Long-lived frost fighters

How the winter hardiness trial was done

The level, full sun planting site was in a public area of the Arboretum (lat. 44.5°N, long. 93.3° W; American Horticulture Society Heat Zone 5; USDA PHZ 4b; Hayden loam, pH 7.6) consisting of 10 full sun and two shade beds: each approximately 1200 ft2. In July, 1987, 159 different grasses and sedges were obtained from Bluemount Nurseries (Monkton, Maryland) and Kurt Bluemel Nurseries (Baldwin, Maryland) and planted for the original collection and trials. An additional 233 kinds have since been added, from additional commercial sources primarily Emerald Coast (Pensicola, Florida), Hoffman Nursery (Rougemount, N.C.), Intrinsic Perennials (Hebron, Illinois), Walters Gardens (Zeeland, Michigan), and North Creek Nursery (Landenberg, Pennsylvania). Plants were obtained in 4-inch to 1-gal containers or as plugs or bare root plants. Plugs and bare root plants were potted and maintained in a greenhouse or under shade cloth until sufficient size to be transplanted into the field. For most selections, a minimum of four plants were trialed in rows with spacing equal to the mature height of each plant. Supplemental water was used only in the first year during establishment. Hand weeding and wood chip mulch were used to control weeds, occasionally, a pre-emergent weed control was applied if high weed pressure was observed. Supplemental fertilizer and pesticides were not used. No winter protection was provided for any of the plants.

In early spring (usually late April or early May), grass tops were burned or cut back to remove the growth residue from the previous year. Plants were divided as needed due to rodent damage or loss of vigor. For some plants, division was never needed while for others, it was necessary as often as every five years.

Winter survival data was collected each spring, usually in early June. A rating scale of 0 = dead, 1 = some dieback, and 2 = no winter injury was implemented. Additional data on spring green up, flowering time, overall appearance, height, width, and pest damage was collected monthly throughout the growing season by the primary author and has been reported elsewhere (Meyer MH. 2012).

Results

Weather data shows the coldest temperatures of -31, -32, -35, and -37 °F during 2019, 1994, 1998, and 1997, respectively. Over the 35 years of this trial, the average minimum winter temperature was -20.5°F, consistent with the -20 to -25°F typical for USDA PHZ 4b. Average annual rainfall from 1987-2022 was 32 inches, and the average growing season or number of frost-free days was 158.

Plants that did not survive and are rated as non-winter hardy for USDA Zone 4b are listed in Table 1. Plants that are hardy are listed in Table 2, along with the number of years in the trial. Scan the QR code on pg. 27 or go to bit.ly/MNgrasses to view the trial results tables.

Challenges with long-term herbaceous trials

Over the 35 years, approximately 30 Chinese silvergrass (Miscanthus) seedlings were found in the adjacent woods and open fields near the original plantings. These seedlings were removed annually. Scouting for seedling grasses in adjacent areas was part of ongoing management. Without this annual scouting and removal, Chinese silvergrass could have increased and become naturalized and perhaps invasive as it has in the eastern U.S. (USDA 2023). No other plants in this trial were found self-seeding outside of the original planting location.

Chinese fountain grass (Pennisetum alopecuroides) was one of the most problematic grasses. Plants often showed large dead sections of winter damage with lopsided growth in the spring. Even so, by fall these plants were filled with inflorescences that disarticulated (fell apart) and self-sowed in nearby turfgrass and planting beds, thus becoming weeds. Reduced seed set in new cultivars of Chinese fountain grass (Hanna and Schwartz 2020a) and Chinese silvergrass (Hanna and Schwartz 2020b; Ranney and Touchell 2016) can reduce weed and invasiveness problems. These reduced seed set cultivars have lived 4 years in our trials.

Maintaining genetic integrity can be challenging in a long-term herbaceous trial due to seedling establishment from nearby plants and variation in vigor among cultivars. This proved especially difficult in purple moorgrass (Molinia arundinacea and M. caerulea) and switchgrass (Panicum virgatum). Annual inspection of each cultivar throughout the growing season and especially during flowering was essential to remove errant seedlings.

Originally, the plant residue or growth from the previous year was burned annually before new growth began each spring, usually in April. Burning rejuvenates perennial grasses and reduces crown and foliage diseases but can kill overwintering Lepidoptera or butterfly larvae. In recent years, the native grasses in this collection have been cut back and the tops removed in early spring, with burning only occurring every third year. This allows overwintering larvae to survive in the base of the grass crown, thus supporting pollination insects (Narem and Meyer 2017).

Variegated cultivars such as Avalanche, Overdam, and El Dorado feather reedgrass (Calmagrostis ×acutiflora) required removal of all-green sections, as did variegated cultivars of other grasses. No fungal pests required management; however, field ants (Formica sp.) did considerable damage in some grasses such as tufted hairgrass (Deschampsia caespitosa) and fall reedgrass (Calamagrostis brachytricha) by building their nests directly in the grass crowns. We used ant baits and often resorted to dividing the entire plant and disruption of the ant colony. Voles (Microtus sp.) required management effort, especially during the fall and winter, when bunch grasses provided the perfect habitat and food source during the winter. Baits and repellents were utilized as annual management efforts against voles.

Why this research matters

Without field trial data, growers and retail garden centers understandably are reluctant to risk offering plants with unknown hardiness to consumers. Having research data enables growers to put correct hardiness information on plant tags and catalogues. Accurate hardiness labeling is enforced in some states which often rely on field trials to determine their ratings (Minnesota Department of Agriculture 2023). We were fortunate to have planted these trials in a public arboretum that had space and resources to dedicate to a long-term research project. This enabled gardeners, as well as growers and retailers, the opportunity to view the plants throughout the seasons and years. Additional detailed features of specific grasses grown in this study and their use in the landscape has been published (Narem and Meyer 2020; Meyer 2012). This winter hardiness information demonstrates the survival likelihood of many landscape grasses in the upper midwestern U.S. and will help inform growers, retailers, and gardeners as well as consumers when selecting grasses to grow in cold climates.

Mary Hockenberry Meyer is professor emeritus and Grass Collection curator at University of Minnesota Landscape Arboretum in Chanhassen, Minnesota. Brandon Miller is an assistant professor at University of Minnesota.