**Members**

Members that collaborated to generate this roadmap:

Keaton Bell (University of Texas at Austin), Shashi Kanbur (SUNY Oswego), Kelly Hambleton (Villanova University)

**Primary subgroup contact:**

Kelly Hambleton (Villanova University - Kelly.hambleton@villanova.edu)

**Subgroup MAF engineer: **

Keaton Bell (University of Texas at Austin - keatonb@astro.as.utexas.edu)

### Subgroup Primary members

- Keaton Bell - University of Texas at Austin
- Macio Catelan - PUC-Chile
- Kelly Hambleton - Villanova University
- Arne Henden - AAVSO
- Shashi Kanbur - SUNY Oswego
- Nicolas Mauron - IN2P3 CNRS
- Edward Schmidt - University of Nebraska
- Alistair Walker - NOAO

### Subgroup Secondary members

- David Arnett - University of Arizona
- David Bersier - Liverpool John Moores University
- Kem Cook - LSST Consultant
- Suzanne Hawley - University of Washington
- Mark Huber - University of Hawaii
- Julie Lutz - University of Washington
- Lucas Macri - Texas A&M University
- David Nidever
- Hakeem Oluseyi - Florida Institute of Technology
- Andrej Prsa - Villanova University
- Abi Saha - NOAO
- Dimitar Sasselov - Harvard Smithsonian Center for Astrophysics
- Nicole Silvestri - University of Washington
- Paula Szkody - University of Washington
- Mark Wells - Penn State

**Roadmap Outline**

Pulsating Variables can be broadly divided into two categories: radial and non-radial pulsators. Oluseyi et al (2012) have demonstrated the potential of LSST in discovering new RR Lyraes in the Local Group. This was done with older versions of OpSim and without the MAF framework. Most of the Science Driver goals for radial stellar pulsators rely on estimating how well LSST can observe Cepheids and RR Lyraes and admit a "good" Fourier decomposition of the observed light curves.

### Radial Pulsators:

1) Constraining Models: Two important classes of radial pulsators are Cepheids and RR Lyraes. Both are important as distance and (in the case of RR Lyraes) age indicators and as tracers of galactic structure. In addition, both are vital as important testbeds for theories of stellar pulsation and evolution. Stellar evolutionary tracks that go through the instability strip for Cepheids and RR Lyraes obey a Mass-Luminosity (ML) relation and have a well defined range of effective temperatures and compositions. These M,L,T, X, Y, Z parameters that produce appropriate evolutionary tracks are also dependent on input stellar evolutionary physics such as degree of convective overshoot. These parameters can also be input into stellar pulsation codes that produce theoretical light curves. The quantitative structure of these theoretical light curves can be compared with observations (e.g. by Fourier decomposition). Thus new multi-wavelength observations of RR Lyraes and Cepehids can be used, in addition to existing data, to place strong constraints on theoretical pulsation models. The theoretical pulsation models are already computed.

2) Distance Scales: Newly discovered Cepheids will be important to constrain the Cepheid PL/PC/PW relations around the Local Group. Coupled with existing accurate absolute calibrations of the distance to the LMC (Pietrzynski et al 2013) and upcoming calibrations of the Galactic PL (Riess et al 2011) relation. This will lead to significantly more accurate distance scales around the Local Group.

Current Work:

We are taking well observed Cepheids and RR Lyraes in the Magellanic Clouds/Galaxy/Stripe 82 as our "models" of well observed Cepheids and RR Lyraes. These "model light curves" are characterized by the parameters of a Fourier expansion (Bhardwaj et al 2014). Since we know the positions of these Cepheids and RR Lyraes, we can try to recover these observations in OpSim simulations and estimate the error on the parameters of the Fourier expansion of the recovered observations and compare with published results. This will be done in a MAF framework and points to one possible metric and hence update the work of Oluseyi et al (2012).

This will take about 3 months and can be started now - collaboration with members of the “Stellar” group is possible.

For Cepheids, we can take well observed Cepheids with periods across the range of a PL: e.g. 5 days, 10 days, 15 days, 20 days. and see how OpSim recovers the PL relation at different distances. This has strong collaborations with the cosmology group. This can be started now and can take a couple of months or so. Can the LSST improve the usefulness of the first overtone Cepheid PL relation so that the fundamental and first overtone Cepheid PL relation can be used in tandem? Collaboration with the Cosmology subgroup would be very useful.

Consult with Lund et al (2016) for other possible metrics.

Experiment with other period detectors including AoV, Conditional Entropy Methods.

3) Investigate other classical pulsators like type II Cepheids and Miras using similar approaches to those described above. This is a longer term activity and may take 6 months or so.

### Nonradial pulsators:

1) Detecting variables:

The photometric precision of LSST will uncover the regions and amplitudes of stellar pulsations in 6-filter/proper motion/parallax space as an observational pulsation H-R diagram analogous to <a href="

http://astro.phys.au.dk/~jcd/HELAS/puls_HR/.">http://astro.phys.au.dk/~j... At the least, this will generate an extensive catalog of pulsating variables for in-depth follow-up. In many cases, it may be possible to reach asteroseismic conclusions from LSST data themselves. Since stars are observed to be pulsationally variable at some point during all major phases of stellar evolution, asteroseismic results for a large number of stars will better constrain our understanding of the structural evolution of stars.

2) Characterizing variables:

- Model the relative amplitudes of pulsation modes of different spherical degrees in different LSST filters to establish the expected signal of this effect and the usefulness of multi-color observations to guide mode identification.

- Model the effect of projected area variations from geometric distortions of both pressure- and gravity-mode pulsations on measured color amplitudes and pulsation phases, determining the expected signal of this effect.

- Determine the high-frequency limit on the recoverability of pulsations in multi-modal variables and assess for which classes of variable (partial) frequency solutions will be possible for (effective Nyquist frequency for non-uniform sampling).

- Develop statistical/mathematical methods to measure overall pulsational power and individual pulsation amplitudes from distributions of measured magnitudes in severely undersampled light curves (superNyquist signals).

- For pulsator types where these quantities for individual objects will not be well constrained, determine the numbers of objects that must fall in a similarity “bin” that must be observed in order to develop an ensemble picture of pulsation properties. In “binned” parameter space, determine if all stars in particular bins pulsate, and therefore to what extent conclusions about these pulsating stars are generalizable to all stars.

3) Testing realistic models/methods:

- Resample the Kepler data as LSST (OpSIM) and see what pulsational details can we recover.

- Test characterization methods on similar data sets (Stripe 82, CSS, PTF, ZTF, …).

4) Advocating this science:

- Develop MAF metrics that evaluate how well the proposed LSST cadences capture pulsation science.

- Tie the performance (numbers of objects detected/characterized) to specific astrophysical results at major milestones of survey operations (e.g., years 1, 3, 5, 10).

### Near Term Goals:

- Use known RR Lyrae and Cephieds, with detailed observations, to determine an optimal MAF metric and determine recoverable information in comparison to published results.

- Determine the optimal MAF metric for obtainig high-frequency pulsations in multi-modal variables and assess for which classes of variable (partial) frequency solutions will be possible for (effective Nyquist frequency for non-uniform sampling).

- Determine the MAF metric for creating the optimal cadence such that the window function does not impead data analysis.

- Resample the Kepler data as LSST (OpSIM) and to determine the types of pulsaters that are recovered and detail of information.

### Long Term Goals:

- Use known classical pulsators like type II Cepheids and Miras, with detailed observations, to determine an optimal MAF metric and determine recoverable information in comparison to published results.

- Create a classification algorithm to identify different pulsators.

- Develop statistical/mathematical methods to measure overall pulsational power and individual pulsation amplitudes from distributions of measured magnitudes in severely undersampled light curves (superNyquist signals).