The River Research Group at the University of Oregon has strengths in river restoration, river environment monitoring and modeling, watershed hydrology and geomorphology, human-river interrelationships, and theoretical fluvial science. Some of our current and past projects are below.
Channel Restoration Monitoring, Middle Fork John Day River, Oregon
A stronghold for wild salmon, the Middle Fork of the John Day River (MFJD) has been affected by historical gold mining, overgrazing by cattle, channel armoring and straightening, and riparian vegetation reduction. For several decades it has been a priority basin for aquatic habitat restoration, and a number of channel restoration projects (engineered log jams, removal of rip-rap and rock structures, planting of woody vegetation on the banks and floodplain, and channel re-meandering) have been completed and are planned. Since 2008 we have been monitoring the effectiveness of these restoration projects in terms of geomorphology and physical habitat. (Other groups are monitoring other aspects of the restoration, including fish, aquatic macroinvertebrates, and water temperature.)
We have two main research goals.
• To understand the degree to which the restoration projects have achieved their ecological goals, and to understand why or why not. We are particularly interested in providing results that can inform the next generation of restoration work on the MFJD and other similar rivers.
• To develop cost-effective methods for river monitoring using new technology such as LiDAR, tablet computers, UAVs, structure-from-motion photogrammetry, and high-resolution aerial imagery.
Through repeat measurements, channel change following project construction is monitored. Monitoring techniques used in the MFJD include the following:
• Channel cross-section morphology (erosion/aggradation, W:D, incision)
• Bed material characteristics (surface and volumetric gravel counts, D50, D95, % fines, embeddedness, armoring)
• Channel planform characteristics (sinuosity, lateral migration)
• Longitudinal profile change
• Fish cover
• Pool frequency and depth
• Log structure morphology and channel effects (difference of DEM survey)
We conduct field work for several weeks each summer. The overall monitoring effort is directed by the MFJD Intensively Monitored Watershed program. Funding has been provided by the Oregon Watershed Enhancement Board, NOAA, and the Confederated tribes of the Warm Springs Reservation.
(McDowell [faculty research], Goslin [PhD dissertation])
Next-Generation Riverscape Mapping and Monitoring
Recent scientific advances demonstrate that river dynamics and habitats vary in the downstream direction not only in a smooth, gradual way, but also in a non-smooth, complex manner at a variety of spatial and temporal scales. This mosaic-like “riverscape” is a serious challenge for current theories regarding how rivers change through space and time, where habitats for specific organisms are or are not connected with one another, and how rivers ought to be managed. A knowledge gap exists in that our current tools and approaches for mapping rivers are providing inadequate characterizations, and we need to develop and test new methods that apply to a broad range of river environments. For this NSF-EAGER project we are cross-comparing two new field approaches, a simple drone-boat compared to the expertly-collected data using a fully outfitted research cataraft, which will allow us to elucidate the complex patterns in the structuring of rivers. Our tests will be carried out in rivers in the Pacific Northwest, but the methods will be applicable to a wide array of rivers worldwide. Our research objectives are to: (1) Develop and test a new, small drone-boat platform capable of collecting basic longitudinal river data with minimal user interactions (2) Compare and validate the drone boat data with expertly-collected data using a fully-outfitted research cataraft, and (3) See if it is feasible to employ a citizen science approach to mapping and monitoring of longitudinal riverscapes by deploying the drone boat with watershed groups in Oregon.
Spatial Patterning of Flow in Complex Perennial River Systems, Willamette River, Oregon
This research focuses on the topic of remote sensing of river flow variables and how these compare with those estimated by different types of models. This comparison is crucial for predicting river flows and flow levels and how these will impinge on society and infrastructure, such as in large floods. The research is driven by the following research questions:
(1) What are the spatial patterns of flow in a large river basin during different-sized flow events and what are the landscape configurations that explain these patterns?
(2) What are the significant differences and reasons for those differences between simulated flow patterns and those observed via high-resolution imagery?
(3) How does the spatial patterning of large river flow affect the quality of flow estimates made from remote observation platforms?
First, the high-resolution, river-length air photos taken by the US Army Corps of Engineers and others at the height of a large 1996 flood, coupled with topobathymetry, provides excellent data for testing models of high magnitude river flow patterning (such as discharge, water surface elevation and slope, water extent, depth, velocity, shear stress). Second, imagery flown as part of the 2015 AirSWOT campaign provides excellent data for testing models of medium magnitude flows. Third, instrumentation and associated field measurements of discharge and water surface extents should provide the needed data to test models of low- and medium-magnitude river flow patterning during the project period. Fourth, river models such as HEC-GeoRAS and LISFLOOD-FP, commonly used for management decisions and in global hydrology, are applicable to multiple model runs with different software and different parameters can provide an understanding of model uncertainty and sensitivity. Landsat and other satellite-based products are the key data from which we can estimate these spatially-varying ground conditions as well as estimates their likely errors. Finally, geospatial data coupled with multivariate statistical approaches allow us to characterize and understand the differences between model simulations and the observations at different discharge levels. (Fonstad [faculty research])
Social Construction of Floodplains in the United States
This research aims to gain a better understanding of the actors who can enact change on Flood Insurance Rate Maps (FIRMs) in the United States National Flood Insurance Programs (NFIP) and the relationship between socio-economic factors and successful changes. While all changes in updating FIRM flood zones must be approved by technical experts, proposed changes can be initiated by technical experts or property owners and other community members. But the relative influence of technical expert versus property owner-initiated change in the program is unknown. The research investigates this unknown by examining areas of change in the Special Flood Hazard Area (the area predicted to have a 1% chance of flooding annually) over different iterations of available FIRMs where changes can be attributed to either technical experts or property owners. These changes will then be compared to socio-economic data from US Census Data and assessed property value data to see if economic inequality is related to the ability to enact successful changes and how property values are influenced by changes in flood designation over time. The results of this research will provide theoretical contribution to who, why, and how these regulatory maps are updates, as well as a practical contribution to ongoing debates on FIRMs being “out of date” in their depiction of flood hazard (Lea [PhD dissertation]).
Willamette River Fieldwork in Support of NASA SWOT and AirSWOT
Sonar-based bathymetric mapping and hydraulic modeling of the Willamette River provides vital information that has been used to test estimates made by NASA’s AirSWOT instrument, a test instrument used to evaluate the SWOT satellite scheduled for launch in 2020. SWOT (Surface Water Ocean Topography) will map the extent and dynamics of earth’s surface water around the globe every 22 days. Algorithms for mapping river depth and discharge are being tested in different basins around the world, but the Willamette is a primary test basin for midlatitude river systems. (Fonstad [faculty research], Zettler-Mann [student research])
Human-River Interactions in Megafan Settings, India
This project focuses on understanding the linkages between structural geometry, morphometric variability, erosional and depositional processes and the scale of human interference that relates to the instability of channels in three megafan settings in actively aggrading basins of the Himalayan mountain front. One of the goals is to map the extent of floodplain disconnection across these megafans. Primary tools in this effort include the use of geographic information systems (GIS), remote sensing and easily obtainable topographic and climate data. This ongoing work will have important implications for the understanding of the spatial variability of water-logging and regional flood hazard. (Goswami [PhD dissertation])
Tropical Mountain River Dynamics, Pacuare River, Costa Rica
This research is focused on the geomorphology and sediment transport dynamics of tropical mountain rivers. Reach by reach sediment budgets are being estimated and compared to local geology and downstream fluvial processes on the Rio Pacuare. Modeling will be done to predict impacts of climate change and dam construction scenarios on the geomorphology of these dynamic systems. Data collection and analysis includes a unique combination of traditional survey techniques and Structure From Motion photogrammetry. (Lind [PhD dissertation])
Mountain Stream Dynamics and Mapping Techniques Development, Scott Creek Basin, Oregon
This project focuses on understanding the development of channel geometry and habitats in a high energy, High Cascade stream environment. To measure the stream channels longitudinally in such an extremely complex environment, new mapping techniques based on ground-based structure from motion and Kinect structured light mapping are being developed and tested. A combination of centimeter-scale mapping and agent-based modeling approaches will allow digital stream ecosystem simulation at the organism scale. (Fonstad [faculty research])
Digitizing Channel Change during Floods in the Umatilla River, Oregon
This project was conducted in 2003-05. The overall goal was to understand the spatial pattern and controls of channel changes resulting from large floods (>25 yr recurrence interval), and to determine whether human response to historic floods, such as bank hardening, limited the ability of rivers to change in subsequent floods. Large floods on alluvial rivers produce geomorphic changes that create and renew fish habitat and help regenerate riparian vegetation. Floods renew spawning gravels, scour and deepen pools, and create side channels and other off-channel features. In addition, cottonwoods and other streamside trees important to stream ecology depend on floods for regeneration. Using historical aerial photos, documentary evidence, and government agency records, we reconstructed and analyzed spatial patterns of geomorphic change on long continuous river reaches, constructing change maps in GIS. We also constructed GIS coverages of riverside flood protection structures — levees, revetments, and other human-built structures that restrict channel change. The study rivers were the Umatilla River and Applegate River in Oregon.