; An Adaptive Approach to
Learning Center
Plans & pricing Sign in
Sign Out

An Adaptive Approach to


  • pg 1
									       An Adaptive Approach to

Restoring Culturally-Important Plants at

  Grand Portage National Monument

                    NPS Photos

        Final report submitted: March 2011

       By Joy B. Zedler and James M. Doherty, with Botany 670 Students
                            Contributions as follows:
Joy Zedler was the principal investigator on the project, taught Botany 670 (Adaptive
Restoration Lab), wrote the bulk of Part 3, and compiled/edited the entire report. James Doherty
(UW-Madison) sampled vegetation in the Grand Portage National Monument (GRPO) meadow in
2009, mapped elevation and species distributions, wrote Part 2 and the management experiments
(Part 3, Phase 4), and revised significant portions of the document based on input from Brandon
Seitz (of GRPO). Deb Pomroy (UM-Duluth) contributed a plant species list at the restoration
site in summer 2009. Jenn Ipsen (UW undergraduate) propagated sweetgrass under varied water
and nutrient treatments at UW, where Dr. Mo Fayyaz and his staff provided space and sweetgrass
care in the Botany greenhouse.

The Fall 2009 Adaptive Restoration Lab students devised several experiments that were revised
and incorporated into this report. Contributors were Kayla Arendt, Michael Chang, Jason
Janzen, Karen Cardinal, Jason Londo, Blaine Wesley, Hadley Boehm, Kristin Millies, Kyle
Magyera, and Nick St. Martin.

The Fall 2008 Adaptive Restoration class contributed literature reviews for the ―Species Profiles‖
document that accompanies this restoration plan, and Fall 2009 students updated their reviews.
Contributors in 2008 were: Erik Olson, Carol Warden, Matt Simon, Stephanie Judge, Sarah
Ruth, Angelique Edgerton, Alissa Santurri, James Doherty, Julie Pignotti, and Laurel
Wilson. Updates in 2009 were contributed by: Kayla Arendt, Michael Chang, Jason Janzen,
Karen Cardinal, Jason Londo, Blaine Wesley, Hadley Boehm, Kristin Millies, Kyle
Magyera, and Nick St. Martin, with additional references from the MS thesis of Michael Healy.

                             University of Wisconsin-Madison
                                   Botany Department
                                    430 Lincoln Drive
                                   Madison, WI 53706


   This report was prepared under the Great Lakes Northern Forest Cooperative Ecosystem
   Studies Unit Task Agreement J6150090004, Cooperative Agreement CAH6000082000,
   between the National Park Service and the University of Minnesota.


        Students in Botany 670 developed understanding of culturally-important plants and
restoration ecology while contributing to this report. Three components provided a model
―service-learning‖ opportunity:

       (1) Literature reviews of the target restoration species, namely, caraway, chives,
           sunchokes, sweetgrass, reed canary grass, smooth brome, Greene‘s rush and Vasey‘s
           rush (Part 1, ―Species profiles‖),
       (2) Proposed experimental restoration of desired species, and
       (3) Plans for the removal of weedy invasive species.

      The restoration planning process benefited students of the University of Wisconsin (UW)
Botany 670 (Adaptive Restoration Lab, 2 Credits) in Fall of both 2008 and 2009.

        Brandon Seitz supplied a single block (about 1 square foot) of sweetgrass sod from the
GRPO meadow, which led to additional undergraduate research. Of the 15 undamaged seeds
contained on sweetgrass inflorescences in that sod block, 6 germinated and produced clones with
six unique genotypes. Jen Ipsen (Directed Study, 2 Credits) propagated the six genotypes of
sweetgrass in the UW Botany greenhouse and tested growth responses to water levels and
nutrients. In the process, she expanded the plants to over 50 pots of sweetgrass (photo below;
report provided to Seitz in May 2010). The pots of sweetgrass were donated to GRPO in
November 2010 for further propagation and use in restoring the meadow.

TABLE OF CONTENTS (headings hyperlinked to their location in the report)

                             Part 1. Species Profiles
Four culturally-important plants

Two native plant species that might assist sweetgrass restoration
      Greene’s rush
      Vasey’s rush

Two invasive species that need to be controlled
     Smooth brome
      Reed canary grass

                      Part 2. Meadow Vegetation Survey

              Part 3. Adaptive restoration and management plans

Phase 1:    Garden experiments
Phase 2:    Meadow-edge experiments
Phase 3:    Adaptive restoration within the meadow
Phase 4:    Adaptive management within the meadow
Summary of experiments by phase and priority

Discussion and Recommendations
Landscape restoration
Summary of recommendations
Highlighting ecological functions

Original project scope of work

                             Part 1. Species Profiles

            Culturally important plants that could be restored at the Grand Portage meadow.

               Compiled by students of Botany 670 Adaptive Restoration Lab
                      Edited by Joy B. Zedler, Professor of Botany
                            University of Wisconsin – Madison

This report was prepared under the Great Lakes Northern Forest Cooperative Ecosystem Studies
Unit Task Agreement J6150090004, Cooperative Agreement CAH6000082000, between the
National Park Service and the University of Minnesota.

SCIENTIFIC NAME: Hierochloe hirta (Schrank) Borbás Hierochloe comes from ieros for
"sacred" and chloe for grass

TAXONOMIC NOTES: We are confident that GRPO sweetgrass is Hierochloe hirta not
Hierochloe odorata (L.) P. Beauv. The former taxa has been recorded present in and around
Cook County MN, whereas the latter has not and is circumboreal in distribution (USDA).
However, Hierochloe hirta has often been referred to as Hierochloe odorata and still is, in some
taxonomic references (as noted by the Freckmann Herbarium). Some of the literature cited in this
report refers ambiguously to ―Hierochloe odorata‖ or ―sweetgrass‖. In such cases, we cannot
resolve which species was intended, so we caution that experiments and accounts of ―Hierochloe
odorata‖ or ―sweetgrass‖ may or may not pertain to the ecology of the related species, Hierochloe
hirta, found at GRPO.
        We are uncertain if GRPO sweetgrass is Hierochloe hirta subsp. hirta (Schrank) Borbás
or Hierochloe hirta subsp. arctica (J.Presl) G.Weim. The Freckmann Herbarium website
recognizes only Hierochloe hirta subsp. arctica in the state of Wisconsin, suggesting that may be
the predominant subspecies near GRPO. Still, out of precaution, in this report we simply refer to
the species, Hierochloe hirta.
        Finally, although Hierochloe hirta is an accepted nomenclatural synonym, the preferred
scientific name for the species is now Anthoxanthum hirtum (Schrank) Y. Schouten & Veldkamp
(ITIS 2010). We used Hierochloe hirta here because the genus is better-known. Likewise, while
the accepted common name is northern sweetgrass, common names for sweetgrass are redundant
or ambiguous and reflect common usage, rather than taxonomic singularity. In this document, we
refer to Hierochloe hirta as ―sweetgrass‖ rather than ―northern sweetgrass‖ to indicate the species
present at GRPO meadow in a way that is familiar and recognizable.

COMMON NAMES: northern sweetgrass, sweetgrass, vanilla grass, holy grass, Seneca grass
and alpine sweetgrass; Ojibwe names: Wiingashk (wii = spicy or aromatic, also to bind or twist;
gashk = herb, according to Lynch and Lupfer 1995); Wiishkobi-mashkosi (as listed by Pomroy in
part 2)

CONTRIBUTORS: Erik Olson, Carol Warden, Matt Simon, Stephanie Judge, Kayla Arendt,
Michael Chang, Jason Janzen

Drawing of Hierochloe hirta subsp. arctica by Agnes    Photographer: Robert W. Freckmann. Pictured:
Chase from Norman C. Fassett's Grasses of Wisconsin.   Hierochloe hirta subsp. arctica (Freckmann Herbarium)
(1951), in: Freckmann Herbarium

DESCRIPTION: Sweetgrass is a native perennial grass. The culms are semi-erect, up to 30
inches tall, and arise from slender, creeping rhizomes. Leaves are few, rough-edged and have
shiny, hairless undersides. Often it has a reddish-purple color near the base of the plant. The
highly-prized longer leaves that grow on sterile shoots late in the season can exceed 18 inches in
length. The very early season inflorescence is an open, pyramid-shaped, golden brown panicle
with slender branches. Spikelets have 3 florets with awnless lemmas; glumes are thin, translucent
and nearly equal in length. The fruit is a caryopsis (USDA). Sweetgrass spreads on slender,
creeping rhizomes. Leaf clumps arise from belowground tissue and wither soon after flowering.
Flowers are in pyramid-shaped clusters (Foster 2000). Lynch and Lupfer (1995) report taller
plants, exceeding three feet in Ontario.

CULTURAL USE: Sweetgrass is culturally important to many people, both in Europe and North
America. American Indians widely used sweetgrass as incense for ceremonies. The Ojibwe
braided sheaves of sweetgrass into a circle; ―when the head touches its roots the braid has no
beginning or end, only the continuity, renewal, and eternity symbolized by the circle‖ (Lynch and
Lupfer 1995). In Northern Europe, Hierochloe was strewn before church doors on saint's days
(Freckmann Herbarium 2011). Tea has been used for coughs, sore throats, chafing, venereal
infections, to stop vaginal bleeding, and expel afterbirth (Foster 2000). Windburn and chapping
were treated through an infusion of sweetgrass stems soaked in water or a salve of sweetgrass
water and grease. The sweetgrass water was also used as eyewash. Further listed uses include

perfume, hair wash, bedding, basketry, dermatological aid, cold remedy, cough medicine,
eyewash, febrifuge, respiratory aid, analgesic, insecticide, veterinary aid, decoration and
adornment (USDA).

NATIVE DISTRIBUTION: Sweetgrass (i.e., Hierochloe hirta) occurs throughout the northern
U.S.; its range includes Alaska and throughout the U.S. excluding the Southeast (USDA).
Normally, it is not found in pure stands but among other grasses and shrubs in mid-successional
communities (USDA).         Sweetgrass is threatened due to over-collection (Foster 2008) and
urbanization (Hurley et al. 2008).

ENVIRONMENTAL PREFERENCES: Goldsmith and Murphey (2008) found that sweetgrass
was an environmental generalist; however, it was sensitive to competition from other species,
making it most common where there was little competition. Sweetgrass prefers open, moist areas
and grows best in low-nutrient soil where competitors cannot outgrow it (Lynch and Lupfer
1995). Sweetgrass usually inhabits moist ground on shores (fresh or brackish), meadows, low
prairies, at the edges of woods, bogs and marshes (USDA). Sweetgrass is a cool-season species
that is extremely cold hardy. It goes dormant in cold weather and resprout once ground
temperatures reach 40 oF. In the Northeast, Midwest and Great Lakes regions, flowering begins
in the spring. It needs moderate amounts of light (White 2002), and can grow in partial shade and
full sun. An overstory helps reduce stress from transpiration at high temperatures, as sweetgrass is
not drought tolerant (USDA).

RESTORATION: In a field experiment, Shebitz and Kimmerer (2005) found great restoration
potential for sweetgrass because it is easily transplanted and reproduces vigorously. Sweetgrass is
a useful plant for wetland and riparian restoration and protection/renovation of springs.
Sweetgrass has potential to reduce erosion on moderately sloping hillsides with seeps (Shebitz
and Kimmerer 2005); it controls soil erosion by increasing the amount of large water-stable
aggregates in the soil (An et al. 2008).
        Sweetgrass spreads vigorously by creeping rhizomes. In the spring these rhizomes
produce inconspicuous fruiting stems with sparse, short leaves. Longer leaves develop later from
separate sterile basal shoots. Although sweetgrass can reproduce by seed, it is mostly infertile,
producing few seedheads that contain few seeds. Seed germination follows a period of cold
temperatures. Late fall, late winter, or early spring is the best time to plant the sweetgrass seed
(USDA). Sweetgrass development from seed can be very slow. This coupled with the infertile
nature of the species explains why plant division is the recommended method of reproducing
sweetgrass. Divide by separating ramets that develop from the rhizomes of a spreading plant.
Newly separated ramets do best if placed in the shade for 2–3 weeks while their roots establish.
After this, transplant at 1-foot centers into areas of partial shade to full sun (USDA). Keep the
soil moist but not saturated. Fertilize with a balanced lawn-starter 2-3 times during the
establishment year and years following harvests. Organic fertilizers may also be used at 5 lbs. per
100 square feet. Sweetgrass should be weeded at least every other year.
        Longer leaves of the sterile shoots may be harvested several times during the year;
however, the mid-season growth is considered superior. These leaves are gathered by grasping
the shoots firmly near the ground and pulling until they break from the rootstock an inch or two
below the surface of the soil (USDA). In Grand Portage it is common practice to cut the leaves at
least one inch above the surface (B. Seitz, pers comm.).

        Lynch and Lupfer (1995) recommended growing sweetgrass from plugs in isolated garden
plots to avoid competition with lawn grasses in areas with moist sandy soil with at least a half day
of sunlight. Shebitz and Kimmerer (2005) provided methods for growing sweetgrass in beds at a
garden scale. Plantings of sweetgrass in small (20-sq.ft = 2.25-m2) plots with hairy vetch
generated abundant tall blades that are desired by basketmakers; that method of cultivation was
not labor intensive. Reestablishment of sweetgrass offered members and visitors of
Kanatsiohareke a means to continue traditional practices associated with sweetgrass and benefit
economically by selling baskets and medicine made with sweetgrass.


An, S., Y. Huang, and F. Zheng. 2008. Aggregate characteristics during natural revegetation on
        the Loess Plateau. Pedosphere 18:809-816.
Foster, Steven, and James A. Duke. 2000. Eastern/Central Medicinal Plants and Herbs,
        Peterson Field Guides. 354.
Freckmann Herbarium (Robert W. Freckmann Herbarium, University of Wisconsin-Stevens
        Point). ―Hierochloe hirta subsp. arctica‖. n.d. Web. February 2011.
ITIS (Integrated Taxonomic Information System). ―Anthoxanthum hirtum‖. 2010. Web. February
Lynch, B., and B. Lupfer. 1995. Sweet Grass (Wiingashk) Project. Project Report. Great Lakes
        Indian Fish & Wildlife Commission. Odanah, WI.
Shebitz, D.J., and R. W. Kimmerer. 2005. Reestablishing roots of a Mohawk community and a
        culturally significant plant: Sweetgrass. Restoration Ecology 13:257-264.
White, D. 2002. Growth and clonal integration of sweetgrass (Hierochloe odorata) in western
        Montana. M.S. Thesis. University of Montana.
USDA (USDA, NRCS PLANTS Database). ―Hierochloe hirta subsp. arctica‖. National Plant
        Data Center, Baton Rouge, LA 70874-4490 USA. n.d. Web. February 2011.

Abstracts and research highlights regarding sweetgrass:

An, S., Y. Huang, and F. Zheng. 2008. Aggregate characteristics during natural revegetation on
        the Loess Plateau. Pedosphere 18:809-816.
Abstract: Field investigations and laboratory analysis were conducted to study the characteristics
of soil water-stable aggregates during vegetation rehabilitation in typical grassland soils of the
hilly-gullied loess area. The relationship between water-stable aggregates and other soil properties
was analyzed using canonical correlation analysis and principal component analysis. The results
show that during the natural revegetation, the aggregates >5 mm dominated and constituted
between 50% and 80% of the total soil water-stable aggregates in most of the soil layers. The 2–5
mm aggregate class was the second main component. The mean value of water-stable aggregates

>5 mm within the 0–2 m soil profile under different plant communities decreased in the following
order: Stipa grandis > Stipa bungeana Trin. > Artemisia sacrorum Ledeb. > Thymus mongolicus
Ronn. > Hierochloe hirta (Schrank) Borbás, organic matter, and total N were the key factors that
influenced the water stability of the aggregates. Total N and organic matter were the main factors
that affected the water stability of the aggregates > 5 mm and 0.5–1 mm in size. The contents of
Fe2O3, Al2O3, and physical clay (< 0.01 mm) were the main factors which affected the water
stability of the 1–2 and 0.25–0.5 mm aggregates.

Hurley, P., Halfacre, A., Levine, N., et al. 2008. Finding a ―disappearing‖ nontimber resource:
        Using grounded visualization to explore urbanization impacts on sweetgrass basketmaking
        in greater Mt. Pleasant, South Carolina. The Professional Geographer 60:556-578.
Abstract: Despite growing interest in urbanization and its social and ecological impacts on
formerly rural areas, empirical research remains limited. Extant studies largely focus either on
issues of social exclusion and enclosure or ecological change. This article uses the case of
sweetgrass basketmaking in Mt. Pleasant, South Carolina, to explore the implications of
urbanization, including gentrification, for the distribution and accessibility of sweetgrass, an
economically important nontimber forest product (NTFP) for historically African American
communities, in this rapidly growing area. We explore the usefulness of grounded visualization
for research efforts that are examining the existence of "fringe ecologies" associated with NTFP.
Our findings highlight the importance of integrated qualitative and quantitative analyses for
revealing the complex social and ecological changes that accompany both urbanization and rural

Shebitz, D.J., and R. W. Kimmerer. 2005. Reestablishing roots of a Mohawk community and a
        culturally significant plant: Sweetgrass. Restoration Ecology 13:257-264.
Abstract: The restoration potential of Sweetgrass (Anthoxanthum nitens (Weber) Y. Schouten
and Veldkamp) was evaluated through a field experiment conducted on Kanatsiohareke, a
Mohawk farm, and at the LaFayette Experiment Station near Syracuse, New York. The effects of
competition reduction and two cover crops on Sweetgrass reestablishment success were
examined. Sweetgrass was planted under four treatments: Sweetgrass alone; with existing, old-
field vegetation; with a cover crop of Hairy vetch (Vicia villosa); and with a cover crop of Annual
(Italian) ryegrass (Lolium multiflorum). The experiment consisted of five replicates of the four
treatments at both LaFayette and Kanatsiohareke. Sweetgrass biomass, height, reproduction rate,
and survivorship were greatest in plots that were weeded to eliminate competition and in plots
with Hairy vetch as a cover crop. A cover crop of Annual ryegrass resulted in reduced Sweetgrass
growth and production.
Highlights: used cover crops (hairy vetch, annual rye grass) to assist with establishment; hairy
vetch cover crop increased biomass, height, reproduction rate, and survivorship; rye grass reduced
growth and reproduction; sweet grass is easily transplanted; it can reproduce vigorously
vegetatively; when grown with hairy vetch, its traits are beneficial for ethnobotanical uses and
harvesting is easier.

Goldsmith, F.B., and S. L. Murphy. 1980. The ecological requirements of Hierochloe odorata in
       Nova Scotia. Holarctic Ecology 3:224-232.
Abstract: Hierochloe odorata (L.) Beauv. is restricted to the upper zone of salt marshes and
rarely becomes dominant. In the field, the species was found not to have strict N, P, K

requirements. It grew in a range of pH values from 4.3 to 7.9, tolerated salinities up to 500 mhos
and favored soils with a moisture content from 25 to 30%. It grew in situations with a mean water
table between 14 and 28 cm below the surface. Natural shading in the field was found to increase
the heights of plants by about 30%. The application of fertilizer in the field increased the height of
sweet grass but also stimulated the growth of associated species. In the greenhouse, the effects of
different soils and fertilizers on plant growth were assessed. Inorganic fertilizer (12:18:12)
application produced more leaves and tillers than organic fertilizer (6:2:0). Sweet grass grows in a
zone of reduced competition between the dune species and the salt-marsh species. Near the salt-
marsh, the species may be limited by high salt concentration. The low levels of competition
offered by cultivation and the responsiveness of the species to fertilizer suggests that cultivation
of Hierochloe odorata may be successful.
Highlights: nutrient requirements flexible; range of pH, 4.3-7.9; tolerates low levels of salinity
(500 mhos); soil moisture content 25-30%; water table depth 14-28 cm; natural shading found to
increase height by 30%; inorganic fertilizer produced the better results than organic; cultivation
may be successful.

White, D. 2002. Growth and clonal integration of Sweetgrass (Hierochloe odorata) in western
        Montana. M.S. Thesis. University of Montana.
Highlights: high moisture content and light availability are very important (especially during
growing season); typically found in meadows with some overstory; seems to prefer an open
canopy w/protection from excessive heat and desiccation; transplanting entire genets would work
best in harsh conditions; transplanting ―plugs‖ would work well in less harsh conditions

SCIENTIFIC NAMES: Helianthus tuberosus L. and H. tomentosus Michx.

COMMON NAMES: Jerusalem artichoke, topinambour, sunroot, sunchoke, earth apple
Ojibwe name: A'skibwan'

TAXONOMIC NOTE: We use the common name ―sunchokes‖ hereafter.

CONTRIBUTORS: Eric Booth, Alison McCleary, Andrea McMillen, Amanda Munsey, Karen
Cardinal, Jason Londo, Blaine Wesley

        http://ecochildsplay.com/2008/02/           http://www.umassgreeninfo.org/fact_sheets/

DESCRIPTION (primarily from Kays and Nottingham 2008): Sunchoke can reproduce both
sexually (seed production) and asexually (tuber production). It is a polyploid with 120
chromosomes that most likely originated from the hybridization of two species. Sunchoke is
perennial as it produces reproductive tubers in addition to seeds. However, it is often grown in
agriculture as an annual using the seed tuber for propagation. It is considered a colonizer that can
invade open areas and occupy areas with native and cultivated species.
        The development of individual plants can be separated in to five states: emergence and
canopy development, rhizome formation, flowering, tuberization and senescence. After seed
tubers have been dormant for a sufficient period, the new growth cycle begins with the sprouting
of the seed tuber and the emergence of individual sprouts. Sprouting will generally occur when
soil temperatures reach 2 to 5 °C, and will occur faster with higher temperatures. When planting
cultivars, emergence in spring usually begins 3 to 5 weeks after planting.
        During early stages of the development cycle, the plants invest carbon and nutrient
resources aboveground. After sprouting, stem growth is followed by branching and leaf
development. The size of the stem and branching depend on population density and photoperiod.
Later in the season, sunchokes store resources belowground as rhizomes and tubers. Rhizomes

allow plants to expand laterally into adjacent areas. High population density can reduce rhizome
formation, as can soil compaction. Compacted soil also makes it difficult to harvest the tubers.
        Tubers are a specialized reproductive storage organ. They develop at the distal end of the
rhizome. The shape and size of the tubers can greatly vary with growing conditions and the time
of development. A long-season cultivar goes into dormancy approximately 32 weeks after the
seed tuber sprouts. This process does not occur simultaneously throughout all tubers in a clone,
but gradually. Sunchoke tubers are freeze tolerant by the end of October, prior to leaf senescence.
By late October tubers can tolerate –5° C and by mid December, tolerance increases to –11°C.
        The photoperiod stimulates flower production of sunchokes. The photoperiod in the more
northern latitudes of the U.S. where H. tuberosus originates is favorable for extensive stem
growth during the summer and flower and tuber formation in the fall. The onset of flowering
ranges from 69 to 174 days after planting, and remain intact from 21 to 126 days. The flower has
10 – 20 yellow petals with a yellow-orange center with a maximum diameter of 3 1/2 inches. In
order for seeds to be produced, the flowers must be cross-pollinated.
        Seed production is highly variable with location, clone, and year. Wild clones may
produce up to 50 seeds per flower head. Wild populations generally produce more flowers, more
seeds, and have higher seed viability than cultivated plants. The seeds display a strong dormancy
at maturity, which can be overcome with stratification. Wild clones are more variable than
cultivated clones. With greater seed production comes increased genetic diversity and greater
potential for dispersal.

CULTURAL USE (from Kays and Nottingham 2008): Sunchoke is one of the oldest cultivated
crops in America with its first documented use being in 1605. It was then brought to France and
England, where it became popular by 1617. It was a very important carbohydrate crop in Europe
until the introduction of the potato. As early as 1622, Tobias Venner of England described
sunchokes as a root that was often eaten with ―butter, vinegar, and pepper, by itself, or together
with other meats (p. 19, Kays and Nottingham 2008).‖ It was not commonly cultivated in
America earlier than 1623 due to its reputation of being fit only for the poor. After that, various
social classes consumed the artichokes more frequently. The tubers were eaten as a source of
carbohydrates, however, the introduction of the potato led to a decrease in sunchoke consumption.
Sunchokes are also a good source of dietary fiber and fructose due to their storage of inulin
instead of starch. The tubers are very high quality for food and feed uses and also contain
abundant minerals such as iron, calcium, and potassium,. In addition, they are a good source of
vitamins B and C, and beta-carotene.
        Today, sunchokes have a wide variety of chemical and nutritional uses. Sunchoke can be
used as a vegetable, animal feed, production of biofuel, and a source of inulin, which occurs in the
tubers. This compound has been a popular bulking agent since the 1990‘s, when artificial
sweeteners became increasingly common and their use in baking caused a significant decrease in
volume. When added to baked goods such as bread, inulin improved softness of the crumb,
prolonged shelf life, and increased bread volume. Additionally, inulin is used as a low calorie fat
replacer in some meats.
        Sunchokes are also important in fermentation reactions. It is used commercially to
produce various common reagents, such as ethanol, acetone, butanol, 2,3-butanediol, lactic acid,
and succinic acid.
        A health benefit is that sunchokes help combat obesity and diabetes. Sunchokes are
considered to be nutraceutical supplements, meaning they provide medical or health benefits.

Much of the benefit is due to inulin, which is fermented in the colon and selectively alters the
microflora found there. It promotes bifidobacteria, which eliminate undesirable microbes in the
gut. Sunchokes are also considered to be beneficial for mineral absorption. Hundreds of European
products, especially yogurt, contain inulin for its nutraceutical properties. One claim that a
particular yogurt had cholesterol-lowering properties was even upheld in court.
        Inulin from sunchokes is rarely used medicinally unless it is in an extremely pure,
fractionated form. Pure inulin is utilized to test for renal failure by injection to determine
glomular filtration rate since inulin is not secreted or reabsorbed in the kidney. The concentration
of inulin in urine and plasma can help determine renal function.
         Because sunchoke tubers are rich in carbohydrates and the crop requires little irrigation, it
is being considered as an alternative food crop (Heiermann et al. 2009) and basis for biofuel
production including ethanol and biodiesel (Cheng et al. 2009). Because of its versatility, demand
for sunchokes has increased dramatically.

NATIVE DISTRIBUTION: North-central USA. This plant thrives best in temperate regions
(Kays and Nottingham 2008). It is generally accepted that the sunchokes originated in Canada,
where it was historically more broadly distributed than at present. Its exact center of origin is not
known, because it was widely cultivated by past civilizations and because it is not easy to track

ENVIRONMENTAL PREFERENCES: Sunchoke needs a growing season of at least 4 to 5
months with 125 frost-free days. The highest tuber yields have been found in sandy loam soils
with good fertility but sunchoke is well known for its tolerance to a wide range of soil conditions
including marginal lands with low fertility. In areas with high precipitation, finer-grained soils
could lead to water-logged soils and lower production. Optimal soil pH is in the range of 4.5 to
8.6 but slightly alkaline soils are the most favorable for production. In soils with low fertility, the
addition of high-phosphate fertilizer can increase tuber yields. Mounding soil around the base of
the plant can favor tuber formation by increasing the amount of buried stem where tubers can
        Overall, sunchokes require very little management to achieve good tuber yields. However,
some conditions and techniques can increase tuber yields; these include soil treatment, irrigation,
optimal planting density (2 to 4 plants per square meter), time from planting to harvest, and weed
control (Kays and Nottingham, 2008). Although sunchokes can grow in the shade, direct and full
sunlight is necessary for optimal yields (Kays and Nottingham, 2008). Most cultivars require an
annual average temperature between 6 and 26C. It is much more tolerant of frost than maize and
other crops; it has even been grown in Alaska.
        Sunchoke is ―moderately tolerant‖ of salt compared to other species; however, high
concentrations decrease plant growth. The degree of salt tolerance by seedlings is highly
dependent upon region of origin (Long et al. 2009).
        Nitrogen is a major factor determining growth of sunchoke. However, too much fertilizer
is not ideal for any species. A recent study linked photosynthetic activity to nitrogen fertilizers in
an attempt to improve nitrogen use efficiency (Michalek and Sawicka 2008). Although this
species is known for its tolerance of dry conditions, irrigation can have a large positive effect on
tuber yield. Excess irrigation, however, can lead to excessive top growth and reduced tuber yield.

RESTORATION: Sunchokes should be easy to reintroduce to areas with suitable soil, moisture

and light. Due to the fact that individual plants can grow up to 3 meters in height, adequate
spacing between individuals is important to reduce shading effects. With its tall stems and
shallow roots, plants can be susceptible to damage by wind. Techniques to address this problem
include the use of support stakes/poles, building up soil around the stems, and cutting off the tops
of the plant above 1.5 meters. Due to its size and height, the shade produced by this species leaf
canopy prevents weeds from establishing at the base. However, it typically takes 2 months for
sufficient shade to develop and during that time, weeds can be a problem. In a field study in NE
Portugal, Rodrigues et al. (2007) concluded that the best planting density was 2 plants per square
meter in both irrigated and rain-fed conditions. Under the irrigated condition, nitrogen
fertilization significantly increased tuber yield only when seed-tubers were used. Lower yields of
tuber were found with botanical seed.


Heiermann, M., M. Plöchl, B. Linke, H. Schelle, and C. Herrmann. 2009. Biogas Crops – Part I:
       Specifications and Suitability of Field Crops for Anaerobic Digestion. Agricultural
       Engineering International: the CIGR Ejournal. (XI) Manuscript 1087.
Kays, S.J. and S. F. Nottingham. 2007. Biology and Chemistry of Jerusalem Artichoke.
       Helianthus tuberosus L. New York, NY: CRC Press
Long, X. H., J. H. Chi, L. Liu, Q. Li, and Z. P. Liu. 2009. Effect of seawater stress on
       physiological and biochemical responses of five Jerusalem artichoke ecotypes.
       Pedosphere. 19:208–216.
Michalek, W., and B. Sawicka. 2008. Photosynthetic activity of Helianthus tuberosus L.
       depending on soil and mineral fertilization. Polish Journal of Soil Science. XLI/2.
Rodrigues, M.A., L. Sousa, J.E. Cabanas, M. Arrobas. 2007. Tuber yield and leafmineral
       composition of Jerusalem artichoke (Helianthus tuberosus L.) grown under different
       cropping practices. Spanish Journal of Agricultural Research 5:545-553
Yun Cheng,Wenguang Zhou, Chunfang Gao, Kenneth Lan, Yang Gao and QingyuWu. 2009.
       Biodiesel production from Jerusalem artichoke (Helianthus tuberosus L.) tuber by
       heterotrophic microalgae Chlorella protothecoides. Journal of Chemistry Technology and
       Biotechnology. 84:777–781

                       WILD AND CULTIVATED CHIVES
SCIENTIFIC NAME: Allium schoenoprasum var. sibiricum (L.) Hartman and A.
schoenoprasum L.

COMMON NAMES: English: wild chives, chives; French: Ciboulette/Civette; Italian:
Cipollina; Japanese: Ezo-negi; Spanish: Cebolleta; Portugese: Cebolinha-francesa (Wiersema and
Leon 1999)

TAXONOMIC NOTE: Just as we refer to northern sweetgrass as ―sweetgrass,‖ we refer to wild
chives as ―chives‖ because that name is commonly used and recognized.

CONTRIBUTORS: Sarah Reuth, Angelique Edgerton, Alissa Santurri, Hadley Boehm, and
Kristin Millies

           Photo by Susan R. Crispin,       USDA-NRCS PLANTS Database / Britton, N.L.,
           Michigan Natural Features        and A. Brown. 1913. An illustrated flora of the
           Inventory                        northern United States, Canada and the British
                                            Possessions. Vol. 1: 497.

DESCRIPTION: Wild chives (Allium schoenoprasum var. sibiricum) is a culturally important
species found at Grand Portage National Monument. Cultivated chives (var. schoenoprasum) and
wild chives (var. sibricum) have been grown widely for a variety of uses including food flavoring,
medicine, religious significance and ornamentation (Fenwick and Hanley 1985). Chives are
perennial monocots of the family Liliaceae (USDA). They are related to garlic and onions;
however, they do not have underground storage structures like the rest of their family members
(Fenwick and Hanley 1985). Chives are the only Allium member native to the Arctic, North
America, and Eurasia (Fenwick and Hanley 1985).

        Allium schoenoprasum var. schoenoprasum: The cultivated variety is an erect perennial
forb with a characteristic onion-like odor. It grows to heights of 4 to 20 inches (Plantlife.org
2008) and has an 8-20-innch (20-50 cm) tall stem (Freckmann Herbarium). It has a stem-like bulb
(Plantlife.org 2008) that is stout and round with membranous outer layers, which grows up to 10
mm thick (Freckmann Herbarium). Flowers are purple to pink and occur in dense, rounded umbel
inflorescences. Individual flowers are 6-parted and 1/3 to 1/2 inches long (Plantlife.org, 2008) and
are tubular-bell-shaped (Freckmann Herbarium). Each flower head has approximately 30 flowers,
with 2 lanceolate to ovate bracts. Allium schoenoprasum var. schoenoprasum blooms between
May and August (Freckmann Herbarium).
        Fruits of Allium schoenoprasum var. schoenoprasum are 3-lobed (Freckmann Herbarium)
capsules (Plantlife.org 2008) with 3 to 6 honey-combed seeds (Freckmann Herbarium). Leaves of
Allium schoenoprasum var. schoenoprasum are along the stem (Plantlife.org, 2008) and
cylindrical (Freckmann Herbarium). They are alternate, typically found in pairs, and grow to be 1
to 7 mm thick with the upper leaves sheathing the stem for approximately 1/3 to 1/2 of its length
(Freckmann Herbarium). Individual leaves are hollow and round in cross-section (Plantlife.org

        Allium schoenoprasum var. sibiricum: Compared to cultivated chives, the wild variety
has coarser, shorter leaves and fewer bulbs (Tardif and Morriset 1990). Further information on
the life history and distribution of the wild variety was available from the Minnesota DNR
3# and the USDA PLANTS database http://plants.usda.gov/java/profile?symbol=ALSCS.

CULTURAL USE: The Chinese first recorded the use of chives in 3000 B.C. It has been said
that Marco Polo brought chives back with him from the East. In western cultures chives began to
appear in gardens and food dishes in the 1500s (Rayment, 2009). Medieval people used chives to
ward off evil by hanging bunches around their homes. Gardeners planted chives around the
border of their gardens to keep harmful insects at bay. Gypsies also used chives in fortune-telling
(Rayment, 2009).
        Chives had a wide range of traditional medical uses in the Middle East, specifically
recorded in Tajikistan and Uzbekistan. They were used to cure headache, colds and stomach
problems. Chives were applied fresh or after boiling directly to the skin, the bulbs were dried,
crushed and snuffed, or simply consumed (Keusgen et al. 2006).
        In the New World, wild chives were used as appetite stimulants and digestive aids
(Freckmann Herbarium). The Cree People in Saskatchewan used wild chives for food by adding
them to boiled fish, using them for flavoring, and eating them fresh (Leighton 1985). The Ojibwe
and the Sioux Peoples used chives in their native wild rice dishes, called ―Manomem‖ (Wild Rice
Recipies from Minnesota). As medicine, wild chives were used to treat coughs and colds, sinus
problems and stomach worms, as dressings for wounds, and to treat a variety of other injuries. As
cough and cold medicines, plant juices were boiled into a syrup or the bulb was coated in sugar
and eaten. As sinus treatments, dried bulbs were burned for use as a fumigant or ground and
inhaled through the nose. For stomach worms, chives were soaked in water for 12 hours, and the
water was consumed on an empty stomach. To treat wounds, the juice of chives was applied to
sphagnum moss and used as an antiseptic dressing. Crushed chive bulbs were further used for a
variety of injuries, including insect bites, burns and snakebites (Freckmann Herbarium).

        Although modern science has disproven some of the traditionally-speculated medicinal
uses of chives, other uses have been validated. Cultivated species show inhibition of tumor cell
growth, and a high intake of Allium is associated with reduced risk of esophageal, stomach, and
prostate cancers (Štajner et al. 2008). The plants also contain vitamins A and C and calcium.
Kidney and bladder weaknesses have been treated with chives (Ecological Gardens Newsletter
        Today, wild chives are eaten as a vegetable in salads and meat dishes. They can be eaten
fresh, pickled or cooked (Freckmann Herbarium). Although enjoyed by humans, chives display
slight toxicity if ingested in high amounts by some animals including cats and cattle (Wiersema
and Leon, 1999). Cats are attracted to chives because they look like grass (Copley 2009). Chives
can be useful as a pest repellant, especially for Japanese beetles. Plantings with parsley, tomato,
beets, carrots, apples, and roses have proven to be beneficial for both parties. Due to their fibrous
roots, chives are beneficial for building soil and limiting erosion (Ecological Gardens Newsletter

RESTORATION: Plant seeds or bulbs in areas with ample light. Wild chives is a hardy plant; it
will grow in nearly any soil (Craig 2006). Wild chives is adapted to dry and sunny conditions,
and is tolerant to drought and oxidative stress (Egert and Tevin, 2002). It prefers full sun or partial
shade (Craig 2006).
        Chives is susceptible to fungal diseases including downy mildew and onion smut
(Fenwick and Hanley 1985). A recent study in California reported the occurrence of leaf blight in
chives (Koike et al. 2009).


Copley, J. Plants that are safe for cats. Cat Care. 2009. Web. October 2009. <http://cat-
Craig, W.J. 2006. Chive Talkin‘. Vibrant Life 22:20.
Ecological Gardens Newsletter. Minneapolis, MN. 2004. Web. October 2009.
Egert, M. and M. Tevin. 2002. Influence of drought on some physiological parameters
       symptomatic for oxidative stress in leaves of chives (Allium schoenoprasum).
       Environmental and Experimental Botany 48:43-49.
Fenwick, G.R. and A. B. Hanley. 1985. The genus Allium. 1. Critical Reviews in Food Science
       and Nutrition 22:199-271.
Freckmann Herbarium (Robert W. Freckmann Herbarium, University of Wisconsin-Stevens
       Point). ―Allium schoenoprasum‖. n.d. Web. February 2011.
Keusgen, M., R. M. Fritsch, H. Hisorie, P. A. Kurbonova, and F. O. Khassanov. 2006. Wild
       Allium species (Alliaceae) used in folk medicine of Tajikistan and Uzbekistan. Journal of
       Ethnobiology and Ethnomedicine 2 (8). Available online. <
Koike, S. T., L. J. du Toit, and M. L. Derie. 2009. A leaf blight of chive caused by Botrytis
       byssoidea in California. Plant Disease 93:844.
Leighton, A. L. 1985. Wild plant use by the Woods Cree (Nih*ithawak) of east-central
       Saskatchewan. Ottawa: National Museums of Canada.

Rayment, W. J. ―History of chives‖. 2007. Web. October 2009.
Štajner, D., R. Igic, B. M. Popovic, and D. Malencic. 2008. Comparative study of antioxidant
        properties of wild growing and cultivated Allium species. Phytotherapy Research 22:113-
Tardif, B. and P. Morisset. 1990. Clinal morphological variation of Allium schoenoprasum in
        Eastern North America. Taxon 39:417-429.
USDA (USDA, NRCS PLANTS Database). ―Allium schoenoprasum‖. National Plant Data
        Center, Baton Rouge, LA 70874-4490 USA. n.d. Web. October 2009.
Wiersema, J. H. and B. Leon. 1999. World Economic Plants: a standard reference. Boca Raton:
        CRC Press. p. 27.
Wild Chives - Allium schoenoprasum. Plantlife.org. 2008. Web. October 2008.


Amarawardana, L., P. Bandara, V. Kumar, J. Pettersson, V. Ninkovic, and R. Glinwood. 2007.
       Olfactory response of Myzus persicae (Homoptera : Aphididae) to volatiles from leek and
       chive: Potential for intercropping with sweet pepper. Acta Agriculturae Scandinavica
       Section B-Soil and Plant Science 57:87-91.
Borlongan, J. 2008. Chives (Allium schoenoprasum). 2008. Web. September 2008.
Brueckner, B., and H. Perner. 2006. Distribution of nutritive compounds and sensory quality in
       the leafs of chives (Allium schoenoprasum L.). Journal of Applied Botany and Food
       Quality 80:155-159.
Engelke, T., and T. Tatlioglu. 2000. The wi gene causes genic male sterility in Allium
       schoenoprasum. Plant Breeding 119(4), 325-328.
Fenwick, G. R., and A.B. Hanley. 1985. Genus Allium. Part 2. CRC critical reviews in food
       science and nutrition 22:273-377.
Fenwick, G. R., and A. B. Hanley. 1985. The genus Allium. 3. CRC critical reviews in food
       science and nutrition 23(1):1-73.
Friesen, N., and F. R. Blattner. 2000. RAPD analysis reveals geographic differentiations within
       Allium schoenoprasum L. (Alliaceae). Plant Biology 2:297-305.
Han, C., K. Pan, N. Wu, J. Wang, and W. Li. 2008. Allelopathic effect of ginger on seed
       germination and seedling growth of soybean and chive. Scientia Horticulturae 116: 330-
Holmes, D. S., and S. M. Bougourd. 1989. B-chromosome selection in Allium schoenoprasum.
       I. Natural populations. Heredity 63:83-87.
Kapolna, E., and P. Fodor. 2007. Bioavailability of selenium from selenium-enriched green
       onions (Allium fistulosum) and chives (Allium schoenoprasum) after 'in vitro'
       gastrointestinal digestion. International Journal of Food Sciences and Nutrition 58:282-
Michigan Natural Features Inventory. Rare Species Explorer. 2007. Web. September 2009.

Nguyen, N. H., H. E. Driscoll, and C. D. Specht. 2008. A molecular phylogeny of the wild onions
       (Allium; Alliaceae) with a focus on the western North American center of diversity.
       Molecular Phylogenetics and Evolution 47:1157-1172.
Stevens, J.P., and S.M. Bougourd. 1988. Inbreeding depression and the outcrossing rate in
       natural populations of Allium schoenoprasum L. (wild chives). Heredity 60:257-261.
USDA (USDA, NRCS PLANTS Database). ―Allium schoenoprasum var. sibiricum‖. National
       Plant Data Center, Baton Rouge, LA 70874-4490 USA. n.d. Web. September 2008.
University of Connecticut Ecology and Evolutionary Biology Plant Growth Facilities. ―Allium
       schoenoprasum (Alliaceae)‖. n.d. Web. September 2008.
Wild Rice Recipes from Minnesota. 1999. Web. October 2009.


COMMON NAMES: caraway, caraway seed (Plants for a Future, 1996)

CONTRIBUTORS: James Doherty, Julie Pignotti, Laurel Wilson

   USDA-NRCS PLANTS Database / Britton, N.L.,
    and A. Brown. 1913. An illustrated flora of the
    northern United States, Canada and the British    Steve Hurst @ USDA-NRCS PLANTS Database
              Possessions. Vol. 2: 659.

NATIVE DISTRIBUTION (Plants for a Future 1996): Research suggests that it is native to
Southeast England. However, it is rarely naturalized in Britain. It is also found in India, Iraq,
Kurdistan, Spain, and Turkey.

LIFE HISTORY (Kiviniemi 2008a): Caraway is a short-lived monocarp or ―facultative
biennial‖ (annual varieties exists, biennial variety is often able to persist for several years even
amidst grazing disturbance). It flowers from June to July, and the seeds ripen from July to August
and it flowers only once--at a size that maximizes fitness. Its nectar and pollen attract insect
pollinators. It has hermaphrodites, but can out-cross. Wind can also act as a pollinator and
pollination is likely not a limiting factor so much as having the resources and energy to make
pollen (Bouwmeester and Smid 1995). Seeds are mericarps with no apparent features to enhance
dispersal. Continuous disturbance is important, as is characteristic of early successional habitats.

CULTURAL USE: The seed is an antiseptic, antispasmodic, aromatic, carminative, digestive-
aid, emmenagogue, expectorant, galactogogue, and stimulant (Plants for a Future 1996). It can be
chewed raw for the almost immediate relief of indigestion and can also be made into infusions
(GardenGuides.com). The seed is also used in the treatment of bronchitis, diverticulitis,
bronchitis and is an ingredient of cough remedies. Moreover, the seed is also said to increase the
production of breast milk in nursing mothers, ease menstrual pain, eliminate flatulence, treat
failing vision, and curb the loss of appetite (Plants for a Future 1996; Medical Uses of Caraway
        Caraway has a long history of use as a household remedy, especially in the treatment of
digestive complaints. Its antispasmodic action soothes the digestive tract and its carminative
action relieves bloating caused by wind. It is often added to laxative medicines to prevent griping
(Plants for a Future 1996).
        Caraway was also used as medicine by Sumerians, Egyptians, and as currency by sixth-
century Persians. In Germany caraway bread was thought to ward off demons, and caraway seeds
were placed in coffins. The seeds have also been thought to prevent theft and were included in
love potions in Medieval times (Methias 2002)
        Caraway can be eaten raw or cooked. Its spicy flavor is enjoyed in bread, salads, and
vegetables. It is high in fat and protein and is often chewed after a meal in order to sweeten the
breath and relieve heartburn. An essential oil from the seed is used as a flavoring in ice cream,
candy, and soft drinks. Crushed seeds are brewed into a tea (has a soothing effect on digestion).
The root is cooked for use as a vegetable. Raw leaves are used to flavor soups; young leaves can
be used in salads, and older leaves can be cooked as spinach.

NATIVE DISTRIBUTION: Caraway prefers moist meadows, arable land, and waste places
from lowland to mountain elevations. Specifically, the plant prefers light (sandy), medium
(loamy) and heavy (clay) soils. It grows well in cultivated beds and ordinary garden soils. Also, it
tolerates a pH between 4.8-7.6 (Plants for a Future 1996). Experiments demonstrated healthy
caraway populations on well-drained dry-mesic grassland soils (Kiviniemi 2008a). Plants grow
well in either full sun or partial shade (Plants for a Future 1996). It has been demonstrated that
full sun produces more essential oils (Bouwmeester et al. 1995a) and perhaps better-tasting seeds

RESTORATION: Caraway grows well in ―semi-natural‖ areas and disturbance is not a problem
for this plant. Despite the fact that caraway seeds lack any specialized mechanism for dispersal
and they do not persist in long-lived seed banks. The results of a sowing experiment in 23
grasslands in Sweden demonstrated establishment by caraway at every site (Kiviniemi 2008b).
Because the meadow at Grand Portage National Monument is ruderal or ―semi-natural‖ it should
be little problem to sustain this species. Caraway is deep rooted and can help break up compacted
sub-soil, which likely underlies the wet meadow areas that were historically used for cattle
pasture. It grows well with other plants, especially those that are shallow-rooted (except fennel
and wormwood). The flowers attract parasitic wasps, which prey on aphids and help to reduce
populations of insect pests (Plants for a Future 1996).


Bouwmeester, H. J., and H. G. Smid. 1995. Seed yield n caraway (Carum carvi) .1. Role of
        pollination. Journal of Agricultural Science 124,235-244.
Bouwmeester, H. J., J. A. R. Davies, H.G. Smid, and R. S. A. Welten. 1995a. Physioloigical
        limitations to carbon yield in caraway (Carum carvi L). Industrial Crops and Products
Bouwmeester, H. J., H. G. Smid, and E. Loman. 1995b. Seed yield in caraway (Carum carvi) .2.
        Role of assimilate availability. Journal of Agricultural Science 124,245-251.
Evenhuis A, B. Verdam, M. Gerlagh, and H. M. Goossenvandegeijn. 1995. Studies on major
        diseases of caraway (Carum carvi ) in The Netherlands. Industrial Crops and Products
GardenGuides.com. ―Caraway (Carum carvi)‖. 2008. Web. October 2008.
Government of Saskatchewan, Agriculture. ―Caraway‖. 2006. Web. December 2008.
Kiviniemi, K. 2008a. Remnant population dynamics in the facultative biennial Carum carvi in
        fragmented semi-natural grasslands. Population Ecology. Available online.
Kiviniemi, K. 2008b. Effects of fragment size and isolation on the occurrence of four short-lived
        plants in semi-natural grasslands. Acta Oecologica-International Journal of Ecology
Langberger, M.W. and A.R. Davis. 2002. Temporal changes in floral nectar production,
        reabsorption, and composition associated with dichogamy in annual caraway (Carum
        carvi; Apiaceae). American Journal of Botany 89:1588-1598.
Machowicz-Stefaniak, Z. and E. Zalewska. 2008. Biodiversity of fungi colonizing different parts
        of caraway (Carum carvi L.). Electronic Journal of Polish Agricultural Universities. 11(1)
Medicinal uses of Caraway. 1999. Web. October 2008.
Methias, M.E. ―Caraway‖. Fall 2002. Web. October 2008.
Plants for a Future: Edible, medicinal, and useful plants for a healthier world. 1996. Web. October
        2008. <http://www.ibiblio.org/pfaf/cgi-bin/arr_html?Carum+carvi>.

                                         GREENE’S RUSH

SCIENTIFIC NAME: Juncus greenei Oakes and Tuck.

CONTRIBUTORS: Kayla Arendt, Michael Chang, Jason Janzen

USDA-NRCS PLANTS Database / Britton, N.L., and A.
Brown. 1913. An illustrated flora of the northern United   Photograph: Emmet J. Judziewicz
States, Canada and the British Possessions. Vol. 1: 472.

DESCRIPTION: Greene‘s rush is a perennial herb with a basal tuft of grass-like leaves that
grow to 70 cm (LJWC). Each plant grows up to 30 culms from short, densely branched rhizomes.
The leaves are dark green, leathery, and terete. The terminal inflorescences are usually congested
and up to 8 cm long, surpassed by a primary bract. The flowers consist of two bracteoles, and 3
mm dark-green lanceolate inner and outer tepals. The 3-locular ellipsoid capsules are dark brown
and range from 1-4 mm long. The dark tan lunate seeds range from 0.5-0.7 mm in diameter. The
species is diploid with 40 pairs of chromosomes for every mature plant (FNA). Greene‘s rush
flowers during June-September and produces fruit from July-October (Burns 1984).

NATIVE DISTRIBUTION: Greene‘s rush grows naturally in northeast North America from
New Brunswick to New Hampshire, mostly near the coast. It also is found inland from southwest
Ontario to Indiana, Iowa, and Minnesota (Burns 1984).

         Greene‘s rush grows well in soils that are moist or dry, sandy, and well drained. The
species tolerates disturbances in pine lands, near lake shores, among sand dunes (FNA), swales,
fields, clearings, inter-dune depressions, and along railroads and roadsides (Burns 1984).

RESTORATION: Greene‘s rush is susceptible to over-shading by woody species during
succession. Its recovery potential is presumed good since it appears able to colonize disturbed
areas (Burns 1984). It is identified as endangered in Ohio and Vermont and as extirpated in
Pennsylvania (USDA).


Burns, J. 1984. Juncus greenei fact sheet. Ohio Department of Natural Resources, Division of
       Natural Areas and Preserves. Available online.
FNA (Flora of North America). ―Juncus greenei‖. n.d. Web. October 2009.
LJWC (Ladybird Johnson Wildflower Center). ―Juncus greenei‖. n.d. Web. October 2009.
USDA (USDA, NRCS PLANTS Database). ―Juncus greenei‖. National Plant Data Center, Baton
       Rouge, LA 70874-4490 USA. n.d. Web. October 2009.

                                          VASEY’S RUSH
SCIENTIFIC NAME: Juncus vaseyi Engelm.

CONTRIBUTORS: Kyle Magyera and Nick St. Martin

                               USDA-NRCS PLANTS Database / Britton,
                                 N.L., and A. Brown. 1913. An illustrated
                               flora of the northern United States, Canada
                                 and the British Possessions. Vol. 1: 471.

DESCRIPTION: Juncus vaseyi has a grass-like structure and small, 6-petal flowers. Its stems
are upright and erect and 20-30 cm tall. Leaves arise near the base of the plant; they are erect,
terete, slender, and up to 30 cm long. Flowers occur in groups of 5-30 on a branching limb off of
the base. It has 6 sepals and petals per flower, which are pale green to greenish-tan with 3-valved
capsules. J. vaseyi is considered to be anemophilous (wind pollinated). Individuals have with
hair-like structures along stigmatic surfaces that help collect air-borne pollen (Judd et al. 1999).
Seeds are tan, elliptical in shape, and 1-1.5 mm long with a tail at each end. Note that Haines
(2003) reported a range of seed lengths as 1.7-2.2 mm. Flowering and fruiting occur in July,
although capsules contain seed through September indicating a slow release over at least two
months. J. vaseyi is a perennial species that reproduces vegetatively via short, compact rhizomes.
         Juncus vaseyi is morphologically similar to its close relative, J. greenei. In addition, J.
vaseyi may hybridize with J. tenuis resulting in the hybrid J. x oronensis. Juncus x oronensis has
been found growing with J. vaseyi at known locations in Maine (McKenny 2004).

NATIVE DISTRIBUTION: Although dispersed by wind, Juncus vaseyi is mainly propagated by
water. A New England study suggests that water may be a primary factor in long-distance
movement of propagules. Also, wet seeds adhere to waterfowl, which transport the species
between watersheds (Haines 2003).
       Juncus vaseyi is an early successional plant that colonizes sites after disturbance, but is

commonly outcompeted by woody plants during succession (McKenny 2004). Although found in
a variety of north temperate and boreal habitats, Vasey‘s rush is a very rare plant associated with
Great Lakes Basin coastal plain marshes dominated by grasses and rushes (Kost 2000). It
typically occurs on sand deposits associated with postglacial lakes and outwash channels. The
sandy soils of the coastal plain marsh are strongly to very strongly acidic and nutrient deficient.
Organic deposits of peat, muck, or sandy peat, generally of the folist type, are sometimes found
above sandy substrate. In some locations, a clay layer occurs several meters below the sand (Kost
        J. Vaseyi tends to occur in areas with hydric soils and is frequently present in wetland
communities such as wet meadows, shrub carrs, and shallow marshes. It can also be found in
wetter or drier conditions, in some regions. For example, in northeastern United States, J. vaseyi
inhabits dry-mesic mixed conifer hardwood forests, and in Michigan, it is common in seasonally-
wet open areas on rocks and sand, typically in partially acidic substrates (Voss 1972). Along the Lake
Michigan shoreline, this species is frequently found in spring runoff channels dominated by
scattered jack pine. Its variability indicates a tendency to occupy sites with seasonal saturation
and inundation. Some observers have correlated greater growth with full sun (McKenny 2004).
        Common associates of J. vaseyi are bluejoint grass (Calamagrostis canadensis), cordgrass
(Spartina pectinata), other rushes (Juncus spp.), sedges (Carex sp.), twig-rush (Cladium
mariscoides), shrubby cinquefoil (Dasiphora fruticosa), swamp milkweed (Asclepias incarnata),
big bluestem (Andropogon gerardii), Indian grass (Sorghastrum nutans), Sullivant's milkweed
(Asclepias sullivantii), purple milkweed, swamp thistle (Cirsium muticum), eastern prairie fringed
orchid (Platanthera leucophaea), marsh blazing star (Liatris spicata), whorled loosestrife
(Lysimachia quadrifolia), grass-of-Parnassus (Parnassia palustris), smooth hedge nettle (Stachys
tenuifolia), swamp rose (Rosa palustris), Missouri ironweed (Vernonia missurica), Joe-pye weed
(Eupatorium maculatum), common bone set (Eupatorium perfoliatium), and spike-rush
(Eleocharis acicularis), according to the Michigan Natural Features Inventory.

RESTORATION: Restoring habitat is a key to sustaining this species, because many of its
former habitats have been degraded and lost to agriculture, development, hydrological alterations,
and fire suppression (Michigan Natural Features Inventory). Once reintroduced, habitat
management largely consists of maintaining the appropriate hydrology and disturbance regimes.
Because J. vaseyi prefers open habitat, woody vegetation needs to be controlled. Effective tools
for sustaining herbaceous vegetation are prescribed burns and manual brush removal.


Brooks, R. E., and S. E. Clements. 2000. Juncaceae. Pages 211–267 in Flora of North America,
       Volume 22. Oxford University Press, New York, NY.
DeFilipps, Robert A. 1964. A taxonomic study of Juncus in Illinois. American Midland
       Naturalist 71:296-319
Haines, A. 2003. Juncus vaseyi Engelm.Vasey‘s Rush. Conservation and Research Plan for New
Judd, W. S., C. S. Campbell, E. A. Kellogg, and P. F. Stevens. 1999. Plant Systematics: A
       Phylogenetic Approach. Sinauer Associates, Inc., Sunderland, MA.
Kost, M. A. 2000. Natural community abstract for coastal plain marsh. Michigan Natural
       Features Inventory, Lansing, MI. 5 pp.

USDA Forest Service (eastern division). 2004. Conservation Assessment for Juncus vaseyi
       Engelmann (Vasey‘s rush). Hiawatha National Forest. Available online. <
Michigan Natural Features Inventory. ―Juncus Vaseyi – Vasey‘s Rush‖. 2007. Web. March 2011.
Voss, E.G. 1972. Michigan Flora. Vol. 1. Cranbrook Institute of Science and University of
       Michigan Herbarium. pp. 381-386.

                                        SMOOTH BROME
SCIENTIFIC NAME: Bromus inermis Leyss.

COMMON NAMES: smooth brome, awnless brome

CONTRIBUTORS: Karen Cardinal, Jason Londo, Blaine Northcraft


DESCRIPTION: Bromus inermis is a cold season, perennial grass that spreads by seeding and
aggressive rhizomes. The average stem height is anywhere from 2 to 4 feet, and produces many
basal and stem leaves that can vary in size from 4 to 10 inches. The leaves of smooth brome often
have a characteristic transverse ―W‖ wrinkle near their tips (USDA). The flowering heads of
smooth brome have 4-10 branching spikes, with each spike being 1-2 inches in length. Roughly
10 blunt florets make up each spike, with the florets turning to a purple-brown color as the mature
from June to August (ODNR).
         Smooth brome is a cool season grass that grows early in spring to late in fall. In the
Chicago area, new shoots emerge as early as mid-March and flowering primordial are first seen in
early April. Stems elongate from early May to late April, with the boot stage starting in mid-to-
late May in Wisconsin and Illinois. Smooth brome is fully headed and blooming in early June in
Illinois, and mid-June and early July in Alberta. Seeds ripen in July in Illinois and in August in
Alberta. During the spring boot stage, carbohydrate levels are lowest, increasing as internodes
elongate until plant begins heading. In Wisconsin, maximum total stored carbohydrates occurred

at seed maturity (mid-July), with small peaks in early spring (pre-boot) and fall (Sather 2009).
Smooth brome is open-pollinated and self-incompatible. Plants produce 156 to 10,080 viable
seeds per plant from 47 to 160 seed heads per plant. Seed predators include Itonidid midges and
Chalcid flies. One study observed ants sequestering the seeds, which then resulted in new smooth
brome patches on anthills (Sather 2009).
         Cold stratifications for 1-14 days allows for early spring shoot growth, followed by
vernalization during a possible cold spell, leading to the development of flowering panicles in the
same year (Sather 2009). Before or during primordial formation a minimum of 5 to 14 leaves
must develop in order to flower in any year, and a critical daylength of 13 hours must occur for
flowering, according to greenhouse studies. In comparison the flowering daylength is 17 to 18
hours when temperatures are warm, and in Alberta studies, little flowering was observed when air
temperatures fell below 60 degrees Fahrenheit (Sather 2009).
         Smooth brome forms rhizomes and sod. Within 5 days of seed germination the first
adventitious roots develop, followed by rhizome formation as early as 3 weeks after germination.
The majority of the root mass develops near the surface associated with creeping rhizomes.
However, the root system can reach depths of 4.7 feet, contributing to drought resistance.
Greenhouse and field tests in Wisconsin suggest that smooth brome roots grow best in silty
compared to sandy soils, and that it experiences high mortality in organic soils. Artificial aeration
of silt loam has been shown to promote growth. However, the relationship between soil aeration
and smooth brome mortality has not been investigated. Researchers estimate that inundation
tolerance is limited to 24 days, specifically during the spring (Sather 2009).

CULTURAL USE: Smooth brome is a useful food for livestock in the form of hay, pasture, or
silage because it is high in protein and relatively low in crude-fiber content. It is also used in
erosion control, due to its extensive rhizome/root system and sod forming characteristics (USDA).

NATIVE DISTRIBUTION: Bromus inermis is a Eurasian species ranging from France to
Siberia. In 1884, it was introduced into the United States by the California Experiment Station
(Sather 2009). Smooth brome grows commonly on roadsides, riverbanks, and edges of fields,
woods and pastures. Within the United States many agricultural strains have developed from the
natural ―northern‖ and ―southern‖ strains, with the southern strain being more tolerant to heat and
drought (Sather 2009). Bromus inermis is now found in all of North America excluding Florida
and Alabama according to the USDA distribution map.

MANAGEMENT IN RESTORATION SITES: Smooth brome is most invasive in cooler
climates, but it is resistant to temperature extremes as well as drought. It becomes very
susceptible to disease in high humidity. The plant tolerates pH 6.0 to 7.5 but grows best in soil
that is slightly acidic, to slightly alkaline. It prefers clay loam that is well drained but is also able
to grow in course-textured soils with adequate moisture and fertility. It grows poorly in soils that
are high in soluble salts (USDA). Overstory vegetation is another limiting factor. Seed
production, numbers of shoots and rhizomes, and dry weight of all plant parts are reduced by
shade (Sather 2009).

        The ability of smooth brome to establish early in the growing season allows it to out-
compete native grasses and become highly invasive. Smooth brome is unable to grow as a pure
stand without heavy applications of nitrogen in early spring and fall.
        Smooth brome can be used to improve soil in some restoration contexts. Its heavy sod is
an exceptional soil conditioner, and it can be plowed out after 3 to 4 years (USDA). The rhizomes
are observed to live one year. Old brome fields become ―sod bound‖ with nitrogen deficiency and
reduced shoot density (Sather 2009). Combinations of mechanical and chemical controls might
be the most effective. A spring burn followed by an herbicide (glyphosate) reduced densities of
smooth brome. However, native species were negatively affected (Grilz and Romo 1995).
        Burning after shoots emerge in the late spring sometimes favors native warm-season
species over smooth brome; however, if done too early it can favor smooth brome. A single burn
is not sufficient and may result in full recovery of smooth brome densities in following years.
However, repeated burning during late stages of growth have been effective in controlling
biomass (Willson and Stubbendiek 1997, Howard 1996). Clipping smooth brome after its period
of growth can be effective as well as mowing in April (Willson 2000). While growth rates are
highest when daytime temperatures are around 18.3-24.9 degrees C, growth responses have been
shown to vary considerably with cultivar and region (Howard 1996).
        An herbicide that contains the active ingredient glyphosate (e.g., Roundup® or
Glypro®) can be effective when applied to dense populations in April or May. However,
extreme care is needed to avoid non-target species since glyphosate is not a selective herbicide.
Atrazine has also been shown to reduce the productivity and density of smooth brome, but with
significant collateral damage also done to restored native warm-season grasses when the herbiced
was applied in late April (Willson 2000). In both cases, the negative impacts of herbicides on
wildlife need to be considered and caution used in application.


Grilz, P. L. and J. T. Romo. 1995. Management considerations for controlling smooth brome in
        fescue prairie. Natural Areas Journal 15:148-156.
Howard, J. L. 1996. ―Bromus inermis‖. Fire Effects Information System, U.S. Department of
        Agriculture, Forest Service, Rocky Mountain Research Station, Fire Sciences Laboratory.
        Available online. <http://www.fs.fed.us/database/feis/>.
ODNR (Ohio Department of Natural Resources, Division of Natural Areas and Preserves).
        ―Bromus inermus‖. Invasive Plants of Ohio. n.d. Web. October 2009.
Sather, N. 2009. Element stewardship abstract for awnless brome, smooth brome. The Nature
        Conservancy. Available online.
USDA (USDA, NRCS PLANTS Database). ―Bromus inermis‖. National Plant Data Center, Baton
        Rouge, LA 70874-4490 USA. n.d. Web. September 2008.
Willson, G. D. and J. Studdendiek. 2000. A provisional model for smooth brome management in
        degraded tallgrass prairie. Ecological Restoration 18:34-38.
Willson, G. D., and J. Stubbendiek. 1997. Fire effects on four growth stages of smooth brome
        (Bromus inermis Leyss.). Natural Areas Journal 17:306-312.

                                   REED CANARY GRASS
SCIENTIFIC NAME: Phalaris arundinacea L.

CONTRIBUTORS: Hadley Boehm and Kristin Millies; supplementary references from Michael Healy.

      USDA-NRCS PLANTS Database / Hitchcock,
     A.S. (rev. A. Chase). 1950. Manual of the grasses
     of the United States. USDA Misc. Publ. No. 200.     Steve Hurst @ USDA-NRCS PLANTS Database
                     Washington, DC.

DESCRIPTION: Reed canary grass is a cool-season, C3, clonal, perennial member of the grass
family (Poaceae). It inhabits temperate regions in Europe, Asia and North America (Kercher and
Zedler, 2004a), and it has invaded inland temperate wetlands in North America (Galatowitsch et
al., 1999). The European type is considered to be more aggressive (WDNR). Both Eurasian and
native ecotypes are present in the U.S., however, there is no reliable method to tell them apart
        Reed canary grass grows approximately 0.5-2 m in height with flat, rough leaf blades
approximately 9-24 cm long and 0.5-2 cm wide (Apfelbaum and Sams, 1987; WDNR). It flowers
greenish purple to beige from April-August (WDNR). Flowers occur on spikelets arranged on 8-
41 cm long panicles in dense clusters on 5-31 cm branches (WDNR). Reed canary grass has long,
membranous ligules, which help distinguish it (WDNR). Seeds are shiny, brown, ellipsoid and
0.25 cm long (WDNR).
        Reed canary grass reproduces both by seed and creeping rhizomes (WDNR). Most early
shoots arise from rhizomes (Evans and Ely 1941). It also produces new shoots at culm nodes
(Miller et al. 2008b).

ENVIRONMENTAL PREFERENCES: Reed canary grass is adapted to a wide range of
extremes of soil moisture (Galatowitsch et al. 1999) and is present throughout North America
(USDA). It tolerates many hydrological conditions thriving where there are frequent fluctuations
in water level, particularly where there is long, early flooding in combination with other
disturbances (Galatowitsch et al. 1999, Kercher and Zedler 2004b, Kercher et al. 2007). Reed
canary grass shows optimal growth on fertile, moist organic soil in full sunlight; however, it can
also grow on dry soils in upland habitat or in partial shade (WDNR). The growth of reed canary
grass is accelerated by disturbance and increased levels of nitrogen (Kercher and Zedler 2004a,
Herr-Turoff and Zedler 2005, Green and Galatowitsch 2002).
        High annual seed yields are common (Lavergne and Molofsky 2004). Seeds ripen in June,
and are capable of germination immediately after ripening with no dormancy requirement
(WDNR, Apfelbaum and Sams 1987). Reed canary grass is able to form highly productive
monocultures using this strategy. Seeds germinate best in moist soils (Lavergne and Molofsky
        Reed canary grass is one of the first plants to develop leafy shoots in spring; it has a
second growth spurt in fall (WDNR, Evans and Ely 1941). Following initial shoot development in
spring, its rhizome biomass develops very rapidly in a dense belowground mat (Evans and Ely
1941). Rhizomes grow near the surface filling rhizosphere gaps and extending the area it occupies
with new shoots from the rhizomes (Evans and Ely 1941). Reed canary grass outcompete other
species for light in spring, then supports that growth by obtaining nutrients with its extended
rhizomes (Reinhardt and Galatowitsch 2005).

CULTURAL USE: Reed canary grass has been an important cultivated forage grass in northern
temperate regions for approximately 200 years (Galatowitsch et al. 1999). The Eurasian type has
been selected because of its vigor, and has been planted in the U.S. for forage and erosion control
since the 1800s (WDNR, Lavergne and Molofsky 2004). Reed canary grass is also grown for
silage and grass fodder for cattle (Apfelbaum and Sams 1987). Today it is still being planted for
livestock, and on steep slopes of ponds and created wetlands (Kading et al. 1990, WDNR). The
grass is used for restoration of waters and degraded soils, as well as wastewater treatment
(Lavergne and Molofsky 2004).
        Reed canary grass is a source of biofuel (requiring an excessive heating and drying period)
and biogas (Burval 1997, Geber 2002, Lamptey et al. 2003). In northern Europe it has been used
in the pulp and paper industry (Finell et al. 2002).

MANAGEMENT IN RESTORATION SITES: The aggressive nature of reed canary grass
makes it a problematic invasive plant in many wetlands and disturbed habitats. Reed canary grass
is capable of invading many types of wetlands including: marshes, wet prairies, sedge meadows,
fens, stream banks, seasonal wet areas, and disturbed areas (WDNR). Growth of reed canary grass
is accelerated in recently disturbed areas and it frequently becomes monotypic (Kercher and
Zedler 2004a, WDNR).
         Reed canary grass reduces native plant growth (Green and Galatowitsch 2002) by quickly
producing a aboveground biomass early in the season, then spreading rapidly belowground
(Adams and Galatowitsch 2005). Reed canary grass then decreases species richness and diversity
(Kercher et al. 2004a, c).

         When grown vigorously on stream banks, reed canary grass increases sediment deposits,
which limits water circulation. In wet sedge meadows, this also decreases the soil organic content
and microstructure (Lavergne and Molofsky 2004).
         Reed canary grass exhibits plastic growth strategies depending upon environmental
conditions including anoxia (Galatowitsch et al. 1999). Height, size and shape are not necessarily
correlated with geographic distribution (Apfelbaum and Sams 1987). Since Eurasian and North
American types cannot be readily distinguished, more genetic information is needed to determine
if planted stocks for forage and erosion control are contributing to invasion (Galatowitsch et al.
         Reed canary grass control measures are costly (Apfelbaum and Sams 1987). Multiple
methods exist to control it, however entirely effective control methods are unknown. Some
mechanical methods include covering with a mat to block sunlight, burning, mowing, discing or
plowing (Derr 2008, WDNR). Some chemical control measures include use of glyphosate in
aquatic habitat, or DaplonTM or Trichloroacetate (TCA)TM in non-aquatic habitat (WDNR).
Burning has not been shown to be effective in reducing aboveground biomass, though it did
reduce seed bank (Adams and Galatowitsch 2008). Fire is not very productive in helping native
species reinvade, and reed canary grass rapidly recolonizes (Foster and Wetzel 2005). Herbicide
has been shown to reduce reed canary grass for a short time if applied appropriately (Adams and
Galatowitsch 2006, Foster and Wetzel 2005). Additional techniques such as planting cover crops
and adding sawdust amendments to reduce light (Iannone and Galatowitsch 2008), and the use of
new types of herbicides (Healy 2009) have been tried with limited results.
         Management efforts should be concentrated during early stages of reed canary grass
establishment (Reinhardt and Galatowitsch 2005). Strategies that mimic natural disturbances such
as fire or flooding will likely not be effective (Galatowitsch et al. 1999).


Adams, C. R., and S. M. Galatowitsch. 2005. Phalaris arundinacea (reed canary grass): Rapid
        growth and growth pattern in conditions approximating newly restored wetlands.
        Ecoscience 12:569-573.
Adams, C. R., and S. M. Galatowitsch. 2006. Increasing the effectiveness of reed canary grass
        (Phalaris arundinacea L.) control in wet meadow restorations. Restoration Ecology
Apfelbaum, S. I., and C. E. Sams. 1987. Ecology and control of reed canary grass (Phalaris
        arundinacea L.). Natural Areas Journal 7:69-74.
Burvall, J. 1997. Influence of harvest time and soil type on fuel quality in reed canary grass
        (Phalaris arundinacea L.). Biomass and Bioenergy 12:149-154.
Derr, J. F. 2008. Common Reed (Phragmites australis) response to mowing and herbicide
        application. Invasive Plant Science and Management 1:12-16.
Evans, M. W., and J. E. Ely. 1941. Growth Habits of Reed Canary Grass. Journal of the American
        Society of Agronomy 33:1017-1027.
Finell, M., C. Nilsson, R. Olsson, R. Agnemo, and S. Svensson. 2002. Briquetting of fractionated
        reed canary-grass for pulp production. Industrial Crops and Products 16:185-192.
Foster, R. D., and P. R. Wetzel. 2005. Invading monotypic stands of Phalaris arundinacea: A test
        of fire, herbicide, and woody and herbaceous native plant groups. Restoration Ecology

Galatowitsch, S. M., N. O. Anderson, and P. D. Ascher. 1999. Invasiveness in wetland plants in
        temperate North America. Wetlands 19:733-755.
Geber, U. 2002. Cutting frequency and stubble height of reed canary grass (Phalaris arundinacea
        L.): influence on quality and quantity of biomass for biogas production. Grass and Forage
        Science 57:389-394.
Green, E. K., and S. M. Galatowitsch. 2002. Effects of Phalaris arundinacea and nitrate-N
        addition on the establishment of wetland plant communities. Journal of Applied Ecology
Healy, M.T. 2009. Adaptive restoration of wetlands dominated by reed canary grass (Phalaris
        arundinacea): a case study for integrating research with land care and restoration. M.S.
        Thesis. Nelson Institute. University of Wisconsin – Madison.
Herr-Turoff, A., and J. B. Zedler. 2005. Does wet prairie vegetation retain more nitrogen with or
        without Phalaris arundinacea invasion? Plant and Soil 277:19-34.
Kading, H., G. Weise, W. Kreil, and K. D. Robowsky. 1990. The forage quality of reed canary
        grass (Phalaris arundinacea L) at grassland sites. Archives of Agronomy and Soil Science
Kercher, S. M., A. Herr-Turoff, and J B. Zedler. 2007. Understanding invasion as a process: The
        case of Phalaris arundinacea in wet prairies. Biological Invasions 9:657-665.
Kercher, S. M., and J. B. Zedler. 2004a. Multiple disturbances accelerate invasion of reed canary
        grass (Phalaris arundinacea L.) in a mesocosm study. Oecologia 138:455-464.
Kercher, S. M., and J. B. Zedler. 2004b. Flood tolerance in wetland angiosperms: a comparison of
        invasive and noninvasive species. Aquatic Botany 80:89-102.
Kercher, S.M., Q.J. Carpenter, and J.B. Zedler. 2004c. Interrelationships of hydrologic
        disturbance, reed canary grass (Phalaris arundinacea L.), and native plants in Wisconsin
        wet meadows. Natural Areas Journal 24:316-325.
Lamptey, J. N. L., R. T. Plumb, and M. W. Shaw. 2003. Interactions between the Grasses
        Phalaris arundinacea, Miscanthus sinensis and Echinochloa crus-galli, and Barley and
        Cereal Yellow Dwarf Viruses. Journal of Phytopathology-Phytopathologische Zeitschrift
Landstrom, S., L. Lomakka, and S. Anderson. 1996. Harvest in spring improves yield and quality
        of reed canary grass as a bioenergy crop. Biomass and Bioenergy 11:333-341.
Lavergne, S., and J. Molofsky. 2004. Reed canary grass (Phalaris arundinacea) as a biological
        model in the study of plant invasions. Critical Reviews in Plant Sciences 23:415-429.
Miller, T. W., L. P. Martin, and C. B. MacConnell. 2008b. Managing reed canarygrass (Phalaris
        arundinacea) to aid in revegetation of riparian buffers. Weed Technology 22:507-513.
USDA (USDA, NRCS PLANTS Database). ―Phalaris arundinacea‖. National Plant Data Center,
        Baton Rouge, LA 70874-4490 USA. n.d. Web. October 2009. <
WDNR (Wisconsin Department of Natural Resources). ―Reed Canary Grass Factsheet‖. USGS
        Northern Prairie Wildlife Research Center. 2006. Available online.

Supplementary references from Healy (2008):

Adams, C. R., and S. M. Galatowitsch. 2006. Increasing the effectiveness of reed canary grass
        (Phalaris arundinacea L.) control in wet meadow restorations. Restoration Ecology
Basil V. Iannone Iii, S. M. G. 2008. Altering Light and Soil N to Limit Phalaris arundinacea
        Reinvasion in Sedge Meadow Restorations. Restoration Ecology 16:689-701.
Berg, T. 1982. Seed dormancy in local populations Of Phalaris arundinacea L. Acta Agriculturae
        Scandinavica 32:405-409.
Craft, C., K. Krull, and S. Graham. 2007. Ecological indicators of nutrient enrichment, freshwater
        wetlands, Midwestern United States (US). Ecological Indicators 7:733-750.
Geber, U. 2000. Nutrient removal by grasses irrigated with wastewater and nitrogen balance for
        reed canarygrass. Journal of Environmental Quality 29:398-406.
Geber, U. 2002. Cutting frequency and stubble height of reed canary grass (Phalaris arundinacea
        L.): influence on quality and quantity of biomass for biogas production. Grass and Forage
        Science 57:389-394.
Green, E. K., and S. M. Galatowitsch. 2001. Differences in wetland plant community
        establishment with additions of nitrate-N and invasive species (Phalaris arundinacea and
        Typha x glauca). Canadian Journal of Botany-Revue Canadienne De Botanique 79:170-
Herr-Turoff, A., and J. B. Zedler. 2007. Does morphological plasticity of the Phalaris arundinacea
        canopy increase invasiveness? Plant Ecology 193:265-277.
Howe, H. F. 1995. Succession and fire season in experimental prairie plantings. Ecology 76:1917-
Howe, H. F. 2000. Grass response to seasonal burns in experimental plantings. Journal Of Range
        Management 53:437-441.
Johnson, D. A., A. F. Rayment, and E. Donefer. 1976. Nutritive evaluation of reed canarygrass
        (Phalaris arundinacea L). Canadian Journal of Animal Science 56:838-838.
Kellogg, C. H., and S. D. Bridgham. 2004. Disturbance, herbivory, and propagule dispersal
        control dominance of an invasive grass. Biological Invasions 6:319-329.
Kilbride, K. M. a. F. L. P. 1999. Integrated pest management to control reed canarygrass in
        seasonal wetlands of southwestern Washington. Wildlife Society Bulletin 27:292-297.
Kim, K. D., K. Ewing, and D. E. Giblin. 2006. Controlling Phalaris arundinacea (reed
        canarygrass) with live willow stakes: A density-dependent response. Ecological
        Engineering 27:219-227.
Klimesova, J. 1994. The effects of timing and duration of floods on growth of young plants of
        Phalaris arundinacea L And Urtica dioica L - an experimental study. Aquatic Botany
Larsson, S. 2006. Supply curves of reed canary grass (Phalaris arundinacea L.) in Västerbotten
        County, northern Sweden, under different EU subsidy schemes. Biomass and Bioenergy
Larsson, S. H., M. Thyrel, P. Geladi, and T. A. Lestander. 2008. High quality biofuel pellet
        production from pre-compacted low density raw materials. Bioresource Technology

Lavergne, S., and J. Molofsky. 2007. From the Cover: Increased genetic variation and
       evolutionary potential drive the success of an invasive grass. Proceedings of the National
       Academy of Sciences 104:3883-3888.
Lefor, M. W. 1987. Phalaris arundinacea L (Reed Canary Grass - Gramineae) As A Hydrophyte
       In Essex, Connecticut, USA. Environmental Management 11:771-773.
Leger, E. A. 2008. The adaptive value of remnant native plants in invaded communities: an
       example from the Great Basin. Ecological Applications 18:1226-1235.
Lindig-Cisneros, R., and J. B. Zedler. 2002b. Relationships between canopy complexity and
       germination microsites for Phalaris arundinacea L. Oecologia 133:159-167.
Morrison, S. L., and J. Molofsky. 1998. Effects of genotypes, soil moisture, and competition on
       the growth of an invasive grass, Phalaris arundinacea (reed canary grass). Canadian
       Journal of Botany-Revue Canadienne De Botanique 76:1939-1946.
Morrison, S. L., and J. Molofsky. 1999. Environmental and genetic effects on the early survival
       and growth of the invasive grass Phalaris arundinacea. Canadian Journal of Botany-
       Revue Canadienne De Botanique 77:1447-1453.
Paveglio, F. L., and K. M. Kilbride. 2000. Response of vegetation to control of reed canarygrass
       in seasonally managed wetlands of southwestern Washington. Wildlife Society Bulletin
Perkins, T. E., and M. V. Wilson. 2005. The impacts of Phalaris arundinacea (reed canarygrass)
       invasion on wetland plant richness in the Oregon Coast Range, USA depend on beavers.
       Biological Conservation 124:291-295.
Relyea, R.A. 2005. The impact of insecticides and herbicides on the biodiversity and productivity
       of aquatic communities. Ecological Applications 15:618-627.
Relyea, R.A. 2005. The leathal impact of roundup on aquatic and terrestrial amphibians.
       Ecological Applications 15:1118-1124.
Vymazal, J., and L. Kropfelova. 2005. Growth of Phragmites australis and Phalaris arundinacea
       in constructed wetlands for wastewater treatment in the Czech Republic. Ecological
       Engineering 25:606-621.
Wisconsin Reed Canary Grass Management Working Group. 2009. Reed Canary Grass (Phalaris
       arundinacea) Management Guide: Recommendations for Landowners and Restoration
       Professionals. Available online. <http://www.ipaw.org/invaders/reed_canary_grass/RCG-

  Part 2. Meadow Vegetation
   Contributed by James Doherty, edited by Joy Zedler

Photos by Ray Barnes, taken during the 2009 vegetation survey.


Baseline knowledge about the composition and relative abundance of species in a plant
community is required to formulate relevant research questions, restoration strategies, and
management objectives. In this project, our objective in sampling the meadow was to provide a
―snapshot‖ of existing vegetation to guide the recommendations in this report and basic
management activities at GRPO. Below, we provide data on the frequency and abundance of
       • target species (the culturally significant species explicitly mentioned in the project goals
         and species profiles above),
       • common species (dominant species have the potential to strongly influence community
         structure and function), and
       • nuisance species (invasive and weed species that should be monitored and removed if
         they appear to suppress other species).

        The data presented below, and in the accompanying appendices, will serve as a starting
point for regular vegetation monitoring surveys at the GRPO meadow. Our approach was to
sample 1-meter-square quadrats in a regular grid across the entire meadow, excluding areas where
trees were present. Future monitoring efforts can be scaled up or down according to resource
availability and the concerns of managers, but each successive survey can, and should, be refined
by the information gained in previous surveys.


From July 5th to July 7th 2009, Deb Pomroy (Assistant Scientist, Olga Lakela Herbarium,
University of Minnesota-Duluth) inventoried the meadow flora, recording all plant species she
encountered. Pomroy collected specimens of taxa that were not readily identifiable and Ray
Barnes also photographed such species. Pomroy later used the specimens and photographs to
complete identification to the greatest possible taxonomic resolution (all taxa listed in Appendix
1). The floral inventory survey augmented the quadrat-based sampling described below.
        From July 6th to July 9th 2009, James Doherty and Brandon Seitz established a 10 x 10-m
sampling grid by setting flags 10 m apart along the road adjacent to the meadow, then walking a
perpendicular line into the site and setting flags 10 m apart along the way. Seitz used a Brunton
compass mounted on a tripod to keep Doherty on a consistent bearing. Doherty used a Sonin
electronic distance-measuring tool (or measuring tape when lines of sight were obscured) to place
flags at proper distance intervals from the road. In total, the grid had 202 points. At each point,
Brandon Seitz used a high-precision Trimble GeoExplorer Global Positioning System (GPS) to
determine position and elevation. The GPS data were post-corrected and quality-assured by Seitz.
        At each of the 202 points, Doherty recorded the presence and cover of species within a
1x1-m quadrat, with no stipulation that plants be rooted inside the quadrat. Doherty recorded
abundance using four classes:
        high (species covering 76-100% of the plot),
        medium (species covering 51-75% of the plot),
        low (species covering 25-50% of the plot), and
        lower (species covering about 20-25% of the plot).

       Cover was estimated as the percent area of the quadrat where plants would block vertical
transmission of light to the soil. The cover of understory leaves (beneath the top layer of
vegetation) were assessed independently, thus, multi-layered vegetation could have total cover
>100%. Species over 20% cover were identified and recorded as present, but species under 20%
cover were not recorded because rare species make up the numerical majority of most plant
communities, and their identification and cover-estimation adds considerable time to sampling.
       Species of special interest were always recorded as present, regardless of cover, to provide
more information about their populations. Species of interest for restoration were:
       northern sweetgrass (Hierochloe hirta),
       sunchoke (Helianthus tuberosa),
       caraway (Carum carvi),
       wild chives (Allium sibiricum),
       Greene‘s rush (Juncus greenei),
       Vasey‘s rush (Juncus vaseyi), and
       green bulrush (Scirpus atrovirens; which often co-occurred with sweetgrass in the field).

Species of interest for potential eradication were:
       cow vetch (Vicia cracca),
       reed canary grass (Phalaris arundinacea),
       Canada thistle (Cirsium arvense),
       smooth brome (Bromus inermus), and
       tree and shrub species that might hinder restoration to an open, fire-dominated grassland.

        To ensure that the cover estimation was repeatable, Doherty, Seitz, Pomroy and Barnes
conducted the first 19 plots as a team. After discussing the species that all thought had >25%
cover, they independently assigned each species a cover ranking and then stated their estimates in
tandem until they reached consensus on the cover class. The remaining plots were sampled by
Doherty. At the end of each sampling day (July 6, 7, 8, and 9), Doherty re-sampled the first plot
that he had sampled that day, checking for consistent cover class assignments. On the rare
occasion when he differed by a cover class for a species, he used the second-sample data. As
needed during sampling, Pomroy and Seitz helped Doherty identify unknowns (to genus, if not to


Elevation ranged from approximately 183 m to 192 m (Universal Transverse Mercator) across the
meadow, with topographic heterogeneity including depressions and a general slope toward Lake
Superior (Figure 1.1). Across all points taken, the mean and standard deviation associated with
vertical precision was 0.16 ± 0.12 m with a range of 0.1 – 0.9 m, and for horizontal precision 0.13
± 0.07 m with a range of 0.1 – 0.5 m; precision is shown by point in Appendix 2.

Figure 1.1. Map of elevations of the GRPO meadow performed with ArcMap GIS software. Elevation
classes were forced, such that breaks occurred at whole-meter intervals. Note that initial sampling used a
5-m grid. This was later changed to a 10-m grid in order to ensure completion in the amount of time

       Pomroy identified 108 taxa in the meadow, and Doherty encountered 86 of those taxa
within the 202 sampled plots (sampled plots shown in Appendix 3). Species differed in their
mean elevation and in the range of elevations where they occurred (Figure 1.2 and 1.3).

                Juncus greenei
            Lotus corniculatus
          Eurybia macrophylla
                      Salix sp.
            Equisetum arvense
      Epilobium angustifolium
             Solidago gigantea
              Prenanthes alba
             Ranunculus acris
               Cornus sericea
         Mertensia paniculata
                Poa pratensis
                  Carum carvi
         Taraxicum officianale
           Heracleum lanatum
           Fragaria virginiana
                 Rubus ideaus
          Trifolium hydbridum
             Agropyron repens
     Calamagrostis canadensis
              Phleum pratense
             Agrimonia striata
           Thalictrum dioicum
             Rubus pubescens
            Trifolium pratense
                  Vicia cracca
              Bromus inermis
                  Rosa blanda
        Doellingeria umbellata
              Vicia americana
          Solidago canadensis

                                 185    186        187        188         189        190           191
                                                         Elevation (m)

     Figure 1.2. Mean elevation of species occurring in more than 10 sampled plots. Error bars =
     standard deviation.

     Hierochloe hirta = odorata
             Scirpus atrovirens
          Populus tremuloides
                Juncus greenei
                Cornus sericea
          Helianthus tuberosus
                    Cornus sp.
                      Picea sp.
          Populus balsamifera
                Cornus rugosa
                   Carum carvi
        Phalaris arundinaceae
             Agropyron repens
             Betula papyrifera
                   Vicia cracca
               Cirsium arvense
               Bromus inermis

                                  185   186    187        188         189      190       191        192
                                                          Elevation (m)

       Figure 1.3. Mean elevation of “species of interest” in sampled plots. Error bars indicate
       standard deviation.

       The above summaries of vegetation are complemented by maps showing the distribution
and abundances of select species within the meadow (Appendices 4, 5, and 6).
       Mean cover (visually estimated in coarse cover classes) was much more evenly distributed
among species than was frequency (Table 1). Typical native plant communities have far fewer
dominants than rare species, but high evenness may be an artifact of sampling only relatively
abundant species (those with cover >20%). The most commonly occurring species, the native
Rosa blanda, was found in half of the plots sampled in this meadow. The exotic Poa pratensis
was a close second. The species with the highest mean cover, as estimated by cover classes, was
Lotus corniculatus, an invasive, exotic, weedy species. The other commonly occurring species
that was classified as an invasive, exotic, weedy species was Bromus inermis, which was already
recognized as a species of concern, though many native species coexist with it in this site.

Table 1. Frequency, mean cover (from cover class midpoints), and invasive/exotic/weed status of 30 species
occurring in >10 of the 202 sampled plots. Three infrequent species are also listed as unwanted; as they have
potential to become invasive or weedy at GRPO.

Common species                    Frequency         Mean          Listed as        Listed as      Listed as
                                  (# of plots       cover        invasive by       exotic by      weedy by
                                  occupied)          (%)          MN-DNR            USDA           USDA
Rosa blanda                          100              39              N                N              N
Poa pratensis                         85              30              N                Y              Y
Equisetum arvense                     77              30              N                N              Y
Carum carvi                           71              31              N                Y              Y
Phleum pratense                       65              32              N                Y              Y
Ranunculus acris                      60              24              N                Y              Y
Vicia Americana                       49              25              N                N              N
Rubus pubescens                       44              42              N                N              N
Lotus corniculatus                    43              47              Y                Y              Y
Calamagrostis canadensis              42              37              N                N              N
Fragaria virginiana                   35              29              N                N              Y
Solidago gigantean                    34              26              N                N              N
Cornus sericea                        31              31              N                N              N
Rubus idaeus                          31              32              N                Y              N
Bromus inermis                        29              35              Y                Y              Y
Thalicrum dioicum                     27              36              N                N              N
Agropyron repens                      26              30              N                Y              Y
Doellingeria umbellata                22              25              N                N              N
Trifolium pratense                    22              33              N                Y              N
Taraxacum officinale                  20              24              N                Y              Y
Solidago canadensis                   18              32              N                N              Y
Vicia cracca                          18              20              Y                Y              N
Epilobium angustifolium               17              26              N                N              Y
Heracleum lanatum                     17              30              N                N              Y
Mertensia paniculata                  17              33              N                N              N
Eurybia macrophylla                   16              31              N                N              N
Agrimonia striata                     14              33              N                N              N
Juncus (cf.) greenei                  14              41              N                N              N
Prenanthes alba                       13              26              N                N              N
Trifolium hybridum                    11              28              N                Y              N
Infrequent unwanted spp.
Cirsium arvense                        7              23               Y               Y                Y
Phalaris arundinacea                   2               3               Y               Y                Y
Hieraceum aurantiacum                  1              23               Y               Y                Y

The GRPO meadow vegetation is a novel collection of native (19 common spp.) and nonnative
(11 common spp.) plants, most of which are widespread in distribution throughout the meadow.
The unique composition of the meadow may be owed to its long history of human use, and is not
necessarily an obstacle to future restoration and management there (one target species, caraway, is
itself exotic). Long-term monitoring should indicate whether or not co-existence between native
and non-native species is sustainable. If exotic or weedy species increase in frequency and
abundance across sampling periods (based on monitoring data), or if such species are already
relatively common in the site, managers should take action to remove them. In particular this
survey indicates that several invasive species of potential concern (Lotus corniculatus,
Ranunculus acris, Bromus inermus, and others) and several abundant shrub species that might be
removed if monitoring data indicate shrub encroachment.
        Based on this survey, the area with sweetgrass and green bulrush on the upper slope is
likely an area of groundwater seepage with low nutrients and moist soil. That environment may
favor sweetgrass, and, despite co-occurring in only two quadrats, sweetgrass and green bulrush
generally seem to occur near each other (see species maps in Appendix 4). In future efforts to
restore sweetgrass, green bulrush could potentially be used as an indicator or facilitator or
sweetgrass, along with Greene‘s rush and Vasey‘s rush.
        Finally, there are several ways in which future monitoring efforts could build on the
survey discussed above or adapt survey methods to fit resource constraints:

       • Our sampling points, which were recorded with GPS, should be re-used to avoid
         confounding changes in vegetation over time with change in sampling location, and also
         to save time and effort
       • The survey can be target-species-centric. We adjusted sampling for a subset of species
         (i.e., we always recorded target species presence regardless of whether or not it had
         >20% cover in the plot). The same approach can be used to focus data-collection on
         invasive or undesired species in-lieu of, or in addition to, a full survey of the vegetation.
       • More repeatable measures of vegetation could be used. We opted to assess the cover of
         understory (under-canopy) species, which added some effort and complication to the
         cover estimation process. A more straight-forward approach would be to estimate cover
         from above the herbaceous canopy and assign a cover for each species from 0-100% or
         with cover classes, such as: 0-25%, 26-50%, 51-75%, 76-100%.
       • Sampling could be scaled down as needed. Any monitoring is better than no monitoring.
         In a given year it may only be possible to re-visit the GPS points where target species or
         invasive species were present in 2009 and assess presence-absence. Still, that effort
         would answer a basic question: is that species‘ population fairly persistent?

         Thus, the vegetation survey above does not need to be repeated exactly to generate useful
monitoring data. The sampling methods used in 2009 could be changed to improve the quality or
utility of monitoring data collected, and still allow GRPO managers to track changes in vegetation
over time.

Part 3. Adaptive Restoration and
       Management Plans

  Satellite image of GRPO meadow. Provided by Brandon Seitz, NPS.

                                        Contents of Part 3

 A broad view of restoration
 The restoration goals
 Management opportunities and constraints

Adaptive restoration of four culturally-important species
 Monitoring to sustain the meadow vegetation
     - Meadow composition
     - Caraway monitoring and management
     - Reed canary grass eradication
     - Smooth brome control

Phase 1: Garden experiments
       Experiment 1.1: Test of caraway‘s effect on other target species
       Experiment 1.2: Test of chives‘ light limitation
       Experiment 1.3.1: Test of sunchoke light limitation
       Experiment 1.3.2: Test of sunchoke nutrient limitation
       Experiment 1.4.1: Test of sweetgrass seedling establishment
       Experiment 1.4.2: Test of sweetgrass facilitation by Greene‘s rush

Phase 2: Meadow-edge experiments
       Experiment 2.1: Test of chives and sunchoke light limitation

Phase 3: Adaptive restoration within the meadow
       Experiment 3.1: Test of young sunchoke requirement for light
       Experiment 3.2: Comparison of methods to expand the existing sweetgrass population
       Experiment 3.3: Test of sweetgrass establishment following tree removal
       Experiment 3.4: Test of sweetgrass facilitation by Greene‘s rush

Phase 4: Adaptive management within the meadow
       Experiment 4.1: Comparison of caraway harvest methods
       Experiment 4.2: Comparison of site-scale management activities
       Suggested future study: Managing meadow drainage

Summary of experiments by phase and priority

Discussion and Recommendations
Landscape restoration
       - Multifunctionality
       - Heterogeneity of spatial pattern
       - Restore and sustain biodiversity of the existing meadow vegetation
Summary of recommendations for monitoring and restoration
Highlighting ecological functions: Education and community involvement

References cited


This document lays out an adaptive restoration process and plan for restoring four culturally-
important plant species to a 3.5-acre meadow at Grand Portage National Monument, northeastern
Minnesota. The four culturally-important species are
       sweetgrass (Hierochloe hirta),
       sunchokes (a.k.a. jerusalem artichoke; Helianthus tuberosus),
       chives (Allium schoenoprasum var. sibiricum), and
       caraway (Carum carvi).

       Two additional native plant species might assist sweetgrass restoration; these are
       Greene‘s rush (Juncus greenei) and
       Vasey‘s rush (Juncus vaseyi).

       Because restoration efforts will be hindered by unwanted aggressive plants that are
already present in the meadow, the plan also addresses two invasive species; these are
       smooth brome (Bromus inermis) and
       reed canary grass (Phalaris arundinacea).

        Of these, the more widespread invader is smooth brome. This grass spreads both by seed
and aggressive rhizomes. The rhizomes establish early in the growing season and allow it to out-
compete native grasses (USDA).
        Relevant literature on each of the above eight species is summarized in Part 1, ―Species
Profiles.‖ Of the four culturally-important species, sweetgrass is highly valued for its use in
ceremonies and crafts, while chives, caraway and sunchokes are edible.
        Sweetgrass is widely used by American Indians as incense for ceremonies and many
medicinal purposes including the treatment of coughs, sore throats, chafing, venereal infections,
vaginal bleeding, and to expel afterbirth (Foster 2000). Additional uses include perfume, hair
wash, bedding, and basketry (USDA).
        Cultivated chives (Allium schoenoprasum var. schoenoprasum) and wild chives (var.
sibiricum) are herbs that have been grown widely for use in flavoring food, medicine, religious
ceremonies and ornamentation (Fenwick and Hanley 1985), as well as appetite stimulation and
digestive aid (USDA). The Ojibwe and Sioux people used chives in their native wild rice dishes,
called ―Mahnomem‖ (Ojibwe name for wild rice).
        Caraway is a spice with a long history of use as a household remedy, especially in treating
digestive complaints. Crushed seeds can be brewed into teas to soothe digestion via its
antispasmodic properties and its carminative action, which relieves bloating. It is often added to
laxative medicines to prevent griping pain. The edible seeds have long been used to flavor sweet
and savory foods. The seeds are high in fat and protein (Plants for a Future 1996).
        Sunchokes (Jerusalem artichoke) is a staple that is among the oldest cultivated crops in
North America (first documented use in 1605). Sunchokes appear to have originated in Canada,
where its historical distribution was broader than at present. Its center of origin is difficult to trace
because it was highly cultivated for perhaps centuries and because it is not easy to track
genetically (Kays and Nottingham 2008).

A broad view of restoration

        Over recent decades, ecological restoration goals have evolved from specific to broad
targets as practitioners and scientists developed deeper understanding of the challenges involved
in aiming for specific ―historical states.‖ In the United States, the public land surveys of the mid-
1800s offered quantitative data that were used as community-composition targets for woody
vegetation at the time of European settlement. But these early records were time-specific and not
representative of the full range of dynamics that native vegetation experience (Pickett and Parker
1994). The records also focused on trees, making it difficult to set targets for the herbaceous and
animal components of biotic communities. Furthermore, the long time periods needed to
reproduce forests and savannas and the difficulty of mimicking either the composition or size
distributions indicated by the records made the ―pre-settlement model‖ unattractive or unsuitable
for many projects.
        Pre-settlement composition is also an inadequate target for those who aim to restore
ecosystem services, such as flood abatement or carbon sequestration. And because the Clean
Water Act of the early 1970s recognized the water-quality-improvement services of wetlands, the
mitigation of lost wetland area quickly focused on rapid restoration of non-woody vegetation in
ponds, marshes, and wet meadows.
        Meanwhile, in Europe, a different type of restoration effort emerged, namely, the
restoration of cultural landscapes with culturally-valuable species. In Switzerland, for example,
Andreas Gigon (personal communication while a guest speaker at UW; see also Gigon and
Langenauer 1998) advocated the conservation and restoration of Sibirian iris to ensure that
historic settlements—both human structures and associated vegetation—would be sustained for
their heritage value. GRPO‘s interest in conserving the cultural history and associated culturally-
important plant species is a similar approach.

The restoration goals

        The restoration goals for this project at Grand Portage National Monument (GRPO) focus
on culturally-valuable species, with additional elements of composition- and ecosystem-service-
based targets. Each of the four culturally-important targets is valued for a different purpose
(ceremonial and craft usage, herb, spice, and staple). Their restoration to GRPO is not proposed
as garden plots of plantations but re-incorporation into an existing vegetated meadow, which in
itself provides a culturally-important service as the historical site of portaging, which gives the
monument its name. Other ecosystem services of the meadow likely include groundwater
conservation, erosion control, and sequestration of carbon in the soil.
        This unique combination of targets is proposed for a unique site, namely a 3.5-acre
meadow that is neither a garden nor a wildland; instead, the restoration site has been modified by
humans for centuries, dating back to the earliest explorers and fur traders. Some of the valued
species are still present in the meadow. One (caraway) is weedy and the others are rare.
        The restoration of these four culturally-important species is entirely consistent with the
mission of Grand Portage National Monument, which ―protects, commemorates, and interprets a

reconstructed fur depot of the North West Company, a rendezvous site for international commerce
and canoe route for transcontinental exploration, Native heritage, natural scene and history of
cross cultural contact and accommodation between traders, Ojibwa and other participants in the
fur trade‖ (Grand Portage National Monument/Minnesota: Final General Monument
Plan/Environmental                     Impact                    Statement                    -

Management opportunities and constraints

        The GRPO staff has indicated willingness to undertake restoration experimentally, as well
as to use fire, mowing, and herbicides in managing the meadow. Volunteers could be engaged to
grow plants in pots and manage weeds. Experiments in gardens and in the meadow would,
however, require continual maintenance and intensive care. The benefit would be improved
understanding of conditions suitable for each species‘ growth. The costs would be space and time
devoted to growing plants under experimental conditions. In the meadow, the tending of field
experiments might require mowed trails to allow data collection in experimental plots. Because
sweetgrass is a desirable species, plants needed to be protected until they can sustain harvesting.
        We recommend an education program, along with garden and meadow experimentation to
ensure that local people become interested in the project and that visitors become aware of efforts
to sustain culturally important species. GRPO could use signs, displays within the visitor center,
and possibly a recipe book that indicates how chives, sunchokes and caraway have been used
historically and can be used again in a sustainable manner. Without interest by local people in the
culturally important species, any plan to sustain them in the meadow could fail. Suggestions for
an outreach program appear at the end of this plan.

             Adaptive restoration of four culturally-important species
This adaptive restoration plan aims to manage the existing population of caraway and restore and
sustain populations of the rare chives, sunchokes, and sweetgrass in the GRPO meadow.
Caraway needs to be managed so it does not become overly abundant, and the three rare species
need to be augmented to achieve sustainable populations. We recommend an adaptive approach,
because there are several unknowns that need to be known in order to sustain these four species at
desired levels. The following unknowns can be answered via experimentation in the garden,
meadow edge, and meadow:

       • Caraway: When would caraway need to be controlled? How could caraway seeds be
        harvested to benefit GRPO and local people without hindering restoration efforts?

       • Chives: How much light is needed for plantings to survive and reproduce?

       • Sunchokes: Does light limit growth of young sunchokes? Would nutrient addition or
        mowing facilitate restoration of sunchokes?

       • Sweetgrass: How can seeds be germinated in order to grow more genotypes for
        subsequent vegetative propagation? Which environmental and biotic factors restrict the
        expansion of existing patches? Where and how can GRPO establish new patches in the

        Below we outline our recommendations for monitoring the meadow vegetation and
establishing four phases of experimentation. The monitoring component will be developed
primarily on the basis of information presented in Part 2 of this report. The experiments will be
developed by indicating: the hypothesis being tested and the reason it should be tested, the
priority for conducting the experiment, factors that should be constant, factors to be varied (i.e.,
treatments), measures to be taken, methods, and analysis. The four phases of experimentation are:

       (1) Experimentation in the garden to determine the environmental factors that restrict
           growth of the three rare species and to provide propagules for restoration, followed by,

       (2) Experimentation at the meadow edge to serve as further tests of pant growth,
           demonstration exhibits, and plots that could sustain some level of harvesting,

       (3) Experimental restoration of three rare species in the meadow, and

       (4) Experimental harvest of caraway and management at the site scale.

        We recommend that monitoring and experimentation be overseen by an ―Adaptive
Restoration Task Force‖ (hereafter, Task Force), which would supervise data collection, interpret
and archive the data, and use the information to guide the overall restoration effort at GRPO. We
propose that GRPO begin by monitoring the meadow vegetation by routinely resampling the plots
established by James Doherty in 2009. Next, we propose the testing of factors that influence each
species using experimental plantings in the garden. A second phase of experimental plantings is

proposed for the meadow edge, where species that require high light conditions could be
established to examine their growth requirements and illustrate the outcomes for staff, as well as
visitors. The plots with the most vigorous growth could then be harvested and used for more
extensive plantings. A third phase of experiments would then test reintroduction and
reestablishment procedures within the meadow. Finally, a fourth phase of experiments would
help guide eventual management of individual target species by testing the effect of
ethnobotanical use and eventual management of the site by comparing the effect of large-scale
management strategies on the vegetation as a whole.
        If the experiments are undertaken in sequence, over several years, the Task Force would
need to meet semi-annually to guide the collection of data, interpret the findings, and decide how
to use the findings from each phase to guide the next. For example, results from the garden
would indicate how best to set up the meadow-edge phase, and results from the meadow-edge
would help guide subsequent, larger-scale experimental plantings.

Monitoring to sustain the meadow vegetation

       - Meadow composition

        We recommend regular resampling of the 202 plots (each 1 m2) that were sampled in 2009
by James Doherty and Deb Pomroy. To track changes in the vegetation, annual sampling can
focus on target species, with special attention to three weedy species (caraway, smooth brome,
and reed canary grass) and three rare species (chives, sunchokes and sweetgrass). Volunteers
might be easily trained to identify these six species, and perhaps count of the number of other
species (without identifying them all). More detailed sampling of every species in every plot
could be done at 5-yr intervals by expert botanists. The 2009 vegetation sample would serve as
the reference point for subsequent expansion of invasive species (annually) and the contraction of
native species distributions and overall status of biodiversity (every 5 years).
        GRPO staff could use GPS equipment or install stakes or flags as visible markers for both
the 202 permanent plots and locations of invasive species and experimental plots. If visible
markers are placed during the first treatment in spring, subsequent measurements of progress
would be simplified. At a minimum, weedy species should be assessed once in mid-season to
allow mid-course correction of weed-control measures. A second measurement near the end of the
growing season would help the Task Force decide on management approaches for the next spring.
        Annual data summaries will allow the Task Force to track the abundances of each of the
four culturally important species and unwanted invasive plants. The patterns of change would
then inform the restoration process. For example, if the caraway population were to expand for
several successive years, the Task Force would want to consider control measures, and if invasive
species rapidly increase their populations at the expense of native species, the Task Force could
decide when and how to take action.
        At the same time, annual information on target species distributions will help the Task
Force judge the effectiveness of restoration actions. While each experiment that we propose
includes procedures for assessment of outcomes of alternative restoration approaches, the
meadow monitoring will indicate whether reintroduced species are spreading beyond the
experimental plantings. Expansion would be a strong indicator that restoration is effective. The

Task Force could then adjust the restoration program or increase effort to employ the most
effective approach more extensively.
        Chives, sunchokes and sweetgrass would be considered ―restored to the meadow‖ when
populations are judged to be ―sustained.‖ That judgment will require data for several successive
years across variable weather conditions (dry years and wet years). It will also require data that
show whether invasive species are under control (populations being reduced and any new patches
able to be eradicated). If monitoring data indicate that the initial methods are not providing the
desired outcomes (i.e., rare species are not sustained and/or invasive species are not controlled),
the Task Force could adapt the management plan and apply more aggressive measures (e.g., more
plantings in a wider variety of environmental conditions).

       -   Caraway monitoring and management

        We recommend that caraway be monitored--and managed if it shows signs of becoming
too abundant, i.e., expanding to the detriment of other culturally important species or desired
native plants. If the long-term monitoring of meadow vegetation indicates a consistently
expanding population both in area and cover, then GRPO staff should be ready with control
measures. Because this species is short-lived or biennial, seed-collection could be used to deplete
the seed bank; the proposed caraway-harvest experiment in Phase 4, below, lays out a way to
assess the effectiveness/sustainability of such harvest-based controls.
        The collection of seeds would be feasible if local volunteers were interested in using the
seeds in cooking or if seeds could be sold as a ―spice souvenir‖ at the visitor center. However,
widespread dissemination of seeds could inadvertently cause invasions elsewhere. An
informational label would be needed to explain why caraway might become invasive if seeds are
planted in gardens and not managed by clipping inflorescences or weeding outside the garden.
More certain, however, would be to provide non-viable seeds for cooking or souvenirs.
        Finding a way to provide non-viable seeds that are still flavorful, could simultaneously
train GRPO personnel in the adaptive approach: Ask which treatment will work, test alternatives,
evaluate results, and adopt the most effective treatment (or repeat the process if no treatment
provides the desired outcome). A microwave oven could be used to determine what
time/temperature setting prevents caraway seed germination while still retaining the characteristic
aroma and taste. Adopting the treatment that kills all seed and retains flavor would ensure that
customers do not introduce caraway into non-native areas.

       Testing ways to make seeds nonviable for use as a spice souvenir:

       Objective: To find time/temperature conditions sufficient to kill caraway seeds while
           retaining flavor. Volatile flavors could be lost by microwaving seeds at too high a
           temperature or for too long at a lower temperature.
       Hypothesis: Some combination of microwaving time and temperature will kill seeds
           without losing flavor.
       Priority: Low, because it is not yet clear that the caraway population needs to be
           controlled by collecting seeds.
       Constants: Seed harvested from the meadow or from garden plots.
       Variables: Temperature setting, time that seeds are microwaved.

       Measure: Count the number of seeds that germinate under each treatment; rank taste
           among treatments.
       Method: Microwave the seeds at low vs. high temperature and short vs. long duration
           (start with 4 treatments, 200 seeds per treatment, 800 seeds total—estimate number per
           volume and set treatments as tablespoon measures). The four treatments would be
           low+short, low+long, high+short and high+long. After treatment, have four
           volunteers taste the results and rank flavor from highest to lowest. Each volunteer
           would taste the same amount of seeds (quarter teaspoon?) from each treatment.
       Next, test germination rates of the remaining 400 seeds by placing 25 seeds from a
           treatment on moist paper toweling folded into a small, shallow, labeled container (e.g.,
           petri dish or jar lid with plastic wrap). Place labeled container under warm, well-lit
           conditions (e.g., windowsill). Four treatments, each with four replicate containers will
           result in 16 containers. Randomize positions weekly so that the replicates of a
           treatment are distributed within the growing space and not confounded by consistently
           higher light or warmer temperatures.
       Analysis: Sum the flavor rankings for each treatment to see if any treatment shows a
           strong loss of taste. Tabulate the number of seeds germinated over a 4-week time
           period and identify the treatment with the lowest germination rate. Did the difference
           in temperature suggest a greater effect than the difference in duration?
       If all treatments are still producing seedlings, higher temperatures or longer durations of
           microwaving are needed. The Task Force would need to meet to discuss how to repeat
           the experiment, i.e., which new treatments should be tested to identify lethal
       The optimal treatment would kill seeds but not degrade flavor.

       - Reed canary grass eradication

         Reed canary grass (Phalaris arundinacea) is an aggressive invader that will detract from
the sustainability of other important species at GRPO. Strains of reed canary grass have been
introduced from Europe for livestock forage and streambank erosion control, and the exotic
strains tolerate a wide range of hydrological conditions, thrive under frequent flooding, and
commonly invade disturbed wetlands (Galatowitsch et al. 1999). Reed canary grass thrives on
fertile, moist organic soil in full sunlight, but it also occurs in partially shaded areas and uplands
(WDNR). Its growth is accelerated by disturbance and increased levels of nitrogen (Green and
Galatowitsch 2002, Kercher and Zedler 2004a, Herr-Turoff and Zedler 2005).
         We recommend that reed canary grass be eradicated immediately, because it does not yet
have a large population at GRPO. If allowed to persist, it will reduce native plant growth (Green
and Galatowitsch 2002) by producing aboveground biomass early in the season, then spreading
vegetatively belowground (Adams and Galatowitsch 2005). As reed canary grass expands, it
decreases native species richness (Kercher et al. 2004c). Due to its ability to outcompete native
species, management efforts should be concentrated during early stages of reed canary grass
establishment (Reinhardt and Galatowitsch 2005). Now is the time to aim for eradication.
         Multiple control methods are available, but none is entirely effective. Mechanical methods
include blocking sunlight with a mat or black plastic, burning, mowing, disking or plowing (Derr
2008, WDNR). Chemical control measures include use of glyphosate in aquatic habitat, or

DaplonTM or Trichloroacetate (TCA)TM in non-aquatic habitat (WDNR). Herbicide has been
shown to reduce reed canary grass for a short time if applied appropriately (Adams and
Galatowitsch 2006, Foster and Wetzel 2005, Apfelbaum and Sams 1987). Healy and Zedler
(2010) found that sethoxydim (VantageTM) could stunt but not kill reed canary grass. Strategies
that mimic natural disturbances such as fire or flooding are unlikely to be effective (Galatowitsch
et al. 1999).
         Due to its small patch size in the GRPO meadow, the entire clone of reed canary grass
could be covered with black plastic for 2-3 years. Alternatively, an herbicide (e.g., glysophate, as
Round-Up) could be applied selectively to kill the invader. In either case, we recommend a late
spring burn to remove the senesced overwintering canopy, which could otherwise puncture the
plastic covering or absorb herbicide. The black plastic will ―solarize‖ seedlings or resprouts that
could develop if light penetrates tears in the plastic. Multiple years of treatment will be needed to
deplete reserves of rhizomes and dormant buds, as well as the seed bank, which will likely persist
for several years. Continued monitoring and spot treatment would help managers prevent re-
establishment of seedlings.
         If using herbicide, allow reed canary grass to regrow following control burning; then hand
treat it using the glove application technique (plastic glove inside a knit glove soaked in
herbicide). This will suppress the reed canary grass stems, but it will not kill seeds or dormant
buds belowground. Monitoring will be needed to identify and promptly treat new patches of reed
canary grass.
         Neither solarization nor glyphosate is a selective technique for reed canary grass control
(WIRCGWG 2009). A plan will be needed to introduce desired species after treatment is judged
to be effective. Aggressive, tall native plants with ample belowground biomass would be most
competitive. Sunchokes have these attributes and would be a suitable replacement ―matrix‖
species, adding understory plants to absorb light that passes through the sunchoke canopy.
         We do not recommend attempting to control reed canary grass using fire, mowing or other
methods that would increase light or add nutrients to the meadow. Burning increases light
penetration to the soil, which favors reed canary grass growth and expansion. Fire is useful in
removing the thatch prior to herbicide application, but used alone, burning cannot control reed
canary grass. More information is in the ―Species Profiles‖ document and the Reed Canary Grass
(Phalaris arundinacea) Management Guide (WIRCGWG 2009).

       - Smooth brome control

        Smooth brome is highly invasive in the Grand Portage meadow. In the 2009 vegetation
survey, it occurred in 29 (~14%) of the 202 surveyed plots. A suitable goal is to eradicate it to
allow culturally significant species to thrive in its place. Experimentation could delay treatment
and allow further expansion.
        The methods that decrease growth of smooth brome include clipping, mowing, and
chemical application early in spring, it sprouts during cool weather. If it sprouts earlier than other
desired species at GRPO, managers could capitalize on this trait by treating smooth brome as
soon as it emerges and before other species have emerged. We recommend early mowing to
reduce its competitive advantage, and subsequent application of glyphosate to kill resprouts. Due
to the wide distribution of smooth brome in the meadow, this would require more resources and
increased monitoring of survival, with follow-up hand spraying or glove application.

       Potential management strategies to eliminate smooth brome as quickly as possible are:
      Mow and clip the smooth brome in April or as soon as growth begins.
      Apply glyphosate in spring as a spot-treatment using the glove technique. Be cautious
       around other species.
      Repeat clipping, mowing, and herbicide as necessary

        Replanting following the control of smooth brome could be difficult if allelopathins
remain in the soil. Allelopathy is the production and release of certain biochemicals that affect
the growth and development of other organisms. Grant and Sallans (1964) observed that smooth
brome root extract inhibited root growth in the majority of legumes and its own root structure. It
is possible that the decomposing biomass releases allelopathins, but we did not find support for
the work of Grant and Sallans (1964). Removing the root system, as discussed in the ―Species
Profiles‖ document, would disrupt the surrounding plants and soil across much of the meadow.
Therefore, we do not recommend cultivation unless smooth brome cannot be controlled using the
above approach.

        - Other weedy species

        Three species of potential concern (Canada thistle, birdsfoot trefoil and exotic vetch)
could be addressed at the same time as smooth brome.
       Canada thistle (Cirsium arvense), from Eurasia, was present in only 7 quadrats and
              prominent in only 5.
       Birdsfoot trefoil (Lotus corniculata), from Europe, was one of the most common (43
              occurrences in 202 quadrats) and abundant (considered dominant in 32 quadrats)
              species of the meadow.
       Exotic vetch (Vicia cracca), from Eurasia, had 18 occurrences and was less common and
              had lower cover than the native (V. americana). V. cracca was considered a
              dominant in only 3 of the 202 survey quadrats.

        Because the above species are not yet widespread or widely dominant, their control will be
easier now than in the future. Long-term monitoring will help track their distributions and allow
the Task Force to decide if control efforts need to increase.

       Shrubs can also become invasive in meadows, including native shrubs. There are two
potentially increasing species in the GRPO meadow:
       Red-osier dogwood (Cornus sericea = C. stolonifera) is a native shrub that can be
                invasive in wetlands. It occurred 31 times in the 202 quadrats, being prominent in
                28 and dominant in 14.
       A native rose (Rosa blanda) was the most common species in the 202 quadrats (100
                occurrences, with prominence in all 100 quadrats and dominance in 69 quadrats.
                This species is unarmed (prickles only at the base).

        Without information that indicates increasing occurrences or dominance over time, neither
species can be considered invasive at GRPO. The preferred approach would be to track their
distributions annually, along with the target species, so the Task Force can evaluate the data and
consider the need for management. If shrub cover is shown to be increasing at the expense of
herbaceous plant diversity (species richness or evenness of cover), the Task Force could
recommend management to reduce woody plant cover. Management tools for shrubs include
control burning and/or mowing. Cutting and herbiciding would not likely be necessary unless a
new, highly aggressive, exotic shrub were to colonize the site.

        A caution is that exotic species are continually being introduced to restoration sites and
natural areas, dispersed by wind, animals, water, and people, not necessarily in that order. A
recent review (Catford et al. 2009) summarizes three general causes of species invasiveness:
―propagule pressure,‖ abiotic and biotic factors. A take-home message is that managers must be
on the alert for new invasions of exotic species and prepared to take action upon colonization.

                    Phase 1: Experimentation in the garden

        Caraway‘s effect on other species is not well understood. Its canopy is not dense, so it
might have little ability to compete for light. On the other hand, it forms a tap root that could
sequester belowground resources that would otherwise support other desired species.
        If this weedy species is shown to expand consistently (for multiple years) in the meadow,
it could negatively affect other culturally-important and native plant species at GRPO. A garden
experiment would improve understanding of caraway‘s effects on the structure and function of the

       Experiment 1.1: Test of caraway’s effect on other target species

       Hypothesis: Mature caraway plants will reduce growth of rare species.
       Priority: Low, because the need to manage the species is easily determined by
           monitoring. Because this is a biennial plant, management could simply involve seed
           collecting or fall burning before seeds are released.
       Objective: Test the effect of caraway on each rare species (chives, sunchokes, sweetgrass)
           to determine if it is necessary to reduce caraway abundance. Each experiment would
           run two years in order for caraway to reach maturity.
       Constants: Eighteen plots (each at least 1-m2) with caraway. Six plots for each rare
           species (i.e., caraway + chives, caraway + sunchokes, caraway + sweetgrass; x6
           replicates). If garden space is limited, the sweetgrass test would have priority. Each
           plot with caraway would receive the same number of seeds, e.g., 200.
       Variables: Caraway left to grow and caraway pruned (cut to ground level mid-season).
       Methods. Seed caraway into all 18 plots. Add a uniform number of ramets of the rare
           species to 6 randomly selected plots per species. Maintain uniform densities of the rare

            species and measure height (tallest stretched stem/leaf in each plot) and estimate cover
            (0-25, 25-50, 50-75, and 75-100% cover). In year two, randomly select three of the six
            plots with the same rare species and prune the caraway (cut to the ground level mid-
            season before flowering). Measure height and estimate cover of the rare species with
            and without caraway pruning.
         Analysis: Compare height and cover of rare species with and without caraway pruned. If
            rare species grow better where caraway is pruned, caraway could be pruned in the
            meadow where it co-occurs with rare species.

        We recommend annual evaluation of the meadow monitoring data for total occurrences
and percent cover of caraway. An increase in caraway and decrease in desired species within the
same plots will indicate the need for caraway control. Either seed collection or a fall control burn
would reduce caraway population expansion, once its soil seed bank is depleted.
        If the monitoring program indicates that the population of caraway is declining, measures
could be taken to preserve the genetic diversity that might be present. Seeds taken when the
population is large population could be stratified (refrigerated with moist paper toweling in sealed
plastic bags) for use in a future restoration effort, should the Task Force determine that caraway is
becoming too rare in the meadow. Seeds could be collected annually to replenish the stratified
reserve, as viability will likely decline during storage time. A long-term seed viability experiment
would determine if it is necessary to store seeds annually. For such an experiment, keep the older
seeds for several years, removing 100 each year to test germination.
        To build up a seed supply, set aside an area in the garden for at least two years, when he
biennial plants would mature and produce seeds for a ―reserve‖ population. If seeds remain viable
for multiple years in moist, cold storage, the reserve bank could be replenished less often. It is
possible, however, that long-term storage would select for a subset of genotypes.

        Chives have been present in the garden at GRPO but not in the meadow. In recent years,
chives have only been reported in one small patch within an equipment yard. Propagules will
need to be grown for transplantation experiments. GRPO‘s Heritage Gardener has cultivated
chives in the garden and her experience should guide propagation of the species.
        Soil in garden plots should be warm and well drained for optimal growth (Davies 1992).
Because chives are an ―edge species‖ that does not compete well for light, the garden plots will
require weeding (Davies 1992, Organ 1960). A shade-tolerance experiment would inform
subsequent efforts to reintroducing chives into the meadow, by suggesting sites for planting
chives and management the surrounding vegetation.

         Experiment 1.2: Test of chives’ light limitation:
         Hypothesis: Chives grow better in full sunlight than in shade.
         Priority: High, because the experiment will help guide plantings at the meadow edge
             while producing the necessary bulbs for transplantation.
         Objective: Compare growth of chives under three light levels.
         Constants: Plot or pot size, number of seeds or transplants in each plot or pot, temperature,
             soil, nutrients and water supply.

       Variable: Shade, i.e., none, one layer, and two layers of shade cloth.
       Measure: Growth of plants, assessed nondestructively as height (tallest stretched height in
          a pot), number of leaves, and flowering (number of flowering stalks) at the end of the
          growing season. Nurseries supply shade cloth with estimated percent shade; however,
          for comparison with conditions in the meadow, it would be useful to measure light
          under each shade cloth treatment using a light meter.
       Methods: Build frames for plots or use tomato cages to suspend shade cloth about 2 feet
          above each pot. Assuming that pots will be easier to use, plant one bulb in each 1-ft-
          dia pot. Randomly assign each pot to a shade-cloth treatment, with 5 pots of chives
          per treatment. Each pot should have its own cage and shade cloth, and the position of
          pots would be randomized monthly to ensure independence of experimental units.
          Assess plant responses as height, number of leaves, and number of flowering stalks per
          pot. Weight of seeds and bulbs at the end of the experiment would assess differences
          in reproduction. Retain seeds and bulbs for later introduction to the meadow.
       Analysis: Compare averages for each response variable to identify the treatment with
          greatest growth (height, leaf density) and reproduction (seeds and of bulbs).
          Reproduction potential will be of greatest interest initially, in order to find the
          conditions that produce the most propagules for transplantation to the meadow edge.


        Sunchokes are not widely distributed at GRPO. Only 2 occurrences were recorded in the
meadow during the 2009 vegetation survey of 202 plots. Sunchokes were considered ―dominant‖
in only one of those 1-m2 plots. Thus, to meet the restoration goals, sunchokes will need to be
reintroduced to the meadow in suitable places.
        Literature indicates that this species is best grown from tubers with 2-3 nodes and at least
3 eyes for each desired plant. It is best to plant in spring or fall at a density of 2-4/m2. Weeding
around young plantings aids establishment, but once plants are taller than their neighbors, they
should not require weeding.
        While sunchokes are known to tolerate a variety of growing conditions, they thrive in
sandy loam with good fertility. Adding nitrogen and phosphorus should increase the size of the
tubers produced and the biomass of the plants. Nutrients supplied to young sunchokes promote
the growth of aboveground biomass, while later application of nutrients facilitates the growth of
tubers and rhizomes, allowing plants to spread vegetatively (Kays and Nottingham 2008). A
meadow experiment could involve nutrient addition, but we caution that broad application of
nutrients would likely favor invasive species, particular grasses. It might be wise to grow
sunchokes in pots with fertile soil for transplantation to the meadow as competitive individuals.
        Sunchoke establishment and/or growth might be limited by competitors. A garden
experiment could test this hypothesis using GRPO meadow soil. A meadow-edge experiment
could identify early-season competitive interactions or rule out early aboveground competition as
a limiting factor in sunchoke restoration/expansion. Later, in phase 3, we recommend using
information from the garden experiment to decide how and when to mow vegetation in meadow
plots where conditions best match those of the existing sunchoke occurrences.

Experiment 1.3.1: Test of sunchoke light limitation

Hypothesis: Young plants will grow better in full sunlight than in shade.
Priority: High, to determine if sunchoke ramets (sprouted tubers) can be produced easily
    by growing them with ample light at the meadow edge.
Measure: Compare height growth and tuber production of sunchoke transplants with three
    levels of shade, as in the chives garden experiment.
Constants: Sprouted tubers (propagated from local stock), pot size, water supply, soil
    (from the site), amount of N and P added where specified.
Variables: Shade cloth absent or present in one layer or two layers. Ramets (sprouted
    tubers) will differ in size, so measure the wet weight of each tuber before planting and
    keep track of the starting weight for each pot.
Measure: Plant height, flowering, and tuber production at end of growing season.
Methods: Build frames to suspend shade cloth or use tomato cages. Plant each pot with
    one tuber. Add the cage and randomly assign a shade cloth treatment to each pot (5
    pots per treatment) and position the pot randomly in the growing space. Ideally, each
    pot would have its own frame and shade cloth, and the position of pots would be re-
    randomized monthly to ensure independence of experimental units.
Analysis: Compare plant responses as height, flowering, and tuber wet weight. Subtract
    starting tuber weight from end-of-season weight to calculate tuber production.
    Initially, tuber production will be the most relevant measure for predicting the
    potential for future harvesting. Later, plant height will be useful in deciding where to
    restore the species to the meadow and how to manage surrounding vegetation.

Experiment 1.3.2: Test of sunchoke nutrient limitation

Hypothesis: Adding nitrogen (N) will increase shoot growth.
Alternative hypothesis: If the soil is deficient in phosphorus (P) or rich in calcium (which
    ties up P), shoot growth will be greatest with both N and P added.
Priority: Low, unless ramets grow poorly when given ample light (based on the above
Constants: Tubers (propagated from local stock), pot size, water supply, soil (from the
    site), amount of N and P added where specified
Variables: Nutrients added (N, N+P, and P). Evaluate soil samples from the meadow edge
    and the meadow and use the above experimental results to select fertilizer treatments
    for experimental plots along the meadow edge.
Measure: Plant height and tuber production.
Methods: First, test soil samples from the garden and meadow edge and meadow to
    determine baseline N and P levels. We recommend three nutrient-addition treatments,
    an N+P fertilizer, an N fertilizer, such as ammonium nitrate holding the total N
    constant, and a P fertilizer, such as bone meal, holding the total P constant. Establish
    15 large pots or plots with one tuber each; add the nutrients as indicated above, with 5
    pots or plots per treatment. Grow outdoors in spring under uniform light for one
    growing season.

       Analysis: Compare plant height and tuber production (number and wet weight) for each
          treatment. A later experiment could effects of fertilizing in summer to increase tuber
          and rhizome formation.
       Retain all tubers for transplantation to the meadow edge or meadow experiments.

        Sweetgrass grows well from ramets, but seedlings are needed to increase genetic diversity.
We know that vegetative ramets do not grow best with waterlogged soil, but we have not tested
the need for (or tolerance of) waterlogged soil on seedling germination and early seedling growth.
        Thus, we recommend first growing new plants from seed. In our experience at UW, six of
15 seeds produced seedlings that expanded vegetatively in the greenhouse. The 15 seeds came
from inflorescences contained in a single block of sod (about 40 cm square) provided by Brandon
Seitz in fall 2008. We understand that GRPO has contracted out the growing of sweetgrass
propagules for restoration purposes (Seitz, pers. comm.).

       Experiment 1.4.1: Test of sweetgrass seedling establishment

       Hypothesis: Sweetgrass seedlings develop best in wet soil.
       Priority: High for developing genetically diverse ramets for subsequent vegetative
       Constants: Seed source (use local source to obtain locally adapted genotypes), light
           supply, temperature.
       Variables: Soil moisture (wet vs. drained pots).
       Measure: Rates of seedling establishment and seedling growth.
       Methods: Place equal numbers of seeds in shallow pots (4‖ deep) with drainage holes;
           roughen the soil surface, and randomly assign to wet vs. drained treatments (5 pots per
           treatment). For the wet treatment, place the pot in a plastic saucer that will retain
           water. Randomly distribute the 5 wet and 5 moist pots in a sunny place within the
           garden. Water all pots with the same amount of water—enough for water to collect in
           the saucers of the ―wet‖ pots.
       Analysis: Compare averages of seedlings at 4 weeks and their maximum heights in wet
           vs. drained pots.

       Experiment 1.4.2: Test of sweetgrass facilitation by Greene’s rush

      Because we did not encounter Vasey‘s rush in the meadow vegetation survey, we
recommend testing the ability of Greene‘s rush to facilitate sweetgrass growth.

       Hypothesis: Greene‘s rush facilitates sweetgrass growth.
       Alternative hypothesis: Greene‘s rush merely indicates wetland conditions that favor
       Priority: Medium priority, because Ipsen‘s (2010) greenhouse experiment at UW
           indicated that sweetgrass grows well and spreads rapidly in monoculture.

       Constants: Pot size (1-ft diameter), transplants per pot, temperature, soil, nutrients and
          water supply.
       Variable: Neighbor identity (growth of sweetgrass alone and with another wetland plant
       Measure: Length of tallest sweetgrass stretched leaf, measured monthly; expansion of
          clone diameter over the growing season. Count flowering stalks if present.
       Methods: Provide moist soil and full sun in the garden, and grow sweetgrass ramets alone
          and in pots with and without ramets of Greene‘s rush. The same experiment can be
          done with horsetail, bulrush, and Vasey‘s rush, if available. Each mixed-species pot
          would have one ramet of sweetgrass and one ramet of the other species. Each
          sweetgrass pot would have two sweetgrass ramets (to control for planting density).
          We suggest 10 replicates per treatment.
       Analysis: If the rush facilitates sweetgrass growth, sweetgrass ramets would average
          greater height and/or a larger area of ramets at the end of the growing season. Compare
          relative growth rates (sweetgrass in pot with a neighbor as a proportion of sweetgrass
          grown alone). If a neighbor increases sweetgrass growth or flowering, it would be
          considered a ―facilitator‖ that could be planted along with sweetgrass in meadow
          reintroduction tests.

        If locally adapted genotypes of Vasey‘s rush can be located, it would be useful to develop
techniques for seeding it into the meadow and for transplanting its ramets. Tests could begin in
the garden, varyiny moisture, as in the sweetgrass experiment. Once enough ramets are available,
the above facilitation experiment could be initiated. If Vasey‘s rush facilitates sweetgrass but will
not grow well in meadow soil, further experimentation would be needed to identify limiting
factors in the field. Most of the meadow does not support wetland vegetation, but below, we
suggest removing a few trees along with their root systems and soil to lower the soil surface to
levels closer to the water table. Tree removal areas could become suitable sites for Greene‘s rush,
Vasey‘s rush and other potential facilitators of sweetgrass growth.
        As discussed in Phase 3, several graminoid species could be planted across the elevation
gradient from edge to bottom of the depression, assuming that the deeper areas of the depression
will be wetter than the shallower edges. If growth increases in relation to soil moisture for
sweetgrass and other desired graminoids, the number of tree removals could increase to
accommodate a larger sweetgrass restoration effort.

       Phase 2: Meadow-edge experiments

If chives and young sunchokes are not shade tolerant, they might not compete well with taller
species, which would help explain their absence/rarity in the meadow. If chives grow best with
full sunlight at the meadow edge, it is not likely that it will thrive in the meadow. The meadow
edge might then become the target location for restoration of chives. If young sunchokes respond
to full sunlight at the meadow edge, then transplantation to the meadow could occur with mowing
or other management to increase light until sunchokes are well established. The area where
meadow-edge experiments could be implemented is mapped in Figure 3.1.

       We have higher expectations for sunchoke growth than for chives in the meadow edge,
because sunchokes were found nearby in the 2009 survey. Soil conditions are likely suitable.
Also, because sunchokes are much taller than chives, it is less likely that light limits sunchokes, at
least where there are ample nutrients.
       We suspect that sunchokes are now rare due to historical harvesting. The species is widely
grown for domestic and commercial purposes (see ―Species Profiles‖). It is a conspicuous plant
that would be easily located and depleted by anyone who knows the species. Alternatively, the
sunchokes that are present in the meadow could be a variety that requires more light and nutrients
than varietys used in agriculture. An initial test of light requirements could eliminate the need for
more complicated experientation with nutrients in the garden.
       If the following experiment indicates that shade limits sunchoke establishment, the garden
experiment would be unnecessary, and meadow restoration could begin sooner.

   Experiment 2.1: Test of chives and sunchoke light limitation
   Hypothesis: Chives and young sunchokes will grow fastest in full sunlight.
   Priority: High if garden experiment indicates that light is limiting and high because the
       meadow-edge plantings would attract the interest of visitors.
   Constants: Plot size, soil, chives bulb density or sunchoke tuber density.
   Variable: Two light levels, created by mowing and not mowing prior to planting chives.
       During the experiment, it might be necessary to hand-clip resprouting vegetation to
       maintain full sunlight in the mowed plots.
   Methods: We recommend establishing ten 1x2-m experimental plots in a row parallel to the
       fence around the south edge of the meadow. An area south of the fence in full sun and
       next to the mowed lawn would be suitable. Random assign 5 plots to be mowed and 5
       unmowed. In each plot, equal numbers of seedlings or bulbs of chives and seedlings or
       small tubers of sunchoke would be planted and monitored for survival and new growth.
       These plantings would need to be about 30 cm apart so they do not interact with one
       another during their establishment and early growth. Use a bulb planter to remove sod;
       add the bulb or tuber; and replace loose dirt, discarding the roots of other species.
       Measure heights of plants at monthy intervals or at least at the end of the growing season
       (maximum stretched leaf).
   Analysis: Compare growth with and without mowing. A positive response to mowing by
       either chives or sunchokes would justify a larger experiment with mowed strips within the
       meadow (described below). The first step, however, is to test shade tolerance so the Task
       Force can base their decision on the experimental outcomes.

Phase 3: Adaptive restoration within the meadow
We recommend introducing sunchokes, chives and sweetgrass into the meadow using
experimental approaches that will answer key questions about what limits their populations. We
suggest that sunchokes will need relief from competition from taller plants, that chives might
require full sunlight and be unable to compete with meadow vegetation and that sweetgrass might
be confined to moist, low-nutrient microsites. Thus, we recommend experiments involving
mowing and tree-removal. Competitors likely reduce the abundance of all three target species so
that introducing them to areas that have been mowed or had trees removed will temporarily shift
the competitive advantage from taller meadow species to ramets of newly planted target species.
        If mowing and follow-up clipping are needed to propagate chives, the meadow edge might
need to become the permanent site for chives restoration. Alternatively, annual mowing of
selected meadow plots would need to become a permanent management practice for chives. If the
latter, we suggest that the Task Force decide if long-term mowing would be possible for the
meadow before introducing chives to areas that are currently fully vegetated. If long-term
mowing is both required for chives and a possibility for management, then chives could be
reintroduced into plots that are identified for long-term mowing (possibly in conjunction with
site-scale management experiments proposed in Phase 4, below).
        If long-term mowing is not possible, chives could still be introduced to areas of tree
removal within the meadow, in depressions resulting from excavating tree stumps and large roots.
If the soil in the resulting depressions is depleted of nutrients or saturated with water or otherwise
limiting to growth of meadow vegetation, the depressions might be more suitable for sustaining
short plants, such as chives and sweetgrass.
        Short-term mowing (i.e., for one or two years) should suffice to establish sunchokes in the
meadow, since this species already occurs there and might have become rare due to
overharvesting. Short-term mowing of existing vegetation and planting of sunchoke tubers
should give this desired species a competitive edge during the vulnerable establishment period.
Once tubers are large and numerous, it should thrive without continued mowing.
        For sweetgrass, we hypothesize that wet soil is a primary limiting factor, followed by
competition from taller neighbors, and that reestablishment will be most effective if initiated in
the wetter, southwest portion of the meadow, where Greene‘s rush occurs. We also hypothesize
that the removal of trees and roots to create shallow depressions will expand habitat for
sweetgrass. Potential locations for each of the following experiments are mapped in Figure 3.1.

       Experiment 3.1: Test of young sunchoke requirement for light

        Following garden experiments that indicate a negative effect of shade on young
sunchokes, we suggest an additional mowing experiment to test for improved establishment
within the meadow. Because we recommend using the trees as sites for other restoration
experiments (below), mowed areas and access routs should not be near trees; instead, leave wide
buffers around trees in the meadow. The first mowing experiment could be small, in keeping
with the small number of propagules that will be available for transplanting. Subsequent mowing
could employ larger areas as more sunchoke propagules become available.
        We suggest mowing strips in well-drained areas, leaving wetter areas for sweetgrass
restoration. Mowing is intended to reduce shade for planting sunchoke seedings and/or small

tubers in areas within and outside the mowed strips. The methods would be the same for chives,
if mowing could be a permanent management action.

       Hypothesis: Mowing will increase survival and growth of the target species.
       Priority: Medium. If the meadow-edge experiment already supports the hypothesis, then
           mowing could be adopted as a management practice without further experimentation.
       Constants: Number of propagules planted, genotype (locally adapted), soil.
       Variables: Mowed or unmowed vegetation (plant cover and species). Note that if soil is
           variable across the meadow, that paired mowed-unmowed plots could be designated
           within areas of similar soil.
       Measure: Count number of propagules that survive. Record average height of the tallest
           stretched leaf of each plant, and note if each plant flowered (count the number of
           inflorescences). It would also help to assess the species richness and composition of
           each experimental plot to ensure that paired plots are similar and to relate sunchoke
           outcomes to neighboring plants.
       Methods: Mow strips that are at least six feet (~2 m) wide. Plant seedlings or small tubers
           in 1-m2 plots with and without a mowed canopy in a paired-plot experimental design.
           For each site, establish three 1-m2 plots within and outside the mowed strip. Add three
           propagules to each plot to test the effect of shading by neighbors and to compare
           survival and growth across the range of conditions present in the well-drained subarea.
       Analysis: Compare target species‘ establishment with and without mowing other
           vegetation. Express the number of ramets that establish and mean height and
           flowering in each unmowed plot as a percent of the same response variable in its
           paired mowed plot. This will scale the data so that the average effect of mowing can
           be determined for each response variable. If establishment is satisfactory but the effect
           is small, then mowing might not be warranted. If the establishment rate is low and not
           related to mowing, then other factors need to be explored (e.g., nutrient addition).

        Results of the mowing experiment will indicate the species‘ light requirements in
restoration sites. If mowing on a small scale increases survival and growth of chives and/or
sunchokes, then mowing and plantings could be expanded elsewhere within the well-drained
areas of the meadow.
        Mowed strips could serve as access routes for volunteers or visitors. This could be a
positive outcome, if volunteers are attracted to the project, or a negative outcome, if visitors
overharvest either of the planted species. Each year, the Task Force should review findings and
decide how best to proceed.

       A different type of experiment is proposed to determine the optimal environmental
conditions for expansion of the current sweetgrass population. For this, we propose altering
conditions along the edge of the existing sweetgrass clone(s). The current population has an edge
of about 60 meters where mowing would increase light and reduce competition for water and
where shallow excavation would eliminate neighboring species roots and increase soil moisture.

       Experiment 3.2: Comparison of methods to expand the existing sweetgrass population

       Hypothesis: Sweetgrass is constrained by aboveground competition, belowground
           competition, or both.
       Priority: High, because the most likely place that the population can expand is near where
           it already occurs.
       Constants: Plot size, proximity to edge of sweetgrass clone.
       Variables: Mowing to remove shoots of neighbors and increase light availability;
           excavation of shallow depressions. The latter would combine two factors, namely,
           removing neighbors‘ roots and increasing surface soil moisture by lowering the soil
           surface and bringing it closer to the groundwater level).
       Measure: Expansion of sweetgrass clone into each treatment plot (distance outward from
           a marked starting line).
       Methods: Mow along the edge of the existing continuous sweetgrass to reduce
           competition for light and excavate shallow depressions, setting up three treatments
           (mow; excavate; control = no treatment). The current sweetgrass population has two
           adjacent patches with a combined edge of about 60 m. This would allow up to 12
           contiguous plots, each 5 m long. We recommend four replicates. Treatment plots (5m
           long x 1m wide) would be established within four 15-m-long blocks, each block
           having all three treatments in random positions.
       Analysis: Compare sweetgrass expansion with in the three treatments: control, mowing
           and shallow excavation. If sweetgrass expand only with mowing, then competition for
           light is implicated, although mowing would also reduce competition for water and
           nutrients. If only the shallow depressions allow sweetgrass to expand, then root
           competition or moisture are implicated. If the both mowing and depressions increase
           sweetgrass expansion, the species may be limited by competition both above- and

        Management to expand the sweetgrass population by long-term mowing would seem
feasible, but we suspect that a limited area with moist soil will confine sweetgrass to the
southwest corner of the meadow. Increased expansion into the shallow depressions and
observations of greater soil moisture where neighboring roots are excavated would support the
hypothesis that sweetgrass is a stronger competitor under wetter, less nutrient-rich conditions.
Rather than proposing the grading of the meadow to lower the soil surface, we suggest removing
trees and their roots in search of suitable microhabitats for sweetgrass espansion.

       Experiment 3.3: Test of sweetgrass establishment following tree removal

        Because sweetgrass is short-statured, it is likely outgrown by taller, shade-producing
species, and because it is associated with certain wetland conditions (e.g., groundwater seepages
with low nutrient status), its competitive ability is probably limited in drier sites and in nutrient-
rich siges, both of which would favor taller, more aggressive upland plants.
        To increase habitat for sweetgrass, we suggest the experimental removal of selected
evergreen trees, including their roots, to create shallow depressions. The depressions should allow
water to accumulate or at least lower the soil surface closer to the water table. We suggest

beginning with the removal of 3-5 trees where the soil is already moist to wet. Trees could be
chosen based on soil moisture conditions that are relatively high or where well points indicate that
the water table is near the surface.

       Hypothesis 1: Shallow, ―tree-removal depressions‖ will provide suitable conditions for
           sweetgrass. Suitable conditions would include one or more of the following: reduced
           competition from neighbors, increased soil moisture, or a temporary advantage by
           being planted (speeding establishment relative to competitors that would need to
           disperse into the depressions and establish from seed).
       Hypothesis 2: If indicated by the garden experiment, rare wetland plants (Greene‘s rush,
           Vasey‘s rush, or others) will facilitate establishment and expansion of sweetgrass).
       Priority: High, because some tree removal is a restoration objective, and the experiment
           will indicate if sweetgrass can be readily restored by planting in bare depressions.
       Constants: Size of depressions (depth and diameter), density of sweetgrass plugs and
           density of plugs of potential facilitators (if the Task Force decides that the garden
           experiment warrants such tests).
       Variables: Location of sweetgrass plantings within the depression (surrogate for elevation
           and soil moisture) and presence or absence of potential facilitators.
       Methods: To assess location effects, we propose comparing 4-5 plugs near the bottom
           center with 4-5 plugs near the upper edge. The grid of sweetgrass ramets would have
           to be sparse enough to test expansion rates (50 cm). Plant two ramets per grid point
           (either 2 sweet grass ramets or 1 sweetgrass + 1 potential facilitator). Adding mulch
           between plugs would reduce colonization by unwanted species. Collect soil cores (1‖
           diameter, 6‖ deep) at center and edge of depression, 3-5 cores per location; place in
           air-tight plastic bags; label and assess moisture conditions. All cores to be compared
           must be taken on the same day, at least once after sweetgrass planting and once at the
           end of the growing season. Randomly select ramets to monitor in each location; mark
           with flags.
       Measure: Height (maximum length of stretched leaves) and initial diameter of each ramet
           selected for monitoring at planting time, during mid summer and at the end of the
           growing season. Measure basal diameters of the ramet (or clone, one it has expanded
           vegetatively) on north-south and east-west axes. Use average diameter per clone to
           compare expansion rates by location. Measure soil moisture by weighing the cores;
           then oven dry 24 hr at and reweigh; soil moisture = weight loss on drying divided by
           the final dry weight of the soil (bag not included). Photos of each planted depression
           will be useful.
       Analysis: Compare heights and clonal expansion rates for central (deeper) and edge
           (shallower) locations within each depression. If deeper locations have greater height
           and expansion rates, future tree removals could involve larger areas with deep
           excavation. If the potential facilitator(s) merely indicate the presence of wet soil, then
           no difference in clonal expansion of sweetgrass expansion would expected with and
           without neighbors.

       Experiment 3.4: Test of sweetgrass facilitation by Greene’s rush

        Greene‘s rush occurs in the northwest corner of the meadow near the two patches of
sweetgrass. If the garden experiment indicates that Greene‘s rush facilitates sweetgrass, then a
meadow experiment could test this concept further. We suggest planting sweetgrass ramets with
and without Greene‘s rush in small plots where neither species occurs in the meadow, but where
soil moisture is relatively high. Other wetland plant indicators could be used to select suitable
plots (6-10 plots with sweetrass, of which 3-5 would be with and 3-5 without Greene‘s rush). If
Greene‘s rush merely indicates the presence of wet soil, then no difference in clonal expansion of
sweetgrass expansion would expected. These meadow plots would need to be mowed prior to
planting to reduce competition and potentially weeded to allow establishment by sweetgrass and
Greene‘s rush ramets.

       Hypothesis: Greene‘s rush facilitates sweetgrass growth.
       Priority: Low unless the garden experiment indicates a neighbor effect. Not needed if the
           tree removal experiment is already underway with and without wetland-plant
       Constants: Plot size and ramet density.
       Variables: Presence or absence of Greene‘s rush.
       Methods: Select 3-5 locations within the meadow where soils are moist. At each location,
           select two similar plots, each about 1 m2. Randomly select one of each of the paired
           plots to have one ramet of sweetgrass and one ramet of Greene‘s rush and the other to
           have two ramets of sweetgrass (to control for ramet density). The paired plots will
           help control for any location effect within the meadow.
       Measure: Monitor initial height and ramet diameters and expansion rates as above. If
           possible, measure soil moisture in each co-planted area once after sweetgrass planting,
           and once at the end of the growing season.
       Analysis: Compare expansion with and without Greene‘s rush. For each pair of plots,
           calculate height increase and diameter increase with Greene‘s rush divided by the
           same variables without Greene‘s rush. A consistent pattern of greater height or greater
           diameter expansion with Greene‘s rush would indicate a facilitator function for this
           neighbor, and future co-plantings would be warranted. Similar heights and expansion
           rates would suggest that Greene‘s rush is merely an indicator of suitable habitat for
           sweetgrass. Pictures of the plot will be useful.

        Mowing meadow vegetation and removing trees are two experimental manipulations that
will create a more heterogeneous landscape with more diverse microsites. Open canopies and wet
depressions will make the meadow more spatially heterogeneous and able to sustain a higher
diversity of culturally important species. It is also possible that weedy species will colonize the
disturbed soils and that continual weeding will be needed.
        These experiments are intended to increase populations of culturally important species,
but both mowing and tree-removal depressions may have the effect of creating favorable
microhabitats for addition native species. More variety in microhabitats could result in greater
diversity of native plant species as well as insect, mammal and birds at GRPO.

Figure 3.1. Potential placement of several Phase 3 experiments. Drawings are nearly to scale. Blue =
unmowed controls, red = mowed treatments, yellow = excavated treatments, and orange = sites
suggested for tree removal followed by sweetgrass planting and soil monitoring (coring) in the
resulting depression. See text for detailed recommendations concerning implementation of these and
other experiments.

            Phase 4: Adaptive management within the meadow
The first three phases of experimentation should indicate suitable environments for individual
species (sunchoke needs full sunlight, sweetgrass needs moist soil, etc.) while testing the
effectiveness of management actions to provide needed environments (mowing around young
sunchokes, felling trees and excavating tree roots to create wet depressions for sweetgrass, etc.).
Those experiments should direct short-term management in support of the establishment and early
growth of target species in the meadow.
        Larger-scale and longer-term management should mimic traditional ethnobotanical
practices, and alternative strategies can be tested within an adaptive management framework.
Management of mature populations of target species, and management of meadow as a whole, can
be guided by field experiments that test the effects of ethnobotanical practices before scaling up
for use throughout the site.
        Recruiting local volunteers to test the effects of harvesting will promote ethnobotanical
use of the target species, help explain decision-making by managers (if, for example, test-
harvesting leads to a localized population crash and harvesting must be limited), and introduce
harvesting in a limited/sustainable way. Harvest experiments would help managers directly
assess the effects of using target species and compare effects of different ethnobotanical practices.
If park managers and volunteers engage in, and learn from, management activities, extractive uses
could become sustainable and serve the overall management goal:

       Management goal: Improve the people’s relationships with plants in the meadow
       area; the former site of the 19th century village of Grand Portage.

        Presently, caraway is the only target species that is common enough in the meadow to be
managed at a large scale (caraway was fifth most frequent and fifth most abundant among all
species in the 202 quadrats surveyed in 2009). Like all the target species in this project, caraway
has traditionally been harvested by local people. Below, we describe a split-plot experiment to
test the effects of harvesting seeds (every year vs. every other year) and whole-plant removal of
flowering individuals (for seeds, leaves, and taproots). The experiment would allow volunteers to
participate in an ongoing experiment, engage in a cultural activity, and take home edible plant
parts (seeds and taproots, noting the aforementioned caution about spreading a potential weed
species). Managers would get a sense for what level of harvest is sustainable and how the plant
could be effectively controlled if it becomes too invasive. Future harvest of sunchokes,
sweetgrass, and chives could begin through similar experimental trials.
        In addition to target-species management, it is necessary to consider ethnobotanical
management at the whole-site scale. The meadow is now mowed most years and burned
intermittently. Fire is the more traditional mode of American grassland management, and
managers and residents are enthusiastic about burning. However, mowing is more reliable
because it is generally less expensive, less weather dependent, and requires fewer people and less
staging. Thus, a switch from the status quo (mowing) to a more ethnobotanically-based practice
(fire) will require strong support and persuasive arguments from the Task Force. We propose
test-burns in subsections of the meadow. Burning in an experimental manner for several years
would give managers a preview of changes in vegetation that could be expected with long-term
burning, and the results would help managers decide if frequent whole-site burns are desirable.

       Experiment 4.1: Comparison of caraway harvest methods

        Caraway has a 2-year life cycle, which may or may not be compatible with potential rates
and methods of harvest. Managers of the GRPO meadow could face either of two problems with
caraway: it becomes too abundant (weedy and competitive with other target species) or it
becomes over-harvested and the population declines. Hence, our proposed harvest-test includes
treatments that are severe enough to extirpate/exhaust caraway locally (and thus control carwaway
where it is unwanted) and other treatments that are low-impact and likely sustainable (to guide
cultural use practices). We suggest three progressively damaging harvest treatments: biennial
seed-head harvest, annual seed-head harvest, and annual whole-mature-plant harvest, including
leaves and taproot (both of which are edible). Harvest effects will be monitored over several
years by comparing the number of seed heads in each treatment plot to the number in an un-
harvested plot.

       Hypothesis: The number of potentially harvestable seed heads will be highest in the
           unharvested control plots, second-highest in the biennial seed-head harvest plots,
           third-highest in the annual seed-head harvest plots, and lowest in the annual whole-
           mature-plant harvest plots.
       Priority: High, because this would provide an immediate way to connect local people to a
           culturally significant plant and develop interest in restoration/ethnobotanical
           management; especially important if vegetation monitoring shows caraway to increase
           continually in frequency of occurrence and abundance.
       Constants: Plot size and starting number of seed heads per plot (as close as possible
           within experimental blocks (clusters of 4 plots)).
       Variables: Harvest treatment.
       Methods: Select 3-5 locations within the meadow where caraway occurs at high and fairly
           uniform density. At each location, select a square of 20x20-m, sub-divide that space
           into four plots. Randomly assign each of the following treatments to one of the four
           plots: control, biennial seed-head harvest, annual seed-head harvest, and annual whole-
           mature-plant harvest. These four plots comprise an experimental block (if one
           location with caraway reacts differently, it should affect the entire block of treatment
           plots). In late July (or whenever caraway seeds are ripe at GRPO), visit each block
           and do the following: 1) record the number of seed heads present in each plot 2)
           collect all caraway seed heads in the biennial- and annual-seed-head-harvest plots and
           3) collect the taproot and shoots of all mature individuals (with seed-heads) in the
           whole-mature-plant harvest plot. The following year, in late July, repeat all the above
           but without harvesting seed heads in the biennial treatment (and every other year
           thereafter, for at least 3 years).
       Measure: Number of seed heads in each plot (before harvest treatments are administered
           each year, as a measure of how much caraway is still harvestable and reproducing).
           Optionally, could measure leaf heights of several individuals in each plot or seed yield
           of the three harvested plots to check for declining plant height or seed yield over each
           year (if a high-resolution scale is available to weigh small seeds).

       Analysis: Compare the mean number of seed heads for each treatment. If harvest
          treatments results in significantly fewer seed heads, that harvest method may be
          unsustainable when repeated in a given area. But such harvest methods would also be
          useful if control of caraway is desired.

       Experiment 4.2: Comparison of site-scale management activities

         Prescribed burning is an attractive alternative to mowing for managing the GRPO
meadow. Burning has the potential to reduce shrub cover and favor native, fire-adapted species.
But assessment of outcomes can be complicated by: missing pre-set goals (fire did not extirpate
invasive species X) and confounding effects of ―year‖ due to the all-or-nothing nature of burning
(e.g., comparing conditions when burned in 2011 to not-burned in 2010).
         The goal of this experiment is to test the effect of fire vs. mowing by setting up sub-plots
within the meadow that would be burned whenever the whole site is mowed. The vegetation
would be assessed in burned plots and in paired, unburned control plots with similar vegetation.
The main objective is to find out how the abundances of species (especially weedy, invasive, and
target species) respond to fire. By burning only a portion of the site at high frequency, the
possible negative consequences of extirpating soil invertebrates or exhausting vegetation are
limited in scope; also, the relatively small area for each burn (400 m2) should decrease the
difficulty in managing the burn and the number of people needed to control the fire.

       Hypothesis: Species‘ abundances will differ significantly in burned vs. unburned plots.
           Shrub species will likely have lower cover in burned plots, whereas grasses (both
           wanted and unwanted) will likely have higher cover in burned plots.
       Priority: Low, relative to experiments that will help restore target species to the meadow.
       Constants: Plot size, season and year of mow/burn.
       Variables: Prescribed burn or not.
       Methods: Select 3-5 locations throughout the meadow. At each location, select two plots
           with similar vegetation, each plot should be 20x20 m (400 m2). Plots should be set up
           with a 2-3 m buffer between them, and they should be at approximately the same
           elevation (to reduce elevation effects). Randomly select one of each of the paired
           plots to serve as the burn plot and the other as the control plot. The paired plots will
           help control for differences in vegetation within the meadow, and at least one pair of
           plots should include high cover of shrubs. When the plots are first set up and marked,
           designate a 5x5-m subplot in the center of each plot to be used for vegetation
           monitoring. This design will minimize edge-effects and allow a relatively quick
           survey, which should be done with GPS since any plot markers would be damaged by
           burning or mowing.
               When the entire meadow is mowed: the burn plot will be mowed around and the
           no-burn-plot will be mowed. After mowing is complete, the burn plots should be
           ignited. When the entire meadow is burned, both plots should be ignited; and when
           the entire meadow goes unmowed and unburned, both plots should go unmowed and
       Measure: Within each subplot, sample initial presence and cover of all species. Re-
           sample species presence and cover every year while the experiment is running.

       Analysis: Compare changes in species presence and cover between the burned and
          unburned plots in each location. Differences in cover of species that occur in most
          locations can be analyzed statistically, as could differences in cover of functional
          groups (shrubs vs. forbs vs. grasses). Because presence/absence of species would be
          collected it would also be possible to tell if species richness (a proxy for diversity)
          were also changing in response to fire.
              However, this is essentially a descriptive experiment, so the most valuable
          information may come from individual data points. For example, (1) in the location
          with highest initial shrub-cover the burn treatment may extirpate most but not all shrub
          species; (2) in the location with highest initial smooth-brome-cover the burn treatment
          may lead to a progressive increase in that exotic grass (likewise for other grasses, or
          for bird‘s foot trefoil which should have an advantage in N-poor conditions); (3) in
          some locations burning may lead to an increase in target or associate species (e.g.,
          Greene‘s rush or green bulrush).
        Harvesting caraway (and eventually other species) and prescribed burning will promote
ethnobotanical practices through adaptive management. While the focus of this project is the
restoration of the culturally-significant target species, the future of the target species in the GRPO
meadow requires careful implementation of plant-harvesting and consideration of multiple
management techniques.
        Species-based experiments, such as Experiment 4.1, could be used to assess the
sustainability of alternative methods of harvesting sweetgrass, sunchoke, and chives once those
species become harvestable. By first testing harvesting at a small scale, the potential to deplete
the population of a target species is minimized. Similar experiments could be done to assess the
effectiveness of efforts to control existing populations of undesirable species, such as smooth
brome (e.g., test the effects of hand-pulling vs. mowing vs. herbicide).
        At the whole-site scale, additional management methods of interest to managers could be
tested using the design for Experiment 4.2. For example, rather than burning, the treatment plot
could be mowed twice as often as the rest of the meadow, or raked after mowing as the rest of the
meadow goes un-raked. In either case effects would be evaluated similarly: testing for
differences in target and invasive species‘ presence or abundance between treated and untreated
plots. Such a comparison will offer more robust support for the use or disuse of a given
management practice, as the number of test locations and the number of test years increases.

Figure 1.5. Potential placement of Phase 4 experiments. Drawings are nearly to scale.
Blue = unmanipulated controls, red = whole-plant removal of caraway, green = annual
collection of caraway seed heads, lavender = biennial collection of caraway seed heads,
and dark-red/brown = burn plots (or other site-scale management treatment to be
tested). See text for detailed recommendations concerning the implementation.

       Suggested future study: Managing meadow drainage

        The road along Lake Superior and two drainage ditches or depressions that lead to culverts
under the road indicate that hydrological conditions have been altered. Although we have no
information on the timing and impacts of these two features, we can speculate how each has
affected water flow, soil moisture and the biota. The hard-surface road likely increases surface-
water runoff locally, funneling water toward the culverts. The elevated roadbed likely impounds
runoff from the meadow and funnels that water toward the culverts. The road and traffic likely
affect animals by inhibiting their movement or causing roadkills. Here we focus on the
hydrological impacts.
        The drainage ditches would have been constructed to deplete soil water or convey runoff
to culverts or both. Regardless of their purpose, the depressions modify hydrological conditions
and could reduce the area suitable for sweetgrass restoration. Ipsen‘s (2010) experiments with
moisture conditions indicate that the GRPO sweetgrass genotypes grow well in moist soil. Also,
species distributions within the meadow (Part 2 of this report) indicate that seepages are not
        Without adequate information on soil moisture and drainage, we suggest that the effects of
the drainages and culverts be investigated and the utility of adding water control structures
considered (e.g., those of Agridrain; www.agridrain.com/watercontrolproductsinlet.asp). With
water control structures, outlets could be open to accommodate spring runoff and prevent
flooding the road and closed to reduce surface- and ground-water depletion at other times of the
        Water control structures at each culvert would allow managers to vary drainage.
Monitoring surface soil moisture and ground water levels in several wells would allow the Task
Force to test the effects of reduced drainage on hydrological conditions in the meadow, thereby
increasing the potential to manage sweetgrass adaptively. A reasonable hypothesis is that
blocking the culvert during normal growing seasons would enhance and expand conditions for
sweetgrass. Impounded water could still be released as needed to prevent flooding the road. The
current condition (year-round drainage) would seem to hinder, rather than facilitate, the
restoration of sweetgrass at the GRPO meadow.

Summary of experiments by phase and priority

   Priority Monitoring           Phase 1         Phase 2         Phase 3             Phase 4
                                 Garden          Meadow          Meadow              Meadow
                                                 edge            Restoration         Mgmt.
   High   Annually for      Chives‘ light        Chives and      Expanding the       Comparison
          caraway, reed     requirement;         sunchoke        existing            of caraway
          canary grass and                       light           sweetgrass          harvest
          smooth brome;     Sunchoke             requirement     population;         methods
                            light                (if supported
          At 5-year         requirement;         by the          Establishing
          intervals for the                      garden          sweetgrass in
          202 vegetation    Sweetgrass           experiment)     tree-removal
          quadrats          seedling                             depressions
                            in wet soil
   Medium                   Does                                 Mowing
                            Greene‘s rush                        improves growth
                            facilitate                           of young
                            sweetgrass?                          sunchokes (if the
                                                                 exp‘t indicates
                                                                 light limitation)

   Low      Annual or            Does caraway                    Does Greene‘s       Comparison
            biennial             reduce                          rush facilitate     of site-scale
            monitoring of        growth of                       sweetgrass?         management
            all species in all   other desired                                       activities
            202 plots            species?

                                 Is sunchoke

                         Discussion and Recommendations

Landscape restoration
All restoration efforts take place within a larger landscape context, and the condition of the
landscape often dictates the degree to which restoration can achieve its goals. For example,
GRPO‘s mostly-undeveloped watershed likely provides clean surface water and ground water in
quantities that are similar to historical levels. A more developed watershed would discharge
excess water of lower quality. The GRPO landscape can also accommodate the use of burning to
manage vegetation, whereas an urban landscape would have ordinances to control smoke. And an
undeveloped landscape will likely retain more of the native animals that disperse and feed on
plants than would an agricultural or urban landscape. For all of these reasons, restoration efforts
should have high potential at GRPO, and the meadow restoration effort can be planned with
broader perspectives in mind, namely, landscape restoration within the boreal region of North
        Lovell and Johnston (2009) proposed three overriding objectives for restoring landscapes,
namely multifunctionality, heterogeneity, and biodiversity. Each is relevant to the restoration of
culturally important species in GRPO‘s 3.5-acre meadow. (1) Multifunctionality is a condition
for sustainability. Neither a landscape nor a 3.5-acre meadow can persist in perpetuity if it does
not support multiple functions, such as providing environmental, social, and economic services.
(2) Heterogeneity of spatial pattern is essential for basic ecological interactions that help to
sustain biodiversity (e.g., one species facilitating another and predator-prey dynamics that keep
pest populations in check). (3) Restoration and protection of biodiversity is an overall goal,
because biodiversity includes species that were introduced and cultured by early settlers and that
retain values that are relevant to Grand Portage as a National Monument. To these, we add a
fourth objective, namely (4) adaptive restoration or ―learning while restoring.‖ Experimental
manipulations and observations of outcomes will show how best to sustain desired species and
control the weedier plants.
        Our proposed experiments are intended to enhance ecosystem functions while adding
spatial heterogeneity and increasing biodiversity support in the meadow. Thus, we characterize
three subareas of the meadow (wetter patches, drier terrain, and edges with high light), and we
propose different restoration actions to favor wetland, upland and light-requiring species. In this
way, the biodiversity-support functions of each subarea will complement one another.
        The approach we propose for restoring culturally important species would increase
multifunctionality, heterogeneity, and biodiversity, as follows:

       - Multifunctionality

        The restoration of culturally important species to sustainable levels will support recreation
and esthetic appeal, as well as community involvement. Social functions would increase as visual
quality changes and as educational opportunities arise. Heritage value will be enhanced when the
populations of historically abundant species become sustainable and as interpretive signage and
leaflets portray the roles that these species played in past centuries. Economic functions will be
enhanced as edible plants (chives, caraway, sunchokes) become more available, and as sweetgrass
becomes more abundant for ceremonial use. Economic development would be fostered by the

provision of sweetgrass for use in weaving and other artwork which may be sold by local artisans.
Sustainable harvesting of sweetgrass is a suitable goal for the restored population at GRPO.

       - Heterogeneity of spatial pattern

        The meadow is relatively flat and mostly unwooded. Two of the management actions that
we recommend (mowing strips and selectively removing a subset of trees) would contribute to the
heterogeneity of the landscape. Neither open areas nor wet depressions occur in the meadow. By
creating them, we would make the site more spatially heterogeneous. Creating more diverse
microhabitats throughout the meadow should increase the potential to support larger populations
of the three rare and culturally important species and also a greater diversity of native species.

       - Restore and protect biodiversity of the existing meadow vegetation

        The meadow and nearby terrain has had a long history of natural and human disturbances.
Over millennia, these included glacial cover and retreat, lake-level fluctuations, changing
groundwater quantity and quality, and wildfire. Over recent centuries, humans added burning,
logging, firewood collection, trampling, grazing, livestock corrals, habitation, gardening,
inadvertent species introductions, and disposal of gray water and wastes. As a result, the soils
have likely been altered by additions and removals of both mineral and organic materials, as well
as excavations that would have exposed organic layers to oxidation and trampling that would
have compacted layers that might otherwise have remained aerated. It is difficult to imagine all
the changes that the groundwater, soil structure and soil microorganisms have undergone since
humans first camped at this site along Lake Superior.
        Neither the abiotic nor biotic conditions are pristine. Thus, the meadow can be considered
a novel environment with novel plant assemblages made up of both native and introduced species.
It is dominated by herbaceous perennial plants, but meadow vegetation is not likely to persist
without fire or mowing, because the climate and surrounding vegetation indicate that ecological
succession would otherwise proceed toward boreal forest. Shrubs and some trees are already
present among the herbaceous vegetation. The novel meadow will need continual management to
remain a meadow.
        In our 2009 sampling and observations, the GRPO meadow was found to support 107
plant species, of which 86 were recorded in at least one of our 202 sampling plots (1-m2
quadrats). Of the 86 species found in the meadow, 71 were native plant species, 15 were
introduced species, 16 were weedy, 6 were invasive according to Minnesota Department of
Natural Resources and USDA. 63 were classified as abundant. Caraway, sunchokes and
sweetgrass were classified as dominant or prominent in 89, 2 and 3 plots, respectively. Of the
culturally-important species, three were rare (chives, sunchokes, and sweetgrass) and one
(caraway) was abundant and potentially invasive. In 2009, caraway occurred in 71 plots,
sweetgrass was found in four, sunchokes in two, and chives in none. Greene‘s rush (a potential
indicator of habitat suitable for sweetgrass or a potential facilitator of sweetgrass growth) was not
found in the plots or in extensive surveillance of the meadow. Vasey‘s rush (Juncus vaseyi) was
encountered in no plots, nor was it observed anywhere else in the meadow.
        Additionally several undesirable species were identified and the most prominent consisted
of smooth brome (Bromis inermis), bird‘s foot trefoil (Lotus cornicalatus), red-osier dogwood
(Cornus sericea), and willow (Salix spp.) which were classified as dominant or prominent in 43,

75, 42, and 36 quadrats respectively. The two most invasive species (reed canary grass [Phalaris
arundinacea] and smooth brome) threaten the diverse plant community if left unchecked. Reed
canary grass was confined to small patches, but smooth brome occurred in about 14% of the plots.
Eradication efforts are needed before either population has an opportunity to expand further.
        The restoration of four target species has the potential to affect existing vegetation. For
example, if sunchokes expand, shorter species might be outshaded. Simultaneously, species that
are already present might negatively affect restoration efforts, since all four target species prefer
full sun/partial shade (Kays and Nottingham 2008, White 2002, Craig 2006, Plants for a Future
1996). Trees and tall shrubs could shade or otherwise outcompete short herbaceous targets either
as seedlings (all four targets) or as adults (chives, sweetgrass). Dr. Deb Pomroy identified 13
shrub species in 2009, and within plots sampled for vegetation the most common willows and
dogwoods were most commonly encountered (occurring in 20% and 18% of plots, respectively).
We understand that GRPO aims to remove some of the trees that are present within the meadow
and between the meadow and Lake Superior.

Summary of recommendations for monitoring and restoration

        We recommend establishing an ―Adaptive Restoration Task Force‖ to oversee monitoring
and experimentation, supervise data collection, interpret and archive the data, and use the
information to guide the overall restoration effort at GRPO.
        We recommend monitoring the 202 plots that were sampled in 2009, resampling annually
for target species (desired and weedy species) and at 5-yr intervals to assess changes in native
species distributions and overall biodiversity.
        We recommend that reed canary grass be eradicated immediately, because it does not yet
have a large population at GRPO. We recommend controlling smooth brome throughout the
        We recommend adaptive monitoring, restoration, and management of the target species as
        (1) Long-term monitoring of the meadow vegetation, using the 2009 survey as a baseline
            and refining methods to suit changing priorities.
        (2) Experimentation in the garden to determine the environmental factors that restrict
            growth of the three rare target species and to provide propagules for restoration.
        (3) Experimentation at the meadow edge to serve as further tests of pant growth,
            demonstration exhibits, and plots that could sustain some level of harvesting.
        (4) Experimental restoration of three rare species in the meadow.
        (5) Experimental harvest of caraway and management at the site scale.

Highlighting ecological functions: Education and community involvement
Education and community involvement could enhance the adaptive restoration of culturally-
important species to the GRPO meadow. Here, we suggest ways to interest local people in the
project and to make visitors aware of efforts to sustain culturally important species. To encourage
awareness and participation in the restoration effort, GRPO staff could develop displays, leaflets,
and items for purchase in the visitor center.

        Display: For the visitor center, we suggest a poster and dried specimens (pressed plants
or a dry bouquet) describing the four culturally-important species (chives, sunchokes, sweetgrass
and caraway), using information in the literature reviews that accompany this restoration plan.

        Leaflets: Local people and visitors would likely appreciate information about the four
target species. Anecdotes about their history at the site, if known, would also be of interest. A
map could show where chives, sunchokes and sweetgrass are being grown on the meadow edge,
where visitors are welcome to inspect and occasionally harvest plantings.
        Caraway should be highlighted for its cultural and potential economic value, but the
public should also be aware of its potential to become invasive.
        Reed canary grass and smooth brome are target species for eradication, and visitors should
be informed on how to identify and prevent the spread of these species.
A leaflet could ask visitors not to spread them to other sites.
        After mowing and tree removal experiments have been completed, we suggest that an
updated leaflet explain the results of the experimental process and call for volunteers to expand
the restoration approach that is judged to be the most effective.

        Items for purchase: A recipe booklet for chives, sunchokes and caraway could spark the
interest in culturally important plants by both local residents and visitors. Recipes for caraway
could motivate volunteers to collect seed if the species becomes overly abundant. Many recipes
using caraway are available on culinary web sites; they include breads, crackers, cookies, cheese
spreads and a ―seed cake.‖ Recipes could be tested, modified, and published in a booklet for sale
in the visitor center. Samples of the foods could also be sold during events that attract large
numbers of visitors.
        Packets of caraway seed with a GRPO label might be an attractive souvenir for visitors.
As indicated above, the seeds should be microwaved to render them non-viable to reduce chances
of establishing invasive populations elsewhere and to sterilize material intended for use in food.
The label would need to indicate that the seeds are not viable.

       Signage: Signs would be useful at the meadow edge to describe experiments that are
underway or planned and to call for volunteers to assist in the restoration work. Such signs would
highlight the culturing of chives, sunchokes and sweetgrass along the meadow edge.
       While the meadow-edge experiments are underway, signs and tour guides could indicate
the importance of each species, the impacts of overharvesting, and the ongoing restoration of rare
species to the meadow.

        Once the meadow-edge experiments are finished, local residents could be permitted to
harvest some plants for cultivation elsewhere. If the meadow-edge experiments indicate that all
three species benefit from high-light conditions, and if the mowed-meadow strips indicate that
management is needed to sustain these species, the Task Force might encourage human trampling
around selected patches within the meadow.
        Each year, the Task Force could update tour guides with new information on the condition
of the meadow and the progress of the restoration effort. The educational effort should benefit
local youth, visitors, and community members and encourage people to volunteer to help the
restoration and educational efforts. Including local people in the execution and monitoring of the
experiments would promote help preserve the cultural uses and, for sweetgrass, traditional

Adams, C. R., and S. M. Galatowitsch. 2005. Phalaris arundinacea (reed canary grass): Rapid growth
        and growth pattern in conditions approximating newly restored wetlands. Ecoscience 12:569-573.
Apfelbaum, S. I., and C. E. Sams. 1987. Ecology and control of reed canary grass (Phalaris arundinacea
        L.). Natural Areas Journal 7:69-74.
Catford, J. A., R. Jansson, and C. Nilsson. 2009. Reducing redundancy in invasion ecology by
        integrating hypotheses into a single theoretical framework. Divesity and Distributions 15:22-40.
Craig, W.J. 2006. Chive talkin‘. Vibrant Life 2006: 22:20.
Davies. D. 1992. Alliums: The Ornamental Onions. Timber Press, Portland, Oregon.
Derr, J. F. 2008. Common Reed (Phragmites australis) response to mowing and herbicide application.
        Invasive Plant Science and Management 1:12-16.
Foster, R. D., and P. R. Wetzel. 2005. Invading monotypic stands of Phalaris arundinacea: A test of fire,
        herbicide, and woody and herbaceous native plant groups. Restoration Ecology 13:318-324.
Foster, Steven, and J. A. Duke. 2000. Eastern/Central Medicinal Plants and Herbs. Peterson Field
        Guides. 354.
Galatowitsch, S. M., N. O. Anderson, and P. D. Ascher. 1999. Invasiveness in wetland plants in
        temperate North America. Wetlands 19:733-755.
Gigon, A., and R. Langenauer. 1998. Blue Data Books: An encouraging new instrument for
        restoration and conservation. Applied Vegetation Science 1:131-138.
Grand Portage National Monument/Minnesota: Final General Monument Plan/Environmental Impact
        Statement. Available online. <http://www.nps.gov/grpo/parkmgmt/upload/GRPOGMP.PDF>
Grant, E. A., and W. G. Sallans. 1964. Influence of plant extracts on germination and growth of
        eight forage grasses. Journal of the British Grassland Society 19:191-197.
Green, E. K., and S. M. Galatowitsch. 2002. Effects of Phalaris arundinacea and nitrate-N addition on
        the establishment of wetland plant communities. Journal of Applied Ecology 39:134-144.
Growing Taste. ―Sunchokes‖. A home food-gardening resource. 2009. Web. October 2008.
Healy, M. G., and J. B. Zedler. 2010. Setbacks in replacing Phalaris arundinacea monotypes with sedge
        meadow vegetation. Restoration Ecology 18: 155-164.
Herr-Turoff, A., and J. B. Zedler. 2005. Does wet prairie vegetation retain more nitrogen with or without
        Phalaris arundinacea invasion? Plant and Soil 277:19-34.
Ipsen, J. 2010. Effects of varying water and fertilizer levels on Hierochloe hirta (northern
        sweetgrass) with implications for restoration. Biology 152 Independent Research Project
        Report. University of Wisconsin-Madison.
Kays, S.J. and Nottingham S.F. (2007). Biology and Chemistry of Jerusalem Artichoke (Helianthus
        tuberosus L). New York, NY: CRC Press.
Kercher, S. M., and J. B. Zedler. 2004a. Multiple disturbances accelerate invasion of reed canary grass
        (Phalaris arundinacea L.) in a mesocosm study. Oecologia 138:455-464.
Kercher, S.M., Q.J. Carpenter, and J.B. Zedler. 2004c. Interrelationships of hydrologic disturbance, reed
        canary grass (Phalaris arundinacea L.), and native plants in Wisconsin wet meadows. Natural
        Areas Journal 24: 316-325.
Organ, J. 1960. Rare Vegetables for Garden and Table. Faber and Faber, London, U.K.
Plants for a Future: Edible, medicinal, and useful plants for a healthier world. 1996. Web. October
        2008. <http://www.ibiblio.org/pfaf/cgi-bin/arr_html?Carum+carvi>.

Reinhardt Adams, C., and S. M. Galatowitsch. 2008. The transition from invasive species control to
        native species promotion and its dependence on seed density thresholds. Applied Vegetation
        Science 11:131-138.
Seitz, B., and D. Cooper. 2009. Restoring Ethnobotanically Significant Species at Grand Portage
        National Monument. Contract between GRPO and UW-Madison. See scope of work (below).
Tumbledown Farm. ―Jerusalem artichokes (sunchokes)‖. 2009. Web. October 2008.
USDA (USDA, NRCS PLANTS Database). National Plant Data Center, Baton Rouge, LA
        70874-4490 USA. n.d. Web. March 2011. <http://plants.usda.gov/>.
White, D. 2002. Growth and clonal integration of sweetgrass (Hierochloe odorata) in western
        Montana. M.S. Thesis. University of Montana.
Wild Rice Recipes from Minnesota. 1999. Web. October 2009.
WIRCWG (Wisconsin Reed Canary Grass Management Working Group). 2009. Reed Canary Grass
        (Phalaris arundinacea) Management Guide: Recommendations for Landowners and Restoration
        Professionals. Available online. <http://www.ipaw.org/invaders/reed_canary_grass/RCG-
WDNR (Wisconsin Department of Natural Resources). ―Reed Canary Grass Factsheet‖. USGS
        Northern Prairie Wildlife Research Center. 2006. Available online.

           Scope of work: Restoring ethnobotanically significant species
                      at Grand Portage National Monument

    Principal Investigator (PI): Joy B. Zedler, Professor of Botany and Aldo Leopold Chair in
          Restoration Ecology, University of Wisconsin, 302 Birge Hall, 608-262-8629,

 NPS Project Manager: David Cooper, Chief of Resource Management, Grand Portage National
          Monument, 170 Mile Creek Rd., Grand Portage, MN 55605, 218-387-2788,

Introduction and Background: Plants that were demonstrably introduced during the fur trade
era or were historically used by the indigenous Ojibwe culture into recent times are part of the
historic, cultural landscape the Monument seeks to preserve. Consequently, some "exotics" for
which we have conclusive evidence were introduced during the fur trade era or were historically
"domesticated" by Ojibwe residents are a public trust and must be preserved. While substantial
scientifically rigorous data exist on the management of fire, disturbance and invasive species in a
native landscape, very few data exist on management of fire, cultural disturbance and invasive
species in a cultural landscape interspersed with species that are ethnobotanically significant to
GRPO such as sweet grass (Hierochloe hirta), jerusalem artichoke (Helianthus tuberosus), wild
chives (Allium schoenoprasum var. sibiricum) and caraway (Carum carvi). To overcome these
uncertainties, this project will use adaptive restoration, a scientifically formalized process of
‗learn by doing‘.

Adaptive restoration begins by identifying what needs to be known to restore a place (Zedler and
Callaway 2003, Zedler 2005). Key questions are then addressed through the design of testable
field experiments formulated to accomplish restoration objectives. This approach to resource
management will be facilitated through an intensely cooperative process of experimental design
with Joy Zedler as primary investigator. Resource managers at GRPO will then take advantage of
this scientifically rigorous experimental design by testing a number of different management
techniques. As management techniques are tested, a natural progression through the adaptive
restoration model will accomplish resource management objectives.

This work is critical as changes in the environment of the lakeshore unit at GRPO have led to a
significant alteration in the conditions that once fostered the persistence of these ethnobotanically
significant species. Where these species were once found to be vigorous, they are now being lost
to a litany of potential factors including: thatch accumulation, cumulative anthropogenic
hydrological adjustments, succession, invasive species encroachment and allelopathic pollens and
functions. As these negative factors accumulate, these plant species held in public trust by GRPO
are increasingly under-represented within a cultural landscape still frequented by the indigenous
people that created it. It is critical that GRPO respond to this management problem proactively
with sound, scientifically rigorous management techniques to ensure that in the future it is not
forced to react to local extirpation.


1)     Synthesize ethnobotanical and natural histories of: sweet grass (Hierochloe odorata),
     jerusalem artichoke (Helianthus tuberosus), wild chives (Allium schoenoprasum var.
     sibiricum) and caraway (Carum carvi).
2)   Synthesize natural histories of species integral to adaptive restoration such as: sweet grass
     (Hierochloe odorata), jerusalem artichoke (Helianthus tuberosus), wild chives (Allium
     schoenoprasum var. sibiricum) and caraway (Carum carvi), Greene‘s rush (Juncus greenei),
     vasey‘s rush (Juncus vaseyi), smooth brome (Bromus inermis), reed canary grass (Phalaris
3)   Document and map the presence, abundance and distribution of aforementioned species in a
4)   Develop an experimental design for restoring and managing desired species.
5)   Prepare a comprehensive report that documents the adaptive restoration experimental design,
     ethnobotanical and natural histories of plants, methodology, rationale, results, analysis,
     discussion and recommendations for future vegetation management.

Methods: Described as ―learning while doing‖, adaptive restoration is an approach to resource
management that is highly cooperative and depends on new information garnered from the
research and testing of each phase in the adaptive restoration model. An ArcGIS baseline map
will be developed cooperatively with the PI in order to document existing conditions in the field.
The PI will design a scientifically rigorous sampling methodology to lead GRPO resource
management staff, a local botanist and a graduate student in the collection of all data necessary
for the map. GRPO staff will provide labor, data, and equipment to develop the map, and NPS
Great Lakes Network (GLKN) staff will process and document GPS data collected in the field so
it is NPS compliant metadata. NPS GPS data will be used to attribute each component of the map
such as microtopography and distributions of: sweetgrass, chives, caraway, Jerusalem artichoke,
reed canary grass, Canada thistle, trefoil, shrub cover and exotic vetch.
        With the assistance of a local botanist employed at the University of Minnesota Duluth
and a graduate student, the PI will lead GRPO resource management staff in the collection of
detailed field data on sweetgrass assemblages. NPS staff will work with the PI collecting field
data and will provide equipment necessary to spatially attribute the data. Field data will provide
the basis for GRPO‘s experimental planting of sweetgrass with and/or without associated species.
Data analysis and interpretation will be conducted by the PI with assistance from the GRPO
Resource Assistant. The PI and GRPO staff will cooperatively develop an experimental design
for adaptive restoration management. GRPO resource management staff will make final decisions
on the prescription of chemical, cultural and mechanical treatments. GRPO staff will work with
the PI and graduate students to develop a comprehensive suite of potential treatments and
management recommendations based on scientific literature review conducted by graduate
students under the direction of the PI, and ethnographic interviews conducted by GRPO staff.
These treatments and recommendations will be initially vetted by GRPO through an internal
determination of management constraints. GRPO staff will then assist the PI in further directing
graduate student participants in providing scientific justifications for inclusion or exclusion of
remaining experimental treatments. Based on results, treatments will be reassessed and those
found to be compatible with successful restoration of sweetgrass and other ethnobotanically
significant species will be incorporated into an experimental design for adaptive restoration

management. The GRPO archeologist will be present during all ground disturbance associated
with data collection in the field to maintain strict cultural resource protection and compliance.
Timeline: The project commences spring 2009. Protocols, maps and experimental treatment
recommendations will be presented in a draft report for review by Aug. 30, 2011. The park will
have two weeks to review and comment, after which time the primary investigator will have two
weeks to revise the draft and present the final report by Sept. 30, 2011. The investigator is
encouraged to present the draft sooner than Aug. 30, 2011 if time permits. Products: Primary
products for the project include:
 ArcGIS baseline map of microtopography showing distributions of: sweetgrass, chives,
    caraway, Jerusalem artichoke, reed canary grass, Canada thistle, trefoil, shrub cover and
    exotic vetch.
 Detailed field data collection, analysis and interpretation on sweetgrass assemblages (i.e.,
    associated species) for the purpose of testing replanting with and/or without associated
 Recommendations and scientific justification for a suite of experimental chemical, cultural
    and mechanical treatments aimed at:
     1) increasing and subsequently maintaining the population of sweetgrass
     2) maintaining without decrease populations of chives, caraway and Jerusalem artichoke
     3) decreasing the distribution and extent of reed canary grass, Canada thistle, trefoil, shrub
         cover, exotic vetch and any other species that may be deemed important by the PI
 Protocol for monitoring efficacy of experimental chemical, cultural and mechanical treatments
    and distribution/extent of aforementioned species through time.
 Experimental results and interpretation of sweetgrass propagation across a moisture gradient.
 Final report in standard scientific format detailing: literature review, protocol and
    development, methodology, results, interpretation, and recommendations.

        An electronic version of the final report must be submitted; text should be in Microsoft
Word and data tables in MS Excel or Access. An adobe acrobat pdf file may be submitted in
addition to, but not as a substitute for, the MS Word document.
        In addition, the GLNF CESU host university, University of Minnesota, will receive
electronic copies of all related project products upon completion for posting on the GLNF CESU
website www.cesu.umn.edu (or the URL where the project is located, whichever is appropriate).
Project products will be provided in electronic format to NPS Research Coordinator, Jerrilyn
Thompson, Jerrilyn_Thompson@nps.gov

References cited in the scope of work:

Zedler, J. B., and J. C. Callaway. 2003. Adaptive restoration: A strategic approach for integrating
   research into restoration projects. Pp. 167-174 in D. J. Rapport, W. L. Lasley, D. E. Rolston,
   N. O. Nielsen, C. O. Qualset, and A. B. Damania, eds. Managing for Healthy Ecosystems.
   Lewis Publishers, Boca Raton, Florida.
Zedler, J. B. 2005. Restoring wetland plant diversity: A comparison of existing and adaptive
   approaches. Wetlands Ecology and Management 13:5-14.

To top