Towards an automated data cleaning with deep learning in CRESST.

The CRESST experiment employs cryogenic calorimeters for the sensitive measurement of nuclear recoils induced by dark matter particles. The recorded signals need to undergo a careful cleaning process to avoid wrongly reconstructed recoil energies caused by pile-up and read-out artefacts. We frame this process as a time series classification task and propose to automate it with neural networks. With a data set of over one million labeled records from 68 detectors, recorded between 2013 and 2019 by CRESST, we test the capability of four commonly used neural network architectures to learn the data cleaning task. Our best performing model achieves a balanced accuracy of 0.932 on our test set. We show on an exemplary detector that about half of the wrongly predicted events are in fact wrongly labeled events, and a large share of the remaining ones have a context-dependent ground truth. We furthermore evaluate the recall and selectivity of our classifiers with simulated data. The results confirm that the trained classifiers are well suited for the data cleaning task.

Secular equilibrium assessment in a CaWO4 target crystal from the dark matter experiment CRESST using Bayesian likelihood normalisation.

CRESST is a leading direct detection sub-GeVc-2 dark matter experiment. During its second phase, cryogenic bolometers were used to detect nuclear recoils off the CaWO4 target crystal nuclei. The previously established electromagnetic background model relies on Secular Equilibrium (SE) assumptions. In this work, a validation of SE is attempted by comparing two likelihood-based normalisation results using a recently developed spectral template normalisation method based on Bayesian likelihood. Albeit we find deviations from SE in some cases we conclude that these deviations are artefacts of the fit and that the assumptions of SE is physically meaningful.

Natural radionuclides as background sources in the Modane underground laboratory.

The Modane underground laboratory (LSM) is the deepest operating underground laboratory in Europe. It is located under the Fréjus peak in Savoie Alps in France, with average overburden of 4800 m w. e. (water equivalent), providing low-background environment for experiments in nuclear and particle physics, astrophysics and environmental physics. It is crucial to understand individual sources of background such as residual cosmic-ray flux of high-energy muons, muon-induced neutrons and contributions from radionuclides present in the environment. The identified dominant sources of background are radioactive contamination of construction materials of detectors and laboratory walls, radon contamination of the laboratory air, and neutrons produced in the laboratory. The largest neutron contribution has been identified from (α, n) reactions in low Z materials (10-7-10-4 n s-1 Bq-1) and from spontaneous fission of 238U (1.1× 10-6 n s-1 Bq-1).

Geant4-based electromagnetic background model for the CRESST dark matter experiment.

The CRESST (Cryogenic Rare Event Search with Superconducting Thermometers) dark matter search experiment aims for the detection of dark matter particles via elastic scattering off nuclei in CaWO 4 crystals. To understand the CRESST electromagnetic background due to the bulk contamination in the employed materials, a model based on Monte Carlo simulations was developed using the Geant4 simulation toolkit. The results of the simulation are applied to the TUM40 detector module of CRESST-II phase 2. We are able to explain up to ( 68 ± 16 ) % of the electromagnetic background in the energy range between 1 and 40 keV .

Environmental radionuclides as contaminants of HPGe gamma-ray spectrometers: Monte Carlo simulations for Modane underground laboratory.

The main limitation in the high-sensitive HPGe gamma-ray spectrometry has been the detector background, even for detectors placed deep underground. Environmental radionuclides such as 40K and decay products in the 238U and 232Th chains have been identified as the most important radioactive contaminants of construction parts of HPGe gamma-ray spectrometers. Monte Carlo simulations have shown that the massive inner and outer lead shields have been the main contributors to the HPGe-detector background, followed by aluminum cryostat, copper cold finger, detector holder and the lead ring with FET. The Monte Carlo simulated cosmic-ray background gamma-ray spectrum has been by about three orders of magnitude lower than the experimental spectrum measured in the Modane underground laboratory (4800 m w.e.), underlying the importance of using radiopure materials for the construction of ultra-low-level HPGe gamma-ray spectrometers.

Monte Carlo simulation of background characteristics of a HPGe detector operating underground in the Gran Sasso National Laboratory.

Monte Carlo (MC) simulation of background components of an ultra-low background high purity germanium (HPGe) detector operating in a deep underground laboratory was carried out. The results show that the background of the HPGe detector is about two orders of magnitude higher than the MC prediction when accounting only for cosmic-ray induced background. The difference is due to natural radioactivity in the parts surrounding the Ge detector. To get reasonable agreement between MC simulations and the experiment, a contamination in the parts surrounding the Ge crystal from 40K, 208Tl and 214Bi of 0.1mBqkg-1 was required to include in the simulations.

Simulation of background characteristics of low-level gamma-ray spectrometers using Monte Carlo method.

A computing code based on the CERN's GEANT4 has been used to compute cosmic-ray background components of low-level Ge-spectrometers, optionally equipped with an anticosmic shield made of a plastic scintillation detector. The results show that the background of the Ge detector placed at 100 m w.e. (water equivalent) has decreased by a factor of 30 and 100 without and with anticosmic shielding, respectively, when compared with a surface laboratory without anticosmic shielding.