreted dwarf galaxies and looking at their kinematics and chemical abundances. In particular, they look at how these stars cluster together when their kinematics are plotted.

But how reliable is this method? This is the main question in my project. The challenge in using observations is that we cannot see the accretion processes as they happen. However, with simulations, we can accomplish exactly that. Using the Caterpillar simulations, we see the evolution of Milky Way-like galaxies from right after the Big Bang to the present day. As a result, we can trace exactly which dwarf galaxies got devoured by the Milky Way.

From this, we then use the kinematic clustering technique used by the observers on our simulated stars/galaxies. Since we know exactly which stars were accreted, we can validate and evaluate the current kinematic clustering techniques used on stellar observations. 


Environmental differences in ultra-faint dwarf galaxy formation

 Summer 2020

See poster above for more information.  For this project, I worked with Professor Anna Frebel and Kaley Brauer at the MIT Kavli Institute for Astrophysics and Space Research (MKI). I presented this work to MKI scientists and students at the end of Summer 2020. 

See also: 

(Clockwise) Red elliptical, blue spiral, red spiral, blue elliptical. Image from Galaxy Zoo.


Investigating the origins of red spiral galaxies in the IllustrisTNG simulations

 Spring 2020

 Galaxies are broadly classified by color and morphology. Color can be used as a crude indicator of the age of the galaxy’s stellar content. A blue galaxy is generally younger and more active at star formation because its emission is dominated by hot, blue stars which are relatively short-lived and therefore young. In contrast, a red galaxy is older and inactive because its stellar population is mostly composed of dim, red stars which have already burned through their fuel. On the other hand, a galaxy’s morphology–whether it is spiral or elliptical–contains information about its dynamical history.

 Interestingly, most observed galaxies follow a color-morphology relation such that spiral galaxies are typically blue and elliptical galaxies are typically red. However, data from the Sloan Digital Sky Survey reveal hundreds of red spiral galaxies–a population which deviates from the classification scheme.

Now, how exactly did these red spirals form? Because of their anomalous nature, red spirals might reveal new insights about galaxy formation. For this project, I used data from IllustrisTNG, a state-of-the-art cosmological hydrodynamic simulation.  From the galaxies in IllustrisTNG, we identified red spirals using color and morphological criteria. From there, we plotted their physical properties to see how they formed. 

 For this project, I worked with Professor Mark Vogelsberger and Hui Li at the MIT Kavli Institute for Astrophysics and Space Research. This project was under MIT’s Undergraduate Research Opportunities Program (UROP).

See also: 


Communicating science to the public through video: Experiences and techniques from a Breakthrough Junior Challenge winner

Andales, H.D. (2018)
Proceedings of the 36th Samahang Pisika ng Pilipinas (Physics Society of the Philippines) Physics Conference


More than ever, scientists are compelled to share relevant scientific knowledge to the public in order to address misinformation, inform decision-making, increase public support for scientific ventures, or promote widespread appreciation for science. Through accessible media production technology and social media, it is now much easier for scientists to be communicators and for the public to be receivers.

In this talk, I will elaborate my experiences, strategies, and lessons in video-based science communication as the sole researcher, writer, producer, and creator of two short Physics videos: (1) Feynman’s Path Integrals and (2) Relativity and the Equivalence of Reference Frames. The former was awarded the Special Prize during the 2016 Breakthrough Junior Challenge. Meanwhile, the latter video was declared the Winner of the 2017 edition of the same competition, where it bested 11,000 entries from 178 countries. This second video has accumulated over 5,000,000 views across all platforms.

The talk will also detail how I created the videos, how the public received and interpreted the videos’ information, as well as how scientists can use video for science communication.

See also:
H. D. Andales. Communicating science to the public through video: Experiences and techniques from a Breakthrough Junior Challenge winner, Proceedings of the Samahang Pisika ng Pilipinas 36, SPP-2018-INV-1C-02 (2018)


N-body simulation of celestial bodies with differentiable motions in 2D space

Andales, H.D. & Roque, P.J.C. (2017)
Proceedings of the 5th International Meeting of Complex Systems

My poster for the internship’s culminating poster presentation; received Best Poster Award



Fractal Visualization in Mathematica & Parallel Computing in C

Andales, H.D. (2015)
National Institute of Physics Interns’ Poster Presentations

In the summer after 9th grade, I interned at the National Institute of Physics in Diliman, Quezon City, Philippines. I worked with Professor Francis Paraan in the Structure and Dynamics (SanD) Lab.  During the two-week program, I completed two projects: MandelZoom and Function Integration Through Rectangle Rule.

MandelZoom is a fractal visualization tool created using Mathematica while Function Integration computes integrals numerically using MPI (message passing interface) in C. At the end of the internship, I received the Best Poster Award out of the >20 interns in the program.

SanD interns presenting their posters. Image from SanD.