Machine Learning Analysis of Hydrologic Exchange Flows and Transit Time Distributions in a Large Regulated River

Hydrologic exchange between river channels and adjacent subsurface environments is a key process that influences water quality and ecosystem function in river corridors. High-resolution numerical models were often used to resolve the spatial and temporal variations of exchange flows, which are computationally expensive. In this study, we adopt Random Forest (RF) and Extreme Gradient Boosting (XGB) approaches for deriving reduced order models of hydrologic exchange flows and associated transit time distributions, with integrated field observations (e.g., bathymetry) and hydrodynamic simulation data (e.g., river velocity, depth). The setup allows an improved understanding of the influences of various physical, spatial, and temporal factors on the hydrologic exchange flows and transit times. The predictors also contain those derived using hybrid clustering, leveraging our previous work on river corridor system hydromorphic classification. The machine learning-based predictive models are developed and validated along the Columbia River Corridor, and the results show that the top parameters are the thickness of the top geological formation layer, the flow regime, river velocity, and river depth; the RF and XGB models can achieve 70% to 80% accuracy and therefore are effective alternatives to the computational demanding numerical models of exchange flows and transit time distributions. Each machine learning model with its favorable configuration and setup have been evaluated. The transferability of the models to other river reaches and larger scales, which mostly depends on data availability, is also discussed.

‘River Water DO Mechanics & New DO-Sag Equation’

DO modeling by Streeter Phelps equation [1] is most popular method to determine the water quality of a River. To compute DO by Streeter Phelps equation River coefficients k1 and k2 (de-oxidation and re-oxygenation) are required. Determination of these coefficients is tedious because it requires field observation of river velocity and depth over a long period of time at river site. To avoid maximum field work in calculating DO of River water DO Modeling approach is developed by combining Lab analysis of water samples DO with field data, e.g. river velocity and depth. Streeter Phelps (1925) developed the 1st important water quality model describing the BOD-DO relationship in a stream. In their pioneering work the simplest system was considered, in which biodegradable waste is discharged to the stream and consumes oxygen, atmospheric reaeration being the only source of oxygen. The model is based on complicated solution of differential equation for above process. The equation is derived assuming River coefficients k1 and k2 as exponential function of time variation. The authors have simplified the derivation of DO-Sag equation [4] by replacing the exponential function with a quadratic polynomial. To explain the use of new equation, authors have defined the geometry of DO curve known as ‘River Water DO Mechanics’. Also in this paper, new equation is applied to make ‘Shivnath River water DO Model’ with data taken by the author as part of his Ph. D. research work. The results justify the acceptance of new modified equation for River Water Quality Assessment.