Heejun Chang

Date of Award

Fall 12-13-2013

Document Type


Degree Name

Master of Science (M.S.) in Geography



Physical Description

1 online resource (ix, 157 pages)


Ecosystem services -- Oregon -- Tualatin River Watershed, Ecosystem services -- Oregon -- Yamhill County Region, Water resources development -- Environmental aspects -- Oregon -- Willamette River Watershed, Land cover -- Economic aspects -- Oregon -- Willamette River Watershed, Climatic changes -- Oregon -- Willamette River Watershed -- Forecasting




Humans make decisions within ecosystems to enhance their well-being, but choices can lead to unintended consequences. The ecosystem services (ES) approach supports decision-making that considers all environmental goods and services. Many challenges remain in the implementation of the ES approach like how specific ES vary through space and time. We address this research problem using the Tualatin and Yamhill river basins in northwestern Oregon as a study area. Freshwater ES are quantified and mapped with the spatially-explicit ES modeling tool, Integrated Valuation of Ecosystem Services and Tradeoffs (InVEST). In chapter II, we develop a simple urban land cover change modeling approach with selected stakeholder input. The products of this analysis are used in part to answer the question of how the freshwater ES of water yield, nutrient retention, and sediment retention will change in the future, and how their distribution potentially will change? In chapter III, these ES are modeled in InVEST using the land cover scenarios and three downscaled global climate models. The base period is 1981 to 2010 and the future period is 2036 to 2065. The models are calibrated to empirical estimates, and display different sensitivities to inputs. Water yield increases with higher rainfall but decreases with the highest temperature scenario. Nutrient export and retention estimates are positively correlated. In the Tualatin basin, more urban lands generally lead to increases in nutrient exports and retention. The effect is reversed in the Yamhill basin from much larger agricultural exports. Sediment exports and retention increase with higher winter rainfall but are negatively spatially correlated due to topographic effects. Simulation of a landscape scale installation of riparian buffers leads to decreases in exports and increases in retention. The distribution of the provision of freshwater ES remains unchanged throughout the scenarios. With few parameters in each InVEST model, all display a high degree of sensitivity. Parameterization is subject to high uncertainty even with calibrated values. We discuss the assumptions and limitations of InVEST's freshwater models. The spatially explicit nature of InVEST is its main advantage. This work coupled with other analyses in the study area can facilitate the identification of tradeoffs amongst ES leading to better ecosystem management.

Persistent Identifier