Publication

Aerosols and the imaging atmospheric Cherenkov technique

Connolly, Michael
Citation
Abstract
The ability to extract energetic γ-rays from the overwhelming high energy cosmic ray back- ground has opened up a new window of observation on the non-thermal universe. By utilis- ing an atmospheric technique VERITAS, an array of four imagining atmospheric Cherenkov telescopes, detects astrophysical gamma radiation in the energy range 85 GeV to >30 TeV. As the atmosphere is an integral part of the detector understanding the effects of aerosols is important. This thesis reports on a novel aerosol extinction estimation technique that utilises an opti- mised and repurposed ceilometer. Taking advantage of water vapour absorption correction for the 905-910 nm laser and optimised ceilometer data quality cuts an independent mea- surement is provided for aerosol extinction pro le up to 5 km above ground level. The inherent uncertainties are as low as ± ∼5-7%, with a high duty cycle > 95% for dusk to dawn operation in the absence of clouds. From close to 6 years of ceilometer data (Decem- ber 2011 to June 2017) aerosol optical depth is now known to seasonally increase more than 4-fold regularly from mid-winter to mid-summer, with a corresponding increase in overall atmospheric aerosol extinction ∼7-8%. During rare episodes of heavy aerosol loading at- mospheric aerosol extinction increases by more than 15%. A historical aring episode, during a period of elevated aerosol loading, of the blazar Mrk 421 in April 2013 was analysed with new elevated aerosol extinction instrument response functions. For the elevated aerosol extinction analysis the following result was obtained; I = 6.8±0.6×10−10 cm−2s−1TeV−1, Γ = −1.75±0.06, Eo = 1.45±0.14, χ2/NDF = 42.86/31. This result lies within experimental accuracies for the result obtained with thenormal aerosol extinction analysis. However, it is noted that below 237 GeV the ele- vated aerosol analysis shows a marked decline in signi cance σ . The spectral plot deviates slightly from a power-law with exponential cutoff below 237 GeV, but the differences are small. An in depth examination of systematic uncertainty in reconstructed energy arising from el- evated aerosol loading has yielded a year-on-year value of ∼5%, while the Mrk 421 April 2013 aring episode yielded an uncertainty in reconstructed energy of ∼6-7%. This analy- sis was not carried out for the next source examined, PKS 1441 +25. A soft VHE γ-ray source from April 2015 was analysed, PKS 1441 +25, whose data set was partially taken during elevated aerosol loading. For the elevated extinction analy- sis the following result was obtained; I = 8.9×10−12±1.3×10−12 cm−2s−1TeV−1 , Γ = −5.81±0.57, χ2/NDF = 4.98/3. This result also lies within experimental accuracies for the result obtained with the normal aerosol extinction analysis. In conclusion, the aerosol extinction correction applied to the Mrk 421 and PKS 1441 +25 data sets do not bene t noticeably from the elevated aerosol correction developed in this thesis, implying that VERITAS data taken during periods of elevated aerosol loading is not in need of re-analysis.
Funder
Publisher
Publisher DOI
Rights
Attribution-NonCommercial-NoDerivs 3.0 Ireland