THE GEOMORPHIC AND ECOLOGICAL INFLUENCE OF LARGE WOODY DEBRIS IN

STREAMS AND RIVERS

by

Neil S. Lassettre (nlass@uclink4.berkeley.edu) Department of Landscape Architecture and Environmental Planning

University of California Berkeley, CA 94720 USA

Richard R. Harris (rrharris@nature.berkeley.edu)

Department of Environmental Science, Policy, and Management University of California

Berkeley, CA 94720-3114

March 2001 Abstract. Large pieces of wood, such as logs, stumps, and large branches are an important ecological component of streams flowing through coast redwood (Sequoia sempervirens) forests of Northern California. The distribution and abundance, geomorphic role, and ecological role of large woody debris (LWD) vary through the channel network. This review summarizes pertinent literature and describes how the physical and ecological characteristics of wood change from small to large streams. Streamside and in-channel management practices influence LWD recruitment and transport processes, and significantly alter its functions. Management decisions can be better defined through a process-based classification that describes the influence of wood through the channel network.

Introduction Logs, stumps, and branches that enter and are transported by rivers and streams are important influences on channel morphology and aquatic ecology. Large woody debris (LWD), generally defined as wood ≥ 10 cm diameter and ≥ 1 m length, obstructs streamflow, stores and distributes sediment, and creates channel features, such as pools, riffles, and waterfalls. Wood intercepts organic matter traveling downstream, allowing this material to be processed by instream organisms. Macroinvertebrates and fish occupy and use pools and riffles as habitat, and sediment deposition provides sites for riparian forest regeneration.

The ecological importance of wood is not limited to the instream environment. On the forest floor, woody debris is both a nutrient source and a habitat element for plants, insects, and vertebrates. In estuaries and near shore ocean environments, LWD provides nursery habitat, protection, and a nutrient source for various organisms. Reviews by Harmon et al. (1986), Sedell et al. (1988), and Triska and Cromack (1980) thoroughly treat these topics.

This review focuses on the instream geomorphic and ecological roles of LWD, highlighting effects occurring at the reach level and basin-wide along the channel network from headwaters to large rivers. When possible, the examples are drawn from redwood (Sequoia sempervirens) ecosystems. However, since such examples do not exist for every situation, studies from other systems are used to demonstrate basic concepts, while recognizing possible local differences.

This literature survey is divided into six general subject areas: 1) characteristics, distribution, and transport of LWD, 2) LWD and channel morphology; 3) LWD and stream ecology; 4) management impacts on LWD; 5) synthesis and a proposal for stream classification and; 6) management recommendations. The first section examines regional LWD loading, general trends of log size through the stream network, and the influences and mechanisms of LWD transport. Section two discusses the influence of LWD on sediment storage, channel dimension and stability, and basin-wide effects of wood on pool formation. Section three explores how wood affects nutrient retention, the distribution of macroinvertebrates and salmonids, and the presence of riparian forest patches. The fourth section looks at how timber harvest, flood control, and road maintenance affect the distribution and abundance of instream LWD. The fifth section includes a synthesis of the morphological and ecological effects occurring throughout the channel network. This synthesis results in a proposed designation of LWD influence zones based on channel size and gradient. The last section provides a discussion on how timber harvest and road maintenance may be modified to accommodate LWD input and transport.

In the course of preparing this review, all available literature concerning LWD in stream systems, regardless of geographic location, was consulted. A listing and summary of interesting results from all studies is contained in the appendix (Tables A-1 through A-13).

2

Characteristics, Distribution, and Transport of LWD LWD loading and distribution. Large woody debris loading refers to the weight or volume of wood per square meter of stream channel (kg/m2 or m3/m2). Along the Pacific Coast, streams flowing through the redwood forests of Northern California exhibit the highest levels of instream wood loading (Table 1, Harmon et al. 1986, Bilby and Bisson 1998). Massive redwood logs contribute huge amounts of biomass to stream channels, and in some reaches loading is dominated by a single piece of woody debris (Tally 1980). LWD loading generally decreases as one moves to more northern forests, with southeastern Alaska streams that flow through sitka spruce forests (Picea sitchensis) displaying the lowest abundance of instream wood (Table 1, Bilby and Bisson 1998). When compared with other regions across North America, forests of the Pacific Coast have the greatest levels of terrestrial and aquatic woody debris (Harmon et al. 1986).

Table 1. Loading of LWD (m3/m2) in streams flowing through unmanaged forests along the Pacific Coast from Northern California to Southeastern Alaska. Loading is greatest in coast redwood (Sequoia sempervirens) systems of Northern California and generally decreases as one moves to more northern forests.

LOCATION FOREST TYPE LOADING REFERENCE Alaska • Sitka spruce (Picea sitchensis), western

hemlock (Tsuga heterophylla), • 0.019 m3/m2 Harmon et al. 1986

Alaska • Sitka spruce (Picea sitchensis), western hemlock (Tsuga heterophylla), red alder (Alnus rubra)

• 0.061 m3/m2 Robison and Beschta 1990

British Columbia, Canada

• Sitka spruce (Picea sitchensis), western hemlock (Tsuga heterophylla)

• 0.068 m3/m2 Harmon et al. 1986

Oregon • Douglas fir (Pseudotsuga menziesii) • 0.057 m3/m2 Harmon et al. 1986 Northern California • Douglas fir (Pseudotsuga menziesii) • 0.028 m3/m2 Harmon et al. 1986 Northern California • Coast Redwood (Sequoia sempervirens) • 0.155 m3/m2 Harmon et al. 1986 Northern California • Coast redwood (Sequoia sempervirens) • 0.181 m3/m2 Keller and MacDonald 1983

Within a stream network, the abundance and distribution of LWD is strongly influenced by channel size and dominant input mechanism (Bilby and Ward 1989, Beechie and Sibley 1997). Debris loading is greatest in low order streams (smallest) and generally decreases in the downstream direction (Figure 1, Keller and Swanson 1979, Keller et al. 1985, Lienkaemper and Swanson 1987, Robison and Beschta 1990, Montgomery et al. 1995). Narrow, low order channels lack the streamflow necessary to transport and redistribute most fallen logs (Swanson and Lienkaemper 1978). Wood enters through windthrow, bank erosion, and mass wasting, and unless transported by debris torrents or flood flows, remains randomly distributed in the channel (Table 2, Swanson et al. 1976, Keller and Swanson 1979). Logs may remain stationary for long periods of time (greater than 200 years in some cases), but will eventually leave through transport, or decay from decomposition and physical abrasion (Tally 1980, Harmon et al. 1986, Murphy and Koski 1989).

3

Table 2. The distribution and mechanisms of LWD input and transport from low to high order streams. Jams range from randomly oriented single pieces to large, clumped accumulations within the stream and on bars and floodplains. In small streams, wood enters and moves along the stream primarily during heavy storms, which trigger landslides and provide adequate flow to transport logs. In larger streams and rivers, LWD mobility increases, leaving wood in distinct jams in the water or on bars.

LOW ORDERS INTERMEDIATE ORDERS HIGH ORDERS DISTRIBUTION • Random

• Single piece • Clumped in jams within

streams • Clumped in jams on bars

and floodplains INPUT MECHANISMS

• Windthrow • Bank erosion • Mass Wasting

• Windthrow • Bank erosion • Fluvial transport

• Bank erosion • Fluvial transport

TRANSPORT MECHANISMS

• Debris flows • Flood flows

• Flotation

• Flotation

Figure 1. Schematic representation of the general trend of LWD characteristics along the channel network. Debris loading and jam frequency is highest in small, low order streams and generally decreases in higher orders. The piece size, jam size, and transport tend to increase along the same gradient. The relationship shows that small channels contain more wood per channel area than large channels, but larger channels have a greater capacity to transport and redistribute LWD pieces.

Intermediate order channels are wide and deep enough to move and redistribute in-channel wood (Swanson and Lienkaemper 1978, Bilby and Ward 1989). The main input mechanisms are windthrow, bank erosion, and fluvial transport from upstream reaches (Table 2, Keller and Swanson 1979). Wood accumulates within the channel or on meander bends in irregularly spaced but distinct jams that have profound influence on stream morphology and ecology (Fetherston et al. 1995, Abbe and Montgomery 1996, Hogan et al. 1998). Jam frequency (number of jams per meter) decreases as channels become larger and are more able to transport wood, while at the same time the size of LWD jams increases (Keller and Tally 1979, Bisson et al. 1987). The trend of increasing LWD jam size down the channel network runs counter to the trend of decreasing LWD loading, however, both are a consequence of the increasing capacity of streams to transport wood (Figure 1).

In high order streams, most wood enters through bank undercutting and fluvial transport. The wood deposits on gravel bars or terraces along the river margin (Table 2, Keller and

LOADING JAM FREQUENCY

High

Low

DISTANCE FROM HEADWATERS

PIECE SIZE JAM SIZE TRANSPORT

High

Low

4

Swanson 1979, Bisson et al. 1987). Wood collecting on bars or islands is frequently out of contact with the low flow channel and may have a limited effect on channel morphology (Keller and Swanson 1979). Fluvially deposited LWD still provides important ecological services for salmonids, such as escape cover, and for riparian forest development (Harmon et al. 1986, Fetherston et al. 1995). In low gradient rivers and streams, channel width and sinuosity are main factors controlling the abundance and distribution of LWD accumulations (Nakamura and Swanson 1994). Wide, unconstrained reaches bordered by floodplains and terraces possess abundant storage and depositional sites for transported logs. Reaches flowing through wide valleys develop secondary channels at the base of terraces and along valley walls that trap wood mobilized during large floods. The mouths of secondary channels may be significant LWD storage sites (Swanson and Lienkaemper 1979, Nakamura and Swanson 1994). Constrained reaches have less extensive floodplain and valley regions and are thus less able to accumulate LWD. Another factor contributing to greater abundance of LWD in sinuous channels is riparian forest development. Riparian forests develop on floodplains and terraces, providing a direct source of wood to stream channels (Lienkaemper and Swanson 1979, Nakamura and Swanson 1994). Confined channels also have a greater capacity to transport wood downstream during high flows (Keller and Swanson 1979).

LWD size and volume. As LWD loading decreases down the channel network, the average size (length and diameter) and volume of instream wood increases (Figure 1, Bilby and Ward 1989, 1991). Channel width is a dominant influence on measures of LWD loading, as it is directly related to channel area (Beechie and Sibley 1997). Low order streams have small channel area and a low capacity to transport in-channel wood, leading to high debris loads. Channel width also controls the average size of wood in streams (Bilby and Ward 1989, 1991, Robison and Beschta 1990). Stable pieces of wood remain stationary during normal to high flows. As channels become wider and deeper, the average size of a stable piece of wood increases. Pieces shorter than bankfull width and with a diameter less than bankfull depth are more likely to be transported out of a reach by streamflow (Bilby 1984, Braudrick et al. 1997). In progressively larger channels, a larger proportion of small pieces is removed, leading to a greater average size of in-channel LWD (Bisson et al. 1987, Bilby and Ward 1989).

Influences on LWD transport. The stability of LWD depends on the physical characteristics of the piece and piece orientation within the channel. Log length and diameter are major controls on the movement of wood in rivers and streams (Swanson et al. 1976, Bilby 1984, Nakamura and Swanson 1994). Most transported pieces are shorter than bankfull width, indicating that length may be a rough estimate of wood susceptibility to transport (Bilby 1984, Lienkaemper and Swanson 1987, Nakamura and Swanson 1994). Shorter pieces move more easily than longer pieces, as they encounter fewer instream obstructions and have less contact with bank regions, leaving fewer opportunities for pieces to deposit and accumulate (Bilby 1984). In low to intermediate order streams (3 m to 25 m bankfull width), Lienkaemper and Swanson (1987) found that all transported pieces travelling more than 10 m were shorter than bankfull width. Nakamura and Swanson (1994) also observed that most transported pieces were shorter than bankfull width, while 20% of un-transported pieces were longer than bankfull width. Rootwads increase the stability of logs by increasing the surface area available for

5

snagging on instream obstructions and potentially increasing log diameter to greater than the average bankfull depth (Sedell et al. 1988).

Log position influences the degree to which a piece is exposed to streamflow and potentially transported. Important positional factors contributing to the stability of in-channel wood are the percent of the piece anchored to the bank, the proportion of the piece in the water, and the angle of orientation to flow (Bryant 1983). Burial of one end in the streambed or along the banks reduces piece exposure to streamflow and significantly increases LWD stability. Pieces buried at both ends are extremely stable and may remain in place for decades (Bilby 1984). Increased stability also results when LWD only partially resides in the stream, with one end either buried or simply resting above the bankfull channel.

Within the channel, pieces oriented increasingly perpendicular to flow are transported more easily than pieces oriented in a downstream direction (Bryant 1983). Bryant (1983) observed a southeastern Alaska stream and developed criteria that identified the most stable pieces of wood as those with 70% of their mass anchored to the stream bank, 15% of their mass touching the water, both ends buried, and oriented 30 degrees to flow. However, precise relationships between physical characteristics, piece orientation, and wood transport are yet to be established. Braudrick and Grant (2000) performed flume experiments with pieces shorter than bankfull width and found that orientation to flow, presence of a rootwad, log density, and log diameter were the most influential factors in LWD transport. Braudrick et al. (1997) simulated wood movement through a system, using flume experiments, and observed three distinct transport regimes based on the degree of piece congestion. The transport regime depended on the rate of log input compared to stream discharge. The results showed that low order channels with high rates of wood input and relatively low discharge have a congested transport regime, while high order channels with low wood input rates and greater discharge are uncongested.

LWD transport mechanisms. The mechanisms of wood transport vary down the stream network with respect to channel size (Table 2, Keller and Swanson 1979). Small, steep, low order tributaries primarily move wood through debris torrents triggered during heavy rainfall and flood flows (Swanson et al. 1976, Keller and Swanson 1979, Nakamura and Swanson 1993). Debris flows usually originate in 1st and 2nd order channels with bank slopes that exceed 50%; torrents are rare in intermediate order streams with moderately sloping banks (Swanson et al. 1976, Swanson and Lienkaemper 1978). Torrents bring material from hillslopes into stream channels, transporting logs and sediment short distances to form large accumulations, or transport wood and sediment over longer distances to be deposited in lower gradient reaches (Keller and Swanson 1979). In some cases, debris flows scour existing in-channel debris and sediment, leaving a bedrock channel that is devoid of complexity (Swanson et al. 1976, Swanson and Lienkaemper 1978). In intermediate to high order channels, flotation is the main transport mechanism. Flotation of logs occurs infrequently in small channels, mainly during unusually high flows that also trigger debris torrents (Swanson et al. 1976, Singer and Swanson 1983). The transport of wood during elevated flows is most common in larger streams, and is an important LWD recruitment mechanism in reaches that have relatively low input of trees from the adjacent riparian forest (Keller and Swanson 1979).

6

LWD and Channel Morphology Sediment storage. Logs and stumps deflect streamflow, creating low energy environments that encourage the deposition of sediment upstream of the obstruction (Lisle 1986b). Anchored or attached pieces can protect and strengthen streambanks, which serves to reduce rates of bank erosion and sediment delivery into the channel (Keller and Swanson 1979). LWD obstructions create sediment storage sites that buffer the movement of bedload sediment through a system. In low order streams, in-channel LWD produces open storage sites that collect and store sediment during high flows (Keller et al. 1985). The storage capacity of these LWD-created sites can be quite large. Tally (1980) estimated that wood stored 200 years worth of bedload in undisturbed reaches flowing through Redwood National Park, with space for another 100 years worth of sediment yield. In the Oregon Coast Ranges, Swanson et al. (1976) estimated that the sediment yield of forested streams was less than 10% of the material in storage, and found that in one 100 m section of stream, wood trapped and stored 230m3 of sediment. The storage capacity of LWD has also been observed in Idaho batholith streams, where logs accounted for 34% of channel obstructions and 50% of sediment storage (Megahan 1982).

The displacement or removal of log steps significantly increases sediment transport out of low order stream reaches. Removal of wood from streams reduces the amount of available storage sites and decreases streambed roughness, leading to higher rates of sediment transport. Winter high flows transported 5000m3 of sediment out of a 250 m stream reach in western Washington after Beschta (1979) experimentally removed all pieces of in-channel wood. Bilby (1981) removed wood from a New Hampshire stream and saw a 500% increase in sediment export over the next year. In steep mountain reaches, log steps are a natural influence on hydraulic geometry and play an important role in channel slope adjustment (Swanson et al. 1976, Keller and Swanson 1979, Tally 1980, Heede 1985a, b). Log steps removed from an Arizona stream were eventually replaced by gravel bars, which assumed control over the drop in elevation (Heede 1985a,b). Increased bedload movement of coarse sediment was required to offset the removal of naturally occurring log steps.

The relative contribution of LWD obstructions to sediment storage decreases down the stream network (Figure 2, Table 3). In channels <7 m wide, Bilby and Ward (1989) found that 40% of debris pieces were associated with sediment accumulations. An examination of progressively wider streams (with decreasing gradient) revealed that <30% of debris pieces in channels between 7 m to 10 m, and <20% of debris pieces in channels >10 m were associated with sediment accumulations. Thus, in high gradient reaches, LWD obstructions have a greater role in bedload storage and in controlling the release of sediment (Nakamura and Swanson 1993). Logs are the primary structural component in the formation of step pools that dissipate stream energy, forming low energy depositional sites just upstream of the created waterfall and on the margins of the resulting plunge pool (Heede 1972, Bilby and Bisson 1998).

7

Table 3. The predominant storage sites and pool types associated with LWD from low to high order streams. Step pools form in small channels and accumulate sediment just upstream of the obstruction. In wider streams, accumulations obstruct flow and create low energy depositional environments downstream. Large rivers accumulate sediment in hydraulically formed floodplains and gravel bars.

LOW ORDERS INTERMEDIATE ORDERS HIGH ORDERS STORAGE SITES • Upstream of log steps • Downstream of single

pieces and log steps • Floodplains and gravel

bars POOL TYPES • Step pools • LWD formed lateral scour • Hydraulically formed

lateral scour

Narrow, low order streams often lack significant floodplains and gravel bars, which are important sediment storage sites for intermediate to high orders. In wider streams and large rivers, obstructions create zones of locally high shear stress and a low energy depositional site just downstream (Table 3, Lisle 1986b, Fetherston et al. 1995, Abbe and Montgomery 1996). These channel margin gravel bars and mid-channel islands accumulate additional LWD and sediment, allowing surfaces to build in elevation. The surfaces are sites for riparian regeneration and forest development, further stabilizing the accumulated sediment (Fetherston et al. 1995, Abbe and Montgomery 1996).

In low gradient reaches, wood has a limited effect over sediment accumulation and transport, mainly providing temporary storage (Figure 2, Table 3, Nakamura and Swanson 1993). Large woody debris may actually limit sediment storage by influencing and retarding floodplain and bar development (Smith et al. 1993b). LWD deflects the channel thalweg, resulting in irregular wood controlled lateral scour and increased local sediment export (Smith et al. 1993b, Nakamura and Swanson 1993). After removing log obstructions from a southeastern Alaska stream, Smith et al. (1993b) observed that streamflow initially mobilized sediment causing an increase in net sediment yield. After the channel re-adjusted, the loss of turbulence and increased resistance associated with bar formation resulted in a more regular sequence of bars that stored greater amounts of sediment.

8

Figure 2. Schematic representation of the general trend of the effects of LWD on channel morphology along the stream network. The sediment storage capacity, influence on channel gradient, and influence on pool formation is greatest in small streams and decreases in higher orders. The influence of LWD on channel width increases along the same gradient, but has a limited effect on the width of large rivers.

Channel dimension. The presence of LWD influences local channel gradient and channel width. In low orders, logs form steps that create waterfalls where the stream abruptly drops in elevation. The presence of steps and waterfalls dissipates stream energy, making the stream less able to transport sediment (Heede 1985a, b). Step forming LWD accounts for 30% to 80% of the elevation drop in moderate to steep gradient streams along the coasts of Northern California and Oregon (Swanson et al. 1976, Keller and Swanson 1979, Keller and Tally 1979). As channel gradient decreases, and channels widen, fewer pieces of wood are able to span the channel and form steps, decreasing the occurrence of waterfalls and lessening the control LWD has on elevation (Figure 2, Keller et al. 1985, Bilby and Ward 1989). In very steep reaches, however, wood may rest on top of large boulders and have a limited effect on sediment storage, waterfall formation, and gradient control (Keller and Tally 1979).

LWD increases local channel width in streams not confined by bedrock (Figure 2, Swanson et al. 1976, Bryant 1980, 1983). Sediment deposits upstream of wood accumulations, or downstream of wood accumulations in low gradient reaches, widen and decrease channel depth (Keller and Swanson 1979). Partially spanning fallen trees deflect the thalweg laterally, causing the stream current to diverge and widen the stream channel (Bisson et al. 1987). In moderate to low gradient streams along the Pacific Coast, studies show that the scour around single pieces of wood and large jams widens the channel by 50% to 200% over average bankfull width (Keller et al. 1985, Keller and Swanson 1979, Bryant 1980, 1983, Nakamura and Swanson 1993). In low gradient southeastern Alaska streams, Robison and Beschta (1990) found that the difference in average bankfull width between streams was best explained by the volume of LWD per 100 m of channel in study reaches. Streams with the greatest volume of wood along their length were the widest.

SEDIMENT STORAGE CHANNEL GRADIENT POOL FORMATION

High

Low

DISTANCE FROM HEADWATERS

CHANNEL WIDTH

High

Low

9

In low to intermediate orders, large jams of wood have the greatest overall influence on channel morphology, while individual pieces influence the bed between jams (Hogan et al. 1998). LWD jams are initiated by the presence of key pieces that are of sufficient length and width to trap transported logs. Jams persist for many years, and as they age, channel morphology increases in complexity (Hogan et al. 1998). After formation, sediment accumulates upstream of the jam, initially resulting in a less complex, sediment laden channel. As wood ages and decays, the porosity of the jam increases, creating differential scour and diverse channel forms that become more complex as the jam further deteriorates. Old growth forests have relatively low rates of jam formation, thus jams of varying age, and a complex channel morphology. Harvested or disturbed basins have accelerated rates of jam formation, thus jam age is less varied, leading to a less complex channel morphology (Hogan et al. 1998). Local stream gradients upstream of reaches controlled by key-LWD or LWD jams are lower than channel averages, contributing to a stepped profile, and local channel widths upstream of key-LWD of LWD jams can be significantly greater (Nakamura and Swanson 1993).

Influence on channel stability. LWD stabilizes channels by dissipating stream energy, armoring stream banks, and creating sites of local scour and fill (Keller and Swanson 1979). In low order streams, log steps create cascades, which reduce available energy by increasing channel roughness and leave less energy to erode and scour bed and banks (Heede 1985a, b). Steps are a major control on bank stability, and removal or reduction in size and abundance of these features through management or timber harvest can destabilize channels (Beschta 1979, Adenlof and Wohl 1994). After removing log steps from reaches in low to intermediate sized streams, Beschta (1979) and Bilby (1984) observed severe erosion that completely modified the stream channel through scour and fill. In low gradient alluvial reaches, wood and boulder obstructions act to stabilize the location of pools and bars during high flow (Lisle 1986b). The obstructions create backwater eddies, which are low energy depositional sites that reduce local shear stress and stabilize bed material (Smith et al. 1993a, b).

LWD also contributes to channel instability. Logs divert streamflow and force lateral cutting in streams with minimal bedrock influence (Keller and Swanson 1979, Swanson and Lienkaemper 1979). Lateral stream erosion leads to bank undercutting and eventual slumping into the channel. In streams impacted by debris flows, wood and sediment accumulations divert flow against adjacent steep banks and can initiate failure of the affected hillslope (Swanson et al. 1976). Wood carried by flood flows reduces bank cohesion by battering low banks and severely abrading streamside vegetation (Swanson and Lienkaemper 1979). In high orders, large jams commonly induce channel migration and cause major changes in streambed topography (Hickin 1984, Nakamura and Swanson 1993).

Influence on pool formation. In step-pool and plane bedded (steep to moderate gradient) stream reaches, wood obstructions are the primary element in pool formation (Figure 2, Montgomery et al. 1995, Abbe and Montgomery 1996, Beechie and Sibley 1997, Montgomery and Buffington 1997). Step-pool and plane bedded reaches are found in steep to moderate gradients, and studies along the Pacific Coast show that 20% to 80% of pools are formed in association with LWD (Lisle 1986b, Andrus et al. 1988, Carlson et

10

al. 1990, Robison and Beschta 1990, Nakamura and Swanson 1994, Montgomery et al. 1995, Wood-Smith and Buffington 1996, Montgomery and Buffington 1997).

In narrow, steep gradient streams, a large proportion of fallen logs fully span the channel and obstruct streamflow to form cascades and plunge pools (Table 3, Bilby and Ward 1989, 1991, Robison and Beschta 1990, Richmond and Fausch 1995). Other instream obstructions, such as boulders, are active in the formation of cascades and plunge pools, but wood exerts the greatest control over pool creation (Keller and Swanson 1979).

In wider, low gradient streams, fewer logs are able to span the channel, so most pieces extend partway across and are oriented diagonal to flow (Gregory et al. 1993, Richmond and Fausch 1995). Scour pools form adjacent to partially spanning pieces, as streamflow is forced under and around obstructions. Dominant pool types shift from step-pools to scour pools down the channel network from step-pool to plane bedded reaches (Table 3, Bilby and Ward 1989, 1991, Robison and Beschta 1990, Richmond and Fausch 1995). Bilby and Ward (1989) surveyed western Washington streams and found that plunge pools were the most common (40%) pool type in streams <7 m bankfull width, while scour pools were the most common (60%) pool types in stream >10 m bankfull width. Bilby and Ward (1991) witnessed the same general shift in pool types among both logged and undisturbed basins.

Further down the channel network, in pool-riffle reaches, the interaction between streamflow and sediment transport is the most important element in pool formation (Leopold et al. 1964, Montgomery and Buffington 1997). The presence of wood has a lesser effect on pool formation and the frequency of LWD associated pools generally decreases with increasing channel width (Bilby and Ward 1991). In stream gradients ranging from 0.01 to 0.05, Montgomery et al. (1995) and Beechie and Sibley (1997) found a strong relationship between the number of pools and LWD loading, while in stream gradients <0.01 to <0.02 the relationship was significantly weaker. Smith et al. (1993a, b) removed wood from a low gradient southeastern Alaska stream and found that pool spacing was similar before and after. The results indicate that in low gradient rivers and streams, hydraulic factors rather than channel obstructions exert a greater control over the formation of pools (Swanson and Lienkaemper 1978, Montgomery et al. 1995, Beechie and Sibley 1997). In very large rivers, most wood deposits in the active channel on bars, islands, or in secondary channels, out of contact with the low flow water surface (Piegay et al. 1999). In these cases, pool formation in the main channel may be independent of LWD presence, but wood along shallow channel margins or in small secondary channels does create pools that are important refuge for salmonids (Bisson et al. 1987).

LWD and Stream Ecology Nutrient dynamics. Allochthonous material from the riparian forest is a primary energy source for undisturbed rivers and streams (Gregory et al. 1991). Wood obstructions potentially store large amounts of externally derived organic matter (Figure 3). In Rocky Mountain streams of various successional stages, Trotter (1990) found that reaches with LWD stored twice as much organic matter as reaches where wood was experimentally removed. Bilby and Likens (1980) removed LWD from a low order stream in New

11

Hampshire and observed significant increases in the export of dissolved organic carbon (18% increase), fine particulate organic matter (<1mm; 638% increase), and coarse particulate organic matter (>1mm; 138% increase). During high discharges, the same reaches showed a 500% increase in the export of fine particulate organic matter and coarse particulate organic matter (Bilby 1981).

Salmon carcasses are an important source of organic matter for stream systems (Cederholm and Petersen 1985, Cederholm et al. 1989). In western Washington streams, twenty-two species of mammals and birds were found to consume coho salmon (Oncorhynchus kisutch) carcasses (Cederholm et al. 1989). Woody debris plays an important role in making nutrients from dead fish available by preventing the rapid transport of carcasses downstream. Cederholm and Petersen (1985) found a positive relationship between LWD loading and the number of retained coho salmon carcasses. Branches and logs physically obstruct the movement of dead fish, and the distance a carcass drifted downstream in a controlled release experiment was inversely related to the amount of LWD. In a separate experiment, pools were the most common carcass deposition sites and most pools that retained dead fish were formed by LWD (Cederholm et al. 1989).

12

Figure 3. Schematic representation of the general trend of the effects of LWD on stream ecology down channel network. Wood-associated nutrient storage and insect abundance is highest in small, low order streams and generally decreases in higher orders. The importance of LWD for salmonid habitat is equally important throughout the basin, although the role of wood varies. Wood decomposes more quickly in the terrestrial environment, and hence more quickly in higher orders where LWD frequently resides partway or completely out of the channel. The influence of LWD on riparian forest development decreases with increasing channel confinement that reduces active floodplain area.

Accumulations of organic matter behind log debris dams are sites for the uptake of nutrients. The sediment associated with LWD is high in organic matter content and maintains locally elevated respiration rates. Hedin (1990) found that community respiration was three times higher in sediment behind log debris dams than elsewhere on the streambed. As an attachment substrate for microorganisms that uptake nutrients, wood debris is active in nitrogen and phosphorous removal (Aumen et al. 1990).

Wood is a potential nutrient source for stream ecosystems. Logs contribute fine particulate organic matter through macroinvertebrate processing and physical abrasion. In streams with high debris loads, logs are a greater source of fine particulate organic matter than leaves or pine needles (Ward and Aumen 1986). The physical production of fine particulate organic matter is greatest in winter, when logs are shattered by high flow and through collisions with mobilized substrate.

The decomposition of wood into basic nutrient forms occurs very slowly in water (Sedell et al. 1988). Fungi and macroinvertebrate decomposers require an aerobic environment, but waterlogging prevents the deep penetration of oxygen into the wood interior. Pieces partly submerged beneath the water surface show variable rates of decomposition, with sections exposed to air breaking down more rapidly. In small streams that are shallow and narrow, wood is partly or completely submerged only during the high flows of winter and spring. Wood decomposition rates may be much more rapid in the summer, when pieces are exposed to air (Sedell et al. 1988). Intermediate stream orders are wide and deep enough to submerge logs during high and low flow periods, and may have relatively slow wood decomposition rates. In large rivers, wood is deposited on banks and islands frequently out of contact with the water surface, possibly encouraging rapid rates of LWD break down (Figure 3).

NUTRIENT STORAGE INSECT ABUNDANCE

High

Low

DISTANCE FROM HEADWATERS

WOOD DECOMPOSITION FOREST DEVELOPMENT

High

Low

SALMONID HABITAT

13

Physical abrasion and macroinvertebrate grazing gradually expose more wood surface area and increase oxygen penetration. As logs decompose, the concentration of essential nutrients, such as nitrogen, increases partly through nitrogen fixation (Sedell et al. 1988). The nitrogen fixation occurring on fallen wood may account for 5% to 10% of the annual nitrogen supplied to streams (Sedell et al. 1988).

Macroinvertebrates. The macroinvertebrate fauna associated with LWD ranges from obligate restriction to purely opportunistic use (Dudley and Anderson 1982). In coastal and Cascades Oregon streams, Dudley and Anderson (1982) found 45 taxa closely associated with wood and over 80 taxa facultatively associated. Obligate groups are mostly xylophagous and rely on wood as a direct nutrient source. Obligates are mainly comprised of borers and gougers, which remove and process log particles. Most obligate groups are found on soft, rotten wood or grooved, textured wood with a large surface area. Major xylophagous species are the wood gouging caddisfly Heteroplectron californicum and elmid beetle Lara avara. Anderson et al. (1978) found caddisflies to be the most conspicuous and diverse wood associated insects, however, the density and abundance of L. avara was more strongly associated with the amount of wood available, irrespective of stream size. Higher densities of xylophagous insects occur on hardwoods than on conifers (Anderson et al. 1978). Hardwoods have a higher nutritional value due to greater microbial activity and nitrogen content (Sedell et al. 1988). Two possible mechanisms for wood exploitation by macroinvertebrates are consumption to obtain digestible carbon and nitrogen from microbial flora to meet energy and nutrient requirements, and cultivation and retention of a gut flora to provide nutrients and aid in digestion of wood fiber (Anderson et al. 1978).

Facultative groups use wood as an attachment substrate for filter feeding and grazing, as refugia during high flows and protection from predators, and as an oviposition site. Wood acts as a stable feeding platform for net spinning caddisflies and may channel flow along surface features to maximize filter-feeding efficiency (Dudley and Anderson 1982, Harmon et al. 1986). Wood is also used in case construction, particularly by Limnephilidae caddisflies (Merritt and Cummins 1996). Smooth, firm wood is a suitable attachment site for filter feeders and macroinvertebrates grazing on biofilm (Dudley and Anderson 1982). In New York, Hax and Golladay (1993) found macroinvertebrate densities on wood substrates to be higher than densities on leaves and other detritus. Densities were correlated with the amount of biofilm, which was highest on logs. Biofilm is a food source for grazing aquatic insects. Collector/gatherers occurred in higher proportions on leaves, while collector/filterers preferred wood attachment sites.

The primary basis of association for most facultative taxa is cover from predators and refuge from abiotic stress (Dudley and Anderson 1982). Grooves and cracks support high densities of macroinvertebrates presumably seeking escape cover, as many times predators dominate the total biomass within these microhabitats. Most groups are found on grooved or textured surfaces (Dudley and Anderson 1982). In Australia, O'Connor (1991) compared different wood surfaces and found that grooved snags supported higher densities of aquatic insects. The results supported a habitat complexity hypothesis predicting an increase in species richness with increasing habitat complexity. Physically complex substrates provide more habitat opportunities for algae, microbes, and

14

macroinvertebrates. Depositional sites upstream of log obstructions may support locally high macroinvertebrate densities as organisms gather to process organic material. Log obstructions also act as refugia during high discharge by creating areas of low water flux and low bedflow (Palmer 1996). These areas retain existing populations and encourage deposit of drifting organisms.

Caddisflies and Diptera commonly pupate in moist or saturated logs in the channel or along the stream margin (Harmon et al. 1986). Limnephilid caddisflies deposit egg masses on damp wood, while hydropsychids prefer submerged branches or overhanging wood.

The sequence of colonizers on a newly fallen piece of wood follows the stage of decay (Sedell et al. 1988). Fresh logs are primarily attachment substrates for algae and microbes, which in turn support populations in the grazer and collector/gatherer functional feeding groups. Early colonizers include chironomid midges and scraping mayflies, such as Cinygma spp. and Ironodes spp. (Merritt and Cummins 1996). As wood decays through decomposition and physical abrasion, populations of xylophagous macroinvertebrates colonize the wood surface. Gougers and borers, such as the caddisfly H. californicum and the elmid beetle L. avara, graze on the soft wood, creating a mottled surface, which increases the surface area available for algae attachment and speeds microbial decomposition. In later stages of decay, chironomid and tipulid (Liposothrix spp.) detrivores continue decomposition into basic nutrient elements. However, the role of aquatic macroinvertebrates in wood decomposition and processing is limited compared to terrestrial insects, with aquatic organisms consuming less than 5% of all available wood (Pereira et al. 1982, Sedell et al. 1988).

The abundance and community structure of wood associated macroinvertebrates changes in relation to stream characteristics (Table 4, Dudley and Anderson 1982). High gradient reaches with coarse substrates and exposed logs support an abundant fauna of xylophilous insects, comprised of borers, gougers, scrapers, and groups colonizing wood surfaces and crevices. In moderate gradient reaches, the accumulation of silt and fine organic matter excludes many gougers, borers, and scrapers, limiting the fauna to collectors and predators. In large rivers, physical abrasion diminishes most populations. Additionally, wood may be deposited on high banks and terraces, out of contact with the water surface and unavailable to aquatic invertebrates (Dudley and Anderson 1982, Piegay et al. 1999).

Table 4. Wood associated insect fauna and role of LWD in salmonid habitat from low to high order streams. Coarse substrates, high debris loads, and exposed logs support an abundance of aquatic insects in high order streams. The accumulation of fine silt and sediment and deposit of wood out of the channel in lower gradients excludes most wood associated insects. The role of wood in salmonid habitat varies with channel width and gradient.

LOW ORDERS INTERMEDIATE ORDERS HIGH ORDERS INSECT FAUNA • Borers

• Gougers • Scrapers • Collectors • Predators

• Collectors • Predators

• Limited, wood out of channel

ROLE IN SALMONID HABITAT

• Pool formation • Cover

• Pool formation • Cover • Spawning

• Cover

15

Salmonids. Along the Pacific Coast, wood is crucial in creating and maintaining the area and complexity of salmonid habitat (Figure 3, Table 4, Bisson et al. 1987). An important role of LWD in forming salmonid habitat is the creation of pools. Pools provide fish with a low energy environment to minimize energy expenditures in running water, provide maximum exposure to drifting food organisms, and are used as escape cover by individuals avoiding predators (Bisson et al. 1987, Bjornn and Reiser 1991). In moderate to high gradient streams, LWD is a primary factor controlling channel morphology, particularly pool formation (Keller and Swanson 1979, Beechie and Sibley 1997). Most pools are formed in association with LWD and wood-formed pools occupy a large proportion of the total stream surface area and total stream volume (Fausch and Northcote 1992, Crispin et al. 1993). Other instream obstructions, such as boulders, interact with flowing water to form pools. However, in addition to altering bed morphology, debris-formed pools provide other habitat values, such as cover and nutrient trapping, not offered by boulder-formed pools. Thus, streams with high volumes of wood typically support higher fish densities. In southeastern Alaska, summer periphyton biomass and the volume of instream debris best modeled the winter density of coho salmon and LWD volume alone best modeled the summer and winter density of Dolly Varden (Murphy et al. 1986).

Large woody debris obstructions also create zones of differential scour and deposit, leading to the creation of gravel bars, which are used by salmonids as spawning habitat. After adding logs to western Oregon streams, House and Boehne (1986) observed a 25-fold increase in gravel bar area. In a separate study, Crispin et al. (1993) observed a 4-fold increase in coho salmon (Oncorhynchus kisutch) spawning activity.

In the early to mid twentieth century, managers manipulated trout populations by installing log structures in order to increase habitat area and wood cover (Tarzwell 1937, Boussu 1954). Recent studies use manipulation to demonstrate the influence of woody debris loading on salmonid populations and fish biomass. Stream reaches with reduced levels of in-channel wood support lower salmonid populations compared to reaches with near natural LWD loading. Despite using selective techniques to remove wood from a southeastern Alaska stream and observing no significant changes in average channel width or depth, Dolloff (1986) found that the abundance of coho salmon and Dolly Varden (Salvelinus malma) was lower in cleared reaches. Similarly, Bryant (1985) found that the densities of coho salmon, steelhead trout (O. mykiss), and Dolly Varden were consistently lower in reaches with reduced LWD loading. Coho fry were particularly sensitive to the presence of wood, decreasing in density as the volume of individual LWD accumulations decreased. After removing LWD, Elliott (1986) noted a decrease in benthos abundance and an elimination of most instream overhead cover. The reduced food supply and increased susceptibility to displacement in high flows contributed to the numerical decline of Dolly Varden. Log structures installed in northern Colorado streams increased pool volume and total cover, which encouraged immigration of fish from nearby unmanipulated reaches, leading to increases in the abundance and biomass of age 2+ brook trout (S. fontinalis), brown trout (S. trutta), and rainbow trout (Gowan and Fausch 1996).

16

An important function of instream wood is providing high flow refuge during winter. Salmonids prefer logs, tree roots, and undercut banks as cover, and streams with abundant cover elements support high winter densities of fish (Heifetz et al. 1986, Tschaplinski and Hartman 1986, Dolloff and Reeves 1990, Bjornn and Reiser 1991). Ideal winter cover sites combine low velocity, shade, and three-dimensional complexity (McMahon and Hartman 1989). Quinn and Peterson (1996) found that overwintering survival of coho was correlated with LWD abundance and volume and habitat complexity at the end of the summer, prior to winter high flows. Cederholm et al. (1997) found little or no change in spring and fall coho populations after log additions, but saw a significant increase in winter populations. The results indicate the importance of LWD in providing winter habitat. In high flows, reaches with an abundance of complex, LWD created pools retain the greatest amount of fish (Harvey 1998).

Salmonids occupy stream positions adjacent to structures that increase habitat complexity (Murphy et al. 1986). In British Columbia, 99% of coho fry and 83% of steelhead parr occupied positions near mid-channel rootwads in drought, normal, and high flow conditions (Shirvell 1990). Coho remained near the shore, while steelhead preferred more distant wood associated sites. An artificial channel experiment showed that coho abundance increased in areas of high overhead cover complexity (McMahon and Hartman 1989). Young of the year fish prefer LWD derived lateral habitats. Moore and Gregory (1988) increased lateral habitat by 2.4-fold through the addition of woody debris and increased the number of age 0 cutthroat trout (O. clarki) by 2-fold. McMahon and Holtby (1992) found that complex woody debris pieces also influence the distribution and abundance of fish in streams and estuaries. Eighty percent of coho smolts occurred within 1 m of debris pieces and 95% occurred within 2 m of debris pieces. The results support the need to retain LWD for smolt production in both habitat types.

Riparian habitat. The age structure of floodplain riparian forests reflects the history of LWD deposition and fluvial disturbance. In moderate to steep gradient streams of the Pacific Northwest, LWD parallel or oblique to flow influences the development of forested floodplains (Fetherston et al. 1995). Downed logs trap alluvium and colluvium in upstream near-bank sites, providing areas for riparian seedling establishment. The logs create low velocity environments where sediment and organic material deposit, speeding soil development and providing nutrients for riparian stands. Woody debris additionally provides nutrients as nurse logs for young trees, and in this capacity also creates an elevated establishment site that minimizes competition between seedlings and other forest floor vegetation. Established forests eventually contribute LWD back to the stream during disturbance events, continuing the process of riparian forest development (Bilby and Bisson 1998).

In exposed channel bars, LWD protects downstream riparian sites by obstructing streamflow and shielding vegetation from high flows, allowing species such as alder to become established (Sedell et al. 1988). The sediment deposition, nutrient deposition, and protection afforded by LWD helps streamside vegetation quickly reach a mature stage and better withstand floods.

17

As channel confinement increases, the influence of LWD on riparian forest development and distribution lessens due to a general decrease in the size of wood-associated sediment deposition sites (Figure 3, Bilby and Ward 1989, Bilby and Bisson 1998). Narrow channels also lack extensive floodplain area and are more prone to debris flows that remove soil and vegetation, further limiting riparian forest development (Fetherston et al. 1995, Bilby and Bisson 1998).

In low to moderate gradient reaches, wood accumulates in large distinct jams within the channel and on the floodplain, providing abundant sites for forest establishment (Bisson et al. 1987, Abbe and Montgomery 1996). In northwest Washington, Abbe and Montgomery (1996) observed a diverse riparian forest structure dotted with anomalous old growth patches. The old growth patches were associated with LWD accumulations located in alluvial terrain characterized by frequent disturbance. The accumulations formed around key member logs, which were originally the largest streamside trees, creating distinct, stable LWD jams. Downstream of these bar area jams (BAJs), three factors facilitate the development of riparian forest: 1) local flow deceleration and decreased basal shear stress, 2) sediment deposition, 3) abundant accumulation of organic matter. Log jams create upstream arcuate gravel bars and downstream central gravel bars, which become sites for forest patch establishment and are stable as long as the wood accumulations exist. The observation of forest patches within the zone of active channel migration suggests that some LWD structures provide long-term refugia for floodplain riparian communities and remain stable despite repeated integration into the active channel.

Management Impacts on LWD Timber harvest. Timber harvest activities in streamside forests can directly affect wood input (Table 5, Swanson and Lienkaemper 1978, Bilby and Bisson 1998). Clearcut logging, often with minimal buffer strips surrounding the stream channel, was a common management practice throughout the Pacific Northwest and Alaska until the early to late 1980s (Dominguez and Cederholm 2000). The harvesting of streamside forests may temporarily reduce or eliminate LWD recruitment to the stream (Bryant 1980). The recovery time for input to return to pre-harvest conditions may be quite long. Fifty years after logging, debris from the current stand of a western Oregon stream contributed only 14% of total LWD volume and only 7% of the wood from the current stand contributed to pool formation (Andrus et al. 1988). The results indicate that some second growth stands must grow at least 50 years before trees contribute LWD in sizes and amounts similar to old growth forests. A decay model calibrated in southeastern Alaska predicted a 70% reduction in wood 90 years after clear-cutting, and that full recovery exceeded 250 years (Murphy and Koski 1989).

Logs derived from second growth forests are smaller in diameter and have less volume than old growth LWD, contributing to lower instream loading in logged streams (Bilby and Ward 1991, Ralph et al. 1994). Second growth wood loads tend to be comprised of deciduous riparian species and small conifers that degrade more easily and have less of an effect on long-term channel morphology (Dominguez and Cederholm 2000).

18

A comparison between unlogged, moderately, and intensively logged catchments found that undisturbed streams contained more logs in the largest size categories (>50 cm diameter) than managed streams (Ralph et al. 1994). Intensively harvested streams displayed reduced average pool depth and area, and a significantly lower proportion of stream area occupied by pool habitat. These characteristics were related to the smaller size of in-channel wood and a modified spatial distribution where a large proportion of logs were located out of the low flow channel or did not interact with stream flow.

Table 5. The effect of certain management practices on the characteristics and abundance of LWD within stream systems. Timber harvest temporarily reduces input or changes the physical characteristics of subsequent inputs. Flood control and road maintenance activities generally result in the removal of in-channel wood.

MANAGEMENT PRACTICE

EFFECT REFERENCES

• Temporary reduction in LWD input Bryant 1980, Andrus 1988, Murphy and Koski 1989

• Second growth input smaller, less rot resistant with less profound effects on physical habitat

Bilby and Ward 1991, Wood-Smith and Buffington 1996, Ralph et al. 1994

• Removal of logging residue simplifies physical habitat by failing to distinguish between naturally occurring habitat-forming logs and leftover material

Swanson et al. 1976, Swanson and Lienkaemper 1978, Beschta 1979, Bryant 1980, Keller and MacDonald 1983, Bilby 1984, Bisson et al. 1987, Bilby and Ward 1989

• Extremely large amounts of logging material reduces intragravel flow, increases biological oxygen demand, reduces space available for invertebrates, and blocks fish migration

Hall and Lantz 1968, Narver 1970, Brown 1974

• Destabilization of hillslopes and increase in debris avalanches

Swanson and Lienkaemper 1978

• Narrow buffer strips (<20 m to 30 m) potentially reduce wood input

McDade et al. 1990, Van Sickle and Gregory 1990

Timber harvest

• Buffer strips adjacent to clearcuts have higher occurrence of windthrow and are depleted of large wood sources rapidly

Reid and Hilton 1998

• Remove wood to decrease channel roughness, increase conveyance, and maintain flood capacity

Marzolf 1978, Young 1991, Gippel et al. 1996

Flood control and road maintenance

• Remove wood and clear jams to keep culverts and bridges free of debris and reduce structural damage during storms

Singer and Swanson 1983, Diehl 1997

Streams flowing through second growth forests have a lower frequency of LWD associated pools and fewer channel spanning logs than old growth streams, leading to a scour pool dominated system (Bilby and Ward 1991). Thus, in low to mid-order streams the percentage of LWD formed waterfalls and the control of wood on gradient is decreased by timber harvest. Old growth logs are larger and retain more bedload sediment and fine organic debris. Fine organic debris influences the physical characteristics of large jams and may contribute to an increased diversity of pool types in old growth streams (Bilby and Ward 1991). Changes in wood loading and abundance significantly alter stream morphology. Wood-Smith and Buffington (1993) showed that pool frequency, pool depth, and local shear stress were significantly different in logged versus unlogged streams.

Near-stream logging influences natural LWD input processes. Depending on the method, harvest activities destabilize hillslopes and increase the likelihood of debris avalanches (Swanson and Lienkaemper 1978). Buffer strips are a common technique to reduce logging effects on forests and streams. Most LWD inputs come from within 20 m to 30

19

m of the stream channel and buffers more narrow than this zone of input potentially reduce the amount of available logs (McDade et al. 1990, Van Sickle and Gregory 1990). Buffer strips adjacent to clearcuts are exposed to higher wind velocities, increasing the occurrence of windthrown logs to the stream channel (Reid and Hilton 1998). Higher rates of windthrow may lead to rapid depletion of available wood from the remaining adjacent forest, increasing short-term LWD input, but decreasing long-term input.

Woody debris clearance. During logging operations, leftover woody material accumulates on hillslopes and within streams. Historically in the Pacific Northwest, salvage logging operations removed instream wood to recover leftover material (Bisson et al. 1987). Many regulations also recommended the removal of log jams, which were perceived by fisheries biologists as impediments to anadromous fish migration. Such activities proceeded without rigorous evaluation to distinguish between natural and logging derived LWD, or between naturally occurring logjams and unnatural barriers to fish migration. Many pieces of wood were unnecessarily removed leading to the simplification of physical habitat (Bilby and Bisson 1998).

Stream cleaning reduces the abundance of large stable, habitat forming logs, leaving smaller pieces that are more easily mobilized in moderate to high flows (Swanson et al. 1976, Swanson and Lienkaemper 1978, Bryant 1980, Keller and MacDonald 1983). These small pieces collect behind remaining large logs and build massive debris dams that impede flow and increase rates of lateral stream erosion. The mobilized LWD also weakens existing stable logjams, contributing to dam failure and the initiation of debris torrents.

In moderate to high gradient streams, logs play an important role in bedload storage (Figure 2), and the removal of LWD eliminates potential storage sites (Beschta 1979, Bilby 1984, Bilby and Ward 1989). The decrease in storage capacity and subsequent release of sediment simplifies physical habitat by filling in the deepest pools, reducing pool area, and smoothing channel gradient (Sullivan et al. 1987, Dominguez and Cederholm 2000). Debris removal affects salmonid populations by decreasing the amount of available hydraulic cover available during winter high flows, and by reducing stream wetted width and perimeter (Dolloff 1986, Elliott 1986). Undisturbed reaches have more available habitat area, contain greater supplies of food, and contain more fish of all sizes than altered reaches. Smaller fish and less available hydraulic cover combine to cause an increase in the displacement of fish in high flows. Streams with reduced levels of LWD, either from instream removal or streamside harvest that decreases wood input are often modified with artificial structures designed to mimic the hydraulic and habitat forming effects of LWD (House and Boehne 1986, Riley and Fausch 1995, Wallace et al. 1995).

Alternatively, an excessive amount of logging material left in the stream may be damaging to fish populations. Fine debris lying on the gravel surface impedes interchange between intragravel flow and surface water, reducing subsurface dissolved oxygen levels (Hall and Lantz 1969, Narver 1970, Brown 1974). Reduced oxygen availability retards the development of salmonid embryos within the gravel. The decomposition of wood increases biological oxygen demand, further reducing available

20

dissolved oxygen (Narver 1970). Tannins and lignin-like substances released by wood decomposition produce yellow and brown pigments that absorb photosynthetically active radiation, possibly reducing periphyton growth. Small pieces of wood and bark occupy interstitial pores, reducing the available living space for stream invertebrates (Narver 1970). Very large human induced accumulations of wood prevent upstream migration of anadromous salmonids (Brown 1974). Much historical management of LWD in logged streams concentrated on the removal of excess debris to allow fish passage (Bilby and Bisson 1998).

Flood control and road maintenance. In systems influenced by human infrastructure, road maintenance and flood control activities affect the abundance of large wood (Table 5). Logs and riparian vegetation increase channel roughness, reduce conveyance, and are commonly removed by managers to maintain flood capacity (Marzolf 1978, Singer and Swanson 1983, Young 1991, Gippel et al. 1996). Wood mobilized during high flows frequently becomes trapped on channel spanning bridges and culverts, leading to road overtopping and eventual structure failure. Managers clear jams to keep structures free of obstructions and reduce damage to river crossings. Such management may completely eliminate LWD or remove only the largest pieces that pose the greatest hazard, but which are most important to habitat formation.

Synthesis and a Proposal for Stream Classification Synthesis. The geomorphic and ecological effects of LWD gradually change downstream through the channel network (Figures 2 and 3). The abundance of wood in a stream or river reflects the balance between wood entering and wood leaving a particular reach (Keller and Swanson 1979). The main controls on the geomorphic effects of LWD are piece size, channel gradient, and channel width. The ecological effect of LWD arises from log characteristics and channel morphology. Logs trap and retain nutrients, create habitat for fish and new riparian growth, and act as a direct nutrient source and attachment site for aquatic insects. This review has thus far concentrated on geomorphic and ecological effects as they occur in varying stream orders. The effects of LWD in stream systems may be better understood when integrated with existing process-based stream classification systems to create a classification based on the influence of instream wood. Such a classification system will aid in the management of instream LWD and help determine the sensitivity of stream channels to changes in LWD characteristics.

Montgomery and Buffington (1997) developed a channel classification system based on geomorphic processes. The system integrates channel processes with the spatial arrangement of specific reach morphologies. In channels composed of alluvial substrates, they identify several reach morphologies that occur in a downstream direction with reducing gradient: cascade, step-pool, plane-bed, and pool-riffle (Figure 4, Table 6). The reach morphologies reflect basin wide trends in sediment transport and storage capacity. Higher gradient reaches have a high transport capacity relative to sediment supply and function as transport zones that deliver sediment to lower gradient reaches. Reach substrate and morphology result from the balance between sediment supply and transport capacity. Other reach types arise when external influences, such as LWD, are present in stream channels and affect sediment input and output processes.

21

In some cases, LWD obstructions force sediment-poor bedrock channels into becoming alluvial channels, owing to the sediment storage capacity of in-channel wood. Montgomery and Buffington (1997) recognize two forced reach morphologies in alluvial channels: forced pool riffle, and forced step pool. The forced morphologies extend beyond the range of free forming analogous channel types into higher and lower gradient sections (Figure 4). Forced pool-riffle reaches span the gradient range for pool-riffle and plane-bed reaches and arise when most pools and bars form in response to LWD obstructions. Forced step-pool reaches arise when most elevation controlling steps are formed by channel spanning logs. The recognition of forced-reach morphologies as distinct types acknowledges the control LWD has on bed morphology.

Table 6. The gradient range and general characteristics of reach morphologies in alluvial channels (Data taken from Bisson and Montgomery 1996 and Montgomery and Buffington 1997).

CASCADE STEP-POOL PLANE-BED POOL RIFFLE GRADIENT • 0.08 to 0.30 • 0.04 to 0.08 • 0.01 to 0.04 • 0.001 to 0.02

BEDMATERIAL • Boulder • Cobble/boulder • Gravel/cobble • Gravel

CONFINEMENT • Confined • Confined • Variable • Unconfined

Figure 4. Generalized long profile of alluvial channels showing spatial arrangement of reach morphologies, including forced step-pool and forced pool-riffle morphologies. Forced morphologies extend beyond the gradient range of free-formed counterparts. Gradient ranges of forced morphologies depicted above are interpreted from Montgomery et al. (1995) and Beechie and Sibley (1997). The classifications are based on geomorphic processes and reflect basin wide trends in sediment transport and storage (Figure adapted from Montgomery and Buffington 1997).

The river continuum concept is an ecological framework that describes adjustments of the biological community in response to physical conditions (Figure 5, Vannote et al. 1980). Energy sources and processes change systematically from headwaters to large rivers. Low order streams tend to be narrow and shaded by streamside vegetation, resulting in low rates of instream autotrophic growth. Headwater communities rely on allochthonous inputs from leaves, pine needles, and pieces of wood as energy sources. In these reaches, the autotrophic production to heterotrophic respiration ratio (P:R) is less than one, and is reflected in algal and macroinvertebrate community structure. Algal growth is limited and the macroinvertebrate community is comprised mainly of shredder and collector

cascade

pool-riffle

step-pool

plane-bed

DISTANCE FROM HEADWATERS

forced pool-riffle

forced step-pool

22

functional feeding groups, which process external inputs. In intermediate order streams, channel widening reduces overhead canopy cover allowing rates of autotrophic production to exceed respiration (P:R >1). Periphyton is the dominant energy source and macroinvertebrates shift to grazers and collectors. Grazers scrape and process algae from solid surfaces, such as boulders and logs. In large rivers, shade is further reduced and channels become wider and deeper, but generally more turbid. The high turbidity and increased depth reduces light penetration and autotrophic production (P:R <1). Dissolved material and fine particles transported from upstream are the dominant energy sources and macroinvertebrate communities are comprised mainly of collectors.

The role of LWD is not explicitly addressed within the river continuum concept. However, the influence of wood on nutrient retention and macroinvertebrate communities is quite important (Bilby and Likens 1980, Wallace et al. 1995). As nutrients travel downstream, they are incorporated by stream organisms. After incorporation, processed nutrients are eventually released and again become available for uptake by stream organisms. The process by which nutrients are used and released, called spiraling, is influenced by the presence of LWD. Fallen logs delay the transport of sediment and organic matter downstream. By physically obstructing particulate and dissolved nutrients, LWD reduces the spiraling length, or distance between uptake and release (Bilby and Likens 1980, Newbold et al. 1982). Without storage behind wood accumulations, allochthonous input is transported downstream without being reduced into smaller particles by physical and biological action. Unobstructed material may be unavailable to organisms, such as macroinvertebrates, in intermediate to high order streams (Vannote et al. 1980, Bisson et al. 1987). Wood obstructions store large amounts of organic matter, the primary energy source in undisturbed streams. Thus, the presence of wood obstructions has an effect on available nutrients and biological community structure. Down the channel network the storage capacity of LWD decreases, but wood may act as growth substrate for periphyton, an attachment site for filter feeders, or as a direct food source, further influencing macroinvertebrate community structure (Dudley and Anderson 1982, Bilby and Ward 1989). In large rivers, where woody debris deposits on banks and terraces largely out of contact with the water surface, the influence on macroinvertebrate community structure and nutrient availability is limited. Unlike Montgomery and Buffington (1997) reach morphologies, the river continuum concept does not recognize ecological stream reaches that are forced by woody debris obstructions. Considering the above relations, stream ecology is most sensitive to the influence of LWD in low to intermediate stream orders.

23

Figure 5. The relationship between channel width, canopy, respiration, and macroinvertebrate community structure. As channels increase in width, the relative shading by riparian canopy decreases, and turbidity, caused by concentration of suspended sediment, increases. In narrow streams, the autotrophic production to heterotrophic respiration ratio is less than 1, as most production comes from allochthonous inputs. As the stream widens and shade decreases, water is clear enough to support periphyton growth and autochthonous production dominates. In large channels, turbidity and depth reduce algal growth, and production is allochthonous input transported from upstream. The relative proportion of macroinvertebrate functional feeding groups changes in response to the dominant energy source. Collectors and shredders process leafy input in small channels, grazers scrape periphyton from substrates in mid-order reaches, and collectors gather transported material in high order rivers (figure adapted from Vannote et al. 1980).

A stream classification proposal. Managers should emphasize restoring and maintaining habitat-forming processes associated with large wood (Beechie and Bolton 1999). Figures 6 and 7 describe a new classification system derived from this review of scientific literature (Figures 2 and 3) and existing process based geomorphic and ecological classifications (Figures 4 and 5). These wood influence zones, based upon position within the channel network, should not be confused with influence zones described by Robison and Beschta (1990) and O'Connor and Ziemer (1989), which are based on lateral channel position. Low to intermediate order reaches make up the single piece/debris flow zone where closely spaced, solitary logs and debris flows have the greatest effect on channel morphology and ecology. Single pieces falling across the channel accumulate sediment and allochthonous material in upstream storage sites, and in general are the major control on morphology and ecology in small streams (Montgomery and Buffington 1997). Wood moves infrequently, but usually in catastrophic bursts triggered by heavy storms. Such events may transport massive amounts of LWD downstream and cause landslides and debris flows that bring wood into the channel (Swanson et al. 1976, Singer and Swanson 1983). In very steep gradients, solitary logs may lie above the channel, resting on large boulders, having minimal effect on channel

CANOPY COVER

High

Low

DISTANCE FROM HEADWATERS

CHANNEL WIDTH TURBIDITY

High

Low

PRODUCTION/RESPIRATION

P/R < 1

P/R > 1

P/R < 1

Shredders

Grazers

Collectors

Predators

Shredders

GrazersCollectors

Predators

Predators

Collectors

24

morphology (Keller and Tally 1979). However, the sediment storage capacity of log obstructions may also force high gradient bedrock channels into an alluvial reach morphology or cascade channels into a forced step-pool morphology (Montgomery et al. 1996, Montgomery and Buffington 1997).

Intermediate order streams comprise the intermediate jam zone (Figure 6). Larger channels have the capacity to redistribute wood in irregularly spaced jams. Wood generally occurs in multiple piece jams and the spacing between accumulations increases downstream. The relative confinement influences the predominant pool type and reach morphology (Figure 7). Confined channels have limited floodplain area, thus wood deposits within the channel, building jams that create a forced step-pool reach morphology. Highly confined channels have a greater capacity to transport wood downstream, and may have low wood loading, minimizing log influence on morphology and ecology (Bilby and Bisson 1998). Moderately confined to unconfined channels have floodplains and gravel bars that act as LWD storage sites. The channel margin location of LWD jams encourages the formation of lateral scour pools within a forced pool-riffle reach morphology. The intermediate jam zone then can be further classified into moderately confined and unconfined subtypes that produce different possible reach morphologies.

Higher order streams and rivers make up the large jam zone (Figure 6). Wide, deep channels transport and deposit wood on floodplains and gravel bars within the active channel, but may be out of contact with the water surface (Figure 7). Jams in this zone tend to be the largest within the channel network, but are widely spaced and have a limited effect on gradient and sediment storage. Most sediment is stored in geomorphic elements, such as floodplains and terraces. Channel margin deposits of wood encourage the formation of lateral scour pools within a forced pool-riffle morphology. Woody debris jams provide salmonids with hydraulic and escape cover during seasonal migrations. Possibly the most important role of LWD in the large jam zone is riparian forest development. Large jams provide sites for vegetation colonization, forest island growth, and forest floodplain development (Fetherston et al. 1995).

Management Recommendations Forest management. Possibly the first step in improving the management of LWD in California stream systems is to recognize the different roles it plays in different parts of the watershed. The stream classification proposed above explicitly does that. It is equally important to understand the mechanisms by which LWD gets into streams, its potential for stability and the limitations on managers for recruiting and maintaining LWD. Timber harvest in and near streamside zones can potentially influence direct inputs of LWD to streams and can also influence indirect inputs through mechanisms such as inner gorge landslides and debris flows. However, timber harvesting in streamside zones probably has more significance on small to intermediate streams. In upper watershed areas, removal of forest from unstable upland sites that periodically contribute large quantities of LWD associated with mass wasting can also have lasting impacts. In view of these issues, the current system of regulating streamside management based on stream size, i.e., larger streams get more protection, may not achieve the desired results.

25

Figure 6. Location of LWD influence zones in relation to general trends occurring along the channel network and in context to geomorphic and ecological classification systems (figures adapted from Vannote et al. 1980 and Montgomery and Buffington 1997).

SEDIMENT STORAGE CHANNEL GRADIENT POOL FORMATION

High

Low

CHANNEL WIDTH

High

Low

NUTRIENT STORAGE INSECT ABUNDANCE

High

Low

High

Low

WOOD DECOMPOSITION FOREST DEVELOPMENT

SALMONID HABITAT

pool-riffle

step-pool

cascade

plane-bed

forced pool-riffle

forced step-pool

SINGLE PIECE/DEBRIS

JAM ZONE INTERMEDIATE

JAM ZONE LARGE JAM

ZONE

DISTANCE FROM HEADWATERS

P/R < 1 P/R < 1

P/R > 1

CHANNEL WIDTH TURBIDITY

PRODUCTION/RESPIRATION

High High

Low Low

CANOPY COVER

26

Figure 7. The general characteristics of LWD influence zones. Single pieces and debris flows dominate in small streams, while progressively larger jams dominate in higher orders. Intermediate jam zones can be further classified according to channel entrenchment, with confined reaches tending toward step-pool morphologies and unconfined reaches tending toward pool-riffle morphologies.

To ensure future supplies of LWD to stream channels, buffer strips serving as reservoirs of wood supply should be wide enough to encompass the zone of LWD input, typically within 20 m to 30 m of the stream channel (Lienkaemper and Swanson 1987, McDade et al. 1990, Van Sickle and Gregory 1990). Some researchers have argued for larger buffers, based on susceptibility of buffer strips next to clear-cuts to blow-down and rapid depletion of available streamside wood (Reid and Hilton 1998). The use of a selectively logged fringe buffer adjacent to the streamside buffer may serve to reduce abnormally high rates of windthrow and preserve natural input rates. Any selective cutting within buffer strips should leave an abundant supply of the largest trees for recruitment (Murphy and Koski 1989, Abbe and Montgomery 1996). Large in this context refers to trees that can produce LWD big enough to interact with flow and influence channel morphology. To be cautious, the very largest trees should always be retained (Fetherston et al. 1995, Abbe and Montgomery 1996). There is no formula for determining the amount of wood required in a given stream so conservative judgment is warranted. Species, diameter, and wood decay rates influence the amount of wood recruitment potentially necessary (Murphy and Koski 1989).

SINGLE PIECE/DEBRIS FLOW ZONE

DISTRIBUTION • Random • Single piece

INPUT MECHANISMS

• Windthrow • Bank erosion • Mass Wasting

TRANSPORT MECHANISMS

• Debris flows • Flood flows

STORAGE SITES • Upstream of log steps

POOL TYPES • Step pools WOOD ASSOCIATED INSECT FAUNA

• Borers • Gougers • Scrapers • Collectors • Predators

ROLE IN SALMONID HABITAT

• Pool formation • Cover

INTERMEDIATE JAM ZONE

CONFINED DISTRIBUTION • Clumped in

jams within streams INPUT MECHANISMS

• Windthrow • Bank erosion • Fluvial

transport TRANSPORT MECHANISMS

• Flotation

STORAGE SITES • Single logs and log jams

POOL TYPES • Step pools WOOD ASSOCIATED INSECT FAUNA

• Collectors • Predators

ROLE IN SALMONID HABITAT

• Pool formation • Cover • Spawning

UNCONFINED DISTRIBUTION • Clumped in

jams within streams INPUT MECHANISMS

• Windthrow • Bank erosion • Fluvial

transport TRANSPORT MECHANISMS

• Flotation

STORAGE SITES • Single logs and log jams

POOL TYPES • LWD formed lateral scour

WOOD ASSOCIATED INSECT FAUNA

• Collectors • Predators

ROLE IN SALMONID HABITAT

• Pool formation • Cover • Spawning

LARGE JAM ZONE DISTRIBUTION • Clumped in

jams on bars and floodplains

INPUT MECHANISMS

• Bank erosion • Fluvial

transport TRANSPORT MECHANISMS

• Flotation

STORAGE SITES • Flood plains and gravel bars

POOL TYPES • Hydraulically formed lateral scour

WOOD ASSOCIATED INSECT FAUNA

• Limited, wood out of channel

ROLE IN SALMONID HABITAT

• Cover

LWD INFLUENCE ZONES

27

Table 7. The possible management implications of preserving LWD input, transport, and presence within the stream channel. MANAGEMENT

PRACTICE IMPLICATION REFERENCES

• Buffer strips should be wider than zone of LWD input McDade et al. 1991, Van Sickle and Gregory 1990

• Fringe buffers can protect streamside buffers from premature wood depletion

Reid and Hilton 1998

• Selective management in buffers should consider future input required based on instream surveys

Bilby and Ward 1989, Murphy and Koski 1989

• Selective management should leave large trees that will be stable and influence channel morphology

Fetherston et al. 1995, Abbe and Montgomery 1996

• Active management of buffer zones can increase recruitment of certain species and sizes of wood

Beechie and Sibley 1997

• Removal of logging debris best dealt with by selective removal

Bryant 1983, Bilby 1984, Gurnell et al. 1995

• Knowledge of habitat conditions, and the size and abundance of LWD required to maintain conditions must be considered when removing instream wood

Bryant 1983, Bilby 1984

Timber harvest

• Characteristics of unmanaged streams should guide re-introduction of wood

Smith et al. 1993a, b, Montgomery et al. 1995, Abbe and Montgomery 1996, Beechie and Sibley 1997, Montgomery and Buffington 1997

• Must gain quantitative understanding of effect of wood on flood heights and how moves through a system

Young 1991, Braudrick et al. 1997, Braudrick and Grant 2000

• Design and modify bridges and culverts to allow for passage of woody debris

Diehl 1997, Flanagan et al. 1998

Flood control and road maintenance

• Develop management that recognizes ecological value and impact of wood on human infrastructure and public safety

Singer and Swanson 1983, Piegay and Landon 1997

Forest managers should seek to increase the recruitment of certain species, primarily conifers which produce the largest and longest lasting LWD. This may involve active management of deciduous riparian zones to promote conifer establishment and growth (Beechie and Sibley 1997). This strategy should be considered in relation to position within the channel network. Small channels (<10 m width) can form pools around smaller pieces of wood (<20 cm), such as alder logs. Large to intermediate channels require greater diameter logs to form pools (>60 cm). Data on variations in the size and amount of woody debris with changing stream size could be used to develop plans for numbers and sizes of trees to be achieved (Bilby and Ward 1989).

LWD clearance. In cases where logging debris is to be removed from a stream, it is best dealt with by selective maintenance to ensure channel stability and ecological function (Gurnell et al. 1995). An adequate amount of leftover LWD is essential to maintain wood related habitat structure (Lisle 1986a). Uncleaned streams flowing through logged or clear-cut reaches may exhibit higher levels of wood debris loading and subsequently have a greater abundance of LWD associated habitats (Lisle 1986a, Lisle and Napolitano 1998). Froehlich (1973) found that some harvest methods increased in-channel LWD abundance by >1000% over natural levels. In southeastern Alaska, Lisle (1986a) found that debris dams were more frequent in clear-cut reaches that were not cleaned after harvest than in undisturbed reaches. The clear-cut reaches showed greater residual pool depth and length, and the remaining LWD stored more bedload sediment than logs in undisturbed reaches. Leftover wood eventually deteriorates through decomposition and physical abrasion and streams eventually decrease in physical complexity if LWD is not replaced through natural inputs (Lisle and Napolitano 1998).

28

Indiscriminate removal of in-channel LWD for any reason can have major influence on channel processes (Beschta 1979, Bilby 1984). Knowledge of habitat conditions and the size and abundance of stable LWD required to create and maintain these conditions are considerations when removing instream wood. Bilby (1984) developed a dichotomous key based on local channel conditions and size of LWD to determine whether to remove or leave logs in the channel. Bryant (1983) used an inventory of piece length, percent of piece in the water, angle of orientation to flow, and location of anchor point to determine piece stability and suggested general removal guidelines based on the age of debris and stream gradient.

LWD placement. In streams with a paucity of LWD, re-introduction or placement of logs as instream structures provides a short-term solution to maintain wood created habitats until natural recruitment processes recover (Sullivan et al. 1987, Gurnell et al. 1995). The characteristics of unmanaged streams, such as the variability in species, size, and spacing of LWD accumulations and the goals of the project should guide the re-introduction of wood (Dominguez and Cederholm 2000). Dominguez and Cederholm (2000) developed a flow chart for determining candidate streams for rehabilitation based on fish habitat needs and characteristics of the stream and surrounding forest. Any logs added should be structured to mimic effects of natural obstructions in streams. Due to its random location, all stable logs may not be active in pool formation (Montgomery et al. 1995). Through careful placement of existing logs, managers may be able to engineer jam configurations that more efficiently enhance pool formation. The sensitivity of stream channels to wood addition varies with channel gradient. Moderate gradient reaches are highly sensitive to the presence of LWD (Sullivan et al. 1987, Montgomery et al. 1995, Abbe and Montgomery 1996, Beechie and Sibley 1997, Montgomery and Buffington 1997). Beechie and Sibley (1997) predict that increases in log abundance will lead to more rapid increase in the number of pools and pool area in moderate slope channels than in lower slope channels. The presence of pools in low gradient reaches may be independent of LWD presence (Smith et al. 1993a, b). The placement of instream structures maintains pools in the short-term, but ignores dynamic ecological and physical processes. Improved management of streamside forests offers the most promise for developing valuable and productive riparian systems (Elmore and Beschta 1989). Effective management of LWD will depend on information relating vegetative and physical characteristics of riparian areas to input of LWD (Bilby and Ward 1989).

Flood control and road maintenance. Flood control programs that encourage the removal of woody debris rarely undergo technical scrutiny, and do not always have a significant effect on flood levels (Williams and Swanson 1989, Young 1991, Dudley et al. 1998). Channel roughness depends on many factors, and the claims of reduced frequency and duration of over bank flooding, used to justify debris removal, are not unequivocally proven in the field (Gippel 1995). In Australia, Young (1991) found that the average amount of wood in lowland rivers seldom had a significant effect on flood levels, and suggested that larger pieces may be rearranged for hydraulic reasons and habitat preservation. Clearance of wood to maintain bridges and culverts is usually undertaken during low flow conditions, yet much wood enters streams and rivers during storms, through debris avalanches, bank erosion, and windthrow, reducing the effectiveness of removal programs (Singer and Swanson 1983). The use of hydraulic

29

models may aid in the planning of debris management programs. However, in re-introducing LWD to stream channels as part of riparian restoration or instream habitat enhancement, managers must understand the geomorphic context of wood function to guide the effective placement for long-term log stability (Braudrick et al. 1997). A quantitative understanding of wood movement is needed to assess stability of re-introduced wood and to prevent unstable logs from becoming a safety hazard.

The alternative is to design bridges and culverts that allow passage of woody debris. Diehl (1997) examined drift damage to bridges across the United States and suggested adequate freeboard, wide spans, solid piers, rounded pier noses, and pier placement out of the path of drift to reduce jam accumulation. Flanagan et al. (1998) suggested methods to determine the potential of culvert crossings to impede downstream movement of logs based on the ratio of culvert width to channel width (w*), degree of upstream widening and stream approach angle to the culvert. Culverts with w* <1 are prone to clogging. Channel widening immediately upstream of culverts encourages ponding and creates eddies that orient wood perpendicular to flow and promote formation of large jams. In streams approaching crossings at a severe angle, wood cannot rotate parallel to flow and is less likely to pass through the culvert. Surveys on the size distribution of instream wood should also be included in studies assessing log passing capacity and considered before construction or modification of structures (Singer and Swanson 1983).

The management of wood in basins that are directly influenced by human infrastructure must balance ecological and safety concerns. Piegay and Landon (1997) proposed a system that recognizes the impact of LWD on public safety and human infrastructure, and the ecological value of wood. The sectored LWD maintenance plan was based on mapping of three areas: 1) woody debris supply areas, 2) areas where log jam formation is most probable, 3) areas of floodplain occupation to determine human vulnerability to flooding. The plan allowed for recruitment of LWD and the formation of jams in areas that posed little safety hazard. Intensive maintenance of instream wood occurred near flood prone areas and upstream of bridges.

Conclusion Pieces of large wood are an important component in the ecology of streams flowing through the forests of Northern California. The physical and ecological roles of logs, stumps, and large branches vary through the channel network. In designing a management plan, one must not only consider the various roles, but also the processes by which LWD enters and is transported downstream. The LWD influence zones described in this paper could be used to define management objectives regarding forest management (LWD input), flood control and road maintenance (LWD transport), and instream rehabilitation projects (LWD loading). More specific management objectives must come from an understanding of habitat requirements and the limiting factors of sensitive species, and the characteristics of individual streams. The uncertain nature of ecological studies requires adequate monitoring, the results of which should be integrated into an adaptive management process that is designed by managers to maintain or improve ecological processes.

30

APP

EN

DIX

Tabl

e A

-1.

The

char

acte

ristic

s an

d di

strib

utio

n of

LW

D in

var

ious

geo

grap

hic

loca

tions

and

cha

nnel

net

wor

k lo

catio

ns (a

rrang

ed in

chr

onol

ogic

al o

rder

).

LOC

ATIO

N

CH

ANN

EL N

ETW

OR

K P

OSI

TIO

N

RES

ULT

S R

EFER

ENC

ES

Wes

tern

Ore

gon

• 0.

10 to

0.3

0 ch

anne

l gra

dien

ts

• LW

D in

sm

all s

tream

s ra

ndom

ly d

istri

bute

d •

LWD

eas

ily tr

ansp

orte

d in

larg

er ri

vers

lead

ing

to s

ize

sorti

ng a

nd a

ccum

ulat

ions

in

dist

inct

jam

s

Swan

son

et a

l. 19

76

Wes

tern

Ore

gon

• 0.

02 to

0.5

0 ch

anne

l gra

dien

ts

• D

ebris

load

ing

high

est i

n sm

all,

stee

p st

ream

s, d

ecre

asin

g do

wns

tream

•

In 1

st a

nd 2

nd o

rder

stre

ams

LWD

rand

omly

loca

ted

beca

use

stre

ams

too

smal

l to

redi

strib

ute

• In

3rd

to 5

th o

rder

stre

ams

flow

s la

rge

enou

gh to

redi

strib

ute

debr

is to

form

dis

tinct

ac

cum

ulat

ions

that

dire

ctly

affe

ct c

hann

el w

idth

•

In la

rge

river

s LW

D th

row

n on

isla

nds

or o

n ba

nks,

hav

ing

little

influ

ence

on

chan

nel,

exce

pt d

urin

g hi

gh fl

ows

Swan

son

and

Lien

kaem

per

1978

Indi

ana

Nor

th C

arol

ina

Ore

gon

• 0.

001

to 0

.40

chan

nel g

radi

ents

•

Load

ing

(kg/

m2 ) d

ecre

ased

with

incr

easi

ng c

hann

el w

idth

, wat

ersh

ed a

rea,

stre

am

orde

r, an

d de

crea

sing

cha

nnel

gra

dien

t Ke

ller a

nd S

wan

son

1979

Nor

thw

est

Cal

iforn

ia

• 0.

01 to

0.4

0 ch

anne

l gra

dien

ts

• 1.

0 km

2 to 2

7 km

2 dra

inag

e ar

eas

• 2nd

thro

ugh

4th o

rder

stre

ams

• R

edw

ood

debr

is u

sual

ly d

omin

ates

tota

l loa

ding

•

Load

ing

(m3 /m

2 ) dec

reas

ed a

s dr

aina

ge a

rea

and

wid

th in

crea

sed

• D

ebris

acc

umul

atio

ns in

low

er re

ache

s la

rger

, mor

e co

mpl

ex, a

nd s

pace

d fu

rther

ap

art t

han

in u

pper

reac

hes

Kelle

r and

Tal

ly 1

979,

Tal

ly

1980

, Kel

ler a

nd M

acD

onal

d 19

83, K

elle

r et a

l. 19

85

New

Eng

land

•

1.5

m to

7m

wid

e ch

anne

ls

• Fr

eque

ncy

of d

ebris

dam

s de

crea

sed

from

1st o

rder

(20

to 4

0 da

ms

per 1

00m

) to

2nd

orde

r (10

to 1

5 da

ms

per 1

00m

) to

3rd o

rder

(1to

6 d

ams

per 1

00m

) stre

ams

Like

ns a

nd B

ilby

1982

Wes

tern

Ore

gon

• 0.

03 to

0.3

7 ch

anne

l gra

dien

ts

• 3.

5 m

to 2

4 m

ban

kful

l wid

ths

• 0.

1 km

2 to 6

1 km

2 dra

inag

e ar

eas

• 1st

thro

ugh

5th o

rder

stre

ams

• Am

ount

of L

WD

gen

eral

ly d

ecre

ased

from

sm

all t

o la

rge

chan

nels

Li

enka

empe

r and

Sw

anso

n 19

87

Wes

tern

W

ashi

ngto

n •

0.13

cha

nnel

gra

dien

t (<7

m

chan

nel w

idth

) •

0.08

cha

nnel

gra

dien

t (7m

to 1

0m

chan

nel w

idth

) •

0.03

cha

nnel

gra

dien

t (>1

0m

chan

nel w

idth

)

• M

ean

diam

eter

, len

gth,

and

vol

ume

of L

WD

incr

ease

d w

ith in

crea

sing

cha

nnel

wid

th

• Fr

eque

ncy

of o

ccur

renc

e (#

pie

ces/

m) d

ecre

ased

with

incr

easi

ng c

hann

el w

idth

•

Cha

nges

rela

ted

to in

crea

sed

capa

city

of l

arge

r stre

ams

to m

ove

woo

d do

wns

tream

•

Hig

her p

ropo

rtion

of w

ood

inpu

t rem

aine

d in

the

stre

am c

hann

el a

s st

ream

siz

e de

crea

sed

Bilb

y an

d W

ard

1989

Sout

heas

t Ala

ska

• <0

.03

chan

nel g

radi

ents

•

0.7

km2 to

55

km2 d

rain

age

area

s •

1st th

roug

h 4th

ord

er s

tream

s

• Ab

unda

nce

of L

WD

and

vol

ume

per c

hann

el le

ngth

(m3 /m

) inc

reas

ed w

ith in

crea

sing

st

ream

siz

e •

LWD

load

ing

(m3 /m

2 ) dec

reas

ed w

ith in

crea

sing

ban

kful

l wid

th

Rob

ison

and

Bes

chta

199

0

Wes

tern

W

ashi

ngto

n •

3 m

to 2

4 m

ban

kful

l cha

nnel

w

idth

s •

Aver

age

LWD

vol

ume

incr

ease

d w

ith in

crea

sing

stre

am s

ize

(ban

kful

l wid

th)

• LW

D a

bund

ance

(# p

iece

s/m

) dec

reas

ed w

ith in

crea

sing

ban

kful

l wid

th

Bilb

y an

d W

ard

1991

Uni

ted

King

dom

•

110

km2 d

rain

age

area

•

Den

sity

of L

WD

jam

s (n

umbe

r of d

ams

per 1

00m

and

num

ber o

f dam

s pe

r 500

m)

decr

ease

d do

wns

tream

from

hea

dwat

ers

and

with

incr

easi

ng c

hann

el w

idth

•

Abun

danc

e of

par

tial s

pann

ing

dam

s in

crea

sed

in d

owns

tream

dire

ctio

n •

Deb

ris lo

adin

g (k

g/m

2 ) dec

reas

ed in

dow

nstre

am d

irect

ion

Gre

gory

et a

l. 19

93

Sout

heas

tern

Fr

ance

•

0.00

12 to

0.0

018

chan

nel g

radi

ents

•

3700

km

2 dra

inag

e ar

ea

• 6th

ord

er ri

ver

• LW

D d

iffer

entia

lly d

epos

ited

on b

anks

dep

endi

ng o

n flo

w a

nd fo

rest

con

ditio

ns

Pieg

ay 1

993

Wes

tern

Ore

gon

• 0.

022

aver

age

chan

nel g

radi

ent

• 23

m a

vera

ge c

hann

el w

idth

•

60 k

m2 d

rain

age

area

• C

hann

el w

idth

and

sin

uosi

ty w

ere

mai

n fa

ctor

s co

ntro

lling

LWD

sup

ply

and

dist

ribut

ion

• N

umbe

r and

vol

ume

of L

WD

hig

hest

in w

ide

sinu

ous

reac

hes

Nak

amur

a an

d Sw

anso

n 19

94

31

Tabl

e A

-1.

The

char

acte

ristic

s an

d di

strib

utio

n of

LW

D in

var

ious

geo

grap

hic

loca

tions

and

cha

nnel

net

wor

k lo

catio

ns (a

rrang

ed in

chr

onol

ogic

al o

rder

).

LOC

ATIO

N

CH

ANN

EL N

ETW

OR

K P

OSI

TIO

N

RES

ULT

S R

EFER

ENC

ES

• 5th

ord

er c

hann

el

Sout

hwes

t Ala

ska

Wes

tern

W

ashi

ngto

n

• 0.

002

to 0

.085

cha

nnel

gra

dien

ts

• 2.

5m to

38m

cha

nnel

wid

ths

• N

umbe

r of L

WD

pie

ces

per m

2 dec

reas

ed w

ith in

crea

sing

cha

nnel

wid

th

• Lo

gs in

larg

er c

hann

els

wer

e m

ore

read

ily tr

ansp

orte

d •

Inve

rse

prop

ortio

nalit

y be

twee

n po

ol s

paci

ng a

nd L

WD

freq

uenc

y

Mon

tgom

ery

et a

l. 19

95

Nor

th C

entra

l C

olor

ado

• 0.

005

to 0

.065

cha

nnel

gra

dien

ts

• 4

m to

10

m b

ankf

ull w

idth

•

2 km

2 to 3

0 km

2 dra

inag

e ar

eas

• 1st

thro

ugh

3rd o

rder

stre

ams

• Ab

unda

nce

(pie

ces

of L

WD

/ m) g

reat

er in

sm

alle

r stre

ams

whe

n so

rted

by d

rain

age

area

and

wid

th

• Lo

wer

per

cent

age

of p

iece

s sp

anne

d ch

anne

l in

larg

er s

tream

s (s

orte

d by

dra

inag

e ar

ea a

nd w

idth

) tha

n in

sm

alle

r stre

ams

• Pe

rcen

tage

of L

WD

lyin

g pe

rpen

dicu

lar t

o st

ream

cha

nnel

dec

reas

ed w

ith in

crea

sing

dr

aina

ge a

rea

• Pe

rcen

tage

of d

ebris

pie

ces

in w

ette

d ch

anne

l dec

reas

ed in

larg

er s

tream

s •

Perc

enta

ge o

f LW

D a

bove

wet

ted

chan

nels

incr

ease

d in

sm

alle

r stre

ams

• LW

D ra

ndom

ly d

istri

bute

d in

stre

ams

<5.0

m w

idth

and

clu

mpe

d in

to ja

ms

in >

5.0

m

in w

idth

Ric

hmon

d an

d Fa

usch

199

5

Nor

thw

est

Was

hing

ton

• 0.

002

to 0

.05

chan

nel g

radi

ents

•

5 m

to 2

0 m

cha

nnel

wid

ths

• 2

km2 to

120

km

2 dr

aina

ge a

reas

• Lo

adin

g an

d ab

unda

nce

decr

ease

d w

ith in

crea

sing

cha

nnel

wid

th

• C

hann

el w

idth

is d

omin

ant i

nflu

ence

on

num

ber o

f pie

ces

of L

WD

/m2

Beec

hie

and

Sibl

ey 1

997

Sout

heas

tern

Fr

ance

•

0.00

12 to

0.0

018

chan

nel g

radi

ents

•

3700

km

2 dra

inag

e ar

ea

• 6th

ord

er ri

ver

• LW

D d

istri

butio

n on

mea

nder

ban

ks re

late

d to

ang

le o

f ban

k to

flow

, hei

ght o

f flo

w,

and

pres

ence

of s

econ

dary

cha

nnel

s Pi

egay

and

Mar

ston

199

8

Spai

n •

0.08

to 0

.157

5 ch

anne

l gra

dien

ts

• 3

m to

8 m

cha

nnel

wid

ths

• 0.

4 km

2 to 6

4 km

2 dra

inag

e ar

eas

• 1st

thro

ugh

3rd o

rder

stre

ams

• Ab

unda

nce

and

load

ing

decr

ease

d in

dow

nstre

am d

irect

ion

Elos

egi e

t al.

1999

Sout

heas

tern

Fr

ance

•

0.00

3 to

0.0

08 c

hann

el g

radi

ents

•

1650

km

2 dra

inag

e ar

ea

• M

ost L

WD

in a

ctiv

e ch

anne

l on

high

bar

s •

LWD

dep

osit

cont

rolle

d by

dep

osit

site

mor

phol

ogy

and

prox

imity

of L

WD

sou

rces

Pi

egay

et a

l. 19

99

32

Tabl

e A

-2.

The

mob

ility

and

trans

port

mec

hani

sms

of L

WD

in v

ario

us g

eogr

aphi

c lo

catio

ns a

nd c

hann

el p

ositi

ons

(arra

nged

in c

hron

olog

ical

ord

er).

LOC

ATIO

N

CH

ANN

EL N

ETW

OR

K P

OSI

TIO

N

RES

ULT

S R

EFER

ENC

ES

Wes

tern

Ore

gon

• 0.

10 to

0.3

0 ch

anne

l gra

dien

ts

• Ab

ility

of s

tream

s an

d riv

ers

to m

ove

LWD

dep

ends

on

size

of r

iver

and

woo

d di

men

sion

s •

Larg

e riv

ers

trans

port

mos

t LW

D, w

hile

sm

all s

tream

s m

ove

only

sm

all d

ebris

sho

rt di

stan

ces

befo

re d

epos

ition

on

chan

nel o

bstru

ctio

ns

• Ve

ry la

rge

debr

is in

sm

all c

hann

els

trans

porte

d th

roug

h de

bris

torre

nts

• To

rrent

s ra

re in

>2nd

to 3

rd o

rder

stre

ams,

as

stee

p gr

adie

nts

are

requ

ired

to m

ove

debr

is

Swan

son

et a

l. 19

76

Wes

tern

Ore

gon

• 0.

02 to

0.5

0 ch

anne

l gra

dien

ts

• D

ebris

mov

es th

roug

h flo

tatio

n at

hig

h flo

ws

or th

roug

h de

bris

flow

s •

Deb

ris fl

ows

orig

inat

e in

1st a

nd 2

nd o

rder

cha

nnel

s (>

50%

gra

dien

t, <0

.2 k

m2

drai

nage

are

as)

• 3rd

to 5

th o

rder

stre

ams

(4 k

m2 to

60

km2 d

rain

age

area

s) w

ide

enou

gh to

floa

t lar

ge

debr

is a

nd d

ebris

jam

s at

hig

h flo

ws

Swan

son

and

Lien

kaem

per

1978

East

ern

Was

hing

ton

• 0.

015

chan

nel g

radi

ent

• 11

.5 m

ban

kful

l wid

th

• LW

D m

ovem

ent d

epen

ded

on le

ngth

and

dia

met

er o

f woo

d •

Dis

tanc

e tra

vele

d by

LW

D in

vers

ely

rela

ted

to p

iece

leng

th

• An

chor

ed p

iece

s m

ore

stab

le in

hig

h flo

ws

Bilb

y 19

84

Indi

ana

Nor

th C

arol

ina

Ore

gon

• 0.

001

to 0

.40

chan

nel g

radi

ents

•

Deb

ris to

rrent

s m

ain

trans

port

mec

hani

sm in

sm

all,

high

gra

dien

t stre

ams

• Fl

otat

ion

mai

n tra

nspo

rt m

echa

nism

in la

rge,

low

gra

dien

t stre

ams

Kelle

r and

Sw

anso

n 19

79

Wes

tern

Ore

gon

• 0.

03 to

0.3

7 ch

anne

l gra

dien

ts

• 3.

5 m

to 2

4 m

ban

kful

l wid

ths

• 0.

1 km

2 to 6

1 km

2 dra

inag

e ar

eas

• 1st

thro

ugh

5th o

rder

stre

ams

• W

ood

mov

emen

t occ

urre

d in

larg

er s

tream

s du

ring

mon

itorin

g pe

riod

• D

ista

nce

mov

ed d

epen

ded

on p

iece

leng

th in

rela

tion

to b

ankf

ull w

idth

•

All p

iece

s m

ovin

g >1

0 m

wer

e sh

orte

r tha

n ba

nkfu

ll w

idth

Lien

kaem

per a

nd S

wan

son

1987

Wes

tern

Ore

gon

• 0.

03 to

0.2

1 ch

anne

l gra

dien

ts

• 7

m to

25

m b

ankf

ull w

idth

s •

2nd th

roug

h 5th

ord

er s

tream

s

• D

ebris

flow

s re

dist

ribut

ed L

WD

in s

teep

, low

- ord

er s

tream

s •

Floo

ds re

dist

ribut

ed L

WD

in m

ediu

m to

hig

h or

der s

tream

s N

akam

ura

and

Swan

son

1993

Wes

tern

Ore

gon

• 0.

019

to 0

.028

cha

nnel

gra

dien

ts

• 9

m to

71

m c

hann

el w

idth

s •

5th o

rder

stre

ams

• M

ost t

rans

porte

d pi

eces

sho

rter t

han

bank

full

wid

th

• 20

% o

f unt

rans

porte

d pi

eces

long

er th

an b

ankf

ull w

idth

•

LWD

leng

th to

cha

nnel

wid

th u

sefu

l mea

sure

of s

usce

ptib

ility

to tr

ansp

ort

Nak

amur

a an

d Sw

anso

n 19

94

Wyo

min

g •

0.04

1 to

0.0

55 c

hann

el g

radi

ents

•

6.4

m to

7 m

low

flow

cha

nnel

w

idth

s

• LW

D le

ss s

tabl

e in

bur

ned

drai

nage

due

to in

crea

sed

runo

ff an

d pe

ak fl

ows,

and

de

crea

sed

bank

sta

bilit

y Yo

ung

1994

NA

• N

A •

Unc

onge

sted

, sem

i-con

gest

ed, a

nd c

onge

sted

woo

d tra

nspo

rt ba

sed

on ra

tio o

f log

in

put (

Qlo

g) to

stre

am d

isch

arge

(Qw)

• Pr

opos

ed a

bilit

y of

cha

nnel

to re

tain

woo

d is

func

tion

of d

ebris

roug

hnes

s, w

hich

va

ries

with

cha

nnel

and

log

dim

ensi

ons

Brau

dric

k et

al.

1997

Spai

n •

0.08

to 0

.157

5 ch

anne

l gra

dien

ts

• 3

m to

14

m b

ankf

ull w

idth

s •

0.8

km2 to

69

km2 d

rain

age

area

s •

1st th

roug

h 3rd

ord

er s

tream

s

• Lo

adin

g(m

3 /m2 ) d

ecre

ased

in d

owns

tream

dire

ctio

n •

Log

mob

ility

incr

ease

d in

larg

er s

tream

s El

oseg

i et a

l. 19

99

33

Tabl

e A

-3.

The

effe

ct o

f LW

D o

n th

e st

orag

e an

d tra

nspo

rt of

bed

load

in v

ario

us g

eogr

aphi

c lo

catio

ns a

nd c

hann

el n

etw

ork

posi

tions

(arra

nged

in c

hron

olog

ical

ord

er).

LOC

ATIO

N

CH

ANN

EL N

ETW

OR

K P

OSI

TIO

N

RES

ULT

S R

EFER

ENC

ES

Wes

tern

Ore

gon

• 0.

10 to

0.3

0 ch

anne

l gra

dien

ts

• LW

D a

ccum

ulat

ions

trap

ped

230

m3 o

f sed

imen

t in

100

m re

ach

Swan

son

et a

l. 19

76

Wes

tern

Ore

gon

• 0.

02 to

0.5

0 ch

anne

l gra

dien

ts

• An

nual

sed

imen

t yie

ld le

ss th

at 1

0% o

f mat

eria

l in

stor

age

• W

oody

mat

eria

l prim

ary

stor

age

elem

ent

Swan

son

and

Lien

kaem

per 1

978

Nor

ther

n C

alifo

rnia

•

0.01

to 0

.048

cha

nnel

gra

dien

ts

• 6.

4 m

to 9

.6 m

cha

nnel

wid

ths

• D

ebris

sto

red

sedi

men

t com

pris

ed 4

0% o

f cha

nnel

are

a

Kelle

r and

Tal

ly 1

979,

Ta

lly 1

980,

Kel

ler a

nd

Mac

Don

ald

1983

, Kel

ler

et a

l. 19

85

Wes

tern

W

ashi

ngto

n •

0.07

cha

nnel

gra

dien

t •

Acce

lera

ted

eros

ion

of s

tore

d se

dim

ent (

5000

m3 ) a

fter L

WD

rem

oval

from

250

m re

ach

Besc

hta

1979

New

Ham

pshi

re

• 0.

21 a

vera

ge c

hann

el g

radi

ent

• 3.

0 m

ave

rage

cha

nnel

wid

th

• 50

0% in

crea

se in

the

expo

rt of

sto

red

sedi

men

t afte

r rem

oval

of L

WD

Bi

lby

1981

Cen

tral I

daho

•

0.21

ave

rage

cha

nnel

gra

dien

t •

5.0

m a

vera

ge c

hann

el w

idth

•

1st th

roug

h 3rd

ord

er s

tream

s

• Lo

gs a

ccou

nted

for 5

0% o

f tot

al s

edim

ent s

tore

d an

d 34

% o

f all

chan

nel o

bstru

ctio

ns

Meg

ahan

198

2

East

ern

Was

hing

ton

• 0.

015

chan

nel g

radi

ent

• 11

.5 m

ban

kful

l wid

th

• R

educ

tion

in s

edim

ent s

tora

ge (i

ncre

ased

sco

ur) a

fter l

oggi

ng a

nd s

tream

cle

anin

g Bi

lby

1984

Ariz

ona

• 0.

07 to

0.0

9 ch

anne

l gra

dien

ts

• R

emov

al o

f log

ste

ps le

d to

incr

ease

d be

dloa

d m

ovem

ent

• Lo

g st

eps

even

tual

ly re

plac

ed g

rave

l bar

s H

eede

198

5a, 1

985b

Nor

ther

n C

alifo

rnia

•

0.00

6 to

0.0

14 c

hann

el g

radi

ents

•

12 m

ave

rage

cha

nnel

wid

th

• LW

D o

bstru

ctio

ns s

tabi

lize

grav

el c

hann

els

by c

ontro

lling

loca

tion

of p

ools

and

bar

s Li

sle

1986

a

Alas

ka

• 0.

01 to

0.0

9 ch

anne

l gra

dien

ts

• 2

m to

6 m

act

ive

chan

nel w

idth

•

Deb

ris a

cted

to s

tore

sed

imen

t thr

ough

the

crea

tion

of lo

w-e

nerg

y de

posi

tiona

l en

viro

nmen

ts

Lisl

e 19

86b

Wes

tern

W

ashi

ngto

n •

0.13

cha

nnel

gra

dien

t (<7

m

chan

nel w

idth

) •

0.08

cha

nnel

gra

dien

t (7

m to

10

m

chan

nel w

idth

) •

0.03

cha

nnel

gra

dien

t (>1

0 m

ch

anne

l wid

th)

• 40

% o

f LW

D a

ssoc

iate

d w

ith s

edim

ent a

ccum

ulat

ions

in 0

.13

chan

nel g

radi

ent (

<7 m

ch

anne

l wid

th)

• <3

0% o

f LW

D a

ssoc

iate

d w

ith s

edim

ent a

ccum

ulat

ions

in 0

.08

chan

nel g

radi

ent (

7 m

to

10 m

cha

nnel

wid

th)

• <2

0% o

f LW

D a

ssoc

iate

d w

ith s

edim

ent a

ccum

ulat

ions

in 0

.03

chan

nel g

radi

ent (

>10

m

chan

nel w

idth

)

Bilb

y an

d W

ard

1989

Wes

tern

W

ashi

ngto

n •

3 m

to 2

4 m

ban

kful

l cha

nnel

w

idth

s •

Sedi

men

t ret

entio

n by

LW

D d

ecre

ased

with

incr

easi

ng s

tream

siz

e •

Amou

nt o

f sed

imen

t ret

aine

d w

as a

ssoc

iate

d w

ith a

ge a

nd v

olum

e of

woo

d Bi

lby

and

War

d 19

91

Wes

tern

Ore

gon

• 0.

03 to

0.2

1 ch

anne

l gra

dien

ts

• 7

m to

25

m b

ankf

ull w

idth

s •

2nd th

roug

h 5th

ord

er s

tream

s

• LW

D w

as d

omin

ant s

edim

ent s

tora

ge c

ontro

l in

mod

erat

e to

ste

ep s

tream

s •

LWD

pro

vide

d te

mpo

rary

sed

imen

t sto

rage

in lo

w g

radi

ent r

each

es

Nak

amur

a an

d Sw

anso

n 19

93

Sout

heas

t Ala

ska

• 0.

01 a

vera

ge c

hann

el g

radi

ent

• 3.

9 m

ave

rage

cha

nnel

wid

th

• Fo

ur fo

ld in

crea

se in

bed

load

tran

spor

t and

ban

kful

l flo

w a

fter L

WD

rem

oval

Sm

ith e

t al.

1993

a

Sout

heas

t Ala

ska

• 0.

008

to 0

.010

cha

nnel

gra

dien

ts

• 4

m a

vera

ge c

hann

el w

idth

•

Initi

al in

crea

se in

bed

load

tran

spor

t afte

r LW

D re

mov

al

• C

hann

el re

adju

sted

to fo

rm s

erie

s of

regu

larly

spa

ced

bars

Sm

ith e

t al.

1993

b

Col

orad

o •

0.11

ave

rage

cha

nnel

gra

dien

t •

3.5

m a

vera

ge c

hann

el w

idth

•

LWD

maj

or c

ontro

l on

the

stor

age

and

rele

ase

of b

edlo

ad

Aden

lof a

nd W

ohl 1

994

Wes

tern

W

ashi

ngto

n •

0.10

to 0

.15

chan

nel g

radi

ents

•

3 m

to 4

m c

hann

el w

idth

s •

LWD

maj

or c

ontro

l on

sedi

men

t sto

rage

and

tran

spor

t O

'Con

nor 1

994

Belg

ium

•

0.04

6 ch

anne

l gra

dien

t •

0.75

m c

hann

el w

idth

•

2nd o

rder

stre

am

• Lo

g ja

ms

redu

ced

mov

emen

t and

tran

spor

t of b

ed m

ater

ial

Assa

ni a

nd P

etit

1995

Nor

thea

st

Cal

iforn

ia

• 0.

02 to

0.0

8 ch

anne

l gra

dien

ts

• 2

m to

12

m b

ankf

ull w

idth

•

LWD

min

imal

effe

ct o

n se

dim

ent s

tora

ge

Berg

et a

l. 19

98

Nor

thw

est

Was

hing

ton

• 0.

24 a

vera

ge c

hann

el g

radi

ent

• 2.

7 m

ave

rage

cha

nnel

wid

th

• LW

D m

ain

com

pone

nt o

f sed

imen

t sto

ring

obst

ruct

ions

G

rizze

l and

Wol

ff 19

98

34

Tabl

e A

-4.

The

effe

ct o

f LW

D o

n ch

anne

l dim

ensi

on in

var

ious

geo

grap

hic

loca

tions

and

cha

nnel

net

wor

k po

sitio

ns (a

rrang

ed in

chr

onol

ogic

al o

rder

). LO

CAT

ION

C

HAN

NEL

NET

WO

RK

PO

SITI

ON

R

ESU

LTS

REF

EREN

CES

W

este

rn O

rego

n •

0.10

to 0

.30

chan

nel g

radi

ents

•

LWD

con

trolle

d 30

% to

50%

of d

rop

in e

leva

tion

• LW

D c

ause

d pr

onou

nced

loca

l wid

enin

g of

stre

ams

Swan

son

et a

l. 19

76

Indi

ana

Nor

th C

arol

ina

Ore

gon

• 0.

001

to 0

.40

chan

nel g

radi

ents

•

Scou

r aro

und

jam

s in

crea

sed

chan

nel w

idth

by

mor

e th

an 5

0%

• In

low

gra

dien

t stre

ams

LWD

con

tribu

ted

to fo

rmat

ion

of m

eand

er c

utof

fs

• In

hig

h gr

adie

nt s

tream

s se

dim

ent s

tora

ge b

y ch

anne

l spa

nnin

g w

ood

crea

ted

step

ped

prof

ile

• LW

D c

ontro

ls 3

0% to

80%

of d

rop

in e

leva

tion

Kelle

r and

Sw

anso

n 19

79

Nor

thw

est

Cal

iforn

ia

• 0.

01 to

0.4

0 ch

anne

l gra

dien

ts

• 1.

0 km

2 to 2

7 km

2 dra

inag

e ar

eas

• 2nd

thro

ugh

4th o

rder

stre

ams

• W

idth

at d

ebris

acc

umul

atio

ns w

as tw

o or

mor

e tim

es g

reat

er th

an c

hara

cter

istic

wid

th

of c

hann

el

• 60

% o

f dro

p in

ele

vatio

n as

soci

ated

with

LW

D s

teps

•

Exte

nsiv

e po

nded

sed

imen

t ups

tream

LW

D

• W

ood

had

less

effe

ct o

n el

evat

ion

in s

teep

er s

ectio

ns w

here

deb

ris re

sts

of a

bove

bo

ulde

rs

• LW

D c

reat

ed v

arie

ty o

f cha

nnel

dep

ths

• Ef

fect

on

elev

atio

n de

crea

sed

with

dec

reas

ing

chan

nel g

radi

ent

Kelle

r and

Tal

ly 1

979,

Tal

ly

1980

, Kel

ler a

nd

Mac

Don

ald

1983

, Kel

ler e

t al

. 198

5

Sout

heas

t Ala

ska

• 42

km

2 dra

inag

e ar

ea

• LW

D s

igni

fican

t inf

luen

ce o

n ch

anne

l wid

th, a

nd p

ool a

nd b

ar lo

catio

n Br

yant

198

0 So

uthe

ast A

lask

a •

15 k

m2 to

75

km2 d

rain

age

area

s •

4th th

roug

h 5th

ord

er s

tream

s •

LWD

acc

umul

atio

ns d

ecre

ased

30

year

s af

ter l

oggi

ng

• LW

D in

fluen

ced

chan

nel w

idth

, and

poo

l and

bar

loca

tion

Brya

nt 1

983

Ariz

ona

• 0.

07 to

0.0

9 ch

anne

l gra

dien

ts

• Lo

g st

eps

cont

rolle

d ch

anne

l gra

dien

t in

stee

p st

ream

s •

LWD

par

t of n

atur

al h

ydra

ulic

geo

met

ry o

f ste

ep s

tream

s H

eede

198

5a, b

Sout

heas

t Ala

ska

• <0

.03

chan

nel g

radi

ents

•

0.7

km2 to

55

km2 d

rain

age

area

s •

1st th

roug

h 4th

ord

er s

tream

s

• Va

riatio

ns in

loca

l ban

kful

l wid

th a

ssoc

iate

d w

ith v

olum

e of

inst

ream

LW

D

Rob

ison

and

Bes

chta

199

0

Wes

tern

Ore

gon

• 0.

03 to

0.2

1 ch

anne

l gra

dien

ts

• 7

m to

25

m b

ankf

ull w

idth

s •

2nd th

roug

h 5th

ord

er s

tream

s

• W

idth

s up

stre

am o

f key

LW

D a

nd L

WD

jam

s w

ider

than

cha

nnel

ave

rage

s in

med

ium

or

der c

hann

els

• G

radi

ents

ups

tream

of k

ey L

WD

and

LW

D ja

ms

low

er th

an c

hann

el a

vera

ges

in

med

ium

ord

er c

hann

els

• C

hann

els

with

key

LW

D a

re 1

.5 ti

mes

wid

er th

an c

hann

els

with

out k

ey L

WD

Nak

amur

a an

d Sw

anso

n 19

93

Uni

ted

King

dom

•

11 k

m2 d

rain

age

area

•

LWD

acc

umul

atio

ns c

ontri

bute

d to

ban

k su

bsid

ence

lead

ing

to w

iden

ing

in s

tream

ch

anne

l D

avis

and

Gre

gory

199

4

Coa

stal

Brit

ish

Col

umbi

a, C

anad

a •

0.01

to 0

.07

chan

nel g

radi

ents

•

10 m

to 3

0 m

ban

kful

l wid

ths

• 1

km2 to

28

km2 d

rain

age

area

s

• LW

D ja

ms

grea

test

influ

ence

on

over

all c

hann

el m

orph

olog

y •

Indi

vidu

al p

iece

s im

porta

nt to

cha

nnel

mor

phol

ogy

and

smal

l sca

le fe

atur

es b

etw

een

jam

s •

Cha

nnel

s in

crea

sed

in c

ompl

exity

as

LWD

jam

s ag

e an

d de

terio

rate

•

Old

gro

wth

fore

sts

had

low

rate

of j

am fo

rmat

ion,

lead

ing

to ja

ms

of v

ario

us a

ges

and

com

plex

cha

nnel

form

s •

Dis

turb

ed fo

rest

s ha

ve a

ccel

erat

ed ra

tes

of ja

m fo

rmat

ion,

alte

ring

chan

nel

mor

phol

ogy

Hog

an e

t al.

1998

35

Tabl

e A

-5.

The

effe

ct o

f LW

D o

n po

ol fo

rmat

ion

and

spac

ing

in v

ario

us g

eogr

aphi

c lo

catio

ns a

nd c

hann

el n

etw

ork

posi

tions

(arra

nged

in c

hron

olog

ical

ord

er).

LOC

ATIO

N

CH

ANN

EL N

ETW

OR

K P

OSI

TIO

N

RES

ULT

S R

EFER

ENC

ES

Wes

tern

Ore

gon

• 0.

02 to

0.5

0 ch

anne

l gra

dien

ts

• In

ste

ep s

tream

s, L

WD

mai

n fa

ctor

in d

eter

min

ing

char

acte

r of a

quat

ic h

abita

t •

In lo

w g

radi

ent s

tream

s, h

ydra

ulic

fact

ors

regu

late

d aq

uatic

hab

itat

Swan

son

and

Lien

kaem

per

1978

N

orth

wes

t C

alifo

rnia

•

0.01

to 0

.048

cha

nnel

gra

dien

ts

• 1.

0 km

2 to 2

7 km

2 dra

inag

e ar

eas

• 2nd

thro

ugh

4th o

rder

stre

ams

• D

ebris

enh

ance

d po

ols

incr

ease

d in

abu

ndan

ce w

ith in

crea

sing

cha

nnel

gra

dien

t •

Pool

spa

cing

dec

reas

ed w

ith in

crea

sing

gra

dien

t Ke

ller a

nd T

ally

197

9, T

ally

19

80, K

elle

r and

Mac

Don

ald

1983

, Kel

ler e

t al.

1985

So

uthe

ast A

lask

a •

0.00

1 to

0.0

3 ch

anne

l gra

dien

t •

2nd th

roug

h 4th

ord

er s

tream

s •

Mos

t poo

ls fo

rmed

by

debr

is

• Po

ol v

olum

e an

d de

bris

vol

ume

dire

ctly

cor

rela

ted

Mur

phy

et a

l. 19

86

Nor

thw

est

Cal

iforn

ia

• 0.

006

to 0

.014

cha

nnel

gra

dien

ts

• 12

m a

vera

ge c

hann

el w

idth

•

26 k

m2 d

rain

age

area

• LW

D m

ajor

com

pone

nt in

poo

l for

mat

ion

and

long

-ter m

cha

nnel

sta

bilit

y Li

sle

1986

a

Alas

ka

• 0.

01 to

0.0

9 ch

anne

l gra

dien

ts

• 2

m to

6 m

act

ive

chan

nel w

idth

•

0.5

km2 to

2.0

km

2 dra

inag

e ar

eas

• D

ebris

form

ed 4

6% o

f poo

ls in

fore

sted

reac

hes

• D

ebris

form

ed 8

6% o

f poo

ls in

cle

ar-c

ut s

tream

s du

e to

hig

her d

ebris

load

s Li

sle

1986

b

Wes

tern

Ore

gon

• 0.

02 to

0.0

6 ch

anne

l gra

dien

ts

• 5.

5 m

to 8

m b

ankf

ull w

idth

s •

6.5

km2 d

rain

age

area

• 70

% o

f all

pool

s cr

eate

d pr

imar

ily b

y w

oody

deb

ris

Andr

us e

t al.

1988

Wes

tern

W

ashi

ngto

n •

0.13

cha

nnel

gra

dien

t (<7

m

chan

nel w

idth

) •

0.08

cha

nnel

gra

dien

t (7

m to

10

m

chan

nel w

idth

) •

0.03

cha

nnel

gra

dien

t (>1

0 m

ch

anne

l wid

th)

• Pl

unge

poo

l mos

t com

mon

(40%

) of a

ll LW

D fo

rmed

poo

ls in

stre

ams

<7m

•

Scou

r poo

ls m

ost c

omm

on (6

0%) o

f all

LWD

form

ed p

ools

in s

tream

s >1

0m

Bilb

y an

d W

ard

1989

East

ern

Ore

gon

• 0.

02 to

0.0

7 ch

anne

l gra

dien

ts

• 2

m to

6 m

ban

kful

l wid

th

• 7

km2 to

25

km2 d

rain

age

area

s

• D

ebris

ass

ocia

ted

with

64%

of p

ools

in a

ll st

udy

stre

ams

Car

lson

et a

l. 19

90

Sout

heas

t Ala

ska

• <0

.03

chan

nel g

radi

ents

•

0.7

km2 to

55

km2 d

rain

age

area

s •

1st th

roug

h 4th

ord

er s

tream

s

• LW

D fo

rmed

65%

to 7

5% o

f all

pool

s •

Shift

in p

ool t

ypes

from

plu

nge

to la

tera

l sco

ur in

sm

all t

o la

rge

stre

ams

Rob

ison

and

Bes

chta

199

0

Wes

tern

W

ashi

ngto

n •

3 m

to 2

4 m

ban

kful

l cha

nnel

w

idth

s •

Freq

uenc

y of

LW

D a

ssoc

iate

d po

ols

decr

ease

d w

ith in

crea

sing

stre

am s

ize

• Ty

pe o

f poo

ls a

ssoc

iate

d w

ith L

WD

cha

nged

with

stre

am s

ize:

sco

ur p

ools

incr

ease

d an

d pl

unge

poo

ls d

ecre

ased

with

incr

easi

ng s

tream

siz

e

Bilb

y an

d W

ard

1991

Wes

tern

Te

nnes

see

• 13

,000

km

2 dra

inag

e ar

ea

• Ph

ysic

al h

abita

t div

ersi

ty s

trong

ly c

orre

late

d w

ith p

rese

nce

of L

WD

Sh

ield

s an

d Sm

ith 1

992

Sout

heas

t Ala

ska

• 0.

008

to 0

.01

chan

nel g

radi

ents

•

4 m

ave

rage

cha

nnel

wid

ths

• 1.

5 km

2 dra

inag

e ar

ea

• Po

ol s

paci

ng s

imila

r bef

ore

and

afte

r LW

D re

mov

al

• Po

ols

form

ed b

y no

n-er

odib

le o

bstru

ctio

ns b

ut la

cked

cov

er a

nd h

ydra

ulic

com

plex

ity

of w

ood

form

ed p

ools

Smith

et a

l. 19

93a,

b

Wes

tern

Ore

gon

• 0.

022

aver

age

chan

nel g

radi

ent

• 23

m a

vera

ge c

hann

el w

idth

•

60 k

m2 d

rain

age

area

•

5th o

rder

cha

nnel

• 17

% o

f LW

D in

wid

e si

nuou

s ch

anne

ls a

ffect

ed p

ool f

orm

atio

n, 2

to 5

tim

es h

ighe

r th

an o

ther

cha

nnel

type

s N

akam

ura

and

Swan

son

1994

Sout

hwes

t Ala

ska

Wes

tern

W

ashi

ngto

n

• 0.

002

to 0

.085

cha

nnel

gra

dien

ts

• 2.

5 to

38

m c

hann

el w

idth

s •

73%

of p

ools

form

ed b

y LW

D

• 40

% o

f woo

d in

fluen

ced

pool

form

atio

n •

Pool

spa

cing

inve

rsel

y re

late

d to

LW

D lo

adin

g in

pla

ne-b

ed, p

ool-r

iffle

, and

forc

ed

pool

riffl

e ch

anne

ls.

• Po

ol s

paci

ng in

crea

sed

with

incr

easi

ng c

hann

el w

idth

•

>0.0

1 gr

adie

nt c

hann

els:

poo

l for

mat

ion

sens

itive

to L

WD

load

ing

• <0

.01

grad

ient

cha

nnel

s: p

ool f

orm

atio

n no

t as

sens

itive

to L

WD

load

ing

Mon

tgom

ery

et a

l. 19

95

36

Tabl

e A

-5.

The

effe

ct o

f LW

D o

n po

ol fo

rmat

ion

and

spac

ing

in v

ario

us g

eogr

aphi

c lo

catio

ns a

nd c

hann

el n

etw

ork

posi

tions

(arra

nged

in c

hron

olog

ical

ord

er).

LOC

ATIO

N

CH

ANN

EL N

ETW

OR

K P

OSI

TIO

N

RES

ULT

S R

EFER

ENC

ES

Nor

th C

entra

l C

olor

ado

• 0.

005

to 0

.065

cha

nnel

gra

dien

ts

• 4

m to

10

m b

ankf

ull w

idth

•

2 km

2 to 3

0 km

2 dra

inag

e ar

eas

• 1st

thro

ugh

3rd o

rder

stre

ams

• 76

% o

f poo

ls fo

rmed

by

LWD

•

4% to

20%

of L

WD

form

ed p

ools

•

Hig

her p

ropo

rtion

of L

WD

in s

mal

ler s

tream

s fo

rmed

poo

ls

• LW

D p

erpe

ndic

ular

to fl

ow in

sm

all s

tream

s •

LWD

dia

gona

l to

flow

in la

rge

stre

ams

Ric

hmon

d an

d Fa

usch

199

5

Nor

thw

est

Was

hing

ton

• 0.

005

to 0

.01

chan

nel g

radi

ents

•

30 m

to 8

0 m

ban

kful

l wid

ths

• 75

km

2 to 2

25 k

m2 d

rain

age

area

s

• 70

% o

f all

obse

rved

poo

ls a

ssoc

iate

d w

ith L

WD

jam

s •

LWD

ass

ocia

ted

pool

s pr

ovid

e ot

her h

abita

t val

ues

(cov

er a

nd n

utrie

nt tr

appi

ng) n

ot

asso

ciat

ed w

ith p

ools

form

ed o

ther

way

s

Abbe

and

Mon

tgom

ery

1996

Nor

thw

este

rn

Nev

ada

• 0.

02 to

0.0

4 ch

anne

l gra

dien

ts

• 10

% o

f LW

D p

iece

s fo

rmed

poo

ls

• LW

D p

ositi

ve e

ffect

s on

poo

l for

mat

ion

and

qual

ity

Mye

rs a

nd S

wan

son

1996

a,

b So

uthw

est A

lask

a •

0.00

17 to

0.0

224

chan

nel g

radi

ents

•

4 m

to 2

5 m

ban

kful

l wid

ths

• 1

km2 to

40

km2 d

rain

age

area

s

• LW

D a

ssoc

iate

d w

ith 8

0% o

f poo

ls in

und

istu

rbed

stre

ams

• LW

D a

ssoc

iate

d w

ith 5

5% o

f poo

ls in

dis

turb

ed s

tream

s W

ood-

Smith

and

Buf

fingt

on

1996

Nor

thw

est

Was

hing

ton

• 0.

002

to 0

.05

chan

nel g

radi

ents

•

5 m

to 2

0 m

cha

nnel

wid

ths

• 2

km2 to

120

km

2 dra

inag

e ar

eas

• Po

ol fo

rmat

ion

in m

oder

ate

to s

teep

slo

pe c

hann

els

high

ly s

ensi

tive

to p

rese

nce

of

LWD

•

Pool

s in

low

slo

pe c

hann

els

form

ed b

y m

echa

nism

s ot

her t

han

LWD

Beec

hie

and

Sibl

ey 1

997

Sout

hwes

t Virg

inia

•

0.01

to 0

.06

chan

nel g

radi

ents

•

5 m

ave

rage

cha

nnel

wid

th

• N

umbe

r of p

ools

in lo

w g

radi

ent r

each

es in

crea

sed

afte

r LW

D a

dditi

ons

• N

umbe

r of p

ools

in h

igh

grad

ient

reac

hes

unch

ange

d af

ter L

WD

add

ition

s

Hild

erbr

and

et a

l. 19

97

Uni

ted

King

dom

•

0.01

3 av

erag

e ch

anne

l gra

dien

t •

2 m

to 5

m b

ankf

ull w

idth

•

Pool

s cl

osel

y as

soci

ated

with

LW

D

• C

hann

els

with

LW

D h

ave

mor

e po

ols

than

cha

nnel

s w

ithou

t LW

D

Gur

nell

and

Swee

t 199

8

Sout

heas

tern

Fr

ance

•

0.00

3 to

0.0

08 c

hann

el g

radi

ents

•

1650

km

2 dra

inag

e ar

ea

• Fe

w p

ools

ass

ocia

ted

with

LW

D

• Po

ol s

ize

inde

pend

ent o

f LW

D m

ass

• LW

D a

ccum

ulat

ions

mod

erat

ely

influ

ence

d ch

anne

l mor

phol

ogy

Pieg

ay e

t al.

1999

37

Tabl

e A

-6.

The

effe

ct o

f LW

D o

n nu

trien

t dyn

amic

s in

var

ious

geo

grap

hic

loca

tions

and

cha

nnel

net

wor

k po

sitio

ns (a

rrang

ed in

chr

onol

ogic

al o

rder

). LO

CAT

ION

C

HAN

NEL

NET

WO

RK

LO

CAT

ION

R

ESU

LTS

REF

EREN

CES

N

ew H

amps

hire

•

0.21

cha

nnel

gra

dien

t •

3 m

ave

rage

cha

nnel

wid

th

• 1st

thro

ugh

3rd o

rder

stre

ams

• In

crea

se in

org

anic

car

bon

trans

port

afte

r woo

d re

mov

al

• 18

% in

crea

se in

DO

C e

xpor

t, 63

2% in

crea

se in

FPO

C e

xpor

t, 13

8% in

crea

se in

C

POM

exp

ort a

fter w

ood

rem

oval

•

Dec

reas

e in

num

ber o

f deb

ris d

ams

dow

n th

e st

ream

net

wor

k: 3

3.5

per 1

00 m

(1st

orde

r), 1

3.7

per 1

00m

(2nd

ord

er),

2.5

per 1

00m

(3rd

ord

er)

Bilb

y an

d Li

kens

198

0

New

Ham

pshi

re

• 0.

21 c

hann

el g

radi

ent

• 3m

ave

rage

ban

kful

l wid

th

• 2nd

ord

er s

tream

• In

crea

se in

FPO

M a

nd C

POM

(500

%) e

xpor

t dur

ing

high

dis

char

ge a

fter L

WD

re

mov

al

• Tw

igs

and

FPO

M a

re im

porta

nt fo

r ret

entio

n of

mat

ter,

fillin

g cr

acks

and

hol

es in

da

ms

to m

ake

them

wat

ertig

ht a

nd in

crea

se h

ydra

ulic

effe

ctiv

enes

s •

Dec

reas

e in

LW

D c

ontro

l on

elev

atio

n fro

m 1

st th

roug

h 3rd

ord

er s

tream

s: 5

2% o

f el

evat

ion

drop

con

trolle

d by

woo

d in

1st o

rder

stre

ams,

46%

of e

leva

tion

drop

co

ntro

lled

by w

ood

in 2

nd o

rder

stre

ams,

10%

of e

leva

tion

drop

con

trolle

d by

woo

d in

3rd

ord

er s

tream

s

Bilb

y 19

81

Nor

thw

est

Was

hing

ton

• 0.

014

(0.0

1 to

0.0

2) c

hann

el

grad

ient

s •

7.7

km2 b

asin

are

a

• G

ener

al p

ositi

ve tr

end

in L

WD

/m2 a

nd n

umbe

r of c

oho

salm

on (O

ncor

hync

hus

kisu

tch)

car

cass

es re

tain

ed

• C

arca

sses

impo

rtant

sou

rce

of n

utrie

nts

for a

quat

ic a

nd te

rrest

rial o

rgan

ism

s

Ced

erho

lm a

nd P

eter

son

1985

Wes

tern

Ore

gon

• 0.

10 c

hann

el g

radi

ent

• 6

km2 d

rain

age

area

•

Woo

dy d

ebris

is s

ourc

e fo

r app

roxi

mat

ely

90g/

m/y

r fin

e pa

rticu

late

org

anic

mat

ter

(FPO

M)

• Fo

r sys

tem

s w

ith la

rge

accu

mul

atio

n of

LW

D a

nd F

WD

, woo

d pr

oces

sing

can

be

sign

ifica

nt s

ourc

e of

FPO

M p

ool

• FP

OM

from

woo

d po

tent

ial t

o be

gre

ater

than

that

gen

erat

ed fr

om le

af a

nd n

eedl

e lit

ter p

roce

ssin

g •

Prod

uctio

n of

FPO

M g

reat

er in

win

ter f

rom

phy

sica

l abr

asio

n

War

d an

d Au

men

198

6

Paci

fic N

orth

wes

t, U

SA

• N

/A

• W

ood

deco

mpo

ses

mor

e sl

owly

in w

ater

than

on

land

•

Wat

erlo

ggin

g pr

even

ts d

eep

pene

tratio

n of

O2 i

n w

ood

• D

ecom

posi

tion

incr

ease

s as

gra

zing

or a

bras

ion

incr

ease

s; a

llow

s O

2 to

pene

trate

w

ood

• C

once

ntra

tion

of c

arbo

n an

d ni

troge

n in

crea

ses

as w

ood

deco

mpo

ses

• In

crea

se in

nitr

ogen

con

cent

ratio

n as

car

bon

use

incr

ease

s, a

nd th

roug

h ni

troge

n fix

ing

mic

roor

gani

sms

• N

itrog

en fi

xatio

n on

woo

d ac

coun

ts fo

r 5%

to 1

0% o

f ann

ual n

itrog

en s

uppl

y to

st

ream

•

Slow

dec

ompo

sitio

n of

inst

ream

woo

d in

fluen

ces

stab

ility

of L

WD

and

max

imiz

es

role

in h

abita

t for

mat

ion

Sede

ll et

al.

1988

Nor

thw

est

Was

hing

ton

• Lo

w g

radi

ent

• 15

m to

24

m c

hann

el w

idth

•

3rd o

rder

stre

ams

• D

ista

nce

carc

asse

s dr

ifted

inve

rsel

y re

late

d to

deb

ris lo

ads

• La

rge

and

smal

l org

anic

deb

ris m

ost i

mpo

rtant

inst

ream

rete

ntio

n el

emen

t for

ca

rcas

ses

• LW

D fo

rmed

poo

ls w

ere

mos

t im

porta

nt c

arca

ss d

epos

ition

site

Ced

erho

lm e

t al.

1989

Nor

ther

n O

rego

n •

N/A

•

At re

ach

scal

e w

ood

not s

igni

fican

t in

Nitr

ogen

and

Pho

spho

rous

upt

ake

• At

sm

alle

r sca

les,

woo

dy d

ebris

act

ive

in N

itrog

en a

nd P

hosp

horo

us re

mov

al fr

om

wat

er

• W

oody

deb

ris p

rovi

des

surfa

ce a

rea

for m

icro

orga

nism

s th

at u

ptak

e nu

trien

ts

Aum

en e

t al.

1990

38

Tabl

e A

-6.

The

effe

ct o

f LW

D o

n nu

trien

t dyn

amic

s in

var

ious

geo

grap

hic

loca

tions

and

cha

nnel

net

wor

k po

sitio

ns (a

rrang

ed in

chr

onol

ogic

al o

rder

). LO

CAT

ION

C

HAN

NEL

NET

WO

RK

LO

CAT

ION

R

ESU

LTS

REF

EREN

CES

N

ew H

amps

hire

•

N/A

•

Org

anic

deb

ris d

ams

impo

rtant

site

s of

org

anic

mat

ter r

eten

tion

and

accu

mul

atio

n •

Com

mun

ity re

spira

tion

3 tim

es h

ighe

r (23

4mg

C/m

2 /day

) in

sedi

men

t ass

ocia

ted

with

or

gani

c de

bris

dam

s •

Org

anic

mat

ter c

onte

nt (u

p to

5 c

m d

epth

) sig

nific

antly

hig

her i

n se

dim

ent a

ssoc

iate

d w

ith d

ebris

dam

s (a

vera

ge =

1.7

kg/m

2 ) •

Res

ults

sug

gest

deb

ris d

ams

are

foca

l site

s of

met

abol

ic a

ctiv

ity in

thes

e he

adw

ater

st

ream

s •

Met

abol

ic a

ctiv

ity m

ay b

e un

dere

stim

ated

, as

mea

sure

d to

a d

epth

of o

nly

5cm

Hed

in 1

990

New

Mex

ico

• 0.

17 to

0.1

9 ch

anne

l gra

dien

ts

• R

each

es w

ith w

ood

stor

ed m

ore

orga

nic

mat

ter t

han

reac

hes

with

out w

ood

• D

ecre

ase

in a

vera

ge v

eloc

ity a

nd in

crea

se in

stre

am w

idth

and

dep

th a

fter w

ood

addi

tion

• In

crea

se in

ave

rage

vel

ocity

and

dec

reas

e in

ave

rage

wid

th a

nd d

epth

afte

r woo

d re

mov

al

• Pa

rticl

e re

tent

ion

rate

s in

crea

sed

in s

tream

s w

ith d

ams

or w

ood

• D

ams

stop

ped

parti

cles

, def

lect

ed fl

ow, c

hang

ed w

ater

vel

ocity

, and

alte

red

wid

th

and

dept

h of

the

stre

am

• R

eten

tion

may

dep

end

on c

hann

el s

tabi

lity;

rete

ntio

n de

crea

ses

with

dec

reas

ing

chan

nel s

tabi

lity

Trot

ter 1

990

39

Tabl

e A

-7.

The

effe

ct o

f LW

D o

n aq

uatic

mac

roin

verte

brat

es in

var

ious

geo

grap

hic

loca

tions

and

cha

nnel

net

wor

k po

sitio

ns (a

rrang

ed in

chr

onol

ogic

al o

rder

). LO

CAT

ION

C

HAN

NEL

NET

WO

RK

LO

CAT

ION

R

ESU

LTS

REF

EREN

CES

O

rego

n •

0.01

7 to

0.3

5 ch

anne

l gra

dien

ts

• LW

D lo

adin

g (k

g/m

2 ) dec

reas

ed w

ith in

crea

sing

stre

am o

rder

•

Cad

disf

ly w

ere

mos

t con

spic

uous

and

div

erse

inse

cts

on w

ood

• Tr

icho

pter

a (H

eter

ople

ctro

n ca

lifor

nicu

m) a

nd C

oleo

pter

a (L

ara

avar

a) w

ere

maj

or

inve

rtebr

ates

ass

ocia

ted

with

woo

d su

bstra

tes

• L.

ava

ra s

trong

ly a

ssoc

iate

d w

ith a

mou

nt o

f woo

d av

aila

ble,

irre

spec

tive

of s

tream

si

ze

• Lo

wer

inse

ct b

iom

ass

on w

ood

com

pare

d to

leaf

pac

k •

Two

poss

ible

mec

hani

sms

for w

ood

expl

oita

tion:

1) w

ood

cons

umpt

ion

to o

btai

n di

gest

ible

car

bon

and

nitro

gen

from

mic

robi

al fl

ora,

2) c

ultiv

atio

n an

d re

tent

ion

of g

ut

flora

Ande

rson

et a

l. 19

78

Was

hing

ton

O

rego

n C

alifo

rnia

• N

/A

• C

ontin

uum

of f

auna

l ass

ocia

tion

on L

WD

rang

es fr

om o

blig

ate

rest

rictio

n to

pur

ely

oppo

rtuni

stic

use

•

Sequ

ence

of c

olon

ists

par

alle

l sta

te o

f woo

d de

cay

• H

igh

grad

ient

trib

utar

ies

had

reas

onab

ly g

ood

faun

a of

bor

ers/

goug

ers

and

seve

ral

othe

r gro

ups

•

Whe

re g

radi

ent d

ecre

ased

, silt

atio

n an

d ac

cum

ulat

ion

of o

rgan

ic fi

ne p

artic

les

excl

uded

man

y go

uger

s an

d sc

rape

rs; m

uch

woo

d bu

ried

whe

re D

O w

as lo

w a

nd

fung

al a

ctiv

ity lo

w

• In

mai

n st

em, a

bras

ion

dim

inis

hed

the

role

of i

nver

tebr

ates

in w

ood

degr

adat

ion,

mos

t w

ood

occu

rred

in la

rge

jam

s, d

epos

ited

on s

hore

, and

was

una

vaila

ble

to a

quat

ic

form

s •

Woo

d as

soci

ated

inve

rtebr

ate

com

mun

ity a

nd it

s im

pact

on

woo

d w

as d

epen

dent

on

phys

ical

regi

me

• W

ood

qual

ity a

nd te

xtur

e im

porta

nt in

det

erm

inin

g sp

ecie

s co

loni

zatio

n •

66%

of c

lose

ly a

ssoc

iate

d gr

oups

(obl

igat

e) fo

und

in s

oft,

rotte

n w

ood,

30%

on

groo

ved,

text

ured

woo

d, a

nd <

10%

on

firm

sub

stra

tes

• 20

% o

f fac

ulta

tive

grou

ps fo

und

in s

oft,

rotte

n w

ood,

70%

on

groo

ved,

text

ured

woo

d,

and

10%

on

firm

woo

d •

Func

tiona

l fee

ding

gro

ups

asso

ciat

ed w

ith w

ood

text

ure;

sm

ooth

sur

face

s fo

r at

tach

men

t (fil

tere

rs) a

nd g

raze

rs, s

oft w

ood

form

bor

ers

• Fa

culta

tive

spec

ies

may

use

woo

d m

ore

as re

fuge

from

pre

datio

n th

an fo

r atta

chm

ent

Dud

ley

and

Ande

rson

198

2

Was

hing

ton

O

rego

n C

alifo

rnia

• N

/A

• Fe

edin

g is

mai

n pr

oces

s af

fect

ing

inse

ct m

edia

ted

fragm

enta

tion

• M

ost i

mpo

rtant

inse

cts

in d

egra

datio

n pr

oces

s ar

e bo

rers

and

gou

gers

•

Com

pare

d to

terre

stria

l situ

atio

ns, d

iver

sity

and

den

sity

of b

orer

s an

d go

uger

s is

low

du

e to

lack

of O

2

Pere

ira e

t al.

1982

Sout

heas

t Ala

ska

• 1.

9 m

ave

rage

cha

nnel

wid

th

• 9

km2 c

hann

el w

idth

•

6.5

km lo

ng

• Be

ntho

s de

nsity

, # o

f drif

t inv

erte

brat

es, a

nd #

of i

nver

tebr

ates

eat

en b

y D

olly

Var

den

(S. m

alm

a) d

ecre

ased

afte

r woo

d re

mov

al

• Be

ntho

s bi

omas

s de

crea

sed

by o

ne-th

ird d

urin

g tre

atm

ent p

hase

•

Inve

rtebr

ate

drift

dec

reas

ed s

igni

fican

tly a

fter d

ebris

rem

oval

•

Deb

ris u

sed

as s

ubst

rate

s us

ed b

y al

l ben

thic

taxa

, rem

oval

dire

ctly

redu

ced

abun

danc

e of

ben

thos

Ellio

tt 19

86

40

Tabl

e A

-7.

The

effe

ct o

f LW

D o

n aq

uatic

mac

roin

verte

brat

es in

var

ious

geo

grap

hic

loca

tions

and

cha

nnel

net

wor

k po

sitio

ns (a

rrang

ed in

chr

onol

ogic

al o

rder

). LO

CAT

ION

C

HAN

NEL

NET

WO

RK

LO

CAT

ION

R

ESU

LTS

REF

EREN

CES

Pa

cific

Nor

thw

est,

USA

•

N/A

•

Lite

ratu

re re

view

•

Sequ

ence

of c

olon

ists

par

alle

ls s

tage

of d

ecay

•

New

woo

d is

prim

arily

hab

itat f

or a

lgae

, mic

robe

s, g

raze

rs, a

n co

llect

ors

• C

olon

izat

ion

prom

otes

dec

ompo

sitio

n an

d co

loni

zatio

n by

woo

d sh

redd

ers

(cad

disf

lies,

sto

nefli

es)

• Te

xtur

ed s

urfa

ce p

rovi

des

habi

tat f

or o

rgan

ism

s •

Fung

al g

row

th a

nd e

ffect

of w

ood

bore

rs s

peed

s de

cay

• D

etriv

ores

and

ear

thw

orm

s co

ntin

ue d

ecom

posi

tion

• Th

ree

mos

t im

porta

nt w

ood

proc

esso

rs c

onsu

me

only

2%

of a

ll av

aila

ble

woo

d

Sede

ll et

al.

1988

Arka

nsas

•

1st th

roug

h 3rd

ord

er s

tream

s •

Tric

hopt

era

foun

d in

gre

ater

den

sitie

s in

ben

thic

hab

itats

than

on

LWD

, but

thre

e fa

milie

s fo

und

in g

reat

er d

ensi

ties

on L

WD

•

Woo

d as

soci

ated

Tric

hopt

eran

s fo

und

on h

ighl

y de

caye

d w

ood

and

on L

WD

with

ro

ugh

text

ure,

pro

vidi

ng a

ttach

men

t site

s fo

r net

spi

nnin

g ca

ddis

flies

Philli

ps 1

994

Nor

th C

arol

ina

• 0.

03 to

0.0

6 ch

anne

l gra

dien

ts

• Po

ol fo

rmat

ion

afte

r log

add

ition

•

Inve

rtebr

ate

abun

danc

e an

d bi

omas

s in

crea

sed

sign

ifica

ntly

afte

r log

add

ition

•

Scra

per a

nd c

olle

ctor

/filte

rer a

bund

ance

incr

ease

d af

ter l

og a

dditi

on

• Se

cond

ary

prod

uctio

n of

scr

aper

s an

d fil

tere

rs d

ecre

ased

•

Shift

s in

func

tiona

l gro

up a

bund

ance

and

bio

mas

s ac

cent

uate

the

impo

rtanc

e of

loca

l ab

iotic

fact

ors

in p

atch

stru

ctur

e of

inve

rtebr

ate

com

mun

ities

Wal

lace

et a

l. 19

95

Vict

oria

, Aus

tralia

•

N/A

•

Top

of s

nags

sup

porte

d m

ore

mac

roin

verte

brat

es s

peci

es th

an o

ther

sur

face

s •

Gro

oved

sna

gs s

uppo

rted

sign

ifica

ntly

mor

e m

acro

inve

rtebr

ates

•

Res

ults

of s

tudy

sho

w im

porta

nce

of s

nag

habi

tat i

n st

ream

s an

d riv

ers

• R

esul

ts s

uppo

rt ha

bita

t com

plex

ity h

ypot

hesi

s w

here

spe

cies

rich

ness

incr

ease

s w

ith

habi

tat c

ompl

exity

O'C

onno

r 199

1

New

Yor

k •

30 m

to 5

0 m

cha

nnel

wid

th

• 4th

ord

er s

tream

•

Inve

rtebr

ate

dens

ities

gre

ater

on

woo

d th

an o

n le

aves

•

Filte

rers

dis

prop

ortio

nate

ly c

olon

ized

woo

d an

d co

llect

or/g

athe

rer f

avor

ed le

aves

•

Larg

er p

erce

ntag

e of

scr

aper

s on

leav

es th

an o

n w

ood

• D

ensi

ties

corre

late

d w

ith b

iofil

m d

evel

opm

ent,

whi

ch w

as g

reat

er o

n w

ood

• As

woo

d so

ftene

d th

roug

h ph

ysic

al a

nd b

iolo

gica

l for

ces,

woo

d be

cam

e m

ore

phys

ical

ly c

ompl

ex, o

fferin

g m

ore

mic

roha

bita

ts

• Su

rface

com

plex

ity a

nd s

tabi

lity

cont

ribut

ed to

incr

ease

d ta

xon

dens

ity o

n w

ood

Hax

and

Gol

lada

y 19

93

Nor

ther

n Vi

rgin

ia

• 20

m a

vera

ge c

hann

el w

idth

•

4th o

rder

stre

am

• D

ebris

dam

s st

ored

fine

sed

imen

t and

leaf

y de

bris

•

Abun

danc

e of

Chi

rono

mid

and

Cop

epod

beh

ind

dam

s le

ss li

kely

to d

eclin

e du

ring

flood

s •

Oth

er p

atch

type

s sh

owed

75%

to 9

5% re

duct

ions

in C

hiro

nom

id a

nd C

opep

od

abun

danc

e af

ter f

lood

s •

Deb

ris re

mov

al d

id n

ot im

pede

faun

al re

cove

ry

• R

esul

ts s

ugge

st d

ebris

dam

s as

soci

ated

with

fine

sed

imen

t and

leaf

y de

bris

hav

e po

tent

ial t

o ac

t as

refu

gia

durin

g hi

gh fl

ow

• Pa

tche

s ac

cum

ulat

ing

anim

als

wer

e ch

arac

teriz

ed b

y lo

w w

ater

flux

and

low

bed

flow

, lik

ely

cont

ribut

ing

to re

tent

ion

or p

assi

ve d

epos

ition

of a

nim

als

Palm

er 1

996

Sout

hwes

t Virg

inia

•

0.01

to 0

.06

chan

nel g

radi

ents

•

5 m

ave

rage

cha

nnel

wid

th

• Be

nthi

c m

acro

inve

rtebr

ate

abun

danc

e di

d no

t inc

reas

e af

ter l

og a

dditi

ons

• N

et a

bund

ance

of P

leco

pter

a, C

oleo

pter

a, T

richo

pter

a, a

nd O

ligoc

haet

a de

crea

sed

whi

le E

phem

erop

tera

incr

ease

d w

ith in

crea

se in

poo

l are

a

Hild

erbr

and

et a

l. 19

97

41

Tabl

e A

-8.

The

effe

ct o

f LW

D o

n fis

h (p

redo

min

antly

sal

mon

ids)

hab

itat a

nd p

opul

atio

ns in

var

ious

geo

grap

hic

loca

tions

and

cha

nnel

net

wor

k po

sitio

ns (a

rrang

ed in

chr

onol

ogic

al

orde

r).

LOC

ATIO

N

CH

ANN

EL N

ETW

OR

K P

OSI

TIO

N

RES

ULT

S R

EFER

ENC

ES

Mic

higa

n •

N/A

•

Log

stru

ctur

es in

crea

sed

habi

tat a

rea

and

incr

ease

d Br

ook

trout

(Sal

velin

us fo

ntin

alis

) ab

unda

nce

Tarz

wel

l 193

7

Mon

tana

•

1 m

cha

nnel

wid

th

• Ab

unda

nce

and

biom

ass

of B

rook

trou

t (Sa

lvel

inus

font

inal

is),

rain

bow

trou

t (O

ncor

hync

hus

myk

iss)

, bro

wn

trout

(Sal

mo

trutta

) inc

reas

ed a

fter a

dditi

on o

f arti

ficia

l w

ood

cove

r •

Abun

danc

e an

d bi

omas

s de

crea

sed

afte

r bru

shy

cove

r rem

oval

Bous

su 1

954

Briti

sh C

olum

bia,

C

olum

bia

• 10

km

2 dra

inag

e ar

ea

• Ar

eas

with

logs

, und

ercu

t ban

ks, a

nd d

eep

pool

s fil

led

with

upt

urne

d tre

e ro

ots

(and

ot

her f

ores

t deb

ris) c

onta

ined

coh

o sa

lmon

(Onc

orhy

nchu

s ki

sutc

h)

Tsch

aplin

ski a

nd H

artm

an

1983

Cen

tral I

llinoi

s •

>0.0

01 c

hann

el g

radi

ent

• 3

m to

5 m

cha

nnel

wid

th

• M

ore

spec

ies,

mor

e in

divi

dual

s, a

nd la

rger

fish

cap

ture

d in

reac

hes

with

woo

d •

Hig

her i

nver

tebr

ate

abun

danc

e in

reac

hes

with

woo

d •

Fish

pre

fere

nce

for d

ebris

fille

d re

ache

s re

late

d w

ith fo

od a

vaila

bilit

y •

Woo

d pr

ovid

ed s

ubst

rate

for m

acro

inve

rtebr

ates

•

Woo

d pl

ays

mul

tidim

ensi

onal

role

in s

truct

ure

and

func

tion

of s

tream

eco

syst

ems

Ange

rmei

er a

nd K

arr 1

984

Paci

fic N

orth

wes

t •

N/A

•

Lite

ratu

re re

view

on

hist

oric

al p

roce

sses

and

man

agem

ent o

f LW

D, g

eom

orph

ic

func

tions

, cha

nnel

mor

phol

ogy,

influ

ence

on

fish

habi

tat,

and

man

agem

ent o

f LW

D

inpu

ts in

to s

tream

s

Sede

ll et

al.

1988

Sout

heas

t Ala

ska

• 4th

thr

ough

5th o

rder

stre

ams

• D

olly

Var

den

(Sal

velin

us m

alm

a), s

teel

head

trou

t (O

ncor

hync

hus

myk

iss)

, coh

o sa

lmon

(Onc

orhy

nchu

s ki

sutc

h)

• W

here

deb

ris a

bsen

t, de

nsiti

es o

f coh

o sa

lmon

(Onc

orhy

nchu

s ki

sutc

h) c

onsi

sten

tly

low

er

• Ba

ckw

ater

s an

d si

de c

hann

els

asso

ciat

ed w

ith L

WD

had

hig

her d

ensi

ties

of c

oho

salm

on (O

ncor

hync

hus

kisu

tch)

than

mai

n ch

anne

l are

as w

ith d

ebris

•

Den

sity

of c

oho

salm

on (O

ncor

hync

hus

kisu

tch)

fry

decr

ease

d as

siz

e of

ac

cum

ulat

ion

decr

ease

d •

Mid

-cha

nnel

deb

ris, r

egar

dles

s of

siz

e su

ppor

ted

low

er d

ensi

ties

of c

oho

salm

on

(Onc

orhy

nchu

s ki

sutc

h) th

an ro

otw

ads

alon

g th

e ba

nk

Brya

nt 1

985

Sout

heas

t Ala

ska

• 0.

035

to 0

.05

chan

nel g

radi

ents

•

2 m

ave

rage

cha

nnel

wid

ths

• N

o si

gnifi

cant

diff

eren

ce in

ave

rage

cha

nnel

dep

th o

r wid

th b

etw

een

clea

red

and

uncl

eare

d re

ache

s •

In g

ener

al, u

ncle

aned

stre

am s

ectio

ns c

onta

ined

mor

e fis

h of

all

size

s th

an c

lear

ed

sect

ions

•

Des

pite

use

d of

sel

ectiv

e te

chni

ques

, num

bers

and

pro

duct

ion

of b

oth

coho

sal

mon

(O

ncor

hync

hus

kisu

tch)

and

Dol

ly V

arde

n (S

alve

linus

mal

ma)

wer

e re

duce

d by

re

mov

al o

f LW

D

Dol

loff

1986

Sout

heas

t Ala

ska

• 1.

9 m

ave

rage

cha

nnel

wid

th

• 9

km2 c

hann

el w

idth

•

6.5

km lo

ng

• R

emov

al o

f log

ging

deb

ris re

duce

d w

ette

d w

idth

and

are

as fr

om 8

74 m

2 an

d 72

6 m

2 •

Dol

ly V

arde

n (S

alve

linus

mal

ma)

pop

ulat

ion

decr

ease

d fro

m 7

48 to

100

, tw

o ye

ars

afte

r deb

ris re

mov

al

• D

ecre

ase

in fi

sh s

ize

afte

r cle

aran

ce

• Be

fore

deb

ris re

mov

al, f

ish

occu

pied

mid

cha

nnel

poo

ls a

nd ta

ngle

s of

deb

ris

• Af

ter r

emov

al, f

ish

occu

pied

bac

kwat

ers

and

stre

am m

argi

ns

• R

emov

al o

f log

ging

deb

ris c

hang

ed m

orph

olog

y, b

riefly

redu

ced

food

sup

ply

and

elim

inat

ed m

ost c

over

, red

ucin

g #

and

mea

n le

ngth

of f

ish,

leav

ing

smal

l fis

h su

scep

tible

to d

ispl

acem

ent

• D

ecre

ase

in b

enth

os a

bund

ance

, red

uced

am

ount

of f

ood

eate

n by

Dol

ly V

arde

n (S

alve

linus

mal

ma)

and

may

hav

e co

ntrib

uted

to n

umer

ical

dec

line

Ellio

tt 19

86

42

Tabl

e A

-8.

The

effe

ct o

f LW

D o

n fis

h (p

redo

min

antly

sal

mon

ids)

hab

itat a

nd p

opul

atio

ns in

var

ious

geo

grap

hic

loca

tions

and

cha

nnel

net

wor

k po

sitio

ns (a

rrang

ed in

chr

onol

ogic

al

orde

r).

LOC

ATIO

N

CH

ANN

EL N

ETW

OR

K P

OSI

TIO

N

RES

ULT

S R

EFER

ENC

ES

Sout

heas

tern

Al

aska

•

0.02

to 0

.03

chan

nel g

radi

ents

•

2nd th

roug

h 4th

ord

er s

tream

s •

Mos

t fis

h, re

gard

less

of s

peci

es o

r age

, wer

e in

poo

ls

• Am

ount

s of

LW

D a

nd c

obbl

e si

gnifi

cant

ly in

crea

sed

in o

ccup

ied

pool

than

in

unoc

cupi

ed p

ools

•

LWD

cau

sed

73%

of a

ll po

ols

• Ju

veni

le s

alm

onid

s (in

win

ter)

pref

erre

d po

ol w

ith c

over

(upt

urne

d ro

ots,

acc

umul

atio

n of

logs

, cob

ble

subs

trate

) •

Amou

nt o

f pre

ferre

d w

inte

r hab

itat d

epen

ded

on a

mou

nt o

f LW

D

• C

lear

cut r

each

es h

ad le

ss L

WD

and

less

poo

l are

a •

Buffe

r stri

ps p

rote

ct h

abita

t by

prov

idin

g LW

D

Hei

fietz

et a

l. 19

86

Ore

gon

• 9.

3 km

2 dra

inag

e ar

ea

• 0.

03 a

vera

ge c

hann

el g

radi

ent

• Fe

wer

pie

ces

of L

WD

in lo

gged

sec

tion

(6 p

iece

s pe

r 100

m) c

ompa

red

to in

un

dist

urbe

d se

ctio

n (4

6 pi

eces

per

100

m)

• LW

D m

ost i

mpo

rtant

fact

or c

ontro

lling

diffe

renc

es in

cha

nnel

mor

phol

ogy:

LW

D

crea

ted

stai

r ste

p m

orph

olog

y, s

econ

dary

cha

nnel

s, a

nd m

eand

ers

• U

ndis

turb

ed s

ectio

ns, L

WD

impr

oved

poo

l qua

lity

and

# of

poo

ls

• Af

ter L

WD

add

ition

s, g

rave

l abu

ndan

ce in

crea

sed

by 2

33%

, gra

vel a

rea

incr

ease

d 25

fo

ld

• 3

times

mor

e co

ho (O

ncor

hync

hus

kisu

tch)

(0.8

8/m

2 ) and

ste

elhe

ad (O

ncor

hync

hus

myk

iss)

(0.5

9/m

2 ) in

undi

stur

bed

sect

ions

•

Posi

tive

corre

latio

n be

twee

n fis

h po

pula

tions

(spa

wni

ng a

nd re

arin

g ha

bita

t) an

d LW

D

Hou

se a

nd B

oehn

e 19

86

Alas

ka

• 0.

01 to

0.0

9 ch

anne

l gra

dien

ts

• 2

m to

6 m

cha

nnel

wid

ths

• 0.

5 km

2 to 2

km

2 dra

inag

e ar

ea

• D

ebris

dam

freq

uenc

y co

rrela

ted

with

deb

ris lo

adin

g •

Deb

ris d

ams

form

ed 8

6% o

f poo

ls in

cle

arcu

t rea

ches

and

47%

of p

ools

in u

nfor

este

d re

ache

s •

Deb

ris d

ams

effe

ctiv

e at

mai

ntai

ning

dep

th in

low

flow

(res

idua

l dep

th)

Lisl

e 19

86a

Was

hing

ton

• 28

0 km

tota

l stre

am le

ngth

for

basi

n •

Salm

onid

dis

appe

aran

ce a

fter M

ount

St H

elen

s er

uptio

n •

LWD

rela

ted

habi

tats

mad

e up

1%

of t

otal

in n

ew c

hann

els,

6%

to 1

2% in

old

affe

cted

ch

anne

ls, a

nd 1

2% to

16%

in o

ld u

naffe

cted

cha

nnel

s •

Stre

ams

with

littl

e am

ount

of L

WD

, als

o ha

d de

crea

sed

amou

nt o

f poo

l are

a •

Deb

ris p

rodu

ces

cove

r but

als

o re

duce

s w

ater

vel

ocity

Mar

tin e

t al.

1986

Briti

sh C

olum

bia,

C

anad

a •

Artif

icia

l cha

nnel

s 4.

9 m

(l) b

y 0.

9 m

(w) b

y 0.

6 m

(d)

• C

oho

salm

on (O

ncor

hync

hus

kisu

tch)

abu

ndan

ce in

crea

sed

as c

over

com

plex

ity

incr

ease

d •

Win

ter c

over

for c

oho

(Onc

orhy

nchu

s ki

sutc

h) c

ombi

ned

low

vel

ocity

, sha

de, a

nd 3

di

men

sion

al c

ompl

exity

McM

ahon

and

Har

tman

19

89

Sout

heas

tern

Al

aska

•

0.10

to 0

.30

chan

nel g

radi

ent

• 5

m to

7 m

cha

nnel

wid

th

• W

inte

r coh

o sa

lmon

(Onc

orhy

nchu

s ki

sutc

h) fr

y de

nsity

bes

t mod

eled

by

sum

mer

pe

riphy

ton

biom

ass

and

volu

me

of in

stre

am d

ebris

•

Parr

abun

dant

whe

re d

ebris

was

abu

ndan

t •

In b

oth

sum

mer

and

win

ter,

dens

ity o

f Dol

ly V

arde

n (S

alve

linus

mal

ma)

par

r dire

ctly

re

late

d to

LW

D v

olum

e •

Cov

er (L

WD

and

und

ercu

t ban

ks) m

ore

impo

rtant

for a

ll pa

rr in

win

ter

Mur

phy

et a

l. 19

86

Briti

sh C

olum

bia,

C

anad

a •

10 k

m2 d

rain

age

area

•

Red

uctio

n in

vol

ume

and

stab

ility

of L

WD

2 y

ears

afte

r log

ging

•

Afte

r log

ging

, buf

ferin

g of

stre

am e

nerg

y re

duce

d, c

ohes

ion

of s

tream

bank

s re

duce

d •

Incr

ease

in s

and

led

to re

duce

d gr

avel

qua

lity

and

decr

ease

d si

tes

avai

labl

e fo

r chu

m

(Onc

orhy

nchu

s ke

ta) a

nd c

oho

salm

on (O

ncor

hync

hus

kisu

tch)

spa

wni

ng

• Be

tter g

row

th b

y yo

ung

coho

sal

mon

(Onc

orhy

nchu

s ki

sutc

h) a

nd c

utth

roat

trou

t (O

ncor

hync

hus

clar

ki)

• LW

D p

rovi

ded

over

win

ter c

over

•

Dec

reas

ed d

ensi

ties

of c

oho

(Onc

orhy

nchu

s ki

sutc

h) le

d to

fast

er g

row

th ra

te, l

arge

r si

ze, a

nd h

ighe

r win

ter s

urvi

val

Har

tman

et a

l. 19

87

43

Tabl

e A

-8.

The

effe

ct o

f LW

D o

n fis

h (p

redo

min

antly

sal

mon

ids)

hab

itat a

nd p

opul

atio

ns in

var

ious

geo

grap

hic

loca

tions

and

cha

nnel

net

wor

k po

sitio

ns (a

rrang

ed in

chr

onol

ogic

al

orde

r).

LOC

ATIO

N

CH

ANN

EL N

ETW

OR

K P

OSI

TIO

N

RES

ULT

S R

EFER

ENC

ES

Ore

gon

• 0.

10 a

vera

ge c

hann

el g

radi

ent

• 3.

2 av

erag

e ch

anne

l wid

th

• 5.

4 km

2 dra

inag

e ar

ea

• 3rd

ord

er s

tream

• 2.

4 fo

ld in

crea

se in

late

ral h

abita

t gav

e 2.

2 fo

ld in

crea

se in

num

ber o

f age

-0 c

utth

roat

tro

ut (O

ncor

hync

hus

clar

ki)

• 86

% re

duct

ion

in la

tera

l hab

itat g

ave

83%

redu

ctio

n in

age

0 c

utth

roat

trou

t (O

ncor

hync

hus

clar

ki)

• Si

gnifi

cant

ly h

ighe

r tro

ut p

rodu

ctio

n in

late

ral h

abita

t of m

anip

ulat

ed s

ectio

n th

an in

co

ntro

l sec

tion

and

LWD

redu

ced

sect

ions

(95%

and

24%

) •

Func

tiona

l lat

eral

hab

itat c

omes

from

LW

D in

tera

ctio

n w

ith s

tream

bed

Moo

re a

nd G

rego

ry 1

988

Sout

heas

t Ala

ska

• <2

m w

ide

stre

ams

• Lo

gs, b

ranc

hes,

und

ercu

t ban

ks, d

ense

ove

rhea

d ve

geta

tion,

and

larg

e bo

ulde

rs w

ere

prim

ary

type

s of

cov

er u

sed

by c

oho

salm

on (O

ncor

hync

hus

kisu

tch)

and

Dol

ly

Vard

en (S

alve

linus

mal

ma)

of a

ll ag

e cl

asse

s

Dol

loff

and

Ree

ves

1990

Briti

sh C

olum

bia,

C

anad

a •

108

km2 d

rain

age

area

•

Coh

o sa

lmon

(Onc

orhy

nchu

s ki

sutc

h) fr

y (9

9%) a

nd s

teel

head

(Onc

orhy

nchu

s m

ykis

s) p

arr (

83%

) occ

upie

d po

sitio

ns n

ear m

id-c

hann

el a

rtific

ial r

ootw

ads

in a

ll flo

w

cond

ition

s (d

roug

ht, n

orm

al, a

nd fl

ood)

•

Coh

o pr

efer

red

near

sho

re p

iece

s, w

hile

ste

elhe

ad p

refe

rred

mor

e di

stan

t site

s •

Youn

g fis

h pr

efer

red

slow

wat

er (8

0%) a

nd lo

w li

ght (

20%

) env

ironm

ents

for

prot

ectio

n fro

m h

igh

curre

nts

and

pred

ator

s

Shirv

ell 1

990

Briti

sh C

olum

bia,

C

anad

a •

0.00

5 to

0.0

4 ch

anne

l gra

dien

ts

• 2

m to

5 m

ave

rage

ban

kful

l wid

th

• 1.

03 to

1.3

7 si

nuos

ity

• Se

ctio

ns w

ith L

WD

rem

oved

wer

e le

ss s

inuo

us, w

ider

and

sha

llow

er a

nd h

ad le

ss

cove

r for

fish

than

una

ltere

d se

ctio

ns

• Po

ols

wer

e 9%

to 2

5% o

f tot

al v

olum

e at

bas

eflo

w in

rem

oval

sec

tions

, 61%

to 6

9% o

f to

tal v

olum

e in

una

ltere

d se

ctio

ns

• N

umbe

r of p

iece

s of

LW

D d

irect

ly in

fluen

cing

cha

nnel

mor

phol

ogy

was

gre

ater

in

unal

tere

d se

ctio

ns

• LW

D fo

rmed

72%

to 8

0% o

f poo

ls in

all

sect

ions

•

Dep

ths

of p

ools

gre

ater

in u

nalte

red

sect

ions

•

Afte

r sur

face

flow

cea

sed

in s

umm

er, 9

6% o

f poo

l vol

ume

asso

ciat

ed w

ith L

WD

•

In h

ighe

r gra

dien

t rea

ches

(0.0

15 to

0.0

4 gr

adie

nt) m

ost w

ood

orie

nted

per

pend

icul

ar

to fl

ow

• In

low

er g

radi

ent r

each

es (>

0.01

5) m

uch

high

er p

erce

ntag

e (4

0%) o

f dia

gona

l or

ient

atio

n •

Stan

ding

cro

p of

age

1+

and

olde

r sal

mon

ids

was

muc

h lo

wer

in c

lear

ed s

ectio

ns th

an

in u

nalte

red

sect

ions

•

Biom

ass

of a

ge 1

+ an

d ol

der f

ish

corre

late

d w

ith p

ool v

olum

e, s

ectio

n vo

lum

e, m

ean

dept

h, a

nd le

ngth

of o

verh

ead

cove

r •

Fish

bio

mas

s st

rong

ly c

orre

late

d w

ith p

ool v

olum

e an

d de

pth

• U

ncle

ared

sec

tions

had

five

tim

es th

e cu

rrent

sta

ndin

g cr

op in

cle

ared

sec

tions

Faus

ch a

nd N

orth

cote

199

2

Briti

sh C

olum

bia,

C

anad

a •

Buffe

red,

cle

arcu

t, es

tuar

ine

reac

hes

• LW

D in

fluen

ced

abun

danc

e of

coh

o sm

olts

(Onc

orhy

nchu

s ki

sutc

h) in

stre

am a

nd

estu

ary

• C

oho

smol

t (O

ncor

hync

hus

kisu

tch)

dis

tribu

tion

high

ly c

lum

ped

arou

nd d

ebris

; 80%

w

ithin

5 m

of d

ebris

; 95%

with

in 1

m

• Su

ppor

ts th

e ne

ed to

reta

in L

WD

for s

mol

t hab

itat i

n st

ream

s an

d es

tuar

ies

McM

ahon

and

Hol

tby

1992

Ore

gon

• 27

km

2 dra

inag

e ar

ea

• 5th

ord

er s

tream

•

Stre

am s

urfa

ce a

rea

and

wat

er v

olum

e in

crea

sed

74%

and

168

% a

fter l

og a

dditi

on

• Su

rface

are

a of

poo

l and

off

chan

nel h

abita

t inc

reas

ed in

trea

ted

sect

ions

•

Spaw

ning

act

ivity

four

tim

es h

ighe

r in

trea

ted

sect

ion

Cris

pin

et a

l. 19

93

Albe

rta, C

anad

a •

23 m

cha

nnel

wid

th

• 0.

5 m

cha

nnel

dep

th

• 57

1 km

2 dra

inag

e ar

ea

• 4th

ord

er s

tream

• Fr

y de

nsiti

es h

ighe

r in

sect

ions

with

fine

woo

dy d

ebris

(FW

D) a

dditi

ons

• FW

D m

ay h

ave

prov

ided

stru

ctur

ally

com

plex

hab

itat,

used

as

refu

ge fr

om p

reda

tors

an

d as

site

s fro

m w

hich

fora

ging

fora

ys b

egan

Cul

p et

al.

1996

44

Tabl

e A

-8.

The

effe

ct o

f LW

D o

n fis

h (p

redo

min

antly

sal

mon

ids)

hab

itat a

nd p

opul

atio

ns in

var

ious

geo

grap

hic

loca

tions

and

cha

nnel

net

wor

k po

sitio

ns (a

rrang

ed in

chr

onol

ogic

al

orde

r).

LOC

ATIO

N

CH

ANN

EL N

ETW

OR

K P

OSI

TIO

N

RES

ULT

S R

EFER

ENC

ES

Mar

ylan

d •

Sube

stua

rine

envi

ronm

ent

• D

ensi

ties

of e

pibe

nthi

c fis

h in

crea

sed

sign

ifica

ntly

in s

ites

with

LW

D

• LW

D p

rovi

ded

refu

ge fr

om p

reda

tion

for e

pibe

nthi

c fis

h an

d in

verte

brat

es

Ever

ett a

nd R

uiz

1993

Nor

th C

arol

ina

• 0.

045

to 0

.06

chan

nel g

radi

ents

•

4 m

to 5

m c

hann

el w

idth

•

7 km

2 to 2

1 km

2 wat

ersh

ed a

rea

• 1st

thro

ugh

3rd o

rder

stre

am

• H

ighe

r abu

ndan

ce o

f LW

D in

und

istu

rbed

stre

ams

• U

ndis

turb

ed re

ache

s ha

d 32

% to

46%

of h

abita

t uni

ts a

ssoc

iate

d w

ith L

WD

•

Broo

k tro

ut (S

alve

linus

font

inal

is),

rain

bow

trou

t (O

ncor

hync

hus

myk

iss)

, bro

wn

trout

(S

alm

o tru

tta) h

ad h

ighe

r den

sity

and

bio

mas

s in

und

istu

rbed

reac

hes

with

LW

D

• La

rger

ave

rage

siz

e of

LW

D in

und

istu

rbed

reac

hes

• M

ore

pool

s an

d rif

fles

of s

mal

ler s

ize

in u

ndis

turb

ed re

ache

s

Fleb

be a

nd D

ollo

ff 19

95

Nor

ther

n C

olor

ado

• 0.

01 to

0.0

24 c

hann

el g

radi

ents

•

3 m

to 6

m c

hann

el w

idth

s •

Pool

vol

ume

incr

ease

d af

ter d

rop

stru

ctur

es in

stal

led

• Am

ount

of c

over

incr

ease

d si

gnifi

cant

ly a

fter a

dditi

on o

f log

dro

p st

ruct

ures

•

Trea

tmen

t sec

tions

wer

e de

eper

and

had

low

er v

eloc

ity th

an c

ontro

l sec

tions

•

Abun

danc

e an

d bi

omas

s of

age

2 a

nd o

lder

incr

ease

d si

gnifi

cant

ly a

fter i

nsta

llatio

n of

dr

op s

truct

ures

•

Log

drop

stru

ctur

es in

crea

sed

abun

danc

e of

adu

lt br

ook

trout

(Sal

velin

us fo

ntin

alis

) an

d po

ssib

ly s

urvi

val,

alth

ough

dat

a no

t con

clus

ive

Rile

y an

d Fa

usch

199

5

Was

hing

ton

• 2.

4 m

cha

nnel

wid

th (a

rtific

ial

chan

nels

) •

Foun

d no

evi

denc

e th

at s

umm

er p

opul

atio

ns w

ere

attra

cted

to b

rush

y de

bris

, nor

did

th

e pr

esen

ce o

f bru

sh in

fluen

ce s

urvi

val o

r gro

wth

, com

para

ble

to fr

ee ra

ngin

g co

ho

salm

on (O

ncor

hync

hus

kisu

tch)

•

Affin

ity o

f coh

o sa

lmon

(Onc

orhy

nchu

s ki

sutc

h) fo

r woo

dy d

ebris

incr

ease

s in

win

ter

Spal

ding

et a

l. 19

95

Nor

ther

n C

olor

ado

• 0.

01 to

0.0

3 ch

anne

l gra

dien

ts

• 3

m to

6 m

ave

rage

cha

nnel

wid

th

• Po

ol v

olum

e an

d to

tal c

over

incr

ease

d af

ter L

WD

add

ition

•

Abun

danc

e an

d bi

omas

s of

adu

lt of

bro

ok tr

out (

Sal

velin

us fo

ntin

alis

), br

own

trout

(S

alm

o tru

tta),

rain

bow

trou

t (O

ncor

hync

hus

myk

iss)

incr

ease

d in

man

ipul

ated

re

ache

s •

Incr

ease

d ab

unda

nce

and

biom

ass

resu

lted

from

imm

igra

tion

from

sur

roun

ding

re

ache

s •

Log

wei

rs in

crea

sed

abun

danc

e an

d bi

omas

s of

fish

in th

e ab

senc

e of

any

m

anag

emen

t lim

itatio

ns

Gow

an a

nd F

ausc

h 19

96

Was

hing

ton

• 0.

002

to 0

.015

cha

nnel

gra

dien

ts

• 38

km

2 dra

inag

e ar

ea

• Po

ol s

paci

ng s

igni

fican

tly c

orre

late

d w

ith L

WD

abu

ndan

ce

• Su

rviv

al o

f coh

o sa

lmon

(Onc

orhy

nchu

s ki

sutc

h) c

orre

late

d w

ith L

WD

abu

ndan

ce a

nd

volu

me

• Po

sitiv

e re

latio

nshi

p be

twee

n ha

bita

t com

plex

ity (a

bund

ance

of L

WD

and

poo

l sp

acin

g) a

t the

end

of t

he s

umm

er a

nd o

verw

inte

r sur

viva

l

Qui

nn a

nd P

eter

son

1996

Wyo

min

g •

2 m

to 5

m c

hann

el w

idth

•

39%

of p

ools

form

ed b

y LW

D, 6

4% o

f cut

thro

at tr

out (

Onc

orhy

nchu

s cl

arki

) rel

ocat

ed

to th

ese

pool

s Yo

ung

1996

Was

hing

ton

• 0.

02 c

hann

el g

radi

ent

• 10

.0 m

cha

nnel

wid

th

• 3rd

ord

er s

tream

• Po

ol a

rea

incr

ease

d af

ter L

WD

add

ition

•

Abun

danc

e of

coh

o sa

lmon

(Onc

orhy

nchu

s ki

sutc

h) d

urin

g sp

ring

and

win

ter s

how

ed

no re

spon

se to

enh

ance

men

t •

Win

ter a

bund

ance

of c

oho

(Onc

orhy

nchu

s ki

sutc

h) in

crea

sed

sign

ifica

ntly

afte

r en

hanc

emen

t

Ced

erho

lm e

t al.

1997

Nor

ther

n C

alifo

rnia

•

0.02

2 ch

anne

l gra

dien

t •

27.5

km

2 dra

inag

e ar

ea

• 3rd

ord

er s

tream

• Po

ols

form

ed b

y LW

D s

cour

con

tain

ed m

ore

cutth

roat

trou

t (O

ncor

hync

hus

clar

ki)

(0.2

5/m

) tha

n po

ols

form

ed b

y bo

ulde

r sco

ur (0

.16/

m)

• R

eten

tion

of c

utth

roat

trou

t (O

ncor

hync

hus

clar

ki) g

reat

est i

n co

mpl

ex (L

WD

) poo

ls

durin

g hi

gher

flow

s •

Pres

ence

of L

WD

had

littl

e of

no

effe

ct o

n tro

ut m

ovem

ent o

r gro

wth

dur

ing

low

or

mod

erat

e di

scha

rge,

food

ava

ilabi

lity

was

the

mai

n fa

ctor

con

trollin

g m

ovem

ent a

nd

grow

th

Har

vey

1998

45

Tabl

e A

-8.

The

effe

ct o

f LW

D o

n fis

h (p

redo

min

antly

sal

mon

ids)

hab

itat a

nd p

opul

atio

ns in

var

ious

geo

grap

hic

loca

tions

and

cha

nnel

net

wor

k po

sitio

ns (a

rrang

ed in

chr

onol

ogic

al

orde

r).

LOC

ATIO

N

CH

ANN

EL N

ETW

OR

K P

OSI

TIO

N

RES

ULT

S R

EFER

ENC

ES

Nor

th C

arol

ina

• 0.

08 to

0.1

6 ch

anne

l gra

dien

ts

• 3.

5 m

ave

rage

cha

nnel

wid

th

• 11

.3 k

m2 d

rain

age

area

•

3rd o

rder

stre

am

• Lo

wer

deb

ris lo

adin

g (e

spec

ially

of l

arge

r siz

es) i

n di

stur

bed

reac

hes

• H

igh

prop

ortio

n of

hab

itat u

nits

form

ed w

ithou

t LW

D in

dis

turb

ed re

ache

s •

Low

pro

porti

on o

f hab

itat f

orm

ed b

y 2

or m

ore

piec

es o

f woo

d in

dis

turb

ed re

ache

s •

71%

of p

ools

and

riffl

es o

ccup

ied

by tr

out i

n di

stur

bed

reac

hes

vers

us 9

0% in

re

fere

nce

reac

hes

• H

abita

t uni

ts w

ith L

WD

sup

porte

d si

gnifi

cant

ly m

ore

trout

than

uni

ts w

ithou

t

Fleb

be 1

999

46

Tabl

e A

-9.

The

effe

ct o

f LW

D o

n rip

aria

n ha

bita

t in

vario

us g

eogr

aphi

c lo

catio

ns a

nd c

hann

el n

etw

ork

posi

tions

(arra

nged

in c

hron

olog

ical

ord

er).

LOC

ATIO

N

CH

ANN

EL N

ETW

OR

K L

OC

ATIO

N

RES

ULT

S R

EFER

ENC

ES

Paci

fic N

orth

wes

t, U

SA

• N

/A

• LW

D p

rote

cts

ripar

ian

site

s w

here

ald

er a

nd o

ther

spe

cies

bec

ome

esta

blis

hed

• Fa

llen

trees

pro

tect

veg

etat

ion

in e

xpos

ed c

hann

el b

ars

• D

owne

d tre

es p

rote

ct v

eget

atio

n an

d cr

eate

site

s w

here

sed

imen

t and

org

anic

m

ater

ial d

epos

it •

Dep

ositi

on o

f sed

imen

t and

org

anic

mat

ter b

oost

s so

il de

velo

pmen

t •

LWD

hel

ps s

tand

s re

ach

mat

ure

stag

e an

d be

tter w

ithst

and

flood

s •

Low

gra

dien

t riv

er s

egm

ents

hav

e ab

unda

nt L

WD

form

ing

depo

sitio

nal a

reas

that

pr

ovid

e si

tes

for f

utur

e flo

odpl

ains

to c

olon

ize

Sede

ll et

al.

1988

Paci

fic N

orth

wes

t, U

SA

• <0

.05

to >

0.18

cha

nnel

gra

dien

ts

• C

hang

e in

siz

e, fr

eque

ncy,

loca

tion,

and

long

evity

of d

epos

ition

al s

ites

form

ed b

y LW

D in

fluen

ces

succ

essi

onal

dyn

amic

s of

ripa

rian

vege

tatio

n •

Esta

blis

hmen

t of f

ores

ted

flood

plai

ns a

ssoc

iate

d w

ith L

WD

par

alle

l or o

bliq

ue to

ch

anne

l •

LWD

trap

s co

lluvi

um a

nd a

lluvi

al s

edim

ent

• LW

D fu

nctio

ns a

s nu

rse

logs

to p

rovi

de n

utrie

nts

and

as e

leva

ted

site

to m

inim

ize

com

petit

ion

betw

een

seed

lings

and

fore

st fl

oor v

eget

atio

n •

Nur

se lo

gs a

lso

act a

s re

fuge

of y

oung

ripa

rian

spec

ies

• As

cha

nnel

con

finem

ent i

ncre

ases

, the

influ

ence

of L

WD

on

ripar

ian

fore

st d

istri

butio

n de

crea

ses

due

to d

ecre

ase

in a

rea

of a

ctiv

e flo

odpl

ain

• Fo

rest

ed fl

oodp

lain

s co

mm

only

form

beh

ind

dow

nstre

am o

f LW

D o

n se

dim

ent

depo

sits

•

Vege

tatio

n co

loni

zes

both

LW

D a

nd lo

w v

eloc

ity z

ones

•

The

age

stru

ctur

e of

fore

sted

floo

dpla

ins

refle

cts

hist

ory

of L

WD

dep

ositi

on a

nd fl

uvia

l di

stur

banc

e •

Accu

mul

atio

ns o

f LW

D fo

rm d

epos

ition

al s

ites

whe

re v

eget

atio

n co

loni

zes

and

beco

mes

est

ablis

hed

in a

n ot

herw

ise

unho

spita

ble

fluvi

al e

nviro

nmen

t

Feth

erst

on e

t al.

1995

Nor

thw

est

Was

hing

ton

• 0.

005

to 0

.01

chan

nel g

radi

ents

•

30 m

to 8

0 m

ban

kful

l wid

th

• 75

km

2 to 2

25 k

m2 d

rain

age

area

s

• R

ipar

ian

fore

sts

alon

g la

rge

allu

vial

cha

nnel

s ge

nera

lly c

hara

cter

ized

as

youn

g an

d ho

mog

enou

s tre

e co

mm

uniti

es th

at re

flect

dis

turb

ance

•

But,

obse

rved

div

erse

ripa

rian

fore

st s

truct

ure

and

anom

alou

s gr

owth

pat

ches

•

Thre

e fa

ctor

s fa

cilit

ate

ripar

ian

fore

st d

evel

opm

ent d

owns

tream

of L

WD

bar

are

a ja

ms:

1) l

ocal

flow

dec

eler

atio

n an

d de

crea

sed

basa

l she

ar s

tress

, 2) s

edim

ent

depo

sitio

n, 3

) an

abun

dant

acc

umul

atio

n of

org

anic

mat

ter o

n th

e su

rface

•

Obs

erva

tion

of o

ld g

row

th ri

paria

n pa

tche

s w

ithin

act

ive

chan

nel m

igra

tion

sugg

ests

so

me

stru

ctur

es re

mai

n st

able

des

pite

repe

ated

inte

grat

ion

into

the

activ

e ch

anne

l •

LWD

pro

vide

long

-term

refu

gia

for f

lood

plai

n rip

aria

n co

mm

uniti

es fo

rmin

g an

omal

ous

old-

grow

th ri

paria

n fo

rest

pat

ches

in a

lluvi

al te

rrain

cha

ract

eriz

ed b

y fre

quen

t di

stur

banc

e

Abbe

and

Mon

tgom

ery

1996

47

Tabl

e A

-10.

The

effe

ct o

f tim

ber h

arve

st o

n LW

D c

hara

cter

istic

s in

var

ious

geo

grap

hic

loca

tions

and

cha

nnel

net

wor

k po

sitio

ns (a

rrang

ed in

chr

onol

ogic

al o

rder

). LO

CAT

ION

C

HAN

NEL

NET

WO

RK

PO

SITI

ON

R

ESU

LTS

REF

EREN

CES

W

este

rn O

rego

n •

0.06

to 0

.65

chan

nel g

radi

ents

•

1 m

to 1

8 m

cha

nnel

wid

ths

• 0.

02 k

m2 to

30

km2 d

rain

age

area

s

• Lo

ggin

g m

etho

ds n

ear a

nd a

cros

s dr

aina

ges

can

chan

ge th

e am

ount

and

con

ditio

n of

in

-cha

nnel

deb

ris

• C

onve

ntio

nal t

ree

fallin

g ad

ded

the

mos

t am

ount

of l

oggi

ng m

ater

ial t

o th

e st

ream

, fo

llow

ed b

y ca

ble

assi

st d

irect

iona

l fal

ling,

and

con

vent

iona

l fal

ling

with

buf

fers

•

Attit

ude

of lo

ggin

g cr

ew p

laye

d la

rges

t par

t in

dete

rmin

ing

amou

nt o

f sla

sh re

mai

ning

Froe

hlic

h 19

73

Wes

tern

Ore

gon

• 0.

10 to

0.3

0 ch

anne

l gra

dien

ts

• C

lear

cut

ting

and

stre

am c

lean

ing

may

redu

ce a

bund

ance

of l

arge

sta

ble

piec

es,

incr

ease

con

cent

ratio

n of

sm

all u

nsta

ble

piec

es th

roug

h in

put a

nd m

obiliz

atio

n of

in

stre

am L

WD

, red

ucin

g ov

eral

l inp

ut

Swan

son

et a

l. 19

76

Wes

tern

Ore

gon

• 0.

02 to

0.5

0 ch

anne

l gra

dien

ts

• Ti

mbe

r har

vest

may

incr

ease

del

iver

y of

LW

D to

stre

am c

hann

el b

y in

crea

sing

pr

obab

ility

of d

ebris

ava

lanc

hes

• Ti

mbe

r har

vest

may

alte

r siz

e an

d di

strib

utio

n on

deb

ris in

stre

ams

• In

crea

sed

occu

rrenc

e of

deb

ris to

rrent

s in

cle

ar c

ut a

reas

•

Rem

oval

of s

tabl

e la

rge

debr

is m

ay m

obiliz

e sm

alle

r log

s an

d in

itiat

e do

wns

tream

to

rrent

s •

Man

agem

ent a

ctiv

ities

redu

ce L

WD

by

thin

ning

and

har

vest

ing,

whi

ch re

duce

s lo

ng

term

deb

ris lo

adin

g

Swan

son

and

Lien

kaem

per

1978

Sout

heas

tern

Al

aska

•

42 k

m2 d

rain

age

area

•

Tim

ber h

arve

st im

pose

d ch

ange

s in

stre

am c

hann

el re

late

d to

acc

umul

atio

ns o

f LW

D

• N

atur

al a

ccum

ulat

ions

app

eare

d st

able

com

pare

d to

logg

ing

debr

is

• Ab

senc

e of

old

gro

wth

fore

st a

long

stre

amba

nk e

limin

ated

LW

D re

crui

tmen

t •

As n

atur

al a

ccum

ulat

ion

deca

ys, p

ools

may

be

repl

aced

by

riffle

s •

Nat

ural

deb

ris a

ccum

ulat

ions

are

affe

cted

by

larg

e flo

atab

le d

ebris

Brya

nt 1

980

Nor

ther

n C

alifo

rnia

•

0.01

to 0

.40

chan

nel g

radi

ents

•

1.0

km2 to

27

km2 d

rain

age

area

s •

2nd th

roug

h 4th

ord

er s

tream

s

• LW

D in

trodu

ced

durin

g h

arve

st a

re s

mal

ler a

nd m

ore

num

erou

s th

an n

atur

al in

put

• Ti

mbe

r har

vest

der

ived

pie

ces

are

mor

e m

obile

and

enc

oura

ge fa

ilure

of d

owns

tream

da

ms

• Su

bseq

uent

nat

ural

inpu

ts a

re s

mal

ler t

han

old

grow

th s

peci

es

• 2nd

gro

wth

tree

s le

ss ro

t res

ista

nt th

an o

ld g

row

th re

dwoo

ds

• Lo

adin

g in

dis

turb

ed s

tream

s is

low

er a

nd c

ompr

ised

of h

emlo

ck a

nd ta

noak

•

Dis

turb

ed w

ater

shed

s ha

d gr

eate

r am

ount

s of

deb

ris s

tore

d se

dim

ent,

refle

ctin

g th

e gr

eate

r ext

ent t

o w

hich

sto

rage

com

partm

ents

are

fille

d in

logg

ed re

ache

s

Kelle

r and

Mac

Don

ald

1983

Sout

heas

tern

Al

aska

•

0.01

to 0

.09

chan

nel g

radi

ents

•

2 m

to 6

m c

hann

el w

idth

•

0.5

to 2

.0 k

m2 d

rain

age

area

s

• D

ebris

dam

s m

ore

frequ

ent i

n cl

ear c

ut s

tream

s •

Tota

l res

idua

l poo

l len

gth

and

dept

h gr

eate

r in

clea

r cut

stre

ams

• LW

D s

tore

d m

ore

sedi

men

t in

clea

r cut

reac

hes

• R

esul

ts s

ugge

st im

porta

nce

of lo

ggin

g de

bris

and

exi

stin

g de

bris

from

cle

arcu

ts

• LW

D le

ft af

ter l

oggi

ng c

onst

itute

s av

aila

ble

supp

ly o

f woo

d cr

eate

d st

ruct

ure

• R

emov

al o

f woo

d de

plet

es L

WD

bef

ore

ripar

ian

recr

uitm

ent r

esum

es

Lisl

e 19

86a

Nor

ther

n C

alifo

rnia

•

0.04

to 0

.22

chan

nel g

radi

ents

•

1st th

roug

h 2nd

ord

er s

tream

s •

Gre

ater

abu

ndan

ce o

f org

anic

deb

ris d

ams

in c

ontro

l stre

ams

than

in s

tream

s ad

jace

nt

to c

lear

cut

s or

har

vest

ed re

ache

s •

Incr

ease

d le

vels

of L

WD

load

ing

follo

win

g ha

rves

t des

tabi

lize

exis

ting

debr

is d

ams

or

chan

ge d

am c

hara

cter

istic

s th

roug

h ad

ditio

n or

mob

ilizat

ion

of w

ood

O'C

onno

r 198

6

Wes

tern

Ore

gon

• 0.

02 to

0.0

6 ch

anne

l gra

dien

ts

• 5.

5 m

to 8

m b

ankf

ull w

idth

s •

6.5

km2 d

rain

age

area

• Af

ter l

oggi

ng, d

ebris

from

cur

rent

sta

nd c

ontri

bute

d 14

% o

f tot

al L

WD

vol

ume

and

7%

of L

WD

from

cur

rent

sta

nd c

ontri

bute

d to

poo

l for

mat

ion

• R

esul

ts in

dica

te th

at tr

ees

mus

t gro

w b

eyon

d at

leas

t 50

year

s be

fore

sta

nds

cont

ribut

e LW

D in

am

ount

s si

mila

r to

old

grow

th fo

rest

s

Andr

us e

t al.

1988

48

Tabl

e A

-10.

The

effe

ct o

f tim

ber h

arve

st o

n LW

D c

hara

cter

istic

s in

var

ious

geo

grap

hic

loca

tions

and

cha

nnel

net

wor

k po

sitio

ns (a

rrang

ed in

chr

onol

ogic

al o

rder

). LO

CAT

ION

C

HAN

NEL

NET

WO

RK

PO

SITI

ON

R

ESU

LTS

REF

EREN

CES

W

este

rn

Was

hing

ton

• 0.

13 c

hann

el g

radi

ent (

<7 m

ch

anne

l wid

th)

• 0.

08 c

hann

el g

radi

ent (

7 m

to 1

0 m

cha

nnel

wid

th)

• 0.

03 c

hann

el g

radi

ent (

>10

m

chan

nel w

idth

)

• H

ighe

r LW

D a

bund

ance

in o

ld g

row

th th

an 2

nd g

row

th o

r cle

ar c

ut

• Po

ol ty

pes

chan

ged

with

sta

nd a

ge c

lass

: sco

ur p

ools

mor

e fre

quen

t in

2nd g

row

th a

nd

clea

r cut

, and

dom

inan

t poo

l typ

es in

all

stre

am s

izes

•

Freq

uenc

y of

LW

D a

ssoc

iate

d po

ols

sign

ifica

ntly

gre

ater

in o

ld g

row

th s

ites

• Pe

rcen

t of L

WD

form

ing

wat

erfa

lls a

nd p

ropo

rtion

of e

leva

tion

drop

con

trolle

d by

LW

D

grea

test

in o

ld g

row

th s

ites

• Ar

ea o

f LW

D a

ssoc

iate

d se

dim

ent g

reat

est i

n ol

d gr

owth

site

s •

Mor

e fin

e or

gani

c de

bris

reta

ined

by

old

grow

th s

ites

• Fi

ne o

rgan

ic d

ebris

may

con

tribu

te to

the

dive

rsity

of p

ool t

ypes

in o

ld g

row

th s

tream

s •

Man

y ch

ange

s in

LW

D o

ccur

sho

rtly

afte

r (<5

yr) a

fter h

arve

st

Bilb

y an

d W

ard

1989

Sout

heas

tern

Al

aska

•

0.04

to 0

.29

chan

nel g

radi

ents

•

8.2

m to

20

m c

hann

el w

idth

•

99%

of L

WD

sou

rces

with

in 3

0 m

of c

hann

el, 3

0 m

buf

fer a

long

bot

h ba

nks

shou

ld

mai

ntai

n LW

D in

put

• M

odel

pre

dict

ed L

WD

redu

ced

by 7

0% n

inet

y ye

ars

afte

r cle

ar c

uttin

g, re

cove

ry to

pre

-lo

ggin

g le

vel e

xcee

ds 2

50 y

ears

•

Stre

amsi

de m

anag

emen

t sho

uld

prov

ide

supp

ly o

f LW

D s

tabl

e en

ough

to p

rovi

de fi

sh

habi

tat

Mur

phy

and

Kosk

i 198

9

East

ern

Ore

gon

East

ern

Was

hing

ton

• 0.

02 to

0.0

7 ch

anne

l gra

dien

ts

• 2

m to

6 m

ave

rage

cha

nnel

w

idth

s

• LW

D v

olum

e at

logg

ed s

ites

1 to

2 fo

ld h

ighe

r tha

n un

logg

ed s

ites

due

to lo

ggin

g re

sidu

e an

d bl

owdo

wn,

but

ove

rall

amou

nts

wer

e si

mila

r C

arls

on e

t al.

1990

Ore

gon

• 1st

thro

ugh

3rd o

rder

stre

ams

• 70

% o

f LW

D c

ame

from

20

m o

f cha

nnel

, 11%

cam

e fro

m 1

m o

f the

cha

nnel

•

Prob

abilis

tic m

odel

bas

ed o

n tre

e he

ight

and

den

sity

•

Mod

el u

sed

to in

terp

ret e

ffect

of w

idth

of b

uffe

r stri

ps o

n fu

ture

LW

D re

crui

tmen

t

McD

ade

et a

l. 19

90

Ore

gon

• 0.

13 c

hann

el g

radi

ent

• 12

ave

rage

ban

kful

l wid

th

• 3rd

ord

er s

tream

• Pr

obab

ilistic

mod

el p

redi

ctin

g nu

mbe

r and

vol

ume

of L

WD

inpu

t •

Inpu

ts a

ggre

gate

d w

ith re

spec

t to

dist

ance

, tre

e he

ight

, and

spe

cies

•

Dem

onst

rate

s im

porta

nce

of c

onsi

derin

g st

and

com

posi

tion,

not

just

stre

am s

ize

in

desi

gnin

g bu

ffer z

ones

Van

Sick

le a

nd G

rego

ry

1990

Sout

heas

tern

Al

aska

•

0.0

02 to

0.0

2 ch

anne

l gra

dien

ts

• 1

km2 to

30

km2 d

rain

age

area

•

4 m

to 2

5 m

cha

nnel

wid

ths

• C

hann

el u

nit d

istri

butio

n di

ffers

mar

kedl

y be

twee

n di

stur

bed

and

undi

stur

bed

site

s •

55%

of w

ette

d ch

anne

l are

a in

poo

l in

und

istu

rbed

reac

hes

and

34%

in d

istu

rbed

re

ache

s •

Diff

eren

ces

in c

hann

el u

nit d

istri

butio

n at

tribu

ted

to d

iffer

ence

in p

ool a

ssoc

iate

d w

ood

• LW

D lo

adin

g gr

eate

r in

undi

stur

bed

stre

ams

• Fr

eque

ncy

of p

ool t

ypes

diff

er b

etw

een

dist

urbe

d an

d un

dist

urbe

d re

ache

s, im

plyi

ng

effe

ct o

f lan

d us

e •

Pris

tine

chan

nel c

ondi

tion

can

be d

iscr

imin

ated

from

man

aged

cha

nnel

s by

ana

lyzi

ng

geom

orph

ic v

aria

bles

Woo

d-Sm

ith a

nd

Buffi

ngto

n 19

96

Wes

tern

W

ashi

ngto

n •

0.00

4 to

0.1

1 ch

anne

l gra

dien

ts

• 0.

5 km

2 to 1

4 km

2 dra

inag

e ar

eas

• U

nlog

ged

stre

ams

had

high

er fr

actio

n of

woo

dy d

ebris

in la

rger

siz

e ca

tego

ries

• In

tens

ive

timbe

r har

vest

ing

sign

ifica

ntly

dec

reas

ed fr

actio

n of

stre

am a

rea

in p

ools

•

Inte

nsiv

e tim

ber h

arve

st re

duce

d po

ol d

epth

and

poo

l are

a •

Inte

nsiv

ely

harv

este

d ba

sins

: 50%

of w

ood

outs

ide

low

flow

cha

nnel

, 33%

inte

ract

ed

with

flow

, 6%

com

plet

ely

with

in c

hann

el

• M

oder

atel

y ha

rves

ted

basi

n: 2

5% o

f woo

d ou

tsid

e lo

w fl

ow c

hann

el, 5

0% in

tera

cted

w

ith fl

ow, 1

9% fu

lly w

ithin

cha

nnel

•

Unl

ogge

d ba

sins

: 25%

of w

ood

outs

ide

chan

nel,

66%

inte

ract

ed w

ith c

hann

el, 3

2%

fully

in c

hann

el

• St

ream

s flo

win

g th

roug

h ol

d gr

owth

fore

sts

reta

ined

mor

e pi

eces

of s

mal

ler w

ood

• N

ewly

recr

uite

d LW

D w

as le

ss s

tabl

e in

mod

erat

e an

inte

nsel

y ha

rves

ted

stre

ams

and

less

like

ly to

con

tribu

te to

con

tribu

te to

hab

itat s

truct

ure

with

less

of a

role

in b

edlo

ad

stor

age

Ral

ph e

t al.

1994

49

Tabl

e A

-10.

The

effe

ct o

f tim

ber h

arve

st o

n LW

D c

hara

cter

istic

s in

var

ious

geo

grap

hic

loca

tions

and

cha

nnel

net

wor

k po

sitio

ns (a

rrang

ed in

chr

onol

ogic

al o

rder

). LO

CAT

ION

C

HAN

NEL

NET

WO

RK

PO

SITI

ON

R

ESU

LTS

REF

EREN

CES

N

orth

Car

olin

a •

0.01

to 0

.10

chan

nel g

radi

ents

•

5 m

to 1

1 m

cha

nnel

wid

ths

• 1st

thro

ugh

4th o

rder

stre

ams

• LW

D v

olum

e in

crea

sed

linea

rly in

stre

ams

asso

ciat

ed w

ith la

te s

ucce

ssio

nal t

hrou

gh

old

grow

th fo

rest

s •

Inst

ream

load

ings

bui

ld a

nd s

tabi

lize

in la

ter s

tage

s of

suc

cess

ion

• St

ream

s flo

win

g th

roug

h yo

unge

r for

ests

rely

on

LWD

from

sta

nd g

ener

atin

g di

stur

banc

es; i

nstre

am L

WD

not

rela

ted

to fo

rest

age

•

Amer

ican

che

stnu

t (C

asta

nea

dent

ata)

maj

or L

WD

com

pone

nt in

mid

-suc

cess

iona

l st

ream

s

Hed

man

et a

l. 1

996

Nor

ther

n C

alifo

rnia

•

4.2

km2 to

5 k

m2 d

rain

age

area

s •

LWD

incr

ease

d af

ter l

oggi

ng d

ue to

resi

dual

tree

s ad

jace

nt to

the

stre

am o

r in

buffe

r st

rips

• D

ougl

as fi

r pro

vide

d gr

eate

st a

mou

nt o

f LW

D in

sec

ond

grow

th s

yste

ms

• LW

D a

ssoc

iate

d po

ols

and

debr

is ja

ms

high

er in

logg

ed s

tream

s

Surfl

eet a

nd Z

eim

er 1

996

UK

• 0.

013

aver

age

chan

nel g

radi

ent

• 2nd

thro

ugh

3rd o

rder

stre

ams

• N

umbe

r of d

ebris

dam

s re

cove

red

7 to

8 y

ears

afte

r rem

oval

•

The

num

ber o

f hyd

raul

ical

ly a

ctiv

e da

ms

did

not r

ecov

er a

s th

e nu

mbe

r and

siz

e of

po

ols

was

low

er th

an b

efor

e re

mov

al

Gur

nell

and

Swee

t 19

98

Nor

ther

n C

alifo

rnia

•

0.02

cha

nnel

gra

dien

t •

7.7

m a

vera

ge b

ankf

ull w

idth

•

3.8

km2 d

rain

age

area

•

3rd o

rder

stre

am

• Lo

ggin

g in

crea

sed

inpu

t of L

WD

from

blo

wdo

wns

, lea

ding

to in

crea

sed

stor

age

of b

ed

mat

eria

l, in

crea

sed

num

ber a

nd to

tal v

olum

e of

poo

ls, a

nd te

mpo

raril

y in

crea

sed

fine

sedi

men

t •

Incr

ease

d LW

D v

olum

es in

crea

sed

size

and

abu

ndan

ce o

n po

ols

• In

crea

se in

LW

D a

nd L

WD

ass

ocia

ted

habi

tats

may

be

tem

pora

ry a

s in

-cha

nnel

woo

d de

cays

and

nat

ural

inpu

t sou

rces

reco

ver

• R

each

es b

orde

red

by c

lear

cuts

and

buf

fer s

trips

may

lose

sed

imen

t sto

rage

, poo

l vo

lum

e, a

nd h

abita

t com

plex

ity a

s LW

D in

puts

dec

line

Lisl

e an

d N

apol

itano

199

8

Wes

tern

W

ashi

ngto

n •

<0.0

2 ch

anne

l gra

dien

ts

• 3.

0 to

12

km2 d

rain

age

area

s •

7 m

to 1

6 m

cha

nnel

wid

th

• D

iam

eter

s of

old

gro

wth

LW

D la

rger

than

2nd

gro

wth

LW

D

• Fo

rest

ry p

ract

ices

: con

vers

ion

of o

ld g

row

th c

onife

rous

to d

ecid

uous

spe

cies

, ins

tream

sa

lvag

e, a

nd s

tream

cle

anin

g ha

ve a

ltere

d LW

D c

hara

cter

istic

s •

60%

redu

ctio

n in

LW

D fo

r log

ged

old

grow

th s

ites

betw

een

1982

and

199

3 •

Ove

r 10

year

per

iod,

num

ber o

f pie

ces

iden

tical

but

LW

D v

olum

e de

crea

sed

as a

n in

crea

se in

sec

ond

grow

th L

WD

was

insu

ffici

ent t

o of

fset

loss

of o

ld g

row

th L

WD

•

Dia

met

er o

f 2nd

gro

wth

LW

D in

crea

sed

from

198

2 to

199

3, b

ut s

till s

mal

ler t

han

old

grow

th L

WD

•

Initi

al ra

pid

loss

of o

ld L

WD

follo

win

g tim

ber h

arve

st c

ause

d by

des

tabi

lizat

ion

and

yard

ing,

cha

nnel

des

tabi

lizat

ion

and

trans

port,

and

long

-term

dec

ay

• C

hara

cter

istic

s of

LW

D fr

om s

econ

d gr

owth

LW

D m

uch

diffe

rent

from

old

gro

wth

co

nife

rs

McH

enry

et a

l. 19

98

Nor

ther

n C

alifo

rnia

•

0.08

cha

nnel

gra

dien

t •

3 m

to 2

0 m

cha

nnel

wid

th

• LW

D s

igni

fican

tly lo

wer

in re

cent

ly lo

gged

bas

ins

than

in o

ld g

row

th re

dwoo

d sy

stem

s N

apol

itano

199

8

Nor

ther

n C

alifo

rnia

•

N/A

•

Pres

ence

of c

lear

cut

s ne

xt to

buf

fer s

trips

can

incr

ease

occ

urre

nce

on w

indt

hrow

for a

di

stan

ce o

f 150

m

• R

esul

ts s

ugge

st u

se o

f a c

ore

buffe

r, gr

eate

r tha

n on

e tre

e he

ight

, fla

nked

by

a fri

nge

buffe

r to

prot

ect c

ore

buffe

r fro

m a

bnor

mal

ly h

igh

rate

s of

win

dthr

ow

Rei

d an

d H

ilton

199

8

50

Tabl

e A-

11.

The

effe

ct o

f LW

D m

anag

emen

t on

fish

habi

tat i

n va

rious

geo

grap

hic

loca

tions

and

cha

nnel

net

wor

k po

sitio

ns (a

rrang

ed in

chr

onol

ogic

al o

rder

). LO

CAT

ION

C

HAN

NEL

NET

WO

RK

PO

SITI

ON

R

ESU

LTS

REF

EREN

CES

C

entra

l Ore

gon

• 0.

71 k

m2 to

30

km2 d

rain

age

area

s •

Burn

ing

of s

lash

pos

sibl

y in

crea

sed

stre

am te

mpe

ratu

res,

resu

lting

in m

orta

lity

of

coho

and

cut

thro

at

• D

ebris

on

grav

el s

urfa

ce p

reve

nted

inte

rcha

nge

betw

een

surfa

ce a

nd in

tragr

avel

w

ater

, red

ucin

g di

ssol

ved

oxyg

en

• D

ecom

posi

tion

of d

ebris

incr

ease

d bi

olog

ical

oxy

gen

dem

and,

furth

er re

duci

ng

avai

labl

e di

ssol

ved

oxyg

en

• To

pre

vent

redu

ctio

n in

dis

solv

ed o

xyge

n, k

eep

all d

ebris

out

of c

hann

el th

roug

h us

e of

a b

uffe

r stri

p

Hal

l and

Lan

tz 1

968

Paci

fic N

orth

wes

t, U

SA

• N

/A

• Ex

cess

logg

ing

deriv

ed L

WD

may

blo

ck fi

sh p

assa

ge o

r del

ay fi

sh m

igra

tion

in v

-sh

aped

cha

nnel

s •

LWD

may

des

tabi

lize

stre

ambe

d, p

artic

ular

ly s

paw

ning

gra

vel

• Sm

all d

ebris

(woo

d fib

er, b

ark,

leav

es) f

ills in

ters

tices

of g

rave

l, re

duci

ng li

ving

spa

ce

for s

tream

inve

rtebr

ates

•

Smal

l deb

ris h

as h

igh

BOD

and

CO

D, d

ecre

asin

g di

ssol

ved

O2

conc

entra

tion

• Sm

all d

ebris

on

grav

el re

duce

s in

ters

titia

l flo

w, a

nd in

crea

ses

BOD

and

CO

D,

decr

ease

d D

O in

tragr

avel

wat

er, i

nflu

enci

ng d

epth

of e

mbr

yos

and

alev

ins

• Ta

nnin

s an

d lig

nin

like

subs

tanc

es p

rodu

ce y

ello

w a

nd b

row

n co

lors

that

cou

ld a

bsor

b ph

otos

ynth

etic

ally

act

ive

radi

atio

n w

ithin

a fe

w in

ches

of t

he s

urfa

ce

Nar

ver 1

970

Paci

fic N

orth

wes

t, U

SA

• N

/A

• Ex

cess

logg

ing

deriv

ed F

WD

may

dec

reas

ed d

isso

lved

O2,

incr

ease

BO

D, a

nd

rest

rict a

erat

ion

of w

ater

•

Exce

ss lo

ggin

g de

rived

LW

D p

rese

nts

barri

ers

to fi

sh m

ovem

ent,

can

caus

e or

co

ntrib

ute

to "f

lush

out

s," m

ass

mov

emen

ts th

at s

cour

stre

ams

to b

edro

ck a

nd re

duce

co

mpl

exity

Brow

n 19

74

Wes

tern

W

ashi

ngto

n •

0.01

to 0

.08

chan

nel g

radi

ents

•

1 km

2 to

30 k

m2 d

rain

age

area

s •

Salm

onid

bio

mas

s 1.

5 tim

es g

reat

er in

logg

ed s

ectio

ns

• Lo

gged

wat

ersh

eds

cont

aine

d hi

gher

per

cent

ages

of a

ge 0

+ st

eelh

ead

(Onc

orhy

nchu

s m

ykis

s) a

nd a

ge 0

+ cu

tthro

at (O

ncor

hync

hus

clar

ki)

• D

ebris

rem

oval

follo

win

g tim

ber h

arve

st in

crea

sed

riffle

are

a •

Juve

nile

ste

elhe

ad (O

ncor

hync

hus

myk

iss)

and

cut

thro

at (O

ncor

hync

hus

clar

ki)

poss

ibly

incr

ease

d du

e to

pre

fere

nce

for r

iffle

hab

itats

Biss

on a

nd S

edel

l 198

4

Ore

gon

• N

/A

• St

ruct

ures

incr

ease

d di

vers

ity o

f stre

ambe

d, tr

appe

d gr

avel

, cre

ated

gra

vel b

ars

and

pool

s •

Spaw

ning

occ

urre

nce

of c

oho

and

stee

lhea

d(O

ncor

hync

hus

myk

iss)

incr

ease

d in

m

odifi

ed s

ectio

ns

• C

ontro

l sec

tion

supp

orte

d hi

gher

den

sitie

s of

juve

nile

coh

o (O

ncor

hync

hus

kisu

tch)

, st

eelh

ead

(Onc

orhy

nchu

s m

ykis

s), a

nd c

utth

roat

(Onc

orhy

nchu

s cl

arki

)

Hou

se a

nd B

oehn

e 19

85

Wes

tern

W

ashi

ngto

n •

N/A

•

Wat

er tr

ansp

orta

tion

and

stor

age

of lo

gs s

ever

ely

impa

cts

phys

ical

, che

mic

al, a

nd

biol

ogic

al fu

nctio

ning

of s

tream

s Se

dell

and

Duv

all 1

985

Sout

heas

t Ala

ska

• 0.

035

to 0

.053

cha

nnel

gra

dien

t •

2 m

ave

rage

cha

nnel

wid

th

• U

ncle

aned

stre

am s

ectio

n co

ntai

ned

mor

e fis

h of

all

size

cla

sses

than

cle

aned

se

ctio

ns

• N

umbe

rs a

nd p

rodu

ctio

n of

coh

o (O

ncor

hync

hus

kisu

tch)

and

Dol

ly V

arde

n (S

alve

linus

mal

ma)

redu

ced

by L

WD

rem

oval

•

Hab

itat s

impl

ifica

tion

affe

cted

age

1+

and

olde

r fis

h m

ore

than

age

0+

• LW

D p

rovi

ded

impo

rtant

hyd

raul

ic c

over

dur

ing

win

ter

Dol

loff

1986

51

Tabl

e A-

11.

The

effe

ct o

f LW

D m

anag

emen

t on

fish

habi

tat i

n va

rious

geo

grap

hic

loca

tions

and

cha

nnel

net

wor

k po

sitio

ns (a

rrang

ed in

chr

onol

ogic

al o

rder

). LO

CAT

ION

C

HAN

NEL

NET

WO

RK

PO

SITI

ON

R

ESU

LTS

REF

EREN

CES

So

uthe

ast A

lask

a •

1.9

m a

vera

ge c

hann

el w

idth

•

9 km

2 cha

nnel

wid

th

• 6.

5 km

long

• R

emov

al lo

ggin

g de

bris

redu

ced

wet

ted

wid

th a

nd a

reas

from

874

m2 a

nd 7

26m

2 •

Dol

ly V

arde

n (S

alve

linus

mal

ma)

pop

ulat

ion

decr

ease

d fro

m 7

48 to

458

to 1

00 tw

o ye

ars

afte

r deb

ris re

mov

al

• D

ecre

ase

in fi

sh s

ize

afte

r cle

aran

ce, b

efor

e de

bris

rem

oval

, fis

h oc

cupi

ed m

id

chan

nel p

ools

and

tang

les

of d

ebris

, afte

r rem

oval

occ

upie

d ba

ckw

ater

s an

d st

ream

m

argi

ns

• R

emov

al o

f log

ging

deb

ris c

hang

ed m

orph

olog

y, b

riefly

redu

ced

food

sup

ply

and

elim

inat

ed m

ost c

over

, red

ucin

g nu

mbe

r and

mea

n le

ngth

of f

ish,

leav

ing

smal

l fis

h su

scep

tible

to d

ispl

acem

ent

• D

ecre

ase

in b

enth

os a

bund

ance

, red

uced

am

ount

of f

ood

eate

n by

Dol

ly V

arde

n (S

alve

linus

mal

ma)

and

may

hav

e co

ntrib

uted

to n

umer

ical

dec

line

Ellio

tt 19

86

Sout

heas

tern

Al

aska

•

0.02

to 0

.03

chan

nel g

radi

ents

•

2nd to

4th o

rder

stre

ams

• M

ost f

ish

rega

rdle

ss o

f spe

cies

or a

ge w

ere

in p

ools

•

Amou

nts

of L

WD

and

cob

ble

high

er in

occ

upie

d po

ols

than

in u

nocc

upie

d po

ols

• LW

D c

ause

d 73

% o

f all

pool

s •

Juve

nile

sal

mon

ids

(in w

inte

r) pr

efer

red

pool

with

cov

er (u

ptur

ned

root

s, a

ccum

ulat

ion

of lo

gs, c

obbl

e su

bstra

te)

• Am

ount

of p

refe

rred

win

ter h

abita

t dep

ends

on

amou

nt o

f LW

D

• C

lear

cut r

each

es le

ss L

WD

and

poo

l are

a •

Buffe

rs p

rote

ct h

abita

t by

prov

idin

g LW

D

Hei

fietz

et a

l. 19

86

Ore

gon

• 9.

2 km

2 dra

inag

e ar

ea

• 0.

03 a

vera

ge c

hann

el g

radi

ent

• LW

D in

logg

ed s

ectio

n (6

/100

m) s

how

ed a

n in

crea

sed

in u

ndis

turb

ed s

ectio

n (4

6/10

0m)

• LW

D m

ost i

mpo

rtant

fact

or c

ontro

lling

diffe

renc

es in

cha

nnel

mor

phol

ogy

LWD

cr

eate

d st

air s

tep

mor

phol

ogy,

sec

onda

ry c

hann

els,

and

mea

nder

s •

Und

istu

rbed

sec

tions

, LW

D im

prov

ed p

ool q

ualit

y an

d #

of p

ools

•

Afte

r LW

D a

dditi

ons,

gra

vel i

ncre

ased

by

233%

, gra

vel a

reas

incr

ease

d 25

fold

in

area

•

3 tim

es m

ore

coho

(0.8

8/m

2 ) and

ste

elhe

ad (0

.59/

m2 ) i

n un

dist

urbe

d se

ctio

ns

• Po

sitiv

e co

rrela

tion

betw

een

fish

popu

latio

ns (s

paw

ning

and

rear

ing

habi

tat)

and

LWD

Hou

se a

nd B

oehn

e 19

86

Sout

heas

tern

Al

aska

•

0.10

to 0

.30

chan

nel g

radi

ent

• 5

m to

7 m

cha

nnel

wid

th

• W

inte

r fry

den

sity

bes

t mod

eled

by

sum

mer

per

iphy

ton

biom

ass

and

volu

me

of

inst

ream

deb

ris

• Pa

rr ab

unda

nt w

here

deb

ris w

as a

bund

ant

• In

bot

h su

mm

er a

nd w

inte

r, de

nsity

of D

olly

Var

den

(Sal

velin

us m

alm

a) p

arr d

irect

ly

rela

ted

to L

WD

vol

ume

• C

over

(LW

D a

nd u

nder

cut b

anks

) mor

e im

porta

nt fo

r all

parr

in w

inte

r

Mur

phy

et a

l. 19

86

Nor

ther

n C

olor

ado

• 0.

01 to

0.0

24 c

hann

el g

radi

ents

•

3 m

to 6

m c

hann

el w

idth

s •

Pool

vol

ume

afte

r dro

p st

ruct

ures

inst

alle

d •

Amou

nt o

f cov

er in

crea

sed

sign

ifica

ntly

afte

r add

ition

of l

og d

rop

stru

ctur

es

• Tr

eatm

ent s

ectio

ns w

ere

deep

er a

nd h

ad lo

wer

vel

ocity

than

con

trol s

ectio

ns

• Ab

unda

nce

and

biom

ass

of a

ge 2

and

old

er c

utth

roat

(Onc

orhy

nchu

s cl

arki

) in

crea

sed

sign

ifica

ntly

afte

r ins

talla

tion

of d

rop

stru

ctur

es

• Lo

g dr

op s

truct

ures

incr

ease

d ab

unda

nce

of a

dult

broo

k tro

ut (S

alve

linus

font

inal

is)

and

poss

ibly

sur

viva

l, al

thou

gh d

ata

not c

oncl

usiv

e

Rile

y an

d Fa

usch

199

5

Nor

th C

arol

ina

• 0.

03 to

0.0

6 ch

anne

l gra

dien

ts

• Po

ol fo

rmat

ion

afte

r log

add

ition

•

Enha

nced

redu

ctio

n up

stre

am fr

om d

ebris

dam

s •

Inve

rtebr

ate

abun

danc

e an

d bi

omas

s in

crea

sed

sign

ifica

ntly

afte

r log

add

ition

•

Scra

per a

nd c

olle

ctor

/filte

rer a

bund

ance

incr

ease

d af

ter l

og a

dditi

on

• Se

cond

ary

prod

uctio

n of

scr

aper

s an

d fil

tere

rs d

ecre

ased

•

Shift

s in

func

tiona

l gro

up a

bund

ance

s, b

iom

ass,

and

abu

ndan

ce a

ccen

tuat

e th

e im

porta

nce

of lo

cal a

biot

ic fa

ctor

s in

stru

ctur

e of

inve

rtebr

ate

com

mun

ities

with

in

patc

hes

Wal

lace

et a

l. 19

95

52

Tabl

e A-

11.

The

effe

ct o

f LW

D m

anag

emen

t on

fish

habi

tat i

n va

rious

geo

grap

hic

loca

tions

and

cha

nnel

net

wor

k po

sitio

ns (a

rrang

ed in

chr

onol

ogic

al o

rder

). LO

CAT

ION

C

HAN

NEL

NET

WO

RK

PO

SITI

ON

R

ESU

LTS

REF

EREN

CES

N

orth

ern

Cal

iforn

ia

• 4.

2 km

2 and

4.7

km

2 dra

inag

e ar

eas

• In

crea

se in

LW

D v

olum

e af

ter t

imbe

r har

vest

•

LWD

incr

ease

d ha

bita

t are

for s

teel

head

(Onc

orhy

nchu

s m

ykis

s), c

oho

(Onc

orhy

nchu

s ki

sutc

h), a

nd P

acifi

c G

iant

sal

aman

ders

(Dic

ampt

on te

nebr

osus

) •

As lo

ggin

g de

rived

LW

D d

ecre

ases

thro

ugh

deca

y, a

nd a

s na

tura

l inp

uts

reco

ver,

habi

tat c

ondi

tions

may

not

per

sist

Nak

amot

o 19

98

Tabl

e A

-12.

The

effe

ct o

f roa

ds o

n LW

D c

hara

cter

istic

s in

var

ious

geo

grap

hic

loca

tions

and

cha

nnel

net

wor

k po

sitio

ns (a

rrang

ed in

chr

onol

ogic

al o

rder

). LO

CAT

ION

C

HAN

NEL

NET

WO

RK

PO

SITI

ON

R

ESU

LTS

REF

EREN

CES

O

rego

n •

N/A

•

Logg

ing

debr

is in

stre

am c

hann

els

is a

maj

or c

ontri

buto

r to

flood

dam

age

• Lo

ggin

g re

sidu

e te

nds

to b

e co

ncen

trate

d in

dra

ws

and

depr

essi

ons

that

car

ry w

ater

du

ring

flood

con

ditio

ns

• Pr

oble

m o

f woo

d tra

nspo

rt th

roug

h ch

anne

ls c

ompo

unde

d by

road

way

s th

at p

lace

re

stric

tion

on m

ovem

ent

• C

ulve

rts a

nd b

ridge

s ob

stru

ct w

ood

mov

emen

t and

cau

se d

amag

e to

road

sys

tem

•

Dam

age

to ro

ad s

yste

m is

maj

or it

em in

cal

cula

tion

of fl

ood

loss

es

• C

ost o

f pro

blem

is u

nkno

wn,

but

incl

uded

dam

age

to ro

ads

and

desi

gn o

f stru

ctur

es to

pa

ss d

ebris

•

Asid

e fro

m re

mov

ing

all d

ebris

, alte

rnat

ives

are

to m

odify

or b

uild

stru

ctur

es to

pas

s w

ood,

mec

hani

cally

rem

ove

larg

e ja

ms,

bui

ld ro

ads

to a

ct a

s de

bris

bar

rier

Froe

hlic

h 19

70

N/A

•

N/A

•

Cle

arin

g an

d sn

aggi

ng d

one

for t

hree

reas

ons:

dra

in fl

oodp

lain

s fo

r agr

icul

tura

l de

velo

pmen

t, pr

otec

t citi

zens

from

floo

ds, m

aint

ain

navi

gabl

e w

ater

way

s •

Cle

arin

g an

d sn

aggi

ng in

hig

h or

der s

tream

s re

duce

s ro

ughn

ess

coef

ficie

nt

Mar

zolf

1978

Wes

tern

Ore

gon

• 0.

02 to

0.5

0 ch

anne

l gra

dien

ts

• H

ighe

r lev

els

of d

ebris

torre

nt a

ctiv

ity in

are

as w

ith h

igh

road

are

a re

lativ

e to

fore

st

area

Sw

anso

n an

d Li

enka

empe

r 19

78

Cen

tral C

alifo

rnia

•

104

km2 d

rain

age

area

•

Floo

d da

mag

es s

igni

fican

tly re

duce

d if

all b

ridge

s ke

pt fr

ee o

f log

jam

s •

No

prac

tical

way

to p

reve

nt a

ll tre

es fr

om e

nter

ing

chan

nel s

ince

mos

t com

e fro

m

debr

is fl

ow tr

igge

red

by s

torm

s •

Onl

y fe

asib

le s

olut

ion

to p

reve

nt a

ccum

ulat

ion

at b

ridge

s is

to m

odify

or r

epla

ce

brid

ges

son

that

logs

will

flow

und

er w

ithou

t cat

chin

g •

Stre

am c

lear

ance

is n

ot a

sol

utio

n as

mos

t log

s co

me

form

deb

ris fl

ows

that

occ

ur

durin

g th

e st

orm

•

Stud

ies

shou

ld in

clud

e lo

g pa

ssin

g ca

paci

ty o

f all

dow

nstre

am b

ridge

s

Sing

er a

nd S

wan

son

1983

N/A

•

N/A

•

LWD

and

veg

etat

ion

rem

oved

to in

crea

se h

ydra

ulic

cap

acity

Sh

ield

s an

d N

unna

lly 1

984

Cal

iforn

ia

• N

/A

• C

lear

wat

er d

esig

n of

floo

d co

ntro

l cha

nnel

igno

res

or u

nder

estim

ates

role

of f

loat

ing

debr

is in

incr

easi

ng fl

ood

elev

atio

ns

• D

ebris

impe

des

or b

lock

s sm

all t

o m

ediu

m s

ized

cul

verts

in s

mal

l to

med

ium

siz

ed

culv

erts

in s

mal

ler s

tream

s, a

nd b

ridge

s an

d la

rger

cul

verts

in m

ain

chan

nels

, cau

sing

si

gnifi

cant

incr

ease

s in

floo

d el

evat

ions

•

Con

veya

nce

prog

ram

s th

at e

ncou

rage

rem

oval

of w

ood

rare

ly u

nder

go te

chni

cal

scru

tiny;

an

appr

oach

that

con

side

rs h

ydro

logi

c, g

eom

orph

ic, a

nd b

iolo

gica

l fac

tors

is

need

ed

Willi

ams

and

Swan

son

1989

N/A

•

N/A

•

Cro

ssin

gs s

houl

d be

des

igne

d to

tran

smit

debr

is d

owns

tream

Fu

rnis

s et

al.

1991

53

Tabl

e A

-12.

The

effe

ct o

f roa

ds o

n LW

D c

hara

cter

istic

s in

var

ious

geo

grap

hic

loca

tions

and

cha

nnel

net

wor

k po

sitio

ns (a

rrang

ed in

chr

onol

ogic

al o

rder

). LO

CAT

ION

C

HAN

NEL

NET

WO

RK

PO

SITI

ON

R

ESU

LTS

REF

EREN

CES

Au

stra

lia

• 0.

46 m

(w) b

y 8

m (l

) by

0.27

(d)

artif

icia

l cha

nnel

•

The

aver

age

amou

nt o

f LW

D fo

und

in A

ustra

lian

low

land

rive

rs s

eldo

m h

as a

si

gnifi

cant

on

flood

leve

ls

• W

here

larg

e lo

g ja

ms

form

or a

t cha

nnel

con

stric

tion

will

effe

cts

be s

igni

fican

t •

Whe

re c

onsu

mpt

ive

uses

redu

ce fl

ow le

vels

, LW

D re

mov

al m

ay b

e re

quire

d m

ore

frequ

ently

to p

reve

nt ja

m fo

rmat

ion

• As

a c

ompr

omis

e be

twee

n LW

D re

mov

al fo

r hyd

raul

ic re

ason

s an

d ha

bita

t pr

eser

vatio

n, it

may

by

poss

ible

to re

arra

nge

LWD

for m

inim

al h

ydra

ulic

influ

ence

Youn

g 19

91

Aust

ralia

•

3540

km

2 dra

inag

e ar

ea

• R

esul

ts s

ugge

st th

at d

e-sn

aggi

ng w

ould

redu

ce b

ankt

op fl

ood

heig

ht b

y 0.

2%

• Av

erag

e de

bris

pie

ces

wou

ld h

ave

min

imal

influ

ence

on

flow

hyd

raul

ics,

alth

ough

la

rge

piec

es m

ay b

e si

gnifi

cant

•

Re-

intro

duct

ion

of w

ood

into

cle

ared

rive

rs u

nlik

ely

to re

sult

in lo

ss o

f con

veya

nce

Gip

pel e

t al.

1996

Uni

ted

Stat

es

• N

/A

• Fl

oatin

g de

bris

incr

ease

s la

tera

l for

ces

on b

ridge

s, p

rom

otin

g sc

our

• D

rift a

ccum

ulat

ion

depe

nds

on c

hann

el, b

asin

, and

brid

ge c

hara

cter

istic

s •

Drif

t acc

umul

ates

aga

inst

brid

ge p

iers

, and

obs

tacl

es s

epar

ated

by

narro

w g

aps

trap

drift

mos

t effe

ctiv

ely

• G

aps

betw

een

two

piec

es a

re n

ot b

lock

ed u

nles

s lo

gs c

an s

pan

pier

to p

ier

• Su

gges

ted

desi

gn fe

atur

es to

redu

ce d

rift a

ccum

ulat

ion:

ade

quat

e fre

eboa

rd, w

ide

span

s, s

olid

pie

rs, r

ound

ed p

ier n

oses

, and

pie

r pla

cem

ent o

ut o

f pat

h of

drif

t (th

alw

eg)

Die

hl 1

997

Okl

ahom

a •

N/A

•

Deb

ris e

ntra

pmen

t inc

reas

ed M

anni

ng's

n b

y 39

%

• Ef

fect

of d

ebris

on

flow

resi

stan

ce d

ecre

ased

with

incr

ease

d di

scha

rge

• R

esul

ts li

mite

d to

cha

nnel

<0.

6 m

dep

th (f

lood

plai

ns o

r low

ord

er c

hann

els)

Dud

ley

et a

l. 19

98

Cal

iforn

ia

• N

/A

• Pl

uggi

ng o

f cul

verts

by

LWD

is a

com

mon

failu

re m

echa

nism

•

Piec

es in

itiat

ing

plug

ging

are

ofte

n no

t muc

h lo

nger

than

cul

vert

diam

eter

•

Cul

verts

siz

ed e

qual

to th

e ch

anne

l wid

th w

ill pa

ss a

sig

nific

ant p

ortio

n of

woo

d •

Woo

d de

bris

cap

acity

of c

ulve

rts c

an b

e as

sess

ed b

y ta

king

ratio

of c

ulve

rt di

amet

er

to c

hann

el w

idth

(w*)

•

Cro

ssin

gs w

ith lo

w w

* val

ues

(<1)

are

pro

ne to

deb

ris p

lugg

ing

• In

let b

asin

con

figur

atio

n al

so in

fluen

ces

woo

d pl

uggi

ng; c

hann

el w

iden

ing

upst

ream

of

culv

erts

pon

ds w

ater

and

cre

ates

edd

ies

that

orie

nts

woo

d pe

rpen

dicu

lar t

o flo

w a

nd

prom

oted

acc

umul

atio

n of

deb

ris p

iece

s, fo

rmin

g la

rge

jam

s up

stre

am o

f the

cul

vert

• W

hen

chan

nel e

nter

s cu

lver

t at a

n an

gle,

woo

d ca

nnot

rota

te p

aral

lel t

o flo

w a

nd p

ass

thro

ugh

culv

erts

Flan

agan

et a

l. 19

98

54

Tabl

e A

-13.

The

impl

icat

ions

of L

WD

man

agem

ent a

ctio

ns in

var

ious

geo

grap

hic

loca

tions

and

cha

nnel

net

wor

k po

sitio

ns (a

rrang

ed in

chr

onol

ogic

al o

rder

). LO

CAT

ION

C

HAN

NEL

NET

WO

RK

PO

SITI

ON

R

ESU

LTS

REF

EREN

CES

W

este

rn

Was

hing

ton

• 0.

07 c

hann

el g

radi

ent

• D

owns

tream

impa

cts

shou

ld b

e co

nsid

ered

in a

sses

sing

trad

e-of

fs a

ssoc

iate

d w

ith

debr

is re

mov

al a

nd in

crea

sed

sedi

men

tatio

n Be

scht

a 19

79

Sout

heas

t Ala

ska

• 15

km

2 to 7

5 km

2 dra

inag

e ar

eas

• 4th

thro

ugh

5th o

rder

stre

ams

• Le

ngth

of p

iece

, per

cent

of p

iece

in w

ater

, ang

le o

f orie

ntat

ion

to fl

ow, a

nd lo

catio

n of

an

chor

poi

nt a

ll in

fluen

ce s

tabi

lity

of in

stre

am w

ood

• Su

gges

t gen

eral

gui

delin

es b

ased

on

age

of d

ebris

and

stre

am g

radi

ent

Brya

nt 1

983

East

ern

Was

hing

ton

• 0.

015

chan

nel g

radi

ent

• 11

.5 m

ban

kful

l wid

th

• In

disc

rimin

ant r

emov

al o

f LW

D h

as m

ajor

sho

rt te

rm in

fluen

ces

on c

hann

el s

tabi

lity

• Kn

owle

dge

of s

ize

and

othe

r fea

ture

s co

mm

on to

sta

ble

debr

is p

rovi

ded

as b

asis

for

man

agem

ent g

uide

lines

•

Prop

osed

dic

hoto

mou

s ke

y to

det

erm

ine

whe

ther

to le

ave

or re

mov

e LW

D

Bilb

y198

4

Paci

fic N

orth

wes

t •

N/A

•

Activ

e st

ream

side

veg

etat

ion

can

incr

ease

fish

bio

mas

s by

man

agin

g fo

r lar

ger l

ight

fle

cks

and

mai

ntai

ning

inpu

t of L

WD

Se

dell

and

Swan

son

1984

Paci

fic N

orth

wes

t •

N/A

•

In w

ood

rein

trodu

ctio

n m

ust c

onsi

der t

hree

thin

gs: 1

) wha

t kin

d of

woo

d is

des

irabl

e, 2

) w

here

sho

uld

LWD

be

plac

ed, 3

) how

muc

h w

ood

is e

noug

h •

Smal

l acc

umul

atio

ns o

f var

ious

siz

ed p

iece

s of

woo

d pr

ovid

e m

ost c

ompl

ex a

nd b

est

utiliz

ed h

abita

ts in

all

size

s of

rive

rs a

nd s

tream

s •

Woo

d ac

cum

ulat

es a

t trib

utar

y ju

nctio

ns, w

here

val

ley

floor

wid

ens,

at c

hann

el b

ends

, an

d at

geo

logi

c co

nstri

ctio

ns

• St

ream

reac

hes

in w

iden

ed v

alle

y flo

ors

and

area

s be

low

trib

utar

y ju

nctio

ns c

an b

e be

tter e

nhan

ced

by in

trodu

ctio

n of

mul

tiple

tree

s in

diff

eren

t con

figur

atio

ns

Sede

ll et

al.

1988

Paci

fic N

orth

wes

t •

N/A

•

Leav

e un

dist

urbe

d bu

ffer s

trip

of o

ld g

row

th ti

mbe

r to

ensu

re lo

ng-te

rm s

uppl

y lo

ng,

larg

e di

amet

er lo

gs

• Le

ave

pred

eter

min

ed fr

actio

n of

tim

ber i

n bu

ffer s

trip

that

is a

dequ

ate

to s

atis

fy s

tream

ha

bita

t nee

ds a

nd a

llow

woo

d to

ent

er th

roug

h na

tura

l pro

cess

es

• R

emov

e tim

ber f

rom

stre

amsi

de m

anag

emen

t zon

e on

dou

ble

rota

tion

basi

s, w

here

tre

es a

re h

arve

sted

eve

ry o

ther

rota

tion

(100

to 1

50 y

ears

) •

Des

ign

and

engi

neer

buf

fers

that

mai

ntai

n LW

D in

put,

prov

ide

mix

of r

ipar

ian

spec

ies

• Va

riabl

e w

idth

buf

fers

offe

r cha

nce

to ta

ilor s

tream

side

buf

fers

to lo

cal s

tream

co

nditi

ons

• Se

lect

ivel

y lo

gged

buf

fers

mus

t lea

ve s

tabl

e lo

gs fo

r con

tribu

tion

to th

e st

ream

cha

nnel

•

LWD

is s

carc

e w

hen:

1) l

ack

of q

ualit

y po

ols

2) la

ck o

f LW

D s

tora

ge s

ites

3) p

rese

nce

of u

nsta

ble

debr

is w

hich

pos

es e

nviro

nmen

tal h

azar

d 4)

lack

of e

scap

e co

ver;

scar

city

be

st d

eter

min

ed b

y be

fore

and

afte

r stu

dies

•

LWD

is to

o ex

cess

ive

whe

n: 1

) blo

cks

fish

mig

ratio

n 2)

impa

irs w

ater

qua

lity

3)

pres

ence

of u

nsta

ble

debr

is w

hich

pos

es e

nviro

nmen

tal h

azar

d 4)

inte

rfere

nce

with

re

crea

tiona

l use

s

• N

eed

thor

ough

long

-term

test

ing

Biss

on e

t al.

1987

Paci

fic N

orth

wes

t •

N/A

•

LWD

add

ition

s ca

n be

effe

ctiv

e in

enh

anci

ng fi

sh h

abita

t in

the

shor

t-ter

m

• St

ruct

ures

mim

ic e

ffect

of n

atur

al o

bstru

ctio

ns in

stre

ams

• Th

e ne

ed fo

r stru

ctur

al e

lem

ents

var

ies

with

cha

nnel

gra

dien

t: 1)

<0.

01 g

radi

ent

chan

nels

hav

e lo

w to

mod

erat

e ne

ed, 2

) 0.0

1 to

0.0

10 g

radi

ent c

hann

els

have

hig

h ne

ed, 3

) >0.

10 g

radi

ent c

hann

els

have

low

nee

d

Sulliv

an e

t al.

1987

55

Tabl

e A

-13.

The

impl

icat

ions

of L

WD

man

agem

ent a

ctio

ns in

var

ious

geo

grap

hic

loca

tions

and

cha

nnel

net

wor

k po

sitio

ns (a

rrang

ed in

chr

onol

ogic

al o

rder

). LO

CAT

ION

C

HAN

NEL

NET

WO

RK

PO

SITI

ON

R

ESU

LTS

REF

EREN

CES

W

este

rn

Was

hing

ton

• 0.

13 c

hann

el g

radi

ent (

<7 m

ch

anne

l wid

th)

• 0.

08 c

hann

el g

radi

ent (

7 m

to 1

0 m

cha

nnel

wid

th)

• 0.

03 c

hann

el g

radi

ent (

>10

m

chan

nel w

idth

)

• Pr

oced

ures

to e

nsur

e LW

D s

uppl

y m

ust c

onsi

der:

1) c

lean

ing

pres

crip

tions

leav

e ap

prop

riate

am

ount

of w

ood

of p

rope

r siz

e in

stre

am, 2

) rem

aini

ng s

tream

side

ve

geta

tion

shou

ld p

rovi

de fo

r lon

g-te

rm in

put

• LW

D d

ata

mus

t be

appl

ied

to s

imila

r stre

ams

• D

ata

on v

aria

tions

in s

ize

and

amou

nt o

f woo

dy d

ebris

with

cha

ngin

g st

ream

siz

e co

uld

be u

sed

to d

evel

op p

lans

for n

umbe

rs a

nd s

izes

of t

rees

to b

e re

tain

ed

• Ef

fect

ive

man

agem

ent w

ill de

pend

on

info

rmat

ion

rela

ting

vege

tativ

e an

d ph

ysic

al

char

acte

ristic

s of

ripa

rian

area

to in

put o

f LW

D

Bilb

y an

d W

ard

1989

N/A

•

N/A

•

Stru

ctur

es te

nd to

lock

stre

ams

into

a re

lativ

ely

fixed

loca

tion

and

cond

ition

•

Impr

oved

man

agem

ent o

f stre

amsi

de v

eget

atio

n of

fers

mos

t pro

mis

e fo

r dev

elop

ing

valu

able

and

pro

duct

ive

ripar

ian

syst

ems

Elm

ore

and

Besc

hta

1989

UK

• N

/A

• LW

D m

anag

emen

t sho

uld

be u

nder

take

n w

ith k

now

ledg

e of

nat

ural

LW

D c

ondi

tions

•

Logg

ing

shou

ld m

inim

ize

disr

uptio

n to

cha

nnel

pro

cess

es

• M

anag

emen

t sho

uld

optim

ize

mai

nten

ance

of h

abita

t and

min

imiz

e ec

olog

ical

di

stur

banc

e •

Som

e ar

eas

may

requ

ire a

dditi

on o

f LW

D

Gre

gory

and

Dav

is 1

992

N/A

•

N/A

•

No-

net l

oss

polic

y in

fore

sted

bas

ins

in te

mpe

rate

USA

sho

uld

incl

ude

four

goa

ls: 1

) m

aint

ain

geom

orph

ic c

ompl

exity

2) r

etai

n LW

D a

long

cha

nnel

3) m

aint

ain

ripar

ian

vege

tatio

n 4)

pre

serv

atio

n of

hyd

rolo

gica

l fun

ctio

n of

wat

ersh

ed

• St

ream

cla

ssifi

catio

n sy

stem

s ba

sed

on g

eom

orph

ic c

hara

cter

istic

s ca

n pr

ovid

e m

eans

of

iden

tifyi

ng c

ritic

al ri

paria

n ar

eas

for r

esou

rce

use

• To

re-e

stab

lish

larg

e rip

aria

n tre

es s

ugge

st p

lant

ing

fast

gro

win

g co

nife

rs a

nd th

inni

ng

of a

lder

(Aln

us ru

bra)

ove

rsto

ry to

pro

mot

e gr

owth

of t

rees

Brya

nt a

nd S

edel

l 199

5

N/A

•

N/A

•

Hyd

raul

ic m

odel

s ca

n be

use

d to

hel

p pl

an d

ebris

man

agem

ent p

rogr

ams

• C

hann

el ro

ughn

ess

depe

nds

on m

any

fact

ors,

it is

unl

ikel

y th

at th

e ap

proa

ch o

f m

easu

ring

chan

ges

in ro

ughn

ess

coef

ficie

nt w

ill yi

eld

univ

ersa

l rel

atio

ns b

etw

een

debr

is ty

pe, q

uant

ity, a

nd h

ydra

ulic

effe

ct

• C

laim

of r

educ

ed fr

eque

ncy

of d

urat

ion

of o

verb

ank

flood

ing

used

to ju

stify

deb

ris

rem

oval

is n

ot u

nequ

ivoc

ally

pro

ven

in fi

eld

Gip

pel 1

995

N/A

•

N/A

•

Prob

lem

of d

ebris

man

agem

ent r

esul

ting

from

logg

ing

best

dea

lt w

ith s

elec

tive

rem

oval

to

ens

ure

mai

nten

ance

of c

hann

el s

tabi

lity,

fish

acc

ess,

and

fish

hab

itat

• G

uide

lines

for r

emov

al ta

ke in

to a

ccou

nt a

mou

nt, s

ize

and

age

of d

ebris

pre

sent

•

Effe

ct o

f pre

viou

s de

bris

cle

aran

ce c

an b

e re

vers

ed b

y ad

ditio

n of

deb

ris to

the

river

•

Varia

bilit

y in

siz

e an

d sp

acin

g of

LW

D a

ccum

ulat

ions

that

occ

ur lo

cally

in u

nman

aged

ch

anne

ls s

houl

d gu

ide

intro

duct

ion

of L

WD

•

Intro

duct

ion

of L

WD

is s

hort-

term

sol

utio

n, p

rodu

ctio

n of

nat

ural

inpu

ts is

pre

fera

ble

• M

aint

ain

buffe

r zon

e to

ens

ure

supp

ly o

f woo

d •

Activ

e m

anag

emen

t of b

uffe

r zon

e ca

n yi

eld

addi

tiona

l ben

efits

suc

h as

ligh

t, te

mpe

ratu

re, f

low

, sed

imen

t tra

nspo

rt, a

nd s

edim

ent c

ondi

tions

Gur

nell

et a

l. 19

95

Sout

hwes

t Ala

ska

Wes

tern

W

ashi

ngto

n

• 0.

002

to 0

.085

cha

nnel

gra

dien

ts

• 2.

5 to

38

m c

hann

el w

idth

s •

<40%

of i

n-ch

anne

l LW

D a

ctiv

e in

poo

l for

mat

ion,

so

effe

ctiv

e rip

aria

n zo

ne

man

agem

ent m

ust c

onsi

der n

atur

al in

effic

ienc

y in

recr

uitin

g do

min

ant L

WD

•

Nat

ural

recr

uitm

ent o

f man

y lo

gs is

requ

ired

to p

rovi

de a

pie

ce th

at is

situ

ated

to fo

rm a

po

ol,

• Th

e si

ze a

nd p

lace

men

t of L

WD

may

be

engi

neer

ed to

mor

e ef

ficie

ntly

cat

alyz

e po

ol

form

atio

n •

Cou

ld u

se p

ool s

paci

ng (c

hann

el w

idth

s pe

r poo

l) ag

ains

t LW

D fr

eque

ncy

(pie

ces

per

met

er) t

o de

fine

man

agem

ent o

bjec

tives

rega

rdin

g po

ol s

paci

ng a

nd L

WD

freq

uenc

y

Mon

tgom

ery

et a

l. 19

95

56

Tabl

e A

-13.

The

impl

icat

ions

of L

WD

man

agem

ent a

ctio

ns in

var

ious

geo

grap

hic

loca

tions

and

cha

nnel

net

wor

k po

sitio

ns (a

rrang

ed in

chr

onol

ogic

al o

rder

). LO

CAT

ION

C

HAN

NEL

NET

WO

RK

PO

SITI

ON

R

ESU

LTS

REF

EREN

CES

N

orth

wes

t W

ashi

ngto

n •

0.00

5 to

0.0

1 ch

anne

l gra

dien

ts

• 30

m to

80

m b

ankf

ull w

idth

s •

75 k

m2 to

225

km

2 dra

inag

e ar

eas

• M

anag

emen

t act

iviti

es m

ust e

nsur

e ad

equa

te re

crui

tmen

t of l

arge

st L

WD

from

ripa

rian

fore

st

• Se

lect

ive

rem

oval

of l

arge

st tr

ees

from

ripa

rian

and

flood

plai

n fo

rest

s w

ill ha

ve m

ajor

im

pact

s on

in-c

hann

el h

abita

t cha

ract

eris

tics

Abbe

and

Mon

tgom

ery

1996

Nor

thw

est

Was

hing

ton

• 0.

002

to 0

.05

chan

nel g

radi

ents

•

5 m

to 2

0 m

cha

nnel

wid

ths

• 2

km2 to

120

km

2 dra

inag

e ar

eas

• Pr

edic

t dec

line

in n

umbe

r and

are

a of

poo

ls in

low

to m

oder

ate

slop

e ch

anne

ls w

ith

redu

ctio

n in

LW

D

• G

iven

sam

e de

crea

se in

LW

D, p

redi

ct g

reat

er d

ecre

ases

in n

umbe

r and

are

a of

poo

ls

in m

oder

ate

slop

e ch

anne

ls th

an lo

w s

lope

cha

nnel

s •

Expe

ct d

ecre

ases

in a

bund

ance

of s

peci

es th

at s

how

stro

ng p

refe

renc

e fo

r poo

ls a

s re

arin

g lo

catio

ns (O

ncor

hync

hus

kisu

tch)

and

incr

ease

s in

abu

ndan

ce o

f spe

cies

sui

ted

to re

arin

g in

riffl

e en

viro

nmen

ts (O

ncor

hync

hus

myk

iss)

•

Afte

r log

ging

exp

ect i

ncre

ases

in n

umbe

r and

are

a of

poo

ls to

be

mor

e ra

pid

in

mod

erat

e sl

ope

chan

nels

than

in lo

w s

lope

cha

nnel

s •

Whe

n LW

D p

iece

s pe

r met

er e

xcee

ds 0

.4, p

ool s

paci

ng is

less

sen

sitiv

e to

incr

ease

d lo

adin

g •

Foun

d re

latio

nshi

p be

twee

n ch

anne

l wid

th a

nd s

ize

of L

WD

form

ing

pool

s •

In c

hann

els

<10

m w

idth

, log

s of

20

cm d

iam

eter

form

ed p

ools

; LW

D fr

om s

econ

d gr

owth

will

form

poo

ls s

oone

r in

smal

l cha

nnel

; poo

l abu

ndan

ce m

ay in

crea

se a

fter 2

5 ye

ars

thro

ugh

inpu

ts o

f dec

iduo

us v

eget

atio

n •

In c

hann

els

20 m

wid

e, p

ools

do

not f

orm

unt

il LW

D e

xcee

ds 6

0cm

; sig

nific

ant

incr

ease

s in

poo

l num

ber a

nd a

rea

may

not

beg

in fo

r 75

year

s un

til th

e re

crui

tmen

t of

larg

e co

nife

rs

• M

ay b

e po

ssib

le to

acc

eler

ate

reco

very

of p

re-lo

ggin

g co

nditi

ons

thro

ugh

man

agem

ent

of ri

paria

n fo

rest

•

Thin

ning

may

ben

efit

larg

e st

ream

s w

hich

requ

ire la

rge

logs

to fo

rm p

ools

•

Thin

ning

may

not

ben

efit

smal

ler s

tream

s, a

s sm

all w

ood

is s

uffic

ient

to fo

rm p

ools

Beec

hie

and

Sibl

ey 1

997

N/A

•

N/A

•

Deb

ris ro

ughn

ess

coul

d be

qua

ntifi

ed fr

om a

kno

wle

dge

of c

hann

el g

eom

etry

, and

tree

he

ight

and

dia

met

er

• St

udie

s sh

ow h

ow w

ood

stru

ctur

e an

d tra

nspo

rt sh

ould

cha

nge

as a

func

tion

of p

ositi

on

with

in th

e ch

anne

l net

wor

k; th

is g

eom

orph

ic c

onte

xt n

eeds

to b

e un

ders

tood

to g

uide

ef

fect

ive

plac

emen

t of w

ood

for l

ong-

term

sta

bilit

y •

Qua

ntita

tive

unde

rsta

ndin

g of

woo

d m

ovem

ent i

s ne

eded

to a

sses

s st

abilit

y of

woo

d ad

ded,

and

to p

reve

nt c

onge

sted

tran

spor

t fro

m p

osin

g en

viro

nmen

tal h

azar

d

Brau

dric

k et

al.

1997

Sout

heas

tern

Fr

ance

•

0.02

8 to

0.0

31 c

hann

el g

radi

ents

•

1640

km

2 •

Met

hodo

logi

cal a

ppro

ache

s ca

n be

pro

pose

d to

hel

p m

anag

ers

iden

tify

reac

hes

on

whi

ch v

eget

atio

n co

rrido

rs c

ould

be

cons

erve

d, re

habi

litat

ed, o

r use

d fo

r sof

t m

aint

enan

ce

• D

evel

opm

ent o

f inc

reas

ingl

y m

atur

e rip

aria

n fo

rest

, due

in p

art t

o in

cisi

on a

nd

decr

ease

d la

tera

l mig

ratio

n, h

as le

d to

incr

ease

d tra

nspo

rt of

woo

d de

bris

, inc

reas

ing

risk

of lo

g ja

m fo

rmat

ion

• N

eed

to m

aint

ain

late

ral c

onne

ctiv

ity b

etw

een

river

and

ripa

rian

vege

tatio

n •

Prop

ose

a se

ctor

ed m

aint

enan

ce p

lan

base

d on

map

ping

of w

oody

deb

ris s

uppl

y ar

eas,

are

as w

here

log

jam

form

atio

n is

mos

t pro

babl

e, a

nd fl

oodp

lain

occ

upat

ion

to

dete

rmin

e vu

lner

abilit

y to

floo

ding

Pieg

ay a

nd L

ando

n 19

97

Paci

fic N

orth

wes

t •

N/A

•

Man

agem

ent a

ppro

ach

mus

t rec

ogni

ze p

roce

sses

resp

onsi

ble

for L

WD

del

iver

y •

Appr

oach

mus

t als

o ad

dres

s m

anag

emen

t rel

ated

cha

nges

in L

WD

cha

ract

eris

tics

• R

ecom

men

d ad

aptiv

e m

anag

emen

t pro

cess

Bilb

y an

d Bi

sson

199

8

57

LITERATURE CITED

Abbe, T. B. and D. R. Montgomery. 1996. Large woody debris jams, channel hydraulics and habitat formation in large rivers. Regulated Rivers: Research and Management 12: 210-221.

Adenlof, K. A. and E. E. Wohl. 1994. Controls on bedload movement in a Sub-Alpine Stream of the Colorado Rocky-Mountains, Usa. Arct. Alp. Res. 26: 77-85.

Anderson, N.H., J.R. Sedell, M. Roberts, and F.J. Triska. 1978. The role of aquatic invertebrates in processing of wood debris in coniferous forest streams. The American Midland Naturalist 100(1): 64-82.

Andrus, C. W., B. A. Long, and H. A. Froehlich. 1988. Woody debris and its contribution to pool formation in a coastal stream 50 years after logging. Canadian Journal of Fisheries and Aquatic Management 45: 2080-2086.

Angermeier, P. L. and J. R. Karr. 1984. Relationships between woody debris and fish habitat in a small warmwater stream. Transactions of the American Fisheries Society 113: 716-726.

Assani, A. A. and F. Petit. 1995. Log-Jam Effects On Bed-Load Mobility from Experiments Conducted in a Small Gravel-Bed Forest Ditch. Catena 25: 117-126.

Aumen, N. G., C. P. Hawkins, and S. V. Gregory. 1990. Influence of woody debris on nutrient retention in catastrophically disturbed streams. Hydrobiologia 190: 183-192.

Beechie, T.J. and S. Bolton. 1999. An approach to restoring salmonid habitat forming processes in Pacific Northwest watersheds. Fisheries: 24(4): 6-15.

Beechie, T. J. and T. H. Sibley. 1997. Relationships between channel characteristics, woody debris, and fish habitat in northwestern Washington streams. Transactions of the American Fisheries Society 126: 217-229.

Berg, N., A. Carlson, and D. Azuma. 1998. Function and dynamics of woody debris in stream reaches in the central Sierra Nevada, California. Canadian Journal of Fisheries and Aquatic Sciences 55: 1807-1820.

Beschta, R. L. 1979. Debris removal and its effects on sedimentation in an Oregon Coast Range stream. Northwest Science 53: 71-77.

Bilby, R. E. 1981. Role of organic debris dams in regulating the export of dissolved organic and particulate matter from a forested watershed. Ecology 62: 1234-1243.

Bilby, R. E. 1984. Removal of woody debris may affect stream channel stability. Journal of Forestry 82: 609-613.

Bilby, R. E. and P. A. Bisson. 1998. Functioning and distribution of large woody debris. River Ecology and Management. In: River Ecology and Management, R. J. Naiman and R. E. Bilby, eds: 324-346. Springer, New York, New York, USA.

58

Bilby, R. E. and G. E. Likens. 1980. Importance of organic debris dams in the structure and function of stream ecosystems. Ecology 61: 1107-1113.

Bilby, R. E. and J. W. Ward. 1989. Changes in characteristics and function of woody debris with increasing size of streams in western Washington. Transactions of the American Fisheries Society 118: 368-378.

Bilby, R. E. and J. W. Ward. 1991. Characteristics and function of large woody debris in streams draining old growth, clear-cut, and second-growth forests in southwestern Washington. Canadian Journal of Fisheries and Aquatic Sciences 48: 2499-2508.

Bisson, P. A., R. E. Bilby, M. D. Bryant, C. A. Dolloff, G. B. Grette, R. A. House, M. L. Murphy, K. V. Koski, and J. R. Sedell. 1987. Large woody debris in forested streams in the Pacific Northwest: past, present, and future. Streamside Management: Forestry and Fishery Interactions. E. O. Salo and T. W. Cundy. Seattle, Washington, University of Washington, Institute of Forest Resources: 143-190.

Bisson, P. A. and D. R. Montgomery. 1996. Valley segments, stream reaches, and channel units. Methods in Stream Ecology. Hauer R. and G.A. Lamberti, Acdemic Press, San Diego, California, USA.

Bisson, P.A. and J. R. Sedell. 1984. Salmonid populations in streams in clearcut vs. old-growth forests of western Washington. In: Fish and Wildlife Relationships in Old Growth Forests, W.R. Meehan, T.R. Merrel, T.A. Hanley, eds. American Institute of Fisheries Biologists, Juneau, Alaska, USA.

Bjornn, T. C. and D.W. Reiser. 1991. Habitat requirements of salmonids in streams. American Fisheries Society Special Publication 19: 83-138.

Boussu, M. F. 1954. Relationship between trout populations and cover on a small stream. Journal of Wildlife Management 18: 229-239.

Braudrick, C. A. and G. E. Grant. 2000. When do logs move in rivers? Water Resources Research 36: 571-584.

Braudrick, C. A., G. E. Grant, Y. Ishikawa, and H. Ikeda. 1997. Dynamics of wood transport in streams: A flume experiment. Earth Surface Processes and Landforms 22: 669-683.

Brown, G. W. 1974. Fish habitat. Environmental effects of forest residues management in the Pacific Northwest: a state-of-knowledge compendium. O. P. Cramer. Portland, Oregon, USDA Forest Service. PNW-124: E1-E15.

Bryant, M. D. 1980. Evolution of large, organic debris after timber harvest: Maybeso Creek, 1949 to 1978. USDA Forest Service, General Technical Report, PNW-101.

Bryant, M. D. 1983. The role and management of woody debris in west coast salmonid nursery streams. North American Journal of Fisheries Management 3: 322-330.

Bryant, M. D. 1985. Changes 30 Years after Logging in Large Woody Debris and Its Use by Salmonids. USDA Forest Service, General Technical Report, RM-120.

59

Bryant, M.D. and J.R. Sedell. 1995. Riparian forests, wood in the water, and fish habitat complexity. Pages 202-223 in N.B. Armantrout and J.R. Sedell (eds) Condition of the World's Aquatic Habitats. Proceedings of the World Fisheries Congress, Theme 1, American Fisheries Society, Bethesda, MD.

Carlson, J. Y., C. W. Andrus, and H. A. Froehlich. 1990. Woody debris, channel features, and macroinvertebrates of streams with logged and undisturbed riparian timber in northeastern Oregon, USA. Canadian Journal of Fisheries and Aquatic Sciences 47: 1103-1111.

Cederholm, C. J., R. E. Bilby, P. A. Bisson, T. W. Bumstead, B. R. Fransen, W. J. Scarlett, and J. W. Ward. 1997. Response of juvenile coho salmon and steelhead to placement of large woody debris in a coastal Washington stream. North American Journal of Fisheries Management 17: 947-963.

Cederholm, C. J., D. B. Houston, D. L. Cole, and W. J. Scarlett. 1989. Fate of coho salmon (Oncorhynchus kisutch) carcasses in spawning streams. Canadian Journal of Fisheries and Aquatic Sciences 46: 1347-1355.

Cederholm, C. J. and N. P. Peterson. 1985. The retention of Coho salmon Oncorhynchus kisutch carcasses by organic debris in small streams. Canadian Journal of Fisheries and Aquatic Sciences 42: 1222-1225.

Crispin, V., R. House, and D. Roberts. 1993. Changes in instream habitat large woody debris and salmon habitat after the restructuring of a coastal Oregon stream. North American Journal of Fisheries Management 13: 96-102.

Culp, J. M., G. J. Scrimgeour, and G. D. Townsend. 1996. Simulated fine woody debris accumulations in a stream increase rainbow trout fry abundance. Transactions of the American Fisheries Society 125: 472-479.

Davis, R. J. and K. J. Gregory. 1994. A new distinct mechanism of river bank erosion in a forested catchment. Journal of Hydrology 157: 1-11.

Diehl, T. H. 1997. Potential Drift Accumulation at Bridges. US Department of Transportation, Federal Highway Transportation, FHWA-RD-97-028.

Dolloff, C. A. 1986. Effects of stream cleaning on juvenile coho salmon and Dolly Varden in southeast Alaska [USA]. Transactions of the American Fisheries Society 115: 743-755.

Dolloff, C. A. and G. H. Reeves. 1990. Microhabitat partitioning among stream-dwelling juvenile coho salmon, Oncorhynchus kisutch, and Dolly Varden, Salvelinus malma. Canadian Journal of Fisheries and Aquatic Sciences 47: 2297-2306.

Dominguez, L. G. and C. J. Cederholm. 2000. Rehabilitating stream channels using large woody debris with considerations for salmonid life history and fluvial geomorphic processes. In: Sustainable Fisheries Management: Pacific Salmon, E. E. Knudsen. C. R. Steward, D. D. MacDonald, J. E. Williams, and D. W. Reiser, eds: 545-563. Lewis Publishers, New York, New York, USA.

Dudley, S. J., J. C. Fischenich, and S. R. Abt. 1998. Effect of woody debris entrapment on flow resistance. Journal Of The American Water Resources Association 34: 1189-1197.

60

Dudley, T, and N. H. Anderson. 1982. A survey of invertebrates associated with wood debris in aquatic habitats. Melanderia 39: 1-21.

Elliott, S. T. 1986. Reduction of a Dolly Varden [Salvelinus malma] population and macrobenthos after removal of logging debris. Transactions of the American Fisheries Society 115: 392-400.

Elmore, W. and R. L. Beschta. 1989. The fallacy of structures and the fortitude of vegetation. California Riparian Systems Conference, University of California, Davis: 117-119. USDA Forest Service General Technical Report, PSW-110.

Elosegi, A., J. R. Diez, and J. Pozo. 1999. Abundance, characteristics, and movement of woody debris in four Basque streams. Archiv Fur Hydrobiologie 144: 455-471.

Everett, R. A. and G. M. Ruiz. 1993. Coarse woody debris as a refuge from predation in aquatic communities an experimental test. Oecologia (Berlin) 93: 475-486.

Fausch, K. D. and T. G. Northcote. 1992. Large woody debris and salmonid habitat in a small coastal British Columbia stream. Canadian Journal of Fisheries and Aquatic Sciences 49: 682-693.

Fetherston, K. L., R. J. Naiman, and R. E. Bilby. 1995. Large woody debris, physical process, and riparian forest development in montane river networks of the Pacific Northwest. Geomorphology 13: 133-144.

Flanagan, S. A., M. J. Furniss, T. S. Ledwith, S. Theissen, M. Love, K. Moore, and J. Ory. 1998. Methods for inventory and environmental risk assessment of road drainage crossings. USDA Forest Service, Pacific Southwest Region.

Flebbe, P. A. 1999. Trout use of woody debris and habitat in Wine Spring Creek, North Carolina. Forest Ecology and Management 114: 367-375.

Flebbe, P. A. and C. A. Dolloff. 1995. Trout use of woody debris and habitat in Appalachian wilderness streams of North Carolina. North American Journal of Fisheries Management 15: 579-590.

Froehlich, H. A. 1970. Logging debris: managing a problem. Forest Land Uses and Stream Environment, Corvallis, Oregon: 112-117. Oregon State University.

Froehlich, H.A. 1973. Natural and man-caused slash in headwater streams. Loggers Handbook XXXIII. Pacific Logging Congress, Portland, Oregon, USA.

Furniss, M. J., T. D. Roelofs, and C. S. Yee. 1991. Road construction and maintenance. American Fisheries Society Special Publication 19: 297-325.

Gippel, C. J. 1995. Environmental hydraulics of large woody debris in streams and rivers. Journal of Environmental Engineering-ASCE 121: 388-395.

Gippel, C. J., B. L. Finlayson, and I. C. O'Neill. 1996. Distribution and hydraulic significance of large woody debris in a lowland Australian river. Hydrobiologia 318: 179-194.

Gowan, C. and K. D. Fausch. 1996. Long-term demographic responses of trout populations to habitat manipulation in six Colorado streams. Ecological Applications 6: 931-946.

61

Gregory, K.J. and R.J. Davis. 1992. Coarse woody debris in stream channels in relation to river channel management in woodland areas. Regulated Rivers: Research and Management 7: 117-136.

Gregory, K. J., R. J. Davis, and S. Tooth. 1993. Spatial-distribution of coarse woody debris dams in the Lymington Basin, Hampshire, UK. Geomorphology 6: 207-224.

Gregory, S.V., F.J. Swanson, W.A. McKee, and K.W. Cummins. 1991. An ecosystem perspective of riparian zones. BioScience 41(8): 541-551.

Grizzel, J. D. and N. Wolff. 1998. Occurrence of windthrow in forest buffer strips and its effect on small streams in northwest Washington. Northwest Science 72: 214-223.

Gurnell, A. M., K. J. Gregory, and G. E. Petts. 1995. The Role of Coarse Woody Debris in Forest Aquatic Habitats - Implications for Management. Aquatic Conservation: Marine and Freshwater Ecosystems 5: 143-166.

Gurnell, A. M. and R. Sweet. 1998. The distribution of large woody debris accumulations and pools in relation to woodland stream management in a small, low- gradient stream. Earth Surface Processes and Landforms 23: 1101-1121.

Hall, J. D. and R. L. Lantz. 1968. Effects of logging on the habitat of coho salmon and cutthroat trout in coastal streams. Symposium on Salmon and Trout in Streams, University of British Columbia: 355-375. Institute of Fisheries, University of British Columbia.

Harmon, M. E., J. F. Franklin, F. J. Swanson, P. Sollins, S. V. Gregory, J. D. Lattin, N. H. Anderson, S. P. Cline, N. G. Aumen, J. R. Sedell, G. W. Lienkamper, et al. 1986. Ecology of Coarse Woody Debris in Temperate Ecosystems. Advances in Ecological Research. A. MacFadyen and E. D. Ford. New York, NY, Harcourt Brace Jovanovich. 15: 133-302.

Hartman, G. F., J. C. Scrivener, and T. E. McMahon. 1987. Saying that logging is either "good" or "bad" for fish doesn't tell you how to manage the system. The Forestry Chronicle 159-164.

Harvey, B. C. 1998. Influence of large woody debris on retention, immigration, and growth of coastal cutthroat trout (Oncorhynchus clarki clarki) in stream pools. Canadian Journal of Fisheries and Aquatic Sciences 55: 1902-1908.

Hax, C.L. and S.W. Golladay. 1993. Macroinvertebrate colonization and biofilm development and leaves and wood in a boreal river. Freswater Biology 29:79-87.

Hedin, L. O. 1990. Factors controlling sediment community respiration in woodland stream ecosystems. Oikos 57: 94-105.

Hedman, C. W., D. H. Van Lear, and W. T. Swank. 1996. In-stream large woody debris loading and riparian forest seral stage associations in the southern Appalachian Mountains. Canadian Journal of Forest Research 26: 1218-1227.

Heede, B. H. 1972. Flow and Channel Characteristics of Two High Mountain Streams. USDA Forest Service, Research Paper, RM-96.

Heede, B. H. 1985a. Channel adjustments to the removal of log steps: and experiment in a mountain stream. Environmental Management 9: 427-432.

62

Heede, B. H. 1985b. Interactions between streamside vegetation and stream dynamics. USDA Forest Service, General Technical Report, RM-120.

Heifetz, J., M. L. Murphy, and K. V. Koski. 1986. Effects of logging on winter habitat of juvenile salmonids in Alaskan [USA] streams. North American Journal of Fisheries Management 6: 52-58.

Hickin, E. J. 1984. Vegetation and river channel dynamics. Canadian Geographer 28: 111-126.

Hilderbrand, R. H., A. D. Lemly, C. A. Dolloff, and K. L. Harpster. 1997. Effects of large woody debris placement on stream channels and benthic macroinvertebrates. Canadian Journal of Fisheries and Aquatic Sciences 54: 931-939.

Hogan, D. L., S. A. Bird, and M. A. Hassan. 1998. Spatial and temporal evolution of small coastal gravel-bed streams: influence of forest management on channel morphology and fish habitats. Gravel-Bed Rivers in the Environment. P. C. Klingeman, R. L. Beschta, P. D. Komar and J. B. Bradley. Highlands Ranch, Colorado, Water Resources Publications: 1701-1720.

House, R. A. and P. L. Boehne. 1985. Evaluation of instream structures for salmonid spawning and rearing in a coastal Oregon stream. North American Journal of Fisheries Management 5: 283-295.

House, R. A. and P. L. Boehne. 1986. Effects of instream structures on salmonid habitat and populations in Tobe Creek, Oregon. North American Journal of Fisheries Management 6: 38-46.

Keller, E. A. and A. MacDonald. 1983. Large Organic Debris and Anadromous Fish Habitat in the Coastal Redwood Environment: The Hydrologic System. California Water Resources Center, University of California, Davis, Technical completion report.

Keller, E. A., A. MacDonald, T. Tally, and N. J. Merritt. 1985. Effects of large organic debris on channel morphology and sediment storage in selected tributaries of Redwood Creek, northwestern California. Geomorphic Processes and Aquatic Habitat in the Redwood Creek Basin, Northwestern California. K. M. Nolan, H. M. Kelsey and D. C. Maron. Vickburg, MS, US Geological Survey: 29.

Keller, E. A. and F. J. Swanson. 1979. Effects of large organic material on channel form and fluvial processes. Earth Surface Processes and Landforms 4: 361-380.

Keller, E. A. and T. Tally. 1979. Effects of large organic debris on channel form and fluvial processes in the Coastal Environment. Adjustments of the Fluvial System: Tenth Annual Geomorphology Symposium, Binghamton, New York: 361-390. Kendall Hunt Publishing.

Leopold, L. B., M. G. Wolman, J.P. Miller. 1964. Fluvial Processes in Geomorphology. W.H. Freeman and Company, San Francisco, CA, USA.

Lienkaemper, G. W. and F. J. Swanson. 1987. Dynamics of large woody debris in streams in old-growth Douglas-fir forests. Canadian Journal of Forest Research 17: 150-156.

Likens, G. E. and R. E. Bilby. 1982. Development, Maintenance, and Role of Organic-Debris Dams in New England Streams. USDA Forest Service, General Technical Report, PNW-141.

63

Lisle, T. E. 1986a. Effects of woody debris on anadromous salmonid habitat, Prince of Wales Island, southeast Alaska [USA]. North American Journal of Fisheries Management 6: 538-550.

Lisle, T. E. 1986b. Stabilization of a gravel channel by large streamside obstructions and bedrock bends, Jacoby Creek, northwestern California. Geological Society of America Bulletin 97: 999-1011.

Lisle, T. E. and M. B. Napolitano. 1998. Effects of recent logging on the main channel of North Fork Caspar Creek. In: Proceedings of the conference on coastal watersheds: The Caspar Creek Story, R. R. Zeimer eds. USDA Forest Service General Technical Report PSW-GTR 168: 81-86.

Martin, D. J., L. J. Wasserman, and V. H. Dale. 1986. Influence of riparian vegetation on posteruption survival of coho salmon [Oncorhynchus kisutch] fingerlings on the west-side streams of Mount St. Helens, Washington [USA]. North American Journal of Fisheries Management 6: 1-8.

Marzolf, G. R. 1978. The Potential Effects of Clearing and Snagging on Stream Ecosystems. US Fish and Wildlife Service, Biological Services Program, FWS/OBS-78/14.

McDade, M. H., F. J. Swanson, W. A. McKee, J. F. Franklin, and J. Van Sickle. 1990. Source distances for coarse woody debris entering small streams in western Oregon and Washington [USA]. Canadian Journal of Forest Research 20: 326-330.

McHenry, M. L., E. Shott, R. H. Conrad, and G. B. Grette. 1998. Changes in the quantity and characteristics of large woody debris in streams of the Olympic Peninsula, Washington, U.S.A. (1982-1993). Canadian Journal of Fisheries and Aquatic Sciences 55: 1395-1407.

McMahon, T. E. and G. F. Hartman. 1989. Influence of cover complexity and current velocity on winter habitat use by juvenile coho salmon (Oncorhynchus kisutch). Canadian Journal of Fisheries and Aquatic Sciences 46: 1551-1557.

McMahon, T. E. and L. B. Holtby. 1992. Behavior, habitat use, and movements of coho salmon (Oncorhynchus kisutch) smolts during seaward migration. Canadian Journal of Fisheries and Aquatic Sciences 49: 1478-1485.

Megahan, W. F. 1982. Channel Sediment Storage Behind Obstructions in Forested Drainage Basins Draining the Granitic Bedrock of the Idaho Batholith. USDA Forest Service, General Technical Report, PNW-141.

Merritt, R.W. and K.W. Cummins eds. 1996. An Introduction to the Aquatic Insects of North America. Kendall-Hunt Publishing Company, Dubuque, Iowa, USA.

Montgomery, D. R., T. B. Abbe, J. M. Buffington, N. P. Peterson, K. M. Schmidt, and J. D. Stock. 1996. Distribution of bedrock and alluvial channels in forested mountain drainage basins. Nature 381: 587-589.

Montgomery, D. R. and J. M. Buffington. 1997. Channel-reach morphology in mountain drainage basins. Geol. Soc. Am. Bull. 109: 596-611.

64

Montgomery, D. R., J. M. Buffington, R. D. Smith, K. M. Schmidt, and G. Pess. 1995. Pool Spacing In Forest Channels. Water Resources Research 31: 9.

Moore, K. M. S. and S. V. Gregory. 1988. Response of young-of-the-year cutthroat trout to manipulation of habitat structure in a small stream. Transactions of the American Fisheries Society 117: 162-170.

Murphy, M. L., J. Heifetz, S. W. Johnson, K. V. Koski, and J. F. Thedinga. 1986. Effects of clear-cut logging with and without buffer strips on juvenile salmonids in Alaskan streams [USA]. Canadian Journal of Fisheries and Aquatic Sciences 43: 1521-1533.

Murphy, M. L. and K. V. Koski. 1989. Input and depletion of woody debris in Alaska Streams and implications for streamside management. North American Journal of Fisheries Management 9: 427-436.

Myers, T. J. and S. Swanson. 1996. Long-term aquatic habitat restoration: Mahogany Creek, Nevada, as a case study. Water Resources Bulletin 32: 241-252.

Myers, T. J. and S. Swanson. 1996. Temporal and geomorphic variations of stream stability and morphology: Mahogany Creek, Nevada. Water Resources Bulletin 32: 253-265.

Nakamoto, R. J. 1998. Effects of timber harvest on aquatic invertebrates and habitat in North Fork Caspar Creek. In: Proceedings of the conference on coastal watersheds: The Caspar Creek Story, R. R. Zeimer eds. USDA Forest Service General Technical Report PSW-GTR 168: 87-96.

Nakamura, F. S. and F. J. Swanson. 1993. Effects of coarse woody debris on morphology and sediment storage of a mountain stream system in western Oregon. Earth Surface Processes and Landforms 18: 43-61.

Nakamura, F. and F. J. Swanson. 1994. Distribution of coarse woody debris in a mountain stream, western Cascade range, Oregon. Canadian Journal of Forest Research 24: 2395-2403.

Napolitano, M. B. 1998. Persistence of historical logging impacts on channel form in mainstem North Fork Caspar Creek. In: Proceedings of the conference on coastal watersheds: The Caspar Creek Story, R. R. Zeimer eds. USDA Forest Service General Technical Report PSW-GTR 168: 97-102.

Narver, D. W. 1970. Effects of logging debris on fish production. Forest Land Uses and Stream Environment, Corvallis, Oregon: 101-111. Oregon State University.

Newbold, J. D., J. W. Elwood, R. V. O'Neill, and W. Van Winkle. 1982. Nutrient spiraling in streams: implications for nutrient limitation and invertebrate activity. American Naturalist (120): 628-652.

O'Connor, M. D. 1986. Effects of Logging on Organic Debris Dams in First order Streams in Northern California. Master of Science. University of California, Berkeley, CA, 90pp.

O'Connor, M.D. 1994. Sediment Transport in Steep Tributary Streams and the Influence on Large Organic Debris. PhD Dissertation. University of Washington, Seattle, WA. 254pp. revised

65

O' Connor, M. D. and R. R. Zeimer. 1989. Coarse wood debris ecology in a second growth Sequoia sempervirens forest stream. California Riparian Systems Conference, University of California, Davis: 165-171. USDA Forest Service General Technical Report, PSW-110.

O'Connor, N. A. 1991. The effects of habitat complexity on the macroinvertebrates colonising wood substrates in a lowland stream. Oecologia 85: 504-512.

Palmer, M. A., P. Arensburger, A. P. Martin, and D. W. Denman. 1996. Disturbance and patch-specific responses: The interactive effects of woody debris and floods on lotic invertebrates. Oecologia (Berlin) 105: 247-257.

Pereira, C.R.D, N. H. Anderson, and T. Dudley. 1982. Gut content analysis of aquatic insects from wood substrates. Melanderia 39:23-33

Phillips, E. C. 1994. Habitat preference and seasonal abundance of trichoptera larvae in Ozark streams, Arkansas. Journal of Freshwater Ecology 9: 91-95.

Piegay, H. 1993. Nature, mass and preferential sites of coarse woody debris deposits in the lower Ain Valley (Mollon reach), France. Regulated Rivers: Research and Management 8: 359-372.

Piegay, H. and N. Landon. 1997. Promoting ecological management of riparian forests on the Drome river, France. Aquatic Conservation: Marine And Freshwater Ecosystems 7: 287-304.

Piegay, H. and R. A. Marston. 1998. Distribution of large woody debris along the outer bend of meanders in the Ain River, France. Physical Geography 19: 318-340.

Piegay, H., A. Thevenet, and A. Citterio. 1999. Input, storage and distribution of large woody debris along a mountain river continuum, the Drome River, France. Catena 35: 19-39.

Quinn, T. P. and N. P. Peterson. 1996. The influence of habitat complexity and fish size on over-winter survival and growth of individually marked juvenile coho salmon (Oncorhynchus kisutch) in Big Beef Creek, Washington. Canadian Journal of Fisheries and Aquatic Sciences 53: 1555-1564.

Ralph, S. C., G. C. Poole, L. L. Conquest, and R. J. Naiman. 1994. Stream channel morphology and woody debris in logged and unlogged basins of western Washington. Canadian Journal of Fisheries and Aquatic Sciences 51: 37-51.

Reid, L. M. and S. Hilton. 1998. Buffering the buffer. In: Proceedings of the conference on coastal watersheds: The Caspar Creek Story, R. R. Zeimer eds. USDA Forest Service General Technical Report PSW-GTR 168: 71-80.

Richmond, A. D. and K. D. Fausch. 1995. Characteristics and function of large woody debris in subalpine Rocky Mountain streams in northern Colorado. Canadian Journal of Fisheries and Aquatic Sciences 52: 1789-1802.

Riley, S. C. and K. D. Fausch. 1995. Trout population response to habitat enhancement in six northern Colorado streams. Canadian Journal of Fisheries and Aquatic Sciences 52: 34-53.

66

Robison, E. G. and R. L. Beschta. 1990. Coarse woody debris and channel morphology interactions for undisturbed streams in southeast Alaska, USA. Earth Surface Processes and Landforms 15: 149-156.

Sedell, J. R., P. A. Bisson, F. J. Swanson, and S. V. Gregory. 1988. What we know about large trees that fall into streams and rivers. From the forest to the sea: a story of fallen trees. Maser, C, R.F. Tarrant, J.M Trappe, and J.F. Franklin, USDA Forest Service General Technical Report PNW-GTR 229.

Sedell, J. R. and W. S. Duval. 1985. Influence of Forest and Rangeland Management on Anadromous Fish Habitat in Western North America, Volume 5: Water Transportation and Storage of Logs. USDA Forest Service, General Technical Report, PNW-186.

Sedell, J.R. and F.J. Swanson. 1984. Ecological characteristics of streams in old-growth forests of the Pacific Northwest. Pages 9-16 in Meehan, W.R., T.R. Merrell Jr., and T.A. Hanley (eds) Fish and Wildlife Relationships in Old-Growth Forests. Proceedings of a symposium sponsored by Alaska Institute of Fishery Research Biologists, Northwest Section of the Wildlife Society, and Alaska Council on Science and Technology, American Institute of Fishery Research Biologists, Morehead City, NC.

Shields, F. D. and N. R. Nunnally. 1984. Environmental aspects of clearing and snagging. Journal of Environmental Engineering 110: 152-165.

Shields, F. D. and R. H. Smith. 1992. Effects of large woody debris removal on physical characteristics of a sand-bed river. Aquatic Conservation: Marine and Freshwater Ecosystems 2: 145-163.

Shirvell, C. S. 1990. Role of instream rootwads as juvenile coho salmon (Oncorhynchus kisutch) and steelhead trout (Oncorhynchus mykiss) cover habitat under varying streamflows. Canadian Journal of Fisheries and Aquatic Sciences 47: 852-861.

Singer, S. and M. L. Swanson. 1983. Soquel Creek Storm Damage Recovery Plan. US Soil Conservation Service, Plan,

Smith, R. D., R. C. Sidle, and P. E. Porter. 1993a. Effects on bedload transport of experimental removal of woody debris from a forest gravel-bed stream. Earth Surface Processes and Landforms 18: 455-468.

Smith, R. D., R. C. Sidle, P. E. Porter, and J. R. Noel. 1993b. Effects of experimental removal of woody debris on the channel morphology of a forest, gravel-bed stream. Journal of Hydrology 152: 153-178.

Spalding, S., N. P. Peterson, and T. P. Quinn. 1995. Summer distribution, survival and growth of juvenile Coho Salmon under varying experimental conditions of brushy instream cover. Transactions of the American Fisheries Society 124: 124-130.

Sullivan, K., T. E. Lisle, C. A. Dolloff, G. E. Grant, and L. M. Reid. 1987. Stream channels: the link between forests and fishes. Streamside Management: Forestry and Fishery Interactions. E. O. Salo and T. W. Cundy. Seattle, Washington, University of Washington, Institute of Forest Resources: 39-97.

67

Surfleet, C.G. and Ziemer, R.R. 1996. Effects of forest harvesting on large organic debris in coastal streams. Proceedings of the Conference On Coast Redwood Forest Ecology and Management: 134-136. Arcata, California, USA.

Swanson, F. J. and G. W. Lienkaemper. 1978. Physical consequences of large organic debris in pacific northwest streams. USDA Forest Service, General Technical Report, PNW-69.

Swanson, F.J. and G.W. Lienkaemper. 1979. Interactions among fluvial processes, forest vegetation, and aquatic ecosystems, South Hoh River, Olympic National Park, Washington. Proceedings of the Second Conference on Scientific research in the National Parks. San Francisco, Ca Nov. 26-30 1979. Vol. 7: 23-34.

Swanson, F. J., G. W. Lienkaemper, and J. R. Sedell. 1976. History, physical effects, and management implications of large organic debris in western Oregon streams. USDA Forest Service, General Technical Report, PNW-56.

Tally, T. 1980. The Effects of Geology and Large Organic Debris an Stream Channel Morphology and Process for Streams Flowing Through Old Growth Redwood Forests in Northwestern California. Doctor of Philosophy. University of California, Santa Barbara, Santa Barbara, CA, 273.

Tarzwell, C. M. 1937. Experimental evidence on the value of trout stream improvement in Michigan. Transactions of the American Fisheries Society 66: 177-187.

Triska, F. J. and K. Cromack. 1980. The role of wood debris in forests and streams. Forests: Fresh Perspectives from Ecosystem Analysis, Corvallis, Oregon: 171-190. Oregon State University Press.

Trotter, E. H. 1990. Woody debris, forest-stream succession, and catchment geomorphology. Journal of the North American Benthological Society 9: 141-156.

Tschaplinski, P. J. and G. F. Hartman. 1983. Winter distribution of juvenile coho salmon (Oncorhynchus kisutch) before and after logging in Carnation Creek, British Columbia, and some implications for overwinter survival. Canadian Journal of Fisheries and Aquatic Sciences 40: 452-461.

Van Sickle, J. and S. V. Gregory. 1990. Modeling inputs of large woody debris to streams from falling trees. Canadian Journal of Forest Research 20: 1593-1601.

Vannote, R.L., G.W. Minshall, K.W. Cummins, J.R. Sedell, and C.E. Cushing. 1980. The river continuum concept. Canadian Journal of Fisheries and Aquatic Sciences 37:130-137.

Wallace, J. B., J. R. Webster, and J. L. Meyer. 1995. Influence of log additions on physical and biotic characteristics of a mountain stream. Canadian Journal of Fisheries and Aquatic Sciences 52: 2120-2137.

Ward, G. M. and N. G. Aumen. 1986. Woody debris as a source of fine particulate organic matter in coniferous forest stream ecosystems. Canadian Journal of Fisheries and Aquatic Sciences 43: 1635-1642.

68

Williams, P.B. and M. L. Swanson. 1989. A new approach to flood protection design and riparian management. California Riparian Systems Conference, University of California, Davis: 40-46. USDA Forest Service General Technical Report, PSW-110.

Wood-Smith, R. D. and J. M. Buffington. 1996. Multivariate geomorphic analysis of forest streams: Implications for assessment of land use impacts on channel condition. Earth Surface Processes and Landforms 21: 377-393.

Young, M. K. 1994. Movement and characteristics of stream-borne coarse woody debris in adjacent burned and undisturbed watersheds in Wyoming. Canadian Journal of Forest Research 24: 1933-1938.

Young, M. K. 1996. Summer movements and habitat use by Colorado River cutthroat trout (Oncorhynchus clarki pleuriticus) in small, montane streams. Canadian Journal of Fisheries and Aquatic Sciences 53: 1403-1408.

Young, W. J. 1991. Flume study of the hydraulic effects of large woody debris in lowland rivers. Regulated Rivers: Research and Management 6: 203-212.