Graduate Thesis Or Dissertation

 

Ecohydrologic Connections of Rangeland Ecosystems in Central and Eastern Oregon, USA Public Deposited

Downloadable Content

Download PDF
https://ir.library.oregonstate.edu/concern/graduate_thesis_or_dissertations/pn89dg07z

Descriptions

Attribute NameValues
Creator
Abstract
  • In light of water scarcity and the impacts of climate change, there is an increased need to understand the interaction between land use characteristics and ecohydrologic processes in semiarid regions. Additionally, many semiarid and arid regions face various land management challenges, including woody plant encroachment, decreased snowpack, and increased stream temperatures. Therefore, a more comprehensive understanding of these processes is necessary for informing short-term land management approaches and long-term planning to help protect the resiliency of these systems. The overarching goal of the studies presented here was to examine the ecohydrologic connections and environmental characteristics at two semiarid watersheds. The research for this dissertation sought to assess these connections in the context of two significant land management concerns: western juniper encroachment and increasing stream temperatures. This dissertation is divided into four chapters. The research for the first two chapters of this dissertation took place at the Camp Creek Paired Watershed Study (CCPWS) in central Oregon, USA. This is a long-term study site established in 1993 in order to research the ecohydrologic impacts of western juniper (Juniperus occidentalis) encroachment and removal. The first chapter compares the seasonal water balance of a western juniper-dominated watershed to that of a sagebrush-dominated watershed over a period of eight years. The second chapter examines multiple approaches to estimating evapotranspiration (ET) and characterizes the relationship between two vegetation indices [Normalized Difference Vegetation Index (NDVI) and the Normalized Difference Moisture Index (NDMI)] and soil moisture, ET, and springflow characteristics. CCPWS consists of two adjacent watersheds of similar size and orientation and a riparian valley site located downstream of both watersheds. The majority of western juniper was removed from one watershed in 2005 and 2006 (‘Mays WS’), and big sagebrush (Artemisia tridentata) is the dominant overstory vegetation. Western juniper is the dominant overstory in the other watershed (‘Jensen WS’). Western juniper encroachment has been linked to reduced herbaceous productivity, altered soil moisture characteristics, altered streamflow timing, and soil erosion. Additionally, western juniper is costly and labor-intensive to remove. Cattle grazing is a key land use in central OR. In many semiarid regions of OR, reductions in herbaceous vegetative production and water availability can have ecological and economic impacts. In order to better understand the potential hydrologic impacts of western juniper at this study site, a seasonal water balance approach was used. Eight years of streamflow, springflow, soil moisture, shallow groundwater levels, and meteorological data (precipitation, air temperature, and solar radiation) were measured. The Water Table Fluctuation Method was used to calculate shallow aquifer recharge. The Hargreaves-Samani equation was used to calculate potential evapotranspiration (PET). Seasonal ET was calculated using PET and as the sink term in the water balance approach. ET accounted for the largest portion of the water budget for both watersheds, although springflow and streamflow were greater at the sagebrush-dominated watershed compared to the juniper-dominated watershed. For both watersheds, greater groundwater recharge occurred and deep percolation occurred in snow-dominated years compared to rain-dominated years, even when total annual precipitation amounts were similar. Specific data regarding ET are very limited in this region and in many semiarid areas. For the second chapter, satellite-based remote sensing data and readily available sources were used to examine monthly ET, PET, NDVI, and NDMI for both watersheds at CCPWS. Additionally, the Soil and Water Assessment Tool (SWAT) was used to model monthly ET for both watersheds. Environmental indicators, specifically springflow, soil moisture, NDVI, and NDMI, related to ET were also examined. . A small unpiloted aerial vehicle (UAV) was used to collect thermal infrared and multispectral imagery (red, green, blue, near-infrared, and red-edge wavelengths) at a small plot in each watershed, which was used to calculate ET and NDVI at an hourly scale for a small plot at each watershed. Considerable variability in seasonal and annual ET patterns and totals was found across the different watershed-scale ET models examined in this study. In general, ET rates peaked in May and June, but this was not the case for all models. For most of the watershed-scale ET models examined, total ET was greater at Jensen WS than Mays WS. A significant correlation was found between SWAT-modeled ET and NDMI, NDVI, and volumetric water content at Jensen WS. At Mays WS, a significant correlation was found between SWAT-modeled ET and volumetric water content, springflow, NDMI, and NDVI. A significant correlation was found between plot-scale hourly NDVI and ET. NDVI and springflow were also found to be significant predictors of ET at the plot scale. Research for the third chapter also sought to characterize aspects of the water balance. Multiple land use practices and land cover types are present at this study site. Sagebrush steppe, including western juniper and perennial grasses, accounts for a large portion of the watershed. Ponderosa pine and mixed conifer forests dominate in the middle reaches of the watershed. Cattle grazing and forestry are two primary land use practices at this study site. For the third manuscript, SWAT was used to model the monthly water balance for a 1280 ha watershed in eastern OR. A combination of on-site weather measurements (precipitation, air temperature, relative humidity, and solar radiation) and PRISM datasets were used to create the SWAT model. Two years were used as a “warm up” period for the model. A 10 m DEM was used for watershed delineation, National Land Cover Database (NLCD) data were used for land cover classification, and State Soil Geographic (STATSGO) data were used for soil type identification. The SWAT-calibration and uncertainty program (SWAT-CUP) was used for calibration, sensitivity analysis, and validation. Streamflow data from 2021 and plant available water content (PAWC) from 2018, 2019, and 2021 were used for calibration. Streamflow data were limited and therefore only PAWC measurements were used for validation. Sensitivity analysis was conducted for calibration simulations using streamflow only, streamflow and PAWC, and PAWC only data. Mean annual precipitation across the watershed from 2018 through 2021 was 377 mm yr⁻¹. ET accounted for the majority of the output of the water balance at 253 mm yr⁻¹, followed by water yield (123 mm yr⁻¹). Total modeled aquifer recharge was 10 mm yr⁻¹. Based on the sensitivity analysis, parameters related to snow cover, canopy cover, soil characteristics, and curve number were among the most influential parameters. The fourth chapter characterizes stream temperature dynamics along a small spring-fed stream and tributary and builds upon the previous research in land cover and water balance characterization at the study site in eastern Oregon. In the Pacific Northwest of the U.S., increased stream temperatures are of particular concern because of their negative impacts on cold-water fish (such as salmonids). While the link between land cover change and stream temperature has been widely researched, particularly in more humid regions, more information is needed to understand these interactions in semiarid climates. Stream temperature measurements were taken along the longitudinal gradient. Both the stream and tributary originate at a small spring. The daily mean, minimum, and maximum stream temperature were calculated along with the seven-day moving average (7DA), the seven-day moving average of the daily maximum (7DADM), and the diurnal range of stream temperatures. Land cover classification was performed using an object-oriented support vector machine approach. The land cover type (forested, sagebrush/shrubland, herbaceous, or non-vegetated) was examined within a 30 m buffer along the stream. A support vector regression (SVR) approach was used to examine the relationship between stream temperature characteristics (specifically the 7DADM and the diurnal range) and environmental characteristics (mean air temperature, dew point temperature, vapor pressure deficit, SWAT-modeled springflow, and land cover characteristics). Excluding the headwater sites, stream temperatures were generally greater at lower elevation sites compared to higher elevation sites, but this did not hold true for all seasons or all locations. Water temperatures at the headwater springs varied very little across seasons or years. The average diurnal range in stream temperature of other sites varied between 1.8 and 5.8 °C and did not demonstrate an association with elevation. The SVR models indicated that air temperature, followed by sagebrush steppe land cover and forest cover were the primary predictors for 7DADM or diurnal stream temperatures. The SVR model for 7DADM (R²=0.83) performed better overall than the SVR model for diurnal stream temperature (R²=0.55). The research profiled in this dissertation addresses the need for more research into the ecohydrologic processes in semiarid regions. Additionally, this research examined multiple approaches that can be applied in data-limited environments. The use of readily available data, such as PRISM, remote sensing imagery, or ET datasets, can help address these gaps in data, particularly when combined with in situ data. Results of this study contribute to the existing body of knowledge regarding the relationship between ecohydrologic processes and land use characteristics, which can provide insight into future research and land management decisions.
License
Resource Type
Date Issued
Degree Level
Degree Name
Degree Field
Degree Grantor
Commencement Year
Advisor
Rights Statement
Publisher
Peer Reviewed
Language

Relationships

Parents:

This work has no parents.

In Collection:

Items