Saltcedar, giant reed and yellow starthistle (figure 1a-c) are often identified as the worst invasive species in the western states and are listed as the top noxious weed species in California by the California Exotic Pest Plant Council (CalEPPC 1999, Kelly, Wagner and Warner, 1998). In a recent national survey conducted by polling hundreds of cooperators in every state, USDA-APHIS determined that both saltcedar and yellow starthistile were national priority weed control targets (Wendel, personal communication 1999). In this proposal, saltcedars (Tamarix ramosissima and T. parviflora) and giant reed (Arundo donax) are being studied together as they often occur in the same habitats, compete for similar resources, and either species will fill empty niches if one is controlled while ignoring the other. In fact, these two plants were characterized as the Deadly Duo in a Cooperative Extension Workshop on their management (Bell 1999). Saltcedar is an invasive shrub to medium sized tree, while giant reed is a tall (30 feet) perennial grass that grows from a large rhizoid mass. Both of these species are of Eurasian origin and were purposefully introduced into North America by early settlers (Brock 1994, Robbins et al. 1951). However, over the last century, these two plants have escaped cultivation and have spread extensively throughout riparian areas in the southwestern US. Although Arundo donax is widely distributed, it is most heavily concentrated in southern and central California and along the Rio Grande from New Mexico through Texas (Cheatham and Johnson 1995, Tracy and DeLoach 1998), while saltcedar is found infesting nearly every perennial drainage in the arid and semi-arid regions of the western US (figure 2). To date, saltcedar has replaced native riparian forest and scrub communities in approximately 1.2 million acres in 23 states from sea level to 7,000 feet elevation (Everitt 1980, Robinson 1965). Although only preliminary economic assessments have been completed on these species, Zavaleta (2000) estimates that saltcedar induced losses to water, agriculture and wildlife populations will cost an estimated $16 Billion to US consumers, producers and the natural environment over a 55 year evaluation period.

Individually and in combination, Tamarix spp. and Arundo donax out compete native vegetation often forming dense monotypic thickets; they provide poor habitat for other flora and fauna; they cause increased siltation of stream and riverbeds and alter channels in ways that eliminate valuable fish habitats; they are excessive water consumers; they increase the likelihood of fire; and most importantly, they alter ecosystem dynamics that further favor their growth and
development at the expense of agricultural production and native species (Bayer et al. 1999, Bell 1997, 1999). Although originally used for erosion control on many western waterways, both species are found to cause extensive flooding by growing in and along waterways, thus forcing flood waters out of the primary channels and into critical banks, bridges and other physical structures. This flooding often ends in extensive property damage, habitat loss and further expansion of the invasive species range (Bell 1998, Vartanian 1998). For these reasons, a huge amount of time, effort and resources have and are continuing to be spent for their physical removal with little to no long-term control possible. Costs of removal vary but are often several thousand dollars per acre in areas that are rapidly reinfested if sustained efforts are not maintained over many years (see Bell 1998).

Where these two riparian invaders leave off, another destructive exotic plant species, yellow starthistle (YST), Centaurea solstitialis, often begins, as it grows right to the edge of stream channels where it can be found adjacent to both saltcedar and giant reed (figure 1c). Thus these three species provide almost a continuous expanse of exotic plant infestation in many western habitats, especially in California where land managers would benefit from planning control procedures for all these species simultaneously. Yellow starthistle infests over 42 million acres in California alone, with additional infested areas in the adjoining western states of Oregon, Idaho, Nevada, Washington, and to a lessor extent, Arizona and New Mexico. The impact of this plant is severe and diverse, especially in highly infested areas where it reduces productivity of agricultural land in many ways. It is particularly damaging for ranchers as cattle avoid feeding on it due to the spiny nature of the developing buds and flowers, and as it produces chemical compounds that cause brain lesions and death in horses. The environmental impact of YST is of great importance not only to grazing wildlife but also to native plants. Its extensive root system is able to reach water deep in the soil and commonly reduces available moisture, allowing YST to out-compete many native plants. This impact on native plants is critical as their demize further allows the increase of invasion potential of this and other exotic weeds in nondisturbed areas. This shift in local flora and reduction of soil moisture is also thought to increase fire hazard in infested areas, which is already a serious problem in most western states.

Based on extensive assessments by the Cooperative Extension Service, several state departments of agriculture and natural resources, USDA-APHIS and several US Department of Interior Agencies, these three exotic species are thought to be the three most serious exotic plant invaders in riparian and adjacent upland habitats. Although Tamarix and Arundo are restricted primarily to riparian zones, YST grows from streamside habitats to upland areas infesting a variety of range, roadside and pasture environments. In all of these cases, a variety of control methods including cultural, physical and chemical are available, however, they are expensive, difficult to implement, and often harmful to other plants and animals in the associated habitats. The great economic and environmental importance of these weeds has mobilized a variety of groups to join the fight against these noxious exotic invaders. Several teams of research and land management personnel are actively involved dealing with these species at local and regional levels. In fact all of the Consortium team members are involved in combating one or more of these species though their agency of employment or through involvement in groups like Team Arundo del Norte, Cal EPPC, or others. Through the efforts of various subgroups involved with the Consortium, biological control efforts are being developed for each of these pest plants as one of the primary tools for their long-term management.

Although each of these target weeds is in a different developmental state in regard to biological control efforts, they are all underway. USDA-ARS is currently engaged in developing and evaluating many aspect of the biological control of these three species. They are doing so in conjunction with many cooperators who have expertise to better integrate biologically-based control methods with other weed control technologies. To-date, pest status assessments, faunistic surveys and foreign exploration have been initiated for Arundo donax. Extensive foreign exploration and host-specificity testing has been conducted for Tamarix spp. and the first natural enemies of this pest plant were released into field cages last summer (see Gould 2000). Due to its long-term pest status, YST has been the target of an intensive biological control program by USDA-ARS, the California Department of Food and Agriculture (CDFA), and the University of California. So far 7 different natural enemies (6 approved and one by accidental introduction) were released for control of this invasive weed. Although some success is at hand, more biological data is needed to quantify and adequately assess the biological control impact of these agents. It is also important that the affect of these agents is taken into consideration when developing management strategies for other invasive exotic species in adjacent areas and when considering the addition of more new biological control agents with existing natural enemies.

As can be seen from the list of cooperating agencies and individuals listed in the "Key Personnel and Their Project Responsibilities" (Section H), the Consortium is currently composed of a highly qualified and diverse team of specialists from all disciplines that are necessary to make this project successful. Since the Consortium has been functional for approximately two years, it is well organized (see figure 3), has several operational committees, and has completed several significant products including detailed implementation plans for biological control agent release and evaluation in six different states. As depicted in figure 3, the consortium has three Project Co-coordinators (Carruthers, DeLoach and Otteni) that currently oversee program management and coordination. They are advised by an Ad-hoc Management Committee of 15 members that sets program direction, reviews project progress and determines needed subcommittees and their membership. Currently, the Consortium has four standing committees that provide the operational expertise to the group (the Monitoring Committee, the MOU, Research and Coordination Committee, the Public Information and Technology Transfer Committee, and the Revegetation Committee). The responsibilities of the committee are relatively self explanatory and will not each be described in detail, however, as an example, the roles, responsibilities and products of the monitoring committee will be described as they provide the basis for some of the implementation methods of this proposal. The monitoring committee actually consists of an overall committee and four subcommittees composed of specialists in the areas of entomology, plant ecology, wildlife biology and assessing the physical environment. Based on programmatic needs as outlined by the USDI-FWS and USDA-APHIS, the different subcommittees developed and interfaced four different monitoring plans that would systematically provide the data and other relevant information required for the implementation of Task 3 of this proposal. The detailed monitoring protocols of each subcommittee (insects, vegetation, wildlife and abiotic factors) ranged from 4 to 44 pages in length and discussed all the relevant information and methods to allow accurate and precise characterization of these components at each of the 13 proposed field test sites. Through two interactive workshops held in San Diego, CA (Feb 1999) and Albuquerque, NM (Oct, 1999), the Consortium formalized these plans and accepted them as the bases for program implementation (see Gould 2000, and Task 3 methods for plan details).

Due to the diverse nature of the Consortium participants and their programmatic backgrounds and interests, this process was an interesting and effective team building experience for all involved. Clearly, establishment and effective interaction of a group as large and diverse as this Consortium requires an extensive amount of time to form, develop overall common goals, and then become operationally functional. Our efforts over the past two years have shown that our Consortium is both serious and effective and we have accomplished many cooperative goals and objectives already with no formal outside supporting resources. To date, each individual and participating group has basically paid their own way in the development of the Consortium and in helping to set the common goals, objectives and methods. We would suggest that this level of commitment and effective collaboration document our ability to work effectively together and to accomplish the wider goals as listed in this proposal.

In summary, based on the private and multi-agency opinions expressed in the Consortium, it is clear that if the objectives of this proposal can be met, several significant accomplishments will be achieved. First, successful integrated biologically-based control of one or more of the target invasive weed projects would be a major economic and environmental achievement. Each of the targeted species causes $100s of millions of losses each year to a diverse set of stakeholders that range from small farmers to large government land management agencies. Current control methods such and herbicide application, physical removal, burning, and/or other methods are expensive, harmful to non-target species and do not provide long-term control. In many areas economic use of the affected land is no longer economically viable and unfortunately, these noxious species are currently further expanding their range at an alarming rate. We feel that the outputs of this project (see Approach) hold significant potential to provide many new methods that will allow partial, if not total, control of these pest plants, and help to slow their spread and overall damage. We also believe that the methodologies used in the coordination of this effort and the outcomes of the detailed research studies will provide excellent guidance to others interested in conducting large-scale collaborative efforts of this type. Last, and possibly most important, we feel that a new inter-agency risk/benefit evaluation process for biological control assessment will provide both the conservation biology and the agricultural/ resource management community with better assurances that exotic natural enemies are being safely evaluated, regulated and used to benefit all aspects of our North American heritage.

Scientific Significance

The research aspects of this proposal are aimed at providing a scientific bases for the management of exotic and invasive pest plant species through the collection of detailed mechanistic data that characterize and explain the biological responses of multiple species interacting in complex ecosystems. That is to say, that we hope to understand invasive weed biology, the biological and physical factors that allow them to become invasive, their interactions with native flora and fauna and with potential introduced natural enemies that are being considered for their control, and the effects that these natural enemies may have on non-target organisms. We hope to conduct detailed experimental investigations that will characterize many of these factors at the genetic, physiological, population and community level of interest. Clearly, not all aspects of these systems can be examined at that level of detail, however, critical threads of information relating factors of biological significance will be investigated at multiple levels of interest and synthesized through systems level technology such as simulation modeling. We also plan to develop and apply remote sensing technologies, geographic information systems and dynamic information delivery systems to aid in our understanding and management of invasive plants. We are particularly interested and excited about developing a more cause and effect understanding, and thus a predictive ability to determine how exotic herbivores mechanistically interact with their natural hosts and thus be better able to pre-assess their impact in the areas of introduction. This is especially important for developing safe and effective biological control.

Practical Significance

From a practical perspective, we believe that our efforts will lead to more effective methods of managing saltcedar, giant reed and yellow starthistle. We expect that a better understanding of the invasive nature and ecosystem level impacts will be developed for each of these target weed species. Based on this knowledge, new methods will be combined into effective integrated weed management programs for local groups, small and large farmers, tribal lands and others. We are particularly interested in working with local weed management associations, local Cooperative Extension personnel, and local land management agency representatives to implement new management methods at a watershed or ecosystem level. In addition, we believe that this program will aid others in combating exotic invasive species by providing an effective Consortium format within which successful research and implementation can be conducted. We feel that it is important that such a consortium be developed out of a programmatic need rather than just an organization developed to acquire resources. It is important that such a consortium work effectively to develop tools that can be delivered to customers that are in direct need of action oriented products.