The independent discovery of planet candidates around low mass stars and astrophysical false positives from the first two TESS sectors

TL;DR

This paper presents an independent detection pipeline applied to TESS data, discovering new small planet candidates around low mass stars and validating their potential as targets for future atmospheric studies.

Contribution

The paper introduces a new detection pipeline and applies it to TESS data, identifying and statistically validating new small planet candidates around low mass stars.

Findings

Detected 8 new planet candidates smaller than 4 Earth radii.

Identified 7 probable blended eclipsing binaries and 8 known TOIs.

Validated the pipeline's effectiveness for finding sub-Neptune-sized planets.

Abstract

Continuous data releases throughout the TESS primary mission will provide unique opportunities for the exoplanet community at large to contribute to maximizing TESS's scientific return via the discovery and validation of transiting planets. This paper introduces our independent detection pipeline of periodic transit events along with the results of its inaugural application to the recently released 2 minute light curves of low mass stars from the first two TESS sectors. The stellar parameters within our sample are refined using precise parallax measurements from the GAIA DR2 which reduces the number of low mass stars in our sample relative to those listed in the TESS Input Catalog. In lieu of the follow-up observations required to confirm or refute the planetary nature of transit-like signals, a validation of transit-like events flagged by our pipeline is performed statistically. The resulting vetted catalog contains seven probable blended eclipsing binaries, eight known TOIs, plus eight new planet candidates smaller than 4 Earth radii. This work demonstrates the ability of our pipeline to detect sub-Neptune-sized planet candidates which to-date, represent some of the most attractive targets for future atmospheric characterization via transmission or thermal emission spectroscopy and for radial velocity efforts aimed at the completion of the TESS level one requirement to deliver 50 planets smaller than 4 Earth radii with measured masses.

Figures

Tables

Table 1. Table 1: Stellar photometry for the 1599 TICs in our sample of low mass stars

TIC	$T$	$G_{\text{BP}}$	$G_{\text{RP}}$	$J$	$H$	$K_{\text{S}}$
2733611	11.121	11.200 $\pm$ 0.001	13.106 $\pm$ 0.001	9.958 $\pm$ 0.020	9.312 $\pm$ 0.025	9.137 $\pm$ 0.023
2758962	11.695	11.872 $\pm$ 0.002	13.888 $\pm$ 0.002	10.466 $\pm$ 0.023	9.842 $\pm$ 0.023	9.637 $\pm$ 0.021
2761472	12.546	12.672 $\pm$ 0.003	15.983 $\pm$ 0.003	10.665 $\pm$ 0.023	10.102 $\pm$ 0.025	9.755 $\pm$ 0.020
2761836	12.998	13.167 $\pm$ 0.002	15.986 $\pm$ 0.002	11.397 $\pm$ 0.024	10.759 $\pm$ 0.025	10.476 $\pm$ 0.025
2762148	12.716	12.720 $\pm$ 0.001	15.495 $\pm$ 0.001	11.060 $\pm$ 0.023	10.561 $\pm$ 0.023	10.246 $\pm$ 0.019

Table 2. Table 2: Stellar parameters for the 1599 TICs in our sample of low mass stars

TIC	$\alpha$ (J2000)	$\delta$ (J2000)	$\varpi$	$d$	$M_{K_{\text{S}}}$	$R_s$	$T_{\text{eff}}$	$M_s$	$\log g$
	[deg]	[deg]	[mas]	[pc]		[R⊙]	[K]	M⊙	[dex]
2733611	356.1299	-15.5194	16.763 $\pm$ 0.053	59.654 $\pm$ 0.188	5.259 $\pm$ 0.024	0.5650 $\pm$ 0.0169	3880 $\pm$ 49	0.6115 $\pm$ 0.0053	4.720 $\pm$ 0.026
2758962	356.2124	-15.1250	13.530 $\pm$ 0.050	73.911 $\pm$ 0.272	5.293 $\pm$ 0.022	0.5590 $\pm$ 0.0166	3768 $\pm$ 49	0.6076 $\pm$ 0.0051	4.727 $\pm$ 0.026
2761472	356.5801	-11.9453	43.288 $\pm$ 0.103	23.101 $\pm$ 0.055	7.937 $\pm$ 0.021	0.2160 $\pm$ 0.0065	3028 $\pm$ 49	0.1799 $\pm$ 0.0021	5.024 $\pm$ 0.027
2761836	356.5473	-14.0693	15.645 $\pm$ 0.060	63.920 $\pm$ 0.246	6.447 $\pm$ 0.026	0.3804 $\pm$ 0.0116	3259 $\pm$ 49	0.4094 $\pm$ 0.0052	4.890 $\pm$ 0.027
2762148	356.5982	-15.9872	28.845 $\pm$ 0.083	34.668 $\pm$ 0.100	7.547 $\pm$ 0.020	0.2519 $\pm$ 0.0075	3262 $\pm$ 49	0.2249 $\pm$ 0.0027	4.988 $\pm$ 0.027

Table 3. Table 3: Descriptions of the free parameters controlling the performance of ORION

Parameter	Definition	Default Value	Summary of Behavior
Linear search parameters
$\Delta t$	Linear search temporal bin width	30 minutes	Decreasing $\Delta t$ will improve sensitivity to ultra-short period
			planets while increasing the ORION runtime number of signals
			for confusion in the periodic search stage.
$D$ grid	Grid of transit durations considered during the	$\{1.2,2.4,$	Decreasing the minimum duration improves sensitivity to
	linear search stage	$4.8\}$ hours	ultra-short period planets but makes ORION more
			susceptible to stochastic features with short timescales.
S/N${}_{\text{thresh}}$	Minimum linear search S/N of a transit-	5	Decreasing S/N${}_{\text{thresh}}$ will result in more light curve features
	like event		being flagged as false positives thus creating more signals for
			confusion within the periodic search stage.
Periodic search parameters
$f_{\text{P}}$	Maximum relative difference between two	0.01	Increasing $f_{\text{P}}$ makes fewer period pairs consistent with
	periods to be flagged as a multiple		being multiples thus increasing the sensitivity to resonant
			planet pairs while simultaneously increasing the number
			of single planets misidentifed as a resonant pair.
Automated light curve vetting parameters
$c_1$	Minimum transit S/N	8.4	Increasing $c_1$ prevents the detection of some small planets
			but will also significantly reduce the number false positives
			due to residual systematics.
$c_2$	Minimum number of MADs from the	2.4	Similar behavior to $c_1$.
	out-of-transit flux that the difference in
	median in and out-of-transit fluxes
	must exceed
$c_3$	Minimum fraction of in-transit points	0.7	Increasing $c_3$ may result in more accurate determinations of
	below $Z+{\sigma }_Z$		correct periods but will also cause some transits to be
			discarded if residual noise is also present during the transit.
$c_4$	Minimum fraction of in-transit points	0.1	Increasing $c_4$ increases sensitivity to asymmetric transit
	prior to $T_0$		shapes such as those from disintegrating planets.
$c_5$	Minimum number of MADs for a flare	8	Increasing $c_5$ makes flare detection smore robust but at the
	above the flux continuum		risk of missing some lower amplitude flares.
$c_6$	Minimum number of successive points	2	Increasing $c_6$ decreases sensitivity to flares of short relative
	within a flare duration		to the light curve cadence.
$c_7$	Assumed M dwarf flare duration	30 minutes	Increasing $c_7$ decreases the sensitivity to long duration flares
			while increasing the assumed fraction light curve fraction that
			is contaminated by flares.
$c_8$	Number of transit durations from $T_0$	4	Decreasing $c_8$ decreases the probability of an transit being
	which cannot be affected by flares		contaminated by flares.
$c_9$	Maximum time from the light curve	4.8 hours	Increasing $c_9$ descreases the probability of a transit being
	edges to not be rejected due to possible		affects by light curve systematics at its edges but also narrows
	contamination at the edges.		the baseline over which transits can be found.
$c_{10}$	Minimum autocorrelation of flux residuals	0.6	Increasing $c_{10}$ improves the robustness of transit detections
			but decreases the detection sensitivity in light curves with
			imperfect systematics corrections.
Eclipsing binary vetting parameters
$c_{\text{EB},1}$	Maximum $r_p/R_s$ of a transit-like event	0.5	-
$c_{\text{EB},2}$	Maximum planet radius	30 R⊕	-
$c_{\text{EB},3}$	Maximum transit duration,	$D(30{\text{ R}}_{\oplus },$	-
	$D(r_p,P,a/R_s,Z,i)$	$P,a/R_s,Z,i)$
$c_{\text{EB},4}$aaThese eclising binary (EB) parameters are intended to identify favorable EBs rather than transiting planets. Planetary signals of interest are rejected if any of the EB criteria are satisfied.	Minimum occultation S/N of an EB (Eq. 7)	5	Decreasing $c_{\text{EB},4}$ makes a larger fraction of occultations
			consistent with being due to an EB rather than a transiting
			planet.
$c_{\text{EB},5}$aaThese eclising binary (EB) parameters are intended to identify favorable EBs rather than transiting planets. Planetary signals of interest are rejected if any of the EB criteria are satisfied.	Minimum fraction of iterative occultation	0.5	Increasing $c_{\text{EB},5}$ makes transit-like events more robust as
	searches consistent with an EB		the probability of being flagged as an EB is reduced.
$c_{\text{EB},6}$aaThese eclising binary (EB) parameters are intended to identify favorable EBs rather than transiting planets. Planetary signals of interest are rejected if any of the EB criteria are satisfied.	Minimum ingress plus egress time fraction	0.9	Increasing $c_{\text{EB},6}$ makes fewer transit-like events consistent
	of $D$ for a ‘V’-shaped transit		with having a ‘V’-shaped transit.

Table 4. Table 4: Model parameter priors

Parameter	Prior
GP hyperparametersaaGP hyperparameter priors used during de-trending (i.e. with zero mean model) and during the simultaneous systematics plus transit modeling.
Covariance amplitude, $\ln a_{\text{GP}}$	$𝒰(-20,0)$
Exponential timescale, $\ln \lambda /$days	$𝒰(-3,10)$
Coherence, $\ln \Gamma$	$𝒰(-5,5)$
Periodic timescale, $\ln P_{\text{GP}}/$days	$𝒰(-3,10)$
Transit model parameters
Orbital period, $P$ [days]	$𝒰(0.9,1.11)\cdot P_{\text{opt}}$bbThe designation ‘opt’ is indicative of the optimized parameter values from the maximum likelihood model used for parameter initialization.
Time of mid-transit, $T_0$	$𝒰(-1.11,1.11)\cdot P_{\text{opt}}+T_{0,\text{opt}}$
$[$BJD-2,457,000$]$
Scaled semimajor axis, $a/R_s$	$𝒰(0.58,1.70)\cdot {\left(a/R_s\right)}_{\text{opt}}$
Planet-star radius ratio, $r_p/R_s$	$𝒰(0,1)$
Orbital inclination, $i$	$𝒰(-1,1)\cdot i\left({\left(a/R_s\right)}_{\text{opt}},b=1\right)$ccThe function $i(a/R_s,b)=a\cos i/R_s$ returns the orbital inclination given $a/R_s$ and the impact parameter $b$ which is constrained to in our transit models.
Single transit model parameters
Orbital period, $P$ [days]	$𝒥(1,100)\cdot P_{\text{inner}}$ddThe designation ‘inner’ is indicative of the inner-most orbital period permissible for a single transit event over the TESS baseline.
Time of mid-transit, $T_0$	$𝒰(-3,3)\cdot D$
$[$BJD-2,457,000$]$
Scaled semimajor axis, $a/R_s$	$𝒥(1,100)\cdot {\left(a/R_s\right)}_{\text{inner}}$
Planet-star radius ratio, $r_p/R_s$	$𝒰(0,1)$
Orbital inclination, $i$	$𝒰(-1,1)\cdot i\left({\left(a/R_s\right)}_i,b=1\right)$

Table 5. Table 5: vespa input parameters and inferred probabilities of transiting planet and astrophysical false positive models for our 24 objects of interest

IDs		vespa Input		vespa Results
OI	TOI	maxrad	secthresh	${\text{P}}_{\text{EB}}$	${\text{P}}_{\text{EB}2}$	${\text{P}}_{\text{HEB}}$	${\text{P}}_{\text{HEB}2}$	${\text{P}}_{\text{BEB}}$	${\text{P}}_{\text{BEB}2}$	FPPaaThe transiting planet false positive probability.	DispositionbbPossible dispositions of objects of interest are a planet candidate (PC), a putative planet candidate (pPC), a single transit event (ST), a putative single transit event (pST), an unclassifed astrophysical false positive (AFP), or any of the scenarios $i$. The putative dispositions have FPP $\in [0.1,0.9]$ whereas remaining candidates and astrophysical false positives have FPP and $>0.9$ respectively.
12421862.01	198.01	34.755	9.9$\times {10}^{-4}$	2.0$\times {10}^{-9}$	4.9$\times {10}^{-5}$	1.1$\times {10}^{-14}$	9.8$\times {10}^{-8}$	5.8$\times {10}^{-2}$	3.7$\times {10}^{-2}$	9.6$\times {10}^{-2}$	PC
47484268.01	226.01	34.147	4.8$\times {10}^{-3}$	1.3$\times {10}^{-7}$	3.8$\times {10}^{-3}$	1.4$\times {10}^{-9}$	1.6$\times {10}^{-4}$	3.9$\times {10}^{-1}$	2.8$\times {10}^{-1}$	6.8$\times {10}^{-1}$	pPC
49678165.01	-	75.094	7.7$\times {10}^{-3}$	9.4$\times {10}^{-27}$	1.4$\times {10}^{-14}$	2.4$\times {10}^{-26}$	3.5$\times {10}^{-14}$	9.1$\times {10}^{-3}$	7.1$\times {10}^{-3}$	1.6$\times {10}^{-2}$	ST
92444219.01	-	35.723	3.8$\times {10}^{-3}$	4.1$\times {10}^{-15}$	6.7$\times {10}^{-8}$	4.4$\times {10}^{-15}$	1.6$\times {10}^{-8}$	2.6$\times {10}^{-1}$	1.6$\times {10}^{-1}$	4.1$\times {10}^{-1}$	pST
100103200.01	-	47.832	7.0$\times {10}^{-4}$	5.1$\times {10}^{-19}$	4.3$\times {10}^{-12}$	5.0$\times {10}^{-33}$	2.9$\times {10}^{-20}$	3.8$\times {10}^{-3}$	2.0$\times {10}^{-3}$	5.7$\times {10}^{-3}$	pPCccOI 100103200.01 is assigned the pPC disposition despite having a FPP because of its proximity to the comparably bright TIC 100103201 (see Fig. 9).
100103201.01	-	47.941	8.1$\times {10}^{-4}$	2.2$\times {10}^{-73}$	1.7$\times {10}^{-67}$	5.1$\times {10}^{-132}$	3.7$\times {10}^{-91}$	6.2$\times {10}^{-1}$	3.5$\times {10}^{-1}$	9.7$\times {10}^{-1}$	BEB
100103201.02	-	47.941	8.1$\times {10}^{-4}$	2.2$\times {10}^{-73}$	1.7$\times {10}^{-67}$	5.1$\times {10}^{-132}$	3.7$\times {10}^{-91}$	6.2$\times {10}^{-1}$	3.5$\times {10}^{-1}$	9.7$\times {10}^{-1}$	BEB
141708335.01	-	33.606	3.6$\times {10}^{-3}$	6.7$\times {10}^{-39}$	2.9$\times {10}^{-31}$	3.4$\times {10}^{-35}$	8.6$\times {10}^{-26}$	6.3$\times {10}^{-1}$	3.3$\times {10}^{-1}$	9.5$\times {10}^{-1}$	BEB
206660104.01	-	38.276	8.4$\times {10}^{-4}$	1.7$\times {10}^{-65}$	8.6$\times {10}^{-39}$	2.5$\times {10}^{-87}$	1.2$\times {10}^{-40}$	1.3$\times {10}^{-2}$	9.3$\times {10}^{-3}$	2.2$\times {10}^{-2}$	PC
231279823.01	-	36.346	5.3$\times {10}^{-4}$	7.7$\times {10}^{-43}$	7.9$\times {10}^{-47}$	7.5$\times {10}^{-55}$	4.0$\times {10}^{-78}$	5.1$\times {10}^{-1}$	4.9$\times {10}^{-1}$	$1.0\times {10}^0$	BEB
231702397.01	122.01	36.485	6.9$\times {10}^{-3}$	7.5$\times {10}^{-18}$	8.7$\times {10}^{-6}$	1.8$\times {10}^{-23}$	5.1$\times {10}^{-9}$	2.3$\times {10}^{-3}$	2.7$\times {10}^{-3}$	5.0$\times {10}^{-3}$	PC
234994474.01	134.01	37.794	7.1$\times {10}^{-4}$	1.2$\times {10}^{-51}$	8.5$\times {10}^{-34}$	3.9$\times {10}^{-107}$	4.2$\times {10}^{-61}$	9.6$\times {10}^{-4}$	1.9$\times {10}^{-4}$	1.2$\times {10}^{-3}$	PC
235037759.01	-	39.567	1.9$\times {10}^{-2}$	-	-	-	-	-	-	-	AFPddThe MCMC for OI 235037759.01 failed to converge so we broadly classify it as an astrophysical false positive (AFP) based on the prevalence of nearby bright sources from GAIA (see panel ‘v’ in Fig. 9). We are unable to classify the object as a particular type of astrophysical false positive.
238027971.01	-	37.085	1.4$\times {10}^{-3}$	2.5$\times {10}^{-67}$	7.5$\times {10}^{-92}$	5.6$\times {10}^{-76}$	4.1$\times {10}^{-79}$	3.4$\times {10}^{-1}$	6.6$\times {10}^{-1}$	$1.0\times {10}^0$	BEB2
260004324.01	-	35.803	1.1$\times {10}^{-3}$	3.3$\times {10}^{-44}$	4.1$\times {10}^{-17}$	1.4$\times {10}^{-61}$	8.8$\times {10}^{-20}$	5.3$\times {10}^{-2}$	2.8$\times {10}^{-2}$	8.1$\times {10}^{-2}$	PC
262530407.01	177.01	40.509	8.0$\times {10}^{-4}$	6.5$\times {10}^{-38}$	3.2$\times {10}^{-41}$	7.6$\times {10}^{-45}$	5.5$\times {10}^{-44}$	1.8$\times {10}^{-7}$	5.7$\times {10}^{-7}$	7.5$\times {10}^{-7}$	PC
278661431.01	-	48.894	5.1$\times {10}^{-3}$	2.4$\times {10}^{-31}$	1.4$\times {10}^{-37}$	3.6$\times {10}^{-48}$	6.9$\times {10}^{-55}$	3.8$\times {10}^{-1}$	4.8$\times {10}^{-1}$	8.6$\times {10}^{-1}$	pPC
279574462.01	-	146.950	1.3$\times {10}^{-2}$	-	1.7$\times {10}^{-86}$	1.0$\times {10}^{-304}$	1.9$\times {10}^{-77}$	3.0$\times {10}^{-1}$	7.0$\times {10}^{-1}$	$1.0\times {10}^0$	BEB2
303586421.01	-	32.084	6.2$\times {10}^{-3}$	7.8$\times {10}^{-53}$	6.2$\times {10}^{-47}$	3.1$\times {10}^{-67}$	1.5$\times {10}^{-31}$	3.3$\times {10}^{-1}$	5.6$\times {10}^{-1}$	9.0$\times {10}^{-1}$	BEB2
305048087.01	237.01	34.445	8.7$\times {10}^{-3}$	2.6$\times {10}^{-30}$	1.4$\times {10}^{-8}$	4.8$\times {10}^{-27}$	8.0$\times {10}^{-8}$	1.8$\times {10}^{-2}$	1.4$\times {10}^{-2}$	3.2$\times {10}^{-2}$	PC
307210830.01	175.01	38.817	7.7$\times {10}^{-4}$	1.7$\times {10}^{-5}$	8.0$\times {10}^{-3}$	9.0$\times {10}^{-10}$	7.5$\times {10}^{-5}$	9.2$\times {10}^{-3}$	4.3$\times {10}^{-3}$	2.2$\times {10}^{-2}$	PC
415969908.01	233.01	32.000	2.3$\times {10}^{-3}$	1.5$\times {10}^{-11}$	1.1$\times {10}^{-5}$	1.4$\times {10}^{-18}$	2.0$\times {10}^{-8}$	6.0$\times {10}^{-3}$	5.5$\times {10}^{-3}$	1.2$\times {10}^{-2}$	PC
415969908.02	-	32.000	2.3$\times {10}^{-3}$	2.5$\times {10}^{-19}$	2.4$\times {10}^{-10}$	4.9$\times {10}^{-28}$	2.7$\times {10}^{-13}$	2.6$\times {10}^{-3}$	2.0$\times {10}^{-3}$	4.7$\times {10}^{-3}$	ST
441056702.01	-	35.795	1.9$\times {10}^{-3}$	5.5$\times {10}^{-19}$	5.3$\times {10}^{-9}$	1.8$\times {10}^{-37}$	6.0$\times {10}^{-17}$	2.5$\times {10}^{-2}$	1.7$\times {10}^{-2}$	4.2$\times {10}^{-2}$	PC

Table 6. Table 6: Planetary parameters for our 16 vetted candidates

OI	TOI	$P$	$T_0$	$a/R_s$	$r_p/R_s$	$i$	$r_p$	$S$	DispositionaaPossible dispositions of objects of interest are a planet candidate (PC), a putative planet candidate (pPC), a single transit event (ST), or a putative single transit event (pST). The putative dispositions have FPP $\in [0.1,0.9]$ whereas the remaining candidates have FPP .
		[days]	[BJD-2,457,000]			[deg]	[R⊕]	[S⊕]
12421862.01	198.01	20.4282 $\pm$ 0.0042	1376.8028 $\pm$ 0.0023	68.05${}_{-9.86}{}^{}{}_{}{}^{+6.21}$	0.033${}_{-0.002}{}^{}{}_{}{}^{+0.001}$	90.24${}_{-0.40}{}^{}{}_{}{}^{+0.44}$	1.59${}_{-0.10}{}^{}{}_{}{}^{+0.09}$	2.7$\pm$ 0.2	PC
47484268.01	226.01	20.2833 $\pm$ 0.0048	1378.7855 $\pm$ 0.0042	56.34${}_{-8.28}{}^{}{}_{}{}^{+9.13}$	0.090${}_{-0.006}{}^{}{}_{}{}^{+0.004}$	90.62${}_{-0.19}{}^{}{}_{}{}^{+0.32}$	3.82${}_{-0.25}{}^{}{}_{}{}^{+0.20}$	1.4$\pm$ 0.1	pPC
49678165.01	-	35.64${}_{-14.61}{}^{}{}_{}{}^{+49.78}$	1371.1945 $\pm$ 0.0038	74.13${}_{-26.55}{}^{}{}_{}{}^{+102.92}$	0.10${}_{-0.00}{}^{}{}_{}{}^{+0.01}$	90.00${}_{-0.36}{}^{}{}_{}{}^{+0.36}$	3.02${}_{-0.19}{}^{}{}_{}{}^{+0.21}$	0.3${}_{-0.2}{}^{}{}_{}{}^{+0.3}$	ST
92444219.01	-	39.52${}_{-16.90}{}^{}{}_{}{}^{+45.64}$	1342.4729 $\pm$ 0.0037	49.71${}_{-19.54}{}^{}{}_{}{}^{+55.87}$	0.07${}_{-0.00}{}^{}{}_{}{}^{+0.00}$	89.99${}_{-0.41}{}^{}{}_{}{}^{+0.40}$	3.07${}_{-0.14}{}^{}{}_{}{}^{+0.13}$	0.5${}_{-0.3}{}^{}{}_{}{}^{+0.6}$	pST
100103200.01	-	18.4462 $\pm$ 0.0024	1380.0894 $\pm$ 0.0018	51.40${}_{-5.97}{}^{}{}_{}{}^{+9.45}$	0.035${}_{-0.001}{}^{}{}_{}{}^{+0.002}$	90.45${}_{-0.23}{}^{}{}_{}{}^{+0.33}$	1.97${}_{-0.09}{}^{}{}_{}{}^{+0.11}$	4.2$\pm$ 0.3	pPC
206660104.01	-	13.4507 $\pm$ 0.0081	1379.7842 $\pm$ 0.0079	27.84${}_{-1.05}{}^{}{}_{}{}^{+1.82}$	0.021${}_{-0.001}{}^{}{}_{}{}^{+0.001}$	90.00${}_{-0.49}{}^{}{}_{}{}^{+0.44}$	1.21${}_{-0.07}{}^{}{}_{}{}^{+0.08}$	5.2$\pm$ 0.4	PC
231702397.01	122.01	5.0789 $\pm$ 0.0010	1349.4338 $\pm$ 0.0031	31.23${}_{-4.64}{}^{}{}_{}{}^{+3.05}$	0.077${}_{-0.005}{}^{}{}_{}{}^{+0.004}$	90.55${}_{-1.12}{}^{}{}_{}{}^{+0.96}$	2.80${}_{-0.19}{}^{}{}_{}{}^{+0.18}$	7.5$\pm$ 0.7	PC
234994474.01	134.01	1.4013 $\pm$ 0.0001	1345.6507 $\pm$ 0.0010	8.68${}_{-1.10}{}^{}{}_{}{}^{+0.73}$	0.021${}_{-0.001}{}^{}{}_{}{}^{+0.001}$	88.84${}_{-3.67}{}^{}{}_{}{}^{+2.67}$	1.39${}_{-0.06}{}^{}{}_{}{}^{+0.07}$	145.1$\pm$ 11.0	PC
260004324.01	-	3.8157 $\pm$ 0.0007	1357.9369 $\pm$ 0.0036	17.89${}_{-2.73}{}^{}{}_{}{}^{+1.98}$	0.021${}_{-0.002}{}^{}{}_{}{}^{+0.001}$	91.12${}_{-2.14}{}^{}{}_{}{}^{+1.23}$	1.15${}_{-0.09}{}^{}{}_{}{}^{+0.07}$	24.6$\pm$ 1.9	PC
262530407.01	177.01	2.8540 $\pm$ 0.0001	1364.7070 $\pm$ 0.0004	16.69${}_{-2.17}{}^{}{}_{}{}^{+1.86}$	0.033${}_{-0.001}{}^{}{}_{}{}^{+0.001}$	90.95${}_{-1.11}{}^{}{}_{}{}^{+1.45}$	1.87${}_{-0.08}{}^{}{}_{}{}^{+0.08}$	39.7$\pm$ 3.3	PC
278661431.01	-	17.6317 $\pm$ 0.0048	1343.9299 $\pm$ 0.0025	46.41${}_{-0.38}{}^{}{}_{}{}^{+0.26}$	0.094${}_{-0.002}{}^{}{}_{}{}^{+0.002}$	89.98${}_{-0.08}{}^{}{}_{}{}^{+0.12}$	2.82${}_{-0.10}{}^{}{}_{}{}^{+0.11}$	1.1$\pm$ 0.1	pPC
305048087.01	237.01	5.4310 $\pm$ 0.0014	1376.9753 $\pm$ 0.0037	35.04${}_{-5.12}{}^{}{}_{}{}^{+3.47}$	0.073${}_{-0.006}{}^{}{}_{}{}^{+0.004}$	90.33${}_{-1.14}{}^{}{}_{}{}^{+0.80}$	1.67${}_{-0.14}{}^{}{}_{}{}^{+0.12}$	3.6$\pm$ 0.4	PC
307210830.01	175.01	3.6893 $\pm$ 0.0001	1374.6508 $\pm$ 0.0005	22.34${}_{-2.52}{}^{}{}_{}{}^{+1.64}$	0.039${}_{-0.001}{}^{}{}_{}{}^{+0.001}$	90.35${}_{-0.58}{}^{}{}_{}{}^{+0.88}$	1.34${}_{-0.04}{}^{}{}_{}{}^{+0.05}$	12.0$\pm$ 0.9	PC
415969908.01	233.01	11.6658 $\pm$ 0.0056	1376.9247 $\pm$ 0.0040	45.23${}_{-7.04}{}^{}{}_{}{}^{+4.77}$	0.046${}_{-0.003}{}^{}{}_{}{}^{+0.003}$	90.11${}_{-0.73}{}^{}{}_{}{}^{+0.67}$	1.90${}_{-0.13}{}^{}{}_{}{}^{+0.13}$	3.8$\pm$ 0.3	PC
415969908.02	-	52.74${}_{-20.31}{}^{}{}_{}{}^{+56.33}$	1381.0703 $\pm$ 0.0042	98.07${}_{-33.09}{}^{}{}_{}{}^{+90.89}$	0.05${}_{-0.00}{}^{}{}_{}{}^{+0.00}$	90.01${}_{-0.37}{}^{}{}_{}{}^{+0.36}$	2.02${}_{-0.20}{}^{}{}_{}{}^{+0.22}$	0.5${}_{-0.3}{}^{}{}_{}{}^{+0.5}$	ST
441056702.01	-	6.3424 $\pm$ 0.0093	1371.8111 $\pm$ 0.0055	20.88${}_{-3.40}{}^{}{}_{}{}^{+3.80}$	0.031${}_{-0.003}{}^{}{}_{}{}^{+0.003}$	89.81${}_{-1.24}{}^{}{}_{}{}^{+1.67}$	1.97${}_{-0.19}{}^{}{}_{}{}^{+0.19}$	25.0$\pm$ 2.2	PC

Table 7. Table 7: Metric values indicating the feasibility of a variety of follow-up programs for our 16 vetted candidates

OI	TOI	$J$	$P$	$r_p$	$m_p$aaPlanet masses are estimated from the planet radius using the deterministic version of the mass-radius relation from Chen & Kipping (2017).	$K$	$\Omega$bb$\Omega$ is a diagnostic metric that is indicative of the observing time required to characterize a planet’s RV mass (Cloutier et al., 2018). $\Omega =r_p/P^{1/3}$ where $r_p$ is given in Earth radii and $P$ in days.	$T_{\text{eq}}$ccPlanetary equilibrium temperature is calculated assuming zero albedo and full heat redistribution via $T_{\text{eq}}=T_{\text{eff}}\sqrt{R_s/2a}$.	TSMddThe transmission spectroscopy metric from (Kempton et al., 2018). See Sect. 5.3.2 for the definition.	ESMeeThe emission spectroscopy metric from (Kempton et al., 2018). See Sect. 5.3.2 for the definition.
			[days]	[R⊕]	[M⊕]	[m/s]		[K]
12421862.01	198.01	8.65	20.43	1.59	3.16	1.19	0.58	358	53.4	0.6
47484268.01	226.01	10.85	20.28	3.82	14.03	5.93	1.40	302	69.0	0.9
49678165.01	-	11.14	35.64	3.02	9.40	4.40	0.92	212	57.7	0.1
92444219.01	-	10.49	39.52	3.07	9.68	3.35	0.90	238	51.1	0.1
100103200.01	-	7.50	18.45	1.97	4.54	1.61	0.74	397	99.6	2.0
206660104.01	-	8.36	13.45	1.21	1.92	0.76	0.51	419	5.9	0.7
231702397.01	122.01	11.53	5.08	2.80	8.26	6.35	1.63	461	69.3	3.2
234994474.01	134.01	7.94	1.40	1.39	2.52	1.98	1.24	965	14.2	9.8
260004324.01	-	8.80	3.82	1.15	1.62	0.98	0.74	619	7.6	2.1
262530407.01	177.01	8.17	2.85	1.87	4.16	2.74	1.32	698	119.8	10.1
278661431.01	-	10.89	17.63	2.82	8.36	5.14	1.08	288	86.2	0.7
305048087.01	237.01	11.74	5.43	1.67	3.44	4.06	0.95	384	66.2	1.3
307210830.01	175.01	7.93	3.69	1.34	2.37	2.15	0.87	517	26.4	6.7
415969908.01	233.01	9.94	11.67	1.90	4.26	2.21	0.84	387	55.5	1.1
415969908.02	-	9.94	52.74	2.02	4.74	1.48	0.54	234	36.4	0.1
441056702.01	-	9.89	6.34	1.97	4.53	2.17	1.06	622	39.8	2.4

Equations

$$M_{K_{\text{S}}}=K_{\text{S}}-\mu -A_{K_{\text{S}}},$$

$$\begin{gathered} \ln \mathcal{L}(\theta ,\Theta )= \\ -\frac{1}{2}[(\mathbf{f}-\mathbf{\mu }(\theta ){)}^{\text{T}}\cdot \text{K}(\Theta {)}^{-1}\cdot (\mathbf{f}-\mathbf{\mu }(\theta )) \\ +\ln \det \text{K}+N\ln 2\pi ] \end{gathered}$$

$$k_{i,j}=a_{\text{GP}}^2\exp \left[-\frac{|t_i-t_j{|}^2}{2{\lambda }^2}-{\Gamma }^2{\sin }^2\left(\frac{\pi |t_i-t_j|}{P_{\text{GP}}}\right)\right].$$

$$m(t)=\{\begin{array}{ll}1-Z& \text{if }{T}_0-\frac{D}{2}\le t\le T_0+\frac{D}{2}\\ 1& \text{otherwise,}\end{array}$$

$${\text{S/N}}_{\text{transit}}=\frac{Z}{{\sigma }_{f,\text{transit}}}\sqrt{n_{\text{transit}}(P,T_0,\mathbf{t})}$$

$$\frac{Z_{\text{occ}}}{{\sigma }_{\text{occ}}}$$

$$\frac{Z-Z_{\text{occ}}}{\sqrt{{\sigma }_{f,\text{transit}}^2+{\sigma }_{\text{occ}}^2}}$$

$$\begin{gathered} K_{\text{RV}}=2.4\text{ m/s}\left(\frac{m_p}{5{\text{ M}}_{\oplus }}\right){\left(\frac{M_s}{0.5{\text{ M}}_{\odot }}\right)}^{-2/3} \\ {\left(\frac{P}{10\text{ days}}\right)}^{-1/3}. \end{gathered}$$

$$J$$

$$\Omega$$

$$\Omega =\left(\frac{r_p}{{\text{ R}}_{\oplus }}\right){\left(\frac{P}{\text{day}}\right)}^{-1/3}$$

$$\text{TSM }=f\cdot 10^{-0.2J}\cdot \left(\frac{r_p^3{T}_{\text{eq}}}{m_p{R}_s^2}\right)$$

$$\text{ESM }=c\cdot 10^{-0.2K_{\text{S}}}\cdot \frac{B_{7.5}(T_{\text{day}})}{B_{7.5}(T_{\text{eff}})}\cdot {\left(\frac{r_p}{R_s}\right)}^2$$

Full text

1 Introduction

With our current observational capabilities, nearby transiting planets offer the best targets to characterize exoplanetary systems in detail. By their proximity many of these planets are amenable to follow-up observations to, for example, refine their radii and orbital ephemerides from the ground (Stefansson et al., 2017; Cooke et al., 2018) and from space (Broeg et al., 2013; Gaidos et al., 2017), measure planetary masses via precision radial velocities (Cloutier et al., 2018), and study their atmospheric compositions, dynamics, and thermal structures (Louie et al., 2018; Kempton et al., 2018). NASA’s Transiting Exoplanet Survey Satellite (TESS; Ricker et al., 2015), which launched on April 18 2018, is a purpose-built survey observatory and currently offers the best opportunity to discover nearby transiting planets smaller than Neptune around stars within $\sim 1000$ pc (Stassun et al., 2018). Indeed TESS has already produced a number of new confirmed planet detections (Esposito et al., 2018; Gandolfi et al., 2018; Huang et al., 2018b; Trifonov et al., 2018; Vanderspek et al., 2018) in addition to its set of TESS Objects of Interest or TESS ‘alerts’.

TESS features four refractive lens that provide a combined wide field-of-view of $24^{\circ}\times 96^{\circ}$ (i.e. $\sim 2300$ square degrees) for a single sector. The primary TESS mission splits the sky into 26 equal sectors (13 per hemisphere) anchored on the ecliptic poles and extending towards the ecliptic plane where fields at the lowest ecliptic latitudes ($\sim 63$% of the sky) will be continuously monitored for $\sim 27$ days. Conversely, fields centered at the galactic poles ($\sim 2$% of the sky) will be continuously monitored for $\sim 350$ days and overlap with continuous viewing zone of the James Webb Space Telescope (JWST). In total TESS will survey $\sim 85$% of the entire sky over its two year-long survey targeting 2-4$\times 10^{5}$ predominantly bright dwarf stars listed in the TESS Input Catalog (TIC; Stassun et al., 2018) with 2 minute cadence. Full Frame Images for all visible objects within each field will also be released with a 30 minute cadence. From these data products TESS is expected to discover thousands of new transiting exoplanets (Sullivan et al., 2015; Ballard, 2018; Barclay et al., 2018) plus potentially thousands more from a variety of proposed extended missions (Bouma et al., 2017; Huang et al., 2018a). The launch of TESS and its recent large data release marks the beginning of a new era of exoplanetary survey science that will carry on the legacy of the infamous Kepler space telescope which was decommissioned on November 16 2018, after nearly a decade of transformative exoplanet observations and thousands of planet discoveries.

TESS is also unique to past space-based transiting exoplanet survey observatories (e.g. Kepler, CoRoT) in that its bandpass extends further redward into the near-IR: 600-1000 nm. This enables TESS to access more cool M dwarf stars at high signal-to-noise than previous missions. Systems of sub-Neptune-sized planets are common around M dwarfs (Dressing & Charbonneau, 2013; Morton & Swift, 2014; Dressing & Charbonneau, 2015) and are required in order to provide a global view of outcomes of the planet formation process across the Initial Mass Function. Given their lower luminosities relative to Sun-like stars, detecting close-in planets around low mass stars probes a subset of exoplanets with systematically lower equilibrium temperatures including temperate planets orbiting within the habitable zone (Kasting et al., 1993; Kopparapu et al., 2013). Given their relative abundance within the solar neighbourhood (Winters et al., 2015), planet masses around nearby low mass stars may be readily characterized with radial velocities to build up a statistically significant view of the mass-radius relationship for small planets (Weiss & Marcy, 2014; Rogers, 2015; Wolfgang et al., 2016; Chen & Kipping, 2017). The small sizes of low mass stars also works to increase observational signatures of transiting planets thus making their planetary systems of particular interest for the atmospheric characterization of terrestrial to super-Earth-sized planets whose scale heights are expected to be inherently small ($\lesssim$ Miller-Ricci et al., 2009) and thus difficult to detect even with state-of-art instrumentation on-board the up-coming JWST (Morley et al., 2017).

The recent public data release from the first two TESS sectors, processed and validated by the TESS Science Processing Operations Center (Jenkins et al., 2016; Twicken et al., 2018; Li et al., 2018), provides an opportunity for members of the extended exoplanet community to pursue a variety of unique science cases. This includes the search for new transiting planets using transit detection algorithms that are independent of those used by the TESS Science Team and on distinct subsets of stars targeted by TESS. In this study, focus on low mass dwarf stars from the TIC and use GAIA parallaxes to infer precise stellar parameters and the refine the sample probable M dwarf TICs. We then search for transiting exoplanets around these low mass dwarfs in the high cadence TESS light curves using our custom-built transit detection pipeline described herein.

In Sect. 2 we present the derivation of our input target list of low mass TICs. In Sect. 3 we present the details of our transit detection pipeline ORION. In Sect. 4 we present our pipeline results and our supplementary efforts to classify flagged transit-like events via human vetting and statistical validation before culminating our final list of planet candidates and astrophysical false positives. We conclude with a discussion in Sect. 5.

2 Stellar Sample

2.1 Initial stellar sample

Our initial stellar sample is retrieved from version 7 of the TESS Input Catalog (TIC-7) which is accessed via the Barbara A. Mikulski Archive for Space Telescope (MAST) Portal111https://mast.stsci.edu/portal/Mashup/Clients/Mast/Portal.html. Among other parameters, the TIC-7 table contains estimates of each star’s physical parameters (i.e. effective temperatures $T_{\text{eff}}$, surface gravities $\log{g}$, radii $R_{s}$, masses $M_{s}$, etc.), astrometry (either from the Tycho-GAIA astrometric solution; Gaia Collaboration et al. 2016a, b or from Hipparcos), $G$-band magnitude from the GAIA data release 1, and 2MASS photometry (Cutri et al., 2003). To identify putative low mass dwarf stars within the TIC-7, we first restrict our sample to sources flagged as dwarf stars based on their 2MASS colors and the reduced proper motion criterion from Stassun et al. (2018), modified from Collier Cameron et al. (2007). We further restrict our sample to stars whose ‘priority’ is $\geq 10^{-3}$ where the TIC priority metric is based on the relative probability of the TIC of detecting small planetary transits. As such, the priority is dependent on $R_{s}$, the expected photometric precision, the number of TESS sectors in which the TIC will be visible, and its contamination ratio: the ratio of contamination to source flux where contamination is computed over ten TESS pixels from the source ($\sim 3.5$ arcmin).

Next we establish our initial sample of low mass stars based on the physical stellar parameters from the TIC-7 and using the following criteria:

• •

$T_{\text{eff}}$ $\in[2700,4200]$ K,

• •

$\log{g}$ $>4$,

• •

$R_{s}<0.75$ R*⊙*,

• •

$M_{s}<0.75$ M*⊙*.

We note that these criteria are not intended to reflect the exact M dwarf parameter ranges of interest but instead are chosen to be intentionally conservative as to avoid missing any potential M dwarfs prior to their final classification (for use within this study) based on $T_{\text{eff}}$ and near-IR luminosities ($M_{K_{\text{S}}}$ $\in[4.5,10]$; Delfosse et al. 2000; Benedict et al. 2016) that will be refined in Sect. 2.2 using GAIA parallaxes. At this stage we find a total of 93090 TICs that obey our criteria. Of these, 2849 TIC are observed in one or both of TESS sectors 1 and 2 and are depicted in Fig. 1.

2.2 Refined stellar sample based on GAIA DR2

The stellar parameters used to derive our initial stellar sample were obtained from a variety of sources as outlined in Stassun et al. (2018). Effective temperatures within our sample were predominantly obtained from the cool dwarf catalog (Muirhead et al., 2018) or alternatively from spectroscopic catalogs or $V$-$K_{\text{S}}$ colors. Most stellar masses and radii also come from the cool dwarf catalog or from the Torres et al. (2010) spectroscopic relations. Stellar surface gravities follow from measurements of $R_{s}$ and $M_{s}$. The TIC-7 stellar radii are typically known at the level of $\sim 16$% which often dominates the error budget of the measured planetary radii from transit observations. Here we aim to produce a homogeneously-derived set of precise stellar parameters by exploiting the exquisite precision of their 2MASS photometry and stellar parallaxes from the GAIA DR2 Gaia Collaboration et al. (2018) available for the majority of stars in our initial sample.

For the accurate and precise characterization of transiting planets we are principally interested in the measurement of host stellar radii $R_{s}$. Additionally, the derivation of other fundamental parameters such as stellar masses and effective temperatures are of importance for a more complete understanding of the effect that stars can have on their host planetary systems. Here, we re-derive the stellar parameters of our initial sample by deriving their near-IR absolute magnitudes coupled with empirically-derived M dwarf radius-luminosity relations (Mann et al., 2015). We begin by querying the GAIA DR2 archive using the star’s right ascension and declination ($\alpha,\delta$) with a search radius of 10-60 arcseconds. Cross-matching the TIC-7 with the GAIA DR2 data is necessary to obtain each star’s updated parallax from the DR2, additional GAIA photometry (i.e. $G_{\text{BP}}$ and $G_{\text{RP}}$) which was not included in the TIC-7, and point-estimates of their respective measurement uncertainties that we will approximate as Gaussian distributed. Our querying procedure utilizes the astropy.astroquery python package (Ginsburg et al., 2017). Next we identify source matches according to their predicted photometric colors based on the 2MASS-GAIA color-color relations reported in Evans et al. (2018). Explicitly, we use the quadratic polynomial fits from Evans et al. (2018) to predict each of the colors $G-K_{\text{S}}$, $G_{\text{BP}}-K_{\text{S}}$, $G_{\text{RP}}-K_{\text{S}}$ and $G_{\text{BP}}-G_{\text{RP}}$ and then compare those predictions for potential source matches to the measured TIC colors. Each of the four color-color relations is accompanied by a characteristic scatter of $0.3692,0.4839,0.2744,$ and $0.2144$ magnitudes respectively. We claim a source match when all of the calculated colors are within $3\sigma$ of their predicted values. Based on numerous checks of individual known TICs, we determined that such a tolerance is required to ensure accurate source matches. This dispersion is also expected given that higher order effects not taken into account by the polynomial fits, can have stark effects on the accuracy of the photometric predictions.

We proceed with identifying bona-fide low mass TICs within our initial stellar sample by using the GAIA data of matched sources to refine the stellar parameters that were initially used to flag low mass stars. We will classify low mass stars within this study based on their absolute $K_{\text{S}}$-band magnitudes (Delfosse et al., 2000; Mann et al., 2015; Benedict et al., 2016) and effective temperatures which are derived in the coming sub-sections. We focus our analysis on the $K_{\text{S}}$-band in this study due to the reduced effects of dust extinction at near-IR wavelengths compared to in the visible. The absolute $K_{\text{S}}$-band magnitude is

[TABLE]

where $K_{\text{S}}$ is the source’s $K_{\text{S}}$-band apparent magnitude, $\mu=5\log_{10}{(d/\text{1 pc})}-5$ is the distance modulus given the distance to the source $d$, and $A_{K_{\text{S}}}$ is the source extinction in the $K_{\text{S}}$-band. Therefore, in order to compute $M_{K_{\text{S}}}$ for our stellar sample we must first obtain the parameters $d$ and $A_{K_{S}}$.

2.2.1 Stellar distances from GAIA

The GAIA DR2 reports precise stellar parallaxes $\varpi$ for the majority of stars in our initial sample The typical parallax uncertainty for the stars in our sample is $\sim 0.2$%. As noted by numerous authors (e.g. Bailer-Jones, 2015; Astraatmadja & Bailer-Jones, 2016; Luri et al., 2018), reliable distances to the majority of stars in the GAIA DR2 cannot be obtained by simply inverting the stellar parallax. Given $\varpi$ values with posterior probability density functions (PDFs) that are presumed Gaussian distributed, and are therefore fully described by their mean values and $1\sigma$ dispersions, the non-linearity of the transformation from $\varpi$ to $d$ will result in an asymmetric $d$ posterior PDF whose skewness is dependent on the absolute $\varpi$ measurement value and its signal-to-noise (Luri et al., 2018). By the proximity of the majority of sources in the TIC-7, their parallaxes are measured with high precision such that the resulting $d$ PDF obtained using the standard formula ($d/$ pc) = ($\varpi$/arcsec, can be well-approximated as a Gaussian distribution (Bailer-Jones et al., 2018). The median relative distance uncertainty therefore corresponds to that of the measured parallax (i.e. $\sim 0.2$%). The maximum a-posteriori (MAP) $d$ value and its $1\sigma$ uncertainty are then propagated to the calculation of $\mu$ which we will ultimately use in Eq. 1 to calculate $M_{K_{\text{S}}}$ after the extinction coefficient are obtained (see Sect. 2.2.2).

There are known systematic effects in the GAIA astrometric solution in the form of a non-zero parallax zero-point that is dependent on the source position, $G$-band magnitude, and possibly color (Lindegren et al., 2018). In computing $d$ from $\varpi$ we first apply a simple correction by adding the globally-averaged parallax zero-point of 29 $\mu$as (Lindegren et al., 2018) to the verbatim stellar parallaxes from the GAIA DR2.

2.2.2 Source extinction estimates

The source extinction is dependent on the source’s location on the sky and particularly on its proximity to the galactic plane where the dust column density is highest. To estimate the $K_{\text{S}}$-band extinction for each source we utilize the mwdust package (Bovy et al., 2016) which queries one of three E(B-V) reddening maps (i.e. Drimmel et al., 2003; Marshall et al., 2006; Green et al., 2015) based on the applicability of each map to the input source position. Given the source’s galactic coordinates ($l,b$), GAIA distance, and uncertainties as input, mwdust queries the reddening maps and returns the extinction coefficient $A_{\lambda}=R_{\lambda}\text{E(B-V)}$ in the desired band using the extinction vector scaling $R_{K_{\text{S}}}=0.31$ from Schlafly & Finkbeiner (2011). Uncertainties in $A_{K_{\text{S}}}$ are derived from the $d$ measurement uncertainty and from inherent uncertainties in the value of $R_{K_{\text{S}}}$ (Green et al., 2018, e.g.) which we attempt to account for via the quadrature addition of a 30% fractional uncertainty on $R_{K_{\text{S}}}$ following the methodology of (Fulton & Petigura, 2018).

2.2.3 Deriving the set of refined stellar radii

Combining the retrieved values of $K_{\text{S}}$, $\mu$, and $A_{K_{\text{S}}}$ into Eq. 1 returns the distribution of $M_{K_{\text{S}}}$ for all of the 2489 stars in our initial sample for which 2MASS photometry and GAIA parallaxes are available.

Calculations of M dwarf stellar radii from their bolometric magnitudes would require $K_{\text{S}}$-band bolometric corrections which for cool stars, are known to often suffer from comparatively large inaccuracies ($T_{\text{eff}}$ $\lesssim 4100$ K; Berger et al., 2018). We therefore adopt the alternative approach from Berger et al. (2018) which used the empirically-derived M dwarf radius-luminosity relation (RLR) from Mann et al. (2015) to update M dwarf stellar radii in the Kepler field using GAIA distances. The fitted RLR uses a quadratic function to map $M_{K_{\text{S}}}$ to directly measured $R_{s}$ from interferometry and parallaxes. Because we are interested in deriving a self-consistent sample of low mass stellar radii, we restrict our analysis to TICs with $M_{K_{\text{S}}}$ values that are applicable to the Mann et al. (2015) RLR which is valid for M dwarfs with $M_{K_{\text{S}}}$ $\in(4.6,9.8)$. This condition will be used to establish our final sample of low mass dwarf stars following the derivation of $T_{\text{eff}}$ within our initial sample. The radii inferred from the RLR have a fractional residual dispersion of 0.0289 $R_{s}$ which we add in quadrature to the radius uncertainty propagated from $M_{K_{\text{S}}}$.

Fig. 2 compares the TIC-7 stellar radii (compiled from various input sources) with those derived from GAIA distances and and the M dwarf RLR from Mann et al. (2015). The relation is largely one-to-one but with a slight translation of the updated $R_{s}$ distribution to larger radii ($\sim 3.7$% median increase). The effect is already known (Berger et al., 2018) and is the result of many sources having their distance measures increased following the release of the GAIA DR2 parallaxes. More notably for the measurement of transit planet radii is the significant reduction in the fractional radius uncertainty as evidence in the histogram included in Fig. 2. The typical fractional radius uncertainty $\sigma_{R_{s}}/R_{s}$ within our updated sample is $\sim 4-5$ smaller than in the TIC-7. The median fractional uncertainty on our GAIA-derived stellar radii is $\sim 3$%.

2.2.4 Deriving the set of refined stellar effective temperatures

Similarly to the RLR, (Mann et al., 2015) also parameterized an empirically-derived M dwarf temperature-color-metallicity relation (TCMR). Our sample contains both 2MASS and GAIA photometry so we use these photometric systems to infer $T_{\text{eff}}$ for the stars in our newly refined sample. Specifically, we adopt the fitted TCMR from Mann et al. (2015) which is cubic in $G_{\text{BP}}$-$G_{\text{RP}}$ and quadratic in $J$-$H$. The latter color is used as a proxy for metallicity (Leggett, 1992; Johnson et al., 2012; Mann et al., 2013; Newton et al., 2014). The TCMR used here has a residual temperature dispersion of 49 K which we add in quadrature to the uncertainties in $T_{\text{eff}}$ propagated from the input photometric uncertainties.

2.2.5 Deriving the set of refined stellar masses

We revise the stellar masses using the empirically-derived M dwarf mass-luminosity relation (MLR) from Benedict et al. (2016). Their fitted relation uses a quartic function to map $M_{K_{\text{S}}}$ to directly measured $M_{s}$ from dynamical analyses of binary star systems and is valid for M dwarfs with $M_{K_{\text{S}}}$ $\leq 10$. This valid range of $M_{K_{\text{S}}}$ is consistent with the range required for the Mann et al. (2015) empirical relations. The five fitted coefficients that parameterize the MLR all have an associated uncertainty which we sample from, along with $M_{K_{\text{S}}}$ sampled from a Gaussian distribution, to infer the $M_{s}$ PDF. Point estimates of each TIC’s $M_{s}$ and uncertainty comes from the MAP of the PDF and the average of its $16^{\text{th}}$ and $84^{\text{th}}$ percentiles.

2.2.6 Final stellar sample

Using the refined stellar parameters obtained from cross-matching putative low mass stars from the TIC-7 with the GAIA DR2, we now construct our final stellar sample as stars that obey the following criteria:

• •

$M_{K_{\text{S}}}$ $\in(4.6,9.8)$

• •

$T_{\text{eff}}$ $-\sigma_{T_{\text{eff}}}<4000$ K,

• •

$\log{g}$ $+\sigma_{\log{g}}>4$,

• •

$R_{s}-\sigma_{R_{s}}<0.75$ R*⊙*,

• •

$M_{s}-\sigma_{M_{s}}<0.75$ M*⊙*.

That is that we retain all TICs whose luminosities, effective temperatures, surface gravities, radii, and masses that are consistent with those of late-K to M dwarf parameter ranges (Pecaut & Mamajek, 2013) at the $1\sigma$ level.

Our final stellar sample contains 1599 low mass stars with 537, 694, and 368 observed within TESS sectors 1, 2, and both respectively. The final stellar sample is over-plotted in Fig. 1. The distribution of $T$ in our stellar samples spans 7-15 with a median $T=12.1$. The distribution of effective temperatures extends from 2740-4040 K with a median $T_{\text{eff}}$ $=3450$ K whose approximate spectral type is M3V (Pecaut & Mamajek, 2013). The $T_{\text{eff}}$ distributions from the TIC-7 and our GAIA-derived values are roughly consistent. The stellar radii span 0.15-0.61 R*⊙* with a median $R_{s}=0.42$ R*⊙. The stellar masses span 0.12-0.63 M⊙* with a median $M_{s}=0.47$ M*⊙*. Owing to the increased distances of many TICs in our final sample, the distributions of $R_{s}$ and $M_{s}$ are both translated to slightly larger radii and masses compared to the values listed in the TIC-7.

3 Overview of the ORION transit detection pipeline

Here we present our independent transit detection pipeline ORION that burrows many of the strategies and vetting procedures from established methods focused on transit detections primarily with Kepler and K2 (see references herein). Our pipeline can be thought of as having six sequential steps that take as input the TIC identifier and stellar parameters to produce a set of transiting planet candidates (PCs) with measured orbital periods $P$, times of mid-transit $T_{0}$, scaled semi-major axes $a/R_{s}$, scaled planetary radii $r_{p}/R_{s}$, and orbital inclinations $i$. The six steps within ORION are 1) to obtain the extracted TESS light curves and ancillary data for of input TIC, 2) to derive an initial systematic model for light curve de-trending, 3) to perform a linear search for transit-like events, 4) to perform a periodic search for repeating transit-like events, 5) to subject putative PCs to a set of vetting criteria in an automated way, and 6) to re-model the light curve with a joint systematics plus transit mode for all vetted transit-like events. These stages are described in detail in the succeeding sections.

3.1 TESS light curve acquisition

The execution of ORION on a TIC begins with downloading the star’s publicly available 2 minute TESS extracted light curves and target pixel files for all available sectors. The TESS data is downloaded from the MAST data service222https://archive.stsci.edu/tess/index.html. Only TICs observed at 2 minute cadence are considered at this time with their extracted light curve made available following its processing by the TESS Science Processing Operations Center. Efforts to extract 30 minute light curve data from the TESS Full Frame Images and significantly expand the list of TESS targets accessible to ORION are underway but are reserved for a future study. Target pixels files are principally used to quickly assess the data quality and will be used to infer the TIC’s point spread function during the statistical validation of putative PCs in Sect. 4.2.

For each available sector of data, the chronological vectors of valid observing times $\mathbf{t}$ (i.e. not NaNs), measured fluxes $\mathbf{f}$, and the associated $1\sigma$ flux uncertainties $\boldsymbol{\sigma}_{f}$ are constructed. Fluxes are obtained from the Simple Aperture Photometry Pre-search Data Conditioning extraction which includes artifact mitigation (Smith et al., 2012). These vectors are attributed to the following fields: TIME [BJD], PDCSAP_FLUX [e-/s], and PDCSAP_FLUX_ERR [e-/s]. The flux and flux uncertainty vectors are converted into normalized flux units via division by median($\mathbf{f}$).

3.2 Initial light curve de-trending

Residual systematic effects are clearly visible in the many of the extracted light curves. Due to the inherent photometric and pointing precision of the first TESS sectors, these systematic effects are often largely attributable to astrophysical noise sources such as flicker (Bastien et al., 2013) in Sun-like stars but more commonly for low mass stars from large-scale variability caused by active regions on the rotating stellar surface. As an initial de-trending step to correct for temporally-correlated noise sources from either systematics or intrinsic stellar phenomena, a semi-parametric Gaussian process (GP) regression model is fit to the extracted TESS photometry.

GP regression models provide a flexible and probabilistic framework to model the temporal covariances between photometric measurements following the removal of a mean model $\boldsymbol{\mu}(\theta)$ which is parameterized by the set of observable parameters $\theta$ (e.g. orbital period, time of mid-transit, etc.). The posterior PDFs of the $\theta$ elements can be sampled simultaneously with the GP hyperparameters $\Theta$ which parameterize the residual covariances through the covariance matrix K($\Theta$) and are fit by optimizing the $\ln$ likelihood function

[TABLE]

along with appropriately chosen priors on the parameters in $\theta$ and $\Theta$. Here, $N$ is the number of photometric measurements in the light curve. Similar routines based on GP regression have been adopted for K2 systematic corrections (e.g. Aigrain et al., 2015; Crossfield et al., 2015; Aigrain et al., 2016) and can also be used to infer accurate photometric stellar rotation periods (Angus et al., 2018). For cases in which the origin of the apparent photometric variations are likely dominated by active regions on rotating spotted stars, the resulting photometry will vary non-sinusoidally as the active regions evolve in size, brightness, and location over the observational baseline. This physically motivates the use of a quasi-periodic covariance matrix K${}_{i,j}=\delta_{i,j}\sigma_{f,i}+k_{i,j}$ where $\delta_{i,j}$ is the Kronecker delta and $k_{i,j}$ is the covariance kernel of the form

[TABLE]

The covariance kernel is parameterized by the four hyperparameters $\Theta=\{a_{\text{GP}},\lambda,\Gamma,P_{\text{GP}}\}$ where $a$ is the correlation amplitude, $\lambda$ the exponential timescale, $\Gamma$ the coherence scale, and $P_{\text{GP}}$ the periodic timescale of the correlations. Moreover, a quasi-periodic covariance kernel is favorable for cases in which the origin of apparent photometric variations are dominated by systematics which need not have a strong periodic component. In this limit, the coherence parameter $\Gamma$ approaches small values such that the covariance kernel becomes well-approximated by a squared exponential kernel with a single effective timescale. Note that because systematic and astrophysical noise sources within the GP noise model are not distinguished, the fitted hyperparameter values are unable to be used to interpret the origin of the photometric variability. Furthermore, during the remainder of this paper the covariance structures modelled by the GP will be solely referred to as ‘systematics’ despite the possibility that their (partial) origin may be astrophysical.

The logarithmic hyperparameters are initialized and subsequently optimized in an iterative manner and are performed on each TESS sector independently assuming a null mean function (i.e. $\boldsymbol{\mu}=\mathbf{0}$). The periodic GP timescale $P_{\text{GP}}$ is initialized by peaks in the Lomb-Scargle periodogram (LSP; Scargle, 1982) of the extracted light curve whereby in each of the iterations performed, $P_{\text{GP}}$ is initialized to the $i^{\text{th}}$ most significant peak in the LSP where $i$ is the iteration index $\in[1,10]$. Periods within 5% of a 1 day are excluded due to the inherent LSP noise at sampled frequencies close to 1 days*-1*. Following the use of GP regression modelling for radial velocity (RV) activity mitigation in Dittmann et al. (2017b), where the physical source of activity is largely common between the optical RVs and broadband TESS photometry, $\ln{\lambda}$ is initialized to $\ln{3P_{\text{GP}}}$. In each iteration $\ln{\Gamma}$ is initialized to [math] and $\ln{a_{\text{GP}}}=\ln{\text{max}(|\mathbf{f}_{\text{bin}}-\text{median}(\mathbf{f}_{\text{bin}})|)}$ where $\mathbf{f}_{\text{bin}}$ is the vector of binned photometric points whose temporal bin width is set such that a single periodic GP timescale is sampled by at least eight measurements.

For each iteration in each TESS sector, the uniquely initialized GP hyperparameters are optimized using the scipy.optimize.minimize python function to minimize the negative $\ln{\mathcal{L}(\Theta)}$ from Eq. 2 given the Jacobian of $\ln{\mathcal{L}(\Theta)}$ with respect to the hyperparameters in $\Theta$. During optimization the $\ln$ GP hyperparameters are bounded by broad uninformative priors which are explicitly reported in Table 4. Broad $\ln$ uniform priors enable the generalization of the ORION de-trending method across all of the input TESS light curves which greatly benefit from semi-parametric modelling given the wide range of covariance timescales exhibited by TICs in photometry. Given an optimized set of hyperparameters, the resulting GP posterior PDF is a $N$-dimensional multi-variate Gaussian distribution whose mean function is taken to be a potential systematic correction. The mean function of the GP from the iteration whose optimized hyperparameters maximize $\ln{\mathcal{L}(\Theta)}$ is assigned as the initial systematic correction and is used to de-trend the photometry prior to the search for periodic transit events.

Fig. 3 depicts two examples of the results of this iterative de-trending procedure over individual TESS sectors for TICs 235037759 and 262530407. The accuracy of each mean GP regression model, with maximum $\ln$ likelihood hyperparameters, is clearly exhibited. The systematics model for TIC 235037759 is required to be much more aggressive than that for TIC 262530407 given the star’s large photometric variability with a peak-to-peak amplitude in the binned light curve of 280,000 ppm and a 85,000 ppm rms. Unlike in the raw light curve, the de-trended light curve lacks any low frequency variations and exhibits a significantly reduced rms of 12,000 ppm.

The TIC 262530407 light curve also exhibits photometric variability albeit with a much lower peak-to-peak amplitude and rms of 1700 and 1000 ppm respectively. After de-trending, the rms is slightly reduced to 810 ppm. The most important residual feature of the de-trended light curves is that they appear free of the majority of large-scale systematic effects. This fact will facilitate the linear search for transit-like events with minimal contamination from residual systematic features. Indeed a transiting planet candidate is detected around each of these systems although the putative PC around TIC 235037759 is ultimately favored by an astrophysical false positive interpretation as presented in Sect. 4.3.

One notable limitation to the effectively corrected light curve systematics is the prevalent increase in rms for TIC 235037759 between BJD - 2,4570,00 $\sim 1347-1349$ during a brief period of loss of TESS pointing precision. This is a common feature to many of the TICs observed during TESS sector 1. During this time, the extracted light curve is only partially corrected for the pointing precision loss while the systematic GP model provides only marginal improvements if any at all. To ensure that the GP hyperparameters were not being strongly affected by the anomalous systematics structure during this period, those measurements were masked and the GP hyperparameters re-optimized with the remaining data. The resulting GP systematics only varies marginally from that shown in Fig. 3 for TIC 235037759 such that we are confident that the ORION de-trending is largely robust to the loss in pointing precision for TICs observed in TESS sector 1.

Recall that in the initial de-trending step discussed in this section that the methodology assumes a null mean function which implies that any transient events such as flares or transits are still present during the optimization of the de-trending model. The principal caveat with this methodology is that one cannot guarantee that the GP model does not (partially) capture any of the in-transit light curve deprecations that ORION is searching for. If partially suppressed by the initial GP model, planets will be more difficult to detect due to the reduced signal-to-noise (S/N) of individual transit events. Furthermore, transit events that remain detectable within the de-trended light curve could result in underestimated transit depths and correspondingly smaller planet sizes. To ensure a self-consistent planet+systematics model for putative PC from ORION, the light model is revisited in Sect. 3.6 with the inclusion of a transiting planet mean function in place of the null mean function used during the de-trending step.

3.3 Linear transit search

Next a linear search for individual transit-like events is conducted on the de-trended light curves over their full duration. The following methodology is reminiscent of a number of individual transit event search algorithms (e.g. Box Least Squares, BLS; Kovács et al. 2002, Transiting Planet Search, TPS; Jenkins et al. 2010; Christiansen et al. 2013, 2015, 2016, TERRA; Petigura et al. 2013, Foreman-Mackey et al. 2015). The aim here is to identify high S/N transit-like events along with their associated mid-transit times $T_{0}$, durations $D$, and depths $Z$ which will feed into the search for repeating transit-like events and ultimately the list of putative transiting PCs.

The linear search for transit-like events begins with stepping through a two-dimensional grid of $T_{0}$ and $D$. At each $(T_{0},D)$ grid point a simple box model of the form

[TABLE]

is constructed with fixed $T_{0}$ and $D$. The box depth $Z$ (or mock transit depth) is fit by $\ln$ likelihood maximization and saved along with the value of $\ln{\mathcal{L}}$ given the unique set of parameters $\{T_{0},D,Z\}$. Computing $\ln{\mathcal{L}}$ with Eq. 2 implicitly assumes that the flux uncertainties are Gaussian distributed which allows for the construction of a diagonal covariance matrix K with elements K${}_{i,j}=\delta_{i,j}\sigma_{f,i}$. The linear search along the $T_{0}$ dimension proceeds by stepping through the observation epochs $\mathbf{t}$ in 30 minute bins and assigning $t_{\text{bin},i}$ to $T_{0}$ $\forall$ $i=1,\dots,N_{\text{bin}}$. This fixed binning is the first of many ORION free parameters which are listed in Table 3 along with their default values and brief explanations of their effects. Initial ORION tests on synthetic light curves with injected transit models determined that finer temporal binning did not result in a significant variation in the number of detected high S/N transit-like events. This is likely due to 30 minute bins being more comparable to typical transit durations of the types of planets that can be detected in 27-54 day baselines. The $D$ dimension is sampled on a much coarser grid given that the precision of the box model parameters are not yet required to infer planet properties but only to identify epochs at which transit-like events are likely to have occurred. Explicitly, the adopted linear search $D$ grid contains three possible transit durations of either 1.2, 2.4, or 4.8 hours.

This procedure produces a $N_{\text{bin}}\times 3$ matrix of transit times and durations with each entry having an associated $\ln$ likelihood and optimized transit depth $Z$. From the $\ln$ likelihoods a S/N spectrum as a function of transit times is computed for each $D$ value considered. By translating the $\ln$ likelihoods by their median value and normalizing by their median absolute deviation (MAD), the aforementioned linear search S/N spectrum versus transit times is calculated. The linear search S/N spectrum is analogous to the Signal Detection Efficiency in the BLS algorithm. The conversion from $\ln$ likelihoods to the ad hoc S/N spectrum centered around zero such aids in its interpretation as each TIC’s spectrum can be searched in absolute terms. In adopting the median and MAD $\ln$ likelihood values over the mean and standard deviation, the S/N is less sensitive to contamination by stochastic, short timescale photometric features and results in a S/N spectrum whose baseline is dominated by the light curve’s inherent photometric precision. An example linear search S/N spectrum is shown in Fig. 4 for a fixed duration of 1.2 hours. We note that referring to the linear search S/N spectra as a S/N is a misnomer given that its values over $T_{0}$ can be negative. However, we find this language to be a clear descriptor of the quantity’s aim and its interpretability.

The S/N spectra are then searched for high S/N transit-like events analogously to the individual significance events in the primary Kepler mission which are combined and flagged as Threshold Crossing Events (Jenkins et al., 2010). High S/N transit-like events are flagged as peaks in the linear search S/N spectrum when S/N${}_{\text{thresh}}\geq 5$. All $n_{\text{T}}$ transit times $\mathbf{T}_{0}$ with S/N exceeding S/N${}_{\text{thresh}}$ for any value of $D$, are compiled into a set of potential transit-like events. Because transit times in the linear search are sampled on a fixed grid (i.e. $\mathbf{t}$ binned to $\Delta t$), each transit time is refined by Gaussian smoothing the light curve around $\pm 2D$ of $T_{0}$ using scipy.ndimage.filters.gaussian_filter and updating $T_{0}$ to the central time of the box model minimum before proceeding to search for periodic events that may be indicative of transiting PCs. In the example shown in Fig. 4, a PC exists with $\sim 1.4$ day orbital period. Six out of the nineteen transit events that occur within the sector 1 baseline are detected above S/Ntextthres. This includes two consecutive transits between 1341 and 1343 BJD - 2,457,000 which are used in the subsequent section to infer its possible period equal to the time difference between the two events.

3.4 Periodic transit search

The chronologically sorted set of $n_{\text{T}}$ high S/N transit times in $\mathbf{T}_{0}$ are used to construct a matrix of differential transit times with elements P${}_{i,j}=T_{0,i}-T_{0,j}$ $\forall$ $i,j=1,\dots,n_{\text{T}}$. A separate matrix is populated for each unique value of $D$. An example of P is shown in Fig. 5 for TIC 234994474 for the fixed duration of 1.2 hours. The $n_{\text{T}}\times n_{\text{T}}$ matrix P represents potential transit periods of PCs whose individual transit events may be separated in time by any of the off-diagonal elements of P or some multiple thereof. By its construction, the matrix P is skew-symmetric implying that only the positive non-zero matrix elements below the main diagonal are valid periods for consideration, or $\sum_{i=1}^{n_{\text{T}}-1}i$ periods. Because this search if for repeating transit-like events, it is required that $n_{\text{T}}>1$. In this way, more than two transit events are not required in order to detect a repeating putative PC. This fact extends the ORION detection sensitivity to nearly the full observational baseline or $\sim 27$ days for the majority of TICs.

Because each linear search with a unique $D$ is independent of the others, the P matrices of differential transit times are considered together. In this way a single master set of periods is compiled whose elements are referred to as Periods-of-Interest (POIs). Recall that each POI has an associated time of mid-transit $T_{0}$, duration $D$, depth $Z$ and $\mathcal{L}$ from the linear search. At this stage, the computationally tractable box transit model is substituted in favor of a more physical transit model whose model parameters are initialized using the box model parameters before optimization. The Mandel & Agol (2002) transit model is used through its implementation within the batman python package (Kreidberg, 2015) to compute model realizations given the input parameters $\theta=\{P,T_{0},a/R_{s},r_{p}/R_{s},i,e,\omega,a_{\text{LDC}},b_{\text{LDC}}\}$ where $e$ is orbital eccentricity, $\omega$ is the argument of periastron, and $\{a_{\text{LDC}},b_{\text{LDC}}\}$ are the quadratic limb darkening coefficients. In practice, only the parameters $\theta=\{P,T_{0},a/R_{s},r_{p}/R_{s},i\}$ are optimized by assuming circular orbits and fixing the quadratic limb darkening coefficients in the TESS bandpass to the values interpolated from the Claret (2017) grid over $T_{\text{eff}}$, $\log{g}$, and assuming solar metallicity. The $\theta$ parameters are optimized using the same routine which was used to optimize the GP hyperparameters during the initial de-trending stage (see Sect. 3.2). With each POI’s optimized transit model, the $\ln$ likelihood of the data is computed for use during the succeeding steps aimed at identifying repeating transit-like events from the initial set of POIs.

A series of cuts is performed on the set of optimized POIs to identify the most likely independent periods within the data. The first cut is to remove repeated period multiples. If POIs with integer period multiples (i.e. $2P_{i},3P_{i},\dots$) are indeed due to a transiting planet, then those POIs are likely to be manifestations of the same object. The exact $P$ value that is retained from a set of apparent period multiples is that with the largest $\ln$ likelihood. Any arbitrary pair of POIs ($P_{i},P_{j}$) is flagged as a period multiple if $P_{i}$ and $n\cdot P_{j}$ are within $f_{\text{P}}=1$% for any $n=2,\dots,n_{\text{transit}}$ where $n_{\text{transit}}$ is the number of transits can occur within the TIC’s observational baseline given $P_{i}$ and $T_{0,i}$. One important caveat to the removal of integer multiple periods is that resonant multi-planet systems are undetectable within the periodic transit search because all but one of the POIs will always be rejected in favor of its maximum $\ln$ likelihood multiple.

Similarly, because the set of POIs are derived from peaks in the linear search S/N spectrum, rational multiples of each POI must also be sampled (i.e. $P_{i}/2,P_{i}/3,\dots$). Consider a S/N spectrum derived from a light curve containing a single transiting planet with $n_{\text{transit}}>2$ but whose individual transits are only marginally detectable due to their amplitude relative to the photometric precision. Consider in this case that only a fraction of transit events detected during the linear search. The detection of only some transit events may result in a misidentified POI that is an integer multiple times greater than the underlying true period if one or more intermediate transit-events go undetected due to the effects of random noise. Therefore fractional multiples of each POI must be considered. These new periods are equal to $P_{i}/n$ for all integers $n$ resulting in a reduced period greater than or equal to the minimum orbital period considered by ORION: 0.5 days. The $\ln$ likelihood of the data under the box model with reduced period $P_{i}/n$ and remaining parameters $\{T_{0,i},D_{i},Z_{i}\}$ is calculated to be compared with the $\mathcal{L}$ value for the model with $P_{i}$. Here, the latter three parameters are fixed regardless of the input period. In the search over period multiples retains the period with the largest $\ln$ likelihood. The lower period limit of 0.5 days is imposed to limit the number of rational multiples of each POI that are investigated and because the temporal bin width used during the linear search stage already limits the sensitivity to short-period planets whose transit durations are comparable to the 30 minute bins used therein. Limiting the periodic search to orbital periods $<0.5$ is unlikely to result in a large number of missed transits owing to the intrinsically low occurrence rate of ultra-short-period planets ($\lesssim 1$%; Sanchis-Ojeda et al., 2014; Adams et al., 2016).

The result of the cuts to the initial set POIs are a set of repeating transit-like events which may or may not correspond to a transiting planet or some other form or periodic astrophysical source such as an eclipsing binary. The next steps in ORION are to vet the surviving POIs for systematic false positives and eclipsing binaries given their distinctive light curve features that can largely be vetted in an automated way.

3.5 Automated planet vetting

3.5.1 Automated vetting based on light curve features

Here, POIs are automatically vetted using a set of eight vetting criteria which investigate the flagged transit-like features in the de-trended light curve (see Sect. 3.4). This automated vetting stage is intended to identify false or insignificant transit-like events and thus provides a preliminary list of putative PCs prior to more selective human vetting and statistical vetting for astrophysical false positives.

The automated vetting criteria are controlled by the set of free parameters $\{c_{i}\}$ for $i=1,\dots,10$ which are described below and are included in the summary Table 3. The adopted values of these parameters controls the performance of ORION in terms of its detection sensitivity and false positive rate and were derived from early ORION executions on both archival Kepler and simulated TESS light curves333https://archive.stsci.edu/tess/ete-6.html prior to the first TESS data release. We do not however make any significant claims of their optimality.

The eight automated vetting criteria are defined as follows:

It is required that each POI’s transit depth $Z$ from its optimized transit model be $>0$. 2. 2.

The transit S/N is

[TABLE]

where $\sigma_{f,\text{transit}}$ is the photometric precision over the transit duration timescale and acts as a proxy for the Combined Differential Photometric Precision (CDPP${}_{\text{transit}}$; Christiansen et al., 2012). The number of observed transits is $n_{\text{transit}}$ given the vector of observations $\mathbf{t}$, the POI’s orbital period $P$, and mid-transit time $T_{0}$. The S/N${}_{\text{transit}}$ is required to be $>c_{1}=8.4$. 3. 3.

The transit parameters $P$, $a/R_{s}$, $Z$, and $i$ are used to phase-fold the light curve and compute the transit duration (Winn, 2010) such that the in-transit points, including those in ingress or egress, can be isolated. It is required that the difference in the median in and out-of-transit fluxes exceed $c_{2}=2.4$ median absolute deviations of the out-of-transit flux. 4. 4.

If a misidentified POI happens to be a less than the true period, then the phase-folded light curve will appear to contain out-of-transit points in-transit. This is combated by requiring that the number of in-transit points lying below $Z+\sigma_{Z}$ (where $\sigma_{Z}$ is the $1\sigma$ uncertainty on the transit depth) accounts for at least the a $c_{3}=0.7$ fraction of all in-transit points. 5. 5.

It is required that the in-transit sampling be approximately symmetric in time by insisting that the number of points between $T_{1}$ and $T_{2}$ be $\in[50-c_{4},50+c_{4}]$% of the total number of in-transit points (i.e. between $T_{1}$ and $T_{4}$). $c_{4}$ is set to $10$%. 6. 6.

Flare stars such as TIC 25200252 as shown in Fig. 6 are found to result in a number of misidentified transit-events. Flare events are therefore searched within each light curve by first flagging individual flux measurements which are $>c_{5}=8$ median absolute deviations brighter than the median flux baseline. However, by the aforementioned criterion, individual stochastic flux jumps can also mimic flares. It is therefore required that any window over which a possible flare event occurs must contain $>c_{6}=2$ successive bright measurements above the $c_{5}$ threshold in order to identify a flare. Flux measurements occurring within a flare window are identified from the $q^{\text{th}}$ percentile of the light curve flux distribution where $q$ is the fraction of the observational baseline that occurs within a flare’s duration. The total flare duration over the light curve is calculated from the number of detected flares multiplied by the characteristic M dwarf flare duration $c_{7}=30$ minutes (Moffett, 1974; Walkowicz et al., 2011; Hawley et al., 2014). Transit-like events with an identified flare occurring within $c_{8}=4$ transit durations from $T_{0}$ of a POI are vetted as flares. 7. 7.

Visual inspection of a number of TIC light curves observed during sector 1 frequently reveals sharp flux losses at the light curve edges. This signature is often falsely attributed to transit-like signal but is clearly a systematic effect that is not always well-modelled during de-trending stage. Because this edge effect appears to operate over the final $\sim 4-5$ hours of the light curve, POIs with mid-transit times within $c_{9}=4.8$ hours of either the first or final flux measurements are automatically flagged as probable false positives. 8. 8.

The optimized transit models of the POIs that satisfy all of the seven aforementioned criteria are removed from the light curve. This produces a maximally clean light curve whose residuals should only arise from random noise in photometry or from inaccuracies in the independent systematic GP and transit models. The numpy.correlate python function is used to compute the autocorrelation of the residual light curve as a function of time delay as light curves demonstrating large autocorrelations due to imperfect systematic models, can often mimic transit-like events which satisfy the previous vetting criteria. This criterion is particularly important for the use of ORION on K2 light curves which often exhibit significant temporal correlations due to the thrusts used for re-orientation of the spacecraft and its imperfect correction (Vanderburg & Johnson, 2014). It is requires that the autocorrelation function for delays greater than zero be $\leq c_{10}=0.6$.

3.5.2 Automated vetting of eclipsing binaries

POIs that obey all eight automated vetting criteria from Sect. 3.5.1 are passed along to vetting of eclipsing binary (EB) astrophysical false positives. Six free parameters control the performance of the astrophysical vetting procedure: $\{c_{\text{EB},i}\}$ for $i=1,\dots,6$. The EB vetting criteria are adopted from a variety of sources (Batalha et al., 2010; Bryson et al., 2013; Günther et al., 2017; Crossfield et al., 2018) and are used to flag light curve features consistent with being an EB rather than a transiting planet. The EB vetting criteria are defined as follows:

POIs are required to have $r_{p}/R_{s}<c_{\text{EB},1}=0.5$. 2. 2.

It is also required that the inferred companion radius $r_{p}<c_{\text{EB},2}=30$ R*⊕*. 3. 3.

The observed transit duration $D$ is required to be less than the transit duration corresponding to planet with radius $c_{\text{EB},2}=30$ R*⊕*. 4. 4.

Searches for secondary eclipses are conducted by first sampling eclipse duty cycles (i.e. the fraction of the orbit during eclipse) from the Shan et al. (2015) duty cycle PDF (see their Fig. 4). This distribution was derived from a synthetic population of M dwarf EBs based on Kepler binary statistics and is dependent on the population of EB total radii, total masses, orbital periods, and eccentricities. In each POI’s light curve, phase-folded points occurring within a duty cycle centered on phase$=0.5$ are considered for possible contamination by a secondary eclipse. With the in-eclipse points defined, the secondary eclipse depth $Z_{\text{occ}}$ and photometric precision during the occultation $\sigma_{\text{occ}}$ are computed. It is required that EBs satisfy the following conditions:

[TABLE]

Recall that $Z$ and $\sigma_{f,\text{transit}}$ are the transit depth and in-transit photometric precision. $c_{\text{EB},4}$ is set to 5 (Günther et al., 2017) and EBs identified by this criterion are required to have $>c_{\text{EB},5}=50$% duty cycle samples satisfy the above conditions. 5. 5.

EBs also exhibit ‘V’-shaped transits due to the self-luminosity of each companion. To search for ‘V’-shaped transits, the ingress time $T_{12}$ and egress time $T_{34}$ are calculated from the optimized transit model and compared to the total transit duration $D$. ‘V’-shaped transits are required to have $T_{12}+T_{34}$ which are $c_{\text{EB},6}\geq 90$% of $D$. Notably, ‘V’-shaped transits may also be indicative of planetary transits at large impact parameters so ‘V’-shaped transits are not explicitly discarded but are instead assigned a non-definitive disposition based solely on this criterion.

3.6 Joint systematic plus transiting planet modelling

The set of transit-like events that satisfy all of the vetting criteria presented in Sects. 3.5.1 and 3.5.2 are treated as PCs in this, the final ORION stage. At this point the modelling of systematic light curve effects using a 1-dimensional GP regression model from Sect. 3.2 is revisited. The alternative is to simultaneously sample the joint GP plus transit light curve parameter posterior PDF using Markov Chain Monte-Carlo (MCMC) simulations. Explicitly, the light curve model is modified by replacing the previously null mean model $\boldsymbol{\mu}(\mathbf{t})$ with a full transit model containing all putative PCs. Overfitting by the systematic model, which can partially fill in planetary transits, is mitigated by simultaneously modelling systematics and PCs. The resulting joint systematic+planet model is therefore derived in a self-consistent manner with more robust solutions for the transiting PC parameters of interest. MCMC sampling of the transit parameter marginalized posterior PDFs allows us to compute point estimates of their MAP values and uncertainties for later use.

For systems containing $N_{\text{PC}}$ PCs, $4+5N_{\text{PC}})$ parameter PDFs are sampled by continuing to insist on circular orbits and fixed limb-darkening coefficients. Explicitly, the GP hyperparameters $\Theta=\{\ln{a_{\text{GP}}},\ln{\lambda},\ln{\Gamma},\ln{P_{\text{GP}}}\}$ are fit along with the transiting planet parameters $\theta=\{P_{i},T_{0,i},a_{i}/R_{s},r_{p,i}/R_{s},i_{i}\}$ $\forall$ $i=1,\dots,N_{\text{PC}}$. The GP hyperparameters are initialized to their maximum likelihood values from the de-trending stage and are continued to be bounded by the broad uniform priors listed in Table 4. Transit parameters are initialized by their maximum likelihood values assuming fixed GP hyperparameters from de-trending. The adopted prior PDFs on the transit model parameters are also reported in Table 4 for the most common case of fitting transiting PCs with multiple transits over the observational baseline. As we will see in Sect. 4.1.2, some priors will be modified when sampling transit parameters used to model only a single transit.

MCMC sampling is performed with the emcee ensemble-sampler (Foreman-Mackey et al., 2013). One hundred walkers are initialized in small Gaussian balls centered on each parameter’s initial value. Throughout the MCMC, each walker’s acceptance fraction is monitored and warns the user when its mean value over all walkers does not fall within the desired range of 20-60% in either of the burn-in or final sampling stages. The desired duration of each MCMC stage is $\gtrsim 10$ autocorrelation times. Warnings are again produced if the MCMC chains fail to reach this length.

The main ORION output is a list of objects of interest (OIs) along with samples from the model parameter marginalized posterior PDFs and point estimates of each parameter’s MAP value and uncertainties derived from the $16^{\text{th}}$ and $84^{\text{th}}$ percentiles of their 1-dimensional marginalized PDF. The raw light curves and models sampled at observation times $\mathbf{t}$ are also saved to produce summary images such as the example shown in Fig. 7 for OI 234994474.01 whose PC is a known TESS Object of Interest: TOI-134.01, a close-in terrestrial planet being validated by HARPS and PFS RVs (Astudillo-Defru et al., 2019).

4 ORION planet search around low mass stars in TESS sectors 1 & 2

For its inaugural application we apply the ORION transit detection pipeline to the 2 minute extracted light curves from the first two TESS sectors. Overall, ORION produces a set of 121 OIs around 96 of the 1599 low mass TICs in our stellar sample after automated vetting. It is expected that many of these postulated PCs will be false positives due to imperfect corrections of systematic effects or to astrophysical false positives other than eclipsing binaries as EBs have certain distinct photometric features which are flagged during the automated vetting stage. As such, we proceed with manual vetting of all ORION OIs via human inspection of the pipeline’s output. This step is particularly important for newly developed transit search algorithms to develop an understanding of common sources of false positives and if or how they may be corrected in future versions.

4.1 Manual vetting of ORION planet candidates

We conduct a visual inspection of each of the raw, de-trended, and phase-folded light curves for each OI produced by ORION. In this analysis we flag 97/121 PCs as being residual systematic effects which are misidentified as transits. Of those, 19 appear to have been directly affected by measurements obtained between $\sim$ 1347 and 1349 BJD - 2,4570,000 during sector 1 at times when TESS briefly lost much of its pointing precision. This effect is not perfectly corrected for in many of the sector 1 extracted light curves nor by our own systematic modelling. Most of the remaining OIs flagged as false positives are attributable to residual systematics mimicking transits. Visual inspection indicates that these OIs are clearly inconsistent with being a planetary transit.

4.1.1 Peculiar light curve features

The two OIs 63037741.01 and 434105091.01 produced by ORION exhibit peculiar light curve features which do not appear to be consistent with planetary transits or residual systematics. These features are depicted in Fig. 8. Each of the two TIC 63037741 features and the TIC 434105091 feature are clearly seen at a high S/N either above or below the surrounding flux continuum. The TIC 63037741 flux excess feature (‘a’ in Fig. 8) is approximately 4% brighter than the baseline flux and has a temporal evolution which is inconsistent with being a transient flare given its lack of a steep rise in flux followed by an exponential decay (Hawley et al., 2014). The second TIC 63037741 feature (‘b’) has a 6% decrease in flux with a similar structure to feature ‘a’, except that it is inverted and with approximately half of the timescale. Feature ‘b’ also appears to have a slightly longer egress time than its ingress which can be a signpost of an extended transiting object if astrophysical in origin. Lastly, feature ‘c’ around TIC 434105091 shows 2-3 flux dips over the entire ‘transit’ duration lasting $\sim 6$ hours. We remain agnostic as to the exact interpretation of these features but their existence is highlighted here for any interested parties.

4.1.2 Single transit events

The interpretations of the remaining 24/121 OIs as transiting planets are deemed plausible by our manual vetting. These preliminary PC dispositions are based purely on the resemblance of the light curve features to periodic transit events or in some instances, to single transits (ST) events that do not show compelling evidence for periodicity over the observational baseline. We note that ORION is not optimized nor intended to be sensitive to the detection of ST events. However, some ST events with moderate to high S/N can be detected but incorrectly classified as periodic events if at least one other transit time in the linear search S/N spectrum exceeds S/N${}_{\text{thresh}}$, even if that event is resulting from noise (see Sect. 3.3). In this case, the ST event is folded to the time difference between the ST and any of the other S/N events exceeding S/N${}_{\text{thresh}}$. If the ST event has a sufficiently high S/N on its own then the addition of noise by phase-folding to the incorrect period, but correct $T_{0}$, may result in a feature that still passes our automated vetting criteria due solely to the significance of the ST event. The inferred period of such a ST event in ORION will therefore always be less than its true period if the feature is indeed the result of a singly transiting planet.

Three OIs resembling ST events are identified during manual vetting around the TICs 49678165, 92444219, and 415969908. The latter TIC already hosts the known TOI-233.01 at 11.7 days but the proposed ephemeris for TOI-233.01 is inconsistent with the $T_{0}$ of our putative ST event. A more complete discussion of this and the other individual systems is reserved for Sect. 5.1.

For each of the three OIs classified as a putative ST, we refine their transit parameters by re-modelling their transits by isolating the light curve around $10D$ and using MCMC to sample the transit model parameters with just a single transit (Seager & Mallén-Ornelas, 2003). In these simulations the $P$, $T_{0}$, and $a/R_{s}$ priors are modified as listed in Table 4. The orbital period of the ST is further restricted to periods greater than the largest time difference between $T_{0}$ and both edges of the light curve’s baseline. The resulting periods are largely uncertain with their posterior PDFs showing extended tails out to long periods $\gtrsim 100$ days, as is expected for transit models which lack multiple events to constrain $P$. The revied transit parameters for our three putative ST events, derived from the isolated data and with the aforementioned updated priors, are used in place of the transit model solutions produced by ORION with their likely underestimated periods.

4.2 Statistical validation of transiting planet candidates

We are currently not in a position to distinguish between confirmed planets and various astrophysical false positive scenarios in an absolute sense. This is because of the lack of follow-up observations in this study which are ultimately required to validate or disprove the planetary nature of our OIs. Despite the lack of such follow-up observations, it is still advisable to attempt to statistically validate OIs by inferring the relative probabilities of a variety of astrophysical false positive (AFP) scenarios which can be compared to the planetary interpretation. Such considerations are further motivated given that the rate of AFPs in the 2 minute TESS light curves is expected to be significant ($\sim 60$%; Sullivan et al., 2015).

We attempt to statistically validate our 24 OIs around 22 TICs using the PyMultinest (Buchner et al., 2014) implementation of the probabilistic transit validation software vespa (Morton, 2012, 2015) which computes the planetary false positive probability (FPP) for use in establishing the final dispositions of our OIs. vespa considers six AFP scenarios as potential explanations for transit-like signals. These include undiluted eclipsing binaries (EB), hierarchical triple EBs (HEB), and blended background or foreground EBs which are not physically associated with the target (BEB). Each of these scenarios then have two instances with the first assuming the input orbital period and the second assuming twice the input orbital period. We note however that the forthcoming statistical OI interpretations are not treated as abolsutely definitive in lieu of follow-up observations aimed at distinguishing transiting planets from AFPs.

For vespa input we use the TIC’s celestial coordinates ($\alpha$,$\delta$), stellar parameters $T_{\text{eff}}$, $\log{g}$, and $\varpi$, along with the star’s $JHK_{\text{S}}$ photometry. vespa also requires the photometric band in which the putative transit is detected but the code cannot properly handle the TESS bandpass in its current version. Fortunately, the central wavelengths of the TESS bandpass and the Cousins $I_{C}$-band are similar but with the TESS band being much wider (Sullivan et al., 2015). Given the similarity of the $I_{C}$ and SDSS $i$-band, and the compatibility of the latter within vespa, we use $T$ and $K_{\text{S}}$ to derive $i$ using the color relation from Muirhead et al. (2018). We also pass to vespa the OI’s orbital period and planet-to-star radius ratio along with its de-trended light curve following the removal of all candidate transit models that are not associated with the OI being statistically validated. The light curves are phase-folded and restricted to $\pm 3D$ around $T_{0}$ for comparison to light curve models generated under the transiting planet and each AFP scenario.

vespa also requires constraints on the maximum angular separation (maxrad) from the target star that should be searched for potential blending sources. We limit this separation to be less than the median full width at half maximum (FWHM) of the target’s approximate point spread function (PSF). The FWHM of the PSF is derived by fitting a 2-dimensional Gaussian profile to the target image in each target pixel file over time and adopting the median FWHM as vespa input. The median FWHM value among the 22 TICs is $\sim 37$ arcsec or nearly two TESS pixels across. Over such a large field, it is reasonable to expect that many of the OIs may be favored by either the BEB or BEB2 models.

Lastly, vespa requires the maximum permissible depth of a secondary eclipse of an EB (secthresh) to be specified. Recall that attempts within ORION were made to automatically vet EBs among our OIs in Sect. 3.5.2. We therefore expect that vespa is unlikely to detect any probable EBs. Nevertheless, the input secthresh value for each TIC is derived from the box model depths fitted to each transit time in $\mathbf{T}_{0}$ during the linear search stage (see Sect. 3.3). After masking measurements occurring within the PC’s transit window and extrapolating the fitted depths to the PC’s transit duration, we adopt the 95${}^{\text{th}}$ percentile of the depth distribution depths as the value of secthresh (Crossfield et al., 2018). The median secthresh is $\sim 2100$ ppm. The input maxrad and secthresh for each OI are reported in Table 5 along with the results of our vespa calculations.

The true power of vespa is realized when additional follow-up observations such as contrast curves from AO-assisted imaging or photometric follow-up are used to inform the interpretation of transiting PCs. Given the lack of such data in this study, we adopt conservative limits on the interpretation of the resulting vespa probabilities. Similarly, we also do not claim to validate planets with ultra low FPP ($<0.01$; e.g. Montet et al., 2015; Crossfield et al., 2018; Livingston et al., 2018) as we will caution in Sect. 4.3 that vespa results should not be taken absolutely in the absence of follow-up observations. Our limiting values on interpreting FPPs are as follows: OIs with FPP$<0.1$ are classified as PCs. Similarly, OIs are classified as AFPs when FPP $\geq 0.9$ and have their dispositions assigned to the specific AFP model with the highest probability. OIs with intermediate FPPs are classified as putative planet candidates (pPC).

The statistical validation calculations with vespa result in 13/24 of our OIs being classified as PCs plus 3/24 as pPCs. The OIs 49678165.01, 415969908.02, and 92444219.01 correspond to the three ST events detected during the manual vetting stage. We reclassify these objects as STs and a pST respectively. Seven of the remaining eight OIs are favored by either BEB model with 4/7 BEBs and 3/7 BEB2s. The MCMC during the vespa calculation of the lone remaining OI 235037759.01 failed to converge leaving its disposition as of yet undefined. We will show in the following subsection that despite the failure of the FPP calculation, the nature of OI 235037759.01 is likely to be an AFP. Although we are unable to distinguish between the different AFP scenarios. The derived rate of AFPs from this small sample of OIs is $\sim 33\pm 12$% which is somewhat lower than expected AFP rate of 60% from the TESS simulations by (Sullivan et al., 2015).

4.3 Querying GAIA sources to supplement statistical validation calculations

vespa calculations are based on synthetic stellar populations from the TRILEGAL galaxy model (Girardi et al., 2005). These synthetic results can be supplemented by querying the GAIA DR2 in the vicinity of each TIC to investigate the number density and brightness distribution of nearby sources on the sky. In this way, we hope to find supporting empirical evidence for any of the BEB interpretations of OIs with high FPPs as those inferences should be expected if nearby bright sources fall within or near the PSF of the targeted TIC. The resulting maps of GAIA sources around the 22 TICs with OIs in our sample are shown in Fig. 9. Querying the GAIA DR2 is performed identically to the method used in Sect. 2.2 to match TICs with the GAIA DR2 catalog although here we conduct our searches with a fixed radius of 105 arcseconds or $\sim 10$ TESS pixels in diameter.

From Fig. 9 it is clear that many of the statistically favored interpretations as either some form of planet candidate or a BEB are consistent with the lack or prevalence of nearby bright sources to the targeted TIC respectively. All panels with planet candidate OIs in Fig. 9, with the exception of panel ‘l’, show no or minor sources of comparable brightness within or very close to the target PSF edges as to significantly contaminate the measured TIC photometry and consequently result in a probable FP. Similarly, all AFP panels other than ‘q’ and ‘s’ do have at least one neighbouring source of comparable brightness that may be responsible for the favorability of an AFP scenario by vespa. This includes the TIC 235037759 whose vespa calculation failed. Two bright sources are clearly seen to contribute to the flux within the target’s PSF thus supporting the probable interpretation of the OI 235037759.01 transit-like event as being caused by an AFP.

We note however that some discrepancies between the distributions of GAIA sources and our vespa interpretations still persist. Particularly with regards to the TICs 100103200 and 100103201 (panels ‘l’ and ‘p’ in Fig. 9) which strongly favor the PC and BEB models respectively despite being located within 1 TESS pixel of one another on the sky, having very similar brightnesses (i.e. $J=7.50,7.66$), and being located at effectively identical distances (i.e. $d=16.745$ pc). Such properties are reminiscent of an M dwarf binary system. Perhaps naively, we might expect this architecture to favor almost any AFP scenario for both TICs including blends, an EB, or an HEB. Indeed the apparent flux dips which appear qualitatively consistent with a transiting planet around either one of the TICs, is also seen to have a clear manifestation in the light curve of the other as evidenced in Fig. 10. This is almost certainly caused by the overlap of each target’s PSF. However vespa results indicate FPPs that differ by over two orders of magnitude between the two TICs. Perhaps it is feasible, although seemingly unlikely, for TIC 100103200 to host a detectable transiting PC while being blended with the nearby TIC 100103201 whose transit-like events are strongly favored by the BEB scenario. Indeed the transit times of each TIC’s transit-like events flagged by ORION appear out of phase as they do not align nor do they overlap in Fig. 10 implying that transit-like events detected around each TIC by ORION do not affect the transit-like events in the light curve of the other. Even if there are regions of the out-of-transit light curve that are mutually affected. We are therefore left with the questionable interpretation of these OIs by the aforementioned conflicts. We opt to use the vespa results to conclude final dispositions of these OIs but we also exercise caution by modifying the statistical disposition of OI 10010300.01 from a PC to a pPC.

4.4 Population of planet candidates

After manual vetting and statistical validation we are left with sixteen candidate planets. These include ten PCs, two STs, three pPCs, and one pST. Half of our candidates are ‘new’ having not yet been released as TOIs444As of December 19, 2018.. Point estimates of observable and derived planetary parameters for these candidates are reported in Table 6. Fig. 11 also depicts their phase-folded light curves along with the transit models computed using the MAP parameter values from Table 6.

Fig. 12 depicts our candidate population and compares it to the twelve TOIs whose TICs are included in our stellar sample. Our candidates have orbital periods from 1.4-20 days for PCs and from 35-50 days for ST events. Although ST periods exhibit large uncertainties of order the MAP period value. Our candidates have radii between 1.1-3.8 R*⊕* potentially making them targets of interest for the completion of the TESS level one science requirement of delivering 50 planets with $r_{p}<4$ R*⊕* with measured masses. In Sect. 5.3.1 we will discuss the prospects that our candidates have for contributing to the realization of the TESS level one science requirement. We do not detect any hot sub-Neptunes in the photoevaporation desert (Lundkvist et al., 2016) nor any small planets ($\lesssim 1.5$ R*⊕*) on orbits longer than $\sim 20$ days. This is largely attributable to our poor detection sensitivity in that regime due to the limited TESS baselines of just 27 days.

Fig. 12 also depicts our candidates as a function of insolation. The majority of candidates (11/16) experience incident insolation levels $S$ in excess of twice that of the Earth. However five candidates, including all three ST events and two pPCs, are likely more temperate and experience insolations $\lesssim 1.5$ S*⊕*. This $S$ limit marks the ‘recent-Venus’ inner edge of the low mass star habitable zone (HZ; Kopparapu et al., 2013). Our STs also lie within the more conservative HZ definition bounded by the ‘water-loss’ inner edge, where an increase in insolation results in the photolysis of stratospheric water vapor causing the atmosphere to experience rapid hydrogen escape, and the ‘maximum-greenhouse’ outer edge where an increase in CO2 no longer results in a net surface heating due to the increased albedo. Our five temperate candidates may represent attractive targets for the characterization of HZ exoplanets around nearby low mass stars. We will address the prospect of atmospheric characterization of these planets in Sect. 5.3.2.

The distribution of our candidates versus stellar parameters of interest are shown in Fig. 13. All candidates are detected around stars hotter than 3000 K which approximately corresponds to stars earlier than M5.5V (Pecaut & Mamajek, 2013). Most candidates are detected around stars with $T_{\text{eff}}$ $\in[3200,3900]$ K which is largely consistent with the population of confirmed transiting planets recovered from the NASA Exoplanet Archive on December 13, 2018 (Akeson et al., 2013), modulo the TRAPPIST-1 planets (Gillon et al., 2017), GJ 1214b (Charbonneau et al., 2009), and the Kepler-42 planets (Muirhead et al., 2012). The median effective temperature of the candidate-hosting TICs in our sample is 3560 K. A notable dearth of PCs with $r_{p}\gtrsim 2$ R*⊕* exists around stars hotter than $\sim 3500$ K. The cause of this is unlikely to be due to sensitivity losses around these relatively hot (and correspondingly bright) stars and is instead likely attributable to the sharp decrease in the number of stars within our sample at $T_{\text{eff}}$ $\sim 3550$ K (see Fig. 1).

The distributions of previously confirmed planets and our candidates versus $J$-band magnitude reveals that many ($\sim 10/16$) of our candidates orbit systematically brighter stars than the majority of confirmed transiting planets with $J<10$. This directly demonstrates the power of a survey mission like TESS at discovering transiting planets orbiting nearby bright stars which are amenable to forthcoming detailed characterization efforts. The remaining six candidates still orbit moderately bright stars with $J\in[10,12]$. The median $J$ of our candidate-hosting TICs is 9.9. From the number of ORION candidates as a function of $J$, it is clear that the ORION sensitivity to planets orbiting low mass TICs starts to drop off around $J\gtrsim 12$. The large photometric uncertainties in this regime are largely dominated by photon-noise from the target star, zodiacal light, and unresolved background stars (Ricker et al., 2015).

The distance distribution of our candidates is also included in Fig. 13. All of our candidates are found between 10-65 pc with a median distance of 34 pc. Our sample includes five candidates within 25 pc as well as the closest transiting PC around a low mass star to date: TIC 307210830.01 (TOI-175.01) at 10.6 pc which is also hypothesized to contain two additional TOIs not detected by ORION (see Sect. 5.2). Among our eight new candidates not released as TOIs, we detect two candidates within 25 pc. These include the pPC 100103200.01, and the PC 206660104.01. Barring the rejection of the candidates in these systems and the TOIs around TIC 307210830, these three systems of five planet candidates represent 3/8 of the closest transiting planetary systems around low mass stars along with GJ 1132 (Berta-Thompson et al., 2015; Bonfils et al., 2018), TRAPPIST-1 (Gillon et al., 2017; Luger et al., 2017), LHS 1140 (Dittmann et al., 2017b; Ment et al., 2018), GJ 1214 (Charbonneau et al., 2009), and GJ 3470 (Bonfils et al., 2012).

5 Discussion

5.1 Discussions of individual systems with planet candidates

OI 12421862.01. This PC has already been reported as TOI-198.01. The ORION planet parameters are consistent with those of TOI-198.01 with a 20.4 day orbital period and measured radius of 1.6 R*⊕*, making it an interesting target for probing the photoevaporation valley around M dwarfs.

OI 47484268.01. This pPC based on its moderate FPP, has already been reported as TOI-226.01. The ORION orbital period is consistent with that of TOI-226.01 but we derive a 10% smaller planet radius of 3.8 R*⊕* despite the star’s refined radius being 15% larger. At 1.4 S*⊕*, this pPC orbits within the ‘recent-Venus’ habitable zone but remains an attractive target for rapid RV characterization owing to its expected large mass compared to most of the smaller candidate planets (see Sect. 5.3.1).

OI 49678165.01. A new candidate ST event with an estimated period between 21-85 days and a radius of 3 R*⊕. This candidate is likely the coldest object in our sample with a MAP $S=0.3$ S⊕* and a corresponding equilibrium temperature of $T_{\text{eq}}=212$ K assuming zero albedo and perfect heat redistribution.

OI 92444219.01. A pST based on its moderate FPP. This cool 3 R*⊕* candidate is has an estimated period between 22-95 days placing it at $S=0.5$ S*⊕* with an equilibrium temperature $T_{\text{eq}}=238$ K.

OI 100103200.01. The pPC with ultra low FPP but is seen to likely be contaminated by the nearby TIC 100103201 based on the GAIA astrometry. This pPC orbits a $J=7.5$ star with an 18.4 day period and whose size is 2 R*⊕* making it a attractive target for both RV characterization and possibly transmission spectroscopy observations with JWST if impending difficulties of observing bright targets can be strategically mitigated (see Sect. 5.3.2).

OI 206660104.01. A new terrestrial-sized (1.2 R*⊕*) PC at 13.4 days. This PC is the second smallest candidate recovered by ORION and is smaller than most (3/4) of the TOIs missed by ORION (see Sect. 5.2).

OI 231702397.01. This PC has already been reported as TOI-122.01. The ORION orbital period is consistent with that of TOI-122.01 although we derive a 15% larger radius (2.8 R*⊕) corresponding to the 13% larger stellar radius in our sample. Given its large size, the expected mass from the Chen & Kipping (2017) mass-radius relation is 8.26 M⊕* making it an attractive target for rapid RV follow-up despite being relatively dim with $J=11.5$.

OI 234994474.01: This PC has been reported as TOI-134.01 and is being validated with RVs from HARPS and PFS (Astudillo-Defru et al., 2019). The ORION parameters are largely consistent with the TOI-134.01 parameters albeit with a slightly smaller radius of 1.38 R*⊕* (12% reduction) for an unchanged stellar radius. Being by-far the hottest target in candidate sample ($T_{\text{eq}}=965$ K) and orbiting an early M dwarf with $J=7.9$, this PC is one of the best targets for any further RV characterization to search for longer period companions, transmission spectroscopy, and even thermal emission spectroscopy. Regarding the latter, TOI-134.01 is even more favorable than the previously most attractive terrestrial-sized planet for such observations: GJ 1132b (Morley et al., 2017).

OI 260004324.01. A new terrestrial-sized PC which is the smallest in our sample at 1.15 R*⊕* with a 3.8 day period. Its small size and small expected mass of 1.6 M*⊕* around a 0.56 M*⊕* early M dwarf ($J=8.8$) will make this a slightly more challenging target for RV follow-up but may still be of interest for probing the low mass end of the 50 planets smaller than 4 R*⊕* targeted for completion of the TESS level one science requirement.

OI 262530407.01. This PC has already been reported as TOI-177.01. The ORION orbital period is consistent with that of TOI-177.01 although we find a 12% smaller radius of 1.87 $R_{\oplus}$ for an unchanged stellar radius. The short period and host star brightness ($J=8.17$) make this PC an exceptional candidate for transmission spectroscopy observations and is one that is less affected by the NIRISS bright limit of its SOSS mode compared to other close-in super-Earth-sized planets in this catalog.

OI 278661431.01. A new pPC based on its moderate FFP. If validated, the candidate would have an orbital period of 17 days and a radius of 2.8 $R_{\oplus}$. This pPC is temperate at $S=1.1$ S*⊕* and near to the ‘water-loss’ HZ inner edge. This pPC orbits one of the cooler TICs in our candidate-hosting sample ($T_{\text{eff}}$ $=3300$ K) with a correspondingly small radius of 0.28 R*⊙*, thus making it an attractive target for transmission spectroscopy observations.

OI 305048087.01. This PC has already been reported as TOI-237.01. The ORION orbital period is consistent with the 5.4 day period and 1.7 R*⊕* radius of TOI-177.01.

OI 307210830.01. This PC is one of the three reported TOIs around TIC 307210830 (i.e. 175-01, 02, 03). Only TOI-175.01 is detected by ORION for the reasons discussed in Sect. 5.2. The ORION planet parameters for this, the middle planet in this candidate three planet system, are all consistent with those for TOI-175.01. This PC is in the closest planetary system in our sample and is correspondingly an attractive target for RV characterization of individual masses and the RV analysis of the possible resonant pair 175.01/175.02 for comparisons to TTV analyzes from follow-up photometry. This PC is also a viable target for the atmospheric characterization of a terrestrial-sized planet (1.3 R*⊕*).

OI 415969908.01. This PC has already been reported as TOI-233.01. The ORION orbital period is consistent with that of TOI-233.01 but finds a 26% smaller radius (1.9 R*⊕*) in part because of the refined stellar radius being reduced by 21% in our sample.

OI 415969908.02. The ST event detected in the light of TIC 415969908 which already hosts the aforementioned candidate TOI-233.01 at 11.7 days. The estimated period of OI 415969908.02 is 32-108 days which effectively spans the conservative HZ limits from Kopparapu et al. (2013) and whose MAP value is 53 days or $S=0.5$ S*⊕. The measured radius is 2 R⊕* ($r_{p}/Rs=0.05$) making this ST event the most difficult to follow-up from the ground among the three ST events in our candidate catalog.

OI 441056702.01. A new PC with an orbital period of 6.3 days and radius of 2 R*⊕* thus potentially being located within the photoevaporation valley around M dwarfs. Orbiting a moderately bright early-M to late-K dwarf ($T_{\text{eff}}$ = 4030 K) and with an expected mass of 4.5 M*⊕*, this PC represents another attractive target for RV characterization aimed at addressing the TESS level one science requirement.

5.2 TOIs undetected by ORION

Our stellar sample using refined stellar parameters based on GAIA distances, contains ten TICs with TOIs listed as part of the TESS alerts555As of December 19, 2018.. These ten systems host a total of twelve TOIs in nine single PC systems and TIC 307210830 which hosts three TOIs. The ORION results presented in this paper include the independent detection of 8/12 TOIs with an additional ST event around TIC 415969908 which already hosts TOI-233.01. The discrepancies between the TESS alerts and our ORION results are described in what follows.

TIC 259962054. This TIC was observed in the consecutive TESS sectors 1 and 2. The TOI-203.01 has an orbital period of 52 days, longer than any repeating candidate in our catalog. A signal at $\sim 52.1$ days is found in the ORION linear and periodic search stages with an orbital phase that is consistent with that reported for TOI-203.01. This suggests that the TOI-203.01 transit-like signal at 52 days does exist in the light curve. The phase-folded light curve satisfies all but the second criterion from the automated vetting stage (see Sect. 3.5.1) with S/N${}_{\text{transit}}=5.6<c_{1}=8.4$. This is principally because the fitted transit depth within ORION is $Z=1839$ ppm which is just $\sim 73$% of the TOI’s reported depth thus making it difficult to confirm the nature of the repeating signal as due to a transiting planet with just two transits observed.

TIC 307210830. This system contains three TOIs (175.01, 175.02, 175.03) at 3.69, 7.45, and 2.25 days respectively. The innermost planet candidate is not found during the linear and periodic search stages. This is likely caused by the candidate’s depth of 571 ppm (as reported in the TESS alert portal) being less than the median photometric uncertainty of its light curve (i.e. median($\sigma_{\text{f}}$) $\sim 770$ ppm). The two remaining planet candidates were each seen to be detected in the ORION linear search stage owing to their $\sim 3$ times larger transit depths. However, this candidate pair has an orbital period ratio that is within 1% of a 2:1 resonant configuration. Recall that pairs of periods of interest which are that close to an integer period ratio will have one of those periods automatically discarded in the periodic search stage due to the detected periodic signals likely being aliases of the each other rather than being due to two separate planetary candidates.

TIC 316937670. TOI-221.01 has an orbital period of 0.624 days and a low transit depth of 954 ppm. By adopting the reported TOI-221.01 transit depth and duration, we estimate CDPP${}_{\text{transit}}$ and find that $Z/\text{CDPP}_{\text{transit}}\sim 1.1$. Because of this, the results of the ORION linear search only detect a single transit-event above the S/N threshold such that no periodic events can be found. If the linear search S/N threshold is lowered such that multiple transits from TOI-221.01 are detected in the linear and periodic search stages, then the expected S/N${}_{\text{transit}}$ is $\sim 7.2$ which would still be insufficient to detect the PC in a single TESS sector. We are further discouraged by the prospect of lowering the linear search S/N threshold as this would drastically increase the number of FPs purely from the noise.

5.3 Prospects for follow-up observations

5.3.1 Mass characterization via precision radial velocities

The TESS level one science requirement is to deliver at least 50 planets smaller than 4 R*⊕* with measured masses via precision radial velocity (RV) follow-up. All sixteen of our candidates have a measured radius consistent with being $<4$ R*⊕* (see Table 6). Using the empirical mass-radius relation for small planets $<14.26$ R*⊕* from Chen & Kipping (2017), we compute the maximum likelihood masses $m_{p}$ of our planet candidates and single transit events to then infer their expected RV semi-amplitudes using

[TABLE]

These values are shown in the first panel of Fig 14 versus $J$-band magnitude and are accompanied by the simulated TESS yield from Barclay et al. (2018) and the set of confirmed transiting planets from the NASA Exoplanet Archive. Both of the latter samples are restricted to cool host stars (i.e. $T_{\text{eff}}$ $<4200$ K). Many existing high performance RV spectrographs are stable at the level of a few cm/s but RV observations are often limited by photon noise and intrinsic stellar activity at the level of 1 to a few m/s, even for relatively inactive stars (Fischer et al., 2016). The majority of our objects of interest have expected $K_{\text{RV}}$ values in excess of this typical RV sensitivity limit and are reported in Table 7.

Cloutier et al. (2018) calculated the observing time required to complete the TESS level on science requirement based on the expected TESS yield from Sullivan et al. (2015). The 50 TOIs requiring the shortest time commitment to characterizing their planet masses at $5\sigma$ with RVs satisfy the following empirically-derived conditions:

[TABLE]

where

[TABLE]

is a proxy for $K_{\text{RV}}$ that can be computed from transit-derived parameters. Seven out of sixteen of our candidates satisfy Eqs. 9 and 10 and are highlighted in Table 7 and in Fig. 14. If astrophysical false positive scenarios can be ruled out for these OIs, then they represent highly favorable targets for RV follow-up observations and the rapid completion of the TESS level one science requirement. That is, assuming that the stars are not active which is an important charactistic to consider for RV follow-up (Moutou et al., 2017) and is one that is not taken into account in Eqs. 9 and 10.

5.3.2 Atmospheric characterization

TESS will provide many of the best transiting exoplanets for atmospheric characterization in the near future. Kempton et al. (2018) presented a framework to prioritize transiting planets for either transmission spectroscopy or emission spectroscopy observations with dedicated missions like JWST and ARIEL. This framework consists of analytical metrics which quantify the expected S/N ratio of transmission and emission signals from planetary atmospheres.

The transmission spectroscopy metric from Kempton et al. (2018) is

[TABLE]

The TSM represents the expected S/N of 10 hour observing programs with JWST/NIRISS assuming fixed atmospheric compositions for different planet types, cloud-free atmospheres, and a deterministic planet mass-radius relation. The planetary equilibrium temperature in Eq. 12 is $T_{\text{eq}}=T_{\text{eff}}\sqrt{R_{s}/2a}$ where $a$ is the planet’s semimajor axis and is calculated assuming zero albedo and full heat redistribution over the planetary surface. The scale factor $f$ is used to make the TSM non-dimensional and is used to correct discrepancies between the analytical TSM and the detailed simulations from Louie et al. (2018) using the NIRISS simulator for Single Object Slitless Spectroscopy (SOSS) observations. Values of $f$ are reported in Kempton et al. (2018) each of four planet types separately: terrestrials ($r_{p}<1.5$ R*⊕*), super-Earths ($1.5<r_{p}/\text{R}_{\oplus}<2.75$) sub-Neptunes ($2.75<r_{p}/\text{R}_{\oplus}<4$), and giants ($4<r_{p}/\text{R}_{\oplus}<10$). We calculate the TSM for our 16 candidates and report those values in Table 7 and in Fig. 14. Kempton et al. (2018) highly recommends planets for atmospheric characterization (and a-priori RV characterization) based on their TSM values relative to their derived cutoffs (see their Table 1). Five out of sixteen candidates exceed the threshold TSM cutoff as highlighted in Table 7 and should be considered for confirmation as they represent highly attractive targets for transmission spectroscopy observations. We note however that host stars with $J\lesssim 8.1$ start to approach the NIRISS bright limit (depending on its spectral type) and will require specialized fast readout modes and/or the use of detector subarrays to be observed with NIRISS in its SOSS mode (Beichman et al., 2014).

Similarly to the TSM, Kempton et al. (2018) defined the thermal emission spectroscopy metric as

[TABLE]

where $B_{7.5}(T)$ is the Planck function of spectral irradiance evaluated for a given temperature $T$ at 7.5 $\mu$m and $T_{\text{day}}$ is the planet’s day-side temperature assumed to be $1.1T_{\text{eq}}$. The constant $c=4.29\times 10^{6}$ is used to scale the ESM to yield the S/N of the reference planet GJ 1132b (Berta-Thompson et al., 2015; Dittmann et al., 2017a) in the center of the MIRI low resolution spectroscopy bandpass at 7.5 $\mu$m. We calculate the ESM for our 16 candidates and report those values in Table 7 and in Fig. 14. (Kempton et al., 2018) advocate that planets with ESM $>$ ESM${}_{\text{GJ 1132}}=7.5$ should be considered favorable targets for thermal emission spectroscopy observations with MIRI. Within our candidate sample, 2/16 candidates exceed this threshold ESM and represent some of the best targets to-date for the characterization of a terrestrial or super-Earth’s atmosphere with emission spectroscopy for the first time.

5.4 Comparison to yield simulations

Ballard (2018) performed a set of yield simulations focusing on M dwarfs not unlike the stellar population considered in this study. Ballard (2018) derive an ensemble completeness function for M dwarfs observed by TESS based on the simulated TESS yield from Sullivan et al. (2015) which includes details of the TESS footprint, systematics, the photometric error budget, and FP likelihoods. The expected TESS yield around M dwarfs is then derived by applying the completeness as a function of $P$ and $r_{p}$ to M dwarf planet occurrence rates. Said occurrence rate are derived from Dressing & Charbonneau (2015) and corrected for the eccentricity distribution (Limbach & Turner, 2015), dynamical stability (Fabrycky et al., 2012), and multiplicity effects according to the ‘Kepler dichotomy’ (Ballard & Johnson, 2016) of M dwarf planet populations: either high multiplicity systems ($N>5$) with low mutual inclinations or systems with lower multiplicity ($N\sim 1-2$) and high mutual inclinations. The resulting TESS yield around M1-M4 dwarfs is predicted to be $990\pm 350$ planets.

The following back-of-the-envelope calculation reveals how the expected M dwarf planet population to be discovered with TESS compares to our ORION results from the first two TESS sectors. First we note that the Sullivan et al. (2015) stellar population is a synthetic one derived from TRILEGAL galaxy model (Girardi et al., 2005). It contains 200,000 stars targeted by TESS which is effectively the size of the TIC (Stassun et al., 2018). Using the stellar parameters from the TIC-7 we find 53204 TIC M1-M4 dwarfs ($T_{\text{eff}}$ $\in[3200-3700]$ K; Pecaut & Mamajek 2013). Of these, 1624 and 1869 are targeted in sectors 1 and 2 respectively with 2849 being unique TICs. However unlike in the TIC-7, our stellar sample is derived using parallaxes from the GAIA DR2 which results in a distinct population of just 1149 M1-M4 dwarfs observed in either or both of TESS sectors 1 and 2. We use a simple correction factor to the expected TESS yield to account for the fractionally fewer M1-M4 stars that we target for transit searches compared to the TIC-7. This factor is $f=1149/2849=0.40\pm 0.01$ where the $f$ uncertainty is propagated from Poisson statistics.

The predicted number of M dwarf TESS planets discovered in sectors 1 and 2 is $990w\sim 53\pm 19$ where $w=2849/53204$ is the fraction of all M1-M4 dwarfs targeted in those sectors. Correcting this expected number of planets from the first two TESS sectors by the $f$ times fewer M1-M4 dwarfs in our stellar sample compared to in the TIC-7, we find that we are expected to detect $\sim 21\pm 8$ planets. Note that this calculation inherently assumes that the detection completeness of ORION is equivalent to the ensemble completeness derived in Ballard (2018).

If we include putative planet candidates, then recall that the total number of PCs detected around M1-M4 dwarfs in this study is sixteen (see Fig. 13). If each of these candidates become validated planets that this many planets would consistent with the expected number of M dwarf planet detections from Ballard (2018). Although, our yield is somewhat on the lower end of what is expected. If the TESS alert TOIs around M1-M4 dwarfs that remain undetected by ORION in this study are included, then TESS has discovered 19 M1-M4 PCs in its first two sectors. The consistency between the ORION yield and the expected TESS yield speaks highly to TESS’s overall performance compared to its expected completeness from Sullivan et al. (2015) and Ballard (2018), as well as to the outstanding performance that the TESS mission has already achieved so early-on in its lifetime.

Acknowledgements

I would like to thank each of Jo Bovy, René Doyon, Xu (Chelsea) Huang, David Latham, Kristen Menou, Norm Murray, Adiv Paradise, and Marten van Kerkwijk for both motivating discussions as well as informative ones that improved the content of this paper as well as my understanding of the data from GAIA and TESS used throughout. I also thank the Canadian Institute for Theoretical Astrophysics for use of the Sunnyvale computing cluster throughout this work. I would also like to acknowledge that this work was performed on land traditionally inhabited by the Wendat, the Anishnaabeg, Haudenosaunee, Metis, and the Mississaugas of the New Credit First Nation.

To my wife, for preparing the majority of our meals since the TESS data release and for turning the volume down on the TV during evening Leaf games so that I could hear myself think.

This work is supported in-part by the Natural Sciences and Engineering Council of Canada (NSERC).

Special thanks go to NASA and to the entire TESS team for their exhaustive efforts over the years in developing such an amazing observatory that will serve the exoplanet community and dreamers alike for years to come.

This research has made use of the NASA Exoplanet Archive, which is operated by the California Institute of Technology, under contract with the National Aeronautics and Space Administration under the Exoplanet Exploration Program.

This work presents results from the European Space Agency (ESA) space mission GAIA. GAIA data are being processed by the Gaia Data Processing and Analysis Consortium (DPAC). Funding for the DPAC is provided by national institutions, in particular the institutions participating in the Gaia MultiLateral Agreement (MLA). The Gaia mission website is https://www.cosmos.esa.int/gaia. The Gaia archive website is https://archives.esac.esa.int/gaia.

This paper includes data collected by the TESS mission, which are publicly available from the Mikulski Archive for Space Telescopes (MAST). Some of the data presented in this paper were obtained from the Mikulski Archive for Space Telescopes (MAST). STScI is operated by the Association of Universities for Research in Astronomy, Inc., under NASA contract NAS5-26555.