Available at: http://digitalcommons.calpoly.edu/theses/503
Date of Award
MS in Agriculture - Soil Science
Earth and Soil Sciences
Dr. Brent Hallock, Ph.D.
The Santa Rosa Creek watershed is one of the most pristine watersheds on California’s Central Coast. Preserving this watershed is of great interest because it provides rich soils for agriculture, vast rangelands for cattle, and flowing streams for federally threatened species such as steelhead trout. Soil erosion could impact these resources. Using prediction tools, it is possible to study the erosion that could be occurring in a watershed and identify locations which could contribute the highest amounts of sediment. The objectives of this study were to use RUSLE2 and Geographic Information Systems (GIS) to predict soil erosion rates for each soil map unit in every drainage of the upper Santa Rosa Creek watershed and to determine areas where soil erosion could surpass a soil development rates. Environmental and anthropogenic factors that influence soil erosion such as topography, climate, soil, geology, vegetation, and land use, were described for the entire watershed to provide supplementary data used in the RUSLE2 model and to explain erosion in highly erosive areas. Predicted soil erosion rates were studied to determine if correlations exist between other factors such as slope, existing erosion features, and vegetation.
Predicted soil erosion rates calculated using RUSLE2 confirmed that the watershed is healthy and that 98 percent of the drainages are within sustainable soil erosion rates (five tons/acre/year). There were 37 soil map units totaling 1,617 acres (5.6 percent of the entire upper watershed area) with predicted soil erosion rates above a sustainable rate. In Perry Creek watershed, these sites were located on steep slopes tangent to streams. Along the main-stem of Santa Rosa Creek these sites were found in the headwaters where on average slopes are steep, soils are shallow, and rock outcrops exist. There appeared to be no relationship between predicted high soil erosion rates and mapped upland erosion sites, however upland erosion features could not be identified where vegetation canopy restricted view of the soil surface. Additionally, RUSLE2 predicts rill and interrill erosion while upland erosion sites identified using GIS identified larger erosion features, such as gullies. Correlations between predicted soil erosion rates and vegetation formations were confirmed with shrub and tree formations having the highest average predicted soil erosion values. In addition, there was a moderate positive correlation between slope percent and predicted soil erosion (r=0.76), affirming that predicted soil erosion rates increased with increasing slopes.