Parameter Sensitivity and Uncertainty Analysis in Urban Drainage Models

(2006 – current)
Cintia B. S. Dotto

A large number of stormwater pollution generation models are currently used worldwide. Regardless of whether they are physically based or purely statistical, they have low accuracy and a high level of uncertainty. However, little is understood about the sources and magnitude of this uncertainty. The aim of this PhD project is to study uncertainty in stormwater pollution and treatment models in order to define their reliability for different domains and applications. The specific objectives of the research are to assess the total model uncertainties and the relative importance of the main sources of uncertainties on the models and to recommend how to minimize uncertainty in stormwater modelling.

Extensive work on the performance and sensitivity analysis of stormwater models with different levels of complexity has been undertaken by means of a Bayesian approach and long-term high resolution data. We are currently developing and implementing a novel and simple approach for global assessment of model errors by cross-relating the different sources of uncertainty and propagating them through the stormwater models.


Project Partner:

Prof Ana Deletic and Dr David McCarthy

Publications (link)

Implementation Models Program