Updating inflow forecasts using empirical statistical matching for real-time prediction of daily net inflows to Okanagan Lake

Abstract

Accurate predictions of flow periods are important for decision-making within the Okanagan Lake System. A nonparametric method to predict the hydrograph to achieve a closer match with the timing and volume of reservoir inflows during the dominant flow period (February1 to July 31) in Okanagan Lake was developed in this study. The method employed, Real-Time Statistical Matching (RTSM), uses a combination of information from a changing suite of best-fit historical years, existing forecasts, and recent inflow trends. This included a comparison between the current hydrograph against hydrographs derived from historical inflows based on the predicted volume and pattern of the hydrograph. The RTSM-based approach is hypothesized to improve the ability of hydrological models to predict shifts in the general timing of peak net inflows. This makes the RTSM model more robust to both historic and non-historic conditions. The performance of the RTSM-based predictions was compared to the legacy hydrology model based on average timing of historic flows. Results indicate an improvement in predictive accuracy of 10%, 6%, and 80% for Nash-Sutcliffe Efficiency (NSE), root mean squared error to standard deviation ratio (RSR), and percent bias (PBIAS) respectively, which are three different measures of the accuracy of predictions. Further, the success of the Okanagan Fish/Water Management Tool (FWMT) relies on the water and fish managers that use the tool, which extends beyond the quantitative metrics in this study. The authors’ also discussed how the tool’s utility has changed over time from when it was put into practice. It was learned that in practice, the best use of the model was based on the volume-based prediction with the real-time adjustment.

RÉSUMÉ

Des prévisions précises des périodes de débit sont importantes pour la prise de décisions relatives au système du lac Okanagan. Dans le cadre de la présente étude, on a élaboré une méthode non paramétrique pour prévoir l’hydrogramme afin d’obtenir une meilleure correspondance avec le moment et le volume des débits entrants du réservoir durant la période de débit dominant (du 1^er février au 31 juillet) dans le lac Okanagan. La méthode employée d’appariement statistique en temps réel (Real-Time Statistical Matching, RTSM) utilise une combinaison de données provenant d’un ensemble changeant d’années passées qui correspondent le mieux, de prévisions existantes et de tendances récentes en matière de débits entrants. Cela inclut une comparaison entre l’hydrogramme actuel et les hydrogrammes dérivés des apports historiques basés sur le volume prévu et le schéma de l’hydrogramme. L’approche fondée sur le RTSM est supposée améliorer la capacité des modèles hydrologiques à prévoir les changements du moment général des pics nets de débits entrants. Cela rend le modèle RTSM plus robuste relatif aux conditions historiques et non historiques. Le rendement des prévisions basées sur le modèle RTSM a été comparé à celui du modèle d’hydrologie traditionnel fondé sur la chronologie moyenne des débits historiques. Les résultats indiquent une amélioration de la précision des prévisions de 10%, de 6% et de 80%, respectivement, par rapport au coefficient de Nash-Sutcliffe (CNS), au ratio de l’erreur quadratique moyenne sur l’écart-type (RSR) et le pourcentage du biais (PBIAS), qui sont trois mesures différentes de la précision des prévisions. En outre, le succès de l’outil de gestion de l’eau et des poissons de l’Okanagan (Okanagan Fish/Water Management Tool, FWMT) dépend des gestionnaires de l’eau et des poissons qui l’utilisent, ce qui va au-delà des mesures quantitatives de cette étude. Les auteurs ont également discuté de la façon dont l’utilité de l’outil a changé au fil du temps depuis sa mise en pratique. Ils ont appris que, dans la pratique, la meilleure utilisation du modèle était basée sur la prédiction fondée sur le volume avec l’ajustement en temps réel.

Keywords:

• multi-objective trade-off assessment
• environmental flow
• inflow prediction
• endangered species
• decision support system

Acknowledgements

We would like to dedicate this manuscript to one of our co-authors, the late Dr. Kim Hyatt (‘Dr. Sockeye’) who passed away May 25, 2021. His vision on the overarching need for improved daily net inflow forecasts and contributions to early versions of this manuscript were foundational. Kim was one of the key champions behind the success of the Canadian Okanagan Basin Technical Working Group and related efforts to re-introduce and restore Okanagan Sockeye including being the primary visionary behind the creation of FWMT. Kim is largely responsible for building greater awareness of the merits of improving real time water operations at Okanagan Lake Dam as well as many other efforts to recover Sockeye throughout the Pacific Northwest. He has etched a deep desire in many to continue being bridge builders and to uphold the optimistic view of what is possible when we give nature a chance.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

Development of RTSM procedures was undertaken as part of a larger project, generously funded by Douglas County Public Utility District (DCPUD) of Washington State and focused on modernizing the Okanagan Fish/Water Management Tool (FWMT) decision support system. The authors are grateful to Tom Kahler who steadfastly supported the FWMT modernization effort between 2015–2018. The Province of British Columbia also provided funding support specific to refinement and testing of the RTSM model. We also recognize the important contributions of FWMT operations team members who implemented and tested various versions of FWMT during the three-year modernization project, especially Margot Stockwell (ret. Fisheries and Oceans Canada), Dawn Machin and Kari Alex (Okanagan Nation Alliance).