Design For Fish Passage at Roadway - Stream Crossings: Synthesis Report

4 Culverts Inventory, Assessment and Prioritization

How to use this chapter

• Learn the importance of developing a culvert inventory database
• Learn how to assess the ability of existing culverts to pass fish of interest
• Understand the importance of prioritizing how barrier culverts are to be addressed
• See examples of assessment and prioritization schemes

4.1 Culvert Inventory

A culvert inventory can provide knowledge of location, adequacy, and potential cost of replacement/retrofit of roadway-stream crossings within a watershed context. With such knowledge, planners can begin to prioritize and plan for fish passage restoration on a watershed scale. A robust inventory will be invaluable in planning efforts, and many assessment schemes have been created to collect information necessary for the prioritization of crossing replacement (i.e. Clarkin et al. 2003; Taylor and Love 2003; Washington Department of Fish and Wildlife 2000).

4.1.1 Knowledge of Crossing Location

The first step in a program of fish passage restoration is awareness of the problem, including location and condition of waterway crossings. An inventory can be as simple as a listing of the locations of existing roadway-stream crossings, and will ideally include basic survey information. There are two standard methods for completing a culvert inventory, including road- and stream-based approaches. Departments of transportation typically use road-based inventories, while stream-based inventories are usually performed by resource agencies.

4.1.1.1 Road-Based Inventory

A road-based inventory follows a particular road system to identify and evaluate all road stream crossings. This type of inventory is useful to managers requiring knowledge of highway impact on fish passage, and will allow highway dollars to be efficiently spent on the mitigation of fish passage barriers. For example, minor adjustment to culvert inlet or outlet conditions, such as debris jams, rock placement, backwatering, etc., can be made during routine road maintenance. Known barriers can be addressed as part of rehabilitation or reconstruction projects.

Road-based approaches can be very complete, although following a road will invariably miss a number of barriers that exist on side streams or barriers created by minor roads, man made dams, or diversions (Washington Department of Fish and Wildlife 2000).

4.1.1.2 Stream-Based Inventory

A stream-based inventory follows the entire fish bearing system within a watershed or ownership, noting all constructed obstacles (e.g. dams, culverts, water diversions). Further evaluation of these structures provides an understanding of fish passage barriers in a watershed context. This type of inventory will allow analysis of the amount of stream habitat that can be opened up by repairing/replacing a particular culvert. This knowledge will help ensure that program dollars are spent for maximum ecological benefit. These efforts will require cooperation amongst the agencies that have jurisdiction along a stream corridor.

4.1.2 Inventory Goals

To allow prioritization for replacement, more specific site information will be required. A national inventory process created by the Forest Service was designed to answer two questions (Clarkin et al. 2003):

• Does the crossing provide adequate passage for the species and life-stage of concern?
• What is the approximate cost of replacement?

Such knowledge allows a basic understanding of fish impediments, as well as the requirements/plausibility of replacement. Additional information, such as environmental risk, may also be beneficial to planners attempting to prioritize corrections of roadway-stream treatments. Risk assessments may be coupled with fish passage assessment and inventories, but will require additional time and expense. Methods for determining environmental risk are outlined in Methods for Inventory and Environmental Risk Assessment of Road Drainage Crossings (Flanagan et al. 1998).

4.1.3 Data Collection

An initial survey of the culvert and adjoining stream reach will allow a basic understanding of current crossing conditions. This survey will cover a number of site characteristics including culvert and channel measurements and classification, flow data, and watershed conditions. Specific culvert characteristics of interest may include those listed in Table 4.1 from Coffman 2005. It will be useful to have a standardized survey collection method that incorporates collection of all pertinent parameters.

Basic survey techniques are included in Stream Channel Reference Sites: An Illustrated Guide to Field Technique (Harrelson et al. 1994). Examples of fish passage survey application, including forms, explanations of survey points, and data collection are included in Appendix E of National Inventory and Assessment Procedure (Clarkin et al. 2003). Figure 4.1 depicts some typical longitudinal survey points used in a culvert survey.

Figure 4.1 Longitudinal profile survey points (Clarkin et al. 2003)

4.2 Culvert Assessment

4.2.1 Overview

4.2.1.1 Criteria

Before crossing assessment can begin, it is necessary to have a clearly defined set of assessment criteria. Much like culvert design criteria, assessment criteria show regional variability, and generally consider the following elements in order to determine fish passability:

• Depth of flow
• Flow velocity
• Drop heights
• Pool depths
• Culvert length
• Substrate

4.2.1.2 Development

Development of procedures and criteria for culvert assessment should be done by a group of knowledgeable individuals, recognizing program/project goals. Properly designed culvert assessment will provide adequate knowledge of a crossing location and ultimately lead to a robust inventory that will aid in crossing prioritization.

Agreements between State DOTs and Resource agencies can greatly expedite the design and assessment procedure, ensuring that the requirements of all parties are met satisfactorily through a common vision. For example, Alaska and Oregon currently have agreements between their respective resource agencies to epedite permit applications for culvert installations. They also have a shared priority of replacement/repair of fish passage barriers (Venner Consulting and Parsons Brinkerhoff 2004).

4.2.2 Criteria

Assessment criteria will vary depending on fish species present as well as the timing and duration of fish movement. Criteria for adult salmon, for example, will be significantly different from that used for juveniles or trout species (e.g. Robison et al. 1999; Washington Department of Fish and Wildlife 2000).

It is recommended that assessment criteria be developed separately from design criteria (Lang et al. 2004). Typically, design criteria are conservative, so as to provide passage for the weakest swimming individual during a range of design flows. Assessment criteria however, seek to determine the degree to which a crossing is a barrier to fish passage. Crossings that would be labeled inadequate by design standards may only provide a partial barrier to fish passage. As a result, criteria for design and assessment are slightly different, and generally not interchangeable.

4.2.2.1 Degree of Barrier

Assessment allows crossings to be grouped into broad categories of adequacy such as "Passable," "Impassable," and "Indeterminate." Category definitions are expounded to clearly place barriers within a matrix. In California, a culvert that can pass all salmonids during the entire migration period earns a "green" classification, while a culvert that does not meet requirements of strongest swimming fish and life stage present over the entire migration period is classified as "red," analogous to traffic signalization (Taylor and Love 2003). Culverts that cannot be placed in these categories remain in the "gray" or "indeterminate" area, where the crossing may present impassable conditions to some species and life stages at some flows. Further analysis is required in order to ascertain the extent of the barrier.

It is likely that initial surveys will show many culverts to be "indeterminate," where adequacy cannot be determined without a detailed hydraulic analysis (Clarkin et al. 2003). Furthermore, a great number of "impassable" crossings typically ensure that "indeterminate" crossings are never properly analyzed (Furniss 2006).

Culverts falling into the "indeterminate" area are likely barriers to some fish species and life stage. The extent of this barrier incorporates further categorization. Table 4.2 shows barrier categories used in California (Taylor and Love 2003). Assessment criteria are used to prioritize culvert crossings for future replacement, and the degree of barrier is one of many factors used to determine the urgency of culvert replacement/retrofit. Most culverts will present a partial or temporal barrier to fish passage, and an understanding of the degree is useful in assessing the impact of a culvert on the surrounding ecosystem, and in determining the need and urgency of culvert replacement (Furniss 2006).

4.2.3 Existing Procedures

In most situations, site survey and inspection alone will not determine barrier status. While drop heights, substrate, inlet contraction, and slope can be examined, hydraulic analysis will likely be required in order to ascertain flow velocities, flow-depth and pool-depths during design conditions. Assessment is therefore broken into a series of "screens" or "filters", using regionally or locally defined criteria.

4.2.3.1 Coarse Filter

A first pass or "coarse filter" can be used to determine the transparency of the crossing to fish in the natural reach. The basis of this analysis is the presumption that crossings successfully replicating the surrounding natural stream channel conditions will exhibit similar hydraulic conditions, allowing passage for all fish at the flows at which they would be traveling in the natural stream reach. A passable culvert will match natural stream reach characteristics including width, substrate and slope. The coarse filter may also be used to quickly identify obvious barriers such as excessive perching or extreme slope.

4.2.3.2 Regional Screen

If a culvert cannot be clearly categorized as adequate or inadequate using a coarse filter, a subset of regionally defined criteria is used to further clarify culvert adequacy. At this level of analysis, specific fish species criteria are examined to understand culvert impact on the local biota.

4.2.3.3 Examples of Regional Screen Criteria

California's Salmonid Stream Habitat Restoration Manual contains a culvert categorization scheme covering adult and juvenile anadromous salmonids (Taylor and Love 2003). This method combines coarse filter and regionally defined criteria. A flow-chart model (Figure 4.2) helps surveyors place culvert passability into one of three categories: green, gray, and red.

Figure 4.2 Example of a coarse filter and regional screen in customary units ( from Clarkin et al. 2003)
(Green-Gray-Red screen developed for California's anadromous adult and juvenile salmonids)

Figure 4.3 Alaskan fish-passage evaluation criteria, United States Forest Service Region 10, customary units (Flanders and Cariello 2000)

• Green: Condition assumed adequate for passage of all salmonids and lifestages during the entire period of migration.
• Gray: Conditions may not be adequate for all salmonid species or life stages presumed present. Additional analyses are required to determine extent of the barrier for each species and lifestage.
• Red: Conditions fail to meet passage criteria over the entire range of migration flows for even the strongest swimming species and lifestage (adults) presumed present.

4.2.3.4 Flow Chart Filters

Flow-chart categorization has the advantage of providing a simple step-by-step method with variables that are easily interchangeable to meet program needs (Clarkin et al. 2003). Although California addresses all culverts and fish in one chart, additional charts could easily be created to address different species and lifestages of concern. The simplicity of this type of analysis may create a propensity for culverts to fall into the "gray" area (Clarkin 2003).

It should also be recognized that other characteristics not covered in the filter may cause culverts to pose potential barriers, and need to be examined. Examples include breaks in slope, inlet and outlet aprons, crushed inlets or damage to the crossing invert (Taylor and Love 2003).

4.2.3.5 Matrix Filters

Alaska and Oregon compile regional criteria and coarse filter information into a set of criteria that depend on installation type and culvert embedment (Robison et al. 1999; Flanders and Cariello 2000). For example, Alaska's filter for juvenile coho, shown in Figure 4.3, provides a matrix of criteria depending on structure type. This added level of scrutiny may ensure that fewer culverts fall into the "indeterminate" area of passability (Clarkin et al. 2003).

4.2.3.6 Hydraulic Analysis

When barrier status of a culvert cannot be determined after a coarse filter or regional screen, a hydraulic analysis, including a field study, mathematical modeling or direct observation should be completed. This may include situations where baffles or weirs are present. The goal of these studies will be to determine if culverts meet the requirements of target fish species and lifestage.

4.2.4 Recommended Template

Most of the existing criteria were developed from studies focusing on one or two target species, or anadromous species such as Pacific salmon (Bunt et al. 1999; Belford and Gould 1989). This resulted in species-specific guidelines tailored to local fish populations, and resulting inventories and criteria are only truly applicable for the region in which they were developed. For national applicability, a general guidance document has been created to aid in the development of regionally specific inventories.

A National Inventory and Assessment Procedurewas produced by the United States Forest Service San Dimas Technology and Development Center (Clarkin et al. 2003). This included a review of current State procedures and a synthesis of techniques into a standardized, and generally applicable method for assessment and prioritization development. Figure 4.4 depicts a flow chart of a culvert assessment technique. Although the specifics of each level of assessment change slightly depending on regional guidance, the general sequence remains the same. In the first stage, conditions within the culvert are compared to conditions in the undisturbed natural channel. If the culvert does not sufficiently maintain natural reach characteristics, a second pass is conducted, in which surveyors analyze the crossing based on regionally defined passage criteria. If the passability is still not determined, a hydraulic analysis of the crossing is employed.

Figure 4.4 Flow chart for culvert assessment (adapted from Clarkin et al. 2003) (Coarse filter determines if the culvert matches stream reach characteristics; if not, regionally defined criteria determine whether the culvert is appropriate for the fish species and life stage of concern; if not, hydraulic modeling software is utilized)

4.3 Prioritization of Road Crossing culvert retrofits/replacements

4.3.1 Overview

4.3.1.1 Objectives

The objective of a prioritization inventory is not necessarily to rank fish barriers in the order that they should be addressed. However, a basic understanding of culvert barrier location and degree, status, and cost estimates will allow culverts to be addressed in an efficient manner. The following section outlines common criteria, and provides a good starting point for the development of a regionally specific prioritization scheme.

4.3.1.2 Funding Source

Funding source may impact the way that project prioritization is best utilized. Highway dollars may be most efficiently spent by performing a road based inventory and assessment to understand how a particular highway impacts fish passage. With an understanding of culvert barriers in context, fish passage can be addressed in conjunction with other road maintenance, ensuring efficient spending of highway dollars. In the case of project based funding, money may be most efficiently spent on culverts with the greatest ecological impact, and a watershed based inventory may ensure efficient spending.

4.3.2 Criteria

Prioritization criteria should be developed by a group of interdisciplinary experts, and regional criteria will likely vary slightly. Some factors to consider at each crossing include: cost of replacement, fish species present and status, amount and quality of habitat blocked, degree of barrier (total, partial, temporal), proximity to other barrier structures and possibility/cost of crossing failure. There may also be unique characteristics to consider, such as barriers that serve to keep out invasive species, and existing barrier structure that have created habitat upstream. The benefit of replacement will need to be weighed against the possible consequences of no action. An explanation of possible considerations follows.

4.3.2.1 Cost

Cost of replacement/retrofit includes:

• Diversion
• Traffic control, including potential detours
• Design
• Installation
• Maintenance

Crossings may be less costly to replace as roadwork or maintenance occurs near the crossing site, or when other crossings are repaired in the same area.

4.3.2.2 Ecological Significance

The ecological significance of a crossing will include consideration of species present, as well as amount and quality of habitat blocked. A culvert that is blocking an endangered species will require more attention than a comparable barrier that impedes the passage of non-listed species. On the other hand, a barrier culvert might be acting to keep invasive species or diseases from reaching undisturbed native populations.

It is important to understand potential costs and benefits before removing a barrier to connectivity, as impassable crossings may occasionally provide an ecological function that outweighs the benefits of replacement. For example, culverts in a vertically unstable channel may provide elevational control by creating a rigid boundary past which channel incision cannot progress. Removal of a grade control culvert could allow channel incision to progress upstream, possibly affecting fish passage at the structure and habitat quality throughout the reach (Castro 2003). Also, a culvert that is impassable to an invasive species protects upstream species from predation.

4.3.2.3 Habitat Blocked

The quantity of habitat blocked by a particular crossing will be combined with considerations of habitat value to understand the benefits of a potential replacement or retrofit. A culvert blocking access to critical spawning habitat, for example, may require urgent consideration.

The degree of barrier (partial, temporal, total) will also determine the urgency of replacement. All other considerations being equal, a culvert that poses a complete barrier will require more urgent attention than a culvert providing a partial or temporal barrier.

In a situation where anadromous fish spawning access is a concern, a culvert replacement opening 10 km (6.2 mi) of high quality spawning habitat will be made ineffective by a single barrier culvert downstream. In general, it is recommended that culvert replacement progress from downstream to upstream, although in some situations benefits may still be significant for resident fish populations.

A series of partial barriers may combine to effectively block fish from reaching their final destination. Although a culvert that presents a short duration barrier during fish migration may seem like a small problem, a series of delays may mean that spawning fish cannot reach their destination. Regional experience must be used to determine acceptable delay. When prioritizing retrofit and replacement projects, it will be important to establish a crossings context within the watershed.

4.3.2.4 Risk/Significance of Failure

A culvert that is in disrepair or that is severely undersized may have large ecological or hydraulic significance associated with failure. For example, a culvert in Oregon became plugged with debris, causing water to wash out fill, run parallel to the road, and eventually scour out an entire valley wall as it found its own path to the river below (Furniss 2006). Figure 4.5, from Furniss et al, depicts this type of failure, know as diversion (1997).

Figure 4.5 The erosional consequences of diverting stream flow onto non-stream slopes (Furniss et al. 1997)
(Often landslides or debris flows can be triggered by the loading of non-stream slopes with excess water and undermining of slope support by gully erosion)

An understanding of environmental risk requires additional analysis. Methods for Inventory and Environmental Risk Assessment of Road Drainage Crossings (Flanagan et al. 1998), provides a discussion of the potential environmental impacts of culvert failure. This includes a review of past assessment procedures including recommended assessment procedures.

The economic cost of doing nothing can be calculated by many methods including:

• Some Applications of Flood Frequency and Risk Information in Forest Management (Hansen 1987)
• Evaluation of Uncertainty of Flood Magnitude Estimation on Annual Expected Damage Costs of Hydraulic Structures (Bao et al. 1987)

In addition to monetary expense, culvert failure can have significant impacts on habitat quality, possibly allowing a sediment slug to progress downstream, covering spawning habitat or useable areas with fines or silt. Studies of the response of road stream crossings to large flood events in the Pacific Northwest showed that additional failure mechanisms include debris flow, woody debris lodgment and hydraulic exceedance (Furniss et al. 1998).

4.3.3 Existing Prioritization Procedures

State resource agencies in California, Oregon and Washington have implemented prioritization procedures for addressing culvert replacement and rehabilitation. The following examples may not be directly applicable to State DOTs, whose first obligation is to protect the traveling public.

4.3.3.1 California Department of Fish and Game

California Department of Fish and Game employs a ranking system for determining the priority of road stream crossings. Points are awarded to a crossing based on species diversity, extent of barrier, habitat value, risk of failure and current conditions. For example, priority based on species diversity is broken into the following rankings:

• Endangered Species - 4 points
• Threatened or Candidate - 2 points
• Not listed - 1 point

Barrier status is responsible for up to 5 points, high habitat value can result in up to 10 points, and risk of failure up to 5. The result of this prioritization is not intended to provide a list and order of culverts to be addressed (i.e. 30 points fixed first, 28 points fixed second), but gives a list of severity and spatial distribution of crossings to aid planning decisions (Taylor and Love 2003).

4.3.3.2 Oregon Department of Fish and Wildlife

Oregon uses degree of fish blockage and risk of crossing failure to group culverts into one of five categories of prioritization. This allows a general categorization of crossing from Type 1 culverts - which block passage of coho salmon habitat, or have high risk of catastrophic failure, to Type 5 installations - which are on non-fish bearing streams with moderate to high risk of failure (Robison et al. 1999).

4.3.3.3 Washington Department of Fish and Wildlife

Washington Department of Fish and Wildlife's Fish Passage Barrier and Surface Water Diversion Screening Assessment and Prioritization Manual (2000) outlines a Priority Index (PI) ranking system similar to California. Values are assigned to various factors affecting barrier severity, including potential benefits of replacement. Priority is based on barrier status, production potential, habitat blocked, condition of fish stock, projected project cost, and species-specific values. Information is input into a database where prioritization is calculated and culvert inventories are ranked and stored. Since 1991, Washington has inventoried over 4000 km (2500 mi) of state routes, and opened up 595 km (369 mi) of habitat once blocked by barrier culverts (Wilder et al. 2004).

4.3.4 Recommended Template

The Forest Service's National Inventory and Assessment Procedure (Clarkin et al, 2003) has an in-depth discussion of culvert assessment, inventory and prioritization that provides adequate guidance for the development of regional criteria, and this is the recommended reading for those wishing to develop a regional fish passage inventory or culvert assessment procedure. A blank template from Clarkin et al, has been included (Figure 4.6) that allows regional criteria to form a simple coarse filter and regional screen (based on Taylor and Love 2003) (2003).

Figure 4.6 Fill in the blank regional screen based on the California model (Clarkin et al. 2003)