Environ. Sci. Technol. 2010, 44, 4302–4308
Scaling of Economic Benefits from Green Roof Implementation in Washington, DC H A O N I U , †,‡ C O R R I E C L A R K , § J I T I Z H O U , † A N D P E T E R A D R I A E N S * ,‡ School of Environmental and Biological Science and Technology, Dalian University of Technology, Department of Civil and Environmental Engineering, College of Engineering, University of Michigan, and Environmental Science Division, Argonne National Laboratory, Washington, DC
Received August 11, 2009. Revised manuscript received April 9, 2010. Accepted April 26, 2010.
Green roof technology is recognized for mitigating stormwater runoff and energy consumption. Methods to overcome the cost gap between green roofs and conventional roofs were recently quantified by incorporating air quality benefits. This study investigates the impact of scaling on these benefits at the city-wide scale using Washington, DC as a test bed because of the proposed targets in the 20-20-20 vision (20 million ft2 by 2020) articulated by Casey Trees, a nonprofit organization. Building-specific stormwater benefits were analyzed assuming two proposed policy scenarios for stormwater fees ranging from 35 to 50% reduction for green roof implementation. Heat flux calculations were used to estimate building-specific energy savings for commercial buildings. To assess benefits at the city scale, stormwater infrastructure savings were based on operational savings and size reduction due to reduced stormwater volume generation. Scaled energy infrastructure benefits were calculated using two size reductions methods for air conditioners. Avoided carbon dioxide, nitrogen oxide (NOx), and sulfur dioxide emissions were based on reductions in electricity and natural gas consumption. Lastly, experimental and fugacity-based estimates were used to quantify the NOx uptake by green roofs, which was translated to health benefits using U.S. Environmental Protection Agency models. The results of the net present value (NPV) analysis showed that stormwater infrastructure benefits totaled $1.04 million (M), while fee-based stormwater benefits were $0.22-0.32 M/y. Energy savings were $0.87 M/y, while air conditioner resizing benefits were estimated at $0.02 to $0.04 M/y and avoided emissions benefits (based on current emission trading values) were $0.09 M-0.41 M/y. Over the lifetime of the green roof (40 years), the NPV is about 30-40% less than that of conventional roofs (not including green roof maintenance costs). These considerable benefits, in concert with current and emerging policy frameworks, may facilitate future adoption of this technology.
* Corresponding author phone: 734-763-8032; Fax: 734-763-2275; e-mail:
[email protected]. † Dalian University of Technology. ‡ University of Michigan. § Argonne National Laboratory. 4302
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 44, NO. 11, 2010
1. Introduction Rapid urbanization and the associated need for energy and water infrastructure are making it difficult for our aging stormwater systems to comply with the requirements of the Clean Water Act (CWA) (1). At the same time, increased air pollution in cities and increased temperatures due to the urban heat island effect (UHIE) in mid- and high-latitude cities are exacerbating local air quality violations (2-4). Due to the scale of the problem, green infrastructure is increasingly considered part of the solution, with green roofs becoming a scalable technology considered by policymakers, developers, and building owners (5). While extensive green roof systems are considered to be best management practices (BMPs) for stormwater management (6), these technologies also have the potential to reduce thermal gain as compared to conventional roofs (7) and can be effective at intercepting pollutants in urban environments. This benefit adds to the inherently lower pollutant loads in stormwater runoff from green roofs as compared to conventional roofing materials, which have been shown to be substantial contributors of hydrocarbons and metals through leaching (8). The insulating abilities of green roofs in summer have been quantified (9, 10), and are based on their capacity to reduce the heat flux in a building due to lower surface temperatures (11, 12) as well as from the combination of the thermal conductivity and thickness of the soil layers (13). When green roof technology is scaled from individual buildings to the city scale, additional benefits need to be considered such as UHIE mitigation, health impacts, and avoided air pollutants emissions, as well as reduced stormwater infrastructure needs. Rosenzweig et al. (14) found that average roof surface temperatures in New York could be reduced by 0.1-0.8 °C if 50% of the city were greened. Reduced temperatures decrease energy demands for cooling, thereby mitigating emissions from electricity generation, resulting in health benefits (15). The potential for pollutant uptake by green roofs has been quantified through the use of multimedia models with limited validation (5, 16, 17). These aggregate benefits have been used to estimate the economic benefits of green roofs at the building-specific (5, 18) and community-wide scales (19). At this scale, either energy savings or air quality improvement provides the greatest economic benefit (5, 18). At the city scale, avoided air pollution emissions become significant due to energy saved from reduced use of electricity and natural gas (20). Results from prior net present value analysis (NPV) indicate that the breakeven point is estimated at around 8 years under the most advantageous conditions (5, 18). The objectives of this paper are to (i) quantify and scale the environmental benefits of green roofs from individual buildings to the city scale and (ii) incorporate them into an economic framework (NPV) model to calculate the range of breakeven points.
2. Materials and Methods The city scale test bed, Washington, DC. encompasses the greening of 20% of green roof ready buildings, including 0.08 M m2 of buildings with roof areas smaller than 100 m2, 0.26 M m2 of buildings with roof areas 100-200 m2, 0.23 M m2 on buildings with roof areas 200-500 m2, and 1.34 M m2 of buildings with roof areas larger than 500 m2 (Supporting Information (SI) Table S1) (1). Installation Cost for Conventional and Green Roofs. The installation cost gap between conventional and green roofs was based on a commercial building of 1795 m2 (5). The cost 10.1021/es902456x
2010 American Chemical Society
Published on Web 05/12/2010
of a conventional roof replacement varies depending on the thickness of the asphalt and membrane, the thickness of new insulation, permitting, and access. In Washington, DC, many clients prefer to select a TPO (thermoplastic elastomerolefin) or other single-ply system for their conventional roof, and the mean cost is $242/m2 (personal communication, Gregory Long, Capitol Greenroofs). The distribution of green roof installation costs varied more than that for conventional roofs according to design and function. The cost data used in this research were limited to extensive roofs, and were estimated at $306/m2 (standard deviation: $44.56/m2) based on available green roof case data (21) (see SI Table S2). While there are associated maintenance costs for a green roof system (from $13/m2 to $21/m2), they were not included in this analysis because they are only incurred until the plants are established (approximately two years) (22). The characteristics of the green roof used in this study are provided in SI Table S2. Stormwater Benefits. Washington, DC, recently updated its stormwater fee structure to accurately charge according to stormwater generation, resulting in a fee (estimated at $0.33/m2.year) on impervious surfaces to meet the District’s annual $13 M obligations. This new fee went into effect on May 1, 2009, and accounts for the high proportion of government facilities and other large stormwater generators while shifting the stormwater cost burden from other property classes such as multifamily residences (23). Concurrent with the fee increase, an incentive program is being developed to allow for reductions in stormwater fees when best management practices (BMPs) such as green roofs, rain gardens, or other strategies are implemented. Estimates range from granting up to a 50% fee reduction for installing a green roof (24), or a 35% fee reduction similar to the Clean River Rewards program in Portland, Oregon (25). Two scenarios were evaluated in our analysis that combine the fee structure and fee reductions for green roofs: $0.33/m2 fee with a 50% fee reduction and $0.33/m2 fee with a 35% fee reduction. The quantitative assessment of stormwater uptake by green roof systems was based on estimates reported by Deutsch et al. (1) indicating the potential retention of 3.50 × 108 L (9.50 × 107 gal) of stormwater per year. An operational cost of $0.003/L was used based on costs associated with pumping and treatment of wastewater (1). Stormwater infrastructure size reduction benefits were calculated using cost data from a combined sewer overflow (CSO) tunnel project (26). The total budget of this project was estimated at $1.265 billion, including all of the costs (materials, equipment, contingencies, transport, labor cost, overheads, additional parts, site mobilization/clearance), and we assumed a 40 year finance period. The green roof plan associated with the 20-20-20 vision has a capacity to retain about 4.2% of the total stormwater volume (1). Assuming that the benefits would only be realized in material costs, this comprises 23-29% (median value: 26%) of the total cost of a typical tunneling project (27). Energy Savings Valuation. Energy savings are based on decreased natural gas and electricity consumption. Heat flux through the roof was determined using EnergyPlus v3.1.0, a building energy simulator (U.S. Department of Energy, DOE). It was designed to model heating, cooling, lighting, ventilating, and other energy flows, as well as water, in buildings based on climate and building use, material, and size inputs (28). The software has the capability to incorporate a green roof (referred to as an ecoroof) on a building. The ecoroof module accounts for heat flux by means of a 1-dimensional heat transfer model. The EnergyPlus parameters are shown in SI Table S3. While the model considers heat transfer processes within the soil and plant canopy, it does not include soil moisture-dependent thermal properties of the green roof (5, 29). According to the vision plan for Washington, DC stated
TABLE 1. Uncertainty Parameters Used in NPV Analysis uncertainty parameters NOx ($/Mg) SOx ($/Mg) CO2(low estimate) ($/Mg) CO2(high estimate) ($/Mg) stormwater infrastructure cost (M $)
range (n)a 500-3600 (108) 170-1520 (108) 1.2-7.4 (253) 17.4-46.2 (245) 290-367 (2)
mean value 1440 736
source (34) (34)
3.8
(35)
33.7
(36)
328.5
(27)
a
The number of each input parameters used in the net present value (NPV) analysis.
above, we selected four types of buildings with average roof areas in each green-ready building category (55 m2, 125 m2, 270 m2 for residential buildings, and 1795 m2 for commercial building) as a scaling tool to the whole city. Therefore, at the city scale, the building activity type in the model was defined as “office/professional” for federal commercial building simulations which resulted in a default occupancy density of 3.91 people/100 m2 and a default electric plug load intensity of 8.07 W/m2, and “lodging” for institutional building simulations which result in a default occupancy density of 4.31 people/100 m2 and a default electric plug load intensity of 2.69 W/m2. The energy savings were based on the rates of electricity for commercial and residential customers or $132.3/MWh and $127.9/MWh in Washington D C (30). The price for natural gas heating was assumed to be $46.12/MWh for the commercial sector and $54.74/MWh for the residential sector (in 2008) in Washington, DC (31). Avoided emissions of criteria air pollutants from power plants and space-heating equipment due to green roof implementation were determined based on utility-specific emission factors for electricity and natural gas use (20, 32) (SI Table S4). Avoided emissions from buildings due to reduced electricity and natural gas use has been reported previously. For example, Kagel and Gawell (33) used emission factors to quantify an externality analysis of geothermal energy. Similarly, McPherson et al. (20) incorporated and economic valuation of carbon dioxide (CO2), nitrogen dioxide (NO2), PM10, and volatile organic compounds (VOCs) in a cost-benefit analysis of urban forests. The emission reduction mass estimates were translated into market value, via allowance pricing; the mean values over the past decade were selected for this analysis (Table 1). The prices for NOx and SO2 were based on the U.S. Environmental Protection Agency (EPA) allowance market price in 2010, estimated by the CAIR (Clean Air Interstate Rule) project, and were $1440/Mg and $736/Mg, respectively (34). The price of CO2 was based on the mean 2008 trading values on the Chicago Climate Exchange (CCX) and the European Climate Exchange (ECX), at $3.80/Mg (standard deviation: $1.90/Mg) and $33.70/Mg (standard deviation: $6.70/Mg), respectively (35, 36). Lastly, the size reduction of air conditioners due to building energy efficiency improvements from green roofs was considered. The sizing of a building’s heating, ventilating, and air conditioning (HVAC) equipment depends upon surface temperature, which is related to a variety of factors including the number of building stories, typical building occupancy, location, building insulation, fac¸ade reflectivity and emissivity, air convection from wind, and the characteristics of the green roof that affect soil moisture properties. For the purposes of this analysis, we simulated this benefit based on a direct savings in the cooling ventilation losses due to the change in roof surface temperature by using a VOL. 44, NO. 11, 2010 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
4303
TABLE 2. Electricity and Natural Gas Savings at Building and City Scale energy savings electricity
energy savings
natural gas
roof range (m2)
average roof area (scaler) (m2)
building typea
MWh
$
MWh
$
roof area in each building pool (‘000 m2)
>500 200-500 100-200
