Floristic Quality of Native Tallgrass Pastures in Eastern South Dakota

Final Report for GNC08-098

Project Type: Graduate Student
Funds awarded in 2008: $10,000.00
Projected End Date: 12/31/2010
Grant Recipient: South Dakota State University
Region: North Central
State: South Dakota
Graduate Student:
Faculty Advisor:
Dr. Alexander Smart
South Dakota State University
Expand All

Project Information


Historic herbicide use and grazing have influenced natural diversity and quality of native pasturelands in the Great Plains. Floristic quality assessments are useful to assist agencies in prioritizing conservation practices to enhance native grasslands. The objective of this study was to determine the effects of past land use practices on the floristic quality of remnant native pastures in eastern South Dakota and southwestern Minnesota. Floristic quality assessments were conducted on 30 native pastures and categorized by past management practices (herbicide application and grazing intensity). Mean coefficient of conservatism and floristic quality index (FQI) were calculated for each site. Results showed that increased herbicide use and grazing intensity resulted in a lower species richness, mean forb coefficient of conservatism, and FQI. However, grass and grasslike plants were minimally affected. Pastures that were infrequently sprayed with herbicides and lightly grazed consistently had the highest species richness, mean coefficient of conservatism, and FQI. Pastures with no grazing produced similar values to those with moderate grazing. Pastures managed as preserves/wildlife habitat areas had higher FQI than those managed for livestock grazing. The implications of this study should further help ecologists and managers understand the positive and negative effects of grazing practices and herbicide application on tallgrass prairie remnants.


The natural landscape of the eastern Great Plains has been immensely altered since pre-European settlement. Large areas of land have been put into agricultural production and as a result less than 1% native tallgrass prairie remains. Remnant tallgrass prairies display varying degrees of quality due to habitat disturbance and invasion by exotic species. Long-term use of herbicides that control broadleaf species has resulted in decreased forb diversity. Exotic cool-season grasses such as smooth bromegrass and Kentucky bluegrass have increased greatly in the Great Plains. Anthropogenic additions of nitrogen deposition have contributed to the competitiveness of exotic cool-season grasses. It is likely that remnant grasslands of the eastern Great Plains have experienced greater grazing intensities since the removal of the bison, fencing of pastures, and introduction of domestic livestock. Grazing intensity has been shown to decrease the vigor of pasture forage species and increase weedy species. Therefore, tallgrass prairie remnants are a high priority for conservation by natural resource agencies. The objectives of this study were to determine the effects of past management practices, involving varying levels of herbicide application and grazing intensity, and whether the primary land use (either nature preserve/wildlife production or livestock grazing) has had an impact on floristic quality of native prairies pastures in the Prairie Coteau of eastern South Dakota and southwestern Minnesota. The literature suggests that herbicide use and grazing intensity decrease the floristic quality of native pastures and that prairies managed as nature preserves/wildlife areas should have higher floristic quality than pastures managed for livestock grazing.

Project Objectives:

1) Determine floristic quality inventory and effects of past management history on at least 16 privately owned native pastures.
2) Educate producers, extension and NRCS personnel about past management influences on floristic quality of native pastures.
3) Assist NRCS personnel in the development of ecological site descriptions and plant community models for MLRA 102A in eastern SD.


Click linked name(s) to expand/collapse or show everyone's info
  • Alexander Smart


Materials and methods:

Thirty native tallgrass prairie pastures in the Prairie Coteau Ecoregion of eastern South Dakota and southwestern Minnesota were sampled for floristic quality, as the assessment of native plants’ degree of dependence on intact native plant communities, in the field seasons of 2006 and 2009 (Fig. 1). Pasture size averaged 88 ha and ranged from 12 to 810 ha. Field methodology followed standard protocol for floristic quality index (FQI) assessment. Pastures were surveyed in July/August to identify all vascular plant species within a parcel of the pasture until no new species were detected after 10 minutes of additional searching. Once the 10 minute period expired the surveyor moved to a new location within the pasture. This procedure was repeated until no new species were found within the entire pasture. The average time spent surveying pastures was 5 minutes per ha. Each plant is assigned a conservatism value. Conservatism or coefficient of conservatism values are integral values ranging from 0 to 10, with 0 assigned to species typifying disturbed habitats and 10 assigned to the most conservative species, i.e., those occurring strictly in undisturbed habitats. The coefficient of conservatism value thus represents the plant species’ ability to indicate or predict the quality of a natural area. Exotic species are not assigned a value. Mean coefficient of conservatism and FQI values were calculated for each pasture based on floristic composition. The mean coefficient of conservatism for a particular land parcel is thus the average of coefficient of conservatism values for all of the native species occurring on the parcel. Floristic quality index is calculated by the following equation:
FQI= mean C?N,
where mean C is the mean coefficient of conservatism for a site, and N is the total number of native plant species found. This procedure is not density dependent and thus mean C and FQI values for a land parcel are stable over time, at least provided that the management doesn’t change, and given the fact that most prairie plants are perennials and are observed during both wet and dry years.
Herbicide use was categorized as frequent if it was broadcast over an entire pasture annually or up to every 3 to 4 years and infrequent if the pasture was only spot sprayed when and where necessary. Grazing intensity was categorized as no grazing, light, moderate, and heavy relative to the recommended stocking rate for South Dakota rangeland and pasture and based on fall stubble height. Light grazing was usually practiced seasonally (spring or summer) for less than 4 months to achieve utilization of approximately 25% or less of the annual herbage production. Moderate and heavy grazing was usually practiced season-long to achieve an approximate utilization of 50% and greater than 65%, respectively. Ranchers and managers verified that land parcels had never been plowed and provided past management information that allowed us to categorize the combined use of herbicides and grazing intensity into six categories [frequent herbicide-heavy grazing (n=8), frequent herbicide-moderate grazing (n=6), infrequent herbicide-heavy grazing (n=3), infrequent herbicide-moderate grazing (n=3), infrequent herbicide-light grazing (n=6), and infrequent herbicide-no grazing (n=4)]. Primary land use was categorized as either nature preserve/wildlife habitat (n=12) or livestock grazing land (n=18). A one-way analysis of variance was used to compare herbicide-grazing combinations and primary land use effects using PROC GLM (SAS, 2009). Normality of the residuals was verified using the NORMAL and PLOT options in PROC UNIVARIATE (SAS, 2009). The Shapiro-Wilk’s test yielded values close to 1 and the normality plots validated the assumptions of normality for all variables. Levene’s test and Welch’s ANOVA were computed to evaluate equal variances among herbicide-grazing treatments and primary use treatment groups using the HOVTEST and WELCH options in the MEANS statement of PROC GLM (SAS, 2009). When the P-value was <0.05, Welch’s ANOVA was used instead. Means were separated using the PDIFF option in PROC GLM (SAS, 2009) when the P-value for the main effect was P <0.05. Preplanned contrast statements were used to separate the effects of herbicide and grazing intensity among the frequent and infrequent herbicide treated pastures and the heavily and moderately grazed pastures (excluded the infrequent herbicide-lightly grazed and infrequent-no grazing pastures).

Research results and discussion:

The number of grass and grasslike species, grass and grasslike mean C, and grass and grasslike FQI were not statistically different among herbicide-grazing intensities, although heavily grazed pastures averaged 6.5 fewer species than moderately grazed pastures (Table 1). The number of forb species, forb mean C, and forb FQI were significantly different among herbicide-grazing intensities (Table 2). Number of forb species was the highest for pastures that were infrequently sprayed and grazed lightly or moderately. Pastures that were frequently sprayed with herbicide and heavily grazed had the fewest forb species. Forb was not different among pastures where herbicides were infrequently used. Pastures that were frequently sprayed with herbicide and heavily grazed had a lower forb than those that were moderately grazed. Forb FQI was 1.5 times greater on pastures that were infrequently sprayed with herbicide compared to those that were frequently sprayed with herbicide. Prairies where herbicide was infrequently used but grazed lightly had 1.25 times greater forb FQI than those that were not grazed. Moderately grazed pastures had 1.26 times greater forb FQI than those that were heavily grazed.

The overall number of species, overall mean C, and overall FQI were significantly different among herbicide-grazing intensities and had rankings similar to the forbs for the different herbicide-grazing intensity combinations (Table 3). The number of exotic grasses and forbs was not statistically different among herbicide-grazing intensities except that frequent use of herbicides averaged 6.8 fewer exotic forbs than pastures that were infrequently sprayed with herbicide (Table 4).

Preserves/wildlife habitat areas were similar to areas managed for livestock grazing in the number of grass and grasslike species, grass and grasslike mean C, and grass and grasslike FQI (Table 5). Pastures managed primarily for livestock grazing averaged 57.7 fewer forb species, had a 14% lower forb mean C, and a 41% lower forb FQI (Table 5). Overall number of species, overall mean C, and overall FQI were 1.8, 1.1, and 1.5 times greater, respectively, for pastures managed as preserves/wildlife habitat compared to those managed for livestock grazing (Table 5). Pastures managed for livestock grazing averaged 2.1 more exotic grasses and 10.7 fewer forbs than areas managed as preserves/wildlife habitat areas (Table 5).

There are several reasons why herbicide use did not affect grass diversity as much as forb diversity (Tables 1 and 2). The most frequently used herbicides on pasturelands in the Prairie Coteau Ecoregion of eastern South Dakota and southwestern Minnesota are those that control broadleaf plants, integrated pest management coordinator, personal communication). State noxious weed laws require the control of invasive species such as leafy spurge (Euphorbia esula L.) and Canada thistle (Cirsium arvense (L.) Scop.), and these species have high infestation rates in counties of this region. In addition, the forb mean C was higher on lands that were infrequently sprayed with herbicides, suggesting that rare and more sensitive forbs are not able to survive frequent herbicide treatment.

It is well documented that increased grazing intensity decreases floral diversity on native rangelands, and our study found likewise. However, grass and grasslike FQI was not significantly different among grazing intensities (Table 1). The reason for this was that the grass and grasslike mean C was not different between grazing intensities even though heavy grazing averaged 6.5 fewer grass and grasslike species compared to moderately grazed pastures. Numerous stocking rate studies show that taller and mid-size grasses are replaced by shorter species. The low standard error of the grass and grasslike mean C indicates that the most commonly found species have similar C values. In addition, the computation of FQI is not density dependent, meaning an area could have a substantial reduction in species diversity without reducing its FQI. Grazing intensity had a significant impact on forb FQI because numerous forb species are sensitive to overgrazing.

The desire for ecosystem goods and services (i.e., an economic benefit such as beef production or a service providing habitat for flora and fauna) dictate management decisions regarding herbicide use and grazing intensity. Our data set represents the common pastureland uses in this region and our findings of higher FQI values for forbs and for overall species on areas managed as preserves/wildlife habitat areas compared to pastures managed for livestock grazing (Table 5) was similar to pervious findings. Higgins et al. (2001) surveyed 63 tallgrass prairie remnant sites in eastern South Dakota and found that private land had an FQI of 39 compared to 57 for preserves. In northeastern Kansas, Jog et al. (2006) found warm-season hay meadows had an FQI of 32 and warm-season pastures managed for livestock grazing had an FQI of 21. Landowners whose primary objective is livestock grazing tend to use herbicides for weed control more frequently and stock heavier than managers of preserves. To conserve the floristic diversity of remaining native prairie tracts, the cultural and economic incentives behind these management decisions must be understood.

Livestock producers tend to accept exotic grasses whereas managers of preserves dislike exotic grasses because they reduce native species biodiversity. Our data showing 2.1 more exotic grass species on pastures managed for livestock grazing versus areas managed as preserves/wildlife areas (Table 5) support this view. Exotic forbs are a real concern for producers and managers. Pastures that were frequently sprayed had fewer exotic forb species and producers that managed land for livestock grazing also had fewer exotic forb species (Table 5). These data suggest that cultural factors such as “what the neighbors think” along with legal obligations to control noxious weeds are likely responsible for the fewer exotic forbs found on lands where livestock grazing is the primary objective.

Judicious use of herbicide application should be practiced instead of frequent broadcast applications. Fuhlendorf et al. (2009) showed that broadcast application of herbicides on Oklahoma rangeland where forbs comprised 23% of the herbage production did not increase grass and beef production. Their work suggests that a higher economic threshold of weedy broadleaf plants exists. Such thresholds are unknown in the Prairie Coteau Ecoregion, but historic climax plant community theory for this region would indicate that forbs make up approximately 5 to 15% of the plant community in terms of biomass (NRCS 2010). The perception that broadleaf plants reduce grass production in this region may be incorrect.

Economic incentives from conservation agencies need to focus on reducing stocking rate to a light grazing intensity in order to maintain or increase plant diversity. Such incentives exist through the Natural Resources Conservation Service’s Conservation Stewardship Plan. Practices such as deferred grazing and rotational grazing with proper stocking rates, pays producers who qualify (NRCS 2010). These payments are necessary to offset economic drivers to graze heavy.

On the other hand, managers of preserves/wildlife production areas should be aware of the benefits of using livestock to apply periodic disturbances. Pastures that were infrequently sprayed with herbicide and lightly grazed had the highest species richness, mean C, and FQI (Tables 1-3). As long as grazing is kept at a light intensity (<25% utilization of annual herbage production) and applied at the right time, it can be a useful tool to reduce competition of exotic cool-season grasses.

Participation Summary

Educational & Outreach Activities

Participation Summary:

Education/outreach description:

Nelson, M., and A. Smart. 2009. Floristic quality of native northern tallgrass prairie pastures in eastern South Dakota. In Abstracts: Merging Trails: Culture, Science, Innovation. Society for Range Management 62nd Annual Meeting. Feb 8-13, Albuquerque, NM.

Project Outcomes

Project outcomes:

Floristic quality assessments provided useful insight into the effects of past management practices, such as grazing and herbicide application. Floristic quality index values for grasses and grasslike plants were less affected by herbicide use and grazing intensity than those for forbs. Lightly grazed pastures had higher FQI values and greater species richness than pastures that were grazed moderately, heavily, or ungrazed. Frequent use of herbicides reduced FQI values for forbs to a much greater extent than for grasses. Finally, pastures managed as preserves/wildlife production areas had higher FQI values than pastures managed for livestock grazing. If conservation is aimed at improving the floristic quality of remaining native prairies in this region, economic incentives should be promoted to reduce stocking rates and encourage judicious use of herbicides.

Economic Analysis

Not quantified

Farmer Adoption

Not quantified


Areas needing additional study

A follow up economic analysis of herbicide use on pastures needs to be conducted in this region.

Any opinions, findings, conclusions, or recommendations expressed in this publication are those of the author(s) and do not necessarily reflect the view of the U.S. Department of Agriculture or SARE.