CHAPTER II
ANALYSIS OF SOIL DATA USING GEOGRAPHIC INFORMATION SYSTEMS
(GIS)

Geographical variation is integral to soil. Spatial variations in soil properties caused by geologic, chemical and topologic factors are combined with variations caused by hydrologic, edaphologic, biologic, climatologic and temporal factors. Disturbances caused by management practices can influence spatial variation of some soil properties. Changes in soil properties will be reflected in their influence on speciation, growth, and other ecosystem components and activities dependent on the soil. Understanding the reasons for spatial distribution of soil characteristics and the scales of those distributions is important in comprehending soil's role in an ecosystem or management area.

Ideas of the spatial nature of soils have developed with soil science. Dokuchaev introduced the ideas of soil topography (Fridland 1972), and many others have added to the ideas of soil as a feature that varies with geographic factors. Descriptions of soils have depended on defining areals, catenas, soil series and associations from a contiguous "soil mantle." Cartography is used to help define these areas of differing soils. Mapping soils occupied many soil scientists over the years, and advances in mapping technology have been helpful to soil scientists. The use of aerial photography expanded the process of mapping soils. Now, with geographic information system (GIS) computer software, soil mapping is becoming even more refined. Soil surveys such as those of the NRCS are being digitized (SSURGO). These maps contain polygons for soil series with homogenous properties that do not reflect the variability in the site. Rogowski and Wolf (1994) combined variability in bulk density and hydraulic conductivity from geostatistical analysis with soil survey maps to increase the accuracy of the digitized survey maps. Soil properties may be best explained by "fuzzy logic," or clustering techniques that allow a wider or non-discrete definition of soil properties (Burrough 1992). Soil scientists are proceeding from Fridland's (1972) statement that "soil is physically contiguous but geographically discrete," (and the traditional Boolean soil series maps), to the ideas of Thwaites (2000) of a three-dimensional continuum of variations outside of , but including, the traditional vertical "profile." Zhu et al. (1996) used GIS and fuzzy logic to create more accurate soil series maps. Moore et al. (1993) and Johnson et al. (2000) used GIS to relate soil properties to topologic postion. Rahman et al. (1996) used GIS and geostatistics to examine spatial variability of soil properties. The amount of information that can be processed with GIS assists soil scientists to see the soil as reacting to and changing as part of an entire landscape of geologic, hydrologic, edaphologic, biologic and temporal factors.

This chapter shows uses of GIS to map and analyze the spatial aspects of the sites and data acquired in the stumping study discussed in chapter one. The purpose of the study is to understand the effects on soil and trees resulting from stump removal to control Phellinus weirii during forest management operations. The disturbance of the soil caused by the stumping operation changed the values of bulk density and nitrogen concentration between stumped and non stumped plots. The measurements from this study have been analyzed using traditional statistics, but adding location and distance to the calculations adds another dimension to the results. The stumping study concerns comparisons of measured data in specific treated and untreated areas. The areas outside the plots are of little interest, but site variability may influence the results obtained at plots located in certain areas. Whether these differences are the result of stump removal or reflect the underlying variability of the site is a question that can be addressed with GIS analysis.

Variability in forest stands and soils can reduce the effectiveness of large scale management practices. Assembling information in thematic GIS layers makes it possible for managers to compare the many criteria necessary for informed decision making. The maps made from combining these layers provide a great tool for analyzing and visualizing how data collected from individual sample points are related to each other and the trends over an entire site. In a study such as the stumping study, that incorporates soil, forest floor, tree and pathogen data from specific areas, thematic layers created from each variable can be used to compare data in specific areas. Managers can add topographic and soil maps or remote sensing data to their information layers to locate soil properties of interest that will influence their decisions. Thematic maps are presented to display the data acquired in the stump removal study.