Deep Learning Exoplanets Detection by Combining Real and Synthetic Data

Scientists and astronomers have attached Scientists and astronomers have attached great importance to the task of discovering new exoplanets, even more so if they are in the habitable zone. To date, more than 4300 exoplanets have been confirmed by NASA, using various discovery techniques, including planetary transits, in addition to the use of various databases provided by space and ground-based telescopes. This article proposes the development of a deep learning system for detecting planetary transits in Kepler Telescope lightcurves. The approach is based on related work from the literature and enhanced to validation with real lightcurves. A CNN classification model is trained from a mixture of real and synthetic data, and validated only with real data and different from those used in the training stage. The best ratio of synthetic data is determined by the perform of an optimisation technique and a sensitivity analysis. The precision, accuracy and true positive rate of the best model obtained are determined and compared with other similar works. The results demonstrate that the use of synthetic data on the training stage can improve the transit detection performance on real light curves.

Color Dependence of Planetary Transit Depths due to Large Starspots and Dust Clouds: Application to PTFO 8-8695

Although thousands of exoplanets have now been discovered, there is still a significant lack of observations of young planets only a few Myr old. Thus there is little direct evidence available to differentiate between various models of planet formation. The detection of planets of this age would provide much-needed data that could help constrain the planet formation process. To explore what transit observations of such planets may look like, we model the effects of large starspots and dust clouds on the depths of exoplanet transits across multiple wavelengths. We apply this model to the candidate planet PTFO 8-8695b, whose depths vary significantly across optical and infrared wavelengths. Our model shows that, while large starspots can significantly increase the color dependence of planetary transits, a combination of starspots and a large cloud surrounding the planet is required to reproduce the observed transit depths across four wavelengths.

Starspot modelling of the TESS light curve of CVSO 30

Aims. I aim to investigate whether the photometric variability in the candidate host star CVSO 30 can be explained by starspots. Methods. The Transiting Exoplanet Survey Satellite (TESS) light curve of CVSO 30 is separated into two independent non-sinusoidal periodic components. A starspot modelling technique is applied to each of these components. Results. Combined, the two model light curves reproduce the TESS observations to a high accuracy, obviating the need to invoke planetary transits to describe part of the variability.

More planetary candidates from K2 Campaign 5 using TRAN_K2

Context. The exquisite precision of space-based photometric surveys and the unavoidable presence of instrumental systematics and intrinsic stellar variability call for the development of sophisticated methods that distinguish these signal components from those caused by planetary transits. Aims. Here, we introduce the standalone Fortran code TRAN_K2 to search for planetary transits under the colored noise of stellar variability and instrumental effects. We use this code to perform a survey to uncover new candidates. Methods. Stellar variability is represented by a Fourier series and, when necessary, by an autoregressive model aimed at avoiding excessive Gibbs overshoots at the edges. For the treatment of systematics, a cotrending and an external parameter decorrelation were employed by using cotrending stars with low stellar variability as well as the chip position and the background flux level at the target. The filtering was done within the framework of the standard weighted least squares, where the weights are determined iteratively, to allow a robust fit and to separate the transit signal from stellar variability and systematics. Once the periods of the transit components are determined from the filtered data by the box-fitting least squares method, we reconstruct the full signal and determine the transit parameters with a higher accuracy. This step greatly reduces the excessive attenuation of the transit depths and minimizes shape deformation. Results. We tested the code on the field of Campaign 5 of the K2 mission. We detected 98% of the systems with all their candidate planets as previously reported by other authors. We then surveyed the whole field and discovered 15 new systems. An additional three planets were found in three multiplanetary systems, and two more planets were found in a previously known single-planet system.

Atmospheric Characterization via Broadband Color Filters on the PLAnetary Transits and Oscillations of stars (PLATO) Mission

<p>We assess broadband color filters for the two fast cameras on the PLAnetary Transits and Oscillations (PLATO) of stars space mission with respect to exoplanetary atmospheric characterization. We focus on Ultra Hot Jupiters and Hot Jupiters placed 25pc and 100pc away from the Earth and warm Super-Earths placed 10pc and 25pc away. Our analysis takes as input literature values for the difference in transit depth between the broadband lower (500-675nm) wavelength interval (hereafter referred to as ”blue“) and the upper (675-1125nm) broadband wavelength interval (hereafter referred to as ”red“) for transmission, occultation and phase curve analyses. Planets orbiting main sequence central stars with stellar classes F, G, K and M are investigated. We calculate the signal-to-noise ratio with respect to photon and instrument noise for detecting the difference in transit depth between the two spectral intervals. Results suggest that bulk atmospheric composition and planetary geometric albedos could be detected for (Ultra) Hot Jupiters up to ~100pc (~25pc) with strong (moderate) Rayleigh extinction. Phase curve information could be extracted for Ultra Hot Jupiters orbiting K and G dwarf stars up to 25pc away. For warm Super-Earths, basic atmospheric types (primary and water-dominated) and the presence of sub-micron hazes in the upper atmosphere could be distinguished for up to a handful of cases up to ~10pc (manuscript accepted in Experimental Astronomy).</p>