Context. Ultra-hot Jupiters have dayside temperatures at which most molecules are expected to thermally dissociate. The dissociation of water vapour results in the production of the hydroxyl radical (OH). While OH absorption is easily observed in near-infrared spectra of M dwarfs, which have similar effective temperatures as ultra-hot Jupiters, it is often not considered when studying the atmospheres of ultra-hot Jupiters. Ground-based high-resolution spectroscopy during the primary transit is a powerful tool for detecting molecular absorption in these planets. Aims. We aim to assess the presence and detectability of OH in the atmosphere of the ultra-hot Jupiter WASP-76b. Methods. We use high-resolution spectroscopic observations of a transit of WASP-76b obtained using CARMENES. After validating the OH line list, we generate model transit spectra of WASP-76b with petitRADTRANS. The data are corrected for stellar and telluric contamination and cross-correlated with the model spectra. After combining all cross-correlation functions from the transit, a detection map is constructed. Constraints on the planet properties from the OH absorption are obtained from a Markov chain Monte Carlo analysis. Results. OH is detected in the atmosphere of WASP-76b with a peak signal-to-noise ratio of 6.1. From the retrieval we obtain Kp = 232 ± 12 km s−1 and a blueshift of − 13.2 ± 1.6 km s−1, which are offset from the expected velocities. Considering the fast spin rotation of the planet, the blueshift is best explained with the signal predominantly originating from the evening terminator and the presence of a strong dayside-to-nightside wind. The increased Kp over its expected value (196.5 km s−1) is, however, a bit puzzling. The signal is found to be broad, with a full width at half maximum of 16.8−4.0+4.6 km s−1. The retrieval results in a weak constraint on the mean temperature of 2700–3700 K at the pressure range of the OH signal. Conclusions. We show that OH is readily observable in the transit spectra of ultra-hot Jupiters. Studying this molecule can provide insights into the molecular dissociation processes in the atmospheres of such planets.

No abstract

Fourier-based wavefront sensors, such as the Pyramid Wavefront Sensor (PWFS), are the current preference for high contrast imaging due to their high sensitivity. However, these wavefront sensors have intrinsic nonlinearities that constrain the range where conventional linear reconstruction methods can be used to accurately estimate the incoming wavefront aberrations. We propose to use Convolutional Neural Networks (CNNs) for the nonlinear reconstruction of the wavefront sensor measurements. It is demonstrated that a CNN can be used to accurately reconstruct the nonlinearities in both simulations and a lab implementation. We show that solely using a CNN for the reconstruction leads to suboptimal closed loop performance under simulated atmospheric turbulence. However, it is demonstrated that using a CNN to estimate the nonlinear error term on top of a linear model results in an improved effective dynamic range of a simulated adaptive optics system. The larger effective dynamic range results in a higher Strehl ratio under conditions where the nonlinear error is relevant. This will allow the current and future generation of large astronomical telescopes to work in a wider range of atmospheric conditions and therefore reduce costly downtime of such facilities.

Extreme temperature contrasts between the day and nightside of ultra-hot Jupiters result in significantly asymmetric atmospheres, with a large expansion occurring over a small range of longitude around the terminator. Over the course of a transit, WASP-76 b rotates by about 30°, changing the observable part of the atmosphere and invoking variations in the appearance of its constituents. Specifically, during the latter part of the transit, the planet’s trailing limb probes an increasing portion of its inflated dayside, which has a higher atmospheric detectability in transmission. As recently reported, this results in time-variable effects in the neutral iron signal, which are amplified by its possible condensation on the nightside. Here, we study the presence of molecular signals during a transit of WASP-76 b observed with the CARMENES spectrograph and compare the contributions from this planet’s morning and evening terminators. The results are somewhat puzzling, with formal detections of water vapor (5.5σ) and hydrogen cyanide (5.2σ) but at significantly different positions in the KP−Vsys diagram, with a blueshift of −14.3 ± 2.6 km s−1 and a redshift of +20.8−3.9+7.8 km s−1 respectively, and a higher KP than expected. The H2O signal also appears stronger later on in the transit, in contrast to that of HCN, which seems stronger early on. We tentatively explain this by silicate clouds forming and raining out on the nightside of the planet, partially removing oxygen from the upper atmosphere. For atmospheric C/O values between 0.7 and 1, this leads to the formation of HCN at the planet’s morning limb. At the evening terminator, with the sequestered oxygen being returned to the gas phase due to evaporation, these C/O values lead to formation of H2O instead of HCN. Overall, if confirmed, these observations indicate that individual molecules trace different parts of the planet atmosphere, as well as nightside condensation, allowing spatial characterization. As these results are based on a single transit observation, we advocate that more data are needed to confirm these results and further explore these scenarios.

No abstract

Current and future high-contrast imaging instruments require extreme Adaptive Optics (XAO) systems to reach contrasts necessary to directly image exoplanets. Telescope vibrations and the temporal error induced by the latency of the control loop limit the performance of these systems. Optimization of the (predictive) control algorithm is crucial in reducing these effects. We describe how model-free Reinforcement Learning can be used to optimize a Recurrent Neural Network controller for closed-loop adaptive optics control. We verify our proposed approach for tip-tilt control in simulations and a lab setup. The results show that this algorithm can effectively learn to suppress a combination of tip-tilt vibrations. Furthermore, we report decreased residuals for power-law input turbulence compared to an optimal gain integrator. Finally, we demonstrate that the controller can learn to identify the parameters of a varying vibration without requiring online updating of the control law. We conclude that Reinforcement Learning is a promising approach towards data-driven predictive control; future research will apply this approach to the control of high-order deformable mirrors.

Ground-based high-resolution spectroscopy (HRS) has detected numerous chemical species and atmospheric dynamics in exoplanets, most notably ultrahot Jupiters (UHJs). However, quantitative estimates on abundances have been challenging but are essential for accurate comparative characterization and to determine formation scenarios. In this work, we retrieve the atmospheres of six UHJs (WASP-76 b, MASCARA-4 b, MASCARA-2 b, WASP-121 b, HAT-P-70 b, and WASP-189 b) with ESPRESSO and HARPS-N/HARPS observations, exploring trends in eleven neutral species and dynamics. While Fe abundances agree well with stellar values, Mg, Ni, Cr, Mn, and V show more variation, highlighting the difficulty in using a single species as a proxy for metallicity. We find that Ca, Na, Ti, and TiO are underabundant, potentially due to ionization and/or nightside rain-out. Our retrievals also show that relative abundances between species are more robust, consistent with previous works. We perform spatially resolved and phase-resolved retrievals for WASP-76 b and WASP-121 b given their high signal-to-noise observations, and find the chemical abundances in each of the terminator regions are broadly consistent. We additionally constrain dynamics for our sample through Doppler shifts and broadening of the planetary signals during the primary eclipse, with median blueshifts between ∼0.9 and 9.0 km s−1 due to day–night winds. Furthermore, we constrain spectroscopic masses for MASCARA-2 b and HAT-P-70 b consistent with their known upper limits, but we note that these may be biased due to degeneracies. This work highlights the importance of future HRS studies to further probe differences and trends between exoplanets.

acific Gas & Electric Company bution system to serve its large, d e n s e loads in downtown San Francisco and Oakland. The net-is noncondarctive, li Vault temperature transformers, most of which are I. Three-phase secondary curin underground vaults. Each of electromagnetic and rent these transformer vaults contains rudio inter fe ren ce l i : Transformer sudden presone or more transformers. A net-information monitored includes: uses a network distri-Fiberoptics does not IB Protector status sg Protector heat sensor ;~i Transformer temperature s: Vault water level Pw&E) require trenching, and is immune to work systems include over 1,100 sure relay.work system is designed for maximum service reliability, maintaining power to customers even if one of its primary feeders or transformers is down.The network primary feeders are 34.5 or 12 kilovolts. The secondary services are 208Y/l20V, 48OY/277V and 2,4OOY/4,16OV. Most of the spot networks are 48OY/277V due to the high loads and economics of large high rise buildings. The higher voltages result in significant reductions in building wiring costs and the cost of supplying power to the buildings.In 1983, PG&E began its network improvement program to improve the safety and operation of its network distribution system. Low current arcing faults on 480Y/277V secondaries could sustain themselves for extended periods of time and cause extensive damage, more so than at 208 volts. The greater potential allows the arc t o restrike and sustain a fault, instead of just burning in the clear.The program had three objectives:x Prevent total network shutdown from secondary fault D Prevent a catastrophic failure in a building vault z Minimize damage to customer and PC&E facilities. SCADA System RequirementsThe supervisory control and data acquisition (SCADA) system would monitor 500 remote terminal units (RTUs) in San Francisco and 150 RTUs in Oakland. The ' H&L Instruments, Burlingame. California Pacific Gas &Electric Company. San Francisco, California Alarms would be triggered by any abnormal conditions so that corrective action could be promptly taken to avoid or minimize damage.It was important that data be received in a predictable time and that positive, real-time, two-way communications be assured. The communication network had to be self-monitoring. If it failed, the operator needed to be notified immediately so that corrective action could be taken. All means of communication between RTUs and the SCADA master computers were considered.