Elseroad 2007 BCA Restoration Plan_June 2007.pdf

BOARDMAN CONSERVATION AREA

RESTORATION PLAN

Adrien Elseroad Plant Ecologist

The Nature Conservancy 821 SE 14th Avenue Portland, OR 97214

June 2007

Boardman Conservation Area Restoration Plan- June 2007 2

June 26, 2007

Attached is the restoration plan for the Boardman Conservation Area, located near Boardman, Oregon. This is currently the final version of the plan but it may be updated on an as-needed basis as restoration priorities and techniques are refined.

Adrien Elseroad

Plant Ecologist The Nature Conservancy 821 SE 14th Avenue Portland, OR 97214

Boardman Conservation Area Restoration Plan- June 2007 3

TABLE OF CONTENTS

INTRODUCTION

BACKGROUND

SCOPE OF PLAN

EXISTING CONDITIONS

CLIMATE

SOILS

VEGETATION

CAUSES OF DEGRADATION

RESTORATION SITES

TYPES OF RESTORATION SITES

Type 1: Sites dominated by non-native annual plant species Type 2: Sites dominated by non-native annual plant species with a low diversity of native perennial species8 Type 3: Sites dominated by native shrubs with a low diversity of native herbaceous species Type 4: Sites dominated by a diverse array of native plant species with high cheatgrass densities

RESTORATION OBJECTIVES

DESIRED FUTURE CONDITIONS

RESTORATION OBJECTIVES AND STRATEGIES BY SITE TYPE

PRIORITIZATION OF RESTORATION SITES

RESTORATION POTENTIAL

RESTORATION METHODS

PLANT MATERIALS

Planting lists Genetic considerations On-site seed collection Seed cleaning Grass seed increasing Purchasing commercial seed Seeding rates Calculating seed mixtures Transplanting

SITE PREPARATION

Mechanical Herbicide Agricultural crops Fire Grazing

PLANTING METHODS

Drill seeding Broadcast seeding Transplanting Timing of planting

POST-PLANTING WEED CONTROL

RESTORATION SEQUENCING FOR EACH RESTORATION SITE

TYPE 1

TYPE 2

TYPE 3

TYPE 4

Boardman Conservation Area Restoration Plan- June 2007 4

RESEARCH AND MONITORING

PILOT RESEARCH PROJECTS

MONITORING

ADAPTIVE MANAGEMENT

RESTORATION SCHEDULE

BUDGET

RESTORATION COST PER ACRE

EQUIPMENT/PLANT MATERIAL/STAFF NEEDS

Equipment needs Plant material needs Staff needs

SOURCES OF FUNDING

Current funding sources Future funding sources Role of potential partners

LITERATURE CITED

TABLES AND FIGURES

TABLE 1. BOARDMAN CONSERVATION AREA RESTORATION SITE DESCRIPTIONS

TABLE 2. PLANTING LISTS FOR GRASSLAND PLANT COMMUNITIES ON THE BOARDMAN CONSERVATION

AREA FOR HIGH, MEDIUM, AND LOW DIVERSITY PLANTINGS

TABLE 3. PLANTING LISTS FOR SHRUBLAND PLANT COMMUNITIES ON THE BOARDMAN CONSERVATION

AREA FOR HIGH, MEDIUM, AND LOW DIVERSITY PLANTINGS

TABLE 4. SEED COLLECTION AND PLANTING INFORMATION FOR BOARDMAN CONSERVATION AREA

NATIVE SPECIES

TABLE 5. BOARDMAN CONSERVATION AREA RESTORATION COSTS PER ACRE

FIGURE 1. LOCATION OF THE BOARDMAN CONSERVATION AREA

FIGURE 2. TYPES OF RESTORATION SITES ON THE BOARDMAN CONSERVATION AREA

FIGURE 3. PRIORITIZATION OF RESTORATION SITES ON THE BOARDMAN CONSERVATION AREA

Boardman Conservation Area Restoration Plan- June 2007 5

INTRODUCTION

Background

The Boardman Grasslands are one of Oregon’s most important native grasslands. Located in the Columbia Plateau ecoregion in north-central Oregon, the 74,000 acre Boardman Grasslands were identified as a Conservation Opportunity Area proposed for focused conservation investments because they contain some of the best remaining grassland and shrub-steppe in the ecoregion (Oregon Department of Fish and Wildlife 2005).

Included within the Boardman Grasslands is the 22,600 acre Boardman Conservation

Area, owned by Threemile Canyon Farms (Figure 1). The Nature Conservancy has managed this site since 2001 with the objective of protecting and improving native plant communities and their associated wildlife species (Nelson 2004). The Conservation Area provides habitat for sensitive wildlife including the Washington ground squirrel, a candidate species under the Federal Endangered Species Act and listed species under the Oregon Endangered Species Act; the ferruginous hawk, listed as a species of concern under the Federal Endangered Species Act, and critical under the Oregon Endangered Species Act; the loggerhead shrike, listed as vulnerable under the Oregon Endangered Species Act; and the sage sparrow, listed as critical under the Oregon Endangered Species Act (Oregon Natural Heritage Information Center 2004). The Conservation Area is also included in a Multi-Species Candidate Conservation Agreement with Assurances intended to provide benefits to these at-risk species (David Evans and Associates 2004).

Despite containing some of the highest quality grassland and shrub-steppe in the ecoregion, portions of the Conservation Area have been degraded as a result of non-native plant species invasion. Non-native plant species were identified as one of the greatest threats to the Boardman Grasslands (Kagan et al. 2000), because they replace native vegetation and degrade wildlife habitat. Over much of the Conservation Area, non-native species occur at relatively low densities within otherwise high quality habitats. In these areas, herbicide and hand control that targets the invasive species is an effective method of control. The Nature Conservancy manages an active weed control program to address these conditions (Nelson 2002).

However, portions of the Conservation Area are dominated by cheatgrass (Bromus tectorum) and other non-native invasive species. Once established these species can spread into the adjacent high quality areas, posing a serious threat to the at-risk species and their habitats.

Cheatgrass also alters natural fire regimes by creating more abundant and continuous fine fuels that can result in more intense, larger, and frequent fires. Intense fires that burn through high-quality native habitats can convert a diverse multi-story ecosystem of cryptogams, perennial grasses and forbs, and shrubs to a monoculture of cheatgrass and other invasive species that is difficult to reverse without active restoration.

Proactively restoring degraded portions of the Conservation Area offers an opportunity to reduce the threats that invasive species pose to the adjacent high quality habitats as well as increase available native habitat for threatened species. Successfully restoring degraded habitats will be difficult and costly, thus efforts should employ the best available restoration techniques

Boardman Conservation Area Restoration Plan- June 2007 6 and focus on areas that have the greatest probability of success and will provide the greatest long-term benefit.

In this restoration plan, we aim to: 1) identify priority restoration sites on the Boardman Conservation Area, 2) outline restoration approaches for each site, and 3) identify funding and infrastructure needs for accomplishing these objectives. A secondary objective of this plan is to provide a restoration framework that can be used by our conservation partners and adjacent landowners who seek to restore grassland and shrub-steppe on their lands.

Scope of plan

This restoration plan focuses on the Boardman Conservation Area, although the methods outlined in this plan are applicable to nearby properties owned by the Department of Defense, Portland General Electric (PGE), Threemile Canyon Farms and other private lands within the Lower Willow Creek Weed Management Area and the Boardman Grasslands (Figure 1).

Vegetation maps created in 2002 that characterized the ecological condition of undeveloped lands on PGE and Threemile Canyon Farms (Elseroad 2002) can be used to identify priority restoration sites on these properties. Mapping and evaluation is needed to identify restoration sites on the other properties.

EXISTING CONDITIONS

Climate

The Conservation Area and surroundings are semi-arid with hot, low precipitation summers and relatively cold winters. At the Boardman climate station, annual precipitation from 1971-2005 averaged 8.5 inches (Western Regional Climate Center 2006). Up to half of the precipitation falls in winter as snow, and less than ten percent of the total precipitation occurs during the summer months. Monthly mean temperatures range from 41°F to 90°F, with lows below freezing in the winter and highs in excess of 100°F in the summer (Western Regional Climate Center 2006). Southwesterly winds prevail through most of the year, with winds in excess of 25 miles per hour common from March to July.

Soils

The Conservation Area is characterized by nearly level to rolling terraces that are 400 to

900 feet above sea level. Soils are wind-deposited silt or loess that is underlain by basalt. Three major soil associations occur on the Conservation Area: Quincy-Koehler, Sagehill-Taunton, and Warden (Hosler 1983). The Quincy-Koehler soils, which dominate the northern half of the Conservation Area, are moderately to very deep loamy fine sands that formed over alluvial gravel deposits, a hardpan, or basalt. Further south, the Sagehill-Taunton soils are moderately to very deep fine sandy loams formed over lacustrine silt or hardpan. Warden soils, which dominate the southern portion of the Conservation Area, are very deep silt loams and very fine sandy loams formed over lacustrine silt. All soils on the Conservation Area are well drained to somewhat excessively drained and subject to wind erosion (Hosler 1983).

Boardman Conservation Area Restoration Plan- June 2007 7

Vegetation

The Conservation Area is composed of a mosaic of grassland and shrubland plant communities. Dominant native vegetation includes bunchgrass communities composed of bluebunch wheatgrass (Pseudoroegneria spicata ssp. spicata (Agropyron spicatum)), needle and threadgrass (Hesperostipa (Stipa) comata), and Sandberg’s bluegrass (Poa secunda); and shrubland communities dominated by big sagebrush (Artemisia tridentata), antelope bitterbrush (Purshia tridentata), and western juniper (Juniperus occidentalis). Descriptions of plant communities on the Conservation Area are available in Nelson (2004).

Vegetation maps created in 2002 characterized the ecological condition of the plant communities on the Conservation Area based on the abundance of cheatgrass and other non-native plant species (Elseroad 2002). More than half (53%; 12,300 acres) of the Conservation Area was identified as being in medium low or low condition, dominated by cheatgrass and other non-native annual species, with few native perennial species present. Twenty-eight percent of the Conservation Area (6,400 acres) was rated as being in medium condition, where native perennial bunchgrasses are present but cheatgrass and other non-native annual species are widespread as well. Nineteen percent (4,200 acres) were rated as being in medium-high or high condition, dominated by native perennial bunchgrasses with a well-developed soil crust and low abundance of cheatgrass and other non-native species.

Causes of degradation

Livestock grazing, plowing, and wildfires have all contributed to the spread of non-native plant species on the Conservation Area. The Conservation Area was grazed from at least the early 1900’s to May 2005, and most of the highly degraded areas (those in low ecological condition) were heavily used by cattle. In former cattle holding pastures, corrals, watering sites and salt lick sites, virtually all of the native perennial plant species and soil crusts were eliminated by trampling and heavy grazing. In areas of medium ecological condition, moderate cattle use probably facilitated the spread of non-native species by trampling soil crust and creating patches of bare ground. The largest highly disturbed site on the Conservation Area was plowed for wheat production, abandoned, and then colonized by non-native species. Non-native species probably also colonized following high intensity or multiple wildfires. Some areas may be susceptible to non-native species invasion even in the absence of disturbance. On sandy soils, soil crusts do not develop as extensively as on other soil types, allowing non-native species to invade gaps between native species.

RESTORATION SITES

Types of restoration sites

Restoration sites on the Conservation Area can be classified into the following four categories based on the proportion of non-native species and the native species composition at each site: 1) sites dominated by non-native annual plant species; 2) sites dominated by non-native annual plant species with a low diversity of native perennial species; 3) sites dominated by

Boardman Conservation Area Restoration Plan- June 2007 8 native shrubs (big sagebrush or antelope bitterbrush) with a low diversity of native herbaceous species; and 4) sites dominated by a diverse array of native species but with high cheatgrass densities. There are also a few sites on the Conservation Area (~375 acres) that are dominated by crested wheatgrass (Agropyron cristatum), a non-native forage grass, but these sites are not being considered for restoration because crested wheatgrass is not invasive and thus not posing a threat to the adjacent native communities.

Described below are the four types of restoration sites present on the Conservation Area.

Acreage values provided for each type are rough estimates because values are derived from a coarse-scale vegetation map (Elseroad 2002). Additional ground-truthing is needed to determine the extent of each site.

Type 1: Sites dominated by non-native annual plant species

These sites are dominated by cheatgrass and redstem filaree (Erodium cicutarium), although at some sites invasive annuals such as Russian thistle (Salsola kali), tumblemustard (Sisymbrium altissimum), yellow starthistle (Centaurea solstitialis), medusahead rye (Taeniatherum caput-medusae), and bulbous bluegrass (Poa bulbosa) (an invasive perennial), are abundant as well. A few disturbance-adapted native species may also be present at these sites, including gray rabbitbrush (Chrysothamnus nauseosus), hoary aster (Machaeranthera canescens), matchbrush (Gutierrezia sarothrae), and desert fescue (Vulpia microstachys), as well as native annual forbs. These sites were rated as being in low ecological condition during vegetation mapping in 2002 (Elseroad 2002).

Currently, there are 1,428 acres of Type 1 sites on the Conservation Area (Figure 2). The largest Type 1 site is “The Field”, 720 acres that was plowed and used for wheat production in the early 1990’s and subsequently abandoned. It is currently dominated by cheatgrass, non-native annual forbs (predominately yellow starthistle), and native annual forbs. Many small Type 1 sites are scattered throughout the Conservation Area as well, including former cattle watering and salt lick sites, former cattle holding pastures and corrals, and plowed fire lines.

Type 2: Sites dominated by non-native annual plant species with a low diversity of native perennial species

These sites are similar to Type 1 sites in terms of both non-native and native plant species composition, but they also contain Sandberg’s bluegrass, a native perennial bunchgrass. Other native perennial bunchgrass species and most native perennial forb species are lacking. These sites were rated as being in medium-low ecological condition during vegetation mapping in 2002 (Elseroad 2002). Currently, there are 3,619 acres of Type 2 sites on the Conservation Area (Figure 2).

Type 3: Sites dominated by native shrubs with a low diversity of native herbaceous species

These sites are similar to Type 1 sites but are also dominated by big sagebrush or antelope bitterbrush. At some sites, western juniper is present in low densities. Few native perennial forbs or grasses are present, although in some of the sites dominated by big sagebrush, Boardman Conservation Area Restoration Plan- June 2007 9

Sandberg’s bluegrass is present. These sites were rated as being in medium-low or medium ecological condition during vegetation mapping in 2002 (Elseroad 2002). Currently, there are 5,790 acres of Type 3 sites on the Conservation Area (Figure 2).

Type 4: Sites dominated by a diverse array of native plant species with high cheatgrass densities

These sites are grasslands dominated by native perennial grasses and forbs, or shrublands dominated by either sagebrush or bitterbrush and native perennial grasses and forbs, but also have high densities of cheatgrass and other non-native annual species. Currently, there are 6,263 acres of Type 4 sites on the Conservation Area (Figure 2).

RESTORATION OBJECTIVES

Desired future conditions

In the absence of non-native species and human-caused disturbance, grassland and shrub-steppe systems on the Columbia Plateau are characterized by large perennial bunchgrasses, perennial forbs, cryptogamic soil crust, and an open shrub layer, with a fire return interval ranging from 20-70 years (Payson et al. 2000). Entirely re-creating these pre-disturbance conditions on the Conservation Area will not be possible given that large areas are degraded to some degree, and the fact that restoring these semi-arid systems in the presence of highly competitive non-native species will be difficult.

Rather, restoration efforts should focus on: 1) reducing threats to existing high quality native grassland and shrub-steppe; and 2) increasing the proportion of native-dominated grassland and shrub-steppe by restoring degraded sites to the greatest extent possible. Restoring the dominant grasses, forbs, and shrubs at degraded sites will provide the structure of the pre-disturbance plant communities, increase resistance to further weed invasion, and reduce the risk of high intensity, frequent fires. Over time, if soil disturbance is minimized, cryptogamic soil crusts may slowly recover. Ideally, existing conditions can be improved such that the grassland and shrub-steppe systems can be self-sustaining with minimal management inputs and their ecosystem functions can be maintained within their historic range of variability.

Restoration objectives and strategies by site type

At individual sites on the Conservation Area, restoration objectives and strategies will vary depending on the current degree of degradation and the probability of improving current conditions. In high quality native grassland and shrub-steppe, objectives will focus on maintaining current conditions by reducing threats from nearby degraded areas. In contrast, within restoration sites, where it will not be possible to completely eradicate non-native species or re-create pre-disturbance plant communities, objectives will focus on increasing the proportion of native species and reducing non-native species abundances. Listed below are restoration objectives and strategies by site type, in decreasing order of site quality.

Boardman Conservation Area Restoration Plan- June 2007 10

Site type Objectives Strategy

High quality native grassland and shrub-steppe with low non-native species cover (sites in high and medium-high ecological condition)

1. Maintain current native plant and soil crust cover

2. Maintain non-native species cover at less than 10% of total plant cover

1. Within sites, prevent soil disturbance by prohibiting livestock grazing and vehicle access

2. Within sites, control outbreaks of non-native species as needed

3. In adjacent degraded sites, plant native species and control non-native species to reduce spread of non-native species and risk of high intensity fire

Type 4 restoration sites-dominated by a diverse array of native plant species with high cheatgrass densities

1. Maintain or increase current native plant and soil crust cover

2. Maintain non-native species cover at less than 25% of total plant cover

1. Within sites, reduce cheatgrass cover by using herbicides or controlled grazing as outlined in “Restoration Methods”

2. In adjacent degraded sites, plant native species and control non-native species to

Type 3 restoration sites-dominated by native shrubs with a low diversity of native herbaceous species

1. Maintain current native shrub cover

2. Increase native perennial herbaceous cover to at least 50% of total plant cover

3. Decrease non-native species cover to less than 50% of total plant cover

1. Within sites, plant native species and control non-native species as outlined in “Restoration Methods”

2. In adjacent degraded sites, plant native species and control non-native species to

Type 2 restoration sites-dominated by non-native annual plant species with a low diversity of native perennial species

1. Increase native perennial herbaceous

2. Decrease non-native species cover to

Type 1 restoration sites-dominated by non-native annual plant species

1. Increase native perennial herbaceous

2. Decrease non-native species cover to

Boardman Conservation Area Restoration Plan- June 2007 11

Prioritization of restoration sites

Initial restoration efforts on the Conservation Area will focus on reducing threats to adjacent high quality native plant communities. Moody and Mack (1988) modeled the non-native species invasion process and clearly showed the importance of eradicating non-native species in founding populations ("nascent foci"). We plan to take preventative steps to eliminate these founding populations by restoring areas dominated by non-native species, before they spread to adjacent high quality plant communities. Subsequent restoration efforts will focus on improving the ecological condition of other degraded areas by increasing the proportion of native species at those sites.

Each restoration site was assigned a restoration priority, based on the plant species composition of the restoration site and its spatial proximity to high quality native plant communities (Table 1):

1) Very high- Type 1 restoration sites adjacent to existing high quality native grassland and shrub-steppe (rated as being in high or medium high condition). These sites are dominated by non-native annual species which can spread to adjacent areas, and if burned, will experience hotter and faster moving fires that can burn through the adjacent high quality areas as well.

2) High- Type 1 restoration sites adjacent to Type 4 restoration sites. Although Type 4 restorations sites have high cheatgrass densities, they are also characterized by a diverse array of native plant species. As with the high quality native grassland and shrub-steppe, Type 4 sites are threatened by spread of other invasive species and altered fire regimes from Type 1 sites.

3) Medium high- Type 2 restoration sites adjacent to existing high quality native grassland and shrub-steppe (rated as being in high or medium high condition).

4) Medium- Type 3 restoration sites and Type 1 restoration sites located within Type 3 sites.

5) Low- Remaining Type 2 restoration sites not adjacent to existing high quality native grassland and shrub-steppe, and Type 1 restoration sites located within these sites.

6) Very low- Type 4 restoration sites.

Restoration potential

Restoring disturbed areas on the Conservation Area and similar landscapes in the Columbia Plateau ecoregion will be challenging. The semi-arid climate, combined with the presence of aggressive non-native plant species and altered fire regimes, provide difficult conditions for the germination, growth and persistence of species planted as part of restoration efforts. Although there is much written and anecdotal information about how to restore degraded western rangelands, examples of long-term success in restoring these systems is lacking.

Boardman Conservation Area Restoration Plan- June 2007 12

In shrub-steppe restoration sites where native species are planted, seedling emergence is often low, as is second and third year survival, and it is not uncommon for some planted species to not survive at all (Humphrey and Schupp 2002). Survival of planted species is largely driven by climatic factors, primarily soil moisture availability during the first growing season, as well as competition with non-native plant species. At a highly disturbed site at the Yakima Training Center in Washington, no seedlings germinated following drill seeding with a native species mix, and failure was attributed to drought conditions the year the study occurred (Gayaldo 1996). At a restoration site in Utah, no native grass seedlings survived when planted into a cheatgrass dominated site, regardless of the site preparation technique used prior to planting (tilling, harrowing, spraying with glyphosate) (Cox and Anderson 2004). Success was greater at the Fitzner-Eberhardt Arid Lands Ecology Reserve on the Hanford Reach National Monument in south-central Washington, where seedling emergence of aerial and drill seeded native grasses and forbs averaged 3% one year after seeding and 0.7% two years after seeding (Evans and Lih 2005).

The results of these few studies are not encouraging, however there is great potential to increase restoration success by using plant materials that are adapted to the local conditions (many projects have used cultivars rather than local seed sources), improving planting methods, and diligently controlling weeds while the planted species are establishing.

RESTORATION METHODS

Most restoration efforts on the Conservation Area will include the following components:

obtaining plant materials, site preparation, planting, and post-planting weed control. Options and guidelines for each of these components are described below, followed by specific recommendations for each of the four types of restoration sites.

Plant materials

Native species will need to be planted in restoration sites that lack the desired species composition and abundance because many of the native plant species (with the exception of a few disturbance-adapted species) lack mechanisms for quickly re-colonizing disturbed sites.

Semi-arid grassland and shrub-steppe species often do not form persistent seed banks, and even when sites are dominated by native species, there are few viable seeds in the seed bank (Hassan and West 1986). Colonization of native species from the seed bank is even more limited in highly degraded sites. In a cheatgrass dominated shrub-steppe in Utah, there were almost no native perennial seeds in the seedbank (Humphrey and Schupp 2001). Colonization of disturbed sites through natural seed dispersal will also be slow because few of the native species have mechanisms for long-distance dispersal. Even seeds of species that are presumably wind-dispersed, such as squirreltail, often do not disperse far from the parent plants (Marlette and Anderson 1986).

Planting lists

At each restoration site, the suite of species that is planted should be a subset of those present at the site prior to disturbance. In Table 1, the pre-disturbance (or “potential”) plant

Boardman Conservation Area Restoration Plan- June 2007 13 community for each restoration site is identified, based on the native-dominated plant community closest to the site that occurs on the same soil type.

Restoration plantings should contain as diverse a mixture of native grasses, forbs, and shrubs that were present in the pre-disturbance plant community as possible. Diverse plantings are more likely to provide the structure and function of the pre-disturbance plant communities and will increase the weed resistance of the restored plant community. Different species of grasses, forbs, and shrubs have different aboveground plant structure and rooting depths, which can maximize utilization of nutrients and moisture and reduce resources available to weeds (Sheley et al. 2004). Diverse plantings can also minimize the risk of planting failure because of the differences in germination and establishment requirements between species. In semi-arid climates, the high year-to-year climatic variability will probably favor different species in different years.

In spite of the benefits of highly diverse restoration plantings, the availability of plant materials, cost, and the specific conditions at each restoration site will limit planting diversity. In Tables 2 and 3, planting lists for high, middle, and low diversity plantings for each plant community are listed. Planting lists were created by compiling native plant species composition data collected between 2003 and 2005 in vegetation monitoring plots on the Conservation Area.

Species that were not considered for planting include early-successional/ disturbance-adapted native species such as hoary aster (Machaeranthera canescens), gray rabbitbrush (Chrysothamnus nauseosus), green rabbitbrush (Chrysothamnus viscidiflorus), matchbrush (Gutierrezia sarothrae), and annual forbs and grasses (Vulpia spp.), since these species are likely to naturally colonize restoration sites and in many cases are already present.

The low-diversity planting list contains the most abundant species found in the target plant community, the medium-diversity planting list contains additional species that are slightly less abundant in the target plant community, and the high-diversity planting list contains all species that occur in the target plant community (with the exceptions listed above). Squirreltail percentages are listed in greater proportions than are typically found in high-quality plant communities, because it is an early-successional species that may be able to compete with cheatgrass and other annuals while the other native perennial grasses are establishing (Tilley et

al. 2006).

Low-diversity planting lists contain the minimum number of species that should be planted in a restoration site in order to achieve the general structure of the pre-disturbance plant community. If the plant materials are available, additional species from the medium or high diversity planting lists should be added to the low-diversity planting list. In restoration sites where some desired native species are already present (i.e. Sandberg’s bluegrasss, big sagebrush, antelope bitterbrush in Type 2 and 3 sites) but additional species need to be planted, planting lists will need to be adjusted to exclude those desired species already present on the site.

As restoration experience is gained at the Conservation Area, planting lists will probably be updated. Plant species with limited seed production, difficult harvesting and germination requirements, or very low survival in restoration sites, may prove to be impractical for use in

Boardman Conservation Area Restoration Plan- June 2007 14 restoration projects. Proportions of species that are easy to obtain and have high survival rates may need to be increased.

One situation where it may be appropriate to plant a less diverse suite of species than even the low-diversity planting option is in greenstrips. Strategically placed green stripping, where fire-resistant vegetation is planted in 30-100 ft wide bands, are used to slow the spread of fire in areas where fire often originates (such as along roadsides) (Link 2005). Commonly used species in greenstrips are crested wheatgrass and forage kochia (Kochia prostrata) (Mosely et al.

1999). Although native perennial bunchgrasses are sometimes used in greenstrips, forage kochia is considered superior for this purpose because it maintains a higher moisture content during the fire season and competes well with cheatgrass, which is highly flammable (Harrison et al. 2002).

Although forage kochia has been shown to reduce fire intensity and rate of spread, and has even stopped fires altogether in some cases (Harrison et al. 2002), further research is needed to determine the invasive potential of this non-native species. The threat of fire spreading from roads is not currently severe enough to warrant planting green strips on the Conservation Area, but if this threat increases in the future, planting a low-diversity mix of native perennial grasses in green strips may be a safer approach even if it is less effective than forage kochia.

Genetic considerations

The genetic diversity of plant materials can be altered during each stage of a restoration project, including seed collecting, cleaning, increasing, and storage (Rogers and Montalvo 2004).

It is important to conserve the natural genetic condition of the species being planted as well as that of the native species present within the vicinity of the restoration site (Rogers and Montalvo 2004). Specific guidelines for maintaining genetic diversity throughout the restoration process are identified in the following sections.

On-site seed collection

Seed used in restoration projects should be obtained from on-site populations whenever possible. Local genotypes will be adapted to the local environmental conditions and are more likely to successfully establish and re-seed (Lesica and Allendorf 1999). Seed should be collected from at least 200 plants in several different stands over a range of sites to capture the genetic variation among the population (Lippitt et al. 1994, Rogers and Montalvo 2004). The timing of seed collection is also important- seed should be collected over the length of seed maturity to minimize the selection against either early or late maturing genotypes (Rogers and Montalvo 2004).

Seed of forbs, shrubs, and low-density grasses will need to be collected by hand, but some grasses can be collected with a mechanized seed stripper (Table 4). Seed strippers are effective for collecting large quantities of grass seed when there are relatively flat and weed-free patches of the target plant. A mechanized seed stripper pulled by an ATV was used successfully on the Conservation Area in the spring of 2006 to collect Sandberg’s bluegrass and needle-and-threadgrass seed. In Nebraska, combines as well as tractor-mounted seed strippers are used to collect large quantities of both single species and several species at once (Steinauer et al. 2003).

However, on the Conservation Area, collecting multiple species at once with mechanized

Boardman Conservation Area Restoration Plan- June 2007 15 equipment is probably not an option due to the differences in seed ripening, spatial distribution, and growth forms of the various target species.

Following collection, all seed must be air dried to prevent molding. Seed should be spread on tarps in a well ventilated area and turned every few days for 1-2 weeks and then stored in a dry, unheated place that is not accessible to rodents or insects.

Seed cleaning

Most seed will require some degree of cleaning before it can be planted. In Table 4, the reproductive structure that is needed for planting is identified. Seed that will be broadcast planted often requires little cleaning other that separating seeds from capsules or seed heads, but seed that will be drill planted may require further cleaning in order to pass through the drill.

Some initial cleaning to remove non-target species and stems can be done by hand, by filtering the seed through a series of mesh screens of different sizes. For further cleaning, seed should be sent to a seed processing facility since mechanized equipment such as hammer mills or air separators are needed to remove the seed from the chaff or pods.

Grass seed increasing

For most grass species, it will not be possible to collect enough seed on-site to use directly in restoration projects. The one exception may be needle-and-threadgrass, since there are large stands of this species that can be efficiently collected with a seed stripper. For other grass species, seeds collected on-site will need to be “increased”- that is, grown by a commercial grower in single-species production plots. Seed harvested from the production plots can then be planted on-site. To maintain the genetic diversity of the original seed collection, seed should only be harvested from the original seed collection (the first generation) planted by the commercial grower and it should be grown in an environment as similar as possible to the original collection site (Rogers and Montalvo 2004).

Purchasing commercial seed

Seed for many of the plant species native to the Conservation Area, particularly the grasses, is available commercially and is often less expensive than collecting and increasing seed. However, as mentioned earlier, seed originally collected on-site should be used whenever possible to avoid genetic contamination of plants already present on the site and to ensure that the seed is adapted to the local environmental conditions.

Source-identified commercial seed, which originates from plant materials either collected directly from or grown from naturally growing populations occurring in a known geographic area (Aubry et al. 2005), may be an appropriate substitute for seed collected on-site, particularly for species that are highly plastic and exhibit little genetic differentiation among populations.

However, no information is currently available to determine suitable geographic limits for off-site seed purchases. Current efforts by the U.S. Forest Service to delineate seed transfer zones for common Pacific Northwest native species may provide seed purchasing guidelines in the future

Boardman Conservation Area Restoration Plan- June 2007 16 and could result in reduced seed costs if appropriate source-identified commercial seed is available.

Cultivars of native species are widely available but generally should not be used in highly sensitive natural areas such as the Conservation Area. Cultivars are often bred for traits that allow more rapid colonization of highly disturbed sites than local genotypes, but they may be poorly adapted for long-term persistence (Lesica and Allendorf 1999). Cultivars may also compete more aggressively with native species and could transfer genetic material to local populations (Aubry et al. 2005). The risk of using cultivars appears to outweigh potential benefits, but in highly disturbed restoration sites the use of cultivars may warrant further consideration if planting attempts using local seed have failed.

Seeding rates

Seeding rates are usually expressed as the weight of seed per unit area, although sometimes the number of seeds per unit area is used. Recommended seeding rates for drill seeding on western rangelands range from 8-16 lbs/acre or 20-30 seeds/square foot (Monsen and Stevens 2004, Skinner et al. 2006). Broadcast seeding requires 33-50% more seed than drill seeding, but if seeds are covered following broadcasting, only 20% more seed may be needed (Monsen and Stevens 2004). Guidelines for determining seeding rates are not currently available, but ideally seeding rates are low enough to minimize competition between planted species but high enough so that when planted species are established, they can effectively compete with non-native species. At the Conservation Area, initial restoration plantings will use a seeding rate of 12 lbs./acre, but this rate will probably be adjusted over time based on our preliminary results.

Calculating seed mixtures

At a given seeding rate, the amount of seed needed for planting can vary considerably depending on the quality of the seed. Seeding rates should be adjusted based on the amount of pure live seed (PLS) in each seed lot. PLS is a measure of seed quality and is based on the percent purity and germination of the seed. All seed collections should be tested for germination and purity by a certified seed laboratory and labeled with this information as well as the date tested (Monsen and Stevens 2004).

The following example shows how much seed is needed for a seed mixture that is 55% bluebunch wheatgrass, 25% needle-and-threadgrass, and 20% Sandberg’s bluegrass planted at 12 lbs. PLS/acre, after adjusting for seed quality.

% of mix lbs.

PLS/acre purity germination

PLS

bulk seed needed lbs./acre bluebunch wheatgrass 55 6.6 95 85 80.8 8.2 needle-and-threadgrass 25 3 80 75 60.0 5.0

Sandberg's bluegrass 20 2.4 85 90 76.5 3.1

Total 100 12

Boardman Conservation Area Restoration Plan- June 2007 17

The seed mixture was calculated as follows:

1) The amount of PLS needed for each species in the seed mixture:

lbs. PLS/acre= (% desired in seed mixture * total lbs. PLS desired/acre) * 100

2) The amount of pure live seed (PLS) for each species in the mixture:

PLS= (% germination * % purity) / 100

3) The amount of bulk seed needed to achieve the desired PLS seeding rate:

bulk seeding rate= (desired PLS/acre) / (% PLS/100)

Calculating seeding rates based on the number of seeds per unit area adjusts rates for differences in the number of seeds in a pound for each species. In Table 4, the number of seeds per pound for each of the native grasses is listed. To calculate seeding rates based on the number of seeds per pound, the following formula is used:

lbs./acre of seed required= (# seeds/square foot desired * 43,560 feet/square acre)

# seeds/lb. * PLS/100

For each species in the seed mixture, the number of seeds/square foot is adjusted based its percentage in the seed mixture.

Transplanting

In addition to planting seed directly on-site, seed can be grown into plugs by a nursery and then transplanted. Plants grown in nurseries often germinate and grow more rapidly than under field conditions, and are often used for species that experience poor germination in the field, for small sites where rapid establishment is needed, or when little seed is available.

Transplants of grasses can also be used to augment grass seedings. Transplants can accelerate the recovery of large bunchgrasses on restoration sites because they will be more competitive with annual species than will newly germinating grass seedlings.

Transplanting is also commonly used for introducing forbs and shrubs into already established grass plantings. The need to use broadleaf herbicides for weed control following planting grasses precludes planting forbs and shrubs at the same time as grasses. Forbs and shrubs are often transplanted several years later, after the grasses are well established and can compete successfully with non-native forbs. An alternative method to transplanting forbs and shrubs is to broadcast forb and shrub seed into the grass planting, and then harrow to incorporate the seed into the soil (Monsen and Stevens 2004). However, the soil disturbance from harrowing may increase weed abundances and damage grass plantings.

Transplanting will incur higher growing and planting costs compared to direct seeding.

These costs may be outweighed by increased survival, although in one study, transplanting forb species (Crepis spp. and Astragalus purshii) did not result in significantly greater survival than direct seeding (Wirth and Pyke 2003).

Boardman Conservation Area Restoration Plan- June 2007 18

Transplanting rates will vary depending on the desired plant density and the expected survival of each species. Initial restoration efforts on the Conservation Area will use a planting rate of a total of 100 forb or shrub transplants per acre, but rates will probably be adjusted in future years depending on our preliminary results.

Site preparation

Site preparation refers to the process of creating a suitable seedbed for planting, and is often considered the most essential component for a successful restoration (Fitzpatrick 2004).

Common site preparation techniques used in restoration projects include mechanical cultivation, herbicide application, planting a cover crop, prescribed fire, and controlled grazing. The relative effectiveness of different site preparation techniques in increasing the survival of semi-arid grassland and shrub-steppe plant species will vary depending on the conditions at each site, such as the size and composition of the weed seed bank and the climatic conditions in the year the treatment occurs.

At the Conservation Area, site preparation treatments must be effective in both reducing weed competition and maintaining soil moisture to the greatest extent possible while the seeded species are establishing. Site preparation treatments will probably need to be applied multiple times to be effective because high cheatgrass seed densities in the soil seed bank will allow for rapid cheatgrass recovery even if treatments are effective in killing existing plants.

Below, the potential benefits and drawbacks of each site preparation technique for the

Conservation Area are evaluated, followed by specific recommendations for each type of restoration site. Additional sources that provide detailed descriptions of site preparation techniques include Fitzpatrick (2004), Steinauer et al. (2003), and Monsen et al. (2004).

Mechanical

Mechanical site preparation treatments include agricultural cultivation practices such as tilling, disking, plowing, and harrowing. Descriptions of the different types of cultivation equipment used for restoration projects are available in Stevens and Monsen (2004a).

Mechanical treatments are used in sites where: 1) few desired remnant plants are present since treatments will remove most of the existing vegetation, 2) the ground is relatively flat and not rocky, and 3) shrub cover is low. On the Conservation Area, mechanical treatments are most appropriate for sites dominated by non-native species (Type 1 restoration sites), since there is little risk of damaging desirable species.

The effectiveness of mechanical treatments in reducing non-native plant cover depends on the frequency, intensity, and depth of cultivation (Fitzpatrick 2004). Immediate suppression may occur with only one treatment, but non-native species often fully recover before the seeded species become established (Wilson et al. 2004). Repeated cultivation over several seasons may be effective in reducing the abundance of weed seeds in the soil seed bank, but this can also result in increased germination of buried seed because with each treatment additional seed is brought to the soil surface. Tillage of 4-6 inches deep has been recommended as a method for burying cheatgrass seed and preventing it from germinating (Mosely et al. 1999).

Boardman Conservation Area Restoration Plan- June 2007 19

Repeated cultivation and deep tillage are probably not suitable site preparation techniques for the Conservation Area. The frequent high winds will preclude using repeated cultivation since it will result in exposed bare soil for extended periods that will be subject to soil erosion.

Survival guidelines for the Washington ground squirrel will also limit the use of mechanical site preparation since plowing, disking, and tilling are considered detrimental to survival (ODFW 2005). Plowing is also prohibited under the conservation easement held by the state of Oregon.

However, disking is specifically permitted under the MSCCAA (David Evans and Associates 2004), and harrowing is probably acceptable as well as long as it doesn’t result in more soil disturbance than disking. The risk of site preparation treatments to the Washington ground squirrel will need to be assessed at each restoration site and the techniques used will need to be adjusted accordingly.

Mechanical site preparation treatments will probably be needed in restoration sites that have compacted soils. Compacted soils should be loosened prior to planting. One cultivation treatment followed by application of a non-specific herbicide may be effective for both reducing compaction and weed abundances. Mechanical treatments should be implemented in the fall (prior to the season with highest precipitation) to minimize soil moisture loss (Monsen and Stevens 2004). If soils are disked, they may need to be slightly re-compacted prior to planting because loose soil can result in seeds being planted too deep. Soil should be loose on the surface but underlain by firm soil prior to seeding (Monsen and Stevens 2004).

Herbicide

Herbicides can be used as a site preparation tool in all types of restoration sites, but the specific herbicide used and the timing of application will vary depending on the unwanted plant species present and whether desired native species are present at the site. In most cases, more than one herbicide application will be needed. A single herbicide application can initially reduce cheatgrass abundance, but can also result in increased cheatgrass seed production, providing abundant seed for re-colonizing the site (Mosely et al. 1999).

In Type 1 restoration sites, a non-specific herbicide such as glyphosate can be used since these sites are dominated by non-native species. Glyphosate can be applied in the fall to kill newly germinating cheatgrass, either before or after planting native grasses. Glyphosate can also be applied in the spring prior to fall planting. Spring applications of glyphosate should occur when cheatgrass seed is in the early dough stage. At this stage, the seed is not yet viable, and killing the plants will prevent additional seed input into the seed bank.

In Type 2, 3, or 4 restoration sites, where desired native species are also present, the timing or type of herbicide used will need to be adjusted to prevent damage to non-target plants.

Glyphosate applied in the early spring at 4-8 oz/acre to kill cheatgrass will cause little damage to native perennial species (Stevens and Monsen 2004b). Imazapic applied in the fall at 4 oz/acre controlled cheatgrass and other annuals on the Conservation Area without significant damage to native perennial species. An advantage of imazapic compared to glyphosate is that it can control annual species for at least one year. Currently in Oregon, imazapic can only be purchased by

Boardman Conservation Area Restoration Plan- June 2007 20 government agencies; The Nature Conservancy has been able to obtain it through the Gilliam County Weed Department.

One advantage of using herbicides for site preparation is that herbicides do not create soil disturbance, which can prevent erosion and reduce additional weed seed germination. The litter created from the dead plants can also improve soil moisture for establishing native seedlings, although a thick layer of mulch can be detrimental (Vallentine 2004).

Agricultural crops

Agricultural crops can be used as a site preparation tool in Type 1 restoration sites, since there are no desirable native species in these sites. Annual agricultural crops, such as wheat or barley, can compete with non-native species and reduce the weed seed bank if planted for several years before native species are planted. On the Conservation Area, “The Field” is the only Type 1 site that is large enough to plant an agricultural crop.