Cardinal Glen Stream Stabilization Evaluation Report

Executive Summary

The two goals for the Cardinal Glen Stream Stabilization Project (CGSSP) were stated as (1) to 'fix' the stream by reducing, or eliminating, the problem of streambank erosion and (2) to provide Loudon County with a demonstration project that may provide the basis for similar types of projects in the future.

The former goal (to 'fix' the stream) is not easlily measured. The short answer to the question of whether or not EQR's design and construction has 'fixed' the stream is, 'yes'. The rock toes were installed well and appear to be effective at reducing streambank erosion, especially considering the fact that a hurricane-sized event has been through the stream since the construction was completed. Vegetation on streambanks is a very effective measure used to reduce streambank erosion (especially above the bank toe level). There had been problems in getting vegetation established on the streambanks at the time of the writing of this report. These problems are identified in greater detail later in the report, as are associated factors, the most significant of which is constraints due to tree preservation which force graded streambank slopes to be steeper than is ideal for vegetative growth conditions. It should be noted that EQR will be re-planting some areas in December of 2003 with live stakes where this problem has been most persistant. Seeding could be incorporated into this effort with minimal costs to the HOA (less than $100). Hydroseeding of steep slopes could be considered as an action to help vegetative establishment in problem areas, however, this is costly effort, with a likely minimum charge of $1000 (assuming a cost of $1.00 per square yard). If vegetation on the streambanks continues to be a problem in the future, Cardinal Glen may wish to consider a more 'structural' solution to fix the streambank erosion problem (especially near Lots 26 and 27), such as the construction of a segmented block wall (i.e., Versalok, Keystone) on the north side of the stream near Lots 26 and 27. Costs for this type of activity range from $23-$26 per square foot of wall. Using an average of $25 per square foot, and a 50-foot long wall that is 9 feet high (with 1 foot used as a foundation), the total cost for this wold be approximately $11,250. Due to 'hard points' (i.e., rip-rap) at the up and downstream limits of the project, it is unlikely that vertical instability will be a problem in the future. The installation of step pools at the upstream limit could reduce that possibility further.

Previous problems with streambank erosion in the Cardinal Glen HOA were solved using rip-rap (rock) to line the stream. This is an effective and relatively inexpensive fix, however it is not aesthetically pleasing or environmentally friendly. Where the use of rip-rap lining in stream stabilization solutions represents the 'old' philosophy, the 'new' philosophy emphasizes a more natural approach, which includes the use of bioengineering techniques to stabilize streams. Loudon County has adopted this new philosophy, and wished to use the CGSS project to showcase some of these new techiques. Some of the solutions showcased in the CGSS project include:

• Rock toes
• Live staking
• Streambank grading and re-vegetating
• Soil stabilization matting on streambanks

Other solutions originally proposed include biologs, log v-drops, rock cross vanes and rock step pools. Rock toes were used in the place of biologs to provide a more permanent solution. Log vdrops, rock cross vanes and rock step pools, which were not constructed due to cost constraints ($2000 per rock cross vane for instance), are commonly used as vertical control structures and therefore would have been good measures to showcase. In summary, the project does a good job in demonstrating some measures commonly used in stream stabilization, though there are some key measures that were not used that could have improved the project's use as a broad-ranging demonstration project; it is a good first step in the process.

Introduction

History of Project

Site visits were made by several environmental science professionals who provided input on possible actions to mitigate the severe streambank erosion problems. These solutions included:

• Remove tree limbs and grass clippings from streambanks;
• Stabilize streambanks through a combination of biologs, geotextile fabrics and vegetative plantings;
• Remove trees that are in danger of falling due to undercut banks;
• Install a series of step pools at upstream limits of the stream reach to dissipate energy from existing riprap channel; and
• Limit access to the stream to reduce wear on stream banks.

Cardinal Glen

Figure 1. Location of Cardinal Glen development

After receiving proposed solutions from various specialists, the Cardinal Glen HOA decided to take action by issuing an RFP (request for proposal) detailing the scope of work as described in the above bulleted list. After soliciting a number of bids from private consultants and contractors, EQR was chosen for the stabilization work.

The EQR consulting division, headed up by Tim Schueler, studied the site, and had the site surveyed by J.A. Rice, Inc. A set of plans were produced in August of 2002 by this group, which detailed the proposed work. This work included the following proposed solutions:

• Grade banks back to 2:1 slope where possible
• Install rock toes at all outside bends
• Install stabilization blanket and matting on all disturbed slopes
• Vegetate banks with seeding and live staking
• Install v-log drop structures at tributary near upstream end of project area
• Install a drop structure in the manhole near downstream end of project area to lower 18' RCP outfall invert at stream

The construction section of EQR, along with Leo Construction, implemented the plan in March of 2003. The construction was completed by April, 2003.

Field Visit

• Peter O'Reilly : Loudon County Coordinator for FOSR/CGSSP and Staff Aide to Supervisor Bill Bogard
• Jim Babcock : Cardinal Glen HOA former president and current secretary
• Subijoy Dutta, PE : Loudon County resident
• Asad Rouhi : Northern Virginia Soil and Water Conservation Service
• Rick Scaffidi : EQR, Inc.
• Seth Brown, PE : GKY & Associates, Inc. (author of this document)

• The construction access area looked to be in good shape [author], see Figure 2.
• According to the plans, the 18" culvert outlet invert was to be approximately 0.8' higher in elevation than the stream invert. Due to rock outcroppings that were unexpected, Leo Construction had to place the structures at elevations that were 1'-2.5' higher than planned (see Figure 3) [Scaffidi].
• The rock toe in downstream-most bend on north side of stream (STA 0+05 to 0+40) is in good shape (see Figure 4). Note in Figure 4 the amount of woody debris built-up at the toe.
• A question was presented by Dr. Rouhi about the use of rock toes instead of a 'softer' and more bioengineered technique like coir logs or biologs. Mr. Scaffidi stated that due to the high amount of shade in the area, it would have been difficult to establish enough vegetation in the biolog to make it a stable measure to use. Mr. Schueler stated that shade was one reason the rock toes were proposed, with another reason being the velocities and shear stresses that were expected in the stream, which were more suited for the use of rock toes, and not biologs. Mr. Schueler provided backing caclulations for this claim.
• The rock toe on the south side of stream directly upstream of the 18' culvert (STA 0+40 to 0+75) is in good shape [author] (see Figure 5). The streambank, however, has not established adequate vegeative cover yet (again, see Figure 5). Mr. Scaffidi noted that several slopes will get additional planitings in December, 2003 (100 more plantings in total). Also, some slopes, such as the one shown in Figure 5, were to be cut at a 2:1 slope, but were actually graded closer to a 1.5:1 or 1:1 slope so as to not disturb trees on the stream banks.
• The rock toe on north side of stream between STA 0+65 and 1+45 is in good shape [author] (see Figure 6). Vegetation is more well-established in along this bank of stream. Note, however, the sediment build up near the rock toe. This section of stream experienced the most amount of grading, as the lateral alignment of the stream was 'pushed' towards the outside bend. [Mr. Scaffidi]
• Dumped grass clippings, and other dumped woody debris is still a problem in the area, as can be seen in Figure 7.
• The small tributary that enters the stream from the north near STA 1+60 was slightly realigned to provide better conveyance in the downstream direction and alleviate pressure on the opposing stream bank (see Figure 8) [Mr. Scaffidi]. Also, the plans called out 3 vlog drop structures in this tributary, however, due to budget constraints, these structures were not constructed. [Mr. Scaffidi]
• The rock toe located on the south side of the stream between STA 2+00 to 1+45 is in good shape [author] (see Figure 9). Again, this streambank does not have adequate vegetation established. Also, there has been a failure of the original erosion control matting on the slopes, as can be seen in Figure 10. New matting was placed in the area of the failed matting [Mr. Scaffidi].
• This area may have problems with vertical stability in the future, due to a high slope in the stream's longitudinal aligment immediately upstream of this area. [Dr. Rouhi]
• One alternative that should be considered for areas with very steep streambanks is to incorporate some type of retaining structures, such as a retaining wall. [Mr. Dutta]

Photographs from Field Visit

Figure 2 : Construction Access Area

Figure 3 : 18" RCP amd Manhole Structure

Figure 4 : Rock Toe and Woody Debris

Figure 5 : Rock Toe and Unvegetated Streambanks

Figure 6 : Rock Toe and Sediment Deposition

Figure 7 : Woody Debris along Streambanks

Figure 8 : Realigned Tributary Entering Stream

Figure 9 : Rock Toe and Failed Streambank Stabilization Matting

Figure 10 : Rock Toe and Failed Streambank Stabilization Matting

Summary of Observations and Implications

Lack of adequate vegetation on streambanks.

This is the most pervasive problem in the stream reach, especially for the streambanks located on the south side of the stream. Unvegetated streambanks are susceptible to erosion, which has been historically a problem in this stream reach, so it is imperative that vegetation become established quickly. The lack of established vegetation is most likely due to shade from trees not removed from south side of stream. One solution for this problem may be to use a shadetolerant seed mix to establish grass growth, although it should be noted that shade-tolerant seed mixes are more costly than standard seed mixes (exact costs depend on specific seed mix required). Another potential reason for the lack of vegetation on streambanks is the steepness of the banks. Seed mixes broadcast by hand can be easily washed off of steep slopes. One method to minimize seed wash-off is to use hydroseeding to broadcast the seed mix. This adds a mixture of liquid-based nutrients to help hold the seed in place and provides a fertilizer. As was stated in the Executive Summary, however, the cost for this method of broadcasting is not trivial. EQR also stated that 100 more live stakes were to be installed in December of 2003. This should help in the establishment of vegetative growth. If vegetation establishment is not attained, property owners of lots 26 and 27 may wish to consider a more structual solution, such as the construction of a segmented-block wall (Keystone, Versalok, etc.). These types of walls have been successfully used in many riverine-type applications. The estimated cost of a wall that is 50 feet long and 9 feet high (1 foot embedded in the ground for a foundation) is $11,250. Elevation between culvert outfall and streambed is greater than planned. The constructed elevation drop downstream of the 18' RCP could cause erosion of the streambed in the immediate vicinity, as well as place increased stress on opposing streambank. The contractor, Leo Construction, did place rip-rap below the outfall, and a rock toe is in place opposite the outfall, so the potential for erosion is small. This area should be observed closely, and if existing rock is displaced frequently in the future, consideration should be made to introduce larger-sized rip-rap in this area. Also, there is a backwater effect in this area during high flows due to the downstream road crossing, which should help to absorb the energy associated with high flows from the culvert outfall.

Small areas of streambank stabilization matting have failed.

It is not uncommon for streambank stabilization matting to fail in certain locations, especially on steep slopes, which is the case where there has been a failure in Cardinal Glen. Some matting has been replaced by new matting. Establishment of vegetation will help keep matting in place. There are isolated areas of sediment build-up. In the area where sediment build-up is evident, the stream channel alignment was shifted slightly. The sediment build-up is an attempt by the stream to maintain the original alignment. Also, it shows that this area is a sediment sink (meaning that it is an area where sediment will be deposited), which usually means that the cross-section is too large or the slope is too flat to convey sediment efficiently. One possible source for this sediment is the transitional area between the upstream highly-sloped area and the lower-sloped natural stream channel reach. Also, the unprotected streambanks upstream of this section could be the source of sediment. In general, this sediment shouldn't be a problem in this area, unless the upstream banks are not adequately protected in the near future.

Potential vertical stability problems in the future.

Prior to the development of the design, several outside experts suggested that step pools or grade control structures be used in the design. It does not appear to the author that there is an existing problem with vertical stability (i.e., down cutting of the streambed). 'Hard points' up and downstream of the reach in the form of rip-rapped channel protect the vertical stability of the stream reach. The rip-rapped channels have slopes approaching 4.5%-5%, which is high for a natural channel system, however, due to the rip-rap lining these channels can accommodate higher velocities and shear stresses associated with high-sloped streams. The reach of stream not protect by rip-rap has been the focus of work. This reach has a slope of approximately 0.9%, which is within the 'normal' range for natural stream slopes, and the endpoints of this natural stream are the rip-rap lined channel, which should effectively 'fix' the stream invert in place. The only area of concern, in terms of vertical stability, is the upstream end of the natural section of stream. This area has the most potential for streambed erosion due to the immediate change in slope. The author agrees with the experts who stated that step pools should be used in this area to help dissipate energy in this transitional area. It should be noted that in a letter dated October 16th, 2001 from Mr. Schuler to Mr. O'Reilly, it is stated that five rock cross-vanes (in stream structures that provide vertical stabiliy for streams) were proposed to be part of the design. In the above-mentioned phone conversation, Mr. Schueler stated that these were removed from the plan design due to budgetary constraints. The cost per structure was estimated by Mr. Schuler to be $2000.

Rock toes were used instead of biologs for toe protection.

According to the calculated velocities and shear stresses predicted, either rock toes or biologs could have been used for toe protection. The author agrees that biologs do not perform as well in applications with large amounts of shade, however, biologs have a lifetime of 1-3 years, which should be an adequate amount of time to establish vegetation. Also, the design specs for biologs show that they can withstand the velocity and shear stresses that have been predicted for this stream, and cost approximately half as much to purchase and install as rock toes. That said, biologs do eventually degrade completely, and are therefore considered a temporay solution. Also, rock toes provide more protection than biologs. Therefore rock toes provide a better chance of success (or rather, a smaller chance of failure), especially in the long-term.

A stream assessment was not performed.

In most stream restoration designs, a stream assessment is performed in order to better understand how the channel is reacting to the 'channel-forming discharge' (i.e., bankfull discharge). Understanding this helps the designer provide a geomorphically stable planform and cross-sectional geometry in the design. Mr. Schueler stated that an assessment was not performed for this project. It should be noted that the designer, Mr. Schueler, did inform the Cardinal Glen HOA that the proposed project was limited to a stream stabilization project, and is NOT a full-blown stream restoration project. This is a cruicial point, since the scope inherent with a stream stabilization (rather than a stream restoration effort) is limited, and often times, geomorphic assessments are NOT performed. Also, due to the many constraints of the project, a planform realignment was not feasible, and proper cross-sectional geometry design was not allowed (due to the desire of the HOA to preserve and protect the trees near or on the streambanks, and again, costs). Although a stream assessment is not necessary in a project of this scope, it would help to better defend the proposed design, and should be used in future projects, if full-blown stream restoration is performed.