SUITMA 2003 Nancy - Landscape model for mapping disturbed soils in urban areas

Soils of Urban, Industrial, Traffic, Mining and Military Areas
SUITMA 2003 Nancy

Landscape model for mapping disturbed soils in urban areas

SUITMA

This abstract is about one of the papers of the Methodology for the study of urban soils and classification theme of the SUITMA 2003 Nancy symposium.

Joyce Mack Scheyer,i

Berman D. Hudson.i

• i - Natural Resources Conservation Service.

Introduction

The mapping and application of urban soil information follows an adaptation of the traditional soil-landscape paradigm. The Natural Resources Conservation Service has a federal mandate for urban conservation and technical assistance in addition to traditional work on soils of range, forests, and cropland. Applications of soil survey specific to human-modified soils in urban areas include metal-contaminated soils, urban agriculture, storm water management, and erosion control from construction sites. Our objective was to review historical methods for soil survey in urban areas of the USA and to record some of the tacit knowledge of the urban soillandscape paradigm. We summarize tools and methods that are useful for urban soil mapping.

Adapting the Soil-Landscape Paradigm for Urban Areas

The challenge set forth by Hudson of capturing the paradigm out of the head of the experienced soil surveyor - in this case for urban soils - is still essential for effective transfer of soil information to the decision-maker. The other critical input is the knowledge base which includes "tacit knowledge" (Hudson, 1992) of the soillandscape model for the area under investigation. The interactive effect of changes in landform and changes in soil materials or layering takes place in multiple dimensions across space and time. A critical, but regionally sensitive, input value for SoLIM is "data on the soil formative environmental conditions" or the relative state of each of the soil forming factors in a format available to GIS. These factors might include climate, parent material, elevation, slope aspect, slope gradient, remote sensing of vegetation (leaf area index, tree canopy coverage), distance to streams, bedrock and surficial geology…(Zhu et al., 2001). Field soil surveyors with experience in rural areas often express an inability to imagine urban soils existing among the buildings and pavement of the city. This attitude reflects a need to better understand urban soils by definition and a training opportunity in pattern recognition. Hudson (1992) stated this need in terms of new visual experiences that can be perceived more readily if some comprehensive framework of knowledge has been established into which the new visual experience can be fitted. First one must learn to see the (urban) landscape as a separate entity. For the paradigm to be successful it must then capture the imagination of a group of scientists and attract them away from competing modes of scientific activity. It also must be nonpecific and open-ended enough to leave many interesting problems for practitioners to solve (Hudson, 1992).

Pre-mapping studies are the key to understanding the altered landscapes and natural soil relationships in Baltimore City (Levin, 1991). Levin's work was notable for the extensive documentation of "tacit knowledge" behind the mapping. Housing density also is a factor in a model of urban hydrology (TR-55) commonly used to compare infiltration and runoff before and after urban disturbance (USDA-NRCS, 1986) where a larger lot size is credited with more "good" lawn. TR-55 runoff curve numbers by housing density and vigor of coverare based on a increasing runoff (RCN) by decreasing infiltration (HSG) as a reflection of soil condition and cover type.

Catenas are a traditional tool for soil survey that can be adapted to convey the urban soil-landscape paradigm. The catena concept as developed by Milne and described by Hall and Olson (1991) had two variants: (i) all soils of a catena are formed in a single material, and (ii) soils of a catena may be formed in two or more materials. Soils of a catena differ in the first of these because of (i) drainage conditions, (ii) differential transport and deposition of eroded material, or (iii) leaching, translocation, and re-deposition of mobile chemical constituents. In the second variant, a geologic factor of multiple-parent materials is added. Either catena variant can describe urban soils. The catena concept can be adapted to show varying depths to contamination below the surface from aerial deposits on a previous landscape or from differential placement of toxic fill materials. Distribution of metals may reflect the landform of the soil surface at the time of deposition (vertical discontinuities), or areas of cut and fill where contaminated materials were added into existing soils (horizontal discontinuities). Ground penetrating radar is useful in non-destructive plotting of catenas related to depth to bedrock and compacted soil layers (Doolittle et. al. 1997). Block diagrams of the urban landscape before and after human intervention can show potential effects of artificial barriers or landscaping on relief as a soil-forming factor. An overlay of urban development as buildings, deep roadbeds, and storm-drains would illustrate the human factor in barriers to hydrologic flow in the landscape.

Urban Soil Series Describe Disturbed Soils on Human-Constructed Landforms

Human-constructed landforms have been the basis for identifying new soil series in urban areas for over 20 years. The earlier surveys often grouped disturbed soils with paved urban land due to the extensive disturbance and the difficulty in separating out the tracts of land expected. Nine examples of published soil surveys with significant urban areas are listed in the references. Naming and describing urban soil series is a challenging task using soil taxonomy where map units may include 1) chemical contamination, 2) significant contents of non-soil materials with unpredictable behavior over time, and 3) artificial landforms.

Urban soils were named and described as series in the St Louis County and City Survey Area but were mapped only as part of complexes with urban land (Benham, 1982). They were like the soils in Charlotte County, FL and Cumberland and Hoke Counties, NC that consist mainly of soil material reworked as fill over truncated or buried soils (Henderson, 1984; Hudson, 1984). The content of non-soil materials such as glass, cinders, and brick was low and confined to the deep substratum where it was unlikely to affect soil behavior for most land uses. Human contributions were considered at that time to be of relatively inert physical objects not in conflict with particle-size classification and to have little effect on soil chemical transformations. This pattern of new urban series consisting largely of reworked soil materials and minimal content of artifacts continued for South LaTourette Park on Staten Island, New York (Hernandez and Galbraith, 1997). Four of their five soils were basically reworked and transported soil materials and only the Greatkills soil introduced the concept of significant content of non-soil materials (household garbage). One soil survey in Florida described over 4% of the county land as formed by earthmoving operations on of dredged fill materials over natural layers in areas designated for urban development (Henderson, 1984). As these young soils develop, their genesis will reflect these differences in parent material and the impact of the newly created landform position and soil depth on internal soil processes.

Recommendations

The advantage of the similarity model for urban areas is that it represents soils simultaneously in both the spatial domain and the parameter domain. Individual soils can be represented not in relation only to the bestfitting central concept but as more detailed intergrades between the modal values of the central concepts (Zhu et. al., 2001). The intergrades in urban areas may include small areas with critical differences where two soil properties combine to enhance (or delete) a particularly high-risk feature. Local citizen groups often are concerned about these high risk environmental issues and dependent on agencies such as USDA-NRCS for objective, scientific information for their decision-making. Improved soillandscape models for urban areas will help us to meet this customer need.

Networking is essential during the early stages of each urban soil survey for pre-mapping and knowledge-base development. Collaboration among agencies and sometimes across national borders can improve data layers for geology, vegetation, and urban infrastructure. Field trials are needed to improve estimates of urban soil behavior. This includes expanded sampling of urban soils to refine defaults in models. The field trials would follow protocols of soil survey investigations (NSSH, 2002) in the format of field sampling guides for community groups of non-scientists. The trials might help us to build a database of site-specific data with links to landscape model.

Increased dialogue with colleagues from many nations can help us to adapt US Soil Taxonomy to describe the genesis and behavior of urban soils (such as the ICOMANTH and SUITMA workgroups). Within these networks of scientists the research and mapping protocols focus on urban soils as a subset of anthropogenic soils. The urban areas mapped in the USA are a small but rapidly growing percentage of the soil survey workload. We can stretch our resources by sharing the task of describing urban soil-landscapes through a well-expressed paradigm built on decades of field experience.

References

• Benham, K.E. 1982. Soil Survey of St. Louis County and St. Louis City, Missouri. U.S. Gov. Print. Office, Washington, D.C.
• Bowman, Roy H. 1973. Soil Survey of the San Diego Area. U.S. Gov. Print. Office, Washington, D.C.
• Brown, J.H. and S.T. Dyer. 1985. Soil Survey of Montgomery County, Maryland. U.S. Gov. Print. Office, Washington, D.C.
• Burghardt, W. and C. Dornauf (ed). Proceedings of First International Conference on Soils of Urban, Industrial, Traffic, and Mining Areas. University of Essen, Germany. July 12-18, 2000.
• DeKimpe, C.R. and J-L. Morel. 2000. Urban Soil Management: A Growing Concern. Soil Sci.165:31-40
• Doolittle, J.A., L.A. Hernandez, and J.M. Galbraith. 1997. Using Ground-penetrating Radar to Characterize a Landfill Site. Soil Survey Horizons, 38:60-67.
• Hall,G.F. and C.G. Olson. 1991. Predicting variability of soils from landscape models. p. 9-24 In M.J. Mausbach and L.P. Wilding (ed). Spatial Variability of Soils and Landforms. SSSA Special Publication 28. Soil Sci. Soc. Am., Madison, WI.
• Henderson, W.G., Jr. 1984. Soil Survey of Charlotte County, Florida. U.S. Gov. Print. Office, Wash.,D.C.
• Hernandez, L. and J.Galbraith. 1997. USDA-NRCS. Soil Survey of LaTourette Park, Staten Island. USDA-Natural Resources Conservation Service, Lincoln, Nebraska and Cornell University.
• Hudson, B.D. 1984. Soil Survey of Cumberland and Hoke Counties, North Carolina. U.S. Gov. Print. Office, Wash.,D.C.
• Hudson, B.D. 1990. Concepts of Soil Mapping and Interpretation. Soil Survey Horizons 31:63-73.
• Hudson, B.D. 1992. The soil survey as a paradigm-based science. Soil Sci. Soc. Am. J. 56:836-841.
• ICOMANTH. 2003. Program plan and circular letters. International Committee on Anthropogenic Soils, U.S. Soil Taxonomy. http://clic.cses.vt.edu/icomanth/
• Levin, Maxine J. 1991. Soil Mapping In An Urban Environment - Baltimore City MD. In Proceedings for Human-Influenced and Disturbed Soils Conference, University of New Hampshire, Durham, NH, Nov.23, 1991
• Levin, M. 1998. Soil Survey of Baltimore City, MD. U.S. Gov. Print. Office, Wash. D.C.
• Soil Survey Staff. 1999. Soil Taxonomy. 2nd ed. NRCS. U.S. Dep. Agric. Handb. 436. U.S. Gov. Print. Office, Wash., D.C.
• Smith, Horace. 1976. Soil Survey of Washington, District of Columbia. U.S. Gov. Print. Office, Wash.,D.C.
• USDA-SCS.1986. Urban Hydrology for Small Watersheds. Technical Release No. 55 (TR-55) http://www.wcc.nrcs.usda.gov/water/quality/common/tr55/tr55.html
• Zhu, A.X., B. Hudson, J. Burt, K. LubichK. Lubich, and D. Simonson. 2001. Soil Mapping Using GIS, Expert Knowledge, and Fuzzy Logic. Soil Sci. Soc. Am. J. 65:1463-1472.