Authentic Astronomical Research as Science Teacher Professional Development

Authentic Astronomical Research as Science Teacher Professional Development

In the summer of 2019, four science teachers, members of the EXES Teacher Affiliate program of the University of Texas at Austin, participated in a week-long session at McDonald Observatory, Ft. Davis, Texas. The purpose of this research was two-fold: to give teachers the experience of collecting data using the 2.1 meter Otto Struve Telescope and to use that experience to create new curricula for the classroom. The data collection and observation was performed by a small group of cohort members.

The EXES teacher group was founded by Dr. Mary Kay Hemenway at the University of Texas at Austin (now retired) as part of the education and public outreach (EPO) for the Echelon-Cross- Echelle Spectrograph (EXES) instrument designed to fly on the Stratospheric Observatory for Infrared Astronomy (SOFIA). Currently the program is managed by Dr. Keely Finkelstein in the Department of Astronomy at the University of Texas at Austin and recently additional funding has come from Dr. Chris Sneden in the astronomy department. Much of Dr. Sneden’s work is in stellar chemistry using high-resolution spectroscopy.

Stellar Alchemy: Nucleosynthesis

Stellar nucleosynthesis is how stars produce elements heavier than hydrogen. Stars consist almost entirely of hydrogen when they form. There are a variety of ways to convert hydrogen into heavier elements. Fusion is the main process of helium creation in a star like the sun, but where do all the other elements come from? In recent decades, work in astronomy has been able to narrow down the mechanisms that create elements from the lighter stuff to the honking big nuclides like lanthanum.

One of the most interesting discoveries of the last decade was the multi-messenger analysis of a neutron star merger ​(Abbott, Abbott, Abbott, Acernese, Ackley, Adams, Adams, Woudt, et al., 2017)​ and ​(Abbott, Abbott, Abbott, Acernese, Ackley, Adams, Adams, Zweizig, et al., 2017)​. Having data from such an event covering multiple types of light, meant the interplay between the different physical phenomena could be explored. This type of merger is now thought to be one source of many heavy elements. One pathway for the formation of heavy elements if the r-process of neutron capture ​(Sneden, Cowan, & Gallino, 2008)​. The formation of heavy elements in earlier stellar populations leads to the elements presence in modern stellar populations. One group of heavy elements that can be seen in stellar atmospheres, and therefore a perfect task for high-resolution spectroscopy, are the lanthanide elements.

Where are the stars with lanthanum?

There is an effort by astronomers to characterize the stars in our galaxy that contains lanthanides. The ratio of heavier-then-hydrogen elements to hydrogen is called metallicity. If we compare the metallicity of star to the abundance of a known lanthanide element, europium, we can then compare different populations of stars. The stars that like the sun and have a lot of metals, the stars with some metals but less than the sun, and those stars that are very metal poor.

If each category of metallicity can also have a corresponding lanthanide abundance, it will help scientists explain where the heavy elements came from and how early in the history of the galaxy they appeared.

We were tasked with finding stars with fewer metals than the sun but also containing a noticeable amount of europium. This is a stellar population that needs more observations and we willingly accepted Dr. Sneden’s challenge to fill in the gaps.

The entire group was tasked with identifying target stars that are from this under-analyzed population of stars. We turned to the PASTEL catalog for our targets​(Soubiran, Le Campion, Cayrel De Strobel, & Caillo, 2010)​.

We discussed the sorts of stellar parameters that would work best and then matched that with stars visible from McDonald Observatory during our window. What would be the visual magnitude limit for the telescope and instrument combination. We were given some stellar parameters to use in the PASTEL database to limit our targets to stars that matched our physical criteria such as surface temperature, surface gravity, and metallicity.

Sample Target Stars

TYC 5701-0019-1	18 51 09.0081	-11 48 21.609	10.53	4998	2.4	-1.44
V* SX Her	16 07 27.2521	+24 54 29.931	8.6	4165	0.2	-1.8
TYC 5562-00446-1	14 18 32.4340	-08 18 24.950	11.17	4217	0.46	-1.51
HD 141531	15 49 16.4985	+09 36 42.429	9.08	4280	0.7	-1.68
HD 165195	18 04 40.0714	+03 46 44.726	7.3	4383	0.8	-1.7

Learning Science By Doing Science

We spent the first day following the directions of Dr. John Kuehne. Dr. Kuehne is an instrumentation expert and understands the subtleties of the Sandiford Echelle Spectrograph. The SES is used mounted at the cassegrain focus of the 2.1 meter Otto Struve Telescope at McDonald Observatory.

Before observing

• Open the dome to equilibrate the instruments
• Take flat and bias frames
• Take thorium argon frames
• Confirm coolant tank filled

Throughout the Night

• Instrument checks
• Fill CCD coolant
• Check the dome slit alignment
• Watch for weather changes
• Running the workstation
• Target selections
• Taking images
• Calculate exposure time
• Monitor telescope
• Detailed note keeping
• Stay awake!

Data Reduction

• Transfer FITS files (IRAF flattened files)
• Subtract flats and bias
• Find wavelength ranges
• Mark spectral orders
• Normalize spectra
• Python visualization (Google Colab Python Notebook)

The initial reduction was handled using IRAF. Once the spectra were flattened, we started looking for a specific absorption line, the first excited state of europium, in the target spectral range. Using Moore’s solar spectrum catalog, we located the europium line at 6645.1 Å by using the neutral nickel line at 6643.6 Å as an anchor ​(Moore, Charlotte E.; Minnaert, M. G. J. & Houtgast, J., 1966)​. Once this spectral order was flattened or normalized, we wrote a Python script to visualize the spectra for comparison between stars. The deeper the absorption line, likely, the higher the abundance of Eu II for that star.

Results

Detection of Eu II among our targets confirms need for further study of the presence of the lanthanides in low metallicity halo stars. The algorithmic analysis to determine the stellar parameters from the spectrum will be part of future work.

Python VS IRAF for Spectroscopy

The Image Reduction and Analysis Facility (IRAF) was the de facto standard for doing astronomy data reduction for tasks such as photometry and spectroscopy. The National Optical Astronomy Observatory has stopped updating and supporting IRAF, although many people still use the software and many community members still act to support the use of IRAF. One of the most common alternatives is using Python with well-established packages and well-known algorithms to reduce data. None of us knew IRAF when we started this project but we learned to use it and made it work. We have been learning to do some of the work in Python and plan to create more resources using Python in the future.

We would like to like to thank 2 scientists in particular for help with IRAF and in making Python work for our needs. Special thanks to Kaitlin Rasmussen, graduate research assistant at Notre Dame University in astrophysics who gave us detailed IRAF instructions for echelle spectra reductions which proved helpful again and again. Some of our reduced and flattened spectra are available here. And special thanks to Dr. Shruti Tripathi, post doctoral fellow at St. Mary’s University who provided detailed Python help in processing and visualizing echelle spectra. A version of the spectra viewer we wrote is available here. Note, we only flattened the spectral order that was of interest to us for Eu I detections.

Curricular Materials

We plan to create lab materials and classroom activities that use our data so students can get a sense of how the research process goes in astronomy.

Changes In Teaching Practice

Each of us describes this research experience as transformative. Besides the hands-on astronomy research skills we gained, we learned a lot of how to learn from mistakes. We learned to depend on each other and to delegate and when to ask for help. Some of the changes we have experienced in the classroom can be found in the narratives below. Much of our experiences match that found in literature from Rebull ​(Rebull et al., 2018)​, Buxner ​(Buxner, 2010)​, and Herrington ​(Herrington, Luxford, & Yezierski, 2012)​.

In what ways was the observing run a form of inquiry science learning?

• In the sense that we didn’t have a clear answer to the question we were asking and also how to answer the question, this was inquiry science learning.
• The observing run had a clear goal but limited instruction in how to reach that goal. The participants were relied upon to devise both strategies and techniques for how to gather data to reach the goal. Our principal investigator was available to provide guidance and occasional correction, but did not provide an “answer” as there was not one “right way” to accomplish our goal.

What aspects of the observing run changed your view of science as a way of knowing?

• The open-ended nature of the questions we asked showed science as a way of knowing. The fact that we went down blind alleys and had to backtrack while knowing this process was necessary is another. The interconnected nature of some of the aspects of astronomy showed that science as a way of knowing is critical in the philosophical approach to answering questions using science.
• The observing run reminded me that science is, often times, more about questioning than knowing. It is difficult to know the right answer, and that answer can be reached many ways, and usually needs to revised after further research. Science should be more about attempting to know, which is possible and happening all the time, than about “knowing” which is often impossible in the short term.

What aspects of the observing run changed your view of science as a body of knowledge?

• One thing I walked away with is how science as a body of knowledge is fragile “at the edges”. The amount of certainty we have about most aspects of science is fairly low when we look in great detail or look for underlying principals. Also, the iterative nature of science as a body of knowledge was clearer. Also the systematic way in which we depend on past knowledge and the quality of that data is more apparent. It is also humbling and awe-inspiring to consider we may play a small role in adding to the science as a body of knowledge through careful application of data analysis of our observations.
• Although we were not doing anything revolutionary in our run, we were still contributing to the body of knowledge that is astronomy. Science as a body of knowledge is continuously growing and it can be added to by all members of the community, not only PhD holders. Anyone attempting to answer a scientific question with authenticity can contribute to the body of knowledge. This also means that science is a body of knowledge is not immutable. This is something teachers often fail to explain to their students.

What aspects of the observing run changed your view of science as a set of skills?

• The data reduction process is almost unattainable in astronomy, especially in spectroscopy. The purpose built telescope, instrument, software packages, and analysis tools show the degree to which one must become, at least somewhat, a subject matter expert in these specific tools. It is likely that many of our developed skills would apply to other systems and also that the mastery of these tools translate into skills that apply outside this specific scientific context. Also, the importance of computing, coding, seeing science as modeling, and algorithmic thinking can’t be ignored.
• In order to accomplish our goal we were required to learn and occasionally devise techniques and skills in order to take data effectively and to later analyze it. These are not “natural” skills but specific to the science objectives and required specific learning.

How has the observing run changed the way you use questioning in the classroom?

• Knowing that science “at the edges” doesn’t have clear answers and also that pushing a learner to answer questions in creative and nuanced ways is key has affected the use of questioning during class time. Applying a longer and appropriate wait time is one big change. I just want to wait for more answers and more complicated answers and more creative answers. I want answers that have elements that are wrong and also right. I want to have students answers build off previous answers. It is a collaborative form of questioning and answering.
• The observing run reminded me that not having an answer is not just ok, but how most science is accomplished. It is normal and expected to struggle through a problem and attempt a solution without knowing if it is correct. In the class, instead of attempting to teach prescribed methodologies and techniques, which do not lead to growth, we should instead allow our students to rely on the skills and knowledge they have previously developed to figure out future questions and solve problems.

AAS 235 Poster

Acknowledgements

Thanks to Dr. John Kuehne for his generous training and observing support. Additionally thanks to the other members of the UT EXES Teacher Associate program for their help and support. This program was originated by Dr. Mary Kay Hemenway, retired senior lecturer at University of Texas at Austin as the SOFIA EXES Teacher Associate Program. This project was funded by NSF Grant No. AST-1616040.

Bibliography