EPA/600/R-13/295 | December 2013 | www.epa.gov/ged
United States
Environmental Protection
Agency
Neighborhood Scale Quantification
of Ecosystem Goods and Services
» ;
Office of Research and Development
Gulf Ecology Division
-------
EPA/600/R-13/295
December 2013
Neighborhood scale quantification of ecosystem goods and
services.
Marc Russell, Aarin league, Federico Alvarez, Darrin Dantin, Michael Osland, Jim Harvey,
Janet Nestlerode, John Rogers, Laura Jackson, Drew Pilant, Fred Genthner, Michael Lewis,
Amanda Spivak, Matthew Harwell, and Anne Neale
This report was prepared by the U. S. Environmental Protection Agency, Office of Research and
Development, National Health and Environmental Effects Research Laboratory, Gulf Ecology
Division. This document has been reviewed in accordance with U.S. Environmental Protection
Agency policy and approved for publication. Mention of trade names, products, or services does
not convey, and should not be interpreted as conveying, official EPA approval, endorsement or
recommendation.
This report should be cited as:
Russell, M., A. Teague, F. Alvarez, D. Dantin, M. Osland, J. Harvey, J. Nestlerode, J. Rogers, L.
Jackson, D. Pilant, F. Genthner, M. Lewis, A. Spivak, M. Harwell, and A. Neale. 2013.
Neighborhood scale quantification of ecosystem goods and services. U.S. Environmental
Protection Agency, Office of Research and Development, Gulf Ecology Division, Gulf Breeze,
Florida. EPA/600/R-13/295. November 2013
-------
TABLE OF CONTENTS
Introduction 1
Methods and Results 2
Section 1. Ecosystem Services 6
Air Pollution Removal 6
Shading 9
Carbon Sequestration 14
Nitrogen Removal 16
Section 2. Ecosystem Goods 20
Walkability and Access to Green space 20
Aesthetic Value of Residential Trees 26
Water Feature Viewscapes 27
Section 3. Neighborhood Scale Value of Ecosystem Goods and Services 30
Biodiversity and Ecosystem Goods and Services Resilience 32
People as Part of the Ecosystem 34
Discussion 36
Conclusions 40
References 43
-------
INTRODUCTION
Humans both alter and benefit from ecosystems in many ways. In many places human presence
dominates the landscape, especially in urban and agricultural settings. Every place on Earth is
directly or indirectly affected in some way by humans. As a result, many now feel that humans
should be included in the definition of ecosystems, while others still think of human activity as
an extrinsic source of stress on ecosystems. Human actions often feed back through multiple and
often complex interacting pathways to change the ecosystem goods and services that benefit
humans. This feedback becomes part of natural, highly complex, cyclic processes. The concept
of ecosystem goods and services (EGS) provides a view of ecosystems that is human-centric; and
thus, makes it easier to consider human stress on ecosystems as intrinsic. Ecosystem goods and
services close the feedback loops that link human actions to human costs and benefits from
ecosystems. Mapping of ecosystem goods and services is useful for making this information
available to the general public, their representatives, and scientists. Here, we present mapped
inventories of ecosystem goods and services production at a neighborhood scale within the
Tampa Bay, FL region. Comparisons of the inventory between two alternative neighborhood
designs are presented as an example of how one might apply EGS concepts to land use decisions
at this scale.
Ecosystem goods and services for an area of land/seascape are dependent on ecosystem type, the
presence of human made complementary resources, such as a means of transportation or the
presence of residential buildings, and the impact of human and natural stressors on that area.
Changes in EGS can be estimated using a strictly supply side view or can take into account
human demand functions. The supply side method can estimate changes in ecological structure
and function due to replacement of an ecosystem by another ecosystem type at the landscape
scale; e.g., forest change to agricultural land causes a net change in the landscape's ecological
functions that then may change the supply of EGS and derived benefits. These EGS supplies
only become realized and valuable when one accounts for the connections between source areas
and human beneficiaries. These landscape replacement related changes in EGS supply tend to
track linearly with the amount of ecosystem replacement; the rate of change being wholly
dependent on the specific types of ecosystem replacement and not on their interaction with
beneficiaries. A more complete and meaningful assessment needs to account for changes in the
spatial arrangement of complementary factors, such as location of human residences, water flow
paths, and transportation networks that are paramount to turning these potential EGS into
realized EGS with actual benefits to identifiable human beneficiaries.
Ecosystem goods and services are those ecological structures and functions that humans can
directly relate to their state of well-being. Ecosystem goods and services include, but are not
limited to, a sufficient fresh water supply, fertile lands to produce agricultural products, shading,
-------
air and water of sufficient quality for designated uses, flood water retention, and places to
recreate. The US Environmental Protection Agency (USEPA) Office of Research and
Development's Tampa Bay Ecosystem Services Demonstration Project (TBESDP) modeling
efforts organized existing literature values for biophysical attributes and processes related to
EGS. The goal was to develop a database for informing mapped-based EGS assessments for
current and future land cover/use scenarios at multiple scales. This report serves as a
demonstration of applying an EGS assessment approach at the large neighborhood scale (-1,000
acres of residential parcels plus common areas).
Land cover/land use replacement based assessment of EGS has to be linked to specific spatially
explicit landscape units that are monitored or modeled through time. The National Land Cover
Dataset (NLCD) is a good national scale example of the type of required geospatial data
available for use in EGS production assessments but it limits temporal assessments due to its
decadal update schedule and spatial assessments due to its restricted number of resolved land
cover types (Homer et al. 2007). A dataset that alleviates these two problems, at least for
assessments in Florida, is the Florida Land Use/Cover Classification System (FLUCCS)
(SWFWMD 2012) dataset that is updated much more often and classifies almost twice as many
specific land use/cover types as the NLCD (Figure 1). Combinations of the FLUCCS dataset
with supplementary information housed in the NLCD's percent canopy cover database, county
residential parcel boundaries, state transportation networks, and digital elevation models allowed
us to identify where most of the ecosystems responsible for the production of EGS are located on
the landscape at a neighborhood scale.
METHODS AND RESULTS
We estimated EGS production for two alternative neighborhood scale development scenarios.
Scenario A was based on the FishHawk Ranch development in the Alafia River basin of Tampa
Bay (Figure 2). Scenario A represents an example of a relatively extensive "green" development
that occurred over a period of almost 20 years. Ranched scrub-shrub land was converted to areas
of light, medium, and dense residential housing with associated roads, schools, parks, and other
infrastructure. Scenario B uses the exact spatial boundary as Scenario A, but relocated over an
area in East Tampa that we use as a proxy for a traditional blocked neighborhood layout. Our
comparison between scenarios is only meant to illustrate how one can complete a neighborhood
scale assessment of EGS differences and should in no way be considered as an endorsement of
either development approach by USEPA. These types of comparisons could be considered
alongside other benefit cost analysis factors during neighborhood planning.
-------
Agriculture
Grassland
Forested
I Unforested Wetland
Water
Extractive
Development
0 5 10
20
i Kilometers
N
A
Figure 1. Simplified Florida Land Use/Cover Classification System map for the Tampa Bay
region, 2006.
-------
Scenario A
Scenario B
Figure 2. Location and aerial photograph of Scenario A and B developments.
We applied ecosystem service process rates and ecosystem goods stock estimates obtained from
an extensive literature review to biophysical features on the landscape using various spatial
datasets. Biophysical attributes, considered here as direct measurements or secondary indicators
of final ecosystem goods and services, include carbon sequestration for mitigation of climate
change, nitrogen removal for maintaining downstream water quality, atmospheric pollution
removal for maintaining higher quality air, shading for maintaining lower heating and cooling
energy costs, and water viewscapes and presence of accessible green spaces for maintaining
physical and mental well-being. It should be noted that while ecosystem goods and services seem
to overlap, the majority of ecosystem services (ES) are valued on a per year basis since they
continue to function or produce as long as they are not significantly disturbed while ecosystem
goods (EG) are valued as a stock at a specific point in time and are generally a result of
-------
ecological processes in the past. For ES that generate regional-scale benefits (e.g., nitrogen
removal and carbon sequestration) we present potential value, while those that are more locally
enjoyed (shading and air pollution removal) are weighted by estimated use and are closer to a
realized value of benefits. Realized valuation of ES most often requires a further assessment of
how they deliver benefits to specific beneficiaries through time so the calculation of realized
value for regional and global scale ES requires connecting sources of production to beneficiaries
in spatial and temporal scales that are beyond the scope of this report. Ecosystem goods do not
generate further value unless human demand increases, and or stocks are increased by continued
ecological production.
Final ecosystem goods and services (FEGS) are biophysical features that human beneficiaries
can directly relate to and would, theoretically, be willing to pay to maintain even in the absence
of any other biophysical change (Johnston and Russell 2011; Landers and Nahlik 2013).
Biophysical attributes, and the processes producing them, relevant for assessing final ecosystem
goods and services at the neighborhood scale are summarized in Table 1. These FEGS are then
translated into derived human benefits using various valuation methods (Table 1). In most cases
the beneficiaries are local residents, however for the larger-scale processes of nitrogen removal
and carbon sequestration the beneficiary groups, such as downstream water users or those
affected by climate change, are associated with both local, but also watershed and global scale
boundaries. We also present the spatial arrangement of biodiversity at the neighborhood scale
since it is difficult to fully translate this biophysical measure into an EGS that directly benefits
human well-being using a common currency such as US dollars.
Table 1. Summary of neighborhood scale metrics used to estimate ecosystem goods and services
and valuation method used to estimate global, regional, and locally derived benefits.
Metric
Tree canopy coverage
Tree canopy coverage
(South side of residential
property)
Rate of carbon
sequestration
Rate of denitrification
Walking distance to open
green spaces, trails, and
parks
Number of viewable
mature trees
Number of viewable water
features
Ecosystem Service
(FEGS)
Atmospheric pollution removal
(Clean air)
Shading (Shade)
Atmospheric regulation
(Stabilized climate)
Nutrient removal (Clean water)
(Accessible green spaces)
(Viewable, aesthetically
pleasing trees)
(Viewable water)
Benefit
Increased respiratory
health
Decreased energy use
More predictable
climatic patterns
Water of sufficient
quality is available to
meet designated uses
Increased opportunity
to recreate
Increased mental
health and well-being
Increased mental
health and well-being
Valuation Method
Avoided medical costs
Avoided energy costs
Avoided social costs
Replacement costs
Hedonic pricing
Hedonic pricing
Hedonic pricing
-------
Biophysical and value maps were produced for two alternative development scenarios, one based
on the 2009 FishHawk Ranch development and the other using an identically shaped and sized
area from East Tampa as an example of what this area could have been developed into if a more
traditional block neighborhood layout had been used.
The following sections present spatial estimates of biophysical attributes and resulting benefit
valuation estimates for EGS in 2009 for Scenario A (Figure 3) and an alternative development
pattern Scenario B (Figure 4) taken from East Tampa and representative of a more traditional
blocked development pattern. These two neighborhoods were chosen to represent maximal
differences between traditional and "green" oriented development patterns. Comparisons
between scenario EGS production and values serve to illuminate the tradeoffs society can
consider as areas are developed or redeveloped to meet growing housing needs. Differences in
EGS can be positive or negative depending on how the landscape is modified during
development and how humans interact with remaining or newly constructed neighborhood
biophysical features.
SECTION 1. ECOSYSTEM SERVICES
AIR POLLUTION REMOVAL
Air pollutants are removed when tree canopy intercepts pollutants in the atmosphere. Air
pollution removal service through time yields cleaner air, which is important for maintaining
human respiratory health. The rates of pollutant removal are a function of the downward flux of
the pollutant and the resistance of the canopy vegetation (Nowak et al. 2006). The canopy
coverage of our scenarios were determined by spectral analysis of remotely sensed images,
which identified pixels that reflect light in a pattern indicative of tree canopy vegetation. The
1 m2 resolution coverage of tree canopy was combined with established pollutant attenuation
rates for carbon monoxide, ozone, particles, sulfur dioxide, and nitrous oxide to calculate the
total air pollutant removal (Nowak et al. 2006).
-------
Land Use ^M Unforested Wetland
Q^l Developed ^H Forested ^^^^^ — —
Agriculture \///A Wetland 0 1
HI Grassland || Water
N
1 Km A
2 A
Figure 3. Aerial photo (a) and simplified Florida Land Use Cover Classification System (b) map
for Scenario A in 2009.
-------
(a)
Land Use ^H Unforested Wetlands
j Developed ^H Forested
I | Agriculture ^H Water
N
"A
Figure 4. Aerial photo (a) and simplified Florida Land Use Cover Classification System (b) map
for Scenario B in 2009.
-------
The rate of air pollutants removal (Figure 5) is the sum of the various pollutants attenuated by
tree canopy. The rate of downward pollutant flux for each of the species of interest was
calculated by iTree model algorithms (Nowak et al. 2008), and the relative air pollutant costs to
human health was used to translate the various pollutant attenuation rates to decreases in human
health impact costs (Murray et al. 1994). The estimated value for attenuation of the selected
pollutants in 1994 US dollars was $959/ton carbon monoxide, $6,752/ton ozone, $l,653/ton
sulfur dioxide, $4,508/ton particulate matter (PM10), and $6,752/ton nitrogen dioxide (Murray
et al. 1994). We applied this 1994 estimate to our 2009 scenarios without year-specific
corrections for inflation.
The total air pollutant removal was calculated as:
Air Pollution Removed = £j Valuei * FluxRatei * %Can * CellSize
where /' represents the individual pollutants, Value is the decrease in costs associated with a
decrease in pollutant species, FluxRate is the removal rate of each pollutant by specific species,
%Can is the percent canopy cover in each section of the landscape, and CellSize is the area of
each section in square meters. One meter resolution percent canopy coverage maps were used to
calculate the total air pollution removed in 1994 US dollars ($) per year using the raster
calculator function in ArcGIS 9.3 (Figure 5). Our Scenario A neighborhood was estimated to
have an air pollution removal service of 7,997 kg of pollutants per year while Scenario B had
8,849 kg of pollutants per year. This service is estimated to be worth $0.39 and $0.43 million US
per year for Scenario A and B, respectively.
SHADING
The shading service enjoyed by each residential parcel was calculated from the percent canopy
cover situated to shade buildings (Figure 6). The production of shade is considered an ecosystem
service that continues to provide humans with lower energy needs for cooling. To quantify the
amount of shade provided by trees, the center of each parcel was determined and a 25 meter
radius semi-circle was drawn on the southern side, representing the area most relevant for
shading the south facing section of buildings in the Northern Hemisphere. Then, using a
remotely sensed image, the number of 1 m2 canopy pixels was determined in this shade-
providing semi-circle. The number of pixels was then translated into large shade tree equivalents
by dividing by the canopy area of a representative large tree within this community (80 m2). It is
important to note that shading service is only provided by the ecosystem when houses or other
structures are present, therefore, before development of this area no shading service could be
provided even though plenty of shade may have been present.
-------
Total: 7997 kg/yr
Air Pollutants
Removed
g/m2/yr
Value of Air
Pollutants Removed
$/m2/yr
Figure 5. Development scenarios A and B for air pollution attenuation (Al and Bl)
corresponding Ecosystem Service Value (A2 and B2).
10
-------
I I Parcel
South Shading of the Parcel
25 meters zone from center of parcel
Center of Parcels Tree Canopy ^
- Meters
0 5 10
A
| Tree Canopy Cells ShadingZone
Figure 6. Example of shade tree coverage estimation from satellite imagery (top panel)
summarized for the southern side of a parcel (bottom panel).
The amount of canopy in a semi-circle from southwest to southeast and corresponding cost
savings from shading decreasing energy use were estimated (Figure 7). Shading by trees on the
southeast and southwest side of a residential house is estimated to reduce energy use by upwards
of 350 kWh per year per 80 m2 of canopy (Simpson and McPherson 1996). Much of this savings
takes place in hotter summer months when energy reductions can get as high as 80 kWh per
month per 80 m2 of south side tree canopy cover (Donovan and Butry 2009; Huang et al. 1987).
These values, however, were calculated from residential parcels using almost three times as
many kWh of energy as the average Tampa Bay region resident's use. Thus, energy savings for
11
-------
Tampa Bay local residents would be approximately one third of the published value or
116.7 kWh a year per 80 m2 of south side tree canopy cover. Tampa Electric's 2012 electricity
rate was estimated at 9.718 cents per kWh based on an average residential customer using 1,200
kWh per month on a two-tiered fuel and energy cost rate from Tampa Electric (2012). Cost
savings per 80 m2 of tree canopy, assuming a decrease of 26.3 kWh per summer month
(Donovan and Butry 2009; Huang et al. 1987) or 116.7 kWh per year (Simpson and McPherson
1996) for the average resident in the Tampa Bay region, would equate to close to $3 per month
in summer months and $12 a year. We applied this 2012 estimate to our 2009 scenarios without
year specific corrections for inflation. Cost savings will not scale linearly for residents using
more than the 1,200 kWh average since per kWh energy costs are higher after the first 1,000
kWh of use. Tampa Electric customers using 1,200 kWh per month with the equivalent of three
large 80 m2 trees to the west-southwest of their residence would be estimated to save up to 80
kWh per month during summer months from afternoon and evening shading (Simpson and
McPherson 1996). This is equivalent to 7% of summertime energy costs. Our Scenario A
neighborhood was estimated to have a total shading service of 14,724 shade trees while Scenario
B had 16,843 shade trees. This service is estimated to be worth $0.18 and $0.20 million US per
year for Scenario A and B, respectively.
12
-------
Total: 14,724 Shade Trees
A2
Total: $0.18 Million/yr
Total: $0.20 Million/yr
Value of
Shade Trees
$/m2/yr
0.01 -0.03
0.04 -0.07
0.08 -0.11
0.12 -0.15
0.16-0.20
Kilometers
Figure 7. Development scenarios A and B for number of shade trees per parcel (Al and Bl) and
corresponding Ecosystem Service value (A2 and B2).
13
-------
CARBON SEQUESTRATION
Carbon sequestration was estimated by assigning the average rate published in peer-reviewed
literature to specific land use types (Table 2). Literature values were used from studies conducted
in similar climates and landscapes. However, there is a scarcity of literature reporting carbon
sequestration rates for developed areas, including residential, institutional, commercial,
transportation, utility, and communication areas. Therefore, the rate of carbon sequestration for
the urban land use classes was estimated in the canopy and lawn areas for these land use areas by
the following equation:
Urban Carbon Sequestration
= ((1 — %/mp) * LawnRate + %Canopy * UrbanTreeRate)
where %Imp is the average percent impervious surface and %Canopy is the average percent
canopy coverage for the land use category, LawnRate is the average published rate of carbon
sequestration for lawns in g C m"2 yr"1 (Bandaranayake et al. 2003; Gebhart et al. 1994; Qian and
Follett 2002), and UrbanTreeRate is the published rate for typical Florida urban trees in g C m"2
yr"1 (Nowak and Greenfield 2009). Each scenario was reclassified into the carbon sequestration
flux rates in ArcGIS 9.3, and then multiplied by the grid cell area using the spatial analyst
extension's raster calculator. The result was the rate of carbon sequestration in grams carbon
removed per year per cell (Figure 8). Thus, rates represent averages from several different
studies with various degrees of accuracy. It should be noted that the carbon sequestration rates
are carbon incorporated into biomass or net primary production and do not reflect long-term
carbon storage or burial rates as this carbon becomes incorporated into soil or wood products.
The value of carbon sequestration was estimated using the social cost of carbon. The social cost
of carbon is an estimate of the monetized damages associated with an incremental increase in
carbon emissions for a given year. It is intended to include, but is not limited to, changes in net
agricultural productivity, human health, property damages from increased flood risk, and the
value of ES. The dollar value of carbon reductions in the form of the greenhouse gas carbon
dioxide was estimated as $20 per ton ($0.01 per Ib) of carbon dioxide in 2010 (US Government
2010). We applied this 2010 estimate to our 2009 scenarios without year-specific corrections for
inflation. Even without year-specific cost adjustments, the $20 per ton of carbon dioxide
represents a conservative estimate that has been recalculated as 12 times larger using a less
severe discount rate more appropriate for intergenerational cost-benefit analysis (Johnson and
Hope 2012). Carbon sequestration rates were multiplied by the social cost of carbon estimate to
arrive at the total value of this ecosystem service that benefits humans by moderating climate
change (Figure 8). Our Scenario A neighborhood was estimated to have a carbon sequestration
service of 3,807 million kg C per year while Scenario B had 1,115 million kg C per year. This
14
-------
service is estimated to be worth $0.76 and $0.22 million US per year for Scenario A and B,
respectively.
Table 2. Land Use Specific Carbon Sequestration.
Description
Residential Low Density
Residential Med Density
Residential High Density
Commercial And Services
Institutional
Recreational
Open Land
Cropland And Pastureland
Other Open Lands
Herbaceous
Shrub And Brushland
Upland Coniferous Forest
Pine Flatwoods
Hardwood Conifer Mixed
Streams And Waterways
Lakes
Reservoirs
Stream And Lake Swamps
Freshwater Marshes
Wet Prairies
Emergent Aquatic Vegetation
Intermittent Ponds
Transportation
Communications
Utilities
FLUCCS
1100
1200
1300
1400
1700
1800
1900
2100
2600
3100
3200
4100
4110
4340
5100
5200
5300
6150
6410
6430
6440
6530
8100
8200
8300
Carbon
Fixed into
Biomass
Map Value
[g C/m2/yr]
148
139
91
57
73
128
133
423
673
743
945
698
698
660
180
397
368
808
618
142
142
142
96
106
133
Reference
See Methods
See Methods
See Methods
See Methods
See Methods
See Methods
See Methods
(Ajtayetal. 1979)
(Ajtay et al. 1979; Milesi et al. 2005)
(Ajtayetal. 1979)
(Ajtayetal. 1979)
(Ajtay et al. 1979; Kroeger 2008)
(Ajtay et al. 1979; Clark et al. 1999)
(Ajtay et al. 1979; Kroeger 2008)
(Ajtayetal. 1979)
(Carpenter et al. 1998; Carrick et al. 1993)
(Carpenter et al. 1998; Carrick et al. 1993)
(Lugoetal. 1988)
(Smith and De Laune 1983)
(Kroeger 2008)
(Kroeger 2008)
(Kroeger 2008)
See Methods
See Methods
See Methods
15
-------
Total: 3,807 million kg C/yr
A2
Total: S0.76 Million fyr
Total: 1,115 million kg C/yr
Carbon Sequestration
g C/nV/yr
508 - 864
865 -1620
1,621 - 3,808
3,809 - 6,278
6,279 - 8,505
Total: $0.22 Million /yr
Value of Carbon
Sequestration
$/mVyr
0.01
0.01-0.02
0.02 - 0.03
0.03 - 0.08
0.08-0.17
Figure 8. Development scenarios A and B for carbon sequestration (Aland Bl) and
corresponding Ecosystem Service value (A2 and B2).
NITROGEN REMOVAL
Excess reactive nitrogen in water results in eutrophication and ground water contamination
(Vitousek et al. 1997). Nitrogen removal helps maintain downstream waters at a sufficient
quality for the designated use of the water body. Appreciable amounts of nitrogen can be
removed from the landscape through enzymatic denitrification. Scientific literature has reported
estimates of this landscape process for land use types, based on case studies in settings in the
Florida area or in similar landscapes. Similar to carbon, the rate of denitrification was estimated
using literature rates assigned to undeveloped areas and calculated rates for urbanized areas
(Table 3). As denitrification occurs in the soil, the mass of nitrogen denitrified in the previous
area was calculated by:
16
-------
Urban Denitrification = (1 — %/mp) * DenLawn * CellSize
where DenLawn is the denitrification rate published for urban lawns (Raciti et al. 2011). The
%Imp, percent impervious surface, was derived from aim2 resolution land cover map. Similar
to the method used for estimating carbon sequestration, each land use area was reclassified using
the denitrification flux rates (Table 3) multiplied by area. The result was the rate of nitrogen
removed via denitrification in grams nitrogen removed per year (Figure 9).
Costs for removing a pound of nitrogen in water coming from various sources range from less
than $10 to as high as $855. Costs increase as the nitrogen becomes harder to route towards
treatment areas and as simpler, more cost efficient mechanisms for removing nitrogen need to be
replaced by more centralized advanced waste water treatment facilities. Compton et al. (2011)
reviewed the cost of removing nitrogen from a wide range of sources and concluded that costs
ranged from $1.22 - $43.54 per pound of nitrogen ($2.71 - $96 kg"1). Abatement costs of
reducing nitrogen from point sources are estimated as $8.16 per pound ($18 kg"1) of nitrogen
(Birch et al. 2011). We use $8.16 per pound as our conservative estimate of what it would cost to
replace the ecosystem service of removing nitrogen for the purpose of maintaining usable water
based on using traditional waste water treatment to remove nitrogen from upstream point sources
(Figure 9). We applied this 2011 estimate to our 2009 scenarios without year-specific corrections
for inflation. Our Scenario A neighborhood was estimated to have a carbon sequestration service
of 5.143 million kg N per year while Scenario B had 1.321 million kg N per year. This service is
estimated to be worth $0.93 and $0.24 million US per year for Scenario A and B, respectively.
Several lifecycle estimates, however, including upgrading and maintaining existing or building
additional advanced wastewater treatment facilities and drainage structures to remove nitrogen,
put the cost as high as $855 per pound ($388 kg"1) of nitrogen removed (Roeder 2007). This
higher ecosystem replacement value may be more appropriate than our more conservative
number if one wants to illustrate the potential future value of bay habitats under a scenario of
increasing demand for nitrogen removal.
17
-------
Table 3. Land Use Specific Denitrification Rates.
Description
Residential Low Density
Residential Med Density
Residential High Density
Commercial And Services
Institutional
Recreational
Open Land
Cropland And Pastureland
Other Open Lands
Herbaceous
Shrub And Brushland
Upland Coniferous Forest
Pine Flatwoods
Hardwood Conifer Mixed
Streams And Waterways
Lakes
Reservoirs
Stream And Lake Swamps
Freshwater Marshes
Wet Prairies
Emergent Aquatic Vegetation
Intermittent Ponds
Transportation
Communications
Utilities
1100
1200
1300
1400
1700
1800
1900
2100
2600
3100
3200
4100
4110
4340
5100
5200
5300
6150
6410
6430
6440
6530
8100
8200
8300
[g N/m2/yr]
1.3
1.13
0.85
0.62
0.87
1.4
1.4
0.72
0.82
0.06
0.06
0.12
0.12
0.19
20.73
12.29
7.5
25.5
28.26
25.48
26.22
17.44
1.2
1.16
1.4
Reference
See Methods
See Methods
See Methods
See Methods
See Methods
See Methods
See Methods
(Barton et al. 1999; Espinoza 1997; Robertson et
al. 1987; Tsai 1989)
(Barton et al. 1999; Tsai 1989)
(Tsai 1989)
(Tsai 1989)
(Barton et al. 1999; Robertson et al. 1987)
(Barton et al. 1999; Robertson et al. 1987)
(Barton etal. 1999)
(Pina-Ochoa and Alvarez-Cobelas 2006;
Seitzinger et al. 2006)
(James et al. 201 1; Pina-Ochoa and Alvarez-
Cobelas 2006; Seitzinger 1988)
(Brenner et al. 2001; Seitzinger 1988)
(Martin and Reddy 1997; Pinay et al. 2007;
Seitzinger 1994; Walbridge and Lockaby 1994)
(Ensign et al. 2008; Martin and Reddy 1997;
Pinay et al. 2007; Reddy et al. 1989; Seitzinger
1994)
(Ensign et al. 2008; Martin and Reddy 1997;
Pinay et al. 2007)
(Ensign et al. 2008; Martin and Reddy 1997)
(Ensign et al. 2008)
See Methods
See Methods
See Methods
18
-------
Total: 5.14 million kg N/yr
Total: 50.93 Million /yr
Total: 132 million kg N/yr
Denitrification Rate
N g/m'/yr
0.06-0.19
0.20-0.87
0.88 -1.40
1.41 - 7.50
7.51-28.26
Total: $0.24 Million /yr
Figure 9. Development scenarios A and B for denitrification rates (Al and Bl) and
corresponding Ecosystem Service value (A2 and B2).
Value of
Denitrification Rate
$/mVyr
0.01
0.01-0.02
0.02-0.03
0.03-0.13
0.13-0.51
19
-------
SECTION 2. ECOSYSTEM GOODS
There are numerous physical things that ecosystems produce that are of benefit to humans. Some
of these ecosystem goods are structural components of the environment and are direct inputs into
the market system, such as edible fish, wild berries and wild game, as well as wood for timber
production. Other ecosystem structural components are harder to break down into discrete,
tangible units, and so are often not marketable as such. Some ecosystem goods are mosaics of
ecosystem attribute that, when combined, generate something greater than the sum of their parts.
Such mosaics include areas of green space that provide opportunities for recreation, wildlife and
other green/blue landscapes providing pleasant views, and even biologically diverse areas that
may provide greater stability in the production of other ecosystem goods. Most of the time,
benefits from ecosystem goods to humans are only manifested when humans physically or
emotionally interact with a tangible component of the ecosystem. This interaction usually takes
place on local scales such as having greenspace within a comfortable walking distance or being
able to look out your window and see a tree or lake. This required close proximity can be found
within a neighborhood and should be accounted for at that scale. Interaction with ecosystem
goods at the neighborhood scale is often dependent on how individual parcels are arranged.
Thus, most ecosystem good's value is wholly dependent on demand for, and current levels of use
of, that good. Lack of demand or inaccessibility equates to zero ecosystem good value in most
cases. Value of ecosystem goods is estimated at a given time and place like a stock, unlike
ecosystem services, which have rates of value production and are, thus, more dynamic. The
temporal scales and associated valuation approach is really what separates the concept of
ecosystem goods from ecosystem services. Most ecosystem services are valued using estimates
of what it would cost to replace beneficial biophysical functions using conventional means, while
ecosystem goods are typically valued using willingness to pay valuation approaches, with the
value being interpreted as what individuals are willing to pay for a set quantity or condition of
something at a specified point in time. Here we quantify the value of several locally important
ecosystem goods at the neighborhood scale.
WALKABILITY AND ACCESS TO GREEN SPACE
The availability of green spaces for recreation is a valuable attribute for neighborhood residents
as is their access to commercial destinations. Walk Score (http://www.Walk Score.com/) helps
people find a walkable place to live (Sidebar 1). Walk Score is a number between 0 and 100 that
indicates the walkability of any location (Figure 10).
20
-------
Walk Score Description
90 -100 Walker's Paradise
Daily errands do not require a car
70 - 89 Very Walkable
Most errands can be accomplished on foot.
50 - 69 Somewhat Walkable
Some amenities within walking distance
25 - 49 Car-Dependent
A few amenities within walking distance
0 - 24 Car-Dependent
Almost all errands require a car
Figure 10. Description of Walk Score
We used the Walk Score algorithm to quantify the
walkability of each neighborhood in Scenario A and B
(http://www.walkscore.com/methodology.shtml). We found
that Walk Score does not currently include many of the
smaller parks or green space access trails that are evident in
Scenario A so we digitized the neighborhood, roads,
sidewalks, and trails and recalculated the distance to park
metric. For Scenario B we present walk scores as sections,
without including small parks not evident from aerial
photography, since there are so many more roads with no
clear grouping.
Sidebar 1:
Walk Score (www.Walk
Score.com) uses Google maps
to compute the distance
between residential addresses
and nearby destinations. The
Walk Score algorithm looks at
destinations in nine categories
and awards points for each
destination that is between
one-quarter mile and one and a
half miles of the subject
residential property:
• grocery stores
• restaurants
• shopping
• coffee shops
• schools
• parks
• banks
• bookstores
• entertainment
(http://www.Walk Score.com/
Accessed 5/2012)
Walk Scores for Scenario A are reflective of how easy or
difficult it is to walk to a suite of amenities including
shopping, entertainment, and parks. Fligher scores are more
preferable for walkers. Distance to parks in Scenario A has a
different pattern than Walk Score since the Walk Score web site currently only incorporates
larger publicly available park location datasets, while our analysis included our own hand
digitization of nature trails and small neighborhood parks (Figure 11).
Walk Scores are generally higher in Scenario B's community configuration (Figure 12) than in
Scenario A, but the distances to parks are longer because of the lack of easily identifiable green
trails and pocket parks. It is, however, somewhat difficult to identify small green spaces, that
serve the same role as formal pocket parks. Ballparks or other open spaces associated with
schools were not included as publically assessable green spaces in this analysis since they are not
considered as being conserved in at least a semi-natural state, many being part of school grounds
that may not be open to the public for recreation. A gridded neighborhood structure does not
necessarily preclude inclusion of small, easily accessible green spaces and the choice of a
21
-------
different location for Scenario B would have most likely influenced our distance to park scores.
Higher overall Walk Scores in Scenario B, however, are reflective of easier access to other
amenities such as shopping.
Kilometers
Distance to Park [Miles]
0.41 -0.50
0.31 -0.40
— 0.26 - 0.30
023-0.25
0.17-0.22
0.13-0.16
0.06-0 12
0.04-0.05
Figure 11. Walk Scores (upper panel) and distance to park (lower panel) for Scenario A streets.
22
-------
Kilometers
Distance to Park [Miles]
0.61 -1.00
0.51 -0.60
0.41 -0.50
0.31 -0.40
0.28-0.30
0.22-0.27
0.21
0.20
Figure 12. Walk Scores (upper panel) and distance to park (lower panel) for Scenario B sections.
23
-------
The walkability of a neighborhood is reflected in the value of property. In a study of 15 separate
housing markets it was determined that for every one point increase in Walk Score the value of
property increases between $700 and $3,000 depending on particular housing markets and local
preferences (Cortright 2009). In the Jacksonville, FL market, a one point increase in Walk Score
equated to an increase of $809 per property. Cortright estimated an average home value of
$179,873 with an average size of 1,660 ft2 in Jacksonville in 2007. Jacksonville's median Walk
Score was 36 with a 25th and 75th percentile of 20 and 51. The estimated value increase from an
identical home with a median Walk Score of 36 versus one in the 75th percentile with a Walk
Score of 51 was $12,951. These values place Jacksonville on the lower end of market price and
in the middle range of Walk Scores as compared to other assessed cities (Cortright 2009). The
use of marginal value increase estimates for each increase in Walk Score, based on the
Jacksonville market, means our estimates of our scenario neighborhood's Walk Score value
should be thought of as conservative since our study areas have generally higher home values but
with a similar Walk Score range as the Jacksonville market.
Distance and ease of travel on foot to parks is one of nine metrics used to calculate the Walk
Score Index. Distance to parks and or green space is weighted as having a 1/15 influence on
Walk Score along with several other amenities, but distance to grocery stores and restaurants has
a 3/15 weight, and distance to shopping has a 2/15 weight. Each single unit of increase in Walk
Score for a property is roughly equivalent to saying that property has $54 ($809/15) of increased
value due to greater access to green space. We applied this 2009 estimate to our 2009 scenarios.
We determined the per-unit distance to greenspace relationship to Walk Score value independent
of the other amenities by measuring the distance to parks noted by Walk Score of 49 different
streets in our Scenario A neighborhood.
The graph below illustrates that for every tenth of a mile increase in distance to a park we
estimate there is a respective decrease in Walk Score of approximately one unit (Figure 13). This
relationship was used as a calibration of the Walk Score's distance decay function for access to
greenspace in our neighborhood analyses. Using the conservative value estimates from
Jacksonville, every tenth of a mile (about one city block) to green space beyond the 0.25 mile
minimum walking distance is thus equivalent to a $54 decrease in a property's assessed real
estate value. The Walk Score calculation assumes that being closer than 0.25 miles to a park or
green space does not add further value to a home than if that home was at 0.25 miles away. Few,
if any, residential areas were found to be more than 1.5 miles away from green space in either
scenario thus we did not have to limit value generation to just parcels within this maximum easy
walking distance.
24
-------
= -10.543x + 35.051
R2 = 03346
0.25 0.45 0.65 0.85 1.05 1.25
Walking Distance to Greenspace (miles)
1.45
Figure 13. Distance to green space's relationship to Walk Score for Scenario A streets.
The sums of estimated increase in residential parcel value ($54 for each tenth of a mile closer
than 1.5 miles) generated by nearby green spaces per street or neighborhood area were
apportioned to each street or neighborhood's closest accessible green space access point and then
divided by that corresponding green space's area (Figure 14). Green spaces included both natural
areas reserved for walking trails and parks, but not ball fields. Our Scenario A neighborhood was
estimated to have a level of access to the ecosystem good of green space worth $7.74 million US
while Scenario B had $0.84 million US. Much of the value in Scenario A was generated by
parcels within walking distance of a trail entrance or small pocket park that were not as easily
identified in Scenario B.
25
-------
Total: S7.74 mil lion
Figure 14. Development Scenarios A and B for distance weighted accessible green space value
derived from residential parcels within walking distance.
AESTHETIC VALUE OF RESIDENTIAL TREES
There are several benefits from ecosystem goods that are best summarized as a marginal increase
in value, such as the fact that humans have been shown to be willing to pay an additional 1% in
property costs per large tree present within view when they are assessing a property (Anderson
and Cordell 1988). We calculated the area of canopy cover in every parcel and divided that by
80 m2, the area of a typical mature tree in our Scenario A neighborhood (Figure 15), to estimate
the number of viewable large trees or smaller trees with equivalent canopy cover per parcel. For
Scenario B we divided each parcel's canopy cover by 120 m2 to correct for the older age of this
neighborhood, and consequently larger more mature tree canopies. Our Scenario A
neighborhood parcels were estimated to have a total tree canopy cover equivalent to 10,000
26
-------
mature trees while Scenario B had canopy cover equivalent to 31,500 mature trees. This
ecosystem good is estimated to be worth $22.63 and $3.79 million US for Scenario A and B,
respectively. The large difference here is mainly driven by lower property values for parcels in
Scenario B that are almost 3 times lower on average than in Scenario A. The correction factor we
chose for the older tree canopy structure in Scenario B could also be too large (Figure 21).
WATER FEATURE VIEWSCAPES
Residences in Scenario A that enjoy water views were identified by hand selecting parcels that
have an unobstructed line-of-sight to a lake, reservoir, or pond water feature, taking into account
views being blocked by other residences and areas of vegetation (Figure 16). The valuation
approach for water features is similar in theory to that used for estimating the value of benefits
derived from viewable mature trees. Landscape psychologists Kaplan and Kaplan (1989) state
that "Water is a highly prized element in the landscape" and a large-scale evaluation of hedonic
pricing valuation concluded that 8-10% of the value of houses overlooking water can be
attributed to the pleasant view that water features offer (Luttik 2000). Similarly Schultz and
Schmitz (2008) estimated, from a large sample of homes with views of artificial lakes, that these
houses had premiums that ranged between 7.6-8.3% due to the view. If we assume that 8% of the
value of each parcel in this area of the country is attributable to water views, then each water
feature generates, on average, around $30,000 of value per parcel having a water view, with the
actual number depending on each parcel's assessed value for the 2009 distribution of home
values. This ecosystem good dominates the total value attributed to the three ecosystem goods
assessed in this study. Each water feature's total value was estimated as the sum of value
generated from all parcels with views of that water feature. The value of a water view can be
thought of as the present value of that natural capital that would be lost if that water feature was
removed or its view-ability degraded or blocked. This ecosystem good is estimated to be worth
$538.95 and $0.32 million US for Scenario A and B, respectively.
27
-------
Total: 10,000 Mature Trees
*^^^^-.. •- .-
Sv-..,:: -pi
Total: S22.63 Million
Figure 15. Development Scenarios A and B for trees per parcel (Al and Bl) and corresponding
Ecosystem Good value (A2 and B2).
28
-------
Aesthetic Value of
Reservoirs
140 - 624
625 - 984
985 - 1244
1245 - 2552
2553-5,124
Figure 16. Per square meter value of parcels with views of reservoirs in Scenario A (Al) and
Scenario B (Bl). The corresponding reservoir Ecosystem Good value (A2 and B2) is the sum of
value generated by all parcels within view divided by the reservoir area.
29
-------
SECTION 3. NEIGHBORHOOD SCALE VALUE OF ECOSYSTEM GOODS
AND SERVICES
Adding the three ecosystem goods values together (green space, viewable trees, and water views)
allows us to compare estimates of, albeit not complete, cumulative values of ecosystem attributes
for each scenario. The three ecosystem goods are valued at just over $571 million for Scenario A
and just over $5 million for Scenario B (Figure 17). Likewise, the sum of the four ecosystem
service values (air pollution removal, tree shading, nitrogen removal, and carbon sequestration)
yields a rough estimate of $2.3 million worth of yearly production for the existing development
and $1 million worth for the alternative scenario (Figure 18). Maps of these values allow us to
see the spatial distribution of both ecosystem goods and ecosystem service value throughout the
neighborhood (Figure 19).
Ecosystem Goods ($ in 2009)
Tree Views Water Views Green Space Total EG Value
Ecosystem Good Source
Figure 17. Total value of three ecosystem goods for each scenario.
30
-------
Ecosystem Services ($ per year)
0.0
Air Pol. Rein. Shading N Rem. C Seq. Total ES Value
Ecosystem Service Source
Figure 18. Total annual value of four ecosystem services for each scenario. (Air Pol. Rem. is Air
pollution removal, N Rem. is Nitrogen removal, and C Seq. is Carbon sequestration).
31
-------
Total: $571 Million
PS
Value of Ecosystem
Services
$/mz/yr
0.01 -0.02
0.03 - 0.07
0.08-0.15
0.16-0.20
0.21-0.95
• __
,•1
K!"!=
Figure 19. Development Scenarios A and B for cumulative Ecosystem Goods value in 2009 (Al
and Bl) and annual Ecosystem Services value production (A2 and B2).
BIODIVERSITY AND ECOSYSTEM GOODS AND SERVICES RESILIENCE
The final ecosystem attribute presented is biodiversity. Biodiversity holds no direct use value to
humans, other than existence value that is usually attributed to specific charismatic species, but it
has been postulated as providing a form of insurance against fluctuations in the production of all
other ecosystem goods and services. Baumgartner (2007) concluded, using a stylized conceptual
model combining ecological and economic factors, that biodiversity can serve as natural
insurance for risk adverse ecosystem goods and services managers. Quantification of the value of
biodiversity remains beyond current economic or ecological capability and is thus simply
presented here as a relative change using the metric of species abundance. Species diversity was
mapped by applying individual species abundance and presence/absence data (Table 4) from
Layne et al. (1977) to each FLUCCS land type for each scenario (Figure 20). The distribution of
land use/cover was then used to quantify the relative value of biodiversity for our two scenarios.
32
-------
It should be noted that the type of species noted for each FLUCCS type by Layne et al. (1977)
was not used to weight the relative contribution to this metric of biodiversity and so species
generally thought of as residential neighborhood nuisances such as feral cats, hogs, or alligators
hold the same value in quantifying this indicator of ecosystem services resilience as song birds,
bats, or butterflies. Quantification of the dollar value of biodiversity is beyond the scope of this
paper but an assessment of biodiversity in each scenario is included here to allow comparison of
our ecosystem goods and service values to this more familiar indicator of ecosystem integrity.
Table 4. Land Use Specific Species Richness
Description
[Number of species]
Reference
Residential Low Density
Residential Med Density
Residential High Density
Commercial And Services
Institutional
Recreational
Open Land
Cropland And Pastureland
Other Open Lands
Herbaceous
Shrub And Brushland
Upland Coniferous Forest
Pine Flatwoods
Hardwood Conifer Mixed
Streams And Waterways
Lakes
Reservoirs
Stream And Lake Swamps
Freshwater Marshes
Wet Prairies
Emergent Aquatic Vegetation
Intermittent Ponds
Transportation
Communications
Utilities
1100
1200
1300
1400
1700
1800
1900
2100
2600
3100
3200
4100
4110
4340
5100
5200
5300
6150
6410
6430
6440
6530
8100
8200
8300
16
16
16
16
43
43
43
37
22
17
50
64
50
59
68
76
85
(Layne etal. 1977)
(Layne etal. 1977)
(Layne etal. 1977)
(Layne etal. 1977)
(Layne etal. 1977)
(Layne etal. 1977)
(Layne etal. 1977)
(Layne etal. 1977)
(Layne etal. 1977)
(Layne etal. 1977)
(Layne etal. 1977)
(Layne etal. 1977)
(Layne etal. 1977)
(Layne etal. 1977)
(Layne etal. 1977)
(Layne etal. 1977)
(Layne etal. 1977)
33
-------
Figure 20. Biodiversity as a resilience indicator for ecosystem goods and services production for
development scenario A and B.
PEOPLE AS PART OF THE ECOSYSTEM
There can be inherent tradeoffs between having higher density residential areas, such as that
presented by our alternative development Scenario B, and maintaining a high level of benefits
from ecosystem goods and services. Selective placement and preservation of natural features
during development planning could greatly increase the potential long-term benefits to residents
from ecosystem services. Consideration of how to optimize the use of existing ecosystem
features during development, such as how to orient property to maximize benefits of shade trees
and how to place accessible and/or viewable green spaces and water features, may produce more
34
-------
sustainable neighborhoods with residents whose well-being is maintained by more than human
built components of the ecosystem. Our two development patterns are noticeably different in the
number of residences they contain. Scenario A has 4,068 parcels with an average assessed value
close to $175,000 while Scenario B has 3,603 more parcels but with only one-third of the
average assessed value (Figure 21). Scenario A parcels have an average acreage of 0.21 per
parcel for a total acreage of 846.97 while Scenario B has 7,671 parcels with a mean acreage of
0.19 per parcel for a total acreage of 1,422. To standardize the two scenarios' residential parcel
footprints we would have to replace approximately 757 acres of open, forested, and or wetland
area in Scenario A with residential parcels. This conversion would result in approximately $350
per acre less annual carbon sequestration and nitrogen removal services for a total of around
$265,000 per year in lost value for global and watershed beneficiaries. This is roughly one-form
our estimated difference between the two scenarios for these two services. This loss in value
may, however, be offset by dramatic increases in local value from residential tree viewscapes
and access to green space depending on where and how the additional parcels were located in
respect to forested areas and or parks. There are other complicating factors that can explain
differences in average parcel value, such as the age of property, building materials, proximity to
schools and crime, etc., so this relatively simplistic assessment of ecosystem goods and services
differences should be used alone to explain value differences. Also, one of the potential
advantages of denser developments does not present itself at neighborhood scales. The less
extensive footprint produced by having smaller parcels or multifamily buildings may leave
surrounding areas more undeveloped, assuming equal regional population numbers, than a
development pattern that is more sprawling in nature. Many ecosystem services are produced on
scales beyond the neighborhood, such as watershed, airshed, or region. Neighborhood scale
comparison results should only be considered alongside broader regional spatial changes in
ecosystem goods and services outside the neighborhood of interest. We only present these
ecosystem goods and services analyses to demonstrate an approach for including them in
development related decisions involving tradeoffs among ecosystem goods and services.
35
-------
-109
llO-217
218-303
304-486
487 -1565
Figure 21. Parcel value for development scenario A and B.
DISCUSSION
While not a comprehensive comparison of the potential benefits derived from two alternative
neighborhood development strategies, this study demonstrates, at the neighborhood scale, an
approach for quantifying differences in the value of EGS. This study looked at both generation
and delivery of benefits at the neighborhood scale. We also acknowledge that there are local
benefits from ecosystem goods and services generated from beyond the confines of the
neighborhood as well as regional and globally delivered benefits derived from the local area. The
EGS used in this demonstration piece are also not all inclusive, but were selected to be
representative of the types of ecosystem attributes that generate benefits to humans. Ecosystem
goods and services not included in this analysis include soil fertility and pollinator habitats
supporting agricultural production activities, access to aquatic ecosystems for recreational
36
-------
activities such as boating or fishing; recruitment and production of recreational and commercial
produce; and precipitation, retention and infiltration by landscapes for recharging water supplies
in aquifers and preventing flooding among others.
Ecosystem goods can potentially provide a large amount of value to residents at the
neighborhood scale. The highest values in this study came from the aesthetics of water features
and access to green spaces. The annual rates of ecosystem services production are of a seemingly
much lower value than the value wrapped up in ecosystem goods but comparing them at a given
point in time is unfair since services continue to produce value through time to both local and
remote beneficiaries. Comparison of our two scenarios, which both include ecosystems with
attributes that are ecosystem goods and processes that are ecosystem services, helps illustrate this
point.
There are two main components of the landscape that provide both ecosystem services and
ecosystem goods. Areas where water ponds or flows together to form streams, rivers, and other
water features are valued for water views but also produce conditions favorable for nitrogen
removal and carbon sequestration. Forested or other vegetated areas provide the structures
needed to produce value as an ecosystem good by providing opportunities to access green space
and for their aesthetic value while also functioning as pollution, carbon, and excess nitrogen
removal zones. When the total value of ecosystem goods and services in Scenario A are
examined we can determine the point in time when the accumulating value of annually produced
ecosystem services will surpass the mostly static value of ecosystem goods.
For Scenario A, the value of ecosystem goods is approximately 250 times that of a year's worth
of ecosystem services production. Another way of saying this is it will be around the year 2260
before Scenario A's ecosystem services generate a cumulative value equivalent to what is
inherent in their ecosystem goods in the year 2009. Much of the value of ecosystem goods is
generated at the initial building of a development. Scenario A, which is based on a neighborhood
developed around 1990, could be assumed to have already enjoyed about 15-20 years of
ecosystem service production and that it may take another 240 years before accumulated value of
ecosystem services surpasses the initial generation of ecosystem goods value during
neighborhood development. In comparison, Scenario B currently has about one-half the annual
production of ecosystem services as Scenario A, but because of the low value of ecosystem
goods present in Scenario B it will only take 5 years or so before the cumulative value of
ecosystem services production surpasses the value of the ecosystem goods. The difference in
annual production also implies that the differences in ecosystem goods and services value
generated by each scenario will compound over time. Communities should consider the temporal
aspect of value production from ecosystem services when making planning decision related to
sustainability goals.
37
-------
Out of the three ecosystem goods and four ecosystem services included in this study, the
quantification of all three ecosystem goods (accessible green space, and viewable trees and water
features) and one of the ecosystem services (reduced energy costs from shading) at the
neighborhood scale required local and high resolution spatial data such as land cover, roads,
trails, canopy cover, impervious surface, and sight lines combined with fairly customized or
manual GIS operations that are not easily obtained or implemented for broader areas. Carbon
sequestration and nitrogen removal process rates are easier to estimate on larger scales but are
harder to accurately value at the neighborhood scale. Carbon sequestration, for example, is only
beneficial, and thus generates value to humans indirectly through its influence on mitigating
rapid climate change. This realization or "delivery" of ecosystem service value to human
beneficiaries takes place in a dispersed way through the atmosphere. Nitrogen removal processes
only generate value to specific beneficiaries as an ecosystem service if excess nitrogen, that
otherwise would affect human health directly through the water supply or indirectly through
decreased production of other ecosystem goods due to nitrogen's influence on downstream
ecosystems, is being removed. The benefits of nitrogen removal are either delivered to
beneficiaries through stream drainage networks using downstream ecosystems or to upstream
beneficiaries that otherwise would have to worry about reducing their nitrogen loads to the
system through engineered solutions. Thus, beneficiaries have to be connected to ecosystem
service production areas to actually benefit in the same way as beneficiaries are connected to
ecosystem goods through transportation networks or viewsheds. A more complete valuation of
ecosystem goods and services at the neighborhood scale would have to consider both local and
remote beneficiaries of ecosystem goods and services produced in that defined area.
The value of ecosystem goods are not really separated from ecological functions since they are
often the result of ecosystem processes that took place in the past (e.g., tree and vegetative
growth, soil production, etc.). The difference between ecosystem goods and ecosystem services
is thus a temporal distinction with ecosystem goods assessed at a given point in time as a stock
while ecosystem services are assessed through time as rates. Here we defined the value of
ecosystem goods at a specific moment in time, but in reality their production through ecological
processes continues to be generated. Essentially, we are placing a value on ecosystem function
rates that happened in the past. If we could define how long it took to produce the ecosystem
good, then a production rate could, theoretically, be calculated. Taking the current value of a tree
and dividing it by the tree's age is used as an example of this. The value at any given time has
been produced from the growth of trees over many years. This growth rate could be considered
alongside ecosystem service rates instead of valuing the ecosystem good at a specific point in
time. A laural oak, a typical tree planted by developers in the Tampa Bay region, for example,
takes around 20-30 years to reach a mature size. Average 2009 detached house property value in
the FishHawk Ranch neighborhood was $385,028 according to www.city-data.com (accessed
June, 2012). Thus, each large tree, by generating 1% of that property value, may increase a
38
-------
property's value by as much as $3,850 just by being present and viewable on each parcel. We
applied this 2009 value estimate to our 2009 scenarios and divided by the area of each parcel.
Assuming a 20-year maturation period, the annual production of ecosystem good value from a
maturing tree equates to around $192.50 per year per tree. This method, albeit with many
assumptions, provides one way to compare the produced value of an ecosystem good, in this case
roughly $2 m"2 year"1, to other annual rates of ecosystem service value production as presented
above.
Alternatively, if we could estimate the value of ecosystem services using future states or stocks
of ecosystem attributes that are beneficial to humans at those specific points in time, we could
avoid the difficulties in trying to sum their values presented by valuing ecosystem goods as
stocks and ecosystem services as rates. While valuation of ecosystem goods and services at a
specific point in time would be less confusing than dealing with rates and stocks, this approach
would require many assumptions about future values.
In residential neighborhoods, where little of the preexisting landscape remains, it becomes
difficult to think of many of the ecosystem goods and services presented in this report as actually
derived from nature. Water retention ponds, for example, are a human construct. Many local
parks are landscaped green spaces with vegetation different than existed pre-development. Most
street and yard trees are planted post development with few previously existing mature trees
remaining. Since these features required human intervention should they be included in value
estimates for ecosystem goods and services? The answer to this question has dramatic
ramifications for valuation estimates of ecosystem goods and service production. Water feature
views generate the majority of ecosystem goods value in this study. If we were to discount those
features wholly created by humans (e.g., retention ponds) then the combined value of the
ecosystem goods present in Scenario A is reduced from over 80 to only 10 times more than that
generated by the annual production of ecosystem services. This shift in how value is generated
from ecosystem goods to annually produced ecosystem services could make significant
differences to decisions on how to best manage these natural assets based on value generated
from their ecosystem goods and services. The ecosystem goods and services paradigm developed
from a need to account for those things in nature that we derive benefits from and are not already
accounted for in our existing economic systems and markets. Thus, human input into producing
ecosystem goods or services that is quantifiable as part of the existing economy and markets,
such as fuel, equipment and labor costs, should theoretically be subtracted from value estimates
for any ecosystem attribute. The net benefit derived from an ecosystem attribute after subtracting
inputs from the market economy are nature's contribution within full cost benefit accounting.
This concept is currently being discussed within the ecosystem goods and services discipline, so
implementation is beyond the scope of this study.
39
-------
An added complication for estimating the value of ecosystem goods and services production in
specific ecosystems is that each will fall along a potentially non-linear production curve that is
dependent on various gradients of characteristics impinging on the system. These relationships
between external factors and functional responses are commonly referred to as ecological
production functions. While an assessment of changes in ecosystem goods and services using an
ecosystem replacement approach, such as used in this study, requires average values for that
ecosystem type's production of each EGS, an assessment that takes into account ecological
production functions requires estimates of production of each ecosystem good or service along at
least one gradient. This added complexity in developing ecological production functions
multiplies as one begins to assess the totality of an ecosystem's suite of EGS production, many
of which respond to multiple gradients. This type of assessment requires complex system
dynamics models operating both in space and through time and with the ability to account for
multiple spatial connections and ecological function interactions.
CONCLUSIONS
There is something that draws humans to neighborhoods developed with a consideration of green
space. Residents often pay higher prices for homes close to accessible green spaces and with
pleasant views of outdoor landscapes. The ecosystem goods and services paradigm provides us
with a defensible, transparent, and objective way of quantifying some of these relationships
between humans and their environment. The utility of the methods we describe in this report is
highest when used in relative comparisons between alternative management strategies or
scenarios.
The application of our methods to a defined geographical space at a given moment in time helps
to illustrate the differences between ecosystem goods and ecosystem services and how related
benefits are delivered to humans. Tradeoffs between fostering the immediate production of value
to residents from ecosystem goods versus sustaining the long-term production of ecosystem
services can be assessed in a spatially explicit manner while taking into account which, where
and when beneficiaries might realize benefits from nature. A developer might, for example, want
to assess the tradeoffs between maintaining an area as a functional forested wetland continuously
providing several widely distributed benefits through time versus opening it up to become a
viewable water feature providing a discrete but concentrated increase in value to local
beneficiaries.
The neighborhood scale quantification of ecosystem goods and services demonstrates the types
of data and customized workup required for accessing benefit tradeoffs associated with
commonplace realistic decisions. Assessment at this scale presents some challenges and there are
few "out-of-the-box" tools that can meet those challenges. This scale, however, works well for
40
-------
identifying beneficial components of the landscape that are relevant to humans and manageable.
The approach presented in this report is well suited to informing small-scale decisions such as
where to locate a walking trail or how to route water flow from street runoff. We propose that the
additional effort to generate geospatial data relevant at the neighborhood scale, and the
somewhat time intensive methods needed to translate from biophysical measures into value to
humans, is warranted since the results are easily relatable and informative for many real decision
contexts.
41
-------
42
-------
REFERENCES
Ajtay, G.L., P. Ketner, and P. Duvigneaud. 1979. Terrestrial primary production and phytomass.
In Global carbon cycle: SCOPE 13, ed. B. Bolin, E.T. Degens, S. Kempe and P. Ketner,
129-182. New York, NY: John Wiley and Sons.
Anderson, L.M., and H.K. Cordell. 1988. Influence of trees on residential property values in
Athens, Georgia (USA): a survey based on actual sales prices. Landscape and Urban
Planning 15: 153-164.
Bandaranayake, W., Y.L. Qian, WJ. Parton, D.S. Ojima, and R.F. Follett. 2003. Estimation of
soil organic carbon changes in turfgrass systems using the CENTURY model. Agronomy
Journal95: 558-563.
Barton, L., C. McLay, L. Schipper, and C. Smith. 1999. Annual denitrification rates in
agricultural and forest soils: a review. Soil Research 37: 1073-1094.
Baumgartner, S. 2007. The insurance value of biodiversity in the provision of ecosystem
services. Natural Resource Modeling 20: 87-127.
Birch, M.B.L., B.M. Gramig, W.R. Moomaw, O.C. Doering III, and CJ. Reeling. 2011. Why
metrics matter: evaluating policy choices for reactive nitrogen in the Chesapeake Bay
watershed. Environmental Science and Technology 45: 168-174.
Brenner, M., C.L. Schelske, and L.W. Keenan. 2001. Historical rates of sediment and nutrient
accumulation in marshes of the Upper St. Johns River Basin, Florida, U.S.A. Journal of
Paleolimnology 26: 241-257.
Carpenter, S.R., J.J. Cole, J.F. Kitchell, and M.L. Pace. 1998. Impact of dissolved organic
carbon, phosphorus, and grazing on phytoplankton biomass and production in
experimental lakes. Limnology and Oceanography 43: 73-80.
Carrick, H.J., F.J. Aldridge, and C.L. Schelske. 1993. Wind influences phytoplankton biomass
and composition in a shallow, productive lake. Limnology and Oceanography 38: 1179-
1192.
Clark, K.L., H.L. Gholz, J.B. Moncrieff, F. Cropley, and H.W. Loescher. 1999. Environmental
controls over net exchanges of carbon dioxide from contrasting Florida ecosystems.
Ecological Applications 9: 936-948.
Compton, J.E., J.A. Harrison, R.L. Dennis, T.L. Greaver, B.H. Hill, S.J. Jordan, H. Walker, and
H.V. Campbell. 2011. Ecosystem services altered by human changes in the nitrogen
cycle: a new perspective for US decision making. Ecology Letters 14: 804-815.
Cortright, J. 2009. Walking the walk: How walkability raises home values in US cities.
43
-------
Donovan, G.H., and D.T. Butry. 2009. The value of shade: Estimating the effect of urban trees
on summertime electricity use. Energy and Buildings 41: 662-668.
Electric, T. 2012.
Ensign, S., M. Piehler, and M. Doyle. 2008. Riparian zone denitrification affects nitrogen flux
through a tidal freshwater river. Biogeochemistry 91: 133-150.
Espinoza, L.A. 1997. Fate of nitrogen and metals following organic waste applications to some
Florida soils, University of Florida Gainsville, FL.
Gebhart, D.L., H. Johnson, H. Mayeux, and P. H.W. 1994. The CRP increases soil organic
carbon. Soil Water Conservation 49: 488-492.
Homer, C., J. Dewitz, J. Fry, M. Coan, N. Hossain, C. Larson, N. Herold, A. McKerrow, J.N.
VanDriel, and J. Wickham. 2007. Completion of the 2001 National Land Cover Database
for the Conterminous United States. Photogrammetric Engineering and Remote Sensing
73: 337-341.
Huang, Y.J., H. Akbari, H. Taha, and A.H. Rosenfeld. 1987. The potential of vegetation in
reducing summer cooling loads in residential buildings. Journal of Climate and Applied
Meteorology 26: 1103-1116.
James, T.R., W. Gardner, M. McCarthy, and S. Carini. 2011. Nitrogen dynamics in Lake
Okeechobee: forms, functions, and changes. Hydrobiologia 669: 199-212.
Johnson, L.T., and C. Hope. 2012. The social cost of carbon in U.S. regulatory impact analyses:
an introduction and critique. Journal of Environmental Studies and Sciences 2: 205-221.
Johnston, R.J., and M.J. Russell. 2011. An operational structure for clarity in ecosystem service
values. Ecological Economics 70: 2243-2249.
Kaplan, R., and S. Kaplan. 1989. The experience of nature: A psychological perspective:
Cambridge University Press.
Kroeger, T. 2008. Economic benefits of conserving natural lands: case study. Collier County
pine and swamp lands, Florida.: Prepared for the Doris Duke Charitable Foundation.
Washington, DC: Defenders of Wildlife.
Landers, D., and A. Nahlik. 2013. Final Ecosystem Goods and Services Classification System
(FEGS-CS). Office of Research and Development Washignton, DC: U.S. Environmental
Protection Agency.
Layne, J.N., J.A. Stallcup, G.E. Woolfenden, M.N. McCauley, and D.J. Worley. 1977. Fish and
wildlife inventory of the seven-county region included in the Central Florida Phosphate
Industry Area-Wide Environmental Impact Study.: Prepared by the Archbold Biological
Station for the U.S. Fish and Wildlife Service under Contract 14-16-0009-77-005.
44
-------
Lugo, A.E., S. Brown, and M.M. Brinson. 1988. Forested wetlands in freshwater and salt-water
environments. Limnology and Oceanography 33: 894-909.
Luttik, J. 2000. The value of trees, water and open space as reflected by house prices in the
Netherlands. Landscape and Urban Planning 48: 161-167.
Martin, J.F., and K.R. Reddy. 1997. Interaction and spatial distribution of wetland nitrogen
processes. Ecological Modelling 105: 1-21.
Milesi, C., S. Running, C. Elvidge, J. Dietz, B. Turtle, and R. Nemani. 2005. Mapping and
modeling the biogeochemical cycling of turf grasses in the United States. Environmental
Management 36: 426-438.
Murray, F.J., L. Marsh, and P. A. Bradford. 1994. New York State energy plan, vol. II: issue
reports. Albany, NY: New York State Energy Office.
Nowak, D.J., D.E. Crane, and J.C. Stevens. 2006. Air pollution removal by urban trees and
shrubs in the United States. Urban Forestry and Urban Greening 4: 115-123.
Nowak, D.J., D.E. Crane, J.C. Stevens, R.E. Hoehn, J.T. Walton, and J. Bond. 2008. A ground-
based method of assessing urban forest structure and ecosystem services. Arboriculture &
Urban Forestry 34: 347-358.
Nowak, D.J., and EJ. Greenfield. 2009. Urban and community forests of the Mid-Atlantic
Region New Jersey, New York, Pennsylvania: United States Department of Agriculture
Forest Service Northern Research Station.
Pifia-Ochoa, E., and M. Alvarez-Cob el as. 2006. Denitrification in aquatic environments: A
cross-system analysis. Biogeochemistry 81:111-130.
Pinay, G., B. Gumiero, E. Tabacchi, O. Gimenez, A.M. Tabacchi-Planty, M.M. Hefting, T.P.
Burt, V.A. Black, C. Nilsson, V. lordache, F. Bureau, L. Vought, G.E. Petts, and H.
Decamps. 2007. Patterns of denitrification rates in European alluvial soils under various
hydrological regimes. Freshwater Biology 52: 252-266.
Qian, Y., and R.F. Follett. 2002. Assessing soil carbon sequestration in turfgrass systems using
long-term soil testing data. Agronomy Journal 94: 930-935.
Raciti, S., A. Burgin, P. Groffman, D. Lewis, and T. Fahey. 2011. Denitrification in suburban
lawn soils. Journal of Environmental Quality 40: 1932-1940.
Reddy, K., W. Patrick Jr, and C. Lindau. 1989. Nitrification-denitrification at the plant root-
sediment interface in wetlands. Limnology and Oceanography: 1004-1013.
Robertson, G.P., P.M. Vitousek, P.A. Matson, and J.M. Tiedje. 1987. Denitrification in a
clearcut Loblolly pine (Pinus taeda L.) plantation in the southeastern US. Plant and Soil
97: 119-129.
45
-------
Roeder, E. 2007. A range of cost-effective strategies for reducing nitrogen contributions from
onsite sewage treatment and disposal systems: Bureau of Onsite Sewage Programs.
Schultz, S., and N. Schmitz. 2008. Viewshed analyses to measure the impact of lake views on
urban residential properties. The AppraisalJournall6: 224-232.
Seitzinger, S.P. 1988. Denitrification in freshwater and coastal marine ecosystems: Ecological
and geochemical significance. Limnology and Oceanography 33: 702-724.
Seitzinger, S.P. 1994. Linkages between organic matter mineralization and denitrification in
eight riparian wetlands. Biogeochemistry 25: 19-39.
Seitzinger, S.P., J.A. Harrison, J.K. Bohlke, A.F. Bouwman, R. Lowrance, B. Peterson, C.
Tobias, and G. Van Drecht. 2006. Denitrification across landscapes and waterscapes: A
synthesis. Ecological Applications 16: 2064-2090.
Simpson, J.R., and E.G. McPherson. 1996. Potential of tree shade for reducing residential energy
use in California. Journal of Arboriculture 22: 10-18.
Smith, C.J., and R.D. De Laune. 1983. Nitrogen loss from freshwater and saline estuarine
sediments. J. Environ. Qual. 12: 514-518.
SWFWMD. 2012. 2009 land use/cover features categorized according to the Florida Land Use
and Cover Classification System (FLUCCS). Southwest Florida Water Management
District.
Tsai, Y.-H. 1989. Factors affecting denitrification kinetics in selected Florida soils, University of
Florida Gainesville, FL.
US Government. 2010. Interagency working group on social cost of carbon. Technical support
document: social cost of carbon for regulatory impact analysis under Executive Order
12866.
Vitousek, P.M., J.D. Aber, R.W. Howarth, G.E. Likens, P.A. Matson, D.W. Schindler, W.H.
Schlesinge, and D.G. Tilman. 1997. Human alteration of the global nitrogen cycle:
sources and consequences. Ecological Applications 7: 737-750.
Walbridge, M.R., and E.G. Lockaby. 1994. Effects of forest management on biogeochemical
functions in southern forested wetlands. Wetlands 14: 10-17.
46
-------
United States
Environmental Protection
Agency
PRESORTED STANDARD
POSTAGES FEES PAID
EPA
PERM IT NO. G-35
Office of Research and Development
Gulf Ecology Division
Gulf Breeze, Florida
Official Business
Penalty for Private Use
$300
------- |