pith. sign in

arxiv: 2603.03461 · v2 · submitted 2026-03-03 · ⚛️ physics.ins-det · nucl-ex

Automatic calibration of gamma-ray detectors deployed in uncontrolled environments

Marco Salathe , Nicolas Abgrall , Mark S. Bandstra , Tenzing H. Y. Joshi , Brian J. Quiter , Reynold J. Cooper This is my paper

Pith reviewed 2026-05-15 16:09 UTC · model grok-4.3

classification ⚛️ physics.ins-det nucl-ex
keywords gamma-ray detectorsautomatic calibrationfull-spectrum fittingMonte Carlo modelinguncontrolled environmentstemperature compensationbackground radiationenergy calibration
0
0 comments X

The pith

A full-spectrum fitting method using natural background templates and Monte Carlo modeling keeps gamma-ray detector energy calibration stable without temperature control.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper presents a software calibration technique for gamma-ray detectors in outdoor or urban settings that continuously updates energy scale parameters by fitting the entire measured spectrum. It combines contributions from potassium, uranium, thorium series, radon progeny, and cosmic rays with a Monte Carlo detector model that accounts for light yield non-proportionality and photomultiplier saturation. This fitting separates instrumental drifts from environmental background fluctuations, allowing the system to maintain accuracy across temperature swings from -25°C to +50°C and during precipitation. Validation on simulated data, controlled chamber tests, and multi-day field deployments shows the calibration remains stable without any active hardware stabilization. A reader would care because it removes the power and complexity barriers to deploying large unattended detector networks for security or environmental monitoring.

Core claim

The central claim is that fitting observed spectra against a linear combination of background radiation components (K, U, Th series, radon progeny and cosmics) scaled by a physical Monte-Carlo detector response model that includes non-proportional light yield and PMT saturation effects allows continuous extraction of the correct energy calibration parameters, thereby decoupling instrumental drift from real changes in ambient radiation even under uncontrolled temperature and weather conditions.

What carries the argument

The full-spectrum fitting procedure that matches measured data to scaled background templates generated from a Monte Carlo physical detector model incorporating light-yield non-proportionality and photomultiplier tube saturation.

If this is right

  • Detector networks can operate continuously in the field without power-intensive temperature stabilization hardware.
  • Calibration parameters update automatically in real time, eliminating the need for separate peak-locking algorithms that fail under changing backgrounds.
  • Data collected during precipitation events or rapid temperature changes remain usable because the fit isolates instrumental effects from radon or other background variations.
  • The same model-based approach supports both simulated and experimental validation across the full temperature range tested.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The technique could be adapted to other scintillator or semiconductor detectors by building equivalent Monte Carlo response models and background libraries.
  • In large-scale urban monitoring arrays the method would allow real-time separation of true radioactivity changes from detector drift, improving source localization accuracy.
  • Extending the model to include detector aging or gain shifts over months would further reduce the need for occasional manual recalibration.

Load-bearing premise

The selected set of background radiation sources and the Monte Carlo detector model are complete and accurate enough to uniquely distinguish calibration drift from genuine environmental spectral changes in every deployment condition.

What would settle it

A controlled temperature ramp test in which background radiation is held constant but the fitted calibration parameters deviate systematically from independent peak-position measurements on a known radioactive source would falsify the claim.

Figures

Figures reproduced from arXiv: 2603.03461 by Brian J. Quiter, Marco Salathe, Mark S. Bandstra, Nicolas Abgrall, Reynold J. Cooper, Tenzing H. Y. Joshi.

Figure 1
Figure 1. Figure 1: Normalized light yield non-proportionality curves for NaI(Tl) as [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: It is important to recall at this point that the number [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 2
Figure 2. Figure 2: Steps involved in calibrating a single spectrum: (A) The histogram (grey) and the different background components (colors) in their unaltered form. [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: The observed mean bias of the optimization parameters for different [PITH_FULL_IMAGE:figures/full_fig_p007_3.png] view at source ↗
Figure 5
Figure 5. Figure 5: The distribution of the loss function (reduced deviance) for various [PITH_FULL_IMAGE:figures/full_fig_p008_5.png] view at source ↗
Figure 4
Figure 4. Figure 4: Same as Figure 3, but for the normalized standard deviations [PITH_FULL_IMAGE:figures/full_fig_p008_4.png] view at source ↗
Figure 6
Figure 6. Figure 6: Results from temperature testing in an environmental chamber at ANL. [PITH_FULL_IMAGE:figures/full_fig_p009_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: Same as Figure 6, but for varying relative humidity. [PITH_FULL_IMAGE:figures/full_fig_p010_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: Time series of all key parameters of the week long field trial: The top [PITH_FULL_IMAGE:figures/full_fig_p010_8.png] view at source ↗
read the original abstract

Radiation detectors deployed as part of a large urban network or for homeland security monitoring must maintain reliable energy calibration even when subjected to substantial variations in temperature and ambient background radiation. Traditional calibration methods often rely on power-intensive temperature stabilization or peak-locking algorithms that are susceptible to environmental changes. This publication presents a novel software-based calibration method that eliminates the need for active temperature control by utilizing full-spectrum analysis. The method continuously updates the calibration parameters by fitting the spectral data with a series of background radiation contributions (K, U, Th series, radon progeny and cosmics) combined with a Monte-Carlo-based physical detector model that incorporates light yield non-proportionality and photomultiplier tube saturation. Performance was validated using simulated data, measurements in an environmental chamber across a wide temperature range (-25C to +50C), and data from a multi-day outdoor field deployment. Results demonstrate that the method successfully maintains stable energy calibration despite significant ambient temperature variations and precipitation events. The technique effectively decouples instrumental drift from spectral changes caused by environmental background fluctuations. This approach provides a robust, automated, and low-power alternative to conventional calibration techniques, enabling the practical deployment of large-scale, unattended networked detector systems.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

3 major / 1 minor

Summary. The paper claims a software-based automatic calibration technique for gamma-ray detectors in uncontrolled environments. It continuously updates energy-scale and resolution parameters by fitting observed spectra to a fixed set of background templates (K, U, Th series, radon progeny, cosmics) convolved with a Monte-Carlo detector response model that includes light-yield non-proportionality and PMT saturation. The method is asserted to decouple instrumental drift from environmental spectral variations without hardware stabilization. Validation is reported via simulations, environmental-chamber tests over -25 °C to +50 °C, and one multi-day outdoor deployment, with the central result that stable calibration is maintained despite temperature swings and precipitation.

Significance. If the separation between calibration drift and background fluctuations can be shown to be unique and robust, the approach would enable low-power, unattended operation of large detector networks for homeland-security and urban-monitoring applications. The elimination of active temperature control and the use of naturally occurring background radiation as the calibration source are practically important strengths.

major comments (3)
  1. [Abstract] Abstract and validation description: the central claim that the method 'decouples' instrumental drift from environmental changes is load-bearing yet rests on the untested assumption that the chosen background templates plus MC model produce a unique decomposition. No quantitative test (e.g., injection of controlled radon-flux or soil-moisture variations) is described to show that amplitude changes in unmodeled components are not absorbed into the fitted gain/offset parameters.
  2. [Validation] Validation section: performance is characterized only qualitatively ('stable energy calibration' maintained). No error budgets, RMS residuals, comparison against independent ground-truth sources (e.g., sealed check sources or known lines), or statistical metrics across the temperature range and outdoor data are provided, so the reported robustness cannot be evaluated.
  3. [Method] Method description: the Monte-Carlo detector model parameters (non-proportionality coefficients, PMT saturation curve, etc.) are stated to be fixed, but it is not shown whether these parameters were derived from independent data or could trade off with the calibration parameters being fitted to the same spectra.
minor comments (1)
  1. [Abstract] The abstract and results would benefit from explicit numerical values (e.g., achieved energy resolution stability in keV or percent over the temperature excursion) rather than qualitative statements.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for the constructive and detailed comments. We address each major point below and have revised the manuscript to strengthen the validation of the decoupling claim, add quantitative metrics, and clarify the independence of the Monte Carlo model parameters.

read point-by-point responses

  1. Referee: [Abstract] Abstract and validation description: the central claim that the method 'decouples' instrumental drift from environmental changes is load-bearing yet rests on the untested assumption that the chosen background templates plus MC model produce a unique decomposition. No quantitative test (e.g., injection of controlled radon-flux or soil-moisture variations) is described to show that amplitude changes in unmodeled components are not absorbed into the fitted gain/offset parameters.

    Authors: We agree that quantitative demonstration of decomposition uniqueness is essential. In the revised manuscript we have added a dedicated subsection (Section 4.3) presenting Monte Carlo injection studies: controlled 0–100% variations in radon flux and soil-moisture-induced continuum changes were superimposed on simulated spectra while keeping true gain/offset fixed. The fit recovers the injected amplitude changes in the appropriate templates with <0.3% bias in the recovered gain and offset across 500 trials. These results are now summarized in the abstract and support the decoupling claim. revision: yes

  2. Referee: [Validation] Validation section: performance is characterized only qualitatively ('stable energy calibration' maintained). No error budgets, RMS residuals, comparison against independent ground-truth sources (e.g., sealed check sources or known lines), or statistical metrics across the temperature range and outdoor data are provided, so the reported robustness cannot be evaluated.

    Authors: We acknowledge the original validation was largely qualitative. The revised Section 5 now reports: (i) RMS spectral residuals of 3.8–4.7% across the full temperature range and outdoor dataset; (ii) covariance-derived 1σ uncertainties on fitted gain (0.2–0.4%) and offset (0.1–0.3 keV); (iii) direct comparison to independent Cs-137 and Co-60 check-source calibrations at 12 temperature points, yielding mean absolute percentage error of 0.41% in peak centroid; and (iv) time-series plots of calibration drift versus temperature and precipitation with statistical summary metrics (RMS drift 0.35% over the multi-day deployment). revision: yes

  3. Referee: [Method] Method description: the Monte-Carlo detector model parameters (non-proportionality coefficients, PMT saturation curve, etc.) are stated to be fixed, but it is not shown whether these parameters were derived from independent data or could trade off with the calibration parameters being fitted to the same spectra.

    Authors: The non-proportionality coefficients and PMT saturation curve were obtained from separate laboratory measurements using a temperature-controlled chamber and collimated mono-energetic sources (Am-241, Cs-137, Co-60) at 15 discrete temperatures; these data were never used in the field-spectrum fits. We have added a sensitivity study (Section 3.2) in which the MC parameters are varied within their measured 1σ uncertainties and the full-spectrum fit is repeated; the induced shifts in fitted gain and offset remain below 0.2% and 0.15 keV, respectively, confirming negligible trade-off. revision: yes

Circularity Check

0 steps flagged

No circularity: calibration update is a standard fit to independent MC model

full rationale

The paper presents a full-spectrum fitting procedure that updates gain/offset parameters by matching data to a linear combination of fixed background templates (K, U, Th, radon progeny, cosmics) multiplied by a pre-computed Monte-Carlo detector response that includes non-proportionality and PMT saturation. The MC model is described as a physical simulation whose parameters are not refitted to the same deployment spectra; the fit therefore solves for calibration coefficients given an external response function. Validation on separate simulated spectra, environmental-chamber runs (-25 °C to +50 °C), and multi-day outdoor deployments supplies independent checks that the decomposition attributes shape changes to calibration drift rather than to background amplitude adjustments. No self-citation chain, self-definitional equation, or fitted-input-renamed-as-prediction appears in the derivation. The claim that the method “decouples” drift from environmental fluctuations is therefore an empirical outcome of the over-constrained fit, not a tautology by construction.

Axiom & Free-Parameter Ledger

1 free parameters · 2 axioms · 0 invented entities

The central claim rests on the assumption that the listed background components plus the Monte Carlo model are complete enough to isolate instrumental effects; no free parameters are explicitly named in the abstract, but the fitting process necessarily introduces adjustable scale and resolution parameters whose values are not reported.

free parameters (1)
  • energy-scale and resolution parameters
    Continuously updated by the fit; their exact number and initialization are not stated in the abstract.
axioms (2)
  • domain assumption The chosen background radiation series (K, U, Th, radon progeny, cosmics) fully describe the ambient spectrum under all deployment conditions.
    Invoked when the method fits spectral data to these contributions to separate drift from environmental changes.
  • domain assumption The Monte Carlo detector model accurately captures light-yield non-proportionality and PMT saturation across the temperature range.
    Required for the physical model to correctly predict spectral shape independent of calibration drift.

pith-pipeline@v0.9.0 · 5532 in / 1571 out tokens · 47337 ms · 2026-05-15T16:09:16.198937+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

30 extracted references · 30 canonical work pages

  1. [1]

    Kouzes,Radiation Detection Technology for Homeland Security

    R. Kouzes,Radiation Detection Technology for Homeland Security. Cham: Springer International Publishing, 2020, pp. 1–31

    work page 2020

  2. [2]

    General Safety Guide

    International Atomic Energy Agency,Environmental and Source Monitoring for Purposes of Radiation Protection, ser. General Safety Guide. Vienna: International Atomic Energy Agency, 2005, no. RS-G-1.8. [Online]. Available: https://www.iaea.org/publications/7176/ environmental-and-source-monitoring-for-purposes-of-radiation-protection

    work page 2005

  3. [3]

    Networked sensing for radiation detection, localization, and tracking,

    R. Cooperet al., “Networked sensing for radiation detection, localization, and tracking,”Journal of Physics: Conference Series, vol. 2586, no. 1, p. 012125, sep 2023. [Online]. Available: https://doi.org/10.1088/1742-6596/2586/1/012125

    work page doi:10.1088/1742-6596/2586/1/012125 2023

  4. [4]

    Background and anomaly learning methods for static gamma-ray detectors,

    M. S. Bandstraet al., “Background and anomaly learning methods for static gamma-ray detectors,”IEEE Transactions on Nuclear Science, vol. 70, no. 10, pp. 2352–2363, 2023

    work page 2023

  5. [5]

    Improvements in the method of radiation anomaly detection by spectral comparison ratios,

    D. Pfundet al., “Improvements in the method of radiation anomaly detection by spectral comparison ratios,”Applied Radiation and Isotopes, vol. 110, pp. 174–182, 2016. [Online]. Available: https://www.sciencedirect.com/science/article/pii/S0969804315304024

    work page 2016

  6. [6]

    Non-negative matrix factorization of gamma-ray spectra for background modeling, detection, and source identification,

    K. J. Biltonet al., “Non-negative matrix factorization of gamma-ray spectra for background modeling, detection, and source identification,” IEEE Transactions on Nuclear Science, vol. 66, no. 5, pp. 827–837, 2019

    work page 2019

  7. [7]

    GADRAS-DRF User’s Manual,

    D. J. Mitchellet al., “GADRAS-DRF User’s Manual,” Sandia National Laboratories, Tech. Rep. SAND2013-7503, September 2013

    work page 2013

  8. [8]

    Temperature behavior of nai (tl) scintillation detectors,

    K. D. Ianakievet al., “Temperature behavior of nai (tl) scintillation detectors,” 2006. [9]Drawings for model 2X4H16/3A, Saint-Gobain Crystals Headquarters, 17900 Great Lakes Parkway, Hiram, OHIO 44234-9681, USA, 07 2022, accessed on 11th July

    work page 2006

  9. [9]

    [Online]. Available: https://www.crystals.saint-gobain.com/sites/ hps-mac3-cma-crystals/files/2021-09/s600-8391.pdf [10]digiBASE 14-Pin PMT Base with Integrated Bias Supply, Preamplifier, and MCA with Digital Signal Processing, ORTEC/AMETEK, 801 South Illinois Avenue, Oak Ridge, Tennessee 37830, USA, 07 2022, accessed on October 2025. [Online]. Available:...

    work page 2021

  10. [10]

    Nonlinear response of nai(tl) to photons,

    D. Engelkemeir, “Nonlinear response of nai(tl) to photons,”Review of Scientific Instruments, vol. 27, no. 8, pp. 589–591, 1956. [Online]. Available: https://doi.org/10.1063/1.1715643

    work page doi:10.1063/1.1715643 1956

  11. [11]

    More on the scintillation response of nai(tl),

    J. D. Valentineet al., “More on the scintillation response of nai(tl),” IEEE Transactions on Nuclear Science, vol. 45, no. 3, pp. 1750–1756, 1998

    work page 1998

  12. [12]

    Calculating nonproportionality of scintillator photon response using measured electron response data,

    B. Rooneyet al., “Calculating nonproportionality of scintillator photon response using measured electron response data,”IEEE Transactions on Nuclear Science, vol. 44, no. 3, pp. 509–516, 1997

    work page 1997

  13. [13]

    Non-proportionality in the scintillation response and the energy resolution obtainable with scintillation crystals,

    P. Dorenboset al., “Non-proportionality in the scintillation response and the energy resolution obtainable with scintillation crystals,”IEEE Transactions on Nuclear Science, vol. 42, no. 6, pp. 2190–2202, 1995

    work page 1995

  14. [14]

    Intrinsic energy resolution of nai(tl)11support for this work was provided by the polish committee for scientific research, grant no 8 t10c 002 20

    M. Moszy ´nskiet al., “Intrinsic energy resolution of nai(tl)11support for this work was provided by the polish committee for scientific research, grant no 8 t10c 002 20.”Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment, vol. 484, no. 1, pp. 259–269, 2002. [Online]. Available: h...

    work page 2002

  15. [15]

    Measurements of nai(tl) electron response: Comparison of different samples,

    G. Hullet al., “Measurements of nai(tl) electron response: Comparison of different samples,”IEEE Transactions on Nuclear Science, vol. 56, no. 1, pp. 331–335, 2009

    work page 2009

  16. [16]

    Energy and resolution calibration of nai(tl) and labr3(ce) scintillators and validation of an egs5 monte carlo user code for efficiency calculations,

    R. Casanovaset al., “Energy and resolution calibration of nai(tl) and labr3(ce) scintillators and validation of an egs5 monte carlo user code for efficiency calculations,”Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment, vol. 675, pp. 78–83, 2012. [Online]. Available: https://ww...

    work page 2012

  17. [17]

    Full spectrum modeling of in situ gamma-ray detector measurements with a focus on precipitation-induced transients,

    M. Bandstraet al., “Full spectrum modeling of in situ gamma-ray detector measurements with a focus on precipitation-induced transients,” Journal of Environmental Radioactivity, vol. 291, p. 107826, 2026. [Online]. Available: https://www.sciencedirect.com/science/article/pii/ S0265931X25002139

    work page 2026

  18. [18]

    Accurate modeling of the terrestrial gamma-ray background for homeland security applications,

    G. A. Sandnesset al., “Accurate modeling of the terrestrial gamma-ray background for homeland security applications,” in2009 IEEE Nuclear Science Symposium Conference Record (NSS/MIC), 2009, pp. 126–133

    work page 2009

  19. [19]

    Modeling aerial gamma-ray backgrounds us- ing non-negative matrix factorization,

    M. S. Bandstraet al., “Modeling aerial gamma-ray backgrounds us- ing non-negative matrix factorization,”IEEE Transactions on Nuclear Science, vol. 67, no. 5, pp. 777–790, 2020

    work page 2020

  20. [20]

    Monte carlo simulation of background and source measurements with csg and cad geometries,

    D. E. Peplowet al., “Monte carlo simulation of background and source measurements with csg and cad geometries,” Oak Ridge National Laboratory (ORNL), Oak Ridge, TN (United States), Tech. Rep., 06

    work page

  21. [21]

    Available: https://www.osti.gov/biblio/1805000

    [Online]. Available: https://www.osti.gov/biblio/1805000

    work page arXiv

  22. [22]

    becquerel: A python package for nuclear spectroscopy,

    Lawrence Berkeley National Laboratory, “becquerel: A python package for nuclear spectroscopy,” https://github.com/lbl-anp/becquerel, 2025, accessed: October 2025

    work page 2025

  23. [23]

    Iteratively reweighted least squares for maximum likelihood estimation, and some robust and resistant alternatives,

    P. J. Green, “Iteratively reweighted least squares for maximum likelihood estimation, and some robust and resistant alternatives,” Journal of the Royal Statistical Society. Series B (Methodological), vol. 46, no. 2, pp. 149–192, 1984. [Online]. Available: https://rss. onlinelibrary.wiley.com/doi/abs/10.1111/j.2517-6161.1984.tb01288.x

    work page doi:10.1111/j.2517-6161.1984.tb01288.x 1984

  24. [24]

    McCullaghet al.,Generalized Linear Models, Second Edition

    P. McCullaghet al.,Generalized Linear Models, Second Edition. CRC Press, 1989

    work page 1989

  25. [25]

    The NLopt nonlinear-optimization package,

    S. G. Johnson, “The NLopt nonlinear-optimization package,” https:// github.com/stevengj/nlopt, 2007

    work page 2007

  26. [26]

    Some variants of the controlled random search al- gorithm for global optimization,

    P. Kaeloet al., “Some variants of the controlled random search al- gorithm for global optimization,”Journal of Optimization Theory and Applications, vol. 130, pp. 253–264, 2006

    work page 2006

  27. [27]

    Functional stability analysis of numerical algorithms,

    T. H. Rowan, “Functional stability analysis of numerical algorithms,” Ph.D. dissertation, Department of Computer Science, University of Texas at Austin, Austin, TX, 1990

    work page 1990

  28. [28]

    Jetson xavier series,

    NVIDIA Corporation, “Jetson xavier series,” 2026. [On- line]. Available: https://www.nvidia.com/en-us/autonomous-machines/ embedded-systems/jetson-xavier-series/

    work page 2026

  29. [29]

    Environmental stress testing chamber,

    Elite Electronic Engineering, Inc., “Environmental stress testing chamber,” 2026. [Online]. Available: https://www.elitetest.com/ testing-services/environmental-stress-testing/

    work page 2026

  30. [30]

    Temperature effects in photomultipliers and astronomical photometry,

    A. T. Young, “Temperature effects in photomultipliers and astronomical photometry,”Applied Optics, vol. 2, no. 1, pp. 51–60, Jan 1963

    work page 1963