North Central Washington

Risky Business: Invasive species management on National Forests - A review and summary of needed changes in current plans, policies and programs

Chapter 1. Policies: Policies and plans must be linked to performance measures

Report Published by Kettle Range Conservation Group (www.kettlerange.org/weeds/)
February, 2001
Return to the Main Document Table of Contents

This section provides a discussion of national and regional policies, plans and programs that deal with invasive species. Most of these policies originate at the national level and serve to guide programs through all levels of federal agencies. Federal policies and plans derive from a number of Acts, agencies, and authorities pertaining to invasive species (Invasive Species Council, 2000, Appendix 2). The combination results in a hodge-podge of policies, some of which are only incidentally concerned with invasive species. An urgent need for change in Forest Service policies was stated by Jim Wells of the General Accounting Office (GAO) (2000), during hearings before the House of Representatives on June 29, 2000:

Comparing the actions that the Forest Service has taken to improve its financial management and reporting with its lack of progress in becoming more accountable for its performance illustrates the low priority that the agency has assigned to providing the Congress and the public with a better understanding of its performance ...

To provide the Congress and the public with a better understanding of what it accomplishes with appropriated funds, the Forest Service will need to make performance accountability a priority within the agency.

Section A. Policies and plans need clearly defined goals and objectives linked to performance measures.

A large number of international treaties and statutes guide policy for the control of invasive species. Beginning with the Lacey Act in 1900, these statutes were enacted primarily to address specific problem areas, rather than affording a generalized approach to controlling invasions. This system of policies has created a polyglot of different approaches to invasive species problems, gaps in legal authority and potential contradictions in control measures, resulting in programs that are inefficient and ineffective at slowing plant invasions in the United States.

Under Executive Order 13112, February 3, 1999, the importance of goals and objectives in directing invasive species policies was strongly stated:

Within 18 months after issuance of this order, the Council shall prepare and issue the first edition of a National Invasive Species Management Plan (Management Plan), which shall detail and recommend performance-oriented goals and objectives and specific measures of success for Federal agency efforts concerning invasive species.

Current land management policies, which operate under land protection plans, are subject to many constraints, such as funding. Plans are meant to follow a consistent approach and be available for public review. However, accountability is often skipped. The National Park Service (NPS), the Bureau Of Land Management (BLM) and U.S. Forest Service (Forest Service) are the largest federal land management agencies with responsibilities for protecting public lands from plant invasions. These agencies operate under different sets of regulations and performance standards. With respect to the Forest Service, the National Forest System Land and Resource Management Planning, 47 FR 43037, Sept. 30, 1982, defines goal (in the terms of project goals) as:

A concise statement that describes a desired condition to be achieved sometime in the future. It is normally expressed in broad, general terms and is timeless in that it has no specific date by which it is to be completed. Goal statements form the principal basis from which objectives are developed.

When a project has the potential to cause environmental or public harm, such as the spread of invasive species or use of herbicides, procedural requirements help to insure that documents and the Agencies themselves take this into account. Such projects are typically reviewed through a planning document that discloses the impacts and specifies mitigation measures, subject to judicial recourse if public interests are not met.

For instance, procedural requirements are specified in the National Environmental Policy Act (NEPA), which establishes policy, sets goals (Section 101), and provides means (Section 102) for carrying out policies. Action-forcing provisions insure that federal agencies comply with the procedures and achieve the goals of the Act and achieve the substantive requirements of Section 101. NEPA procedures require that environmental information is available to public officials and citizens before decisions are made and before actions are taken. NEPA directs that federal agencies shall to the fullest extent possible,

Implement procedures to make the NEPA process more useful to decision-makers and the public . . .

Planning on National Forests is guided by the National Forest System Land and Resource Management Planning (LRMP) regulations (Forest Service, 1999b), which in the latest set of proposed revisions state,

The set of documents that comprise the land and resource management plan must clearly display the goals, objectives, standards, guidelines, and other decisions made at different geographic and temporal scales that apply to the plan area. . .

. . . goals, objectives, standards, or guidelines in special area plans be incorporated into the land and resource management plans as plan decisions.

In essence, there is to be a clear set of directions established, which drive the management decisions of the agencies. Disclosure requirements, such as those in NEPA, provide a means of assessing whether decisions conform with policy direction at all levels.

Despite such readily available guidance provided by NEPA and the National Forest Management Act (NMFA), there is no specific intent to guide invasive species policies. Consequently, projects on public lands lack the specific and consistent guidance and authority necessary for effective programs. The GAO (1997b) answered Congressional requests for information on the underlying causes of inefficiency and ineffectiveness in the decision-making process used by the Forest Service in carrying out its mission by stating that, “agreement does not exist on the agency’s long-term strategic goals”. Essentially, there is a lack of consistency within the agency regarding its own mission surrounding public land management. The same problem also exists with regard to its long-term strategic goals for invasive species management.

This lack of consistency has been recognized regionally in the Proposed Decision for the Interior Columbia Basin Ecosystem Management Project (ICBEMP, 2000), which states,

Uncoordinated efforts throughout the project area have been ineffective against noxious weeds. Noxious weed strategy (ies) need to be consistently implemented project-area wide to reduce the negative impacts of noxious weeds. This objective (B-O11) hinges on a project-area-wide integrated weed management strategy being developed….

Despite clear articulation of the problem, the ICBEMP document falls short of providing specific guidance for the development of the IWM strategy that it calls for.

This failure to provide a coordinated, consistent strategy for managing invasive species permeates all levels of the Forest Service (FS) and Bureau of Land Management (BLM), although the National Park Service has made notable attempts to develop innovative plans. Planning documents lack goals specific to invasive species that would guide the choice of actions toward successful outcomes. Lacking policy direction, program goals are more often described in terms of resource outputs with mitigation prescribed as an after-thought for the control of invasive species. Without clearly defined goals, performance measures, which would be linked to proposed actions, cannot be determined. Performance measures are important because they account for thresholds beyond which proposed actions are unacceptable and must be changed or restricted.

For example, the goals for controlling new invaders might be eradication; for established invaders in Wilderness areas it might be prevention; and for large infestations of established invaders it might be containment, quarantine, or tolerance of a quantity determined by a damage threshold. Each situation should have objectives, methods, and performance measures specific to the goals for the lands involved. Such goals and objectives should relate back to the agency’s long-term strategy for stewardship of public lands, and should be disclosed as part of NEPA documentation.

However, existing Forest Plans were only written with loosely defined goals couched in general descriptions of “desired future conditions”. These do not provide quantifiable criteria for the presence or management of invasive species. The end result is that Forest Service implementation of programs to control invasive species has resulted in the indiscriminate spraying of herbicides across large landscapes without any measures of program success or fiscal accountability. To date, performance is measured only by acres treated, with no accounting of the effectiveness of the measures or impacts of treatments on other resources.

Invasive species management continues to lag behind other areas of National Forest management in measured performance of programs and project effectiveness. Part of the blame lies with Forest Plans and projects that were written without clear goals, or with ill-defined goals, e.g., “the ‘implied’ future condition is to have an absence of any new invader” (Okanogan NF, 1999, p.5).

Unfortunately, the primary goals of Forest Service land management plans operate under the oftentimes conflicting principles of multiple use and sustained yield, which requires managers to provide high levels of six “renewable” resources. The fact that project implementation under the multiple use doctrine guiding Forest Service actions may conflict with regulations under the NFMA or NEPA has been successfully applied in the courts to halt unsound projects. However, ultimately it will be up to the Forest Service to improve its public accountability.

Almost 3 years later, the Chief of the Forest Service observed that the change in culture had not occurred. In his February 16, 2000, testimony he stated that to restore the agency’s credibility with the Congress and the American people, the Forest Service must change its culture, recognizing that it cannot be an effective resource manager if it is not first accountable for taxpayer money and for its own actions on the landscape (GAO, 2000).

To date, there is no single set of goals and objectives which guide Forest Service management of invasive species. In the Forest Service, goals seem to be an afterthought to funding. According to the General Accounting Office (2000):

To become more accountable for its performance, the Forest Service will need to link its budget and organizational structures as well as its budget allocation criteria, forest plans, and performance measures to its strategic goals, objectives, and strategies. However, the agency is still years away from completing these linkages.

In response to Congressional prodding through the Government Performance and Results Act of 1993 (referred to as the Results Act), the Forest Service Strategic Plan was revised beginning in fiscal year 2000 to address the problems with invasive species. This new plan states some admirable goals and objectives for invasive species, albeit somewhat grudgingly couched in the language of forest health.

Increase the amount of forests and rangelands restored or maintained to a healthy condition with reduced risk and damage from fires, insects and diseases, and invasive species . . . Prevent the spread of invasive species . . . Measure trends in acres at extreme risk from fire, insects, diseases, and invasive species . . . Implement an invasive species detection and monitoring program.

Goals and objectives reflect the political concerns of the time they were enacted. Furthermore, the Forest Service expertise is largely vested in timber management and fire control. Existing invasive species programs in the Forest Service are piggybacked onto existing programs such as timber or range management, and as a consequence they lack a sound decision-making framework. These ad hoc programs skew the legitimacy of goals for invasive species management and sever the links between national policies and performance measures. Goals lack accountable budget targets, rendering some decisions irrational. Successful management of invasive species is dependent on rational approaches, long-term planning and commitment such as that embodied in Integrated Weed Management (IWM), e.g., Hoglund (1991),

IWM is a decision-making process, which selects, integrates and implements weed control based on predicted ecological, sociological and economic consequences.

This National Park Service Strategic Plan for 2000-2005 covers all of the National Park Service lands, reflecting its mission to preserve resources and serve the public. All goals are shaped by this mission statement. The Park Service (August, 1996) Strategic Plan for Managing Invasive Nonnative Plants, includes strategies for invasive species and also identifies targets within that strategy,

Strategy: Prevent invasion. Target: Modify National Park Service policy and guidelines to include nonnative plant management issues, as needed.

Strategy: Manage invasive non-native plants. Target: Reduce populations of invasive nonnative plants through an integrated pest management program that incorporates chemical, biological, cultural, and physical (mechanical) operations.

It is important to note that policy objectives should be formulated in terms of control, not blind treatment. The National Park Service states its service-wide goals as measurable outcomes (results), embedding the performance measure into each long-term goal and stating its annual goals in the same way, to show clear and direct relationships between long-term goals and annual goals. Annual goals are simply one-year increments of the long-term goals. For example, the long-term goal for Exotic Species states that by September 30, 2005, exotic (nonnative) vegetation on 6.5% of target acres of parkland is contained (167,000 of 2,656,000 acres). The annual goal for 2001 parallels that long-term goal: By September 30, 2001, exotic vegetation on 1.3% of targeted parkland is contained (33,000 of 2,656,000 acres). The Park Service bases its goals and targets on the appropriations that can reasonably be expected. Goals are directly related to budget requests on a goal-by-goal basis.

Case example: Boulder Creek on the Okanogan NF in Washington

Federal land management agencies have recently prepared a large number of weed control projects, e.g., the Okanogan NF Integrated Weed Management Environmental Assessment (EA) (1997, 1999). These documents do not originate from a set of overall planning goals for managing invasive species. Instead, they are ad hoc measures appended to existing range, timber or other programs (Forest Service, 1988; Mediated Agreement, 1989; Forest Service, 1992) and applied piecemeal by local authorities.

In lieu of appropriate project goals for invasive species management, the Okanogan NF EA cited the desired future condition and historic range of variability given in the Forest Plans, which are otherwise lacking in the mention of invasive species. Nonetheless the EA claimed that the desired future condition had contained within it an “implied” goal which was an, “absence of any new invader noxious weed species” (p. 5). On this basis, the EA then proceeded to build a case for selecting an alternative (p. 6).

. . . the desired future condition is vegetation with structural diversity . . . wildlife needs would be met by maintaining vegetation within the historic range of variability. . . .

The alternatives proposed different control “strategies”, which in reality were merely tactics. The differences between the alternatives did not involve a choice between a variety of treatments, but merely whether or not to use herbicides. In actuality, no substantive non-chemical measures were described or undertaken, except for some weed pulling by convict crews (a measure not described in the original document).The document contains considerable detail aimed at demonstrating that all measures except chemical treatments would be ineffective or harmful. Offhand and anecdotal statements were frequently given without citation, e.g., non-chemical methods would allow toxic species to persist; and public health risks from chemical treatments would be remote and limited to dermal contact. The EA presented a biased analysis in which the two rejected alternatives without chemical treatments were made to appear frivolous and harmful to the environment, while the chosen alternative was contrasted favorably by overstating the expected outcome and minimizing or omitting any mention of associated untoward effects (p. 107):

This [chemical] alternative would improve water quality by controlling noxious weed populations contributing to sedimentation . . . This alternative would improve fish habitat by improving the sediment indicator . . .

The treatment involved spraying non-selective herbicides along 34 roadside areas, each covering many miles. Most of the vegetation along these roads was predominantly native or beneficial non-native vegetation and many miles of roadsides with no weeds at all were also treated. The result was that many miles of roads were unnecessarily denuded or defoliated.The projected benefits of the control program were not borne out by actual outcomes; e.g., the herbicides actually damaged beneficial native vegetation and denuded roadsides and streambanks, causing increased erosion. But, from the flawed analysis given in the EA, the chemically oriented alternative was made to appear to be the safest and most effective. Despite public sentiment against the project, a large Congressional appropriation given without constraints to this Forest appears to have been the hidden agenda that guided approval of the flawed EA. In examination of this and similar Forest Service projects, an obvious flaw is the failure to clearly state goals and objectives required prior to making a decision. Had destruction of miles of native vegetation been a project goal, then the project could have been deemed successful.

It is true that some weed infestations were temporarily reduced in this operation, but non-target vegetation was reduced even more; spray trajectories left resistant weeds unharmed while reaching far beyond the roadsides into native communities. Long-term goals did not indicate whether re-infestation from soil seed banks and migration (Lajeunesse, 1997) would be within the range of effective control given the extremely large amount of areas treated and high budget costs of the 1997 program ($300,000, which did not cover monitoring expenses). Thus, there is little assurance that the so-called goal,“absence of any new invader noxious weed species”, is realistic.

Solutions

Plans need a sound, consistent framework formaking decisions.

Efforts to successfully control plant invasions do not always follow a consistent or rational approach; these efforts are most likely to fail (Hobbs and Humphries, 1994).

Testifying before the U.S. House of Representatives on the Forest Service’s approach to making decisions, Barry T. Hill, Associate Director, Energy, Resources, and Science Issues, Resources, Community, and Economic Development Division, GAO (1997) stated:

Our report on the Forest Service’s decision-making identifies an organizational culture of indifference toward accountability. The agency’s historically decentralized management and recently increased flexibility in fiscal decision-making have not been accompanied by sufficient accountability for expenditures and performance. As a result, inefficiency and waste have cost taxpayers hundreds of millions of dollars, and opportunities for both ecological and economic gains have been lost through indecision and delay. Past efforts by the Forest Service to change its behavior have not been successful. Decision-making within the agency is broken and in need of repair.

The GAO also recommended specific recommendations for improving Forest Service decision-making (GAO, 1997b):

GAO recommends that the Chair of the Council on Environmental Quality change the Council’s regulations for implementing the National Environmental Policy Act to require, rather than merely allow, federal agencies to tier plans and projects to broader-scoped studies.

It is useful to use National Park Service planning efforts as a yardstick for agency progress in developing plans for invasive species control. Hiebert and Stubbendieck (1993) explained the need for an analytical decision-making approach to invasive species management, which was aimed primarily at the National Park Service, but which applies equally well to other land management agencies:

Several sound reasons exist for using an analytical approach as the basis of prioritizing exotic species. One of the basic reasons for using a decision analysis process is to get scientists involved in the decision-making process ... . . . If an analytical approach was not employed, decisions would most likely be based on the opinion of an individual or a group of individuals or decisions would be based on precedent. Granted, many field ecologists have a good idea of which exotic species are impacting natural ecosystem processes or impacting species composition. However, decisions based on judgment alone are rarely based on defined criteria, do not usually document the reasoning processes, and give no assurance that the full array of significant factors was considered. Such decisions may suffer from personal biases and political whims. Decisions are hard to defend if challenged, and proposals for funding are hard to justify. Decisions based on precedent may be easier to defend but are not responsive to the variation in exotic species or natural system interactions over space and time. Thus, priorities set for managing exotic species based on precedent may not reflect current ecological and economic realities.

First and foremost should be a recognition that invasive species are as much a symptom, as they are a cause of poor forest health. The failure to address invasions from such a holistic viewpoint contributes to declining ecosystem integrity by perpetuating the very conditions that lead to invasions in the first place, e.g., ground disturbances, or spreading seeds through the consequences of management.

The failure to address the underlying causes of plant invasions results from the reluctance of the Forest Service to approach problems in a rational, analytical manner which links goals to expected outcomes. In many plant invasions, effective, inexpensive control measures have been ignored (Mack et al., 2000), resulting in invasions which cost thousands of times more to control (Pimentel, 1999), even to the point of irretrievable loss of resources (Turner et al., 1994; Van Wilgen and Richardson, 1985; Clarke et al., 1984). Unless the ecological causes of the plant invasions are addressed and understood within a framework for making decisions, weed control projects are doomed to fail.

Faced with public opposition and lacking sound guidance, project planners may cobble inappropriate measures from other programs with conflicting goals. One promising approach to invasive species management that, unfortunately, has been widely misinterpreted and misapplied, is Integrated Weed Management (IWM). IWM can afford managers with a wide variety of options (Wooten, 1999), but as applied by the Forest Service and other government agencies, it is more often used as a thinly veiled rationale for the overzealous use of herbicides.

The emerging paradigm of ecosystem management is beginning to take hold among land management agencies (Noss, 1999; Wooten, 1999; Appendix A). If land management agencies truly want to stem the tide of invasive species, then management programs going to have to look at the big picture afforded by ecosystem management. A recent paper by the Ecological Society of America suggests the importance of considering ecosystems (Mack et al., 2000).

Control of biotic invasions is most effective when it employs a long-term, ecosystem-wide strategy rather than a tactical approach focused on battling individual invaders. Prevention of invasions is much less costly than post-entry control.

Invasive species plans and policies could benefit greatly by using principles of ecosystem management. For instance, National Park Service managers are directed to manage not only for individual species, but to maintain all the components and processes of naturally evolving park ecosystems (Hiebert and Stubbendieck, 1993). The requirements of ecosystem management vary. However, there are a number of consistent factors, which can be noted. Ecosystem management has pitfalls if attention to time and distance scales is not accounted for. Mistakes carry the burden of an entropy cost (see Appendix A).

Ecosystem management considers the following components (Noss, 1999; Appendix A):

1) Long-term sustainability.
2) Clear, operational goals.
3) Sound ecological models and understanding.
4) Understanding of complexity and interconnectedness.
5) Recognition of the dynamic character of ecosystems.
6) Attention to context and scale.
7) Consideration of humans as ecosystem components.
8) Adaptability and accountability.

Humanity can no longer pretend that land management can exist without considering ecosystem processes, many of which are of inestimable value to our way of life. According to Daily et al. (1999), “Based on available scientific evidence, we are certain that ecosystem services are essential to civilization.” Policies and plans should rely on objectives to control invasive species based on a rational decision-making framework, which can reasonably attain stated goals without significant negative impacts. Such a framework could be based on principles of true Integrated Pest Management, and its plant counterpart, Integrated Weed Management, or IWM (Wooten, 1999b; Appendix A):

True IPM is an interdisciplinary system of techniques for controlling invasive plants that is both practical and environmentally sensitive.

Components of a true IPM program should include ( Appendix A):

A wide variety of pest control options is considered in true IPM with preference for:

When combined with ecosystem management, IWM can be used as a powerful tool to accomplish the goals and objectives of invasive species management (Wooten, 1999; Appendix A). A wealth of alternatives to chemical controls exist (Wooten, 1999c). To wage a successful campaign against plant invasions, plans and decisions need to be in place. A decision hierarchy must be capable of responding to emergency risks from actual or potential infestation problems. Such a hierarchy should flow smoothly from National goals, policies, plans, programs and specific project objectives. Management procedures need to anticipate the need for flexibility by incorporating the ability to make decisions based on expected outcomes, and when necessary, to adjust those procedures. The implementation of IWM is dependent on specific land use goals that manage ecosystems, not just administrative boundaries. Its use requires more than just proclamations, but should include detailed descriptions of how goals will be accomplished within ecosystems.

Another important part of invasive species management is coordination with other public and private entities. The Land and Resource Management Plan, 47 FR 43037, Sept. 30, 1982, § 219.9 (a)(7), requires that,

The regional guide shall contain a description of measures to achieve coordination of National Forest System, State and Private Forestry, and Research programs.

The need for coordinated activities is reiterated in terms of invasive species by the General Accounting Office (2000):

Moreover, the agency has no plan to better link its research division and state and private programs (see fig. 1) to the national forests to identify and address stewardship issues--such as wildfires, insects and diseases, and noxious weeds--that do not recognize the forests’ administrative boundaries.

Currently, the individual Forests and have incompatible goals with their regional offices. Regional goal statements include far more documented impacts, such as health protection measures, drift regulations and applicator certification, that do not get translated into Forest EAs. For example, the Forests are dependent on state certified applicators who are generally unaware of new Regional regulations.

Regional and federal offices maintain large databases which are unavailable on the Forests, but which contain important technical references, EPA hazard memos, EPA environmental effects database, EPA health effects database, and other documents that only get incorporated by reference into EAs. In addition to being a violation of the public trust, the lack of information makes it difficult to mount a rapid response to new invaders using the best available science.

Case example: Boulder Creek on the Okanogan NF in Washington

The lack of coordination between different government entities is evidenced on the Okanogan NF by the signing of a Biological Evaluation (Molesworth, 1999) for herbicide use which proposed using label violation as a method of protecting threatened and endangered fish species, in the same paragraph with a statement assuring label adherence:

Surfactants, are often more toxic than the herbicide and will not be used within the riparian buffers...The EA specifies that herbicide label specifications will be followed and specifies that weather and soil conditions that will be met before herbicides are used.

When asked why this occurred, the applicator pinned the blame squarely on Forest Service directions. Although this resulted in a Notice of Correction from the Washington Department of Agriculture (Washington State Department of Agriculture. 2000. Case File 051C-99), the implementation involved other transgressions.

The Okanogan NF also attempted to protect aquatic species without considering the limitations inherent in control programs. The Integrated Weed Management Environmental Assessment (EA) (1997) specified buffers to protect aquatic species:

The mitigation measures for herbicide use include 100' buffer widths along steelhead, bull trout, spring chinook, and westslope cutthroat habitat where no picloram will be applied and require that a Forest Service inspector work with the contractor at all times. Within this buffer and within 50' of all other streams glyphosate will be hand applied.

Yet, the treatments were obviously applied within riparian areas using a sprayer, as weeds were consistently missed while prominent native vegetation along the water was killed or damaged ( Photo 4, p. 3). The National Marine Fisheries Service (NMFS) was contacted, and came to observe the site, however their investigation was never concluded.

Had the operation used hand application in riparian areas as specified, less herbicide would have been used, and this would have been evident from a much reduced application area and avoidance of non-targeted native plants. However a site visit and photographs taken after the application clearly show the width of the swath was wider and more even and continuous than would have resulted from hand application (Wooten, 1999d). Areas practically devoid of any vegetation were treated along with the rest of a five-mile stretch of road ( Photo 4, p. 3). The treatment affected native plants far more than weeds, sometimes missing weeds completely, while spraying over them onto native plants on the streambanks below the road as far as 30 feet beyond the road ( Photo 4, p. 3; Wooten, 1999d). The Forest Service claimed that a Forest representative was on hand during the treatment, but if so, then they must have allowed the contractor to preferentially treat native plants from a truck in violation of the EA.

If the agency cannot be relied upon to coordinate its activities with other agencies and with its own different branches, then there is little reason for the public to approve the use of dangerous chemicals by these same people.

Case example: Wenatchee National Forest in Washington

The Wenatchee NF Environmental Assessment (EA) for noxious weeds (1998) states as its legal direction the Federal Noxious Weed Act of 1974 that, “establishes and adequately funds an undesirable plants management program through the agency’s budgetary process.”

Unfortunately, agency programs cited in the EA are not linked to policy goals, but instead to regional guides for project implementation (Forest Service, 1988; Forest Service, 1992), which constrain plans to the analysis of various tactics for vegetation control without an overarching approach toward achieving federal goals, budgets, coordination, or long-term environmental sustainability.

The EA then goes on to design an entire Forest program within the EA, in excess of the scope of implementation and budgetary authority cited as direction. However admirable, the stated purpose and need do not follow the line of authority from policies to plans to programs to decisions. Instead, the EA concludes that an Integrated Weed Management program (IWM) should be implemented on the District, without really following the regional guidelines for following an IWM program.

IWM was defined by the Forest Service (1988) as a five-step process: (1) conduct a site analysis; (2) select a strategy; (3) design the project; (4) take action; and (5) monitor. It is arguable that this definition of IWM is incomplete. There is no consistent procedure to get from step 1 to step 2, and lacking such authority, the Forest Service inevitably concludes that herbicides will be the chosen “strategy”, then concocts a project design to implement this “strategy”.

Unfortunately, the flaw in this reasoning has resulted in the Wenatchee NF being unable to implement basic weed control measures because the so-called “IWM program” relied too much on herbicide treatments, which exceeded the program budget.

Case example: Wenatchee National Forest in Washington

Lacking a sound program for managing invasive species, the National Forests are floundering in a sea of weeds. Liz Tanke, botanist and Chelan Field Representative for the Northwest Ecosystem Alliance, documented the following cases of inappropriate management on the Wenatchee NF:

In Swakane Canyon, Dalmatian toadflax occurs in multiple large populations, diffuse knapweed occurs along roads and Canada thistle occurs in riparian areas, much of this in bighorn sheep habitat and mule deer winter range on both state and NF land. Toadflax and knapweed are in places where vehicles will pick up the seeds during the fall hunting season and during spring greenup, when deer and sheep will pick up seeds. At least two signboards are available to post notices about weeds, but neither are used to warn the public of the weed infestations.On the Mad River, diffuse knapweed is thick at Pine Flat and Camp 9 trailheads, and is spreading into the Entiat Roadless Area and Chiwawa LSR along the trail. Canada and bull thistle, oxeye daisy, yellow sweetclover and a few patches of whitetop have also been observed along the trail. Volunteers have been pulling the knapweed and as many other weeds as possible for at least four consecutive years, but some of the thistles along the trail and the knapweed at the trailheads need a more concentrated effort. Volunteers have been told not to publicize this activity because it is not legal; there is no NEPA document to cover it. Yet the Forest Service has known about the volunteer effort for several years and despite requests, has not helped make the work legal. Trail crews busy with trail clearing ride their motorbikes past the weeds. When asked, they have admitted that they have not received training in noxious weed management.

On Echo Ridge on the north shore of Lake Chelan, Dalmatian toadflax is spreading within the off-road area where mountain bikers and cross-country skiers use multiple loop trails. Infestations are thick along several miles of approach road and along one edge of parking area. Nothing was done about the weeds in 1999. The road and parking area were mowed once this year which helped reduce the height of weeds, but trails were mowed after weeds went to seed. Toadflax plants growing over roadcuts are going to seed. A large signboard at the trailhead does not have weed information. Volunteers offering to lead weed-pulling outings with bikers and skiers have not gotten replies from the Forest Service offices.

Solutions

Principles of ecosystem management should include:

1) Long-term sustainability.
2) Clear, operational goals.
3) Sound ecological models and understanding.
4) Understanding of complexity and interconnectedness.
5) Recognition of the dynamic character of ecosystems.
6) Attention to context and scale.
7) Consideration of humans as ecosystem components.
8) Adaptability and accountability.

Formulate plans and policies using principles of true Integrated Weed Management (Appendix A). Components of an IWM program should include:

A wide variety of pest control options should be considered in IWM with preference for:

Formulate plans and policies to manage invasive species using principles of Integrated Weed Management within a framework of ecosystem management (IWM-EM, Appendix A). The application of IWM-EM includes the following principles:

1) Maintenance of long-term ecosystem sustainability should be the prime goal.
2) Planning needs to address the causes of biological invasions.
3) Program budgets need to be based on performance evaluations and monitoring.
4) Planning and implementation needs to be site-specific.
5) Planning and implementation needs to be species-specific.
6) Planning and implementation need to address prevention.

Formulate plans and policies based on an analytical approach.

Programs need to treat the causes of invasions, not the symptoms, if they are to succeed. The National Environmental Protection Act (40 CFR § 1508.20) requires that impacts be lessened through:

(a) Avoiding the impact altogether by not taking a certain action or parts of an action. (b) Minimizing impacts by limiting the degree or magnitude of the action and its implementation.

Land managers must address the problems that are the root causes of plant invasions. Such problems are diverse, but they result in the type of resource degradation that predisposes and reinforces the ability of plants to invade ecosystems, which then leads to further impacts, resulting in an inescapable whirlpool of environmental effects, as illustrated in the diagram below (Figure 1).
image
Figure 1. The cycle of plant invasions perpetuated by continued disturbances.

It rarely occurs that the causes of weed spread are attended to. Instead, land managers mitigate and treat the consequences of invasive species spread, not the impact. In essence, they continue to focus on the symptoms.

By treating only the symptoms of poor ecosystem health, land managers insure that programs will fail to control invasive species. In fact, such maintenance measures lack acknowledgment of the causes of plant invasions and therefore will never be a viable means to eradicate the source of infection. What it does do, however, is reinforce the agency’s dependence on herbicides.

The invasion of non-native plant species has resulted in a spate of recent decisions that attempt to justify a perceived need for the use of chemicals to manage impacts. Unfortunately the effects of the treatment are sometimes worse than what triggered them (e.g., the presence of weeds). When added to the prior effects, the result is a dilemma requiring perpetual management to maintain ecosystems from total collapse. These so-called “maintenance strategies” (Forest Service, 1988) are condoned and even encouraged by management schemes intended to insure funding, but which lack long-term sustainability.

In order to comply with regulations in the NEPA to limit negative environmental impacts, land managers must begin to eliminate and reduce the causes of invasive species spread. Planning documents that blame impacts on invasive species, and then subsequently fail to acknowledge the causes of those invasions, only contribute to increased degradation (BLM Weed Team):

Weeds often start in sites where ecosystem processes are disrupted such as trailheads, trails, wildlife bedgrounds, overgrazed areas, and campgrounds.

Planning documents that cite the impacts of weeds as a need for aggressive control measures further violate NEPA and mislead the public. By treating only the symptoms of resource degradation, managers insure that symptomatic control programs will fail to protect resources. In failing to look at the big picture of invasive species as part of ecosystems, program managers have abdicated their responsibility to avoid detrimental environmental effects of projects. Belsky and Gelbard (2000) found that,

“Most of the current recommendations in management plans for stopping non-indigenous plant invasions on public lands in the West focus on preventing landscape-level introductions of weed seeds by washing vehicles and using weed-free livestock feed. Although useful, these strategies are similar to rearranging the deck chairs on the Titanic.”

Plant invasions are themselves a symptom of poor ecosystem health and poor stewardship of public lands.

For instance, overgrazing has been shown to degrade soils through compaction, reductions in soil decomposers, and lowered soil hydrologic conductivity; all of which appear to favor weedy species over native bunchgrasses (Belsky, 1995). Weed invasions are reinforced by a positive-feedback system involving selective grazing, trampling disturbances, destruction of microbiotic crusts and loss of native species (O’Brien et al., unpublished). Conversely, relatively healthy native shrub-steppe lands which still retain native species, such as the U.S. Department of Energy’s Hanford Site in eastern Washington and a semi-isolated plateau known as The Island in central Oregon, are also relatively free of non-indigenous plant species, except along roads (Belsky and Gelbard, 2000). Roadless lands on National Forests and other ownerships also act as strongholds for native communities (Almack et al., 1993).

Without any mandate to seek long-term solutions to problems of invasive species, managers have concocted an array of complex bureaucratic procedures to appear efficient while providing short-term fixes to ameliorate the problem. The Forest Service is well aware of this, as shown by a letter from a private seed supplier that was circulated on the Forest Service computers (Dalpiaz, 1994):

I think that the idea that native plant material is more expensive than exotic materials is a very big fallacy ... What will work best in the long run for overall ecosystem health will always be cheaper in the long run from a management standpoint. Eurasian Wheatgrasses and Smooth Brome might be quick, cheap “visual management” solutions in the short run, but just because they green up an area quickly doesn’t mean it’s fixed.

Case example: Okanogan NF

The Forest Service continues to plead for public recognition of the impacts caused by the invasion of non-native plant species (ICBEMP, 2000):

Cheatgrass and other exotic plant infestations have simplified species composition, reduced biodiversity, changed species interactions and forage availability, and reduced the system’s ability to buffer against change or act as wildlife strongholds in the face of long-term environmental variation.

Yet cheatgrass is not even classed as a noxious weed in most states and is legally allowed in so-called “100% noxious weed-free” seed mixtures which are specified for project use by the Okanogan NF.

The “Respect the River” program was initiated to restore degraded salmon habitat along the Chewuch River. Implementation involved liberal applications of “native” seed mixtures on a number of disturbed recreational sites. Because the sites were hardened campsites with extremely poor soil quality, the chances of successful restoration was not high, and many of the initial plantings failed, except for a single species of grass which managed to survive—cheatgrass.

Cheatgrass is one of the most destructive invasive species in the Inter-mountain West (Monsen, 1994; Mack, 1986), yet some states don’t even classify it as a noxious weed. Although the Washington State Department of Agriculture certifies weed-free seed, it does not require the Forest Service to examine the certification or even require a certification. Regulations to certify seed are inconsistent from one state to the next, and allow any potential new weed species to be present in a mix, so long as they are not classified as Noxious in the receiving state.

The problem is that the causes of degradation—loss of resources due to spreading populations of invasive species—are not being attended to. Instead, the “Respect the River” program treated the symptoms of the problem—unhappy campers. Because our existing framework for controlling invasive species is based on a “noxious weed” list that is primarily designed to protect agricultural enterprises, ecosystem restoration efforts are failing and in some cases causing more harm than good.

Case example: Boulder Creek on the Okanogan NF in Washington

Because of the rugged terrain on the Okanogan NF, many roads on the Methow District are located along streams and rivers where they have become festering sores in the fragile glacial soils. The Forest acknowledges that, “We know that some road fills in the Boulder watershed have little to no vegetation on them and have surface erosion occurring at rates higher than we would like.” (O’Neal, 2000).

To cope with the problem of eroding road banks, the Okanogan NF often uses non-native forage grass plantings to meet road vegetation guidelines. However, the problem of road bank erosion may be worsened by the use of forage grasses, not only because these are aggressive and invasive in their own right, but also because high-protein grasses attract deer and free-ranging livestock to sites where they grow. The result is that the cycle of Fig. 1 is perpetuated: soil disturbance leading to invasive species, leading to more soil disturbance, ad infinitum. The cause of the problem which should be addressed is the cycle of disturbance. But, the Forest is treating the symptom—lack of roadside vegetation—through an inappropriate treatment strategy of planting more invasive species.

The result is that native shrub-dominated communities have been replaced by orchard grass, Kentucky bluegrass, timothy and intermediate and crested wheat. Roadsides have been converted into linear pastures ( Photo 1, p. 3), and the increased trampling disturbance provides a seedbed for new weed seeds hitch-hiking along with livestock and cars; trampling and driving helps scarify the seed-coats and cover the seeds with soil; and the ruts and hoof prints provide a convenient microniche to hold water during early development of the weeds.

These artificial ecosystems are self-reinforcing cycles of destruction, which can only be controlled by breaking the disturbance cycle, and restoring a self-sustaining vegetative community. Existing programs such as treating these sites with herbicides as the Okanogan NF did in 1999, contribute to further loss of beneficial native communities, and greater soil loss, while wasting valuable public funds. Weeds on the Okanogan NF are a symptom of poor forest health brought on by years of roading, logging and livestock grazing, but until these causes are addressed, symptomatic treatments will continue to fail.

Solutions

Section B. Projects and plans must incorporate measurable standards and guidelines.

The recently created National Invasive Species Council (2000) has made a strong recommendation for creation of standards and management framework for invasive species:

In 1997, 500 scientists and resource managers wrote to the Vice President and requested action on invasive species ... The team prepared a review of the issue with recommendations, foremost among them that an executive order be issued providing standards and a framework for continuing action. On February 3, 1999, President Clinton issued Executive Order 13112 on Invasive Species . . .

During scoping from the National Invasive Species Council Research, Information Sharing, Documentation and Monitoring Working Group (2000), the importance of developing standards was addressed:

Information sharing and documentation issues address the fact that our efforts related to invasive species have been hampered by the fragmented state of information systems addressing invasive species. A major challenge identified was to develop standards and protocols that shared core information (species names, locations, collections, effective practices, experts, etc.) needed for synthetic national assessments that can be reported in a cost-effective and transparent way, while still maintaining local control and responsibility for the fundamental data.

Standards and guidelines should be reasonable and should be implemented consistently across all National Forest lands. Decisions are being issued with statements that certain actions will occur, such as monitoring, reseeding, etc., when in fact such objectives have no means of accomplishment.

At a minimum, standards and guidelines should include damage and action thresholds for management and control of invasive species. The use of thresholds is a tenet of Integrated Pest Management, which uses pesticides only within predetermined action and damage thresholds (Quarles, 1999; Hoglund et al., 1991; Lanier, date unknown).

Damage thresholds refer to limits on the amount of impacts to humans and the environment that will be tolerated. Damage thresholds are determined by the amount of change that occurs in the environment or in humans as a result of a natural or management-induced process. The Forest Service uses a very narrow definition of damage threshold, limited only to the damage caused by invasive species (Okanogan NF 1997, 1999). Damage caused by the Forest Service is excluded from planning documents. Damage thresholds for chemical impacts are not presented in Forest Service planning documents because they are a constraint on the status quo. By forcing managers to limit their actions within acceptable limits, setting damage thresholds for control measures has the potential to upset the free rein which managers have had in the past in accomplishing targets regardless of the consequences.

Action thresholds are determined in relation to damage thresholds (Hoglund et al., 1991). Action thresholds set the minimum allowable quantity of an invasive species beyond which control measures will be undertaken. The use of action thresholds is dependent on both the management goals for an area as well as the calculated threat from each invading species. For instance, aggressive rhizomatous grasses might be highly tolerated in pastures and given a high action threshold; whereas in tree plantations, the action threshold might be lower and in Wilderness areas it might be zero, or not tolerated.

In conjunction with the development of action thresholds, which would determine when control or preventive actions were mandated, a national list of potential and known invasive species for which damage thresholds would apply needs to be compiled, along with the rationale for a species inclusion, and the action threshold it would receive for treatment. The Invasive Species Council (2000, Section 4) described the creation of such a list within a few years:

Invasive species must be detected and identified before they become widespread. No comprehensive national system is in place for detecting and responding to incipient invasions. Key elements needed in such a system are accessing current scientific and management information; facilitating identification . . .

After years of neglect and inefficient spending on ineffective programs, the revised Forest Service Strategic Plan (Forest Service, 1999) performance measures for invasive species for the year 2006 is still nebulous,

Acres infested with targeted invasive species remain unchanged or are diminished.

In contrast, the National Park Service has spent considerable amount of time and energy analyzing the management factors for various weeds (Hiebert and Stubbendieck, 1993), which the Forest Service would do well to heed:

Managers must not only be concerned with the level of impact that an exotic can cause but must also consider the impact of removing the species. Removal can often disturb areas that are easily colonized by the same or other exotic species (Westman, 1990).

Case example: Boulder Creek on the Okanogan NF in Washington The Okanogan NF Environmental Assessment on noxious weeds (1997) proposed treating half of 10,000 weed infested acres with herbicides.

The public was concerned about the deleterious effects from using these herbicides, and said so in numerous letters, e.g., under issues contributed by the public (p. 15), “herbicides could damage or kill non-target species, even if used according to label instructions.” Forest Service reviews of the herbicide glyphosate (Syracuse Environmental Research Associates, 1996) indicate that it can be toxic, e.g., “herbicide treatments have the potential to be lethal to fish, amphibians, and aquatic invertebrates if the concentration of the application exceeds the tolerance of the organism and becomes toxic.”

Since herbicides are acknowledged to be toxic to a wide variety of organisms, the appropriate action would have been for the EA to analyze the extent of these risks through reference to the available literature, and then to establish threshold concentrations that would be measured during implementation to mitigate the harm that would occur as a result of using chemicals in the environment.

However, instead of analyzing the risks posed by applying toxic chemicals on over 5,000 acres, the EA brushed off these concerns, and cited a single out-of-date reference (Norris et al., 1991) to demonstrate that proposed chemicals will have low risks, in contradiction of the statements made about toxicity.

The EA did not include an analysis of the likely effects from herbicides, stating that this information was purportedly incorporated by reference to regional and national analyses, such as those performed during registration of a pesticide by the Environmental Protection Agency. Unfortunately, these references are generally unavailable to the decision-maker, the District staff, and the public at large.

The EA did not provide damage thresholds for chemical effects as required in the Competing Vegetation EIS, FEIS and ROD (Forest Service, 1988), but nonetheless stated that the herbicide applications would meet the Aquatic Conservation Strategy Objectives of PACFISH (Forest Service and BLM, 1995) and INFISH (Forest Service, 1995) and the impact on fish from herbicides would be “little”.

With no damage thresholds to measure effects, the assumption that impacts will be “little” is without justification. The EA goes on to print misleading and ridiculous claims for benefits, stating that,

Native plant species would dominate existing population centers. This would improve and maintain wildlife habitat. Slight possibility of damage to non-targeted animal species from chemicals.

Building on this flawed analysis, a subsequent two-page Biological Evaluation was prepared and found there would be “no effect” on listed fish. The EA and BE are not posing reasonable alternatives, they are building a case for herbicide use. But based on the flimsy evidence given, the entire process can be seen as nothing but a house of cards. Clearly the “no effect” determination of the BE occurred without substantial review, and the actual, likely consequences were never described in the document.

In fact, the actual likely consequences that went undocumented included: (1) use of the wrong herbicide (picloram, not glyphosate) in riparian areas as documented in Forest Service memoranda and photographs (Bennett, 1999: Figure 6, and statement); (2) a killed or damaged swath of primarily native vegetation 3-10 feet wide along roadsides; and (3) increased streambank erosion due to removal of beneficial native plant cover (Wooten, 1999d).

The risks of ground-based herbicide applications exceeding damage thresholds to nontarget species was not presented. The lack of standards and guidelines meant that environmental impacts went undocumented. Reliable information that was available was ignored, resulting in a biased decision. Ultimately, the environment was degraded more by the cure than the cause, however the lack of standards and guidelines for protecting the environment insured that the public and decision-makers would never know the true extent of the damage.

Solutions

Section C. Programs must be held accountable to budgets.

The goals and objectives of programs and projects should specify a timeline and budget for project accomplishment. Program budgets need to consider long-term funding limitations before deciding on an implementation plan. The current U.S. budget of $251 million for prevention of alien species invasions is estimated to be one-quarter of one percent of the yearly U.S. damage costs from all invasive species (National Invasive Species Council Risk Analysis and Prevention Working Group, 2000).

Existing budgets for invasive species are run as ad hoc extensions of grazing or timber departments which lack a comprehensive framework for handling species invasions (Okanogan NF, 1999). According to the General Accounting Office (2000):

. . . the Forest Service continues to develop its annual budgets and to allocate most appropriated funds to its field offices primarily on the basis of its nine National Forest System programs. (See fig. 1.) These programs are not linked to the Forest Service’s strategic goals, objectives, or strategies or to the way that work is routinely accomplished on the national forests.

Once funding is obtained, there is little incentive to follow through with plans. Existing budgetary policies within the Forest Service are lax and they act as a disincentive to cost-effective programs by basing future budget allocations on past spending, without a hard look at how funds were spent.

Case example: Budget considerations

According to information received under a March, 2000 FOIA request from Kettle Range Conservation Group to the Region 6 Forest Service, the Okanogan NF received $300,000 for managing invasive species in 1999, which was used to apply herbicides over approximately 5,956 acres. But despite the large sum of money allocated by Congress, the Okanogan NF stated that three times this amount, or $1 million, would be necessary to control just one third of the existing weed population (Okanogan NF, 1997, p. 91). Funding on just the first year was completely exhausted before completing required monitoring specified in the EA, however, another appropriation of $300,000 was granted to the Okanogan in 2000.

At the March 29th, 2000 meeting in Bend, Oregon, between the Forest Service and parties to the Mediated Agreement (1989), Wenatchee NF botanist, Terry Lillybridge, described procedures used to get more funding for noxious weed programs. It turned out that the Okanogan and Colville NFs got special allocations through lobbying efforts of the Okanogan County Weed Supervisor at the office of U.S. Senator Gorton. From this, it is apparent that the Forest Service is relying on under-the-table deals and pork barrel politics to secure funding, rather than proceed through its own internal processes.

Perhaps the federal funding process for managing invasive species needs to be changed. To undertake such a large program without assurances of budget accountability or long-term effectiveness is an irresponsible waste of taxpayer money. As described in the EA (p. 91), $300,000 is but a fraction of the $1.8 million that would be required to treat all the first-, second-, and third priority sites, and there is no guarantee that this experiment would work.

In fact, the following observations indicate that the project is failing to control weeds and protect public resources. Untreated populations continue to threaten the Forest, and new invaders have already returned to the treated sites (Wooten, 1999d). Native plant communities were the unintended targets of many treated sites. Spray applicators treated areas at their convenience, without Forest Service inspectors present to oversee their methods or postings of required public warnings (from a FOIA response of notes of Forest Service Contract Inspector Bauman, July 7, 2000). Herbicide labels were routinely violated in a number of ways: disregard for riparian buffers (Bennett, 1999); use of restricted herbicides on livestock forage (Wooten, 1999d); and through Forest Service instructions to applicators to violate labels and use of incorrect formulations.

Case example: GAO investigation into Forest Service performance

During hearings before the House of Representatives, an investigation by the Inspector General of the General Accounting Office (2000) described how a Forest Service worker reported spraying five acres of road with herbicide, instead of the 18 acres actually sprayed, because had the correct figure been reported, subsequent budgets would have been restricted to discounted prices obtained for that year’s purchase of herbicide (Hughes, 2000).

Solutions

Section D. Policies, plans, and programs needadjustments based on periodic evaluations.

In testimony before the Subcommittee on Forests and Forest Health, Committee on Resources, House of Representatives, on June 29, 2000, the General Accounting Office (2000) reported,

In addition, even though forest plans are intended to serve as a basis for developing future budget proposals, the Forest Service has not determined how or if the national forests will blend agency-wide objectives and strategies with local priorities in revising their plans. Moreover, instead of developing new performance measures and improving existing ones to better align them with its strategic goals and objectives and its on-the-ground projects and work activities, the agency is relying on old program-based performance measures.

Proposed new rule changes in the National Forest Management Act (Forest Service, 1999b) underscore the importance of adaptive management in relation to monitoring:

In addition, as efforts continue to adopt the principle of adaptive management to guide natural resource stewardship, greater emphasis needs to be placed on evaluating resource conditions and monitoring trends over time. Consistent with the 1990 Critique as validated by the Committee of Scientists’ report, the proposed rule emphasizes monitoring and evaluation so that management can be adapted as conditions change over time.

The Hearings over the Results Act brought these problems to the attention of the House of Representatives (Hill, 1997):

Our report on the Forest Service’s decision-making identifies problems in the agency’s data and information systems dating back 17 years. These problems include (1) not adequately monitoring the effects of past management decisions to more accurately estimate the effects of similar future decisions and to modify decisions when new information is uncovered or when preexisting monitoring thresholds are crossed and (2) not maintaining comparable environmental and socioeconomic data that are useful and easily accessible. We and others have recommended steps that the Forest Service could take to improve its data and systems, but it has deferred action on these recommendations ... . . . In conclusion, Madam Chairman, the inefficiencies and ineffectiveness of the Forest Service’s decision-making, combined with the agency’s reluctance to change, give urgency to implementing the Results Act. The agency’s plan should provide the starting point for establishing the measures and annual target levels to be used in assessing the Forest Service’s progress toward achieving strategic goals. However, the draft plan’s silence on the Forest Service’s rationale for its strategic goals, its management approach, and the likely effects of its policy choices on multiple uses on the national forests has contributed to a stalemate on the agency’s strategic goals which threatens successful implementation of this landmark legislation.

The National Forest Management Act (NFMA) requires that Forest Plans must contain “monitoring and evaluation requirements that will provide a basis for a periodic determination and evaluation of the effects of management practices” on forest resources. 36 CFR § 219.11(d). To effectively monitor the impacts of management actions, each Forest Supervisor is required to “obtain and keep current inventory data appropriate for planning and managing” forest resources (ibid. at § 219.12 (d)). Field units operate within regional and national guidelines, which are in turn dependent on receiving meaningful feedback from the units. There are strong indications that existing weed management programs are failing in many respects, however this information is not being relayed beyond individual Forests, which in some cases are not even performing required monitoring. Policy-makers should be provided with program evaluations in a timely fashion, so that plans can be adjusted when necessary.

Case example: Boulder Creek on the Okanogan NF in Washington

The Okanogan NF Environmental Assessment (EA) on noxious weeds (1997) was accompanied by a Biological Evaluation (BE) written two years after the 1997 EA and 3 years after a study provided for the Forest Service to review studies on the effects of the herbicide glyphosate (Syracuse Environmental Research Associates, 1996). In a March 3, 1999 letter from Janice McDougle to Forest Service Region 5, various reasons were given for denying a request to include an additional 78 citations to those in the review. These include the conclusion that based on a cursory review by the Forest Service Washington Office (McDougle, 1999), the new information,

. . . would not alter anticipated impacts or consequences on aquatic organisms, water quality, plants and food, occupational exposure, drift, soil, and nontarget species from those previously disclosed. We suggest that the individuals submitting the list of references be asked to explain, in detail, why the studies referenced would significantly change or alter any of the consequences that the Forest Service has or will disclose to the public.

During appeals of the EAs for noxious weeds (Okanogan NF, 1999; Colville NF, 1998), the Regional Office went on record to deny requests to incorporate additional references to the toxicity of glyphosate made since the SERA was issued 5 years ago, stating,

. . . the peer review process used by SERA [Syracuse Environmental Research Associates] for Forest Service pesticide risk assessments includes qualified scientific experts outside of the Forest Service. Their comments would already be incorporated into the final document you have retrieved.

Even if these assumptions were invalid, they should have been caught by monitoring. However, monitoring was not performed in most of the treated areas.

Following the implementation of the 1997 weed control program, photographs and surveys by independent investigators clearly show that the amount of nontarget vegetation killed was significant (Wooten, 1999d). Had the EA disclosed the likely effects, as it should have, a “no effect” determination might have been untenable. In that case, this disclosure would have altered the consequences, as the document could not implemented until consultation with lead agencies under the Endangered Species Act (ESA) occurred. This also would have altered the consequences in that oversight might have prevented the spraying of Tordon 22K© into at-risk streams and possibly would have prevented habitat degradation for threatened and endangered bull trout and salmon.

But the EA could not address the likely effects from herbicides because this information was not incorporated into regional and national analyses, which are generally unavailable on the Districts anyway. The EA only briefly mentions effects from herbicides and then primarily under “Issues Contributed by the Public . . . herbicides could damage or kill non-target species, even if used according to label instructions . . .” (p. 15). The EA cites a single out-of-date reference (Norris et al., 1991) to justify the chosen alternative, despite numerous references in possession of the Forest Service (Syracuse Environmental Research Associates, 1996) showing a greater range of herbicide toxicity.

What should have occurred was to research the available literature and analyze whether the basis for the invasive species program was on a sound foundation. Instead, the EA makes projections based on invalid assumptions, e.g., “Formulations containing herbicides picloram or glyphosate would be applied with implementation of this alternative. Both chemicals are rapidly diluted and tend not to bioaccumulate.” (p. 107). This is not an honest appraisal of available research, it is blatant rationalization. Only a single reference is cited, and that one is out of date. Furthermore the summarization lacks quantification (“rapidly”and “tends”leave much to the imagination).

The failure of the Okanogan NF to control invasive species lies largely with their lack of regard for following procedures. The Forest had numerous opportunities to question its motives and methods, but in each case blundered ahead without assurances that projects would be effective. Until the Forest Service is willing to change the way it does business, we will continue to see poor performance records.

Case example: Poor record keeping on Okanogan NF and Colville NF

The Okanogan and Colville National Forests applied herbicides over approximately 5,956 and 3,791 acres, respectively, in recent single years (Okanogan NF, 1997; Colville NF, 1998). A Freedom of Information Act Request (FOIA) was sent to the Regional Office by Kettle Range Conservation Group on March 14, 2000, requesting figures for the amount of acreage treated. The response took months for a reply to be received, which was a single page answer, stating the number of acres treated with herbicides as 810 acres for the Okanogan, and 1,777 acres for the Colville.

In trying to determine the reason for the large discrepancy with the Environmental Assessments, the Kettle Range Conservation Group sent another FOIA request for the number of treated acres to the Colville NF and received the following reply:

This item . . . was not kept on Forest files and has just recently been mailed to us from the Regional Office.

It is interesting that the Colville NF says it did not have these required records because the information was supposedly at the Regional Forest, which supposedly came from the Colville NF in the first place. Unfortunately, the public will probably never know the amount of acres treated with herbicide on the Colville NF more accurately than between 1,777 and 3,791 acres. The Okanogan NF responded to a FOIA request for the number of acres treated with chemicals in 1999 by sending an Accomplishment Report, but only from the Tonasket District, indicating that a total of 1,937 acres was treated. Again, the actual amount treated, which will probably never be known, was somewhere between 1,937 and 5,956 acres, however the Regional Office blithely went on record to say that only 810 acres were treated on the Okanogan NF. However important it might be, the need for adjustments to plans, programs and policies, cannot occur in a logical fashion until adequate record keeping occurs on the National Forests. In the meantime, poor record keeping serves to tell the public and policy-makers that everything is just fine.

Case example: The injunction in Region 6 Forest Service

In 1984, litigation over health and environmental damage on private lands caused by aerial herbicide spraying by the Forest Service and BLM resulted in a five-year injunction against using pesticides in Region 6. Following the injunction, the Forest Service did practically nothing to monitor or control invasive species on public lands, while allowing and even causing them to spread widely.

When interviewed by Audubon author Ted Williams (1997), University of Wyoming weed scientist Tom Whitson gave his rationale for the government’s lapse of weed management as the inability of bureaucracies to cope with adversity:

Why hire professional people and limit them to the point where they can’t operate? . . . In government, progress is slow anyway; and when some of these groups start blocking it with a bunch of environmental-impact statements and lawsuits, it [all weed management] stops.

A more considerate explanation for the lack of invasive species management during this period might take note that when the existing program was excised from the timber budget, there was no weed management program at all left in its place.

Case example: Hells Canyon Research Natural Area on the Wallowa-Whitman NF

Williams (1997) describes careless Forest Service reseeding efforts following the 1988 Tee Pee fire in Hell’s Canyon, which resulted in a disastrous explosion of yellow star thistle (Centaurea solstitialis) in a Research Natural Area.

Despite the fact that the existing injunction against herbicide use in the Region was lifted the year after reseeding the burn, managers with failing programs found an excuse to blame the injunction rather than admit to their superiors that they were responsible for the infestation.

The initial control efforts were ineffective, despite costs of over $200,000 (Bob Williams, Wallowa Whitman NF, personal communication). However, detailed accounts of the reasons for the failure have since been destroyed (Forsgren, 2000), even though such information could be very informative. According to Williams (1997), “Managers knew they weren’t going to eradicate yellow star-thistle . . .” The reasons for the failure of previous herbicide efforts are now concealed from the public and future managers who might otherwise make beneficial changes in the arena of invasive species management.

Lacking reports that would inform planners of the optimum course of action, hundreds of thousands of additional dollars have recently been allocated to annually treat 14 sites on 5,000 acres in Hell’s Canyon, using aerial herbicide applications (Wallowa-Whitman NF et al., 1998). Although other methods are briefly mentioned, the proposal is clearly aimed at using herbicides first, and prevention last. Ironically, the government proposal fails to mention requirements to prioritize prevention and perform site-specific analysis and monitoring which were stipulated in the Mediated Agreement (1989) between the USDA and plaintiffs Northwest Coalition for Alternatives to Pesticides (O’Brien, 1989). In all these failed attempts to control invasive species, the common thread has been a lack of sound planning.

Solutions