3. The Mrk 501 data sample
The analysis is based on the same data sample as used in Paper 1, corresponding to a total exposure time of 110 h, between March and October of 1997. Mrk 501 was observed in the so-called wobble mode, with the source positioned in declination away from the optical axis of the telescopes, alternating every 20 min. For background subtraction, an equivalent region displaced by the same amount in the opposite direction is used. The separation of of these on-source and off-source regions is large compared to the angular resolution of the telescope system. In total, the sample comprises about 38,000 -ray events.
To select -ray candidates, the same loose cuts were applied as in Paper 1. In particular, the reconstructed shower direction had to be within from the source, and the mean scaled width parameter had to be less than 1.2. Events were accepted up to a maximum impact parameter of 200 m from the central telescope CT3. Events with larger impact parameters frequently suffer from truncated images in the cameras and, due to small angles between the stereo views, have larger uncertainties on the shower parameters. Since the analysis presented in the following emphasizes the control of systematic errors, it was felt that in this case the advantage of having clean events and a well-defined effective area at large energies - basically all events within the 200 m radius trigger above a few TeV - outweighs the gain in statistics which could have been achieved by accepting all events 1. For the following analysis a software threshold of two or more triggered telescopes, each with at least 40 recorded photoelectrons was used.
Properties of this Mrk 501 data sample were examined in detail in Paper 1.
© European Southern Observatory (ESO) 1999
Online publication: August 25, 1999