Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
8 Habitat Management and Rehabilitation f From tributaries to mainstem rivers and from headwaters to estuaries through- out the Pacific Northwest, habitat changes have resulted from the cumulative effect of various land uses, cultural developments, and other factors operating on a wide range of temporal and spatial scales. Specific cause-effect relationships involved in habitat degradation are not easy to decipher. Restoration strategies for improving habitats are not always clear, and there is often disagreement among professional biologists, other technical specialists, land-managers, policy- makers, and the general public regarding how to proceed. Programs to restore aquatic habitats have increased in recent years as the extent and magnitude of impacts of management activities and cultural practices on wetlands, streams, rivers, and estuaries have become more widely recognized. In a recent review of aquatic-ecosystem restoration, the National Research Coun- cil (1992a:17) defined restoration as the reestablishment of predisturbance aquatic functions and related physical, chem- ical, and biological characteristics. Restoration is different from habitat cre- ation, reclamation, and rehabilitation it is a holistic process not achieved through the isolated manipulation of individual elements. The present committee concurs with that definition of restoration and emphasizes that in-channel restoration of aquatic habitats can seldom be accomplished with- out considering the accompanying riparian zone or other portions of a watershed that affect the aquatic system. WATERSHED INFLUENCES An essential concept in stream ecology is that terrestrial components of the 204
HABITAT MANAGEMENT AND REHABILITATION 205 environment have a profound influence on aquatic systems. Those components can include watershed characteristics, sediment production, and nutrient cycling. In the upper reaches of the Columbia River Basin, the hydrology of mountainous watersheds is dominated by snowmelt regimes that tend to produce distinct sea- sonal hydroperiods with predictable frequency and duration. In contrast, coastal watersheds along the Pacific tend to have a rain-dominated hydrology with sharply defined storms that are much less predictable in timing, duration, and magnitude. The hydrologic regimes for the two geographic areas are remarkably different, as they are for other areas of the region, but anadromous salmon have adapted their life strategies to all of them. In the Pacific Northwest, many dams and stream diversions and a wide variety of land uses have contributed to altering the flow regimes of streams and rivers. The natural hydrologic regime associated with a particular watershed has an important influence on instream functions and processes and habitat characteristics. Hence, where the natural disturbance pat- tern has been altered through dams, irrigation diversions, or other flow modifica- tions, some degree of restoration of the natural hydrologic regime might be required before restoration of aquatic functions and habitat characteristics can occur (Hill et al. 1991~. In estuarine and depressional wetlands, re-establishing hydroperiods within the range of natural conditions is critical to the restoration of such systems (Kusler and Kentula 19903. Similarly, a variety of land-use practices (e.g., timber harvesting and reading in steep terrain, grazing, dryland farming and agriculture, and urban development and construction) can increase sediment production or alter the timing of its entry into stream systems (Everest et al. 1987, Swanson et al. 1987, McNabb and Swanson 19901. Reducing sedimentation is often another prerequisite for restor- ing aquatic habitats. In some instances, the transport of sediment to downstream reaches can be hindered or prevented. Dams of various sizes from small stockwater ponds too numerous to inventory to mainstem Columbia River dams- all tend to impound and store sediment from upstream reaches, preventing the normal downstream transport and storage of fluvial sediment by a stream or river system. The larger of such structures also tend to flatten hydrograph peaks and so further alter instream sediment transport, particularly that of coarser bedload sediments. Where hydrologic and sedimentation regimes have been seriously altered, a fundamental objective of restoration would be to restore the dynamics of natural flow regimes enough to re-establish the processes and functions of both . . aquatic and r~par~an ecosystems. The riparian environment associated with Pacific Northwest streams and rivers is of primary importance to the functioning of aquatic systems (Johnson et al. 1985, Mutz and Lee 1987, Salo and Cundy 1987, Abell 1989, Gresswell et al. 1989, Chaney et al. 19901. Riparian zones can have pronounced effects on the biological, chemical, and physical components of the aquatic system. From a biological perspective, the structure and species composition of streamside veg- etation influence the local characteristics of both the channel and its aquatic
206 UPSTREAM: SALMON AND SOCIETY IN THE PACIFIC NORTHWEST habitat. Vegetation provides shade and moderates stream temperatures, provides a source of carbon for instream organisms through annual leaf-fall, and, at least for forested systems, provides the large woody debris that is often a key habitat feature (discussed in Chapter 7~. However, riparian vegetation does much more. The underground root systems of streamside vegetation bind soil particles and provide stability to stream banks, thus influencing channel structure. Plant stems and near-surface roots provide flow resistance during periods of overbank flow and thus promote sediment deposition and floodplain development. Root sys- tems, in combination with large woody debris, provide channel roughness ele- ments that not only promote sediment storage but encourage the hydraulic exchange of streamflow and hyporheic (underground) flows. Chemical transfor- mations of various kinds occur in the highly variable oxidation-reduction envi- ronment associated with riparian areas. Because streamside vegetation has such an important influence on the characteristics and productivity of aquatic habitats, restoration of these habitats requires a commitment to the restoration of riparian vegetation functions and processes where they have been substantially altered by human influences. HABITAT-MANAGEMENT OPTIONS The influx of Euro-Americans has altered the characteristics and functioning of stream systems in most freshwater salmon habitats in the Pacific Northwest. The last two centuries and particularly the last 50 years have seen rapid transition of watersheds. Although the degree of alteration varies widely throughout the region, habitat impacts and losses are common. The variability of impacts indi- cates that various options (Figure 8-1) are needed for improving conditions for sustaining anadromous fish populations. Protection Anadromous salmon have adapted over thousands of years to types of habi- tats that existed in the Pacific Northwest before Euro-American settlement. They thrived in naturally functioning aquatic-riparian ecosystems. Where wild popu- lations continue to survive and maintain healthy populations, the protection of intact and functional aquatic-riparian habitats should have a high priority; where protection is the desired management option to sustain fish and other aquatic organisms, human influences need to be prohibited or minimized. Restoration Where aquatic-riparian habitats have been degraded by human activities but have the potential to recover the characteristics that make them functionally equivalent to a pristine system, restoration might be the management target; a
HABITAT NAGEMENT AND REHABILITATION a.., I,............. Management Actions Ecosystem Conditions u, s ._ o ._ A) ._ E .. A) At: s An - ._ U. m 207 Protection: Preservation of areas that are ecologically intact and healthy. Restriction, to the extent possible, of human activities that seriously affect aquatic and riparian ecological functions. The strategy is intended to protect aquatic-riparian systems that are in good condition so that naturally regenerative processes can continue. Restoration: A. Natural restoration Removal of sources of anthropogenic disturbance in altered aquatic-riparian ecosystems to allow natural processes to be the primary agents d recovery. The strategy is to allow the natural disturbance regime to dictate the speed of recovery in areas that have a high probability of returning to a fully functional state without human intervention. B. Actively managed restoration Restoration of dysfunctional aquatic-riparian ecosystems to a state within the range of natural conditions by activey managing some aspects of habitat recovery. The strategy is to combine elements of natural recovery with management activities directed at accelerating development of self-sustaining, ecologically healthy ecosystems. Rehabilitation: Re-establishment of naturally self-sustaining aquatic-riparian ecosystems to the extent possible while acknowledging irreversible changes-such as dams, permanent channel changes due to urbanization and roads, stream-channel incision, floodplain losses, and estuary losses "might permit only partial restoration of ecological functions. The strategy is to combine natural and active management approaches in areas where ecological recovery is possible and to use substitution approaches where ecological seH-suh`iciency cannot occur. Substitution: A. Enhancement Deliberately increasing the abundance or functional importance of selected habitat characteristics as desired. Such modifications might be outside the range of conditions that would occur naturally at a site. The strategy involves technological intervention and substitution of artificial for natural habitat elements. Enhancement activities might shift aquatic-riparian ecosystems to another state in which neither restoration nor rehabilitation will be achieved. B. Mitigation An attempt to offset habitat losses by improving or creating aquatic-riparian habitat somewhere else or by replacement of lost habitat onsite. The strategy involves extensive use of technological intervention and replacement of natural habitats with artificially created habitats. Deg raclatl on: Existing or continued loss of aquatic-riparian habitat and ecological functions due to human activities. The strategy is to continue present practices and accept continued habitat loss. to U} o _ ~ ~ ._ .O ~ ~ ._ = ~ ~ AS .m Cal Q ._ 1 Cal ._ Cal Cal - ~ ~n o E ._ ~ ~ an tn 0 ~ C.) ~ , FIGURE 8-1 Alternative approaches to habitat management based on existing watershed conditions and desired level of improvement.
208 UPSTREAM: SALMON AND SOCIETY IN THE PACIFIC NORTHWEST return to natural streamflow and sediment input might be possible. Generally, such areas have experienced adverse impacts of historical watershed and land- use practices; if the land use can be modified to reduce or eliminate potential impacts on the watershed? s hydrology and sediment production, the prognosis for long-term re-establishment of a natural disturbance regime would be good. In other instances, natural hydrological and sediment-production processes have been not been impaired, but npanan vegetation and channel structure have been changed, and the recovery of riparian vegetation to a more natural condition might be the management goal (Table 8-13. TABLE 8-1 Management Strategies Appropnate for Different Habitat Objectives Management Area Aquatic-Riparian Management Strategy Class 1 waters Bear the closest resemblance to waters unaltered by modern human activities, Protection 1) Identify aquatic-riparian ecosystems in good ecological condition through some contain a complete set of native biota. and method of watershed analysis. have a high degree of natural protection. Each contains a complete set of native fauna, 2) Implement measures designed to prevent a diversity of habitats, and enough area to adverse human impacts, including the maintain viable populations of the largest proscription of all potentially damaging and most mobile species. Management goal: activities within the ecosystem (e.g., no new keep as pristine as possible, recognizing that roads in roadless areas). some biotic change is inevitable or necessary. Class 2 waters Restoration Modified by human activity but contain 1) Inventory riparian and aquatic habitat mainly native organisms and have reasonable characteristics throughout watersheds. potential to be restored to Class 1. Manage ment goal: maintenance of natural diversity and prevention of further degradation, but allow potentially compatible uses (e.g., low impact recreation, selective logging, . . . nonrlparlan grazing). 2) Establish substantial undisturbed riparian zones adjacent to streams, lakes and wetlands where future high impact management activities will occur; allow natural recovery of ecosystem functions (e.g., federally proposed ' PacFish" buffers). 3) Remove high impact anthropogenic disturbances that are now occurring (e.g., fencing riparian zones in grazing areas). 4) Identify opportunities for accelerating the development of desired ecological conditions by actively managing riparian zones (use caution, however) or reconnecting rivers with their floodplains.
HABITATMANAGEMENT AND REHABILITATION TABLE 8-1 Continued 209 Management area Aquatic-riparian management strategy Class 3 waters Appear natural, but their biotic communities have been significantly and probably irreversibly altered. Unlikely ever to be restored to Class 1 but can be refuges for native species or migration corridors for anadromous species. Vulnerable to change and cannot be relied upon for long-term preservation of species. Management goal: maintenance of supplemental populations and gene pools, sources of organisms to stock restored waters, and "wild" areas that can sustain fairly heavy public use. Class 4 waters Artificial aquatic refuges created and/or managed for protecting species that otherwise would likely become extinct. Simulate original environments but require continuous management and monitoring; should be regarded as temporary solutions for saving species or for providing back-up populations for species with limited wild populations. Management goal: short-term back-up for Class 2 and 3 waters. Class 5 waters Artificial refuges with no attempt to recreate natural conditions. Management goal: maintain small areas with some semblance of important habitat characteristics; often species-directed. Rehabilitation/Enhancement 1) Inventory riparian and aquatic habitat characteristics throughout watersheds. 2) Perform watershed-scale evaluation that considers (a) the geomorphic setting of the river and its valley, (b) the natural disturbance regime of the region, (c) the historical patterns of anthropogenic disturbances, and (d) condition of the watershed relative to some type of unmanaged reference site. 3) Prioritize and implement, at the watershed scale, plans to achieve habitat goals. Enhancement/Mitigation 1) Inventory riparian and aquatic habitat characteristics throughout watersheds. 2) Perform watershed-scale evaluation that considers (a) the geomorphic setting of the river and its valley, (b) the natural disturbance regime of the region, (c) the historical patterns of anthropogenic disturbances, and (d) condition of the watershed relative to some type of unmanaged reference site. 3) Prioritize and implement, at the watershed scale, plans to achieve habitat goals. Mitigation 1) Inventory riparian and aquatic habitat characteristics throughout watersheds. 2) Perform watershed-scale evaluation that considers (a) the geomorphic setting of the river and its valley, (b) the natural disturbance regime of the region, (c) the historical patterns of anthropogenic disturbances, and (d) condition of the watershed relative to some type of unmanaged reference site. 3) Prioritize and implement, at the watershed scale, plans to achieve habitat goals. Source: Adapted in part from the aquatic diversity management area concept of Moyle and Yoshiyama (1994).
210 UPSTREAM: SALMON AND SOCIETY IN THE PACIFIC NORTHWEST Management policies directed at the restoration can potentially proceed along two major pathways: natural restoration (referred to as "passive restoration" by Kauffman et al. 1993) and active restoration. In natural restoration, removal of the sources of anthropogenic disturbances is all that is necessary for full restora- tion of the system. For example, where agricultural practices occur in riparian areas, the cessation of such practices might allow the long-term re-establishment of riparian vegetation and associated functions. Natural disturbances would com- bine with the establishment, growth, and succession of riparian plants to assist in restoring aquatic habitats. Similarly, the removal of grazing from streamside zones or a change to grazing policies that allow full recovery of riparian plants along streams might be all that is needed to restore aquatic-riparian functions for many rangeland streams. For forested riparian systems that have previously experienced great amounts of logging, the establishment of no-harvest buffers might provide for the restoration of aquatic-riparian functions and characteristics. The time required for restoration to occur will depend on the local conditions (e.g., species of riparian plants, climate, geomorphic characteristics of stream and valley, and hydrologic disturbance pattern). However, the intent of natural resto- ration is to use fully the natural abilities of physical processes (channel adjust- ments, bank building, scour and fill, etc.), chemical processes (nutrient transfor- mations), and biological processes (establishment, growth, and succession) to restore the functioning of aquatic-riparian systems. Because of the multitude of variables, microclimates, and other conditions associated with recovering ripar- ian systems, natural restoration allows those processes to occur at a level dictated by the local capability of the system. The first step in active restoration also involves the removal or elimination of activities that are causing degradation. However, where monitoring or observa- tion indicates that recovery will not be complete or might require much time, additional management practices can be considered. Active restoration incorpo- rates practices designed to fill an ecological void or accelerate natural recovery. For example, large woody debris may have been removed from the channel when a forested riparian zone might have been harvested years earlier. Although the growth rates and species composition of the second-growth riparian forest could be providing the desired functions within their natural range, the scarcity of large wood in the stream might not be overcome for many decades. In this situation. addition of large woody debris in configurations normally expected for the stream might be undertaken. In other instances, if riparian areas have been used for crop production, eliminating agricultural practices might initiate natural recovery of riparian functions and aquatic habitat; but because native plant species that would be characteristic of the local riparian system are infrequent, natural restoration could take a very long time, so planting native species of riparian plants obtained from locally adapted genetic stock might accelerate recovery. In rangeland areas where prolonged grazing and other practices have caused the disappearance of
HABITATMANAGEMENT AND REHABILITATION 211 willows, cottonwoods, or other key riparian plants along streams, active manage- ment might be needed to re-establish a native vegetation community. The practices involved in active management can vary widely, but the intent is to assist or accelerate the restoration of aquatic-riparian functions and related physical, chemical, and biological characteristics that support natural communi- ties and maintain aquatic productivity. Such practices are intended to aid in re- establishing a sustainable aquatic-riparian ecosystem and aquatic habitat that will be, for all practical purposes, functionally equivalent to pristine conditions. Rehabilitation In many wetlands, streams, rivers, and estuaries in the Pacific Northwest where habitat alteration and loss have been extensive, restoration itself is not feasible. Natural disturbance regimes might have been altered to such an extent that there is little opportunity for restoration. For example, hydrologic and sedi- ment transport regimes could have been affected by dams, irrigation diversions, changes in fire frequency, conversion of lands to agricultural practices, etc.; introduced plants could have replaced native riparian species; channel incision could have lowered local groundwater tables and affected hyporheic interchanges with the stream; estuaries could have been filled; and road construction, agricul- tural practices, or urban development could have reconfigured channel sinuosity or shifted stream location. In many of those situations, self-sustaining aquatic- riparian ecosystems that can provide important habitat for anadromous salmon might still be possible, but increased levels of human effort (time, money, and management persistence) will probably be required because of the extent and magnitude of the changes. Habitat management can then be directed toward re- establishing self-sustaining conditions that are able to provide some of the eco- logical requirements of anadromous fishes i.e., rehabilitation. Rehabilitation of habitat features can occur, but full restoration to predisturbance functions and characteristics is unlikely (Figure 8-1~. An important aspect of the rehabilitation option is that not all the predistur- bance aquatic-riparian functions can be restored. For example, if a dike had been constructed along a stream's edge, moving the dike back from the channel would allow the return of streamside vegetation and some floodplain functions, such as temporary storage of floodwaters, sediment deposition on floodplain terraces, and improved interactions between stream and groundwater. Even though the channel might not be able to develop full predisturbance sinuosity, the prior density of side channels, or full floodplain functions, a major improvement in aquat~c-riparian functions and characteristics would be achieved by the reposi- tioning of the dike. Rehabilitation projects constitute important opportunities for developing improved and sustainable habitats for salmon and other aquatic-ripar- ian biota.
212 Enhancement UPSTREAM: SALMON AND SOCIETY IN THE PACIFIC NORTHWEST Substitution Substitution approaches to habitat management (Figure 8-1) are generally directed toward selectively altering or modifying habitat features to offset the effects of anthropogenic impacts. In the last decade, numerous instream habitat enhancement projects have been undertaken throughout the Pacific Northwest. Many involve the placement of gabions (wire baskets filled with rock), boulders, riprap, large wood, or other large structures in stream channels. The addition of such roughness elements might be outside the range of conditions-with respect to size, spatial distributions, orientation, etc.-that would be expected to occur natu- rally in the setting. For example, the placement of boulders or logs in meadow systems that historically did not have these large roughness elements is outside the range of expected natural conditions. Exotic materials, including gabions and geotextile fabrics, are commonly used in enhancement projects. For lakes, the addition of fertilizers to boost production would constitute an enhancement ap- proach. Enhancement can provide important opportunities for improving fish habitat in some instances, but often it has not been very successful in improving conditions that sustain productivity. The National Research Council's 1992 re- port on the restoration of aquatic ecosystems includes a strongly worded caution about the practice of enhancement (pp. 222-223~: Practitioners of species-centered stream management generally introduce artif;- cial structures into stream and river environments to modify banks, channel, bed, or current in hopes of improving salmonid or other game fish productivity. When this work is done without a profound understanding of the interactions among stream hydrology, fluvial geomorphology, and fish, the least detrimental consequence may be that mechanical structures emplaced in the stream at con- siderable expense and trouble could be of limited durability and longevity. The term enhancement implies improvement and betterment of a system, but it is important to be aware that if ecosystem needs are misinterpreted, a stream- habitat enhancement project can actually shift an aquatic ecosystem from one degraded state to another (Figure 8-13. For example, fully spanning logs with bank revetments might be placed in a stream that is deficient in pools to provide additional pool habitat. If the new pools are without cover and losses to predation are increased, if the logs create waterfalls that become barriers to juvenile or adult fish movements, or if the channel can no longer adjust to high flows and sediment transport by altering sinuosity or creating natural pools, the long-term conse- quences of the enhancement project might simply be another form of habitat degradation.
HABITAT MANAGEMENT AND REHABILITATION Mitigation 213 Where habitat losses are unavoidable, mitigation is a management option that attempts to minimize or offset the effects of loss by creating new habitat. Although the concept of mitigation is relatively simple, its application often is not. Losses of habitat at a particular site are seldom balanced by mitigation at another site; i.e., replacement of physical habitat rarely includes replacement of all relevant ecological interconnections. The ecological ledger is not linear and cannot be "balanced" on the basis of simple tabulations of side channels, lengths of streams, or areas of spawning gravels created. If a key component of a required habitat feature is severely altered or destroyed, there might be no suit- able means of mitigating that impact. Habitat-management projects tend to focus on the characteristics and needs of a specific stream reach, but the condition of individual habitats, stream reaches, and entire tributary systems within a watershed must be considered for restora- tion planning to be effective. It might do little good to invest time and money in restoring spawning gravel for salmon if rearing habitats are scarce or downstream barriers severely limit accessibility to them. It might do little good to emphasize riparian restoration when excessive sediment production from upstream land uses or streamwater withdrawals are adversely affecting instream habitats. Habitat managers need to be aware of how habitat alterations affect the life-history stages of salmon and other aquatic organisms on various spatial and temporal scales. They must also understand how conditions change as a result of a wide range of anthropogenic perturbations and natural disturbance patterns. The complexity of environmental factors and options available to habitat managers presents a major challenge to the reduction of anthropogenic impacts. WATERSHED ANALYSIS Fish habitat upstream in a river basin might be little affected by human activities, but habitat lower in a watershed could experience the cumulative ef- fects of a wide range of human activities or other factors. Understanding factors that affect the availability and quality of aquatic habitat at lower sites is much more difficult. In addition, the importance of various contributing factors will probably change in emphasis or magnitude from watershed to watershed. Thus, there is an increasing need to understand cumulative effects not only on a site- specific basis, but also across entire watersheds. Only through a broad geo- graphic perspective can the unique qualities of each watershed and their spatial and temporal effects on aquatic habitats be effectively understood. A recent development in forest-management planning has been the proce- dure of watershed analysis to evaluate resources and the potential environmental impacts of land-management proposals. The general goal of watershed analysis is to combine habitat-inventory information with environmental-hazard assess
214 UPSTREAM: SALMON AND SOCIETY IN THE PACIFIC NORTHWEST meets over a relatively large area, usually encompassing a fourth- to sixth-order stream network, so that land-use prescriptions can be based on stewardship ob- jectives and opportunities for habitat restoration can be identified on somewhat larger geographical scales than are normally used. For forested basins, watershed analysis can lead to management prescriptions that provide greater environmen- tal protection than standard forestry rules. The procedure was created in the Timber, Fish, and Wildlife Program in Washington State to address the cumula- tive effects of logging-related activities and has been incorporated into the state's forest-practices laws (Washington Forest Practices Board 1993J. Idaho is devel- oping a cumulative-effects analysis and control process designed to protect water quality from forested watersheds so that beneficial uses are supported (Idaho Department of Lands 19941. A comprehensive study is under way in Oregon to identify the cumulative effects of forest practices on air, soil, water, fish, and wildlife (Beschta et al. 1995), and recent revisions to Oregon's forest-practices rules include a watershed-analysis option under certain circumstances. At the federal level, management guidelines proposed by the Forest Ecosystem Man- agement Assessment Team (FEMAT) for regulation of national forests in west- ern Oregon, western Washington, and northern California include watershed analysis as an important underpinning for future resource prescriptions and man- agement (FEMAT 1993~. Watershed analysis as now envisioned requires a landscape-scale perspec- tive, although it is too early to determine whether its application will result in substantive improvements in habitat protection or in a more comprehensive ap- proach to aquatic-habitat restoration. Its success will ultimately depend on how managers translate habitat-inventory and hazard-assessment information into pre- scriptions and on the strength of the commitment to effective monitoring (the Washington watershed-analysis procedure encourages monitoring but does not require it). Watershed analyses and assessments now being implemented (Wash- ington Forest Practices Board 1993) and those suggested by FEMAT (1993) are designed to promote efficient regulation, continued use of forest resources for timber production, and the protection of forested ecosystems. Considering entire watersheds in the regulation of forest practices has many benefits, but it also presents several practical and conceptual difficulties (Wash- ington Forest Practices Board 1993:v): 1. Watershed ecosystems involve a complex dynamic between many water- shed and biological processes operating at many spatial scales. Scientific un- derstanding of these processes is limited, and comprehensive reliable techniques for evaluating watersheds are lacking. 2. The physical and biological characteristics of a watershed and sub-areas within it reflect the local geology, terrain, climate, vegetation and so on. Con- sequently, every watershed is unique, with its own distribution of these factors as well as effects due to the history of past disturbance including natural events or land use.
HABITAT MANAGEMENT AND REHABILITATION 3. Because of these differences in landscape features, the sensitivity of water- sheds and sub-a~eas within them to forest practices also varies from place to place. While one location may generate no likelihood of local or cumulative effects from an activity, the same activity conducted in the same way in another location with heightened sensitivity could have both local and cumulative im pacts. 215 Although those difficulties appear to represent barriers or constraints to water- shed analysis, they actually provide compelling arguments for doing it. Spatial and temporal variability; the dynamic interactions between physical and biologi- cal processes; the unique attributes of each watershed (geology, terrain, climate, vegetation, fish populations, history of land use, and natural disturbances); the local sensitivities of watersheds, sub-areas, and stream reaches to management practices; and other factors all suggest that some form of comprehensive analysis is necessary for ecologically sensitive management planning. A simple "stream- reach analysis" is obviously inadequate for understanding watershed-scale con- cerns, processes, and cumulative effects that are driven by land-use practices or by institutional and social programs. The use of watershed analysis for improving habitat should be directed at providing the public and managers with information that will identify a range of issues and opportunities for Pacific Northwest streams. Because a wide range of spatial and temporal scales needs to be considered for a given watershed, such analysis is expected to yield both strategic and tactical approaches to improving habitat. Several types of information should be considered in the analysis: . Spatial context. The highly varied nature of fish habitat requires that geomorphic characteristics of aquatic habitats be identified throughout a basin. The setting of various stream reaches (e.g., constrained vs. unconstrained, sinu- ous vs. braided, incised vs. unincised, and bedrock-controlled vs. alluvial flood- plains) can constitute an important spatial context from which to understand both reach and watershed scale problems, issues, and habitat-management opportuni- ties. Without a spatial perspective, inappropriate and counterproductive habitat manipulations or alterations of selected reaches might be selected. A geographic information system iGIS) might be useful for displaying and analyzing some types of spatial information, but the large variability in conditions associated with specific stream reaches indicates that on-the-ground assessments of habitat and related factors should have high priority in watershed analysis. · Temporal context and disturbance regimes. Characterizing the temporal variability of flow patterns and sediment yields, long-term channel adjustments, climatic patterns, vegetation succession patterns, fire history, or other factors provides an important perspective on the dynamic features of a particular water- shed and stream system. The peak-flow regime of a given watershed is especially important because the instream biota, riparian vegetation, and many channel- forming processes are fundamentally tied to the frequency and magnitude of
216 UPSTREAM: SALMON AND SOCIETY IN THE PACIFIC NORTHWEST hydrologic events. In other instances, the occurrence of low flows can affect habitat quality and productivity, and managers need to recognize the implications of recurring drought and its effects on habitat dynamics and quality. For most watersheds, long-term data are generally lacking or are of varied quality; extrapo- lation of temporal information from sources outside a watershed is often neces- sary. Managers need to develop an understanding of the randomness of natural- disturbance regimes and incorporate that understanding into strategies for the maintenance or improvement of sustainable aquatic habitats. · Riparian vegetation and reference sites. Because of the importance of riparian vegetation to many channel and aquatic-habitat characteristics, an under- standing of riparian plant communities is fundamental. Furthermore, the mecha- nisms by which riparian vegetation interacts with natural-disturbance patterns and land-use practices need to be thoroughly understood on stream-reach and watershed scales. Because many riparian plant communities have been affected by historical land-use practices, reference sites that consist of ecologically intact and functional aquatic-riparian systems should be identified. It may not be pos- sible to restore or rehabilitate all aquatic habitats in a watershed to the same functional level, but reference sites are necessary to increase understanding the complex interactions between streamside vegetation, channel characteristics, and aquatic habitats. They provide fundamental information on types of processes, functions, and desired future conditions of intact riparian systems. · History of impacts. Institutional, scientific, or social records of human land use and changes in aquatic and riparian ecosystems in a given watershed are often incomplete. However, a thorough understanding of historical practices and their effects is very helpful. Important insights into current conditions are often gained when historical information is developed and used to understand the mag- nitude and extent of human perturbations. It is important for both society and fishery managers to understand the magnitude and extent of changes that have occurred over periods of decades or longer. Watershed analysis requires gathering a large amount of inventory informa- tion to obtain an improved understanding of how spatial and temporal patterns, cause-effect relationships and other interactions, and cumulative effects occur in a particular watershed or stream reach and how they affect aquatic habitats. In watershed analysis, the best available technical understanding and scientific in- formation can be focused on a particular watershed. The process can also high- light limitations of databases, lack of information, monitoring needs, and so on. Most important, watershed analysis can be adapted to the goals of managing a particular resource or group of resources. It can provide information about man- aging a specific reach of stream; in the case of anadromous salmon, whose range transcends institutional and land-ownership boundaries, results of a watershed analysis can be joined with other socioeconomic considerations to develop habi- tat-management priorities on a watershed scale.
HABITAT MANAGEMENT AND REHABILITATION 217 Watershed analysis can provide a relatively clear understanding of existing aquatic resources and factors that affect them, but social preferences and institu- tional constraints might preclude implementation of particular solutions to habi- tat problems. For example, dewatering of a stream by irrigation diversions seem- ingly could be solved by shifting to irrigation techniques that are more efficient and less consumptive. Any water savings might be retained in the stream to maintain summer rearing habitat. However, improved water-use efficiencies might instead allow a landowner to increase the amount of area under irrigation with no net change in the amount diverted. If the streamwater were no longer used by the landowner, it might simply be diverted by another landowner imme- diately downstream with a junior water right. In essence, what could have seemed like a relatively simple approach to solving an instream problem begins to in- volve important institutional barriers. The institutional constraint of "use it or lose it," which is so deeply etched in western water law, precludes what in some instances might be a simple solution to an instream habitat problem. When the original water laws were promulgated, concerns for fish and aquatic habitat had a low priority. Similar institutional constraints occur with respect to other re- sources and land-management practices. Unless the role and importance of vari- ous social, economic, institutional, and population factors that affect aquatic habitats are considered, the potential for effective maintenance or improvement of habitat for anadromous salmon and other aquatic organisms will be greatly constrained. Whereas watershed analysis might provide important resource perspectives previously unavailable to land managers, it is important to point out that water- shed analysis is currently designed only for drainages with forestry operations (e.g., Washington Forest Practices Board 1993~. There are no institutional or legal means of applying the watershed-analysis approach to the management of nonforest lands. And it is not known whether the methods used to assess the consequences of forest management and provide recommendations for habitat restoration are fully applicable to streams and lakes surrounded by land used in other ways. Thus, there is no institutionally sanctioned means of assessing habi- tat and identifying opportunities for restoration in the case of over multiple own- erships encompassing a variety of land uses. The freshwater-habitat needs of anadromous fish, however, do not stop at forest boundaries. OPPORTUNITIES AND CHALLENGES Numerous opportunities await those interested in improving and restoring aquatic habitats in the Pacific Northwest. In some instances, substantial recovery can be accomplished by the simple cessation of streamside management practices that cause local degradation of freshwater habitats; restoration might proceed rapidly and be easily observed, or useful recovery might take a long time. Most degraded aquatic habitats take years or even decades to have their natural produc
218 UPSTREAM: SALMON AND SOCIETY IN THE PACIFIC NORTHWEST lion fully restored. Given current knowledge about habitat restoration and reha- bilitation and the large amount of freshwater habitat in need of improvement, habitat managers should be more concerned about whether a particular degraded habitat is improving (moving in the right direction) and less concerned about whether it has attained a specific desired condition. Aquatic habitats are complex and dynamic; they vary over large temporal and spatial scales in response to local structure, vegetation, and unpredictable natural disturbances. Attempting to "mi- cro-manage" individual stream reaches to meet a perceived potential is probably unrealistic and unneeded. Some of the changes in terrestrial and aquatic ecosystems in the Pacific Northwest caused by Euro-American development are permanent, some are de- clining in importance with respect to their effects on aquatic habitats, and others are growing in importance. Nevertheless, degraded aquatic habitats can be im- proved in large portions of the Northwest's stream systems. Most of the length of a stream network comprises relatively small streams, which provide water, nutri- ents, organic matter, and sediment to downstream reaches. Minimizing adverse impacts on these small streams and restoring their ecological connections are not only technically feasible but of paramount importance if conditions in many larger streams and rivers are to be improved. Habitat managers often view restoration practices from spatial and temporal perspectives that are too limited, do not match the life histories of the salmon species of concern, and fail to address important ecological processes that are responsible for maintaining natural productivity. The approach to habitat im- provement has often involved introducing habitat elements to streams and lakes in an attempt to boost productivity, e.g., placing structures in streams, excavating spawning channels or riverine ponds, or adding fertilizer to rearing lakes. Such habitat enhancement is used when there is evidence that existing habitat is defi- cient in some way and when it is believed that creation of new conditions will eliminate a major bottleneck in salmon production. Enhancement, based on a "limiting-factor" analysis of the current situation (Reeves et al. 1989), assumes that information about the factors controlling the abundance of the populations in question is sufficient for managers to rehabilitate degraded habitat in a cost- effective manner. In many cases, however, habitat rehabilitation has been under- taken without either formal or informal analysis of limiting factors but instead has been based on the presumed extent of degradation and the efficacy of existing restoration methods. In other cases, habitat is modified with the intent of increas- ing productivity beyond what would be expected in normal conditions. The whole-lake fertilization project of British Columbia's Salmon Enhancement Pro- gram, which was intended to improve survival and growth of juvenile sockeye salmon in oligotrophic (nutrient poor, unproductive) lakes (Hyatt and Stockner 1985', is an example of an attempt to increase salmon production above natural levels. Despite the large amounts of time and money that have been devoted to
HABITAT MANAGEMENT AND REHABILITATION 219 habitat restoration and enhancement by federal agencies, state agencies, and oth- ers, few projects have been shown unequivocally to increase salmon populations (e.g., Hilborn and Winton 19931. Some of the reasons for the failure to demon- strate success are the high inherent variability of freshwater salmon production, erroneous assumptions about production bottlenecks, including the ocean, and the time required to demonstrate the effect of restoration, and unwillingness to monitor biological responses to project implementation effectively (Lichatowich and Cramer 1980, Hall and Knight 1981, Sedell and Beschta 1991, Hilborn and Winton 1993~. A central difficulty for many habitat-alteration projects is a general failure to match the scale of the project to the scale of life histories of salmon in a river basin. In many cases, the scales do not match. For example, many habitat- restoration projects in Oregon and Washington are targeted at increasing the rearing capacity of streams for echo salmon or steelhead. The methods might involve constructing pools and placing large woody debris in the channel (House and Boehne 1986) or, when coho are the species of concern, creating riverine ponds for overwintering (Cederholm et al. 1988~. It is easy to demonstrate that enhanced habitat is occupied by target species, but such projects are typically applied only to a small portion of the total drainage network inhabited by salmon because of their high cost and the boundaries imposed by land ownership. As a result, overall benefits (if any J of restoration and enhancement projects cannot provide measurable signals in terms of increased smolt production or run size. The "limiting-factor" approach has been used by fishery biologists to assess habitat deficiencies and as a basis for proposing enhancement projects. It pro- vides a means of identifying site-specific habitat defects, such as excessive water temperature, lack of pools, insufficiency of large woody debris, excess of fine sediment, or lack of flushing flows. Those might be important factors, but the approach has often resulted in additions of whatever was believed to be in short supply (Beschta et al. 1991, Beschta et al. 1993, Kauffman et al. 1993~. The simple alteration of physical features in streams or lakes does not necessarily promote restoration of habitat when riparian and watershed management prac- tices continue to exert their effects on the aquatic ecosystem. Attempts at im- proving habitats by adding in-channel roughness elements without eliminating management practices that are causing habitat degradation are likely to fail. For example, some short-term habitat benefits might be achieved by adding large woody debris to streams, but the benefit can be only temporary from an ecologi- cal perspective unless riparian management practices ensure the long-term re- cruitment of large woody debris from the riparian zone. Limiting-factor analyses often ignore the hydrogeomorphic features of a particular stream. For example, boulders, logs, or other roughness elements might be used in inappropriate locations, such as wet meadows, where they were probably never present. Habitat treatments that have been developed in one ecoregion are freely transferred to another or transferred from one channel type to
220 UPSTREAM: SALMON AND SOCIETY IN THE PACIFIC NORTHWEST another with little regard to the geomorphic context of a stream reach or the character of the streamside vegetation (e.g., House and Boehne 1986, Seehorn 19921. The apparent inability of limiting-factor analysis to incorporate a more holistic perspective of instream habitat might underlie the current shift toward "ecosystem management" that is being proposed for federal forest ownership in the Pacific Northwest (FEMAT 19931. The concept of "desired future conditions" (DECs) has been suggested as a means of defining habitat goals, such as the number of pools per mile of stream or the abundance of large woody debris. Identifying DFCs requires that the charac- teristics of fully functional aquatic ecosystems be known for each site. DFCs might overemphasize the physical attributes of a stream reach unless sufficient recognition is given to other factors or temporal processes that resulted in the attributes (e.g., species composition, structure, and successional patterns of ripar- ian vegetation, natural flooding regimes, and fire) or to how a particular reach of stream operates within a larger spatial context. Although DFCs on a stream- reach scale might remain a component of habitat-improvement goals, care must be exercised in their application across varied ecological and structural templates. Landscape-scale DFCs are needed to protect the integrity of local breeding popu- lations where trends within river basins indicate repeated patterns of habitat alteration. Some examples of DFCs include · Increased percentage of riparian zones with late-successional forest char- acteristics where early-successional forests are dominant. · Increased percentage of reaches with free-flowing discharge regimes in river basins where flows are largely controlled. · Increased connections of rivers to floodplains and side channels where riverbanks have been extensively contained within dikes and levees. Reduced watershed erosion where human activities have accelerated sedi . ment inputs. · Increased areas of high-quality habitat throughout river basins that pro- vide protection to local demos. In recent years, increasing attention has been directed at understanding and improving riparian habitat in Pacific Northwest streams (e.g., Elmore and Beschta 1987, Salo and Cundy 1987, Gresswell et al. 1989, Chaney et al. 1993, FEMAT 19933. Because naturally functioning riparian ecosystems are crucial to sustain- able and productive aquatic habitats, riparian plant communities along stream channels are often protected by setting a specific distance from the channel within which anthropogenic disturbances are minimized or excluded. That can be useful for planning purposes, but the specific width of buffer strips, riparian reserves, or streamside management areas might change locally, depending on structural features, stream size, desired level of protection, and location within the drainage system. As one goes downstream, discharge increases and flood
HABITAT MANAGEMENT AND REHABILITATION 221 plains widen and riparian reserves or buffer strips tend to increase (FEMAT 19933. Both large and small streams need space to allow their channels to adjust to continually changing flow and sediment loads. Where artificial confinement of channels and degraded riparian areas have occurred, the ability of a channel to respond to natural events is severely constrained. Streams given room to interact with streamside vegetation can usually be recovered, but restoration of aquatic habitat where roads, buildings, riprap, berms, or other types of structures en- croach on a stream is often unsuccessful. The level of risk associated with a specific management practice or restora- tion effort should be considered and acknowledged. For example, in the design of a bridge or other structure of high value, a safety factor is often incorporated into the final design to account for unknown factors and the longevity of the structure. Similarly, the dimensions of riparian protection zones should also include safety factors to allow for natural disturbances, uncertainties about the ecosystem of interest, and changes in public values. If an additional margin for error is allowed, the probability of habitat improvement becomes greater and options for future management decisions are increased. PROPERTY RIGHTS AND HABITAT PROTECTION ON PRIVATE LANDS Land ownership carries the right to undertake a wide range of activities. But streams, fish, and other organisms that use aquatic environments are generally considered to be publicly owned resources. Where private lands abut streams, rivers, wetlands, and estuaries, the demarcation between private and public rights can become ambiguous. Ecosystem boundaries are spatially irregular and can change with time. Owing to their wide-ranging life-cycle requirements, salmon often cross many property lines during their migrations to and from spawning and rearing areas. Although much of their time in freshwater might be spent in streams, rivers, and lakes on publicly owned lands, salmon can reside in or pass through privately owned lands. Recent efforts to protect habitat in federal forests have led to formulation of an aquatic-conservation strategy that calls for a net- work of wide buffer strips adjacent to all streams and lakes (FEMAT 19933. Depending on the number of tributaries in a watershed, such a system of buffer strips can account for up to about half the total land area. The aquatic conserva- tion strategy in FEMAT (1993) has been endorsed by the Snake River Salmon Recovery Team (1994) as an appropriate habitat-conservation measure for en- dangered salmon in the Snake River basin. If applied to privately owned forests inhabited by salmon, the system of large buffer strips recommended by FEMAT (1993) would undoubtedly impose severe financial hardships-reduced return on investment for shareholders in forest-products companies or reduced incomes for small woodlot owners. Applying similar restrictions to nonforest landowners in rangeland, agricultural, or urban areas could produce similar hardships.
222 UPSTREAM: SALMON AND SOCIETY IN THE PACIFIC NORTHWEST Yet there is little doubt that over the last century land and water uses on many privately owned lands have continued to degrade aquatic habitat and re- sulted in loss of the natural production capacity of these waters (Lichatowich 1989, Thomas et al. 1993, Moyle and Yoshiyama 19941. Uniform and consis- tently applied habitat-conservation strategies are not practiced on the scale of river basins, the scale most relevant to the metapopulation structure of Pacific salmon. The dilemma is clear. How can private-property rights be respected while adequate habitat is provided for salmon across the landscape? The committee believes that progress toward solving the dilemma is possible and recommends that attention be given to developing a more equitable and more uniform system of habitat-protection requirements on private ownerships across all land uses, establishing joint planning groups for entire river basins (or subbasins), where private landowners can participate in land-use policy deci- sions, investigating various incentives for landowners to practice improved envi- ronmental stewardship, and expanding programs that involve the public in moni- toring and habitat-conservation projects. Those steps would benefit not only salmon but virtually all public values associated with aquatic-riparian ecosys- tems. At present, different environmental laws apply to privately owned lands under different types of management. Forestry operations are regulated by state forest-practices acts. Range and agricultural lands are regulated by voluntary best-management-practices or provisions of water-quality laws. Urban-industrial lands are regulated by pollution-control and local greenway ordinances. The intent of such laws might or might not include safeguarding the ecosystem pro- cesses necessary to maintain aquatic productivity; for example, some laws re- quire landowners to do only what is necessary to make water safe for human consumption. The committee suggests that habitat for salmon and other impor- tant aquatic resources will benefit greatly from a system of environmental man- agement that transcends the type of use for which private lands are zoned and that acknowledges the need to protect interactions between aquatic and terrestrial ecosystems. A system need not require identical protection measures (e.g., buffer- strip width) in every situation. Local conditions and landscape-level consider- ations can provide landowners with some management flexibility. However, more emphasis must be given to protecting the quality of the land (and especially the riparian zone) that affects freshwater ecosystems if environmental conditions are to improve and habitat degradation is to be reversed. New management systems might take the form of an integrated land-use practices act that applies to all private-land uses and that includes provisions for protecting both riparian zones and water quality. The committee believes that providing greater consis- tency in protecting aquatic habitat on private lands should have high priority in state and local governments. A second means of improving habitat protection on privately owned lands is to involve property-owners more fully in environmental-policy matters so that
HABITATMANAGEMENT AND REHABILITATION 223 they have more ownership in habitat-conservation decisions. Property-owners are often concerned with environmental quality and participate in outdoor recre- ational activities, such as hunting and fishing. Landowners might resent being told what they can and cannot do with their land, but they usually understand the value of having abundant natural resources nearby. They might be reluctant to embrace federal and state regulations administered by bureaucrats who are not familiar with local conditions, but they often take pride in displaying improved environmental stewardship to their friends and neighbors. As opposed to being told "this is how you must do it," private landowners are usually more responsive to "this is what you should protect, but you can be creative in helping to design protection measures that fit your situation." Local planning organizations and conservation boards (e.g., soil-conservation districts) could be useful forums for stimulating private-landowner involvement. A few local property owners prac- ticing improved habitat conservation can act as examples and catalysts for others. On the watershed or river-basin scale, private landowners can participate in joint planning organizations that help to set environmental policy and promote environmental stewardship. Many such organizations have formed in Oregon, Washington, and British Columbia with the goal of improving fish and wildlife habitat through rehabilitation projects and increased riparian protection (Krueger 1994, C.L. Smith 1994~. Membership often includes representatives of federal and state fish and wildlife agencies, conservation organizations, tribes, private landowners, and commercial fishers. In the face of such diverse interests, progress can be slow (Halbert and Lee 1990), but those involved value having a seat at the table and are more likely to accept responsibility for reaching consensus on private-property issues (Pinkerton 1993~. A third approach to improving habitat on privately owned land is an ex- panded system of conservation incentives. Incentives can include tax deductions for riparian protection, conservation easements, and cost-sharing programs for restoration projects. Some of the most successful examples have been associated with rangeland where cost-shared riparian fencing has lowered livestock damage to streamside areas (Elmore 19921. Conservation easements can also provide financial incentives for habitat protection. Some incentive programs have not succeeded, however. Oregon instituted a Riparian Tax Incentive Program (RIPTIP) in the 1970s to provide tax relief to private landowners for increasing riparian protection. The value of the tax incentive in many cases was insufficient to encourage property-owners to participate; less than 50 miles of streams had received additional protection when the program was terminated. Most property- owners felt that the program did not provide enough reward to offset the financial losses that resulted from forgone management opportunities. The failure of RIPTIP suggests that incentive programs should not be based strictly on tradeoff between environmental and economic interests, because the latter will most often prevail. Finally, the public can be encouraged to participate more fully in monitoring
224 UPSTREAM: SALMON AND SOCIETY IN THE PA CIFIC NORTHWEST programs and habitat-restoration projects. Dedication to habitat protection is increased when people take part in long-term habitat monitoring (Hellmund 19931. They begin to recognize the importance of less-visible but nonetheless critical linkages between aquatic and terrestrial ecosystems. Regulatory agencies often fail to enlist public involvement in monitoring programs, believing that the programs must remain solely in the hands of technical specialists. But simple habitat features can be monitored by lay persons, and public participation can be a powerful tool not only for expanding habitat databases but also for environmen- tal education. A better-educated public is far more likely to recognize the long- term value of protecting aquatic resources through individual actions that pro- mote environmental stewardship than a public that tends to lay the problem at someone else's doorstep. Likewise, management organizations can enlist the aid of property-owners in carrying out habitat-restoration activities, such as replanting native vegetation in riparian zones. Public interest in enhancing salmon is generally high, as evidenced by the growing number of local fishery enhancement organizations. Many of the projects have emphasized small-scale artificial production tech- niques (e.g., egg boxes and small rearing ponds), but it should be possible to generate more interest in restoring degraded habitat, particularly in riparian zones. Such projects might not produce immediate benefits in returning adult salmon, but private landowners can take comfort in the knowledge that their efforts will help to maintain natural productivity so that their children will be able to enjoy wild fish. BURDEN OF PROOF Many of the improvements in watershed-management practices that have occurred in past decades have required that substantial impacts or harm be dem- onstrated before changes would occur. For instance, early stream-temperature studies showed that removal of riparian shade during logging could have impor- tant adverse affects on summertime stream temperatures (e.g., Brown and Krygier 1970), but only after the evidence was clear were shade requirements instituted. Other studies found an association between accelerated sediment production and particular types of roads and logging practices (for summaries of research, see Ice 1985, Swanson et al. 19871. Management practices associated with forested areas have continued to evolve when research results linked particular practices to potentially adverse changes in sediment production, channel stability, water quality, flow regimes, and fish habitat. Changes in forest practices came only after major impacts had been identified. Furthermore, the burden of proving damage was generally on state agencies that were claiming that particular prac- tices had detrimental effects. Similar situations exist for rangelands and agricul- tural areas; important impacts of management practices must be well documented
HABITAT MANAGEMENTAND REHABILITATION 225 (e.g., continued violation of a water quality standard) before changes in the practices are considered. A major change with regard to the burden of proof has apparently occurred on many federally owned forest lands (FEMAT 19931. Relatively large riparian buffer strips are intended to provide high levels of protection for fisheries and other riparian-dependent wildlife species. Although the dimensions and configu- rations of the riparian reserves can be changed, such change can occur only after a watershed analysis has been conducted (FEMAT 19933. Alterations in riparian buffers, particularly decreases in width, are allowed only if it can be demon strated that alterations will not adversely affect water quality, wildlife, fisheries, or other aquatic organisms. This shifting in the burden of proof is a major change in how forest resources on federal lands are managed. The extent to which the shift will carry over onto state-owned or privately owned lands or other types of land use is not known. HABITAT MANAGEMENT AND FISHERIES MANAGEMENT There is critical interaction between habitat management and fisheries man- agement. Many habitat alterations do not directly extirpate populations, but instead reduce survival rates so as to reduce sustainable exploitation rates. Deter- minations of sustainable exploitation rates are based on analyses of long-term population performance; therefore, there is increasing risk of catch rates remain- ing too high as habitat and survival deteriorate more rapidly than productivity assessments can be updated. It is important that habitat degradation and reduced ocean productivity not be used as excuses to continue overfishing, especially where information gathering and adaptive responses are delayed even if manage- ment agencies are determined to respond as wisely as possible. Greater attention should be paid by fisheries managers to trends in condition of freshwater habitat.