Analysis of Spawning Gravel Availability

Introduction

This purpose of this chapter is to address the long-term goal of stabilizing and enhancing the viability of the spawning habitat for spring and fall run salmon and steelhead trout in Big Chico Creek and Lindo Channel, specifically with respect to gravel beds suitable for spawning.

Big Chico Creek's Managed Flows

Big Chico Creek emerges onto the Chico fan at the Five-Mile Recreation Area on the northeast side of the Central Valley from a foothill canyon. Flows at Five-Mile are regulated for flood control by diversion of high flows from a single stilling basin into Big Chico Creek and two bypass channels: Lindo Channel and the Sycamore Creek Bypass Channel. The flow control occurs by three diversion structures shown in Figure 1.

View of stilling basin at Five-Mile Area during a flow of approximately
thirty-five cubic feet per second (cfs).

Figure 1

The invert elevations of Big Chico Creek and Lindo Channel diversion structures are similar, allowing both to carry summer low flows. However, the amount of water entering Lindo Channel, in particular during low flows, are strongly controlled by the size, height, and configuration of the gravel bar which forms within the stilling basin immediately upstream of the Big Chico Creek flow control structure. The included photo shows the stilling basin at the Five-Mile Area during a flow of approximately 35 cubic feet per second (cfs). From May to November of each year, there is little to no flow in Lindo Channel. Only large winter flows spill over the Sycamore Weir, when water is significantly ponded behind the Big Chico and Lindo Channel Flow Control Structures. Engineering plan and detail drawings of the diversion structures at the Five-Mile Area are provided in Figures 2 and 3.

Flows are controlled by pool and flood control maintenance and by the temporal variability of high-energy gravel transporting floods. The DFG has stated that it considers Big Chico Creek as the primary resource and has been reluctant to split flows between Big Chico Creek and Lindo Channel during the summer months (Mitchell Swanson & Associates [MSA], 1994, p38)

Flood Control: Primary Purpose of Hydraulic Management

The Chico and Mud Creeks and Sandy Gulch Improvement and Levee Construction Project was completed by the Corps of Engineers in 1965 to provide flood protection for the growing City of Chico. This work is detailed in the summary report "Operation and Maintenance Manual for Chico and Mud Creeks and Sandy Gulch, Sacramento River and Minor Tributaries Project, California, 1965". Since that time, the public interest and natural resource needs have changed to include streamside vegetation and instream flows for salmonid runs.

In the 1993 Flood Emergency Management Act (FEMA) study prepared for the City of Chico, Schaaf & Wheeler (S&W) estimated the 100-year flow in Big Chico Creek to be 14,000 cfs, with 4,000 cfs entering Lindo Channel and 4,600 cfs passing under the Esplanade. Other studies indicate that the 100-year flow of Big Chico Creek could be less than 14,000 cfs. The COE estimated the 100-year flow to be 16,000 cfs (COE, 1965). For the purpose of this study, 3-month, 6-month, 1-, 2-, 5-, and 25-year flows were based on daily average and maximum flows collected at Big Chico Creek gauging stations between 1975 and 1999.

Figure 2


Figure 3

The Downstream Effects of Dams

Aside from flood control, downstream effects of dams were of little concern during the design and construction of most historic dams in the western United States. Changes in channel morphology, fish populations or riparian vegetation were often unanticipated or were not taken seriously. Downstream effects include channel scouring and impacts on the biological ecosystem.

Altered Sedimentation Patterns

In an undisturbed watershed, there is a dynamic equilibrium between supply and outflow of sediment in the stream system. Sediment to a stream is derived from three primary sources in a watershed: (1) mass wasting in which soil, rock, and other debris are moved down slope by gravity, (2) surface erosion by wind, water and chemical processes and (3) stream bank and stream channel erosion, in which sediment is entrained, transported, and re-deposited. This sediment load dictates the form and habitability of the stream and disruption of the load greatly affects the species dependent on it.

Sediment is an important and vital component of in-stream fish habitat; gravel, cobbles, and organic debris form the critical components. Salmonids are dependent upon stream reaches with sorted and well-distributed gravel to spawn successfully (a stream reach is a section of a stream extending from one point to another). The gravel must be reasonably free of fine sediment, such as clay and silt, in order for eggs and embryo to be sufficiently oxygenated and thus survive and emerge as fry. Young fry further depend on gravel and cobble areas for escape cover.

Human activity in a watershed, such as dam construction, disrupts the sediment budget. Dam construction limits the downstream magnitude and volume of flows, increases upstream supply and reduces the downstream supply of coarse and fine sediment. When large volume flows capable of moving coarse sediment encounter the pool/stilling area at the Five-Mile Area, the velocity and lift force is reduced such that the coarse sediment previously transported as bedload is deposited on the upstream side of the dam. The finer sediment remains entrained in the water column. Water leaving the Five-Mile stilling area to Big Chico Creek contains a reduced coarse sediment bedload, significantly diminishing downstream transport and supply. At Chico's One-Mile Recreation Area, the flow is again impeded and another portion of the bedload, along with suspended load, is deposited on the upstream side of that dam. Water leaving the One-Mile Area pool during high flows scour out the finer particles from the downstream bed, remove entire deposits of spawning gravels, and reduce particle size diversity.

Present Conditions

Factors affecting salmon production and survival include:

1. Spawning gravel supply and gravel quality
2. Juvenile rearing habitat availability and condition
3. Streamflow during fry, juvenile, and adult migration periods
4. Predation by non-native fish species
5. Bay/delta and ocean mortality (predation, pumping, and sport/commercial harvesting)
6. Water temperatures and water quality

Chinook salmon generally spawn in water from one to three feet deep. Other criteria include water velocities of 1 to 3 feet per second, a gradient of 0.2 to 1.0 percent, and substrate from 0.5 to 10 inches thick dominated by 1 to 3-inch cobble (DFG, 1998). Additional information regarding spawning, incubation, and rearing periods in Big Chico Creek and Lindo Channel are provided in Aquatic / Biotic Inventory.

Hydrology

Salmonids are dependent upon different flows during various life stages; their presence, absence, and movements are influenced by flows. Therefore, hydrographic data are useful for fisheries assessment.

For all gauged streams a hydrograph can be generated. A hydrograph is a graph showing, for a given point on a stream, the discharge, stage, velocity, or other property of water with respect to time (Bedient & Huber, 1988, p69). Depiction of long periods such as an annual hydrograph can be used to determine low flow, summer base flow, winter base flow, and flood discharges.

Data presented in hydrographs for Big Chico Creek were obtained from the California Department of Water Resources (DWR) stream gage station AO-42105 located 50 feet upstream from the intersection of Rose Avenue and Bidwell Avenue. Data presented in hydrographs for Lindo Channel were obtained from DWR's stream gage station AO-0165 located at the right abutment of the Cussick Avenue Bridge, 2.25 miles northwest. Flow data presented in this study comprise data collected between the periods of October 1975 to March 1999.

In addition to hydrographs, flood frequency curves were generated based on four periods that correspond to the four spawning periods of Chinook salmon and steelhead trout. Flood frequency is the average number of times a flood of a given magnitude is likely to occur over a specified number of years. The spawning periods for these species are consistent with the ECR and include the following:

• Spawning Period 1 - February (Steelhead Trout)
• Spawning Period 2 - January to February (late fall run Chinook)
• Spawning Period 3 - Mid September to October (spring run Chinook)
• Spawning Period 4 - Late October to December (fall run Chinook)

For each spawning period, average and maximum daily flows were tabulated and utilized in the hydrologic modeling and associated sediment transport analysis discussed below.

Annual Hydrographs for Big Chico Creek

Hydrograph characteristics (magnitude, duration, timing, and frequency) were obtained by plotting and inspecting annual hydrographs for each water year (Appendix A). An example hydrograph for Big Chico Creek for Water Year (WY) 1994-1995 is shown in Figure 4.

Figure 4. Annual Hydrograph for Big Chico Creek for WY 1994-1995.

Seasonal streamflow patterns can be fairly predictable, but specific flow magnitudes, durations, and frequencies are unpredictable due to runoff patterns produced by storms, droughts, and snowmelt. As is shown in Figure 4, winter storms produced high flows in January and February. Streamflow variability is important to overall river ecosystem health.

In addition to the hydrographs, annual instantaneous peak discharges at each stream were compiled and plotted by return period (the time interval for which an event of a given magnitude will occur once on the average). The flow data were then fit to the Gumbel Distribution (Bedient et al, 1988) and plotted to produce annual maximum flood frequency curves. During the four spawning periods addressed in this section, annual maximum flood flows for Big Chico Creek ranged from a low of 8 cfs in WY1979 to a high of 1,210 cfs in WY1995. An example flood frequency curve for Big Chico Creek during Spawning Period I is provided in Figure 5.

From this curve, a 2-, 5-, and 25-year return period maximum discharge for each spawning period was used in the hydrologic and sediment transport analysis. For example, flows of a 2-, 5- and 25-year maximum discharge in Big Chico Creek for Spawning Period 1 was 462, 773, and 1,242 cfs, respectively. In addition to these higher flows, annual daily flows over the past 24 years were averaged to obtain 1-year, 6-month, and 3-month flows to be used in the hydrologic and sediment transport analysis. Hydrographs and frequency curves generated from the 1975-1999 data are provided in Appendix A.

Figure 5. Flood Frequency Curve for Big Chico Creek during Spawning Period 1 (February).

These flows were selected to determine the amount of erosion or bed mobility that would occur at these flows for gravel sizes of 1/16-inch, 1-inch, and 4-inches in mean diameter. These gravel sizes were selected based on the range of gravel sizes from 0.5- to 4-inches of mean diameter that are adequate for salmonid spawning habitat (DFG, 1997, p. VII-47).

Annual Hydrographs for Lindo Channel

A hydrograph for Lindo Channel for WY1995-1996 is shown in Figure 6. During this water year, flow was present in Lindo Channel from mid-December 1995 to early June 1996 (typical due to the natural diversion of flow to Big Chico Creek during the summer and fall seasons). (see Existing Management Plans and Water Quality Monitoring Data chapters.)

Similar to Big Chico Creek, a 2-, 5-, and 25-year return period maximum discharge for each spawning period was used in the hydrologic and sediment transport analysis. From Figure 5, flows of a 2-, 5-, and 25-year maximum discharge in Lindo Channel for Spawning Period 1 was 808, 1,656, and 2,297 cfs, respectively (see Figure 7). Hydrographs and frequency curves generated from the 1975-1999 data are provided in Appendix A.

Figure 6. Annual hydrograph for Lindo Channel during WY1995-1996.

Figure 7. Flood frequency curve for Lindo Channel during Spawning Period 1 (February)

As previously mentioned, these flows were selected to determine the amount of erosion or bed mobility that would occur at these flows for gravel sizes of 1/16-inch, 1-inch, and 4-inches in diameter.

Fluvial Geomorphology

Geomorphology should be considered at both the watershed and reach scales. This study focused on the reaches of Big Chico Creek and Lindo Channel below the Five-Mile Recreation Area. However, discussion of the geomorphology of the whole watershed is presented in other sections of the ECR. An analysis of flow and bedload transport sediment was conducted to assure sufficient sediment is conveyed to maintain channel capacity and prevent adverse impacts to banks and riparian vegetation from sediment accumulation in the channel.

Geomorphic Processes

Geomorphic processes include the following:

• Flow Regime. Flows in streams consist of subcritical (referred to as tranquil or upper stage) flow, supercritical (referred to as rapid or lower-stage) flow, and/or mixed flow regimes. For this analysis, the flow was assumed to be subcritical based on the gradual slopes of the channel bed downstream of the Big Chico Creek and Lindo Channel flow control structures.
• Bed Mobility. Bed mobility is important for restoring and maintaining alluvial river morphology. Mobilization initiates bedload transport and routing, discourages riparian vegetation from colonizing, periodically cleanses spawning gravel deposits, and rejuvenates alluvial features.
• Sediment Budget. Sediment budget is the change in sediment storage defined by the difference between the amount of sediment entering a stream and the amount leaving a stream.
• Gravel Supply. Gravel/coarse sediment supply represented by coarse sand to small boulder sized particles are transported as bedload (particles that are in almost continuous contact with the streambed when transported, also known as saltation). Gravel losses occur when gravel input is less than the output or when storage decreases.
• Fine Sediment Transport. Fine sediment transport is by suspension (particles that are suspended in the water column and rarely contact the streambed).

During wet water years, substantial bedload transport, transport of large bedload particles, floodplain deposition, side channel creation, and significant channel migration can occur.

The dominant geomorphic processes during a normal water year are flows associated with moderate winter floods and snowmelt runoff that transport sands and moderate volumes of coarse bedload. This results in limited turnover of gravel deposits and modest channel migration.

Dominant geomorphic processes associated with dry water years are small winter floods and modest snowmelt runoff that transports sand in secondary alluvial features and minor course bedload. This results in little to no channel migration.

Aggregate Source Inventory

As mentioned previously, the ideal range of spawning gravels is between 0.5- to 4-inches in mean diameter (DFG, 1997, p. VII-47). To determine the gravel size distribution at several spawning gravel sites in Big Chico Creek, an Aggregate Source Inventory was conducted by the California Department of Water Resources (DWR) in November 1997 (DWR, 1997). The sites were located along the Big Chico Creek at Highway 32, below the Five-Mile Area Flow Control Structure, and at Rose Avenue. To analyze the bulk sample data from each site, DWR plotted the percent by weight passing each sieve as a curve on a semi-logarithmic scale. The plotted data show the percentages of particles retained and passing through each sieve. Particle distribution curves for these three sites are provided in Appendix A.

The shape and location of a curve shows the general particle size distribution characteristics of the gravel. A very steep curve, with no tail, indicates relatively well-sorted, uniform gravel with a small range of particle sizes. Conversely, a low-slope curve indicates poorly sorted gravel with a wide range of particle sizes.

For example, Figure 8 shows a fairly steep curve, indicating relatively uniform gravel. Along the x-axis of the graph, grain size in millimeters is arranged in logarithmic succession. Along the y-axis the arithmetic scale is divided into percentage values of the cumulative percent finer by weight of grain sizes ranging from 0.01 millimeters (mm) to greater than 512 mm. Particle size diameters were determined values of 95%, 50%, and 5% were determined from the curves. The D₅₀ represents the median grain size at which 50% of the sample is coarser and 50% is finer. The D₉₅ is the grain size at which 95% of the sample is finer. The D₉₅ and D₅ dimensions fall two standard deviations from the median.

The range of acceptable spawning gravel sizes on each curve is represented as the area that lies between the solid lines (Figure 8). Summaries of the range of gravel sizes are presented in the following table.

The gravel sizes ranged from 20 mm to 100 mm (or approximately 1 to 4 inches) in mean diameter. Based on gravel sizes alone, this data indicates that these sites contain gravel sizes that represent good potential for salmonid spawning habitat.

Figure 8

Bedload Forfeiture at One-Mile and Five-Mile Flow Control Structures

Five-Mile Area Flow Control Structures

The Five-Mile flood control system is designed to create a pool in the stilling basin thereby allowing controlled flows through the Big Chico Creek and Lindo Channel flow control structures and the Sycamore Bypass Channel. During high flow periods Upper Big Chico Creek exits the narrow foothill canyon at very high velocities carrying a large bedload until it encounters the Five-Mile Area stilling basin. Velocity and bedload mobilizing capacity is significantly reduced, allowing for the larger, entrained sediment to quickly fall out of the water column depositing the large gravel just upstream of the Five-Mile Area Flow Control Structures. During the next high flow period the previously deposited gravels flow in the direction of least resistance, bypassing Big Chico Creek proper and flowing down Lindo Channel or Sycamore Bypass instead.

View looking upstream at the Big Chico Creek flow control structure.

One-Mile Area Flow Control Structure

The flow control structure at the One-Mile Recreation Area (also known as Sycamore Pool) was constructed within Big Chico Creek in 1929. It consists of a 700-foot long in-stream, flow-through swimming pool with a flashboard dam. The dam is fitted on the north side with a concrete-step pool/flashboard fish ladder that DFG found to be adequate for year round fish passage. The One-Mile Area flashboard dam lies on a concrete apron and is formed by an angle iron frame holding removable timber flashboards. From mid-September to mid-May, the flashboards are removed in anticipation of high winter flows. During high flow periods pooling of water behind the One-Mile channel restriction occurs and allows entrained/suspended sediment to drop out of the water column. However, with a portion of the bedload and suspended having already been deposited at the Five-Mile Area results in a minimum amount of gravel to cobble-sized sediment being deposited at the One-Mile Area.

The city of Chico's current maintenance practices include cleaning out sediment and debris at the start of the summer season, setting up the dam, and performing weekly cleaning. In the past, the DFG and Regional Water Quality Control Board (RWQCB) raised concerns over the city's spring-cleaning practices, which result in notable discharges of suspended sediment and organic debris (Mitchell Swanson and Associates, 1994). During 1997, the city of Chico designed and installed a box culvert beneath the Sycamore Pool to decrease downstream turbidity from pool cleaning activities.

View looking upstream at the downstream side of One-Mile Dam.

Biological Resources

Big Chico Creek and Lindo Channel are salmonid streams that are utilized by migratory members of the salmon/trout family for spawning and/or rearing. Specific salmonoid runs, spawning, incubation, and rearing periods are discussed below.

Anadromous Fisheries of Big Chico Creek Watershed

Four salmonoid runs use Big Chico Creek and Lindo Channel for spawning: three runs of Chinook Salmon and one of Steelhead Trout. Chinook salmon and steelhead trout are anadromous, having a river-to-ocean, ocean-to-river life cycle. Young salmon migrate to sea shortly after emerging from spawning gravels, and spend most of their life in coastal waters where there are abundant food supplies. Fall run Chinook adults return to spawning grounds in the fall and early winter, after 1.5 to 3.5 years in the ocean. The female digs a trough (also known as a redd) in river gravels and deposits her eggs as the male fertilizes them. Juveniles emerge from spawning gravels between December and April and rear in deep, slow portions of the creek. Historically, high spring flows during the snowmelt runoff helped juveniles migrate out to sea before high summer temperatures made river conditions less hospitable. Figure 9 and Figure 10 show the life cycle of Chinook salmon with respect to the annual hydrograph flows for Big Chico Creek and Lindo Channel, respectively.

Procedures Used for Hydrologic Modeling and Geomorphologic Analysis

The overall objective of the Big Chico Creek Watershed Alliance's Watershed Management Strategy is to improve the spawning and migration habitat for spring run and fall run salmon and steelhead trout in Big Chico Creek and Lindo Channel. To accomplish this objective, one task was to perform a hydrology and sediment transport evaluation to assist in the preparation of a gravel placement and maintenance plan.

Hydrologic Modeling

Topographical cross-section data

Channel cross-sections are a very important part of hydrologic analysis. A quick walk up Big Chico Creek or Lindo Channel reveals that even over short distances stream cross-sections can take on a wide variety of shapes and sizes. Additional data of particular importance comprise wetted-perimeter, slope of stream channels, roughness, and average velocity.

A Flood Insurance Study was conducted for FEMA on behalf of the City of Chico in August 1993 (S&W, 1993). The purpose of the study was to conduct hydrologic modeling for streams in the Chico area. The study presented discharge estimates for Big Chico Creek and Lindo Channel, and documented the hydrologic calculations behind these estimates. S&W based their

study on photogrammetric contouring/cross-sectioning to obtain cross-sectional data with a resolution of 2-foot elevation contour intervals. The cross-sectional data and other model parameters used in the hydrologic modeling and sediment transport analysis are discussed in detail below.

Prior to hydrologic modeling, a topographic survey of approximately twenty additional cross-sections was made to supplement the S&W cross-sections. The additional cross-sections were approximately equidistant; some co-located with the FEMA cross-sections, were generally representative of the nature of the creek, and were surveyed in the field by a Professional Land Surveyor to an accuracy of 0.5-foot. The co-located cross-sections were generally similar and are provided in Appendix B. Figure 11 shows the stream cross-section locations along Big Chico Creek and Lindo Channel.

HEC-RAS Modeling

The COE Hydrologic Engineering Center's River Analysis System (HEC-RAS) model was used to determine flow velocities and depths as a function of distance along the streams for different flows. Based on these, erosion factors were determined in Big Chico Creek and Lindo Channel. Channel cross-sections were specified at various intervals and these were used to estimate the size of gravel that would be mobilized for different flow conditions (as discussed earlier in this chapter). The HEC-RAS outputs are included as Appendix C.

Sediment Bedload Transport Analysis

The potential quantities of sediment capable of being transported were estimated as a function of distance using analytical formulae for the different modes of transport. The shear stress, Shield's parameter, and bedload transport were calculated from the results of the HEC-RAS outputs. The calculated Shield's parameter was then compared to the critical Shield's parameter to determine if erosion would occur for particular gravel sizes at a particular cross-section. Results of these calculations are discussed in this section and are included in Appendix D.

Bedload Parameters

The result of the HEC-RAS modeling generates data, which are necessary in calculating the amount of bedload transport along Big Chico Creek and Lindo Channel. HEC-RAS outputs include the slope of the energy grade line, channel velocity, flow area, and other parameters for each cross section. From these parameters and an assumed gravel size (e.g., 1/16-inch, 1-inch, and 4-inches in mean diameter), the hydraulic radius (R_h), shear stress (??), Shield's parameter (?), Reynolds number (R_e), critical depth (y_c), and bedload transport (q_s) were calculated.

In addition, the bedload transport potential (cubic feet or cubic yards) was calculated by multiplying the bedload transport (ft²/sec) by the width of the channel (ft) by the duration of the maximum discharges during each spawning period (seconds). For this study, a channel width of 30 feet and 60 feet were assumed for Big Chico Creek and Lindo Channel, respectively.

Figure 11

Bedload Shear Stresses

Assuming steady flow, calculations were performed for water surface slope and cross-sectional area for each flow. The estimated cross sectionally averaged boundary shear stress is:

(1)

Equation 1 is the most common way to estimate boundary shear stress. The boundary shear stress is necessary in order to calculate the Shield's parameter and ultimately, the bedload transport.

Modeling Transport with Shield's Parameters

There are a large number of methods that have been developed to estimate bedload transport rates. The first question, however, is whether sediment motion will occur and the most recognized method for establishing this is the Shield's diagram (Vanoni, 1975). The most important parameter is the Shield's parameter, defined as follows:

(2)

Sediment motion occurs when the Shield's parameter is larger than the critical Shield's parameter, Y_c. The critical Shield's parameter is the hydraulic threshold at which channelbed surface particles begin to mobilize, occurring when the drag force exceeds the gravitational force resisting downstream motion of the particle. For fully developed and turbulent flow (Reynolds number greater than 450), which could be expected in the streams of interest, the critical Shield's parameter (Y_c) is approximately constant and equal to 0.06 (Vanoni, 1975). Therefore, sediment motion occurs when Y > Y_c. Comparing Sheild's parameter with the predicted critical Shield's parameter provides a method to estimate whether conditions could be achieved at a given discharge and hydraulic setting.

Particle sizes of 1/16-, 1-, and 4-inches in diameter were used to calculate the Shield's parameter. The interest lies with gravel and coarse sand with the relevant modes of transport being rolling and saltation, also referred to as bedload. A widely accepted formula for bedload transport is that of Meyer-Peter and Müller (Madsen, 1991):

(3)

The above equations can be combined to give:

	(4)
	(5)

The transport rate given by the above equation is a potential transport rate; the actual rate depends on the availability of sediment, either in place or transported from upstream. Also, if the transport exceeds the upstream supply, the channel bottom will drop, reducing velocities and further transport. Thus, a balance is established, which can only be simulated using a full sediment transport model and transient flow records. The results of this evaluation allow estimation of the transport of different sized gravel introduced at different locations within the streams.

Hydrologic Modeling Results

The results of the HEC-RAS model runs include cross section profile data and the necessary parameters to analyze sediment bedload transport in reaches of Big Chico Creek and Lindo Channel. Three flow scenarios for each spawning period was used to generate the HEC-RAS modeling data. Results are presented in Appendix C.

Rationale for Site Selection

One viable option to increase salmonid spawning habitat is to place gravel in areas that do not experience significant transport potential for coarser gravels (i.e., 1- to 4-inch gravels) but do have sediment transport probability for fine-grained sediments (i.e., 1/16-inch gravels). Based on this approach, the reaches of Big Chico Creek and Lindo Channel were evaluated for potential areas of gravel placement. It is also important to mention that there are other factors in addition to hydrology and gravel size distribution to consider regarding a site's viability for salmonid spawning habitat. These other factors include stream cover, streamside vegetation, water temperature, etc., and are discussed in other chapters of the ECR.

Results of Sediment Bedload Transport Analysis

The results of the sediment transport modeling are presented in Appendix D. For the basis of this analysis, gravel sizes of fine-grained and coarse gravels were used to assess areas of sediment transport potential. Based on these results, each cross section on Big Chico Creek and Lindo Channel was evaluated for high, medium, and low bedload transport areas.

Big Chico Creek Reaches

Generally, higher flows correspond to a higher calculated Shield's parameter. For example, average Shield's parameters for a 1-inch gravel ranged from 0.06 to 0.19 for flows of 38 to 1,374 cfs, respectively. These values compare to a critical Shield's parameter of 0.06 (Vanoni, 1975), above which sediment motion will occur. At a flow of 460 cfs, energy grade (or channel) slopes ranged from 7.0E-06 to 0.12. Higher channel slopes resulted in higher shear stresses, and have higher Shield's parameters.

At a flow of 460 cfs, channel velocities ranged from 0.20 to 9.4 feet per second (fps). Higher channel velocities resulted in higher shear stresses. Reaches that had high shear stresses and Shield's parameters resulted in high bedload transport potential. Locations where the model predicted potential bedload transport was categorized into high, medium, and low bedload transport potential (or high, medium, and low potential energy) for each cross-sectional area. It should be stressed that the volumes presented in Appendix D have the potential for movement at that particular cross-section. However, this does not necessarily correlate to the amount of bedload that is being deposited downstream and therefore is available for use. More sophisticated transport modeling (beyond the scope of this study) must be performed to determine the mass balances of the stream ecosystem.

There were several locations noted along Big Chico Creek where the model predicted medium to high transport potential for gravels equal to or less than 4-inches in diameter at various flows. These are areas where gravel, if deposited, would be transported further downstream. Therefore, it may be useful to place gravel at these areas for entrainment. However, they would not be considered necessarily suitable for spawning redds. The locations consist of the following:

• Immediately downstream of the Big Chico Creek flow control structure
• Immediately downstream of the Manzanita Bridge
• Between Madrone Avenue and Crister Avenue
• Immediately downstream of Highway 99 footbridge
• Immediately downstream of Glenwood Avenue

In addition to areas that were likely to have low to high sediment transport potential, the HEC-RAS model predicted stream reaches that experience little to no sediment transport. There were several areas on Big Chico Creek where the HEC-RAS model predicted little to no transport of coarser gravels (1- to 4-inch) and medium to high transport potential for fine-grained sediments. These were the potential gravel placement areas recommended, and are discussed in the gravel placement plan section. It is important to note that these areas are potential gravel placement areas based on the HEC-RAS model. As mentioned previously, there are many other factors that influence the selection of a stream reach for salmonid spawning habitat viability.

Lindo Channel Reaches

Generally, higher flows correspond to a higher calculated Shield's parameter. Average Shield's parameters for a less than or equal to 2-inch gravel ranged from 0.02 to 0.08 for flows corresponding to 13 to 3,082 cfs, respectively. At a flow of 141 cfs, channel slopes ranged from 2.07E-4 to 0.075. Higher channel slopes resulted in higher shear stresses.

At a flow of 141 cfs, channel velocities ranged from 0.79 to 6.0 fps. Reaches that had high Shield's parameters also had high bedload transport potential. As mentioned previously, locations where the model predicted bedload transport were categorized into high, medium, and low bedload transport potential.

There were several locations along Lindo Channel where the model predicted moderate to high bedload transport potential for gravels less than 4-inches in diameter for various flows. The locations consist of the following:

• Immediately downstream of the Lindo Channel flow control structures
• Immediately downstream of the Manzanita Bridge
• Immediately downstream of Highway 99 Bridge
• Downstream of Cussick Avenue
• Downstream of Nord Avenue

Currently there appears to be adequate coarse gravel in Lindo Channel for spawning habitat. However, spawning habitat is limited due to the lack of summer and fall flows in Lindo Channel. Although Mitchell Swanson and Associates has recommended to the City to consider diverting water to Lindo Channel during low flow months, DFG has stated that it desires to manage Big Chico Creek as the primary resource. Until this priority changes, focusing on gravel placement in Big Chico Creek is recommended.

RESTORATION THROUGH GRAVEL MANAGEMENT

Gravel Placement Plan

Presently, a reduced percentage of the available gravel passes through the Big Chico Creek and Lindo Channel flow control structures due to the gravel deposition in the stilling basin when gravel-transporting flows occur. Since the flood control project was initiated at Five-Mile Area in 1963, gravel has been removed from the stilling basin and used as road cover and for other construction purposes (MSA, 1994, p63). Removal of gravel from the stilling basin was discontinued in 1995 due to concern by DFG about downstream sedimentation.

The purpose of a gravel placement plan is to develop an adaptive management plan for improving gravel areas for salmonid spawning and migration. Optimal locations for gravel placement, gravel placement volumes, timeline, and a maintenance program are discussed below.

Gravel Placement Functions

Just as water is essential to the Big Chico Creek Watershed, gravel and cobbles are important elements of channel geomorphology. The coarse gravel sediment forms riffles, pools, and other alluvial features that provide salmonid spawning and rearing habitat.

Although it is commonly thought that fine sediments are detrimental for the river ecosystem, this is not always the case. For example, fine sediments deposit on the inside of migrating meander bends and encourage riparian revegetation (McBain & Trush, 1998, p294). However, chronic fine sediment loading can greatly increase instream fine sediment storage rather than floodplain storage, which can severely impact salmonid habitats.

Introducing potential bedload with a significant component of fine sediment is not recommended to improve salmonid spawning and rearing habitat. Based on the preferred spawning sizes for fall and spring-run Chinook salmon as well as steelhead trout, the suitable gravel size range for placement is 0.5 to 4 inches mean diameter (MSA, 1994; DFG, 1997,  p.VII-47).

The lower reaches of Big Chico Creek and Lindo Channel were evaluated for potential areas of gravel placement. As previously mentioned, there are several other factors in addition to hydrology and bedload transport that should be considered in the selection of viable salmonid spawning habitat locations (e.g., stream cover, streamside vegetation, water temperature, etc.). These issues are addressed separately within the ECR.

Optimal Locations for Gravel Placement

The use of gravels for placement as spawning gravel beds was evaluated at appropriate reaches below Five-Mile Area Big Chico Creek and Lindo Channel flow control structures. Based on the HEC-RAS model results, several potential sites for gravel placement were found.

Big Chico Creek Reach Locations

There were several areas on Big Chico Creek where the model predicted little to no transport of coarser gravels (1- to 4-inch) and moderate to high transport of fine-grained sediments. Placement of gravel downstream of certain pools is proposed to eventually create stream riffle habitat. These potential gravel placement areas are shown as cross-sections on Figure 12 and include the following:

• Cross-section 5430 to 5420, approximately half way between Manzanita Bridge and the Big Chico Creek flow control structure.
• Cross-sections 4500, 4480, and 4460, downstream of Manzanita Bridge.
• Cross-section 4370 to 4360, upstream of Highway 99 Bridge.
• Cross-section 4280, upstream of One-Mile Area's Sycamore Pool. Although the model shows that this location may be a possible location for gravel placement, introduction of gravel immediately upstream of the One-Mile Area is not recommended due to increased sedimentation within the pool and resultant maintenance efforts required for removal.
• Cross-section 4260, downstream of One-Mile Area's Sycamore Pool. Bank erosion is evident on the south side of the stream in this area. One option may be to place gravels in this area of high sediment transport potential (high potential energy) to transport gravels downstream where the model predicts minimal coarse gravel transport potential. A secondary benefit of this option would be to reduce the undercutting of the concrete apron and streambank erosion in this area.
• Cross-section 4070 to 4066, upstream of the Warner Street Bridge between Acardian Avenue and Citrus Avenue.
• Cross-section 4010 to 3049, between the Warner Street Bridge and Rio Chico Street.
• Cross-section 3016 to 2990, downstream of the Warner Street Bridge between Rio Chico Street and the Railroad Tracks.
• Cross-section 2910, downstream of the Nord Avenue Bridge.

Figure 12

Field observations of these locations generally indicated the presence of pools and riffles that could be enhanced for salmonid habitat. In most cases, adequate stream cover exists. Several of these areas are also consistent with MSA's findings of potential gravel placement areas, such as below Big Chico Creek flow control structure, below Manzanita Road, and below the One-Mile Area's Sycamore Pool flashboard dam (MSA, 1994, p65).

Although the HEC-RAS model included reaches downstream of Glenwood Avenue to the junction of Big Chico Creek and Lindo Channel, the data is not reliable due to the large distances between these cross-sections.

It is important to note that the potential gravel placement reaches identified are solely based on the modeling effort. As mentioned previously, many other factors should be considered in order to select specific stream reaches for salmonid spawning habitat restoration.

Lindo Channel Reach Locations

Potential gravel placement areas on Lindo Channel are shown as cross-sections on Figure 12 and include the following:

•  Cross-section 5310 to 5290, downstream of the Manzanita Avenue Bridge.
•  Cross-section 5150 to 5130, downstream of the Floral Avenue Bridge.
•  Cross-section 5070 to 5040, between the Mangrove Avenue Bridge and Highway 99 Foot Bridge.
•  Cross-section 5010 to 2670, between The Esplanade Bridge and the Mangrove Avenue Bridge. However, the model predicted coarse gravel movement at moderately higher flows for a few of the cross-sections in this reach (e.g., cross-sections 2685 and 2640). As such, these particular cross-sections may be inadequate for gravel placement.

There are several potential sites between The Esplanade and Cussick Avenue. Possible cross-sections include 2515, 2500, and from 2485 to 2455.

•  Cross-section 2355 to 2315, immediately downstream Cussick Avenue.
•  Cross-section 2270, located upstream of Guynn Avenue.
•  Cross-section 2140, upstream of the Nord Avenue Bridge.
•  Cross-section 2075 to2020, downstream of the Nord Avenue Bridge.

Although there are potential gravel placement areas on Lindo Channel, spawning habitat is limited due to the lack of summer and fall flows. Unless conditions were to change, it is recommended that the focus on gravel placement be directed towards Big Chico Creek.

Methods for Gravel Placement

Spawning gravel for salmon should be clean, creek-run from 0.5- to 4-inches diameter. It is proposed that gravel would be dumped at a staging area on the bank and then picked up and placed with a small front-end loader. It is recommended that naturally deposited gravels be used from the Five-Mile Area stilling basin above Big Chico Creek and Lindo flow control structures. In addition, it is recommended that the gravel portion of sediments removed from the One-Mile Area's Sycamore pool-cleaning effort be screened and used.

One placement option is to input gravel in the area of high water velocity to allow the stream to more naturally distribute the gravel downstream during high flows. Possible areas include downstream of the Big Chico Creek and Lindo Channel flow control structures and the other potential stream reaches mentioned above. An area of active bank erosion may also be a good candidate site for gravel placement because the stream has already demonstrated high energy and the ability to move substrate material (Cross-section 4260 is one such site). However, the sites selected must have good equipment access and have adequate transport capacity so that gravels do not cause flow obstruction and potential flooding.

Gravel Placement Volumes

Proposing initial placement volumes without specific placement locations is not recommended. The volume of gravel placement for the areas recommended in this chapter will vary from site to site. However, it is recommended that a small amount of gravel be placed initially and monitored to assess whether the placed gravels are stable or transported downstream. This is consistent with an adaptive management approach. Generally, volumes should be characterized based on the thickness of the added gravel. A thickness of twice the mean gravel diameter seems reasonable. For example, placement of a 2-inch gravel at a particular cross section in Big Chico Creek would translate to a volume of 0.4 cubic yards per linear foot of stream (assuming 30-foot width).

In 1994 MSA made recommendations to place small amounts of gravel at the downstream end of pools where it could immediately be useable as spawning gravel. In general, Mitchell, Swanson & Associates recommended that approximately 0.25 to 0.5 cubic yards of gravel would be required per site (MSA, 1994, p65).

Timeline for Initial Gravel Placement

In order to minimize the disturbance of the benthic invertebrate community and salmonids in Big Chico Creek, it is recommended that placement of gravel occur in the late summer. Placement during this time would also benefit from low flows and have the least impact on the salmonid spawning and rearing periods.

Initially, it is recommended that gravel placement occur once per year. Gravel placement frequency may be revised based on observations in actual bedload transport and the results of gravel placement monitoring. Again, this approach is consistent with the WMS goal of operating an adaptive management program.

Maintenance Program

Butte County has historically removed loads of gravel at the Five-Mile Recreation Area and placed them on levee roads or used them for other construction purposes. In 1992, the City of Chico took over gravel removal and flood control management in the stilling basin from Butte County (MSA, 1994, p20).

After a high flow or watershed disturbance, fine sediment may be deposited in spawning gravel substrates. In order to avoid deposition of fine sediment, this study recommended choosing stream reaches where fine sediment transport potential was identified. However, periodic maintenance might be required to reduce fine sediment in spawning areas (i.e., spawning riffles). Plowing the gravel with a ripper attachment on a tractor then adding fresh gravel can do this. Gravel ripping has been used before as a method of loosening and cleaning gravels by the DFG in the Sacramento River, by the USFWS on the Feather River, and by the Turlock and Modesto Irrigation Districts (in cooperation with the DFG) on the lower Tuolome River below La Grange Dam (Vyverberg et al, 1997).

Monitoring Methods

The following monitoring methods are recommended:

• Visual survey of bed material.
• Cross-sectional surveys at various reaches to monitor channel degradation and aggradations.
• Tracer gravels to monitor travel distance downstream. Different colored and sized gravels can be placed in gravel placement areas to assess the physical horizontal transport of gravels over time.

The purpose of monitoring is to assess the accuracy of the measures performed and to modify the adaptive management plan as needed.

Gravel Replenishment

After the gravels are initially placed in the streams, it is recommended that gravel be replenished every year during the late summer to minimize impacts to the stream ecosystem. It is recommended to use gravels that are deposited in the stilling basin at the Five-Mile Area for gravel replenishment below the Big Chico Creek and Lindo Channel flow control structures. In addition, it is recommended that the gravels from the fine and coarse-grained sediments removed from the One-Mile Area Sycamore Pool be considered for gravel replenishment below the One-Mile Dam. This would require some type of mechanical separation and could be implemented following the annual pre-summer cleaning event at Sycamore Pool.

Long-Term Maintenance Recommendations

Long-term maintenance depends upon the success of the initial and on-going gravel placement, monitoring, and maintenance recommendations presented above. It is recommended that feedback from these activities be evaluated by the appropriate agencies including the DFG, RWQCB, City of Chico, Butte County, and other interested parties consistent with the BCCWA's adaptive management objective.

It has already been recommended that a combined fish passage and structural design investigation be undertaken to identify a solution to problems associated with the Lindo Channel flow control structure (MSA, 1994, p68). The damaged concrete apron immediately downstream of the dam prevents fish from entering the fish passage channel. In addition to structural design changes, increasing flows to Lindo channel and enhancement of the vegetation in the stream corridor should be considered to promote additional salmonid spawning habitat.

References

Bedient, Philip B. & Huber, Wayne C. (1988). Hydrology and Floodplain Analysis. Addison-Wesley Publishing Co., Reading, Massachusetts.

CH2M Hill. (1991). Guide to Upper Sacramento River Chinook Salmon Life History. Prepared for U.S. Bureau of Reclamation Central Valley Project.

Collier, Michael, Webb, Robert H., and Schmidt, John C. (1996). Dams and Rivers, A Primer on the Downstream Effects of Dams. U.S. Geological Survey, Tucson, Arizona. Circular 1126.

Flosi, Gary, et al. (1997). California Salmonid Stream Habitat Restoration Manual. California Department Fish and Game.

Madsen, O.S. (1991). Mechanics of Cohesionless Sediment Transport in Coastal Waters. Proceedings, Coastal Sediments. 1991. American Society of Civil Engineers Specialty Conference, Seattle, Washington.

McBain & Trush. (1997). Trinity River Maintenance Flow Study Final Report. Prepared for Hoopa Valley Tribe Fisheries Department. Hoopa, California.

McCuen, Richard H. (1998). Hydrologic Analysis and Design. Second Edition. Prentice Hall, Inc., New Jersey.

Mitchell Swanson and Associates. (1994). Hydrology Management Plan for Big Chico Creek, One- and Five-Mile Dam Facilities, and Lindo Channel. Prepared for City of Chico Parks Department.

Schaaf & Wheeler. (1993). Flood Insurance Study for Big Chico Creek and Lindo Channel, Chico, Butte County, California. Prepared for Federal Emergency Management Agency.

State of California. (1997). The Resources Agency Department of Water Resources, Division of Engineering, Civil Engineering Canals and Levees. Soils and Concrete Laboratory Report No. 97-35, Northern District, Big Chico Creek.

U.S. Army Corps of Engineers. (1965). Operation and Maintenance Manual for Chico and Mud Creeks and Sandy Gulch, Sacramento River, and Major and Minor Tributaries Project, California.

U.S. Army Corps of Engineers, Hydrologic Engineering Center. (1997). HEC-RAS River and Reports on Engineering Practice No. 54.

Vyverberg, Kris, et al. (1997). Lower American River Chinook Salmon Spawning Habitat Evaluation, an Evaluation of Attributes Used to Define the Quality of Spawning Habitat

Analysis System, Hydraulic Reference Manual.

Vanoni. (1975). Sedimentation Engineering. American Society of Civil Engineers Manuals and Reportson Engineering Practice No. 54.

Vyverberg, Kris, et al. (1997). Lower American River Chinook Salmon Spawning Habitat Evaluation, an Evaluation of Attributes Used to Define the Quality of Spawning Habitat.

Available Upon Request

APPENDIX A - BIG CHICO CREEK AND LINDO CHANNEL

FLOW DATA

APPENDIX B - CROSS-SECTION DATA

APPENDIX C - HEC-RAS MODELING RESULTS

APPENDIX D - SEDIMENT TRANSPORT RESULTS