If you would like more information, please contact Larry Band, University of North Carolina, Bob Costanza, University of Maryland, or Dave Nowak, U.S. Forest Service
Our conceptual model of the urban landscape fuses the approaches of a watershed based input-output scheme with that of a patch dynamics system. This is done by viewing any watershed as a hydrologically connected set of patches with temporally and spatially variable attributes. The set of patches may be destroyed and reformed by disturbance processes, and the state of any patch is at least partially dependent on surrounding and distant patches as influenced by drainage and advection/diffusion processes. A significant attribute of urban landscapes is that disturbance in the form of development and redevelopment can alter not only the patterns and characteristics of patches, but also the topography and drainage sequence to the extent that the horizontal redistribution of water and nutrients can locally be redefined. This degree of topographic and drainage change typically does not occur in other landscapes on the same time scales. Distinct assemblages of vegetation, urban form and socioeconomic characteristics may form an urban catena in which surface patterns may be described at scales ranging from upland to bottomland sequences along a single hillslope, or through the full watershed structure.
The Gwynns Falls watershed follows a pattern common in many older US cities in which a dense urban region with a central commercial area around the river mouth or waterfront is surrounded by poorer residential and commercial areas, which in turn are surrounded by more affluent and less dense suburbs. The distribution and heterogeneity of patch types within each zone must be considered in terms of the different typical soils, vegetation structure and other land cover, and the drainage sequence to understand and compute the flux of carbon, water and nutrients across the landscape and between land surface and atmosphere.
The simulation models that we are parameterizing and operating over this scale range (RHESSYS, GFLM and UFORE) are fully consistent with our conceptual view of the landscape, described above. The set of model approaches we are using is designed to incorporate different levels of surface heterogeneity into the patch structure, and will provide both a conceptual approach and a framework to collect, organize and analyze information sampled across the urban landscape.
RHESSYS operates at the hillslope level and explicitly incorporates variation in patch structure and function down a drainage sequence from hillslope crest to the riparian zone. RHESSys can be scaled to larger watersheds by progressively aggregating the landscape into larger, more complex "hillslopes" and transferring spatially explicit patch heterogeneity between hillslopes into within hillslope distribution functions.
For the spatially explicit GFLM, the modeled landscape is partitioned into a spatial grid of square unit cells. The model is hierarchical in structure, incorporating an ecosystem?level "unit" model that is replicated in each of the unit cells representing the landscape. The unit model itself is divided into a set of model sectors that simulate the important ecological dynamics at a daily time step. While the unit model simulates ecological processes within a unit cell, horizontal fluxes across the landscape occur within the domain of the broader spatial implementation of the unit model to form the GFLM. Such fluxes are driven by cell?cell head differences of surface water and of ground water in saturated storage. Within this spatial context, the water fluxes between cells carry dissolved and suspended materials, determining water quality in the landscape. The ecological model is linked to an economic model which predicts the probability of land use change within the area. The economic model allows human decisions to be modeled as a function of both economic and ecological spatial variables.
Combining the two modeling approaches -- the fine scale, spatially explicit combination RHESSys, which incorporates ecological and physical processes, with the coarser scale GFLM, which incorporates ecosystem processes at the 0.4 km2 scale and combines ecological and hydrological processes with economic processes, will allow a new level of understanding the functioning of urban watersheds. Applying these modeling approaches together with an empirical assessment of the spatial heterogeneity of social, ecological, and physical processes in an urban region will determine whether the heterogeneity at fine scales acts as a mechanism for the fluxes at the coarser scales. Evaluating the entire metropolitan area, and subsets represented by watersheds contrasting in land cover, management, and socio-economic processes will add substantially to the ability to understand long-term dynamics of urban ecosystems and the downstream environments they affect.
A third modeling effort (the Urban Forest Effects or UFORE model) is designed to quantify urban vegetation structure, biogenic volatile organic compound emissions, vegetation carbon storage and sequestration, tree transpiration, dry deposition to vegetation of gaseous and particulate pollutants, insect and disease potentials, and the effects of vegetation of local building energy and use. The model uses locally collected field data along with local hourly pollution and weather data to estimate urban forest structure and ecosystem services and values. Model analyses have been performed in numerous cities across the globe.
The UFORE model has been incorporated in the i-Tree modeling suite and renamed i-Tree Eco. The model is available for free at www.itreetools.org. The i-Tree model is based on a partnership among the US Forest Service, Davey Tree Expert Company, National Arbor Day Foundation, Society of Municipal Arborists, and the International Society of Arboriculture. Spatial integration and mapping within i-Tree has begun, with many projects currently under way. This work merges the field data assessments with the aerial cover maps to help illustrate the distribution of ecosystem services. This work is also integrating field-derived ecosystem service information with several spatial modeling components to help determine how forest cover and ecosystem services will likely change in the future and which forest structures are best to optimize ecosystem services to sustain environmental quality and human health.
Various spatial tools in development include programs to determine the highest priority planting areas based on tree and human population distribution, how air pollution concentration may vary based on locations relative pollutant sources in cities (risk to exposure); and how tree and impervious locations can affect local air temperatures. This temperature mapping will be integrated with other programs to help determine how changes in tree cover (and air temperatures) affect air pollution removal and concentrations, volatile organic compound emissions, and human comfort. Other programs in development will project tree populations through time to illustrate what the new cover would look like in the future and the associated changes in ecosystem services and values due to the cover changes. The ultimate goal of these spatial tools is to help planners or managers determine optimal landscape structure to sustain or improve environmental quality and human health at the local to regional scale in urban or urbanizing landscapes.