Speaker
Description
Airborne gamma spectrometry measures desired radiation quantities usually from relatively long distances, often from sources spread over large areas. Values measured over one data acquisition period (scan) represent the average over a specific area and, in general, may not agree with the value at the corresponding coordinates on the surface. This “spatial resolution” depends on the height and speed of flight, source size and energy, and sampling frequency, i.e., time interval of one scan. Request for good spatial resolution with the requirement of quick coverage of large monitored areas leads to the requirement of a high sampling frequency, with a time of scan typically 1 second. It is in contradiction with the needs of reasonable statistics in measured spectra with respect to the next data processing and analysis, namely in UAS-borne spectrometry, when the size of detectors is limited by the available payload. This leads to the fact that in airborne spectrometry, it is often necessary to work with spectra with poor, even extremely poor, statistics.
Contribution discuses statistic errors of quantities derived from the airborne spectrometry data, its dependence on the arrangement of measurements and possibilities of its reduction. The idea of generating and processing “most probable spectra” created using perturbation techniques is also discussed. The experimental data were measured using an airborne spectrometer D230A (Georadis s.r.o., Brno, Czech Republic) equipped with a pair of 2”x2” NaI(Tl) detectors. Spectra were processed by full spectra analysis based on the spectra unfolding and LSQ fitting.
This work is co-financed from the state budget by the Technology Agency of the Czech Republic and Ministry of the Environment under the Environment for life Programme within the project SS06010467.