GIS descriptive modeling of General Land Office surveyor tree data and spatial relationships with river valleys and their potential fire shadow effects in Polk and Dallas Counties, Iowa.

摘要

Fire was a controlling force in the Iowa landscape long before Euro-Americans took up residence among its forests and prairies (Anderson 1998). Rivers and river valleys are features that influenced the spread of these fires (Leitner and others 1991, Anderson 1998). Based on previous studies in the midwest, the primary wind direction during fire season was from the south and west (Leitner and others 1991, Anderson 1998). As fire spread, river valleys acted as fire breaks and decreased fire frequency and intensity in protected areas to the north and east (Leitner and others 1991). In this study several spatial and statistical measures were used to examine evidence for this view of the spatial relationships between rivers, landforms, and fire within Polk and Dallas Counties, Iowa. Mean direction and distance to General Land Office (GLO) timber on both the north/east (protected) and south/west (exposed) sides of rivers was calculated using Euclidian distance and direction functions in ArcGIS. Timber was determined to be present in larger quantities in the protected zone in Polk County, and in the exposed zone in Dallas County. Timber was also determined to be farther away from GLO rivers on average, in exposed zones in both counties. These results do not support the hypothesis that timber was present in greater quantities and was further away from GLO rivers to the north and east, and demonstrate the need for research in additional areas. GLO tree species composition (witness, bearing, line, and meander trees) was also analyzed to determine if edaphic and vegetative factors including soil characteristics, slope, and GLO vegetation type were similar on both sides of GLO rivers. Based on this analysis, it does not appear that edaphic factors are different on protected and exposed sides of GLO rivers, so they do not provide an explanation for differences in tree species distribution. Importance values were calculated for each species within each zone by summing relative density and relative dominance values (Bolliger and others 2004). In the protected zone, species with the highest importance values included White Oak, Elm, and Burr Oak. In the exposed zone, species with the highest importance values included Burr Oak, White Oak, and Elm. For trees lacking genus or species identifiers, edaphic and vegetative factors were analyzed to determine a likely genus or species. Results of this analysis for 37 trees identified only as 'maple' resulted in 26 trees designated as Silver Maple, and 11 trees identified as Sugar Maple. For two trees identified only as 'Black', one was identified as Black Walnut, and one was identified as Black Oak. For the 35 trees identified only as Ash, 26 were identified as Green Ash, and 9 were identified as White Ash. Future research should include the expansion of this study to additional parts of Iowa, analyzing species composition based on fire tolerance, and measuring the statistical significance of differences in composition in protected and exposed zones.