INTRODUCTION - Modelling Uncertainty in Flood Forecasting Systems - page 9

• To explore the advantages and possibility of hybrid techniques of uncertainty
modelling and probability-possibility (or fuzzy) transformations;
• To develop necessary computer codes to implement the developed methodology
and to validate them by application to real-world problems.
1.3.4 Application examples
In recent years the European Commission (EC) has funded several projects involving
studies of river basin and flood crisis management and mitigation including techniques
for flood forecasting, warning and information dissemination. Some examples of recent
EC funded projects are AFORISM (A Comprehensive Forecasting System for Flood Risk
Mitigation and Control), 1991-1994; EUROflood, 1992-1994; RIBAMOD (River Basin
Modelling, Management and Flood Mitigation), 1996-1998; TELEFLEUR (Telemetric
Assisted Handling of Flood Emergencies in Urban Areas), 1998-2000; EFFS (European
Flood Forecasting System), 2000-2003; and OSIRIS (Operational Solutions for the
Management of Inundation Risks in the Information Society), 2000-2003.
The significance of uncertainty management in flood crisis was realised in the OSIRIS
project, which resulted in the development of uncertainty assessment modules for the
pilot flood forecasting systems of the project. Most of the present research was financed
by this project. The application examples used in the present research are for two of the
three pilot sites of this project namely, flood forecasting models for the Klodzko
catchment in Poland and the Loire River in France.
1.4 Structure of the thesis
Chapter 2 reviews mathematical models used in flood forecasting and sheds some light on
the issue of uncertainty assessment of the outputs from these models. This chapter also identifies
uncertainty types and sources with respect to flood forecasting and reviews the commonly used
uncertainty representation theories. Chapter 3 summaries methods of uncertainty analysis based
on probability theory and fuzzy set theory and also reviews methods of probability-possibility
transformations.
Chapter 4 presents the contribution of the present research in the field of uncertainty
estimation in general and uncertainty in flood forecasting in particular. This involves the
development of a methodology for the treatment of uncertainty due to time series inputs
in flood forecasting (based on Monte Carlo technique and the fuzzy Extension Principle),
an improvement to the existing first-order second moment method, development of
qualitative uncertainty scales using best-case and worst-case scenarios and results of an
investigation on hybrid techniques for modelling uncertainty and probability-possibility
(or fuzzy) transformations. Chapters 5 and 6 present the application of the developed
techniques in operational flood forecasting systems for the Klodzko catchment (Poland)
and the Loire River (France), respectively. Chapter 7 is devoted to conclusions and
recommendations. Appendix I includes definitions of fuzzy sets, defuzzification methods
and fuzzy arithmetic. References are included at the end of the thesis.