Soutenance de thèse : Santiago SANDOVAL
Revisiting stormwater quality conceptual models in a large urban catchment: online measurements, uncertainties in data and models
Doctorant : Santiago SANDOVAL
Laboratoire INSA : DEEP
Ecole doctorale : ED162 : Mécanique, Énergétique, Génie Civil, Acoustique (MEGA)
Different hypotheses about traditional Total Suspended Solids (TSS) stormwater models from data validation and hydrological concepts have been explored in a 185 French urban catchment, including online data from 365 rainfall events. Four sampling strategies during rainfall events are simulated and compared to online monitoring. Recommended sampling time intervals are of 5 min, with average sampling errors between 7 % and 20 % and uncertainties in sampling errors of about 5 %, depending on the sampling interval. The probability of underestimating the cross section mean TSS concentration in the sewer system is estimated by two methodologies, were one shows more realistic TSS underestimations (about 39 %). A power law describing the TSS as a function of flow rate is revealed. In the hydrological context, a parameter estimation strategy is proposed by analyzing the variability of parameters obtained by event-based Bayesian calibrations, based on clusters and graphs representations. A single model structure might be able to reproduce at least two different hydrological conditions. A methodology aimed to calculate “mean” areal rainfall estimation is proposed, based on a hydrological model and flow rate data. Regarding TSS modelling, the performance indicators of the traditional Rating Curve (RC) model are superior to different transfer Functions (TFs), with flow rate, rainfall or “mean” areal rainfall as the model input. The potential missing representation of an essential process in the RC model is found to be independent of antecedent dry weather period. A Bayesian reconstruction method indicates that a potential missing process in the RC description is hardly interpretable in terms of a unique state of virtual available mass over the catchment that is decreasing over time, as assumed by a great number of traditional models. Furthermore, the reconstructed processes are highly unrepeatable regarding their shape, besides having a low transferability to other rainfall events.
Salle IOA - Bâtiment Freyssinet - GCU - INSA Lyon - Villeurbanne