Landsburg_Final_Report_12-31-10.pdf

ANALYSIS OF ALTERNATIVES

Landsburg Dam Modification Upgrade / Cedar River LWD Management Plan

Prepared for

Seattle Public Utilities Watershed Management Division Cedar River Watershed 19901 Cedar Falls Road SE Seattle, Washington 98045

December 31, 2010

Prepared by

1501 Fourth Avenue, Suite 1400 Seattle, WA 98101-1616

(206) 438-2700

In Association with Herrera Environmental Consultants, Inc.

Project No. 33761745 i

K:\City of Seattle Public Utilities\Large On Call Program - 2006\Landsburg LWD\Reports\Analysis Report\Final\Landsburg Final Report 12-31-10.doc

CONTENTS

1.0 INTRODUCTION ................................................................................................................. 1-1

1.1 Project Goals and Objectives ............................................................................... 1-1

1.2 Scope of Work ..................................................................................................... 1-1

1.3 Evaluation Criteria ............................................................................................... 1-3

1.4 Reference Information and Available Data ......................................................... 1-4

2.0 BACKGROUND ................................................................................................................... 2-1

2.1 Relevant History .................................................................................................. 2-1

2.2 Survey Datum ...................................................................................................... 2-2

2.3 Basin Hydrology and Municipal Watershed........................................................ 2-2

2.4 Flow Frequency ................................................................................................... 2-3

2.5 Hydraulic Information ......................................................................................... 2-4

2.6 Riparian Conditions ............................................................................................. 2-4

3.0 DESCRIPTION OF TIERED ALTERNATIVES ................................................................. 3-1

3.1 No Action / Existing Condition ........................................................................... 3-1

3.2 2006 Proposed Dam Modification ....................................................................... 3-4

3.3 High Flow Bypass Spillway................................................................................. 3-5

3.4 Log Boom Upstream of the Dam....................................................................... 3-13

3.5 In-channel Log Retention Systems .................................................................... 3-18

3.6 In-channel LWD Stabilization Using ELJs........................................................ 3-19

4.0 GEOMETRIC ANALYSIS OF WOOD PASSAGE VERSUS BLOCKAGE...................... 4-1

4.1 Wood Mobility..................................................................................................... 4-1

4.2 No Action / Existing Conditions.......................................................................... 4-2

4.3 2006 Dam Modification (Widened Gate) ............................................................ 4-2

4.4 High Flow Bypass Spillway – Options A and B.................................................. 4-2

4.5 Summary .............................................................................................................. 4-2

5.0 ECONOMIC ANALYSIS ..................................................................................................... 5-1

5.1 Capital Costs ........................................................................................................ 5-1

5.2 Operations and Maintenance Costs (O&M) ........................................................ 5-3

5.3 Major Repair and Replacement Costs (MRR) ..................................................... 5-3

5.4 Present Value Cost Estimate Summary ............................................................... 5-3

6.0 RISK EVALUATION ........................................................................................................... 6-1

6.1 Risk Model........................................................................................................... 6-1

7.0 CONCLUSIONS AND RECOMMENDATIONS ................................................................ 7-1

7.1 Conclusions.......................................................................................................... 7-1

7.2 Recommendations................................................................................................ 7-4

8.0 REFERENCES ...................................................................................................................... 8-1 ii

K:\City of Seattle Public Utilities\Large On Call Program - 2006\Landsburg LWD\Reports\Analysis Report\Final\Landsburg Final Report 12-31-10.doc

TABLES

Table 2-1. Flow Frequency data from 1918 to 1996 for the Cedar River at Landsburg ........... 2-3 Table 3-1. Comparison between Present Bypass Concepts and MWH 2005 Alternative 3 ..... 3-11 Table 4-1. Unstable Wood Passing Dam or Blocked at Opening (no Upstream Log

Boom) ....................................................................................................................... 4-3 Table 5-1. Opinion of Probable Project Cost.............................................................................. 5-2 Table 5-2. Present Value Summary ............................................................................................ 5-4 Table 6-1. Assigned Response Probabilities for the Existing Dam Conditions ......................... 6-3 Table 6-2. Probability of LWD Blockage at Dam for Alternatives (Year 2105, 500-yr

Event)........................................................................................................................ 6-5 Table 6-3. General Probabilities of Damage at Existing Dam.................................................... 6-7 Table 6-4. Assigned Outcome Probabilities (Existing Structure, 100-year Event) .................... 6-8 Table 6-5. Outcome Consequences............................................................................................. 6-8 Table 6-6. Total Expected Annual Risk Cost per Alternative .................................................. 6-11 Table 6-7. Total Present Value Risk Cost per Alternative........................................................ 6-12 Table 7-1. Project Alternatives Summary................................................................................... 7-4

FIGURES

Figure 2-1. Wood Loading Upstream of Landsburg by Volume per 100 meters ....................... 2-6 Figure 2-2. Unstable Instream Wood versus Flow Rate as Predicted by HEC-RAS ................. 2-9 Figure 3-1. Existing Conditions and “At Dam” Alternatives Plan View ................................... 3-2 Figure 3-2. Large Woody Debris with a Spillway-blocking Rootwad (Herrera 2007). ............. 3-3 Figure 3-3. Examples of Possible Configurations for Spillway-Blocking Debris (Herrera

2007). ........................................................................................................................ 3-5 Figure 3-4. Proposed High Flow Bypass Spillway Plan – Option A: Rock Channel................. 3-7 Figure 3-5. High Flow Bypass Sections – Option A: Rock Channel.......................................... 3-8 Figure 3-6. Proposed High Flow Bypass Spillway Plan – Option B: Rock Key........................ 3-9 Figure 3-7. High Flow Bypass Sections – Option B: Rock Key .............................................. 3-10 Figure 3-8. At-Dam Log Boom Plan ........................................................................................ 3-14 Figure 3-9. At-Dam Log Boom Plan Details ............................................................................ 3-15 Figure 3-10. Typical At-Dam Log Boom Detail (Worthington Products, 2010) ..................... 3-16 Figure 3-11. In-Channel Log Retention Structure Conceptual Placement ............................... 3-21 Figure 3-12. In-Channel Log Retention Structure Conceptual Details..................................... 3-22 Figure 3-13. Woody ELJ Structure Conceptual Details ........................................................... 3-23 Figure 6-1. Copy of Calculation: Risk...................................................................................... 6-10 Figure 6-2. Copy of Calculation: Present Value Costs (Risk and Capital)............................... 6-11

APPENDICES

Appendix A. Project Approach and Evaluation Criteria Report Appendix B. LWD Inventory and Wood Loading Appendix C. Opinion of Probable Costs Appendix D. Risk Analysis

1-1

K:\City of Seattle Public Utilities\Large On Call Program - 2006\Landsburg LWD\Reports\Analysis Report\Final\Landsburg Final Report 12-31-10.doc

1.0 INTRODUCTION

1.1 PROJECT GOALS AND OBJECTIVES

The goal this project is to analyze alternatives that reduce the risk of significant damage or complete loss of the Landsburg Dam caused by blockage from the accumulation of large woody debris (LWD) during a high flow event. Specifically, flow events with a recurrence interval of 500 years will be evaluated, in accordance with the Washington State Department of Ecology Dam Safety requirements (Ecology, 2007). The main project objective is to determine the best alternative for LWD management that meets Dam Safety regulations.

Management of LWD is necessary to prevent a prolonged disruption in the ability of the facility to deliver two thirds of Seattle’s drinking water supply. Improving worker safety during the handling of flood debris at the dam is also an important issue. Currently, operations staff and contractors hired to remove and manage LWD that becomes lodged in the dam spillway during high flows may be subject to unsafe conditions due to the limited access available for removing material.

Without flood passage improvements (such as increased spillway capacity to address gate blockages, LWD management in the watershed, and/or flood debris handling improvements) the dam and associated infrastructure remain at risk of damage or loss during large-magnitude flow events. The costs to replace the structure, should it fail during such an event, greatly exceed the estimated costs of the LWD management alternatives considered in this report. The new fish passage facility and spring water source for the Landsburg Sockeye Salmon Hatchery could also be damaged if the dam were to fail. Additionally, the potential exists for the water supply diversion to be out of service for months following a failure of the dam, which could preclude the City’s ability to deliver water from the Cedar River for even longer periods, assuming that major in-river repair work would need to hold-off until the summer, low-flow season.

Another important aspect is to uphold the goals of the Habitat Conservation Plan (HCP). Items specific to the HCP are maintaining fish passage, minimizing the new road construction, and protecting water quality.

In summary, the project goals and objectives are as follows:

Reduce risk to the Dam. Allow safe passage of water and wood up to and including the 500-year recurrence flow as required by Dam Safety Regulations (Ecology 2007).

Improve worker safety handling flood debris at the Dam.

Uphold the Habitat Conservation Plan (HCP) goals.

1.2 SCOPE OF WORK

The scope of work for the project included reviewing previous work to determine the level of risk at the existing dam configuration due to the accumulation of LWD during a 500-year flow

1-2

K:\City of Seattle Public Utilities\Large On Call Program - 2006\Landsburg LWD\Reports\Analysis Report\Final\Landsburg Final Report 12-31-10.doc event. The scope of work also includes an evaluation of risk-reduction measures using a tiered alternatives approach which considers the best alternative or combination of alternatives to meet the project goal of passing the 500-year recurrence flow and associated LWD.

At the onset of the project, four initial alternatives were identified, including:

Alternative 1: In-Channel Stabilization of LWD.

Alternative 2: In-Channel Retention System/Log Boom System.

Alternative 3: 2006 Proposed Dam Modification.

Alternative 4: Southbank Roughened Channel/High Flow Bypass.

However, this approach was revised early in the project. On October 29, 2009, during a site visit and workshop with SPU staff the order of the tiered project approach was reversed. The tiered alternatives evaluation will start downstream, at the dam instead of upstream, within the watershed.

The bottom-up, tiered approach is composed of the following elements:

1. Alternative 1: No Action/Existing Condition:

o Evaluate the capacity of the dam under existing conditions.

2. Alternative 2: 2006 Proposed Dam Modification:

o Evaluate the risk reduction of the proposed dam modification option that would combine two 20-foot wide bays into one 44-foot wide bay.

o Address the residual risk at the dam by specifying the volume of wood required to be bypassed and/or retained.

o Evaluate improved worker safety by considering the proposed on-structure debris removal/manipulation equipment, the potential need for heavier equipment than currently proposed, the location and type of equipment, and the option for the remote operation which would improve worker safety during higher flows.

3. Alternative 3: High Flow Bypass Spillway:

o Evaluate a roughened channel/high flow bypass with fuse plug to be used during extreme events ranging between the 100- and 500-year recurrence flow. The new spillway design would include a fuse plug, or a “soft” left abutment with grade control and scour protection embedded at depth within erodible fill. Two options were included:

Alternative 3A represents an open channel spillway that is lined with rock along the entire length of the spillway channel.

1-3

K:\City of Seattle Public Utilities\Large On Call Program - 2006\Landsburg LWD\Reports\Analysis Report\Final\Landsburg Final Report 12-31-10.doc

Alternative 3B consists of a rock key which provides protection of the dam, but would allow the surrounding soils (to the left when looking downstream) to erode during a dam overtopping event.

4. Alternative 4: Log Boom Upstream of the Dam:

o Evaluate a single log boom system upstream of the dam. Identify fail-safe options that would provide protection from catastrophic failure of this type of system.

o The system would be designed such that it does not disrupt fish passage.

5. Alternative 5: In-channel Log Retention System:

o Evaluate the alternative of utilizing in-channel log retention systems/log booms in the watershed using pile-supported partial spanning log structures.

o Structure locations must consider access for both construction and for regular maintenance (wood removal). The structures would be required to provide passive protection during large events (i.e., no wood removal during large events).

Although HCP goals include road decommissioning, the limited construction of short spur roads to the log-retentions structures would be considered.

6. Alternative 6: In-channel Stabilization of LWD using Engineered Log Jams o Consider in-channel stabilization of LWD using Engineered Log Jams (ELJs).

The Project Approach and Evaluation Criteria Report (URS/Herrera, 2009), presented in Appendix A, documents the information gathered and the decisions made during the site visit/meeting on October 29, 2009.

1.3 EVALUATION CRITERIA

The most important criterion in the evaluation of alternatives is the effectiveness in reducing risk to dam. Evaluation criteria were used as a screening tool to refine alternatives and will not be used to rank various alternatives. Environmental benefits were evaluated qualitatively. The evaluation criteria are organized in the following triple-bottom-line categories:

Financial

Capital costs.

Simplicity of design and operation.

Operations, maintenance, and repair.

Replacement period.

Annual risk (costs based on the likelihood and cost of the consequence of failure).

1-4

K:\City of Seattle Public Utilities\Large On Call Program - 2006\Landsburg LWD\Reports\Analysis Report\Final\Landsburg Final Report 12-31-10.doc

Social

Effectiveness (dam safety, maintains water supply for customers, downstream effects).

Improved worker safety.

Environmental

Habitat benefits.

Consistency with the HCP.

Adverse impacts (degree of mitigation required).

Cost estimates, presented in Section 5, include the following items for each alternative:

Planning-level/conceptual capital costs (low, most likely, and high); Operations and maintenance costs; and Lifecycle costs that include lifespan and/or the replacement period of the alternative being considered.

The risk analysis, presented in Section 6, includes annual risk costs for each alternative based on the likelihood and cost of the consequence of failures.

1.4 REFERENCE INFORMATION AND AVAILABLE DATA

The following key information related to the watershed and dam was referenced in this evaluation:

Wood Inventory within the Cedar River Riparian Corridor (Herrera, 2005; Updated by SPU in 2007).

Preliminary Engineering Study Final Report (MWH, 2005).

2006 Proposed Dam Modification Drawing and Basis of Design Report (MWH, 2006).

Large Woody Debris Management Plan (Herrera, 2007).

HEC-RAS models of the Cedar River (WEST Consultants, 2008).

Updated Wood Mobility Analysis Spreadsheet (SPU, 2009).

2-1

K:\City of Seattle Public Utilities\Large On Call Program - 2006\Landsburg LWD\Reports\Analysis Report\Final\Landsburg Final Report 12-31-10.doc

2.0 BACKGROUND

The dam was originally constructed as a timber crib structure in 1901, with a water supply line constructed along the left bank. The existing concrete and gated diversion dam was constructed in 1935 to replace the original timber crib structure. The concrete spillway structure is 26 feet high and 130 feet wide. The structure supports four radial gates, each 20 feet wide by 6.5 feet high, and a new 20-foot wide tip-out gate, which was constructed as part of the recent fish passage improvements. A fish ladder facility is located on the right abutment (looking downstream). Temporary fish hatchery facilities and the associated spring water supply are located just upstream and downstream of the left abutment. With its low height and its shallow impoundment of about seven to eight feet of depth, the dam does not create significant storage, nor does it pose significant safety concerns for downstream communities if a dam breach were to occur.

2.1 RELEVANT HISTORY

Almost all of the riparian forest in the watershed above the dam was clearcut harvested within the last 80 years. Debris loading at the dam was not a major concern until a blockage of the dam occurred in 1990 during an estimated 100-year recurrence flow event (10,800 cfs). During the 1990 flow event, debris blockage in the forebay of the dam resulted in elevated water levels that caused severe erosion at both abutments. Management efforts by operations staff to remove the LWD that accumulated during this flood event were successful in limiting damage to the dam.

Approximately 920 cubic yards (cy), or 700 cubic meters (m3) of LWD are estimated to have accumulated at the dam in 1990 (Dave Beedle, personal communication). The partial blockage of the radial gates of the dam during this event reduced the dam conveyance from a design capacity of 17,400 cfs (in excess of the 500-year recurrence flow) to a capacity of approximately 10,000 cfs (roughly corresponding to the 100-year event).

Based on the 2007 LWD Management Plan prepared by SPU by Herrera, wood loading in the Cedar River above Landsburg (and delivery of this wood to Landsburg) is expected to double over the next 100 years. Based on a forest-growth model prepared by SPU, the size of trees delivered to the river will also increase, but will not be large enough to provide key logs necessary to form stable logjams. Therefore, the risk of blockage by LWD at the dam will increase if no action is taken. With the projected increase in the volume and size of trees recruited to the Cedar River upstream of the dam, SPU is concerned with the effects of increased LWD size and loading on the operation of the dam and intake facility.

Since 1990, SPU has evaluated several dam modification alternatives to safely pass flow and LWD through or around the dam during an extreme event. These included modifications to the existing radial gates and an emergency spillway through the left abutment (MWH, 2006). The preferred alternative selected by SPU included combining two spillway gates to form a 44-foot-wide gate. Based on a preliminary assessment of LWD loading (Herrera, 2007), the dam modification option was determined to be insufficient. The MWH study in 2005 was based on the available information at that time before the assessment report on LWD was completed by

2-2

K:\City of Seattle Public Utilities\Large On Call Program - 2006\Landsburg LWD\Reports\Analysis Report\Final\Landsburg Final Report 12-31-10.doc

Herrera in 2007. Herrera’s report identified LWDs with rootwad sizes large than expected for the proposed modification, which leads the determination that the proposed dam modification option is insufficient to pass LWD and the flood. The conclusions lead SPU to this reevaluation and study.

2.2 SURVEY DATUM

Elevations referenced in this report are based on the North American Vertical Datum of 1988 (NAVD 88) unless otherwise specified.

2.3 BASIN HYDROLOGY AND MUNICIPAL WATERSHED

The headwaters of the Cedar River are located at the crest of the Cascade Range and drain northwesterly into Chester Morse Lake and the Masonry Pool. Runoff from the upper watershed is regulated at the Masonry Dam and flows into the lower Cedar River via the dam valves, spillways, and penstocks of the City’s Cedar Falls hydroelectric plant located approximately two miles downstream of the Masonry Dam. The Masonry Dam was constructed to a height of 215 feet (65.5 meters) in 1914 to increase reservoir storage and replace the original timber crib structure constructed in 1904. The dam increased the capacity of the lake as a water supply reservoir and increased the hydroelectric capacity of the Cedar Falls Power Plant operated by Seattle City Light. The City operates the reservoir facilities primarily as a water supply source but also for hydroelectric power and to maintain target instream flow levels for the benefit of downstream fish populations. Although the Masonry Dam was not financed or built for flood control purposes, dam management strategies include flood control operations to protect the health and safety of people and their property along the lower Cedar River.

Downstream of the Masonry Dam, the Cedar River flows over Lower Cedar Falls and through the steep bedrock gorge that marks the range front of the Cascade Mountains and the transition to the glacial deposits of the lower Cedar River. Taylor Creek and tributaries draining the southern flank of Taylor Mountain contribute unregulated flow to the lower Cedar River. Peak flow in lower Cedar River occurs in the late fall and winter months and is dominated by stormwater runoff; the lowest flows are typically measured in the summer. In the spring, runoff from snowmelt is more significant in the higher elevations of the Cedar River basin, upstream of the Masonry Dam.

The Landsburg Dam is operated as a run-of-the-river facility, which means that there is essentially no flow regulation. Flow that is not diverted for water supply passes over the dam.

Water levels behind the dam are controlled by four radial Tanner gates and one tip-out gate measuring 20 feet (6 meters) wide by 6.5 feet (2 meters) high. The replacement of two gates with a single gate measuring 13.5 feet (4.1 meters) high by 44 feet (13.4 meters) wide has been proposed by SPU to improve flood conveyance and debris passage at Landsburg. The backwater pool created by the diversion dam submerges the screen house forebay and intake structures, which divert water to the Lake Youngs aqueduct. Flow at Landsburg is regulated primarily by the operation of the Masonry Dam at Chester Morse Lake, approximately 13 miles (21 km) upstream of Landsburg.

2-3

K:\City of Seattle Public Utilities\Large On Call Program - 2006\Landsburg LWD\Reports\Analysis Report\Final\Landsburg Final Report 12-31-10.doc

SPU currently supplies water to 1.3 million customers throughout King County. Residents of Seattle receive their water from the Cedar River, the Tolt River, and three wells in the Highline Well Field. Serving these customers requires an annual average supply of 150 million gallons of water per day. Water from the Cedar River supplies approximately 60% of SPU’s total needs.

2.4 FLOW FREQUENCY

An analysis of flow frequency (Table 2-1) was performed for SPU by Raytheon (1998a) for the Cedar River at Landsburg. The flow-frequency analysis was performed using historical flows measured at U.S. Geological Survey stream gauge 12117500, which is located 1.8 miles upstream of the Landsburg Dam. This stream gauge was installed in 1895 and is the oldest continuously operated gauge in western Washington. Flow frequencies for conditions after the dam construction were determined by Raytheon (1998a) from a Log-Pearson III analysis of historical flow data for the period 1918 to 1996. Peak annual flows from years prior to 1914 do not reflect the current flow regulation by the Masonry Dam and include three of the four largest flow events recorded at Landsburg. The largest recorded flow (14,200 cfs) occurred on November 19, 1911, after the failure of the timber crib dam at Chester Morse Lake. This discharge is equivalent to about a 300-year flood under the current hydrologic regime.

Table 2-1. Flow Frequency data from 1918 to 1996 for the Cedar River at Landsburg

Return Period (years) Peak Flow (cfs)

2 2,400 5 3,900 10 5,100

25 7,000

50 8,700 100 10,700 200 13,000 500 16,000

1,000 18,500

(Raytheon 1998a)

Under the current conditions of flow regulation at the Masonry Dam, the 100-year discharge is approximately 10,700 cfs. The 1990 Thanksgiving Day flood (10,800 cfs) exceeded the estimated 100-year discharge and all previous discharge records since the completion of the Masonry Dam in 1914. The 500-year discharge of the Cedar River at Landsburg is approximately 16,000 cfs.

An update to the hydrology, to include flow records from 1996 through the present day, may be considered in the future by SPU. The flood frequency data presented in Table 2-1 have been used for HEC-RAS modeling (WEST Consultants, 2008), alternatives design studies (MWH

2-4

K:\City of Seattle Public Utilities\Large On Call Program - 2006\Landsburg LWD\Reports\Analysis Report\Final\Landsburg Final Report 12-31-10.doc

2001, 2005, 2006), and the 90% Dam Modification Design (MWH 2007); therefore, the analyses conducted for this evaluation will use flows listed in Table 2-1.

2.5 HYDRAULIC INFORMATION

2.5.1 Dam Conveyance Hydraulics

A hydraulic engineering analysis of the dam (MWH, 2005) indicates that the current dam structures can pass the 500-year recurrence flow event assuming no LWD loading. The existing spillways can pass up to about 17,400 cfs before the upstream water level exceeds the abutment sections of the dam at an elevation of 555.5 feet. A comparison of the rating curves (MWH, 2005) with the wood loading that induced overtopping during the 1990 flood suggests that blockage by 920 cy (700 m3) of LWD in 1990 was equivalent to blockage of two of the five existing bays.

2.5.2 Watershed Hydraulics

A HEC-RAS hydraulic model of the Cedar River upstream of the dam was developed for SPU by WEST Consultants (2008). SPU used the model output to complete an analysis of instream wood stability using the 2005 wood inventory. The HEC-RAS model evaluated steady flow conditions for recurrence flows of 1.2-, 1.6-, 2-, 5-, 10-, 20-, 50-, 100-, and 500-years. The HEC-RAS model also evaluated unsteady flow conditions using hydrographs developed for recurrence flows of 1.2-, 2-, 5-, 10-, 20-, and 100-years.

Model inputs show that flow within the study reach is controlled by releases from Masonry Dam.

For instance, the 500-year flow in the mainstem only increases by 6% (941 cfs) at Taylor Creek, the largest tributary between the Masonry and Landsburg dams.

A hydrograph of the design event was required to quantify wood loading at the dam. Because the HEC-RAS model only evaluated steady flow for the 500-year event, a hydrograph for this event was generated by scaling the 1990 hydrograph up to the 500-year instantaneous peak of 16,000 cfs. The scaled hydrograph is presented in Appendix B.

2.6 RIPARIAN CONDITIONS

Before European settlement, riparian forests along the lower Cedar River and much of the western flank of the Cascade Range contained a diverse mix of multi-age, conifers and minor hardwoods. Historical forest clearing along the entire reach of the lower Cedar River modified the natural riparian forest conditions. Forest clearing in the watershed began during the 1880s in the western lowlands around Landsburg and generally proceeded eastward to less accessible forests in the higher elevations of the upper watershed. By 1915, nearly the entire lower watershed had been logged. The City of Seattle started a process of land purchase and land condemnation within the watershed in the early 1900s. The City started a replanting program in the project area in the 1920s to establish tree cover in burned-over areas and along the river. By

2-5

K:\City of Seattle Public Utilities\Large On Call Program - 2006\Landsburg LWD\Reports\Analysis Report\Final\Landsburg Final Report 12-31-10.doc the 1940s, the City had purchased most of the land within the hydrographic boundary between Cedar Falls and Landsburg. Riparian forests have recovered to mid-seral stands since the 1940s.

The subsections that follow describe the riparian forest conditions under existing and projected (100-year growth); the inventory of instream wood measured in 2005 and 2007; and the predicted wood mobility based on the 2005 inventory and 100-year projections of instream wood loading.

2.6.1 Forest Conditions

Forest conditions in the lower watershed show a predominance of early to mid-seral stands less than 85 years old. Minor patches of riparian forest younger than 25 years old occur locally along the lower Cedar River. A remnant patch of original old-growth forest can be found on the left bank immediately upstream of the dam. The dominant riparian tree species along the lower Cedar River include Douglas-fir, western red cedar, western hemlock, Sitka spruce, red alder, and big-leaf maple. The City planted much of the Sitka Spruce along the river to address water quality concerns.

2005 Summary

SPU began long-term monitoring of riparian forest conditions in 2005 with the establishment of 30 permanent riparian plots at 15 sites along the Cedar River and selected tributary streams. In each plot, data for all trees with a diameter at breast height (DBH) of at least 5 inches (13 centimeters) were recorded. Collected data included species, DBH, height, crown ratio (crown as a percentage of height), and decay class for snags. The number of trees in each plot was used to calculate tree density. The results of the 2005 sampling (Herrera, 2007) were used as input for a forest growth model in order to project future riparian forest conditions and changes in the size and distribution of future wood recruitment. In general, the size distribution of trees measured in riparian plots is right-skewed toward larger sizes (in terms of both diameter and length). The median DBH of trees in existing forest plots is 5.9 inches, with a maximum DBH of 7.7 feet.

Future - Projected 2105

SPU estimated future forest conditions using the forest vegetation simulator (FVS) model distributed by the U.S. Forest Service. Results of the model are described by Herrera (2007) and summarized herein. The FVS model calculates forest growth and timber yield based on initial forest conditions and algorithms calibrated to specific geographic areas. Wood histograms generated from the current riparian plots (2005) and from 100-year projections of forest conditions simulated with the FVS model (2105) show a considerable shift in the size distribution of the riparian forest over the next century. The trees with the largest diameter in the 2005 distribution are old snags (dead trees) that are expected to decay over the next century and disappear from the distribution and reduce the degree of skewing. By 2105, the forest distribution is expected to include fewer small-diameter trees and more trees of intermediate diameter and length as riparian forests along the lower Cedar River continue to recover from twentieth-century forest clearing.

Although 100-year projections using the FVS model predict a decline in the fraction of the largest-sized trees, increases in the number of intermediate-sized trees is expected to yield a 2.8-fold increase in overall wood volume within the riparian forest by 2105. The expected increase in wood volume within the riparian forest implies an increase in the rate of wood recruitment over the next century.

2.6.2 Inventory of Instream Wood

2005 Wood Inventory

An inventory of instream LWD was conducted in July 2005 between Landsburg and Cedar Falls (River Kilometer (RK) 34.9 to RK 55.0 (or River Mile (RM) 21.7 to RM 34.2)) to determine the character and volume of instream wood that could potentially move downstream to Landsburg during large flow events. Wood volumes were calculated in units of cubic meters (m3) of wood per 100 meters of channel length for comparison with reference conditions reported by previous studies. A full description of the field methods and the results of the 2005 inventory are presented by Herrera (2007).

The 2005 inventory counted 871 pieces or 2,935 m3 (3,840 cy) of LWD along 20 kilometers (12.4 miles) of the lower Cedar River. Local wood loading rates ranged from zero to 196 m3/100 meters (233 cy/100 yards), with the highest rates measured along the reaches affected by blowdown during the 2003 windstorm (RK 40.8 (RM 25.5) and RK 42.2 (RM 26.2). Figure 2-1 illustrates the LWD loading upstream of Landsburg. Overall, the average rate of wood loading measured along the 20-kilometer (12.4 mile) project reach is 14.7 m3/100 meters (17.5 cy/100 yards). The average rate of wood loading in terms of the number of pieces of LWD is 4.3 pieces/100 meters (3.9 pieces/100 yards).

35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54

River kilometer

L W

D d e n s it y m 3 /1 m

Figure 2-1. Wood Loading Upstream of Landsburg by Volume per 100 meters

2-6

K:\City of Seattle Public Utilities\Large On Call Program - 2006\Landsburg LWD\Reports\Analysis Report\Final\Landsburg Final Report 12-31-10.doc

2-7

K:\City of Seattle Public Utilities\Large On Call Program - 2006\Landsburg LWD\Reports\Analysis Report\Final\Landsburg Final Report 12-31-10.doc

The current trend in wood loading (i.e., whether instream wood volumes are increasing or decreasing with time) was evaluated by comparing the results of the 2005 LWD inventory with wood loading rates reported by Fox (2003) for rivers throughout western Washington that recruit wood from young and old-growth riparian forests. The comparison with old-growth riparian forests provides some insight into possible pre-settlement wood loading in the lower Cedar River and the potential for long-term wood loading as the riparian forest recovers from historical timber harvesting and instream wood removal.

The average volumetric rate of wood loading measured along the 20-kilometer project reach (14.7 m3/100 meters) is roughly half the reference loading rate of 30 m3/100 meters (35.7 cy/100 yards) measured by Fox (2003) for rivers that regularly recruit wood from riparian forests that are younger than 150 years. In terms of the number of LWD pieces, the rate of wood loading upstream of Landsburg (4.3 pieces/100 meters) is only one-tenth the average loading rate measured by Fox (2003). Additionally, the average number of LWD pieces in a logjam measured in the 2005 inventory (less than 5 pieces) was considerably smaller than the average group size of 20 measured by Fox (2003). The comparison of the 2005 inventory with reference conditions indicates that the rates of instream wood loading along the lower Cedar River will increase over the next century.

2007 Inventory Update and Comparison to 2005

A second instream wood inventory was completed in August 2007 along the same 20-kilometer reach as the 2005 inventory. The 2007 inventory occurred after an approximately 10-year recurrence flow in November 2006 and a wind storm in December 2006. The 2007 inventory yielded 734 pieces of LWD, which is approximately 130 pieces fewer than the 2005 inventory.

The apparent reduction in the number of instream pieces can be attributed to minor changes in field protocols for determining if wood falls within the bankfull channel, the shearing off of the tops of trees recruited during the 2003 blowdown event that were within the bankfull channel in 2005, underestimating the number of hidden pieces in large logjams, and wood mobilized downstream of Landsburg during the November 2006 flow event. Results of the 2007 inventory were not analyzed to the same extent as were the results of the 2005 inventory; therefore, the exact cause of the discrepancy in wood loading between the inventories has not been determined.

Future - Projected 2105

Results of the 2005 wood inventory indicate that blowdown is the dominant mechanism for wood recruitment in the project reach. Based on the results of the FVS model, the volume of wood within the riparian forest available for recruitment is expected to double or triple over the next 100 years. Assuming that larger trees would be as susceptible to blowdown as those included the current forest composition, wood recruitment by this dominant recruitment mechanism should deliver more wood to the lower Cedar River over the next century and increase the volume of instream wood that might accumulate at Landsburg during the design flood event. Based on relationships between the size distribution of the 2005 riparian plots and the distribution of instream wood, Herrera (2007) used the results of the FVS model to predict the size distribution of instream wood present in 2105. The analysis predicts that the size distribution of instream LWD will coarsen to include more pieces larger than 0.8 meters (31.5 inches) in diameter and fewer pieces of LWD with diameters less than 0.8 meters. Additionally, 2-8

K:\City of Seattle Public Utilities\Large On Call Program - 2006\Landsburg LWD\Reports\Analysis Report\Final\Landsburg Final Report 12-31-10.doc the volume of instream wood is expected to double by 2105 if wood loading is to keep pace with loading rates measured by Fox (2003).

2.6.3 Wood Mobility and Loading at Landsburg

The database developed from the 2005 wood inventory was used by SPU along with the HEC- RAS model output to predict wood mobility for various recurrence flows up to the 500-year event. Maximum potential wood volumes mobilized by various flows were calculated by SPU using the instantaneous flow rates and hydraulic conditions for the steady model results. The procedure involved calculating the buoyancy of each log based on the local flow depth predicted by the HEC-RAS model. Logs that were fully buoyant were classified as unstable. For this study, the volume of existing (2005) unstable wood was doubled to simulate 2105 conditions (Herrera 2007). The total number of unstable logs and the volume of unstable wood for each flow scenario are plotted against discharge in Figure 2-2.

Also plotted in Figure 2-2 as a point, is the wood loading reported by SPU during the approximately 100-year event in 1990. This historical information provides a key calibration point for determining the maximum wood loading and additional conveyance requirements at the dam during the 500-year event. For instance, if a 500-year event were to occur with the same wood loading as in 1990, additional overflow capacity would be needed to pass approximately 6,000 cfs.

The quantification of wood loading at the dam during the 500-year event under 2105 conditions requires both an assessment of the quantity of unstable wood in the system and consideration of LWD-mobilizing flow, distance to the dam, and flow velocity. Using a velocity of 1.66 m/s (5.4 feet per second (fps)) derived from half of the average flow velocity (wood tends to move slower due to drag on water, bed, and banks) and a distance of 20 km, it would take 3.5 hours for all the unstable wood to reach Landsburg. Based on the 500-year hydrograph developed for this study, the minimum flow during this 3.5-hour duration is about 15,000 cfs. A linear interpolation between the 100-yr and 500-yr wood loading provided by SPU results in an estimated 8,000 m3 (10,500 cy) of wood delivered to the dam during the 500-yr event for conditions expected in 2105.

Figure 2-2. Unstable Instream Wood versus Flow Rate as Predicted by HEC-RAS

2-9

K:\City of Seattle Public Utilities\Large On Call Program - 2006\Landsburg LWD\Reports\Analysis Report\Final\Landsburg Final Report 12-31-10.doc

3-1

K:\City of Seattle Public Utilities\Large On Call Program - 2006\Landsburg LWD\Reports\Analysis Report\Final\Landsburg Final Report 12-31-10.doc

3.0 DESCRIPTION OF TIERED ALTERNATIVES

The bottom-up tiered approach begins with the downstream-most alternative and works up stream. The alternatives are listed as follows:

Alternative 1: No Action/Existing Condition.

Alternative 2: 2006 Proposed Dam Modification.

Alternative 3A: High Flow Bypass Spillway – Rock Channel.

Alternative 3B: High Flow Bypass Spillway – Rock Key.

Alternative 4: Log Boom Upstream of the Dam.

Alternative 5: In-channel Log Retention System.

Alternative 6: In-channel Stabilization of LWD using Engineered Log Jams.

Figure 3-1: Existing Conditions and “At Dam” Alternatives Plan View, presents Alternatives 1, 2, 3A, 3B, and 4. Alternatives 5 and 6 would be located within the watershed upstream and would be placed in multiple locations depending on wood loading and desired level of protection. The six alternatives are described in the following subsections.

3.1 NO ACTION / EXISTING CONDITION

3.1.1 Description

Without wood loading, the existing spillway gates at the dam are capable of passing the 500-year flow and flows up to 17,400 cfs under fully functioning conditions. Based on the flood response in 1990, the dam is capable of passing a maximum flow of approximately 10,000 cfs (roughly a 100-year flow event) and debris loading of approximately 700 m3 to remove a portion of debris from the dam. During the 1990 event, equipment was mobilized to the dam to remove debris that was causing the left abutment to be overtopped and eroded. The equipment was able to remove enough debris to restore conveyance through the dam and arrest the overtopping and erosion within hours of arrival at the site (SPU staff, personal communication).

The dam has successfully passed all flow events since 1990, including LWD loading. This has been accomplished by using a logging contractor to mobilize large track-mounted excavators to the dam and remove LWD as it accumulates at the dam forebay. This has been successful to date in maintaining adequate conveyance through the dam. The largest flow event since 1990 was an approximately a 10-year event in 1995. During the 2003 event, approximately 300 m3 (400 cy) of LWD was removed from the dam forebay.

According to the record of wood removal since 1990, it has been shown that events up to the 10-year event can be managed by mobilizing equipment to the site for log removal. Based on an available forebay storage capacity of approximately 700 m3 (920 cy) of woody debris (approximated by the 1990 event) and the ability to remove approximately 300 m3 (400 cy) of woody debris with equipment, it is assumed that approximately 1,000 m3 (1,300 cy) of debris could be managed without damage and with some blockage for a peak flow of approximately 10,000 cfs or a 500-year flow with diversion of approximately 6,000 cfs around the sides of the dam.

3.1.2 Capacity / Hazard Assessment

The hazard assessment completed by Herrera (2007) indicates that the existing dam is at serious risk of blockage and overtopping during the 100-year flood event under existing and projected LWD loading scenarios. The risk to the dam is related not only to the overall volume of accumulated debris, but to the inability of the radial gates to pass even the smallest percentile logs. For instance, the wedging of one log in a single gate could cause a cascading effect, whereby additional wood could accumulate in the gate and render the gate inoperable. This is illustrated in Figure 3-2.

Figure 3-2. Large Woody Debris with a Spillway-blocking Rootwad (Herrera 2007).

3-3

K:\City of Seattle Public Utilities\Large On Call Program - 2006\Landsburg LWD\Reports\Analysis Report\Final\Landsburg Final Report 12-31-10.doc

3-4

K:\City of Seattle Public Utilities\Large On Call Program - 2006\Landsburg LWD\Reports\Analysis Report\Final\Landsburg Final Report 12-31-10.doc

3.2 2006 PROPOSED DAM MODIFICATION

3.2.1 Description

This alternative includes a new 8-foot high by 44-foot wide radial gate (in Bays 3 and 4). The Downstream Passage Gate (Bay 2) and existing Bays 1 and 5 radial gates would remain in-place and operational under this alternative. This alternative is described in studies and preliminary designs developed by MWH (2005, 2006, and 2007) and presented in plan view on Figure 3-1.

The modified spillway would increase the total free flow spillway capacity by 5% to about 18,000 cfs at an upstream water level of 555.5 feet. The capacity of the new spillway is 7,700 cfs versus the 7,000 cfs for the two existing gates to be removed. According to MWH (2005), given the estimated 500-year recurrence flow discharge of 16,000 cfs, with the new spillway (unobstructed), the equivalent of slightly less than one of the remaining three bays can be obstructed before the abutments are overtopped.

A new debris boom would be installed from the right abutment upstream of the intake area and would extend to the new spillway. The boom would aid SPU personnel in positioning floating debris to the larger opening. A new trash raking system would be included to manage debris.

The rake track would be replaced across the new gate bay.

The steel superstructure over Bays 3 and 4 would be replaced with new structural beams, decking and handrails. The deck would be extended about 20-feet or as needed to allow the rake to park off the deck to maintain free access across the spillway. All weather equipment access to the left abutment would be provided from the current access road.

3.2.2 Capacity / Hazard Assessment

Based on the hazard assessment completed by Herrera (2007), the proposed dam modification would not significantly reduce the existing risk at the dam during a 100-year event. The proposed modification would maintain the same vertical clearance and would, therefore, be prone to the same risk of rootwad blockage against the radial gate as the existing spillway gate configuration, as shown on Figure 3-2. Additionally, blockage of the proposed 44-foot wide spillway could be bridged by the 50th percentile log with or without rootwads (55 feet long) as shown in Figure 3-3. Logs of this size without rootwads that become blocked during flood events with a recurrence interval of approximately 10 years or less could be manipulated or moved with equipment to allow them to pass through the gates; however, the manipulation of logs under more extreme events (when flood conveyance through the dam depends on log removal) would prove difficult or impossible.

Wood loading for the dam modification during a 500-year flow event was simulated by routing unstable logs downstream at 1.66 m/s (5.5 fps) based on their distance above the dam as recorded during the 2005 inventory. The simulation assumed that all unstable logs with a rootwad greater than 6.5 feet in diameter and half of the logs with rootwads less than 6.5 feet and log length greater than 44 feet would not pass through the 44-foot wide gate opening. The results show that 1,000 m3 (1,300 cy) of wood (86 logs) would accumulate at the dam and would be unable to pass through the 44-foot wide gate during the first hour of a 500-year flow event. This estimate does not account for the accumulation of smaller logs due to blockage.

Figure 3-3. Examples of Possible Configurations for Spillway-Blocking Debris

(Herrera 2007).

3.3 HIGH FLOW BYPASS SPILLWAY

3.3.1 Description

A conceptual design for a roughened open channel high flow bypass spillway on the left bank abutment was developed (Figure 3-1). Initially, the geometric dimensions were based on the existing site constraints, including the available on site footprint, the existing radial gate configuration, and expected tailwater conditions during flood events.

The bypass was evaluated in conjunction with existing radial gates under partial blockage with the objective to pass the 500-year event flow and LWD, without damage to the operation and safety of the dam and associated structures. The dimensions of the spillway, including width, side slopes, and invert elevation, were adjusted from the initial dimensions to optimize flow conditions in order to optimize the passage of LWD.

This evaluation resulted in a trapezoidal open channel with a bottom width of 100 feet, side slopes of two horizontal to one vertical (2H:1V), and an invert elevation of 547.5 feet which would allow an overall flow depth of eight feet. This concept is defined as Alternative 3A: High

3-5

K:\City of Seattle Public Utilities\Large On Call Program - 2006\Landsburg LWD\Reports\Analysis Report\Final\Landsburg Final Report 12-31-10.doc

3-6

K:\City of Seattle Public Utilities\Large On Call Program - 2006\Landsburg LWD\Reports\Analysis Report\Final\Landsburg Final Report 12-31-10.doc

Flow Bypass Spillway – Rock Channel and is…