Land Use History and Soils: Impacts on Woody Cover AssessmentsEPA Grant Number: FP916378
Title: Land Use History and Soils: Impacts on Woody Cover Assessments
Investigators: Browning, Dawn M.
Institution: University of Arizona
EPA Project Officer: Graham, Karen
Project Period: January 1, 2004 through December 31, 2006
Project Amount: $110,476
RFA: STAR Graduate Fellowships (2004) RFA Text | Recipients Lists
Research Category: Academic Fellowships , Fellowship - Terrestrial Ecology and Ecosystems , Ecological Indicators/Assessment/Restoration
The objective of this research project is to examine patch density and patch-size class distributions in conjunction with total percent land cover to allow for appropriate assessment of the functional consequences of vegetation change. Changes in land cover and land use are recognized as predominant drivers of global change. One of the most striking land cover changes in arid and semi-arid rangelands over the past 150 years has been the proliferation of trees and shrubs at the expense of perennial grasses. Changes in the abundance and spatial pattern of patches of the dryland shrub, velvet mesquite (Prosopis velutina [Woot.]), on uplands in the semidesert grasslands of southeastern Arizona were quantified using a time series (1936, 1966, and 2002) of aerial photographs. The 2-km2 study area on the Santa Rita Experimental Range spanned contrasting soil types (coarse sandy loams [CSL] versus more developed sandy clay loams [SCL]) and encompassed contrasting management treatments designed to reduce woody plant cover (removal of mesquite trees in 1935 by cutting and application of herbicides from 1960 to 1965).
Shrub and nonshrub patches were delineated using a supervised classification with a texture layer and dynamics were assessed using patch-based metrics generated in FRAGSTATS and ArcGIS. From a coarse-scale perspective, woody cover across the entire landscape appeared relatively stable over the 66-year period (30.6, 31.2, and 26 percent for 1936, 1966, and 2002, respectively). This apparent stability, however, masked cover changes and dynamics associated with land use and soil properties at finer scales. On untreated portions of the 1936 landscape, woody plant cover on SCL (26.2 percent) was less than that of untreated CSL (36.0 percent). Although woody cover in 1966 was greater on control than on herbicide treated areas (33.2 percent versus 23.1 percent), shrub patch densities were comparable, reflecting differences in patch-size class distributions. Woody cover on herbicide-treated areas changed little from 1966 (23.1 percent) to 2002 (21.4 percent), whereas cover on control areas decreased from 33.2 percent to 26.8 percent. During this same period, there was a marked increase in patch densities on both soils and all treatment types with the greatest increase on the SCL cut site. However, the four-fold increase in the number of small (<10 m2) patches did not offset decreases in cover that resulted from the loss of a few large patches. The net result was a decrease in total woody plant cover. This example illustrates how quantifying woody plant cover with aerial photography or high-resolution satellite imagery without taking soil or land use history into account can generate misleading and incorrect assessments of vegetation structure.