Protocols and Methods
Soil samples were taken to a depth of 6 inches with a standard soil corer (diameter = .75 inches). Grassy material was removed before drying and testing for P.
The Bray-1 test for plant-available P was conducted by the University of Wisconsin Soil and Plant Analysis Lab.
GPS longitude and latitude: Data points for measuring soil P and associated factors were randomly located and stratified by zone according to the combination map - with approximately 70 data points per zone. Location and address of each point was determined using the Madison and Dane County parcel GIS layers. Permission was requested from landowners to take a soil sample. If permission was denied (2 cases out of 330) or if there was no one present at the location, a coin toss was used to determine movement one parcel to the right or to the left along the same road.
URG Zone versus URG location: Urban-rural gradients were calculated to be continuous variables, but categorized into five zones for ease of analysis (Urban, Suburban, Suburban Fringe, Agricultural Fringe, and Agriculture).
Sample_loc: sample locations within each sampling site were haphazard.
Simplified_topo: refers to topography immediately surrounding the area where the sample was taken.
URG_Distance: This urban-rural gradient is an attempt to define a gradient as close to the direct distance method used in other gradient studies, while taking Dane County's unique geography into account. Since urban and suburban development in Dane County appears to follow a pattern along major roads, I used an analog for driving time to the center of downtown (marked by the state Capitol building) as a modified and indirect distance gradient.
The map of urban-rural gradient by modified distance from the capitol (the modified-distance map) was developed with a grid size of 100 m. The location of the capitol was marked. The driving time to the capitol building was calculated by summing a cost of moving across the grid from one cell to the next in the direction of the capitol building: low cost along major roads, medium cost along minor roads, and high cost for moving from a grid cell with no roads towards the nearest road. Total cost for moving from every 100 x 100 m grid cell to the capitol was calculated based on this algorithm and mapped. I then split this continuous gradient into zones for mapping purposes. The five urban-rural gradient zones were determined by standard deviation of the gradient. Any grid cell between 3 and 2 standard deviations below the mean gradient value was considered to be urban. Any grid cell with a value between 2 and 1.5 standard deviations below the mean was suburban. Similarly, between 1.5 and 1 standard deviations below the mean was considered suburban fringe, 1 to 0.5 standard deviations below the mean was agricultural fringe, and from 0.5 standard deviations below the mean to 3 standard deviations above the mean was called agricultural. Grid cell values were then rescaled to range between 0 and 1, with 0 being at the state capitol building and 1 being the cell with the longest travel time to the capitol.
URG_Population: Population density was chosen because it may be a representation of neighborhood type, land use, and age of the development, all factors I strongly suspected would play a role in determining soil P concentrations. Population density and housing density information was available from the Applied Population Lab in the Rural Sociology Department at the University of Wisconsin-Madison (http://www.ssc.wisc.edu/poplab/) and from the U.S. Census homepage (http://www.census.gov/). Population density data were available from the census web page by census block. Coverages were reclassified to a 100 m grid by averaging the population density values for all census blocks (weighted by area) within each cell. Zone was calculated for each grid cell: any cell with a population equivalent to or greater than 2687 people per square kilometer was classified as urban, between 1737 and 2687 people per square kilometer was suburban, between 750 and 1736 was suburban fringe, between 25 and 749 was agricultural fringe, and up to 24 people per square kilometer was considered agricultural. These numbers match closely with the U. S. Census definition of urban core areas, which is approximately 2590 people per square kilometer (http://www.census.gov). The grid cell values were then rescaled to range between 0 and 1, with 0 being the lowest population density and 1 being the highest.
Apparent zone: Apparent zone was determined by visual inspection using a predetermined set of definitions of each zone. For example, urban sites were those with highest housing density or some industrial use, suburban were those sites of moderate housing density and residential character, suburban fringe were newer (5-10 years old) residential developments of low housing density and relatively large houses, agricultural fringe were older residential developments of low density, and agricultural were those areas that were actively farmed (See Appendix A of Elena Bennett's dissertation (Bennett 2002) for more information on apparent zone including visual photographs of each zone).
Information about the protocols used to measure plant available Bray-1 P and total P, in addition to other nutrients, at SPAL can be found at: http://uwlab.soils.wisc.edu/.