REU Physical Sciences Program

PS REU Web Pic

Physical Science Research Experience for Undergraduates Program

The Research Experience for Undergraduates Program in Physical Sciences (Earth and Planetary Sciences and Astrophysics) is funded by the National Science Foundation. The Museum's Division of Physical Sciences—in collaboration with the City University of New York (CUNY)—is pleased to offer summer undergraduate research opportunities in Astrophysics, and Earth and Planetary Sciences.

Our program brings approximately eight students to the American Museum of Natural History in New York City each summer for a ten-week experience working with our curators, faculty, and post-doctoral fellows. Students receive a $5000 traineeship stipend, as well as per diem costs for housing and meals, relocation expenses, and transportation subsidies. Housing is made available at nearby Columbia University.

In addition to conducting original research projects throughout the summer, students participate in a series of weekly meetings at which they discuss their research, present informal progress reports, and engage in discussions and seminars regarding scientific research, graduate school, and research career opportunities. At the conclusion, they deliver oral presentations of their work and prepare publication quality research papers. The program is open to all students who are U.S. citizens or permanent residents, in any two or four year undergraduate degree program, who will not have completed a bachelor's degree before September 1, 2021.  

Who Should Apply

All students in the program must be U.S. citizens, U.S. nationals, or permanent residents of the United States. Students must be entering or continuing in an Associates or Baccalaureate degree program following their summer internship. As part of the National Science Foundation's commitment to broadening participation in STEM fields, we especially encourage students who come from community colleges, undergraduate-only institutions, and minority-serving institutions to apply.  Applications will be accepted from January 1-January 31, 2022.


For assistance with application process, contact [email protected]

2022 Physical Sciences Project Titles 

(includes Earth and Planetary Sciences and Astrophysics)

Note: We encourage interested students to check the web site frequently since we expect to update it with additional projects through the winter


Crystallization of Barred Olivine Chondrules


Advisor: Michael Weisberg (Professor, CUNY) with Mabel Gray (CUNY)

 B. Photo: Magnified chondrule

Chondrules are small igneous rocks found in primitive meteorites called chondrites. They formed as free floating droplets in the early Solar System, either in the early Solar Nebula or in vapor clouds produced from collisions of early-formed planetesimals. Chondrules show a wide range of textures with barred olivine being among the most unusual. Barred olivine chondrules have been of interest to petrographers and experimentalists. Their conditions of formation can provide important constraints for understanding the chondrule-forming environment. This project will be to re-evaluate the textural variations among the barred olivine chondrules, make comparisons to experimental studies and understand the crystallization of barred olivine chondrules as a function of their thermal histories. The work will include use of the petrographic microscope, scanning electron microscopy with backscatter electron imaging, electron microprobe for element mapping, and possibly electron backscattered diffraction. The student will receive a "crash course" in meteoritics as part of the project.


Diagnosing Filamentary Structure of Atomic Gas in the Milky Way

Illustration of the Milky way as seen from the side.
This is what the Milky Way might look like from the side. It's like a giant disk with a bump in the middle!
Sideview of Milky Way, ©Paternstro

Advisors: Dr. Mordecai Mark Mac Low  with Dr. Rowan Smith, U. Manchester; Dr. Juan Diego Soler (MPIA/U. Rome)

Radio observations of atomic hydrogen in the Milky Way Galaxy reveal long filamentary structures. Although the majority of these structures are aligned parallel to the Galactic plane, there are regions where they are instead oriented perpendicular to the plane. It is hypothesized that these perpendicular regions contain multiple supernova blast waves combining to drive a superbubble bursting out of the Galactic disk. In this project the student will apply the statistical analysis tools developed for the observations to synthetic data derived from numerical simulations of the Milky Way that include gas and supernova explosions, as well as much other relevant physics. In the simulations, the times and positions of supernova explosions are recorded, allowing a direct test of the hypothesis to be made.


High Energy Astrophysics

Milkyway expansion with red and blue shift.
A depiction of energies in the sky above 1 GeV.
Copyright AMNH

Advisor: Tim Paglione

Gamma-rays probe the most energetic processes in the universe. Most gamma-rays are created by light and matter interacting with cosmic rays, particles accelerated to nearly the speed of light by supernova explosions and the pulsars they leave behind. Our group uses over a decade of data from the Fermi Gamma-Ray Space Telescope to stack signals from any and all potential sources of gamma-rays including pulsars, hot stars, and a variety of interacting binaries and other exotica (even Jupiter!).


Identifying the Biogeochemical Signals of Eddies in the Bermuda Atlantic Time Series

A cyclonic/anti-cyclonic depiction of the Gulf Stream in the month of May off the coast of Florida and the Caribbean
Surface relative vorticity and tracks of modeled eddies in the Gulf Stream region at the end of May.  The boundaries of the analysis region (black lines) and the location of BATS site (cross) are shown.
Dr. Nathalie Goodkin

Advisor: Dr. Nathalie Goodkin (EPS)

Mesoscale eddies are prevalent in the Sargasso Sea near the Bermuda Atlantic Time Series (BATS) station.  While eddies have themselves been well studied, little is known about how eddies propagate on-shore.  Evidence for the influence of eddies exists in the paleo-proxy records near Bermuda (Goodkin et al., 2008, Goodkin et al., 2015).  In this project, the student will mine the BATS data combined with eddy tracks to identify months when eddies were most prevalent in the Sargasso Sea.   Using the BATS data, the geochemical signal of the identified eddies will be found and compared to paleo-proxy records to evaluate the potential of paleo proxies to identify years of increase eddy frequency.  The student will be invited to contribute to field work in Bermuda though this is dependent on the COVID situation.


Calibrating Corals for North Atlantic Oscillation Reconstructions

Two images, one a map of the world indicating winter surface temperature and another image of a coral surrounded by fishes.
a) Map of correlations between winter surface temperature and the North Atlantic Oscillation.  Circle indicate published marine records and the star indicates the location of this record. B) Photo of a coral that has been sampled in Tobago.
Dr. Nathalie Goodkin

Figure 1: a) Map of correlations between winter surface temperature and the North Atlantic Oscillation.  Circle indicate published marine records and the star indicates the location of this record. B) Photo of a coral that has been sampled in Tobago. 

Advisor: Dr. Nathalie Goodkin (EPS)

Marine carbonates record chemical changes based on the environment in which they form.  In corals this includes the ratio of strontium to calcium (Sr/Ca) and oxygen isotopes (δ18O). Sr/Ca changes in coral skeleton based on the temperature of the water and δ18O changes with both sea surface temperature and salinity.  By examining these chemical records back through time we are able to unlock earth’s climate variability including systems like the North Atlantic Oscillation (NAO).  The NAO is an atmospheric oscillation that drives surface temperature, storm tracks and wave heights for example from the United States to Europe leaving a clear imprint on sea surface temperature (Fig. 1a).  Many reconstructions exist from trees and corals, but no records exist from the negative focal point in the southern Caribbean Sea.  In this project, the student will study the chemical proxies of a coral from Tobago (Fig. 1b) to calibrate the key environmental proxies using gridded instrumental data of sea surface temperature and salinity and self-generated chemical proxy data. Work will be conducted in collaboration with Dr. Nathalie Goodkin and graduate student, Ross Ong.


An inference methodology to predict solar neutrino flux


An image of Sol up close, with noticeable surface features.
A depiction of the surface of a star.
Eve Armstrong

Advisor: Eve Armstrong

Neutrinos are elementary particles created in abundance in the core of the Sun, and play a role in the nuclear fusion reactions responsible for releasing light.  To better understand the physics of those reactions, it is important to know the evolution of neutrino “flavor” throughout the Sun.  Neutrinos come in at least three flavors, a property that defines the manner in which they interact with the surrounding neutrons, protons, and electrons – and a neutrino’s flavor evolves as it propagates.  We aim to map the complete flavor evolution of neutrinos as they emanate from the Sun’s core outward. 

This aim is a challenge, for two reasons.  First, neutrinos interact weakly with matter and thus are extremely difficult to detect.  The data obtained by current detectors are sparse.  Second, current methods of studying the physical model of flavor evolution – specifically, numerical integration techniques – possess limitations, including rigidity and high computational cost.  For these reasons, we are developing a means to augment those existing codes, via an inference formulation.  Inference is a means to efficiently optimize a model given available data.  We use a specific type of inference, called data assimilation, that is designed for the case of extremely sparse data.

The student will use Python to learn two distinct techniques for solving a physical model: 1) numerical integration and 2) optimization-based inference.  Our aim this summer is to demonstrate that the latter technique formulates the mathematical problem in a manner that renders the physics easier to access.  The student will also learn to use Unix and a supercomputing cluster.

Please sign up for the mailing list for updates: