Why does the detector generate negative concentrations?
The detector analyses the raw spectra based on the methods and settings selected in the spectral processing setting tab. By default, this is set to Full Spectrum Analysis (FSA) and the number of spectra to sum is set to 1. This means that each collected raw spectrum is analysed for radionuclide concentrations by using FSA. FSA is a method that fits standard spectra to the data, which is purely a mathematical analysis technique. The number of times a standard spectrum fits in the data is a representation of the concentration of that radionuclide. Because the fitting is a mathematical operation, it can result in the best fit that has a negative number of certain standard spectra, resulting in the reporting of a negative radionuclide concentration.
Often negative concentrations are found for 137Cs in areas where there is no 137Cs present. The average value of 137Cs is close to zero, but the concentrations for individual raw spectra can fluctuate around zero.
Another possible encounter of negative concentrations can be found if the spectrum is not well defined. Depending on the detector size and the radionuclide concentration in the area, a single 1-second spectrum may not contain sufficient counts to accurately determine the radionuclide concentration. A good indication of an insufficient number of counts in a single spectrum is if the uncertainty in the individual data points is of similar magnitude as the resulting radionuclide concentrations. A solution to this problem is increasing the Spectra to sum setting in the spectral processing tab. Alternatively, the fitmode can be set to non-negative least Full Spectrum Analysis (NNLS FSA), this analysis only includes standard spectra that result in a positive radionuclide concentration.
1. Increasing the Spectra to sum settings in the spectral processing setting tab.
2. Use the non-negative least Full Spectrum Analysis (NNLS FSA) fitting scheme