.|  Baltimore Ecosystem Study
Urban Riparian Ecology
(Much of this text is taken from: Groffman, P.M., D.J. Bain, L.E. Band, K. T. Belt, G.S. Brush, J.M. Grove, R. V. Pouyat, I.C. Yesilonis and W. C. Zipperer. 2003. Down by the riverside: Urban riparian ecology. Frontiers in Ecology and Environment 6:315-321.)
Riparian areas (stream sides) are considered to be "hotspots" of ecological function in many landscapes. Because they occur between land and water, they form unique ecosystems that act as "buffer zones" between uplands and streams. Research on riparian ecology has been diverse, including analysis of habitat requirements for rare species of plants and animals, prevention of the movement of pollutants from uplands into streams and regulation of stream temperature and physical structure. Human: environment interactions are also pronounced in riparian zones, ranging from extreme manipulation for transportation, energy generation, and agricultural needs to a desire to maintain the ecological functions listed above.
The vast majority of riparian ecological research has been in agricultural and forested watersheds, with particular interest in the ability of riparian zones to prevent the movement of nitrate (NO3-) from agricultural uplands into coastal waters. Urban effects on riparian ecology have received relatively little attention and has emerged as an integrative topic in the BES, with studies addressing how changes in hydrology associated with urbanization have altered soil, vegetation and microbial processes, the history of human use and abuse of riparian ecosystems and the effects of stream restoration on multiple functions.
Urban Riparian History

Figure 1. Early colonial land use in the lower portion of the Gwynns Falls Watershed. The inset shows the location of the larger map in relation to Baltimore City and the watershed. The areas in gray are properties incorporated into Georgia (shown with thick line) in 1732 (years of the earlier grants are indicated on the properties). Notice the emphasis of the Georgia claim on riparian zones (cf. to the earlier grants adjacent to water.) Dotted lines show the approximate paths of the buried Gwynns Run and the Calverton Raceway, both major engineered disruptions of the riparian hydrology. Also shown on the inset are the locations (represented by stars) of the ten mills established by 1776.

Given that access to water for drinking, irrigation and transportation has a strong influence on the development of human settlements, it is not surprising that riparian areas have complex histories of human activities, with varied effects on ecosystem structure and function. In the main BES study watershed (Gwynns Falls), human interactions with the riparian zone have shifted from an early emphasis on transportation (up to early 1700s), to an extended and persistent period of industrial use (1730 - present), and finally to recent use as parks and green space (recent past). Early settlement of the Baltimore region was largely concentrated on areas immediately adjacent to navigable waters, with early recognition of the value of riparian areas. "Mill Haven", a small, strategically-named parcel near the mouth of the Falls, was granted in 1696. In 1732, "Mill Haven" was incorporated into "Georgia," part of the Baltimore Company, an iron refining operation (Figure 1). The 1732 Georgia boundaries were contorted to include unclaimed riparian areas. This early industrial activity likely caused forest clearance for charcoal and cropping throughout much of the lower Gwynns Falls, leading to relatively early cycles of upland soil erosion and riparian deposition.
Soon after formation of the Baltimore Company, mills were established along much of the Gwynns Falls. Mill worker housing was built close to the mills in or near the riparian areas. Later, as the manufacturing process became more power intensive, mill "races" were extensively engineered. This level of engineering and water diversion altered riparian hydrology in the Gwynns Falls throughout much of the 1800s. Human habitation and continued sedimentation from upland crop production likely altered the chemistry and geomorphology of the floodplains.
The Gwynns Falls continued to support industrial development well into the 1900s with convenient, inexpensive waste disposal. There was particular concern about the discharge of meat packing wastes directly into the stream. In 1881, owners of the Mt. Clare Mill (near the mouth of the stream) sued upstream meat packers because "this blood and offal, naturally and necessarily by the flow of the stream, makes its way into the appellants mill race . . . whereby the water in the race and its banks are mixed with and covered by said animal matter . . . the said matter decomposing and creating an offensive smell". The massive loading of untreated waste to the Gwynns Falls throughout the 1800s is difficult to compare with the current setting, where even non-point sources are regulated.
As the valley was intensely developed for industrial purposes, flood risks to riparian property grew. The flood of 1868 killed at least 100 people and destroyed manufacturing facilities throughout the area. Hurricanes Agnes in 1972 and David in 1979 inflicted millions of dollars of damage in the Gwynns Falls valley alone. With the advent of cheaper fossil fuel energy sources and stricter waste disposal regulations, the risk of losses to flooding began to outweigh necessary capital re-investments in riparian industrial facilities. Many former establishments were abandoned and after 1972 most riparian industrial and residential activity ended.
The Olmstead brothers (famous landscape architects) analyzed the Baltimore regional park system in 1904 and advocated purchase of the Gwynns Falls valley from the mouth to the city limits for use as parkland. The complete purchase of this land was never completed, as industrial activity was still quite strong at that time. Now, nearly 100 years later, with much of the former agricultural and industrial activity absent, the Revitalizing Baltimore project has developed a parkland trail running from the mouth of the Gwynns Falls to the city, much as the Olmstead brothers envisioned.
Urban Riparian Hydrology
The most obvious hydrologic changes associated with urbanization are the engineering of stream channels, with the replacement of natural features with concrete channels and streambank stabilization efforts designed to resist increased flood flows. Extensive piped storm drainage networks often completely bypass riparian zones putting large amounts of water, derived from impervious surfaces quickly, and with increased frequencies, directly into streams. A result of this altered hydrology is that incision or "downcutting" is a common feature of urban stream channels. Downcutting results from large volumes of water scouring out sediment that has accumulated during agricultural and/or residential construction land use phases in the watershed (Figure 2). Incision is especially marked in watersheds with old and/or stable urban land use, where there are few sources of sediment to replace material scoured by high flows. There is thus tremendous variability in the condition of urban streams depending on historic patterns of development, redistribution of sediments within streams and hydrogeologic conditions in the watersheds. However, we suggest that over time, urban watersheds move toward stable land use with high amounts of impervious cover and low sediment production that leads to stream incision in most locations, in what is now called "urban stream syndrome."

Figure 2. Conceptual diagram showing streams and associated riparian zones in forested and urban watersheds. High runoff during storm events leads to incised stream channels in urban watersheds, which in combination with reduced infiltration in impervious urban uplands, can lead to reduced riparian groundwater levels.

Stream incision, in combination with reduced infiltration in impervious urban uplands can reduce riparian groundwater levels (Figure 2), which can have dramatic effects on soil, plant and microbial processes. Water table level is a critical controller of riparian ecosystem structure and function, influencing soil type (i.e. the presence of wetland or hydric soils) and plant communities (wetland and upland/wetland transition plants) and the unique fauna (e.g. salamanders) that depend on the presence of specific soils and plants. Groundwater level is also a key controller of the ability of riparian zones to prevent the movement of pollutants from uplands into streams, regulating the interaction of groundwater-borne NO3- with near surface soils that support plant and microbial processes that consume this ion. We suggest that riparian "hydrologic drought" caused by lowered water tables is a general effect of urbanization that likely occurs in many cites.
Urban Riparian Soils
The hydric soils that are common in riparian wetlands are characterized by a series of visual "redoximorphic" features that are a product of chemical reactions, primarily the oxidation, reduction and solubilization of iron. Under anaerobic conditions, iron is reduced, solubilized and leached away, producing grey (or gleyed) colors in the soil profile. Redoximorphic features are used to classify soil "drainage classes" that index long-term average water table conditions at a site. These features take many years to develop and change so they are poor indicators of short-term water table dynamics and they often persist for many years in sites where water tables have dropped, e.g., in urbanizing watersheds.

Figure 3. . Profiles of riparian zone soils before and after enhanced rates of sediment deposition associated with agriculture and residential construction, and lowered water tables associated with urbanization.

Hydrologic changes associated with urbanization cause soil patterns in urban riparian zones to diverge from what is expected for hydric wetland soils (Figure 3). For example, the lowering of the water table associated with urbanization can create aerobic conditions in soils that have been mapped as hydric. Moreover, dynamic cycles of erosion and deposition in watersheds can alter the normal sequence of horizons (layers) in riparian zone soils. In agricultural watersheds, high erosion and frequent flooding lead to increased sedimentation, which buries surface organic horizons. In a survey of riparian soils in the Gwynns Falls watershed, which was previously dominated by agriculture, we observed numerous buried surface horizons and evidence of relic redoximorphic features near what was once the surface organic horizon (Figure 3). The combination of increased sedimentation and the lowering of the water table has necessitated a change in the drainage classification of these soils, with the loss of a hydric soil designation in many cases.
Alteration of riparian soil profiles can have dramatic effects on processes that produce and consume NO3-. We have observed high levels of NO3- and nitrification (a microbial process that produces NO3- ) and low rates of denitrification (an anaerobic process that consumes NO3-) in aerobic, urban riparian soil profiles with water tables deep in the soil profile compared to forested reference watershed riparian soils with shallow water tables (Groffman et al. 2002, Groffman and Crawford 2003, Gift et al. 2010, Bettez and Groffman 2012). These results suggest that soil and hydrologic changes associated with urbanization can cause riparian zones to be sources rather than sinks for NO3- in urban watersheds.
Urban Riparian Vegetation
Vegetation plays several critical roles in riparian zones, including maintenance of stream temperature, provision of woody debris to create stream habitat and uptake of NO3- from shallow groundwater. Given that the regeneration and growth of riparian vegetation is adapted to the flooding regime of the adjacent stream and to water table levels, the hydrologic and soil changes associated with urbanization described above should have dramatic effects on riparian vegetation.
In the BES, we have examined natural and human-altered links between hydrology, soil and vegetation in riparian zones. Comparisons of vegetation between the rural/suburban (upper) and urban (lower) sections of the watershed show distinct patterns across an urban to rural gradient. In the lower, more urban section of the watershed, wetland tree species are either absent or occur as small stems while upland species are abundant, in mixed sizes. A comparison of the number of wetland and upland species in the mostly urbanized Gwynns Falls riparian zone with non-urbanized Piedmont floodplains throughout Maryland shows approximately twice as many upland species in the urban floodplain than in non-urbanized floodplains. The majority of shrubs in riparian zones through the Gwynns Falls are upland species. For herbaceous species, frequencies of upland and wetland species are about equal in the upper and middle regions of the watershed, but upland species are more common in the more urban lower floodplains by a factor of greater than two.
Our vegetation analysis clearly shows the effect of riparian hydrologic drought induced by urbanization. The riparian zone is becoming drier throughout the Gwynns Falls watershed, with the lower urbanized watershed providing the driest habitats, more favorable for germination and growth of upland species. Such a shift could have dramatic effects on ecosystem services provided by vegetation in the riparian system. Long term research will likely reveal a similar loss of wetland riparian species as similar hydrologic changes accompany suburban and urban development in the upper reaches of the watershed.
Urban Riparian Ecosystems
Urbanization is a human-driven process and research in urban riparian ecology must explicitly consider human values and behavior. The integration of social science with physical and biological sciences is one of the clear frontier challenges being addressed in the new urban LTER sites in Baltimore and Phoenix, and in other urban ecology research efforts.
In the BES we use the "Human Ecosystem Framework" as a tool for integration of social science with biological and physical science disciplines. The framework outlines explicit interactions between critical biophysical resources (soil, water, vegetation) and social drivers that depend on and influence these resources. An example of an integrative "human riparian ecosystem" approach in action is the creation of the Gwynns Falls trail, a 14-mile stream valley trail system opened in June 1999 (Figure 4). The trail connects over 30 neighborhoods in west and southwest Baltimore with parklands, unique urban environmental features, cultural resources and historic landmarks (www.gwynnsfallstrail.org).

Figure 4. The Gwynns Falls trail map. From www.gwynnsfallstrail.org

The creation of the Gwynns Falls trail is an excellent example of how social, physical and biological sciences can be integrated in an effective way in an urban ecosystem. The motivation for creation of the trail was social and biophysical degradation. A U.S. Army Corps of Engineers study had chronicled extensive degradation of streams and riparian zones in Baltimore City (poor riparian and in-stream habitat, stream bank and bed stability problems, low water quality). At the same time, neighborhoods were undergoing socio-economic decline, with the loss of nearly 50% of the population of Baltimore City between 1940 and 1990 and the creation of over 60,000 abandoned houses and lots. In 1994, the US Forest Service established the Revitalizing Baltimore (RB) project to carry out community forestry, watershed restoration and educational projects centered on conservation stewardship and outdoor experiences. One of the main objectives of RB was to develop the idea that ecological revitalization can stimulate socio-economic revitalization by bringing people in underserved (poor) neighborhoods together through community forestry and stream restoration projects. These projects foster the development of community cohesion, which leads to community interest in improved city services. Increases in services leads to improvements in environmental and socio-economic conditions and creates positive feedback for neighborhood revitalization, reversing the negative spiral of population loss with consequent environmental and social degradation which leads to further population loss.
The idea for the Gwynns Falls trail emerged from a series of stream restoration projects and was intended to serve as a highly visible focal point for stream and neighborhood revitalization in Baltimore City. The idea was brought to fruition by the Parks & People Foundation, a non-profit group that explicitly works at the interface between humans and the environment, with a focus on creating recreational opportunities for Baltimore City residents. Parks & People was able to coordinate fundraising activities from multiple sources and the trail was opened with great fanfare in 1999.
As more people use the trail, important feedbacks develop. As people experience the stream and riparian zone, they become aware of this unique and valuable natural resource, which increases demand for maintaining the resource. Humans are thus functioning as a regulatory feedback mechanism in the ecosystem, much as a complex web of interactions maintains stability (resistance and resilience) in unmanaged forest ecosystems.
Bain, D.J. and G.S. Brush. 2004. Placing the pieces: Reconstructing the original property mosaic in a warrant and patent watershed. Landscape Ecology 19:843856.
Bettez, N. D. and P. M. Groffman. 2012. Denitrification potential in stormwater control structures and natural riparian zones in an urban landscape. Environmental Science & Technology 46:10909 - 10917.
Gift, D. M., P. M. Groffman, S. S. Kaushal, and P. M. Mayer. 2010. Denitrification potential, root biomass, and organic matter in degraded and restored urban riparian zones. Restoration Ecology 18:113-120.
Groffman, P.M., D.J. Bain, L.E. Band, K. T. Belt, G.S. Brush, J.M. Grove, R. V. Pouyat, I.C. Yesilonis and W. C. Zipperer. 2003. Down by the riverside: Urban riparian ecology. Frontiers in Ecology and Environment 6:315-321.
Groffman, P.M. and M.K. Crawford. 2003. Denitrification potential in urban riparian zones. Journal of Environmental Quality 32:1144-1149.
Groffman, P.M., N.J. Boulware, W.C. Zipperer, R.V. Pouyat, L.E. Band, M.F. Colosimo. 2002. Soil nitrogen cycling processes in urban riparian zones. Environmental Science & Technology 36:4547-4552.
Walsh, C.J., A.H. Roy, J.W. Feminella, P.E. Cottingham and P.M. Groffman. 2005. The urban stream syndrome: current knowledge and the search for a cure. Journal of the North American Benthological Society 24:706-723.
This research was supported by funding from the NSF Long-term Ecological Research (LTER) Program. This material is based upon work supported by the National Science Foundation under Grant No. 1027188. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation.