Skip to Main content Skip to Navigation

Suivi et modélisation à haute résolution des flux hydriques d’une toiture végétalisée

Abstract : Green roofs are multifunctional type of Nature-Based Solutions that provide different ecosystem services among which the reduction and detention of the urban drainage outflow are the most important from the aspects of hydrology and stormwater management. As for various scientific fields, the issue of scales also appears as rather important scientific question in case of hydrology, and thus in case of green roofs. The idea behind it is to find a proper way of treating spatio-temporal variabilities of different processes involved in green roofs at larger scales, without masking heterogeneity characteristic for smaller scales. This is rather important for green roof designers, since the homogenization (averaging) in both space and time domain can impact the results of modeling significantly, providing unreliable insight into the hydrological performances of green roofs. This way, predictions of hydrological responses at larger urban (sub)catchment scales are also affected, which prevents from meeting regulation rules adopted by local authorities in charge of stormwater management.In order to improve reliability of hydrological predictions, various thorough investigations were performed on Green Wave, a green roof of the Bienvenüe building located close to Ecole des Ponts ParisTech, in suburban area of Paris. Firstly, different physical properties of the Green Wave substrate were measured in laboratory (specimen scale). The laboratory investigation of the hydraulic properties of the unsaturated / saturated Green Wave substrate, were carried out by means of the newly developed apparatus and the innovative methodology for determination of the hydraulic conductivity function.Furthermore, on the specimen scale, spatial variability of the soil density field obtained using X-ray CT scanner is analyzed using Universal Multifractals, a theoretical framework convenient for characterizing both spatial and temporal variabilities of different geophysical fields. As a result of the investigation, new methodology and analytical functions for describing different soil properties such as the grain / pore size distribution, water retention curve and the hydraulic conductivity function, are derived. The obtained analytical functions proved to be able to interpret rather well the experimentally determined properties of the Green Wave substrate, and other soil types taken from the literature.On the green roof scale, in-situ conditions were investigated using detailed monitoring system installed on Green Wave, where three main water balance components are measured: rainfall rate, water content indicator and drained discharge. Results showed that based on the multifractal analysis of temporal variabilities of three mentioned components, where the indicator of water content is measured by means of the network of TDR sensors distributed along the roof slope, it is possible to go beyond the standard investigation of the rainfall-runoff ration and to analyze the impact of roof inclination on the lateral water movement within the substrate. The mentioned analysis showed that the roof inclination does not affect the peak outflow, allowing development of a new one-dimensional analytical hydrological model.The proposed model is based on a cascade of non-linear reservoirs, where the leakage from each reservoir is described by means of the analytical function of hydraulic conductivity, also developed in this work. The model was proved as an adequate alternative for numerical solving of Richards equation in terms of accuracy and reliability, but also as a significant improvement from the aspect of computational efficiency. As such, it can be further used to efficiently treat spatial heterogeneity of green roofs at the scale of a single roof and larger, allowing reliable investigation of hydrological impacts of this type of Nature-Based Solutions on the urban catchment scale.
Document type :
Complete list of metadata
Contributor : ABES STAR :  Contact
Submitted on : Thursday, May 27, 2021 - 11:06:09 AM
Last modification on : Tuesday, June 1, 2021 - 3:20:56 PM
Long-term archiving on: : Saturday, August 28, 2021 - 6:39:48 PM


Version validated by the jury (STAR)


  • HAL Id : tel-03238705, version 1



Filip Stanic. Suivi et modélisation à haute résolution des flux hydriques d’une toiture végétalisée. Hydrologie. Université Paris-Est, 2020. Français. ⟨NNT : 2020PESC1012⟩. ⟨tel-03238705⟩



Record views


Files downloads