This section refers to the methodology used to identify the intervention opportunities for green infrastructure in Chelsea MA and to estimate their quantitative impact to the water system performance on an annual basis. Our framework included computational methods from the fields of green infrastructure and urban water management in combination with statistical analysis, geographic information systems, and algorithmic modeling.
However, the Zofnass Information Tool is not a modeling tool that performs hydrological simulations. Thus, it cannot substitute the hydrology-hydraulic models or the on-site assessments that certified professionals use and perform.
Our objective was to achieve the meaningful computational abstraction that describes the degree of the impact of green infrastructure to the water system’s annual performance, so that it serves the learning scope of this website. You can read below the set of assumptions and the summary of limitations regarding the computational framework.

1. Green Infrastructure Opportunities

1.1 Green Roofs

Based on data retrieved from Assessor’s datasets of MassGIS 2015, as intervention opportunities are identified the buildings with flat roofs, of minimum rooftop area of 1,000 sq.ft (93 sqm). The buildings are residential (condo, apartment 4 story, apartment 8 story, and public housing) and land use data intersecting buildings (e.g. industrial, commercial, urban public and institutional). Since no data was available on the structural capacity of buildings to withstand the weight of green roofs, the buildings built in 1980 or after were selected as such.

The stormwater runoff estimates used a mean green roof coverage of 75% for each rooftop area, while the rest 25% is considered as impervious surface. The sub-catchment area is defined by the 100% of the green roof area. For this sub-catchment area the local runoff is calculated based on the annual runoff volume reduction rate of 0.45 as defined by Virginia DEQ Stormwater Design Specification No. 5.

The opportunities are prioritized according to macro intervention strategies. Primarily sorted in groups their building use and ownership: public institutional, industrial, commercial, and residential. Secondarily, buildings in each group are sorted from the largest to the smallest rooftop area.

1.2 Street Planters

Based on data from MassDOT, as intervention opportunities, the road classifications 4 and 5 (e.g. major road and minor road) were extracted. The roads with slope less than 5 percent were selected based on 3 meter contours topography. We also assume that 30 percent of the sidewalk’s entire length is covered with street planters. We assigned sidewalk widths of 5 ft for circulation which can meet ADA requirement and a minimum of for street planter implementation. The suitable selected streets exclude sites with existing contamination problems (21E sites), sites with activity and use limitations (AUL sites) as defined by EPA for the City of Chelsea.

The sub-catchment area is defined as 2 times the street planter area. For this sub-catchment area the local runoff is calculated based on an annual runoff volume reduction rate of 0.40 as defined by Virginia DEQ Stormwater Design Specification No. 9.

The required depth to seasonal high water table (2.3 feet or more) is a critical constraint. Therefore the opportunities are prioritized in two groups: first in areas of high water table depth (EPA for the City of Chelsea) and in the rest areas where we assume raised planters. Secondarily, the streets in each group are sorted from the largest to the smallest total planter area.

1.3 Permeable Pavement

Based on data retrieved from Assessor’s Parcels (Level 3) datasets of MassGIS 2015, as intervention opportunities are identified the parking lot parcels. Site constraints were determined by percentage of slope with less than 5 percent based on 3 meter contours topography analysis. The suitable selected areas exclude sites with existing contamination problems (21E sites), sites with activity and use limitations (AUL sites) as defined by EPA for the City of Chelsea.

The sub catchment area is equal to 100% of the permeable pavement area. For this sub-catchment area the local runoff is calculated based on the annual runoff volume reduction rate of 0.45 as defined by Virginia DEQ Stormwater Design Specification No. 7.

1.4 Rainwater Harvesting

To define opportunities, land use and zoning parcel data were used to assign pitched roofs for residential and institutional buildings in MassGIS 2015 by selecting buildings with non-flat rooftops. Rain barrels were assigned to institutional and residential parcels.


2. Water System

2.1 Computation method

The annual average performance of system flows is estimated based on alterations in the volume of urban runoff caused by green infrastructure interventions. We used open historical data from the Massachusetts Water Resources Authority (MWRA) which provides water supply and wastewater services to the City of Chelsea, and formulas from the literature for data that were not available. Average monthly data were gathered in million gallons per day (MGD) over the calendar years 2010-2014 for the water use, wastewater flow, stormwater inflow, groundwater infiltration, sanitary flow, and rainfall, while no respective recorded data is available for the runoff and drainage flows in Chelsea.

2.2 Water Supply

The initial demand for imported potable water in Chelsea from MWRA is considered 3.08 MGD, the average monthly water use for the years 2010-2014 based on MWRA data. The unaccounted water it is considered always as of 12% of the purchased imported water for every month due to non-availability of monthly data. Accordingly, we estimate the average end-use water consumption as of 2.71 MGD that is considered steady.

2.3 Stormwater Inflow

Historical data analysis has revealed a very strong linear relationship between stormwater inflow and rainfall (correlation coefficient of 0.914) based on monthly MWRA data over the years 2010-2014. Since the Simple runoff coefficient method was used, the runoff and rainfall have also a linear relationship. For annual average flows, the relation stormwater inflow = 0.35(runoff) was used as the result of a least squares regression analysis on the same data sets. The annual rainfall and runoff are converted from inches to gallons by the assumption that one inch of rain falling on 1 acre of ground is equal to about 27,154 gallons (U.S. Geological Survey).

2.4 Wastewater Flow

The total wastewater flow of Chelsea is the sum of stormwater inflow, groundwater infiltration and sanitary flow. The sanitary flow is considered as a constant of 2.80MGD as always in the 2010-2014 MWRA data . For the estimation of wastewater flow we assume the groundwater infiltration as steady of 1.14 MGD (average of MWRA monthly data) since no significant correlation with the rainfall was revealed.

2.5 Drainage

The remaining amount of surface stormwater (runoff - stormwater inflow) is considered as inflow to the drainage system. Based on observations on terrain slope analysis and other feedback we assumed that Chelsea doesn’t receive a considerable amount of stormwater from the neighboring communities.

3. Stormwater model

3.1 Urban Stormwater Runoff

For the purpose of estimating the average annual runoff, the rational runoff coefficient method was used. The impervious cover that is used concerns the area within the administration boundaries of Chelsea. The original impervious fraction for Chelsea is 75.08% corresponding to 1.66 sq.miles (4.299 of impervious area out of a total area of 2.21 sq.miles (5.723 as in Stormwater Regulated Municipalities by EPA.The annual rainfall is considered constant (44.90 inches) throughout the calculations based on the average monthly data by NOAA for the period 2010-2014.

3.2 Impact from GI Opportunities

When new intervention opportunities of green infrastructure (GI) are accounted, the calculation method is modified in order to better capture their contribution in reducing runoff. The Simple method is still used for the area of Chelsea (global), but now it excludes the sub-catchment areas of the added green infrastructure. The runoff for the GI. sub-catchment areas is calculated in the local level by the percentage of annual reduction as defined above. The total urban runoff is the sum of the global and local results.


The computational framework that identifies green infrastructure opportunities and quantifies their impact to the urban water system is a first attempt for an integrated computational approach on aggregated infrastructure projects for cities. The methodology we used is based on public geospatial data and information from water authorities, so the results are sensitive to any issues these datasets might have. This framework includes literature review-based assumptions, but it can be further enhanced by employing more advanced models for computing stormwater runoff, such as the curve number or the non-linear runoff method.
In the near future, the deployment of sensors in the city (e.g. the internet of things) is likely to provide better and more diverse data on the urban environment, green infrastructure, and water systems. Then, our computational framework could provide better predictions, reduce the margins of error, refer to shorter time frames, and determine as well the financial values of the interventions.
For the moment, we see our computational method to have a value to show the potential of an integrated systemic approach and facilitate the collaboration for different stakeholders involved in cities.


This is a beta version of the Zofnass Information Tool website, in the process of being tested before its official release. This being so, the Zofnass Program does not give any warranties, whether express or implied, as to the suitability or usability of the website, its software or any of its content.

The Zofnass Program for Sustainable Infrastructure will not be liable for any loss, whether such loss is direct, indirect, special or consequential, suffered by any party as a result of their use of the Zofnass Information Tool website, its software or content.

Should you encounter any bugs, glitches, lack of functionality or other problems on the website, please contact us on email. Your help in this regard is greatly appreciated.

Copyright © 2016 President and Fellows of the Harvard College, and Zofnass Program for Sustainable Infrastructure, All rights reserved.