Optimization Programming for Stormwater Control Measures: Methods for Sizing and Selection

The Study Problem and Research Objectives. Fouling of water quality in receiving urban storm runoff is chronic in metropolitan areas across the USA and large cities worldwide. These urban areas have well-known problems of polluted storm runoff and urban flooding. Urban storm runoff exhibits deleterious physical-chemical-biological characteristics, such as bacteria, trash content, large biochemical-oxygen-demand (BOD), oil & grease, toxic sediments, water-borne pathogens, suspended solids (SS) and total dissolved solids (TDS), heavy metals, and nutrient content that degrade water quality of receiving waters. The contamination of urban runoff is the result of a number of natural and anthropogenic processes: changing and rapid expanding population and land use within urban areas; vulnerable receiving water bodies that become contaminated from degraded storm runoff that hinders their hydrologic, ecologic, and socioeconomic functions, and intense rainfall events with pronounced seasonal and inter-annual variability of storm intensity (typical in the western United States). State and federal regulations on Total Maximum Daily Loads (TMDLs) of pollutants to natural waters from urban storm runoff are not met by most cities in the USA. This thesis presents a modeling and experimental study using one of the best available data sets on urban land use, soils, groundwater, streets, storm conveyance infrastructure, non-point and point sources of pollution, rainfall, and Stormwater Control Measures (SCMs) technologies. This thesis proposes a novel approach to (1) model, screen and or evaluate urban areas using Geographic Information Systems (GIS) with the purpose of selecting appropriate SCMs in watershed hot spots, (2) select suitable SCMs to be deployed for the capture of and treatment or retention of urban runoff using Optimization Programming, (3) implement and test the proposed research in a large urban area with pervasive urban runoff pollution, and (4) field test SCMs for which there is limited or no information on treatment efficiency to assess their runoff-cleaning potential. Optimization methods are developed and presented in this research work to minimize the total cost of SCM deployment while satisfying constraints on (i) the total cost of deployment, (ii) SCM capacities, (iii) volumetric balance at SCM sites, (iv) stormwater volumes at arbitrary sites, (v) unit Operational, Maintenance, and Replacement (OMR) cost, and (vi) water-quality and quantity at monitoring locations. Two alternative optimization models for SCM siting and sizing are presented in this thesis:• Linear Programming (LP) for optimal sizing of SCMs relies with a linear programming formulation. In addition, the Binary Linear Integer Programming (BLIP) for optimal selection of SCMs based on a binary (0,1) linear integer programming formulation.• Nonlinear Programming (NLP) for optimal sizing and selection of SCMs uses mixed (binary-real) nonlinear integer programming formulation.Summary of the Research Method. The key tasks accomplished in this thesis are: • Modeling and determination of priority catchments for SCM deployment based on SCM;• Development and application of a SCM Optimization programing model to select from within the GIS-evaluated SCMs those that are most cost effective in reducing the pollutant loads and concentration in urban runoff; • Nonlinear model’s results have been field tested by implementing its SCMs selection with site-specific data;• Conducted field experiment to evaluate the pollution removal efficiency of percolation wells and vegetated swales. The implementation of GIS-based electronic maps to classify areas according to their conditions: heavy traffic, impervious soils, high rainfall intensity for design storms, high urban density, and steep topography. The selected SCMs are tested in chosen priority catchments to assess model-predicted performance with field performance. The research task consists of field-experimenting with dry wells and vegetated swales. These types of SCMs appear to have good performance potential in permeable soils, and swales exhibits high aesthetic distinction and value. These SCMs are field tested to determine their pollutant removal efficiency for selected indicator pollutants (for example, total suspended solids). Broad Impacts of the Proposed Research. This thesis overarching hypothesis is that the sequential application of (1) computer based modeling in screening of high priority urban catchments and cost effective SCMs, and (2) optimization programming, can be used successfully in (i) identifying catchments with high SCM indices to urban runoff pollution and (ii) selecting the most cost-effective SCMs to reduce runoff pollution. This novel research concept represents a trend setting approach in combating urban runoff pollution in the United States, and in other places where the resources and know-how for urban runoff pollution control are in high demand and required. Intellectual Merit of the Research. This is a novel research attempt to develop and integrate novel analysis of pollution and suitability for SCM deployment with optimization programming method for the selection of the types and sizes of SCMs in urban catchments