betaFeedback

WorksPhysics and Astronomy

Calibrated Period-Luminosity Relations for Mira variables in the Large Magellanic Cloud in the Vera C. Rubin Observatory zyzy-band Photometric Systems

Abstract

The Vera C. Rubin Observatory's Legacy Survey of Space Time (LSST) is expected to discover numerous Mira variables in nearby galaxies, enabling their application for the cosmic distance scale. Miras are excellent distance indicators at infrared wavelengths, and therefore their period-luminosity (PL) relations in reddest zyzy photometric systems will be crucial in the LSST era. We present first empirical zyzy-band PL relation for 85 oxygen-rich (O-rich) Miras located in the Large Magellanic Cloud based on time-series observations collected using the Dark Energy Camera (DECam) at the Víctor M. Blanco 4-m Telescope. The DECam photometry was directly calibrated using the Rubin Observatory Data Preview 2 (DP2) data. The observed PL relations were found to exhibit a break around a period of ∼300\sim 300 days, and therefore we only derived and calibrated the PL relations for Miras with a period shorter than 300300 days. We also explored the Rubin LSST DP2 data for known Mira variables in three Local Group dwarf galaxies. The zyzy-band light curves of Miras in IC 1613 are too sparsely sampled, whereas the DP2 observations of NGC 6822 and the Sagittarius Dwarf Irregular Galaxy cover only a short temporal baseline. For an O-rich Mira in NGC 6822, we obtain a distance modulus μ=23.32±0.24μ= 23.32\pm0.24~mag in good agreement with the literature. Significantly improved temporal coverage is expected in the first official Rubin LSST data release that will enable a robust validation of the PL relation in these nearby systems.

Introduction

The Mira variables (hereafter Miras) are long-period pulsating stars with a period ranging from ∼100\sim100 to ∼1000\sim1000 days. As late-type cool asymptotic-giant-branch (AGB) stars, the spectral-energy distribution (SED) for Miras peaks around the KK-band, making this filter an excellent choice for constructing the period-luminosity (PL) relation for distance scale applications. Indeed, most of the Mira’s PL relations derived in the literature are in the KK-band or similar near-infrared (NIR) wavelengths (for example, Bhardwaj et al. [8]; Ita et al. [19]; Sun et al. [41, 42]; Lewis et al. [24]; Sanders [35]; Konchady et al. [22]; Bhardwaj et al. [7], and reference therein).

As argued in Whitelock [46] and Huang [16], Miras can be competitive to classical Cepheids for distance scale applications, such as providing independent measurements of the Hubble-Lemaitre constant H0H_0 (for example, see Huang et al. [17], for the application of Miras found in M101). Recently, Bhardwaj et al. [7] provided NIR calibration of Miras in the Galactic globular clusters and used those as an independent anchor to determine a 3.7% measurement of H0H_0. One advantage of using Miras is that these intermediate to old age stars are found in both spiral and elliptical galaxies, in contrast to young Cepheids which can only be found in spiral galaxies. For general reviews of Miras and their applications to the cosmic distance scale, the readers may refer to Huang [16] and Skowron & Soszyński [38].

Beginning 30 June 2026, the NSF-DOE Vera C. Rubin Observatory Legacy Survey of Space and Time (hereafter Rubin LSST, Ivezić et al. [20]) will routinely observe the southern sky in six ugrizyugrizy filters for a decade. As one of the scientific targets within the Rubin LSST Transients and Variable Star science collaboration (Hambleton et al. [15]), numerous (∼104\sim10^4 to ∼105\sim10^5) Miras are expected to be discovered in nearby galaxies (Macri [25]). Clearly, Rubin LSST lacks a KK-band filter, and its closest or longest wavelength filter for Rubin LSST is the yy-band. Therefore, it is desirable to derive and calibrate the PL relation for Miras in the yy-band, such that these empirical calibrations can be applied to the Rubin LSST-discovered Miras for dis-

Email: [email protected] tance scale applications. We note that there is no empirically calibrated yy-band PL relation for Miras available in the literature, except for the synthetic PL relations presented in Iwanek et al. [21].

As a result, we initiated a project to derive the yy-band PL relation using the known Miras located in the Large Magellanic Cloud (LMC). This is because LMC has a well-measured distance to the precision of the 1% level ([34, 23]), and LMC is located within the visibility area of Rubin LSST. Furthermore, LMC hosts a large population of Miras with well-determined periods from the Optical Gravitational Lensing Experiment (OGLE, Soszyński et al. [40]), and hence the periods and other pulsation properties are already known. In particular, we adopted a “clean” sample of LMC Miras from Ou & Ngeow (2022), composed of oxygen-rich (O-rich) variables1 with well-behaved (i.e. “regular”) light curves for the derivation of the yy-band PL relation.

Ideally, time-series data obtained from Rubin LSST could provide necessary light curve data sets for the LMC Miras. However, Rubin LSST saturates around y∼14y \sim14 mag, which is roughly the same as the mean yy-band magnitude for LMC Miras (Iwanek et al. [21]). Clearly, saturated Rubin LSST photometry would affect the light curve quality and hence the derived mean magnitudes. To remedy this, we observed the LMC Miras using the wide-field DECam (Dark Energy Camera, Flaugher et al. [11]), mounted on the Víctor M. Blanco 4-meter Telescope. This was done by taking a series of shallow images so that the LMC Miras were not saturated. Since Miras are intrinsically red, the DECam observations have also covered the zz-band to include the (z−y)(z-y) color-term during the photometric calibrations. Because of this, we have also derived the zz-band PL relation in addition to the yy-band PL relation.

DECam Observations

Given the large field-of-view (FOV) for DECam, about 3 square degrees, most of the target ∼400\sim400 regular O-rich LMC Miras can be covered within 13 DECam pointings (see Figure 1). A sequence of zyzy-band images with exposure time of 5 and 10 seconds, respectively, were taken for each of the DECam pointings to avoid saturation in the LMC Miras. With such exposure time in the zyzy-band, together with overheads, the observation time to cover 13 DECam pointings would take about 20 minutes in a given night. This 20 minute observation represents one epoch on the light-curve, and multiple epoch observations were carried out in the NOIRLab 2024B and 2025B Semesters. Because such a short observation time is needed per night but is repeated in multiple nights, our “short-and-repeat” remote observation experiments were carried out within the DECam Alliance for Transients (DECAT) consortium (for a further description, see Graham et al. [13]).

Distribution of 13 DECam pointings toward the LMC

Figure 1. The distribution of 13 DECam pointings toward LMC to cover the selected O-rich LMC Miras (in yellow circles, selected from Ou & Ngeow 2022) as much as possible. The green and red squares represent the guiding and focusing CCD chips on the DECam, respectively. This plot was prepared using the Aladin software ([9]).

Due to multiple constraints, including the availability of the yy-band filter on DECam, the nights awarded for the DECAT consortium, priority of the targets within the DECAT pool, etc., our program was carried out in 11 and 3 nights in the 2024B and 2025B Semesters, respectively, by the DECAT remote observers. The 11 nights in the 2024B Semester include 4 nights in September 2024 and 7 consecutive nights in January 2025. In the 2025B Semester, our program was carried out one and two nights in August and September 2025, respectively. Note that on 14-16 January 2025, the observations were repeated in these three nights. Altogether, a total of 17 epochs of observations have been taken for our program. For a small number of Miras, the number of epochs could be as much as 34, because they are located at the overlapped regions of two DECam pointings.

Photometric Calibration

Preparing the DECam Images

All DECam observed images were processed via the DECam community pipeline (Valdes et al. [43]), and processed images with InstCal types were downloaded for photometric measurements. To save disk space, we have used MissFITS (Marmo & Bertin [27]) to split the DECam multiextension FITS images into individual CCD chips, and only the CCD chips that contain our target LMC Miras have been kept. Using Swarp (Bertin et al. [6]), we further median combined all CCD chips in the same DECam pointings with the same CCD numbers to create a deep detection image for a given DECam pointing and CCD chip combination. These processed images are used in the following subsection.

Preparing the Reference Catalog

Prior to the official data releases of the Rubin LSST, the Vera C. Rubin Observatory has released two datasets based on the observations taken during the on-sky commissioning campaign and the early optimization phase, known as Data Preview 1 (Vera C. Rubin Observatory Team et al. [45]) and Data Preview 2 (hereafter DP2, [44, 30]).2 As shown in Figure 2, LMC is included in the zyzy-band sky coverage of DP2, providing a natural reference catalog to calibrate the DECam photometry.

Sky coverage of DP2 in the $z$-band and $y$-band

Figure 2. The top and bottom panel display the sky coverage of DP2 in the zz-band and yy-band, respectively. The colorbars represent the number of DP2 visits on each LSSTCam observations. The 13 DECam pointings on LMC are marked as orange circles, along with the locations of three dwarf galaxies investigated in § 6 shown as red symbols. The cyan rectangles represent the DP2 coadded tracts.

However, as can be seen from Figure 2, there are no coadded images at the location of LMC, implying that there are no data in the DP2’s Object table. Therefore, we queried the Source table, through the Rubin Science Platform (RSP), to obtain the zyzy-band photometry for the point-like sources (selected using the extendedness = 0 condition) located within the DECam pointings. Furthermore, we also excluded sources with any of the following flags set to “TRUE” in the Source table: invalidPsfFlag, pixelFlags_bad, pixelFlags_cr, pixelFlags_edge, pixelFlags_interpolated, pixelFlags_nodata, pixelFlags_offimage, pixelFlags_saturated, pixelFlags_suspect. A caveat of using the Source table is that a given source would have multiple and unique sourceID in this table if it has been observed multiple times (or multiple visits).

releases, where the early DP2 only includes the coadd images and all of the catalog products. In contrast, the full DP2 will also include the visit-level images. Since in this work we only used the catalog products, we simply refer the early DP2 as DP2.

Therefore, we further split the downloaded Source catalogs according to the unique visit IDs.

Using the deep median combined images of the same DECam pointings and CCD chips, prepared in § 3.1, we cross-match the split Source catalogs with the sources detected in these DECam detection images to build the DP2-LMC reference catalog. For matched sources that have multiple visits in a given filter, we took the weighted average of the available DP2 photometry, else we adopted the single-visit DP2 measurements. However, DP2 did not uniformly cover the entire LMC but with patchy observations, resulting ∼5%\sim5\% of the images to have no DP2 sources and ∼36%\sim36\% of the images to have fewer than 20 reference sources of DP2. These images were excluded in the subsequent photometric calibration.3

Running the Photometric Calibration

For the remaining images, we used the Source-Extractor ([5]) and

PSFEx ([4]) software suite to measure the instrumental point-spread function (PSF) magnitudes, mPSFm^{PSF}, in the DP2 reference sources. To ensure that good quality photometry was used in the calibration, we further filtered out those mPSFm^{PSF} with errors larger than 0.1 mag and possess Source-Extractor’s flag values larger than zero. We also excluded sources that exhibit variability in magnitudes observed in two different DP2 visits. We kept those images that have 20 or more DP2-LMC reference sources after such filtering. The photometric calibration involved iteratively solving the following equations (following the same approach as done in [29]), separately on the zyzy-band images, using the final selected DP2-LMC reference sources:

zLSST−zPSF=ZPz+Cz(z−y)LSST,(1)z^{LSST} - z^{PSF} = ZP_z + C_z(z-y)^{LSST}, \tag*{(1)}
yLSST−yPSF=ZPy+Cy(z−y)LSST,(2)y^{LSST} - y^{PSF} = ZP_y + C_y(z-y)^{LSST}, \tag*{(2)}

where ZPmZP_m and CmC_m represent the zero-points and color-terms in filter m={z,y}m = \{z, y\}, respectively. The colors in the DP2-LMC reference sources ranged from ∼−1.5\sim-1.5 mag to ∼2.3\sim2.3 mag.

The mean values of the color-terms on these images are ⟨Cz⟩=0.534\langle C_z\rangle= 0.534 and ⟨Cy⟩=0.100\langle C_y\rangle= 0.100, with a dispersion of σz=0.204\sigma_z = 0.204 and σy=0.217\sigma_y = 0.217, in the zyzy bands, respectively. We have further removed a small fraction of these images that have a color-term deviating more than 3σ3\sigma from the mean values, as the calibrated photometry from these images becomes outlier epochs in the light curves.

Light Curves and Color Curves

For the photometrically calibrated images, we extracted the instrumental PSF magnitudes for those LMC Miras located within these images in the same way as the reference stars, and calibrated their photometry via equations 1 and 2. Since a pair of zyzy-band images was taken near simultaneously, which allowed the (z−y)LSST(z-y)^{LSST} color for a given Mira to be solved:

(z−y)LSST=(ZPz+zPSF)−(ZPy+yPSF)1−Cz−Cy.(3)(z-y)^{LSST} = \frac{(ZP_z + z^{PSF}) - (ZP_y + y^{PSF})}{1-C_z-C_y}. \tag*{(3)}

The solved color, (z−y)LSST(z-y)^{LSST} ranges from ∼0.0\sim0.0 mag to ∼1.3\sim1.3 mag, was then applied to equations (1) and (2) to calibrate the zyzy-band magnitudes for a given LMC Mira. Using such an approach, we can construct not only the zyzy-band light curves but also the color curves. During the construction of the light and color curves, we have also excluded a pair of zyzy-band magnitudes if either one of them has a photometric error larger than 0.1 mag (they also tend to be outliers in the light and/or color curves). Therefore, a given Mira would have the same number of data-points (NN) in both of the zyzy bands. Two examples of the zyzy-band light curves and color curves are presented in Figure 3. We further removed those Miras that have five or less data-points left on the light and color curves, as they are too sparse to have a meaningful analysis.

Figure 3. Examples of the zyzy-band light curves and (z−y)(z-y) color curves for two LMC Miras observed with DECam. The OGLE ID, numbers of data-points (NN), and the OGLE periods are given on top of each panel. The solid curves are the fitted light curves or color curves using m=⟨m⟩+Asin⁡(2πΦ+ϕ)m = \langle m\rangle+ A\sin(2\pi\Phi+ \phi), where ⟨m⟩\langle m\rangle is the mean value and Φ\Phi represents the pulsations phases from 0 to 1. The phase shift ϕ\phi is forced to be the same for both of the zyzy-band light curves (but not for the color curve). For these two LMC Miras, we have also included the DP2 photometry, queried from the Source table and linked the matched sources to the Visit table for obtaining the modified Julian date (MJD), as squared symbols.

After applying various selection criteria, as mentioned previously, there were about 100 LMC Miras left in our sample. The published pulsation periods (PP) were used to fold their zyzy-band light curves and the color curves. Given the sparse nature of our light/color curves, we fit these phased curves using a sinusoidal function to determine the mean magnitudes. We visually inspected the fitted light and color curves, and excluded those Miras that exhibit unrealistic large amplitudes. This happened in cases where data-points clustered in a narrow range of the pulsation phases. As a result, our final sample consists of 85 LMC Miras.

We have also queried the DP2 zyzy-band light curves, via the RSP on the Source table, for our entire LMC Mira sample. However, when using a search radius of 0.5″, only about half of them have at least one measurement in the zz- or yy-band.4 Furthermore, the DP2 photometry seems to agree with our fitted light curves only in very few cases. An example is shown in the left panel of Figure 3, where the zyzy-band photometry queried from the DP2’s Source table was located very close to the fitted curves. In many cases, the DP2 photometry is either too faint or too bright. An example is displayed in the right panel of Figure 3 — even though there is one data-point that agrees with the fitted curve, there are two outlier data-points at y∼15.5y \sim15.5 mag. Since the LMC Miras are quite bright for the Rubin Observatory, it is possible that DP2 photometry is not optimized for bright sources yet. Therefore, we did not include the DP2 photometry when fitting the sinusoidal curves.

The PL Relations

In Table 1, we summarize the mean magnitudes and mean colors for our final sample of LMC Miras. These mean magnitudes and mean colors were derived based on the fittings of a sinusoidal curve (see the caption in Figure 3). In the same Table, we also included the values of PP and E(V−I)E(V-I) adopted from Ou & Ngeow (2022, and reference therein). For extinction and reddening corrections, we adopted Rz=1.323R_z = 1.323 and $R_y = 1.088 ([36]), together with a conversion of E(B−V)=E(V−I)/1.237E(B-V) = E(V-I)/1.237 ([39]). In the upper panel of Figure 4, we compare the (extinction-corrected) II-band PL relations between the full sample adopted from Ou & Ngeow (2022) and our final selected sample. As can be seen from the entire sample, the II-band PL relation exhibits a break around log⁡P∼2.4\log P \sim2.4 to ∼2.5\sim2.5. By adopting a break period at P0=300P_0 = 300 days ([8]), we fit the II-band PL relation separately for the short (P<P0P < P_0) and long (P>P0P > P_0) period Miras in the form of I=aI+bI(log⁡P−log⁡P0)I = a_I + b_I(\log P - \log P_0).

OGLE IDPP (days)E(V−I)E(V-I)NN⟨z⟩\langle z\rangle⟨y⟩\langle y\rangle⟨(z−y)⟩\langle(z-y)\rangle
OGLE-LMC-LPV-00743216.370.1791614.009 ±\pm 0.00213.776 ±\pm 0.0030.292 ±\pm 0.003
OGLE-LMC-LPV-00902296.250.1771613.793 ±\pm 0.01013.313 ±\pm 0.0120.473 ±\pm 0.007
OGLE-LMC-LPV-00990321.410.1241713.081 ±\pm 0.00412.859 ±\pm 0.0030.254 ±\pm 0.005
OGLE-LMC-LPV-01059199.350.1171714.063 ±\pm 0.00313.852 ±\pm 0.0030.196 ±\pm 0.006
OGLE-LMC-LPV-01199132.810.0621715.010 ±\pm 0.00314.732 ±\pm 0.0030.296 ±\pm 0.005
⋯\cdots⋯\cdots⋯\cdots⋯\cdots⋯\cdots⋯\cdots⋯\cdots

Table 1. Mean zyzy-band Magnitudes and Colors for the LMC Miras. **Note—**This table is available in its entirety in machine-readable form. A portion is shown here for guidance regarding its form and content.

Figure 4. Upper Panel: Comparison of the II-band PL relation for the full sample of LMC O-rich Miras as presented in Ou & Ngeow (2022) (open gray circles) and our final selected sample (filled black circles). The dashed and solid lines are the fitted PL relations to the full and our selected samples, respectively. Middle and Lower Panels: The zyzy-band PL relations for our final selected sample. The solid lines are the fitted PL relations to the short period Miras, as given in equation (4) and 5. In all panels, the mean magnitudes have been corrected for extinction. The vertical dotted line indicates the adopted break period of P0=300P_0 = 300 days.

When we compared the II-band PL relations fitted for the full sample and our final selected sample, we found that the PL relations derived for the short period Miras are in good agreement. This suggests that our final selected sample can be used to derive the short period zyzy-band PL relations, as given below:

⟨z⟩0=14.05(±0.09)−0.92(±0.45)[log⁡P−log⁡P0](4)\langle z\rangle_0 = 14.05(\pm0.09) - 0.92(\pm0.45)\left[\log P - \log P_0\right] \tag*{(4)}
⟨y⟩0=13.55(±0.06)−2.07(±0.34)[log⁡P−log⁡P0](5)\langle y\rangle_0 = 13.55(\pm0.06) - 2.07(\pm0.34)\left[\log P - \log P_0\right] \tag*{(5)}

with a dispersion of σz=0.26\sigma_z = 0.26 mag and σy=0.20\sigma_y = 0.20 mag, respectively. However, discrepancies in the PL relations were found for the long period Miras (as shown in the upper panel of Figure 4). Furthermore, there are only a few of the long period LMC Miras left in our sample, and they exhibit a larger dispersion than their short period counterparts. Therefore, we did not derive the long period zyzy-band PL relations.

Dwarf Galaxies Miras Observed in DP2

Since the short-period O-rich Miras in the Galactic field (e.g. [35]) and globular clusters (e.g. [7]) are too bright for the Rubin Observatory, the best way to test our derived PL relations is using Miras in the local dwarf galaxies. There are three dwarf galaxies located in the DP2 footprint (see Figure 2) with Miras detected in the past. These three dwarf galaxies are IC 1613 ([28]), NGC 6822 ([48]), and the Sagittarius dwarf irregular galaxy (hereafter SagDIG, [47]), providing a great opportunity to test our derived PL relations directly with the DP2 data.

Using the LSDB framework (Large Scale Database, [10]) within the RSP, we queried the DP2 DiaObject catalog and the nested diaObjectForcedSource tables to extract the zyzy-band light curves for the O-rich Miras. The post-processing of the LSDB framework has also converted the PSF fluxes given in the DP2 tables to AB magnitudes (for details, see [26]), providing a convenient way to construct the light curves. When constructing the light curves, we have also excluded data-points with any of the following flags set to True: psfFlux_flag, pixelFlags_bad, pixelFlags_saturated, pixelFlags_cr, pixelFlags_nodata, pixelFlags_suspect, pixelFlags_edge, and pixelFlags_interpolated.

The value of 23.40 mag is the mid-point of the measurements cited in Table 10 of [32 spanning from 23.24 to 23.56 mag]. The right panel of Figure 5 compares the absolute magnitude of these Miras to those of the LMC, showing that this NGC 6822 Mira is located within the parameter space defined by the LMC Miras.

Figure 5. Left Panel: The DP2 light curve data (black symbols) for an O-rich Mira, with ID 20331 [48], in NGC 6822. The red curve is the fitted sinusoidal curve as done in Figure 3. Right Panel: Comparison of the z-band absolute magnitude for Mira 20331 (black symbol), after correcting for extinction using the [37] extinction and adopting μ6822=23.40\mu_{6822}=23.40 mag (see text for details), with the LMC O-rich Miras (red circles). A distance modulus of μLMC=18.48\mu_{LMC}=18.48 mag [34] was used to convert the observed mean magnitudes of LMC Miras to the absolute magnitudes, and calibrate the PL relation derived in equation (4) (solid line). For the NGC 6822 Mira, the shorter error bar represents the maximum and minimum values of μ6822\mu_{6822} listed in [32], and the longer error bar is for the 0.22 mag error in the mean magnitude.

Alternatively, we can evaluate the performance of our derived PL relation by deriving the distance modulus to NGC 6822. Since this Mira has a period of 314 days, slightly longer than P0P_0, if we assumed that the extrapolated z-band PL relation is valid for this Mira, then we obtained a distance modulus of μz=23.32±0.24\mu_z=23.32\pm0.24 mag, which is consistent with the values compiled in [32].

For the remaining 10 O-rich Miras that are not fitted with a sinusoidal curve, we took a weighted mean of their light curve data and treated it as a single-epoch random-phase magnitude. In Figure 6, we compare such single-epoch random-phase magnitudes, after converting those to absolute magnitudes, to the LMC Miras in both zyzy-band PL relations. As expected, they show a larger scatter, especially for the long period Miras, due to incompleteness of phase sampling. Adding longer baseline light curve data from the Rubin LSST would naturally improve the measurements of their mean magnitudes, and hence reduce the scatter on the PL relations.

Figure 6. The PL relations in the zyzy bands for Miras similar to the right panel of Figure 5. O-rich Miras in NGC 6822, IC 1613, and SagDIG are shown along with the LMC Miras. The extinction-corrected ”magnitudes” for the Miras in the dwarf galaxies are based on the equivalent single-epoch random-phase measurements (see text for details). Same as NGC 6822, we adopted the distance modulus of IC 1613 as the mid-point of the values listed in Table 11 of Parada et al. [32], as μ1613=24.34\mu_{1613}=24.34 mag. The distance modulus of SagDIG, μSagDIG=25.27\mu_{SagDIG}=25.27 mag, was adopted from Parto et al. [33].

NGC 6822

NGC 6822 has the highest number of known Miras among the three dwarf galaxies, with 61 Miras detected in [48]. The 11 O-rich Miras in NGC 6822 have 7 to 11 DP2 data-points available on their zyzy-band light curves. However, DP2 yy-band observations on the location of NGC 6822 only spanned ∼3.9\sim3.9 days. This implied that the yy-band light curves data, after folded with Mira’s period, would cluster in a narrow range of the pulsation phases and hence it would be impractical to fit a sinusoidal curve. The data for the zz-band light curves spanned a longer observing time of about 23.1 days. The ranges of the pulsation phases are slightly wider than in the yy-band; however, the fitted sinusoidal curves still returned unrealistic mean magnitudes and/or amplitudes for most of the Miras in NGC 6822. The only exception is Mira with ID 20331, which has a fitted mean magnitude of ⟨z⟩=19.15±0.22\langle z\rangle=19.15\pm0.22 mag and an amplitude of 2.87 mag. Its DP2 zz-band light curve and the fitted sinusoidal curve are shown in the left panel of Figure 5.

To compare this NGC 6822 Mira with the LMC Miras, we adopted a distance modulus of μ6822=23.40\mu_{6822}=23.40 mag to NGC 6822 and converted the extinction-corrected mean zz-band magnitude of this Mira to absolute magnitude.

SagDIG and IC 1613

The only O-rich Mira in SagDIG reported by [47] has a very long pulsation period of 950 days and is undergoing a hot-bottom burning phase. Given its projected location close to NGC 6822 (see Figure 2), this Miras has a very similar number of DP2 zyzy-band data-points as the Miras in NGC 6822. The observing time-spans, on the other hand, are ∼12.3\sim12.3 and ∼37.2\sim37.2 days on the zz- and yy-band light curves, respectively. Given that this Mira has a very long period, the light curve data-points cluster in a narrow range of pulsation phases and are not suitable to fit a sinusoidal curve. As in the case of NGC 6822 Miras, we took a weighted mean for the light curve data of this Mira and treated it as a single-epoch random-phase measurement. Its location in the PL relation is shown in Figure 6.

For IC 1613, 5 of the 11 Miras found in Menzies et al. [28] are O-rich Miras, and the zyzy-band light curve data were available for 4 of them. However, there are only 1 to 2 data-points available on the light curves. Treating them the same way as the Miras in NGC 6822 and SagDIG, we averaged the light curve data if it has two measurements, or adopted the single data-point as the single-epoch random-phase magnitudes. The comparison with the LMC Miras in Figure 6 shows large scatter at the long-period end. Therefore, Mira light curves with limited sampling in nearby dwarf galaxies are not very useful for the distance measurement with current data preview releases of Rubin LSST.

Discussion and Conclusion

In the era of Rubin LSST, the zyzy-band PL relations for Miras would be undoubtedly valuable especially toward the latter half of its decade-long survey when a large number of Mira variables are discovered in numerous galaxies. In this work, we present the first empirical zyzy-band PL relations of a sample of O-rich Miras derived from shallow observation of the LMC with DECam. This is particularly important because the photometry of Rubin LSST is likely to have a saturation problem at the bright end. Indeed, when we examined the DP2 zyzy-band photometry for our sample of Miras, most of the photometric data were found to be spurious, i.e. too faint or too bright.

Our DECam photometry is directly calibrated using the DP2 data; therefore, these zyzy-band PL relations represent the first empirical study in the Rubin LSST photometric system. A drawback of using DP2 photometry is that the current observations toward the LMC are patchy and incomplete, resulting in a large fraction of our targeted LMC Miras to remain photometrically un-calibrated. Nevertheless, we tested the derived PL relations for short period O-rich Miras using the II-band data between the full sample and our final calibrated sample, and both PL relations show a good agreement. Therefore, we derived the zyzy-band PL relations for O-rich Miras with periods shorter than 300 days, after calibrating with the LMC distance (Pietrzyński et al. [34]), as:

Mz=−4.43(±0.09)−0.92(±0.45) [log⁡P−log⁡P0],My=−4.93(±0.07)−2.07(±0.34) [log⁡P−log⁡P0].\begin{aligned} M_z &= -4.43(\pm0.09) - 0.92(\pm0.45)\,[\log P-\log P_0],\\ M_y &= -4.93(\pm0.07) - 2.07(\pm0.34)\,[\log P-\log P_0]. \end{aligned}

In our work, we assume that the zyzy-band PL relations for the O-rich Miras are insensitive to metallicity, since our DECam observations were not designed to investigate the metallicity effect. However, as pointed out by the referee, a recent study by Huang et al. [18] showed that the spectral properties of O-rich Miras could be affected by metallicity and may have an impact on the applications of the zyzy-band PL relations for stellar systems with different metallicities. Bhardwaj et al. [7] found no significant metallicity effect on JHKsJHK_s PL relations for O-rich Miras in star clusters that cover a wide range of mean-metallicities. Therefore, the metallicity dependence of the reddest LSST yy band PL relation may not be significantly different from the JJ band PL relation. A detailed investigation of the metallicity dependence is beyond the scope of this work.

In addition to PL relations, we have also attempted to derive the period-Wesenheit (PW) relation, in which the Wesenheit index is defined as W=y−4.630(z−y)W = y - 4.630(z-y). However, the fitted PW relation carried a large dispersion of 0.42 mag, which is twice that of the yy-band PL relation, and it will not be useful for distance measurement. This is likely because the zyzy-band Wesenheit index is very sensitive to color, with a coefficient of 4.63 in the color term.

There are three dwarf galaxies located within the footprint of DP2, and they are known to host at least one O-rich Miras. This provides a natural testbed for our derived zyzy-band PL relations. However, the time-series zyzy-band observation for IC 1613 is too sparse to fit a light curve. In case of NGC 6822 and SagDIG, which have a higher number of data-points on the zyzy-band light curves, the observations time-spans are too short when compared to the pulsation periods of Miras, and cause the fitted mean magnitudes and/or amplitudes to be unreliable. The only exception is Mira 20331 in NGC 6822, which has a reasonable mean magnitude and amplitude after fitting its zz-band light curve data with a sinusoidal curve. Nevertheless, by adopting the distance modulus of NGC 6822 from the literature, the absolute zz-band magnitude for this Mira agrees with LMC Miras with similar periods. We emphasize that this is a preliminary test, because the DP2 data were taken during the commissioning phase. We expect more complete light curve data to be available by combining the DP2 data and the first Rubin LSST data release (DR1) in the coming years.

Acknowledgments

The authors thank several useful comments and suggestions of the anonymous referee, which helped improve the manuscript. CCN acknowledges the funding from the National Science and Technology Council (NSTC, Taiwan) under the grants 114-2112-M-008-011 & 115-2112-M-008-003. AB acknowledges the funding from the Anusandhan National Research Foundation (ANRF) under the Prime Minister Early Career Research Grant scheme (ANRF/ECRG/2024/000675/PMS). This research was supported by the International Space Science Institute (ISSI) in Bern/Beijing through the ISSI/ISSI-BJ International Team project ID #24-603 – “EXPANDING Universe” (EXploiting Precision AstroNomical Distance Indicators in the Gaia Universe). We sincerely thank A. Rest, and the observers from the DE-CAT team for carrying out the remote DECam observations required for our project. This research used Astropy,5 a community-developed core Python package for Astronomy ([1, 2, 3]).

This project used data obtained with the Dark Energy Camera (DECam), which was constructed by the Dark Energy Survey (DES) collaboration. Funding for the DES Projects has been provided by the US Department of Energy, the U.S. National Science Foundation, the Ministry of Science and Education of Spain, the Science and Technology Facilities Council of the United Kingdom, the Higher Education Funding Council for England, the National Center for Supercomputing Applications at the University of Illinois at Urbana-Champaign, the Kavli Institute for Cosmological Physics at the University of Chicago, Center for Cosmology and Astro-Particle Physics at the Ohio State University, the Mitchell Institute for Fundamental Physics and Astronomy at Texas A&M University, Financiadora de Estudos e Projetos, Fundação Carlos Chagas Filho de Amparo à Pesquisa do Estado do Rio de Janeiro, Conselho Nacional de Desenvolvimento Científico e Tecnológico and the Ministério da Ciência, Tecnologia e Inovação, the Deutsche Forschungsgemeinschaft and the Collaborating Institutions in the Dark Energy Survey.

The Collaborating Institutions are Argonne National Laboratory, the University of California at Santa Cruz, the University of Cambridge, Centro de Investigaciones Energéticas, Medioambientales y Tecnológicas–Madrid, the University of Chicago, University College London, the DES-Brazil Consortium, the University of Edinburgh, the Eidgenössische Technische Hochschule (ETH) Zürich, Fermi National Accelerator Laboratory, the University of Illinois at Urbana-Champaign, the Institut de Ciències de l’Espai (IEEC/CSIC), the Institut de Física d’Altes Energies, Lawrence Berkeley National Laboratory, the Ludwig-Maximilians Universität München and the associated Excellence Cluster Universe, the University of Michigan, NSF NOIRLab, the University of Nottingham, the Ohio State University, the OzDES Membership Consortium, the University of Pennsylvania, the University of Portsmouth, SLAC National Accelerator Laboratory, Stanford University, the University of Sussex, and Texas A&M University.

Based on observations at NSF Cerro Tololo Inter-American Observatory, NSF NOIRLab (NOIRLab Prop. ID 2024B-261264; PI: Chow-Choong Ngeow), which is managed by the Association of Universities for Research in Astronomy (AURA) under a cooperative agreement with the U.S. National Science Foundation.

This publication is based in part on proprietary Rubin Observatory Legacy Survey of Space and Time (LSST) data, and was prepared in accordance with the Rubin Observatory data rights and access policies. All authors of this publication meet the requirements for coauthorship of proprietary LSST data

This material is based upon work supported in part by the National Science Foundation through Cooperative Agreements AST-1258333 and AST-2241526 and Cooperative Support Agreements AST-1202910 and 2211468 managed by the Association of Universities for Research in Astronomy (AURA), and the Department of Energy under Contract No. DE-AC02-76SF00515 with the SLAC National Accelerator Laboratory managed by Stanford University. Additional Rubin Observatory funding comes from private donations, grants to universities, and in-kind support from LSST-DA Institutional Members. This research uses services or data provided by the Rubin Science Platform at NSF-DOE Vera C. Rubin Observatory, which is jointly funded by the U.S. National Science Foundation and the U.S. Department of Energy, Office of Science.

Facilities: Blanco (DECam), Rubin:Simonyi, Rubin:USDAC

Software: Aladin ([9]), astropy ([1, 2, 3]), ds9 ([12]), dustmaps ([14]), MissFITS ([27]), PSFEx ([4]), Swarp ([6]), Source-Extractor ([5])

References

  1. [1]Astropy Collaboration, Robitaille, T. P., Tollerud, E. J., et al. 2013, A&A, 558, A33, doi: 10.1051/0004-6361/201322068
  2. [2]Astropy Collaboration, Price-Whelan, A. M., Sipőcz, B. M., et al. 2018, AJ, 156, 123, doi: 10.3847/1538-3881/aab c4f
  3. [3]Astropy Collaboration, Price-Whelan, A. M., Lim, P. L., et al. 2022, ApJ, 935, 167, doi: 10.3847/1538-4357/ac7c74DOI
  4. [4]Bertin, E. 2011, in Astronomical Society of the Pacific Conference Series, Vol. 442, Astronomical Data Analysis Software and Systems XX, ed. I. N. Evans, A. A. Accomazzi, D. J. Mink, & A. H. Rots, 435
  5. [5]Bertin, E., & Arnouts, S. 1996, A&AS, 117, 393, doi: 10.1051/aas:1996164DOI
  6. [6]Bertin, E., Mellier, Y., Radovich, M., et al. 2002, in Astronomical Society of the Pacific Conference Series, Vol. 281, Astronomical Data Analysis Software and Systems XI, ed. D. A. Bohlender, D. Durand, & T. H. Handley, 228
  7. [7]Bhardwaj, A., Matsunaga, N., Huang, C. D., Riess, A. G., & Rejkuba, M. 2025, ApJ, 990, 63, doi: 10.3847/1538-4357/adf20bDOI
  8. [8]Bhardwaj, A., Kanbur, S., He, S., et al. 2019, ApJ, 884, 20, doi: 10.3847/1538-4357/ab38c2DOI
  9. [9]Bonnarel, F., Fernique, P., Bienaymé, O., et al. 2000, A&AS, 143, 33, doi: 10.1051/aas:2000331DOI
  10. [10]Caplar, N., Beebe, W., Branton, D., et al. 2025, arXiv e-prints, arXiv:2501.02103, doi: 10.48550/arXiv.2501.02103DOI
  11. [11]Flaugher, B., Diehl, H. T., Honscheid, K., et al. 2015, AJ, 150, 150, doi: 10.1088/0004-6256/150/5/150DOI
  12. [12]Fruscione, A., Glotfelty, K., Joye, W., & McDowell, J. 2026, AJ, 172, 116, doi: 10.3847/1538-3881/ae84c1DOI
  13. [13]Graham, M. L., Knop, R. A., Kennedy, T. D., et al. 2023, MNRAS, 519, 3881, doi: 10.1093/mnras/stac3363DOI
  14. [14]Green, G. 2018, The Journal of Open Source Software, 3, 695, doi: 10.21105/joss.00695DOI
  15. [15]Hambleton, K. M., Bianco, F. B., Street, R., et al. 2023, PASP, 135, 105002, doi: 10.1088/1538-3873/acdb9aDOI
  16. [16]Huang, C. D. 2024, in The Hubble Constant Tension, ed. E. Di Valentino & B. D. Brout (Singapore: Springer), 191–218
  17. [17]Huang, C. D., Yuan, W., Riess, A. G., et al. 2024, ApJ, 963, 83, doi: 10.3847/1538-4357/ad1ff8DOI
  18. [18]Huang, C. D., Marengo, M., Capodagli, A., et al. 2026, arXiv e-prints, arXiv:2609.30551, doi: 10.48550/arXiv.2609.30551DOI
  19. [19]Ita, Y., Menzies, J. W., Whitelock, P. A., et al. 2021, MNRAS, 500, 82, doi: 10.1093/mnras/staa3251DOI
  20. [20]Ivezić, Ž., Kahn, S. M., Tyson, J. A., et al. 2019, ApJ, 873, 111, doi: 10.3847/1538-4357/ab042cDOI
  21. [21]Iwanek, P., Kozłowski, S., Gromadzki, M., et al. 2021, ApJS, 257, 23, doi: 10.3847/1538-4365/ac1797DOI
  22. [22]Konchady, T., Macri, L. M., Yan, X., & Huang, J. Z. 2024, MNRAS, 531, 110, doi: 10.1093/mnras/stae1140DOI
  23. [23]Lengen, B., Anderson, R. I., & Cruz Reyes, M. 2026, arXiv e-prints, arXiv:2605.28941, doi: 10.48550/arXiv.2605.28941DOI
  24. [24]Lewis, M. O., Bhattacharya, R., Sjouwerman, L. O., et al. 2023, A&A, 677, A153, doi: 10.1051/0004-6361/202346568DOI
  25. [25]Macri, L. 2017, in European Physical Journal Web of Conferences, Vol. 152, European Physical Journal Web of Conferences (EDP), 07001
  26. [26]Malanchev, K., DeLucchi, M., Caplar, N., et al. 2025, arXiv e-prints, arXiv:2506.23955, doi: 10.48550/arXiv.2506.23955DOI
  27. [27]Marmo, C., & Bertin, E. 2008, in Astronomical Society of the Pacific Conference Series, Vol. 394, Astronomical Data Analysis Software and Systems XVII, ed. R. W. Argyle, P. S. Bunclark, & J. R. Lewis, 619
  28. [28]Menzies, J. W., Whitelock, P. A., & Feast, M. W. 2015, MNRAS, 452, 910, doi: 10.1093/mnras/stv1310DOI
  29. [29]Ngeow, C.-C., & Cheng, C.-Y. 2025, AJ, 169, 58, doi: 10.3847/1538-3881/ad99cfDOI
  30. [30]NSF-DOE Vera C. Rubin Observatory. 2026, Legacy Survey of Space and Time Data Preview 2 [Data set], NSF-DOE Vera C. Rubin Observatory, doi: 10.71929/rubin/3382528. https://www.osti.gov//servlets/purl/3382528
  31. [32]Parada, J., Heyl, J., Richer, H., Ripoche, P., & Rousseau-Nepton, L. 2023, MNRAS, 522, 195, doi: 10.1093/mnras/stad965DOI
  32. [33]Parto, T., Dehghani, S., Javadi, A., et al. 2023, ApJ, 942, 33, doi: 10.3847/1538-4357/aca471DOI
  33. [34]Pietrzyński, G., Graczyk, D., Gallenne, A., et al. 2019, Nature, 567, 200, doi: 10.1038/s41586-019-0999-4DOI
  34. [35]Sanders, J. L. 2023, MNRAS, 523, 2369, doi: 10.1093/mnras/stad1431DOI
  35. [36]Schlafly, E. F., & Finkbeiner, D. P. 2011, ApJ, 737, 103, doi: 10.1088/0004-637X/737/2/103DOI
  36. [37]Schlegel, D. J., Finkbeiner, D. P., & Davis, M. 1998, ApJ, 500, 525, doi: 10.1086/305772DOI
  37. [38]Skowron, D. M., & Soszyński, I. 2026, ARA&A, 64, 103, doi: 10.1146/annurev-astro-122325-024929DOI
  38. [39]Skowron, D. M., Skowron, J., Udalski, A., et al. 2021, ApJS, 252, 23, doi: 10.3847/1538-4365/abcb81DOI
  39. [40]Soszyński, I., Udalski, A., Szymański, M. K., et al. 2009, AcA, 59, 239, doi: 10.48550/arXiv.0910.1354DOI
  40. [41]Sun, Y., Zhang, B., Reid, M. J., et al. 2022, ApJ, 931, 74, doi: 10.3847/1538-4357/ac69e0DOI
  41. [42]Sun, Y., Zhang, J., Zhang, B., et al. 2023, Frontiers in Astronomy and Space Sciences, 10, 1232151, doi: 10.3389/fspas.2023.1232151
  42. [43]Valdes, F., Gruendl, R., & DES Project. 2014, in Astronomical Society of the Pacific Conference Series, Vol. 485, Astronomical Data Analysis Software and Systems XXIII, ed. N. Manset & P. Forshay, 379
  43. [44]Vera C. Rubin Observatory Team. 2026b, The Vera C. Rubin Observatory Data Preview 2, Technical Note RTN-115, NSF-DOE Vera C. Rubin Observatory, doi: 10.71929/rubin/3377440
  44. [45]Vera C. Rubin Observatory Team, Acero-Cuellar, T., Acosta, E., et al. 2026, AJ, 171, 360, doi: 10.3847/1538-3881/ae521f
  45. [46]Whitelock, P. A. 2013, in IAU Symposium, Vol. 289, Advancing the Physics of Cosmic Distances, ed. R. de Grijs, 209–216
  46. [47]Whitelock, P. A., Menzies, J. W., Feast, M. W., & Marigo, P. 2018, MNRAS, 473, 173, doi: 10.1093/mnras/stx2275DOI
  47. [48]Whitelock, P. A., Menzies, J. W., Feast, M. W., Nsengiyumva, F., & Matsunaga, N. 2013, MNRAS, 428, 2216, doi: 10.1093/mnras/sts188DOI

Paper details

vlnc:W_ec3dwj6f
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 0

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Physics and Astronomy

Not recorded for this paper

View statement
Looking for related papers…
Contents