The anomalous Cepheid needles in the RR Lyrae haystack

Over 200 000 RR Lyrae candidates were detected in Gaia DR2. However, most of these are too far to have accurate parallaxes and too faint to have precise photometry from TESS. We wanted to find out how pure the RR Lyrae candidate list is, and how many anomalous Cepheids might hide among them.

We selected stars from the DR2 all-sky variability catalog of Holl et al. (2018) that are brighter than 15 mag in the G_RP band (which is very similar to the passband of TESS). We then selected those that fall on silicon between TESS Sectors 1-22 and are within 4000 pc, and ended up with an initial list of 860 stars. We then added XZ Cet, one of the anomalous Cepheids prototypes to the sample for reference, although it was not classified as variable in DR2.

We calculated the absolute brightnesses with the mwdust code and the `combined19' extinction model which is the combination of four different works (Bovy et al., 2016).

We limited this initial study to stars within 4000 pc. At this value the relative distance errors reach 6-18%, leading to uncertainties in M_G that are comparable or larger than the width of the RR Lyrae PL relation. 

We calculated the Fourier parameters of each TESS light curve to classify them (see on the right), then plotted only those stars whose Fourier-parameters fitted into the RR Lyrae loci based on the TESS light curves:

Our RR Lyrae sample is clearly much less affected by misidentified stars, but still not completely clean. Some sub- and superluminous stars still persist that will require further inspection. 

The goal of our project is to identify anomalous Cepheids (as well as any other types of variables) in the RR Lyrae background and separate them, in order to:

• be able to study nearby members of the enigmatic class of anomalous Cepheids,
• and to determine the Leavitt-Law for RR Lyrae and anomalous Cepheid stars separately.

With the combination of Gaia DR2 distances and TESS light curve information, we were able to create an accurate Leavitt-Law (Period-Luminosity relation) of the nearby RR Lyrae stars. However, the three ACEP candidates we identified in this sample do not separate from the RR Lyrae group. 

We found very few ACEP candidates in this initial sample that prevented is from creating a PL relation for them: we clearly need more. More stars can come from different strategies:

• Further TESS observations: roughly half of the sky has not been covered, nor has we carried out a blind search among the TESS light curves.
• More candidates: multiple surveys identified ACEP candidates.
• Extending the distance. The Gaia DR2 ones did not fit our criteria here. Nor did the ACEP stars identified by Jurkovic (2018). 

The pulsation properties of RR Lyrae and anomalous Cepheids are very similar. It is possible that the PL relations of the two classes overlap in the common period range. The overtone star XZ Cet clearly lies along the extension of the RRc sequence, except it pulsates with a period of 0.82 d (Szabados et al., 2007). In that case it may will be unavoidable to search for the subtle differences in the light curve shapes to separate the two populations

This, however, will be required if we want to understand the Galactic population of anomalous Cepheids and to determine which of these stars are the result of binary evolution and which are single stars that are metal-poor enough to evolve into the instability strip despite being about twice as heavy as RR Lyrae stars.

This study is a first demonstration of the power of the combined Gaia and TESS data when applied to classical pulsators. Continuous photometry also reveals that some stars might have looked like RR Lyraes superficially, but tune out to be something different upon closer inspection.

TESS is a treasure trove of beautiful, unexpected light curves.

Acknowledgements:

L.M. was supported by the Premium Postdoctoral Research Program of the Hungarian Academy of Sciences. The research leading to these results received funding from the LP2014-17 and LP2018-7/2019 Lendület grants of the Hungarian Academy of Sciences. This paper includes data collected by the TESS mission. Funding for the TESS mission is provided by the NASA Explorer Program. This work has made use of data from the European Space Agency (ESA) mission Gaia, processed by the Gaia Data Processing and Analysis Consortium (DPAC). Funding for the DPAC has been provided by national institutions, in particular the institutions participating in the Gaia Multilateral Agreement.

TESS collected full-frame images in 27-day long sectors. We extracted photometry with our custom differential-imaging pipeline based on the fitsh package (Pál 2012). We calculated simple Fourier fits for each light curve, fitting the dominant frequency component above 0.5 d^-1 and its first two harmonics. We then derived the relative Fourier amplitudes (R_i1 = A_i/A₁) and phase differences (ϕ_i1 = ϕ_i-iϕ₁), which are often used to classify the light curves.  We show a few examples below. 

We grouped the light curves into a few broad categories based on their absolute brightnesses and light curve shapes. These were:

• RR Lyrae stars - these form two loci for the fundamental-mode and overtone pulsators.
• Eclipsing binaries - most eclipsing stars 
• Other pulsators (high-amplitude δ Sct, Cepheid)
• Other/unknown

We identified three new nearby anomalous Cepheid candidates. We overlaid the Fourier parameters of the candidates plus that of XZ Cet (the longest-period one in the plot) on top of the stars from the OGLE catalogs. We find that V1949 Cyg (the shortest-period one) is very likely to be an overtone ACEP. The three other stars (HD 297201, UCAC2 21664220, ASAS J064215-2217.9) produce somewhat worse fits.