USGS - science for a changing world

Northern Prairie Wildlife Research Center

  Home About NPWRC Our Science Staff Employment Contacts Common Questions About the Site

Plant Community Patterns on Upland Prairie
in the Eastern Nebraska Sandhills

Results and Discussion

Species Richness

A total of 89 vascular plant taxa was found on the upland sites of the Barta Brothers Ranch which included 22 grass species, 59 forbs, 4 shrubs, 2 cacti, a general sedge category, and Baltic rush (Juncus balticus Willd.). Sedges and western ragweed (Ambrosia psilostachya DC.) were the most common taxa occurring over all topographic positions (Table 1). Mean species richness of the south-facing slopes (17.9), dune tops (18.6), and north-facing slopes (19.1) did not differ statistically (P>0.05). Mean species richness was lower for the interdunal position (11.1), but this may have been a consequence of fewer quadrats sampled per transect in the interdunal valleys than on the slopes or dune tops. However, visual observations made on the study site during June and July support these results indicating lower species richness in the interdunal valleys. A similar decline in species richness from ridge tops and slopes to lowland areas was also reported in the mixed prairie of northcentral South Dakota (Barnes et al. 1983).

Table 1.  Frequency of occurrence by topographical position of plant species with frequencies greater than 2.0 at the Barta Brothers Ranch, Nebraska.  Each mean is based on 87 transects.

Topographic position

Dune top
North-facing slope
South-facing slope
Blue grama
    Bouteloua gracilis
3.9a 1.5b 1.9ab 1.4b
Bluegrasses 1
    Poa spp.
72.5c 3.6a 13.3b 3.0a
Cacti 2
    Opuntia spp.
1.2a 7.8c 5.3b 8.1c
Hairy grama
    Bouteloua hirsuta
3.6a 17.7c 9.9b 16.1c
Indian grass
    Sorghastrum nutans
5.1a 5.4a 6.2a 11.2b
    Koeleria pyramidata
4.6a 19.8c 15.2c 9.7b
Lead plant
    Amorpha canescens
7.6a 18.1b 21.9c 16.0b
Little bluestem
    Schizachyrium scoparium
5.6a 34.3b 40.1c 34.2b
Prairie sandreed
    Calamovilfa longifolia
9.6a 20.9b 13.0a 23.0b
Prairie wild rose
    Rosa arkansana
8.2a 8.1a 9.1a 8.5a
Purple lovegrass
    Eragrostis spectabilis
8.1b 2.0a 3.2a 1.7a
Sand bluestem
    Andropogon hallii
3.6a 15.5c 10.9b 15.4c
Sand dropseed
    Sporobolus cryptandrus
19.0a 18.2a 15.4a 19.6a
Scribner dichanthelium
    Dichanthelium oligosanthes
59.4ab 55.0b 53.1b 64.2a
Sedges 3
    Carex spp.
97.5c 86.3b 81.4a 90.2b
Stiff sunflower
    Helianthus rigidus
0.1a 9.3c 6.9bc 5.6b
Needlegrasses 4
    Stipa spp.
15.7a 49.2c 43.9bc 39.8b
    Panicum virgatum
65.9c 25.3a 29.1ab 34.8b
Western ragweed
    Ambrosia psilostachya
70.6a 76.4a 72.0a 80.3a
White sage
    Artemisia ludoviciana
10.9a 1.5b 3.0b 2.0b
Wilcox panicum
    Dichanthelium wilcoxianum
6.3a 21.6c 28.6d 16.2b
abc Different letters in a row indicate means are different (P < 0.05).
1 Includes Canada bluegrass (Poa compressa) and Kentucky bluegrass (P. pratensis).
2 Includes brittle cactus (Opuntia fragilis) and prairie prickly pear cactus (O. macrorhiza).
3 Includes primarily sun sedge (Carex heliophila) and Schweinitz flat sedge (Cyperus schweinitzii).
4 Includes needle-and-thread (Stipa comata) and porcupine-grass (S. spartea).

Topographic Positions

The STEPDISC procedure indicated that the 18 taxa that played a significant role in discriminating differences in the topographic positions (Figs. 1 and 2) also had the highest frequencies (Table 1). When using the 18 species in the canonical discriminant analysis, the eigenvalues for the first two canonical variates were 4.399 and 0.376 with the first factor accounting for 89.6% of the variation in frequency or counts of species and the second factor accounting for 7.7% of the variation. Most interdunal transects were strongly separated from transects on the other topographic positions by the first variate (Fig. 1), which indicated that vegetation cover on interdunal positions was different from that on other positions. The five taxa that had highly positive coefficients for the first variate were bluegrasses (Poa L. spp.), switchgrass, white sage (Artemisia ludoviciana Nutt.), purple lovegrass [Eragrostis spectabilis (Pursh) Steud.], and sedges (Fig. 2). The two bluegrass species on the study site, Canada bluegrass (Poa compressa L.) and Kentucky bluegrass (Poa pratensis L.), were not easily identifiable as separate species in the vegetative stage of growth and were grouped into a single taxon. Principal species in the sedge grouping were sun sedge (Carex heliophila Mack.) and Schweinitz flat sedge (Cyperus schweinitzii Torr.). Even though sedges had the highest frequency of occurrence on interdunes, they also were common on the other topographic positions (Table 1). Switchgrass and bluegrasses were the two taxa most highly correlated or associated with the interdunal position. Each of these five taxa had significantly higher frequencies on interdunal valleys than on the other three topographic positions (Table 1).

Figure 1

Figure 1.  Canonical discriminant analysis ordination of all transects with 18 species selected from PROC STEPDISC, λ1 = 4.399 and λ2 = 0.376.

Figure 2

Figure 2.  Canonical discriminant analysis ordination of the 18 species based on their correlation structure. Plant species: SBL = sand bluestem, HGR = hairy grama, PSA = prairie sandreed, WRA = western ragweed, IND = Indian grass, SDI = Scribner dichanthelium, SWI = switchgrass, WSA = white sage, BLU = Canada and Kentucky bluegrasses, PLO = purple lovegrass, WPA = Wilcox panicum, LEA = lead plant, JUN = Junegrass, LBL = little bluestem, NEE = needle-and-thread and porcupine-grass, and SSU = stiff sunflower.

Little bluestem had the most negative coefficient for the first variate but lead plant (Amorpha canescens Pursh), needlegrasses (Stipa L. spp.), and the other warm-season grasses, i.e., sand bluestem, hairy grama, and prairie sandreed, also had highly negative coefficients. The needlegrasses included both needle-and-thread and porcupine-grass because of the difficulty of separating these two species when in the vegetative stage of growth. These taxa with negative coefficients for the first variate were in the quadrants associated with slopes and dune tops (Figs. 1 and 2) and had significantly higher frequencies on slopes and dune tops than on interdunal valleys (Table 1).

The first variate is probably related to moisture availability in the upper soil profile. Bluegrasses were the principal taxa (partial R²=0.403) affecting the separation of transects by topographical position. Bluegrasses are shallow rooted and have relatively low water-use efficiency. Shallow-rooted grasses commonly are reported to dominate swales and dry interdunal valleys in the Nebraska Sandhills (Barnes and Harrison 1982, Barnes et al. 1984) as well as in mixed prairie sites in northcentral South Dakota (Barnes et al. 1983). Barnes et al. (1984) concluded that dry valleys and swales in the Sandhills of Arthur County, Nebraska were good habitat for such shallow-rooted species as western wheatgrass, needle-and-thread, and blue grama. Barnes et al. (1984) hypothesized that the finer-textured soils (loamy fine sands) of these lowland areas provided readily available soil moisture early in the growing season within the surface 30 to 40 cm, whereas moisture availability of the fine sands of the slopes and ridges was restricted during the growing season to lower horizons out of reach of the shallow-rooted grasses.

Texture analyses of soils from the four topographic positions at the Barta Brothers Ranch did not differ in particle-size fractions as they were all in the loamy sand textural class with 84 to 89% sand. Soil samples in the interdunes were taken from locations dominated by bluegrasses and sedges. Soil texture would not appear to explain the difference in plant species composition among interdunal and dune top and slope positions. Organic matter content and total carbon content, however, were higher in the soils of the interdunal valleys than of the dune tops and slopes (Table 2). The higher organic matter content could positively influence available water in the upper soil profile in the interdunal valleys. Phosphorous and potassium contents of the interdunal soils also were higher than that of the soils on the slopes and dune tops.

Table 2.  Mean (± S.E.) soil pH, total organic matter (OM), total potassium (K), and total phosphorus (P), in the 0-15-cm and 15-45-cm depth intervals.
Topographic position 

0 - 15 cm

15 - 45 cm

pH OM (%) K (ppm) P (ppm) pH OM (%) K (ppm) P (ppm)
Interdune 6.4
South-facing slope 6.4
North-facing slope 6.4
Dune top 6.3

Livestock grazing also is a likely factor influencing plant cover on the topographic positions. Cattle in a continuous-grazed pasture commonly concentrate their grazing in lowland areas, especially near livestock water sources, and minimize grazing time on slopes (Stubbendieck and Reece 1992). Relatively high grazing pressure on lowland vegetation can be detrimental to tall and mid-grasses and favor grazing-resistant short grasses, such as the bluegrasses, blue grama, and purple lovegrass. The study site used by Barnes et al. (1984) had been grazed by cattle until 3 years before the initiation of their study. The dominance of their valley site by shortgrasses may have been related to a history of cattle grazing.

Switchgrass was the only deep-rooted, warm-season grass associated with the dry, interdunal valleys in our study (Table 1 and Fig. 2). Switchgrass has not been reported previously as a common component of plant communities on dry, interdunal valleys (Burzlaff 1962, Barnes et al. 1984, Bragg 1978). In support of the results of our study (Table 1), we have observed that switchgrass commonly develops dense stands in dry valleys and swales on upland sites throughout the Nebraska Sandhills. Switchgrass is a rapidly maturing warm-season grass that becomes stemmy and lower in palatability by midsummer when cattle will make only light use of it.

Blue grama often is reported to be a dominant species in interdunal valleys (Burzlaff 1962, Barnes and Harrison 1982, Barnes et al. 1984). Frequency of occurrence of blue grama on the Barta Brothers Ranch was low, but frequency of blue grama was higher on the interdunal valleys than on the slopes or dune tops (Table 1). Scribner dicanthelium [Dicanthelium oligosanthes (Schult.) Gould var. scribnerianum (Nash)] was correlated with the first variate, but was a commonly occurring species on all topographic positions (Table 1).

Taxa associated with interdunal valleys were either positively correlated or weakly negatively correlated with the second variate. The second canonical variate, however, indicated a separation of transects on south-facing slopes from transects on north-facing slopes (Fig. 1). Cacti (Opuntia P. Mill spp.), western ragweed, and warm-season grasses, such as prairie sandreed, sand bluestem, Indian grass [Sorghastrum nutans (L.) Nash], and hairy grama, had positive coefficients in relation to the second variate and were correlated with the quadrant in which about 75% of the transects on south-facing slopes were located (Figs. 1 and 2). Prairie sandreed, sand bluestem, hairy grama, cacti, and Indian grass had significantly higher frequencies on south-facing slopes than on north-facing slopes (Table 1). The cacti were comprised primarily of little brittle cactus [Opuntia fragilis (Nutt.) Haw.] but plains prickly pear (Opuntia macrorhiza Engelm.) was frequently encountered. Western ragweed was common on all topographic positions (Table 1). Western ragweed is an opportunistic plant that increases rapidly on upland sites with reduced vigor of the dominant grass species and/or during years of above-average precipitation (Reece, unpubl. data). The high frequency of occurrence of western ragweed on the study site may be related to the recent wet years.

Little bluestem, lead plant, and cool-season grasses, such as Junegrass, Wilcox panicum [Dicanthelium wilcoxianum (Vasey) Freckmann], and needlegrasses, had negative coefficients related to the second variate and were associated with the quadrant in which over 88% of the transects on north-facing slopes were concentrated (Figs. 1 and 2). Frequency of occurrence of little bluestem, lead plant, Junegrass, and Wilcox panicum, as well as the bluegrasses, was higher on north-facing slopes than on south-facing slopes (Table 1).

Aspect plays an important role in affecting habitat for plants. In the northern hemisphere, south-facing slopes have higher light intensity, higher soil-surface temperatures, and more limited moisture availability in surface horizons because of higher evapotranspiration rates (Humphrey 1962). The C4 pathway, lower transpiration rates, and deeper root systems of warm-season grasses, e.g., prairie sandreed and sand bluestem, and forbs, e.g., western ragweed, would appear to make them better adapted than cool-season species to conditions on south-facing slopes. Tolstead (1942) observed that prairie sandreed was most prevalent on south-facing slopes and broad, dry valleys in the Sandhills of Cherry County. Bragg (1978) also reported that prairie sandreed was the most prevalent species on south-facing slopes in Garden County of the southwestern Sandhills.

The C3 pathway of cool-season species, e.g., needlegrasses and Junegrass, is advantageous in environments of lower light intensities, lower temperatures, and high water availability (Doliner and Jolliffe 1979). Needlegrasses and Junegrass are relatively shallow rooted (Weaver 1920, 1954), have lower water use efficiencies, and require readily available moisture at the soil surface during their growing season to be competitive (Barnes and Harrison 1982). Therefore, these two taxa along with other cool-season plants, e.g., Wilcox panicum, appear to be best adapted to north-facing slopes. Bragg (1978) reached a similar conclusion based on the prevalence of needle-and-thread on north-facing slopes of choppy sands range sites in the southwestern Nebraska Sandhills. The strong association of little bluestem with north-facing slopes cannot be readily explained. On sand range sites in the northcentral Nebraska Sandhills, Bragg (1998) also found little bluestem to be more prevalent on north-facing slopes than south-facing slopes, but offered no explanation. We would expect little bluestem to be associated with other deep-rooted, warm-season grasses on south-facing slopes and dune tops. However, in another study in the west central Nebraska Sandhills (Cullan et al., unpubl. data), frequency of occurrence of little bluestem was found to be negatively correlated with prairie sandreed on upland sites.

Even though transects on dune tops consistently had negative coefficients associated with the first variate, 54% of dune top transects were negatively and 46% positively associated with the second variate. Prevalent species on dune tops were a mixture of species with negative and positive coefficients in respect to the second variate. Some species with negative coefficients in respect to the second variate and associated with north-facing slopes had high frequencies on dune tops (Table 1). Frequency of occurrence of needlegrasses and Junegrass on dune tops was similar to that on north-facing slopes but higher than that on south-facing slopes. Some species with positive coefficients in respect to the second variate and associated with south-facing slopes had high frequencies on dune tops (Table 1). Frequency of occurrence of sand bluestem and hairy grama on the dune tops was similar to that on south-facing slopes but higher than that on north-facing slopes. Dune tops were not characterized by a unique species or group of species but appeared to be a transitional area where species common to the opposing slopes mixed.

Three other frequently-occurring species were sand dropseed, prairie wild rose (Rosa arkansana Porter), and stiff sunflower [Helianthus rigidus (Cass.) Desf.] (Table 1). Frequency of occurrence for sand dropseed and prairie wild rose did not differ by topographic position. Sand dropseed is a shallow-rooted, short-lived perennial that is a conspicuous part of upland prairie sites (Pool 1914, Tolstead 1942, Barnes et al. 1984) and is particularly prevalent on areas recovering from improper grazing or drought (Frolik and Shepherd 1940, Burzlaff 1962). As reported by Burzlaff (1962), prairie wild rose generally occurs as uniformly scattered, individual plants throughout the uplands that account for relatively small amounts of total plant cover and biomass production. Stiff sunflower was found almost exclusively on slopes and dune tops. Stiff sunflower is a rhizomatous forb and is a common component of upland vegetation in good and excellent range condition (Frolik and Shepherd 1940).

In summary, topographic position played an important role in plant distribution on grazed, upland prairie in the eastern Nebraska Sandhills. Edaphic factors and grazing have been cited (Barnes et al. 1984, Stubbendieck and Reece 1992) as critical factors affecting plant community distribution across topographic positions. Soil textural classes at our site did not vary greatly among topographic positions and appeared not to be a contributing factor in affecting plant community patterns; however, soil organic matter content was higher in the interdunal valleys. Because cattle tend to concentrate in low-lying areas, long-term, continuous grazing may have played a significant role in the dominance of the grazing resistant, short grasses in the interdunal valleys. The potential influence of grazing on plant community distribution on slopes and dune tops is not obvious. Seasonal patterns of grazing livestock by topographic position has not been studied in the Nebraska Sandhills. Finally, warm-season, tall grasses were dominant on the dune tops and both the south- and north-facing slopes even though the cool-season grasses and sedges were strongly associated with north-facing slopes. The warm-season, tall grasses represented over 55% of the annual aboveground biomass production on the slopes and dune tops (Schacht et al., unpubl. data). Cool-season grasses and sedges, however, were dominant on the interdunal valleys.

Previous Section -- Materials and Methods
Return to Contents
Next Section -- Acknowledgments

Accessibility FOIA Privacy Policies and Notices

Take Pride in America logo logo U.S. Department of the Interior | U.S. Geological Survey
Page Contact Information: Webmaster
Page Last Modified: Saturday, 02-Feb-2013 06:33:41 EST
Reston, VA [vaww55]