DOT Project Number:  90-00-LRTF-506

Fiscal Year:  2005

Award:  $9,000

Principal Investigator:  Dr. Brian Wilsey, Department of Ecology, Evolution and Organismal Biology, Iowa State University, bwilsey@iastate.edu

Summary Report:

WARM SEASON GRASS EFFECTS ON DIVERSITY AND WEED INVASION RESISTANCE

The overall objective of this project is to better understand how high grass dominance develops in restoration plantings.  Seedlings of a single warm-season grass species were planted into each plot, and then a mix of forb and cool-season grasses was added (same mix for each plot) to determine mechanistically how the grasses were suppressing prairie establishment.  Bare ground plots without a grass treatment served as a control for grass effects.  Five species of warm-season grasses were used, and locally collected seed was compared to cultivar seed (see proposal for full description of the design).   Establishment, species diversity, light and water availability are being monitored over time in all plots.  Plots were planted in Monona County in May 2005.

The hypotheses for the experiment are the following:

H₁  Differences will occur in species diversity due to the identity and seed source of the dominant grass species.  Fast growing grass species will cause species diversity to be lower in developing prairie communities.  Cultivars will depress diversity more than non-cultivar seed.
Alternative:  There will be no difference among different dominant grass species.

H₂  The amount of dominance by grasses on recruiting prairie forbs will differ between 5 grass species mixtures and monocultures.  This is because of the lower relative abundance of each species in mixtures as well the more diverse competitive environment in mixture.  Lower relative abundances in mixtures will prevent any one species from developing strong dominance.  Having multiple species may lead to having different micro-patches for forb species to establish.
Alternative:  There will be no difference, or even less establishment in mixtures compared to monocultures due to having a greater proportion of overall niche space filled.

H₃  Grasses will suppress forb diversity compared to bare ground plots.  
Alternative: Grasses will increase forb diversity due to their greater suppression of weeds.

Experimental design

During spring 2005, we set up a new experiment with planted seedlings to test these  hypotheses.  Equal-sized transplanted seedlings rather than seed were used to 1) help to successfully and evenly establish the grass species, and 2) to help to control weeds during the initial grass establishment phase.  Both of these things will help me to better address my objectives.  Plots were weeded for one month until grasses established and the first seedlings begin to emerge from the prairie seed mix.  From this point onward, we have been monitoring weed invasion in addition to grass and forb seedling establishment.  (However, there were very few weeds in plots as of May, 2006.) 

The design is a 6 x 2 factorial treatment design (6 species treatments x 2 seed sources) plus a few bare ground controls with no grasses present (Figure 1). The six grass species treatments are: 1) switchgrass Panicum virgatum, 2) big bluestem Andropogon gerardii, 3) little bluestem Schizachyrium scoparium, 4) side-oats grama Bouteloua curtipendula, and 5) Indian grass Sorghastrum nutans, or 6) all five grass species combined.  The two seed sources are 1) Cultivar seed or 2) Local genotype seed.

Seedlings were planted in blocks on 3 aspects (N, E, and SW) to increase the generality of the study.  These aspects varied the moisture availability from medium (north slope) to low (southwest slope).  There were two 2 replicates of each grass treatment (except for mixtures, see below) on each slope, or a total of  5 species x 2 seed sources x 3 blocks x 2 rep’s for a total of 60 monoculture plots (tests hypothesis 1).  There were 4 replicate mixtures for each of the 2 seed sources on each aspect for a total of 24 5-grass-species mixture plots (tests hypothesis 2).  Four bare ground plots were also included in each block for 12 bare plots.  These will be used to test hypothesis 3.

Each plot was planted with 70 plants to be close to estimates of plant density in a prairie near Sylvan Runkel Preserve (Losure, Wilsey and Moloney in review at Oikos, individual plants were counted shortly after a fire which enabled individual plants to be observed) (Figure 2).  Thus, a total of 5,880 seedlings were germinated, grown and planted (70 plants per plot x 84 plots = 5,880 plants) during winter and early spring 2005.  Seedlings were grown in ISU greenhouses in 2 parts potting soil and 1 part sand, trimmed to be similar size across species, and then acclimatized to field sun and wind for one week before planting.  Seedlings were transported to Monona County in a moving van on May 24, 2005, and planted into plots on May 25 and 26, 2005. 

Plant biomass was not significantly different among species in seedlings that were planted.  Seedling survival, which was checked during the following week was 99.7% (5,862/5,880); dead seedlings were replaced at this time.  Seedlings have established very well.

A seed mix containing the species listed in Table 1 in the proposal was added to each plot on June 15, 2005 and then again in early December, 2005.  Seed was scattered on the snow in December.

Blazing star seedling experiment

Blazing star (in this case, Liatris punctata) is a conservative prairie species with showy purple flowers.  It is being used here to represent conservative species that are highly prized but  difficult to establish in grass-dominated plantings (Losure, Wilsey and Moloney in review).  On April 27, 2006, I added two small two-week old seedlings to each plot.  These seedlings will be harvested in September, 2006 to estimate how the treatments affect conservative species growth and flowering rates.  Shoot and root biomass, flower number,  and height will estimate establishment success, and results will be reported in June, 2007. 

 Results

Prairie species establishment

The first species to emerge from the prairie seed mix was hoary vervain Verbena stricta, and it had enough seedlings to analyze statistically during the first year of emergence.  Other species that were beginning to emerge, but with not enough seedlings (yet) to analyze statistically are ox-eye Heliopsis helianthoides, black-eyed susan Rudbeckia hirta, Canada wildrye Elymus canadensis, tall dropseed Sporobolus asper, purple prairie clover Dalea purpurea, and others.  These species, as well as all other species from the seed mix, will be monitored and reported on in the future.  Gumweed Grindelia squarrosa, skeleton rush Lygodesmia juncea, and whorled milkweed Asclepias verticellata were included in the first weeds to establish, although at a rate too low and infrequent to report on at the present.

Here, I will report on the abundance and percent cover of Verbena stricta, assuming that this species is a general indicator for prairie emergence and establishment (Figures 3 and 4).  Verbena was more abundant and had higher percent cover in plots planted with local genotype seed than in plots planted with cultivar seed (Figure 3, 2006 abundance data, F_1,48 = 5.0, p = 0.03, 2006 percent cover data, F_1,48 = 5.4, p = 0.03).  Interestingly, I also found that Verbena was higher in plots planted with little bluestem and to some extent, side-oats grama than in bare ground plots or plots planted with taller species (Figure 4, 2005, 2006 abundance data, and 2006 percent cover data all significant, p < 0.05).  There were roughly twice as many Verbena plants in plots with these two short grasses compared to plots with the tallgrasses big bluestem, indian grass, or switchgrass.  The fact that twice as many Verbena plants came into little bluestem plots than in bare ground plots suggests that this species is facilitating Verbena establishment.

Figure 3.  Establishment of the prairie forb hoary vervain Verbena stricta (means+SE) in experimental plots planted with either big bluestem (BB), indian grass (IN), little bluestem (LB), side-oats grama (SO) or switchgrass (SW).  Grasses were planted at a common density of 70 plants per m².  Bare ground plots (no grass) of a similar size were also included to compare responses to plots without any grasses planted.  Different letters denote significantly different treatments.

Figure 4.  Differences in Verbena abundances (mean+SE) between plots planted with cultivar seed or local seed (data are averaged across species treatments) during 2005 (a) and spring 2006 (b).

Figure 5.  Differences in Verbena percent cover (mean+SE) between plots planted with cultivar seed or local seed (data are averaged across species treatments) during 2005 (a) and spring 2006 (b).

Figure 6.   Proportion of light at the soil surface (below/above canopy) in plots planted with big bluestem (BB), indian grass (IN), little bluestem (LB), side-oats grama (SO), and switchgrass (SW).  Light was measured with a Decagon ceptometer light meter.  Side-oats grama and little bluestem had more light striking the soil surface during 2006 months (p < 0.05).

Canopy light penetration

The amount of light striking the soil surface under the grass canopies differed significantly across grass species, which probably explained much of the difference in Verbena establishment.  By June 2006, the amount of light striking the soil surface was highest in plots planted with side-oats grama and little bluestem, and was lower in plots planted with taller species (Figure 6).  There were no consistent differences in light availability under cultivar and local seed canopies (Figure 7).

Individual plant characteristics

Individual plants were measured for traits related to light capture and spatial spread to determine if variation existed among grasses.  These variables are expected to be important predictors of species suppression.  Plant height varied significantly among species, but not consistently between cultivar and local seed plants (Figure 8).  Switchgrass was the tallest plant species, and side-oats grama and little bluestem were the shortest species, which follows long standing understanding of plant heights among these species.  Indian grass and switchgrass cultivars were shorter than their local seed counterparts (Figure 8).  The other grass species had cultivars that were taller than their corresponding local seed plants.

Basal area, which is the area of the plant at the soil surface, was estimated during three time periods with the following formula:
area = Br²
where r was estimated by measuring the diameter in two directions for three individual plants per plot.  Diameter measurements were divided by two to get estimates of radius.  Basal area was measured to get a measure of the amount of spatial spread by clones of each species during the establishment phase (year 1 and 2 of the experiment).

Basal area was not correlated with height (p > 0.05).  That is, plants that were tall did not necessarily have higher basal areas and plants that were short did not necessarily have smaller basal areas.  In 2005, the highest basal area was in the shortest species side-oats grama (Figure 9).  The smallest basal areas occurred in indian grass and big bluestem.  By spring 2006, the rankings among species were similar except that switchgrass had passed side-oats grama as the species with the greatest basal area. 

Basal area was significantly greater in cultivar plants than in plants from local seed (Figure 10).  These differences were largely consistent across species.

Combining height and basal area data into a principal components analysis provided an index that described the differences among grass species (data not shown).  Some species could be characterized by being short but wide at the base (side-oats grama), whereas other species could be characterized by being tall but narrow at the base (big bluestem).   These differences in growth structure are likely to be highly important to prairie establishment and weed suppression, both of which will be measured over time as the study continues.

Figure 7.  Light at the soil surface in plots planted with cultivars or local seed.  There was no significant difference between the two.

Figure 8.  Height to the base of the first emerging leaf in grasses planted with different warm season grasses from either cultivar or local seed.  There were significant differences between cultivars in indian grass (IN), side-oats grama (SO), and switchgrass (SW), but they varied in direction.  That is, sometimes the cultivar was taller (SO) and sometimes the local seed plants were taller (SW).

Figure 9.  Basal area in grasses planted in Monona County over three time periods.

Figure 10.  Basal area (size of plant crown at soil surface in cm²) in plants from cultivar or local seed.

Conclusions and recommendations

Warm season grasses are important to prairie establishment for two reasons.  They provide early cover that probably prevents heavy weed build-up (this will be tested by comparing my bare ground plots to the grass plantings).  Prairie restorationists traditionally  assume that the grasses are providing the majority of weed resistance (Packard and Mutel 1997).  In the case of my plots, all plots with grasses planted are relatively weed free.  However, some warm season grasses can become too aggressive, choking out prairie plants as well as weeds (e.g. Sluis 2002, Baer et al. 2004, Martin et al. 2005, Polley et al. 2005, Martin and Wilsey 2006). 

The objective of this study was to study how different grasses affect weed and prairie establishment.  With adequate knowledge of this subject, we can provide management recommendations on which species to use and their seed proportions in initial mixes, and which to add in later years.  For example, a grass that prevents weeds from establishing but that is not so aggressive that it prevents the establishment of prairie forbs and legumes would provide an excellent choice for a native cover crop.  Grass species that are too aggressive could be kept out of the mix until later years. 

Based on my results so far, I tentatively suggest that little bluestem and side-oats grama are the best warm season grasses for seed mixes, especially in areas with relatively low water availability.  Other species could be added at a later date.  Little bluestem and side-oats grama allowed more light to strike the soil surface (important to prairie seedlings), and facilitated the recruitment of Verbena stricta the first species to establish from the seed mix.  Assuming that Verbena is an indicator of how other species will establish, then the best establishment will be in little bluestem and side-oats grama.  (The establishment of other species will be tested in 2006 and beyond). 

Cultivars are planted in Conservation Reserve Program plantings at the national level, as well as in some restorations in other states (Baer et al. 2004).  Because of this, it is important for us to determine scientifically whether cultivars differ from non-cultivars in their growth characteristics and their suppression of other prairie species.  We found here that cultivars reduced the establishment rate of Verbena compared to non-cultivars.  Cultivars were not consistently taller than non-cultivars, but they did have greater basal diameters.  The switchgrass cultivar was especially aggressive, with especially large basal diameters developing by May 2006.  I will continue to study whether cultivars suppress the establishment and diversity of other prairie species over time.

Prairie species establishment, including the planted seedlings of Liatris punctata, as well as soil moisture and light, which is important to prairie establishment (Wilsey and Polley 2003, Martin and Wilsey 2006), is being be measured in 2006 and beyond.

References

Baer, S.G., Blair, J.M., Collins, S.L. & Knapp, A.K. (2004) Plant community responses to resource availability and heterogeneity during restoration. Oecologia, 139, 617-629.

Martin, L.M. and B.J. Wilsey.  (2006)  Assessing grassland restoration success: relative roles of seed additions and native ungulate activities.  Journal of Applied Ecology  in press.

Martin, L.M., Moloney, K.A. & Wilsey, B.J. (2005) An assessment of grassland restoration success using species diversity components. Journal of Applied Ecology, 42, 327-336.

Packard, S. & Mutel, C.F. (1997) The tallgrass restoration handbook for prairies, savannas, and woodlands. Island Press. Washington D.C.

Polley, H.W., Derner, J.D. & Wilsey, B.J. (2005) Patterns of plant species diversity in remnant and restored tallgrass prairies. Restoration Ecology, 13, 480-487.

Sluis, W.J. (2002) Patterns of species richness and composition in re-created grassland. Restoration Ecology, 10, 677-684.

Wilsey, B.J. & Polley, H.W. (2003) Effects of seed additions and grazing history on diversity and productivity of subhumid grasslands. Ecology, 84, 920-931.