Le filtre d’ensemble de Kalman

Table des matières

1.Introduction
1.1 Problématique générale
1.2 Présentation de la zone d’étude
1.2.1 Géographie et morphologie
1.2.2 Climatologie et hydrologie
1.3 Système de prévision opérationnel actuel
2.Revue de littérature
2.1 L’utilité de la prévision d’ensemble en hydrologie
2.2 La valeur économique de la prévision et l’aversion au risque
3.Méthodologie générale 
3.1 Le filtre d’ensemble de Kalman
4.Moving beyond the cost-loss ratio : Economic assessment of streamflow forecast for a risk averse decision maker 
4.1 Abstact
4.2 Introduction
4.3 The economic model and the limits of the cost-loss ratio
4.4 Context
4.4.1 Floods on the Montmorency Watershed
4.4.2 Current forecasting and decision-making process
4.4.2.1 The hydrologicalmodel HYDROTEL
4.4.2.2 Flood alerts
4.4.3 A concurrent flood forecasting framework based on meteorological ensemble forecasts
4.4.3.1 Meteorological ensemble forecasts
4.4.3.2 Data assimilation and state variables uncertainty
4.5 Parametrization of the economic model
4.5.1 Level of risk aversion
4.5.2 Damages d, spending s, and damage reduction b
4.5.3 Warning time and dynamic decision-making
4.6 Performance assessment
4.6.1 Forecast quality
4.6.2 Evaluating the benefits of forecasts
4.7 Results
4.7.1 Assessment of hydrological forecasts relative to observations
4.7.2 Assessment of hydrological forecasts in terms of economic value
4.8 Discussion
4.9 Conclusions
4.10 Appendix A: How the cost-loss ratio implies risk-neutrality
4.11 Appendix B: Properties of the CARA utility function
4.12 Appendix C: Simulation procedure
4.13 Competing interests
4.14 Acknowledgements
Bibliography
5.Synthèse et conclusion