Water Resources Collections and Archives

Aquatic ecosystems in the Upper Klamath Basin (Upper Basin), Oregon are degraded as a result of more than a century of land use alterations due to logging, dams, irrigated agriculture, and cattle grazing. These changes have led to degraded habitat conditions including decreased baseflow, loss of vegetation, increased stream temperature, fish impediments, and nutrient loading. All these factors negatively impact watershed function and resident fish populations, which have experienced severe declines in recent decades. The primary threats to fish populations include habitat loss, degraded water quality, barriers and entrainment, and predation and competition from non-native species. Millions of dollars have been spent since the late-1900’s to restore aquatic habitat in the Upper Basin primarily to improve the distribution and abundance of endangered and threatened fish species. This project details the hydrologic characteristics of three primary tributaries in the Upper Klamath Lake Sub-Basin including Sevenmile Creek, Wood River, and the Williamson River (including spring tributaries). Available discharge data was assembled to plot seasonal fluctuations in flows and identify annual peak flow at different re-occurrence intervals. Stream systems in the UKL Sub-Basin show a range of hydrologic inflow due to groundwater and/or snow-melt run-off. Characteristics of spring-fed vs. run-off dominated stream systems are reviewed and recommendations are made for how to address restoration practices considering the hydrologic and geomorphic characteristics of stream channels.

Livestock grazing in the western United States has lead to riparian ecosystem and stream channel degradation. Establishing fenced-off exclosures is a common management strategy that aims to passively restore these areas, however, relatively few studies have assessed the evolution of exclosed reaches over time. We evaluated temporal trends in channel form and riparian vegetation along a 4.2 km reach of Long Creek (drainage area of 180 km2), a tributary to Sycan Marsh in western Lake County, Oregon. The Nature Conservancy implemented reduced livestock grazing along this reach in 1996 and complete exclosure in 1999. Based on previous studies that documented vegetation establishment and subsequent sediment accumulation on channel banks, we hypothesized that the channel would have narrowed and vegetation would have re-established after eleven years of cattle exclosure.   In October 2011, we surveyed seven previously-established cross sections and compared channel geometry in 2011 to surveys conducted since 1990. We also took photos at these locations and compared them to historical images. We found no consistent trend in channel morphology, but two cross sections demonstrated narrowing, one of which was likely driven by reduced flow velocity from a downstream beaver dam. Vegetation had successfully established along the streambanks, but in addition to the riparian species, upland lodgepole pines were also abundant along the channel and on the floodplain.  Our results suggest that beavers should be encouraged so their dams will increase overbank flow and discourage the invasion of upland plant species. We also propose improvements in monitoring methods to ensure repeatability of cross-sections over an extended monitoring period.  

Abstract

Putah Creek (drainage area = 2,000 km2) drains the slopes of Cobb Mountain in Lake County, flowing 137 km southeastward into the Yolo Bypass near Davis, California. Lower Putah Creek, the 37km reach from the Solano Diversion Dam to the Yolo Bypass, is confined within a flood control channel.  Dry Creek (drainage area = 44km2) joins Putah Creek near Winters, California.  Putah Creek is regulated by water releases from Monticello Dam at Lake Berryessa.  Dry Creek flows only part of the year and has no dams. Southward channel migration of Putah Creek from the 1990’s was threatening Putah Creek Rd., a paved county road following the south bank of the evaluated stream-section.  The new location of the Putah Creek channel also reduced the amount of gravels entering Putah Creek from Dry Creek.  This was significant because dams reduced the amount of course sediment available from upstream, leaving Dry Creek as one of the only natural sources of the gravels important to the ecology of Putah Creek. In 2005, the Lower Putah Creek Coordinating Committee implemented a project to move the channel of Putah Creek northward to its approximate historical course.  The project included the removal of invasive giant reed (Arundo donax), in part because Arundo appeared to have contributed to the unwanted channel avulsion.  We evaluated the project performance towards the goals of 1. protecting Putah Creek Road, 2. keeping invasive plant species out of the area, 3. facilitating natural transport of Dry Creek gravels into Putah Creek, and 4. improving salmonid habitat.  Our evaluation found that 1. Putah Creek has stayed within the general path of the design channel and is not returning to the southern pre-1997 channel which threatened the road, 2. there is a mix of native plant species (e.g. willow, cottonwood) and invasives (e.g. Arundo, blackberry) in the floodplain, 3. There is no physical barrier separating Putah and Dry Creek, 4.  the channel and floodplain provide good habitat complexity for native fish, but fine sediments cover most of the potential spawning gravels.

The simultaneous withdrawal of water from streams for springtime frost protection of grapevines in the Russian River basin can coincide with the emergence of salmonid fry and the rearing of juveniles. These water diversions have contributed to water level declines, which in some instances, have resulted in the stranding mortality of fish. Endangered coho salmon and threatened steelhead trout can become stranded when water levels decrease abruptly and fish seek refuge in the rapidly dewatering gravel.

In response to this issue, the National Marine Fisheries Service (NMFS) has proposed a site-specific method to determine minimum flows to protect salmonids from these effects. This method seeks to identify “high risk” stranding surfaces and determine the stream stage at which they become exposed. In this study, we evaluated the ability of the NMFS protocol to accurately prescribe protective stages. To do this, we analyzed three components of the protocol: its stranding risk classification system, it’s sampling of stranding surfaces and its method of establishing protective stage recommendations.

We evaluated the risk classification system by comparing it to published literature values on salmonid stranding. We assessed the sampling of stranding surfaces by performing the protocol at two sites. NMFS developed the method based on data from a medium-sized drainage (12.6 mi2), so we selected a small drainage (4.6 mi2) and a large drainage (50.2 mi2) to evaluate how effectively the method characterized the variation in potential stranding surfaces in different watershed settings. We evaluated the protocol’s protective stage recommendation by comparing the protective stage from our two surveyed sites to stream stage data for the season of regulation. 

Our assessment has led us to make several recommendations. First, the risk classification system would benefit from consideration of other factors influencing stranding risk and should adjust stranding risk thresholds to better fit the literature. Also, the protocol is weak in its ability to capture within-site variation. We therefore recommend increased sampling of stream reaches and scaled mapping of each site to better define stranding surfaces. These measures should result in improved protective stage recommendations but further studies may be necessary. With these changes, we believe that the NMFS protocol will be an effective tool for protecting fish from being stranded due to vineyard use of water during frost events in the Russian River Watershed.

In Fall 2010, a partnership between the University of California-Berkeley and the cities of Albany and Berkeley completed the third of four restoration phases planned for a 0.6-mile stretch of Codornices Creek in Alameda County, California, between the San Pablo Avenue and UPRR crossings. Originally initiated in the mid-1990s to improve a straightened and channelized ditch, the project objectives were to convey the 100-year flood, improve user access to the creek, and establish an ecologically valuable riparian corridor dominated by native species (reducing invasive non-natives).  We assessed the performance of the third phase of the project during a high flow of 136 cfs on October 5, 2011. We obtained relevant data and information from project designers, and on October 22, 2011, while evidence of the high flow was still fresh, we conducted a detailed topographic survey of the channel, surveyed high water marks, documented conditions with photographs, and mapped site conditions. In addition, we surveyed cross sections and high water marks in the downstream reaches (Phases 1 and 2 of the overall restoration project).  High water marks show floodplain inundation was inconsistent throughout the three reaches, with the October 5 storm flow largely staying within the constructed banks in Phase 3, and overbank flow occurring in Phases 1 and 2.   Our longitudinal profile shows Phase 3 incised up to 2 ft below the design grade in the upstream portions of the reach, and aggraded up to 2 ft at the downstream end. Survey results also confirm that additional vertical channel adjustment occurred during the October 5 flow. This, along with the presence of an active headcut, suggests that the channel is still in the process of finding geomorphic equilibrium. Cross-section monitoring in Phase 3 should proceed into the future to determine whether channel adjustments continue, and as a basis to assess whether more complexity should be introduced to promote aggradation, channel complexity, floodplain inundation, and more ecologically valuable habitat.