Inhalt:
Asteroids are involved in understanding several key issues in Solar System science and the space environment of our planet, such as the conditions of the Solar System during its formation, the delivery of water and organic molecules to Earth, the potential danger of NEA and their role in affecting Earth's climate.Stellar occultation events are a unique opportunity to obtain from the ground very accurate asteroid astrometry, close to the performance of Gaia, and shapes/sizes. When an asteroid hides the light of a star, the uncertainty of its instantaneous position can be similar to that of the target star. By exploiting the accuracy of Gaia DR2 on both asteroids and stars, stellar occultation prediction and exploitation becomes an effective method to systematically collect asteroid astrometry.The improvement of predictions through Gaia DR2 is proven via statistics of real predictions and comparison between stellar occultation predictions with Gaia DR2 for asteroids and other, such as Astorb and MPCORB, to verify which fit better to observed chords of past occultations.At the same time, asteroid occultations can offer the possibility to confirm or discover double stars, in a range of small angular separations very complementary to the resolution accessible to Gaia itself. We will present statistics and simulations showing the improvement expected in the prediction of asteroid occultations thanks to Gaia astrometry, in particular regarding the smaller uncertainties on the proper motion of target stars.Through a bayesian approach, the Bayesian Inference Method (BIM), we determine in the parameter space (duration; centre epoch; flux drop; star brightness) the domain of detectable events from a single site. Our study prepares the exploitation of the 0.5-m robotic telescope at Plateau de Calern" (Southern France) UniversCity, for which we determine the range of asteroid size and star brightness that we expect to reach. This facility will start operations after this work is over. The results obtained regarding the performance were compared with the previously used method to deriving all the relevant parameters (Least Squares Fit), with false positive signals to determine when these are most likely, and with several real observations, to verify the viability of this new method.After this work simulating the expected performance of UniversCity with the available equipment, the plan is to apply these limitations to predicted events and maximize the efficiency of the telescope's use. For that end, and accounting for all these factors, a survey was made to estimate how many events would be observable with a robotic telescope in a 1-year period with the current star and asteroid catalogues. To account for improvements in the asteroid uncertainties thanks to Gaia, for each event we checked what the impact on the likelihood would be if the asteroid had an uncertainty 2, 5, 10 or 20 times smaller, and results for each regime were compiled.We also analyzed the data of the 14 099 asteroids present DR2, how this impacted the semi-major axis (a) uncertainty, and how that would translate into improvements on stellar occultation predictions. This was made for two different weighting schemes, the one used for AstDyS (Farnocchia et al.) and one developed by the team, using observation residuals and occultations from the past to verify that the new weighting scheme would bring an improvement.Thanks to the collaboration with several astronomers, 16 observations were made throughout this work, with the three positives being analyzed with the new bayesian approach, which was also used for a few other observations where the photometric data was shared.
Anmerkung:
Dissertation HAL CCSD 2020
Sprache:
Englisch
