International low impact development conference 2016 mainstreaming green infrastructure

I L I2016 PROCEEDINGS OF THE INTERNATIONAL LOW IMPACT DEVELOPMENT CONFERENCE 2016 August 29–31, 2016 Portland, Maine SPONSORED BY New England Water Environment Association Environm

MainstreaMing green infrastructure

international

Low impact

Development

conference 2016

Proceedings of the international

Low impact Development conference 2016

eDiteD by

I L I

2016

PROCEEDINGS OF THE INTERNATIONAL LOW IMPACT

DEVELOPMENT CONFERENCE 2016

August 29–31, 2016 Portland, Maine

SPONSORED BY

New England Water Environment Association

Environmental and Water Resources Institute

of the American Society of Civil Engineers

EDITED BY

Robert Roseen, Ph.D., P.E., D.WRE

Virginia Roach, P.E

James Houle, Ph.D

Published by American Society of Civil Engineers

1801 Alexander Bell Drive Reston, Virginia, 20191-4382 www.asce.org/publications | ascelibrary.org Any statements expressed in these materials are those of the individual authors and do not necessarily represent the views of ASCE, which takes no responsibility for any statement made herein No reference made in this publication to any specific method, product, process,

or service constitutes or implies an endorsement, recommendation, or warranty thereof by ASCE The materials are for general information only and do not represent a standard of ASCE, nor are they intended as a reference in purchase specifications, contracts, regulations, statutes, or any other legal document ASCE makes no representation or warranty of any kind, whether express or implied, concerning the accuracy, completeness, suitability, or utility of any information, apparatus, product, or process discussed in this publication, and assumes no liability therefor The information contained in these materials should not be used without first securing competent advice with respect to its suitability for any general or specific application Anyone utilizing such information assumes all liability arising from such use, including but not limited to infringement of any patent or patents

ASCE and American Society of Civil Engineers—Registered in U.S Patent and Trademark Office

Photocopies and permissions Permission to photocopy or reproduce material from ASCE

publications can be requested by sending an e-mail to permissions@asce.org or by locating a title in ASCE's Civil Engineering Database (http://cedb.asce.org) or ASCE Library

(http://ascelibrary.org) and using the “Permissions” link

Errata: Errata, if any, can be found at https://doi.org/10.1061/9780784480540

Copyright © 2017 by the American Society of Civil Engineers

All Rights Reserved

ISBN 978-0-7844-8054-0 (PDF) Manufactured in the United States of America

Preface

The International Low Impact Development Conference was held in Portland, Maine

in August of 2016 The Proceedings presented here represent a slice of the interesting and timely content that was presented at the conference The conference this past year highlighted the mainstreaming of Green Infrastructure and Low Impact Development

in municipal programming as well as new and existing work and research in the United States and internationally We are excited to announce that the 2016 LID conference led to a spin-off conference entitled Operations and Maintenance of Stormwater Control Measures that will be coming to Denver, Colorado in November 2016 We hope that these proceedings provide the in-depth information that you are looking for and we look forward to seeing you at the next LID conference in 2018!

Acknowledgments

Preparation and planning are the key to a successfully executed conference so we would like to recognize the hard work of the Conference Steering Committee and also others that are not mentioned here

James Houle UNH Stormwater Center

Technical Program Chair

James Houle UNH Stormwater Center

Technical Program Vice Chairs

Bethany Eisenberg Vanasse Hangen Brustlin Inc

William Arcieri Vanasse Hangen Brustlin Inc

Local Host Chair

Curtis Bohlen Casco Bay Estuary Program Rachel Rouillard

Piscataqua Regions Estuary Partnership

Workshop and Field Trip Coordinator Chair

James Houle UNH Stormwater Center

Workshop and Field Trip Coordinator Vice-Chair

Jami Fitch Cumberland County Soil & Water Conservation District

Past LID conference Member

Scott Struck Geosyntec Finally, we acknowledge and thank the staff of the EWRI of ASCE, who, in the end, make it all happen

Director, EWRI

Brian K Parsons, M.ASCE

Technical Manager, EWRI

Contents

Cistern Performance for Stormwater Management in Camden, NJ 1

Farzana Ahmed, Michael Borst, and Thomas O’Connor

Low Impact Development for Controlling Highway Stormwater Runoff—Performance

Evaluation and Linkage to Cost Analysis 9

Azadeh Akhavan Bloorchian, Jianpeng Zhou, Abdolreza Osouli, Laurent Ahiablame,

and Mark Grinter

How the Implementation of Green City, Clean Waters in Philadelphia Advances

Modeling Capabilities across the Program 16

Eileen Althouse, Edward Lennon Jr., and Julie Midgette

Addressing Water Scarcity in South Africa through the Use of LID 20

L N Fisher-Jeffes, N P Armitage, K Carden, K Winter, and J Okedi

Development of a Low Impact Development and Urban Water Balance Modeling

Tool 29

Steve Auger, Yuestas David, Wilfred Ho, Sakshi Sani, Amanjot Singh, Tim Van Seters,

Chris Davidson, Melanie Kennedy, and Kevin MacKenzie

Simulation of the Cumulative Hydrological Response to Green Infrastructure 43

Pedro M Avellaneda, Anne J Jefferson, and Jennifer M Grieser

Dual Opportunity for Education and Outreach to Evaluate Benefits of GI

Implementation 52

Leslie Brunell and Elizabeth Fassman-Beck

Examination of Empirical Evidence and Refining Maintenance Techniques for GI 65

Amirhossein Ehsaei and Thomas D Rockaway

Comprehensive Benefits Assessment with as a Step toward Economic Valuations of

Ancillary Benefits of Green Storm Water Infrastructure and Non-Structural

Storm Water Quality Strategies in San Diego, California 78

Clem Brown, Richard Haimann, and Chris Behr

A New Method for Sizing Flow-Based Treatment Systems to Meet Volume-Based

Standards 89

Kelly L Havens, Zachary J Kent, and Aaron Poresky

Evaluating the Real Estate Development and Financial Impacts of the San Diego

Region’s Post-Construction Standards and Alternative Compliance Program:

A Multi-Disciplinary Effort 98

Juli Beth Hinds

Estimating Monetized Benefits of Groundwater Recharge from Stormwater

Retention Practices 106

John Kosco, Lisa Hair, Jonathan Smith, and Heather Fisher

Design Parameters for Manufactured Soils Used in Storm Water Treatment,

Wetland Restorations, and LID Projects 114

Geoffrey Kuter, David Harding, and Mike Carignan

Performance Optimization of a Green Infrastructure Treatment Train Using

Real-Time Controls 123

C Lewellyn, B M Wadzuk, and R G Traver

Greening Indiana—One Training at a Time 131

Sheila McKinley

Update to Permeable Pavement Research at the Edison Environmental Center 135

Thomas P O’Connor and Michael Borst

Full-Scale Structural Testing of Permeable Interlocking Concrete Pavement to

Develop Design Guidelines 143

David J Jones, Hui Li, Rongzong Wu, John T Harvey, and David R Smith

Developing Low Impact Development (LID)-Based District Planning (DP)

Techniques and Simulating Effects of LID-DP 155

C H Son, J I Baek, D H Kim, and Y U Ban

Green Infrastructure Performance Model in the Real World: Modeling Natural

and Simulated Runoff Events 163

Stephen White, Tyler Krechmer, Taylor Heffernan, Nicholas Manna,

Elizabeth Mannarino, Chris Bergerson, Mira Olson, and Jason Cruz

Winter Road Salting in Parking Lots: Permeable Pavements vs Conventional

Asphalt Pavements 173

H Zhu, J Drake, and K Sehgal

Cistern Performance for Stormwater Management in Camden, NJ

Farzana Ahmed1; Michael Borst2; and Thomas O’Connor3

1Post-Doctoral Research Fellow, Oak Ridge Institute of Science and Education (ORISE), U.S

Environmental Protection Agency, 2890 Woodbridge Ave., MS-104, Edison, NJ 08837 E-mail:

volume and peak discharge The collected water can be substituted for potable water in some

applications reducing the demand This presentation focuses on five cisterns that were monitored

as part of a capture-and-use system at community gardens The cisterns capture water from

existing rooftops or shade structures installed by CCMUA as part of the project Cistern volumes

varied from 305 gallons to 1,100 gallons The design volume was based on the available roof

drainage area Water level was monitored at 10-minute intervals using pressure transducers and

rainfall was recorded using tipping bucket rain gauges Cisterns were sampled at 6 to 8 week

intervals through the growing season for determination of concentration of microorganisms,

nutrients, and metals The analyses detected antimony, arsenic, barium, copper, lead, manganese,

nickel, vanadium, and zinc Concentration of all these metals were below recommended water

quality criteria for irrigation by EPA guidelines for water reuse The total nitrogen, phosphate,

and total organic carbon concentrations varied from 0.23 to 2.26 mg/L, 0.025 to 1.11 mg/L, and

0.55 to 4.06 mg/L, respectively Large total coliform concentrations were observed in some

samples The presentation will summarize the data for first growing season giving the results

from monitoring the water use and water quality of cisterns

INTRODUCTION

The Camden County Municipal Utilities Authority (CCMUA) has installed several green infrastructure stormwater control measures (SCMs) throughout the City of Camden to reduce the

volume of Combined Sewer Overflows This presentation focuses on five cisterns installed at

community gardens The Rutgers Cooperative Extension Water Resources Program developed

engineering plans and specifications for each of the sites US EPA is monitoring these five

installed cisterns for three consecutive growing seasons This paper summarizes the findings

from monitored cisterns for the first year growing season

Cisterns collect and store rainwater that can be used for household and other uses A gutter and downspout system directs the collected rainwater to the storage cistern Cisterns can be

installed above or below ground Roof harvested rain water has been considered to be one of the

most cost effective sources for various non potable uses like irrigation, toilet flushing, and car

washing (Ahmed, et al 2011)) Cisterns can reduce stormwater runoff volume and peak

discharge rates, and provide an alternative water supply during times of water restriction Factors

that influence the quality and quantity of captured rainwater include: roof geometry (size,

exposure, and inclination), roof material (chemical characteristics, roughness, surface coating,

age, and weatherability), location of the roof (proximity of pollution sources), maintenance

history of the roof, rainfall events (wind speed, intensity, and pollutant concentration), other

meteorological factors (seasons, weather characteristics, and antecedent dry period), and

concentration of substances in the atmosphere (transport, emission, half-life, and phase

distribution) (Abbasi and Abbasi 2011)

The objectives of this study are to: 1) demonstrate the performance of cisterns, 2) determine how the performance of cistern changes during the first three years of operation, and 3) collect

and analyze aqueous samples from cistern for presence of bacteria and other analytes This paper

only presents the quantity and quality analysis from the first growing season

SITE DESCRIPTION AND INSTRUMENTATION

Camden is located in southwestern New Jersey, United States The city is highly urbanized with an aging and overburdened combined sewer system which discharges to the Delaware

River As part of the effort to control combined sewer overflows, CCMUA installed the cisterns

in 2014 and 2015, with capacities ranging from 300 to 1,100 gallons to provide capture-and-use

for irrigating community gardens and existing landscaped areas Since May 2015, US EPA

monitored water collection-and-use at five cistern sites: the Vietnamese Community Garden,

Kaighns Avenue Neighborhood Community Center, Respond Inc., Cooper Sprouts Community

Garden, and St Joan of Arc Church Level loggers (Solinst 3001 LT Levelogger) placed at the

bottom of each cistern record water level at 10-minute intervals At five sites, standalone tipping

bucket rain gauges (Onset model RGD-04) were installed A layout of the location of the

installed sensors are shown in Figure 1

Figure 1 Location of cisterns and rain gauges

Figure 2 Cistern at St Joan of Arc Church site

These cistern tanks as shown in Figure 2, are made of resins that meet FDA specification to ensure safe water storage The black color limits light penetration to reduce the growth of water

borne algae The monitored cisterns are installed near vegetable gardens to provide irrigation

water These cisterns capture roof runoff through the downspout from adjacent buildings At

some sites no building existed, therefore, a shade structure was constructed to supply roof runoff

to the cistern tank If it rains while the cistern is full, the cistern overflows Near the bottom of

each tank a spigot and a hose was installed so that the water can be used At two sites, a pump

was installed with the spigot that helped to draw the water from the tank

Pressure transducer and rain gauge data were used to calculate the fraction of available water used from each cistern Water samples were collected every 6 to 8 weeks after the sensor

installation The samples were analyzed for microorganisms, metals, and nutrients

AVAILABLE WATER USE

For each cistern, the relative use of available water was calculated To calculate the relative volume used, the total water use between consecutive rain events is divided by the available

water volume in the tank For example, in Figure 3, the green line shows the water level depth

inside the cistern tank monitored by pressure transducer at the Kaighns Avenue site The red dots

show the cumulative rain depth from 07/31 to 08/12 Precipitation was recorded on 07/30, 08/07,

and 08/11 Between 07/30 and 08/07 rain events the available water was 1.04 m and no water

was used, so the water use between 07/30 and 08/07 is 0%

Between 08/07 and 08/11 rain event water use was 0.12m and available water was 1.04m So the water use is 2

water Discussions with one of the gardeners suggested that the reason might be lack of pressure

to help to transfer water from the tank to the garden Since no water was used, the roof runoff

overflowed from the cistern after it was full and it did not reduce stormwater runoff volume At

Respond Inc site, the cistern water was also unused The roof downspout was not connected to

the cistern tank until end of September, so the cistern did not collect water until September

Among all five sites, the gardeners from Vietnamese Garden used the cistern water most

frequently

Figure 3 Water use calculation for Kaighns Avenue Site

Table 1 Percent of available water use

available water Roof type Drainage area (sq-ft)

WATER QUALITY ANALYSIS

For first growing season, five site visits were made between June and November to collect water samples from the cisterns For some sites, only four samples were collected due to access

difficulty or empty cisterns A randomly selected site was sampled in duplicate for each round of

sampling The water samples were analyzed for total coliforms, E coli and enterococci as Most

Probable Number (MPN) estimates from the IDEXX Quanti-Tray / 2000 (IDEXX Laboratories,

2013) The water samples were also analyzed for metals by Inductively Coupled Plasma-Mass

Spectrometer method (EPA 200.8), Nitrate-Nitrite Nitrogen by Automated Colorimetry (EPA

0%

method 353.2), Phosphate by Semi-Automated Colorimetry (EPA method 365.1) and Total

Organic Carbon (EPA method 415.3) Microbial and metal analysis were completed by EPA

Region 2 Nutrient analyses were completed by EPA ORD laboratories in Cincinnati, OH

St Joan of Arc Church site had the largest total coliform concentration Figure 2 shows the picture of St Joan of Arc Church, which shows a tree next to the tank Feces from animals that

inhabit in this tree may contribute to the microorganism concentration Overall, the water in all

cisterns exceed the drinking water and recreational water standards Each cistern has a “Do not

drink” sign attached

Table 2 shows the summary of microorganism concentration in the samples collected from cisterns It shows that the total coliform concentration ranges from 60 to more than 242,000

MPN/100 mL E coli and enterococci are below detection limit for some samples Studies from

different researchers showed presence of microorganism in roof harvested rainwater (Lye 1987,

Ruskin and Krishna 1990, Lye 2002) Within the United States, the number of fecal coliform

varied between 0 - 4800 and the highest fecal coliform was reported at the rain water collection

system in Hawaii (Fujioka, Inserra et al 1991, Thomas and Greene 1993) At a site located in

Australia the range was between 0 – 130 (Thomas and Greene 1993) The detection frequency

for total coliform and E Coli was reported as 93% and 3% respectively at some sites in US (Lye

1987, Ahmed, Gardner et al 2011)

Table 2 Summary of microorganism concentration data of aqueous samples from cisterns

Site

Statistical parameter

Microorganism concentration

(MPN/100 mL) Total

coliform

E coli Enterococci

Cooper Sprouts

detection limit

Below detection limit

St Joan of Arc Church

At each Camden site, there is a wide variation in microorganism concentration For each type

of microorganism, data for all five sites were combined to perform ANOVA test Dependent

variable was the microbial concentration and treatment variable was the site and date of event

The combined microorganism data were log normally distributed, so the analyses were

performed on log-transformed data ANOVA test showed that the microorganism concentration

is independent of the location and time (p>0.05)

Table 3 shows the average metal concentration that is above detection limit in aqueous samples from cisterns The metal concentration was compared with the EPA irrigation water

standard for long term use (Manual 1992) and the WHO drinking water standard (EPA 2009)

The average metal concentration was less than the EPA long-term irrigation water standard and

drinking water standard except for Pb in Kaighns Avenue Pb concentration in Kaighns Avenue

is above the drinking water standard A study analyzed the runoff from roofs made of galvanized

metal and reported 0.01 to 1.4 mg/L and 0.42 to 14.7 mg/L leaching of Cu and Pb respectively

(Tobiason 2004, Clark, Steele et al 2008) Whereas, another study reported 0.17 mg/L of Cu,

0.88 mg/L of Zn, and 0.011 mg/L of Pb leaching from roof made of Plywood (Good 1993)

Table 3 Summary of metal concentration data of aqueous sample from cistern

Site name

Statistical parameter

Ba (mg/L)

Cu (mg/L)

Pb (mg/L)

Mn (mg/L)

Ni (mg/L)

Vn (mg/L)

Zn (mg/L)

Detection Limit

Cooper Sprouts

detection limit

Below detection limit

Below detection limit

36

Geometric mean

Kaighns Avenue

detection limit

250

Geometric mean

St Joan

of Arc Church

detection limit

Geometric mean

namese

detection limit

Below detection limit

88.5

Geometric mean

Respond, Inc

detection limit

Geometric mean

represents neither normal distribution nor log-normal distribution The ANOVA test was

performed on untransformed data The test showed that Pb and Zn concentration at Kaighns

Avenue site was significantly larger (p=0.001) than other sites and Cu concentration for St Joan

of Arc Church site was significantly larger (p=0.001) than other sites A site inspection at

Kaighns Avenue and St Joan of Arc Church site revealed that the roof and gutter system is old

and is not maintained This might contribute to elevated Pb, Cu and Zn concentration

Table 4 shows the average nutrient concentration of the aqueous samples from cisterns

According to FAO guidelines, if nitrate concentration is < 5mg/L then there is no restriction on

using the water for irrigation For all sites the nitrate concentration is below 5 mg/L Nitrate,

phosphate and total organic carbon concentration is larger at St Joan of Arc Church cistern water

compared to other sites

Table 4 Summary of nutrient concentration data of aqueous sample from cistern

Site name Statistical

parameter

NH3 as N (mg/L)

NO2 as N (mg/L)

NO3 as N (mg/L)

PO4

(mg/L)

TN (mg/L)

TOC (mg/L)

Detection limit

Cooper Sprouts

Geometric mean

Kaighns Avenue

detection limit

0.72 4.92

Geometric mean

St Joan of Arc Church

Geometric mean

Geometric mean

Respond, Inc

Max No 0.06 Below

detection limit

Geometric mean

CONCLUSION

The water use from the cistern is below expectation If the cistern’s water is not used then there will be overflow from the cistern during rain and the stormwater runoff volume will not be

reduced This may be resolved by ensuring there is adequate pressure at the spigot and improved

education of the user as to the purpose of the cistern and appropriate timing of use

From the first year analysis, it was found that the metal concentrations and nutrient concentrations were below the EPA long-term irrigation water standard and the FDA standard,

respectively The number of total coliform is larger than what was found from previous studies

on rain water harvesting The practice of attaching “Do not drink” sign to cistern should continue

and additional signage advising gardeners to washing the fruits and vegetables before eating is

recommended This would reduce the risk of ingesting microorganisms

EPA plans to collect samples for two additional growing seasons which will help to make a firm conclusion about the water quality

REFERENCES

Abbasi, T and S Abbasi (2011) “Sources of pollution in rooftop rainwater harvesting systems

and their control.” Critical Reviews in Environmental Science and Technology 41(23): 2097–

2167

Ahmed, W., T Gardner, S Toze (2011) “Microbiological quality of roof-harvested rainwater

and health risks: a review.” Journal of Environmental Quality 40(1): 13–21

Clark, S E., K A Steele, J Spicher, C Y S Siu, M M Lalor, R Pitt, J T Kirby (2008)

“Roofing materials’ contributions to storm-water runoff pollution.” Journal of irrigation and

drainage engineering 134(5): 638–645

EPA, U (2009) National primary drinking water regulations EPA 816-F-09-004 U E P

Agency

Fujioka, R., S G Inserra, R D Chinn (1991) The bacterial content of cistern waters in Hawaii

Proceedings of the Fifth International Conference on Rain Water Cistern Systems, Keelung, Taiwan

Good, J C (1993) “Roof runoff as a diffuse source of metals and aquatic toxicity in storm

water.” Water science and technology 28(3–5): 317–321

Lye, D J (1987) Bacterial levels in cistern water systems of northern Kentuky, Wiley Online

Library

Lye, D J (2002) Health risk associated with consumption of untreated water from household

roof catchment systems, Wiley Online Library

Manual, E (1992) Guidelines for water reuse, EPA/625/R-92/004

Ruskin, R H and J Krishna (1990) A preliminary assessment of cistern water quality in

selected hotels and guest houses in the US Virgin Islands

Thomas, P and G Greene (1993) “Rainwater quality from different roof catchments.” Water

science and technology 28(3–5): 291–299

Tobiason, S (2004) “Stormwater metals removal by media filtration: Field assessment case

study.” Proceedings of the Water Environment Federation 2004(4): 1431–1448

Low Impact Development for Controlling Highway Stormwater Runoff—Performance

Evaluation and Linkage to Cost Analysis

Azadeh Akhavan Bloorchian1;Jianpeng Zhou, Ph.D., P.E.2; Abdolreza Osouli, Ph.D., P.E.3;

Laurent Ahiablame, Ph.D.4; and Mark Grinter5

1Ph.D Candidate, Dept of Civil Engineering, Southern Illinois Univ Edwardsville, IL

4Assistant Professor/Grassland Hydrologist, Dept of Agricultural and Biosystems Engineering,

South Dakota State Univ., SD 57007 E-mail: Laurent.Ahiablame@sdstate.edu

5Associate Professor, Dept of Construction Management, Southern Illinois Univ Edwardsville,

IL 62026-1800 E-mail: mgrinte@siue.edu

ABSTRACT

Highways are major source of stormwater runoff due to their large foot print of paved areas

The runoff can lead to many environmental problems such as non-point source pollution, soil

erosion, and flooding The low impact development (LID) practice, through incorporating best

management practice (BMP) elements in linear infrastructure projects such as highways, can

provide a cost-effective and environmentally sound solution for on-site control and management

of stormwater runoff Because of the diversity and variety of site conditions across the country,

an extensive number of factors have to be considered For a given project, factors can include

soil characteristics such as soil type and infiltration rate; site conditions such as surface

vegetation cover, drainage area and pathway, slopes, imperviousness; meteorological conditions

such as rainfall; available land space for BMPs, and costs associated with the installation and

maintenance of BMPs To develop a most cost-effective engineering solution for a given site, a

large number of scenarios need to be analyzed to evaluate the impact of essential factors

aforementioned on the performance of BMPs, which is to be linked to the cost analysis of a

given scenario For practical application, an approach that can be readily deployed for efficient

evaluation of many scenarios in relatively short time is needed Such an approach should be

integrated with considerations for cost analysis Information and reporting in the currently

available LID design manuals and related technical documents about such an integrated approach

linking extensive performance evaluation with cost analysis is limited This paper discussed

results from our study that takes a modeling approach to evaluate the impact of many

aforementioned factors on performance of several BMPs for control of stormwater from

highways, as well as the linkage with cost analysis This study used an idealized catchment and

Personal Computer Stormwater Management Model (PCSWMM) for analysis The modeled

BMPs included bioswale, infiltration trench and vegetated filter strip The analysis results on

newly constructed BMPs indicated an average runoff reduction of up to 100% from the

infiltration trench, of 70-83% and 68-78% for bioswale and vegetated filter strip, respectively

The linkage between the performance and the costs of BMP installation and maintenance for

linear projects were discussed Findings from this study provides valuable information to support

decision-making for selecting and placing cost-effective stormwater BMP for controlling

stormwater runoff from highways and beyond

KEYWORDS: Green infrastructure; LID; modeling; PCSWMM INTRODUCTION

Storm runoff from highways contain suspended solids and many other pollutants Residuals from the use of deicing, salt, and antiskid materials can have negative impacts on vegetation and

soil of surrounding areas, and the water environment receiving the runoff (DEP 2006)

Therefore, on-site management of storm runoff for effective volume reduction and peak flow

attenuation is important for the environmental protection The primary approach to manage

stormwater on-site is the Low impact development (LID) The purpose of LID is to maintain and

restore the hydrologic functionality of a site through alternative urban design and adoption of

Best Management Practices (BMPs) (USEPA 2008; PGC 1999) The US Environmental

Protection Agency (USEPA) defines the best management practices (BMPs) as an engineered

and constructed system that is designed to provide water quantity and quality control of

stormwater (USEPA 1999a) The hydrologic performance of stormwater BMPs is an important

factor in the overall effectiveness of BMPs in reducing potential adverse impacts of urbanization

on receiving waters (Poresky et al 2011) BMPs that are frequently used for roadway runoff

control are linear BMPs which include grassed swales, vegetated filter strips, and infiltration

trench

The performance and effectiveness of stormwater managemnt can be affected by many factors, which include site characteristics such as local climate, soil types and geologic

conditions, groundwater conditions, site topography and grading; watershed characteristics,

project location in watershed, and adjacent land uses Moreover, project characteristics

influencing runoff volume reduction include project type, highway type, the amount of open

space in medians and shoulders, shoulder-width and usage, highway landscaping and vegetation,

and maintenance access (Strecker et al 2015) In order to install and maintain BMPs, factors

including soil characteristics such as soil type and infiltration rate; site conditions such as surface

vegetation cover, drainage area and pathway, slopes, imperviousness and meteorological

conditions such as rainfall; available land space for BMPs, and costs need to be considered as

well

Bioswales, infiltration trenches, and vegetated filter strips have been employed as appropriate BMPs for linear construction projects Measured performance from previous studies show

percentage runoff volume reduction for vegetated filter strips is about 40% to 85% and for

bioswales is between 50% to 94% runoff volume reduction (Hunt et al 2010; Poresky et al

2011; Xiao and McPherson 2009) The difference in performances can be due to the sizes of the

studied BMP, the size of the service area, and the local characteristics such as the site conditions

Studies demonstrated almost 100% mitigation capacity for infiltration trench (Geosyntec 2008;

Caltrans 2004)

Many guides and manuals have already been developed, but many of them might be locally relevant and for some areas there might be no guidance for LID design Regionally, best design

and practices can be accomplished by academics, the engineering corporation, or local

municipalities Due to similarities in climate in some locations, there may be lessons learned

from limited areas that can be used in other areas A comprehensive evaluation is required to

implement and monitor locally relevant demonstrations and basic research resulting in a more

effective stormwater management in long term Considering potential pitfalls of the practices, the

design team should include experts in engineering, hydrology, ecology, economics, policy,

and/or education from both the public (academic and government) and the private sector Also

funding and financial resources are the inseparablepart of these work (Vogel et al 2015)

The literature offer information about the performance of the linear BMPs; monitoring efforts are limited to short-term and as mentioned above they are localized evaluation For practical

application, an approach that can be readily deployed for efficient evaluation of many scenarios

in relatively short time is needed Such an approach should be integrated with considerations for

cost analysis Simulation modeling is one of the practice for evaluating BMPs at various spatial

and temporal scales

In order to use simulation models, there are several tools available Researchers used the Hydrological Simulation Program-FORTRAN (HSPF) to the effects of various LID practices,

land-use changes, and water-management activities on streamflow at multiple spatial scales

(Zimmerman et al 2010), and also to model the grass swale (Ackerman et al 2008) The Soil

and Water Assessment Tool (SWAT) (White et al 2009) and the Long-Term Hydrologic Impact

Assessment- Low Impact Development (L-THIA-LID) models (Ahiablame et al 2012; Liu et al

2015) have been used to analyze the vegetated filter strip at the watershed scale and to estimate

hydrology and water quality, respectively Other studies used the Storm Water Management

Model (SWMM), which is dynamic rainfall-runoff-routing model to simulate hydraulics and

hydrology of the BMPs These studies indicated that the algorithms used in LID control

parameters provide satisfactory results for event-based and continuous simulations (Abi Aad et

al 2010; McCutcheon et al 2013; Sun et al 2014)

Following the discussion above, objectives of this paper are (1) to present a readily deployed approach to evaluate the impact of selected linear BMPs for controlling storm runoff from

highways, and (2) to estimate the cost for construction and maintenance of the BMPs This

approach can help to simulate various highway sites with different soil types, surface covers, and

types of BMPs

APPROACH

The approach used in this paper is the idealized catchment area including half of an lane interstate highway in an urban area and its right of way This idealized catchment area

eight-consisting of four subcatchments such as highway (with 58 ft width and slope of 1.5%),

foreslope (with 21 ft width and slope of 3H,1V), ground surface (with 25 ft width and slope of

0), and backslope (with 14 ft width and slope of 6H:1V) For each soil type, according to the

condition of the foreslope, ground surface, and backslope there are several scenarios needed to

simulate Considered scenarios including pre-BMP, post-bioswale, post-infiltration trench, and

post-VFS Moreover, for each pre and post-BMP, depending on the soil cover, there would be

three soil surface cover conditions: no vegetation cover, turf grass cover, and prairie grass cover

(Akhavan Bloorchian et al 2016) Hence, for each soil type there will be 12 scenarios which

much time and effort (besides the cost) would be devoted if performed in practice than

simulation

In order to evaluate the performance of aforementioned BMPs with 11 soil type and three different soil cover, there are 132 scenarios to build the BMPs on site which considering the

construction and maintenance cost and also seeking the linear locations for each 11 soil types

and three soil surface cover, won’t be feasible; where simulating the models comes into play

PCSWMM, a GIS version of the EPA Storm Water Management Model (EPA SWMM) was used in this study The Green-Ampt model was applied to stimulate the infiltration in the model

One-inch rainfall was used to model all scenarios for BMP performance The evaporation was

assumed to be negligible Consequently, the runoff from the impervious area was either

infiltrated, stored on the surface, or flowed overland For post-BMP scenarios, the BMP was

applied at the ground surface subcatchment; The BMPs were modeled with very typical

dimensions reported in the literature (Strecker et al 2015; SEMCOG 2008; Emanuel et al 2014;

James et al 2010; LLG, 2009; PGC 1999; USEPA 2000) The bioswale width is five ft with 3:1

side slope and the total footprint of bioswale is 14 ft The infiltration trench is three ft wide; and

the vegetated filter strip is 25 ft wide

Figure 1: Idealized catchment area of half of an eight-lane interstate highway

To evaluate the performance in each scenario, the runoff at the outfall of the catchment area for each of scenario was compared to the one-inch rainfall input to the area At the result section,

an example of this approach for one soil type (silt loam) is presented

Cost analysis plays a key role in planning highway projects Estimated short-term and long term (depending on the policy) costs should be determined during the project planning stage to

be compared with the benefits (USEPA, 1999b) Cost details based on Illinois prevailing wage,

material, and equipment rates are broken down into discrete units, adjusted over a range of BMP

geometry variations, and compiled into spreadsheet based cost calculators capable of generating

BMP unit cost, based on a range of BMP dimensions To support the processing, analysis, and

examination of output data produced by the PCSWMM, a result analysis tool or post-processor

module has been incorporated into the system A Microsoft Excel spreadsheet with macros and

interface perform the analysis Depending on the cost of BMP construction and maintenance and

their capacity in runoff volume reduction, and the goal of the design, engineers or decision

makers can choose the type and size of the BMP

RESULTS

Performance of Pre- and Post-BMP

As Figure 2 demonstrates in the catchment area with silt loam, in pre-BMP condition even with no vegetation cover comparing with the 1-inch rainfall input to the system about 35%

runoff reduction occurs Adding BMPs to the catchment area will increase the runoff reduction

depending on vegetation cover up to 88-92% in post-bioswale, to 100% in the post-infiltration

trench, and up to 84-89% in post-vegetated filter strip The result shows turf grass cover, and

prairie grass are more efficiently in capturing 1- inch rainfall comparing to the site with no

surface cover regardless of the pre- or post- BMP condition Therefore, even without

implementing BMPs and by just using more infiltration friendly grass types in the right of way

an increase in runoff reduction can be achieved

Results of cost and performance

The construction cost of the considered BMPs have been determined Construction cost, for a bioswale, infiltration trench, and vegetative filter strip for the simulated sized for the idealized

cross section (Figure 1) are $ 16,291 per 100 linear ft for Bioswale, $ 4379 per 100 linear ft for

infiltration trench and $ 207 per 100 linear ft for vegetated filter strip Bioswale cost

significantly exceeds the two other type of BMPs

Figure 2: Runoff reduction with BMP in an idealized catchment area (silt loam soil)

A rate of $140 per mile for annual mowing costs for bioswale up to 8 ft wide may be used for planning purposes For bioswales between 9 ft and 6 ft wide mowing costs are estimated at

$280 per mile The cleanup task is estimated at $284 per 100 linear feet for bioswales up to 8 ft

wide and $426 for bioswales between 9 ft and 16 ft wide For infiltration trench, routine

maintenance should include herbicide applications to control vegetation and maintain the open

nature of the infiltration trench surface Herbicide application may be required three times

annually to maintain a vegetation-free surface Each herbicide application will cost

approximately $55 per mile of infiltration trench or a total of $165 per year

Vegetative filter strips should be mown and inspected for rills and gullies annually to promote stand health, and exclude woody plants and sediment build-up A side-mounted sickle

type or spinning disc type mower is recommended for annual mowing operations A rate of $55

per acre may reliably be used for planning purposes for a VFS with 25 ft width Estimated

10-year maintenance cost, of bioswales, infiltration trenches, and vegetated filter strips are

approximately 32%, 80%, and 3.4% of construction cost, respectively

CONCLUSION

This study presented an approach for efficient evaluation of linear BMPs performance in reducing highway runoff from 1-inch rainfall, also the relative construction and maintenance cost

for each of them In spite of the on-site practices, this approach would take efficient time and

cost to estimate the performance of linear BMPs for highway and road projects The linear BMPs

considered in this study include bioswale, infiltration trench, and vegetated filter strip Three

different cover conditions, i.e., no vegetative surface cover, turf grass and prairie grass cover, in

the right-of-way, were also taken into account to evaluate the efficiency of vegetated cover in

capturing stormwater runoff

Furthermore, using a bioswale with 5 ft bottom width and 3:1 side slope lead to 88-92%

performance runoff reduction would cost $ 16,291 per 100 linear ft to construct Employing an

infiltration trench with 3 ft width and 7 ft depth resulted in 100% runoff reduction and would

0 10 20 30 40 50 60 70 80 90 100

Pre-BMP Post- Bioswale Post- infiltration

No surface cover Turf grass Prairie grass

cost $ 4379 per 100 linear ft to install in the area Finally implementing vegetated filter strip

with 25 ft width which ends up with 84-89% runoff reduction $ 207 per 100 linear ft for

vegetated filter strip It indicates bioswales and infiltration trenches are relatively narrow, while

vegetative filter strip must be much wider to provide the same effectiveness Also the results

showed prairie grass and turf grass increase the runoff capture by BMP Furthermore, the results

demonstrated that bioswales are the costliest BMPs considered, and vegetative filter strip

construction and maintenance cost are substantially the lowest Regarding performance runoff

reduction, BMPs can supposedly be considered to capture targeted stormwater runoff However,

if sufficient land is available, vegetative filter strips are preferred

REFERENCES

Abi Aad, M., Suidan, M & Shuster, W., (2010) Modeling techniques of Best Management

Practices: rain barrels and rain gardens using EPA SWMM-5 Journal of Hydrologic Engineering, 15(6), pp 434–443

Ackerman, D., & Stein, E D (2008) Evaluating the effectiveness of best management practices

using dynamic modeling Journal of Environmental Engineering, 134, 628–639

Ahiablame, L M., Engel, B A., & Chaubey, I (2012) Representation and evaluation of low

impact development practices with L-THIA-LID: An example for site planning Environment and Pollution, 1(2), 1

Akhavan Bloorchian, A.; Ahiablame, L.; Zhou, J.; Osouli, A., (2016) Modeling BMP and

Vegetative Cover Performance for Highway Stormwater Runoff Reduction, Procedia Engineering 145, 274–280

Caltrans (2004) BMP Retrofit Pilot Program Final Report, California Department of

Transportation, CTSW-RT 01-050

DEP, P (2006) Pennsylvania Stormwater Best Management Practices Manual Commonwealth of

PA Emanuel, R., & Powers, T H (2014) City of Chicago stormwater management ordinance

manual City

GeoSyntec, (2008) Post-Construction BMP Technical Guidance Manual, Storm Water BMP

Guidance Manual, City of Santa Barbara

Hunt, W., Hathaway, J., Winston, R., and Jadlocki, S., (2010) Runoff Volume Reduction by a

Level Spreader–Vegetated Filter Strip System in Suburban Charlotte, N.C J Hydro Eng 15, SPECIAL ISSUE: Low Impact Development, Sustainability Science, and Hydrological Cycle, 499–503

James, W., Rossman, L.E., and James W.R., (2010) User’s Guide to SWMM 5, 13th Edition,

CHI Press Publication

Liu, Y., Ahiablame, L M., Bralts, V F., & Engel, B A (2015) Enhancing a rainfall-runoff

model to assess the impacts of BMPs and LID practices on storm runoff Journal of environmental management, 147, 12–23

LLG, 2IM group, (2009) I-294 Tri-state tollway bioswale stormwater/water quality treatment

from TOUHY Avenue to Sanders road overpass contract plans, the Illinois state toll highway authority, Vol I and II

McCutcheon, M., & Wride, D (2013) Shades of Green: Using SWMM LID Controls to Simulate

Green Infrastructure In 2012 Stormwater and Urban Water Systems Modeling Conference

Pragmatic Modeling of Urban Water Systems, Monograph (Vol 21, pp 289–301)

PGC (1999) Low-Impact Development Design Strategies: An Integrated Design Approach

Prince George County, MD

Poresky, A., Bracken, C., Strecker, E., & Clary, J., (2011) International Stormwater Best

Management Practices (BMP) Database, Technical Summary: Volume Reduction, GeoSyntec Consultants & Wright Water Engineers, Inc

Southeast Michigan Council of Governments (SEMCOG), (2008) Low Impact Development

Manual for Michigan, A Design Guide for Implementers and Reviewers

Strecker, E., Poresky, A., Roseen, R., Soule, J., Gummadi, V., Dwivedi, R., & Littleton, C O

(2015) Volume Reduction of Highway Runoff in Urban Areas: National Cooperative Highway Research Program (NCHRP) Report 802, Transportation Research Board (TRB)

Sun, Y W., Li, Q Y., Liu, L., Xu, C D., & Liu, Z P (2014) Hydrological simulation approaches

for BMPs and LID practices in highly urbanized area and development of hydrological performance indicator system Water Science and Engineering, 7(2), 143–154

US Environmental Protection Agency (USEPA) (1999a) Preliminary Data Summary of Urban

Storm Water Best Management Practices, Office of Water (4303) Washington: 99-012

EPA-821-R-US Environmental Protection Agency (EPA-821-R-USEPA) (1999b) Stormwater Technology Factsheet

Bioretention Environmental Protection Agency, Office of Water

US Environmental Protection Agency (USEPA) (2000) Low Impact Development (LID): A

Literature Review Washington, DC http://water.epa.gov/polwaste/green/upload/lid.pdf

Accessed Sept 7, 2014 Vogel, J R., Moore, T L., Coffman, R R., Rodie, S N., Hutchinson, S L., McDonough, K R.,

& McMaine, J T (2015) Critical review of technical questions facing low impact development and green infrastructure: A perspective from the Great Plains Water Environment Research, 87(9), 849–862

White, M J., & Arnold, J G (2009) Development of a simplistic vegetative filter strip model for

sediment and nutrient retention at the field scale, Hydrol Process 23, 1602–1616

Xiao, Q., McPherson G.E., (2009) Testing a Bioswale to Treat and Reduce Parking Lot Runoff,

University of California, Davis and USDA Forest Service

Zimmerman, M.J., Barbaro, J.R., Sorenson, J.R., & Waldron, M.C (2010) Effects of selected

low-impact-development (LID) techniques on water quality and quantity in the Ipswich River Basin, Massachusetts—field and modeling studies U.S Geological Survey Scientific Investigations Report, 2010–5007 p 113

How the Implementation of Green City, Clean Waters in Philadelphia Advances Modeling

Capabilities across the Program

Eileen Althouse1; Edward Lennon Jr.2; and Julie Midgette3

of green stormwater infrastructure Since the adoption of “Green City, Clean Waters” in June

2011, the modeling efforts to represent, evaluate and guide the program have advanced The

growth of the program has led to an increase in data collection and general information about the

collection system, which helps make the hydrologic and hydraulic (H&H) models better

The H&H models have been developed by Philadelphia Water in EPA SWMM (Environmental Protection Agency Storm Water Management Model) The H&H models

simulate the hydrologic response from the areas served by the combined and separate sanitary

sewer system, as well as the hydraulics of the collection system(s) that convey combined and

sanitary flow to the Water Pollution Control Plants The H&H models are driven by observed

rainfall data, and runoff is calibrated using data from a flow-monitoring program Since the

implementation of “Green City, Clean Waters” began, the flow monitoring program has

increased, and using the monitoring data for model calibration has led to better estimate of

general hydrologic and hydraulic parameter’s in the H&H models, hence leading to decrease in

the uncertainty associated with the model results The program has also led to the acquisition of

other datasets that can be used to inform the H&H models One example is pre-construction

infiltration tests used to validate the saturated hydraulic conductivity determined in SWMM

calibration This information is valuable because one of the primary model parameters adjusted

during calibration is the saturated hydraulic conductivity of the pervious areas Other examples

include the acquisition of high resolution radar-rainfall data This paper and presentation will

explore how the implementation of a large scale green infrastructure program leads to the

accumulation of information that can be applied to other facets of the program, and ultimately

inform planning and design

RAINFALL

As the “Green City, Clean Waters” program implementation advances, the availability of high-resolution rainfall data over the service area becomes critical Rainfall data is used in the

H&H models to estimate volume entering the collection system The H&H models are used in

planning and decision support Rainfall data is also needed for flow monitoring as well as

monitoring studies of green infrastructure Since the implementation of green stormwater

infrastructure is to be distributed throughout the combined sewer areas, a rainfall dataset that has

a good spatial resolution over the land surface is essential to evaluate the impact due to green

stormwater infrastructure The quality of the H&H model results relies heavily on the quality of

the precipitation data A significant effort is made to ensure the best spatial and temporal

precipitation data is obtained, used and maintained

Multiple data sources are used in a tiered approach to improve the coverage and resolution of rainfall data over the service area There are three primary sources of rainfall data used in the

program National Weather Service (NWS) operated Philadelphia International Airport (PHL)

surface observation station, the City-wide rain gage network, and radar rainfall grid data are all

utilized Historically, the H&H models used rainfall data from a network of rain gages

throughout the City of Philadelphia However, radar rainfall data has been acquired to cover the

extent of the model domain, and is used for most of the recent modeling and planning

The longest data record is from the NWS station at PHL This station in southwestern Philadelphia has over 100 years of precipitation data, from 1902 through the present The

National Oceanic and Atmospheric Administration (NOAA), publishes hourly station

climatological data reports Data from this station is useful because it provides a long-term

record of data that can be used for analysis of past conditions and to determine long term

statistics The climatological data such as reported temperature, wind speed and direction, and

snowfall record is also utilized in wet weather event identification

Philadelphia Water maintains a rain gage network consisting of 35 tipping bucket rain gages

The minimum recorded depth of the tipping bucket gages is 0.01 inches The accumulated tip

count is recorded at a 2.5 minute time-step and then converted to a 15-minute time-step for H&H

model input Some of the rain gages are heated and record water equivalent data for snowfall

events Preliminary quality assurance procedures are performed on the raw gage data Bad or

missing data for each rainfall event is identified and flagged on a monthly basis by visual

inspection of the 15-minute accumulated data The measurements at nearby gages are also

compared, and patterns of obvious gage failures are identified Data issues include plugged

gages, field calibration data, and erratic tipping Flagged data for each gage may be filled with

data from the nearby gages using inverse distance squared (IDS) weighting Any neighboring

gage data that are flagged bad or questionable are removed prior to using the data for filling

The H&H models may be analyzed using the data from the rain gage network The rainfall over each sewer catchment in the model is processed using an IDS method, which calculates a

unique rainfall time series for each modeled catchment based on a weighted distance between the

nearest rain gages This improves the spatial resolution over the sewershed so that the effects of

localized rainfall and isolated storms throughout the study area can be simulated The rain gage

data is also useful in model validation Since the model results are compared to trunk sewer

monitoring data on the sewershed scale, it may be beneficial to simulate the models with local

rainfall data that best represents the rainfall that fell over the sewer catchment during the flow

monitoring period

Philadelphia Water acquires radar rainfall data from Vieux and Associates (Vieux) for a grid covering the Greater Philadelphia region, which includes the collection system service area The

dataset is made available as a continuous 15-minute 1 x 1 km gage calibrated grid In

development of the radar rainfall dataset the Philadelphia rain gage data is used to calibrate the

National Weather Service’s Next Generation Weather Radar (NEXRAD) data Through this

method the total rainfall volume reported at the gages is retained, while the spatial variability

provided by the NEXRAD data is accounted for to provide a more detailed record The resulting

radar rainfall dataset has a greater spatial resolution than the rain gage network and can provide a

better estimate of rainfall volumes and timing in the areas that are between the rain gages

One of the primary uses of the radar rainfall dataset is to estimate the rainfall between measured gages in hydrologic models During some of the storms, the highest intensity of

rainfall may pass over modeled sewer catchments that are between rain gages This may happen

during quick and intense local thunderstorms, a characteristic of summer storms in the region

The rain gages may not record the most intense areas of the storm Therefore the IDS rainfall

calculated for a particular catchment between gages may also be under represented However,

when the rain gage data is augmented to include the spatial coverage provided by the radar at the

1 km grid resolution, the higher intensity hot spots between gages will be mostly accounted for

Using the radar rainfall dataset as an input to the models has produced catchment runoff results

that more closely match the timing of sewer flow monitoring data

FLOW MONITORING PROGRAM

The Philadelphia Water has a flow-monitoring program to support the Long Term Control Plan Update and department wide planning and design initiatives In 2011, the budget for flow

monitoring increased from $500,000 to $1.5 million per year Philadelphia Water currently has

around 70 flow meters installed throughout the collection system

Philadelphia Water also maintains a network of permanent level sensors at most of the combined sewer regulating chamber and key hydraulic control points This network is used

during daily operations and maintenance of the collection system, and for validation of the

hydrologic and hydraulic (H&H) models

Flow monitors are deployed at strategic locations throughout the area served by combined sewer and sanitary sewer systems to obtain data on depth of flow and mean flow velocity The

goal of flow monitoring deployments is to cover as much of the sewershed service area as

possible Sometimes difficulties arise when attempting to find an appropriate location to install a

flow meter that will produce good quality data The issues can range from troubles locating a

desired manhole, to large debris build-up in the sewer pipe that causes erratic flow that is unable

to be properly measured As the program is growing so is the number of meters, the meter

installation plan is adapting to maximize data quality and usefulness by strategically placing

meters in optimal locations

From 1999 through July 2016, around 1,300 different manholes have been investigated for potential deployment of a portable flow meter, with the majority of investigation being

performed after 2011 An estimate of 30% of investigations (400) resulted in locations that were

appropriate for flow monitor installations When a meter is installed at a potentially good

location that does not mean high quality data is guaranteed The flow characteristics at the

location can change over time such as build-up in grit/silt or sudden presence of increased flows

or erratic discharging, etc These circumstances cause some meters to be removed and results in

limited usage of the data collected at these sites These problematic sites are identified and

“lessons learnt are documented” which provide feedback and help with future meter installation

plan

The 423 flow monitoring sites that were installed resulted in 80 utilized for hydrologic calibration of the combined system and 57 utilized in the calibration of the sanitary sewer

system

PRE-CONSTRUCTION INFILTRATION TESTS

Infiltration field test data was acquired citywide Prior to design and construction of GSI projects, sites were tested for saturated hydraulic conductivity (infiltration rate measured at the

end of the field test) As of December 2015, this dataset includes results from 425 infiltration

tests The results from the 425 infiltration test have been used to support H&H model calibration

The decision to use a starting point of loam soil parameters (saturated hydraulic conductivity =

0.52 inch/hour) reached by evaluating the distribution of field infiltration test data, which was

centered slightly higher than this soil type (field data test median saturated hydraulic

conductivity = 0.62 in/hour, loam)

The hydrologic calibration of a subcatchment was performed by adjusting the following parameters in that subcatchment; percent of impervious area routed to pervious (referred to as

percent routed), subcatchment width, pervious area depression storage, impervious area

depression storage, and infiltration The approach to model calibration was to match the volume

and timing of the response of impervious area to small and medium storms by adjusting percent

routed and impervious area depression storage prior to adjusting infiltration parameters

After a calibration set was completed and monitored sewershed properties are adjusted as necessary, validation is performed with a second set of monitors The purpose of H&H model

validation is to confirm that the calibration process resulted in good set of calibrated model

parameters Validation monitoring data was used to decide how to extend the results of model

calibration to un-monitored areas Since the infiltration field test data did not appear to be

strongly distributed geographically, it was decided to pool the calibration results together

city-wide Although several methods of extrapolating calibration results were evaluated, the

system-wide medians for the parameters were chosen for extrapolation of to unmonitored

subcatchments

CONCLUSION

The implementation of “Green City, Clean Waters” (GCCW) has led to an increase in data collection The data collected will assist in updating and maintaining the hydrologic and

hydraulic (H&H) models

Since the start of program in 2011, the department’s rain gage network has expanded from 24

to 35 gages, gauge adjusted radar rainfall is acquired, and the flow-monitoring program has

tripled its budget The methodology for modeling green stormwater infrastructure (GSI) is also

evolving as collected data is analyzed and new technologies are explored

REFERENCES

PWD (Philadelphia Water Department) (2014) “Comprehensive Monitoring Plan.”

〈http://www.phillywatersheds.org/doc/Revised_CMP_1_10_2014_Finalv2.pdf〉 (Jan 10, 2014)

PWD (Philadelphia Water Department) (2011) “Green city clean waters.”

〈http://www.phillywatersheds.org/ltcpu/LTCPU_Complete.pdf〉 (Jul 1, 2015)

Addressing Water Scarcity in South Africa through the Use of LID

L N Fisher-Jeffes1; N P Armitage2; K Carden3; K Winter4; and J Okedi5

1Dept of Civil Engineering, Univ of Cape Town, Private Bag X3, Rondebosch 7701, South

Africa

2Dept of Civil Engineering, Univ of Cape Town, Private Bag X3, Rondebosch 7701, South

Africa E-mail: Neil.Armitage@uct.ac.za

3Dept of Civil Engineering, Univ of Cape Town, Private Bag X3, Rondebosch 7701, South

Africa

4Dept of Environmental and Geographical Science, Univ of Cape Town, Private Bag X3,

Rondebosch 7701, South Africa

5Dept of Civil Engineering, Univ of Cape Town, Private Bag X3, Rondebosch 7701, South

Africa

ABSTRACT

By 2030 the Republic of South Africa (RSA), a developing country, is predicted to be severely impacted by physical water scarcity 2015 was one of the driest years on record, leaving

many towns with extremely compromised water supply systems whilst also limiting food

production across the country This is placing pressure on the already fragile economy In order

to avert a future water crisis, the country needs to reduce its reliance on conventional surface

water schemes based on impoundments on rivers This paper presents an initial assessment of the

potential role of low impact development (LID) for the purposes of stormwater harvesting

(SWH) and describes the principal outcomes from two RSA case studies The findings suggest

that SWH could substantially increase the supply of water to urban areas, but that there are a

number of barriers to its wider use that need to be addressed The paper concludes with three

“position statements” describing how SWH, with the assistance of LID, can contribute to:

improving water security and increasing resilience to climate change in urban areas; preventing

frequent flooding; and providing various other additional benefits such as amenity and the

preservation of biodiversity

INTRODUCTION

By 2030 the Republic of South Africa (RSA), a developing country, is predicted to be severely impacted by physical water scarcity 2015 was one of the driest years in decades and

left many towns with extremely compromised water supply systems whilst also limiting food

production across the country While the weakening of the El Nino in 2016 gave some respite,

the effects of the drought placed pressure on the already fragile economy In order to avert a

future water crisis, the country needs to reduce its reliance on conventional surface water

schemes based on impoundments on rivers Within urban areas, municipalities need to find ways

to adapt to, and mitigate the threats from, water insecurity resulting from, inter alia: droughts,

climate change and increasing water demand (driven by both population growth and rising

standards of living) Low Impact Development (LID) is increasingly being promoted in the RSA

as a means of improving the management of stormwater (SW) runoff, but this is largely to

prevent the impacts that diffuse pollution have on receiving water bodies The potential use of

LID for addressing water scarcity – whilst simultaneously meeting its SW management

objectives – has been mooted, but not properly tested, in RSA

This paper begins by defining rainwater harvesting (RWH) and stormwater harvesting (SWH) before moving on to present a brief overview of the well-known international exemplar

of SWH – Singapore It then discusses the links between LID and SWH This is followed by a

discussion of two RSA case studies that have made use of, or have considered the potential of

SWH using LID approaches The paper concludes by presenting three ‘position statements’

related to how SWH – using LID – should be considered in the RSA – and elsewhere These

three position statements form the basis of a strategic framework – currently being formulated –

to promote SWH using LID principles

URBAN RUNOFF AS A SUPPLEMENTARY WATER SOURCE

In much of the world, including the RSA, water for urban areas is conventionally supplied from surface water schemes based on impoundments on rivers These schemes are typically

situated well outside the urban area, and provide high quality water Meantime urban runoff, i.e

SW, is usually disposed via concrete-lined conduits as quickly as possible to the nearest

watercourse This has contributed to their degradation owing to the increased flows and pollutant

loads compared with the pre-development conditions Internationally, it is now recognised that

SW is being overlooked as a potential water resource This is especially relevant in water-scarce

regions where there are limited alternative water supply sources

Rainwater Harvesting (RWH) and Stormwater Harvesting (SWH) are two distinct terms used

to refer to the harvesting of runoff within an urban area While RWH and SWH have broadly

similar benefits, there are also distinct differences (DECNSW, 2006) In this paper, RWH will be

used to refer to the collection and storage of runoff from an individual property – usually from

the roofs of buildings – and its subsequent use within that property SWH, on the other hand, is

the collection and storage of runoff from an urban region and its subsequent use irrespective of

location Whilst RWH is usually carried out by private property owners, SWH is usually

implemented by the local authority There is, however, some overlap between the two

approaches e.g in gated (security) villages or on large commercial, industrial or educational

properties This paper focuses on SWH

RWH has a well-documented history dating back over 8000 years (Pandey et al., 2003), whereas SWH is a relatively more recent approach The best known international example is

probably to be found in Singapore that currently obtains more than 10% of its potable water

needs through SWH (Straits Times, 2016) Commencing in the 1970’s, Singapore was one of the

first cities to harvest SW from its urban catchments for potable end-uses Land use planning and

the design of the collection systems formed an important part of the SWH system plans (Lim et

al., 2011) Restrictions were put in place to eliminate runoff from areas which were expected to

yield poor quality SW (e.g industrial areas) which was rather directed to the ocean Additionally,

LID approaches which not only manage SW quantity but also quality were widely implemented

within the contributing catchment areas to minimise pollutants at the downstream abstraction

point LID may be defined as ‘an approach to land development (or re-development) that works

with nature to manage stormwater as close to its source as possible LID employs principles

such as preserving and recreating natural landscape features, minimizing effective

imperviousness to create functional and appealing site drainage that treat stormwater as a

resource rather than a waste product’ (USEPA, 2012)

LID AND SWH

Urbanisation leads to increased runoff volumes and peak flows and decreased SW quality

compared with the natural environment (Makepeace et al., 1995; AMEC et al., 2001; Marsalek

et al., 2006; Lee et al., 2010) Conventional drainage systems – typically making used of

concrete-lined conduits – are generally designed to eliminate local flood nuisances by conveying

SW as efficiently as possible to the nearest watercourse This approach largely ignores the

impact on the environment from the pollutants typically found in SW (Table 1) that originate

from various sources including, inter alia: vehicle emission and brake linings, atmospheric

deposition; domestic use of fertilisers etc It also ignores the damage through erosion caused by

the increased flow peaks

Table 1 Typical SW pollutants and associated principal treatment processes Pollutants Principal treatment processes

Sediments Sedimentation and filtration Hydrocarbons Biodegradation, photolysis, filtration and adsorption Metals Sedimentation, filtration, adsorption, precipitation and plant uptake Nutrients Sedimentation, nitrification and plant uptake

Litter Good ‘housekeeping’, urban planning and litter traps

Figure 1: Identifying opportunities for SW harvesting in the LID ‘treatment train’

LID – referred to by the British term ‘Sustainable Drainage Systems (SuDS)’ in the RSA –

offers an alternative, more holistic, approach to conventional drainage practices by inter alia:

managing SW volumes and flow rates; reducing the pollution of runoff; offering various amenity

functions; and supporting biodiversity To do this LID makes use of a ‘treatment train’ (Figure 1)

comprising a number of different LID options which are implemented at a range of scales from

the source, through the local, to the regional scale (Armitage et al., 2013) The goal is to achieve

as high a level as possible, within reason, on the ‘SW hierarchy’ – which may be represented as a

four-step ‘pyramid’ (Figure 2) It is necessary to first ensure any design meets the requirements

of the base of the pyramid (managing water quantity) before climbing to the second step

(improving water quality), the third step (amenity) and finally the fourth step (biodiversity) With

SWH, however, there are some special concerns:

Figure 2: The SW design hierarchy The storage requirement: SW needs to be stored until such time as it can be abstracted,

given further treatment (if necessary), and distributed for use Figure 1 highlights the fact that

only five of the common LID options provide significant storage Of these, RWH tanks store

relatively small volumes Permeable pavements have potential for significant water storage,

although their primary function to date has been to improve water quality and increase recharge,

and the scale at which they might be implemented is limited Retention ponds can store quite

large quantities of water Wetlands, while superficially similar to retention ponds, typically have

relatively smaller storage volumes and are more useful for improving SW quality Underground

aquifers, used in conjunction with managed aquifer recharge (MAR), potentially offer large

long-term water storage

The runoff flow rate: High runoff flow rates typically lead to channel erosion which results

in high sediment loads reaching the storage system – either reducing storage volumes in retention

ponds and/or clogging infiltration systems They also increase the need for temporary storage

All the LID options shown in Figure 1 offer, to a greater or lesser degree, some form of

detention If they are used effectively in a treatment train to attenuate runoff flow rates the need

for storage could be reduced

SW quality: In a conventional SW system, the water quality is typically impacted by a range

of pollutants (Table 1) from both diffuse (e.g sediment in gutters) and point sources (e.g leaking

sewers) LID technologies improve water quality through a range of processes (Table 1 and

Figure 1) which can be adopted to remove selected pollutants It is not usually possible for a

single LID intervention to manage the full range of SW impacts For example, whilst wetlands

and retention ponds can remove sediment – a major source of pollution – from SW through

sedimentation, this ultimately results in reduced storage volume and treatment capacity To

remove the sediment from such “wet” systems is typically expensive and it is thus preferable to

limit the sediment load entering these LID options – through, inter alia, the use of swales, filter

strips, bio-retention areas, or including sedimentation fore-bays in the treatment train The

wetland and / or retention pond then provides a final ‘polishing’ step aimed at removing some of

the remaining soluble pollutants – in addition to providing the storage component This was the

approach that was taken in Singapore

Local climate: Local climate may affect how a SWH scheme is operated For example, a

Mediterranean climate results in most of the harvestable SW being available during the wet

winter months when the water demand is low and the reservoirs are typically filling in any case

However, this affords an opportunity to reduce the demand on the city’s reservoirs during winter

and thus increase the rate at, and level to which, they fill – thereby ensuring an increase in the

availability of water during the hot dry summer months

LID / SWH CASE STUDIES IN RSA

While LID as a paradigm is relatively new in the RSA, there is increasing interest in how LID could, and should, form part of the RSA’s approach to water management most notably

through the promotion of SWH This section reviews the Atlantis Water Resource Management

Scheme that utilises MAR and a local unconfined aquifer for the storage, and subsequent use, of

SW harvested from this suburb; and a recent study into the viability of RWH and SWH in the

Liesbeek River Catchment that modelled the use of rainwater tanks and/or retention ponds as the

storage systems in an urban catchment Both are situated in the City of Cape Town, the

legislative capital of the RSA, situated on the Atlantic coastline in the Western Cape Province in

the south-western part of the country

The Atlantis Water Resource Management Scheme (AWRMS): The AWRMS has

supplied potable drinking water to the Cape Town suburb of Atlantis since 1979 (DWAF, 2010)

and provides a useful South African example of harvesting SW on a large scale making use of a

natural unconfined aquifer and MAR to partially treat and store the water (Philp et al., 2008) An

important consideration is that the suburb was designed from the outset to have separate

residential and industrial areas that allowed for the associated separation of high- and low-quality

wastewater effluent SW is directed to carefully located infiltration ponds where it is used to

supplement the natural rain-fed recharge of the underlying unconfined aquifer for later extraction

and use The higher quality treated municipal effluent has also been added to the SW on

occasion The lower quality effluent is disposed through recharge ponds near the coast in such a

way as to create a hydraulic barrier between the cleaner groundwater and the seawater (Murray

& Tredoux, 2004) The AWRMS has successfully ensured a sustainable supply of up to five

million cubic metres per annum to some 65,000 people in Atlantis (King et al., 1991) over some

three decades

The potential for RWH and SWH in the Liesbeek River catchment: Fisher-Jeffes (2015)

undertook one of the few detailed studies of the financial viability of RWH and SWH in the RSA

focusing on the residential areas in the Liesbeek River Catchment in Cape Town This catchment

is approximately 2,600 hectares in extent and is the oldest urbanised river valley in the RSA

(Evans, 2007) It was selected for this study as it incorporated a diversity of land uses, represents

a range of wealth levels, and had the necessary data available for the effective development of

the detailed models required for simulating catchment-wide RWH and SWH Whilst it is

acknowledged that this study does not represent the full range of social, physical and

environmental conditions across the country, there were a number of interesting findings

including, inter alia:

 RWH was generally not a viable option, either as a water resource – only viable economically for less than 10% of households – or as a SW control measure

 The properties that would be most financially incentivised (where the cost of RWH

would be less than the cost of municipal water) are typically those that have higher water demands and are in wealthier suburbs – which are in the minority of households in RSA

 SWH has the potential to reduce the total current residential potable water demand of the catchment by more than 20%

 The greater the scale of the SWH scheme the more cost effective it is

 The adoption of RWH adversely affects the viability of SWH as it reduces the of-scale benefits

economy- SWH has the potential to generate significant additional benefits (amenity, biodiversity, flood control) if designed in line with the principles of LID

 Whilst it would be costly to retrofit the Liesbeek River Catchment, had the catchment been developed from the outset with SWH as an alternative resource in mind the costs would have been significantly reduced

 Of significant concern to water resource planners is the uncertainty of the effects of climate change on water resources in this catchment; the general opinion appears to be that evaporation is expected to increase, while precipitation is expected to decrease (Fisher-Jeffes, 2015) Based on the expected changes in evaporation and precipitation from 31 different climate change scenarios – it appeared very likely that SWH systems (as with other water resource schemes) will be negatively impacted by climate change

Losses could however be reduced through the use of MAR in place of open storage – as

is the case of the AWMRS

DRIVERS FOR, AND BARRIERS TO, SWH IN RSA

Drivers: Wright (1996) appears to have been the first to moot the possibility of the

widespread use of SW as a supplementary water resource in the RSA Recently however, LID –

and its adaptation to SWH – has found policy support at the national level through the

recently-released Integrated Urban Development Framework (IUDF) – a policy document outlining how

best to manage future urbanisation in the RSA (DoCGTA, 2014) This states that the approach to

SW management needs to change: SW must be considered as part of the water cycle; SW should

no longer be ‘just’ an extension of ‘roads’ departments; and SuDS (LID) approaches should be

adopted Added to this has been the influence of the green building industry (GBI) in the

country; and the emergence of Socially Responsible Investment (SRI), i.e the use of capital for

an acceptable return on investment that supports environmental, social and governance issues so

as to influence the manner in which investors or consumers make decisions and the associated

disclosure mechanisms (Nurick & Cattell, 2013) ‘Sustainable Development’ is increasingly seen

to be important in the RSA with the concept of ‘doing the right thing’ acknowledged as an

important driver Furthermore, the publication of the SuDS guidelines (Armitage et al., 2013),

ongoing SuDS / LID research, and various municipal and industry initiatives has seen the

increased adoption of LID principles in the RSA

Barriers: Local climatic factors have a major influence on the viability of SWH; the more

variable the climate, generally the more expensive the water – although, of course, the existence

of SWH increases the resilience of the system to climatic shocks Other barriers in the RSA

identified by Fisher-Jeffes et al (2012) and Ellis et al (2016) include: fragmented water

management institutions; underfunded water service providers; resistance to innovative

approaches; lack of political will; uncoordinated frameworks; lack of long-term vision and

innovation retarding technocratic path dependencies; poor monitoring systems; technical

capacity problems; and lack of local experience and knowledge These resemble many of the

barriers consistently highlighted in Australian literature; particularly by Brown & Farrelly (2009)

who undertook a meta-analysis of existing peer-reviewed, empirical and analytical literature; and

Tjandraatmadja et al (2014) – a more recent but smaller survey-based study in South Australia

Philp et al (2008) further highlight the potential risks associated with large open bodies of water (e.g SWH retention ponds) which include, inter alia: drowning and increases in the

mosquito population (associated with, e.g malaria, dengue and the Zika virus) Risks such as

these could be managed by rather making use of MAR – i.e no standing water – but such an

approach is unavailable where suitable aquifers are not present or where dolomitic formations

(leading to the risk of sink-hole formation) are common There is also the practical problem of

ensuring an adequately high rate of infiltration into the aquifer

POSITION STATEMENTS ON THE PROMOTION OF SWH IN RSA

Fisher-Jeffes et al (in prep.) propose three ‘position statements’ describing how SWH can

contribute to: improving water security and increasing resilience to climate change in urban

areas; preventing frequent flooding; and providing additional benefits to society in the RSA:

1 SWH improves water security Harvested SW may be used in a ‘fit-for-purpose’

manner to meet a variety of end-use water demands If needs be, as in the case of Singapore, it is possible to treat harvested SW to potable standards, although typically

SW will be most economical in the RSA when used for non-potable end-uses including,

inter alia: irrigation, toilet flushing etc

2 SWH prevents flooding Any system that temporarily or permanently detains a

substantial proportion of the runoff from a catchment will have a positive impact on reducing flooding; the implementation of Real Time Control (RTC) systems for the management of the pond outflows could further increase the degree to which storm runoff is attenuated in SWH schemes

3 SW harvesting provides additional benefits An LID approach could offer significant

additional benefits in the RSA including, inter alia: the creation of recreational areas,

flood mitigation, water quality improvement, increased property values etc

These three position statements form the basis of a strategic framework that is currently being formulated to recognise and promote the potential value of SWH in RSA in line with the

‘Dublin principles’ (UN, 1992) including, inter alia: that water is a finite and vulnerable

resource, essential to sustaining life, development and the environment; that water development

and management should be based on a participatory approach, involving users, planners and

policy-makers at all levels; and that water has an economic value in all its competing uses and

should be recognised as an economic good To this end further research is being undertaken to

assess the viability of new SWH schemes in the City of Cape Town – most particularly on the

central area of the Cape Flats The results of these studies will be used to further inform the

development of the proposed framework

CONCLUSIONS

It is clear that SWH can substantially increase the availability of water in urban areas and should thus be a part of the RSA’s strategy towards addressing water scarcity now and into the

future There are, however, a number of barriers to its wider use that need to be addressed While

SW could be treated to potable standards – as has been done in Singapore – this may not be

economically feasible and it may be preferable to use LID to treat runoff to non-potable

standards and use it for ‘fit-for-purpose’ applications such as, inter alia: garden irrigation, toilet

flushing etc LID systems – especially those enhanced with RTC – offer additional benefits such

as: mitigating flooding through storing runoff; possible opportunities for amenity (e.g

recreational areas); as well as supporting biodiversity, e.g through forming green corridors with

indigenous vegetation

REFERENCES

AMEC, Center for Watershed Protection, Debo and Associates, Jordan Jones and Goulding, &

Atlanta Regional Commission (2001) Georgia Stormwater Management Manual Volume 1:

Stormwater Policy Guidebook First Edition United States of America Retrieved from

http://www.georgiastormwater.com/

Armitage, N., Vice, M., Fisher-Jeffes, L., Winter, K., Spiegel, A., & Dunstan, J (2013) South

African Guidelines for Sustainable Drainage Systems (WRC Report: TT558/12) Water

Research Commission (South Africa) South Africa ISBN: 09781431204137

Brown, R R., & Farrelly, M A (2009) Delivering sustainable urban water management: A

review of the hurdles we face Water Science and Technology, 59(5), 839–846

doi:10.2166/wst.2009.028 ISBN: 0273-1223 (Print) ISSN:02731223

DECNSW (2006) Managing Urban Stormwater: Harvesting and Reuse Department of

Environment and Conservation NSW Sydney South, Australia Retrieved May 22, 2012 from www.environment.nsw.gov.au

Duncan (1995) A review of urban stormwater quality processes Cooperative Research Centre

for Catchment Hydrology Australia ISBN: 1876006064 Retrieved June 26, 2012 from http://www.ewater.com.au/archive/crcch/archive/pubs/pdfs/technical199509.pdf

DWAF (2010) The Atlantis Water Resource Management Scheme: 30 years of Artificial

Groundwater Recharge Department of Water Affairs and Forestry Pretoria, South Africa

Ellis, D., Armitage, N P., & Carden, K (2016) Effective Water Sensitive Design Drivers And

Barriers In Water Institute of South Africa (WISA) Biennial Conference Durban, South

Africa

Evans, A (2007) River of Life: River Liesbeek Rivers of the World, Thames Festival Project

Retrieved from http://totallythames.org/images/uploads /documents/River of the World Info Packs/River_Liesbeek_South_Africa.pdf

Fisher-Jeffes, L (2015) The viability of rainwater and stormwater harvesting in the residential

areas of the Liesbeek River Catchment , Cape Town University of Cape Town

Fisher-Jeffes, L., Carden, K., Armitage, N P., & Winter, K (n.d.) Stormwater harvesting :

Improving water security in South Africa ’ s urban areas

Fisher-Jeffes, L., Carden, K., Armitage, N., Spiegel, A., Winter, K., & Ashley, R (2012)

Challenges Facing Implementation of Water Sensitive Urban Design in South Africa In 7th Conference on Water Sensitive Urban Design (pp 1–8) Engineers Australia Melbourne,

Australia, Australia ISBN: 9780858258952

Lee, J., Pak, G., Yoo, C., Kim, S., & Yoon, J (2010) Effects of land use change and water reuse

options on urban water cycle Journal of Environmental Sciences, 22(6), 923–928

doi:10.1016/S1001-0742(09)60199-6 ISSN:10010742 Retrieved from http://linkinghub.elsevier.com/retrieve/pii/S1001074209601996

Lim, M H., Leong, Y H., Tiew, K N., & Seah, H (2011) Urban stormwater harvesting: A

valuable water resource of Singapore Water Practice and Technology, 6(4)

doi:10.2166/wpt.2011.067 ISBN: 1751-231X ISSN: 1751231X

Makepeace, D K., Smith, D W., & Stanley, S J (1995) Urban stormwater quality: Summary of

contaminant data Critical Reviews in Environmental Science and Technology

doi:10.1080/10643389509388476 ISBN: 1064-3389 ISSN:1064-3389 Retrieved July 27,

2014 from http://www.tandfonline.com/doi/abs/10.1080/10643389509388476

Marsalek, J., Karamouz, M., Goldenfum, J., & Chocat, B (2006) Urban water cycle processes

and interactions International Hydrological Programme (IHP) of the United Nations

Educational, Scientific and Cultural Organization (UNESCO)

Murray, E C., & Tredoux, G (2004) Planning water resource management: The case for

managing aquifer recharge In 2004 Water Institute of Southern Africa (WISA) Biennial Conference (pp 430–437) Cape Town, South Africa (2-6 May 2004) ISBN: 1920017283

Pandey, D N., Gupta, A K., & Anderson, D M (2003) Rainwater harvesting as an adaptation

to climate change Current Science, 85(1), 46–59

Philp, M., Mcmahon, J., Heyenga, S., Marinoni, O., Jenkins, G., Maheepala, S., & Greenway, M

(2008) Review of Stormwater Harvesting Practices Urban Water Security Research Alliance Technical Report No 9 Urban Water Security Research Alliance Technical Report No 9

ISSN:1836-5566

Tjandraatmadja, G., Cook, S., Chacko, P., Myers, B., Sharma, A K., & Pezzaniti, D (2014)

Water Sensitive Urban Design Impediments and Potential: Contributions to the SA Urban Water Blueprint - Post-implementation assessment and impediments to WSUD Goyder

Institute for Water Research Adelaide, Australia Retrieved March 21, 2016, from http://www.goyderinstitute.org/uploads/WSUD_Task1_Final_web.pdf

UN (1992) The Dublin Statement on Water and Sustainable Development Report from

International Conference on Water and the Environment (ICWE) Retrieved May 25, 2012, from http://www.wmo.int/pages/prog/hwrp/documents/english/icwedece.html

USEPA (2012) Low Impact Development (LID) United States Environmental Protection

Agency Retrieved May 23, 2012, from epa.gov/owow/NPS/lid/

Wright, A (1996) Urban Stormwater, Correctly Managed, is a resource rather than a nuisance

In Water Institute of South Africa 1996 Biennial Conference & Exhibition Port Elizabeth

Water Institute of South Africa

Development of a Low Impact Development and Urban Water Balance Modeling Tool

Steve Auger1; Yuestas David2; Wilfred Ho3; Sakshi Sani4;Amanjot Singh5; Tim Van Seters6;

Chris Davidson7; Melanie Kennedy8; and Kevin MacKenzie9

1Lake Simcoe Region Conservation Authority, 120 Bayview Pkwy., Newmarket, ON L3Y 3W3

E-mail: S.Auger@lsrca.on.ca

2Toronto and Region Conservation Authority E-mail: YDavid@trca.on.ca

3Toronto and Region Conservation Authority E-mail: WHo@trca.on.ca

4Credit Valley Conservation Authority, 1255 Old Derry Rd., Mississauga, ON L5N 6R4 E-mail:

Ssaini@creditvalleyca.ca

5Credit Valley Conservation Authority, 1255 Old Derry Rd., Mississauga, ON L5N 6R4 E-mail:

Asingh@creditvalleyca.ca

6Toronto and Region Conservation Authority E-mail: TvanSeters@trca.on.ca

7Golder Associates Ltd., 6925 Century Ave., Mississauga, ON L5N 7K2 E-mail:

CAs,” have worked collaboratively to identify preferred low impact development (LID)

stormwater management (SWM) models suitable for meeting typical design criteria The GTA

CAs research and experimentation efforts with various LID SWM case studies will inform the

audience of model criteria and rationale that supports the selection of preferred open-source

and/or commercial license LID modeling tool(s) which effectively demonstrate design targets

have been met The GTA CAs commissioned Golder Associates Ltd to provide programming

support and water resources expertise to support the development of the LID Treatment Train

Tool (LID TTT) for Ontario, presently in a working beta version for review and experimentation

before a planned hard launch in late March, 2017 The primary intention of the LID TTT is to

support more consideration and realization of LID opportunities for a proposed site plan,

throughout the design process undertaken by planners, SWM designers, and other decision

makers for all types of site development or retrofit projects The LID TTT will process

computational results from EPA-SWMM to provide an assessment of stormwater runoff volume

and associated target depths, along with total suspended solids (TSS) and total phosphorus (TP)

reductions for both annual and event based scenarios, throughout the design process A

preliminary water budget assessment for pre and post conditions is also provided in the beta

version of the LID TTT

KEYWORDS: Low Impact Development; Treatment Train; Preliminary Design; Modeling;

Stormwater Volume Retention; Infiltration; Water Balance; Total Phosphorus Reduction; Total

Suspended Solids Reduction

INTRODUCTION

The Low Impact Development Treatment Train Tool (LID TTT) has been developed by Lake Simcoe Region Conservation Authority (LSRCA), Credit Valley Conservation (CVC) and

Toronto and Region Conservation Authority (TRCA) as a tool to help developers, consultants,

municipalities and landowners understand and implement more sustainable stormwater

management planning and design practices in their watersheds

SELECTING A MODEL

Computer models that aim to simulate water and/or nutrient transport over an area can be categorized into the following three main categories: hydrologic, hydraulic and water quality

models (MPCA 2015) Some models may cover multiple categories Each type of model serves a

different purpose and therefore can function on very different principles and inputs Such stark

differences mean users need to undertake careful evaluation of potential models This requires

first a preliminary review of available models to understand their functionalities/limitations and

inputs/outputs and subsequently selecting the model that best satisfies the project objectives

The Minnesota Pollution Control Agency (MPCA) has published a review summary of the most commonly used water modeling software and has categorized them using the following

model type definitions Hydrological models include rainfall-runoff modeling and

reservoir/channel routing, partitioning precipitation through conservation of mass Hydraulic

models consider the mechanics of flowing water through flow rates, flow velocities, and water

surface profiles through waterways, structures and pipes Some may also include functions to

model the water balance (inflow, infiltration, evapotranspiration, storage, and discharge) of green

infrastructure Water quality models consider pollutant loading to surface waters or pollutant

removal by a Best Management Practice (BMP) MPCA (2015) also acknowledge the combined

hydrologic and hydraulic model, rainfall-runoff calculation tools and BMP calculators; the latter

two representing less complex versions of hydrologic and water quality models respectively

Categorized models are further evaluated based on the following characteristics: input

complexity, simulation type(s) (event or continuous), availability in the public domain, functions

to model unsteady flow, type of water quality model (e.g., loading, receiving water, BMP),

BMPs represented, ability to model TP, TSS, and volume dynamics (MPCA 2015) The GTA

CAs used very similar characteristics to evaluate available models and additionally considered

input complexity and adaptability specifically for LIDs

The most complex model with more numerous functionalities is not always the best model to choose The US Environmental Protection Agency (US EPA) recommends choosing the simplest

model that would meet the project objectives (US EPA 2015), since greater number of

parameters can propagate higher uncertainties in the results Therefore, developing clear project

objectives is fundamental to selecting the most fitting and efficient model for the project Some

project objectives can help to quickly shortlist considered models (e.g., availability in the public

domain) while others may warrant more detailed comparisons MPCA (2015) suggest

considering project objectives regarding the following categories: regulatory compliance,

hydrologic process, land use, area to be modeled, intended use, model complexity, and modeler

experience To evaluate how models comply with project objectives, as an alternative to the

checklist approach, the GTA CAs applied a weighting criteria based on the relative importance

of each objective Project members voted on model choice based on their separate evaluation of

shortlisted models scored on weighted categories such as model applicability and relevance for

the management tasks, simulating important variables, suitability for simulation scenario,

resources required for development (time, materials, and expenses), model transparency, ease of

understanding, ease of use, model sensitivity, uncertainty estimations, and adaptability The GTA

CAs further tested and compared shortlisted models using case studies with monitored

input/output data

Elliot and Trowsdale (2006) reviewed 10 existing stormwater models that incorporated LIDs and noted that most models can be used to investigate effects of reducing imperviousness but

gaps in development include broadening the range of contaminants represented, improving

transport routines within streams and treatment devices, including linkages between baseflow

and runoff components of pervious surfaces, addition of calibration and prediction uncertainty

routines and up-scaling results of on-site devices at a catchment level They noted that there is a

trend in introducing LID devices into stormwater models by modifying the existing model,

building new models based on existing models, or documenting how to model LID devices

indirectly using existing parameters Upon evaluation of shortlisted models the GTA CAs

decided to take the former approach of modifying the existing EPA SWMM model to better

satisfy project objectives of CAs in managing stormwater

OBJECTIVES

The LID TTT has been designed to support practitioners with an initial evaluation of the effectiveness of LID SWM features to meet stormwater volume reduction criteria, along with

both Total Phosphorus (TP) and Total Suspended Solids (TSS) load reductions, to support a

preliminary assessment of the treatment train removal efficiency for annual and event based

scenarios, for different LID SWM features considered for a site

The LID TTT was designed to predict overall pollutant reduction computational objectives

The LID TTT will also be used to provide a preliminary water budget to determine the amount of

runoff volume that will be reduced through infiltration, evapotranspiration or re-use, based on

site conditions (including soil texture, vegetation cover), BMP design (e.g impervious to

pervious area ratio, underdrain configuration, storage available for re-use), drainage area treated

by LID, and other site specific factors

3 Non-proprietary (Free to use)

4 User friendly and visually compelling graphic interface

5 Input and computational results are provided in metric units

6 Reasonable representation of range of LID features available for consideration

7 Performance goal adaptability based on site opportunities and constraints

8 Volume of storage required at end of pipe to meet site specific flood control criteria

REVIEW OF DIFFERENT LID MODELS

The GTA CAs started down the path towards developing the LID TTT by performing a comparative review of existing LID modeling tools This comparative work reviewed the B.C

Water Balance Model, the EPA National Stormwater Calculator, EPA-SWMM, GSI-Calc,

Hydrotrek calculator, MIDS GUI Credit Calculator, and WIN-SLAMM The GTA CAs selection

criteria for short listing the LID models considered for a more comprehensive review focused on

the following: 1 Use by practitioners; and 2 Open source /flexibility to support adaptation The

GTA CAs review of the model characteristic and functionality of the short listed models is

EPA

MIDS GUI Credit Calculator

One-dimensional

dimensional N/A

Input for LID

LIDs lumped vs distributed Lumped Distributed Lumped Distributed

Output Temporal Scale Minimum (Daily) Min (User

defined)

Minimum (Daily) Annual Output Spatial Scale Site - Watershed Site –

restrictions)

Yes (w restrictions) These comparative results summarized in Table 1, was supported through the review of two case study examples (1 Mosaik Glenway residential subdivision in Newmarket, and 2 Honda

Campus Canada commercial head office in Markham) where LID features were utilized as the

primary means of providing site stormwater management

The short listed models were reviewed by using the approved SWM reports for each case study to re-evaluate the SWM features using the selected models in Table 1 In the original

SWM report, submitted and approved, the Mosaik Glenway Subdivision was modelled using

Visual OttHymo to evaluate water quantity controls, and the Best Management Practices (BMP)

in series equation to evaluate water quality control The Honda Campus Canada site was

modelled with more conventional modified rational calculations in a conventional spreadsheet,

along with reference to accepted empirical removal efficiencies for the LID features Both site

designs adhered to the LID sizing guidelines from the 2010 CVC / TRCA Planning and Design

the development of the LID TTT The EPA-SWMM modeling results consistently demonstrated

the closest comparison to water quantity and quality predictions presented in the SWM reports

One of the major limitations of all these models when considering an LID Tool for Ontario is that they do not reflect local climatic conditions The effort required to adapt any one of these for

local conditions would have been comparable to building the model interface from scratch

USER INTERFACE

The application is comprised of six key user screens, depicted in the Figure 1 montage

1 Application Start Up screen – allows the user to open a project, create a project or

compare results between two projects

2 Project Start Up screen – allows the user to insert the project information details (i.e.,

project name, location, storm type and water balance targets)

3 The Main Interface screen – allows the user to create, edit and delete elements

4 Element Parameter Input screen – allows the user to enter the details for specific

elements

5 Element Table of Contents screen – displays all the created elements in the Project

6 Compare Results screen – allows the user to view the results of the model run and

compare pre- and post-development scenarios

Figure 1: Montage of LID TTT – User Interface key user screen