The California Legacy Survey. I. A Catalog of 178 Planets from Precision Radial Velocity Monitoring of 719 Nearby Stars over Three Decades

The HIRES (Vogt et al. 1994) has been in operation on the Keck I Telescope since 1994 and has been used to measure stellar RVs via the Doppler technique since 1996 (Cumming et al. 2008). This technique relies on measuring the Doppler shift of starlight relative to a reference spectrum of molecular iodine, which is at rest in the observatory frame (Butler et al. 1996). We consistently set up HIRES with the same wavelength format on the CCDs for each observation and followed other standard procedures of CPS (Howard et al. 2010a). With the iodine technique, starlight passes through a glass cell of iodine gas heated to 50°C, imprinting thousands of molecular absorption lines onto the stellar spectrum, which act as a wavelength reference. We also collected an iodine-free "template" spectrum for each star. This spectrum is naturally convolved with the instrumental point-spread function (PSF) and is sampled at the resolving power of HIRES (R = 55,000–86,000, depending on the width of the decker used). These spectra are deconvolved using PSF measurements from spectra of featureless, rapidly rotating B stars with the iodine cell in the light path. The final, deconvolved intrinsic stellar spectra serve as ingredients in a forward-modeling procedure from which we measure relative Doppler shifts of each iodine-in spectrum of a given star (Valenti et al. 1995). We also used this process to compute uncertainties on the Doppler shifts. The uncertainty for each measurement is the standard error on the mean of the RVs for 700 segments of each spectrum (each 2 Å wide) run through the Doppler pipeline. We distinguish between "pre-upgrade" RVs (1996–2004; ∼3 m s⁻¹ uncertainties) and "post-upgrade" RVs (2004–present; ∼1 m s⁻¹ uncertainties). In 2004, HIRES was upgraded with a new CCD and other optical improvements. We account in the time series modeling for different RVs zero points (γ) for data from the two different eras.

The RVs reported here stem from HIRES observations with a long history. The RVs from 1996 to 2004 are based on HIRES spectra acquired by the California & Carnegie Planet Search (CCPS) collaboration and were reported in Cumming et al. (2008). CCPS continued to observe these stars, but split into two separate collaborations: CPS and the Lick-Carnegie Exoplanet Survey (LCES). This paper principally reports results from 41,804 CPS and CCPS HIRES spectra that were obtained and analyzed by our team during 1996–2020. In addition, we have included RVs computed by our pipeline for 7530 spectra of CLS stars taken by LCES during 2008–2014. These HIRES spectra were acquired with the same instrumental setup as the CPS spectra and are publicly available in the Keck Observatory Archive. Butler et al. (2017) separately published RVs based on the same HIRES observations from CCPS, CPS, and LCES for the 1996–2014 time span. The LCES and CPS Doppler pipelines diverged in ∼2007. Tal-Or et al. (2019) uncovered the 2004 zero-point offset, which we model with two independent offsets. They also claimed two second-order systematics in the LCES 2017 data set: a long-term drift of order <1 m s⁻¹, and a correlation between stellar RVs and time of night with respect to midnight. They estimated the long-term drift by averaging the zero points of three RV-quiet stars on each night, where possible. However, by our estimates, even the quietest stars exhibit 1–2 m s⁻¹ jitter in HIRES time series. Averaging the zero points of three such stars will likely yield a scatter of 1 m s⁻¹ across many nights. Additionally, they did not remove planet RV signals from their data before estimating the linear correlation between RV and time of night, and it is unclear how they derived the uncertainty in that correlation.

The APF-Levy spectrograph is a robotic telescope near the summit of Mount Hamilton, designed to find and characterize exoplanets with high-cadence Doppler spectroscopy (Radovan et al. 2014; Vogt et al. 2014). The facility consists of a 2.4 m telescope and the Levy Spectrometer, which has been optimized for optical Doppler shift measurements. The Doppler pipeline that was developed for Keck-HIRES also extracts RV measurements from APF spectra. Most of the APF data in the California Legacy Survey was collected as part of the APF-50 Survey (Fulton 2017), the stellar sample of which was drawn entirely from the Eta Earth sample. These two surveys have slightly different selection criteria. While both surveys have a distance cut d < 25 pc and luminosity cut M_V < 3, Eta-Earth cuts on apparent magnitude V < 11, whereas APF-50 has V < 7; Eta-Earth cuts on chromospheric activity $\mathrm{log}R{{\prime} }_{\mathrm{HK}}\lt -4.7$, whereas APF-50 has $\mathrm{log}R{{\prime} }_{\mathrm{HK}}\lt -4.95;$ and Eta-Earth cuts on decl. > −30°, whereas APF-50 has decl. > −10°. These stricter cuts were made to ensure higher data quality for the high-cadence APF survey.

The Hamilton Spectrograph is a high-resolution echelle spectrometer, attached to the 3 m Shane telescope on Mount Hamilton. Beginning in 1987, and ending in 2011 with a catastrophic iodine cell failure, the Lick Planet Search program (Fischer et al. 2014) monitored 387 bright FGKM dwarfs to search for and characterize giant exoplanets. This was one of the first surveys to produce precise RVs via Doppler spectroscopy with iodine cell calibration, and yielded RVs with precision in the range 3–10 m s⁻¹. The Lick Planet Search overlaps heavily with the Keck Planet Search and other CPS surveys, since these surveys drew from the same bright-star catalogs.

For each HIRES and APF spectrum from which we measure radial velocities, we also measure the strength of emission in the cores of the Ca ii H & K lines (S-values) following the techniques of Isaacson & Fischer (2010) and Robertson et al. (2014). There is a small, arbitrary offset between the HIRES and APF activity indices. We adopted uniform S-value uncertainties with values of 0.002 and 0.004 for HIRES and APF respectively. We provide activity indices along with our RV measurements. Missing values are the result of sky contamination and/or low SNR.

We collected long-term photometric observations of the subset of our sample that were included in the APF-50 survey (Fulton 2017), in order to search for evidence of rotation-induced stellar activity. We collected these measurements with Tennessee State University's Automated Photometric Telescopes (APTs) at Fairborn Observatory as part of a long-term program to study stellar magnetic activity cycles (Lockwood et al. 2013). Most stars have photometric data sets spanning 15–23 yr. The APTs are equipped with photomultiplier tubes that measure the flux in the Stromgren b and y bands relative to three comparison stars. We combined the differential b and y measurements into a single (b + y)/2 "passband" then converted the differential magnitudes into a relative flux normalized to 1.0. The precision in relative flux is typically between 0.001 and 0.0015. Further details of the observing strategy and data reduction pipeline are available in Henry (1999), Eaton et al. (2003), Henry et al. (2013). We make the photometric data available in a .tar.gz package.

We examined the range of observing cadences and observational baselines within the CLS sample, to determine whether stars without known planets were observed with strategies that differed significantly from those for stars with known planets. Figure 2 shows the distribution of number of observations and observational baselines for three groups of stars: the entire sample, the stars around which we detected planets, and the star around which we did not detect planets. Each of these three samples has a median baseline of 21 yr. Stars with detected planet have a median of 74 observations, compared to 35 observations for stars without detected planets and 41 observations for the entire CLS sample. A factor of two in number of observations will have a small but measurable impact on planet detectability of a given data set—and therefore on its search completeness contours.

Zoom In Zoom Out Reset image size

We developed an iterative approach to a search for periodic signals in RV data in order to generate the CLS planet catalog. We outline this algorithm, which we developed as the open-source Python package RVSearch and have made public alongside the publication of this paper. Figure 5 is a flowchart that lays out each step of the algorithm, and Figure 6 is a visualization of an example RVSearch output, where the top two panels show the final model, and each successive row shows an iterative search for each signal in the model. First, we provide an initial model, from which the iterative search begins. This initial model contains an RV data set and a likelihood function. The natural logarithm of the latter is defined as

$$\begin{eqnarray}&&\mathrm{ln}({ \mathcal L })=-\displaystyle \frac{1}{2}\displaystyle \sum _{i}\left[\displaystyle \frac{{\left({v}_{i}-m({t}_{i})-{\gamma }_{D}\right)}^{2}}{{\sigma }_{i}^{2}}+\mathrm{ln}(2\pi {\sigma }_{i}^{2})\right],\end{eqnarray} \tag{ 1 }$$

where i is the measurement index, v_i is the ith RV measurement, γ_D is the offset of the instrumental data set from which the ith measurement is drawn, and ${\sigma }_{i}^{2}$ is the quadrature sum of the instrumental error and the stellar jitter term of the ith measurement's instrumental data set. Here, m(t_i ) is the model RV at time t_i , defined as

$$\begin{eqnarray}&&m(t)=\displaystyle \sum _{n}K(t| {K}_{n},{P}_{n},{e}_{n},{\omega }_{n},{t}_{{cn}})+\dot{\gamma }(t-{t}_{0})+\ddot{\gamma }{\left(t-{t}_{0}\right)}^{2},\end{eqnarray} \tag{ 2 }$$

where n is a given Keplerian orbit in the model, K(t∣K, P, e, ω, t_c ) is the Keplerian orbit RV signature at time t given RV amplitude K, period P, eccentricity e, argument of periastron ω, and time of inferior conjunction t_c , $\dot{\gamma }$ is a linear trend term, $\ddot{\gamma }$ is a quadratic trend term, and t₀ is a reference time, which we defined as the median time of observation.

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

We used RadVel (Fulton et al. 2018) to fit Keplerian orbits. The initial likelihood model contains either a one-planet Keplerian model with undefined orbital parameters, or a predefined model including trend/curvature terms and/or Keplerian terms associated with known orbital companions. We defaulted to performing a blind search starting with the undefined single-planet model, and we only supply a predefined model if there is evidence for a highly eccentric companion whose period is misidentified by our search algorithm. Several highly eccentric stellar binaries satisfy this criterion, as do two planets: HD 120066 b (Blunt et al. 2019) and HD 80606 b (Wittenmyer et al. 2007b).

Before beginning a blind search, RVSearch determines whether the data merits a trend with curvature, a linear trend, or no trend. It does this by fitting each of these three models to the data, then performing a goodness-of-fit test to decide which model is favored. We measured the Bayesian Information Criterion (BIC) for each of the three models, and computed the ΔBIC between each model. RVSearch selects the linear model if it has ΔBIC = 5 with respect to the flat model, and the quadratic model if it has ΔBIC = 5 with respect to the linear model. We did not perform this test on data sets that contain eccentric companions with orbital periods greater than the data's observational baseline, since such data sets would be better fit with a long-period Keplerian orbit than with linear and parabolic trends. The Bayesian information criterion is defined as

$$\begin{eqnarray}&&\mathrm{BIC}=\mathrm{kln}({{\rm{n}}}_{\mathrm{obs}})-2\mathrm{ln}({ \mathcal L }),\end{eqnarray} \tag{ 3 }$$

where n_obs is the number of observations, k is the number of free model parameters, and $\mathrm{ln}({ \mathcal L })$ is the log-likelihood of the model in question.

Once we provide an initial model, RVSearch defines an orbital period grid over which to search, with sampling such that the difference in frequency between adjacent grid points is $\tfrac{1}{2\pi \tau }$, where τ is the observational baseline. We chose this grid spacing in accordance with Horne & Baliunas (1986), who state that, in frequency space, a Lomb–Scargle periodogram has a minimum peak width of $\tfrac{1}{2\pi \tau }$. For each data set, we searched for periodicity between two days and five times the observational baseline. Searching out to five times the baseline only adds a few more points to the period grid, and it allows for the possibility of recovering highly eccentric, ultra-long-period planet candidates with best-fit orbital period.

The search algorithm then computes a goodness-of-fit periodogram by iterating through the period grid and fitting a sinusoid with a fixed period to the data. We measure goodness-of-fit as the ΔBIC at each grid point between the best-fit, n+1-planet model with the given fixed period, and the n-planet fit to the data (this is the zero-planet model for the first planet search).

After constructing a ΔBIC periodogram, the algorithm performs a linear fit to a log-scale histogram of the periodogram power values. The algorithm then extrapolates a ΔBIC detection threshold corresponding to an empirical false-alarm probability of 0.1%, meaning that, according to the power-law fit, only 0.1% of periodogram values are expected to fall beyond this threshold. This process follows the detection methodology outlined in Howard & Fulton (2016).

If a periodic signal exceeds this detection threshold, RVSearch refines the fit of the corresponding Keplerian orbit by performing a maximum a posteriori fit with all model parameters free, including eccentricity, and records the BIC of that best-fit model. RVSearch includes two hard-bound priors, which constrain K > 0 and 0 < = e < 1. The algorithm then adds another planet to the RV model and conducts another grid search, leaving all parameters of the known Keplerian orbits free so that they might converge to a more optimal solution. In the case of the search for the second planet in a system, the goodness of fit is defined as the difference between the BIC of the best-fit one-planet model and the BIC of the two-planet model at each fixed period in the grid. RVSearch once again sets a detection threshold in the manner described above, and this iterative search continues until it returns a nondetection.

This iterative periodogram search is superior to a Lomb–Scargle residual subtraction search in two key ways. First, this process fits for the instrument-specific parameters of each data set, stellar jitter and RV-offset, as free parameters throughout the search. Second, by leaving the known model parameters free while searching for each successive planet, we allow the solutions for the already discovered planets to reach better max-likelihood solutions that only become evident with the inclusion of another planet in the model.

Note that our search and model comparison process is not Bayesian; we do not use priors to inform our model selection, and we do not sample posteriors, beyond a grid search in period space, until we settle upon a final model. We use the BIC as our model comparison metric because it incorporates the number of free parameters as a penalty on more complex models—which, in our case, corresponds to models with additional planets.

We make RVSearch publicly available alongside this paper via a GitHub repository. See the RVSearch website ²⁰ for installation and use instructions.

We characterized the search completeness of each individual data set, and of the entire survey, by running injection-recovery tests. Once RVSearch completed an iterative search of a data set, it injected synthetic planets into the data and ran one more search iteration to determine whether it recovers these synthetic planets in that particular data set. We ran 3000 injection tests for each star. We drew the injected planet period and $M\sin i$ from log-uniform distributions, and drew eccentricity from the beta distribution described in Kipping (2013), which was fit to a population of RV-observed planets.

We used the results of these injection tests to compute search completeness for each individual data set, and report constant $M\sin i$/a contours of detection probability. Figure 7 shows examples of these contours and the corresponding RVs for three different stars, all early G-type: one with 25 observations, one with 94, and one with 372. We make the 10th, 50th, and 90th percentile completeness contours for each individual star available in the .tar.gz package.

Zoom In Zoom Out Reset image size

Once RVSearch returned max-likelihood estimates of the orbital model parameters for a given data set, we sampled the model posterior using affine-invariant sampling, implemented via emcee and RadVel (Foreman-Mackey et al. 2013; Fulton et al. 2018). We sampled using the orbital parameter basis {$\mathrm{log}P\ K\ {t}_{c}\ \sqrt{e}\sin \omega \ \sqrt{e}\cos \omega$}. We placed uniform priors on all fitting parameters, with hard bounds such that K > 0 and 0 ≤ e < 1. We fit in $\mathrm{log}P$ space to efficiently sample orbits with periods longer than our observational baseline, and in $\sqrt{e}\sin \omega$ and $\sqrt{e}\cos \omega$ to minimize bias toward higher eccentricities (Lucy & Sweeney 1971). We reported parameter estimates and uncertainties as the median and ±1σ intervals.

If a data set is so poorly constrained by a Keplerian model that emcee's affine-invariant sampler cannot efficiently sample the posterior distribution, we instead used a rejection sampling algorithm to estimate the posterior. In these cases, we used TheJoker (Price-Whelan et al. 2017), a modified Markov Chain Monte Carlo (MCMC) algorithm designed to sample Keplerian orbital fits to sparse RV measurements. We chose a flat prior on $\mathrm{log}P$, with a minimum at the observing baseline and a maximum at twenty times the observing baseline. We drew orbital eccentricity from a beta prior weighted toward zero, as modeled in Kipping (2013), in order to downweight orbits with arbitrarily high eccentricity, which can be viable fits to sparse or otherwise underconstraining RV data sets.

We performed a series of tests to vet each planet candidate discovered by our search pipeline. The following subsections each detail one test we perform to rule out one way in which a signal might be a false-positive. We also represent this process with a flowchart in Figure 8, and include a table of all false-positive signals recovered by RVSearch in Table 6.

5.4.1. Stellar Activity, Magnetic/Long-period

5.4.2. Stellar Activity, Rotation/Short-period

5.4.3. Yearly Alias

When we find a signal with a period of a year or an integer fraction of a year, we investigate whether it is an alias of long-period power, or a systematic that is correlated with the barycentric velocity at the time of observation or Doppler fitting parameters. We do this by recomputing the associated RVs using a different template observation. When another template observation was unavailable, we were able to take one using Keck-HIRES during collaborator observing nights. Templates taken in poor observing conditions or when barycentric velocity with respect to the observed star is high can produce systematic errors in the Doppler code. If a search of this new data set returns a nondetection, or detection at a significantly different period, we conclude that this signal is an alias. Figure 9 shows the presence of yearly alias power in our survey, seen in a stack of the final nondetection periodograms of all CLS stars.

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

HD 3765 is a K2 dwarf at a distance of 17.9 pc (Gaia Collaboration et al. 2018). Figure 14 shows the RVSearch results for this star. We recovered a signal with a period of 3.36 yr. Table 1 reports all planet parameters. There is significant periodicity in the S-value time series, but concentrated around a period of 12 yr. Figure 15 shows a Lomb–Scargle periodogram of the S-value time series. Furthermore, we find no correlation between the RVs and S-values. Thus, we label this signal as a confirmed planet, with $M\sin i$ = 0.173 ± 0.014 M_J and a = 2.108 ± 0.033 au. The magnetic activity cycle is too weak for RVSearch to recover, but is evident in the best-fit RV residuals. We used RadVel to model this activity cycle with a squared-exponential Gaussian process, and report MCMC-generated posteriors for both orbital and Gaussian process parameters in Figures 16 and 17.

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

HD 24040 is a G1 dwarf at a distance of 46.7 pc. Figure 18 shows the RVSearch results for this star. It hosts a known gas giant (Wright et al. 2007; Feng et al. 2015) with a semimajor axis that we measured as a = 4.72 ± 0.18 au, an orbital period of 9.53 ± 10⁻⁴ yr, and a minimum mass $M\sin i$ = 4.09 ± 0.22 M_J. We have extended the observational baseline of our HIRES measurements to 21.7 yr, constrained the long-term trend and curvature of the RVs, and discovered a new exoplanet, a sub-Saturn ($M\sin i$ = 0.201 ± 0.027 M_J) on a 1.4 yr orbit (a = 1.30 ± 0.021 au) that is consistent with circular. The S-values are uncorrelated with the RVs of both planet signals, after removing the long-term trend. Figure 19 shows a Lomb–Scargle periodogram of the S-value time series. Table 1 reports all planet parameters.

Zoom In Zoom Out Reset image size

In addition to the newly detected sub-Saturn, we further constrained the known linear trend in the RVs and found evidence for a curvature term as well. RVSearch detected a curvature term with model preference ΔBIC > 10 over a purely linear trend. We measured the linear trend to be 0.00581 ± 0.00044 m s⁻¹ day⁻¹, and the curvature to be − 6.6 × 10⁻⁷ ± 1.2 × 10⁻⁷ m s⁻¹ day⁻¹, a 5.5σ detection. The trend and curvature parameters are slightly correlated in the posterior, but neither is correlated with any of the Keplerian orbital parameters in the model. Therefore, we kept the curvature term that RVSearch selected in our model. This long-term trend is low-amplitude enough that it may be caused by another planet in the system, orbiting beyond 30 au. Gaia astrometry or another two decades of RVs may provide further constraints on this object.

Zoom In Zoom Out Reset image size

HD 26161 is a G0 dwarf located at a distance of 50.0 pc. Figure 20 shows the RVSearch results for this star. Our RVs are consistent with a long-period, eccentric companion, and RVSearch detected this long-period signal. Due to the sparseness of the data and the fractional orbital coverage, traditional MCMC methods fail to return a well-sampled model posterior. Since the data underconstrains our model, we used TheJoker to sample the posterior, which is consistent with an extremely long-period gas giant with minimum mass $M\sin i$ = ${13.5}_{-3.7}^{+8.5}$ M_J, semimajor axis $a={20.4}_{-4.9}^{+7.9}$ au, and eccentricity $e={0.82}_{-0.05}^{+0.06}$. Table 1 reports current estimates of all orbital parameters, and Figure 21 shows their posterior distributions. A Keplerian model is significantly preferred over a quadratic trend, with ΔBIC > 15.

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

The Simbad stellar catalog designates HD 26161 as a stellar multiple. We used Gaia to identify a binary companion with similar parallax and within 60''. This companion has an effective temperature identified from Gaia colors of 4053 K, and a projected separation of 562 au. A stellar companion that is currently separated from its primary by more than 560 au could not cause a change in RV of 100 m s⁻¹ over 4 yr. This curve is far more likely caused by an inner planetary or substellar companion approaching periastron.

Figure 22 shows a sample of possible orbits for HD 26161 b, drawn from our rejection sampling posteriors and projected over the next decade. We will continue to monitor HD 26161 with HIRES at moderate cadence, and have begun observing this star with APF. As we gather more data during the approach to periastron, we can tighten our constraints on the minimum mass, eccentricity, and orbital separation of HD 26161 b.

Zoom In Zoom Out Reset image size

HD 66428 is a G8 dwarf found at a distance of 53.4 pc. Figure 23 shows the RVSearch results for this star. This system has one well-constrained cold Jupiter (Butler et al. 2006b) and an outer companion candidate first characterized in Bryan et al. (2016) as a linear trend. With four more years of HIRES data, we now see curvature in the RVs and a clear detection in RVSearch, and can place constraints on this outer candidate's orbit with a Keplerian model. The Keplerian orbit for the outer candidate is preferred to a parabolic trend with ΔBIC > 30. A maximum likelihood fit gives an orbital period of P = 36.4 yr. However, since we have only observed a partially resolved orbit so far, the orbit posterior in period space is wide and asymmetric. MCMC sampling produces $P={88}_{-49}^{+153}$ yr. Table 1 reports current estimates of all orbital parameters.

Zoom In Zoom Out Reset image size

The model parameters are $M\sin i$ = ${27}_{-17}^{+22}$ M_J, $a={23.0}_{-7.6}^{+19.0}$ au, and $e={0.31}_{-0.13}^{+0.13}$. This orbital companion could be a massive gas giant or a low-mass star, if we only consider constraints from RV modeling. However, Bryan et al. (2016) used NIRC2 Adaptive-Optics images to place upper bounds on the mass and semimajor axis of an outer companion, at a time when it only presented as a linear trend in HIRES RVs. They found an upper bound of ≈100 M_J on mass, not just $M\sin i$, and an upper bound of ≈ 150 au on a. We will continue to monitor this star with HIRES to further constrain the mass and orbit of HD 66428c.

HD 95735 (GJ 411) is an M2 dwarf found at a distance of 2.55 pc. Figure 24 shows the RVSearch results for this star. This system has one known short-period super-Earth, with $M\sin i$ = 3.53 M_⊕ and an orbital period of 12.9 days. Our detection of this planet was driven by high-cadence APF data. This planet was first reported by Díaz et al. (2019), who also noted long-period power in their SOPHIE RV data, but they did not have a sufficiently long baseline or the activity metrics necessary to determine the origin of this power. With our HIRES post-upgrade and APF observations, we have an observational baseline of 14 yr, allowing us to confirm this long-period signal as a planet with $M\sin i$ = 24.7 ± 3.6 M_⊕ and an orbital period P = 8.46 yr. Table 1 reports all planet parameters. Since GJ 4ll is a cool M dwarf, the Lick-Hamilton and HIRES pre-upgrade data are not reliable, because those detectors are not sufficiently high-resolution to capture a cool M dwarf's dense spectral lines (Fischer et al. 2014).

Zoom In Zoom Out Reset image size

Figure 25 shows a Lomb–Scargle periodogram of the HIRES S-value time series. There is a long-period trend in the HIRES S-value time series, with significant power at and beyond 25 yr, but no significant power near the orbital period of the outer candidate. Therefore, we included this candidate in our catalog as a new planet candidate, to be verified and constrained with several more years of HIRES observations.

RVSearch also recovered a highly eccentric, 216 day signal, but this signal correlates with APF systematics. Therefore, we labeled it as a false positive. This systematic remained when we applied RVSearch only to the HIRES post-upgrade and APF data and left out the problematic pre-upgrade and Lick data.

Zoom In Zoom Out Reset image size

HD 107148 is a G5 dwarf at a distance of 49.5 pc. Figure 26 shows the RVSearch results for this star. Butler et al. (2006b) reported a planet with a period of 44 days. They reported periodicity at 77 days, but determined that this was an alias of the 44 day signal. The 77 day signal is significantly stronger in our likelihood periodogram, as seen in Figure 26, and fits the data better than a 44 day Keplerian does, by a significant ΔBIC. This constitutes strong evidence that the true period of this planet is 77 days. We report new orbital parameters for this planet in Table 3.

Zoom In Zoom Out Reset image size

Table 3. Planet Catalog

CPS Name	Lit. Name	$M\sin i$ (MJ)	a (au)	e	References
HD 104067 b	HD 104067 b	${0.202}_{-0.017}^{+0.017}$	${0.2673}_{-0.0033}^{+0.0032}$	${0.247}_{-0.082}^{+0.080}$	Vogt et al. (2000)
HD 10697 b	HD 10697 b	${6.39}_{-0.13}^{+0.13}$	${2.156}_{-0.02}^{+0.02}$	${0.0998}_{-0.0082}^{+0.0082}$	Butler et al. (2006b)
HD 107148 b	HD 107148 b	${0.203}_{-0.015}^{+0.015}$	${0.3668}_{-0.0048}^{+0.0047}$	${0.174}_{-0.075}^{+0.071}$	Butler et al. (2006b)
HD 107148c	HD 107148c	${0.0626}_{-0.0098}^{+0.0097}$	${0.1406}_{-0.0018}^{+0.0018}$	${0.34}_{-0.16}^{+0.13}$	This work
HD 108874 b	HD 108874 b	${1.32}_{-0.047}^{+0.047}$	${1.053}_{-0.016}^{+0.016}$	${0.144}_{-0.016}^{+0.016}$	Butler et al. (2003)
HD 108874c	HD 108874c	${1.14}_{-0.050}^{+0.052}$	${2.83}_{-0.045}^{+0.044}$	${0.265}_{-0.031}^{+0.030}$	Vogt et al. (2005)
HD 114729 b	HD 114729 b	${0.892}_{-0.053}^{+0.053}$	${2.094}_{-0.022}^{+0.022}$	${0.098}_{-0.050}^{+0.048}$	Butler et al. (2003)
HD 114783 b	HD 114783 b	${1.033}_{-0.033}^{+0.034}$	${1.164}_{-0.016}^{+0.016}$	${0.126}_{-0.016}^{+0.016}$	Vogt et al. (2002)
HD 114783c	HD 114783c	${0.66}_{-0.047}^{+0.046}$	${4.97}_{-0.11}^{+0.12}$	${0.114}_{-0.058}^{+0.058}$	Bryan et al. (2016)
HD 115617 b	61 Vir b	${0.0507}_{-0.0036}^{+0.0035}$	${0.215}_{-0.0029}^{+0.0028}$	${0.068}_{-0.047}^{+0.067}$	Vogt et al. (2010)
HD 115617c	61 Vir c	${0.0161}_{-0.0016}^{+0.0017}$	${0.04956}_{-0.00067}^{+0.00066}$	${0.099}_{-0.070}^{+0.091}$	Vogt et al. (2010)
HD 117176 b	70 Vir b	${7.24}_{-0.13}^{+0.13}$	${0.4766}_{-0.0044}^{+0.0043}$	${0.3989}_{-0.0012}^{+0.0011}$	Butler & Marcy (1996)
HD 117207 b	HD 117207 b	${1.87}_{-0.075}^{+0.076}$	${3.744}_{-0.060}^{+0.059}$	${0.142}_{-0.026}^{+0.027}$	Marcy et al. (2005)
HD 11964a b	HD 11964a b	${0.631}_{-0.027}^{+0.027}$	${3.185}_{-0.032}^{+0.032}$	${0.101}_{-0.046}^{+0.044}$	Butler et al. (2006b)
HD 11964a c	HD 11964a c	${0.0766}_{-0.0061}^{+0.0063}$	${0.2315}_{-0.0022}^{+0.0022}$	${0.106}_{-0.071}^{+0.088}$	Wright et al. (2009)
HD 120066 b	HR 5183 b	${3.15}_{-0.17}^{+0.18}$	${25.0}_{-8.3}^{+19.0}$	${0.886}_{-0.056}^{+0.049}$	Blunt et al. (2019)
HD 120136 b	tau Boo b	${4.3}_{-0.075}^{+0.075}$	${0.04869}_{-0.00040}^{+0.00039}$	${0.0074}_{-0.0048}^{+0.0059}$	Butler et al. (1997)
HD 12661 b	HD 12661 b	${2.283}_{-0.063}^{+0.062}$	${0.824}_{-0.011}^{+0.011}$	${0.3597}_{-0.0044}^{+0.0046}$	Fischer et al. (2001)
HD 12661c	HD 12661c	${1.855}_{-0.054}^{+0.054}$	${2.86}_{-0.039}^{+0.038}$	${0.025}_{-0.012}^{+0.012}$	Fischer et al. (2003)
HD 126614 b	HD 126614 b	${0.356}_{-0.030}^{+0.032}$	${2.291}_{-0.027}^{+0.026}$	${0.577}_{-0.065}^{+0.069}$	Howard et al. (2010a)
HD 128311 b	HD 128311 b	${3.25}_{-0.12}^{+0.12}$	${1.742}_{-0.021}^{+0.020}$	${0.196}_{-0.057}^{+0.047}$	Butler et al. (2003)
HD 128311c	HD 128311c	${2.0}_{-0.15}^{+0.15}$	${1.088}_{-0.013}^{+0.012}$	${0.283}_{-0.04}^{+0.04}$	Vogt et al. (2005)
HD 130322 b	HD 130322 b	${1.149}_{-0.036}^{+0.037}$	${0.0929}_{-0.0011}^{+0.0011}$	${0.015}_{-0.010}^{+0.016}$	Udry et al. (2000)
HD 1326 b	HD 1326 b	${0.0171}_{-0.0013}^{+0.0013}$	${0.07321}_{-0.00048}^{+0.00047}$	${0.075}_{-0.052}^{+0.072}$	Howard et al. (2014)
HD 134987 b	HD 134987 b	${1.623}_{-0.049}^{+0.049}$	${0.817}_{-0.012}^{+0.012}$	${0.2281}_{-0.0077}^{+0.0078}$	Vogt et al. (2000)
HD 134987c	HD 134987c	${0.934}_{-0.060}^{+0.063}$	${6.62}_{-0.15}^{+0.16}$	${0.154}_{-0.050}^{+0.049}$	Jones et al. (2010)
HD 136925 b	HD 136925 b	${0.84}_{-0.074}^{+0.078}$	${5.13}_{-0.11}^{+0.12}$	${0.103}_{-0.070}^{+0.094}$	This work
HD 13931 b	HD 13931 b	${1.911}_{-0.076}^{+0.077}$	${5.323}_{-0.091}^{+0.091}$	${0.02}_{-0.014}^{+0.021}$	Howard et al. (2010a)
HD 141004 b	HIP 77257 b	${0.0428}_{-0.0045}^{+0.0047}$	${0.1238}_{-0.002}^{+0.002}$	${0.16}_{-0.10}^{+0.11}$	This work
HD 141399 b	HD 141399 b	${1.329}_{-0.047}^{+0.046}$	${0.693}_{-0.012}^{+0.012}$	${0.0465}_{-0.0068}^{+0.0068}$	Vogt et al. (2014)
HD 141399c	HD 141399c	${1.263}_{-0.048}^{+0.047}$	${2.114}_{-0.037}^{+0.036}$	${0.044}_{-0.013}^{+0.013}$	Vogt et al. (2014)
HD 141399 d	HD 141399 d	${0.452}_{-0.017}^{+0.017}$	${0.4176}_{-0.0073}^{+0.0070}$	${0.053}_{-0.015}^{+0.015}$	Vogt et al. (2014)
HD 141399 e	HD 141399 e	${0.644}_{-0.040}^{+0.042}$	${4.5}_{-0.11}^{+0.11}$	${0.047}_{-0.033}^{+0.052}$	Vogt et al. (2014)
HD 143761 b	rho CrB b	${1.057}_{-0.024}^{+0.024}$	${0.2213}_{-0.0025}^{+0.0025}$	${0.0355}_{-0.0037}^{+0.0035}$	Noyes et al. (1997)
HD 143761c	rho CrB c	${0.0885}_{-0.0056}^{+0.0055}$	${0.4157}_{-0.0047}^{+0.0046}$	${0.044}_{-0.031}^{+0.050}$	Fulton et al. (2016)
HD 145675 b	14 Her b	${4.85}_{-0.14}^{+0.15}$	${2.83}_{-0.041}^{+0.040}$	${0.3674}_{-0.0038}^{+0.0035}$	Wittenmyer et al. (2007b)
HD 145675c	14 Her c	${5.8}_{-1.0}^{+1.4}$	${16.4}_{-4.3}^{+9.3}$	${0.45}_{-0.15}^{+0.17}$	This work
HD 145934 b	HD 145934 b	${2.04}_{-0.23}^{+0.24}$	${4.73}_{-0.21}^{+0.20}$	${0.057}_{-0.039}^{+0.057}$	Feng et al. (2015)
HD 1461 b	HD 1461 b	${0.0208}_{-0.0018}^{+0.0019}$	${0.06361}_{-0.00099}^{+0.00096}$	${0.062}_{-0.042}^{+0.060}$	Rivera et al. (2010)
HD 1461c	HD 1461c	${0.0222}_{-0.0028}^{+0.0028}$	${0.1121}_{-0.0017}^{+0.0017}$	${0.112}_{-0.077}^{+0.10}$	Mayor et al. (2011)
HD 147379A b	HD 147379A b	${0.096}_{-0.012}^{+0.012}$	${0.3315}_{-0.0024}^{+0.0024}$	${0.096}_{-0.068}^{+0.110}$	Reiners et al. (2018)
HD 154345 b	HD 154345 b	${0.905}_{-0.089}^{+0.071}$	${4.272}_{-0.080}^{+0.086}$	${0.038}_{-0.027}^{+0.036}$	Wright et al. (2007)
HD 156279 b	HD 156279 b	${9.5}_{-0.31}^{+0.31}$	${0.5039}_{-0.0084}^{+0.0082}$	${0.64779}_{-0.00066}^{+0.00068}$	Díaz et al. (2012)
HD 156279c	HD 156279c	${9.44}_{-0.32}^{+0.31}$	${5.46}_{-0.093}^{+0.091}$	${0.2597}_{-0.0049}^{+0.0050}$	Bryan et al. (2016)
HD 156668 b	HD 156668 b	${0.0991}_{-0.0077}^{+0.0079}$	${1.57}_{-0.017}^{+0.017}$	${0.089}_{-0.061}^{+0.084}$	Howard et al. (2011)
HD 156668c	HD 156668c	${0.0158}_{-0.0013}^{+0.0013}$	${0.05025}_{-0.00051}^{+0.00050}$	${0.235}_{-0.072}^{+0.072}$	This work
HD 16141 b	HD 16141 b	${0.25}_{-0.017}^{+0.017}$	${0.3609}_{-0.0075}^{+0.0072}$	${0.195}_{-0.051}^{+0.051}$	Marcy et al. (2000)
HD 164922 b	HD 164922 b	${0.344}_{-0.013}^{+0.013}$	${2.149}_{-0.025}^{+0.025}$	${0.065}_{-0.029}^{+0.027}$	Butler et al. (2006b)
HD 164922c	HD 164922c	${0.0451}_{-0.0036}^{+0.0036}$	${0.3411}_{-0.0040}^{+0.0039}$	${0.096}_{-0.066}^{+0.088}$	Fulton et al. (2016)
HD 164922 d	HD 164922 d	${0.0331}_{-0.0031}^{+0.0031}$	${0.2292}_{-0.0027}^{+0.0026}$	${0.086}_{-0.060}^{+0.083}$	This work
HD 164922 e	HD 164922 e	${0.0149}_{-0.0021}^{+0.0021}$	${0.1023}_{-0.0012}^{+0.0012}$	${0.18}_{-0.12}^{+0.17}$	Benatti et al. (2020)
HD 167042 b	HD 167042 b	${1.59}_{-0.13}^{+0.13}$	${1.304}_{-0.043}^{+0.040}$	${0.092}_{-0.057}^{+0.061}$	Johnson et al. (2008)
HD 168009 b	HIP 89474 b	${0.03}_{-0.0037}^{+0.0038}$	${0.1192}_{-0.0018}^{+0.0017}$	${0.121}_{-0.082}^{+0.110}$	This work
HD 168443 b	HD 168443 b	${7.92}_{-0.16}^{+0.16}$	${0.2977}_{-0.0030}^{+0.0029}$	${0.5313}_{-0.00061}^{+0.00062}$	Marcy et al. (1999)
HD 168443c	HD 168443c	${17.76}_{-0.35}^{+0.35}$	${2.881}_{-0.029}^{+0.028}$	${0.2112}_{-0.0013}^{+0.0013}$	Marcy et al. (2001)
HD 168746 b	HD 168746 b	${0.2294}_{-0.0075}^{+0.0078}$	${0.06504}_{-0.00056}^{+0.00054}$	${0.098}_{-0.026}^{+0.026}$	Pepe et al. (2002)
HD 169830 b	HD 169830 b	${2.957}_{-0.070}^{+0.069}$	${0.8131}_{-0.0085}^{+0.0082}$	${0.306}_{-0.013}^{+0.012}$	Naef et al. (2001)
HD 169830c	HD 169830c	${3.51}_{-0.12}^{+0.12}$	${3.283}_{-0.036}^{+0.035}$	${0.257}_{-0.019}^{+0.019}$	Mayor et al. (2004)
HD 170469 b	HD 170469 b	${0.555}_{-0.072}^{+0.075}$	${2.212}_{-0.044}^{+0.040}$	${0.15}_{-0.11}^{+0.16}$	Fischer et al. (2007)
HD 177830 b	HD 177830 b	${1.347}_{-0.098}^{+0.097}$	${1.179}_{-0.043}^{+0.040}$	${0.028}_{-0.017}^{+0.017}$	Vogt et al. (2000)
HD 177830c	HD 177830c	${0.104}_{-0.015}^{+0.016}$	${0.496}_{-0.018}^{+0.017}$	${0.54}_{-0.15}^{+0.12}$	Meschiari et al. (2011)
HD 178911B b	HD 178911B b	${7.07}_{-0.23}^{+0.22}$	${0.34}_{-0.0055}^{+0.0053}$	${0.1132}_{-0.0025}^{+0.0025}$	Zucker et al. (2002)
HD 179949 b	HD 179949 b	${0.966}_{-0.030}^{+0.031}$	${0.04439}_{-0.00046}^{+0.00045}$	${0.016}_{-0.011}^{+0.017}$	Tinney et al. (2001)
HD 181234 b	HD 181234 b	${8.9}_{-0.76}^{+1.90}$	${7.48}_{-0.23}^{+0.39}$	${0.793}_{-0.082}^{+0.086}$	Rickman et al. (2019)
HD 186427 b	16 Cyg B b	${1.752}_{-0.054}^{+0.054}$	${1.676}_{-0.026}^{+0.025}$	${0.6832}_{-0.0031}^{+0.0031}$	Cochran et al. (1997)
HD 187123 b	HD 187123 b	${0.501}_{-0.016}^{+0.016}$	${0.04185}_{-0.00067}^{+0.00065}$	${0.004}_{-0.0028}^{+0.0040}$	Butler et al. (1998)
HD 187123c	HD 187123c	${1.713}_{-0.058}^{+0.058}$	${4.431}_{-0.072}^{+0.071}$	${0.227}_{-0.017}^{+0.016}$	Wright et al. (2009)
HD 188015 b	HD 188015 b	${1.455}_{-0.059}^{+0.060}$	${1.187}_{-0.019}^{+0.018}$	${0.17}_{-0.026}^{+0.025}$	Marcy et al. (2005)
HD 189733 b	HD 189733 b	${1.162}_{-0.035}^{+0.036}$	${0.03126}_{-0.00037}^{+0.00036}$	${0.027}_{-0.018}^{+0.021}$	Bouchy et al. (2005)
HD 190360 b	HD 190360 b	${1.492}_{-0.043}^{+0.043}$	${3.955}_{-0.053}^{+0.052}$	${0.3274}_{-0.0087}^{+0.0081}$	Naef et al. (2003)
HD 190360c	HD 190360c	${0.0674}_{-0.0026}^{+0.0027}$	${0.1294}_{-0.0017}^{+0.0017}$	${0.165}_{-0.028}^{+0.027}$	Vogt et al. (2005)
HD 192263 b	HD 192263 b	${0.658}_{-0.03}^{+0.03}$	${0.154}_{-0.0020}^{+0.0019}$	${0.04}_{-0.027}^{+0.037}$	Santos et al. (2000)
HD 192310 b	HD 192310 b	${0.0451}_{-0.0060}^{+0.0064}$	${0.3262}_{-0.0037}^{+0.0035}$	${0.14}_{-0.093}^{+0.120}$	Howard et al. (2011)
HD 195019 b	HD 195019 b	${3.655}_{-0.090}^{+0.091}$	${0.1376}_{-0.0017}^{+0.0017}$	${0.0198}_{-0.0032}^{+0.0032}$	Fischer et al. (1999)
HD 209458 b	HD 209458 b	${0.665}_{-0.022}^{+0.022}$	${0.04635}_{-0.00070}^{+0.00068}$	${0.0105}_{-0.0074}^{+0.0110}$	Henry et al. (2000)
HD 210277 b	HD 210277 b	${1.236}_{-0.033}^{+0.032}$	${1.123}_{-0.014}^{+0.014}$	${0.472}_{-0.0056}^{+0.0055}$	Marcy et al. (1999)
HD 213472 b	HD 213472 b	${3.48}_{-0.59}^{+1.10}$	${13.0}_{-2.6}^{+5.7}$	${0.53}_{-0.085}^{+0.120}$	This work
HD 216520 b	HIP 112527 b	${0.0326}_{-0.0038}^{+0.0036}$	${0.1954}_{-0.0025}^{+0.0025}$	${0.19}_{-0.12}^{+0.13}$	Burt et al. 2020
HD 217014 b	51 Peg b	${0.464}_{-0.014}^{+0.014}$	${0.05236}_{-0.00079}^{+0.00076}$	${0.0042}_{-0.0030}^{+0.0046}$	Mayor & Queloz (1995)
HD 217107 b	HD 217107 b	${1.385}_{-0.039}^{+0.039}$	${0.0739}_{-0.0011}^{+0.0010}$	${0.1279}_{-0.0016}^{+0.0017}$	Fischer et al. (1999)
HD 217107c	HD 217107c	${4.31}_{-0.13}^{+0.13}$	${5.944}_{-0.086}^{+0.083}$	${0.3928}_{-0.0067}^{+0.0069}$	Vogt et al. (2005)
HD 218566 b	HD 218566 b	${0.198}_{-0.018}^{+0.018}$	${0.6875}_{-0.0085}^{+0.0082}$	${0.268}_{-0.095}^{+0.10}$	Meschiari et al. (2011)
HD 219134 b	HD 219134 b	${0.308}_{-0.014}^{+0.014}$	${2.968}_{-0.037}^{+0.037}$	${0.025}_{-0.018}^{+0.027}$	Vogt et al. (2015)
HD 219134c	HD 219134c	${0.0516}_{-0.0030}^{+0.0032}$	${0.2346}_{-0.0027}^{+0.0026}$	${0.077}_{-0.050}^{+0.055}$	Vogt et al. (2015)
HD 219134 d	HD 219134 d	${0.013}_{-0.0011}^{+0.0010}$	${0.03838}_{-0.00043}^{+0.00043}$	${0.063}_{-0.045}^{+0.070}$	Vogt et al. (2015)
HD 219134 e	HD 219134 e	${0.0243}_{-0.0022}^{+0.0023}$	${0.1453}_{-0.0016}^{+0.0016}$	${0.072}_{-0.051}^{+0.078}$	Vogt et al. (2015)
HD 219134 f	HD 219134 f	${0.0112}_{-0.0014}^{+0.0014}$	${0.06466}_{-0.00073}^{+0.00073}$	${0.16}_{-0.11}^{+0.12}$	Vogt et al. (2015)
HD 22049 b	eps Eri b	${0.651}_{-0.039}^{+0.039}$	${3.5}_{-0.040}^{+0.039}$	${0.044}_{-0.031}^{+0.047}$	Hatzes et al. (2000)
HD 222582 b	HD 222582 b	${7.88}_{-0.24}^{+0.24}$	${1.335}_{-0.02}^{+0.02}$	${0.7615}_{-0.0034}^{+0.0035}$	Vogt et al. (2000)
HD 24040 b	HD 24040 b	${4.05}_{-0.15}^{+0.15}$	${4.708}_{-0.077}^{+0.076}$	${0.0117}_{-0.0082}^{+0.0110}$	Boisse et al. (2012)
HD 24040c	HD 24040c	${0.201}_{-0.027}^{+0.027}$	${1.3}_{-0.021}^{+0.021}$	${0.11}_{-0.079}^{+0.120}$	This work
HD 26161 b	HD 26161 b	${13.5}_{-3.7}^{+8.5}$	${20.4}_{-4.9}^{+7.9}$	${0.82}_{-0.050}^{+0.061}$	This work
HD 28185 b	HD 28185 b	${6.04}_{-0.2}^{+0.2}$	${1.045}_{-0.018}^{+0.016}$	${0.0629}_{-0.0049}^{+0.0042}$	Santos et al. (2001)
HD 285968 b	HD 285968 b	${0.0285}_{-0.0043}^{+0.0043}$	${0.06649}_{-0.00043}^{+0.00042}$	${0.16}_{-0.11}^{+0.14}$	Forveille et al. (2009)
HD 31253 b	HD 31253 b	${0.446}_{-0.063}^{+0.063}$	${1.296}_{-0.025}^{+0.024}$	${0.44}_{-0.17}^{+0.12}$	Meschiari et al. (2011)
HD 32963 b	HD 32963 b	${0.726}_{-0.035}^{+0.036}$	${3.416}_{-0.059}^{+0.058}$	${0.069}_{-0.039}^{+0.040}$	Rowan et al. (2016)
HD 33636 b	HIP 24205 b	${8.92}_{-0.30}^{+0.29}$	${3.238}_{-0.053}^{+0.051}$	${0.491}_{-0.0049}^{+0.0050}$	Vogt et al. (2002)
HD 34445 b	HD 34445 b	${0.658}_{-0.04}^{+0.04}$	${2.106}_{-0.040}^{+0.038}$	${0.103}_{-0.056}^{+0.059}$	Howard et al. (2010a)
HD 3651 b	HD 3651 b	${0.2202}_{-0.0077}^{+0.0077}$	${0.2954}_{-0.0044}^{+0.0043}$	${0.614}_{-0.015}^{+0.015}$	Fischer et al. (2003)
HD 3765 b	HIP 3206 b	${0.173}_{-0.013}^{+0.014}$	${2.108}_{-0.033}^{+0.032}$	${0.298}_{-0.071}^{+0.078}$	This work
HD 38529 b	HD 38529 b	${13.21}_{-0.1}^{+0.1}$	${3.736}_{-0.01}^{+0.01}$	${0.3545}_{-0.0047}^{+0.0048}$	Fischer et al. (2001)
HD 38529c	HD 38529c	${0.876}_{-0.014}^{+0.014}$	${0.1329}_{-0.00035}^{+0.00035}$	${0.26}_{-0.015}^{+0.014}$	Fischer et al. (2003)
HD 40979 b	HD 40979 b	${3.8}_{-0.16}^{+0.16}$	${0.8669}_{-0.0076}^{+0.0074}$	${0.251}_{-0.028}^{+0.028}$	Fischer et al. (2003)
HD 4203 b	HD 4203 b	${1.821}_{-0.077}^{+0.078}$	${1.177}_{-0.022}^{+0.021}$	${0.513}_{-0.014}^{+0.013}$	Vogt et al. (2002)
HD 4203c	HD 4203c	${2.68}_{-0.24}^{+0.99}$	${7.8}_{-0.78}^{+5.40}$	${0.19}_{-0.089}^{+0.290}$	Kane et al. (2014)
HD 4208 b	HD 4208 b	${0.771}_{-0.035}^{+0.036}$	${1.634}_{-0.022}^{+0.022}$	${0.059}_{-0.038}^{+0.044}$	Vogt et al. (2002)
HD 42618 b	HD 42618 b	${0.0478}_{-0.0056}^{+0.0057}$	${0.5337}_{-0.0090}^{+0.0089}$	${0.27}_{-0.12}^{+0.13}$	Fulton et al. (2016)
HD 45184 b	HD 45184 b	${0.0374}_{-0.0038}^{+0.0040}$	${0.0641}_{-0.0011}^{+0.0010}$	${0.18}_{-0.1}^{+0.1}$	Mayor et al. (2011)
HD 45184c	HD 45184c	${0.0345}_{-0.0059}^{+0.0057}$	${0.1095}_{-0.0019}^{+0.0018}$	${0.48}_{-0.24}^{+0.18}$	Udry et al. (2019)
HD 45350 b	HD 45350 b	${1.821}_{-0.070}^{+0.075}$	${1.958}_{-0.030}^{+0.029}$	${0.794}_{-0.012}^{+0.012}$	Marcy et al. (2005)
HD 46375 b	HD 46375 b	${0.2267}_{-0.0087}^{+0.0087}$	${0.03998}_{-0.00060}^{+0.00059}$	${0.063}_{-0.024}^{+0.024}$	Marcy et al. (2000)
HD 49674 b	HD 49674 b	${0.1149}_{-0.0084}^{+0.0083}$	${0.058}_{-0.00084}^{+0.00082}$	${0.06}_{-0.042}^{+0.060}$	Butler et al. (2003)
HD 50499 b	HD 50499 b	${1.346}_{-0.087}^{+0.084}$	${3.847}_{-0.040}^{+0.038}$	${0.348}_{-0.045}^{+0.046}$	Vogt et al. (2005)
HD 50499c	HD 50499c	${3.18}_{-0.46}^{+0.63}$	${10.1}_{-0.84}^{+2.00}$	${0.241}_{-0.075}^{+0.089}$	Rickman et al. (2019)
HD 50554 b	HD 50554 b	${4.35}_{-0.18}^{+0.19}$	${2.265}_{-0.036}^{+0.035}$	${0.459}_{-0.018}^{+0.019}$	Fischer et al. (2002)
HD 52265 b	HD 52265 b	${1.108}_{-0.031}^{+0.031}$	${0.506}_{-0.0056}^{+0.0056}$	${0.213}_{-0.014}^{+0.013}$	Butler et al. (2000)
HD 66428 b	HD 66428 b	${3.19}_{-0.11}^{+0.11}$	${3.455}_{-0.050}^{+0.049}$	${0.418}_{-0.014}^{+0.015}$	Butler et al. (2006b)
HD 66428c	HD 66428c	${27}_{-17}^{+22}$	${23.0}_{-7.6}^{+19.0}$	${0.32}_{-0.16}^{+0.23}$	This work
HD 68988 b	HD 68988 b	${1.915}_{-0.054}^{+0.053}$	${0.07021}_{-0.0010}^{+0.00096}$	${0.1581}_{-0.0031}^{+0.0027}$	Vogt et al. (2002)
HD 68988c	HD 68988c	${15.0}_{-1.5}^{+2.8}$	${13.2}_{-2.0}^{+5.3}$	${0.45}_{-0.081}^{+0.130}$	This work
HD 69830 b	HD 69830 b	${0.0323}_{-0.0020}^{+0.0021}$	${0.0794}_{-0.0013}^{+0.0012}$	${0.112}_{-0.060}^{+0.054}$	Lovis et al. (2006)
HD 69830c	HD 69830c	${0.031}_{-0.003}^{+0.003}$	${0.1882}_{-0.0030}^{+0.0029}$	${0.114}_{-0.079}^{+0.096}$	Lovis et al. (2006)
HD 69830 d	HD 69830 d	${0.0444}_{-0.0056}^{+0.0054}$	${0.645}_{-0.01}^{+0.01}$	${0.104}_{-0.073}^{+0.10}$	Lovis et al. (2006)
HD 72659 b	HD 72659 b	${2.85}_{-0.12}^{+0.12}$	${4.652}_{-0.068}^{+0.067}$	${0.269}_{-0.026}^{+0.018}$	Butler et al. (2003)
HD 74156 b	HD 74156 b	${7.65}_{-0.26}^{+0.27}$	${3.726}_{-0.065}^{+0.064}$	${0.3691}_{-0.0055}^{+0.0055}$	Naef et al. (2004)
HD 74156c	HD 74156c	${1.745}_{-0.061}^{+0.061}$	${0.2844}_{-0.0049}^{+0.0049}$	${0.6536}_{-0.0038}^{+0.0036}$	Naef et al. (2004)
HD 75732 b	55 Cnc b	${0.841}_{-0.026}^{+0.026}$	${0.1162}_{-0.0018}^{+0.0018}$	${0.0048}_{-0.0031}^{+0.0036}$	Butler et al. (1997)
HD 75732c	55 Cnc c	${0.171}_{-0.0067}^{+0.0066}$	${0.2432}_{-0.0039}^{+0.0037}$	${0.048}_{-0.029}^{+0.031}$	Marcy et al. (2002)
HD 75732 d	55 Cnc d	${2.86}_{-0.25}^{+0.25}$	${5.54}_{-0.1}^{+0.1}$	${0.139}_{-0.016}^{+0.015}$	Marcy et al. (2002)
HD 75732 e	55 Cnc e	${0.1475}_{-0.0094}^{+0.0093}$	${0.792}_{-0.013}^{+0.012}$	${0.142}_{-0.066}^{+0.060}$	McArthur et al. (2004)
HD 75732 f	55 Cnc f	${0.0295}_{-0.0013}^{+0.0014}$	${0.01583}_{-0.00025}^{+0.00024}$	${0.036}_{-0.025}^{+0.034}$	Fischer et al. (2008)
HD 7924 b	HD 7924 b	${0.0259}_{-0.0014}^{+0.0014}$	${0.05596}_{-0.00079}^{+0.00075}$	${0.049}_{-0.035}^{+0.044}$	Fulton et al. (2015)
HD 7924c	HD 7924c	${0.0278}_{-0.0019}^{+0.0019}$	${0.1121}_{-0.0016}^{+0.0015}$	${0.1}_{-0.062}^{+0.069}$	Fulton et al. (2015)
HD 7924 d	HD 7924 d	${0.126}_{-0.012}^{+0.013}$	${3.51}_{-0.096}^{+0.10}$	${0.278}_{-0.082}^{+0.081}$	Fulton et al. (2015)
HD 80606 b	HD 80606 b	${4.16}_{-0.13}^{+0.13}$	${0.4602}_{-0.0070}^{+0.0069}$	${0.93043}_{-0.00069}^{+0.00068}$	Naef et al. (2001)
HD 82943 b	HD 82943 b	${1.652}_{-0.076}^{+0.085}$	${0.747}_{-0.011}^{+0.011}$	${0.421}_{-0.015}^{+0.014}$	Mayor et al. (2004)
HD 82943c	HD 82943c	${1.539}_{-0.052}^{+0.052}$	${1.189}_{-0.017}^{+0.017}$	${0.142}_{-0.046}^{+0.042}$	Mayor et al. (2004)
HD 8574 b	HD 8574 b	${1.765}_{-0.073}^{+0.075}$	${0.75}_{-0.012}^{+0.012}$	${0.306}_{-0.022}^{+0.022}$	Perrier et al. (2003)
HD 87883 b	HD 87883 b	${2.292}_{-0.069}^{+0.069}$	${4.073}_{-0.057}^{+0.056}$	${0.7121}_{-0.0079}^{+0.0083}$	Fischer et al. (2009)
HD 90156 b	HD 90156 b	${0.037}_{-0.0060}^{+0.0062}$	${0.2508}_{-0.0037}^{+0.0037}$	${0.16}_{-0.11}^{+0.18}$	Mordasini et al. (2011)
HD 92788 b	HD 92788 b	${3.52}_{-0.1}^{+0.1}$	${0.949}_{-0.013}^{+0.013}$	${0.3552}_{-0.0072}^{+0.0070}$	Fischer et al. (2001)
HD 92788c	HD 92788c	${2.81}_{-0.17}^{+0.18}$	${8.26}_{-0.28}^{+0.37}$	${0.355}_{-0.052}^{+0.057}$	Rickman et al. (2019)
HD 95128 b	47 UMa b	${2.438}_{-0.085}^{+0.086}$	${2.059}_{-0.033}^{+0.031}$	${0.016}_{-0.0080}^{+0.0076}$	Butler & Marcy (1996)
HD 95128c	47 UMa c	${0.497}_{-0.030}^{+0.032}$	${3.403}_{-0.056}^{+0.055}$	${0.179}_{-0.092}^{+0.090}$	Fischer et al. (2002)
HD 95128 d	47 UMa d	${1.51}_{-0.18}^{+0.22}$	${13.8}_{-2.1}^{+4.8}$	${0.38}_{-0.15}^{+0.16}$	Gregory & Fischer (2010)
HD 95735 b	GJ 411 b	${0.0568}_{-0.0083}^{+0.0091}$	${3.1}_{-0.11}^{+0.13}$	${0.14}_{-0.095}^{+0.160}$	Díaz et al. (2019)
HD 95735c	GJ 411 c	${0.00882}_{-0.00098}^{+0.00092}$	${0.07892}_{-0.00054}^{+0.00055}$	${0.095}_{-0.066}^{+0.099}$	This work
HD 97101 b	HD 97101 b	${0.184}_{-0.011}^{+0.010}$	${1.424}_{-0.011}^{+0.011}$	${0.185}_{-0.073}^{+0.069}$	Dedrick et al. submitted
HD 97101c	HD 97101c	${0.0322}_{-0.0038}^{+0.0043}$	${0.2403}_{-0.0017}^{+0.0017}$	${0.28}_{-0.13}^{+0.13}$	Dedrick et al. submitted
HD 97658 b	HD 97658 b	${0.0247}_{-0.0017}^{+0.0018}$	${0.0805}_{-0.0011}^{+0.0010}$	${0.063}_{-0.044}^{+0.061}$	Howard et al. (2011)
HD 9826 b	ups And b	${0.675}_{-0.016}^{+0.016}$	${0.05914}_{-0.00063}^{+0.00062}$	${0.0069}_{-0.0046}^{+0.0072}$	Butler et al. (1997)
HD 9826c	ups And c	${1.965}_{-0.050}^{+0.049}$	${0.8265}_{-0.0088}^{+0.0087}$	${0.266}_{-0.012}^{+0.012}$	Butler et al. (1999)
HD 9826 d	ups And d	${4.1}_{-0.1}^{+0.1}$	${2.518}_{-0.027}^{+0.026}$	${0.294}_{-0.012}^{+0.011}$	Butler et al. (1999)
HD 99109 b	HD 99109 b	${0.474}_{-0.035}^{+0.035}$	${1.122}_{-0.017}^{+0.017}$	${0.06}_{-0.042}^{+0.064}$	Butler et al. (2006b)
HD 99492 b	HD 99492 b	${0.0841}_{-0.0062}^{+0.0061}$	${0.1231}_{-0.0015}^{+0.0014}$	${0.085}_{-0.057}^{+0.070}$	Marcy et al. (2005)
GL 317 b	GL 317 b	${1.852}_{-0.037}^{+0.037}$	${1.1799}_{-0.0076}^{+0.0076}$	${0.098}_{-0.016}^{+0.016}$	Johnson et al. (2007a)
GL 317 c	GL 317 c	${1.673}_{-0.075}^{+0.078}$	${5.78}_{-0.38}^{+0.74}$	${0.248}_{-0.074}^{+0.110}$	Anglada-Escudé et al. (2012)
GL 687 b	GL 687 b	${0.0553}_{-0.0047}^{+0.0047}$	${0.1658}_{-0.0012}^{+0.0012}$	${0.153}_{-0.080}^{+0.077}$	Burt et al. (2014)
HIP 109388 b	GJ 849 b	${0.891}_{-0.036}^{+0.035}$	${2.41}_{-0.017}^{+0.017}$	${0.05}_{-0.030}^{+0.029}$	Butler et al. (2006a)
HIP 109388c	GJ 849 c	${1.079}_{-0.053}^{+0.053}$	${4.974}_{-0.074}^{+0.082}$	${0.099}_{-0.040}^{+0.041}$	Feng et al. (2015)
HIP 22627 b	GJ 179 b	${0.752}_{-0.041}^{+0.041}$	${2.522}_{-0.033}^{+0.034}$	${0.169}_{-0.050}^{+0.048}$	Howard et al. (2010a)
HIP 57050 b	GJ 1148 b	${0.26}_{-0.029}^{+0.030}$	${0.9302}_{-0.0086}^{+0.0088}$	${0.27}_{-0.17}^{+0.14}$	Haghighipour et al. (2010)
HIP 57050c	GJ 1148c	${0.339}_{-0.014}^{+0.014}$	${0.1686}_{-0.0014}^{+0.0014}$	${0.366}_{-0.028}^{+0.027}$	Trifonov et al. (2018)
HIP 74995 b	GJ 581 b	${0.051}_{-0.0019}^{+0.0020}$	${0.04099}_{-0.00044}^{+0.00043}$	${0.02}_{-0.014}^{+0.022}$	Bonfils et al. (2005)
HIP 74995c	GJ 581 c	${0.0159}_{-0.0022}^{+0.0022}$	${0.07358}_{-0.00078}^{+0.00077}$	${0.12}_{-0.084}^{+0.120}$	Mayor et al. (2009)
HIP 83043 b	GJ 649 b	${0.275}_{-0.021}^{+0.022}$	${1.1325}_{-0.0087}^{+0.0087}$	${0.17}_{-0.1}^{+0.1}$	Johnson et al. (2010a)
HIP 57087 b	GJ 436 b	${0.0668}_{-0.0022}^{+0.0022}$	${0.02849}_{-0.0002}^{+0.0002}$	${0.145}_{-0.027}^{+0.027}$	Butler et al. (2004)
BD-103166 b	BD-103166 b	${0.449}_{-0.017}^{+0.017}$	${0.04501}_{-0.00066}^{+0.00064}$	${0.022}_{-0.015}^{+0.024}$	Butler et al. (2000)
HD 175541 b	HD 175541 b	${0.577}_{-0.047}^{+0.049}$	${0.99}_{-0.032}^{+0.030}$	${0.057}_{-0.039}^{+0.053}$	Johnson et al. (2007b)
HD 37124 b	HD 37124 b	${0.65}_{-0.016}^{+0.016}$	${0.5253}_{-0.0032}^{+0.0032}$	${0.05}_{-0.022}^{+0.019}$	Butler et al. (2003)
HD 37124c	HD 37124c	${0.647}_{-0.09}^{+0.08}$	${1.687}_{-0.011}^{+0.011}$	${0.126}_{-0.045}^{+0.047}$	Butler et al. (2003)
HD 37124 d	HD 37124 d	${0.659}_{-0.036}^{+0.034}$	${2.67}_{-0.021}^{+0.022}$	${0.16}_{-0.11}^{+0.12}$	Vogt et al. (2005)
GL 876 b	GL 876 b	${2.108}_{-0.036}^{+0.036}$	${0.2177}_{-0.0018}^{+0.0018}$	${0.0019}_{-0.0013}^{+0.0022}$	Marcy et al. (1998)
GL 876 c	GL 876 c	${0.698}_{-0.013}^{+0.013}$	${0.1363}_{-0.0011}^{+0.0011}$	${0.0055}_{-0.0040}^{+0.0064}$	Marcy et al. (2001)
GL 876 d	GL 876 d	${0.0184}_{-0.0015}^{+0.0016}$	${0.02183}_{-0.00019}^{+0.00018}$	${0.273}_{-0.098}^{+0.093}$	Rivera et al. (2005)
HD 83443 b	HD 83443 b	${0.409}_{-0.019}^{+0.019}$	${0.04067}_{-0.00062}^{+0.00060}$	${0.074}_{-0.032}^{+0.031}$	Butler et al. (2002)
HD 183263 b	HD 183263 b	${3.705}_{-0.10}^{+0.098}$	${1.508}_{-0.02}^{+0.02}$	${0.3786}_{-0.004}^{+0.004}$	Marcy et al. (2005)
HD 183263c	HD 183263c	${7.97}_{-0.22}^{+0.22}$	${6.038}_{-0.085}^{+0.082}$	${0.0595}_{-0.0063}^{+0.0064}$	Wright et al. (2009)

Download table as:  ASCIITypeset images: 1 2 3 4

We also recovered a signal with a period of 18.3 days. There is significant periodicity in the S-value time series, a periodogram of which is shown in Figure 27. However, it concentrated around a period of 6 yr, and there is no significant power near 18.3 days. Furthermore, we find no correlation between the RVs and S-values. Thus, we report this signal as a confirmed planet, with $M\sin i$ = 19.9 ± 3.1 M_⊕ and a = 0.1406 ± 0.0018 au.

Zoom In Zoom Out Reset image size

HD 136925 is a G0 dwarf, found at a distance of 47.9 pc. RVSearch detected two periodic signals in this data set, as seen in Figure 28, at 311 days and 12.4 yr. This data set is currently sparse, with two gaps of several years in the post-upgrade HIRES data, but there is clear long-period variation in the RVs. Keplerian modeling predicts $M\sin i$ = 0.84 M_J for the giant planet.

Zoom In Zoom Out Reset image size

The S-value periodogram seen in Figure 29 shows no significant power beyond 1000 days, suggesting that the long-period HD 136925 b is a real planet. There is broad power around 300 days, overlapping with the period of the inner signal. It is unclear whether this periodicity is caused by real stellar variability or is a product of sparse data. Table 1 reports current estimates of all planet parameters. We need more data in order to clarify our model, and determine whether the inner signal is caused by a planet or a product of stellar activity and sparse data. Therefore, we designated HD 136925 b as a planet, and the inner signal as a probable false positive, to be clarified with continued HIRES observing.

Zoom In Zoom Out Reset image size

HD 141004 is a G0 dwarf found at a distance of 11.8 pc. Figure 30 shows the RVSearch results for this star. A. Roy et al. (2021, in preparation) discovered a sub-Neptune at an orbital period of 15.5 days, with $M\sin i$ = 13.9 ± 1.5 M_⊕, and will report on the analysis of this system in greater detail. Table 1 reports current estimates of all planet parameters.

Zoom In Zoom Out Reset image size

HD 145675 (14 Her) is a K0 dwarf found at a distance of 17.9 pc. Figure 31 shows the RVSearch results for this star. This system has one known cold gas giant, with $M\sin i$ = 5.10 M_J and an orbital period of 4.84 yr, which was first reported in Butler et al. (2003). Wittenmyer et al. (2007a) conducted further analysis with a longer observational baseline of 12 yr, and noted a long-period trend. Wright et al. (2007) used additional RV curvature constraints to show that this trend must correspond to a companion with P > 12 yr and $M\sin i$ > 5 M_J. The observational baseline has since increased from 12 yr to 22, and regular observations with HIRES and APF allow us to place further constraints on this long-period companion. We find $M\sin i$ = ${5.8}_{-1.0}^{+1.4}$ M_J, $P={68}_{-25}^{+64}$ yr, semimajor axis $a={16.4}_{-4.3}^{+9.3}$ au, and eccentricity $e={0.45}_{-0.15}^{+0.17}$. Table 1 reports all planet parameters.

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

Figure 32 shows a Lomb–Scargle periodogram of the HIRES S-value time series. There is strong periodicity in the HIRES S-value time series, peaking around 10 yr, but no significant power near the supposed orbital period of the long-period candidate. These S-values strongly correlate with a third Keplerian signal picked up by our search, also with a period of 10 yr, as seen in the Figure 33, therefore we designate this signal as stellar activity.

Zoom In Zoom Out Reset image size

There is a potential complication owed to a stellar binary candidate. Roberts et al. (2011) conducted a direct-imaging survey of known exoplanet hosts and reported a candidate stellar companion to 14 Her, with a differential magnitude of 10.9 ± 1.0, an angular separation of 4.3'', and a minimum orbital separation of 78 au. This is a single-epoch detection, and therefore could be only a visual binary. Additionally, Rodigas et al. (2011) conducted a deep direct imaging study of 14 Her, to constrain the mass and orbital parameters of 14 Her c, which, at the time, presented only as a parabolic trend in RV data. They used the Clio-2 photometer on the MMT, which has a 9'' × 30'' field of view; the authors only looked at imaging data within 2'', to filter out background stars. Although this deep imaging study did not mention any stellar companion, the candidate reported by Roberts et al. (2011) falls outside of their considered imaging data, which corresponds to a minimum separation of 112.8 au. Wittrock et al. (2017) also found a null binary detection, using the Differential Speckle Survey Instrument (DSSI) at the Gemini-North Observatory. A 6 Jupiter mass object would not have been detected by the above surveys, as they were designed only to rule out stellar companions and therefore used shorter imaging exposures that would miss planetary-mass companions.

Additionally, we used Gaia DR2 to search for bound stellar companions within 10'', and found no such companions. We conclude that 14 Her does not have a bound stellar companion. Therefore, we designated 14 Her c as an eccentric, long-period planet. We will continue to monitor this star with Keck/HIRES and APF, to further constrain the orbit of this planet.

HD 156668 is a K3 dwarf found at a distance of 24.4 pc. Figure 34 shows the RVSearch results for this star. This system has one known short-period super-Earth, with $M\sin i$ = 4.15 M_⊕ and an orbital period of 4.64 days. This planet was first reported by Howard et al. (2011), who also noted a long-period (P ≈ 2.3 yr) signal with insufficient RV observations or additional data for confirmation as a planet. The observational baseline has since increased from five years to fourteen, allowing us to confirm this long-period signal as a planet with $M\sin i$ = 0.167 M_J and an orbital period P = 2.22 yr.

Zoom In Zoom Out Reset image size

There is a strong periodicity in the HIRES S-value time series, peaking around 10 yr, but no significant power near the orbital period of the long-period candidate. If we do not model this activity, a one-year alias signal appears in the periodogram search (Figure 34). The data do not sufficiently constrain a Keplerian fit with a 10 yr period, but we find that a linear trend models the activity well enough to remove the one-year alias from the search. We opt to include this linear trend, which we treat as a nuisance parameter.

HD 164922 is a G9 V dwarf located at a distance of 22.1 pc. Figure 35 shows the RVSearch results for this star. It hosts two known planets: a 0.3 M_J planet with an orbital period of 1207 days (Butler et al. 2006b), and a super-Earth with $M\sin i$ = 14.3 M_⊕ and an orbital period of 75.8 days. This super-Earth was reported by Fulton et al. (2016), who also reported residual power around 41.7 days but did not find it significant enough to merit candidate status. With approximately two more years of HIRES and APF data, we identified the 41.7 day signal as a strong planet candidate and confirmed the 12.5 day planet reported in Benatti et al. (2020). Both planets are of sub-Neptune mass and have eccentricity posteriors that are consistent with circular orbits. The 41.7 day planet has $M\sin i$ = 10.7 ± 1.0 M_⊕ and a semimajor axis a = 0.2294 ± 0.0031 au. The 12.5 day planet has $M\sin i$ = 4.63 ± 0.70 M_⊕ and a semimajor axis a = 0.1024 ± 0.0014 au. Table 1 reports all planet parameters.

Zoom In Zoom Out Reset image size

To validate these candidates, we searched for periodicity in both S-value activity metrics and APT photometry. We found no evidence for stellar rotation in S-values, but estimated a stellar rotation period of 62.1 days from our APT photometry. Figure 36 shows periodograms and a phase-folded curve from this APT analysis, and Figure 37 shows equivalent analysis for HIRES S-values. The 1 yr alias of 62.1 days is 75.8 days, but the 75.8 day planet detection is high-amplitude and clean, without an additional peak near 62 days in any of the RVSearch periodograms. Therefore, within the limits of our activity metrics and vetting process, we ruled out stellar rotation as a cause of the 41.7 day signal.

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

Benatti et al. (2020) used multiple HARPS-N spectral activity indicators to estimate a stellar rotation period of 41.6 days, and they note that this rotation period is to be expected from empirical activity-rotation relationships. Therefore, they determined that the strong 42 day signal present in their HARPS RVs is caused by rotation. However, we find no evidence of significant 42 day periodicity in our analysis of spectral activity indicators or APT photometry, as seen in Figures 36 and 37, and both data sets reflect significant periodicity near 60 days. Since our RV detection of this planet candidate is clean and does not conflict with our activity analysis, we chose to include this signal in our catalog as a planet candidate, to be confirmed or refuted by independent analysis.

HD 168009 is a G1 dwarf found at a distance of 23.3 pc. Figure 38 shows the RVSearch results for this star. A. Roy et al. (2021, in preparation) discovered a super-Earth candidate at an orbital period of 15.5 days, with $M\sin i$ = 10.3 ± 1.1 M_⊕, and will report on the analysis of this candidate in greater detail. Table 1 reports current estimates of all planet parameters.

Zoom In Zoom Out Reset image size

RVSearch also recovered a highly eccentric 1 yr signal, but this signal correlates with APF systematics. Therefore, we labeled it as a false positive. A. Roy et al. (2021, in preparation) will model these systematics in greater detail.

HD 213472 is a G5 dwarf located at a distance of 64.6 pc. Figure 39 shows the RVSearch results for this star. There is an approximately 11 yr gap in RV observations of this star. The first post-upgrade HIRES observation was measured in 2005, shortly after the last pre-upgrade observation, and the second post-upgrade observation was measured in 2016. The 40 m s⁻¹ difference between these two observations prompted the CPS team to begin observing HD 213472 regularly. Together with observations since 2016, and the 13 pre-upgrade HIRES measurements, the data are consistent with a long-period, eccentric, planetary companion. Our periodogram search detects such a long-period signal. Due to the sparseness of the data, traditional MCMC methods fail to return a well-sampled model posterior. We used the rejection sampling algorithm TheJoker (Price-Whelan et al. 2017) to estimate the posterior, and found it to be unimodal. This model is consistent with a very long-period gas giant, with $M\sin i$ = ${3.48}_{-0.59}^{+1.10}$ M_J orbital period $P={46}_{-13}^{+33}$ yr, semimajor axis $a={13.0}_{-2.6}^{+5.7}$ au, and eccentricity of $e={0.53}_{-0.09}^{+0.12}$. Table 1 reports all planet parameters. Figure 40 shows the orbital parameter posteriors generated by TheJoker.

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

To investigate the possibility of a stellar or substellar companion, we compared this Keplerian model to a simple linear trend by computing the ΔBIC between the two max-likelihood models. The Keplerian model is significantly preferred with ΔBIC = 23.7. Additionally, we used Gaia to search for bound companions within 10'', and found no such companions. Therefore, we inferred that HD 213472 b is either a planet or low-mass substellar companion, and not a wide-orbit stellar companion.

Figure 41 shows a sample of possible orbits for HD 213472 b, drawn from our rejection sampling posteriors and projected over the next decade. More HIRES observations will further constrain this object's mass and orbital parameters.

Zoom In Zoom Out Reset image size

HD 68988 is a G0 dwarf found at a distance of 61 pc. Figure 42 shows the RVSearch results for this star. This system has one well-constrained hot Jupiter (Vogt et al. 2002) and an outer companion candidate that was first characterized in Bryan et al. (2016) as a partially resolved Keplerian orbit. With four more years of HIRES data, we can place tighter constraints on this outer candidate's orbit. A maximum likelihood fit gives an orbital period of 49.2 yr. However, since we have only observed a partially resolved orbit so far, the orbit posterior is wide and asymmetric in period space. MCMC sampling produces $P={61}_{-20}^{+28}$ yr. The model parameters are M sin i = ${17.6}_{-2.5}^{+2.4}$ M_J, $a={16.5}_{-3.8}^{+4.8}$ au, and $e={0.53}_{-0.09}^{+0.13}$. Table 1 reports all companion parameters.

RVSearch detects a third periodic signal, with P = 1900 days, which has the same period and phase as the peak period in the S-value time series. This signal also has a low RV amplitude, ∼6 m s⁻¹. Therefore, we designated this signal as a false positive corresponding to stellar activity.