The theory, or conceptual model, guiding the Baltimore Ecosystem Study is derived from several complementary disciplines. The sources include physical sciences, social sciences, civil infrastructure, and ecology. In order to combine relevant ideas and models from these different disciplines, BES researchers have examined analogous "middle level" theories from each of the contributing specialties. In this way, we have avoided incorporating too much detail from each topic, but also avoided addressing each area at too general a level.

The integrated theory will be tested by comparisons across different ecological, social, and physical conditions in metropolitan Baltimore. To conduct these tests, we will construct integrated data sets that include key information about the human communities, ecological processes, and physical functions of the metropolitan area. In addition, new mathematical models will be constructed to apply the theory to the functioning of metropolitan Baltimore as an ecological system.

Patch Dynamics of Watersheds

The integration begins with the watershed approach, which is fundamental to both ecology and hydrology. Watersheds can be divided into patches that have different ecological and physical structures. Flow path, flow rate, and the distribution of soil saturation level depend on the topography and the characteristics and arrangement of patch types. Water flow and quality are regulated by the characteristics and arrangement of vegetation patches in watersheds (Band et al. 1993).

Integration requires dividing watersheds into a hierarchy of smaller basins. The second hierarchy relevant to watershed function is the social-political hierarchy used in planning. It divides large political units into regions, "land use" classes, communities, and households (fig 1). While these patches in a social hierarchy differentiate one community from another based upon a certain set of social characteristics, and are the target of different scales of political and social decision making, they also contain mosaics of land cover types that have specific hydrological and ecological functions. The surface cover or land cover hierarchy can therefore be termed an ecological hierarchy, because of the functional contribution of vegetation, soil and inert surfaces to land cover classes.

Therefore, land cover patches are common to both the watershed and social hierarchies. These functional cover types can link the watershed-hydrological hierarchy with a social hierarchy at different scales (fig 1). The multiple hierarchical nature of watersheds prompts three questions. First, is the hierarchy of patch types used for rural watersheds appropriate for urban and developing regions? Second, are social ecology theories and community forestry tools and

techniques from rural areas appropriate for urban areas? Third, does a hydrological hierarchy correspond to the hierarchy of units used by decision makers (fig 1)?

Heterogeneity and Spatial Patchiness

The function of urban-rural watersheds depends on heterogeneity of the three controlling factors at different scales. Therefore, the first step in linking hydrological, ecological, and social phenomena in a watershed model is to discover patterns and processes within the three realms. Biophysical and social scientists have developed systems for identifying units in their disciplines. Such systems can measure heterogeneity and link comparable social and ecological units and scales (Grove 1996, 1997).

1. HYDROLOGICAL HETEROGENEITY. Patches within a watershed contribute variable amounts of water and nutrients to stream flow, depending upon their position in the watershed and their physical attributes. This concept is represented by the Variable Source Area (VSA) (Black 1991), which is a dynamic model that reflects the abiotic attributes of the watershed, such as seasonal fluctuations in precipitation and temperature, and physical characteristics including topography, soil properties, water table elevation and antecedent soil moisture.

Urban areas are a mosaic of patches, and therefore challenge the integration of hydrology with ecological and social regulators. Therefore, we require a nested sampling design to scale measurements to the plot through subcatchment up to full watershed , with analogous sociocultural units ranging from household through political units (fig 1).

2. ECOLOGICAL HETEROGENEITY. The second component is based upon the relationship between the structure, species composition, and arrangement of vegetation patches, and the resultant heterogeneity of the regulation of hydrologic processes. Although the watershed approach has traditionally dealt with watersheds as relatively homogeneous units, the approach can be enhanced by considering spatial heterogeneity, which is virtually universal (Turner et al. 1991).

Point and non-point sources of nutrients and toxins can alter the species composition and function of stream communities. In addition, urbanization changes habitat, with profound effects on stream biota. Riparian soils, stream substrates, channel and floodplain morphology, and water temperatures can be changed by flow and sediment regimes associated with construction and the creation of impervious surfaces and storm sewer networks. Also important are changes in the magnitude and frequency of physical disturbances (e.g. floods, droughts, channel scour) to which lotic communities are sensitive. Such effects have altered the lotic and riparian ecology of urban-rural watersheds such as the Gywnns Falls.

3. SOCIOCULTURAL HETEROGENEITY. The final component is founded upon the idea that the social characteristics of the watershed—such as the relationship between social differentiation at different scales (e.g. land use, within land use, and within communities) and the extent, distribution, structure and species composition of vegetation— are spatially heterogeneous. Human societies are characterized by social differentiation and hierarchy at different scales (Burch, & DeLuca 1984). Indeed, social differentiation is a central focus of sociology (Grusky 1994). Social scientists have used social identity (age, gender, class, caste and clan) and social hierarchies (wealth, power, status, knowledge and territory) to study how and why human societies become differentiated. Social differentiation is important for human ecological systems because it affects the allocation of critical natural, socioeconomic, and cultural resources (Burch & DeLuca 1984).

Social scientists have linked patterns of social stratification of urban areas and their biophysical conditions (Morrill 1974). Few of these researchers addressed the reciprocal relationships between the social stratification of both people and place (Choldin 1984). However, Logan and Molotch (1987) argue that social stratification has significant environmental implications. According to their framework, key social variables affecting access to power, the allocation of private and public resources, and subsequently the biophysical characteristics of wealthy residential areas include:

1. the presence of homeowners and the absence of renters or absentee landowners;
2. residents who are: a) able to migrate to more desirable and healthy areas; b) effective at community organizing; c) willing to become involved in local politics; and
3. elites who have preferential access to government control over public investment, pollution control and land use decision making.

1. are disproportionately in or next to polluted areas.
2. are unable to migrate to more desirable and healthy areas; and
3. have fewer human resources in terms of leadership, knowledge, tactical and legal skills, and communication networks to manipulate existing power structures.

While Logan and Molotch (1987) and Choldin (1984) describe the relationship between social stratification among communities within land uses and their biophysical characteristics, their analysis also provides the framework for examining the relationship between social differentiation among households within communities and their biophysical characteristics. Research at this level has been conducted in rural areas using community forestry techniques (Chambers 1985), but has been addressed only preliminarily in urban areas (Burch & Grove 1993). For this second hierarchy, three social dimensions may affect the biophysical heterogeneity of communities. These three dimensions are property regimes (Bromley 1991), social order (Burch & DeLuca 1984, Machlis et al. 1994); and access to information which could influence land management.

Integrating Hydrological, Ecological and Sociocultural Processes

To understand the hydrological, ecological, and sociocultural controls of water quantity and quality in an urban-rural watershed, three domains need to be linked. Just as a Variable Source Area analysis can account for heterogeneity in hydrology, so in human-dominated ecosystems, can social attributes be measured with a revised Social Area Analysis (Shevky & Bell 1955, Hamm 1982, Grove 1996) to measure the impact of social heterogeneity on ecological patterns and processes at different hierarchical levels. In essence, this represents a social-ecological VSA that integrates physical, biotic, and social attributes at comparable levels of analysis and provides the basis for a dynamic model to study hydrologic processes using a human ecosystem and landscape approach. The resultant integration can be called a "human ecosystem". The component parts of such a comprehensive ecosystem model are grouped into ecological, socio-cultural and economic resources, and social units that respond and in turn influence the ecological and physical processes of the metropolis.

Significance

The proposed research will test the understanding of the theoretical relationship between patterns and processes of hydrology, ecosystem structure, and social stratification of an urban-rural watershed, based upon distributed hydrology (Band et al. 1996), ecosystem patch dynamics (Pickett and White 1985), and political economy theory of place (Logan & Molotch 1987). Second, it will enhance the theory common to hydrology, ecosystem ecology, and human ecology, and apply forest ecosystem models to human-dominated areas. Thus, this research will test an approach and parameters for developing an integrated regional model for human ecological systems. Further, this research is relevant to nutrient loadings to coastal waters, and the sustainability of ecosystems in developing regions. Specifically, this research can help resolve problems of water quality that policy makers, planners and community advocates must confront to address environmental equity and the linkages between environmental restoration, urban revitalization, and regional planning in a socially and ecologically heterogeneous landscape. Finally, this project will explore whether an integrated understanding of watershed dynamics can be used in collaboration with decision makers to develop socially and ecologically based solutions to the problems decision makers articulate within a heterogeneous urban landscape.

Bibliography

Band, L.E., & I. Moore. 1995. Scale: landscape attributes and geographical information systems. Hydrol. Processes 9:401-422.
Band, L.E., P. Patterson, R. Nemani, & S. Running. 1993. Forest ecosystem processes at the watershed scale: incorporating hillslope hydrology. Forest Meterol. 63:93-126.
Band, L.E., D. Mackay, I. Creed, R. Semkin, & D. Jeffries. 1996. Ecosystem processes at the watershed scale: Sensitivity to potential climate change. Limnol. Oceanogr. 4:928-938.
Black, P.E. 1991. Watershed hydrology. Pentice Hall, Englewood Cliffs.
Bogue, D.J. 1984. Procedure for delimiting ecological community areas. Pp. 27-36 in D.J. Bogue & M.J. White, Eds. Essays in human ecology, 2nd Ed. Vol. 2. The Community and Family Study Center, Univ. Chicago, Chicago.
Bromley, D.W. 1991. Environment and economy: property rights and public policy. TJ Press, Cornwall.
Burch, W.R. 1988. Human ecology and environmental management. Pp. 145-159 in J. Agee & J. Darryll, Eds. Ecosystem management for parks and wilderness. Univ. Washington Press, Seattle.
Burch, W.R., & D. DeLuca. 1984. Measuring the social impact of natural resource policies. Univ. New Mexico Press, Albuquerque.
Burch, W.R., & J.M. Grove. 1993. People, trees, and participation on the urban frontier. Unasylva 44:19-27.
Burch, W.R., & P. Jahnige. 1995. Gwynns Falls Greenway Human Ecology Inventory. Vol. 2, Interpretative analysis. A report to the Park and People Foundation.
Chambers, R. 1985. Rural development, putting the last first. Longman Group, Essex.
Choldin, H. M. 1984. Subcommunities: neighborhoods and suburbs in ecological perspective. Pp. 237-276 in M. Micklin & H. Choldin, Eds. Sociological human ecology: contemporary issues and applications. Westview Press, Boulder.
Cummins, K.W. 1962. An evaluation of some techniques for the collection and analysis of benthic samples with special emphasis on lotic waters. Am. Midl. Nat. 67:477-504.
Firey, W. 1960. Man, mind and land: theory of resource use. The Free Press, Glencoe, Illinois.
Grove, J.M. 1996. The relationship between patterns and processes of social stratification and vegetation of an urban-rural watershed. PhD Diss. Yale Univsersity, New Haven.
---. 1997. New tools for exploring theory and methods in human ecosystem and landscape research: computer modeling, remote sensing and geographic information systems. in H. Cordell & J. Bergstrom, eds. Integrating social sciences and ecosystem management. Sagamore, Champaign.
Grusky, D.B., Ed. 1994. Social stratification: class, race, and gender in sociological perspective. Westview Press, Boulder.
Hamm, B. 1982. Social area analysis and factorial ecology: a review of substantive findings. Pp. 316-337 in A. Theodorson, Ed. Urban patterns: studies in human ecology. Pennsylvania State Univ. Press, University Park.
Logan, J.R., & H. Molotch. 1987. Urban fortunes: the political economy of place. Univ. California Press, Berkeley.
Machlis, G.E., J. Force, & S. Dalton. 1994. Monitoring social indicators for ecosystem management. Coll. Forestry, Wildlife & Range Sci., Univ. of Idaho, Tech. Paper Interior Columbia River Basis Project.
Mackay, D.S., & L. Band. 1994. Extraction and representation of watershed structure including lakes and wetlands from digital terrain and remote sensing information. Trans. Am. Geophys. Union 75:S175.
Morrill, R.L. 1974. The spatial organization of society, 2nd Ed.. Duxbury Press, Duxbuty.
Neville, L.R. 1996. Urban watershed management: the role of vegetation. PhD. Diss. SUNY, College of Environmental Science and Forestry, Syracuse.
Pickett, S.T.A., & P. White, Eds. 1985. The ecology of natural disturbance and patch dynamics. Academic Press, Orlando.
Pouyat, R.V., & M. McDonnell. 1991. Heavy metal accumulation in forest soils along an urban to rural gradient in southern NY, USA. Water, Soil Air Poll. 57-58:797-807.
Pouyat, R.V., M. McDonnell, & S. Pickett. 1995a. Soil characteristics of oak stands along an urban-rural land-use gradient. J. Environ. Qual. 24:516-526.
Pouyat, R.V., M. McDonnell, S. Pickett, P. Groffman, M. Carreiro, R. Parmelee, K. Medley, & W. Zipperer. 1995b. Carbon and nitrogen dynamics in oak stands along an urban-rural gradient. Pp. 569-587 in J. Kelly & W. McFee, Eds. Carbon forms and functions in forest soils. Soil Science Soc. Am., Madison.
Rives, N.W., & W. Serow. 1984. Introduction to applied demography: data sources and estimation techniques. Sage Univ. Press, Newbury Park, CA.
Shevky, E., & W. Bell. 1955. Social area analysis: theory and illustrative application and computational procedure. Stanford Univ. Press, Stanford.
Turner, M.G., & R. Gardner, Eds. 1991. Quantitative methods in landscape ecology: the analysis and interpretation of landscape heterogeneity. Springer-Verlag, New York.