Ecological Archives E095-210-D1

Emily Grman, Tyler Bassett, and Lars A. Brudvig. 2014. A prairie plant community data set for addressing questions in community assembly and restoration. Ecology 95:2363. http://dx.doi.org/10.1890/14-0888.1


Introduction

Understanding the determinants of community structure and ecosystem function are central goals of community ecology. Community assembly, the process by which ecosystems accumulate species after a disturbance, is likely a key driver of community structure and ecosystem function, but understanding the intricacies of multiple interacting drivers during community assembly remains a major challenge (Weiher et al. 2011). In part, this is because community assembly may occur over large spatial and long temporal scales, particularly for long-lived taxa such as perennial plant communities which have been a frequent focus of community ecology research (del Moral 2009, Reich et al. 2012). As a result, melding the spatial and temporal realism of observed communities with the mechanistic insights gained by manipulative experiments has proved challenging.

Ecological restoration, or the process of assisting the recovery of an ecosystem that has been degraded, damaged, or destroyed (SERI 2004), has long been lauded for its potential to explore fundamental questions in ecology (Bradshaw 1987, Jordan et al. 1987), including the determinants of community assembly and ecosystem functioning (Young et al. 2001, Temperton et al. 2004, Bullock et al. 2011). Restoration often occurs at large spatial and temporal scales that are relevant to community assembly. Moreover, restoration is an inherently manipulative endeavor, which ranges from strongly active (e.g., introducing species or modifying disturbance regimes) to relatively passive (e.g., cessation of agricultural tillage). As a result, restored ecosystems provide opportunities to explore how manipulated processes (either introduced or ceased) affect communities and ecosystems with an unusual degree of ecological realism. For example, species pools are thought to be critical in determining the outcome of assembly, influencing community properties such as community composition, diversity, and ecosystem function (Foster et al. 2004, Harrison and Cornell 2008, Kraft et al. 2011). However, most studies examining the role of species pools are either observational, where ecologists gather data on species occurrence in the region, or experimental, conducted in micro- or mesocosms. As an illustration of the unique value of restoration data sets, we recently used the data presented here to examine the role of species pool size in structuring beta diversity in prairies <1 to 100 acres in size and 3–8 years old (Grman and Brudvig 2014).

Understanding the outcome of restoration is critical in its own right. Widespread human impacts have degraded communities and ecosystems, leading to losses of biodiversity, ecosystem services, and habitat for declining species (Wilcove et al. 1998, Foley et al. 2005). Restoration of these degraded lands is a key mechanism to decrease the likelihood of extinctions and improve ecosystem functioning. However, the outcome of restoration is known to be highly variable and unpredictable (Suding 2011), calling into question the utility of restoration for improving habitat and ecosystem functions. Restoration ecologists have begun to quantify and compare the relative importance of various ecological influences and management strategies on restoration outcomes (Brudvig 2011), but much work remains. For example, a recent publication using this data set was the first (of which we are aware) to quantify the relative importance, for community composition and diversity, of four major categories of restoration drivers: historical contingency, landscape context, management decisions, and site characteristics (Grman et al. 2013).

The data sets we present here provide opportunities to explore questions in community ecology, in the context of prairie restorations in southwest Michigan, USA. In the summer of 2011, we surveyed the plant communities in 29 3–8-year-old prairies that had been restored from former agricultural lands by private landowners, nature centers, nonprofit foundations, and land conservancies. We also collected data on a suite of likely drivers of community assembly, including soil characteristics, site size and shape, landscape context, history of prescribed fires, and the seeding density and species composition of seed mixes used. Finally, we clipped aboveground biomass to begin to assess the variation in one important ecosystem function provided by restored prairies: primary productivity. Although prairie plant community assembly likely takes decades or longer to fully play out, 3–8 years is typically a timeframe over which restoration practitioners assess success. Furthermore, plant community composition and ecosystem function change dramatically in the first few years of assembly. After 3–8 years, prairies are dominated by sown perennial prairie species, distributions and abundances of these sown species are structured by environmental gradients thought to be important drivers of community assembly, and prairies of this age illustrate recovery in levels of some ecosystem functions (Baer et al. 2002, Foster et al. 2011).

While there are many important community assembly questions that could be addressed with these data, these data sets also present opportunities to explore the relationships among community composition, diversity, ecosystem function, the abundance of exotic and invasive species, soils, and historical legacies. The data set is unusual in that it includes data on the seed mixes applied to each restoration site, making it uniquely suited to addressing the role of species pools and other questions in community assembly through comparisons of seed mixes and the resulting assembled plant communities.

This data paper contains four data sets. The first contains site-level information including the year of sowing, results of soil sample analyses, prior land-use history, number of prescribed fires since restoration, restoration age at the time of sampling in 2011, seeding densities (g of total, grass, or forb seed added per m²), site size and shape, and landscape context. The second data set is a site-by-species matrix presenting the species composition of the seed mixes used during restoration of each prairie site. The third data set is a plot-by-species matrix presenting the species composition we observed in each of ten plots at each prairie. The fourth gives plot-level data on soil moisture, mass of dried aboveground biomass, and cover of litter and bare ground. The accompanying metadata describe the sampling methods and the data sets in more detail.

Metadata

Class I. Data set descriptors

A. Data set identity: A prairie plant community data set for addressing questions in community assembly and restoration

B. Data set identification code

C. Data set description

Originators: Emily Grman, Tyler Bassett, Lars A. Brudvig (affiliations and addresses given above).

Abstract: By assisting the recovery of disturbed or destroyed ecosystems, ecological restoration plays an important role in biodiversity conservation. Moreover, restoration has been heralded as an “acid test” of ecological understanding, by affording the ability to study community assembly, ecosystem function, and human influence over ecosystems across large spatial and long temporal scales. These data sets report the outcome of community assembly, in terms of plant community composition and structure and one important ecosystem function (aboveground biomass production), in 29 prairie restorations in southwestern Michigan. We also report putative forces shaping the outcome of assembly including the species pools (seed mixes applied during restoration), site conditions, landscape context, and land-use history. Detailed knowledge of each restoration effort, including seed mixes used, is unusual and makes these data sets uniquely suited to addressing questions in community assembly by comparing the sown seeds and resulting assembled plant community. For example, we have used the data to test the role of species pools in determining the diversity of assembling communities. We have also used the data to characterize the relative importance of various drivers of community assembly outcomes during restoration, as a step toward resolving the highly contingent and unpredictable outcomes that plague the field of ecological restoration. We suggest that these data sets may prove useful for addressing additional questions in community ecology through the lens of ecological restoration.

D. Key words: aboveground biomass; community assembly; ecosystem function; prairie restoration; seed mix; species pool.

Class II. Research origin descriptors

A. Overall project description

Identity: A prairie plant community dataset for addressing questions in community assembly and restoration

Originators: Emily Grman, Tyler Bassett, Lars A. Brudvig (affiliations and addresses given above).

Period of study: Field data collection occurred July–September 2011.

Objectives: The primary objective of this data collection was to better understand how species pools (seed mixes), site conditions, landscape context, management, and historical contingency affect the plant community that establishes in prairie restorations. We used a group of prairies that had already been restored by private landowners to ensure that the data set is representative of the range of practices and site conditions that characterize restoration in our region. As such, it provides an important point of comparison for restoration experiments implemented by academic ecologists, who may employ methods more amenable to short time spans and small spatial scales but that do not adequately represent the conditions experienced in restoration efforts. Importantly, we collected data on the seed mix used during each restoration from the restoration practitioner, making this dataset uniquely suited to analyses of the role of the species pool. We designed the sampling scheme to enable comparisons of beta diversity (plot-to-plot variation in community composition) among prairies, holding the distance among plots constant regardless of site size. Secondary objectives included understanding the relationships between diversity and ecosystem function outside the highly manipulative systems where it has previously been investigated (Cardinale et al. 2012).

Abstract: given above.

Sources of funding: Michigan State University (MSU), MSU Plant Sciences Fellowship, and George H. Lauff Research Award.

B. Specific subproject description

Site description: The prairies in our data set are spread across four counties in southwest Michigan (Fig. 1), in the eastern portion of the historical, pre-settlement prairie region (Ladd and Oberle 1995). Mean annual precipitation for the region is 103 cm, with slightly more falling during April–September than in winter. Soils were generally loams and sandy loams, including the following soil series: Brady Sandy Loam, Bronson Sandy Loam, Coloma Loamy Sand, Coloma-Marlette Complex, Gilford Sandy Loam, Hillsdale Sandy Loam, Kalamazoo Loam, Marlette-Oshtemo Complex, Oshtemo Sandy Loam, Schoolcraft Loam, Spinks Loamy Sand, Spinks-Oshtemo Complex, and Thetford Sandy Loam.

Fig1

Fig.1. A map of restoration sites included in this data paper. Counties in southwest Michigan in which the prairies are located are labeled; numbers indicate the unique site identification number (site.id).


All the prairies in our data set were former agricultural fields, probably in a corn-soybean rotation. After the cessation of agriculture, some of the fields went through intervening land use of hayfield/pasture or successional oldfield before being restored to prairie; others were continuously in either tillage or no-till row crops until restoration (this prior land use is indicated in the data set). Each prairie was treated with glyphosate herbicide, then sown with seeds of native prairie species once at the initiation of the restoration using a modified Truax drill. The year of initiation is given in the dataset. Most prairies were sown in the spring; two (Site #20 and #27) were sown in the fall. All seeds were supplied by Native Connections (Three Rivers, MI; www.nativeconnections.net).

Sampling design: We visited all the prairies in July-September 2011, the approximate time of peak aboveground biomass in this system, to measure plant community composition (identity and abundance of species), aboveground biomass, and site characteristics. We set up a 50-m sampling transect centered on the approximate center of the site, with ten 1-m² sampled plots along the transect (every five meters, beginning at 0 meters and ending at 45 meters).

Research methods: At each plot, we visually estimated the percent cover by each vascular plant species using Voss (1972, 1985, 1996). We also estimated the percent cover of litter and bare ground that was visible from above the plant canopy. We clipped all the aboveground biomass in each plot. In the lab we sorted samples to remove litter and previous years' production, then oven-dried and weighed the biomass. We took 8 soil cores (20 cm deep) evenly spaced around the perimeter of each plot and composited the cores to make a plot-level soil sample. After sieving and air-drying the soils, we analyzed their capacity to retain moisture when saturated in the lab and then oven-dried (following methods described in Brudvig and Damschen 2011). We created a site-level composite soil sample by pooling plot-level samples from each site; we sent these site-level samples for analysis (Brookside Laboratories, Inc, New Knoxville, OH) of soil texture (sand, silt, and clay content using the hydrometer method; ASTM 2002), pH (1:1 in water; McLean 1982), organic matter (loss on ignition at 360 degrees; Schulte and Hopkins 1996), total exchange capacity (Ross 1995), and phosphorus (Bray-II extractable; Bray and Kurtz 1945). We gathered data on each site's landscape context within 500 m of the sampling transect's midpoint. Within each landscape buffer, we quantified the proportion of landcover in row-crop agriculture, forest, wetland, perennial grassland, and developed uses using ArcGIS 9.3.1 (ESRI, Redlands, CA), 2010 aerial photographs, and recent Google Maps Street Views. When classifying land covers, we considered perennial herbaceous agriculture (e.g., alfalfa), oldfield, and pasture to be grassland. For agricultural cover, we included only annual row-crop agriculture that was in production in 2011 (the year of analysis). We interviewed landowners to assess the number of prescribed fires since seeds were sown to initiate the restoration. We obtained seed mix data directly from the restoration practitioner. We collected data on the precipitation during the year of planting in the appropriate climate division from the National Climatic Data Center (www7.ncdc.noaa.gov/CDO/CDODivisionalSelect.jsp).

Project personnel: Morgan Chavez, Mike Epperly, Michael Gonczar, Jonathan Landis, Suse LaGory

Class III. Data set status and accessibility

A. Status

B. Accessibility

Storage location and medium: Original paper data sheets are archived by L. A. B.

Contact person(s): L. Brudvig, brudvig@msu.edu

Copyright restrictions: None.

Proprietary restrictions: We request that potential data users contact the authors when using the data. This will assist the authors in documenting data use by the scientific community.

Citation: Please cite these data as Grman, E, T. Bassett, and L. A. Brudvig (2014).

Costs: None.

Class IV: Data structural descriptors

There are four data sets associated with this paper.

SITE INFORMATION

A. Data set file

Identity: SITE INFO.csv. Various information about the history, management, and abiotic characteristics of each restoration site.

Size: 30 rows (including header), 23 columns

Format and storage mode: Comma separated file

Header Information: Table 1 describes the variables given in the header.

Alphanumeric attributes: Mixed

B. Variable information. Table 1 lists and describes the variables in SITE INFO.csv.

Table 1. Variable information for SITE INFO.csv.

Variable identity

Variable definition

Units of measurement

Data type

Min and max
of numeric data

Codes for
character data

site.id

Unique identifier for each restoration site

NA

Integer

1–29

NA

year

Year in which restoration occurred

NA

Integer

2003–2008

NA

clay

Soil percent clay

%

Number

4.04–19.83

NA

silt

Soil percent silt

%

Number

8.35–53.91

NA

sand

Soil percent sand

%

Number

28.95–87.61

NA

ph

Soil solution pH

NA

Number

5.1–6.7

NA

om

Soil organic matter

%

Number

1.18–7.93

NA

tot.ec

Soil total exchange capacity

meq/100g

Number

3.26–14.77

NA

bray.p

Soil Bray-II extractable phosphorus

mg P/kg soil

Number

6–315

NA

land.use

Land use immediately prior to restoration

NA

Character

NA

Tilled=tillage agriculture,
Hay=hayfield or pasture,
Oldfield=abandoned
(not tilled, grazed, or mowed)

burn.history

Number of prescribed fires since restoration

NA

Integer

0–6

NA

site.age

Number of years between restoration and sampling in 2011

NA

Integer

3–8

NA

g.seeds.m²

Total seeding density; mass of all seeds applied per m² during restoration

g/m²

Number

0.68–1.30

NA

g.forb.m²

Forb seeding density; mass of all forb seeds applied per m² during restoration

g/m²

Number

0.003–0.85

NA

g.grass.m²

Grass seeding density; mass of grass seeds applied per m² during restoration

g/m²

Number

0.45–0.81

NA

site.area

Area of site

Number

2230–389,456

NA

edgearea

Site edge:area ratio

m/m²

Number

0.007–0.106

NA

forest500

Cover of forest within 500 m of site center

Number

13,588–557,577

NA

ag500

Cover of agricultural field within 500 m of site center

Number

0–589,396

NA

grass500

Cover of perennial grassland within 500 m of site center

Number

83,505–571,684

NA

wet500

Cover of wetland (forested or herbaceous) within 500 m of site center

Number

0–375,363

NA

dev500

Cover of developed land (such as buildings, roads, mowed lawns, etc) within 500 m of site center

Number

5433–370,367

NA

precip.planting

Precipitation that fell during the year of sowing to prairie

mm

Number

807.7–1212.3

NA

C. Data anomalies. One site (Site #28) had anomalously high phosphorus (315 ppm, whereas the next highest value was 108 ppm). We included this value in the data set because we deemed it likely that the very high P was a legacy of very high manure applications during agricultural land use. While we have no data on the history of manure application, it is plausible that farmers would have decided to apply large amounts to this site to compensate for the site's very sandy, low organic matter soils and low productivity. Two sites (Sites #4–5) are listed as having an oldfield prior land use; they were fallow for only one year after the cessation of tillage agriculture and sowing for restoration. Some sites are very near to each other, and consequently their 500-m buffer zones for classifying surrounding land-use overlap (e.g., Sites #6–15 (except #11 and #14) overlap some portion of their surrounding landcovers; #4–5 overlap with each other; #16–17 overlap with each other; #18–20 overlap with each other; #24–26 overlap with each other; #27–28 overlap with each other). A smaller buffer size would have reduced this problem substantially, but a smaller buffer would be more directly influenced by the size of the restoration itself, a situation we wished to avoid.

SEED MIX COMPOSITION

A. Data set file

Identity: SEED MIX.csv. A site-by-species matrix of the composition of seed mixes applied during restoration.

Size: 30 rows (including header), 137 columns

Format and storage mode: Comma separated file

Header Information: The first variable in the header ("site.id") indicates the unique site identifier. Headers in columns 2–137 name the species included in the seed mixes.

B. Variable information. The first variable ("site") indicates the unique site identifier. Columns 2–137 are numeric, indicating the g/m² of seed of that species included in that site's seed mix. We standardized all species names using the Michigan Flora (Voss 1972, 1985, 1996) for easy comparisons with the plant community data. A few species were not included in the Michigan Flora (Dalea candida, Desmanthus illinoensis, Dracopis amplexicaulis, Echinacea angustifolia, Echinacea paradoxa, Thalictrum pubescens); for these we used the Flora of the Great Plains (Great Plains Flora Association 1986) or the Manual of Vascular Plants of Northeastern United States and Adjacent Canada (Gleason and Cronquist 1991).

C. Data anomalies. For two sites (Sites #9–10), we have data on the total seeding density, grass seeding density, and forb seeding density, but not the abundance of each species included in the seed mix. For those sites, we have indicated species' presence in the mix with "1" and absence with "0"; these sites should be removed from any analyses of the seed mix that use species relative abundance information. The Helianthus strumosus included in these seed mixes was probably either a mixed collection of H. strumosus and H. divaricatus, or a hybrid population of the two species.

PLANT COMMUNITY COMPOSITION

A. Data set file

Identity: PLANT COMMUNITY.csv. A plot-by-species matrix indicating the observed plant community in each of 10 sampling plots in each prairie restoration site.

Size: 291 rows (including header), 254 columns

Format and storage mode: Comma separated file

Header Information: The first variable in the header ("site.id") indicates the unique site identifier (a number from 1–29). The second ("plot") indicates the location along the transect where the plot was located. Headers in columns 3–254 name the species.

B. Variable information. The first variable ("site.id") indicates the unique site identifier (a number from 1–29). The second variable ("plot") is a character variable indicating the plot name, the plot's location along the transect (i.e., m0 is the plot located at the 0 meter mark, m5 is the plot located at the 5 meter mark). Columns 3–254 are numeric, indicating the percent cover of that species observed in the plot. We used Voss (1972, 1985, 1996) as a reference for species names.

C. Data anomalies. We could not identify all specimens to species. We include all observed taxa in the data set, but we chose to be conservative in including ambiguous taxa in our analyses of community composition (e.g., Grman et al. 2013, Grman and Brudvig 2014). For some genera, we could identify some specimens but not others (for example when we had only plant fragments). For unknowns where we were not sure the specimen was unique from those identified to species, we chose to exclude the taxon from analyses of species composition; these are labeled as "unknown.Genus." For other unknowns, where we were sure a specimen was unique from other identified specimens in the genus (or there were no other identified specimens in the genus), we chose to include the taxa in analyses of community composition; these are labeled as "Genus." For example, we know that specimens in the genus Oenothera that could not be identified are not Oenothera biennis (the only identified species in that genus); they are labeled Oenothera and we chose to include them in analyses of species composition. For other genera, we chose to lump all individuals into a single taxon and included it in analyses of species composition (e.g., Acer). Where we could make no identification, we gave each specimen a unique label (e.g., unknown.gram9); we excluded these from analyses of community composition.

PLOT-LEVEL SOIL MOISTURE, LITTER, BARE GROUND COVER, AND ABOVEGROUND BIOMASS

A. Data set file

Identity: PLOT LEVEL.csv.

Size: 291 rows (including header), 4 columns

Format and storage mode: Comma separated file

Header Information: Table 2 describes the variables in the header.

Special characters/fields: NA indicates missing data

B. Variable information. Table 2 lists and describes the variables in PLOT LEVEL.csv.

Table 2. Variable information for PLOT LEVEL.csv.

Variable

Variable definition

Units of measurement

Data type

Min and max
of numeric data

Codes for character data

site.id

Unique identifier
for each restoration site

NA

Integer

1–29

NA

plot

Identifier for plot location
along transect

NA

Character

 

m0 = plot located at 0 meters;
m5 = plot located at 5 meters;
m10 = plot located at 10 meters; etc.

perc.water

Soil moisture

%

Numeric

26.52–62.30

NA

biomass

Dried mass of clipped
aboveground biomass

g/m²

Numeric

67.57–1249

NA

C. Data anomalies. There are six missing aboveground biomass values (indicated by NA) as a result of labeling errors or lost samples.

Class V. Supplemental descriptors

A. Data acquisition. Original data entry files are maintained by the authors.

B. Quality assurance/quality control procedures. Original data entry into Excel (2007) files was double-checked.

C. Computer programs and data-processing algorithms. Site areas, edge:area ratios, and surrounding landscape cover data were analyzed using ArcGIS 9.3.1 (ESRI, Redlands, CA). Data were manipulated in R 2.14.0, library reshape.

D. Publications and results. The following papers make use of some or all of these data sets:

Grman, E., T. Bassett, and L. A. Brudvig. 2013. Confronting contingency in restoration: management and site history determine outcomes of assembling prairies, but site characteristics and landscape context have little effect. Journal of Applied Ecology 50:1234–1243.

Grman, E. and L. A. Brudvig. 2014. Species pool richness increases beta diversity among prairie restorations, but not through enhanced species sorting. Journal of Ecology. DOI: 10.1111/1365-2745.12267.

Acknowledgments

We gratefully acknowledge the support of the Edward Lowe Foundation, the Kalamazoo Nature Center, the Southwest Michigan Land Conservancy, the Fetzer Institute's GilChrist retreat center, and the many private landowners who allowed us access to their prairies and provided information on their history and management. We also thank Jerry Stewart at Native Connections for supplying data on seed mixes and restoration methods.

Literature cited

ASTM D422. 2002. Standard test method for particle-size analysis of soils. ASTM International.

Baer, S. G., D. J. Kitchen, J. M. Blair, and C. W. Rice. 2002. Changes in ecosystem structure and function along a chronosequence of restored grasslands. Ecological Applications 12:1688–1701.

Bradshaw, A. D. 1987. Restoration: an acid test for ecology. Pages 23–29 in W. R. I. Jordan, M. E. Gilpin, and J. D. Aber, editors. Restoration Ecology: A Synthetic Approach to Ecological Research. Cambridge University Press, Cambridge, UK.

Bray, H. R., and L. T. Kurtz. 1945. Determination of total, organic, and available forms of phosphorus in soils. Soil Science 59:39–45.

Brudvig, L. A. 2011. The restoration of biodiversity: where has research been and where does it need to go? American Journal of Botany 98:549–558.

Brudvig, L. A., and E. I. Damschen. 2011. Land-use history, historical connectivity, and land management interact to determine longleaf pine woodland understory richness and composition. Ecography 34:257–266.

Bullock, J. M., J. Aronson, A. C. Newton, R. F. Pywell, and J. M. Rey-Benayas. 2011. Restoration of ecosystem services and biodiversity: conflicts and opportunities. Trends in Ecology and Evolution 26:541–549.

Cardinale, B. J., J. E. Duffy, A. Gonzales, D. U. Hooper, C. Perrings, P. Venail, A. Narwani, G. M. Mace, D. Tilman, D. A. Wardle, A. P. Kinzig, G. C. Daily, M. Loreau, J. B. Grace, A. Larigauderie, D. S. Srivastava, and S. Naeem. 2012. Biodiversity loss and its impact on humanity. Nature 486:59–67.

del Moral, R. 2009. Increasing deterministic control of primary succession on Mount St. Helens, Washington. Journal of Vegetation Science 20:1145–1154.

Foley, J. A., R. DeFries, G. P. Asner, C. Barford, G. Bonan, S. R. Carpenter, F. S. Chapin, M. T. Coe, G. C. Daily, H. K. Gibbs, J. H. Helkowski, T. Holloway, E. A. Howard, C. J. Kucharik, C. Monfreda, J. A. Patz, I. C. Prentice, N. Ramankutty, and P. K. Snyder. 2005. Global consequences of land use. Science 309:570–574.

Foster, B. L., T. L. Dickson, C. A. Murphy, I. S. Karel, and V. H. Smith. 2004. Propagule pools mediate community assembly and diversity-ecosystem regulation along a grassland productivity gradient. Journal of Ecology 92:435–449.

Foster, B. L., E. J. Questad, C. D. Collins, C. A. Murphy, T. L. Dickson, and V. H. Smith. 2011. Seed availability constrains plant species sorting along a soil fertility gradient. Journal of Ecology 99:473–481.

Gleason, H. A., and A. Cronquist. 1991. Manual of Vascular Plants of Northeastern United States and Adjacent Canada. Second edition. New York Botanical Garden, Bronx, New York, USA

Great Plains Flora Association. 1986. Flora of the Great Plains. University Press of Kansas, Lawrence, Kansas, USA.

Grman, E., T. Bassett, and L. A. Brudvig. 2013. Confronting contingency in restoration: management and site history determine outcomes of assembling prairies, but site characteristics and landscape context have little effect. Journal of Applied Ecology 50.

Grman, E. and L. A. Brudvig. 2014. Beta diversity among prairie restorations increases with species pool size, but not through enhanced species sorting. Journal of Ecology DOI:10.1111/1365-2745.12267.

Harrison, S., and H. Cornell. 2008. Toward a better understanding of the regional causes of local community richness. Ecology Letters 11:969–979.

Jordan, W. R., III., M. E. Gilpin, and J. D. Aber. 1987. Restoration ecology: ecological restoration as a technique for basic research. Pages 3–21 in W. R. Jordan, III., M. E. Gilpin, and J. D. Aber, editors. Restoration Ecology: A Synthetic Approach to Ecological Research. Cambridge University Press, Cambridge, UK.

Kraft, N. J. B., L. S. Comita, J. M. Chase, N. J. Sanders, N. G. Swenson, T. O. Christ, J. C. Stegen, M. Vellend, B. Boyle, M. J. Anderson, H. V. Cornell, K. F. Davies, A. L. Freestone, B. D. Inouye, S. P. Harrison, and J. A. Myers. 2011. Disentangling the drivers of β diversity along latitudinal and elevational gradients. Science 333:1755–1758.

Ladd, D. M., and F. Oberle. 1995. Tallgrass Prairie Wildflowers: A Field Guide. Falcon Publishing, Helena, Montana, USA.

McLean, E. O. 1982. Soil pH and lime requirement. Pages 199–223 in e. a. A. L. Page, editor. Methods of Soil Analysis, Part 2. ASA and SSSA, Madison, Wisconsin, USA.

Reich, P. B., D. Tilman, F. Isbell, K. Mueller, S. E. Hobbie, D. F. B. Flynn, and N. Eisenhauer. 2012. Impacts of biodiversity loss escalate through time as redundancy fades. Science 336:589–592.

Ross, D. 1995. Recommended soil tests for determining exchange capacity. Pages 62–69 in J. T. Sims and A. Wolf, editors. Recommended soil testing procedures for the northeastern United States. Northeastern Regional Bulletin #493. Ag Experiment Station, University of Delaware, Newark, Delaware, USA.

Schulte, E. E., and B. G. Hopkins. 1996. Estimation of soil organic matter by weight loss-on-ignition. Pages 21–32 in F. R. Magdoff, M. A. Tabatabai, and E. A. Hanlon, Jr., editors. Soil Organic Matter: Analysis and Interpretation. Special Publication No. 46. Soil Science Society of America, Madison, Wisconsin, USA.

Society for Ecological Restoration. 2004. SER International Primer on Ecological Restoration. Society for Ecological Restoration International Science & Policy Working Group.

Suding, K. N. 2011. Toward an era of restoration in ecology: successes, failures, and opportunities ahead. Annual Review of Ecology and Systematics 42:465–487.

Temperton, V. M., R. J. Hobbs, T. Nuttle, and S. Halle, editors. 2004. Assembly Rules and Restoration Ecology. Island Press, Washington, DC, USA.

Voss, E. G. 1972. Michigan Flora, Part I: Gymnosperms and Monocots. Cranbrook Institute of Science, Bloomfield Hills, Michigan, USA.

Voss, E. G. 1985. Michigan Flora, Part II: Dicots (Saururacae-Cornaceae). Cranbrook Institute of Science, Bloomfield Hills, Michigan, USA.

Voss, E. G. 1996. Michigan Flora, Part III: Dicots (Pyrolaceae-Compositae). Cranbrook Institute of Science, Bloomfield Hills, Michigan, USA.

Weiher, E., D. Freund, T. Bunton, A. Stefanski, T. Lee, and S. Bentivenga. 2011. Advances, challenges and a developing synthesis of ecological community assembly theory. Philosophical Transactions of the Royal Society of London, Series B 366:2403–2413.

Wilcove, D. S., D. Rothstein, J. Dubow, A. Phillips, and E. Losos. 1998. Quantifying threats to imperiled species in the United States. BioScience 48:607–615.

Young, T. P., J. M. Chase, and R. T. Huddleston. 2001. Community succession and assembly: comparing, contrasting, and combining paradigms in the context of ecological restoration. Ecological Restoration 19:5–18.


[Back to E095-210]