Solving river engineering problems typically requires river flow characterization, including the prediction of flow depth, flow velocity, and flood extent. Hydraulic models use governing equations of the flow in motion (conservation of mass and momentum principles) to predict the flow characteristics. However, solving such equations can be substantially expensive, depending upon their spatial extension. Moreover, modeling two- or three-dimensional river flows with high-resolution topographic data for large-scale regions (national or continental scale) is next to impossible. Such simulations are required for comprehensive river modeling, where a system of connected rivers is to be simulated simultaneously. Machine Learning (ML) approaches have shown promise for different water resources problems, and they have demonstrated an ability to learn from current data to predict new scenarios, which can enhance the understanding of the systems. The aim of this paper is to present an efficient flood simulation framework that can be applied to large-scale simulations. The framework outlines a novel, quick, efficient and versatile model to identify flooded areas and the flood depth, using a hybrid of hydraulic model and ML measures. To accomplish that, a two-dimensional hydraulic model (iRIC), calibrated by measured water surface elevation data, was used to train two ML models to predict river depth over the domain for an arbitrary discharge. The first ML model included a random forest (RF) classification model, which was used to identify wet or dry nodes over the domain. The second was a multilayer perceptron (MLP) model that was developed and trained by the iRIC simulation results, in order to estimate river depth in wet nodes. For the test data the overall accuracy of 98.5 percent was achieved for the RF classification. The regression coefficient for the MLP model for depth was 0.88. The framework outlined in this paper can be used to couple hydraulics and ML models to reduce the computation time, resources and expenses of large-scale, real-time simulations, specifically for two- or three-dimensional hydraulic modeling, where traditional hydraulic models are infeasible or prohibitively expensive.

This research presents a fully automated framework for runoff estimation, applied to Philadelphia, Pennsylvania, a major urban area. Trends in global urbanization are exacerbating stormwater runoff, making it an increasingly critical challenge in urban areas. Understanding the fine-scale spatial distribution of local flooding is difficult due to the complexity of the urban landscape and lack of measured data, but it is critical for urban management and development. A one-meter resolution Digital Elevation Model (DEM) was used in conjunction with a model developed by using ArcGIS Pro software to create urban micro-subbasins. The DEM was manipulated to account for roof drainage and stormwater infrastructure, such as inlets. The generated micro-subbasins paired with 24-h storm data with a 10-year return period taken from the National Resources Conservation Service (NRCS) for the Philadelphia area was used to estimate runoff. One-meter land-cover and land-use data were used to estimate pervious and impervious areas and the runoff coefficients for each subbasin. Peak runoff discharge and runoff depth for each subbasin were then estimated by the rational and modified rational methods and the NRCS method. The inundation depths from the NRCS method and the modified rational method models were compared and used to generate percent agreement, maximum, and average of inundation maps of Philadelphia. The outcome of this research provides a clear picture of the spatial likelihood of experiencing negative effects of excessive precipitation, useful for stormwater management agencies, city managers, and citizen.

Urbanization typically results in increased imperviousness which alters suspended sediment yield and impacts geomorphic and ecological processes within urban streams. Therefore, there is an increasing interest in the ability to predict suspended sediment yield. This study assesses the combined impact of urban development and increased precipitation on suspended sediment yield in the Cuyahoga River using statistical modeling. Historical satellite-based land-cover data was combined with precipitation and suspended sediment yield data to create a Multiple Linear Regression (MLR) model for the Cuyahoga watershed. An R² value of 0.71 was obtained for the comparison between the observed and predicted results based on limited land-use and land-cover data. The model also shows that every 1 mm increase in the mean annual precipitation has the potential to increase the mean annual suspended sediment yield by 860 tons/day. Further, a 1 km² increase in developed land area has the potential to increase mean annual suspended sediment yield by 0.9 tons/day. The framework proposed in this study provides decision makers with a measure for assessing the potential impacts of future development and climate alteration on water quality in the watershed and implications for stream stability, dam and flood management, and in-stream and near-stream infrastructure life.