PROFILE Stream Corridor Management in the Pacific Northwest: I. Determination of Stream-Corridor Widths WILLIAM W. BUDD*

FREDERICK R. STEINER

Program in Environmental Science and Regional Planning Washington State University Pullman, Washington 99164-4430, USA

Programs of Regional Planning and Landscape Architecture Department of Horticulture and Landscape Architecture Washington State University Pullman, Washington 99164-4430, USA

PAUL L. COHEN

Design and Development Department City of Bellevue P.O. Box 90012 Bellevue, Washington 98009-9013, USA PAUL R. SAUNDERS

Department of Forestry and Range Management Washington State University Pullman, Washington 99164-6410, USA

Stream corridors are the surface water drainage systems that include both the water body and the adjacent riparian land. Stream corridors can be viewed from a variety of perspectives: (1) as an aesthetic amenity for residential and commercial development, (2) as habitat waters for fish, (3) as wildlife refuges, (4) as the outfall for storm sewers, (5) as important for agricultural uses, (6) as water supplies, and (7) as resources for open space and recreation. Pressures on stream corridors increase whenever land development expands (Coughlin and others 1972). The purpose of this article is to develop a management strategy from a literature review and fieldwork on Pacific Northwest corridors. This approach combines several methods for rating aspects of stream corridors into one common technique. Protecting a major portion of stream corridors as a classified land use would reduce the degradation of riparian ecosystems from development pressures. Changes in land use create problems of reduced water quality and open space, along with a reduction in the quality of riparian zones through loss of vegetation cover and soil erosion. Development can have serious impacts on stream and riparian environments. Sediment, toxic wastes, erosion, fecal pollution, decreased dissolved oxygen, higher water temperatures,

KEY WORDS: Stream corridors; Buffer widths; Environmental planning; Puget Sound; Pacific Northwest; King County (Washington); Watershed management

*Communicating author. Environmental Management Vol. 11, No. 5, pp. 587-597

ABSTRACT / King County, Washington is a part of the rapidly growing Pacific Northwest region. This growth has placed pressure on stream corridors. Past studies about regional stream corridors provide a rich source of information for environmental planners and managers. This article draws on existing literature and case studies to provide guidelines for determining optimal stream corridor widths in a watershed located in King County, Washington.

and week and algae blooms can reduce water quality and seriously impact anadromous fisheries. King County is the most populated county in the Pacific Northwest (Figure 1). The county is extensively developed in the Seattle metropolitan area and the Puget Sound lowlands in the west. In the east are the heavily forested Cascade Mountains. Surface waters generally flow westerly across the county from the forested mountains to the Puget Sound lowlands. Land development is expanding from the western third of the county eastward and into the Cascade foothills. The growth trend in King County is toward new development and construction outside the limits of Seattle in suburban cities and unincorporated areas of the county. Between 1960 and 1980 King County's population increased by 335,000. Of that increase, 183,017 was suburban growth, and 214,959 was growth in unincorporated King County, while the City of Seatde declined by 63,241 people (King County Planning and Community Development 1983). Because population and land develoment are rapidly moving into undeveloped and underdeveloped parts of King County, there is a need to systematically and comprehensively address riparian ecosystem protection through stream corridor management. While federal and state government have broad, entrusted powers, county and city levels of government are better suited to protect their resources comprehensively and to administer and to enforce policies. Local government officials have knowledge of numerous, interrelated issues in their jurisdiction which allows the design and enforcement of policies to be tailored to local needs and capabilities. 9 1987 Springer-VerlagNew York Inc.

588

W.W. Budd and others

INGTON

~

REA

Figure 1. Bear-Evans Creek Study Region, King County, WA. Riparian ecosystems west of the Cascade Mountains in the Pacific Northwest display similar physical, chemical, and biological characteristics. Often within these areas resource management conflicts occur between interests in timber and salmonids. Consequently, much of the research done on stream corridor management addresses forest practices (logging) and the maintenance of salmon habitat. This article evaluates the physical and biological conditions critical to sustaining stream ecosystems. These conditions are based on an analysis of stream corridor needs for fisheries and wildlife. From this analysis, operational buffer widths are suggested for the Bear Creek watershed in King County, Washington. These suggestions form the basis of a practical model for stream corridor management.

Stream Corridors in the Pacific Northwest The stream corridor or riparian ecosystem can be defined as the area of transition between the aquatic zone and the upland zone. Stream corridors therefore contain elements of both aquatic and terrestrial ecosystems. The condition of the riparian ecosystem affects the natural character and water quality of streams, thus impacting fisheries. Wildlife is most abundant along stream corridors because of the proximity of the riparian zones to water (McCormick 1978). For this reason, it is important to consider the impacts of changes in stream corridors on fisheries and wildlife habitat.

Fisheries T h e supply of salmonids for commercial and sports fisheries is the focus of controversy in the Pacific Northwest concerning aboriginal fishing rights, spawning bed protection, migration routes, and hy-

droelectric power development. In attempting to mitigate concerns over these issues, baseline conditions for habitat maintenance are vital to decision making. Considerable research has been conducted on the definition of habitat conditions necessary for self-sustaining salmonid populations (see Milo 1986). This has included work on natural and hatchery populations, which can make generalization difficult. Aside from their own inherent ecological and recreational values, riparian ecosystems are important because of their effects on the biological productivity and water quality of streams. Riparian vegetation affects the physical composition of stream habitat as well as the biological communities of which salmonids are a part (Bottom and others 1983). Up to 90% of the energy processed in headwater streams comes from streamside vegetation (Washington Department of Game 1984). In the Pacific Northwest, salmonid productivity is a strong indicator of the water quality of a stream and the quality of the riparian ecosystem surrounding it. Alterations or impacts to a riparian ecosystem will eventually show themselves in the stream. Impacts are felt most on small streams, because the dilution factor is lowest. The major fish habitat elements influenced by riparian ecosystems and correlated with stream corridor widths are: (1) water temperature, (2) food supply, (3) stream structure, and (4) sedimentation control.

Water Temperature Water temperature is largely influenced by the rate of streamflow, elevation, and the amount of shade. Other stream channel characteristics that affect water temperature are the inflow of surface and groundwater, undercut embankments, organic debris, surface area, depth, and stream velocity. Riparian ecosystems act as reservoirs, storing runoff in soil spaces and wetland areas. This characteristic maintains summer streamflow and keeps water temperatures low as stored water is discharged back to the stream. Riparian vegetation creates a microclimate that helps maintain a more constant temperature, minimizing the extreme highs and lows. The vegetation canopy adjacent to streams shields the water from direct solar radiation and moderates extreme temperature fluctuations during summer. The insulating properties of riparian vegetation may also keep streams from freezing over during the winter months, increasing the winter survival of fish and other aquatic organisms (Bottom and others 1983). The impacts of forest management practices on stream temperature have been documented in several studies (Brown and Krygier 1970). Patch cut water-

Determination of Stream-Corridor Widths

sheds with buffer strips exhibit no increase in temperature attributable to logging activity. Clearcut watersheds, however, have shown a monthly mean maximum increase of 7~176 The peak daily maximum rise in these streams may reach 15.6~ during low flows experienced in late summer (Montgomery 1976). The shading requirements for the maintenance of fisheries habitat are dependent on stream size. Montgomery (1976) found daily temperature variation in undisturbed streams to be approximately 4~ or more. Temperature variation increased to 10~ or higher when all shade along the stream was removed. This effect is mitigated by stream size. Even though streamside vegetation cannot shade much of the stream surface, larger streams are less affected by solar radiation because greater flow volumes mitigate surface temperature increases (Everest and others 1982). Additionally, the bank position of vegetation can be important. Vegetation on southern banks is more important than on northern banks. Brazier and Brown (1973) have defined buffer strip characteristics important to effective regulation of temperature in small streams. Buffer strips have been designed to minimize changes in both stream temperature and the amount of commercial timber (Montgomery 1976). Brazier and Brown (1973) suggest that the effectiveness of buffer strips in controlling temperature changes is independent of timber volume. Neither timber volume nor buffer width alone is an important criterion for water temperature control. In their study, maximum shading capability of the average strip was reached within a width of 25 m (80 ft), and 90% of that maximum was reached within 17 m (55 ft). Specifying standard 31-61-m (100-200 ft) buffer strips for all streams, while it usually assures protection, will include more timber in the strip than necessary. Finally, they suggest that greater Angular Canopy Density (ACD) increases buffer strip shading effectiveness and is correlated with stream temperature control. This is the only criterion foresters have to measure adequate temperature control for the stream without overdesigning the buffer strip (Montgomery 1976). Where the average maximum stream temperature is low and limits growth, salmonid populations benefit from increased solar radiation. Bisson and Sedell (1982) studied salmonid population dynamics in paired unlogged and logged streams flowing through old growth forests. Their studies indicated that during summer low-flow periods the population biomass of cutthroat trout, juvenile steelhead, and juvenile coho salmon increased an average of 1.5 times after logging

589

over adjacent unlogged sections. Among all sites (paired plus unpaired locations) total salmonid biomass averaged 2.0 times greater in clearcuts. Bisson and Sedell (1982) suggest that removal of the forest canopy allows more solar radiation to penetrate to the stream channel, causing temperature increases (Brown and Krygier 1970) and elevated algal growth (Lyford and Gregory 1975). Both processes increase fish production until thermal change and algal accumulation cause reduced production (Bisson and Sedell 1982). Although fish biomass may increase following a clearcut, alteration of physical habitat brought about by cover removal and channel destabilization caused a shift in species and age composition within the community to fish not dependent upon pools and cover (Bisson and Sedell 1982). The conclusion that salmonid biomasses were generally higher in streams following clearcuts must therefore be tempered with the observation that carrying capacity increases were not shared equally among different species or age groups. Similarly, the effect of shading on salmonid populations is variable. Hawkins and others (1983) have shown that salmonids are more abundant in streams without riparian shading than in shaded streams. Other research (Lyford and Gregory 1975, Hall and others 1978, Gregory 1980, Murphy and Hall 1981, Bisson and Sedell 1982, Hunt 1979, Newbold and others 1980) provides convincing evidence that streams with open canopies are more productive than heaviliy shaded streams.

Food Supply The abundance and type of food supply is directly related to the abundance and type of fisheries (Likens and others 1970). The relationships between instream productivity and adjacent forest ecosystems indicate that over 99% of the energy and hydrocarbon in aquatic food webs has its origins in adjacent forest ecosystems (Likens and others 1970, Bormann and others 1969, Bormann and others 1968). Temperature control and stream shading are integrally linked to food supply. Benthic invertebrates, algae, terrestrial insects, leaves, and other organic material contribute food to the vertebrate species and minerals to the water that support fish populations. All these food sources are related either directly or indirectly to riparian vegetation. Invertebrate communities of logged (disturbed) streams have a lower diversity index and higher population than unlogged streams (Erman and others 1977). Weber (1981) recorded higher densities of benthic invertebrates in streams where the forest canopy had been removed. Narver (1972) determined

590

w . w . Budd and others

that increased temperature promoted earlier emergence of fry and increased appetite, and prolonged the fish's growing season. It is therefore not clear whether the increased biomass of several invertebrate food sources is beneficial. Erman and others (1977) found that benthic invertebrate communities in streams without forest buffers were significantly different from communities of unlogged streams, on the basis of both diversity index and ecological distance. On the basis of impacts to invertebrate communities, these authors suggest a probable increase in stream protection as buffer widths increase up m 30 m (98.4 ft).

Stream Structure Stream structure, the physical character of the stream, is the result of various combinations and sizes of pools, riffles, falls, instream cover, and bank stabilization from fallen trees, rootwads, gravel, and boulders. A large portion of the stream structure in forested areas is derived from trees and other vegetation in the riparian zone (Bottom and others 1983). Large woody debris can create the habitat diversity necessary for salmonid production in stream channels and off-channel areas, including the variety of depths, velocities, and substrates utilized during the freshwater phases o f the salmonids life cycle (Everest and others 1982). The necessity for woody debris in streams is often confused by the issue of slash residue left in streams following logging operations. Bisson and Sedell (1982) found that removal of woody debris after logging effectively stripped the riparian zone of necessary organic matter and large woody material necessary to promote stream structure and variety. Steinblums and others (1984) found that leaving a buffer strip during logging operations increased the number of windthrown trees, thereby contributing to the stream structure. The buffer width necessary to provide a natural supply of woody debris to a stream is unknown. However, most woody structure in streams is derived from within 31 m (100 ft) of the bank (Bottom and others 1983).

Sedimentation Control Sedimentation in streams contaminates salmonid gravel spawning beds, alters stream structure, fills in pools, and lowers water quality. Vegetation along stream banks holds the soil in place, preventing instream erosion and stabilizing the channel. During high flows, flood crests are dispersed, water velocity is reduced, and erosive power is dissipated by root masses and other stream structures. Riparian vegeta-

tion filters out fine sediment, debris, oils, and other pollutants such as pesticides and herbicides from upland runoff sources (Bottom and others 1983). In one study, when 30-m (98.4-ft) buffers were left along streams in a logged area, no sedimentation was measured in the stream (Erman and others 1977). However, when buffers were less than 30 m, stream sedimentation caused a change in aquatic insect communities that were the basis of fish diets.

Wildlife Habitat McCormick (1978) indicates that high-quality riparian ecosystems are uniquely characterized by a combination of high species density, high species diversity, high productivity (Johnson and others 1977), high degree of endemism (Hubbard 1977), and a high number of endangered species (Johnson and others 1977). A majority of North American wildlife is depend~nt upon riparian habitats for their survival (Hubbard 1977). There is no habitat type upon which wildlife is more dependent (Ohmart and others 1977). Alteration of wetland and riparian zone habitat has been shown to have significant effects on fish and wildlife populations. One study (Carothers and others 1974) found a 46% reduction in breeding birds in a riparian zone that had been thinned to 25 trees per ha, over what was found in a nearby undisturbed zone with similar habitat (116 trees per ha). These areas are primary activity centers of birds and contain a higher bird density than areas without a water influence (Gill and others 1974). Fish populations have also been found to decrease significantly adjacent to and downstream from wetland and riparian zone alteration, through effects of temperature increase, siltation, debris barriers, introduction of chemicals, and increases in flow fluctuations (US EPA 1976, Gibbons and Salo 1973). Although it is of critical importance, there is insufficient information about the nature of species dependence upon riparian habitat to serve as a basis for resource planning and management (McCormick 1978, Greer 1982). Still less is known of minimal area and critical habitat necessary to support indigenous species. Research (Levin and Paine 1974, McCormick 1978, Simberloff 1976, Tramer and Suherweir 1975, Van Winkle and Martin 1973) suggests that minimal area and shape of wildlife habitat are important determinants of species productivity. Habitat area is especially relevant to stream corridor management (Comarite 1975) and preservation (Diamond 1975) because these habitats often occur as narrow, linear corridors along streams and rivers. For a fixed width,

Determination of Stream-Corridor Widths

habitat area is a function of length. T h e narrower the width the greater the length of habitat required to maintain the system. Studies which correlate riparian wildlife with prescribed stream corridor buffered widths are based on wildlife habitat rather than individual species requirements. A western Washington urban stream assessment by the Washington Department of Ecology (1981) found streamside habitats extended beyond the stream bank a minimum 27.4 m (90 feet) into the uplands.

Buffer Widths Width of the riparian zone varies from stream to stream and along the course of an individual stream. In the P~tcific Northwest, stream buffer widths for each side range from 11 m to 38 m (35 ft to 125 ft) from the high-water mark, depending on the riparian ecosystem element studied. Water temperature control requires 60% to 80% shade on stream surfaces. Sixty to 80% shade when compared to measured ACD can be achieved with 11 to 24.3 m (35 ft to 80 ft) of buffer width by Brazier and Brown (1973) and 23 m to 38 m (75 ft to 125 ft) by Steinblums and others (1984). Food supply of benthic invertebrates has been shown to be unaffected when a minimum of 30-m (98.4-ft) buffer widths remained after logging (Erman and others 1977). T h e contribution of woody debris to stream structure is believed to be derived from within 31 m (100 ft) of the banks of a stream (Bottom and others 1983). Sedimentation control can be achieved by stable stream banks to control instream erosion and the buffer width to control overland erosion. Erman and others (1977) found 30-m buffer widths protected aquatic insect communities from increased sedimentation. T h e Washington Department of Ecology (1981) found wildlife habitat extended beyond the stream bank a minimum 27.4 m (90 ft) into the uplands.

Case Study T o identify environmentally responsive and practical buffer widths for a selected stream in King County, a case study of the B e a r - E v a n s Creek watershed was undertaken. This study was intended to test the general methodology for stream corridor assessment and to provide an initial foundation for policy development on stream corridor management. The Bear Creek area is in rural unincorporated King County on the eastern edge of suburban metropolitan Seattle. T h e survey was conducted in August 1984 and in January 1985. T h e procedure consisted of measuring stream cross-sections a minimum of 15 m from both

591

Table 1. Combinations of streams surveyed. Site conditions Unconstrained Slopes 15%-40% Slopes 40% Wetlands

Types 1-4

Type 5

Site A Site B Site C Site D

Site E Site F Site G Site H

sides of the stream starting at the stream edge. Vegetation, percentage of slope, and soil types were recorded along the cross-section. Soil classifications were determined using the US Department of Agriculture soil survey for King County, Washington. From these data, runoff potential, slippage potential, soil water capacity, and erosion potential were determined. Temperature control determination followed the method outlined in Brazier and Brown (1973). Stream structure assessments were based on the results of Bottom and others (1983). Sedimentation control assessments followed criteria described by Erman and others (1977). Wildlife habitat determinations were made using criteria established by the Washington Department of Ecology (1981). T h e stream classifications used in the study were based on State of Washington Department of Natural Resources (WDNR) stream Types 1 through 5, and King County's three major categories of sensitive areas: (1) slopes 15-40%, (2) slopes greater than 40%, and (3) wetlands. T h e r e are five types of water in the WDNR system. Water Types 1 to 4 are natural water courses, usually perennial, may include wetlands, lakes of sufficient size, and have significant influence on water quality downstream. Type 5 waters include all other waters, may be perennial or intermittent, may lack a well-defined channel, have short periods of spring runoff, and include seepage areas, ponds, and sinks. Type 1 water includes all shorelines defined under the Washington Shoreline Management Act (Revised Code of Washington 90.58). Types 2 to 4 waters have significant influence on water quality downstream, in addition to high value for anadromous fish and a water source for h u m a n use. Type 5 water includes all other natural water courses and wetlands. T y p e 5 streams were treated separately in this study. These are the most numerous and are also the most difficult streams in terms of site-specific requirements to buffer. Table 1 shows eight combinations surveyed in the B e a r - E v a n s Creek watershed. B e a r - E v a n s Creek watershed has moderate to steep slopes that drain into a broad floodplain. T h e watershed drains through B e a r - E v a n s Creek into the Sammamish Slough, Lake Washington, and finally the salt water of Puget Sound. Bear and Evans Creeks are

592

w . w . Budd and others

Table 2. Stream corridor evaluation criteria. Site

Stream a Soilb type classification

A

1

B

2

C

3

D

1

E

5

F

5

G

5

H

5

Earlmont sandy soil Everett sandy loam Alderwood loam Sultan silt loam Kitsap silt loam Alderwood gray. sandy loam Alderwood sandy loam Orcas peat

Runoff~ potential

Slippageb potential

Erosionb potential

Soil-water b Vegetation ~ Temperature d Stream e Sedimentation ~ Wildlifeg capacity cover control structure control habitat

Slow

na

Slight

na

Mediumrapid Rapid

na

na

Severe

Moderatesevere Severe

na

na

na

Moderate flooding na

na

Slowna Slight moderate moderate Moderate Moderate Severe na

Very rapid Ponding

Severe na

Very severe Low

na High

None sparse Dense sparse Dense dense Sparse sparse Sparse none Dense dense

Poor

Poor

Good

Fair

Good

Excellent

Excellent

Excellent

Excellent

Excellent

Fair

Excellent

Poor

Fair

Fair

Fair

Poor

Poor

Good

Poor

Excellent

Excellent

Excellent

Excellent

Dense dense Sparse dense

Excellent

Excellent

Excellent

Excellent

Good

Good

Excellent

Excellent

Sources: aWashington Department of Natural Resources 1976; bSnyder and others 1973; findex reported as overstory/understory; dBrazier and Brown 1973; eBottom and others 1983; german and others 1977; gWashingtonDepartment of Ecology 1981.

wide, shallow streams with low gradients. The tributaries of these streams drain the surrounding valley walls and have moderate to steep gradients. Bear Creek (12.7 river miles) and its four tributaries (13.9 river miles) total 26.6 miles in length. Evans Creek is 8.2 river miles in length. Bear Creek between river mile 3 and 6.5 and Evans Creek between river mile 0 and 1 have been designated "Conservancy" by the King County Shoreline Management Master Program (1981). Conservancy shoreline consists of areas which are primarily free from intensive development. The vegetation canopy varies throughout the watershed. There are areas with few deciduous trees that provide little shade. Conversely, there are thick growths of deciduous and coniferous trees that provide excellent shade. Instream cover consists of overhanging branches and grasses, natural stream debris, and undercut banks. Portions of both streams flow through wetland areas where the stream bank is undefined (METRO 1982).

Results As the literature review above indicates, determining practical stream corridor buffer widths is contingent upon several factors. The protection of anadromous fish in the King County, temperature, habitat, erosion control, and, to a lesser extent, food supply are the environmental variables that form the basis for establishing corridor widths. From our review it is clear that a generic corridor width which would provide habitat maintenance while satisfying human demands does not exist. The determination of corridor widths

involves a broad perspective that integrates potential uses with environmental constraints. In this regard, the determination of appropriate stream corridor widths is not unlike the classification of streams. Stream classification involves assessing a multitude of phenomena interacting over time and in changing environments and is a complex process (Beschta and Platts 1985). Unlike stream classification, however, the determination of a workable management tool need not be optimal. Environmental management is an imperfect science, more of an art, in that sense. However, provided the tools are available, decision makers can ensure that environmental limitations are incorporated directly into that process. Our work presents such a methodology applied to a broad range of stream characteristics for selected reaches of the Bear-Evans Creek watershed in King County. The range of characteristics examined is shown in Tables 1 and 2. In reviewing these stream sections to determine practical corridor widths our focus was to (1) examine the sensitivity of our integrated approach to the stream classification approach adopted by WDNR for stream management, (2) examine the effect of a wide range of stream characteristics on application of our methodology, and (3) develop recommendations for determining stream corridor width based on our assessment for King County. Site A is a WDNR Type 1 stream without steep slopes or adjacent wetlands on Bear Creek (Figure 2). T h e site is located at river mile (RM) 1 near the mouth of the Bear-Evans watershed where the stream exits the watershed into a broad floodplain before emptying into the Sammamish River. At this location the

593

Determination of Stream-Corridor Widths

SITE A

SITE B d TYPE2 STEEPSLOPES 15-40% 228AVENE

IP~

TYPE1 UNCONSTRAINED ATBEARCREEKSHOPPINGCENTER EAST

!

~

SOUTH

J

SEDGE BUTTERCUP THISTLE CROWNGRASS CANARYGRASS

15

10

5

SEDGE CANARYGRASS l BLACKBERRY BUTTERCUP FIELDGRASS SOILS- Earlmont SlowRunoff SlightErosionHazard Subjectto Flooding 0

0

5

10

15

Figure 2. Bear-Evans Creek Study Site A. stream is approximately 7.6 m (25 ft) wide, its broadest in the study area. Soil conservation data indicate that site A is located on Earlmont sandy soils, which have modest stability problems and slight erosion potential (Snyder and others 1973). Stream bank collapse is frequent along this stream section. Vegetation at the cross-section has no upper canopy and a sparse lower canopy; low shrubs and grasses overhang the banks. The sparse vegetation cover results in low ratings for temperature control (shading), existing stream structure (woody debris), and potential wildlife habitat. Sedimentation control is considered good because of the low, side slope gradients and dense mat of grasses to control runoff and slope erosion. A 15-m (50-ft) buffer width from the stream edge is radiated to preserve some of the floodplain, allow larger vegetation to reestablish along the banks for shade, stream structure, and wildlife habitat, and ensure that open space will be preserved in an area under intensive development. Site B is a Type 2 stream with valley side slopes of 15%-40% north and 0 to 2% south (Figure 3). The north slope is heavily wooded and undeveloped, while the south slope is lightly wooded with some clearing from new residential development within 15 m (50 ft) of the stream. The north, undeveloped slopes are underlain with Everett gravelly, sandy loam. These soils have medium to rapid runoff potential and moderate to severe erosion potential (Snyder and others 1973). T h e vegetation has a dense canopy and sparse understory on the north slope and sparse upper canopy and dense understory on the south slope. Temperature control was considered good because the south slope has an understory but little overstory to block solar radiation. Stream structure, sedimentation, and wildlife were rated excellent because the existing stream corridor is wider than that necessary to supply woody

HEMLOCK DOUGLASFIR I W. REDCEDAR BIGLEAFMAPLE SWORDFERN LADYFERN BUTTERCUP

METERS

15

10

5

0 0

5

10

15

Figure 3. Bear-Evans Creek Study Site B.

Figure 4. Bear-Evans Creek Study Site C. debris, mitigate sedimentation, and provide wildlife habitat. A 15-m (50-ft) buffer width from the stream edge is adequate to protect this riparian ecosystem. Site C is a Type 3 stream in a small, heavily wooded ravine with valley side slopes over 40% (Figure 4). The stream is located at RM 7 on a tributary to Evans Creek. The area is wooded with scattered rural residential development. Alderwood and Kitsap soils are found at this site. Slopes are over 40% with rapid runoff, severe slippage potential, and severe erosion potential (Snyder and others 1973). Considerable amounts of woody debris are found in the stream. This is the result of soil creep and slope slump which has undermined trees, resulting in their falling into the stream channel. The ample supply of woody debris on the slopes and in the stream mitigates some of the soil creep, slippage, and erosion, resulting in fair sedimentation control. Temperature control and wildlife habitat are excellent. The density and extent of both upper canopy and understory provide excellent shade to block solar radiation and provide cover

594

w . w . Budd and others

SITE E ~-~ TYPE 5 ~ ~ ~'. UNCONSTRAINED

SITE D

TYPE 1 WETLANDS BEAR-EVANS ROAD

SOUTH

RED ALDER CATTAIL CANARY GRASS RUSHES BLACKBERRY PASTURE SOILS-Sultan Moderate Hazard of Stream Overflow METERS

15

10

5

0

0

5

10

Figure 5. Bear-Evans Creek Study Site D. for widlife. A 15-m (50-ft) buffer width would encompass the steep, severely erosive slopes. Considering the severe to very severe soil hazards and close proximity of residential development, a 15-m (50-ft) buffer width is indicated. Site D is a Type 1 stream with an adjacent wetland (Figure 5). The stream is on Evans Creek located at RM 2.7. The area is a small agricultural (dairy) valley with wooded side slopes. The soils at site D are Sultan silt loam with less than 2% slopes and a moderate hazard of stream overflow (Snyder and others 1973). Site D is similar to Site A in that both are major streams in valley bottoms with similar adjacent landuse patterns (agriculture and pasture). Temperature control is poor at this site because of the lack of vegetation canopy on the southern bank. Stream structure, sedimentation, and wildlife habitat are fair. The absence of a vegetation canopy has minimized potential stream structure and wildlife habitat. A 15-m (50-ft) buffer width would encompass most of the wetland and act as sedimentation control for runoff. Such a buffer width would allow streamside vegetation to reestablish, help shade, mitigate bank erosion, and help prevent agricultural chemicals and fecal matter from entering the stream. Site E is a Type 5 intermittent stream without steep slopes or adjacent wetlands (Figure 6). The stream channel resembles a grass-lined swale with no distinct channel, banks, or water edge. The stream gradient is steep, but the cross-section has less than 15% side slopes. Soils found at site E are Kitsap silt loam on 2% to 8% slopes, with slow to medium runoff, and slight to moderate erosion hazard (Snyder and others 1973). Sedimentation control is good due to the grass matting that covers the side slopes, and the swale bottom filters runoff. Temperature control; stream structure, and wildlife habitat are poor. This is because stream bank

BUTTERCUP RUSHES GRASSES RED ALDER BIGLEAF MAPLE BRAMBLES

SOILS - Kitsap Slow to Medium Runoff Slight to ModerateErosion

METERS

Figure 6. Bear-Evans Creek Study Site E.

Figure 7. Bear-Evans Creek Study Site F. vegetation is inadequate to support the requirements of shading, woody debris, and cover for wildlife. A 7.6-m (25-ft) buffer width appears to be optima! to shade the channel and reduce and filter any overland erosion. A buffer width greater than 7.6 m would not increase the protection of the stream during low flows or dry periods in the summer months. Site F is a Type 5 stream with valley side slopes of 15%-40% (Figure 7). It is located on a tributary of Cottage Creek at RM 0.5. The site is a broad, wooded ravine with new large-lot, residential development on the surrounding uplands. Steep road grades provide access to the wooded hillsides from the Bear Creek valley floor. The soils are Alderwood gravelly sandy loam, which have medium runoff potential, moderate slippage potential, and severe erosion potential (Snyder and others 1973). Temperature control, stream structure, sedimentation control, and wildlife habitat are excellent because of the large, undisturbed stream corridor. For site F, a 7.6-m (25-ft) buffer width would not cover one-third of the slopes, espe-

595

Determination of Stream-Corridor Widths

r

~~STEEP SLOPES SITE G TYpE 5

~;JBII

DOUGLAS W. RED CEDAR RED ALDER VINE MAPLE SALMONBERRY SWORDFERN LADYFERN BUTTERCUP

METERS

1.5

1~

SOILS--Alderwood ~nd Kitsap Rapid to Very Rap~dRunoff Severe to Very Severe Erosion Severe Slippage &

0

()

5

I~)

I'5

SITE H

TYPE5

WETLANDS SEAR CREEK ROAD

WEST

RED ALDER HARD HACK BRACKEN FERN RUSHES BUTTERCUP LADYFSRN

00 METERS

SPIREA RUSHES

5

CANARY GRASS RUSHES BUTTERCUP

[

SOILS--Orcas Peat High Waler Capacity Ponding Runoff Low Erosion 10

15

20

25

30

35

40

50

Figure 8. Bear-Evans Creek Study Site G.

Figure 9. Bear-Evans Creek Study Site H.

cially on soils with severe erosion potential. A 15-m (50-ft) buffer width, or the explicit inclusion of all slopes of 15%-40% with severely unstable soils, appears warranted. Site G is also a Type 5 stream with valley side slopes greater than 40% (Figure 8). This stream is a tributary to Evans Creek located high on the valley walls in a V-shape ravine. Both slopes are heavily wooded with conifers and deciduous trees and several layers of understory. T h e soils are Alderwood and Kitsap with rapid to very rapid runoff, severe to very severe slippage potential, and severe erosion hazard (Snyder and others 1973). Site G, like Site F, has excellent temperature control, stream structure, sedimentation control, and wildlife habitat because of the largely undisturbed stream corridor. Under these conditions sites require a buffer width large enough to protect the unstable soils. Slopes greater than 40% extend a minimum of several hundred feet on either side of the stream. Identification of a specific buffer width must allow for these extreme topographic considerations. Site H is a Type 5 stream in a wetland and is a tributary to Cottage Creek at RM 2.5 (Figure 9). The site is an open wetland. T h e stream is a straight channel 0.6 m (2 ft) wide flanked by small trees and large shrubs. Smaller wetland grasses are found on the wetland fringes. T h e site resembles a wetland channeled to drain the surrounding area. Orcas peat soils are found on the site. These soils characteristically have ponding of runoff, low soil erosion potential, and high water capacity (Snyder and others 1973). The high water capacity of the soils, level side slopes, and existing vegetation provide excellent sedimentation control. Wildlife habitat consists of the extensive wetland ecosystem with some small, dense tree cover along the stream. Temperature control and stream structure are good due to the dense understory of deciduous trees,

even though an overstory is absent. A 7.6-m (25-ft) buffer width at the wetland boundary would provide adequate protection for the stream. As with site G, the extreme topographic features of the site warrant broader control measures to ensure the protection of the wetland areas bordering the stream corridor.

Conclusions Evaluation of stream corridor widths has been as difficult a task as stream classification. In part that has been a limitation of evaluation techniques. Research has focused on selected environmental factors affecting riparian ecosystems, and management methods have evolved from this research which emphasize that single dimension focus. Moreover, this single-dimension perspective has extended to the implementation and planning aspects of stream management (Orsborn and Anderson 1986). Our research indicates that practical determinations of stream corridor widths can be efficiently and easily made using a simple field survey of select reaches of a stream system combined with an analysis of soils, vegetation, physiography, and land-use characteristics. This approach was found applicable over a broad range of stream characteristics. Additionally, the technique was sensitive to extreme environmental conditions, such as extremely steep slopes or extensive wetland regions. A 15-m (50-ft) buffer width was found to be an adequate protection barrier for many reaches of the Bear-Evans Creek watershed. This includes intermittent (WDNR Type 5) as well as continuous flowing (WDNR types 1-3) reaches of this watershed. As expected, the greatest variation in corridor widths was found for intermittent reaches. Under conditions of poor habitat (Site E), extremely severe bank slopes (Sites F and G), and extensive wetlands areas (site H)

596

w.w. Budd and others

practical corridor widths were variable. Where stream bank slopes exceeded 40%, determination o f an appropriate corridor buffer width required additional data. Similarly, corridor buffer widths in wetland regions could not be determined solely on the basis o f the criteria set used in o u r analysis. In both these instances the limitations o f the methodology were readily apparent, thus allowing any prospective manager to identify such reaches as required additional study. It should be noted here that o u r focus was exclusively on providing recommendations which would ensure maintaining existing conditions o f the stream systems being studied. O u r objective was not stream improvement, although this approach could be modified to address such an objective. Several benefits are derived f r o m stream corridors. These areas are enjoyed for aesthetic and recreational reasons. T h e y also serve a variety o f functional h u m a n purposes, including the preservation o f water quality, the minimization o f r u n o f f impacts, and the control of erosion. Additionally, these areas are critical to the maintenance o f fish and wildlife habitat t h r o u g h o u t a region. As a result, the protection and enhancement of stream corridors should be an important public policy goal at all levels o f government. A considerable a m o u n t o f research has been conducted by scientists in the Pacific Northwest and elsewhere on the nature o f stream corridors. Unfortunately, this research has focused on selected aspects o f the m a n a g e m e n t problem and has too infrequently been integrated into the decision-making process. O u r research has shown that this integration can be readily achieved t h r o u g h an interdisciplinary site assessment process. This process requires adopting a holistic perspective towards environmental management. Having d o n e so, local planners can establish control zones which are consistent with ecologic criteria intended to preserve fish and wildlife and which are also sensitive to existing land uses.

Acknowledgments This is scientific paper n u m b e r 7500 o f the Agricultural Research Center, Washington State University, Pullman, WA, Projects 0596 and 0750. T h e authors would like to thank J o h n Orsborn, Julie McQuary, and J o h n Roberts for their comments on early drafts o f this manuscript. We also wish to thank Brenda Stevens and Angie Briggs for typing the manuscript.

Literature Cited Beschta, R. L., and W. S. Platts. 1985. Morphological features

of small streams: significance and function. Water Resources Bulletin 22:369-379, Bisson, P. A., andJ. A. Sedell. 1982. Salmonid populations in streams in clearcut vs old-growth forests of western Washington. Weyerhaeuser Company, Tacoma, WA. 17 pp. Bormann, F. H., G. E. Likens, and J. S. Eaton. 1969. Biotic regulation of particulate and solution losses from a forest ecosystem. BioScience 19:600-610. Bormann, F. H., G. E. Likens, D. W. Fisher, and R. S. Pierce. 1968. Nutrient loss accelerated by dear-cutting of a forest ecosystem. Science 159:882-884. Bottom, D. L., P.J. Howell, and J. D. Rodger. 1983. Final report: fish research project Oregon, salmonid habitat restoration. Oregon Department of Fish and Wildlife, Portland, OR. 155 pp. Brazier, J. R., and G. W. Brown. 1973. Pages 1-9 in Buffer strips for stream temperature control. Forest Research Laboratory, School of Forestry, Oregon State University, Corvallis, OR. Brown, G. W., and J. T. Krygier. 1970. Effects of clear-cutting on stream temperature. Water Resource Research 6:1131-1139. Carothers, S. W., R. R. Johnson, and S. W. Aitchison. 1974. Population structure and social organization of southwestern riparian birds. American Zoologist 14:97-108. Comartie, w.J.,Jr. 1975. The effect of stand size and vegetational background on the colonization of cruciferous plants by herbivorous insects. Journal of Applied Ecology 12:517533. Coughlin, Robert E., T. R. Hammer, T. G. Dickert, and S. Sheldon. 1972. Perception and use of streams in suburban areas: effects of water quality and of distance of residence to stream. Regional Science Research Institute, Philadelphia, PA. 70 pp. Diamond, J.M. 1975. The island dilemma: lessons of modern biogeographic studies for design of natural reserves. Biological Conservation 7:129-146. Erman, Don C.,J. D. Newbold, and K. B. Roby. 1977. Evaluation of streamside buffer strips for protecting aquatic organisms. California Water Resources Center, University of California, Davis, CA. 48 pp. Everest, F.H., N.B. Armantrout, S.M. Keller, W.D. Parante, J.R. Sedell, T.E. Nickelson, J. M. Johnston, and G. N. Haugen. 1982. Salmonids Westside Forest-Wildlife Habitat Relationship Handbook. US Forest Service, Pacific Northwest Forest and Range Experiment Station, Portland, OR. 97 pp. Gibbons, D. R., and E. O. Salo. 1973. An annotated bibliography of the effects of logging on fish of the western United States and Canada. USDA Forest Service, General Technical Report PNW-10. 145 pp. Gill, J. D., R.M. DeGraaf, and J. W. Thomas 1974. Forest habitat management for nongame birds in central Appalachia. USDA Forest Service Research Note NE-192. 6 pp. Greer, M. 1982. Urban wildlife population: ecological evaluation of management and planning. Environmental Management 6:217-229. Gregory, S.V. 1980. Effects of light nutrients, and grazing

Determination of Stream-Corridor Widths

on periphyton communities in streams. Ph.D. thesis, Oregon State University, Corvallis, OR. 151 pp. Hall, J. D., M. L. Murphy, and R. S. Aho. 1978. An improved design for assessing impacts of watershed practices on sn'lall streams. Verl. International Ver. Limnology 20:13591365. Hawkins, C. P., M. L. Murphy, N. H. Anderson, and M.A. Wilzbach. 1983. Density of fish and salamanders in relation to riparian canopy and physical habitat in streams of the Northwestern United States. Canadian Journal of Fisheries and Aquatic Sciences 40:1173-1184. Hubbard, J.P. 1977. Importance of riparian ecosystems: biotic considerations. Pages 14-19 in R. R. Johnson and D. A. Jones (eds.), Importance, preservation and management of riparian habitat: a symposium. US Forest Service, Fort Collins, CO. Johnson, E. R., L.T. Haight, and J. M. Simpson. 1977. Endangered species vs endangered habitats: a concept. Pages 68-80 in R. R. Johnson and D. A. Jones (eds.), Importance, preservation and management of riparian habitat: a symposium. US Forest Service, Fort Collins, CO. King County Planning and Community Development, Planning Division. 1983. 1983--annual growth report. King County Council, Seattle, WA. 191 pp. Levin, S.A., and R.J. Paine. 1974. Disturbance, patch formation, and community structure in Proceedings National Academy of Science 71:2744-2747. Likens, G. A., F. H. Bormann, N. M. Johnson, D. W. Fisher, and R. S. Pierce. 1970. Effects of forest cutting and herbicide treatment on nutrient budgets in the Hubbard Brook watershed-ecosystem. Ecological Monographs 40:23-47. Lyford, J. H., and S.V. Gregory. 1975. The dynamics and structure of periphyton communities in three Cascade Mountain streams. International Association of Theoretical and Applied Limnology 19:1610-1616. McCormick, Frank J. 1978. Position paper in support of a habitat preservation proposal. Memorandum. Office of Biological Services, US Fish and Wildlife Service, Washington, DC. 19 pp. (METRO) Municipality of Metropolitan Seattle. 1982. BearEvans Creek stream resource inventory. Water Quality Division, Technical Report WR-82-2, Seattle, WA. 52 pp. Montgomery, James M., Consulting Engineers, Inc. 1976. Forest harvest, residue treatment, reforestation and protection of water quality. Prepared for US Environmental Protection Agency, National Technical Information Service, Springfield, VA. 273 pp. Murphy, M. L., and J. D. Hall. 1981. Varied effects of clearcut logging on predators and their habitat in small streams

597

of the Cascade Mountains Oregon. Canadian Journal of Fisheries and Aquatic Science 38:137-145. Narver, D. W. 1972. A survey of some possible effects of logging on eastern Vancouver Island streams. Fisheries Resources Board of Canada, 55 pp. Newbold, J. D., D. C. Erman, and K. B. Roby. 1980. Effects of logging on macroinvertebrates in streams with and without buffer strips. Canadian Journal of Fisheries and Aquatic Science 37:1076-1085. Ohmart, R.D., W.O. Deason, and C. Burke. 1977. A riparian case history: the Colorado River. Pages 35-47 in R. R. Johnson and D. A. Jones (eds.), Importance, preservation and management of riparian habitat: a symposium. US Forest Service, Fort Collins, CO. Orsborn, J. F., and J. W. Anderson. 1986. Stream improvements and fish response: a bio-engineering approach. Water Resources BuUetin 22;381-388. Revised Code of Washington. Chapter 90.58. Simberloff, D. S. 1976. Island biogeography theory and conservation practice. Science 191;285-286. Snyder, D. E., P. S. Gale, and R. F. Pringle. 1973. Soil survey, King County area, Washington. USDA, Soil Conservation Service in cooperation with Washington Agricultural Experiment Station. Washington, DC and Pullman, Washington. 100 pp. Steinblums, I.J., H. A. Froehlich, andJ. K. Lyons. 1984. Designing stable buffer strips for stream protection.Journal of Forestry 82:49-52. Trainer, E.J., and D. E. Suhrweir. 1975. Farm woodlots as biogeographic islands: regulation of tree species richness. Bulletin of the Ecological Society of America 56:53. US Environmental Protection Agency. 1976. Forest Harvest, Residue Treatment, Reforestation and Protection of Water Quality. EPA 910/9-76-020. 273 pp. Van Winkle, W., Jr., and D. C. Martin. 1973. A home range model for animals inhabiting an ecotone. Ecology 54:205209. Washington Department of Ecology. 1981. Western Washington urban stream assessment. Washington Department of Ecology, Office of Water Programs, Water Quality Planning. Olympia, WA. 54 pp. Washington Department of Game. 1984. Wetlands and stream corridors: critical areas worth preserving. Washington Department of Game. Olympia, WA. 44 pp. Weber, P. K. 1981. Comparisons of the lower tropic levels of small stream communities in forest and clearcut sites, southeast Alaska. Ph.D. Thesis, University of Washington, Seattle, WA. 228 pp.