Basin assessment is an important component of the President's Forest Plan, yet it has received little attention. Basin assessments are intended both to guide watershed analyses by specifying types of issues and interactions that need to be understood, and, eventually, to integrate the results of watershed analyses occurring within a river basin. Problems facing basin assessment efforts include the lack of agency interest in the process, achieving interdisciplinary analysis, assembling a functioning team, reconciling the needs of different agencies, the perceived complexity of the problem, ensuring a credible result, and soliciting public participation. These problems are shared by watershed analysis efforts, and observations of the watershed analysis process indicate ways that each might be addressed. However, basin assessment may itself serve as the solution to several of these problems for watershed analysis.

The Forest Ecosystem Management Assessment Team (FEMAT 1993) recommended that watershed analyses be carried out to better understand environmental requirements in 50- to 500-km² watersheds, and the Record of Decision for Amendments to Forest Service and Bureau of Land Management Planning Documents Within the Range of the Northern Spotted Owl (USDA and USDI 1994; the Record of Decision and the accompanying Standards and Guidelines are referred to here as the "ROD") formalized the requirement (ROD p. B-21). The 50- to 500-km² range was selected as a compromise between the minimum size needed to evaluate large-scale and downstream effects and the maximum size that reasonably allowed analysis over a 2-month period. However, FEMAT also realized that even a 500 km² watershed is too small to understand many issues. Some of the most important environmental concerns focus on changes occurring on mainstem rivers and in estuaries with drainage basins of tens of thousands of square kilometers. FEMAT therefore specified that two additional levels of analysis would be carried out: river basin assessment and regional assessment (FEMAT 1993 p. V-I-4). The FEMAT report itself was the regional assessment. The basin assessment will be conducted for entire river basins, from their upstream origin to their point of entry to the sea. Basin assessments are intended to provide input to watershed analyses on larger-scale issues, and to integrate the results of multiple watershed analyses as they pertain to those larger-scale issues.

Many of the characteristics of watershed analysis are shared by basin assessments. First, they are to be broad, qualitative descriptions of the types of issues and processes active in the river basin. Second, they are not "decision documents" in the NEPA sense; they will not allocate land or provide land-use recommendations. Future project planning will make use of the assessments as background information, but the assessments will not identify particular projects or set guidelines for activities. Third, they are intended to act as open files. Land-use priorities, resource conditions, and issues of concern will change through time in the river basin, and such changes will be noted as they occur.

Other characteristics make basin assessments quite different from watershed analyses. In the first place, there is far less guidance available for basin assessments. Watershed analysis is discussed extensively in the ROD, and a manual has been prepared that outlines a procedure for completing the analysis (REO 1995). In contrast, basin assessments are mentioned only once in the ROD (p. B-22), and no guidelines exist for their preparation. Second, no activities depend on the completion of basin assessments, so they have not attracted the attention that watershed analysis has. There is little motivation to complete them, and little interest in participating in the assessments. Third, along with the lack of motivation and interest comes a lack of funding.

Some resource experts have questioned the need for a broad assessment that has no specified planning role, while others have questioned how any resource planning and impact assessment could have been carried out in the past without such an assessment. Federal land management in the west has a long history of uncoordinated activities that work against one another. These have usually come about because each agency has independent responsibilities, and each is working hard to achieve its own goals without necessarily assessing the effects of its actions on other agencies or communities. Thus, the Army Corps of Engineers builds levees while the Fish and Wildlife Service restores riparian vegetation, and carefully restored channels are bulldozed into culverts a few months later when roads are upgraded. Basin assessment is intended, in part, simply to avoid the oversights caused when project planners don't consider the context of their projects. Basin assessments and watershed analyses are also intended to identify goals and values that are in conflict in an area and to identify those which cannot be reasonably met by the ecosystem that is present there.

The effort to enhance chinook populations in the Klamath-Trinity river system provides an example of this need. With considerable effort, the Departments of Commerce and Interior worked together to increase chinook escapement to the Trinity in 1993. Three thousand additional fish entered the system, and there was hope that this would be a turning point leading to enhanced stocks in perpetuity. Unfortunately, the Bureau of Reclamation decreased the flow while the eggs were in the gravel, desiccating and killing about 25% of the redds. If either side had had a document outlining the influences, agencies, and potential impacts associated with their actions, the appropriate communication could have been established before it was too late.

The river basin

The river basin was originally selected as a unit of analysis because of the importance of environmental impacts in large rivers and estuaries. In many cases, the overall aquatic health of the basin is strongly affected by conditions along the mainstem river, so impacts to fish stocks in individual watersheds cannot be understood without an appreciation for the influence of changes throughout the larger basin. Downstream changes, in turn, cannot be understood without assessing the influence of multiple tributary watersheds. The environmental impacts of most concern to the most people in many areas are downstream changes in flood frequency, water quality, and channel form, yet these cannot be evaluated at the scale of individual watershed analyses.

However, a river basin--or, for that matter, a watershed--is not an inherently relevant analysis unit for many types of issues. For example, wildlife ranges and vegetation types usually span drainage divides, and economic data are usually reported by political units, such as counties. The only functions which must be evaluated by watershed or river basin are those influenced by channels, but the drainage basin is absolutely essential for these analyses. In addition, the channel is the vehicle for the downstream impacts on wildlife, vegetation, and economic concerns, so a full analysis of wildlife, vegetation, and social impacts would require a watershed-based analysis anyway. Only by examining the full watershed or river basin can cumulative watershed effects be recognized and understood.

The fact that ranges do not correspond to watershed or river-basin boundaries is actually not much of a problem. Every issue or organism carries an implied scale at which analysis is most convenient, and these are different for each issue. Because much of any useful analysis will be devoted to examining the interactions between different organisms and processes, a uniform scale is required for the analysis. Whatever scale is chosen will be an inherently awkward one for all issues but one. However, information from each relevant scale can be superimposed easily on any arbitrary analysis unit. The hydrologically defined unit was selected as the arbitrary unit because it is definable on the ground, it does not change much through time, it would have to be evaluated anyway to understand at least some of the influences on all other components, and most other components are influenced to some extent by drainage divides. Economic communities, for example, tend to interact along transport routes, which ordinarily follow major river valleys.

What a basin assessment should include

Much of the confusion over basin assessment arises from uncertainty about how the information is to be used, and therefore, about what the assessment should include. The Record of Decision describes its role as a context for watershed analysis. The basin assessment is intended to identify and prioritize watersheds for analysis; identify the values and issues and describe their distribution through the river basin; and to describe the pattern of processes relevant to the river basin and interactions between ecosystem components (ROD p. B-22). The FEMAT report (1993) provides more information on assessment objectives. Specifically, the assessment will analyze beneficial uses and ecosystem values for large river basins or physiographic provinces and identify the processes that affect those uses and values. The goals that are outlined include (FEMAT 1993, Chapter V, Appendix I, p. 8):

1. Identify key resource issues and concerns and identify individuals and groups who can speak for those interests

2. Identify the context of the river basin with respect to other large basins (intra-basin/regional issues that cross drainage basin boundaries)

3. Identify ownership patterns, agency boundaries and areas of jurisdiction, wilderness, and other special management areas, historical land use patterns

4. Describe the physiographic provinces in which the basin lies and identify key physical processes and their spatial distribution at this coarse scale.

5. Identify overriding ecological issues and areas, for example Key Watersheds, ecological reserves, species distributions.

6. Prioritize watersheds for analysis.

7. Integrate results from individual watershed analyses and evaluate cumulative effects at the province and river basin scales

8. Provide a general description of physical and biological conditions within the river basin.

Further, "The results of this analysis will define a minimum set of issues and maps that will guide the more detailed individual watershed analyses" (FEMAT 1993, Chapter V, Appendix I, p. 8). The assessment is expected to encompass the full range of past impacts and the full range of likely future impacts. It will "consider time periods long enough to represent rare natural catastrophes such as major floods, fires, windstorms and droughts", and will also "consider the possible effect of potential, but unmeasurable concerns such as global climatic change" (FEMAT 1993, Chapter V, Appendix I, p. 10).

The basin assessments are thus intended to fill two distinct roles: to guide watershed analyses, and to integrate the results of the watershed analyses occurring within the river basin.

Guidance for watershed analyses would take several forms (Table 5-1). First, the perspective of the river basin as a whole is needed to prioritize the basin's watersheds for analysis. In this way, various large-scale issues can be weighed in the prioritization, along with individual agency interests and resources. If prioritization is left to the decision of the principal land-management agency in each watershed, then it will be based on single-agency interests, and it will be difficult to motivate other agencies to participate where they are needed.

A variety of problems are usually recognizable at a basin scale that would not be evident at the scale of a particular watershed, yet which require some information from watershed-scale analyses if they are to be addressed. For example, if mainstem habitat for Pacific salmon is poor along particular reaches, it might become particularly important to assess upstream inputs of sediment in consistent ways so that their role in contributing to mainstem channel changes might be compared. Similarly, if there is a particular water quality problem identified in an estuary, upstream watershed analysts would be put on alert to diagnose the sources of the problem. The basin assessment would thus point out types of information that must be addressed consistently to allow comparison among multiple watershed analyses, and it might also require particular analyses to contribute particular types of information. This information would then provide the basis for the second iteration of the basin assessment.

Table 5-1. Useful tasks for a basin assessment

First iteration (before watershed analyses are begun)

Identify basin-scale and mainstem issues and their distribution

Describe why the issues are issues, from a basin perspective

Identify potential causes and sources for the identified problem

Outline history of development through the basin

Describe the socio-economic setting of the basin

Describe which stocks and endangered species are likely to be in which watersheds

Describe basin-scale resource influences and limitations (e.g. dams)

Identify basin-wide trends

Identify the watersheds that are "special cases" for whatever reason

Prioritize watershed analyses

Identify watersheds likely to have the best economic and restoration opportunities

Identify information sources and catalog basin-scale data

Identify the agencies and administrative units that will need to work on each watershed analyses

Identify the watershed-scale information that will be needed to assess basin-level problems

Describe the types of information for which consistency among watershed analyses is important

Second iteration (after watershed analyses are completed)

Compare and evaluate the trends evident from the watershed analyses

Coordinate information exchange between different watershed analyses

Use watershed analysis results to describe patterns of disturbance throughout the basin

Evaluate issues, changes, and influences in the estuary and mainstem at the level of detail appropriate for watershed analysis

Identify the watersheds that may provide analogs for future conditions in others

At this stage, the basin assessment could also provide a preliminary characterization of what watersheds are most important for what problems or activities, and it will provide basin-level information that sets the context of each watershed within the river basin. Thus, prioritization for stream restoration for fisheries would be based on the overall benefits to the fish stock in question, rather than on the suitability of particular sites in particular streams on particular Forest Service Districts. Those watersheds with the greatest potential to make a difference would be identified at the basin scale with a consensus among all the relevant agencies, and watershed analyses within those watersheds would be tailored to address those opportunities. This guidance for watershed analyses is accomplished through identification of basin-scale issues and through qualitative evaluation of potential watershed-scale influences on those issues. Finally, the socio-economic context for future trends might best be addressed at a basin level, since these results will be applicable to multiple watersheds, and they reflect large-scale socio-economic influences. To fulfill these functions, of course, the basin assessment must be completed before watershed analyses are begun.

The basin assessment is also responsible for more detailed evaluations of issues, processes, and interactions that require a broader view than a watershed analysis can provide (Table 5-1). This type of assessment is likely to require results from at least some of the watershed analyses in a river basin before it can be carried out. These assessments might include evaluations of estuary sedimentation, mainstem flooding, factors affecting stocks at risk, and downstream water quality. At the preliminary stage, that of supporting watershed analysis, the basin assessment would identify such problems, decide what additional information is necessary, and specify that such information be included in the relevant watershed analyses. The evaluation of these problems would then be deferred until a later iteration of the basin assessment, when watershed analysis results are available.

During the second stage of the basin assessment, the assessment has the important role of integrating the watershed analysis results into a coherent view of conditions, issues, and influences operating in different parts of the basin. The basin assessment is in the position of being able to compare and contrast resource responses in different areas, and thus to alert the relevant watershed analysts of useful data sets or analyses that might shed light on particular issues of concern. This information would then be incorporated into later iterations of the watershed analysis.

The basin assessment will incorporate some input from even larger areas. Regional- and province-level goals and issues will be recognized and incorporated. In particular, economic and resource concerns often must be identified over very wide areas, and in some cases, internationally. The significance of regional population centers and transportation networks must be recognized for the analysis, as well as trans-basin water supply and power supply networks, if relevant. Agency interests and concerns are usually directed from a national level, and these must be accounted for in the analysis. Assessment will also identify regional and provincial data sets that may be relevant to the basin or to watersheds within it.

A few basin assessments are currently being worked on. Different assessment teams have taken different approaches, and this allows various approaches to be compared.

The Klamath Basin assessment

The Klamath River Basin occupies more than 30,000 km² in northern California and southern Oregon. The Klamath River drains the forested Cascade Mountains of central Oregon; the semi-arid volcanic plateau of the border region; the steep, unstable gorges of the Klamath Mountains; and the wet forests of the northern California Coast Ranges. The river cuts through four geologic provinces and through as many ecological ones; it drains parts of two states, eight national forests, and hundreds of towns. Potent issues range from water quality problems on agricultural land near Klamath Falls, to water rights on the Modoc Plateau, to Native American fishing rights, to a proposed dam in Oregon, to loss of jobs in the timber industry, to threatened and endangered species, to declining stocks of Pacific salmon, to flood risk. In one way or another, all of these issues are interrelated. And the basin assessment is supposed to address all of this.

The first challenge was to assemble an assessment team from among the 60 community, state, federal, and non-governmental agencies and groups with interests in the area. A core team of 16 groups and individuals was thus identified according to their level of interest and their willingness to devote time and effort to the assessment. This initial phase, and deciding the scope of the initial assessment, required several months of work.

The Klamath assessment is aided by interagency planning activities that had already begun in the basin. A long-term basin plan and a cooperative ecosystem restoration analysis are already under way, with the latter coordinated by an interagency ecosystem restoration office in Klamath Falls. This office, the Forest Service, and Humboldt State University are cooperating in the construction of a "seamless" GIS for the basin.

The assessment team decided to follow a three-part approach. The initial assessment will take less than a year, and is intended to support the watershed analysis efforts that will be undertaken in the basin. However, it is planned to consider only aquatic- and fishery-related issues. Public scoping sessions will be held at three sites throughout the basin to identify issues of concern to the public.

The second phase will be an assessment that incorporates the variety of issues not focused on aquatic biology. This is the actual basin assessment as visualized by FEMAT, and the assessment team expects it to require 2 to 3 years to complete. The final phase is expected to require 10 years, and is described by the assessment team as a basin-scale plan. This, of course, is a NEPA decision document, so it would not fall under the FEMAT expectations for basin assessment. Presumably the planning document will be linked closely with the basin-level planning efforts already being carried out.

The aquatics component of the assessment will identify the issues of concern and their locations. It will describe the types of influences likely to be acting on those issues, and will specify the tributaries of particular importance to fish stocks in the basin. Evidently, the decision has been made that aquatic biological issues are the primary focus of concern in the entire Klamath basin, since the information needed to direct watershed analyses that are not focused on fisheries issues will not be available until after the first round of watershed analyses has been completed.

The Mad River Basin assessment

The assessment for the Mad River basin has taken a very different approach than that for the Klamath. In part this is because the basin is much smaller, is less complex, and has fewer "key players", but the difference in approach also reflects the wider range of expertise and interests present among active members of the assessment team.

The Mad River drains approximately 1300 km² of the Coast Ranges in northwest California. The basin is about 150 km long and averages less than 15 km wide, so it has an awkward shape for assessing wildlife or vegetation communities. The basin supported an important fishery in the past, and it now is an important water and power supplier for north coast communities. Most of the basin is forest or rangeland, and potent issues include the loss of timber-related jobs, decreasing fish stocks, recreation, Native American cultural uses, grazing, water quality, flooding, gravel mining, and channel changes along the mainstem.

The Forest Service is the major public land holder in the basin, and has taken the lead in the assessment. Few other agencies are directly involved, though the BLM, Fish and Wildlife Service, and others are kept informed of the progress of the assessment through a weekly newsletter. The assessment team has identified lack of interest by other agencies and the public as a problem, and suggests that it will be important to quicken their interest by showing how the analysis product will be used and overtly identifying how it will aid the goals of other agencies, private land-owners, and other public interests. One potential selling point is that a coordinated GIS for the basin would benefit all participants.

Assessment is expected to require a year to complete, and will include identification and evaluation of the full range of issues in the basin. Currently most work is suspended because assessment team members are doing double duty as team members on the Pilot Creek watershed analysis. On-going, however, is the description of the socio-cultural history and development of the basin, which will provide the background for assessing biological and physical changes. Problems of interest to the public are being identified through public meetings and interviews. The team has identified an important product of the assessment to be the compilation of existing information for the basin, and this information has been incorporated into the ongoing watershed analysis as it is identified. Currently an ecosystem unit inventory is being carried out for National Forest lands in the basin, and although this is not part of the assessment, results may be useful in the future.

The assessment is expected to include an identification of the basin-scale issues and their distribution through the basin; the identification and distribution of important biological and physical processes in the basin; and directives to watershed analyses about what basin-scale issues will need attention at the watershed scale.

As practitioners gain experience with basin assessment, several problems are emerging. As with watershed analyses, it is difficult to integrate an assessment so that it provides an interdisciplinary view of the area, and it is equally difficult to incorporate the views of the various agencies that have interests in each river basin. There is an institutional lack of motivation for carrying out basin analyses since they have no independently earmarked funding, and since they provide no overt road-block to carrying out activities.

The problem of complexity also plagues basin assessment, just as it has plagued watershed analysis: analysts tend to make the process and the report far more complicated and detailed than it actually needs to be. Unfortunately, there is no authority delegated to answer questions definitively concerning what an adequate assessment should include, and no description of how their adequacy will eventually be judged. The Jackson decision concerning the applicability of the Federal Advisory Committee Act has also been cited as a roadblock to analysis, since much of the issue identification step must be carried out through public interviews and meetings.

The interdisciplinarity problem

A major goal of FEMAT is to have environmental problems be evaluated from an interdisciplinary view, and to ensure that land-use decisions are made by taking into account the full range of influences the decision might have. Thus, interdisciplinary analysis was identified as a major component of watershed analysis and of basin assessment. In effect, FEMAT recognized that no discipline is relevant by itself; appropriate problem solving is possible only when each individual discipline takes a subservient role to the combination of skills that the problem itself demands. This is one of the most important messages of FEMAT, and it is central to the whole concept of ecosystem management. It also appears to be one of the most difficult goals for either basin assessment or watershed analysis efforts to attain.

In the Klamath example, the interdisciplinarity problem was partially avoided by redefining the assessment to focus on a single issue, aquatic ecosystems. Unfortunately, this, itself, is a major manifestation of the problem. Aquatic biological issues will never be resolved without understanding their context and connections with water rights and grazing issues, and without understanding the likely future patterns of resource demands and recreational uses. The Mad River assessment is intended to include the full range of issues present, but is not far enough along to determine how the interdisciplinary linkages will be addressed. If a basin assessment does focus on a single issue, it is likely that the ensuing watershed analyses will need to be redone to incorporate the wider range of issues that become apparent later.

Most watershed analyses include evaluations of issues within several disciplines, because both wildlife and fisheries issues are usually widely recognized in the analysis areas. However, the early examples of completed analyses have generally been of multi-disciplinary rather than inter-disciplinary flavor. Although multiple disciplines are represented, the connections between them are not well described. The reports are presented with subheadings like "landslides", "fish", and "hydrology". It will take careful work to ensure that basin assessments do not follow the same pattern.

An approach that encourages interdisciplinary analysis is to present the information as a historical narrative, much like a Michener novel. Another approach may be to organize the report by public-defined issues, making sure to fully explore the possible influences of each facet of the biological, social, and physical system on the issue under discussion.

Interdisciplinarity might be encouraged if there were an example of the types of information required from each of the disciplines to carry out a typical analysis. Note that the following attempt to provide this information is itself broken down by disciplinary boundaries, and accept this as further confirmation of how difficult it is to achieve a truly interdisciplinary analysis.

The social component of either a watershed analysis or a basin assessment outlines the history of land use in the area and describes the social, economic, physical, and biological reasons for changing land-use patterns; it describes current land-use patterns, the values and expectations associated with them, and their dependence on social, economic, physical, and biological factors; and it predicts future land-use patterns and associated values and expectations on the basis of social, economic, biological, and physical trends. Many of these aspects depend on conditions far removed from the analysis area, so the social analysis often must take into account regional and international influences. Other important aspects of the analysis may be to identify constraints posed by agency culture, to obtain anecdotal descriptions of past resource conditions, and to identify who cares about what in the area.

It is clear from the above description that the social component of the analysis requires a considerable amount of information from the other analysis team members. To explain land-use changes, it is necessary to know how the resource base has changed and will change in the future; thus, physical and biological trends must be identified. Similarly, the effects of management activities on physical and biological conditions must be known. Assessment of expectations and explanation of land-use change must also take into account year-to-year variations in physical and biological conditions and the risk and effects of rare disturbances such as floods or fires. If conflicts between different land uses and values are to be understood and if future use patterns are to be predicted, the biological and physical constraints that control the distribution and abundance of each resource must be known, along with the interdependence of one resource or impact on others. Finally, the socio-economic specialist must learn from the other specialists what types of social and economic information would be of use for other facets of the analysis.

The biological component of either a watershed analysis or a basin assessment describes the past, the present, and predicts the future biological changes in the area as they are related to social, economic, physical, and biological factors. Many of these changes are influenced by occurrences that are far outside the analysis area. Once again, the need for information from other team members is quite evident. The social science specialist provides the historical land-use information needed to explain past biological changes, as well as the social and economic trends necessary for predicting future changes. Other changes and patterns can be explained only with the help of the physical science specialist, who can provide information on disturbance frequencies and the nature, timing, and duration of changes in physical habitat. Each of these specialists will also need information from the biological specialist to aid them in their own analyses, and these information needs must be identified.

The physical science component of either a watershed analysis or a basin assessment describes the past, the present, and predicts future physical changes as they are related to social, economic, physical, and biological factors. With the exception of climatic factors, physical influences are usually located within the analysis area. However, the physical setting can be strongly influenced by social, economic, and biological changes occurring outside the area.

In addition to the relations outlined above, the social science specialist contributes to an understanding of the physical component by providing information on past environmental conditions, as gleaned from interviews. Often, the socio-economic specialist can suggest types of information sources that the physical specialist may not be familiar with, such as photographic archives in local historical museums. The physical specialist must also work closely with the socio-economic specialist to identify the types of physical information needed for the socio-economic analysis. The biological specialist contributes to the physical analysis by describing recovery trends of vegetation after disturbance, evaluating fire patterns and history, and providing information on geomorphologically important animals such as beavers.

The punch line, of course, is that all of the analyses described above--the social, physical, and biological--are actually the same analysis. All components appear in all "analyses", and each description is merely a reorganization of the one before. It doesn't really matter which one of these organizations is selected for the analysis report, as long as all components are considered within that organization.

The teamwork problem

One contributor to the interdisciplinarity problem is that team members must be able to work closely together to make interdisciplinarity happen. Communication between different disciplinary specialists is difficult because of differences in approach, knowledge base, and vocabulary. These differences are made even more profound when different specialists come from different agencies, where cultures, goals, and commitment to the analysis process may be quite different. On-going interdisciplinary and interagency efforts in northern California provide a few lessons that may be useful in formulating effective basin assessment and watershed analysis teams.

Almost uniformly, the biggest block to effective teamwork is personality clashes within the team. In most cases, it is more useful to select people who are effective in working with others than those who may be preeminent in their fields but incapable of subjugating their egos to the needs of a team effort. This is particularly important in an interdisciplinary setting, where everyone's ignorance of everyone else's field must be seen as an opportunity to learn rather than a threat to one's omniscience. Each member has to be interested in the other fields represented and to be willing to make the effort both to learn the appropriate vocabularies and to speak in a language accessible to all. An analysis team of naturalists with different specialties may be far more effective than a team of specialists with different natures. It is particularly important that team members be energetic volunteers for the position, rather than unenthusiastic draftees.

An additional problem arises because team members often continue to shoulder the rest of their workload, and the time they are able to devote to the analysis may not coincide with the schedules of other team members. Analysis is aided if the team can be sequestered in a workplace apart from their customary work site for scheduled work periods. More of the right questions get asked if one is sitting next to the specialists capable of answering them. In any case, it is essential that team members stay informed of what each other are doing. Meetings do not achieve this purpose nearly as well as coffee breaks.

Some teams are at a disadvantage because they do not include all of the people necessary to accomplish the task. Either particular disciplines or representatives of particular agencies are absent. Surprisingly few watershed analysis teams, for example, have either a geomorphologist or a social science specialist as members, even though these have been identified as critical roles. In part this lack reflects the paucity of such specialties among agency employees, although the large budgets devoted to watershed analysis could have been used to recruit the needed specialists. However, the lack is also due to the lack of understanding of how such specialists could aid the analysis, and this lack of understanding also rises from their absence in the agencies. There is no one there to explain why they are needed.

FEMAT is explicit in its requirements that the relevant agencies participate in the analyses, yet it has been difficult to elicit appropriate levels of participation from many of the agencies. This problem will be considered in more detail in the following section.

Teams with ill-defined visions of the goal tend to flounder. Often the intended goal is replaced by the types of goals that individual team members are used to working toward. This seems to be the main source of the "analysis paralysis" currently bogging down various watershed analysis efforts. It is important that the team be guided by someone with a clear vision of the goal, the process, and the product, and that members be willing to forgo their preconceptions and biases and buy into the somewhat alien goal of the analysis. A bold leader is also needed to ensure that the needs of the assessment are placed above the partisan demands of individual agencies.

Problems also arise when different agencies have different agendas and the agency representatives feel more responsible to their agency than to the goal of the assigned task. In the past, this type of agency-centrism has led to the agency version of pork-barrel politics: projects are designed to meet agency needs and abilities rather than the needs of the resource. Once the agency has achieved its own goal, interest in the overall project disappears. This hurdle can be partly circumvented by establishing a dedicated office on "neutral" territory, where team members interact more with each other than with their agency cohorts. If the team answers directly to an interagency oversight committee rather than to their accustomed supervisors, results will be even more effective. In some cases, efforts may be hampered by agency territorialism, where an agency is unwilling to "lose control" of a data set or analysis function. In this case, too, insulation of team members from line officers may be helpful in derailing territorialism.

An effective team thus is one composed of enthusiastic individuals who enjoy working with one another, understand each other's fields and capabilities, and share a vision of the nature of their task. It takes time, experience, and dedication for such a team to become effective. Thus, once an effective team is assembled, it should be carefully fostered. The skill and efficiency created by such a team is likely to out-produce multiple unpracticed work groups both in the quantity and quality of the output. The most efficient approach to managing watershed analysis and basin analysis in an area may well be to form a single dedicated team that works full time on analyses, rather than to assemble a new team for each of the required analyses. The elite team could then call in special expertise where needed for particular analyses, or could coordinate small working groups to handle specific analysis tasks.

The interagency problem

Each agency was founded for a different purpose, has different jurisdictions, different mandates, different goals, different cultures, and is governed by different laws. At the same time, each shares a common motivation of carrying out the tasks that will ensure preservation of its budget, and individuals in each tend to make the decisions that will leave themselves blameless if something goes wrong. Although each agency has its share of idealism and idealists, these voices do not carry far when confronted by the reality of budgets and job security.

In addition, each agency is huge. Performance goals thus have become defined by things that are easy to count, rather than by less tangible successes that may actually mean a lot more. Fisheries specialists are judged by the number of instream structures constructed on their District (irrespective of the longevity or appropriateness of those structures), soils specialists by the number of acres inventoried, and budgets are distributed accordingly. This approach to "accountability" leads directly to agency conflicts, as where the Army Corps is judged by the miles of riprap constructed, while Fish and Wildlife Service is rewarded by miles of riparian zone restored. Taking time out to work on intangibles like watershed analysis and basin assessment is directly counter to agency goals--at least until accountability starts being judged by the number of watershed analyses completed. Even more insidious is the concern by some that analysis may show that some of the agency's pet programs are ineffective and should be abandoned.

To achieve the interagency participation demanded by the ROD, it thus becomes necessary for the lead agency in an analysis to be creative in motivating participation from other agencies. The most effective attraction would be a demonstration that the analysis product will decrease agency workloads in the future, or that it directly contributes to achieving agency goals in some other way. For example, much of the Fish and Wildlife Service workload is devoted to meeting Section 7 requirement for consultation under the Endangered Species Act. If basin assessment and watershed analysis could be shown to decrease the efforts needed to do this in the future, then the Fish and Wildlife Service would have an interest in seeing that the analyses are done rapidly, efficiently, and are of high quality. In particular, they would be motivated to participate early in the process to ensure that the final product will meet their needs as much as possible.

The key to this approach is to understand how each of the key agencies could benefit from an analysis, and to ensure that the analysis prototypes fulfill these roles. It will probably take a few examples of successful analyses to market the process effectively.

Marketing is also necessary within agencies. Mid-level managers in the Forest Service, for example, see analysis as a roadblock to doing the projects that are the measure of their own success. Their goal thus is simply to see the analyses out of the way as soon as possible so that "work" may resume. If the analysis products could be shown to contribute to the selection of appropriate projects, and thus make project design and approval easier, then the managers would become more committed to providing the support needed for a quality product. Much of this support would simply be the message that the agency believes in the process at all levels. In short, there's a difference between compliance and buy-in. Currently, agencies are participating in analysis because they are directed to by the administration. Successful analyses will begin to happen when the agencies participate because of the value of the analysis product.

Meanwhile, we still have the problem that some of the necessary agencies have neither the time nor the inclination to participate in analyses. This is particularly true of the chronically understaffed regulatory agencies. Their goals are not impeded by a lack of an analysis, as are those of the land management agencies; and the regulatory agencies are in the strong position of participating on the Regional Interagency Executive Committee, which reviews proposed actions to ensure that they are consistent with the goals of the ROD (ROD p. 35, par. 4). It is thus very much in the interest of the BLM and Forest Service, for which the analyses are a very potent hoop, to encourage buy-in from the other agencies early in the analysis process. Simply sending a letter of invitation or keeping people informed of meetings by E-mail messages is not adequate, particularly since the lack of response is likely to be greeted by a sigh of relief from management agencies which have found the other agencies difficult to work with in the past. Participation must be actively cultivated through telephone calls and informal briefings. On the other hand, establishment of a seamless network for interagency electronic communications will certainly aid cooperative efforts.

The basin assessment may serve a useful role in facilitating interagency cooperation. This assessment could provide the forum for interagency participation from agencies that are too overworked, under-budgeted, and under-staffed to take on major roles in watershed analyses. At this level, the goals of particular watershed analyses would be mapped out, and the agencies could then choose to participate in the ones most crucial to their own interests. They would have had input and buy-in to the full range of analyses, and would thus be aware of the particular issues they are interested in. All agencies would then play a part in prioritizing and scheduling analysis to optimize their own involvement, and in ensuring that the full range of watershed analyses in the basin will suit their own needs as much as possible. Resource people within the agencies could be identified as technical advisers on particular topics. This type of participation would be made easier if each agency had a directive from its central office that participation was not only permissible but expected, and that the agency will support the efforts and findings of the assessment and analysis teams. First, however, the agency hierarchies must themselves embrace the concept of analysis.

The complexity problem

The scale that must be considered during the evaluation and the level of detail required both depend on the issue being evaluated and on the type of analysis. The basin assessment will address all of the topics outlined in the previous section, but will do so qualitatively. The major issues will be identified, along with the likely causes of the problems. These causes might be associated with particular watersheds or types of sites (e.g. mainstem turbidity caused by active earthflows on Franciscan bedrock in the Peterson, Macon, and East Fork watersheds; or the endangered scarlet mousewort found near springs in the mid-elevation forests of the Beartrap Mountains), but specific sites will not need to be identified. A GIS may be useful for disclosing patterns, but it is certainly not necessary for a basin assessment, and may actually impede progress by making non-useful levels of complexity possible. The overall objective of the assessment is to describe patterns and make generalizations, and thus to avoid the myopic disciplinocentrism generated by an unthinking reverence for detail.

The most important function of the first iteration of a basin assessment is to identify the full range of basin-level issues that will need to be addressed by watershed analyses, and to identify the types of information needed to address them. On a moderate-sized river basin (e.g. 25,000 km²), issue identification is expected to take up to a month and will be carried out primarily through interviews and public meetings. Once issues are identified, the information needs are usually quite evident. A second important function is to catalog the types of information available for the basin. This can be carried out at the same time as issue identification, and will undoubtedly continue even after the first assessment iteration is complete. Initial expectations were that a basin assessment would require 3 to 6 weeks to complete, and this estimate is borne out by reported durations required for issue assessment from ongoing assessments. "Analysis paralysis" begins when assessment teams go beyond the simplistic goals of the assessment and begin to incorporate inappropriate levels of detail and scales of analysis.

The second iteration of a basin assessment will not be carried out until results are available from multiple watershed analyses within the basin. This iteration is likely to take slightly longer than the first. The focus at this time will be to use watershed-scale information to better understand changes occurring along the mainstem, and to integrate results of the analyses into a better description of the overall patterns of change and its causes through the river basin.

Although the approach described above for basin assessment is basically the same as that for watershed analyses, watershed analyses can include more detailed information. But even for watershed analyses, inventory data are not necessary. Instead, general habitat requirements and the types of sites at which the habitat is found might be described. Particular tributary watersheds might be identified as having more or less of the habitat type, but site-level information is unnecessary. Even at this scale, use of a GIS can be more distracting than useful.

Each issue will define the scale at which it will need to be evaluated, so this will vary for each issue. For example, an important cause for the increased interest in mushroom harvesting is to be found in Japanese markets, while the fate of an endemic pupfish population may depend on the feeding habits of a particular herd of cattle during a drought.

The quality control problem

The ROD and FEMAT reiterate that watershed analyses are to be scientifically defensible, but do not explain how the scientific defensibility will be certified. The assumption is that analysis documents will be circulated for scientific review before they are released, and that the review comments and suggestions will be acted upon. Management agencies have begun to lose some credibility in scientific and academic circles because there is a pattern of informed criticism being ignored. Apparently, the solicitation of comments has become the goal, rather than the improvement of the documents critiqued. Conscientious review of documents requires considerable time and effort, and this must be recognized by those soliciting the reviews.

A second quality control issue is how to ensure that the documents will actually be useful, and thus to avoid their taking on the role of the ritualistic NEPA documents that are rarely read. In the case of the first iteration of basin assessments, this might be accomplished by having the existing watershed analysis teams specify the types of information they would benefit from if it were included in a basin assessment. Similarly, the agencies involved could also specify the types of information that would be of most use to them.

Watershed analyses have a broader intended scope of application than basin assessments, and their role is more carefully outlined in the ROD and FEMAT report. In this case, the utility of the product is judged by the extent to which it facilitates later project planning and approval. The intent is that future planning will be guided by the analysis, whereas in the past, analysis was triggered by specific project plans. This constitutes a major shift in approach, and it will take a lot of trial and error before the new way of doing business becomes the natural way of doing business.

The bottom line is that if the administration is serious about its commitment to changing the way federal lands are managed, the agencies' central offices will support the efforts, and the best products will be rewarded. This will provide motivation both to analysis teams and to their agency supervisors to do a good job on the analysis, rather than simply to do a rapid job. Once examples of "good" assessments and analyses are produced, they are likely to become the yardsticks against which others are measured. The interagency committees that are established at provincial and regional levels can perform a quality-control role by comparing various analyses and making sure that the best examples are circulated as models.

The public involvement problem

The FEMAT report and the ROD stress the need for soliciting public involvement, yet as soon as these documents were circulated, they provoked lawsuits contending that public involvement was misused in their preparation. The result has been a sudden fear of soliciting any information from anyone who is not a federal employee. This is unfortunate, since this fear is overlaid on even more deep-seated inclinations to avoid public involvement. Many agencies have come to see the public as uninformed critics intent on vetoing all the agencies' work for no rational reason. At the same time, much of the public has come to view the agencies as juggernauts that will proceed according to their own plans, irrespective of what the public desires.

Public involvement is required at several levels during assessments and analyses. First, the public knows what it cares about, and the agencies do not. Considerable effort must be put into identifying who cares about what in a river basin or watershed. The usual vehicle for NEPA scoping of issues is the public meeting. Participants are self-selected, and each is given a specified time to express their views. The process is legal under FACA and is required by NEPA. Unfortunately, follow-up interviews of participants and non-participants (Reid, in review) suggest that the approach is not adequate for meeting the needs of basin assessments and watershed analyses for several reasons:

1. Self-selection means that the most well-organized groups are best represented, while powerful small groups and diffuse majority opinions may not be represented at all. The result is that some unidentified issues appear at a later, more awkward stage to derail the planning process.

2. So many public scoping sessions are held that it becomes a tremendous time-drain on the few people capable of articulating many of the issues. This also selects for paid representatives of large, well-organized groups, who can share the load within the group and be recompensed for their time.

3. Participants feel undervalued. Each must wait through several hours of other people's thoughts to have the privilege of expressing their own position. Many have the perception that the meetings are designed to avoid wasting the paid bureaucrats' time at the expense of wasting that of the unpaid participants'. A 2-hour meeting might require 6 hours of agency personnel's time and 200 hours of the public's, while each of the participants may be allowed only 2 minutes of input. This is not good public relations.

4. Some of the most well-informed public, such as resource professionals, do not attend such meetings because they fear that participation will be taken as evidence that they are not objective in their work.

5. Some of those who would be able to provide valuable input simply do not hear of the meetings.

6. The public meetings do not foster dialog that allows clarification and expansion of the points being made; rarely are they interactive enough to develop a full understanding of the issues raised.

7. The public meetings are inherently confrontational because they focus on projects, decisions, or land allocations that have already been planned. The "anti's" thus have the most motivation to attend, and this taints both the agencies' and the public's image of public involvement.

For FEMAT purposes, issue identification is proactive. No projects are yet planned, so there is nothing to provoke strong feelings for or against. Even though the identified issues will be used to aid in selecting appropriate projects, public expectations of the efficacy of their input are jaded. This means that there is little motivation for the overworked public to gratuitously attend--and waste their time at--the typical scoping meetings. In addition, the type of information that the assessment team needs about issues is far more detailed than that which can be presented in the scoping session "sound bites". The information required can only be obtained by interactive interviews. This approach requires considerably more effort from agency personnel who have had training in interview techniques. It pays off, however, both in the quality of information obtained and in the message conveyed to the respondent that their opinions are worth the agencies' time. However, care must be taken to follow the guidelines of the Federal Advisory Committee Act (FACA): input can be solicited from individuals, but not selectively from functional representatives of organized groups.

A second function of public involvement for FEMAT is to forge a basis for community participation and buy-in. Analysis documents provide a mechanism for educating the public about the full range of concerns that go into making land-use decisions on public lands, while the analysis process itself provides a mechanism for educating the management agencies about the full range of concerns the public embraces. As the public and the agencies come to better understand each other's quirks and constraints, land-use decisions will become better informed and less confrontational. This function will be unattainable if analysis reports adopt the unreadable style and gratuitous volume of NEPA documents, and if issue identification is carried out without extensive public participation.

Watershed analysis and basin assessment also require considerable input from the public both about the history of changes in the analysis area and about resource conditions in the past. This type of information is available only through interviews.

Analysis also benefits from cooperation with land owners in the analysis area, since the analysis encompasses both public and private lands in the designated area. Encouraging such cooperation poses the same challenges as soliciting interagency cooperation: the public needs to see advantages for their participation before they are likely to commit the time and effort to helping. Several selling points might be that the analysis will greatly simplify analysis of cumulative watershed effects on private lands; that access to private lands and privately held information will make it unnecessary to assume worst-case conditions on the private lands, thus permitting more options for future management; and that information gained during the analysis on public lands will be useful to private land owners for future planning. In addition, participation in broad-scale ecosystem management efforts such as FEMAT may make the imposition of legislation regulating private lands less likely. In other words, if timber companies can show that they are participating in basin-wide evaluation, restoration, and monitoring activities, additional regulatory legislation may be perceived to be unnecessary.

FEMAT and the ROD have introduced several fundamental changes to federal land management in the Pacific Northwest. Among these are that long-term planning and future projects are to be based on the results of basin assessment and watershed analysis, rather than having analysis be triggered by specific projects or planning needs, as in the past; that interdisciplinary analysis is essential for understanding the implications of management decisions; that interagency cooperation is essential for an efficient approach to land management; and that the public is to be cultivated as a partner rather than an adversary. Each of these changes is reflected in the requirements for basin assessment and watershed analysis.

Unfortunately, agencies have tended to jump into the analyses without fully embracing the changes. As a result, many analyses are being carried out by single agencies working in isolation; many are organized along strict, disciplinary lines; many are designed with little public input, or with input gleaned in relatively ineffective ways. In addition, many fail to see the utility of analyses, since they do not fit the outmoded form of providing direct support to project planning; people have overlooked the newly defined role of analysis as the conceptual framework that allows us to understand the system and its interactions well enough that we are capable of designing appropriate land-use plans.

In essence, the process of carrying out the analysis has become the most valuable analysis product. The requirement for analysis has set up a framework for learning more about the systems that the agencies manage. The ROD specifies that analysis will never be complete, that it is an iterative process that will continue to guide the acquisition of new information and will continue to incorporate the information as it becomes available. Understanding the system has thus become a way of doing business, rather than being a discrete goal that can ever be attained. This change has effectively created a new item for attainment reports and job descriptions of agency personnel. No longer can success be judged simply by counting the works done; learning is now part of the job. Why should resource professionals take time away from "work" to attend seminars, participate in journal clubs, spend time in university libraries, or simply talk to others about advances in relevant fields? Because this is the only way that they will keep honed the skills needed to perform their new jobs: learning how the systems they work with work.

FEMAT. 1993. Forest ecosystem management: an ecological, economic, and social assessment. Forest Ecosystem Management Assessment Team.

Regional Ecosystem Office, 1995, Ecosystem analysis at the watershed scale, version 2.2. Regional Ecosystem Office, Portland, OR. US Government Printing Office: 1995 - 689-120/21215 Region 10. 26 p.

Reid, L.M. In review. Evaluating timber management effects on beneficial water uses in Northwest California. Report prepared for California Department of Forestry and Fire Protection.

United States Department of Agriculture and United States Department of the Interior. 1994. Record of Decision for amendments to Forest Service and Bureau of Land Management planning documents within the range of the northern spotted owl; Standards and Guidelines for management of habitat for late-successional and old-growth forest related species within the range of the northern spotted owl.