Cities and the Environment (CATE) Cities and the Environment (CATE) Urban Ecological Stewardship: Understanding the Structure, Urban Ecological Stewardship: Understanding the Structure, Function and Network of Community-based Urban Land Function and Network of Community-based Urban Land Management Management

Urban environmental stewardship activities are on the rise in cities throughout the Northeast. Groups participating in stewardship activities range in age, size, and geography and represent an increasingly complex and dynamic arrangement of civil society, government and business sectors. To better understand the structure, function and network of these community-based urban land managers, an assessment was conducted in 2004 by the research subcommittee of the Urban Ecology Collaborative. The goal of the assessment was to better understand the role of stewardship organizations engaged in urban ecology initiatives in selected major cities in the Northeastern U.S.: Boston, New Haven, New York City, Pittsburgh, Baltimore, and Washington, D.C. A total of 135 active organizations participated in this assessment. Findings include the discovery of a dynamic social network operating within cities, and a reserve of social capital and expertise that could be better utilized. Although often not the primary land owner, stewardship groups take an increasingly significant responsibility for a wide range of land use types including street and riparian corridors, vacant lots, public parks and gardens, green roofs, etc. Responsibilities include the delivery of public programs as well as daily maintenance and fundraising support. While most of the environmental stewardship organizations operate on staffs of zero or fewer than ten, with small cohorts of community volunteers, there is a significant difference in the total amount of program funding. Nearly all respondents agree that committed resources are scarce and insufficient with stewards relying upon and potentially competing for individual donations, local foundations, and municipal support. This makes it a challenge for the groups to grow beyond their current capacity and to develop long-term programs critical to resource management and education. It also fragments groups, making it difficult for planners and property owners to work in partnership with them. The organizational networks are self-contained and do not include business or even legal groups, which may point to a gap between stewardship and environmental justice organizations. Abstract Urban environmental stewardship activities are on the rise in cities throughout the Northeast. Groups participating in stewardship activities range in age, size, and geography and represent an increasingly complex and dynamic arrangement of civil society, government and business sectors. To better understand the structure, function and network of these community-based urban land managers, an assessment was conducted in 2004 by the research subcommittee of the Urban Ecology Collaborative. The goal of the assessment was to better understand the role of stewardship organizations engaged in urban ecology initiatives in selected major cities in the Northeastern U.S.: Boston, New Haven, New York City, Pittsburgh, Baltimore, and Washington, D.C. A total of 135 active organizations participated in this assessment. Findings include the discovery of a dynamic social network operating within cities, and a reserve of social capital and expertise that could be better utilized. Although often not the primary land owner, stewardship groups take an increasingly significant responsibility for a wide range of land use types including street and riparian corridors, vacant lots, public parks and gardens, green roofs, etc. Responsibilities include the delivery of public programs as well as daily maintenance and fundraising support. While most of the environmental stewardship organizations operate on staffs of zero or fewer than ten, with small cohorts of community volunteers, there is a significant difference in the total amount of program funding. Nearly all respondents agree that committed resources are scarce and insufficient with stewards relying upon and potentially competing for individual donations, local foundations, and municipal support. This makes it a challenge for the groups to grow beyond their current capacity and to develop long-term programs critical to resource management and education. It also fragments groups, making it difficult for planners and property owners to work in partnership with them. The organizational networks are self-contained and do not include business or even legal groups, which may point to a gap between stewardship and environmental justice organizations.


Figure 1. Multi-scaled model of Socio-Organizational Ties
Source: Grove et al. 2002 In particular, this paper hopes to shed light on active organizations that are dedicated to using ecological strategies to create, restore, reveal or maintain any part of the urban landscape in six large urban areas in the Northeastern U.S.: Boston, New Haven, New York City, Pittsburgh, Baltimore, and Washington, D.C. These organizations include informal community groups, formal nonprofits, as well as municipal, state, and federal partners. While public, private and civil society entities will be discussed in this paper, each will be distinguished in the overall analysis. In order to support groups' stewardship efforts and improve their effectiveness as agents, a better understanding of their basic functioning as individual organizations and as a network is required. Using data from the Urban Ecology Collaborative (UEC) assessment, this paper examines how these organizations interact with critical biophysical resources (e.g. land, water, soil, air) and social institutions (e.g. government, commerce, education, nonprofits) through the flow of materials, energy and information (e.g. human capital, funding, partnerships, science). The findings challenge three recent debates in urbanism, which claim that participation in civic associations is declining (Putnam 2000); that the urban environmental movement is place-based and fragmented (Harvey 1999) and that there is a waning public interest in issues pertaining to environmental quality (Greenberg 2005).

Human Ecosystem Framework: A context for understanding the challenges of urban stewardship
Urban areas are ecosystems with interdependent resources and flows that are no less complex than wilderness or forested ecosystems (Burch and Grove 1993;Grove and Burch 1997;Pickett et al. 1997;Redman 1999). One might argue that in the urban context, the environment is nested within larger quality of life issues such as public health and well-being, economic development and social justice which are collectively driving social motivations for land based stewardship. The Human Ecosystem Approach is used as a framework to aid in revealing interactions that drive particular system at a particular point in time (see Figure 2). In this sense, the city-as-ecosystem is more than just a clever metaphor. Rather, it allows us to have a holistic understanding of the relationships between individuals, groups, organizations, culture, and norms-not just as sociological concerns, but as key contributors to the biophysical functioning of our cities. While one could choose any number of aspects from this Human Ecosystem Framework (HEF) for study, this paper considers the role of organizations as a critical "cultural resource, for they provide the structural flexibility needed to create and sustain human social systems" (Machlis et al. 1997). Stewardship groups, in particular, are chosen because they are literally agents that interact with both the biophysical resources and the social system of the human ecosystem (see Figure 2). From a practical or managerial standpoint, determining how best to manage the urban ecosystem requires a consideration of these human organizations as vital parts of the urban ecosystem.

Figure 2. Human Ecosystem Framework
Source: Machlis, Force, and Burch 1997 Application of the HEF model to the analysis of the UEC assessment's stewardship organizations frames a number of key challenges that must be considered, which are described below.

Biophysical challenges
The largest percentage of the world's population living in urban areas was recorded at the turn of the 20 th Century. It is no longer a question of whether urbanization affects ecosystem functions but rather, to what degree they do so and how positive externalities can be created within this highly manipulated system (Millennium Assessment 2005). At the metropolitan level, urban growth affects the heterogeneity of the landscape through landcover change and affects the spread of disturbance through invasive species, to name a few critical and documented examples (Alberti 2005). Within cities themselves, there is a range of open space areas from protected wildlife habitats, to contaminated and fallow sites, to highly managed and used parks. Habitats are fragmented, both discontinuous and small in size, yet species diversity can still be quite high in these disturbed landscapes (Niemela 1999). Basic urban infrastructure has major impacts on the environment. Landscape and social ecologists are still on the frontiers of knowledge regarding the management needs of highly urbanized areas. Yet, management and use of the landscape both by public authorities and the private sector continues regardless, despite a lack of understanding of how "best" to support certain ecosystem services. Often urban sites are not managed for biophysical function at all, instead serving social functions as recreation sites and as promoters of neighborhood efficacy (Sampson and Raudenbush 1999). It is in the interest of environmental planners to ascertain where, how and when community-based management of street trees, planter beds, lots, greenways, parks, and forests is occurring. And in urban areas, one simply cannot divorce the sites from their property jurisdiction, regulations, or users.

Social and organizational challenges
Stepping back from concerns about the ecosystem, there is a need for discourses on social capital, resource management and civic environmentalism to engage with the issue of urban stewardship, for it lies at a nexus of these issues. The debate over the proclaimed "death of associations" and accompanying dearth of social capital in American cities cites low membership in traditional civic and social groups like the American Legion, PTA and sports leagues (Skocpol and Fiorina 1999;Putnam 2000). In terms of the HEF model, Putnam is arguing that there is a decline in the (socioeconomic) resource of social capital as a function of where our society is in the current macro social cycles of participation and volunteerism, as influenced by media and technology like the television. While Putnam's hypotheses and methodology has been challenged, his contribution to the public perception of local involvement is great (Edwards and Foley 1998). To this critique, this paper adds another argument. A new class of ecologically-minded nonprofit and community based groups is emerging in urban areas as 69% of the civil society groups surveyed in the UEC assessment were formed in 1980 or later; and 55% of the civil society groups consider their service areas to be at the city or sub-city level. While Skocpol emphasizes the change rather than decline of civic environmental associations in the 1970s, the focus remains on groups organized nationally for direct political purpose (Skocpol and Fiorina 1999). These national organizations have been the basis of environmental organization research and typically have small constitutions at the local level. Both Putnam and Skocpol's work differs from the UEC research, which suggests that vital social organizations emerge and expand from local, place-based and laterally networked issues. At the same time, the UEC findings hint that environmental motivations are nested within larger quality of life issues.
Similarly, activists and scholars alike have proclaimed that we are experiencing "the death of environmentalism", citing the institutionalization of environmental non-profits, fragmentation and their inability to achieve necessary, radical environmental change (Harvey 1999;Shellenberger and Nordhaus 2004). A version of this argument read through the HEF model is: the environmental movement's current cultural resources are inadequate (or, misappropriated) to achieve its goals, given the existing social institutions (government, business) and the social order (power, hierarchy, norms). Authors focusing on national organizations and surveys are typically discussing issues at a particular scale, such as international climate change or environmental quality (Fisher 2004;Fisher and Green 2004;Greenberg 2005). Criticism therefore focuses on policy-oriented and broad membership organizations, which wholly ignores that the rhetoric of "death of environmentalism" is not relevant to community-based stewardship groups that are actively integrating biophysical and social goals. Evidence of this emerges in this assessment as groups straddle the divide between environmental protection and community development. Based on the coding of open-ended reporting of missions and major programs, these groups focus on improvement of environmental quality (22.5%), community development (39.2%), and environmental education (38.3%).

Collaboration challenges
Some of the most visible efforts at collaborative natural resource management occur in high profile land use conflicts in the Western United States. Many forest, rangeland, and coastal managers attempt to achieve stakeholder-inclusive, ecosystem scale management (Weber 2000;Wondolleck and Yaffee 2000;McCreary 2001). However, recent studies have shown that similar patterns of nonoppositional strategies are emerging within the urban frame (Sirianni and Friedland 2001). This suggests that while there may be a wide range of urban environmental actors using multiple strategies, there can be cohesive management and policy-making, given the time and space to negotiate. While partnership strategies and coalitions certainly exist, the concentrated problems-particularly in low income urban communities-of water quality, air quality, soil quality, availability and distribution of open space, and toxics far outpace the political power or organizational capacity of any single group to add them adequately (Bullard 1990). As such, there remains a great deal of work to be done in the coordination of urban ecosystem management. This is not to suggest that all management in cities should be centralized, but these findings suggest the need to recognize and harness the degree of diversity, autonomy and effectiveness among public and private sector stewardship regimes. Any attempt to understand who these groups are, why they are involved with caring for the urban landscape, and what can be done to help them work more effectively in light of the many challenges can increase the likelihood of coordinated urban ecosystem management.

Methods
The assessment was conducted in 2004 by the research subcommittee of the UEC, with supporting funding from the USDA Forest Service. The goal of the assessment was to determine the status of organizations and community-based urban stewardship initiatives operating in selected major cities in the Northeastern U.S. Specifically, it intended to: • "Discover the gaps between biophysical and social resources, organizations, and programs; • Highlight specific stewardship opportunities, priorities and resources in each major city; • Examine the current capacity of organizations to use urban and community forestry activities in the improvement of the physical environment and quality of life issues common to large urban areas; • Determine strategies for the exchange of urban and community forestry tools and techniques." (UEC 2004) There was some slight variation by city in terms of methodology; as the established process was that each city would generate (or use existing) lists of organizations that are currently engaged in urban ecology initiatives. These initiatives could range from tree planting, to open space design, to environmental education, with the common criterion being that the groups must be actively supporting or caring for a particular piece of the urban landscape. From these lists, a sample of organizations was selected for study, stratified by management type, which consisted of: non-profit, federal, state, and local government, for-profit, community-based groups and individuals (usually independent environmental contractors). The outreach strategy to those organizations varied by city: New Haven convened a meeting and distributed surveys in person; Pittsburgh, Washington, D.C., and Boston relied upon emailing and phone outreach. The New York City methodology is described here in greater detail, as it may be most useful as a model for expanding research on a more expansive sampling framework.

New York City Sampling Methodology
The sample of 100 organizations and informal groups for the New York City assessment was drawn from a population of 2,027 groups compiled from the combined stewardship databases, participant rosters, and organizations tracked by the largest urban ecology intermediary groups in the city and in some cases region. This chart represents the groups used for this assessment who were tracking explicit stewardship information. After the databases were assembled, they were merged along all common characteristics and duplicate listings were eliminated. Then two fields "scale" (region; city; borough; neighborhood/block) and "management type" (public agency federal; public agency state; public agency city; for-profit; nonprofit; community group) were ascertained for each group, based on information in the existing databases and input from the staff of organizations maintaining the databases. Some unknowns remained for which management type and scale could not be determined, and these were excluded. The fields were then used to stratify the sample. A four percent sample was taken from all community groups and nonprofits. Because of the limited number of organizations, with many of the natural resource groups being known entities, federal, state, and local agencies were purposively over sampled in the assessment. Forprofit groups were randomly sampled. The sampling is summarized in Table 1. Returned (n=34) 2 4 7 0 12 6 (+2 indv) The 100 selected groups were sent the survey by mail, with a follow-up phone call to answer any remaining questions, followed by a postcard reminder to complete the survey and one final round of calls, all conducted in the summer of 2004. 1 Of the surveyed organizations, 34 completed the survey, eight said the survey was not applicable to their group (because they were actually not engaged in stewardship), and one refused to participate. Clearly, community groups had the lowest response rate, which is not surprising given the challenge of reaching these informal and sometimes temporary groups. It is possible that a number of non-responses were due to groups that no longer exist, given the age of some of the stewardship databases comprising the parent population.
The six cities, combined to survey 135 organizations (34 in New York City, 19 in Baltimore, nine in Boston, 34 in Washington, D.C., 20 in New Haven, and 19 in Pittsburgh), is not comprehensive enough to make any sort of quantitative cross-city comparisons. Because the sample was not drawn randomly, it does not enable the use of predictive statistics (e.g. regressions or means testing) on this dataset. Although this limits the analysis and makes clear the need for further study, the intent of this project was to characterize the basic form and function of an under-studied set of civil society and public actors. Thus, frequencies and percentages will be used to report the overall trends in the data.

Findings and Discussion Organizational Demographics: Management Type and Age of Organization
Organizational demographics are some of the fundamental attributes of these groups, including management type and age. Because the goal of the UEC assessment was to understand local environmental stewardship, rather than solely the role of civil society, we see that there is a mix of organization types included in the results (see Figure 3). However, despite an attempt to be inclusive of government actors, it is evident that civil society actors outnumber them, with nonprofits, community groups, and individuals comprising 73% of the sample. This is likely a reflection of the fact that government agencies are larger and more centralized, while nonprofits and community groups are more local and place-based. So, for example, while there is one New York City Park Department, there are over 600 community gardens and more than 1000 active park-based stewardship groups in New York City. 2 The level of civil society involvement is significant from a managerial standpoint, since it means that resource managers wishing to make changes on a landscape or to improve ecological functioning in a watershed will need to do so in concert with informal and nonprofit groups. However, this does not suggest the absence of public sector involvement as suggested in the case of advocating for citizen monitoring "bucket brigades" (O'Rourke and Gregg 2003). Instead, it may suggest the need to reconsider models for shared stewardship or 'governance' of urban land (Durant 2004).
In fact, one could perhaps make the argument that the hard boundaries of public entities and civil society actors begin to blur at the local level. There are numerous examples of intermediaries: Partnerships for Parks is a public-private entity that is a combination of the New York City Parks Department and the City Parks Foundation, dedicated to supporting community groups in their engagement with parks; GreenThumb is a federal Community Development Block Grant (CDBG) supported program of the New York City Parks Department that offers resources, materials, and technical assistance directly to informal community gardening groups; and the Harbor Estuary Program is a National Estuary Program authorized by the EPA that includes participants "from local, state, and federal environmental agencies, scientists, citizens, business interests, environmentalists, and others" (Program 2002). These intermediaries, organized around particular site types, seem to have a more prominent presence in New York City than the other cities studied, which is a function of the size and complexity of the stewardship network. These organizations differ from the majority of the small nonprofits and groups included in this survey that directly carry out volunteer stewardship. Their primary function is to maintain flows of material, information, and resources. They bear some resemblance to intermediaries that work in other areas of the urban environment, such as large CDCs or nonprofit coalitions that coordinate citywide brownfield inventories, such as the Cleveland Neighborhood Development Coalition (Brachman 2003). Distinctly missing from this assessment is the business community. This is due to both to the nature of the populations from which the samples were drawn and the criterion applied for inclusion in the survey. The New York City parent population (the combined databases of the environmental intermediaries) illustrates the first issue, with just 32 for-profit entities out of the total 2,004 organizations, the business sector is simply not in this stewardship network as we sampled. 3 Second, the baseline criterion applied was that each respondent had to be able to answer the question on site type, to identify a portion of the physical landscape that they manage. This is not to say, however, that the forprofit sector is not involved in the local environment; it is simply not involved in the stewardship function of public lands in the same way as non-profit groups. Rondinelli and London (2003) describe firm-NGO relationships of differing intensities, with the most common being the "arm's length" relationship, which includes corporate donations and employee volunteerism. The survey shows that 18.5% of respondents listed corporate donations as one of their top three sources of funding, the third highest ranked funding source overall. Also, the involvement of corporate volunteers in large-scale one time park clean-up days and other events is quite common. Sustained environmental stewardship, however, is not generally a long-term function filled by these firms unless representatives function in a dual capacity of citizen and business leaders.

Figure 3: Type of Environmental Management
Groups are defined by more than whether they are public or private entities. Organizational culture, which can be understood in a limited way by analyzing missions and major programs, fundamentally contributes to the way a group "does business. " Wilson notes (1989), "Every organization has a culture, that is, a persistent patterned way of thinking about the central tasks of and human relationships within an organization. Culture is to an organization what personality is to an individual. Like human culture generally, it is passed on from one generation to the next. It changes slowly, if at all" (91).
Based on the coding of open-ended reporting of missions and major programs, stewardship groups focus on improvement of environmental quality (22.5%), community development (39.2%), and environmental education (38.3%), showing that the groups have environmental and community values. Situating urban ecological stewardship within the chronology of the environmental movement provides an understanding of how these groups map onto waves of protectionism, conservationism, populist environmental advocacy, and environmental justice (see Figure 4). Generally, urban stewardship organizations are young, with over 90% founded since 1970. This is not surprising, given the rise in urban 'self-help' social movements during the 1960s and 1970s.
Reviewing the data respondent-by-respondent, organizations founded prior to 1960 included government entities like the National Parks Service and the Metropolitan Council of Governments. Comparing civil society and stewardship organizations overall shows that patterns are similar, reflecting the increase across all sectors in environmentalism. The mean founding date of all stewardship groups is late 1981. There is a marked rise in stewardship groups founded since 2000, which may continue to rise given that newer organizations might have been systematically under sampled from a parent population based on databases that are in some cases up to three years old. Further research on these newer organizations is needed.

Organizational Resources: Staff, Budget, Funding Source, and Information
An examination of organizational resources is useful for two reasons: 1) it helps to evaluate one dimension of the capacity of these stewardship organizations to pursue their missions, again framed by the HEF concept of critical resources, and 2) it reveals one layer in the stewardship network, the relationship between funders and recipients, and a capital flow in the HEF model. These resources are examined through questions on staff, budget, funding sources, and information.
Staff size is an important measure of the level of development and formality of an organization, and looking at staff size and community volunteer base together can give a sense of how an organization accomplishes its work and at what scale (see Figure 5). The stewardship groups are generally small in size, with 63.8% of all organizations and 80.7% of civil society organizations having fewer than ten full time staff. The number of organizations with zero full time staff is also notable, with many of the groups operating entirely on a volunteer basis. Groups with zero full time staff were not just the volunteer community groups as one might expect, but were evenly divided between formal nonprofits and informal groups.
Another surprising finding was the large number of groups with zero or less than ten community volunteers, as stewardship is popularly associated with high levels of volunteerism. There were seven civil society groups that reported having both zero full time staff and zero community volunteers, relying upon part time staff, part time volunteer staff, consultants, and contractors. These all-volunteer groups serve the community informally by creating public green space and beautifying neighborhoods, but they count members as the only participants in their programs rather than users of the site. A count of the latter would reveal broader impact more clearly. Fisher and Green (2004) argue that staff capacity (among other endogenous resources) can be a barrier leading to disenfranchisement of civil society organizations and developing countries from international sustainability negotiations and politics. Particularly in large metropolitan areas, local political decisions also require time, resources, savvy, and lobbying, which should limit the ability of stewardship groups to participate. While some stewardship-only groups may not be interested in local politics a priori, they can become engaged when the sites that they manage are threatened, as was true in the 1990s during the closing, auction, bulldozing, and development of a number of community gardens in New York City (von Hassell 2002). In that case, full time staff was not the limiting factor, as these groups tended to rely upon volunteers working through a community organizing process and building coalitions with likeminded garden groups, using outsider tactics like protest and street theatre. In parallel, larger nonprofits like Trust for Public Land used insider tactics, including discussions with the city and the Attorney General and the buying up of auctioned garden sites. Community organizing around threatened gardens is beyond the scope of this paper, it is raised as one example of the way in which crises can politicize even previously non-political stewardship groups (a 'triggering event,' described in (Carmin and Hicks 2002)), at which point the interaction between resources and political participation becomes even more salient.
Budget can be considered one of a group's most fundamental resources (see Figure 6). Budgetalong with volunteer staff and in kind donations-entirely determines the level of possible staffing and on the ground programs. Over 16% of the civil society organizations function with a budget of under $1,000/year, indicating a large, grassroots, under-resourced portion of the network. In contrast, just one organization categorized as a local public agency (a public school environmental group), had a budget of under $1,000/year. These small budget groups include the site-specific stewardship groups, such as community garden groups, school garden groups, neighborhood park "friends of" groups, and environmental "clubs". The network is not entirely without financial resources, however, as over 64% of these organizations have budgets of larger than $100,000/year. The intermediate-sized nonprofit organizations with budgets of $100,000-$500,000 include citywide groups like the New Haven Land Trust and the Boston Toxics Action Center, as well as larger environmental education groups. Those with resources over $1 million include high profile citywide friends-of parks groups like the Pittsburgh Parks Conservancy, as well as nationally significant nonprofits (many of which were located in Washington, D.C.) like American Forests and the America the Beautiful Fund. Seventy-six percent of public agencies have budgets of over $100,000. The ten organizations with budgets over $5 million include the Parks and Recreation departments of these major cities, as well as some county agencies with responsibility for the metro area (e.g. County of Allegheny Department of Parks) and federal groups responsible for the National Mall in Washington, D.C. The diversity of groups even within the mantle of urban ecology stewardship helps to explain the wide range of budgets that are observed. Figure 6 shows the contrast between the budgets of civil society and government groups. Despite the available resources, 49% of groups in the survey identified "lack of funds" as the top barrier to the successful pursuit of their organizational missions (see Figure 7). The second highest barrier was "lack of staff" at 23%, which is at least partially a function of lack of funds. These responses were generated in response to an open question rather than picking a response from a list. Additional barriers include (in rank order): lack of time, bureaucratic barriers, lack of cooperation, and lack of political power. Moreover, respondents were asked if they agreed with the statement "this budget adequately serves my group's needs." Fifty-three percent of respondents disagreed (and 27% were neutral). Therefore, we can conclude that the current allocation of resources is not meeting the needs of the majority of urban ecology organizations. Whether it is an issue of absolute resources or allocation is not known, but it makes the need for leveraging resources all the more important. Indeed, the potential to leverage resource and pursue joint fundraising was one of the motivators behind the formation of the multi-city collaborative (the UEC) that supported the assessment discussed here.

Lack of funds 49%
Lack of staff 23%

Lack of time 8%
Bureaucratic barriers 5% Lack of cooperation 5% Lack of political power 3% Other 7% The question on funding sources asked respondents to select their top three funding sources (unranked); figure 8 shows the percent of all respondents that included each funding source in their top three. Unsurprisingly, municipal government (32.1%), state government (22.6%), and federal government (20.8%) were the top three sources of funding for public agencies. All other sources were ranked highly by no more than 11% of public agencies. Local foundations (42.7%) and private giving/membership (32.9%) are the top two sources for civil society organizations. It would have been useful to separate membership fees from private giving. Further confounding these responses was the separation of fees/program income from giving/membership. Despite these potential wording issues in the assessment tool, it is evident that more than 50% of stewardship groups rely on the financial support of individuals (through fees and donations) as one of their primary funders. All government funding sources combined were selected by 41.6% of respondents as being primary funders. The insufficient budgets and small staff sizes combined with a heavy reliance upon local foundations corroborate assessment research that small stewardship nonprofits lack much-needed support for general operating expenses (Svendsen and Campbell 2005). While there is private foundation funding available to support program expenses, general operating resources are scarce, making organizational growth and sustainability a real challenge. Environmental stewardship organizations are also supported by the private choice of individuals through in-kind and volunteer support. Since they are less reliant on public funding, this contribution should be considered a "source" rather than a "sink" of human and social capital. They should be supported and used as conduits to affect environmental change, rather than ignored or reinvented, as some government-led programs tend to do (Burch and Grove 1993). The HEF model categorizes information as a critical socioeconomic resource. Since the UEC was formed in part to support better information exchange amongst stewardship groups, the survey wanted to determine how easily stewardship groups can access information and "successful models" in their field. 4 Over 72% of all organizations and all civil society organizations agreed that they could access these models. This finding was surprising given the perceived programmatic redundancies and inefficiencies that can be observed amongst small, developing nonprofits. What, then, is the role for government and private foundations interested in supporting research, networking, and information clearinghouses? It seems to suggest that these agencies and funders could be encouraged to move away from the current model of 'technology transfer' and more towards one of capacity building through 'technology exchange.' The issue is less one of availability of technical information and more one of coproduction of knowledge (Fischer 2000). In this case, stewardship organizations reported that the primary resources they provided to community were: information (54%), hands-on training (41.5%) and volunteers (37.8%); see Figure 9. These data are used by groups internally to improve programs and services (58.5%), to satisfy funders' requests (54%) and to create legitimacy and a constituency. Stewardship groups, like all organizations, have networks that connect them to other organizations and actors both vertically and horizontally. For instance, government agencies, funders, and intermediaries interact with stewardship groups by providing funding, technical assistance, information, as well as material resources (such as soil, tools, landscaping equipment, etc). The stewardship groups themselves interact horizontally with other stewards, coalitions, and advocacy nonprofits that share a common interest in urban ecology. Finally, stewardship groups interact directly with individual members, neighborhood residents, schoolchildren, and one-time and sustained volunteers. Groups were asked to describe their existing networks in both directions, in terms of audience and fellow stewardship groups. Determining which partners are considered critical to the functioning of these groups and what groups they would like to work with in the future was considered critical for network analysis.
Since the assessment was implemented in two rounds, with Boston and New Haven conducting outreach in late winter/early spring 2004 and the remaining cities conducting outreach in summer 2004, two different versions of one question were asked. For the first set, the question asked "what is the target audience of your programming?" and respondents were asked to choose all groups that apply. Participants conducting the survey reported confusion over the wording in this question, perhaps because stewardship groups do not consider partners or participants "audiences". Overall, civil society organizations selected: individuals (72.7%), community groups (63.6%), and public agencies (59%) as their top three audiences. The question's intent was reconsidered and its' phrasing reconfigured to ask "with what type of organizations does your group most often work?" Here the distribution of civil society organizations responses shifted away from individuals to other community groups (72%), schools (62.3%), and nonprofits (58.7%) as the top three selected. For public agencies, the top selected partners were schools (61.8%), community groups (61.8%), and nonprofits (61.8%) By operationalizing the question of partnership in multiple ways, the assessment sought to get a better understanding of relatively who works with whom. Respondents were asked to rank other stewardship groups by the frequency with which they partner. The distribution of partners looked very similar between government respondents and civil society respondents. Both sets of groups ranked government groups as the stewardship group with which they most frequently partnered, (consistently/year round for 54% of civil society and 86% of public entities). Both groups tended to work a great deal with nonprofits, though civil society organizations had more interaction with individuals, and both worked infrequently with business groups. The distribution for just the civil society organizations is shown in figure 10. With the exception of the business sector, the majority of respondents reported partnering with all other stewardship groups frequently or consistently. This result could potentially be a function of the survey design and implementation. If anything, though, this question simply reinforces the lack of involvement on the for-profit sector in this capacity. It also reiterates the fact that government agencies (including municipal, state, and federal parks department as well as less obvious groups like water-based or agricultural agencies) are important stewards. The assessment asked respondents to identify and rank up to six organizations or individuals that were "critical to their work" currently. They were also asked to rank the top six individuals or groups with whom they would like to work with, in the future but are not currently. These two questions, taken together, move towards an understanding of the beginnings of a network-though not as loosely defined as the community of common values that Batterbury (2003) describes.
Comparing these responses side-by-side allows us to understand where this network currently stands and the direction in which it may evolve. Current organizations mirrored the responses to the stewardship partner questions, with city agencies and non-profits being the highest ranked responses. Of the non-profits listed, 19 were specifically environmental nonprofits, three were "cultural" nonprofits, and one was a healthcare nonprofit. Of the city agencies, 15 were specifically referring to parks departments of the various cities, which continue to play critical roles in urban environmental stewardship. Other named agencies include health, environmental services, planning, and urban forestry departments. Finally, of the 12 organizations listing state agencies as key partners, 10 of these were state natural resource departments.
For the future, respondents ranked highest a variety of environmental groups, government agencies, and research groups. The high ranking of research as a priority area is surprising, and perhaps suggests the potential for community based or participatory research that takes advantage of the existing close relationship between government agencies and local stewards. Also notable is the rather high rank of business groups; it seems that the stewardship groups are aware of this gap in their network. Both grouped lists are shown in Table 2. Beyond knowing who is in the network or who groups would like to have in the network, the assessment sought to find out what particular networking strategies organizations used to connect with other groups. Here, there was little variation between civil society and government actors. The most commonly used strategies by civil society organizations were: attending local community meetings (76.9%), generating press (71.4%), and participating in regional coalition group (67%). The high response for regional coalition was surprising, given a common perception of a lack of regional information-sharing and formal collaborative entities. Perhaps this reflects some ambiguity of the meaning of the word regional. The partners of the UEC and others are interested in using intermetropolitan coalition in order to affect change in individual cities. Other common strategies listed were attending national conferences (61.6%) and participating in citywide coalitions (57.1%). For government groups, the top three strategies were public-private partnerships (83.3%), participating in regional coalition groups (76.7%) and 73.3% said they attend local community meetings and generate press. Since public-private partnerships did not rank highly on the strategies of civil society organizations, it remains a question as to what are the groups with whom these government actors are partnering. The lowest ranked strategy in both cases was "participate in list servs", reflecting the reliance on face-to-face rather than virtual collaboration. When urban groups can physically meet, they seem to prefer that to virtual communication.

Biophysical & Social Impacts: Scale of Service, Neighborhood, Site Type, Land Jurisdiction
The final aspect from the UEC assessment that is considered here is how these groups' activities play out across the space of the urban landscape in terms of scale of service delivery and areas of stewardship work by neighborhood and site type. The HEF model includes biophysical resources as a major component of the human ecosystem. While this survey did not involve any physical land assessment or inventory of sites, it does capture where and how these groups organize on the landscape to demonstrate where the overlaps and gaps between groups are, which is a first step to establishing the link between organizations and physical resources.  A high number of groups indicated that they work across regions. While this was intended to mean metropolitan areas, upon reviewing the group's missions and self-descriptions, it may have been selected for different reasons. Many of the Washington, D.C. based groups selected "region", perhaps because they thought it better defined the District than did the term "city." Second, a number of watershed, stream, or other groups that were operating on an ecological rather than a political scale, were a selected region because of its more flexible usage. Civil society organizations comprise the strong majority of groups working at the neighborhood, block, and classroom scales, with most government agencies working city and region-wide. This pattern fits with our intuition about the civil society groups, given that most of them are small in terms of staff and resources; many groups have an intensely local focus. Why is it, then, that there is a perceived chasm between environmental interests and community development interests both on the ground in urban neighborhoods and in the academic literature? (Campbell 1996;Evans 2002). Is there a greater role for stewardship of the environment in the stabilization and development of neighborhoods? Research has documented aspects of this function, particularly in terms of open space's impact on property values and the importance of planning for active living to promote healthy communities (Harnik 2000;Frumkin 2003). But there is a need for further exploration of the links between the social act of stewardship/caring for the environment and public health, crime, and social cohesion. Findings from this assessment suggest that the stewardship motivations conflate improving the physical site, inspiring people to positive action and impacting the overall neighborhood. Figure 12 shows the social and environmental impacts that groups reported achieving. While scale explains one dimension of group influence and describes one dimension of group capacity, geographically locating stewardship "spheres of influence" is suggested as a useful tool for the ecological planner, manager, designer and community organizer. For example, as the ecological planner tries to create recreation and nature corridors such as greenways, it is necessary to know both where the potential users and maintainers of these sites are. The community organizer needs to know where the clusters of high and low stewardship activity are for the purposes of focusing her outreach efforts or coalition building, for example. Groups were asked to identify both the neighborhood in which they work as well as the physical boundaries of where the group works (down to the block and street level). Neighborhood information for the New York City groups was geocoded and made into a sample map shown in figure 13. With further refinement at the neighborhood scale, this map could be developed for long-term use by urban environmental managers. Within each city and neighborhood, there exists a diversity of site types. Respondents were asked to select from a list of 36 site types that were developed jointly by the UEC Research Sub-Committee to represent the range of sub-neighborhood site types within the Forest Opportunity Spectrum ( Raciti et al. 2006). Overall, the top ranked sites were park, watershed, protected/natural site, stream/river/canal, and waterfront. Every site type was selected by no fewer than nine respondents. The thirty-six site types can be categorized into four general categories. Designated open space, including both recreational space like playgrounds and recreation parks as well as ecological space like natural protected areas, is the most frequently stewarded site type (34.1%). Water related sites (26.8%) include the expected: streams, waterfronts, estuaries, as well as the less conventional: underground streams and sewersheds. Built environment (20.5%) includes any green space on buildings or building sites, including green rooftops and courtyards, but also vacant lots and brownfields. Neighborhood streetscape (18.6%) includes all of the sites that are not on dedicated open space or building parcels, so this includes street trees and planters, but also highway medians, public right of ways, street ends, and traffic islands. Figure 14 shows the ranking of all the site types that were selected.  The final aspect to consider related to perceived impact is the jurisdiction of the various site types. Given the distribution of site types that includes the built environment and streetscape in substantial numbers, it is clear that stewardship is not just occurring on officially designated and publicly managed open space. In total, publicly held property does comprise the majority of sites on which all stewardship groups work at 57.5% (56.6% for just civil society orgs). Municipal government is the most common landowner of these sites, followed by state, and then federal government. Public managers must heed this presence of independent stewardship groups acting on public lands; for, as many have observed the design, use and meaning of public space is constantly challenged in the modern city (Jacobs 1961;Cranz 1982;Jackson 1984;Rosenzweig and Blackmar 1992) The remainder of sites is divided almost evenly between individually owned land (15%), nonprofit owned land (15%), and business owned land (12%). Managing the city as an ecosystem would require coordinated action across parcels with different management objectives and stewardship groups. Inventorying and making publicly available information on site jurisdiction is one critical first step, even independent of further research on organizations.

Conclusion
This paper begins to describe the nature of local environmental stewardship in large metropolitan areas in the Northeastern United States. Stewards are a mix of a few, larger public agencies operating at the citywide, regional and state scales and many smaller civil society actors, both 501(c) 3 nonprofits and informal community groups operating in ecological regions, across cities, and in specific neighborhoods. This organizational diversity can be viewed as both a source of social capital, in response to Putnam, and evidence of vibrant local environmentalism, in response to Shellenberger and Nordhaus. Public interest in the quality of the environment may in fact be on the rise, in response to Greenberg, but is nested within a larger context of quality of life issues. Finally, Harvey's notion that urban environmental groups are fragmented and inefficient is unresolved. The extent to which these groups will become further fragmented within specific spheres of influence or begin to develop organizational mechanisms in which to partner is unknown at this time. There is a strong underlying assumption made by this paper that without the introduction of a perceived crisis or risk, the only way to harness the capacity of stewardship groups is through deliberate multi-scaled, capacity building networks.
The assessment discovered a dynamic social network of organizations within cities with a reserve of social capital and expertise that could be better utilized. Although not the primary land owner of the sites on which they work, stewardship groups take responsibility for a wide variety of land use types. Outputs include the delivery of public programs as well as site maintenance. Most of the groups work in collaboration with government managers but operate on staffs of zero or fewer than ten, with small cohorts of community volunteers (and potentially large numbers of 'site users'). Resources are scarce and inconsistent, making it a challenge for groups to grow beyond their current capacity to develop long-term programs critical to education and management. This creates an impression of fragmentation which may not be legitimate given that certain events have the potential to unite groups across place and between scales. Stewardship networks are rather self-contained and while the business sector and legal groups are present, they are not sufficient given the critical resources that these groups can provide. This presents a challenge both to stewardship groups themselves (in terms of their own sustainability) and to planners and land managers that attempt to work with these groups. Research partnerships and shared governance structures are two potential means by which this network could be expanded (Durant 2004).
More comprehensive research of these groups is needed to be able to ask second order-questions, like the relationship between ideologies, management type, resources, strategies, and outcomes. Further research is also needed to explore the full breadth and complexity of the stewardship network. This study is a first attempt to understand groups with some affiliation to environmental umbrella organizations, but we recognize that there is a much larger universe of civil society groups for which environmental concerns are nested within other priorities (e.g. green career groups, faith based groups, youth oriented groups). An understanding of the full stewardship network will need to be cultivated in order to support stewards' work in restoring and revitalizing urban ecosystems and human communities.

Literature Cited
Agyeman and Evans, J. and. Evans, T.. 2003 29. Our programs adequately fulfill our stated mission and goals.
30. Our programs adequately meet stated public needs.