Subscribe to the gold package and get unlimited access to Shamra Academy
Register a new userRecovery of TESS Stellar Rotation Periods Using Deep Learning
100
0
0.0
(
0
)
Ask ChatGPT about the research
No Arabic abstract
We used a convolutional neural network to infer stellar rotation periods from a set of synthetic light curves simulated with realistic spot evolution patterns. We convolved these simulated light curves with real TESS light curves containing minimal intrinsic astrophysical variability to allow the network to learn TESS systematics and estimate rotation periods despite them. In addition to periods, we predict uncertainties via heteroskedastic regression to estimate the credibility of the period predictions. In the most credible half of the test data, we recover 10%-accurate periods for 46% of the targets, and 20%-accurate periods for 69% of the targets. Using our trained network, we successfully recover periods of real stars with literature rotation measurements, even past the 13.7-day limit generally encountered by TESS rotation searches using conventional period-finding techniques. Our method also demonstrates resistance to half-period aliases. We present the neural network and simulated training data, and introduce the software butterpy used to synthesize the light curves using realistic star spot evolution.
rate research
Read More
Stellar RV jitter due to surface activity may bias the RV semi-amplitude and mass of rocky planets. The amplitude of the jitter may be estimated from the uncertainty in the rotation period, allowing the mass to be more accurately obtained. We find candidate rotation periods for 17 out of 35 TESS Objects of Interest (TOI) hosting <3 R_Earth planets as part of the Magellan-TESS Survey, which is the first-ever statistically robust study of exoplanet masses and radii across the photo-evaporation gap. Seven periods are 3+ sigma detections, two are 1.5+ sigma, and 8 show plausible variability but the periods remain unconfirmed. The other 18 TOIs are non-detections. Candidate rotators include the host stars of the confirmed planets L 168-9 b, the HD 21749 system, LTT 1445 A b, TOI 1062 b, and the L 98-59 system. 13 candidates have no counterpart in the 1000 TOI rotation catalog of Canto Martins et al. (2020). We find periods for G3-M3 dwarfs using combined light curves from TESS and the Evryscope all-sky array of small telescopes, sometimes with longer periods than would be possible with TESS alone. Secure periods range from 1.4 to 26 d with Evryscope-measured photometric amplitudes as small as 2.1 mmag in g. We also apply Monte Carlo sampling and a Gaussian Process stellar activity model from the code exoplanet to the TESS light curves of 6 TOIs to confirm the Evryscope periods.
We measure rotation periods for 12151 stars in the Kepler field, based on the photometric variability caused by stellar activity. Our analysis returns stable rotation periods over at least six out of eight quarters of Kepler data. This large sample of stars enables us to study the rotation periods as a function of spectral type. We find good agreement with previous studies and vsini measurements for F, G and K stars. Combining rotation periods, B-V color, and gyrochronology relations, we find that the cool stars in our sample are predominantly younger than ~1Gyr.
In order to understand stellar evolution, it is crucial to efficiently determine stellar surface rotation periods. An efficient tool to automatically determine reliable rotation periods is needed when dealing with large samples of stellar photometric datasets. The objective of this work is to develop such a tool. Random forest learning abilities are exploited to automate the extraction of rotation periods in Kepler light curves. Rotation periods and complementary parameters are obtained from three different methods: a wavelet analysis, the autocorrelation function of the light curve, and the composite spectrum. We train three different classifiers: one to detect if rotational modulations are present in the light curve, one to flag close binary or classical pulsators candidates that can bias our rotation period determination, and finally one classifier to provide the final rotation period. We test our machine learning pipeline on 23,431 stars of the Kepler K and M dwarf reference rotation catalog of Santos et al. (2019) for which 60% of the stars have been visually inspected. For the sample of 21,707 stars where all the input parameters are provided to the algorithm, 94.2% of them are correctly classified (as rotating or not). Among the stars that have a rotation period in the reference catalog, the machine learning provides a period that agrees within 10% of the reference value for 95.3% of the stars. Moreover, the yield of correct rotation periods is raised to 99.5% after visually inspecting 25.2% of the stars. Over the two main analysis steps, rotation classification and period selection, the pipeline yields a global agreement with the reference values of 92.1% and 96.9% before and after visual inspection. Random forest classifiers are efficient tools to determine reliable rotation periods in large samples of stars. [abridged]
Magnetic activity in stars manifests as dark spots on their surfaces that modulate the brightness observed by telescopes. These light curves contain important information on stellar rotation. However, the accurate estimation of rotation periods is computationally expensive due to scarce ground truth information, noisy data, and large parameter spaces that lead to degenerate solutions. We harness the power of deep learning and successfully apply Convolutional Neural Networks to regress stellar rotation periods from Kepler light curves. Geometry-preserving time-series to image transformations of the light curves serve as inputs to a ResNet-18 based architecture which is trained through transfer learning. The McQuillan catalog of published rotation periods is used as ansatz to groundtruth. We benchmark the performance of our method against a random forest regressor, a 1D CNN, and the Auto-Correlation Function (ACF) - the current standard to estimate rotation periods. Despite limiting our input to fewer data points (1k), our model yields more accurate results and runs 350 times faster than ACF runs on the same number of data points and 10,000 times faster than ACF runs on 65k data points. With only minimal feature engineering our approach has impressive accuracy, motivating the application of deep learning to regress stellar parameters on an even larger scale
We measure rotation periods and sinusoidal amplitudes in Evryscope light curves for 122 two-minute K5-M4 TESS targets selected for strong flaring. The Evryscope array of telescopes has observed all bright nearby stars in the South, producing two-minute cadence light curves since 2016. Long-term, high-cadence observations of rotating flare stars probe the complex relationship between stellar rotation, starspots, and superflares. We detect periods from 0.3487 to 104 d, and observe amplitudes from 0.008 to 0.216 g mag. We find the Evryscope amplitudes are larger than those in TESS with the effect correlated to stellar mass (p-value=0.01). We compute the Rossby number (Ro), and find our sample selected for flaring has twice as many intermediate rotators (0.04<Ro<0.4) as fast (Ro<0.04) or slow (Ro>0.44) rotators; this may be astrophysical or a result of period-detection sensitivity. We discover 30 fast, 59 intermediate, and 33 slow rotators. We measure a median starspot coverage of 13% of the stellar hemisphere and constrain the minimum magnetic field strength consistent with our flare energies and spot coverage to be 500 G, with later-type stars exhibiting lower values than earlier-types. We observe a possible change in superflare rates at intermediate periods. However, we do not conclusively confirm the increased activity of intermediate rotators seen in previous studies. We split all rotators at Ro~0.2 into Prot<10 d and Prot>10 d bins to confirm short-period rotators exhibit higher superflare rates, larger flare energies, and higher starspot coverage than do long-period rotators, at p-values of 3.2 X 10^-5, 1.0 X 10^-5, and 0.01, respectively.
Log in to be able to interact and post comments
comments
Fetching comments
Sign in to be able to follow your search criteria
