Abstract
PURPOSE City of Atlanta Department of Watershed Management (DWM), through its Clean Water Atlanta (CWA) Program, is required by the ongoing First Amended Consent Decree (FACD) to develop Remedial Measures Reports (RMRs). RMRs are planning and conceptual design documents providing the basis of design for detailed design and construction plans for capacity relief and sewer structural rehabilitation projects. In parallel, DWM developed Watershed Improvement Plans (WIPs) for major watersheds in the City. The WIPs identified stormwater projects to improve water quality and reduce the volume of stormwater discharged to the city's creeks. Custer Avenue, Clear Creek, and Intrenchment Creek are three of Atlanta's combined sewersheds that collectively are served by more than 707,000 linear feet (134 miles) of combined sewers. To reduce CSOs and minimize urban flood risk, the RMRs identified approximately 164,000 linear feet of sewer requiring either capacity relief or sewer rehabilitation with an estimated capital cost of $201.1 million. The RMR defined projects were collectively referred to as 'gray' infrastructure projects. Concurrently, approximately 1,100 stormwater water quality and control infrastructure projects were identified city-wide, through the WIP process, at a total estimated capital cost of approximately $600 million. Collectively WIP projects were termed 'green' infrastructure projects. Considering the significant capital costs of the identified green and gray projects, DWM requested that the Program Management Services Team (PMST), under the lead of Stantec Consulting Services, undertake a project to answer three simple questions: 1. Would it be possible to integrate green infrastructure with the planned gray infrastructure projects to address combined sewer overflow (CSO) and sewer flooding problems in three combined sewersheds? 2. Would it be possible to reduce the capital cost of current RMR plans without impacting future levels of service? 3. Would solution optimization modeling present an opportunity to increase environmental and community benefits in line with wider DWM 'One Water' objectives? METHODOLOGY To answer the questions all applicable green and gray projects to DWM's existing InfoWorks ICMTM hydraulic and hydrologic models. Green projects that would not directly influence combined sewer flows were omitted. Only green projects, if implemented strategically, that could have a beneficial outcome in reducing CSO activations and reducing urban flood risk were included. PMST and DWM have utilized the hydraulic and hydrologic models for a wide range of engineering planning and design activities for more than 20 years and, as a result, the models are well populated, reliable, detailed, and were able to provide an excellent level of detail to start the optimization. The updated InfoWorks ICMTM models were then imported into a cloud-based optimization software, OptimizerTM. OptimizerTM is a genetic algorithm software designed to solve complex multi-criteria problems faster and more efficiently favoring speed of process over absolute accuracy. For this project OptimizerTM was applied to analyze hundreds of thousands of green and gray project combinations to establish the 'best outcome'. After establishing a baseline plan, OptimizerTM varied the size, scale, and combination of all possible solutions to create new plans for comparison against the baseline using three metrics: capital cost, hydraulic performance, and non-cost benefit. Each metric was applied as detailed: Capital Cost. The incremental capital cost of each project was calculated and entered OptimizerTM as a lookup table. During optimization, as project scales and sizes adjust, so do the costs. For gray projects, the increments related to pipe diameter, pipe depth, storage tank volume, etc. and where applicable, construction technique. For green projects, the increments related to the relative size of the solution based on its ability to capture an amount of rainfall runoff volume using rainfall depth and contributing area. Hydraulic Performance. Hydraulic performance was measured by applying hydraulic penalty scores to threshold exceedances in the model's predicted hydraulic grade line (HGL) and flood volume. Penalties were proportional to the exceedances and applied to to every combination of gray and green projects, OptimizerTM tried to achieve the best hydraulic performance by converging on the lowest total hydraulic penalty score. Co-benefits. Infrastructure value is an important criterion for DWM, this metric was designed to evaluate the value of all gray and green infrastructure to ensure OptimizerTM decisions considered the value of all projects. Co-benefits ensured consideration of social, environmental, and economic indicators were included in determining the best outcome. The principles applied to all projects was consistent with DWM's Triple Bottom Line (TBL) categories across all watershed initiatives. With all project information, including metrics uploaded to OptimizerTM, the optimization process was repeated more than 100,000 times to continually seek convergence, each time attempting to find a 'better' green / gray plan based on the metrics. RESULTS Unlike many models, OptimizerTM creates several potential outcomes and there were many combinations of green and gray plans that offered a broad range of possibilities. Some plans performed hydraulically better, some less expensive, and some realized much greater co-benefits; however, no single plan was deemed to be 'optimal'. Results were presented in the form of a three-dimensional (3-D) Pareto graph from which a subset of potential plans were selected. The dimension axis representing capital cost, hydraulic performance, and co-benefits. The 3-D Pareto graph showing possible plans is depicted in Figure 1. DWM were able to select a preferred plan for each combined sewershed. The selected plans and their relative costs are summarized in Table 1. CONCLUSIONS In conclusion, the Green/Gray Optimization project was able to positively answer all three stated questions. 1. Would it be possible to integrate green infrastructure with the planned gray infrastructure projects to address CSO and sewer flooding problems in three combined sewersheds? Yes, it is possible. The project demonstrated there were effective green alternatives for all three combined sewersheds with WIP projects making up 4% of the total capital cost in Intrenchment Creek, 6% in Clear Creek and 31% of the total capital project cost in Custer Avenue. 2. Would it be possible to reduce the capital cost of current RMR plans without impacting future levels of service? Yes, in all three combined sewersheds, cost estimates were lower. Custer Avenue's selected plan was 39% less expensive; Clear Creek's was 16% less expensive; and Intrenchment Creek's was 52% less expensive. In all cases, the hydraulic performance of the selected plans was better than the original RMR plan. 3. Would solution optimization modeling present an opportunity to increase environmental and community benefits in line with wider DWM 'One Water' objectives? Without question, the inclusion of green projects adds enhanced environmental and community benefits to the overall outcomes for DWM and better aligns with City wide objectives to move to a 'One Water' approach that values all water. Further, these co-benefits are added without compromising sewer system performance.
This paper was presented at the WEF Collection Systems Conference in Detroit, Michigan, April 19-22.
Author(s)N. Anderson1; P. Flynn3
Author affiliation(s)Stantec Consulting Inc1; City of Atlanta Department of Watershed ManagementStantec3
SourceProceedings of the Water Environment Federation
Document typeConference Paper
Print publication date Apr, 2022
DOI10.2175/193864718825158366
Volume / Issue
Content sourceCollection Systems
Copyright2022
Word count18