1.1 Background
The University of Montana (UM) is fortunate to have several hundred acres of natural areas in Missoula County. These include dry grasslands and open forests on the slopes of Mt Sentinel adjacent to the main campus, as well as old growth cottonwood galleries along the Bitterroot River at Fort Missoula.

These properties were acquired by UM with the intent of providing areas for natural areas research, education, recreation opportunities, and wildlife habitat. These values are threatened by the presence of invasive plants.

Conservation biologists and ecologists recognize invasive species as a leading threat to North American biodiversity. Noxious weeds, which are a legal category of invasive species, evolved on other continents and were introduced here both intentionally and unintentionally by humans. Impacts of noxious weeds to their invaded ecosystem include loss of biodiversity of native plants, loss of wildlife habitat, increased soil erosion, and loss of agricultural productivity.

The most common noxious weeds in the Missoula Valley are spotted knapweed (Centaurea maculosa), leafy spurge (Euphorbia esula), Dalmatian toadflax (Linaria dalmatica), sulfur cinquefoil (Potentilla recta), common tansy (Tanacetum vulgare) and hound's tongue (Cynoglossum officinale). Combined, these species infest hundreds of acres of UM’s natural areas, reducing the biodiversity and conservation values for which the lands were acquired.

Additionally, the unchecked proliferation of noxious weeds on UM property is a violation of state laws. House Bill 395, passed in 1995, requires all state agencies in Montana to enter into cooperative weed management agreements with their local county weed districts. It also requires that agencies have their weed management plans reviewed by the weed district every 6 years.

1.2 Scope and need

This document provides a vegetation management plan for UM natural areas, including Mt Sentinel and Fort Missoula. It does not address main campus or the landscaped properties including the UM Golf Course, or off campus residence areas. A plan and its implementation is necessary in order to comply with the Montana Code Annotated Section 7-22-2151 (HB 395), which requires that all state agencies within weed control districts enter into cooperative noxious weed management agreements with the respective weed boards. A written plan is also useful to solicit public input to the management process on public lands.

The previous weed management plan for UM land on Mt Sentinel was approved by the University and the Missoula County Weed District in 1998, and implemented beginning in 1999. This document is an update of that plan, and when approved by the Missoula County Weed District, will ensure UM’s ongoing compliance with state law.

1.3 Project areas

This plan addresses UM land on Mt Sentinel (Figure 1) and at Fort Missoula (Figure 2). Detailed descriptions of these properties are included in the appropriate section.

In general, the native plant communities on these properties include bluebunch wheatgrass (Pseudoreugnaria spicata), rough fescue (Festuca scabrella), Idaho fescue (Festuca idahoensis) and junegrass (Koeleria macrantha) as dominant grasses. Silky lupine (Lupinus sericeus), larkspur (Delphinium bicolor), arrowleaf balsamroot (Balsamorhiza saggittata), and penstemon (Penstemon wilcoxii) are common spring wildflowers, and hairy golden aster (Chrysopsis villosa), blanket flower (Gaillardia aristata), and asters (including Aster pansus and A. falcatus) are common late-season wildflowers. Forested sections of Mt Sentinel are dominated by Ponderosa pine (Pinus ponderosa) and Douglas fir (Pseudotsuga menzesii). Forested sections of Fort Missoula are dominated by old growth black cottonwood (Populus trichocarpa).

1.4 Vegetation management objectives

Objectives for the above named areas are to:

1. Maintain and restore the structure and function of native plant communities.

2. Maintain and restore native wildlife habitat.

3. Develop and maintain native plant communities that are resistant to non-native plant invasion.

4. Preserve and maintain the land's aesthetic values.

5. Comply with the Montana Code Annotated Section 7-22-2151, which requires that municipalities within weed control districts enter into cooperative noxious weed management agreements with the respective weed boards.

Thus, goals include:

1. Reduced cover by noxious weeds.

2. Increased cover by native plant species

3. Maintain a clean, healthy environment with diverse native plant assemblages.

While attempting to reach these goals, UM will take a long-term, integrated approach. It will strive to reduce herbicide use over the long term while making measurable progress toward vegetation management goals over the short term. Reliance solely upon broadleaf herbicides without additional tools would be expensive, increase health concerns, and ultimately lead to losses in native broadleaf diversity. An aggressive, creative, and fully integrated management approach is needed.

1.5 Vegetation management strategy: an ecological and social framework

This management plan exists within the context of an overall strategic ecological framework that addresses social and political concerns as well as public and environmental health effects of both weed treatments and noxious weeds themselves in the Missoula Valley. The plan is based on current understanding of the biology and ecology of Missoula Valley grassland species, as well as social experiences during the past 6 years of weed control implementation in these natural areas. Education and participation remain important components of the vegetation management plan.

All plant communities are in a constant state of flux, and the plant communities of the Missoula Valley are constantly changing in response to disturbances, annual variation in weather, abundance of pollinators and herbivores, spread of exotic species, and many other factors. Changes in plant community structure in response to these factors depend largely on plant site availability (or "niche" availability, usually caused by a disturbance ) species availability, and species performance.

Sheley et al. (1996) provide a helpful framework for restoration of noxious weed-infested areas based on these three points. The framework is called "successional weed management (Figure 3). Weed control creates a vacant niche (site availability). A desirable species that is able to establish and grow in an acceptable time frame (performance) must be either already present, or introduced via seeding to occupy that niche (species availability). If the availability and performance of desirable species are not addressed, then it will not be possible to achieve plant community succession in a direction that addresses the management goals of restoration and preservation of native species. In other words, revegetation must be considered along with removal of noxious weeds.

2.0 Elements of the management plan

The University of Montana will adopt an integrated weed management plan in order to reach the stated objectives. Tools will include herbicides, manual removal, burning, biological control agents, mulching, grazing, seeding, education, and monitoring, implemented within a framework of setting long-term objectives and cooperation with local agencies and private landowners. Each of these tools must be considered in the context of specific goals and site conditions (soils, proximity to water, steepness, aspect, biology of target species, size of weed infestation, etc.).

It is important to emphasize that the rehabilitation of these lands will be a long-term process, without quick fixes or simple prescriptions. This plan is intended to be dynamic, and incorporate adaptive management in a document that allows for incorporation of new technologies and evolving understanding of these biological systems. Monitoring and analysis of results are crucial, because they allow incorporation of experience to future efforts.

This weed management plan consists of eleven basic elements:

Setting objectives
Prevention of new infestations
Early detection
Education
Mapping
Baseline monitoring
Control actions
Revegetation with native species
Response monitoring
Re-evaluation
Cooperation with neighboring landowners (public and private)

2.1 Setting objectives.

General objectives for these natural areas were identified in section 1.4, and include reduction in noxious weed cover and increase in native plant cover. More detail for each property is included in the appropriate section of this Plan.

2.2 Prevention of new infestations.

The single most effective method of managing noxious weeds is to prevent them from establishing. The success of this method will depend on public education on ecology of weeds, their spread, and their impacts to biodiversity and wildlife habitat. Success also requires keeping field equipment and vehicles clean of weed seeds and treating existing infestations so that they do not spread into new areas.

2.3 Early Detection.

Following prevention, early detection is important for effective noxious weed control. When infestations are small, there are far more options for successful control, and control at this stage is less expensive. For example, if a patch of 10-100 white top (Cardaria draba) plants is found, they can be hand pulled and checked annually for follow up treatments. If the plants are not detected until they densely occupy several acres, comprising a population of several thousand plants, successful control is unlikely without the use of herbicides, including follow up treatments for several years.

2.4 Education.

Education is necessary for the general public, for decision-makers, and for professional land managers. Employees responsible for implementing the vegetation management plan should receive annual training to acquire and implement new technologies as they become available and share experiences with other land managers.

As with all long-term ecological problems, resolution is unlikely unless the public is educated and aware of the problem and possible solutions. If people do not appreciate negative impacts of noxious weeds on the regional ecology, it is very unlikely that there will be "emotional" or financial support for the objectives outlined in this vegetation management plan. Passage of a mill levy in Missoula County in 2000 specifically for weed management indicates that the community is supportive of vegetation management. However, ongoing education efforts are needed.

Public education should include interpretive and informative signage at trailheads, field trips for adults and schools, press releases, Internet resources, and public presentations. Educated users of UM natural areas are also an important source of early detection for new infestations of weeds. In the last several years, UM has partnered with fellow agency and non profit groups to create Prairie Keepers, an umbrella volunteer group dedicated to stewardship of Missoula’s native plant communities. This has turned out to be a very effective outreach tool, and UM will continue participation in this program. The Prairie Keepers website is http://prairiekeepers.dbs.umt.edu

2.5 Mapping.

Weeds should be mapped at regular intervals (such as every five years) so that progress can be assessed and strategies developed. The Montana State Weed Mapping Guidelines should be used (see Cooksey and Sheley 1998) and managed with Arc View geographic information systems software. It is beyond the scope of this document to outline mapping procedures in detail.

The weeds of Mt Sentinel were last mapped in 2001. A few sample maps are included in appendix A.

2.6 Baseline monitoring.

Before management activities are initiated, pre-treatment data should be collected from established monitoring sections so that changes over time can be determined. This is a way of measuring progress at a finer scale than at the mapping scale. Permanent monitoring transects or plots should be established at each property.

UM has conducted vegetation monitoring along the M trail in 1999-2004, and will again monitor those transects in spring 2006. Results of this monitoring were published in Ecological Restoration (2005).

2.7 Control actions.

There are several actions that can be taken to control weeds, and each is considered on a site-by-site basis in this report.

Hand pulling. This is a useful method for dealing with small infestations, or infestations within a limited area. Hand pulling is most effective on tap rooted weed species such as knapweed or dyer's woad and least effective on rhizomatous species such as leafy spurge and St Johnswort. Since it is very time consuming when used to control more than a few plants, it is best accomplished with unpaid labor. Sources of weed pulling labor for Mt. Sentinel and Fort Missoula have been community volunteer groups, school groups, youth offenders working off community service sentences, students participating in the Flagship Program, Montana Conservation Corps, Americorps National Community Civilian Corps, and public volunteer days on Earth Day (April) and National Public Lands Day (September). UM will continue working with these groups to maximize pulling as a management tool. Upcoming weed pulls and other stewardship events are listed at the Prairie Keepers website, http://prairiekeepers.dbs.umt.edu

Pros: Non-toxic (excluding allergic reactions in many people), fun (for some people), results in education, can be very selective in the appropriate conditions.

Cons: Can result in disturbed bare ground (that provides niches for more weeds) and trampling of non-target species, not fun (for some people), labor intensive.

Mowing. This is a useful tool to reduce flowering and seed set in noxious weeds, and may be helpful in concert with other tools, especially biocontrols and herbicide application.

Pros: Non-toxic (excluding CO emissions), reduces seed production.

Cons: Non-specific (can be detrimental to non-target species), doesn't kill any of the weed species, just suppresses them.

Biological control agents. Biocontrols are insects that are the "natural enemies" of the pest plants. These are located in the home range of the invading plant, screened for specificity, and eventually released onto infestations. The goal of biocontrol programs is to reduce the pest plant density to an acceptable level, so that desirable vegetation can respond positively. Biocontrol is not intended to eradicate noxious weeds.

Biocontrols are appropriate for addressing long-term control of large, established weed infestations of which there is little chance for eradication. Biocontrols are not appropriate for small, new infestations that can often be eliminated via hand pulling or herbicide application

Although biocontrols are often promoted as possessing a lower environmental risk than herbicides, it is important to note that there is a tremendous degree of uncertainty involved in the release of exotic insects. Although insects are screened to ensure that they do no attack non-target plants, there is no way to screen for the impact on trophic level interactions, i.e. interactions between different levels in the food chain such as between plants and herbivores or insects and birds. An example of biocontrols influencing trophic level interactions in western Montana is the relationship between the knapweed gall fly (Urophora sp.) and deer mice (Pearson et al.1999). Establishment of Urophora gall flies has altered deer mouse diets and habitat selection by effecting changes in foraging strategies, and could lead to measurable increases in deer mouse populations.

Many biocontrol agents are available to UM at no cost through the Missoula County Weed District or by having staff collect insects from local sites.. Releases of biocontrol agents should be coordinated with the Weed District.

The following insects are currently available or already established for the following noxious weeds in Missoula County:

Target weed	Biocontrol insects	Establishment stage
Leafy spurge	Aphthona species (flea beetle)	Well established
	H. euphorbia (Spurge hawkmoth)	Well established
Spotted knapweed	Cyphocleonus (root weevils)	Still establishing
	Larinus minutus (flower head weevils)	Well established
	Urophora spp. (seed head gall flies)	Well established
	Agapeta zoegana (root moths)	Well established
Dalmatian toadflax	Gymnatron	Still establishing (difficult)
	Brachypterolus spp. (Ovary feeding beetles)	Established on Mt Sentinel
	Calophasia	Difficult to establish (climate, and disease)
	Mecinus janthenus	Still establishing

Pros: Non-toxic, offers potential of long-term control after release, most host specific of all control methods

Cons: Some degree of ecological risk, takes a long time to show results

Grazing. Grazing with sheep and goats has been well documented to successful control noxious weeds, especially spotted knapweed and leafy spurge (Olsen 1999). Grazing will not eradicate weeds, but can help maintain them at acceptable levels, and can improve the efficacy of herbicide treatments. Two separate goals can be achieved through a grazing program: weeds are controlled and the sheep rancher makes an income.

In previous years, UM worked with a local sheep rancher who provided sheep and their transportation for no charge. UM grazed about 10 sheep at a time adjacent to the M trail for a few weeks each year in 1998, 1999, and 2000, and grazed up to 30 sheep at a time at Fort Missoula. The use of sheep at both sites has proven challenging due to conflicts with domestic dogs, and with the UM Geology Department. Dogs have killed several sheep and harassed many more.

Pros: Non-toxic, can cover a large area, can help reduce fire fuels

Cons: Dogs must be leashed during grazing season, possible erosion if animals are mismanaged, labor intensive, non-specific and can thus negatively impact some native plant species

Herbicides. Tordon 22k; (active ingredient picloram) is the only herbicide that has been used in the last 7 years on UM owned portions of Mt Sentinel. Tordon 22k and RoundUp (Active ingredient glyphosate) have been used at Fort Missoula.

Beginning in 2006, UM will employ Tordon 22k, Round Up, Plateau, and metasulfuron for weed management at either location, depending on conditions. Table 1 discusses of each of these herbicides. Herbicides will be ground applied by truck, ATV, pack stock, or backpack by licensed/certified contractors or staff. Product labels are available on line, and can be accessed using the web addresses listed in Table 2.

This management plan will not rely solely on applications of broadleaf herbicides. Such an approach would ultimately result in loss of native forb diversity. This plan advocates complementing herbicide treatments with biological controls, mechanical approaches, grazing, re-vegetation, and cultural methods to "spread out" the treatment interval and reduce reliance on herbicides. Additionally, spot treatments are preferred to broadcast treatments when feasible, also in order to minimize impacts to non-target species.

Adequate public notification will accompany herbicide applications, including visible signage at trailheads and at the site of application, and at minimum compliance with the City of Missoula pesticide notification ordinance. Areas to be treated are signed at least 24 hours in advance and 48 hours after. UM will also issue press releases prior to spraying.

Pros: High degree of invasive plant control in a short time period, longer interval between treatment than other methods, can be applied to more acres in a shorter time interval than other methods, minimal ground disturbance and erosion.

Cons: Varying toxicity, varying degree of impact to non-target vegetation (native broadleaf plants) depending on rate of application and product used.

Fire. Fire by itself is will not control any species of noxious weed addressed in this plan (Sheley and Petroff 1999). However fire has been shown to improve efficacy of herbicide treatments, especially on spotted knapweed (Sheley et al. 1999a) and leafy spurge (Lajeunesse et al. 1999). The removal of dead standing stems by burning allows for better contact of herbicides with live plant tissue, and improves uptake. Local native plant communities have evolved with fire (Antos et al. 1983), and therefore controlled burns within uninfested areas of grasslands may help to promote native plant vigor and discourage weed infestation (Sheley et al. 1999b)

There is also anecdotal evidence that spring burns can control cheatgrass. UM should try this approach. This will require some additional staff training and monitoring. In a range situation. While burns may kill off annuals like cheatgrass, range fires are typically not hot enough to kill perennial parent plants, and therefore the perennial plants (including weeds) that are present at a site before the burn will still exist after the burn.

Pros: May improve efficiency of herbicide treatments.

Cons: Can promote weed growth, air pollution, risk of fire escaping.

2.8 Revegetation with native species

Depending on the severity of the noxious weed infestation at a given site, revegetation may or may not be required. Typically, revegetation is recommended if the cover by desirable species is less than 30% at the time of weed control (G. Marks and R. Sheley, MSU Extension Service, personal communications).

To preserve genetic diversity and ensure that the plants used in revegetation projects are adapted for local conditions, it is important to use local ecotypes of native species. Ecotypes are populations of a plant species that are genetically adapted for a given set of conditions. Current literature indicates that the use of appropriate ecotypes can significantly improves restoration project success (Knapp and Rice 1996). This is a greater factor for some species than others (Lesica and Allendorf 1999) depending on their reproductive system, genetic variation among and within populations, and the degree of disturbance at the restoration site.

Choices of seed sources are listed below, beginning with the most desirable options and ending with the least desirable.

Local ecotypes of native species (from sources less than 100 miles from Missoula)
Similar ecotypes of native species (from sources over 100 miles from Missoula, but a similar region)
Commercial-source seeds of native species
Commercial- source, cultivars of native species
Native species that are not found on the project areas
Non-native species that do not reproduce beyond the first growing season.
Non-native, persistent species should never be used because they do not promote the management goals of restoring and preserving native species.

2.9 Response monitoring.

Data from permanent monitoring sections established for baseline monitoring (section 2.6) should be collected after control treatments have been applied. This will allow an evaluation of the changes in vegetation in response to weed control efforts.

2.10 Review and Re-evaluation.

Re-evaluation allows for adaptive management, in which managers learn from past experience what is working and what is not, so that future actions can be adapted accordingly. If it doesn't work- don't keep doing it! This plan will be reviewed annually by the UM IPM Committee and every six years by the Missoula County Weed District.

2.11 Cooperation

Communication and active cooperation with neighboring landowners and local agencies will ensure maximum efficiency of vegetation management efforts.

map of Mount Sentinel

3.0 Mount Sentinel

3.1 Historic condition

3.1.1 Vegetation

Mt. Sentinel was historically dominated by bluebunch wheatgrass, Idaho fescue, and rough fescue grasslands. Ravines had more mesic and shrubby plant communities, including serviceberry, chokecherry, and mountain maple. Aerial photos from the mid 1900s indicate that cover by woody species was much lower than the current condition.

3.1.2 Land use

The west face of Mt Sentinel is important habitat for many species of wildlife, including deer, grouse, and mountain lions. The entire west face of the mountain may have been used by Native Americans for grazing livestock. Also, the south end of Mt Sentinel which was recently owned by the Cox family but is now owned by the City of Missoula, was grazed by horses until 2000. However oral accounts indicate that no horse grazing has taken place on the UM portion of Mt Sentinel in recent history.

3. 2 Current condition

3.2.1 Vegetation

The native prairies have been invaded by many species of noxious weeds, including spotted knapweed, leafy spurge, Dalmatian toadflax, sulfur cinquefoil, St Johnswort, and dyer’s woad. A few individuals of hound’s tongue and common tansy have also been found. Musk thistle sometimes occurs above the fire road.

In some areas, treatment of noxious weeds has resulted in increased cover by native plants. In other areas, especially below the hang glider takeoff, sprayed areas filled in almost exclusively with cheat grass and tumble mustard.

3.2.2 Land use

The land was given to the University for recreational use, as an outdoor laboratory and for preservation of open space and scenic beauty. Walking and wildlife viewing are the primary uses of the area by humans. University students, faculty and staff conduct ecology research on Mt Sentinel. The M trail is the one of the most popular hiking trails in the Missoula Valley.

An additional interesting aspect of this property is the existence of the visible beach lines from Glacial Lake Missoula (see Alt and Hyndman for more information). Several faculty members at the University of Montana Department of Geology are involved in a regional project to promote this as a site for study and appreciation of Glacial Lake Missoula. They have expressed concern that grazing could impair the beach lines and have requested that no grazing take place in those areas.

3.3 Management recommendations

These recommendations are made with the end goal of restoring the property to native grassland. Recommendations are organized by weed species, with a geographical discussion as appropriate.

1. Continue efforts to prevent new infestations of weeds not currently found on Mt Sentinel, especially hounds tongue. There is a small patch of hound’s tongue at the Evans Street trailhead, which is partly private property and partly UM property. This site will be visited each year for handpulling and spot spraying with metasulfuron (less than 1/100^th of an acre is infested). Additionally, increased efforts will be made to communicate with the homeowners on either side of the trailhead, so that they can help control hound’s tongue on their own land.

2. Continue efforts to control the new invader dyer’s woad (Isatis tinctoria) via hand pulling, but increase efforts to also include spot spraying with metasulfuron. For the last 12 years, the Montana Native Plant Society has hosted two dyer’s woad pulls in cooperation with UM, the Weed District, the State Dyer’s Woad Project and the public. These efforts have been very successful at keeping this new invader under control. Unfortunately, wet weather in 2005 led to a sharp increase in woad plants on Mt Sentinel and around the state (Monica Pokorny, personal communication). Also, a previously undiscovered patch of dyer’s woad was found near the Evans Street trailhead. The State Dyer’s Woad Project has asked UM to increase handpulling efforts and make some spot applications of metasulfuron to get the infestation under control once again (Appendix A).

For 2006, the hand pulls are scheduled for May xx, June xx and xx. All are welcome to attend.

3. Of weeds that are already established on the mountain, St Johnswort is the least widespread and should therefore be a high priority for control. In 1999, there were about 2-3 acres of St Johnswort about the fire road, directly above the “M.” After 2 summers of attempting to control this patch with handpulling UM sprayed the area with 1.5 quarts of Tordon per acre. This was extremely effective, and since 2001, only minor spot spraying and opportunistic pulling has been needed to keep it under control.

4. The next most widespread weed is leafy spurge. There are currently less than 10 acres (closer to 5 acres) of leafy spurge on the mountain, and it is important to contain it and reduce it while looking for new outbreaks.

Over the past 6 years, we have made good progress in controlling leafy spurge with backpack and horseback applications of Tordon at a rate of 1.5 quarts per acre. This work is labor intensive, but relying on ground application is an effective way to do this, because as workers are hiking from known patch to known patch, they are also looking for new infestations. Leafy spurge spraying takes place at the same time of year and with the same tank mixture as Dalmatian toadflax.

The problem with spraying leafy spurge with Tordon in areas that are heavily degraded is that this frequently leaves the site open for cheatgrass (Bromus tectorum) to establish. Cheatgrass prevents desirable native species from establishing. UM should begin using Plateau on a limited trial basis for leafy spurge control. Plateau can only be applied in the fall (not the spring) because it can damage perennial bunchgrasses when they are actively growing. When applied in the fall, when perennial bunchgrasses are dormant, it can control leafy spurge and cheatgrass. Recommended application rate is 10 oz per acre.
Leafy spurge flea beetles and spurge hawk moths are well established on Mt Sentinel. Hand pulling is not effective on this species because of its freakishly long root system, which can extend more than 20 feet below ground.

5. Next in priority for control is Dalmatian toadflax. This plant occurs all over the mountain, and is highest in density on the south facing slope below the hang glider take off and along the base of the mountain. Control of this plant has to be systematic and ongoing, with the goal of containment in some areas, and suppression/eradication in other areas. Broadcast applications are necessary in some areas, while most of the mountain just needs consistent spot treatments.

The preferred herbicide for toadflax has been Tordon at a rate of 1.5 quarts per acre, applied either in spring or fall on growing tissue. This has worked well in some years and less well in other years. This inconsistent control has been observed by other managers in Montana (Celestine Duncan, Jerry Marks, Andy Kulla, personal communication). UM will begin using Plateau in the autumn on a trial basis (see discussion above for leafy spurge).

Ovary feeding beetles are well established on Mt Sentinel. Stem miners have been released, but as yet are not well established (cite Jeff Copeland’s report).

6. Spotted knapweed, and to a much lesser degree, sulfur cinquefoil, are scattered across the mountain. This is one of the easier noxious weeds to control, except for its broad distribution. It can be successfully hand pulled, sprayed with Tordon at 1 pint per acre, and there are several biocontrol agents established to (hopefully) provide additional control.

Spotted knapweed has been and will continue to be controlled by hand pulling along the M trail, via a successful community outreach program call “Adopt a Switchback.” Each of the 13 switchbacks has been adopted by a different school or community group for long term control. The biggest push of the year comes each April in the form of 150 8^th graders from Washington Middle School, who ride their bikes to the M trail and engage in 2 hours community service weed pulling. The Adopt a Switchback volunteers each take 6-8 students to their switchbacks and have a weed pulling, nature lesson. This project has happened each year since 2000. Washington School and UM staff are happy with this arrangement, and plan to continue it.

In other parts on the mountain (outside of the M trail “special management area”) knapweed is treated with 1 pint per acre Tordon, broadcast or spot sprayed in the fall. Fall application is intended to reduce negative impacts to non-target species. Occasionally, spring crews who are out spot spraying leafy spurge or toadflax will encounter a large patch of knapweed, and these may be opportunistically sprayed as well.

Sulfur cinquefoil is opportunistically sprayed or pulled while knapweed is sprayed or pulled. It is conveniently controlled at the same rate and at the same time as spotted knapweed, although it is more difficult to pull.

3.4 M Trail Special Management Area

When the original Mt Sentinel Weed Management Plan was developed in 1997, there was a lot of public concern about use of herbicides in an area that receives such high recreational use. In response to these concerns, UM designated the M trail and a 20 foot buffer zone on either side as a no spray zone. This zone did allow for spot treatment of leafy spurge and Dalmatian toadflax, so that these hard to control invaders could be addressed before they spread. No changes to this policy are recommended at this time. Hand pulling the knapweed has worked well, and spot treatments of spurge and toadflax are becoming fewer and fewer each year. Native prairies are recovering very well along the M trail under this regime.
4.0 Fort Missoula Riparian Area (Sleven’s Island)

4.1 Historic condition

4.1.1 Vegetation

Aerial photos from the mid 1900s show a black cottonwood forest, with an open mid- and under story. The mid-story would have consisted of mixed deciduous shrubs including chokecherry, hawthorn, and dogwood. Native grasses and sedges, snowberry and false Solomon's seal would dominate the under story.

4.1.2 Land use

Not sure. Apparently people planted some ornamental shrubs in this area during the Army days. Also, this used to be an actual island, but the upstream end of the island was filled in with concrete blocks. (Not sure if this was by the Army or UM).

4.2 Current condition

4.2.1 Vegetation

Deciduous trees and shrubs dominate Sleven's Island. The overstory is composed of old-growth black cottonwood, box elder, and weeping willow, with various shrubs, especially chokecherry, hawthorn, Wood’s rose and currants in the mid-story. Much of the island is dominated by the "box-elder/chokecherry forest" community type according to Hansen et al. (1995). However, since box elder is not native west of the continental divide, it is an artificial assemblage.

Exotic grasses and forbs dominate the understory, including several the noxious weeds leafy spurge, common tansy, hounds tongue, and Canada thistle. Exotic grasses present at the site include quackgrass (Agropyron repens), Kentucky bluegrass (Poa pratense) and reed canary grass (Phalaris arundinacae).

UM graduate student Katherine Warner prepared a GIS map of the community types on the island in summer of 1999, and collected species and coverage data from 10 permanent transects along the length of the island. This data is available in Warner’s thesis, and can be used as the baseline for future monitoring projects.

4.2.2 Land use

The area is used hikers, birdwatchers, and anglers. The UM Native American Studies Department has several sweat lodges at the east end to the island, and these need to be treated with respect since they are religious sites..

4.3 Management recommendations

The University has identified native riparian species as the desired vegetation, and intends to restore the plant communities as much as possible to their historic condition. The black cottonwood stand that currently exists on the Island does not appear to be regenerating. Since the flood regime at the site is different now from when it was a true island, the natural processes maintaining the historic community type may not be functioning. However, removing non-native species and replacing them with natives will be an important step in returning natural ecological structure to Sleven's Island.

Prioritization for control

The first priority, from both legal and ecological standpoints, is the control of noxious weeds. Leafy spurge, common tansy, and hound's tongue are all tremendous threats to diversity of riparian areas. Of these, leafy spurge is the highest priority for control, since it has greatest potential to spread to upland areas. Hound's tongue is also a high priority since it is least widespread, and therefore most easily addressed at this point.

Leafy spurge. Leafy spurge is a deep rooted perennial that reproduces via seeds and rhizomes. Leafy spurge management in riparian areas is problematic. This plant is most effectively controlled with Tordon, but that herbicide is not approved for use near water. 2,4-d is approved for riparian use but does not provide effective control of leafy spurge. Land managers in our area have had good success treating leafy spurge in riparian areas with Rodeo (Wessner, Newman, Sturdevant, personal communications), the formulation of glyphosate that is approved for use near water. This is an option at Sleven’s Island, but applicators must be aware that Rodeo is a non-selective herbicide and will kill or damage any plant that it touches. Leafy spurge flea beetles are established at this location. Hand pulling is not effective.

Hound’s tongue. Hound’s tongue is taprooted biennial that reproduces by seeds. The seeds are covered with Velcro-like hairs that allow them to stick to clothing, hair, passing animals, and so on. Since the summer of 2000, UM has worked with local youth groups to hand pull hound’s tongue plants. Early years focused on removing old “skeletons” with seeds attached. After 2 years or so, the vast majority of pulled plants had not yet gone to seed. The abundance of hound’s tongue at the site has decreased substantially (Marler personal observation). In 2004 and 2005, workers began making spot applications of glyphosate on basal rosettes of hound’s tongue, and this has helped as well. I recommend continuing this combination of handpulling and spot spraying, because it has worked very well so far, and efforts get easier each year. Missoula Youth Homes have been champions of hound’s tongue pulling.

Canada thistle.

Common tansy. Common tansy is a deep rooted and long lived perennial that reproduces via seeds and rhizomes.

Following noxious weed control, removal of caragana, weeping willow and box elder are second in priority. If labor resources are not sufficient for removal of these trees this year, some trees may be girdled. This procedure, which involves cutting through the trees' cambium, will kill the tree slowly. Native species can be planted beneath girdled trees.

The exotic grasses are not a priority for control since they are so widespread, difficult to control, and do not legally require attention. However, where shrubs and forbs are removed mechanically or chemically, aggressive re-vegetation with native species should be done to prevent colonization by the exotic graminoids.

Control methods for noxious weeds in riparian areas, by species.

Species	Proposed management methods
Leafy spurge	Mowing, biocontrol, Plateau trials in study plots, Krenite trials
Spotted knapweed	Pulling, opportunistic spraying as Canada thistle is treated
Canada thistle	Mowing and spraying (with glyphosate)
Hound's tongue	Hand pulling, and spot spray with metasulfuron if needed
Common tansy	Mowing, digging, and spray with glyphosate pr metasulfuron

4.3.2 Recommended species for revegetation

Willow species, dogwoods and cottonwoods can be propagated from cuttings on-site or nearby. Those species, as well as wood's rose, chokecherry, and elderberry are commercially available as well. In moist areas, Nebraska sedge (Carex nebrascensis) could be seeded to slow invasion by exotic graminoids.

5.0 Fort Missoula Upland Areas

5.1 Current condition

5.1.1 Vegetation

All of the UM property at Fort Missoula spotted knapweed. Leafy spurge, sulfur cinquefoil, henbane, baby's breath, quack grass, and smooth brome are also present in very low densities, and present management problems. Small stands of native grasses and red fescue, which was probably seeded during Army days, persist on a limited scale. Red fescue and quack grass are far more common than native grasses. A working plant species list is given in Appendix A.

Siberian elms were planted as "street trees" around the Fort. This species grows quickly, produces abundant seeds, and spreads quickly. The original adult trees have become unhealthy, and the Northern Rocky Heritage Center is actively removing them. Because of their poor health and ability to reproduce rapidly, UM should also remove its elms.

5.2 Desired vegetation

Native dominated bunchgrass prairie is the desired community type. The land has been impacted so severely that it is impossible without historic records to determine the original community composition. A return to the actual historical plant community is unlikely, and an assemblage dominated by native species but including many non-native species is acceptable.

UM has two native plant demonstration projects at Fort Missoula, near the water tower. One is the Nature Adventure Garden, and the other is the Fort Missoula Native Prairie. Both are outdoor classrooms, and both serve as seed sources for restoration of native prairies on Mount Sentinel. Both projects are collaborative efforts between UM, Montana Natural History Center, the Montana Native Plant Society and the Missoula County Weed District.

The Nature Adventure Garden demonstrates a variety of “garden worthy” native plants, ranging from dry prairie species to berry producing shrubs to wetland wildflowers. Several hundred people tour the garden each year via workshops offered through the UM Natural Areas Restoration Program and the Montana Natural History Center. It demonstrates that homeowners can use native plants to meet goals of water conservation, reduced pesticide use, and support of birds and butterflies. A large portion of the maintenance of the garden is done via volunteers.

Immediately adjacent to the Nature Adventure Garden is the Fort Missoula Native Prairie. This area demonstrates a native prairie, as opposed to arid landscaping. It is maintained with the same partnerships as the Garden.

5.3 Management recommendations for upland areas

5.3.1 Prioritization for control

Currently, all the upland areas at Fort Missoula are mowed annually, with the exception of the demonstration areas discussed above and research plots located near the water tower. Mowing should continue, and should be combined with occasional broadcast application of Tordon, along with spot treatments as needed and hand pulling as appropriate.

Small infestations should receive attention first. A few small patches of leafy spurge need annual visits to keep them from spreading (these patches have become much smaller since 2000). Individual plants of henbane and baby's breath appear occasionally, and are be manually removed while there are still fewer than 20 individuals of each. This needs to continue.

The remaining acreage of knapweed infestations should be treated systematically with ground applied Tordon 22K (1 pint per acre), combined with biocontrol releases. Urophora gall flies, Larinus weevils, and Cyphocleonus have all been released at Fort Missoula. The two former species are well established.

Since there is very little native or desired vegetation present at the site, treatment must be paired with revegetation. Since seed source orchards are present at the Fort, out-of-area cultivars or varieties of native species should not be used. At this time, that includes bluebunch wheatgrass, prairie junegrass, Sandberg's bluegrass, yarrow, and sky blue penstemon.

4.0 Literature cited

Alt, D. and D. Hyndman. 1986. Roadside Geology of Montana. Mountain Press Publishing Company. Missoula, MT USA.

Cooksey, D. and R. Sheley. 1998. Mapping noxious weeds in Montana. Montana State University Extension Service, Extension Bulletin 148. Bozeman, MT.

Goodson, N. 1982. Effects of domestic sheep grazing on bighorn sheep populations: a review. Proceedings of the biennial symposium of the Northern Wild Sheep and Goat Council 3:287-313.

Grilz P., and J. Romo. 1995. Management considerations for controlling smooth brome in fescue prairie. Natural Areas Journal 15:148-156.

Knapp, E. and K. Rice. 1996. Genetic structure and gene flow in Elymus glaucus (blue wildrye): Implications for native grassland restoration. Restoration Ecology 4:1-10.

Lajuenesse, in R. Sheley and J. Petroff (eds), Biology and Management of Noxious Rangeland Weeds. Oregon State University Press, Corvallis, Oregon.

Lesica, P. and F. Allendorf. 1999. Ecological Genetics and the Restoration of Plant Communities: Mix or Match? Restoration ecology 7:42-50.

Olsen, B. 1999. Grazing and weeds. Pages 85-96 in R. Sheley and J. Petroff (eds), Biology and Management of Noxious Rangeland Weeds. Oregon State University Press, Corvallis, Oregon.

Pearson, D., K. McKelvey, L. Ruggiero. 2000. Non-target effects of an introduced biological control agent on deer mouse ecology.Oecologia122:121-128.

Sheley, R., T. Svejcar, and B. Maxwell. 1996. A theoretical framework for developing successional weed management strategies on rangeland. Weed Technology10:712-720.

Sheley, R. and J. Petroff. 1999. Biology and management of noxious rangeland weeds. Oregon State University Press. Corvallis, Oregon.

Sheley, R, M. Manoukian, and G. Marks. 1999a. Spotted knapweed. Pages 350-361 in R. Sheley and J. Petroff (eds), Biology and Management of Noxious Rangeland Weeds. Oregon State University Press, Corvallis, Oregon.

Sheley, R, J. Jacobs and M. Carpinellia. 1999b. Preventing noxious weed invasion. Pages 69-71 in R. Sheley and J. Petroff (eds), Biology and Management of Noxious Rangeland Weeds. Oregon State University Press, Corvallis, Oregon.

Rice, P. and C. Toney. 1996. Exotic weed infestations and grassland restoration on Mt Sentinel: recommendations for implementing a management plan. Report to the IPM Committee, University of Montana.

Willson, G. and J. Stubbendieck. 1997. Fire effects on four growth stages of smooth brome (Bromus inermis Leyss.) Natural Areas Journal 17:306-312.

Willson, G. and J. Stubbendieck. 2000. A provisional model for smooth brome management in degraded tallgrass prairie. Ecological restoration 18:34-38.

Appendix A. Letter from the State Dyer’s Woad Project requesting that UM increase management efforts for Dyer’s woad.

Appendix B. Public comment received.

At the April 18^th public hearing:

Will Snodgrass; Mr. Snodgrass referred to several different articles and literature he had researched. His main concern is the health risks involved with using pesticides. He referred to the EPA registration process and that it is illegal to say that a pesticide is approved by the EPA. The EPA only registers the pesticide. He said that the IPM plan did not address the health concerns he referred to. He stated that manufacturer’s claims that pesticides are practically non-toxic are false. He recommended that the health risks also be put on posting that is required before applications are made.

Laureen Folsom; Her comments addressed health issues with pesticides being used. She was concerned about the risks for people pulling weeds in areas that had been previously treated with herbicides. She requested a graph of pesticide use in the last ten years. She noted that pesticides can be tracked in to living areas where there is less chance for them to break down naturally. She stated that the EPA does not approve pesticides. Many are registered without being tested. She also stated that Plateau had a high toxicity and the half life of glyphosate was questionable. She stated that Glyphosate had caused damage to lab animals. She also stated that Picloram did kidney and liver damage, and was highly mobile in soil.

Ms. Folsom also read a letter from Tony Tweedale expressing opposition to the use of pesticides.

Jerry Marks; Complimented Marilyn Marler on her work on the IPM plans. He stated that there will always be controversy with any control program. He stated that adaptive management such as stated in the IPM plan was a strong point of the plan

From Kelly Chadwick via email (4/24/06)

The Natural Area Plan looks good. It is educational and integrated.

Trade name	Active ingredient	Target Weeds	Rate	Comments
Tordon 22k	picloram	Spotted knapweed, sulfur cinquefoil	1 pint/acre	Cannot use near surface water or shallow ground water
		Leafy spurge, Dalmatian toadflax	3 pints/acre	See above, plus results are inconsistent on Dalmatian toadflax
Escort	metasulfuron	Dyer’s woad, common tansy	1 oz/acre	These target weeds either do not respond to picloram (woad) or tend to occur near water (tansy)
Plateau	imazipic	Dalmatian toadflax, leafy spurge, cheatgrass	10 oz/acre	Fall application only to avoid damaging native bunchgrasses
Round Up	glyphosate	Cheat grass, quack grass, smooth brome, knapweed	Per label instructions	Non-selective and must be applied very carefully

Name	Web address for label
Tordon 22k	www.umt.edu/sentinel/tordon_label.pdf
Escort	www.umt.edu/sentinel/escort_label.pdf
Plateau	www.umt.edu/sentinel/plateau_label.pdf
Round Up	www.umt.edu/sentinel/roundup_label.pdf