Skip to content

ENSI-G14 Calculator

logo

python release Docs

Overview

This package is a python 3.9 implementation of the ENSI-G14 directive. The end goal is calculating the radioactive dose experienced by the population in the surroundings of installations susceptible to release radioactive nuclides in the air. For example:

  • nuclear power plants,
  • cyclotrons used for research or medicinal isotope productions,
  • nuclear medicine facilities.

Unofficial implementation

This independent package attempts to implement the complete ENSI-G14 calculation chain. It is not an official ENSI product and has not been endorsed or certified by ENSI.

The package can evaluate scenarios that are prospective (planned or accidental release), or retrospective (often a computation of the long term dose given actual measured releases).

To do so, it implements the infrastructure required to compute:

  • The dispersion factor \(\chi\) (modelled as a Gaussian plume).
  • The nuclide-dependent deposition \(\xi\), computed from fallout \(F\) and washout \(W\).
  • The stored activity in and on the soil, as well as on plant foliage and in roots.

Following from these, the package computes the doses to exposed groups from the following pathways:

  • ingestion of plants, meat, dairy, or water,
  • ground-shine,
  • submersion,
  • inhalation (including from resuspension).

The package handles:

  • The geographical setup, including coordinates of the installation and of the points of interest.
  • The topography (relative elevation between the installation and the points of interest).
  • The calculation of the effective release height of the activity.
  • The computation of the quantities mentioned above (dispersion, deposition, etc).
  • The averaging of the dispersion over measured weather patterns to compute "long term" dose estimates.

Getting started

See the basics and recipes page for installation and small, copy-able examples of the individual components. All code blocks in the example pages are executed in the test-suite.

Short term scenarios

For prospective short term scenarios, the logic has been implemented as prescribed in the directive, with releases happening in sequential slices; the weather varying in a pre-defined way in each slice.

All types of short term scenarios can be easily programmed with the components provided by the package, and we provide a high level interface for a 24h release, see this page.

Long term scenarios

Long term dose estimation assumes a steady state of activity concentration in soil and food. We provide a high level interface to compute dose

  • by chimney emitting the activity,
  • by nuclide,
  • by exposed group (position).

See the long term example page.

Point-Kernel-Integral submersion dose for short distances

The package also implements an opt-in photon point-kernel integral (PKI) and a vectorized PKI-trained surrogate. They evaluate actual photon-line energies and finite-cloud geometry, including short distances and observers away from the plume axis. These are diagnostic alternatives rather than regulatory replacements: ENSI-G14 remains the default, and the PKI paths do not include beta transport or the complete nuclide-level regulatory coefficient model.

See the submersion model overview for the model assumptions, training design, and comparison plots, or go directly to the photon PKI and surrogate example.

For short stacks on or next to buildings, with receptors closer than 200 m (e.g. cyclotron exhausts), see the near-field example: a building-wake extension of the G-14 dispersion, outside the directive.