Periodic Reporting for period 2 - EXACT (Exoplanet Adaptive Characterization with the ELT)
Periodo di rendicontazione: 2022-03-01 al 2023-08-31
Until the 1990's, we did not know of any planet around another star. Even though there was little doubt that they existed, we has no information at all about them. Were they different from the ones in our solar system? Were they rare? Could life as we know it exist elsewhere? It was not yet possible to answer these questions through astronomical observations. That did not prevent society from imagining what these worlds could look like, and the interest for the general public for these questions is neither small, nor recent.
This has changed radically in the space of 30years. We have now detected more than 5000 planets. Many of these detections come with a measurement of the planets 'radius and mass - and thus of their density - and of the distance between the planet and its star. This was enough to flag some of these planets as potentially habitable worlds, as they might have liquid water on their surface.
A key piece of the puzzle is missing: the composition and properties of their atmosphere - if they have one. With this data, the claim that a planet is indeed habitable (or not) could be significantly strengthened. In addition, it would help comparing planets between them, and to compare them with the ones from our solar system. Some limited spectroscopic information can be obtained with relatively small telescopes for a very small fraction of the planets. If the telescope is large enough, however, and if the atmospheric turbulence can be corrected well enough, the planet can potentially be observed next to the star, and light coming from the planet can be analyzed directly, which greatly increases the efficiency of its spectroscopic characterization.
Our most advanced instruments (commissioned in 2014-2015) have been designed to observe planets in this fashion. They do suffer from a very limited spectral resolution, however, and this limits their ability to study exoplanets atmosphere. Fortunately, recent projects aim at connecting them with high-resolution spectrometers. Besides, the next generation of extremely large telescopes (ELT, 4-5 times larger than our largest telescopes) is being developed at the moment, together with instruments that will be capable of both high-angular resolution, and high-spectral resolution. Up until now, only Jupiter-like planets have been imaged. In the 2030's and 2040's, the ELT may observe smaller planets, including rocky planets, potentially similar to the Earth, Venus, or Mars. Doing so will require a more accurate correction of the atmospheric turbulence, and other optical aberrations, in addition to more efficient instruments: few photons will be received, and the instrument's transmission should be as high as possible!
The project adresses these objectives through three different directions:
1- push the planet characterization capability of ELT/HARMONI, the first-light, visible and near-infrared spectro-imager of the ELT. We want to make the best use of HARMONI when it is offered to the community in ~2030.
2- prepare the ELT 2nd generation instruments to make them capable of characterizing rocky planets. We need to develop robust techniques to spot planets 100 times fainter than those that we observe today, while loosing as few photons as possible.
3- demonstrate the higher efficiency of novel high-resolution spectroscopy techniques. Current high-res spectrometers usually loose up to 80-90% of the light!
Each of these objectives would push forward our capability to understand how diverse exoplanets can be, how they form, and to what extent some of them could be habitable.
It will also be a necessary step towards searching for extraterrestrial biological activity.
This would address two major philosophical questions of our time: are we alone, and where do we come from?
This project will also lead to the development of new hardware and software techniques which could find applications in society. Advanced amplitude and phase control of light using novel mechanical-electronical systems could be used in the telecommunication industry, for instance, and high-resolution spectral analysis techniques could find an application in environmental monitoring, to look and quantify greenhouse gas emissions.
This has changed radically in the space of 30years. We have now detected more than 5000 planets. Many of these detections come with a measurement of the planets 'radius and mass - and thus of their density - and of the distance between the planet and its star. This was enough to flag some of these planets as potentially habitable worlds, as they might have liquid water on their surface.
A key piece of the puzzle is missing: the composition and properties of their atmosphere - if they have one. With this data, the claim that a planet is indeed habitable (or not) could be significantly strengthened. In addition, it would help comparing planets between them, and to compare them with the ones from our solar system. Some limited spectroscopic information can be obtained with relatively small telescopes for a very small fraction of the planets. If the telescope is large enough, however, and if the atmospheric turbulence can be corrected well enough, the planet can potentially be observed next to the star, and light coming from the planet can be analyzed directly, which greatly increases the efficiency of its spectroscopic characterization.
Our most advanced instruments (commissioned in 2014-2015) have been designed to observe planets in this fashion. They do suffer from a very limited spectral resolution, however, and this limits their ability to study exoplanets atmosphere. Fortunately, recent projects aim at connecting them with high-resolution spectrometers. Besides, the next generation of extremely large telescopes (ELT, 4-5 times larger than our largest telescopes) is being developed at the moment, together with instruments that will be capable of both high-angular resolution, and high-spectral resolution. Up until now, only Jupiter-like planets have been imaged. In the 2030's and 2040's, the ELT may observe smaller planets, including rocky planets, potentially similar to the Earth, Venus, or Mars. Doing so will require a more accurate correction of the atmospheric turbulence, and other optical aberrations, in addition to more efficient instruments: few photons will be received, and the instrument's transmission should be as high as possible!
The project adresses these objectives through three different directions:
1- push the planet characterization capability of ELT/HARMONI, the first-light, visible and near-infrared spectro-imager of the ELT. We want to make the best use of HARMONI when it is offered to the community in ~2030.
2- prepare the ELT 2nd generation instruments to make them capable of characterizing rocky planets. We need to develop robust techniques to spot planets 100 times fainter than those that we observe today, while loosing as few photons as possible.
3- demonstrate the higher efficiency of novel high-resolution spectroscopy techniques. Current high-res spectrometers usually loose up to 80-90% of the light!
Each of these objectives would push forward our capability to understand how diverse exoplanets can be, how they form, and to what extent some of them could be habitable.
It will also be a necessary step towards searching for extraterrestrial biological activity.
This would address two major philosophical questions of our time: are we alone, and where do we come from?
This project will also lead to the development of new hardware and software techniques which could find applications in society. Advanced amplitude and phase control of light using novel mechanical-electronical systems could be used in the telecommunication industry, for instance, and high-resolution spectral analysis techniques could find an application in environmental monitoring, to look and quantify greenhouse gas emissions.
The three objectives listed above have been addressed in parallel since the beginning of the project, and this has lead to the publication of several scientific articles, and the organization of a workshop.
2 PhD, 1 postdoc, and one engineer have been hired to work on the project. I am explicitly referring to them in this summary.
1 - Exhaustive work has been performed to push the performance of ELT/HARMONI.
1.1 - an optical experiment has been improved and used to reproduce in the laboratory the capability of HARMONI to measure and correct the small, quasi-static optical aberrations that could ultimately limit its performance. It was in particular used to demonstrate the ability to keep the residual error small enough in spite of residual atmospheric dispersion, and atmospheric turbulence. PhD student A. Hours has worked on this part of the project, in collaboration with colleagues at LAM, and in other institutes that are part of the HARMONI consortium.
1.2 - FastCurves, a semi-analytical noise propagation model has been developed by PhD student A. Bidot to estimate the limitation to the spectral analysis of the star and planet signals that will be obtained with HARMONI. This work now enables to estimate the capability of HARMONI to detect and characterize planets in various cases of star magnitude, planet differential magnitude, type, and separation, integration time, etc, and for the various observation modes of HARMONI. All of this could not be done with the classical end-to-end approach, and this type of model can be adapted to many other instruments. Early results indicate that HARMONI could study planets up to 10 times fainter than the ones that have already been observed. The FastCurves package is available online, and it has been actively shared with the community.
1.3 - a second optical experiment has been designed, and partially installed to reproduce in the laboratory the images that ELT/HARMONI will provide, so as to test the data processing algorithms that will later be used on the real data. The installation wad not completed at the end of the period covered by this report, and no results is therefore reported here, apart that no showstoppers have been encountered.
2 - Robust, adaptative amplitude control of a light beam has been tested in the context of the preparation of ELT 2nd generation instrument.
2.1 - a third optical experiment has been designed, installed, and used to successfully test the capability of a commercial micro-mirror array to spatially mask out part of a telescope beam, together with a wavefront control system. This work was performed by PostDoc L. Leboulleux (now a permanent staff at IPAG). This involved the development of a control procedure, which was done by software engineer S. Curaba. The results of this experiment indicate that this device - which was not designed for this application - can indeed be used to adaptively mask the pupil to change the point-spread function of the telescope to enable the observation of planets next to the star. The main limitation that was anticipated is its too high chromaticity, and this was confirmed in our results.
2.2 - PostDoc L. Leboulleux has developed a novel way to design components robust to two different types of severe optical aberrations: co-phasing errors of segmented telescopes, and differential piston errors encountered in low-wind conditions. Her results indicate that these aberrations, while their amplitude should be kept as low as possible, can some extent be countered by her design. Without it, instruments designed to characterize planets can see their performance severely limited.
2.3 - we have identify an alternative type of micro-mirror array, which was recently developed at the Fraunhofer Institute in Dresden, which should be capable of achromatic amplitude control (and possibly phase control as well). More work will be necessary to confirm this.
3 - A first demonstration of a high transmissive, compact, high-resolution spectrometer has been performed, as a first step towards its potential use as a visitor instrument.
3.1 - The VIPA spectrometer developed at IPAG has been calibrated and demonstrated at the Palomar Observatory using the 5m Hale telescope in March and April 2022. This was a collective effort of my team, and specifically involved the work of PhD A. Bidot, and software engineer S. Curaba. This demonstration was delayed about for about 1.5y because of the COVID-19 pandemic, which shut down observatories worldwide for months, and limited their access for an additional year. The test confirmed the data obtained in the lab (spectral resolution, transmission). Moreover, it was especially useful to check that the operation of the spectrometer was quite easy. Its optical interface with the rest of the observatory consists in two optical fibers, which simply need to be connected to the spectrometer. Apart from a few hurdles due to the observatory itself, the optical fine alignment, and the overall setup of the spectrometer took about 10days. Observations were performed over 3 nights, including some hours during which data was obtained with both the VIPA spectrometer, and PARVI, a local spectrometer observing at the same wavelengths.
3.2 - The first demonstration used an engineering grade detector provided on a temporary basis by U. de Montréal. A new scientific grade detector has since been purchased by the project, and modifications have been made to the mechanical structure, and the electronics of the spectrometer to enable its installation. Fibered gas cells have been purchased, and tested in the laboratory. They will be used to perform a fine wavelength calibration in the 2-2.5um spectral range (calibration in the 1.5-1.7um range is performed thanks to a tunable laser also purchased by the project).
This work has lead to several peer-reviewed publications, some of which pertaining to the period covered by this report) are currently under review. This work has also been presented in several international conferences (Spirit of Lyot, SPIE, AO4ELT-7) for which conference proceedings have been written.
2 PhD, 1 postdoc, and one engineer have been hired to work on the project. I am explicitly referring to them in this summary.
1 - Exhaustive work has been performed to push the performance of ELT/HARMONI.
1.1 - an optical experiment has been improved and used to reproduce in the laboratory the capability of HARMONI to measure and correct the small, quasi-static optical aberrations that could ultimately limit its performance. It was in particular used to demonstrate the ability to keep the residual error small enough in spite of residual atmospheric dispersion, and atmospheric turbulence. PhD student A. Hours has worked on this part of the project, in collaboration with colleagues at LAM, and in other institutes that are part of the HARMONI consortium.
1.2 - FastCurves, a semi-analytical noise propagation model has been developed by PhD student A. Bidot to estimate the limitation to the spectral analysis of the star and planet signals that will be obtained with HARMONI. This work now enables to estimate the capability of HARMONI to detect and characterize planets in various cases of star magnitude, planet differential magnitude, type, and separation, integration time, etc, and for the various observation modes of HARMONI. All of this could not be done with the classical end-to-end approach, and this type of model can be adapted to many other instruments. Early results indicate that HARMONI could study planets up to 10 times fainter than the ones that have already been observed. The FastCurves package is available online, and it has been actively shared with the community.
1.3 - a second optical experiment has been designed, and partially installed to reproduce in the laboratory the images that ELT/HARMONI will provide, so as to test the data processing algorithms that will later be used on the real data. The installation wad not completed at the end of the period covered by this report, and no results is therefore reported here, apart that no showstoppers have been encountered.
2 - Robust, adaptative amplitude control of a light beam has been tested in the context of the preparation of ELT 2nd generation instrument.
2.1 - a third optical experiment has been designed, installed, and used to successfully test the capability of a commercial micro-mirror array to spatially mask out part of a telescope beam, together with a wavefront control system. This work was performed by PostDoc L. Leboulleux (now a permanent staff at IPAG). This involved the development of a control procedure, which was done by software engineer S. Curaba. The results of this experiment indicate that this device - which was not designed for this application - can indeed be used to adaptively mask the pupil to change the point-spread function of the telescope to enable the observation of planets next to the star. The main limitation that was anticipated is its too high chromaticity, and this was confirmed in our results.
2.2 - PostDoc L. Leboulleux has developed a novel way to design components robust to two different types of severe optical aberrations: co-phasing errors of segmented telescopes, and differential piston errors encountered in low-wind conditions. Her results indicate that these aberrations, while their amplitude should be kept as low as possible, can some extent be countered by her design. Without it, instruments designed to characterize planets can see their performance severely limited.
2.3 - we have identify an alternative type of micro-mirror array, which was recently developed at the Fraunhofer Institute in Dresden, which should be capable of achromatic amplitude control (and possibly phase control as well). More work will be necessary to confirm this.
3 - A first demonstration of a high transmissive, compact, high-resolution spectrometer has been performed, as a first step towards its potential use as a visitor instrument.
3.1 - The VIPA spectrometer developed at IPAG has been calibrated and demonstrated at the Palomar Observatory using the 5m Hale telescope in March and April 2022. This was a collective effort of my team, and specifically involved the work of PhD A. Bidot, and software engineer S. Curaba. This demonstration was delayed about for about 1.5y because of the COVID-19 pandemic, which shut down observatories worldwide for months, and limited their access for an additional year. The test confirmed the data obtained in the lab (spectral resolution, transmission). Moreover, it was especially useful to check that the operation of the spectrometer was quite easy. Its optical interface with the rest of the observatory consists in two optical fibers, which simply need to be connected to the spectrometer. Apart from a few hurdles due to the observatory itself, the optical fine alignment, and the overall setup of the spectrometer took about 10days. Observations were performed over 3 nights, including some hours during which data was obtained with both the VIPA spectrometer, and PARVI, a local spectrometer observing at the same wavelengths.
3.2 - The first demonstration used an engineering grade detector provided on a temporary basis by U. de Montréal. A new scientific grade detector has since been purchased by the project, and modifications have been made to the mechanical structure, and the electronics of the spectrometer to enable its installation. Fibered gas cells have been purchased, and tested in the laboratory. They will be used to perform a fine wavelength calibration in the 2-2.5um spectral range (calibration in the 1.5-1.7um range is performed thanks to a tunable laser also purchased by the project).
This work has lead to several peer-reviewed publications, some of which pertaining to the period covered by this report) are currently under review. This work has also been presented in several international conferences (Spirit of Lyot, SPIE, AO4ELT-7) for which conference proceedings have been written.
Together with my team, I will work towards completing the three major objectives of the project during the next few years.
I anticipate several next steps:
1 - Completing the estimation of ELT/HARMONI ultimate performance
1.1 - The FastCurves package that has been developed to predict the signal-to-noise ratio that one could obtain with HARMONI (as a function of various parameters) will be used by PhD student S. Martos to explore a vast parameter space, with the aim of determining which observing mode of HARMONI would be best for which planet, and to check whether HARMONI could realistically observe Neptune-like planets, or even some rocky planets, in addition to Jupiter-like planets.
1.2 - The optical experiment aimed at testing the ability of HARMONI to sense and correct optical aberrations will be used to test a new sensing mode designed to enable it to measure twice as large aberrations. This would potentially make it possible to properly sense large differential piston aberrations due to low-wind conditions, which is expected to happen with the telescope, and which can severely decrease the quality of the image.
1.3 - The second optical experiment will be used to reproduce the same type of data that HARMONI will acquire, in various observing conditions. The data will be analyzed using various post-processing techniques, and the results will be compared to what semi-analytical and numerical models have predicted to check the validity of underlining assumptions. This will further guide the estimation of ELT/HARMONI performances.
2 - Developing the hardware and software solutions for the ELT 2nd generation instruments
2.1 - The third optical experiment will continue to be used to demonstrate amplitude and phase control techniques, in particular with the aim of maximizing the signal-to-noise ratio of the planet light collected by a monomode fiber, by locally decreasing the amount of stellar light, or, alternatively, by shaping the beam to minimize the local coupling of the stellar light with the fiber fundamental mode.
2.2 - A prototype micro-mirror array from the Fraunhofer Institute will be developed over 2 years, and eventually tested in the laboratory to check its ability to perform amplitude control in a large spectral range, contrary to the current device, and, if applicable, to perform phase control as well. This is a typical example of a high risk / high gain scenario. This test will require a careful control of the device. On-sky testing of the device (with a telescope) is not currently planned, but this may change in the future.
3 - Using the VIPA spectrometer on-sky, for technical demonstrations, and as a visitor instrument
3.1 - The spectrometer will be tested at the Observatoire de Haute Provence (OHP) where a new adaptive optics system has recently been installed. This will both serve to check the performance of the instrument with its new scientific grade detector, and to develop novel wavefront control techniques to optimize the injection of the target light in the optical fiber that feeds the spectrometer. Tests will be performed in at least two phases, to first characterize on-sky objects such as bright stars, and, later, to characterize off-axis objects such as brown dwarves which will serve as analogues of planets given the small size of the telescope. Observations may be conducted first in a 1.5-1.75um spectral range (as it was done at Palomar Observatory), and later in a 2-2.5um spectral range.
3.2 - Assuming that the results obtained at OHP are satisfying, I will try to ship the VIPA spectrometer to a large telescope (either the Keck, Subaru, or VLT) to characterize the atmosphere of young giant planets. In particular, I envision a partnership with the HiRISE visitor instrument that connects VLT/SPHERE with the VLT/CRIRES spectrometer. VIPA would replace CRIRES, and should provide a 2-3 times higher transmission, allowing to study planets at least one magnitude fainter than it is currently possible, or, alternatively, to increase the signal-to-noise ratio of planets with the same magnitude.
I anticipate several next steps:
1 - Completing the estimation of ELT/HARMONI ultimate performance
1.1 - The FastCurves package that has been developed to predict the signal-to-noise ratio that one could obtain with HARMONI (as a function of various parameters) will be used by PhD student S. Martos to explore a vast parameter space, with the aim of determining which observing mode of HARMONI would be best for which planet, and to check whether HARMONI could realistically observe Neptune-like planets, or even some rocky planets, in addition to Jupiter-like planets.
1.2 - The optical experiment aimed at testing the ability of HARMONI to sense and correct optical aberrations will be used to test a new sensing mode designed to enable it to measure twice as large aberrations. This would potentially make it possible to properly sense large differential piston aberrations due to low-wind conditions, which is expected to happen with the telescope, and which can severely decrease the quality of the image.
1.3 - The second optical experiment will be used to reproduce the same type of data that HARMONI will acquire, in various observing conditions. The data will be analyzed using various post-processing techniques, and the results will be compared to what semi-analytical and numerical models have predicted to check the validity of underlining assumptions. This will further guide the estimation of ELT/HARMONI performances.
2 - Developing the hardware and software solutions for the ELT 2nd generation instruments
2.1 - The third optical experiment will continue to be used to demonstrate amplitude and phase control techniques, in particular with the aim of maximizing the signal-to-noise ratio of the planet light collected by a monomode fiber, by locally decreasing the amount of stellar light, or, alternatively, by shaping the beam to minimize the local coupling of the stellar light with the fiber fundamental mode.
2.2 - A prototype micro-mirror array from the Fraunhofer Institute will be developed over 2 years, and eventually tested in the laboratory to check its ability to perform amplitude control in a large spectral range, contrary to the current device, and, if applicable, to perform phase control as well. This is a typical example of a high risk / high gain scenario. This test will require a careful control of the device. On-sky testing of the device (with a telescope) is not currently planned, but this may change in the future.
3 - Using the VIPA spectrometer on-sky, for technical demonstrations, and as a visitor instrument
3.1 - The spectrometer will be tested at the Observatoire de Haute Provence (OHP) where a new adaptive optics system has recently been installed. This will both serve to check the performance of the instrument with its new scientific grade detector, and to develop novel wavefront control techniques to optimize the injection of the target light in the optical fiber that feeds the spectrometer. Tests will be performed in at least two phases, to first characterize on-sky objects such as bright stars, and, later, to characterize off-axis objects such as brown dwarves which will serve as analogues of planets given the small size of the telescope. Observations may be conducted first in a 1.5-1.75um spectral range (as it was done at Palomar Observatory), and later in a 2-2.5um spectral range.
3.2 - Assuming that the results obtained at OHP are satisfying, I will try to ship the VIPA spectrometer to a large telescope (either the Keck, Subaru, or VLT) to characterize the atmosphere of young giant planets. In particular, I envision a partnership with the HiRISE visitor instrument that connects VLT/SPHERE with the VLT/CRIRES spectrometer. VIPA would replace CRIRES, and should provide a 2-3 times higher transmission, allowing to study planets at least one magnitude fainter than it is currently possible, or, alternatively, to increase the signal-to-noise ratio of planets with the same magnitude.