Do we need trees in Treatment Wetland models ?
2024
Guillaume, Sophie | Pueyo-Ros, J | Comas, J. | Forquet, N | Réduire, valoriser, réutiliser les ressources des eaux résiduaires (UR REVERSAAL) ; Institut National de Recherche pour l’Agriculture, l’Alimentation et l’Environnement (INRAE) | Instituto Catalán de Investigación del Agua = Catalan Institute for Water Research (ICRA) ; Universitat de Girona = University of Girona (UdG) | INRAE | ODE (Office de l'eau de la Martinique) | Caribbean Water & Sewerage Association Inc | Hupanam
International audience
Afficher plus [+] Moins [-]anglais. Treatment Wetlands (TW) are employed in the treatment of wastewater, participating in a decentralised and sustainable urban water management. Their designs and usages vary greatly across the world, which limit the aggregation of comparable data for deeper understanding of the mechanisms ruling over water treatment. We developed a hybrid modelling methodology to combine mechanistic and empirical pre-sizing models, applied to a context of low amount of data and knowledge on Horizontal and Vertical Flow TW. We first collected data from the scientific literature and introduced a data quality validation step providing reliability weights associated with each observation. We secondly trained and compared machine-learning models on the assessed dataset. We finally tested two hybridizations with the tank-in-series model, and predicting an optimized surface with a uncertainty interval. This methodology is reproducible and we believe hybrid models can provide more accurate and reliable predictions, therefore facilitating implementations of TW in urban plannings.
Afficher plus [+] Moins [-]Mots clés AGROVOC
Informations bibliographiques
Cette notice bibliographique a été fournie par Institut national de la recherche agronomique
Découvrez la collection de ce fournisseur de données dans AGRIS
