Reed College Canyon

Valuing the Benefits of Ecosystem Services Generated by the Reed Canyon Restoration Project: 1999-2009

Index | Executive Summary | 1 | 2 | 3 | 4 | References | Appendix A | Appendix B | Appendix C | Appendix D | Appendix E

4. Ecosystem services

Several ecosystem services were identified in the Reed College Canyon: water quality and quantity, air quality, wildlife habitat and recreation. This section summarizes relevant research and values, to the extent possible, the change in these ecosystem services resulting from the restoration of the Reed College Canyon.

4.1 Water Quality and Quantity in the Reed College Canyon

The Crystal Springs system, a tributary to the Johnson Creek watershed, originates in two spring complexes in southeast Portland: the hills above Reed Lake on the Reed College campus and further downstream in bluffs just east of the Eastmoreland Golf Course. Crystal Springs Creek runs 2.5 miles from its origins southwestward joining Johnson Creek one mile from its confluence with the Willamette River in Milwaukie (Lee and Snyder 2009). While the creek is naturally occurring, it has been modified to accommodate urban development, most significantly in 1936 during the construction of Westmoreland Park when it was channelized into a single stream (Lee and Snyder 2009). Natural and artificial lakes within the Reed Canyon and downstream in Eastmoreland capture and slow streamflow, elevating water temperatures noticeably due to reduced flow velocity. Industrial zones and fertilization use below Reed have degraded water quality.

Crystal Springs, as a perennial, groundwater-fed spring, is unique within the Johnson Creek Watershed.  While other systems within the watershed have high seasonal variation in volume, temperature, and potential for flooding and contamination, Crystal Springs demonstrates steady flow throughout the year with relatively constant temperatures (Lee and Snyder 2009).

Many of the salient benefits of Crystal Springs existed well before canyon restoration efforts began in 1999-2000. Crystal Springs contributes nearly half of the total water measured at the confluence of Johnson Creek and the Willamette River in Milwaukie, and research has demonstrated that changes in Crystal Springs streamflow are directly reflected in the outflow of Johnson Creek (Karl Lee, November 9, 2009, personal communication with Tom Fenollosa). Significant changes in water conditions within the Crystal Springs and Johnson Creek watersheds due to restoration work will presumably occur downstream of the Reed Campus and within Johnson Creek, rather than onsite near the headwaters of Crystal Springs. Thus, immediate benefits of the Canyon restoration project, while potentially beneficial to systemic health of the Johnson Creek watershed, may not be measurable on the Reed Campus-except for, perhaps, by the presence of anadromous fish.

4.1.1 Water Quantity

Restoration projects located near the emergence of Crystal Springs are believed to have had little effect on water flows into the Reed Canyon. The land area that contributes surface water to Crystal Springs on the Reed property includes only the immediate bluffs above the Reed Canyon. This area's contributions to streamflow are significant only during particularly hard rainstorms when stormwater runoff from impervious surfaces is greatest (Lee and Snyder 2009).

Over the past decade, streamflow in the Crystal Springs Creek has decreased slowly, from an average of 17 cubic feet per second in 1998-1999 to flows ranging between 10-14 cubic feet per second since 2000 at the mouth, and from 6-7 cubic feet per second to 1-2 feet per second within the Reed Canyon (Lee and Snyder 2009, 11 & 49). This coincides with the canyon restoration project, but USGS Hydrologist Karl Lee (November 9, 2009, personal communication with Tom Fenollosa) states that this decline in streamflow volume was due to groundwater recharge and was unrelated to restoration work on the Reed Campus.

While data from the Reed Campus do not exist before the mid-1990s, historical measurements do exist for the mouth of Crystal Springs Creek. These suggest that the mid-1990s were significantly high-water years throughout Crystal Springs Creek and that present measurements reflect a return to normal water levels (Lee and Snyder 2009). Regardless of decreasing trends in water volume in Crystal Springs, it is clear that Crystal Springs Creek supplies significant volume to Johnson Creek with the proportion of flow increasing during critical summer months when flows of Johnson Creek are at their lowest point. Similarly, because spring supply is steady, erosion and sedimentation damage caused by rapid streamflow change is minimal.

In many urban hydrological systems, including other, more flood-prone areas of Portland like the upstream reaches of Johnson Creek, tree canopy has been an important tool in managing excessive stormwater. This is particularly effective in locations with large areas of impervious surfaces. A study by American Forests (2001, 7) finds that urban tree canopy in the Willamette/Lower Columbia basins reduces the need for stormwater management by 8.5 billion cubic feet, at an estimated monetary benefit of $6 per cubic foot in urban settings. While tree canopy in the canyon does provide valuable water assimilation services, restoration efforts ought to have little effect on the coverage of a forest canopy; specific composition is changing, but vegetative canopy has remained the same.

The Portland Bureau of Environmental Services states that its Grey-to-Green infrastructure initiatives can have a considerable effect on metropolitan hydrology. BES claims that invasive plant removal and native plant revegetation, including work done in the Reed Canyon, reduces stormwater volume, filters stormwater pollutants, provides habitat diversity, and cools the air, pavement, and streams (BES 2009, 1-7). BES predicts that the restoration of 840 acres of public lands under its Grey to Green initiative will, after five years, slow stormwater flow rates by 10 percent, improve groundwater recharge and surface infiltration by 44.5 million gallons per year, increase evapotranspiration by 20.5 million gallons per year, and enhance stormwater reuse by 8.9 million gallons per year (BES 2009, 1-3). Additionally, BES expects significant reductions in streamwater temperature and contamination. However, because restoration usually occurs in natural areas that completely assimilate rainwater, and do not produce stormwater runoff, restoration practices generally do not affect stormwater quantities (BES 2009, 5-2).

4.1.2 Water Quality

With respect to water quality, Crystal Springs presents another somewhat unique case. Groundwater tends to be minimally contaminated and considerably cooler compared to surface water due to soil filtering and generally longer recharge times. Historically, Crystal Springs was considered as a potential source of drinking water for metropolitan residents and as suitable fish spawning habitat (Karl Lee November 9, 2009, personal communication with Tom Fenollosa). Despite high quality water from Crystal Springs, the Johnson Creek system regularly receives very poor ratings from the Oregon Department of Environmental Quality on their Water Quality Index because of high water temperatures during the summer, fecal coliform, and toxins (Oregon DEQ, 2008).

The ECONorthwest study (2009) conducted for the East Buttes area of Portland finds that native plant restoration projects provide significant water filtering and improved water quality, and that the integrity of aquatic and riparian habitat in Johnson Creek and its tributaries would likely contribute to sustaining and enhancing fish populations (ECONorthwest 2009, 40). Similarly, BES's Grey-to-Green culvert-removal program has prioritized Crystal Springs because of its potential for providing salmonid habitat-Crystal Springs is known to be spawning habitat for Coho and Chinook salmon, and steelhead trout (Entrix 2009). BES predicts that culvert removal and repair will open more than 4 miles of stream passage in the Portland metropolitan area and significantly reduce streamwater temperatures (BES 2009, 1-3), all goals shared by the Reed College restoration efforts. Helvoigt and Montgomery report in their 2003 study-though not specific to Crystal Springs or Johnson Creek-that the monthly household willingness-to-pay for improved salmon runs in Oregon is $4.22 (in 2008 dollars) (ECONorthwest 2004, 21), and a 2009 survey performed by Richardson and Loomis finds that households have a one-time willingness-to-pay for cutthroat trout protection in Johnson and Kelley Creeks of $21 (ECONorthwest 2009, 40).

The ECONorthwest report also suggests a correlation between ambient water quality and the value of streamside properties. A Maryland study conducted by Poor et al. (2006) estimates that each additional unit (mg/L) of suspended solids in watercourses reduces average housing prices within the watershed by 0.530%, and increases in waterborne inorganic nitrogen reduce housing value by 8.780% per unit. While these are important studies, these estimates may not be appropriate for valuing the impact of restoration work in the Reed Canyon. Crystal Springs is known to have some of the cleanest water in Portland because it is almost exclusively groundwater-fed, and small marginal increases or decreases may not be considerable enough to alter property values.

Non-native vegetation, especially like that of the pre-restoration canyon, generally provides poor hillside support. This may lead to soil erosion and poor water quality. A study by Ribaudo, cited in ECONorthwest (2009, 44) estimates that offsite benefits of reducing soil erosion, and thus mitigating landslides and watercourse sedimentation, is approximately $4 per ton. However, while Canyon Specialist Zachariah Perry agrees that non-native vegetation does increase erosion, the Reed Canyon has never had a significant erosion problem. He explains that the establishment of a trail system removed from the immediate shoreline has mitigated the little erosion that periodically occurred due to dogs or people entering the canyon. For the purposes of this study, though, erosion control is not significant enough to have a reliable value (Zachariah Perry 2009, personal communication with Tom Fenollosa).

While fertilizer use has been a concern downstream of the Reed Campus, particularly in the industrial sectors west of 28th Avenue and in Westmoreland Park, there has been little evidence of contamination in the Canyon. The Reed Buildings and Grounds department started using slow-release organic fertilizers in the first few years of the restoration program (Zachariah Perry 2009, personal communication with Tom Fenollosa) eventhough elevated mineral levels have never been detected in Crystal Springs Creek. Zachariah Perry has observed increased aquatic plant growth in recent years, but the connection between fertilizer use and plant growth is inconclusive (Zachariah Perry 2009, personal communication with Tom Fenollosa).

Because groundwater sources tend to be of better quality compared to surfacewater-sources, and tend to have more consistent seasonal flow volumes, revegetation and other restoration efforts in the Reed Canyon may not have significant effects on water quantity and quality within Crystal Springs. While Crystal Springs clearly contributes considerably to Johnson Creek, restoration does not appear to have any additional effect. Thus, salient benefits to Crystal Springs hydrology, due to Canyon restoration, are at this point inconclusive.

4.2 Air Quality in the Reed College Canyon

Air pollution is an environmental and health concern. A decrease in air quality is correlated with cancer, skin problems, and respiratory disease (EPA 2009). An improvement in air quality is correlated with an increase in recreation and time spent outside and a reduction in associated health care costs (EPA 2009).

Levels of various air pollutants may be used as a yardstick to gauge air quality. Primary data collection and comparison to other studies was undertaken to investigate the following three pollutants: nitrogen dioxide (NO2), particulate matter, and carbon dioxide (CO2), within the canyon. Using the primary data, values for air quality improvements made by the canyon were estimated using benefit transfers from comparable studies.

Urban trees and shrubs remove gaseous air pollution (Novak et al. 2006). The improvement in air quality occurs as a result of both chemical and physical processes, for example, particulate matter deposes to plant surfaces, ozone and NO2 are absorbed via leaf stomata, and CO2 is consumed during photosynthesis. The canyon restoration project has included a reintroduction of native trees and shrubs. Zachariah Perry estimates that the restoration project has increased the absolute amount of vegetation and tree canopy by 30% (Zachariah Perry, December 14, 2009, meeting with Claire Remington).

4.2.1 Nitrogen Dioxide

Nitrogen dioxide is a major air pollutant. It contributes to environmental concerns like acid rain and eutrophication which is caused by excessive nutrients in water bodies. These nutrients encourage enhanced aquatic plant growth, reducing dissolved oxygen in the water and creating dead zones inhospitable to a diversity of plants and animals (Henry 1993). It also irritates lungs and lowers resistance to respiratory infections. Nitrogen dioxide also plays a major role in the chemical reaction that produces ozone. Ozone, a secondary pollutant, reduces lung function and damages forests.

The atmospheric removal of nitrogen dioxide decreases the production of the secondary pollutant ozone. Ozone is produced from the photochemical reaction of NO2 with volatile organic compounds. The rate of this reaction decreases at lower temperatures, so increasing tree canopy, which lowers ambient air temperatures, can significantly reduce ozone concentrations.

The levels of nitrogen dioxide in the canyon were measured explicitly using passive nitrogen dioxide samplers (Palmes et al. 1976). As shown in Figure 3, a total of 21 test sites were prepared and distributed throughout the canyon and the Reed campus- including a few sites at the outskirts and near busy intersections

Figure 3: NO2 concentration (ppb) at 18 sampling sites

The average concentration of NO2 calculated from samples within the canyon was 11.3 ppb over a two-and-a-half week period (11/22/2009 - 12/10/2009). The average concentration of NO2 calculated from samples outside the canyon boundary, as shown in Figure 3, was 13.4 ppb during the same period. Air quality in the canyon improved by 2.1 ppb NO2 (Appendix D). This decline may be a result of the deposition of NO2 on the plant and absorption by the plant.

The quantity of nitrogen dioxide removed by urban trees (this includes trees and shrubs) in the canyon was measured implicitly by estimating the amount of canopy cover within the canyon boundary. A modeling study by Novak (2006) found a removal rate of 0.7-1.9 g m^-2 for urban trees in Portland, OR. Based on total vegetation cover, the removal rate is: 4.4x10⁴ g/hr^-1-1.2x10⁵g/hr^-1.

The restoration's effort increased the quantity of vegetation in the canyon. This creates a cooling effect and removes NO2 from the ambient atmosphere - this subsequently decreases the production of ozone.

4.2.2 Comparison to Studies Performed in Portland

The data collected can be assigned a monetary value using the benefit transfer technique. ECONorthwest (2004) performed an ecosystem service valuation of the Lents neighborhood in Portland. Their report cites two studies, from the California Energy Commission and the U.S. Office of Management and Budget, in developing a value for NOx reduction: "The calculation of values per ton of pollutant removed, in year 2002 dollars from the CEC and OMG reports include $5,572 to $1,114 per ton of NOx" (ECONorthwest 2004, 18). In 2009 dollars, these values range from $6,377 to $1,275. This values the NOx removal by vegetation in the canyon as a range of $61.2 to $841 per hour.

4.2.3 Comparison to Studies Performed Elsewhere

Studies performed elsewhere provide proxy values for estimating the value of air quality improvements resulting from the canyon restoration projects. These studies primarily focus on urban cities located on the east coast of the U.S. There are intrinsic similarities and differences between the cities studied and Portland. Similarities include: urban lifestyles, residential family areas, and access to open space (including parks) while differences include: demographic characteristics, education levels, and housing markets (U.S. Census 2009).

A study by Zabel and Kiel (2000) finds a large positive marginal willingness to pay for ozone reduction. In Philadelphia, the mean average benefit per household from reducing ozone from 142 to 120 ppm is $537 and the total benefit is $953 million. In Washington, D.C., the mean of the average benefit per household from reducing ozone from 136 to 120 [ppm] is $334 and the total benefit is $526 million (Zabel and Kiel 2000). This value is not directly transferable to residents who own property near the canyon. However, it does suggest that habitation near an ozone sink has a positive effect on the price of that property, holding housing characteristics constant. The magnitude of this value will vary from location to location, but these findings suggest positive values for ozone reduction from the perspective of urban property owners.

4.2.4 Particulates

Particulate matter is produced from the combustion and reaction of gaseous pollutants. It damages lung tissue and causes cancer and premature death (EPA 2009). Prior to 1987, the EPA regulated air quality on the basis of total suspended particulate (TSP). After this date, the EPA replaced this with the PM-10 and PM-2.5 standards. PM-10 refers to particles smaller than 10 micrometers; these particles may be inhaled. PM-2.5 refers to particles smaller than 2.5; these are respirable.

The levels of TSP were measured in the canyon and at various sites around campus. The device used was a handheld condensation particle counter. This device measures particles with diameters smaller than 1 micrometer. These particles are described as ultrafine particles.

Figure 4: Distribution of Locations of Ultrafine Particle Measurement

The limitation of the device used for the analysis is that the data collected are not comparable to other values reported in the literature. The device used measured only ultrafine particles, PM1, (diameter less than one micrometer), whereas the EPA regulates PM2.5 and PM10 (particles with diameters less than 2.5 micrometers and 10 micrometers respectively). However, the data collected can be used to show that levels of a class of particulates known to be harmful to health are lower within the canyon boundary than outside the study area. The average concentration of ultrafine particles measured within the canyon at selected sample sites was 8744 particles/cm³. The average concentration outside the canyon boundary was 19565 particles/cm³. The data collected can thus act as a proxy value. It shows that the concentration of ultrafine particulates is less inside the canyon by a margin of 10821 particles/cm³ (Appendix E).

Studies show that particulate removal reduces health and legal compliance costs (Evans 2004). Using estimates from Evans (2004), the 28-acre canyon is responsible for removing around 2.5 pounds of particulate matter from the air. The ECONorthwest study, using a model developed by American Forests (2003), determined that a one ton reduction of particulate matter is worth $4,519 (Evans 2004). An additional estimate, using the CEC estimates from 1992, estimates that a one ton reduction of particulate matters is worth $1,817 ($2,800 in 2009 dollars).

The quantity of PM-10 removed by vegetation in the canyon was measured implicitly by estimating the canopy cover of vegetation (Figure 5). A modeling study presents the relationship between urban tree canopy cover and pollutant removal (Novak 2006). The study found a removal rate of 1.3-5.2 g m^-2hr^-1 for urban trees in Portland, OR. Based on total vegetation cover of 6.28x10⁸ cm², the removal rate is: 900 to 3,600 tons per hour. This is worth: $252 to $1008 per hour.

Figure 5: Vegetation Coverage of the Reed College Canyon

4.3.5 Comparison to Studies Performed Elsewhere

According to Chattopadhyay (1999), a 25% reduction in particulate pollution creates a benefit to the average household worth $2,037 to $3,350 in 1989-90 dollars ($3,369-$5,541 in 2009 dollars). In addition, the author finds a one unit reduction in particulate matter to be worth $268 to $363 ($443-$600 in 2009 dollars).

4.3.6 Carbon Dioxide

The EPA ruled that carbon dioxide, in addition to other greenhouse gases, is a danger to human health (EPA 2009). The EPA's endangerment finding allows the Obama administration to circumvent Congress and begin issuing regulations on CO2 emissions. The value of carbon sequestration that occurs within the canyon may be estimated using the carbon offset market by using carbon permit prices as a proxy. This method of valuation will become increasingly accurate as prices within the permit market stabilize and the market as a whole becomes more widely used.

The ECONorthwest study values carbon sequestration at $9.50 per year per ton (Evans 2004). One study finds that "the public's WTP to prevent ecosystem impacts [as a result of climate change] is [not significantly] different from 0$ for 'small' impacts. [The WTP] increases sharply as the impact approaches complete ecosystem change" (Layton and Levine 2003, 543). This suggests that the estimate based on the carbon market should be used as a baseline for our valuation of the Reed Canyon. As research into the value of positive externalities associated with carbon sinks improves, this estimate can be refined.

4.4 Wildlife habitat in the Reed College Canyon

4.4.1 Background

The habitat of the Reed Canyon is comprised of the Reed Lake and its surrounding emergent wetland and marsh areas, forested riparian and upland habitats, beaver dams, and the Reed Creek channel. Riparian vegetation shades about half of this channel. Upland areas are mixes of dense tree canopy and dense understory. For a partial list of plant species, see Appendix B. The dense vegetation cover, relatively low disturbance rates, and structural diversity of the vegetation provide habitat for a variety of species. The presence of side pools, year-round flow of water, and gravel substrate within the creek channel make the area suitable for salmonid redds and habitat (Aldofson and Associates Inc 2001).

Restoration of the Reed Canyon greatly improved the biodiversity of the area. The majority of the dominating invasive species that were out-competing native plants were removed, allowing native trees to prosper. Also, before the restoration there was minimal understory due to annual burning. Restoration has included numerous plantings of lower-level plants such as sword ferns, wood sorrel, and fringe cup to create a more structurally complex ecosystem. Animal biodiversity has also improved. While a number of non-native birds such as crows and starlings populated the Reed Canyon before restoration, within one year of commencing restoration songbirds and waterfowl such as widgeons, wood ducks, and mergansers had found a home in the area. Since the restoration frogs and a new species of newt have been spotted in the Reed Canyon (Zachariah Perry, October 14, 2009, meeting with Lauren Bloomquist and Rachel Workin). For a partial list of animals that inhabit the Reed Canyon, see Appendix C. A wildlife habitat assessment by Portland Parks and Recreation summarized the area as having high quality water, food, and cover (Aldofson Associates Inc 2001).

The ecological and economic benefits to a strong, healthy habitat are numerous. Habitat acts as both a benefit in itself as well as a final service and intermediate component for other ecological benefits such as clean drinking water, air quality, intrinsically valued species, and recreation. In itself, high-quality habitat provides us with biodiversity, control of invasive species, in addition to potentially raising home prices.

4.4.2 Biodiversity

The Reed Canyon restoration project increased the biodiversity of the study area. It has been suggested that high species richness can promote resilience and heightened adaptability to human-induced pressures such as climate change, invasive species, and disease (Secretariat of the Convention on Biological Diversity 2009). This provides economic benefit in the form of avoided-costs due to loss of species' productivity because of stress. There is also evidence of increased productivity of varied ecosystems over less diverse systems under the same condition (Secretariat of the Convention on Biological Diversity 2009). A study of the benefits derived from biodiversity in the United States, based on the ecological and genetic services provided, estimates that the economic value of these services is $319 billion per year (Pimentel 1997).

Increasing the biodiversity of the Reed Canyon involved an initial large-scale removal of invasive species as well as continuous monitoring and maintenance (see Section 2.1). This control of the spread of invasive species provides economic benefits through the avoidance of future costs. Noxious weeds cost the state of Oregon $125 million dollars a year in production losses, fire damage, and control costs (Cusack et al. 2009). Personal income loss due to invasive species in Oregon amounts to $83 million per year (Oregon Department of Agriculture 2000). Based on a land index of 7,800 acres owned by the City of Portland, invasive plants have already spread to cover between 13% and 40% of the land area in Portland (Bureau of Environmental Services 2008). The BES Revegetation Team estimates that the cost of invasive removal and subsequent revegetation is $12,000 per acre over five years (Bureau of Environmental Services 2008). The restoration of the Reed Canyon restoration helps avoid these costs by limiting the spread of some of these invasive plants. According to the Oregon Department of Agriculture, the early detection of invasive species, such as that occurring in the Reed Canyon, provides an economic benefit of $34 in savings of future cost impacts for every $1 spent (Oregon Department of Agriculture 2000).

4.4.3 Home Prices: Hedonics

The land surrounding the Reed Canyon, especially the eastern and northeastern sides, contains predominantly single-family residential homes. Hedonic analysis can be used to attempt to understand how much people are willing to pay for a public good that is restricted to a certain spatial or geographic area; looking at whether property values increase as proximity to the Reed Canyon increases will allow us to explore whether Portland residents put a premium on the services that the Reed Canyon provides. Presumably other variables may contribute to the variance in property values; the hedonic price method incorporates variables that control for a house's age, size, etc. to more easily examine other variables, like distance to a wetlands area.

Given that the existence of wildlife habitat is a public good, approximating how much people value it presents much difficulty. With the exception of non-profit organizations like The Nature Conservancy or the Sierra Club, almost no markets exist specifically for wildlife habitat. The fact that people can free ride on government programs and others' private donations makes it likely that there is underinvestment in wildlife habitat. The hedonic method, when applied to the valuation of wildlife habitat, can demonstrate a revealed preference for living near wildlife habitat and, in turn, help us approximate how much people value habitat. It is crucial to note that the hedonic method is limited to values in a specific area; so, although many people may value the habitat and the services it provides, the hedonic method only measures the values of those who live near the study area.

Mahan et al.'s study (2000) finds a positive correlation between property values and proximity to wetlands, streams, and lakes in Portland, using data from 1992 to 1994. According to this study, a 28-acre wetland area like the Reed Canyon is estimated to increase property value by 0.56%. Given that in September 2009, the average value of a single-family residential property within one quarter of a mile of the Reed Creek was $433,986, a 0.56% increase would equal a $2,430 premium. Increasing proximity to a stream by 1000 feet increases property values by 0.21% ($911), while increasing proximity to a wetlands area by 1000 feet increases values by 0.36% ($1562). Finally, increasing proximity to a lake by 1000 feet increases property values by 1.34% ($5,815). Using Mahan's estimates, it is clear that the Reed Canyon, a wetlands area that encompasses the Reed Lake and the Reed Creek, increases property values by at least $911, given its ecological services.

Another study established a positive relationship between open water wetlands and property values in Portland, Oregon (Bin 2005). The capitalization of the value of wetlands into single-family home sale prices may be due, in part, to residents willingness to pay for wildlife habitat, but it is likely that the increased value from proximity to streams, lakes, and wetlands is also due to recreational value and aesthetic value. Although it is clear that Portland residents value wetlands, streams, and lakes, neither Mahan's nor Bin's study attempts to explain what about those features they value-aesthetic value, recreational value, existence value, etc. This problem is reflective of the ultimate difficulty of using hedonic analysis in valuing habitat-the value of habitat is almost impossible to isolate from other values in revealed preference methods.

One way to attempt to isolate the willingness to pay for habitat from other values is to examine changes in property values once a restoration project has been completed. Hedonic analysis in the Bay area of California and Santa Cruz, California found higher property values near government-restored urban streams than values for those that were not restored (Streiner and Loomis 1996). The same study found a one-time premium for restoration projects that revegetated with native plants and removed stream obstructions of about 3% to 5.3% of property value. Higher one-time premiums of about 10.6% to 13% of property value were found for projects that maintained fish habitat, acquired land, and/or established educational trail systems.

The Reed Canyon has focused on revegetation, establishing a trail system, removing stream obstructions, and re-establishing the stream as fish habitat. Although the geographical areas under study are in different states, both areas receive similar levels of precipitation, are on the West coast, and are inland from the coast (see Section 3.7 for an explanation of benefit transfer). Many stream areas in the Bay area are also semi-forested, like the Reed Canyon. Additionally, cities in the Bay area offer public services and amenities very similar to those of Portland, indicating that their residents' values are comparable. So, it is likely that the Reed restoration project increased property values by a similar percentage.

Given a one-time premium of 10.6% to 13% for restoration projects that restored fish habitat and/or established trail systems, it is likely that the Reed restoration project increased property values within a quarter mile of the Reed canyon by $46,003 to $56,418. Given 327 single-family residential properties within 1/4 mile of the Reed Lake, we estimate that residents of each property placed a lower premium of $46,003 on the restoration project for a total value of $15,042,981

Restoration projects, including the Reed Canyon restoration and those studied by Streiner and Loomis, are habitat-centric; they attempt to improve natural areas in the hopes of providing a safe haven for wildlife and improving animal and plant populations after the destruction of habitat due to urbanization. Increased property values after restoration projects attempt to capture the value of this improved wildlife habitat to urban residents. For this reason, the values from Streiner and Loomis's (1996) hedonic study are more applicable to the valuation of wildlife habitat than those from Mahan et al. (2000).

Another study examined the value of improving the quality of riparian corridors and wildlife habitat in Fanno Creek watershed (Netusil 2006). Fanno Creek watershed is in SW Portland and in Tigard, Oregon (Portland Bureau of Environmental Services). Fanno Creek is approximately 10 miles southwest of the Reed canyon in the Johnson Creek watershed. This study used habitat quality rankings and riparian corridor quality rankings from Metro regional government, ranging from Class A to C wildlife habitat (A being highest-value) and Class 1 to 3 riparian corridor quality (1 being highest-value). Figure 6 demonstrates rankings in the Reed Canyon.

Figure 6: Color-coded riparian corridor and wildlife habitat rankings of 3203 SE Woodstock Blvd (Metro 2003)

These rankings were created 3 years after the restoration project had begun, and as is visible in Figure 6, the Reed canyon contains primarily Class 1 riparian corridors and Class A wildlife habitat-the highest-value for both. Unfortunately, Metro's quality rankings do not show changes over time and are from 2003, so there is no way to tell if the Reed restoration project improved habitat or riparian corridor quality rankings. But given personal testimony from Reed staff, it is clear that habitat has improved significantly (see Section 2). Netusil's 2006 study used the hedonic price method to examine the effect on property values of improved quality rankings by Metro's standards; this study found that residents are placing a premium on lots with habitat providing the highest ecological values, those lots being Class 1 and Class A. Given the proximity and similarity of the Fanno Creek watershed to the Reed Canyon, it is extremely likely that property owners living near the Reed Canyon placed a similar premium on the private Reed restoration project and its improvements on wildlife habitat and riparian corridors.

4.4.4 Willingness to Pay: Funding from Outside Organizations

Another method of valuing wildlife habitat is through government funding. Although what the government regards as important and valuable may not be identical with the value of its constituents, it cannot be denied that the government has funding and influence-its values are important. In the realm of public goods, the government is a crucial actor; it is often considered to be the government's job to regulate externalities, improve environmental quality, and use taxation to attempt to combat the free rider problem and provide public goods, like the existence of wildlife habitat.

Table 1: Public and non-profit funding sources for the first phase of the Reed canyon restoration project. (David Frazee Johnson, 2009)

Although only a small portion of the overall funding for the Reed restoration project came from the public sector, the fact that the government is willing to invest in projects similar to Reed's private restoration project shows that the government is willing to pay for wildlife habitat to exist and places a value on the Reed canyon and the high quality wildlife habitat it provides.

Portland's government places a high premium on wildlife habitat. For the 2009-2010 fiscal year, the Portland Bureau of Environmental Services had several watershed-focused programs (see Table 2 for program summaries).

Table 2: 2009-10 Bureau of Environmental Services Programs for Watersheds (City of Portland Management & Finance)

The City of Portland spends millions on revegetation each year as part of their Watershed Revegetation program. The Reed restoration project, which covered the entire 28-acre canyon, has ensured the removal of many invasive species and has organized extensive revegetation efforts. If the City of Portland values its 70-acre Watershed Revegetation program at $2,242,067 total, then, using acres as a metric, an estimate for their value of the revegetation of the Reed canyon is approximately $896,826. It is more difficult to separate the other three programs implemented by the Portland Bureau of Environmental Services (Sustainable Storm water, Science Fish and Wildlife, and Watershed Management) by activity to find an actual dollar value. Regardless, it is clear that the City of Portland's plan to spend over $10.5 million on three watershed restoration programs alone indicates the value the City of Portland places on wildlife habitat and restoration projects.

In 1995, voters approved a Metro bond measure for $135.6 million specifically for acquiring open spaces, parks, and streams (Metro). Again in 2006, voters approved a bond measure for $227.4 million directing Metro to purchase natural areas, parks, and streams (Metro). Metro's 2009-2010 fiscal year budget allocated $60,015,529 to its Natural Areas Fund. This bond measure indicates voters preferences in the region. That a 58.6% majority voted for the measure demonstrates voters willingness to accept current debt in exchange for improved natural areas (Metro). Additionally, Metro as an organization has made a commitment to maintaining the health of the watersheds in Multnomah County and surrounding counties through its Title 13 / Nature in Neighborhoods program, which was given a $15 million budget in the 2009-10 fiscal year. Metro also publishes a report that studies the health of the watersheds and creates recommendations for how to protect and improve wildlife habitat (Hennings 2006). Again, it is difficult to assign a dollar value to the Reed restoration project given how much Metro values watershed health because of the sheer size of the organization and how it organizes its funding; but the fact that voters supported the 1995 and 2006 bond measures indicates voters willingness to pay for habitat in Portland and other nearby cities. The fact that Metro dedicates itself to sustainability and maintaining the watersheds under its jurisdiction also indicates Metro's dedication to conservation and that voters and taxpayers have shown a willingness to pay to support Metro's efforts.

Non-profit organizations like the Bullitt Foundation and the National Fish & Wildlife Foundation also donated $30,000 and $90,000, respectively, to the first phase of the restoration project; these donations indicate each respective organization's willingness to pay for the improved ecosystem services in the Reed Canyon. That two non-profit organizations chose to give money to the Reed restoration project indicates their willingness to pay for the habitat the Reed Canyon provides as well as, to some degree, their donors willingness to pay for the Reed restoration project (and other projects that enhance habitat). Individual donors gave over $250,000 to Reed earmarked specifically for the canyon; though this value doesn't indicate how much of those donations were for the restoration of wildlife habitat, knowledge of the plan for the restoration project (which is habitat-centric) was likely a significant factor in deciding to donate.

4.4.5 Willingness to Pay: Individuals

Individuals are also willing to pay to protect wildlife habitat. One contingent valuation survey (Woodward 2001) concerning the value of wetlands found that, in total, each acre of habitat is worth $374 per year (in 2009 dollars). For the 28 acres of the Reed Canyon, this amounts to $10,472. However, there is additional benefit from habitat based on the species the habitat attracts and the recreation the area provides. For example, Bald Eagles have been spotted in the Reed Canyon. A meta-analysis of contingent valuation surveys determined the economic value of each bald eagle to be an average of $297 (Richardson 2008). By providing habitat for species, the Reed Canyon contributes to the benefit people derive from the existence of species including the bald eagle.

Willingness to pay for wildlife habitat and species can also be examined by looking at membership in groups such as the Portland Audubon Society. The Portland Audubon Society, whose mission is to "[promote] the enjoyment, understanding, and protection of native birds and other wildlife and their habitats" focusing on the local community and the Pacific Northwest, has membership fees ranging from $25.00 at the introductory level to $1,000 annually. According to the 2007-2008 annual report, the 10,000 members of Portland Audubon generated $267,155 in revenue from membership fees (Audubon Society of Portland 2008). That people pay to be members of groups such as the Audubon Society that protect habitat demonstrates a valuation for habitat and its species.

4.4.6 Recreation Value

Since the habitat of the Reed Canyon attracts many species (see Appendix C), it also has recreation value for wildlife watchers such as birdwatchers. Birdwatchers generate substantial economic activity including $31,686,673,000 in 2001 from retail expenditures alone (La Rouche 2003). According to a contingent valuation component of the nationwide National Survey of Fishing, Hunting, and Wildlife Recreation, the experience of wildlife watching, such as bird watching, provides a net economic benefit of $313 per person per year adjusted to 2009 dollars (La Rouche 2003). The Oregon Parks and Recreation Department determined that 20% of Portland's population participated in bird watching. (Oregon Parks and Recreation Department 2003).

There are 327 households with an average of 2.3 residents within mile of the Reed Canyon (U.S. Census Bureau 2000). If 20% of these 752 residents participated in bird watching each year in the Reed Canyon, there would be a net economic benefit of $47,075 per year. It is important to emphasize that the suitability of Reed Canyon as a bird-watching destination occurred mainly after the influx of new bird species due to the restoration of the Reed Canyon. The Reed Canyon habitat provides additional benefit to birdwatchers at other locations, because it is utilized as temporary habitat for migratory birds whose primary habitat is located elsewhere. For more information on the recreation value of the Reed Canyon, see Section 4.5.

4.4.7 Further Research

Further research towards an economic valuation of the Reed Canyon habitat would entail more extensive species inventorying and estimations of the numbers of individuals of these species. A contingent valuation of the Reed Canyon habitat and the species in the Reed Canyon could also be helpful. Overall, economic valuations that better isolate the value of ecosystem services, whether that be aesthetics, recreational quality, air/water quality, or the existence value of carbon sequestration or wildlife habitat, etc., would be helpful in ascribing economic benefit directly to habitat.

4.5 Recreation and Aesthetics in the Reed College Canyon

When considering the current and future importance of natural areas for recreation and aesthetic benefits, it is essential to note that in 2008 half of the world's population (3.3 billion people) lived in urban areas and that this number is expected to grow to 5 billion by 2030 (United Nations 2007). Many cities, like Portland, are experiencing an increase in population and density. Metro, the regional government, has projected the population in 2030 will be between 2.9 and 3.2 million people up from only 1.9 million in 2000 (Metro 2009). The region's Urban Growth Boundary (UGB) will encourage the Portland area's growth in population to result in increased urban density.

With increased density, the planning of city spaces will continue to be a contentious issue. A meta-analysis by Brander and Koetse (2007) determined that areas with higher population density attached more value to open space. This was true across studies using the contingent valuation and hedonic methods. Put simply, when determining the value of natural areas "scarcity and crowdedness matter." This study also found that residents value "urban parks" more highly than other types of open space. At the same time, from an economic perspective, protected open and natural spaces will have a higher opportunity cost as demand for urban development increases. Understanding the benefits of urban open space is an essential component of rational policy making. "Public decision-making requires information on the value of services provided by open spaces in order to make informed trade-offs against the (opportunity) costs of preservation" (Brander and Koetse 2007, 1).

This section aims to highlight the recreation and aesthetic benefits of the Reed Canyon. It is one of Portland's urban open spaces and understanding the value it creates will be helpful for future canyon management and regional policy decisions for similar open spaces. The human (recreational and aesthetic) benefit derived from the canyon depends on its ecological quality and human accessibility. Therefore, the two major factors that will be considered in this section are the quality of recreation/aesthetic opportunities and the degree of accessibility/utilization of the Reed Canyon because these are the most important determinants of the value of recreation and aesthetic benefits.

4.5.1 Atypical: The Reed Canyon Recreational Area

Measuring the recreational benefits of the Reed Canyon requires considering the entire ecosystem because the quality of other ecosystem services (wildlife habitat, improved air quality etc.) affects the level of recreation and aesthetic benefits. As discussed in Sections 2 and 4, the Reed Canyon restoration project has transformed this area in to an ecologically important natural area. The canyon is a "park," but it is atypical even for Portland because of the large percentage of tree canopy cover and natural/native habitat. From an aerial perspective, the Reed Canyon has 59.5% tree canopy coverage. This percentage includes water surface as non-canopy area. From Figure 5 it is clear that the area of the canyon that is not water surface is almost 100% tree canopy.

A report on urban forest canopy cover (Portland Parks and Recreation 2007) shows that many of the urban/natural parks in Portland have less than 50% tree canopy coverage. Somewhat surprisingly, many of Portland's parks offer almost no canopy cover. However, many urban parks are sports-oriented, with large areas for parking, courts, and recreation fields. Reed therefore allows for an atypical recreational/aesthetic experience and future improvements to the canyon should continue to contribute to its unique recreational and aesthetic benefits.

4.5.2 Benefits

Aesthetics and Livability

Several important psychological studies have documented the emotional benefit of a natural view (Kaplan 1995). Similarly, Ulrich (1981) monitored the effect of natural vs. urban aesthetics on human stress levels. Subjects under stress were shown slides that depicted either water scenes, vegetated natural areas or urban environments. Physiological indications of stress decreased when subjects were exposed to the natural aesthetics. When subjects were exposed to a highly urban environment they experienced the same stress levels or stress even increased (Ulrich 1981). Another study measured the effect of a "green view" on hospital patients' ability to recover from gallbladder surgery. The findings showed that patients with rooms facing a park had 10% faster recovery and needed 50% less strong pain-relieving medication compared to patients in rooms facing a building wall (Ulrich 1984).

In essence, there has been an attempt to prove that the "subjective beauty" of green views is actually a psychological fact. If these studies are accurate, the preservation of urban open spaces is necessary for the mental health of a city's population and urban planners should consider these areas to be a psychologically necessary benefit. More research should be done to determine the most efficient ways to incorporate green aesthetic benefits. For example, city planners may need to consider whether park revegetation or street trees have a greater impact on the mental health of a city's residents.

Livability: Recreation

The relationship between access to natural areas and health effects have not been firmly proven for all age groups. A study done in the United Kingdom found no correlation between open space accessibility and the level of physical activity for middle aged adults (Hillsdon et al. 2006). Adult demand for recreation, in this case, was found to be fairly inelastic with one explanation focusing on the high opportunity cost of working adults' time. Studies do indicate that younger demographics are more affected by a change in neighborhood access to recreation space. Bell et al. (2008) find a correlation between the greenness of a neighborhood and the health of surrounding child populations.

Of course, outdoor recreation is desired for more than its psychological and health benefits. There are, indeed, innumerable other benefits. For example, outdoor recreation is certainly enjoyed culturally as a hobby and pastime. There is a sense in which we all acknowledge that running on a treadmill while staring at a picture of the forest will never be a perfect substitute for jogging on an outdoor trail. It is almost impossible to denote every difference between the above examples, but there is an observable and quantifiable willingness to pay for outdoor experiences as the studies below demonstrate.

Quantifying Benefits

There are several key factors that affect the level of recreation and aesthetic benefits that are possible/probable in a given open space. Most of these are indicators are what can be termed "accessibility infrastructure." These factors include:

• proximity to roads
• surrounding population density
• unrestrictive property rights
• trail systems (interactive capacity) and views

The valuation of recreation and aesthetic benefits must rely on indirect measurement because there are no market prices for these ecosystem goods and services. These benefits can be quantified using the nonmarket valuation techniques introduced earlier in this report: travel cost, hedonics, and the contingent valuation method.

Property values in the area surrounding an open space may reflect many of the perceived benefits of recreation and aesthetics. Lutzenhiser and Netusil (2001) use the hedonic price method to estimate the effect of open spaces on home sale price. This study finds that homes within 1,500 ft of a natural area park (where more than 50% of the park is preserved in native and/or natural vegetation) have the largest increase in sale price. "The results show that natural area parks, on average, have the largest statistically significant effect (1% level) of $10,648 in 1990 dollars, on a home's sale price holding all other factors constant" (Lutzenhiser and Netusil 2001, 296). The habitat section of this report used estimates from a hedonic study conducted by Mahan et al. (2000) that measures the increase in property values from being near a body of water.

The extent to which these estimated increases in the sale price of properties can be attributed to expected local recreation and aesthetic benefits is indeterminable from a hedonic study. It is important, however, that the Lutzenhiser and Netusil (2001) study found that the areas with the largest positive increases in sale price were near natural open spaces with large acreage. Anecdotally, the Reed College Admissions office certainly capitalizes on the aesthetic benefits of the Reed Canyon. The blue bridge, specifically, is most often used in catalogues and displays in attempts to attract prospective students.

Contingent valuation surveys have also been used to value recreation at nearby open spaces. Breffle et al. (1998) estimate a neighborhood's willingness to pay to preserve 5.5 acres of undeveloped urban land in Boulder, Colorado. Willingness to pay was measured as a function of distance and income and was found to increase with income and decrease at a decreasing rate with distance from the study site. Median household willingness to pay was $234 ($371 in 2009 dollars), while mean willingness to pay was $294 ($467 in 2009 dollars). Interestingly, within .1 miles, the mean household willingness to pay was found to be $1,197 ($1,900 in 2009 dollars).

Is that how much a student living on campus would be willing to pay to keep the Reed Canyon? How much would a professor with a view of the canyon be willing to pay? Multiplying the estimated willingness to pay of $1,900 by the number of Reed Community members, approximately 1,500 people, generates a total willingness to pay for the preservation of the Reed Canyon of $2,850,000 (in 2009 dollars). However, without designing a contingent valuation study for the campus, the extent to which the willingness to pay is completely attributable to recreation and aesthetic benefits is indeterminable. As explained in Section 2, the initial grant for the Canyon Restoration project came from an alumna who presumably cared more about the existence value of the canyon than her direct use value.

Travel Cost Approaches

Bowker et al. (2004) uses the travel cost method to determine the consumer surplus derived from the Washington & Old Dominion Trail (W&OD), a popular recreation trail in Northern Virginia. The study includes people who visit the site as their primary destination with 84% of respondents stating that their primary use of the site is for recreation and fitness. Consumer surplus estimates for a group were $18.13 while individual consumer surplus estimates were $13.63.

This study is particularly relevant to individuals interested in the valuation of recreation benefits of open spaces in Portland because ECONorthwest (East Butttes Report 2009) relies on the estimates from the Bowker et al. study for its valuation of the Springwater Corridor Trail. The East Buttes report uses the individual consumer surplus estimates from the W&OD study to calculate that the annual consumer surplus associated with biking on the Spring Water Corridor Trail within the study area as $2 million dollars.

The Reed Canyon has an extensive and well maintained trail system. It is not a regional multi-use path like the Springwater Corridor Trail, but it does provide significant recreational opportunities in a densely vegetated natural area. No contingent valuation studies were located that focused on these small-scale recreational paths. The ECONorthwest report (2009) similarly fails to value smaller trail systems, but perhaps the value derived from the Bowler et al. (2004) report could be used as the benefit to bikers on the Reed bridges and pedestrians on both the bridges and the trails. Further study would need to be done to determine if this value is appropriate--the average travel distance in the Bowker et al. study was far greater (20 miles) than the 1.5 mile travel distance assumed for local parks in Portland. However, Reed is similar to the W&OD and Springwater Corridor Trail because the Canyon is an open access resource. Unlike other private schools, Reed is an open campus, with no physical barriers to entry, and does not enforce its property rights over the canyon.

Possible Spillovers for distant recreators

In the future, the Reed Canyon will have a motion sensitive underwater camera in the fish ladder. This camera will help identify the number of fish that enter and leave the canyon using the fish ladder. This information will help determine the benefit of the Reed Canyon as an estuary for anadromous populations. Quantifying and valuing the existence value of a restored species, as well as the benefit to recreational and commercial fishermen, is an important area for further research in coming years. A great deal of the restoration project has focused on improving fish habitat. An estimate of benefits from these investments will be an area for study once more information is available.

Index | Executive Summary | 1 | 2 | 3 | 4 | References | Appendix A | Appendix B | Appendix C | Appendix D | Appendix E