Antonio D'Isanto (31.1.)
Vikas Josji (India) 06.02.2019
Reconstruction and Analysis of Highest Energy γ-Rays and its Application to Pulsar Wind Nebulae ( thesis pdf, 35 MB )
The High Altitude Water Cherenkov (HAWC) γ-ray observatory detects cosmic- and γ-rays in the TeV energy range. HAWC was recently upgraded with a sparse detector array (the outrigger array), which increases the instrumented area by a factor of 4-5 and will improve the sensitivity at energies greater than 10 TeV. This thesis consists of a number of contributions towards the improvement of the performance of HAWC at the highest energies and the study of a prominent high energy source, 2HWC J2019+367. To decide on components of the outrigger array, simulation input is provided. A new Monte Carlo template-based reconstruction method for air shower arrays is developed. It reconstructs the core location and energy of γ-ray showers. The goodness of fit of the method is utilised to separate the cosmic- and γ-ray showers. This method significantly improves the HAWC shower reconstruction and combines the reconstruction of HAWC and the outrigger array. In-depth spectral and morphological studies of 2HWC J2019+367 are performed. 2HWC J2019+367 shows a hint of energy-dependent morphology. A new HAWC source is discovered in the vicinity associated with VER J2016+371. The preferred direction of the X-ray and TeV emission indicates their association, and their combined spectral modelling show that 2HWC J2019+367 is likely to be the TeV pulsar wind nebula of PSR J2021+3651.
Supervisor: Jim Hinton (MPIK)
Antonio D'Isanto (Italy) 31.01.2019
Probabilistic photometric redshift estimation in massive digital sky surveys via machine learning ( thesis pdf, 21 MB )
The problem of photometric redshift estimation is a major subject in astronomy, since the need of estimating distances for a huge number of sources, as required by the data deluge of
the recent years. The ability to estimate redshifts through spectroscopy does not scale with this avalanche of data. Photometric redshifts provide the required redshift estimates at
the cost of some precision. The success of several forthcoming missions is highly dependent on the availability of photometric redshifts.
The purpose of this thesis is to provide innovative methods for photometric redshift estimation. Two models are proposed. The first is fully-automatized, based on the combination of
a convolutional neural network with a mixture density network, to predict probabilistic multimodal redshifts directly from images. The second model is features-based, performing a
massive combination of photometric parameters to apply a forward selection in a huge feature space. The proposed models perform very efficiently compared to some of the most common
models used in the literature. An important part of the work is dedicated to the correct estimation of the errors and prediction quality.
The proposed models are very general and can be applied to different topics in astronomy and beyond. ...
Supervisor: Kai Polsterer (HITS)