Hyperspectral drift, QAQC and Spectral Alignment
- 2 minutes read - 369 wordsHi Everyone,
I would like to make a short comment on drift of the wavelength features in hyperspectral instruments and cross instrument calibration.
For the sake of my sanity and yours I’m only going to address wavelength position/alignment, spectral resolution is a slightly separate issue.
The first point of clarification is that the wavelength standards have a dual purpose:
- to monitor the performance of the instrument and drift
- to allow you to correct the position of the spectral features to a known energy.
Secondly calibration pucks should be:
- spectralon
- doped spectralon
- mylar If you have high performance requirements add 1-3 grey reference disks.
The utility of using rocks in the reference set is low, as their inherent variability makes using the spectra difficult.
To demonstrate we will correct the spectra between two of the most different hylogger instruments the one at the NTGS and the other at GSSA using their QAQC test rocks.
To be clear these instruments are very well calibrated so the adjustment is minimal.
Also the NIR calibration puck is quite different between these two instruments, which makes the correction less reliable than it should be.
To do this correction properly you should use the same set of calibration discs with each instrument.
The figure below shows the effect of wavelength correcting the spectra between the instrument housed at the NTGS and the one at the GSSA.
The correction between instruments is smaller at the lower wavelength range and higher at the longer wavelengths.
The practical effect of this is that you need to be more careful when interpreting the change in the location of the 2200nm feature as this can be potentially an artifact of instrument drift/serial number not composition.
Fig 2.B. from Cloutier et al shows a 5nm shift in the position of Fe-Mg Chlorites a 1.4nm drift between instruments is approximately 30% of the variation, so quite material.
Thanks,
Ben.
References:
Preliminary procedure for performing QAQC checks on HyLogger Test rock data using TSG’s “Headless mode”: Lau et al.Field-Portable SWIR Acquisition, QA-QC, and Processing Guide: McLean Trott et al
Cloutier, J.; Piercey, S.J.; Huntington, J. Mineralogy, Mineral Chemistry and SWIR Spectral Reflectance of Chlorite and White Mica. Minerals 2021, 11, 471. https://doi.org/10.3390/min11050471