This website has moved to The current site will be disabled on 27 April 2018.
You should update any bookmarks to go directly to the new website.

STScI Logo

Space Astronomy Summer Program
Project Proposals

3D Astronomy: Tactile Representation of HST Data

The 3D Astronomy project makes galaxy and star cluster imagery from HST observational programs available in 3D tactile representations. The prints are produced through a new, innovative process developed by Christian, Nota, Grice, Shaheen, and Sabbi and their collaborators for HST1 Cycle20 galactic star clusters (Westerlund 2 [W2] and NGC 602) and extended through the LEGUS multi treasury program. We produce 3D prints of HST data for use by visually impaired individuals, and tactile learning strategies to stimulate a better understanding of Hubble research, astronomy, and STEM in general. Experience in astronomy, assistive technologies and/or programming preferred.

Calibration of Brackett Alpha as a Protostellar Accretion Indicator for JWST

How stars build up their main-sequence masses is not yet fully understood, but the answer lies in the spectra of accreting protostars. Emission lines of atomic hydrogen in these systems have strengths that are proportional to the rate at which mass is accreting onto the star. With mid-IR spectroscopy, JWST will extend the analysis of these lines to protostars that are too deeply embedded in their natal material to have been studied at shorter wavelengths. It will easily detect the Brackett alpha line of atomic hydrogen, but calibration of this poorly understood line before launch is essential for interpreting forthcoming spectra. The student will analyze existing spectra from a ground-based telescope to compare the well-studied Brackett gamma line to Brackett alpha, and deduce the relationship between luminosity and accretion rate for the latter line.

Designing Discoveries – Creating Beautiful, Usable Tools for Astronomers

The MAST archive is an important tool for astronomers conducting research about the universe. It has a tremendous number of features and is an incredibly powerful tool for gathering data. Unfortunately, the website is lacking in usability and visual design. In order to make the archive more learnable and easy to use, it will be redesigned. Because of the wide scope and complexity of this redesign, there are a number of interesting projects the intern may choose to work on. The intern might choose a feature to redesign, creating wireframes and UI Specifications; run a usability test, gathering data on how astronomers use the archive; or they could design and create an updated set of icons used in the interface. Must have visual design experience.

Discovering Things That Go Bump in the Night with Pan-STARRS

We will use machine learning and data science techniques to classify light curves of variable and transient sources—things that go bump in the night—in Pan-STARRS. The Pan-STARRS Medium Deep Survey hosted at STScI is the best source of archival multi-band photometry prior to LSST. However, groups have largely only exploited the data to find supernovae. We will use unsupervised and semi-supervised learning techniques to classify the Pan-STARRS Medium Deep Survey data, creating a catalog of all variable and transient sources with average S/N > 5. We will then apply the algorithms we’ve trained on this archival data to current data taken with Pan-STARRS, to try to discover new sources that also fall within the footprint of the Kepler K2 mission. Prefer a student comfortable with Python, ideally having already done some research with Python.

Enable Outreach Products for the World Stage

The intern will contribute to the production of outreach websites, including The intern will experience the full suite of production practices—organization and management, content curation, development, implementation—for websites and virtual reality products. If the intern plans to go into scientific outreach in the future, they will develop a better understanding of the skills, roles, responsibilities, and requirements for successful online products in the science field. Prefer a student with scientific or engineering background.

Exploring Stellar Binarity of Fast Rotating Red Giants

There is a population of red giant stars that show moderately fast rotation. These stars are interesting because of the possibility that they have been spun up by the engulfment of a massive exoplanet companion. However, this interpretation must first rule out the possibility that the star interacted with a stellar companion. For the proposed project, the student will work with a sample of red giant stars in the field and in open clusters that have previously-measured rotation speeds, and for which multiple radial velocity measurements have been obtained. The student will first conduct simple tests to determine which stars shows evidence of radial velocity variability and will relate this variability (or lack thereof) to other stellar properties (e.g. the rotation velocity). More advanced students may progress to studying individual stars in more detail to measure or constrain the orbital elements of identified stellar companions.

Exploring the Hubble Images with Neural Networks

Hubble has produced many beautiful images of wispy, cloudy, and luminous structures that help us understand how stars and clouds form, how galaxies evolve, and how magnetic fields work in deep space. Unfortunately, there is no way to search the Hubble archive for such images, which makes it difficult for researchers to find interesting targets to explore further. We are developing a prototype tool that would allow users to find images similar to an image of interest, based on deep transfer learning (neural networks). Neural networks are becoming very popular tools in astronomy right now, based on their ability to work directly with complex image data. The summer student could work on many interesting aspects of this project, including: building better deep networks, exploring a range of machine learning clustering algorithms, building graphical user interfaces, developing citizen science tools for making validation data sets, and building science cases for the tool.

Exploring the Role of Galaxy Morphology in the Mass Metallicity Star Formation Rate Relation

The mass-metallicity-star formation rate (M-Z-SFR) relation for galaxies gives insight into the accretion and outflow of gas, which drives the formation and evolution of galaxies. In this project the student will investigate how the morphology of galaxies affects their location on this relation, which is believed to form a plane, and thereby address the role that the structure of galaxies play on the M-Z-SFR relation. In particular, they will investigate what physical properties (such as metallicity, mass, and ionization state) are responsible for the deviation of galaxies from the plane, and what it tells us about galaxy gas reservoirs, inflows, and outflows. Basic Python programming and plotting experience would be very useful but not required.

Galaxy Spectroscopy at Half the Age of the Universe

With the LEGA-C survey, we have obtained about 2000 high-resolution, very high-quality spectra of galaxies. Such spectra allow us to measure the properties of these galaxies with unprecedented precision. The candidate student will assess the quality of the measurements and find correlations among such measurements and other quantities like the morphology and environment of the galaxies. Working on the visualization of such results will be particularly important. The student will acquire knowledge on galaxy evolution, spectroscopy, and statistics. Coding experience in Python would be appreciated but not required.

Illustrating the JWST NIRISS Scientific Revolution

The NIRISS team seeks a student who is interested in learning about the ways in which the NIRISS instrument on JWST will push the boundaries of our current understanding of the universe. The NIRISS instrument is an imager (including aperture masking interferometry) and slitless spectrograph that is expected to revolutionize both the science of exoplanets, by determining the atmospheric composition of transiting planets and directly imaging planets, and early universe science, by being able to get spectra of the earliest galaxies. The summer student will be asked to create narratives on the leading science cases for NIRISS, speaking with experts (inside and outside STScI, facilitated by the mentors) about the current state of the field, the capabilities of NIRISS, and the possibilities for the future. The student will be asked to create these narratives, including enticing graphics and scientifically relevant data, to make the community understand just how groundbreaking this new instrument will be. Specifically: 1) with the NIRISS team and mentor, the student will identify three or four groundbreaking science cases that will move the field forward through NIRISS observations; 2) prioritize the list of science cases based on community excitement and student enthusiasm; 3) research the top priority science topic and build an understanding of where the field is, and specifically how NIRISS will advance this science narrative; 4) create the narrative, including putting together text describing the ways in which NIRISS advances the science, including developing the text and putting together graphics (this can include schematics, real-world science images, and artistic renditions. If the student finishes the top priority narrative, the student can then move to others in the ranked list, and repeat! We are interested in a broad swath of students, including visual arts students.

Improvements to STIS Echelle Flux Calibration

There are a number of issues affecting the STIS echelle flux calibration. These include misalignment of the blaze function, whole spectral orders that are on the detector but missing from the calibration, focus dependent aperture throughput, and some detector edge effects that weren’t calibrated correctly in the first place. The first part of this student project would be to systematically review existing STIS echelle spectra under the supervision of STIS team members to quantify and systematically document these effects—which missing orders are worth adding back to the calibration, what possibly correctable trends can be identified in the blaze function alignment offsets, and which settings are affected by poor calibration in the edge orders. A next step would be to create updates for those problems where the best solution is simply an easy update to the existing calibration files. As results of this analysis are obtained, the student will also be continuously guided towards writing them up in a form suitable for inclusion in an Instrument Science Report on which the student will appear as either co-author or perhaps even first author, depending on the details of their contribution. Such a document would be very valuable to the STIS user community as it could indicate where particular calibration problems are to be expected, and suggest work-arounds. As such, it is likely to generate at least a few citations. While this project will be focused primarily on analysis rather than on the development of software tools per se, the student will be guided to do the analysis in a scripted manner that is easily repeatable and extensible, and may develop some reusable utilities to assist in the analysis. Desired skills: some modest programming experience in Python and some knowledge of astronomical spectra.

Injecting Cosmological Supernovae into HST Images

We are attempting to very precisely measure the expansion of the universe with time using Type Ia supernovae as our tools of measurement. This requires very precisely determining the brightness of the supernovae, and accounting for any errors the telescope and our observing techniques might be introducing. In this project, we will inject artificial supernovae in the HST images that contain our real supernovae, and attempt to recover the brightnesses (photometry) of the artificial supernovae. Since we know the true brightnesses of the artificial supernovae, the differences between what we observe and what we expect will give us an accurate assessment of many of the errors that our observing technique could introduce, and thus allow us to better estimate the expansion history of the universe. Programming and/or Linux experience is preferred but not necessary.

Multi-wavelength Observations of Star Formation in the Magellanic Clouds

Among nearby galaxies, the Magellanic Clouds are some of the best astrophysical laboratories for studies of star formation and the interstellar medium. Their proximity allows us to probe the interplay of gas, dust, and stars in an extragalactic environment on unprecedented scales. Using large surveys of the Magellanic Clouds, we have identified thousands of Young Stellar Objects across the Magellanic Clouds, which are grouped in hundreds of star forming complexes in a wide variety of masses, sizes, and evolutionary states. From these targets, we have selected two regions in the LMC and SMC which will be observed during our James Webb Space Telescope GTO program. This project will study these regions in great detail using multi-wavelength observations, with the aim to improve our census of the stellar content of these regions. The work will provide critical information to be used in the analysis of the upcoming JWST observations.

Science Writing and Communication of Astronomical Discoveries for the News Media

The intern will contribute to the production of written materials describing astronomical research to news media, as related to the Hubble and Webb space telescopes. The intern will experience the full suite of popular writing venues: press releases, fact sheets, web stories, and video scripts. If the intern plans to go into scientific communication in the future, they will develop a better understanding of the skills, roles, responsibilities, and requirements for successfully working with journalists. We're looking for an intern with a journalism background, with an affinity for grasping and translating complex science topics for the public.

Search Engine Optimization for Outreach Websites

In a world where the collective knowledge of generations is at your fingertips, it is easy to be overwhelmed when attempting to find information on a subject matter. In addition to finding the correct information, it is equally important to find the information you’re looking for in a timely manner. Enter Search Engine Optimization, a set of methodologies, strategies, and tactics to enhance the value of search results returned by a search engine. Our front-door website, along with our outreach websites, will all soon be powered by Jahia. Jahia is an enterprise-level content management system, or CMS. Jahia is bundled with Apache Lucene—a high-performance, full-featured text search engine. Lucene includes powerful, accurate, and efficient search algorithms that support ranked searching, many query types, fielded searching, and more. While Jahia provides an out-of-the-box configuration of Lucene that will go live with our websites, we want to utilize the full power of Lucene. To measure the effectiveness of Lucene, a baseline will be established by setting metrics on the resulting data sets from queries performed. By establishing a baseline metric, we can then complete an assessment of performance to be prepared for analysis. The assessment of performance will then allow the student to investigate, compare, and implement additional strategies and algorithms to optimize Lucene beyond the out-of-the-box configuration.

Temperature of an Important, Cold Exoplanet

GJ1214b is the smallest exoplanet whose atmosphere our telescopes can examine using indirect methods—namely transits and eclipses. Many teams have tried to uncover the exoplanet’s physical properties, but to date, the only measured information that we have is its mass and radius. If this exoplanet behaved as we expect using Solar System models, then we should have been able to spectroscopically probe the amount of water and methane in its atmosphere, as well as measure its dayside temperature. I have data that I used to measure its radius to very high precision. This same data can be used to measure its temperature to reasonable precision; it is a much smaller signal. My team tried to detect the temperature of the exoplanet with our data set, but the techniques used at that time were unable to tease the signal from the noise; we published an upper limit. There have been two generations of noise models since then that I would like to apply to the data in order to attempt to constrain the atmosphere’s temperature. Knowing its temperature sets an anchor for its theoretical atmospheric chemistry. This would go a long way for predicting future observations of this exoplanet or other exoplanets similar to it. It would also add only the third measured data point of thousands of lost signals.

Time Dependency of ACS/WFC Geometric Distortion

The HST ACS/WFC imaging instruments have a considerable geometric distortion due to the optical assembly of the HST. In addition to a large geometric distortion, ACS/WFC also has a time-dependent distortion. The size of this change is clearly noticeable over 5 years, reaching about 15 mas (0.3 pixel) off from the original 2002-based distortion solution. The newly derived ACS/WFC geometric distortion (2015) significantly improved the correction for time dependent distortion. Therefore, it is important to examine and monitor the ACS/WFC distortion and forecast the evolution of the scale of the geometric distortion with time using a longer 15-year baseline. Any changes in the linear part of the geometric distortion with time would also be an indication of mechanical, optical, and thermal changes in the ACS/WFC camera itself. The ACS/WFC observations of the stellar cluster 47Tuc over 15-year period will be used to examine the variations of the astrometric X&Y-scale over time. Knowledge of basic Python, IDL, or FORTRAN programming languages, and basic knowledge of IRAF/PyRAF, or any experience with astronomical images analysis, is a plus.

Visualization Tools for Observation Planning

The intern will work with STScI engineering and operations staff to investigate new tools and technologies for potential integration with an artificial intelligence based planning and scheduling system used for HST and JWST operations. Duties include: performing trade studies evaluating new technologies; designing, coding, and evaluating prototype implementations integrating new technologies with existing STScI software; and communicating with engineering staff on the progress of the studies, including the preparation of a final report. Students must have some ability to develop software.