Office of Brownfields and Land Revitalization
Air and Water Quality Impacts of
Broiunfields Redevelopment
United States Environmental
Protection Agency
Office of Solid Waste and
Emergency Response (5101T)
EPA560-F-10-233
April 2011
This fact sheet describes an approach to quantifying the environmental impacts of Brownfields reuse and provides
estimates of impacts on air and water quality of multiple revitalization projects in five municipal regions.m
Overview
A number of previous studies compared the envi-
ronmental performance of specific brownfield
redevelopments with similar projects built on unde-
veloped greenfield sites. These studies examined single
brownfield/infill developments and entailed extensive
site-specific analysis. The comparison sites generally ac-
commodated the same number of residential units and
commercial square footage, but typically occupied more
acreage per employee or residence and were less loca-
tion efficient. A review of 12 of these studies concluded
that brownfield/infill development results in significant
environmental benefits when compared to their green-
field counterparts. However, making broader quanti-
tative assessments applicable to brownfield redevelop-
ment around the country requires a methodology that
is more easily transferable. This study tests an analytical
approach to quantifying the environmental impacts of
multiple redevelopment projects in a given municipal
area in a manner that can be replicated in other regions.
The method was applied to five cities and their surround-
ing areas—Seattle, WA, Baltimore, MD, Minneapolis-St.
Paul, MN, Emeryville, CA, and Dallas-Fort Worth, TX.
Study Approach
For each of the five cities, all known brownfield sites
that benefited from U.S. EPA Brownfields Program
assistance and that have redevelopment completed
or under way were identified. Most of these proper-
ties are in close-in, high density areas. The study also
identified alternative development locations for each of
the brownfield sites, based on prevailing development
trends in the area. That is, it was assumed that had the
development not occurred on the brownfield, it would
have gone to these locations. The environmental per-
formance of both sets of locations were measured and
compared, using metrics such as vehicle use per capita,
air pollutant emissions per capita, personal vehicle en-
ergy use per capita, and stormwater runoff and pollut-
ant loads. The environmental performance measures
were developed with data from regional transportation
demand models, a watershed management model, and
INDEX, a geographical information system (GlS)-based
analytical tool.[2>3]
Development on suburban/exurban sites consumes
more acreage per resident or employee than urban core
project areas. Most but not all of the alternative sites
were located outside the urban core. The study assumed
that these projects were sited on greenfields and would
require 2-4 times the acreage typically used for develop-
ment on brownfield sites. This assumption, believed to
be conservative, is derived from factors drawn from lit-
erature on land use patterns by type of use as well as ex-
perience in the Puget Sound area. Nearly all alternative
locations identified for this study would require more
land to accommodate the same type of development on
brownfield sites.
Five Municipal Areas Included in Study
'Tphe five municipal areas (see table) were selected based
A on several factors, including: the existence of a signifi-
cant number of brownfield properties that met the afore-
mentioned criteria, the availability of information about the
status of redevelopment on the brownfield sites, the availabil-
ity of data that could be used as indicators of local environ-
mental performance, and ecoregion stratification to provide
precipitation profile diversity for the stormwater analysis.
Municipal Areas Studied
City
Seattle
Minn&apolis-
Sainl Paul
Emeryville
Bafoimwe
Dallas-
FL Worth
Total
No. of
Brownfield
Properties
25
37
39
37
25
163
Brownfield
Acreage
87
60
i&3
322
266
8M
Planning Area
4-county area
7-county area
9-county area
Scounli&sS
Baltimore Cily
12-oounly area
Population
In Planning
ATM
(millions)
3.6
2.9
6.1
2.5
6.5
-------
Results
Indicators of environmental performance were estimated
for each of the 163 brownfield sites that met the above
criteria and their hypothetical counterparts. The values
varied widely from site to site, as would be expected given
the wide range of characteristics of the various locations.
Nevertheless, for 90-95% of the sites, the brownfield loca-
tions had superior environmental performance to their
conventional/greenfield counterparts. The results were
averaged for each municipal area and are shown in the at-
tached table. Average results for the five municipal areas
indicate that:
* On average, automobile use by residents and em-
ployees at brownfield locations is estimated to be
substantially lower than at the alternative locations:
• Daily vehicle miles traveled per capita would be
32 - 57% lower.
• Daily vehicle trips per capita would be 16 - 38%
lower.
• Personal vehicle energy use per capita would be
32 - 57% lower.
* Brownfield redevelopments produce 32 - 57% less
carbon dioxide (CO2) emissions per capita relative to
conventional developments.
* Brownfield redevelopments produce 32 - 57% less
air pollutant emissions per capita relative to conven-
tional developments.
* Stormwater runoff for brownfield redevelopments is
estimated to average 43 - 60% less than the green-
field alternatives.
* Brownfield redevelopments also produce substan-
tially lower loads of all pollutants studied, averaging
9 - 80% for conventional pollutants and 59% to 72%
for metals. These estimates do not include the po-
tential reduction in pollutant loads from cleanup of
the brownfield properties, nor some of the pollution
from contaminated properties that would remain
undeveloped if the redevelopment had occurred on
the greenfields instead.
* Based on a literature review, it is estimated that
brownfield sites typically accommodate the same
number of homes and businesses on about 1A (maxi-
mum impact result) to l/z (minimum impact result)
the land typically used at corresponding convention-
al sites. The table at the end of this fact sheet provides
additional detail about these estimates.
Vehicle Miles Traveled
(Miles Per Capita Per Day)
Carbon Dioxide Emissions
(Pounds Per Capita Per Year)
Air Pollutant Emissions
(NOx, CO, HC: Pounds Per Capita Per Year)
Stormwater Runoff
(Percent Reduction for Brownfields)
****** *»«*"** *****
-------
The table at the end of this fact sheet provides additional
detail about these estimates.
Discussion
The air quality results are consistent with the land
use patterns and urban form measures for the areas
studied. On average, neighborhoods where the brown-
field sites are located had higher development density
(population, dwelling units, and employees per gross
acre), more travel accessibility (in terms of distance and
travel time) to other areas, and better access to tran-
sit than the areas where the hypothetical conventional
counterparts are located. For example, the Seattle neigh-
borhoods where the brownfields are located have more
than twice the population density of the conventional
counterpart areas. In addition, the percentage of houses
or jobs within 1A mile of transit in the brownfield neigh-
borhoods is more than double that of the conventional
locations, on average.
The results are also in line with other studies that com-
pare the environmental performance of brownfield/in-
fill development with conventional development. Fur-
thermore, the results are consistent with evaluations of
the relationship between urban built environment and
vehicle use and air emissions using data for wider geo-
graphic areas, such as counties and metropolitan statis-
tical areas. A well-researched summary of this literature
is included in the report Growing Cooler: The Evidence
of Urban Development and Climate Changed The study
estimated that, with more compact development, people
drive 20 to 40% less per year.
Another study using the 1999 Puget Sound Household
Travel Survey and land use measures concluded that resi-
dents make travel choices based on several factors, most
of which are related to time spent traveling, including wait
times, which are, in turn, related to land use patterns.151 In-
creased levels of mixed-use development, retail density, and
street connectivity were associated with lower per capita
emissions and an increased tendency to walk.
Although the results for each city show significant posi-
tive environmental outcomes from building on brown-
fields, the estimates vary from city to city. This variation is
not always empirically explained by a direct comparison
of the average urban form indicators used in this study,
such as population density, employment density, dwell-
ing units per gross acre, and accessibility measures. Di-
rect comparisons of these variables are confounded by
the fact that the results are first differences between the
conventional and brownfield scenarios (i.e., convention-
al scenario less brownfield scenario) and that there are
many factors that vary among cities, such as geographi-
cal barriers, socioeconomic characteristics, existence and
effectiveness of mass transit, physical form of existing
greenfield areas, and economic growth. For example, be-
cause Seattle is surrounded by many bodies of water and
mountains, some of the outlying areas are accessible to
the central city or other destinations only by bridge, ferry,
or circuitous routes. This may explain why the results for
the Puget Sound area indicate considerably greater reduc-
tions in vehicle miles traveled and emissions under the
brownfields redevelopment scenario than for the other
cities.
The estimated percentage reductions in stormwater run-
off and pollutant loads are dependent on precipitation,
soil type, and land use type, which vary within each re-
gion as well as across regions. Generally, the percentage
runoff reduction is greatest when the alternative green-
field site is on forest land, lower for pasture, and even
lower for agricultural land. The percentage reduction also
varies positively with the ratio of greenfield acreage to
brownfield acreage used for the same type and amount of
development. Except for a few pollutants on agricultural
lands in some locations, all the brownfield development
scenarios yielded substantially lower pollutant loads.
While the study properties account for a relatively small
portion of total development acreage in these regions,
their development has been important to the communi-
ties in overcoming obstacles to redevelopment. Brown-
fields and infill redevelopment have significant positive
environmental impacts on water resources and air qual-
ity, particularly greenhouse gas emissions.
References
1. U.S. EPA 2010. Air and Water Quality Impacts of Brownfields Reuse, Office of Solid Waste and
Emergency Response, Office of Brownfields and Land Revitalization, EPA-560-F-10-232,
July 2010.
2. Allen, E., 2008."Clicking Toward Better Outcomes: Experience with INDEX 1994 to 2006,"
Eliot Allen, in Planning Support Systems for Cities and Regions, 139-166, Ed. Richard Brail,
Lincoln Institute of Land Policy, Cambridge, MA, 2008.
3. U.S. EPA 2001. EPA's Smart Growth Index in 20 Pilot Communities: Using GIS Sketch Model-
ing to Advance Smart Growth, EAP 231-R-03-001, August 2001.
4. Ewing et al., 2008. Growing Cooler: The Evidence on Urban Development and Climate
Change, Reid Ewing, Keith Bartholomew, Steve Winkelman, Jerry Walters, Don Chen,
Urban Land Institute, 2008.
5. Frank, L., 2005. Travel Behavior, Emissions & Land Use Correlation Analysis in the Central
Puget Sound, Lawrence Frank, James Chapman, Mark Bradley, T. Keith Lawton, prepared
for the Washington State Department of Transportation, June 2005.
-------
Comparison of Environmental Performance Measures for Five Regions
Environmental
Indicator
Air and Energy
Home bused vehitle miles traveled
Non-home bos*d vehicle miles traveled
Total vehicle miles traveled
Home based vehicle trips
Non-horns based vahicla trips
Total Vehicle trips per capita
Personal vehicle eneigy use
Residential structural energy use
GUI bun JioiiJe (CO^ emission;;
Air Pollutants (NQx, HC, & CO)
Land and Water
Land (.uTjuniplion
Stormwaler runoff
Nitrogen
Phosphorous
Suspended solids
Biological oxygen demand
Chemical oxygen demand
Oil and grease
Metals (lead, copper, zinc, cadmium,
chromium, nickel)
Unit;
rni/cupilu/day
mi/capita/da/
mi/capila/day
mi/capita/day
mi/capita/day
Trips/ai pita/day
MMBtu/ca pita/yr
MMBtu/ca pifa/'yr
Ibs/nesident/yr
Ibs/resident/yr
Aueo
Acne Feel/yr
Ibs/yr
Ibs/yr
Ibs/yr
Ibs/yr
Ibs/yr
Ibs/yr
Ibs/yr
Percent Difference for Brownfieldsas Compared to Conventional/Greenfields
(Conventional Less Brownfields Scenarios as a Percent of Conventional}
Seattle
Area
67%
37%
57%
11%
29%
19%
57%.
6%
57%
57%
5D to 75%
49 to 64%
57 to 71%
6 4 to 78%
65 to 79%
64 to 78%
65 to 79%
65 to 79%
60 to 74%
Twin Cities
Area
32%
34%
32%
'3%
19%
"6%
32%
t.A
32%
32%
50 io 75%
48 to 69%
75 to -17%
31 to -36%
26 to 83%
67 to 83%
71 to 84%
71 to 84%
65 to 78%
Emeryville
Area
53'i>
45%
49%
36?,;,
40%
: a"-.
49%
NA
49%
49%
50 Io 75%
27 to 45%
53 to 69%
77 to -11 3%
79 to -11%
54 to 7 7%
60 to 77%
60 to 77%
53 to 64%
Baltimore
Area
37%
53%
42%
NA
NA
NA
42%
NA
42%
42%
50 to 75%
48 to 70%
1 to 74%
79 to -13%
30 to 80%
65 to 78%
61 Io78%
£5lo80%
62 to 77%
Dallas -Fort
Worth Area
NA
NA
53%
KA
NA
24%
53%
NA
53%
53%
50 Io 75%
43 Io 52%
66 to -46%
77 to -55%
79 to -3%
59 to 78%
66 Io 79%
67 Io 80%
5 4 to 68%
Average
45%
43%
47%
20%
30%
2 4",
47%,
NA
47%
47%
50 Io 75%
43 to 60%
9 to 71%
-31 to 78%
21 to 80%
62 to 79%
65 to 79%
66 to 80%
59 to 72%
Full report available at: www.epa.gov/brownfields/sustain.htm
------- |