A physics-based Gaussian kernel for air pollutant dispersion over large Eulerian grids

Fuente: Zenodo
Saved in:
Bibliographic Details
Main Author: Bessagnet, Bertrand
Format: Recurso digital
Language:English
Published: Zenodo 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866901769509928960
author Bessagnet, Bertrand
author_facet Bessagnet, Bertrand
contents <p>The Gaussian atmospheric dispersion theory is widely used to assess the dispersion of an emitting point sources. For the first time, we generalize this concept to possibly millions of emitting grid cells over large domains, with a high spatial resolution (100 m to some kilometers). The model requires a gridded emission dataset and a minimal number of meteorological variables to reconstruct stability conditions. A combination of a <em>fortran </em>routine embedded in a <em>python </em>architecture makes the module usable in a simple computing infrastructure. This code, called AIRIK (as AtmospheIc dispeRsIon Kernel), is freely available and allows for rapid dispersion analysis of multiple emission sources or to inform metamodels with more realistic dispersion conditions. This approach opens new developments to elaborate more physical convolutional filters in neural network architectures for atmospheric dispersion.</p>
format Recurso digital
id zenodo_https___doi_org_10_5281_zenodo_18212372
institution Zenodo
language eng
publishDate 2026
publisher Zenodo
record_format zenodo
spellingShingle A physics-based Gaussian kernel for air pollutant dispersion over large Eulerian grids
Bessagnet, Bertrand
modelling
dispersion
Gaussian
air quality
<p>The Gaussian atmospheric dispersion theory is widely used to assess the dispersion of an emitting point sources. For the first time, we generalize this concept to possibly millions of emitting grid cells over large domains, with a high spatial resolution (100 m to some kilometers). The model requires a gridded emission dataset and a minimal number of meteorological variables to reconstruct stability conditions. A combination of a <em>fortran </em>routine embedded in a <em>python </em>architecture makes the module usable in a simple computing infrastructure. This code, called AIRIK (as AtmospheIc dispeRsIon Kernel), is freely available and allows for rapid dispersion analysis of multiple emission sources or to inform metamodels with more realistic dispersion conditions. This approach opens new developments to elaborate more physical convolutional filters in neural network architectures for atmospheric dispersion.</p>
title A physics-based Gaussian kernel for air pollutant dispersion over large Eulerian grids
topic modelling
dispersion
Gaussian
air quality
url https://doi.org/10.5281/zenodo.18212372