Enhancing Bayesian methods for radioactive source localization : a parameter study, prior construction and signal smoothing
Dvornik, Aliaksandr LU
; Finck, Robert
LU
and Rääf, Christopher
LU
(2026)
In Journal of Environmental Radioactivity
295.
- Abstract
This study clarifies when Bayesian analysis provides a practical advantage, particularly in large-area searches, hazardous or inaccessible environments, and low signal-to-noise ratio (SNR) scenarios where direct source approach is not feasible. We present a PyMC-based Bayesian framework for localizing a single unshielded gamma-emitting source and estimating its activity, with the ability to switch between generic priors and measurement-derived informed priors. Model performance was evaluated using 1240 synthetic datasets spanning varying source activities, detector-to-source distances, and background levels, and further tested using controlled field experiments. Two workflows were assessed: a rapid single-step analysis using generic... (More)
This study clarifies when Bayesian analysis provides a practical advantage, particularly in large-area searches, hazardous or inaccessible environments, and low signal-to-noise ratio (SNR) scenarios where direct source approach is not feasible. We present a PyMC-based Bayesian framework for localizing a single unshielded gamma-emitting source and estimating its activity, with the ability to switch between generic priors and measurement-derived informed priors. Model performance was evaluated using 1240 synthetic datasets spanning varying source activities, detector-to-source distances, and background levels, and further tested using controlled field experiments. Two workflows were assessed: a rapid single-step analysis using generic priors and a two-step approach in which preliminary estimates of source distance and activity define an informed prior and constrain the effective search area. Informed priors improved parameter stability near detection limits and reduced computation compared with generic priors.This
(Less)
study clarifies when Bayesian analysis provides a practical advantage,
particularly in large-area searches, hazardous or inaccessible
environments, and low signal-to-noise ratio (SNR) scenarios where direct
source approach is not feasible. We present a PyMC-based Bayesian
framework for localizing a single unshielded gamma-emitting source and
estimating its activity, with the ability to switch between generic
priors and measurement-derived informed priors. Model performance was
evaluated using 1240 synthetic datasets spanning varying source
activities, detector-to-source distances, and background levels, and
further tested using controlled field experiments. Two workflows were
assessed: a rapid single-step analysis using generic priors and a
two-step approach in which preliminary estimates of source distance and
activity define an informed prior and constrain the effective search
area. Informed priors improved parameter stability near detection limits
and reduced computation compared with generic priors. Savitzky-Golay
smoothing enhanced SNR and improved robustness in marginal cases but
could not compensate for insufficient signal strength. These results
define practical conditions under which Bayesian localization is
operationally beneficial. Savitzky-Golay smoothing enhanced SNR and improved robustness in marginal cases but could not compensate for insufficient signal strength. These results define practical conditions under which Bayesian localization is operationally beneficial.
- author
- Dvornik, Aliaksandr
LU
; Finck, Robert
LU
and Rääf, Christopher
LU
- organization
- publishing date
- 2026-04
- type
- Contribution to journal
- publication status
- published
- subject
- in
- Journal of Environmental Radioactivity
- volume
- 295
- article number
- 107960
- publisher
- Elsevier
- external identifiers
-
- pmid:41844023
- scopus:105033073903
- ISSN
- 0265-931X
- DOI
- 10.1016/j.jenvrad.2026.107960
- language
- English
- LU publication?
- yes
- id
- b674bb33-5456-4496-98e6-2bf377b06fab
- date added to LUP
- 2026-04-20 15:50:32
- date last changed
- 2026-05-18 19:02:50
@article{b674bb33-5456-4496-98e6-2bf377b06fab,
abstract = {{<p>This study clarifies when Bayesian analysis provides a practical advantage, particularly in large-area searches, hazardous or inaccessible environments, and low signal-to-noise ratio (SNR) scenarios where direct source approach is not feasible. We present a PyMC-based Bayesian framework for localizing a single unshielded gamma-emitting source and estimating its activity, with the ability to switch between generic priors and measurement-derived informed priors. Model performance was evaluated using 1240 synthetic datasets spanning varying source activities, detector-to-source distances, and background levels, and further tested using controlled field experiments. Two workflows were assessed: a rapid single-step analysis using generic priors and a two-step approach in which preliminary estimates of source distance and activity define an informed prior and constrain the effective search area. Informed priors improved parameter stability near detection limits and reduced computation compared with generic priors.This<br>
study clarifies when Bayesian analysis provides a practical advantage, <br>
particularly in large-area searches, hazardous or inaccessible <br>
environments, and low signal-to-noise ratio (SNR) scenarios where direct<br>
source approach is not feasible. We present a PyMC-based Bayesian <br>
framework for localizing a single unshielded gamma-emitting source and <br>
estimating its activity, with the ability to switch between generic <br>
priors and measurement-derived informed priors. Model performance was <br>
evaluated using 1240 synthetic datasets spanning varying source <br>
activities, detector-to-source distances, and background levels, and <br>
further tested using controlled field experiments. Two workflows were <br>
assessed: a rapid single-step analysis using generic priors and a <br>
two-step approach in which preliminary estimates of source distance and <br>
activity define an informed prior and constrain the effective search <br>
area. Informed priors improved parameter stability near detection limits<br>
and reduced computation compared with generic priors. Savitzky-Golay <br>
smoothing enhanced SNR and improved robustness in marginal cases but <br>
could not compensate for insufficient signal strength. These results <br>
define practical conditions under which Bayesian localization is <br>
operationally beneficial. Savitzky-Golay smoothing enhanced SNR and improved robustness in marginal cases but could not compensate for insufficient signal strength. These results define practical conditions under which Bayesian localization is operationally beneficial.</p>}},
author = {{Dvornik, Aliaksandr and Finck, Robert and Rääf, Christopher}},
issn = {{0265-931X}},
language = {{eng}},
publisher = {{Elsevier}},
series = {{Journal of Environmental Radioactivity}},
title = {{Enhancing Bayesian methods for radioactive source localization : a parameter study, prior construction and signal smoothing}},
url = {{http://dx.doi.org/10.1016/j.jenvrad.2026.107960}},
doi = {{10.1016/j.jenvrad.2026.107960}},
volume = {{295}},
year = {{2026}},
}