Nicolas Lauzon

Hybrid fuzzy-mechanistic models for addressing parameter variability

A simulation approach which integrates mechanistic models and fuzzy logic in order to accommodate parameter variability is developed and explored. The approach modifies a mechanistic model, such as a runoff model, for which the values of the parameters are originally fixed. Fuzzy logic is used to redefine the parameter values by varying them as a function of meaningful system indicators, such as inflow, precipitation or temperature in the case of inflow modelling. The modification adds flexibility to the structure of the mechanistic model, by allowing the values of the parameters to be reset at every time step based on the current values of the system indicators. This approach is applied to two different models, a runoff model and an algal concentration model, in order to demonstrate its versatility. The results are indicative of improved performance with the hybrid fuzzy-mechanistic models compared with the purely mechanistic models. In the case of the runoff model, the resulting description of the parameter domain also indicates a possible deficiency of the model structure, that is, a lack of clear distinction between watershed runoff and water retention through routing. The approach may be data intensive, but its implementation is straightforward. A wide range of potential applications of this approach in environmental and natural resources descriptive modelling exists, including: snowmelt modelling, fish habitat modelling, transport modelling, and species migration modelling. However, one must be careful to identify parameter-system indicator relationships that are representative of the system under study, and to avoid extrapolations beyond the known system conditions.

Artificial Intelligence Techniques as Detection Tests for the Identification of Shifts in Hydrometric Data

This paper presents the development of tests based on one artificial intelligence technique, the Kohonen neural network, for the detection of shifts in hydrometric data. Two new Kohonen-based detection tests are developed, the classification and mapping tests, and their performance is compared with that of well-known conventional detection tests. The efficacy of the tests is demonstrated with synthetic data, for which all the statistical properties and induced shifts are known. These synthetic data are designed to replicate hydrometric data such as annual mean and maximum streamflow. The results show that all tests, conventional and Kohonen based, may be considered equally reliable. However, no one test should be used alone because all generate false diagnostics under different circumstances. Within a decision support environment, a pool of tests may be used to confirm or complement one another depending on their known strengths and weaknesses. The Kohonen-based detection tests also perform well when applied to multivariate cases (i.e., testing more than one data sequence at a time), and their performance for multivariate cases is better than that for the univariate cases.