Inter-analyst comparison and reproducibility of apatite fission
track analysis
Murat T. Tamer, Ling Chung, Richard A. Ketcham, Andrew, J.W. Gleadow
Factors influencing data reproducibility of fission-track (FT)
thermochronology can be summarized into three main categories associated
with data acquisition steps. (1) Sample preparation involves mineral
separation, mounting, polishing and etching; (2) data revelation relates
to instrumentation (microscope, LAICPMS, etc.) and software settings;
and (3) execution depends on feature selection by the analyst. Previous
committee reports and studies (Hurford A.J. 1990; Ketcham et al. 2009;
Ketcham et al. 2015; Ketcham et al. 2018) have contributed significant
insights into the reproducibility of fission-track data by comparing
length and age measurements produced by several laboratories using their
own preparation and revelation procedures.
A recent attempt to isolate analyst-specific factors in length
measurement using an image-based approach (Tamer et al. 2019) found that
when two analysts observe the same feature and agree it is a valid
track, measurement reproducibility was very good, though impacted by
etching. Dispersion of individual length measurements was 0.7-1.0 µm
(2σ) for weaker etching and 0.5-0.8 µm for stronger etching, but mean
lengths were always within 0.1 µm of each other. Where the analysts
disagreed more significantly, however, was in finding tracks and
evaluating whether they were valid, sufficiently clear, and sufficiently
etched for measurement, which led to differences of up to
~0.8 µm in mean track length.
This study builds on the image-based approach to encompass more aspects
of the measurement process and increase the number of analysts being
compared. We will look at confined track selection in greater detail,
and also study analyst decisions behind age determination, including the
selection of the region of interest for counting, and identification of
grain-surface features as tracks appropriate for counting. Reflected and
transmitted light image stacks for 41 grains and graticules are
available on a cloud platform Participants will carry out analyses of
these images using their preferred approach, e.g. suitable analytical
software, manual measurements or AI-based analysis. A limited license
for FastTracks (v3.2) will be available for those who would like to
participate but do not have measurement software.
Analysts are asked to fill out a questionnaire about their fission track
experience, conduct track density estimations, confined track length and
Dpar measurements, and especially provide comments on all
grains being analyzed or skipped. FastTracks users are asked to send
the .xml files produced by the software, while other participants are
asked to submit the results using a template. The results will be
entirely anonymous unless the analyst states otherwise. The deadline for
the submission of the results is June 1st, 2022. The results will be
shared on 18th International Conference on Thermochronology.
Image Stacks for 41 grains:
https://drive.google.com/drive/folders/1yUp4bqtEag5wc8CJNG0M20nKDidWCHnQ?usp=sharing
FastTracks V3 limited license:
https://mediaflux.researchsoftware.unimelb.edu.au:443/mflux/share.mfjp?_token=LGT8N3Andqlnu4qx5CBY1128203387&browser=true&filename=FastTracks_3.2.4.zip
.
FastTracks V3 manual:
https://mediaflux.researchsoftware.unimelb.edu.au:443/mflux/share.mfjp?_token=mVLFu6kizDYIKpTamNGH1128203389&browser=true&filename=FastTracks_V3_Manual.pdf
.
Submissions to murat@ies.ac.cn