Native warm season grass variety trial, 2021-2023

Introduction

While introduced, cool-season grasses dominate Virginia pastures, there is growing interest in filling the traditionally slow period of forage productivity during the summer months with alternative forages.

Native warm season grasses (NWSG) can fill typical summer forage deficits. These grasses are productive and nutritious for livestock during the summer months. They are well-adapted to Virginia’s soils, pests, and climate. With deep root systems, these grasses perform well even in mild droughts, and nutrient inputs, such as fertilizer, are less critical compared to the input requirements of their nonnative counterparts. These grasses can even handle a lower pH compared to cool-season grasses.

While these grasses are highly productive in the summer, cool weather diminishes their productivity. Thus, common recommendations for Virginia are to include no more than 20-30% of the grazing system in NWSG species.

In addition to offering these production benefits, NWSG have an important role to play in wildlife conservation. The robust, upright form and open space between plants in a NWSG stand provides the type of habitat required for foraging and nesting by bobwhite quail and other ground nesting birds. These grasses shelter small mammals and birds from predators when left standing overwinter, even after heavy snow events.

Native warm season grasses can provide food for livestock and wildlife alike. Under proper management, NWSG provide highly nutritious forage and can persist in pastures indefinitely.

Unfortunately, their adoption has been minimal. Lack of familiarity, historic challenges with establishment, and misperceptions and uncertainty surrounding nutritional quality and stand management largely account for farmer reluctance to adopt NWSG in Virginia.

While new techniques and improved cultivars have led to more success in the establishment process, native grasses require slight differences in management during establishment and subsequent grazing management practices

In addition, there is little information available on various cultivars and ecotypes of native warm season grasses recommended for forage purposes. Cultivars are defined as selected, improved varieties, while ecotypes are generally defined as varieties selected from a single population, usually in a specific, undisturbed area.

Forage Variety Trial

The purpose of this variety trial was to determine the regional productivity of various NWSG cultivars or ecotypes of four different species of native grasses: big bluestem (Andropogon gerardii), eastern gamagrass (Tripsacum dactyloides), indiangrass (Sorghastrum nutans), and switchgrass (Panicum virgatum).

This trial was established in 2020 at five locations across Virginia. These locations encompass the dominant hardiness zones (Figure 1) and physiographic provinces of Virginia. Soil types and test result values for each location are reported in Appendices 1-5. The five locations of this variety trial were:

1. Middleburg Agricultural Research and Extension Center, Middleburg, Virginia
2. Shenandoah Valley Agricultural Research and Extension Center, Raphine, Virginia
3. Southern Piedmont Agricultural Research and Extension Center, Blackstone, Virginia
4. Southwest Agricultural Research and Extension Center, Glade Spring, Virginia
5. Tidewater Agricultural Research and Extension Center, Suffolk, Virginia

Establishment

A mixture of glyphosate (41%) at 2 qt/ac and 0.5% nonionic surfactant was sprayed on plot areas the fall prior to seeding, except for the Middleburg location which was sprayed one week prior to seedbed preparation. Seedbeds were prepared for planting through discing following by harrowing or rototillage. The plots were then rested for a minimum of one week to allow the soil to settle, with the exception of the Raphine location which was cultipacked immediately following tillage.

Plot sizes were six by ten feet with four replications per cultivar. A Carter forage plot seeder was used to seed all of the species except the eastern gamagrass, which was planted with a single row push corn planter in four strips per plot due to the large seed size of this species. Seeding depth was less than ¼” of an inch for all of the species except the eastern gamagrass, which was planted at ¾-1” depth. All cultivars were planted on a pure live and non-dormant seed basis with 9 lb/ac for the big bluestem, 14 lb/ac for the eastern gamagrass, 9 lb/ac for the indiangrass, and 6 lb/ac for the switchgrass. A fixed amount of pelletized lime was used as a carrier for all of the species and varieties.

All plots were sprayed again with a mixture of glyphosate (41%) at 2 qt/ac and 0.5% nonionic surfactant at planting.
Plots were established on the following dates:

• May 27, 2020: Blackstone
• June 2, 2020: Glade Spring
• June 11, 2020: Raphine
• June 24, 2020: Suffolk
• July 14, 2020: Middleburg

Rainfall from May through July at each location was within 1.5” of the 30-year precipitation mean for those three months.

Approximately two months following planting, plots at all locations except for Blackstone were clipped with a rotary mower to remove weed biomass above the native grass seedlings. The mower was set to a height at or above the tallest height of the native grasses.

Germination Evaluation and Statistical Analysis

Seedling germination was evaluated around sixty days following establishment at Raphine and Blackstone. A 0.5 m2 quadrat was placed directly in the center of each plot, and the number of native grass seedlings was counted within the quadrat.
Seedling count by cultivar was compared within a species using PROC MIXED in SAS Studio, v. 9.4 (SAS Inst., Cary, NC). Differences were considered significant when P ≤ 0.05 and as trends when 0.05 < P ≤ 0.10.

Harvest Management and Statistical Analysis

Plots were harvested once in 2021 at Raphine, Blackstone, and Suffolk. Plots were harvested twice in 2022 at those three locations and once at Middleburg. Plots were harvested once in 2023 at Middleburg and Glade Spring and twice at Raphine, Blackstone, and Suffolk. No fertilizer or soil amendments were applied to the plot area at any location for the duration of this test. After each harvest, all biomass was removed from the plots. The monthly precipitation for the county where each trial was located is presented in Appendix 6.
Cutting height was set to 5” for all plots at harvest. Total biomass harvested from the plot area was recorded, and a subsample was collected for dry matter corrections. The subsamples were dried in a forced oven at 135 °F for a minimum of four days before weighing. The dry matter correction was applied to the total fresh plot weight to calculate dry matter yield per unit area.
Prior to harvest, percent cover of the cultivar and the weed pressure by type of weed (grass, broadleaf, and sedge) was scored in each plot using a modified Daubenmire ranking scale (1: 0-5.0% cover; 2: 5.1-25.0% cover; 3: 25.1-50.0% cover; 4: 50.1-75.0% cover; 5: 75.1-95.0% cover; 6: 95.1-100.0% cover).
Plots were sprayed with 3 qt/ac pendimethalin (38.7%), 1 qt/ac glyphosate (41%) or 1 qt/ac triclopyr (60.5%), and 0.5% non-ionic surfactant in April 2022 to reduce weed competition at all locations.
Forage dry matter yield by cultivar was compared within a species using PROC MIXED in SAS Studio, v. 9.4 (SAS Inst., Cary, NC). Locations were analyzed separately for the tables presented in the appendices. Differences were considered significant when P ≤ 0.05 and as trends when 0.05 < P ≤ 0.10.

Results

Germination

Seedling count by cultivar is shown in Table 1. There tended to be a treatment (cultivar) by location interaction (P=0.0642). Some cultivars had lower germination than others, but by the second year, these differences had an indistinguishable effect on yield, likely due to germination of dormant seed and tillering of seedlings in the first year.

¹ Least significant means

² Standard error

*significantly different from the highest numerical value within the same species based on 0.05 LSD

**tended to be significantly different from the highest numerical value within the same species based on 0.10 LSD

Yield

Seasonal yields (sum of all of the harvests at a location each year) were analyzed within a species across all five locations. There was no treatment (cultivar) by location interaction for any of the species (P>0.05). However, there tended to be treatment by location interaction for indiangrass (P=0.0964).

Mean seasonal yields for each cultivar are presented as averages across all locations in Table 2. There were no significant differences in yields of cultivars within a species. However, cultivar tended to have a significant effect in the comparison of indiangrass due to the poor germination and growth of two cultivars, ‘NC ecotype’ and ‘KY ecotype.’ For these two cultivars, some of the plots were not harvested due to absence of the cultivar of interest.

¹ Least significant means

² Standard error

*tended to be significantly different from the highest numerical value within the same species based on 0.10 LSD

Although differences were not analyzed across species, it may be useful to note the greater yields of switchgrass and eastern gamagrass compared to big bluestem and indiangrass. While the two former species do in general produce more biomass than the two latter species, the yield data presented here are likely biased towards switchgrass and eastern gamagrass because they are earlier maturing species. Due to all species and plots within a location harvested on a single date once or twice a year, the earlier maturing species would indicate greater yields than the later maturing species. Farmers considering certain species for selection should consider nutritive value and palatability, however, not just forage yield.

Seasonal yields are also presented by year for each location in Appendices 7-11. Plots with insufficient cover of the target cultivar for a reliable estimation of yield within the plot were not harvested. Poor establishment of some species at some locations (e.g. indiangrass at Middleburg) resulted in very high standard errors. Thus, caution should be taken when

Germination and eventual productivity of indiangrass ecotypes (‘GA ecotype’ and ‘NC ecotype) were lower than for the other cultivars and ecotypes at four of the locations. This was a common pattern across all locations with the exception of Glade Spring, where indiangrass productivity was poor across cultivars and ecotypes due to poor establishment at this site.

Ecotypes are defined as seeds from an unimproved selection of seed from a given area. These varieties are thought to be better adapted to the region from which they were selected due to their extended period of evolution in that region. Cultivars, however, are improved lines of plants developed through selective breeding processes to target specific desired characteristics. These characteristics may include growth, forage nutritive value, and diseases resistance. As a result, improved cultivars often, but not always, may be expected to be higher yielding than ecotypes.

Conclusions

While this project yielded a substantial amount of information on the productivity of various cultivars and ecotypes of NWSG across Virginia, it also reinforced a few points that anyone interested in getting these species established for forage production or wildlife habitat may find helpful.

Germination of these perennial species can be slow, especially compared to the growth of nonnative annual weeds. Weed control through advanced site preparation and follow-up herbicide or mechanical control is imperative for the success of the planting. In addition, full productivity of these species may not be realized until two seasons following establishment.

Once these species are established, they can be very productive, even without fertilizer or lime applications. In the case of this variety trial, no soil amendments were applied to the plots despite the removal of biomass from the plots following each harvest. In a grazing system, these nutrients would be largely recycled through the grazing animals and thus very little soil amendments would be necessary for the optimum productivity of these species. In a hay production system where the vegetation is removed from the field year after year, it would be advantageous to follow soil test nutrient recommendations when applying fertilizer and lime.

In general, the cultivars matured in the following order: eastern gamagrass, switchgrass, big bluestem, and indiangrass. This may be helpful information when selecting a species based on when forage is most needed and to prevent growing season overlap with the rest of the forage system. It is also not recommended to mix eastern gamagrass or switchgrass with plantings of the other species due to their more rapid rates of maturity. If a mixture of species is desired for forage production purposes, it is helpful to pair species and cultivars together with similar maturity rates so that harvest can be more appropriately timed to the needs of the crop.

Greater biomass yields may be useful in some contexts (maximizing yield for forage and biomass production purposes), but in some situations, such as in wildlife or conservation plantings, too much biomass may not be advantageous. In wildlife habitat plantings, a thinner grass sward may be more beneficial to small birds and mammals building nests and burrows within the stand. In these situations, ecotypes may be a better choice.

Acknowledgements

Seed for this trial was donated by Ernst Conservation Seeds, Gamagrass Seed Company, Garrett Wildflower Seed Farm, Roundstone Native Seed, and Star Seed, Inc.

This trial was made possible by the research farms and crews at each location and with the support of the Virginia Agricultural Experiment Station. A Conservation Innovation Grant from the United States Department of Agriculture Natural Resources Conservation Service provided funding for this work.

¹ Species: BB = big bluestem, EG = eastern gamagrass, IG = indiangrass, SG = switchgrass

² Least significant means

³ Standard error

⁴ Maturity ranking: V = vegetative, E = elongating, R1 = boot stage, R2 = fully emerged, R3 = anthesis, S = mature seed

⁵ Insufficient data to report

* Yield was significantly different from the highest numerical value within the same species based on 0.05 LSD

¹ Species: BB = big bluestem, EG = eastern gamagrass, IG = indiangrass, SG = switchgrass

² Least significant means

³ Standard error

⁴ Maturity ranking: V = vegetative, E = elongating, R1 = boot stage, R2 = fully emerged, R3 = anthesis, S = mature seed

* Yield was significantly different from the highest numerical value within the same species based on 0.05 LSD

** Yield tended to be significantly different from the highest numerical value within the same species based on 0.10 LSD

¹ Species: BB = big bluestem, EG = eastern gamagrass, IG = indiangrass, SG = switchgrass

² Least significant means

³ Standard error

* Yield was significantly different from the highest numerical value within the same species based on 0.05 LSD

** Yield tended to be significantly different from the highest numerical value within the same species based on 0.10 LSD