Communication Dans Un Congrès Année : 2022

GSA of a green roof hydrological model with multivariate inputs and outputs

Résumé

In the last decade, soil imperviousness has been one of the main urban issues in NorthEurope. In case of strong rain events, runoff can lead to the discharge of high volume of water and can cause water system saturation. Among all urban-water regulation systems, Green Roofs (GR) can be used to store and delay the release of rainwater to sewers [3]. GR are also considered as a sustainable solution that offers benefits such as building insulation, urban heat island cooling during summer and air pollution control. Green roofs are composed of a substrate layer (ii) on top of which vegetation (i) is growing. Under the substrate, a geotextile (iii), a drainage layer (iv) and an insulating layer (v) protect the building. The hydrological performances are directly associated to the GR characteristics such as soil parameters of the different layers, dimension, type of vegetation, etc. The outflow of a GR is mainly related to the water content inside the layers. Few models exist to describe the hydrological infiltration throughout soil and they are based on the Richards’ equation. This highly nonlinear partial differential equation describes the water retention capacity and the outflow in un- saturated porous media. The Richards’ equation combined with the Van Genuchten - Mualem model are coded in Hydrus-1D© software to simulate hydrological behavior. This software allows the set up of the green roof structure, boundary conditions, meteorological data and vegetation variables in order to reproduce the green roof real conditions. However, some of the model parameters, as soil parameters or vegetation variables, are challenging to determine as they are difficult to measure accurately through experiments. All the parameter uncertainties are propagated to the water retention capacity simulated and need to be analysed with Global Sensitivity Analysis (GSA). A recent study has been dedicated to analyse the influence of soil parameters as saturated water content, porosity, etc. It has highlighted that three soil parameters of the substrate layer are more influential than the soil parameters of the other layers. These results allow to reduce the number of uncertain soil parameters for the following studies. Nevertheless, it also appears that the variability of the vegetation parameters as crop height and root depth, set to a constant value in this previous study, could have effects on the water retention capacity. In this proposed study, the GR model is dynamic and depends on static uncertain inputs, the soil parameters and on time-varying uncertain inputs, the vegetation parameters. Indeed, a one year period is analysed in order to observe different hydrological phenomenon, and the vegetation parameters as crop height or root depth vary over the seasons. Boundary conditions, meteorological data and water retention data were measured from an in-situ experimental green roof platform located in Tomblaine (France) and are used to simulate water retention with real conditions. The objective of this study is to investigate the influence of soil and vegetation parameters on the water retention capacity over time with GSA approach. The challenge is to generate samples of vegetation parameters that satisfy the random fields distribution. The random fields are assumed independent and normally distributed, defined by their means and covariance functions. One possibility is to resort to the Karhunen-Loeve expansion. Then, to compute sensitivity indices for the time-varying model output, polynomials chaos expansion is applied with a sequential and PCA-based multivariate approach.

abstract_SAMO2022_HEGOAxelle.pdf (67.47 Ko) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)
Licence

Dates et versions

hal-03866160 , version 1 (22-11-2022)

Identifiants

  • HAL Id : hal-03866160 , version 1

Citer

Axelle Hego, Floriane Collin, Hugues Garnier, Rémy Claverie. GSA of a green roof hydrological model with multivariate inputs and outputs. 10th International Conference on Sensitivity Analysis of Model Output, SAMO 2022, Mar 2022, Tallahassee, United States. ⟨hal-03866160⟩
79 Consultations
48 Téléchargements

Partager

  • More