Search

Search Results

• 1. [Article] The effects of land use on mineral flat wetland hydrologic processes in lowland agricultural catchments

  Hydrologic processes within mineral flat wetlands, along with their connections to groundwater and downstream surface water in lowland agricultural catchments are poorly understood, particularly under ...

  Citation × Citation Citation Abstract Copy Title: The effects of land use on mineral flat wetland hydrologic processes in lowland agricultural catchments Author: Marshall, Sarah M. (Sarah Marie) Hydrologic processes within mineral flat wetlands, along with their connections to groundwater and downstream surface water in lowland agricultural catchments are poorly understood, particularly under different land uses. In the three field studies included in this thesis, we examined infiltration, wetland hydroperiod, groundwater recharge dynamics, surface runoff generation, and water quality in mineral flat wetlands using a combination of soil and hydrometric measurements, stable isotope tracers, and water chemistry analysis. Our overarching objectives were to examine, for mineral flat wetlands under native prairie, farmed grassland, and restored prairie land cover: 1) how different land management influences infiltration and wetland hydroperiod at the plot scale, 2) the effects of land use on seasonal groundwater-surface water dynamics at the field scale, and 3) seasonal variation in runoff sources and nutrient transport from native prairie and farmed wetlands at the small catchment scale. At the plot scale, our results suggest that edaphic factors, particularly those related to soil structure, are strongly associated with wetland infiltration and overall hydroperiod across least-altered prairie, farmed, and restored prairie mineral flat wetlands. The hydroperiod metrics we examined were generally more sensitive to level of site disturbance than land use alone. At the field scale, our results indicate that, in spite of land use differences and slight variations in soil stratigraphy, many similarities exist in overall wetland hydroperiod, water sources and evaporation rates for mineral flat wetlands in the Willamette Valley lowlands. Isotopic evidence suggests that the greatest degree of groundwater-surface water mixing occurs in the upper 0.5 m of the saturated soil profile across sites under all land uses. Finally, at the small catchment scale, farmed wetland runoff was isotopically similar to field surface water for most of the wet season, indicating that saturation excess was an important runoff generation process. Prairie wetland runoff was isotopically similar to upstream water throughout the winter, and briefly similar to shallow groundwater and surface water within the wetland in mid-spring. Throughout the wet season, elevated nitrate, sulfate, and chloride concentrations were observed in groundwater and surface water at the farm site, and deeper groundwater at the prairie site. Upstream-downstream runoff chemistry remained similar throughout the wet season at the prairie site. Farm site runoff chemistry reflected the dominant water source within the farm field throughout the wet season. Our findings suggest that, while surface water pathways dominate runoff from wetland flats under farm land use, large wetland flat fields have a high potential to absorb, store, and process nutrients and agrochemicals from on-site and nearby off-site chemical inputs. Mineral flats that maintain wetland hydrology in spite of farm use represent a unique balance between agricultural production and preservation of some of the water storage and delay, and water quality-related ecosystem services once provided at a much larger scale in the Willamette Valley lowlands. We anticipate that results of this work will lead to better understanding of key site-scale edaphic and hydrologic factors to consider when prioritizing and managing sites for restoration, and how site disturbance under a variety of land uses may impact different hydrologic processes and components of the wetland hydroperiod. Additionally, our results provide a better understanding of how land use affects seasonal runoff generation processes in mineral flat wetlands, and the water quality implications of modifying groundwater and surface water connectivity between mineral flats and surrounding surface drainage networks. Title: The effects of land use on mineral flat wetland hydrologic processes in lowland agricultural catchments Author: Marshall, Sarah M. (Sarah Marie) Hydrologic processes within mineral flat wetlands, along with their connections to groundwater and downstream surface water in lowland agricultural catchments are poorly understood, particularly under different land uses. In the three field studies included in this thesis, we examined infiltration, wetland hydroperiod, groundwater recharge dynamics, surface runoff generation, and water quality in mineral flat wetlands using a combination of soil and hydrometric measurements, stable isotope tracers, and water chemistry analysis. Our overarching objectives were to examine, for mineral flat wetlands under native prairie, farmed grassland, and restored prairie land cover: 1) how different land management influences infiltration and wetland hydroperiod at the plot scale, 2) the effects of land use on seasonal groundwater-surface water dynamics at the field scale, and 3) seasonal variation in runoff sources and nutrient transport from native prairie and farmed wetlands at the small catchment scale. At the plot scale, our results suggest that edaphic factors, particularly those related to soil structure, are strongly associated with wetland infiltration and overall hydroperiod across least-altered prairie, farmed, and restored prairie mineral flat wetlands. The hydroperiod metrics we examined were generally more sensitive to level of site disturbance than land use alone. At the field scale, our results indicate that, in spite of land use differences and slight variations in soil stratigraphy, many similarities exist in overall wetland hydroperiod, water sources and evaporation rates for mineral flat wetlands in the Willamette Valley lowlands. Isotopic evidence suggests that the greatest degree of groundwater-surface water mixing occurs in the upper 0.5 m of the saturated soil profile across sites under all land uses. Finally, at the small catchment scale, farmed wetland runoff was isotopically similar to field surface water for most of the wet season, indicating that saturation excess was an important runoff generation process. Prairie wetland runoff was isotopically similar to upstream water throughout the winter, and briefly similar to shallow groundwater and surface water within the wetland in mid-spring. Throughout the wet season, elevated nitrate, sulfate, and chloride concentrations were observed in groundwater and surface water at the farm site, and deeper groundwater at the prairie site. Upstream-downstream runoff chemistry remained similar throughout the wet season at the prairie site. Farm site runoff chemistry reflected the dominant water source within the farm field throughout the wet season. Our findings suggest that, while surface water pathways dominate runoff from wetland flats under farm land use, large wetland flat fields have a high potential to absorb, store, and process nutrients and agrochemicals from on-site and nearby off-site chemical inputs. Mineral flats that maintain wetland hydrology in spite of farm use represent a unique balance between agricultural production and preservation of some of the water storage and delay, and water quality-related ecosystem services once provided at a much larger scale in the Willamette Valley lowlands. We anticipate that results of this work will lead to better understanding of key site-scale edaphic and hydrologic factors to consider when prioritizing and managing sites for restoration, and how site disturbance under a variety of land uses may impact different hydrologic processes and components of the wetland hydroperiod. Additionally, our results provide a better understanding of how land use affects seasonal runoff generation processes in mineral flat wetlands, and the water quality implications of modifying groundwater and surface water connectivity between mineral flats and surrounding surface drainage networks. Close

• 2. [Article] Water quality effects of hyporheic processing

  "PB2000-107420."

  Citation × Citation Citation Abstract Copy Title: Water quality effects of hyporheic processing Author: Wigington, P.J., Landers, Dixon, Fernald, Alexander "PB2000-107420." Title: Water quality effects of hyporheic processing Author: Wigington, P.J., Landers, Dixon, Fernald, Alexander "PB2000-107420." Close

• 3. [Article] Environmental temperature sensing using Raman spectra DTS fiber-optic methods

  This is the publisher’s final pdf. The published article is copyrighted by American Geophysical Union and can be found at: http://sites.agu.org/.

  Citation × Citation Citation Abstract Copy Title: Environmental temperature sensing using Raman spectra DTS fiber-optic methods Author: Torgersen, Thomas, Schladow, S. Geoffrey, Hausner, Mark B., Hatch, Christine E., Selker, John S., Tyler, Scott W., Thodal, Carl E. This is the publisher’s final pdf. The published article is copyrighted by American Geophysical Union and can be found at: http://sites.agu.org/. Title: Environmental temperature sensing using Raman spectra DTS fiber-optic methods Author: Torgersen, Thomas, Schladow, S. Geoffrey, Hausner, Mark B., Hatch, Christine E., Selker, John S., Tyler, Scott W., Thodal, Carl E. This is the publisher’s final pdf. The published article is copyrighted by American Geophysical Union and can be found at: http://sites.agu.org/. Close

• 4. [Article] Power-law residence time distribution in the hyporheic zone of a 2nd-order mountain stream

  This is the publisher’s final pdf. The published article is copyrighted by the American Geophysical Union and can be found at: http://www.agu.org/journals/gl/.

  Citation × Citation Citation Abstract Copy Title: Power-law residence time distribution in the hyporheic zone of a 2nd-order mountain stream Author: Haggerty, Roy, Wondzell, Steven M., Johnson, Matthew A. This is the publisher’s final pdf. The published article is copyrighted by the American Geophysical Union and can be found at: http://www.agu.org/journals/gl/. Title: Power-law residence time distribution in the hyporheic zone of a 2nd-order mountain stream Author: Haggerty, Roy, Wondzell, Steven M., Johnson, Matthew A. This is the publisher’s final pdf. The published article is copyrighted by the American Geophysical Union and can be found at: http://www.agu.org/journals/gl/. Close

• 5. [Article] Subsurface flow of a forested riparian area in the Oregon coast range

  This study was undertaken to gain further understanding of the subsurface hydrology for a stream-adjacent riparian area in Western Oregon's Coast Range. Spatial and temporal fluctuations of the free water surface ...

  Citation × Citation Citation Abstract Copy Title: Subsurface flow of a forested riparian area in the Oregon coast range Author: Reiter, Maryanne This study was undertaken to gain further understanding of the subsurface hydrology for a stream-adjacent riparian area in Western Oregon's Coast Range. Spatial and temporal fluctuations of the free water surface of a toeslope, adjacent riparian area, and stream channel in a forested terrace reach were monitored over a period of one year. A total of 27 piezometers were installed (nine in each of three transects established perpendicular to the stream) in a stream-adjacent terrace and hillslope. These piezometers were monitored from September 26, 1989 to July 25, 1990. Results indicate that the direction of flow within a forested terrace can vary throughout the year. During the drier months of September and October, flow direction was generally towards the stream. However, by November the direction of flow for those areas closest to the stream had begun to change. In some cases the direction of flow changed up to 180° from the September direction. The location of influent/effluent zones along the stream (i.e., zones along the stream where the terrace has a lower or higher hydraulic head than the stream, respectively) also varied throughout the year. During October, 1989 the furthest upvalley stream-adjacent piezometer had hydraulic heads greater than the stream for the entire month, while two other stream adjacent piezometers did not. During normal precipitation in January, 1990 both furthest upvalley and the middle stream-adjacent piezometers had hydraulic heads greater than the stream for the entire month. By July, 1990 only the middle stream-adjacent piezometer had a hydraulic head greater than the stream for the entire month, while the furthest upvalley piezometer had a higher head for 33% of the month and the hydraulic head of furthest downvalley piezometer never exceeded that of the stream. For the Deer Creek study site, subsurface velocities were estimated to be in the range of 10-8 to 10-11 m/s resulting in a 3.8 year minimum travel time for storm water to reach the stream. Thus, soil matrix velocities using the Darcy equation were not sufficient to generate stormflow. However, no overland flow was observed during storm events, indicating that alternative subsurface flow pathways, such as macrochannels, are being utilized. Previous research has suggested that the release of water from terrace storage in small headwater streams is no.t sufficient to maintain baseflow. The results of the Deer Creek study support this conclusion. Furthermore, the results of this study indicate that forested riparian areas are hydrologically complex with respect to both space and time and that oversimplification of these systems may lead to misinterpretation of other processes associated with subsurface water dynamics. Title: Subsurface flow of a forested riparian area in the Oregon coast range Author: Reiter, Maryanne This study was undertaken to gain further understanding of the subsurface hydrology for a stream-adjacent riparian area in Western Oregon's Coast Range. Spatial and temporal fluctuations of the free water surface of a toeslope, adjacent riparian area, and stream channel in a forested terrace reach were monitored over a period of one year. A total of 27 piezometers were installed (nine in each of three transects established perpendicular to the stream) in a stream-adjacent terrace and hillslope. These piezometers were monitored from September 26, 1989 to July 25, 1990. Results indicate that the direction of flow within a forested terrace can vary throughout the year. During the drier months of September and October, flow direction was generally towards the stream. However, by November the direction of flow for those areas closest to the stream had begun to change. In some cases the direction of flow changed up to 180° from the September direction. The location of influent/effluent zones along the stream (i.e., zones along the stream where the terrace has a lower or higher hydraulic head than the stream, respectively) also varied throughout the year. During October, 1989 the furthest upvalley stream-adjacent piezometer had hydraulic heads greater than the stream for the entire month, while two other stream adjacent piezometers did not. During normal precipitation in January, 1990 both furthest upvalley and the middle stream-adjacent piezometers had hydraulic heads greater than the stream for the entire month. By July, 1990 only the middle stream-adjacent piezometer had a hydraulic head greater than the stream for the entire month, while the furthest upvalley piezometer had a higher head for 33% of the month and the hydraulic head of furthest downvalley piezometer never exceeded that of the stream. For the Deer Creek study site, subsurface velocities were estimated to be in the range of 10-8 to 10-11 m/s resulting in a 3.8 year minimum travel time for storm water to reach the stream. Thus, soil matrix velocities using the Darcy equation were not sufficient to generate stormflow. However, no overland flow was observed during storm events, indicating that alternative subsurface flow pathways, such as macrochannels, are being utilized. Previous research has suggested that the release of water from terrace storage in small headwater streams is no.t sufficient to maintain baseflow. The results of the Deer Creek study support this conclusion. Furthermore, the results of this study indicate that forested riparian areas are hydrologically complex with respect to both space and time and that oversimplification of these systems may lead to misinterpretation of other processes associated with subsurface water dynamics. Close

• 6. [Article] Characterization of hydrologic parameters and processes in shrink-swell clay soils

  Vertisols and other vertic-intergrade soils are found all over the globe, including many agricultural and urban areas. These soils are characterized by their cyclical shrinking and swelling behaviors, ...

  Citation × Citation Citation Abstract Copy Title: Characterization of hydrologic parameters and processes in shrink-swell clay soils Author: Stewart, Ryan D. Vertisols and other vertic-intergrade soils are found all over the globe, including many agricultural and urban areas. These soils are characterized by their cyclical shrinking and swelling behaviors, where bulk density and porosity distribution both vary as functions of time and/or soil moisture. In turn, alterations in physical soil parameters become manifest as crack networks, which open during the shrinkage phase and re-seal as the soil swells. As a consequence, when present these soils can significantly impact agriculture and infrastructure, and can act as a dominant control on the local hydrology. Therefore, understanding how water moves through and interacts with these soils is of utmost importance to ensure their proper utilization and management. This study had three objectives: 1) to identify the role of initial soil moisture on infiltration in a rigid (non-swelling) soil; 2) to quantify variation in the soil's hydraulic properties, using single-ring infiltration measurements taken in a vertic soil over a 1.5 year period; and 3) to build on these concepts to examine data from a set of instrumented field plots in a vertic soil located in the Secano Interior region of Chile. Significant findings of the study include 1) the development of a new formulation to describe a soil's wetting front potential (a measure of capillary pull) in terms of initial degree of saturation; 2) the creation of a simple correction to approximate the decrease of wetting front potential in wetter soils, thereby improving the accuracy of the traditional Green and Ampt sorptivity model in such conditions; 3) an equation to accurately estimate hydraulic parameters from short-duration infiltration tests, when steady-state conditions have not been realized; 4) a new theoretical model to estimate crack porosity as a function of soil moisture, which was developed based on the soil shrinkage curve; and 5) modification of the traditional two-term Philip infiltration model for use in vertic soils. The field study also showed that crack networks cause highly complex and non-linear wetting of the soil profile, with water simultaneously infiltrating from the soil surface and from the soil-crack interfaces; that cracks can seal at the surface while remaining open and hydraulically active below the surface, which indicates that surface-based monitoring alone may not be sufficient to predict water movement and soil response in vertic soils; and that the transition between infiltration and runoff may be strongly correlated with the cumulative amount of net precipitation that had reached the soil surface, so that cumulative amount of precipitation has the potential to be a simple yet accurate metric to predict runoff in vertic soils. These findings offer improved understanding of soil-water interactions in vertic soils, and reveal that very simple concepts underlie seemingly complex systems. As a result, the concepts and formulations developed in this study should allow for straightforward integration into other studies and models. Title: Characterization of hydrologic parameters and processes in shrink-swell clay soils Author: Stewart, Ryan D. Vertisols and other vertic-intergrade soils are found all over the globe, including many agricultural and urban areas. These soils are characterized by their cyclical shrinking and swelling behaviors, where bulk density and porosity distribution both vary as functions of time and/or soil moisture. In turn, alterations in physical soil parameters become manifest as crack networks, which open during the shrinkage phase and re-seal as the soil swells. As a consequence, when present these soils can significantly impact agriculture and infrastructure, and can act as a dominant control on the local hydrology. Therefore, understanding how water moves through and interacts with these soils is of utmost importance to ensure their proper utilization and management. This study had three objectives: 1) to identify the role of initial soil moisture on infiltration in a rigid (non-swelling) soil; 2) to quantify variation in the soil's hydraulic properties, using single-ring infiltration measurements taken in a vertic soil over a 1.5 year period; and 3) to build on these concepts to examine data from a set of instrumented field plots in a vertic soil located in the Secano Interior region of Chile. Significant findings of the study include 1) the development of a new formulation to describe a soil's wetting front potential (a measure of capillary pull) in terms of initial degree of saturation; 2) the creation of a simple correction to approximate the decrease of wetting front potential in wetter soils, thereby improving the accuracy of the traditional Green and Ampt sorptivity model in such conditions; 3) an equation to accurately estimate hydraulic parameters from short-duration infiltration tests, when steady-state conditions have not been realized; 4) a new theoretical model to estimate crack porosity as a function of soil moisture, which was developed based on the soil shrinkage curve; and 5) modification of the traditional two-term Philip infiltration model for use in vertic soils. The field study also showed that crack networks cause highly complex and non-linear wetting of the soil profile, with water simultaneously infiltrating from the soil surface and from the soil-crack interfaces; that cracks can seal at the surface while remaining open and hydraulically active below the surface, which indicates that surface-based monitoring alone may not be sufficient to predict water movement and soil response in vertic soils; and that the transition between infiltration and runoff may be strongly correlated with the cumulative amount of net precipitation that had reached the soil surface, so that cumulative amount of precipitation has the potential to be a simple yet accurate metric to predict runoff in vertic soils. These findings offer improved understanding of soil-water interactions in vertic soils, and reveal that very simple concepts underlie seemingly complex systems. As a result, the concepts and formulations developed in this study should allow for straightforward integration into other studies and models. Close

• 7. [Article] Improving soil water determination in spatially variable field using fiber optic technology and Bayesian decision theory

  Achieving and maintaining sustainability in irrigated agriculture production in the era of rapidly increasing stress on our natural resources require, among other essential actions, optimum control and ...

  Citation × Citation Citation Abstract Copy Title: Improving soil water determination in spatially variable field using fiber optic technology and Bayesian decision theory Author: Sayde, Chadi Achieving and maintaining sustainability in irrigated agriculture production in the era of rapidly increasing stress on our natural resources require, among other essential actions, optimum control and management of the applied water. Thus, a significant upgrade of the currently available soil water monitoring technologies is needed. The primary goal of this work was to reduce the uncertainties of spatially variable soil water in the field. Two approaches are suggested: 1) The Bayesian decision model that implicitly accounts for spatial variability at minimal cost based on limited field data, and 2) The Actively Heated Fiber Optic (AHFO) method that explicitly accounts for spatial variability with high sampling density at relatively low cost per measurement point. The Bayesian decision model uses an algorithm to integrate information embodied in independent estimates of soil water depletion to derive a posterior estimation of soil water status that has the potential to reduce the risk of costly errors in irrigation scheduling decisions. The sources of information are obtained from an ET based water balance model, soil water measurements, and expert opinion. The algorithm was tested in a numerical example based on a field experiment where soil water depletion measurements were made at 43 sites in an agricultural field under center pivot irrigation. The results showed that the estimates of the average soil water depletion in the field obtained from the posterior distributions of soil water depletion proved to outperform simple averaging of n soil water depletion measurements, up to n = 35 measurements. For n< 3, the model also provided a 39% average reduction in risk of error derived from non-representative measurements. The AHFO method observes the heating and cooling of a buried fiber optic (FO) cable through the course of a pulse application of energy as monitored by a distributed temperature sensing (DTS) system to reveal soil water content simultaneously at sub-meter scale along the FO cable that can potentially exceeds kilometers in length. A new and simple interpretation of heat data that takes advantage of the characteristics of FO temperature measurements is presented. The results demonstrate the feasibility of AHFO method application to obtain <0.05 m³m⁻³ error distributed measurements of soil water content under laboratory controlled conditions. The AHFO method was then tested under field conditions using 750 m of FO cables buried at 30, 60, and 90 cm depths in agricultural field. The calibration curve relating soil water content to the thermal response of the soil to a heat pulse was developed in the lab. It was successively applied to the 30 and 60 cm depths cables, while the 90 cm depth cable illustrated the challenges of soil heterogeneity for this technique. The method was used to map with high spatial (1m) and temporal (1hr) resolution the spatial variability of soil water content and fluxes induced by the non-uniformity of water application at the surface. Title: Improving soil water determination in spatially variable field using fiber optic technology and Bayesian decision theory Author: Sayde, Chadi Achieving and maintaining sustainability in irrigated agriculture production in the era of rapidly increasing stress on our natural resources require, among other essential actions, optimum control and management of the applied water. Thus, a significant upgrade of the currently available soil water monitoring technologies is needed. The primary goal of this work was to reduce the uncertainties of spatially variable soil water in the field. Two approaches are suggested: 1) The Bayesian decision model that implicitly accounts for spatial variability at minimal cost based on limited field data, and 2) The Actively Heated Fiber Optic (AHFO) method that explicitly accounts for spatial variability with high sampling density at relatively low cost per measurement point. The Bayesian decision model uses an algorithm to integrate information embodied in independent estimates of soil water depletion to derive a posterior estimation of soil water status that has the potential to reduce the risk of costly errors in irrigation scheduling decisions. The sources of information are obtained from an ET based water balance model, soil water measurements, and expert opinion. The algorithm was tested in a numerical example based on a field experiment where soil water depletion measurements were made at 43 sites in an agricultural field under center pivot irrigation. The results showed that the estimates of the average soil water depletion in the field obtained from the posterior distributions of soil water depletion proved to outperform simple averaging of n soil water depletion measurements, up to n = 35 measurements. For n< 3, the model also provided a 39% average reduction in risk of error derived from non-representative measurements. The AHFO method observes the heating and cooling of a buried fiber optic (FO) cable through the course of a pulse application of energy as monitored by a distributed temperature sensing (DTS) system to reveal soil water content simultaneously at sub-meter scale along the FO cable that can potentially exceeds kilometers in length. A new and simple interpretation of heat data that takes advantage of the characteristics of FO temperature measurements is presented. The results demonstrate the feasibility of AHFO method application to obtain <0.05 m³m⁻³ error distributed measurements of soil water content under laboratory controlled conditions. The AHFO method was then tested under field conditions using 750 m of FO cables buried at 30, 60, and 90 cm depths in agricultural field. The calibration curve relating soil water content to the thermal response of the soil to a heat pulse was developed in the lab. It was successively applied to the 30 and 60 cm depths cables, while the 90 cm depth cable illustrated the challenges of soil heterogeneity for this technique. The method was used to map with high spatial (1m) and temporal (1hr) resolution the spatial variability of soil water content and fluxes induced by the non-uniformity of water application at the surface. Close

• 8. [Article] Funneled flow mechanisms in a sloping layered soil: Laboratory investigation

  This is the publisher’s final pdf. The published article is copyrighted by American Geophysical Union and can be found at: http://sites.agu.org/.

  Citation × Citation Citation Abstract Copy Title: Funneled flow mechanisms in a sloping layered soil: Laboratory investigation Author: Braddock, R. D., Selker, J. S., Steenhuis, T. S., Parlange, J.-Y., Kim, J.-S., Boll, J., Walter, M. T., Heilig, A. This is the publisher’s final pdf. The published article is copyrighted by American Geophysical Union and can be found at: http://sites.agu.org/. Title: Funneled flow mechanisms in a sloping layered soil: Laboratory investigation Author: Braddock, R. D., Selker, J. S., Steenhuis, T. S., Parlange, J.-Y., Kim, J.-S., Boll, J., Walter, M. T., Heilig, A. This is the publisher’s final pdf. The published article is copyrighted by American Geophysical Union and can be found at: http://sites.agu.org/. Close

• 9. [Article] Flow processes in the dry regime : the effect on capillary barrier performance

  Engineered capillary barriers typically consist of two layers of granular materials designed so that the contrast in sediment hydrologic properties and sloping interface retains infiltrating water in the ...

  Citation × Citation Citation Abstract Copy Title: Flow processes in the dry regime : the effect on capillary barrier performance Author: Jansik, Danielle P. Engineered capillary barriers typically consist of two layers of granular materials designed so that the contrast in sediment hydrologic properties and sloping interface retains infiltrating water in the upper layer. We report here on the results of two bench-top capillary barrier experiments, and associated modeling. These experiments were conducted to better understand the behavior of capillary barriers in particular with respect to an engineered barrier system proposed at the potential high-level nuclear waste site at Yucca Mountain, Nevada, where different engineered barriers have been under consideration throughout the site development process (Carter and Pigford 2005; Tidwell et al. 2003). We measured hydrologic parameters for both coarse materials using standard methods and found that the two materials had similar hydrologic properties despite being morphologically different (round, uniform vs. angular, non-uniform). The rounded sand provided a better functioning capillary barrier than the angular sand, but neither experiment could be characterized as a perfectly working capillary barrier. In both cases, more than 93% of the infiltrating water was successfully diverted from the lower layer, but infiltration into the underlying layer was observed in both systems. Our experimental results show that capillary barriers designed based on standard hydrologic property measurements may result in invalid prediction of the system behavior. Moreover, our numerical simulations of these experiments show that the predicted capillary barrier performance was highly sensitive to physically realistic variability in model hydrologic parameter values. Based on this work, we believe that other non-continuum processes such as vapor diffusion and film flow contribute to the observed phenomena and are important aspects to consider with respect to capillary barrier design, as well as dry vadose zone processes in general. By applying a theoretical film flow equation representative of sediment surface geometries we were able to show that infiltration into the underlying sediment layer can be dominated by water film flow, a characteristic that is typically not considered in mainstream numerical models. Title: Flow processes in the dry regime : the effect on capillary barrier performance Author: Jansik, Danielle P. Engineered capillary barriers typically consist of two layers of granular materials designed so that the contrast in sediment hydrologic properties and sloping interface retains infiltrating water in the upper layer. We report here on the results of two bench-top capillary barrier experiments, and associated modeling. These experiments were conducted to better understand the behavior of capillary barriers in particular with respect to an engineered barrier system proposed at the potential high-level nuclear waste site at Yucca Mountain, Nevada, where different engineered barriers have been under consideration throughout the site development process (Carter and Pigford 2005; Tidwell et al. 2003). We measured hydrologic parameters for both coarse materials using standard methods and found that the two materials had similar hydrologic properties despite being morphologically different (round, uniform vs. angular, non-uniform). The rounded sand provided a better functioning capillary barrier than the angular sand, but neither experiment could be characterized as a perfectly working capillary barrier. In both cases, more than 93% of the infiltrating water was successfully diverted from the lower layer, but infiltration into the underlying layer was observed in both systems. Our experimental results show that capillary barriers designed based on standard hydrologic property measurements may result in invalid prediction of the system behavior. Moreover, our numerical simulations of these experiments show that the predicted capillary barrier performance was highly sensitive to physically realistic variability in model hydrologic parameter values. Based on this work, we believe that other non-continuum processes such as vapor diffusion and film flow contribute to the observed phenomena and are important aspects to consider with respect to capillary barrier design, as well as dry vadose zone processes in general. By applying a theoretical film flow equation representative of sediment surface geometries we were able to show that infiltration into the underlying sediment layer can be dominated by water film flow, a characteristic that is typically not considered in mainstream numerical models. Close

• 10. [Article] A macroscale measurement and modeling approach to improve understanding of the hydrology of steep, forested hillslopes

  The dominant controls on flow generation in steep, forested hillslopes are poorly understood. This dissertation examined the dominant flow processes operating at the hillslope scale, using a combined ...

  Citation × Citation Citation Abstract Copy Title: A macroscale measurement and modeling approach to improve understanding of the hydrology of steep, forested hillslopes Author: Graham, Christopher Brian The dominant controls on flow generation in steep, forested hillslopes are poorly understood. This dissertation examined the dominant flow processes operating at the hillslope scale, using a combined macroscale measurement and model development and analysis framework. Irrigation experiments at two steep forested hillslopes were conducted to isolate individual hillslope flow components and reveal the dominant controls on flow routing to the stream. A new perceptual model of flow processes at the hillslope scale was developed from these field experiments that included three key components: 1) A connected preferential flow network located at the soil/bedrock interface controls lateral water and solute transport, 2) The bedrock surface controls the subsurface flow routing, and 3) Bedrock is permeable, and acts as a sink for precipitation at the hillslope scale. These components formed the basis of a new, low dimensional numerical model that was used to represent, quantitatively, qualitative experimental findings. A multiple criteria calibration using hydrometric and tracer experimental data was conducted to evaluate the model and determine parameter identifiability. The model was able to adequately reproduce both hydrometric response to precipitation, and a tracer application breakthrough. The model was then used within a virtual experiment framework to test the dominant controls on the whole storm precipitation / discharge threshold relationship at the catchment scale. The modeling experiments showed that the macroscale precipitation / discharge threshold was controlled by a balance between antecedent evaporation (evaporation rate times antecedent drainage time), and geologic factors (bedrock permeability and subsurface storage volume). Overall, these findings suggest that making measurements at the scale of the processes one wishes to understand, and constructing numerical models based on the dominant processes at the same scale may be a first step towards the development of macroscale laws of hillslope behavior. Title: A macroscale measurement and modeling approach to improve understanding of the hydrology of steep, forested hillslopes Author: Graham, Christopher Brian The dominant controls on flow generation in steep, forested hillslopes are poorly understood. This dissertation examined the dominant flow processes operating at the hillslope scale, using a combined macroscale measurement and model development and analysis framework. Irrigation experiments at two steep forested hillslopes were conducted to isolate individual hillslope flow components and reveal the dominant controls on flow routing to the stream. A new perceptual model of flow processes at the hillslope scale was developed from these field experiments that included three key components: 1) A connected preferential flow network located at the soil/bedrock interface controls lateral water and solute transport, 2) The bedrock surface controls the subsurface flow routing, and 3) Bedrock is permeable, and acts as a sink for precipitation at the hillslope scale. These components formed the basis of a new, low dimensional numerical model that was used to represent, quantitatively, qualitative experimental findings. A multiple criteria calibration using hydrometric and tracer experimental data was conducted to evaluate the model and determine parameter identifiability. The model was able to adequately reproduce both hydrometric response to precipitation, and a tracer application breakthrough. The model was then used within a virtual experiment framework to test the dominant controls on the whole storm precipitation / discharge threshold relationship at the catchment scale. The modeling experiments showed that the macroscale precipitation / discharge threshold was controlled by a balance between antecedent evaporation (evaporation rate times antecedent drainage time), and geologic factors (bedrock permeability and subsurface storage volume). Overall, these findings suggest that making measurements at the scale of the processes one wishes to understand, and constructing numerical models based on the dominant processes at the same scale may be a first step towards the development of macroscale laws of hillslope behavior. Close