Emergency Local-Scale Dispersion (ELSI) Software

0549701 - ÚT 2022 RIV DE eng C - Conference Paper (international conference)
Chaloupecká, Hana - Jakubcová, Michala - Kellnerová, Radka - Jaňour, Zbyněk - Jurčáková, Klára
Emergency Local-Scale Dispersion (ELSI) Software.
Air Pollution Modeling and its Application XXVII. Berlin: Springer, 2021 - (Mensink, C.; Matthias, V.), s. 295-299. Springer Proceedings in Complexity. ISBN 978-36-626-3759-3. ISSN 2213-8684. E-ISSN 2213-8692.
[nternational Technical Meeting on Air Pollution Modeling and its Application, ITM 2019 /37./. Hamburg (DE), 23.09.2019-27.09.2019]
R&D Projects: GA TA ČR(CZ) TJ01000383
Institutional support: RVO:61388998
Keywords : dispersion model * wind-tunnel modeling * puffs
OECD category: Meteorology and atmospheric sciences
https://link.springer.com/chapter/10.1007%2F978-3-662-63760-9_42

Operational softwares predicting a situation in case of short-term gas leakages are usually based on simple models because of time demands. But the project COST ES1006 revealed that predictions of these models for the short-term gas leakages can be as inaccurate as one order of magnitude. In contrast, predictions of simple models for continuous sources are not as much inaccurate. Hence, the aim of this paper is to introduce a new operational software for short-term gas leakages. The software is validated on data from experiments of short-term gas releases in a wind tunnel. The data describe the situation in different landscapes, a typical idealized European city centre and a rural area. The model of city centre was composed of buildings with pitched roofs organised into closed courtyards. The model of rural area was simulated by a surface roughness. The puff experiments were repeated a few hundred times for each measurement position at the models to get statistically representative datasets. Concentrations were measured by a fast flame ionisation detector. Ethane was utilized as a tracer gas. The introduced software uses mean concentrations calculated by a simple plume model as one of its inputs. The outputs are probability density functions of puff characteristics. This distinguishes the software from the usually utilized ones in which the outputs are only the ensemble-averaged puff outlines and concentration fields.
Permanent Link: http://hdl.handle.net/11104/0327749