Hydrometric network design using dual entropy multi-objective optimization in the Ottawa River Basin

Abstract

Water resources managers commonly rely on information collected by hydrometric networks without clear knowledge of their efficiency. Optimal water monitoring networks are still scarce especially in the Canadian context. Herein, a dual entropy multi-objective optimization (DEMO) method uses information theory to identify locations where the addition of a hydrometric station would optimally complement the information content of an existing network. This research explores the utility of transinformation (TI) analysis, which can quantitatively measure the contribution of unique information from a hydrometric station. When used in conjunction, these methods provide an objective measure of network efficiency, and allow the user to make recommendations to improve existing hydrometric networks. A technique for identifying and dealing with regulated basins and their related bias on streamflow regionalization is also examined. The Ottawa River Basin, a large Canadian watershed with a number of regulated hydroelectric dams, was selected for the experiment. The TI analysis approach provides preliminary information which is supported by DEMO results. Regionalization was shown to be more accurate when the regulated basin stations were omitted using leave one out cross validation. DEMO analysis was performed with these improvements and successfully identified optimal locations for new hydrometric stations in the Ottawa River Basin.

Entropy Applications to Water Monitoring Network Design: A Review

“…They compared the performance of various regionalization methods including not only a conceptual hydrologic model, but also spatial proximity, physical similarity and their combinations with drainage area ratio. Based on the performance statistics by applying multiple basins, inverse distance weighting coupled with drainage area ratio performed the best, and this conclusion has been adopted in several studies [34,35,37,55].…”

Assessing Spatial Scale Effects on Hydrometric Network Design Using Entropy and Multi‐objective Methods

“…The years 2001-2010 were selected for the analysis. This time period was chosen in order to facilitate comparisons to Werstuck and Coulibaly (2017) where a similar analysis was done on the entire ORB. A 10-year time period was determined to be the recommended length for CR-DEMO analysis of daily streamflow series in Keum and Coulibaly (2017).…”

“…A total of 79 distinct subcatchments were delineated in the watershed. This is more than the seven subcatchments which were identified in the Madawaska Watershed in the previous study (Werstuck and Coulibaly, ) because focusing on a smaller study area allows for finer delineation which would otherwise be difficult due to computational limitations. A number of subcatchment delineations were tested, and this size was chosen because it appeared to give the potential station locations good spatial coverage.…”

“…Combined regionalization dual entropy multi-objective optimization (CR-DEMO), introduced by Samuel et al (2013), has recently been adopted as a method of network augmentation because of its flexibility and robust optimization algorithm. Recently, another information theory-based method called transinformation (TI) analysis has been used in conjunction with CR-DEMO to assess the stations in an existing network and rank them based on their information contributions (Mishra and Coulibaly, 2014;Werstuck and Coulibaly, 2017). This offers another technique for measuring information density in a basin which can be compared to CR-DEMO analysis.…”

“…This research builds on the CR-DEMO method introduced in Samuel et al (2013) and used in Leach et al (2015) and Werstuck and Coulibaly (2017) by assessing the scaling properties of CR-DEMO and modifying the methods for operation on a watershed with limited information. This study highlights the nuances of CR-DEMO operation which should be considered by water resources engineers and hydrologists who plan to use these emerging techniques for evaluating and/or augmenting their hydrometric networks.…”