2017 Progress Report: Performance and Effectiveness of Urban Green Infrastructure: Maximizing Benefits at the Subwatershed Scale through Measurement, Modeling, and Community-Based Implementation

EPA Grant Number: R835555
Title: Performance and Effectiveness of Urban Green Infrastructure: Maximizing Benefits at the Subwatershed Scale through Measurement, Modeling, and Community-Based Implementation
Investigators: McGarity, Arthur E , Hobbs, Benjamin F. , Rosan, Christina , Welty, Claire , Heckert, Megan
Institution: Swarthmore College , University of Maryland - Baltimore County , The Johns Hopkins University , Temple University
Current Institution: Swarthmore College , Temple University , The Johns Hopkins University , University of Maryland - Baltimore County
EPA Project Officer: Packard, Benjamin H
Project Period: October 1, 2013 through September 30, 2017 (Extended to September 30, 2018)
Project Period Covered by this Report: October 1, 2016 through September 30,2017
Project Amount: $1,000,000
RFA: Performance and Effectiveness of Green Infrastructure Stormwater Management Approaches in the Urban Context: A Philadelphia Case Study (2012) RFA Text |  Recipients Lists
Research Category: Watersheds , Water


(1) Evaluate selected Green Infrastructure (GI) demonstration projects in the Philadelphia CSO area; (2) develop methodology for creating zones of green infrastructure (ZGIs); (3) develop quantitative GI benefit-investment functions for each zone; (4) incorporate benefit functions into tools for use by municipal officials and community stakeholders; (5) incorporate STEM learning at multiple levels.

Progress Summary:

Task 1. Subsurface modeling and monitoring; outputs – refined methodologies for performance assessment of GI practices at the site and watershed levels; outcomes: increased national capability for assessing and implementing GI practices.  Continuation of groundwater modeling and instrumentation installation and maintenance focusing on three GI sites within the Wingohocking sewershed: Wakefield Park (rain garden), Waterview Recreation Center (permeable pavement), and Leeds School (tree trenches).  ParFlow.CLM is being used to model the coupled groundwater/surface water/land- atmosphere system at watershed and site scales.    Cores removed from auguring and well drilling is being characterized for hydraulic conductivity and total porosity.  ParFlow.CLM is being used to model the coupled groundwater/surface water/land- atmosphere system at GI sites.  The principal groundwater modeling effort in 2016-2017 was for the Waterview Recreation Center GI site; this work was completed and a manuscript on this work is in preparation (Barnes and Welty, to be submitted to ASCE J. Hydrologic Engineering).  In addition, C. Welty worked with Theo Lim at U. Penn to apply Parflow.CLM to a site in Washington, DC.  This work was published during the reporting period (Lim and Welty, 2017).  Refereed Journal Publications: 1 completed, 1 in preparation; Theses: 1 master’s thesis in preparation;  Conference proceedings and presentations: 6 completed. 

Task 2. Philadelphia Community Engagement through Research Partnerships; outputs – refined methodologies for evaluation of cost effective GI implementation strategies using a bottom-up modeling process; outcomes – enhanced prospects for widespread successful implementation of green infrastructure.  Four meetings of our GreenPhilly Community Advisory Research Board (GCARB) to date.  Over the summer 2016, the Overbrook Environmental Education Center worked with students from Philadelphia high schools on a summer educational program that used “green visual literacy”.  During the 2016-17 academic year, Art McGarity supervised an undergraduate senior thesis by Alexandra Philyaw’17 entitled Modeling and visualizing community benefits of green stormwater infrastructure in West Philadelphia.  Rosan and Heckert have continued to refine the GI Equity Index using the neighborhood around the Village of Arts and Humanities as a test case. In 2016-2017, they presented this research at several conferences and invited talks.  Refereed Journal Publications: 1 completed; Undergraduate Theses: 1 completed; Conference proceedings and presentations: 15 completed, 3 in preparation. 

Task 3. Develop benefit functions for Direct Benefits and Co-Benefits of Green Stormwater Infrastructure; outputs – refined methodologies for accurate life-cycle assessment of GI practices employing multiple benefit metrics including direct benefits and co-benefits; outcomes – increase national capabilities for implementing GI practices at the subwatershed level.  The Python interface to EPA’s Storm Water Management Model (SWMM) software, developed in years 1 - 3 of this project was extended to calculate time series (15 minute resolution) of CSO flow rates during storm events.  These time series were then used to calculate annual CSO volumes.   CSO flow rates are calculated using the positive differences between sewer flow rates at the sewershed outfall and a specified threshold flow rate.  The threshold is chosen using a calibration procedure to provide the best fit between simulated and published values of annual CSO flow volumes.  For our Wingohocking sewershed analyses, we replaced our previously-used seven subcatchment SWMM model with a 145 subcatchment SWMM runoff model provided by the SWMM modeling team at the Philadelphia Water Department (PWD).  We added sewer lines to this SWMM model using GIS layers also provided by PWD.  Our project consultant, AKRF, Inc. was tasked with continuing the developing cost models for use in development of GI benefit functions. The statistical regression model developed during Year 3 was updated with the most recent cost data available.  Underlying mechanisms that cause nonlinearities in CSO reduction response to GI deployment were explored and categorized.  A nonlinear statistical model for predicting annual CSO reductions resulting from GI deployment for the Wingohocking sewershed was incorporated into the StormWISE optimization formulation to solve for optimal combinations of the three specified GI technologies so as to achieve CSO reductions, over the entire feasible range, at minimum investment cost.  We have built on a partnership with the Overbrook Environmental Education Center in West Philadelphia to develop relationships with stakeholders in the community that have enabled us to build another case study for evaluating the application of the StormWISE framework to Philadelphia green infrastructure.  Our partnership with Overbrook stakeholders provides an opportunity to further develop the multiobjective components of StormWISE to include quantification of co-benefits into the analysis.  A prototype web app version of the multiobjective StormWISE model that includes co-benefits as well as CSO reduction benefits is now hosted on the Amazon Web Services cloud computing platform, and it will be used to interact with Overbrook stakeholders.  Refereed Journal Publications: 3 in preparation; Undergraduate Theses: 4 completed; Conference proceedings and presentations: 5 completed, 1 in preparation. 


Task 4. Develop a Stochastic and Multi-Stage StormWISE Model; outputs – refined methodologies for incorporation of uncertainties and an adaptive management process in modeling GI investment strategies; outcomes – increased national capabilities for cost-effective and adaptive implementation of GI practices for urban stormwater management at the subwatershed scale.  The model development subgroup consisting of investigators Hung, Hobbs, and McGarity have developed a methodology based on Two-stage Stochastic Programming with risk constraints and learning for Adaptive Stormwater Management. Conditional Value at risk (CVaR) is chosen the risk metric because of its favorable mathematical properties (coherent and convex).  The Wingohocking watershed is tentatively chosen as the case study site, building on our project’s progress in monitoring and modeling in this watershed to date. The hydrologic characteristics of the watershed will be derived from additional GIS analyses while the uncertainty of GI performance will be evaluated from additional runs of our EPA SWMM simulations. The results of the GIS analysis and SWMM simulation will be fed into the stochastic StormWISE as inputs in the form of coefficients and random distributions. This work will result in a manuscript to be submitted for publication in Spring 2018.  The relationship of GI performance to location and hydrological timing factors is also being investigated.  This analysis aimed at understanding the interactions between precipitation, watershed characteristics, GI siting, and the resulting runoff modifications.  First, we investigate key factors that affect the timing of runoff and basin discharge, including network travel time, time to fill-up GI, and rainfall duration. We then apply dimensionless analysis to characterize the effects of combinations of timing factors upon the generation of combined sewer overflows (CSOs). Third, an analysis is conducted of the effects of installing GI at different locations in a watershed and the resulting CSO volume reductions. Fourth, we investigate the efficacy of GI at various locations given long-term precipitation data from Philadelphia.  The analysis will be repeated for a second city, Atlanta, to explore how the results may vary with geographic location. Finally, we will generate precipitation data under possible future climate change and evaluate the efficacy of GI at various locations to provide insights for stormwater adaptation management. This work will result in a manuscript to be submitted for publication in mid-2018.  Refereed Journal Publications:  1 in preparation; Ph.D. Dissertations: 1 Ph.D dissertation in preparation; Conference proceedings and presentations: 4 completed, 1 in preparation.

Future Activities:

Sensors will be maintained and new data collected at our selected GSI sites in Philadelphia.  Expenditures associated with Task 1 continue, although at a lower level consistent with maintenance of our field sites and ongoing modeling.  Hydrologic modeling studies will continue, monitored data will enable comparisons with model outputs, and results will be used to improve the StormWISE model’s ability to calculate subwatershed runoff reductions resulting from GI implementation.  Community engagement activities associated with Task 2 will proceed by maintaining our partnerships in  Philadelphia neighborhoods.  The equity index model will be applied within the StormWISE modeling framework.   Task 3 will continue to develop the concept of geographic Zones of Green Infrastructure (ZGI) within Philadelphia by combining statistical, simulation, and evolutionary optimization modeling to create benefit functions incorporating community input and further developments in our equity index, moving towards more realistic models implemented in the Philadelphia CSO area using SWMM models that originate with PWD.  Task 4 will build on the tools developed during years 1 through 4 for quantifying and prioritizing community benefits to begin implementation of StormWISE model extensions relating to these benefits to the Philadelphia CSO area.  Development of the user interface for the StormWISE model will continue to be made more interactive to improve its ability to inform stakeholders about GI benefits, costs, and necessary tradeoffs.  Development of the two-stage stochastic model will continue with the goal of integrating adaptive management into the StormWISE modeling framework.

Journal Articles on this Report : 2 Displayed | Download in RIS Format

Other project views: All 56 publications 5 publications in selected types All 5 journal articles
Type Citation Project Document Sources
Journal Article Heckert M, Rosan CD. Developing a green infrastructure equity index to promote equity planning. Urban Forestry & Urban Greening 2016;19:263-270.
abstract available   full text available
R835555 (2016)
R835555 (2017)
R835555 (Final)
  • Full-text: ScienceDirect-Full Text HTML
  • Abstract: ScienceDirect-Abstract
  • Other: ScienceDirect-Full Text PDF
  • Journal Article Lim TC, Welty C. Effects of spatial configuration of imperviousness and green infrastructure networks on hydrologic response in a residential sewershed. Water Resources Research 2017;53(9):8084-8104.
    abstract available  
    R835555 (2017)
    R835555 (Final)
  • Abstract: Wiley-Abstract
  • Supplemental Keywords:

    Equity Index, Batch SWMM, Python Wrapper SWMM, StormWISE, Wingohocking, Tensiometer, Piezometer, Parflow.CLM, pressure head, diurnal signal, adaptive management, stochastic, multi-stage

    Relevant Websites:

    GreenPhilly Research Group Exit

    Progress and Final Reports:

    Original Abstract
  • 2014 Progress Report
  • 2015 Progress Report
  • 2016 Progress Report
  • Final Report