Printer Friendly

Evidence of species and functional group attrition in shrub-encroached prairie: implications for restoration.

INTRODUCTION

Increasing woody encroachment in savanna and grassland communities has been widely reported (Bragg and Hurlbert, 1976; Archer et al., 1995; Hoch and Briggs, 1999; Brown and Archer, 1999; Price and Morgan, 2008). Postulated causes include fire suppression (Gibson and Hulbert, 1987), climate change (Archer et al., 2001), disturbance (Schlesinger et al., 1990), and grazing effects (Van Auken, 2000; Briggs et al., 2005). Consequences of woody encroachment include reduced richness in the herbaceous layer (Lett and Knapp, 2005; Price and Morgan, 2008), reduced annual net primary productivity of dominant C4 species (Heisler et al., 2004), and reduction in biomass and density of herbaceous vegetation (Brown and Archer, 1999).

Changes associated with woody encroachment in grassland habitats have been characterized as a process of community disassembly (Zavaleta et al., 2009). The species attrition hypothesis, originally applied to seed banks (Bond, 1984) and extended to predicting the response of ground layer species richness and traits to woody encroachment in oak woodlands (Taft, 2009), provides a framework for examining changes in shrub and sapling-encroached native grasslands. The species attrition hypothesis as applied here predicts woody encroachment in tallgrass prairie can be expected to result in declining richness and diversity of ground layer species due to the absence of a compensating replacement flora of shade-tolerant species. Many species traits can increase vulnerability to population declines such as growth form (Turner et al., 1996), dispersal type and geographic range (Williams et al., 2005), habitat specialization (Walker and Preston, 2006), life history and phenotype (Leach and Givnish, 1996; Duncan and Young, 2000), and taxonomic group size (Schwartz and Simberloff, 2001). Plant functional groups (FGs), species assemblages with shared traits, have been used effectively to examine resource use in ecosystems, response to distvirbance, and simulated climate change (Symstad, 2002; McLaren and Turkington, 2010; Sivicek and Taft, 2011; Cantarel et al., 2013). We used FGs in this study to link traits such as growth form, life history, and ecophysiology to different levels of woody encroachment.

While previous studies have noted declines in herbaceous diversity with increasing shrub canopy cover (Brown and Archer, 1999; Lett and Knapp, 2005; Price and Morgan, 2008), we expanded previous work by exploring overstory interactions with cover and species richness of FGs. Identifying levels of woody encroachment and the sequence of decline in FG cover and richness for groups, such as perennial forbs and C.t grasses (Heisler et al., 2004), could illuminate the potential for restoration of prairie remnants. Questions examined in this study were: (1) how do patterns of plant species composition, richness, and diversity and FG cover and richness respond to increasing levels of woody encroachment, (2) are there levels of encroachment relevant to restoration opportunities, and (3) is there an order to the pattern of FGs associated with increasing levels of encroachment that can serve as an ecological indicator?

METHODS

STUDY SITE

The study site is a 65 ha (410 m by 1582 m) mosaic of grassland, shrubland, and wetland habitats in Lake County in northeastern Illinois (42.30104N, 87.88278W). Grassland habitats include floristically rich prairie and old field containing numerous prairie species. The site was heavily infested with shrubs and saplings. However, the distribution of woody cover was variable with some areas having high cover and other areas with low cover but high stem density where woody species have not formed a canopy over the ground layer. This study is based on baseline data from permanent vegetation sample plots to be used for tracking restoration progress following shrub and sapling removal. Botanical surveys at this site have identified 429 species of vascular plants (80% native), including four that are listed by the Illinois Endangered Species Protection Board as threatened or endangered (Taft et al., 2013).

VEGETATION SAMPLE DESIGN

A stratified vegetation sampling design was utilized with 10 parallel transects running west to east, each separated by 152 m intervals. Five sample points were established on each transect separated by 76 m. However, a fire station in the northwest corner of the property eliminated one plot on the first transect and 12 plots fell within previously delineated wetland habitats and were omitted from this study. This array provided 37 terrestrial vegetation plots. In addition eight plots were established at the study site in areas of high-quality reference prairie (missed in the stratified sample design). The areas of high-quality prairie were identified in previous botanical surveys (Taft, 2006) and were distinguished based on habitat quality criteria utilized for the Illinois Natural Areas Inventory (White, 1978) and also application of Floristic Quality Assessment (Taft et al., 1997), a procedure shown to be robust for making assessments of prairie habitat quality (Taft et al., 2006). The addition of these reference plots provided a total of 45 independent vegetation sample plots. Specific plot locations for the additional targeted sampling were determined randomly. Vegetation primarily was sampled in 25 [m.sup.2] (5 m X 5 m) plots for woody overstory with ground layer sampled in nested quadrats (1 [m.sup.2]). Composition and stem density of all woody stems >1 m tall and <10 cm dbh were sampled within the 25 [m.sup.2] plots. Trees (woody stems [greater than or equal to] 10 cm dbh) were sampled in 0.02 ha plots but were scarce in the study area (only a few occurred in the shrub-sapling plots). Shrub-sapling canopy cover was determined with digital photography using a hemispherical lens oriented vertically in the plot center to photograph the canopy of the plot area (narrowed to the area of the study plot with a lens tube). Interference from herbaceous cover was minimized by placing the camera on a 70 cm tall tripod. Percent visible sky and leaf area index (LAI) were calculated from these images using HemiView Canopy Analysis Software, ver. 2.1. Percent canopy cover was calculated as 100--% visible sky. Images were classified by adjusting threshold light intensity levels from high-contrast displays compared to continuous tone displays to estimate percent visible sky as accurately as possible.

Ground layer vegetation was sampled with three 1-[m.sup.2] quadrats within each shrub-sapling plot with quadrat placement in the center and the southwest and northeast corners of the shrub-sapling plots. Most vegetation parameters (defined below) for ground layer plots were averages for the three quadrats. Data collected from each quadrat included species presence and percent cover for each species estimated with a modified Daubenmire cover-class scale (0-1%, 1-5%, 5-25%, 25-50%, 50-75%, 75-95%, 95-100%). All plants rooted within each quadrat frame were recorded to species including woody species <1 m tall. Vegetation was sampled from mid-Jun. to mid-Jul. 2009.

FUNCTIONAL GROUPS

Native species were assigned to the following plant FGs based on growth form, life history, and ecophysiology: fern (and fern allies), annual/biennial forb (AB forb), perennial dicot forb (PD forb), perennial monocot forb (PM forb), legume, hemi-parasite, sedge, [C.sub.4] grass, [C.sub.3] grass, and woody (including shrub, tree, and vine); nonnative species were combined as herbaceous (X herb) or woody (X woody). Similar groups were followed by Kindsher and Wells (1995) and Sivicek and Taft (2011).

DATA PREPARATION AND ANALYSIS

Data analysis was conducted using PC-ORD Version 4.34 (McCune and Mefford, 1999) and IBM SPSS Statistics ver. 21.0. Dissimilarity among plots was determined with the Sorensen Index and one prairie reference plot, based on average distance measure among plots, was identified as an outlier and eliminated from further analysis yielding a final total of 44 sample plots. Cluster analysis was used to produce a classification of sites from species sample data to compare differences in overstory structure, explore variation in species composition and richness across the gradient of woody encroachment, and to highlight extraordinary levels of species richness in tallgrass prairie habitat. Flexible sorting with (1 set at 0.25 was used for its optimal grouping characteristics to construct a hierarchical dendrogram based on Sorensen distance measures. Values from indicator species analysis were used to identify taxa modal to vegetation types with significance (P < 0.05) determined from Monte Carlo randomization tests (4999 iterations).

Percent canopy cover and LAI of the shrub-sapling stratum explained far more variance in ground layer species parameters (e.g, species density, richness, diversity, and % cover) than woody stem density. Percent canopy cover and LAI were highly correlated (r = 0.93); however, canopy cover had a nonnormal distribution uncorrected with data transformations and overstory LAI explained slightly more variance in the data. Consequently, LAI was the preferred woody encroachment variable. To highlight the sequential order of FGs with increasing woody encroachment (Question 3), four LAI classes were formed by equal range division:

Class 1 = LAI 0-0.75 (n = 9) -- percent canopy cover range: 10.1-52.4 (avg. 32.3%), Class 2 = LAI >0.75-1.5 (n = 9) -- percent canopy cover range: 62.6-79.0 (avg. 71.4%), Class 3 = LAI >1.5-2.25 (n = 18) -- percent canopy cover range: 78.8-90.4 (avg. 86.2%), Class 4 = LAI >2.25-3.0 (n = 8) -- percent canopy cover range: 92-93.4 (avg. 92.3%).

FGs were then arranged in descending rank order of percent cover and richness of species according to proportions in LAI Class 1.

Species and FG metrics were calculated as follows:

Species Density.-mean native species richness per quadrat, averaged for each plot;

Species Richness.-sum of total native species from all three quadrats in a sample plot;

Shannon-Wiener Index of Diversity (H'n).--[summation] [[p.sub.i] ln([p.sub.i])], where [p.sub.i] is the relative abundance of each native species [based on importance values (IV200) calculated as the sum of relative cover and relative frequency];

% Ground Cover.--sum of cover values for each native species;

% Bare Ground.--mean estimate of bare ground from quadrats in each plot;

Functional Group Density.--mean number of functional groups per quadrat;

Cover for each Functional Group.-sum cover for each species in each functional group (sum of species cover values for all species recorded in plots (combining data from three quadrats)

Species Rchness for each Functional Group.--count of species in each functional group for each plot (sum of species from all three quadrats).

With the exception of PD forb, FG variables based on cover and richness lacked central normal tendency even after transformation.

Question 1.--The interactions between woody overstory (LAI), the independent variable, and species-level dependent variables (native species density, richness, H'n, percent cover, percent bare ground, and FG density) were examined with regression analysis. To determine whether the interactions were linear or quadratic, we modeled the linear effect of LAI following by a quadratic model. If a quadratic function explained a significant amount of additional variance, as determined from hierarchical regression analysis, the quadratic function was used.

FG cover and richness response to increasing levels of woody encroachment were examined with indirect gradient analysis using nonmetric multidimensional scaling (NMS) (Mather, 1976). NMS was used to examine plot variance based on species composition; we then searched for correlations between the ordination and the cover and richness of FGs after rotating the biplot to maximize the fit of the independent variable LAI with the first ordination axis (r = 0.79). NMS was used for its independence from species response models (the data lacked central normal tendency) and optimal graphical representation of community relationships (McCune and Grace, 2002). Vegetation classes identified in duster analysis were used to code plots but do not constrain the ordination. Using the Sorensen distance measure and a random starting configuration, results from 250 random starts of NMS were examined in one to six dimensions. A Monte Carlo test was applied to the randomized runs to determine whether the ordination axes reduced more stress than expected by chance. The strength of the correlations between axes and FG variables was determined with Pearson's correlation coefficient r.

Question 2.--To search for levels of woody encroachment relevant to restoration opportunities we examined trend lines with a significant linear or quadratic function.

Question 3.--Proportions of FGs within LAI classes were examined graphically to characterize the ordered sequence with increasing levels of woody encroachment. Discriminant analysis with automatic forward stepwise analysis was used to identify FGs with significant association to LAI classes. FG cover and richness were used as dependent variables in separate analyses. LAI classes were used as the grouping variable. F-to-remove statistics determined the relative importance of FGs separating LAI classes. The forward stepwise analysis used variables with significance set at 0.05 to enter the model and 0.1 to remove.

RESULTS

VEGETATION TYPES, SPECIES RICHNESS, AND STRUCTURE

Three basic vegetation types were identified at the root of the cluster dendrogram based on species composition: prairie (n = 5), prairie-old field (n = 10), and old field (n = 29). Twenty-nine species (of 106 total) representing 10 FGs were identified as significant indicators in the prairie vegetation type (Table 1). Fifteen species (of 132 total) representing 7 FGs were identified as significant indicators in the prairie-old field vegetation type. Only a single species (out of 174) was identified as a significant indicator in the old field vegetation type.

A total of 212 vascular plant species were recorded in 132 [m.sup.2] quadrats in 44 sample plots stratified throughout the study site. Mean native species density and native species richness were highly correlated (r = 0.97) and each was strongly correlated to mean FG density (r = 0.81). Mean native species density was highest in prairie plots (29.1/[m.sup.2] [+ or -] 0.73, mean [+ or -] SE) with up to 39 native species richness in sample quadrats. Old field plots had under half this total (13.7/[m.sup.2] [+ or -] 1.31); prairie-old field plots were intermediate (21.9/[m.sup.2] [+ or -] 1.43). Mean shrub-sapling stem density per plot (25 [m.sup.2]) was 58.5 [+ or -] 5.1 [22,898/ha]), mean LAI was 1.61 ([+ or -] 0.12), and mean shrub-sapling cover was 73% ([+ or -] 3.6). Twenty-four shrub-sapling species were recorded in 44 plots. Rhamnus cathartica L., Cornus racemosa Lam., Lonicera X bella Zabel., and Viburnum lentago L., in descending rank order, were the dominant species comprising 93% of total stems. Compared to old field plots, the combined prairie and prairie-old field plots have lower LAI and percent shrub-sapling cover but higher stem density (64.1 versus 55.6 stems per plot) due to a predominance of lower, presumably younger, stems that had yet to form a canopy overtopping the ground layer.

GROUND LAYER AND WOODY COVER INTERACTIONS (QUESTIONS 1 AND 2)

Species-level responses.--Increasing shrub-sapling LAI is associated with significant declines in native species richness and native and nonnative species density, and a significant increase in percent bare ground, and the responses are linear (Figs. 1A-D). Hierarchical regression analysis for the relationship between LAI and native species diversity (H'n) indicated that a quadratic function explained 10% more variance compared to a linear response and the additional explained variance was significant (P = 0.01). Decline in H'n is detectable only after LAI exceeds about 1.5 (Fig. 1E), a level associated with mean percent canopy cover of about 75%. There is a precipitous decline in percent native ground cover with increasing LAI that levels off with the highest LAI (Fig. 1F). A quadratic function explained 7% more variance compared to a linear response and the difference was significant (P < 0.01).

FG-level responses.--Native FG density including woody and herbaceous groups and native FG density including only herbaceous groups were negatively associated with LAI (Figs. 2A, B). The relationship between native FGs and LAI, similar to H'n, also was nonlinear with FG decline detectable after about LAI 1.5 (Fig. 2). Hierarchical regression analysis indicated a quadratic function explained 9% more variance compared to a linear response for FGs when including herbaceous and woody taxa in the ground layer (Fig. 2A), and the additional explained variance was significant (P = 0.017). The quadratic function for herbaceous FGs explained 5.2% more variance (Fig. 2B), a marginally significant difference (P = 0.066).

LAI was positively correlated (r = 0.79) with NMS axis 1 following rotation and percent cover of PD forbs, [C.sub.4] grasses, [C.sub.3] grasses, legumes, nonnative herbs, and sedges each were negatively associated with NMS axis 1 (Fig. 3) and the correlations were significant (Table 2). PD forb cover (the only prominent FG that was normally distributed), similar to total native cover (Fig. IE), declined precipitously with increasing LAI but leveled off at the highest LAI. PD forb was the most species rich functional group with 47% of all native species. Species richness totals for PD forbs, [C.sub.4] grasses, [C.sub.3] grasses, PM forbs, hemi-parasites, perennial legumes, and nonnative forbs also were negatively associated with NMS axis 1 (Fig. 4) and the correlations were significant (Table 2).

ASSOCIATIONS OF FUNCTIONAL GROUPS COVER AND RICHNESS WITH INCREASING LAI (QUESTION 3)

Hemi-parasites, [C.sub.4] grasses, legumes, and PD forbs are the functional groups with the greatest proportion of their cover (>50%) in LAI Class 1 plots (Fig. 5A). The rank order of FGs based on proportions in LAI Class 1 are similar for percent cover and richness of species (Spearman's rho 0.87, P = 0.0003, df = 10). Results from forward stepwise discriminant analysis indicated that percent cover and species richness of [C.sub.4] grasses are significantly associated with LAI Class 1. Although PD forb cover is significantly associated with LAI Class 1, nearly 70% of PD forb species richness occurred in LAI classes 2-4 (Fig. 5B). FGs with richness of species significantly affiliated with LAI classes 2-4 are AB forb, Sedge, and Fern.

DISCUSSION

The average native species density recorded in prairie plots (29.1 /[m.sup.2] [+ or -] 0.73) is quite high compared to other reports for tallgrass prairie [e.g, 17.5/[m.sup.2] [+ or -] 0.5 (Dornbush, 2004), 10.1/ [m.sup.2] [+ or -] 4.5 (Milbauer and Leach, 2007), 12.0/0.5-[m.sup.2] (Polley et al., 2005), 15.4/50-[m.sup.2] [+ or -] 1.4 (Collins et al., 2002)]. Differences in species and FG cover and richness associated with increasing overstory shading in this study and reported in numerous other studies (e.g., Lett et al., 2004; Briggs el al., 2005), substantiate what may be the single greatest management concern in tallgrass prairie. Woody encroachment in tallgrass prairie generally is attributable to inadequate fire frequency (Bragg and Hulbert, 1976; Abrams, 1986; Lett and Knapp, 2003; Milbauer and Leach, 2007). For example fire suppression in grasslands can lead to a closed canopy forest in as little as 35 y (Hoch and Briggs, 1999). In contrast frequent burning in the eastern region of the tallgrass prairie is associated with maximum species richness and diversity levels for vascular plants (Bowles and Jones, 2013), in part due to fire controls on shrub and sapling canopy formation. The high density of woody stems in the prairie and prairie-old field plots suggests these areas also are at risk without management intervention. Similar to semi- arid grasslands (Brown and Archer, 1999), the primary factor determining the spread of woody encroachment in the grassland mosaic appears to be the pattern and rate of seed dispersal.

Functional groups were utilized in this study to understand their relationship to LAI/ canopy cover and whether there were patterns of decline related to increasing woody encroachment that suggest loss of ecosystem functions in the absence of fire. As predicted by the species attrition hypothesis, there is evidence of species and functional diversity decline with increasing woody encroachment. For most attributes this change commenced with early stages of woody encroachment while H'n and FG Density declined after intermediate encroachment. Loss of species within FGs can have lasting effects on community structure and function (Hooper and Vitousek, 1997). The loss of functionally unique species likely will affect ecosystem functioning due to the lack of compensating species in the community (Walker, 1995; Tilman et al., 1997). The pattern of FG decline is similar to differences found between degraded and relatively undisturbed prairies (Sivicek and Taft, 2011), where hemi-parasites and nitrogen-fixing species appear to be most affected. Nitrogen-fixing species also are among the greatest declining species in fragmented prairies of Wisconsin (Leach and Givnish, 1996). The sequence of declining FG richness and cover evokes a process of community disassembly resulting from stressor interactions with vulnerable species of functional groups leading to nonrandom declines and losses (Zavelleta et al., 2009).

Following early stages of woody encroachment at this site and more clearly after the canopy of woody species increases, significant declines in many individual FGs can be detected and there appears to be a shift in the ground layer towards the old field assemblage of generalist species. The FGs in this study that appear most sensitive to woody encroachment (C4 grasses, hemi-parasites, legumes, and PD forbs) highlight which FGs to include in targeted augmentation. C4 grasses and PD forbs have been observed to decline with woody encroachment in other studies (Kochy and Wilson, 2000; Briggs et al., 2002; Lett and Knapp, 2003; Mason et al, 2009). C4 grasses at this site were notably absent in the soil seed bank (Kron, 2011) and likely will require augmentation following removal of woody species. While it is recommended to prevent initial stages of woody expansion, in part because of the high labor costs involved in reversing the trend (Lett and Knapp, 2003), the resilience of H'n and FG density suggest efficacious opportunities for restoration remain at portions of this study site characterized by low-to-moderate levels of woody encroachment. Habitat rehabilitation is possible in heavily encroached prairie but species and functional groups likely will be missing.

Acknowledgments.--We extend our thanks to Illinois Tollway for research support. Associate Editor Roger Anderson and two anonymous reviewers provided insightful comments improving the clarity of the paper and their contributions are much appreciated. There is no conflict of interest in the publication of this material.

LITERATURE CITED

ABRAMS, M. D. 1986. Historical development of gallery forests in northeast Kansas. Vegetatio, 65:29-37.

ARCHER, S., D. SCHIMEL, AND E. H. HOLLAND. 1995. Mechanisms of shrubland expansion: land use, climate or C[O.sub.2]. Climatic Change, 29:91-99.

--, T. W. BOUTTON, AND K. A. HIBBARD. 2001. Trees in grasslands: biogeochemical consequences of woody plant expansion, p. 115-137. In: E. D. Schulze, S. P. Harrison, M. Heimann, E. A. Holland, J. Lloyd, I. C. Prentice and D. Schimel (eds.). Global Biogeochemical Cycles in the Climate System. Academic Press, Sand Diego, CA.

BOND, W. J. 1984. Fire survival of Cape Proteaceae-influence of fire season and seed predators. Vegetatio, 56:65-74.

BOWLES, M. L. AND M. D. JONES. 2013. Repeated burning of eastern tallgrass prairie increases richness and diversity, stabilizing late successional vegetation. Ecol. Appi, 23:464-478.

BRAGG, T. B. AND L. C. HULBERT. 1976. Woody plant invasion of unbumed Kansas bluestem prairie. J. Range Manage., 29:19-23.

BRIGGS, J. M., G. A. HOCH, AND L. C. JOHNSON. 2002. Assessing the rate, mechanisms, and consequences of the conversion of tallgrass prairie to Juniperus virginiana forest. Ecosys., 5:578-586.

--, A. K. KNAPP, J. M. BLAIR, J. L. HEISLER, G. A. HOCH, M. L. LETT, AND J. K. MCCARRON. 2005. An ecosystem in transition: causes and consequences of the conversion of mesic grassland to shrubland. Biosci., 55(3):243-254.

BROWN, J. R. AND S. ARCHER. 1999. Shrub invasion of grassland: recruitment is continuous and not regulated by herbaceous biomass or density. Ecology, 80:2385-2396.

CANTAREL, A. A. M., J. M. G. BLOOR, AND J. F. SOUSSANA. 2013. Four years of simulated climate change reduces above-ground productivity and alters functional diversity in a grassland ecosystem. J. Veg. Sri., 24:113-126.

COLLINS, S. L., S. M. GLENN, ANDJ. M. BRIGGS. 2002. Effect of local and regional processes on plant species richness in tallgrass prairie. Oikos, 99:571-579.

DORNBUSH, M. E. 2004. Plant community change following fifty-years of management at Kalsow Prairie Preserve, Iowa, U.S.A. Am. Midi. Nat., 151:241-250.

DUNCAN, R. P. AND R. J. YOUNG. 2000. Determinants of plant extinction and rarity 145 years after European settlement of Auckland, New Zealand. Ecology, 81(11):3048-3061.

GIBSON, D. J. AND L. C. HULBERT. 1987. Effects of fire, topography and year-to-year climatic variation on species composition in tallgrass prairie. Vegetatio, 72(3):175-185.

HEISLER, J. L., J. M. BRIGGS, A. K. KNAPP, J. M. BLAIR, AND A. SEERY. 2004. Direct and indirect effects of fire on shrub density and aboveground productivity in a mesic grassland. Ecology, 85(8):2245-2257.

HOCH, G. A. ANDJ. M. BRIGGS. 1999. Expansion of eastern red cedar (Juniperus virginiana) in the northern Flint Hills, Kansas, p. 9-15. In: J. T. Springer (ed.). Proceedings of the 16th North American Prairie Conference. University of Nebraska, Kearney.

HOOPER, D. U. AND P. M. VITOUSEK. 1997. The effects of plant composition and diversity on ecosystem processes. Science, 277:1302-1305.

KINDSCHER, K. AND P. V. WELLS. 1995. Prairie plant guilds: a multivariate analysis of prairie species based on ecological and morphological traits. Vegetatio, 117:29-50.

KOCHY, M. AND S. D. WILSON. 2000. Competitive effects of shrubs and grasses in prairie. Oikos, 91:385-395.

KRON, Z. P. 2011. Quantitative vegetation patterns, seed limitation, and inferred patterns of functional group attrition in shrub encroached tallgrass prairie. MS. Thesis, University of Illinois at Urbana-Champaign.

LEACH, M. K. AND J. T. GIVNISH. 1996. Ecological determinants of species loss in remnant prairies. Science, 273:1555-1558.

LETT, M. S. AND A. K. KNAPP. 2003. Consequences of shrub expansion in mesic grassland: resource alterations and graminoid responses. J. Veg. Sci., 14:487-496.

--AND --. 2005. Woody plant encroachment and removal in mesic grassland: production and composition responses of herbaceous vegetation. Am. Midi. Nat., 153(2):217-231.

--, --, J. M. BRIGGS, AND J. M. BLAIR. 2004. Influence of shrub encroachment on plant productivity and carbon and nitrogen pools in a mesic grassland. Can, J. Botany, 82:1363-1370.

MASON, T. J., K. FRENCH, AND M. W. LONSDALE. 2009. Do graminoid and woody invaders have different effects on native functional groups?J. Appl. Ecol., 46:426-433.

MATHER, P. M. 1976. Computational methods of multivariate analysis in physical geography. J. Wiley and Sons, London. 544 p.

MCCUNE, B. AND J. GRACE. 2002. Analysis of ecological communities. MjM Software Design, Gleneden Beach, OR. 300 p.

--AND M. J. MEFFORD. 1999. PC-ORD. Multivariate analysis of ecological data. Version 4.34. MjM Software, Gleneden Beach, OR.

MCLAREN, J. R. AND R. TURKINGTON. 2010. Ecosystem properties determined by plant functional group identity. J. Ecol., 98:159-469.

MILBAUER, M. L. AND M. K. LEACH. 2007. Influence of species pool, fire history, and woody canopy on plant species density and composition in tallgrass prairie. J. Toney Bot. Soc., 134:53-62.

POLLEY, H. W., J. D. DERNER, AND B. J. WILSEY. 2005. Patterns of plant species diversity in remnant and restored tallgrass prairies. Restor. Ecol., 13:480-487.

PRICE, J. N. AND J. W. MORGAN. 2008. Woody plant encroachment reduces species richness of herb-rich woodlands in southern Australia. Austral Ecol., 33:278-289.

SCHLESINGER, W. H. J. F. REYNOLDS, G. L. CUNNINGHAM, L. F. HUENNEKE, W. M. JARRELL, R. A. VIRGINIA, AND W. G. WHITFORD. 1990. Biological feedbacks in global desertification. Science, 247(4946):1043-1048.

SCHWARTZ, M. W. AND D. SIMBERLOFF. 2001. Taxon size predicts rates of rarity in vascular plants. Ecol. Lett., 4:464-469.

SIVICEK, V. A. AND J. B. TAFT. 2011. Functional group density as an index for assessing habitat quality in tallgrass prairie. Ecol. Indie. 11:1251-1258.

SYMSTAD, A. J. 2002. An overview of ecological plant classification systems: linking functinoal response and functional effect groups, p. 13-50. In: R. S. Ambasht and N. K. Ambasht (eds.). Modern Trends in Applied Terrestrial Ecology. Kluwer Academic Publishers, Dordrecht.

TAFT, J. B. 2006. Floristic survey and assessment of the Illinois Department of Transportation North Chicago Wedand Mitigation Site, Lake County, Illinois. Center for Wildlife and Plant Ecology. Illinois Natural History Survey Technical Report.

--. 2009. Effects of overstory stand density and fire on ground layer vegetation in oak woodland and savanna habitats, p. 21-39. In: T. F. Hutchinson (ed.). Proceedings of the 3rd Fire in Eastern Oak Forests Conference, 2008 May 20-22, Carbondale, IL. General Technical Report NRS-P-46. Newtown Square, PA: U.S. Department of Agriculture, Forest Service, Northern Research Station.

--, G. S. WILHELM, D. M. LADD, AND L. A. MASTERS. 1997. Floristic Quality Assessment for vegetation in Illinois, a method for assessing vegetation integrity. Erigenia, 15:3-95.

--, C. HAUSER, AND K. R. ROBERTSON. 2006. Estimating floristic integrity in tallgrass prairie. Biological Conservation, 131:42-51.

--, C. CARROLL-CUNNINGHAM, AND D. RUFFATTO. 2013. Results from 2012 vegetation monitoring in prairie and old field habitats following shrub removal at the North Chicago Wetland Mitigation Site, Lake County, Illinois. Illinois Natural History Survey Technical Report 2013 (16).

TILMAN, D., J. KNOPS, D. WEDIN, P. REICH, M. RITCHIE, AND E. SIEMANN. 1997. The influence of functional diversity and composition on ecosystem processes. Science, 277:1300-1302.

TURNER, I. M., S. K. CHUA, Y. S. J. ONG, B. C. SOONG, AND H. T. W. TAN. 1996. A century of plant species loss from an isolated fragment of lowland tropical rain forest. Consent. Biol., 10:1229-1244.

VAN AUKEN, O. W. 2000. Shrub invasions of North American semiarid grasslands. Annu. Rev. Ecol., Evol. S., 31:197-215.

WALKER, B. 1995. Conserving biological diversity through ecosystem resilience. Consent. Biol., 9:747-752.

WALKER, K. J. AND D. C. PRESTON. 2006. Ecological predictors of extinction risk in the flora of lowland England, UK. Biodivers. Consent., 15:1913-1942.

WHITE, J. 1978. Illinois natural areas inventory technical report, Vol. 1. Survey methods and results. Illinois Natural Areas Inventory, Urbana. 426 p.

WILLIAMS, N. S. G., W. J. MORGAN, J. M. MCDONNELL, AND MCCARTHY, M. A. 2005. Plant traits and local extinctions in natural grasslands along an urban-rural gradient. J. Ecol., 93:1203-1213.

ZAVELATA, E., J. PASARI, J. MOORE, D. HERNANDEZ, K. B. SUTTLE, AND C. C. WILMERS. 2009. Ecosystem response to community disassembly. The Year in Ecology and Conservation Biology, 2009: Ann, N.Y. Acad. Sci., 1162:311-333.

SUBMITTED 14 JUNE 2013 ACCEPTED 13 JUNE 2014

JOHN B. TAFT (1) AND ZACHARY P. KRON (2)

Illinois Natural History Survey, Prairie Research Institute, University of Illinois, Champaign 61820

(1) Corresponding author: 1816 S. Oak Street: Telephone: (217) 244-5046; e-mail: taft@inhs.illinois.edu

(2) Present address: Marek Landscaping, 820 E. Knapp St. Milwaukee, WI. 53202; Telephone: (608) 214-0644; e-mail: zkron2@gmail.com

TABLE 1.--Species identified as significant indicators in the
three vegetation types with their corresponding functional groups
and average % cover. A/B = annual/biennial, PD = perennial dicot,
PM = perennial monocot. * = nonnative species. Nomenclature
follows Mohlenbrock (2002)

Vegetation            Species             Functional       Avg. %
type                                        group          cover

Prairie       Erigeron strigosus          A/B Forb          2.03
              Agrostis alba               C3 Grass          8.70
              Bromus kalmii               C3 Grass          0.63
              Schizachyrium scoparium     C4 Grass         39.13
              Andropogon gerardii         C4 Grass         30.53
              Sorghastrum nutans          C4 Grass         27.07
              Commandra umbellata         Hemi-Parasite     0.83
              Castilleja coccinea         Hemi-Parasite     0.43
              Lespedeza capitata          Legume            0.83
              Vida americana              Legume            0.83
              Silphium terebinthinaceum   PD Forb          21.53
              Parthenium integrifolium    PD Forb          19.83
              Helianthus rigidus          PD Forb           6.47
              Potentilla simplex          PD Forb           4.33
              Aster oolentangiensis       PD Forb           4.23
              Solidago rigida             PD Forb           3.23
              Achillea millefolia *       PD Forb           2.50
              Liatris spicata             PD Forb           2.33
              Rudbeckia hirta             PD Forb           1.60
              Zizea aptera                PD Forb           1.00
              Lithospermum canadense      PD Forb           0.97
              Potentilla arguta           PD Forb           0.07
              Allium cernuum              PM Forb           6.30
              Juncus tenuis               PM Forb           1.63
              Juncus interior             PM Forb           0.23
              Carex pellita               Sedge            11.27
              Rubus pensylvanica          Shrub             5.40
              Crataegus sp. (seedling)    Tree Seedling     1.53
Prairie-Old   Daucus carota *             A/B Forb          8.21
 Field        Dichanthelium acuminatum    C3 Grass          0.71
              Melilotus alba *            Legume            0.94
              Medicago lupulina *         Legume            0.13
              Solidago juncea             PD Forb          25.69
              Antennaria neglecta         PD Forb           6.03
              Fragaria virginica          PD Forb           5.28
              Ratibida pinnata            PD Forb           4.20
              Hieracium caespitosum *     PD Forb           3.27
              Aster drummondii            PD Forb           3.25
              Solidago nemoralis          PD Forb           2.21
              Lobelia spicata             PD Forb           0.48
              Sisyrinchium albidum        PM Forb           0.35
              Prunus americana            Shrub             0.55
              Ulmus americana             Tree Seedling     0.55
Old Field     Allium canadense            PM Forb           5.90

TABLE 2.--Correlations (Pearson's r) for cover and species
richness of ground layer functional groups to NMS axis scores
(n = 44) following rotation of shrub-sapling leaf area index
(LAI) to maximize LAI fit with Axis 1. Significant correlations
(P < 0.05) are shown in bold. Percent variance explained in two
ordination axes was 57% (Axis 1) and 20% (Axis 2)

                                            Cover

                              Axis:      1          2

Leaf Area Index                        0.786#     0.046
Annual & Biennial Forbs               -0.040     -0.127
Ferns and Fern Allies                 -0.094      0.065
Perennial Dicot Forbs                 -0.874#    -0.025
Perennial Monocot Forbs               -0.044     -0.200
Perennial Legumes                     -0.448#    -0.328#
Hemi-Parasites                        -0.327#    -0.244
Perennial [C.sub.3] Grasses           -0.515#    -0.091
Perennial [C.sub.4] Grasses           -0.696#    -0.241
Sedges                                -0.458#     0.344#
Native Woody Species                  -0.472#     0.514#
Non-Native Herbs                      -0.487#     0.250
Non-Native Woody Spp.                  0.100     -0.719#

                                       Species richness

                              Axis:      1          2

Leaf Area Index                        0.786#     0.010
Annual & Biennial Forbs               -0.227      0.094
Ferns and Fern Allies                  0.016      0.024
Perennial Dicot Forbs                 -0.805#     0.179
Perennial Monocot Forbs               -0.544#    -0.037
Perennial Legumes                     -0.436#    -0.326#
Hemi-Parasites                        -0.518#    -0.399#
Perennial [C.sub.3] Grasses           -0.567#     0.021
Perennial [C.sub.4] Grasses           -0.683#    -0.237
Sedges                                -0.147      0.173
Native Woody Species                  -0.274      0.258
Non-Native Herbs                      -0.555#     0.264
Non-Native Woody Spp.                  0.153     -0.010

Note: Significant correlations (P < 0.05) are indicated with #.
COPYRIGHT 2014 University of Notre Dame, Department of Biological Sciences
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2014 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Taft, John B.; Kron, Zachary P.
Publication:The American Midland Naturalist
Article Type:Report
Geographic Code:1U3IL
Date:Oct 1, 2014
Words:5608
Previous Article:Reproductive ecology of the endangered Utah endemic Hesperidanthus suffrutescens with implications for conservation.
Next Article:Growth, biomass, and allometry of resprouting shrubs after fire in scrubby flatwoods.
Topics:

Terms of use | Privacy policy | Copyright © 2022 Farlex, Inc. | Feedback | For webmasters |