Research Projects
Comparing grasshopper (Orthoptera: Acrididae) communities on tallgrass prairie reconstructions and remnants in Missouri
Period: March 1, 2020 - Present
Contact: Prairie Fork
Organization: University of Missouri Prairie Fork Trust, Missouri Department of Conservation, US Geological Survey
Funding Source: MDC
Objectives: Comparing grasshopper (Orthoptera: Acrididae) communities on tallgrass prairie reconstructions and remnants in Missouri
JOSEPH P. LAROSE, 1 ELISABETH B. WEBB2 and DEBORAH L. FINKE 1 1Division of
Plant Sciences, University of Missouri, Columbia, MO, USA and 2U.S. Geological Survey, Missouri Cooperative Fish and Wildlife Research Unit, University of Missouri, Columbia, MO, USA
Abstract. 1. Tallgrass prairies, which once occupied a large swath of central North America, face the combined challenges of habitat loss and fragmentation. In Missouri, where less than 1% the historical prairie remains, prairies are being reconstructed from agricultural or wooded land.
2. Invertebrates are often assumed to colonise reconstructions if native vegetation returns; however, the limited mobility of many invertebrates, the isolation of many tall- grass remnants, and the difficulty in establishing prairie plants raise serious questions as to whether invertebrate communities on reconstructed prairies are and will be equivalent to those found on remnant prairies.
3. Grasshoppers (Acrididae) display a range of dispersal capabilities and may be valu- able for assessing the success of prairie restoration for invertebrates.
4. Our first objective was to compare grasshopper communities on reconstructed and remnant prairies and, if differences existed, identify species or functional groups associ- ated with each habitat type. The second objective was to evaluate the effect of time because prairie reconstruction on grasshopper communities to determine if communities on reconstructions are converging with communities on remnants.
5. Our results suggest that prairie reconstructions in Missouri do not support the same communities of grasshoppers as prairie remnants.
6. Grasshopper diversity was generally greater on remnants. Many species had not colonised nearby reconstructions.
7. Communities on prairie reconstructions were characterised by a few long-winged, generalist species that are typically successful in agroecosystems.
8. Further investigation into the habitat disparities driving low grasshopper diversity on reconstructions could help restore the grasshopper community of reconstructions.
Key words. Dispersal, Ecology, Grassland, Orthoptera, Prairies, Restoration, Succession.
Introduction
Grassland ecosystems are in decline globally due to intensifica- tion of land use by humans (Ceballos et al., 2010). Predictably, species that rely on grassland habitat have declined as well, in some cases precipitously (Brennan & Kuvlesky Jr, 2005; Pleasants & Oberhauser, 2013). Besides traditional conservation
Correspondence: Deborah L. Finke, Division of Plant Sciences, 1-33 Agriculture Building, University of Missouri, Columbia, MO 65211, Tel: 573.884.5125, Fax: 573.882.1469. E-mail: finked@missouri.edu
of remaining fragments of native grassland, there are also efforts to reconstruct grassland from agricultural or otherwise disturbed land (Dobson et al., 1997; Török & Helm, 2017). Tallgrass prai- rie reconstruction in Missouri is an excellent example of the ongoing effort to rebuild ecosystems based on information from only small fragments of remnant habitat. Tallgrass prairie once covered close to a third of Missouri but has since been reduced by over 99% (Christisen, 1972) due to mechanised agriculture, urbanisation, and forest expansion (Samson & Knopf, 1994; Wright & Wimberly, 2013). The remaining tallgrass prairie in Missouri consists of small patches within a landscape of row
Published 2019. This article is a U.S. Government work and is in the public domain in the USA. 1
crops, pasture, and woodland (Christisen, 1972; Solecki & Toney, 1986). Conservation organisations such as the Missouri Prairie Foundation and The Nature Conservancy, as well as public nat- ural resource agencies including the Missouri Department of Conservation, have responded by converting agricultural or wooded land back to prairie, a process of unknown duration and with substantial challenges.
The process of tallgrass prairie reconstruction is often quite extensive (Smith, 2010; Kurtz, 2013). Following removal of existing vegetation occupying the land, whether forest, crops, or pasture, managers plant a diverse native seed mix containing grasses and forbs (Smith, 2010; Kurtz, 2013). Management can be intensive, often involving spraying herbicides, burning, mow- ing, and grazing (Smith, 2010). Ascertaining the effectiveness of prairie reconstruction and management practices requires moni- toring animal and plant communities of reconstructions as well as remnant prairies (Kremen et al., 1994; Thom, 2000; Block et al., 2001; Benayas et al., 2009). Remnant prairies represent the best examples of historical tallgrass prairie ecosystems, and thus often serve as target communities for reconstruction efforts. Of particular interest for monitoring and conservation are taxa that may not be able to surmount the obstacles to colonisation posed by habitat fragmentation, such as grasshoppers (Acrididae) (Hjermann & Ims, 1996; Reinhardt et al., 2005; Schultz et al., 2008; Heidinger et al., 2010; Ortego et al., 2015). Grasshoppers do not typically garner extensive conservation attention. Restoration or reconstruction studies focusing on grass- hoppers are less common than on economically beneficial insects like bees (Harmon-Threatt & Hendrix, 2015; Griffin et al., 2017; Tonietto et al., 2017) or publicly popular insects such as butterflies (Ries et al., 2001; Swengel, 2001; Davis et al., 2007; Vogel et al., 2007; Brückmann et al., 2010; Kuefler et al., 2010). Several grass- hopper species are abundant, serious agricultural pests that receive intensive insecticidal control (Lockwood et al., 1988). Despite some species’ abundance, the loss of grasshopper diversity in prai- rie ecosystems is a distinct possibility considering the sudden extinction of the Rocky Mountain locust, Melanoplus spretus (Scudder) (Acrididae: Melanoplinae), which once rivalled the bison in terms of animal biomass on North American prairies (Lockwood, 2004), along with the endemism found in some grass- hopper genera (Hilliard Jr, 2001; Knowles, 2001; Otte, 2012). Indeed, 15 grasshopper species are already listed as species of concern by the Missouri Department of Conservation (Missouri Natural Heritage Program, 2018) and the status of currently unde- scribed or recently described species is largely unknown. We eval- uated grasshopper communities not only for the sake of grasshopper conservation, but also because of their potential as indicators of overall prairie reconstruction health as orthopterans are known to be good indicators of biodiversity in agricultural and grassland areas (Sauberer et al., 2004; Bazelet & Samways,
2011; Alignan et al., 2018).
Colonisation ability in insects is tied to dispersal ability (Zera & Denno, 1997; Thompson, 1999; Lester et al., 2007; Picaud & Petit, 2008) and habitat specificity (Piechnik et al., 2008; Dennis et al., 2011). Grasshopper species display widely variable dispersal capabilities. Some species possess long wings and powerful flight muscles that allow them to disperse long distances; the migratory grasshopper Melanoplus sanguipines (Fabricius) (Acrididae:
Melanoplinae) is capable of dispersing hundreds of kilometres (Pfadt, 1994). Other grasshoppers have poorly developed wings and are only capable of short distance dispersal by hopping (Pfadt, 1994), such as Melanoplus gracilis (Bruner) (Acrididae: Melanoplinae). Grasshoppers with strong dispersal abilities may be better colonisers because they are more capable of immigrating to new habitats. In fragmented landscapes, such as grasslands in the United States and much of Europe, differences in mobility can affect grasshopper community composition and persistence (Marini et al., 2010; Marini et al., 2012), with low mobility grass- hoppers being more likely to go extinct than high mobility grass- hoppers (Reinhardt et al., 2005). In France, wing length to body length ratios of grasshopper species decreased over succession in old fields, suggesting that better dispersing species colonise new habitats before poor dispersers (Picaud & Petit, 2008). A survey of Wisconsin prairies found that reconstructed prairies were occu- pied by a suite of mobile species (Bomar, 2001). Those mobile spe- cies were also habitat generalists. Their relative abundance on reconstructions may reflect the role of habitat generalism in coloni- sation. Diet is a major part of habitat specificity. Grasshoppers dis- play a range of diet breadths (Joern, 1979), and species with broad diets may be better colonisers because they can survive and repro- duce under a variety of resource conditions and habitats (Peterson & Denno, 1998; Piechnik et al., 2008).
There were two objectives of our study. First, we compared grasshopper communities on reconstructions to those on rem- nants by evaluating species richness, diversity, and community composition at four pairs of reconstructed and remnant prairies in Missouri. We sought to identify species or functional groups associated with remnants or reconstructions for use in monitor- ing these reconstructions and remnants in the future. Second, we evaluated the effect of prairie reconstruction age on grasshop- per communities to determine if grasshopper communities on reconstructions were converging with communities on remnants. We hypothesised that grasshopper species richness and diver- sity would be greater on remnants than on reconstructions, although we were aware that insect species richness on remnants is not always greater than on the presumably lower quality recon- structions (Davis et al., 2007; Williams, 2011; Diepenbrock et al., 2013) because of differences in restoration practices and remnant health. In accordance with results from the Wisconsin prairie survey from Bomar (2001), we hypothesised that long- winged grasshoppers and grasshoppers with generalist diets, those feeding on a mixture of grasses and forbs, would be more common on reconstructions than remnants because of increased probability of successfully colonising new habitat. We also hypothesised that smaller, short-winged, specialist species would be more common on remnants than reconstructions because they are not as likely to colonise newly formed habitat
patches.
Materials and methods
Site selection
We sampled four areas (locations) containing both recon- structed and remnant prairies managed by the Missouri
Department of Conservation (Supporting Information Fig. S1). Wah’Kon-Tah Prairie, Linscomb Wildlife Area, and Schell- Osage Conservation Area (hereafter Schell), all within St. Clair County in the Upper Osage Grasslands of southwestern Missouri, contained contiguous remnant and reconstructed prai- ries. Wah’Kon-Tah’s remnant prairie covers 756 ha and its reconstruction covers 160. Linscomb had 41 ha of remnant and 32 of reconstruction. Schell had approximately 1.5 ha of each. The remaining location, approximately 200 km northeast in the Central Dissected Plains in Calloway County, consisted of one remnant, the University of Missouri’s Tucker Prairie (59 ha), and one reconstructed prairie, Prairie Fork Conservation Area (142 ha). These two prairies are one location in our analyses (hereafter North) even though they were separated by 32 km. We included the North location in order to generalise our conclu- sions for reconstructions in various parts of Missouri.
The reconstructed prairies at Wah’Kon-Tah, Linscomb, and North contained reconstructions of different ages. There were ten individual reconstructions at Wah’Kon-Tah initiated between 2002 and 2008. Linscomb contained two reconstructions, one reconstructed in 2007 and the other in 2013. The North location had the greatest range of reconstruction ages: 2004 to 2016. Schell had only one reconstruction site, initiated in 2014. For analyses, reconstruction age was calculated as the years since planting until sampling year.
There are various grassland management practices that could potentially confound invertebrate surveys. We sought to identify patterns independent of management practices, therefore, we excluded tracts scheduled to be hayed, grazed, mowed, or high-clipped in 2016 or 2017 because those practices could alter invertebrate communities during summer months (Humbert et al., 2010). Prescribed burning also affects invertebrate com- munities (Panzer, 2002), but we included burned patches in our site selection because the burns were scheduled for the dormant season, outside of the sampling window.
Due to the heterogeneous and fragmented nature of the native prairie patches available for sampling, we chose to sample tran- sects at locations randomly generated in each remnant and recon- structed prairie. We generated the same number of points on remnants and reconstruction at each location except for North. There were double the number of transects on the North recon- struction because it was larger than the remnant and contained the widest range of reconstruction ages, which we were particu- larly interested in sampling. We used ArcMap 10.3.1 (ESRI 2015) to randomly generate transect sampling points, which were regenerated after 2016 for sampling in 2017. Each point, which represented the centre of a transect, was located >40 m from the prairie edge and > 75 m from another transect. We assumed that a minimum of 75 m between points ensured inde- pendence, because nymphs, which made up the vast majority of grasshoppers captured in standardised sweeps, are incapable of flight. Adult long-winged grasshoppers however, may be capable of traveling distances >75 m (Pfadt, 1994). In 2016 we sampled along 134 transects: 75 on reconstructions and 59 on remnants. In 2017, we reduced sampling intensity at the North location, the most intensely sampled location, by a third in order to redirect effort to another method of collection described in the next section. This lowered total transects to 116: 63 on
reconstructions and 53 on remnants. A summary of transects sampled at each location in 2016 and 2017 are located in Sup- porting Information Table S1.
Collecting
We collected grasshoppers using two methods: standardised sweeping and targeted capture. We performed standardised sweeping, generally accepted as the best grasshopper collection method (Evans et al., 1983; Larson et al., 1999), along 60 m transects centred on the randomly generated points. We per- formed 10 sweeps perpendicular to each transect with a 38 cm diameter net at four spots along the transect, located 15 m and 30 m from the centre of the transect in both directions. The four subsamples were combined into a single sample; thus, each tran- sect consisted of 40 total sweeps. By necessity, seven people per- formed the sweeping over the 2 years. Different sweeping techniques can bias the grasshoppers collected (O’Neill et al., 2002); therefore researchers used and practised the same standar- dised sweeping motion and made an effort to assign sweepers to different prairie types throughout the season. One researcher conducted 45% of the sweeps.
In 2016, sampling occurred once in June, once in July, and once in August/September. Sweeps were conducted on days without consistent precipitation, after dew had evaporated suffi- ciently, when temperature was above 21 oC, and winds were under 15 km/h. The majority of individuals caught in 2016 were nymphs. In 2017, we restricted standardised sweeping to one visit in August/September and we replaced the early-season stan- dardised sweeps with targeted capture to catch more adult grass- hoppers. Targeted capture entailed walking slowly and capturing every adult grasshopper that we observed or flushed. We spent equal time conducting targeted captures at remnant and recon- structed prairies and alternated sampling between the two prairie types several times per day. We spent approximately 100 h con- ducting targeted capture from June to September 2017. The same temperature, wind, and precipitation requirements applied to tar- geted capture.
Grasshoppers were identified to species, or genus for some nymphs and females, using Pfadt (1994), Kirk and Bomar (2003), Ballard (1992), and Song (2009). Voucher specimens were pinned or frozen. Pinned specimens are stored in the Enns Entomological Museum at the University of Missouri, Columbia.
Vegetation
We measured vegetation density and estimated forb to grass ratio during standardised sweeping in August/September of 2017. We measured vegetation density with a modified Robel pole (Benkobi et al., 2000; Uresk & Benzon, 2007) at three ran- domly chosen sweeping points along each transect. At each loca- tion we recorded the lowest decameter on the Robel pole visible from a distance of 4 m and a height of 1 m. We visually esti- mated the forb to grass coverage ratio of a 1.0 × 0.5 m plot at the locations where we measured vegetation density.
Statistical analyses
Do communities on reconstructions differ from those on rem- nants?
Diversity. We compared taxon richness (species plus genera that were not identified further) and diversity in reconstructed and remnant prairies at each location using rarefaction/extrapola- tion (Weibull et al., 2003; Gotelli & Colwell, 2011; Colwell et al., 2012), which resamples species data to estimate richness or diversity at other sample sizes. We conducted all analyses using R version 3.4.0 (R Core Team, 2017). We performed individual-based rarefaction instead of sample-based (Gotelli & Colwell, 2011) using the package iNEXT (Hsieh et al., 2016) because of the addition of the targeted capture grasshopper observations in 2017. Individual-based rarefaction assumes independence for each individual, which is often violated because some individuals are caught at the same location, thus failing to account for clumping. This was the case for the individual-based rarefaction we performed.
The rarefaction/extrapolation curves depict Hill numbers, which are measures of diversity that combine species richness and abundance (Hsieh et al., 2016). The curves show the esti- mated Hill numbers at hypothetical sample sizes, ranging from zero to two-times the actual sample size. We generated rarefac- tion/extrapolation of Hill curves for each year separately as well as lumped together. We plotted the curves using 95% confidence intervals, calculated with the bootstrap method (Colwell et al., 2012).
One of the parameters in Hill number equations is q, which determines the sensitivity to relative frequencies of species (Chao et al., 2014). Hill numbers were calculated for q = 0, 1, and 2. The resulting estimates are species richness (q = 0), Shannon diversity (q = 1), and Simpson diversity (q = 2). The three diversity metrics are influenced differently by relative fre- quencies of species (Chao et al., 2014). Species richness is not influenced by relative frequency, and only refers to the presence of a species. Shannon diversity weights species according to their relative frequencies (Peet, 1975; Routledge, 1979; Chao et al., 2014), meaning communities with highly skewed relative abundances will have lower Shannon diversity than communi- ties with the same number of evenly abundant species (Peet, 1975; Keylock, 2005). Simpson diversity discounts rare species and places greater emphasis on abundant species, making it a good measure of the diversity of dominant species (Keylock, 2005; Chao et al., 2014). Shannon diversity estimates are pre- sented as the exponentials of Shannon indices, and Simpson diversity estimates are presented as inverses of Simpson concen- tration, such that larger numbers represent greater diversity (Hsieh et al., 2016).
Total abundance. We modelled total abundance of grasshop- pers using univariate generalised linear models. A mixed effect model was not appropriate because there were less than the five levels necessary to estimate the among-population variance accurately (Harrison et al., 2018); in this case there were only four locations. We used a negative binomial distribution, which was the best fit for the data based on Dunn-Smyth residual plots. We evaluated the effects of status (reconstruction or remnant),
location (Wah’Kon-Tah, Linscomb, Schell, or North) and the status × location interaction. Only grasshopper data from 2016 were included in the models because we did not collect enough grasshoppers from standardised sweeping in 2017. We started with a model including all variables and interactions, then removed interactions and variables one at time. We conducted analyses of variance (ANOVA) on models with and without var- iables to determine whether variables improved model fit, dis- carding those that did not (P > 0.05) (Blakey et al., 2016; Clarke-Wood et al., 2016). We used the function glm.nb in the package MASS (Venables & Ripley, 2002). We did not perform similar analyses comparing diversity and richness between reconstructions and remnants because grasshopper abundance, diversity, and richness were strongly correlated at each transect.
Community composition. The following analyses were per- formed only on grasshoppers collected in 2016 because of the low number of grasshoppers collected from transects in 2017. Individuals belonging to the genera Orphulella (Acrididae: Gomphcerinae) and Hersperotettix (Acrididae: Melanoplinae) were not identified to species because all individuals captured in 2016 were nymphs that were difficult to identify past genera. To visualise community data, we ordinated grasshopper commu- nities using non-metric multidimensional scaling (NMS) with a Bray–Curtis dissimilarity matrix (Paton et al., 2009; Clarke- Wood et al., 2016). NMS ordination compresses the abundance and species information from each sample and constructs a space of k dimensions based on the differences between samples. Taxa abundances were summed at each transect. Only taxa that occurred in >5% of transects were included in the ordination, thus we included only the 15 most common grasshopper taxa in the ordination. Eight transects (four remnant, four reconstruc- tion) with zero individuals were removed prior to analysis. We used the function metaMDS in package vegan (Oksanen et al., 2016) to run NMS. We used the function dimcheckMDS to choose the number of dimensions (k) and viewed the resulting ordination with the ordirgl function.
To test whether grasshopper communities on reconstructed and remnant prairies were statistically distinct, we constructed multivariate models of 2016 grasshopper abundances using the same 15 taxa used in the ordinations. Models were generated in the package mvabund (Wang et al., 2012), which handles mul- tivariate count data with generalised linear models. The response variables were the abundances of each taxa summed across sam- pling dates for each transect and we used a negative binomial dis- tribution for all models. Explanatory variables included prairie status (remnant or reconstructed), location (Wah’Kon-Tah, Lin- scomb, Schell, and North), edge proximity (distance from tran- sect to closest prairie edge, measured in ArcMAP) and all possible interactions. We started with a model including all vari- ables and interactions, then removed interactions and variables sequentially. We conducted ANOVA on models with and with- out variables to determine whether variables improved the model fit, discarding those that did not (P > 0.05) (Blakey et al., 2016; Clarke-Wood et al., 2016). Because of a significant status × location interaction, we ran individual multivariate models for each location as well. We examined the multivariate model
coefficients for each taxon to identify which taxa contributed to differences between locations.
Table 1. Grasshopper taxa captured in 2016 and 2017. Species typi- cally associated with prairies are denoted with an asterisk.
Functional groups. We grouped grasshoppers by wing length (short, long) and by preferred diet ( grass, forb, mixed) (Table 1),
Grasshopper taxon
Unique
to Diet
Wing length
according to Pfadt (1994), Otte (1981), and Capinera and Sechr- ist (1982). We considered grasshoppers to be short-winged if the
Amphitornus coloradus
(Thomas)
- Grass Long
Arphia sulphurea (Fabricius)* - Mixed Long
species’ wings typically do not extend more than half of abdo- men length. Using the package mvabund, we created multivari- ate models of abundance for each functional group to look for differences between reconstructions and remnants as well between reconstructions of different ages. We again used only the grasshoppers collected at transects from 2016. Explanatory variables for the reconstruction versus remnant analysis included prairie status, location, and edge proximity and all possible inter- actions. We conducted ANOVA on models with and without variables to determine whether the variables improved the model fit, discarding those that did not (P > 0.05) (Blakey et al., 2016;
Arphia xanthoptera (Burmeister)* Campylacantha olivacea (Scudder)
Chortophaga viridifasciata
(DeGeer) Dichromorpha viridis (Scudder)* Dissosteira carolina (Linnaeus)
Encoptolophus sordidus
(Burmeister)*
- Grass Long
- Mixed Short
- Grass Long
- Grass Short
Reconstn Mixed Long
- Grass Long
Clarke-Wood et al., 2016). Due to a significant status × location interaction, we also modelled functional group abundance for each prairie separately.
Hesperotettix - Forbs Long Hippiscus ocelote (Saussure) Reconstn Grass Long Hypochlora alba (Dodge)* Remnant Forbs Short Melanoplus bivittatus (Say) - Mixed Long Melanoplus confusus (Scudder) Remnant Mixed Long
Does reconstruction age affect grasshopper communitieWs?e. used the same methods as those described earlier to analyse the effect of reconstruction age on grasshopper communities, with a few modifications. We modified the ordination comparing rem- nants and reconstructions by adding a heat-map of colour that
was scaled to reconstruction age. Only reconstructed transects
Melanoplus differentialis
(Thomas)
Melanoplus femurrubrum
(DeGeer) Melanoplus flavidus (Scudder)*
- Mixed Long
- Mixed Long
Remnant Forbs Long
were used in models and prairie status was replaced with recon-
Melanoplus gracilis (Bruner) - Grass Short
struction age. Only the 12 most abundant grasshopper taxa
Melanoplus inconspicuous
(Caudell)*
Remnant Forbs Short
found on reconstructions were used in the multivariate models
Melanoplus keeleri (Thomas) - Forbs Long
of age effect.
Melanoplus sanguipines
(Fabricius)
- Mixed Long
Are grasshopper communities on older reconstructions con-
Melanoplus scudderi (Uhler) - Forbs Short
Mermiria bivittata (Serville)* Remnant Grass Long
verging with remnants?. To determine if older reconstructions have converged with remnants, we also classified reconstruc- tions into two broad age classes: ≤5 years and >5 years, and per- formed the same multivariate model analyses of abundance for taxa and functional groups as described in the community com- position and functional group sections above. We repeated those analyses for the reconstruction groups split into ≤9 years and >9 years. Five and nine were chosen as breaks because they represented roughly the bottom and top third of the age range,
Orphulella pelidna (Burmeister)* Orphulella speciosa (Scudder)* Paratylotropidia brunneri (Scudder)*
Phoetaliotes nebrascensis
(Thomas)* Pseudopomala brachyptera (Scudder)*
- Grass Long
Remnant Grass Long Remnant Unknown Short
- Grass Short
Remnant Grass Short
which ensured a sufficient number of transects to perform
Schistocerca alutacea (Harris) - Forbs Long
analyses.
Did vegetation density or the forb:grass ratio affect grasshop-
Schistocerca americana
(Drury) Schistocerca obscura (Fabricius)
- Mixed Long
Remnant Mixed Long
per communities?. We averaged the three measures of vege- tation density and forb:grass ratio for transects in 2017 and compared vegetation on remnants to vegetation on reconstruc- tions with linear models (function lm in package Stats). To eval- uate relationships between vegetation characteristics and grasshopper species abundances, we ran separate models for each vegetation measurement, using the same methods as described in the community composition analysis.
Stethophyma celata (Otte)* Remnant Grass Long
Syrbula admirabilis (Uhler)* - Grass Long
Results
We collected 2435 grasshoppers, 1044 on remnants and 1391 on reconstructions, representing 33 species in 2016 and 2017
combined (Table 1). There was one grasshopper species of note. Melanoplus inconspicuous (Caudell) (Acrididae: Melanoplinae) represents, to our knowledge, a northern range expansion into Missouri. We found the short-winged, early-hatching species at Wah’Kon-Tah Prairie and Linscomb Wildlife Area.
Do communities on reconstructions differ from those on remnants?
Diversity. At the three locations where sampling effort (i.e., number of transects) was equal (Wah’Kon-Tah, Linscomb, Schell), remnant prairies had greater raw taxa richness compared to reconstructions. At the fourth location (North), where sam- pling effort was greater on the reconstructed prairie than the rem- nant, raw species richness was greater on the reconstruction than on the remnant. Rarefaction/extrapolation curves of species rich- ness indicate greater species richness on remnants compared to reconstructions at two locations, Wah’Kon-Tah and Linscomb, based on the lack of overlap in the 95% confidence interval (Supporting Information Fig. S2). Extrapolated and rarefied Shannon diversities were greater, with no overlap in 95% confi- dence intervals, on remnants than reconstructions at Wah’Kon- Tah, Linscomb, and Schell. Simpson diversity confidence intervals overlapped at all locations except Linscomb, where remnants were more diverse. This suggests that the other prairies contained similar numbers and frequencies of the most common species.
Total abundance. The final model for grasshopper abun- dance included a location × status interaction (χ2 = 30.91, P < 0.001). When abundance was modelled separately at each location, status was significant at Linscomb (χ2 = 15.09, P < 0.001) and Wah’Kon-Tah (χ2= 11.41, P < 0.001). Grasshop- per abundance was greater on reconstructions than remnants at Linscomb (Wald χ2 = 4.00, P < 0.001), while there were more grasshoppers on remnants than reconstructions at Wah’Kon- Tah (Wald χ2 = −3.37, P < 0.001).
Community composition. Remnant and reconstruction grasshopper communities from 2016 appeared distinct based on ordinations (Fig. 1; stress = 0.169, k = 3). Multivariate abun- dance models of the same species used in ordination contained a significant status × location interaction (χ2 = 122.5, P < 0.0001), indicating differences in grasshopper community composition between reconstructed and remnant prairies varied by location. When locations were analysed separately, grasshopper commu- nities on remnant and reconstruction prairies differed at Wah’- Kon-Tah (χ2 = 79.4, P < 0.001), Linscomb (χ2 = 75.76, P < 0.001), and North (χ2 = 28.43, P = 0.013) but not at Schell (χ2 = 16.53, P = 0.287). Univariate models of the abundance of two taxa, Campylacantha olivacea (Scudder) (Acridae:Gom- phocerinae) and the genus Orphulella (Acrididae: Gomphoceri- nae), contained significant status × location interactions. The multivariate model without the interaction term supported
Fig. 1. NMS ordination (k= 3) of grasshopper communities. Dots represent communities at transects. Spheres represent 95% confidence intervals around the centroids of reconstructions (black) and remnants (red). [Color figure can be viewed at wileyonlinelibrary.com]
evidence from the ordination that remnant and reconstruction
(Fig. 3, Wald χ2 = 3.09, P = 0.021; Fig. 3, Wald χ2 = 5.01,
1 1
grasshopper communities were distinct (χ2 = 87.76, P < 0.001). Communities also differed by location (χ2 = 273.6, P < 0.001). Status coefficients representing the effect of reconstruction on abundance from the multivariate model without the interaction are presented in Fig. 2. We refrained from interpreting the overall status coefficient for the two taxa responsible for the significant interaction. Melanoplus femurrubrum (DeGeer) (Acrididae: Melanoplinae), Melanoplus differentialis (Thomas) (Acrididae: Melanoplinae), Hesperotettix (Acrididae: Melanoplinae), and Syrbula admirabilis (Uhler) (Acrididae: Gomphcerinae) were all more abundant on reconstructions, while Phoetaliotes nebrascensis (Thomas) (Acrididae: Melanoplinae) and Melano- plus gracilis were more abundant on remnants (Fig. 2). The most common grasshopper across sites was Melanoplus scudderi (Uhler) (Acrididae: Melanoplinae). Among all grasshoppers captured with standardised sweeps and targeted capture over both years, 10 species were unique to remnants, and two species
were found only on reconstructions (Table 1).
Functional groups. Long-winged grasshoppers and mixed
P < 0.001). Grass-eating and short-winged grasshopper abun- dances were strongly associated with remnant prairie at Wah’- Kon-Tah, but not at the other locations (Fig. 3). At all four locations, grasshoppers with mixed diets made up greater pro- portion of total captures (15–79%) on reconstructions than on remnants in both years.
Does reconstruction age affect grasshopper communities?
Ordination (Supporting Information Fig. S3; stress = 0.169, k = 3) indicated that grasshopper communities on reconstruc- tions 5 years and younger were distinct from those over 5 years old, but communities corresponding to all other reconstruction ages appeared to have considerable overlap. The best multivari- ate abundance model contained a significant interaction between age and location (χ2 = 42.84, P = 0.009), but univariate tests found the interaction was only significant for one species, Dichromorpha viridis (Scudder) (χ2 = 11.636, P = 0.043). The model without an interaction showed that grasshopper commu-
diet grasshoppers were more abundant on reconstructions
nities differed by age (χ2
= 42.57, P < 0.001) and location
Fig. 2. Coefficients, with 95% confidence intervals, of the effect of prairie status (reconstruction or remnant) from multivariate models of the abundance of taxa used in ordination. Taxa are listed by abundance, with the most abundant at the bottom. Positive coefficients signify a greater abundance on recon- structions, negative coefficients signify greater abundance on remnants. There are up to five coefficients for each species, representing models for Wah’- Kon-Tah, North, Linscomb, Schell, and the prairies combined with no interaction (overall). Model coefficients with very large standard errors are not shown. There was a significant interaction in univariate abundance models for Campylacantha olivacea and Orphulella. [Color figure can be viewed at wileyonlinelibrary.com]
Fig. 3. Coefficients, with 95% confidence intervals, of the effect of prairie status (reconstruction or remnant) from multivariate models of the abundance of functional groups. Positive coefficients signify a greater abundance on reconstructions, negative coefficients signify greater abundance on remnants. There are up to five coefficients for each species, representing models for Wah’Kon-Tah, North, Linscomb, Schell, and the prairies combined with no interaction (overall). [Color figure can be viewed at wileyonlinelibrary.com]
(χ2 = 161.3, P < 0.001). M. femurrubrum, Melanoplus bivittatus (Say), and first instar Melanoplus (thought to be mainly
M. femurrubrum) abundances were negatively associated with reconstruction age (Supporting Information Fig. S4).
The best model for total grasshopper abundance included age
remnants. Ordination plots displaying reconstruction age indi- cated grasshopper communities on older reconstructions tended to be more similar to remnants in ordination space than younger reconstructions (Supporting Information Fig. S3). The multivar- iate models of those same taxa split at 5 and 9 years contained a
(χ2
= 13.291, P = 0.0003) and location (χ2
= 50.615,
significant interaction between age group (including remnant
1
P . Age
1
prairies) and location (χ2
= 109.1, P < 0.001; χ2
= 110.1,
< 0.0001) was negatively associated with grasshopper 1 1
abundance, meaning grasshopper abundance was lower on older reconstructions. The best model for grasshopper abundance on
P < 0.001), which once again was only significant in univariate
tests for Campylacantha olivacea (χ2 = 17.731, P = 0.031; χ2
1 1
reconstructions grouped by wing length contained age (χ2 = 16.15, P = 0.002) and location (χ2 = 59.55, P < 0.001). Long-
= 16.534, P = 0.034). The models from either age split without the interaction indicated older reconstructions were still distinct
from remnants (Wald χ2 = 7.65, P < 0.001; Wald χ2 = 6.55,
winged grasshoppers were more abundant on more recent 1 1
reconstructions overall and at the three prairies that contained reconstructions of different ages (Supporting Information Fig. S5). There was no effect of reconstruction age on the abun- dance of short-winged grasshoppers (Supporting Information Fig. S5). The best grasshopper abundance models based on diet
contained age (χ2 = 20.5, P = 0.002 and location (χ2 = 110.8,
P < 0.001) after accounting for location (χ2 = 238.4, P < 0.001; χ2 = 266.7, P < 0.001). Examining individual taxa did not yield many significant differences between reconstructions and rem- nants for either age class at either break point. The reduction in sample size substantially limited our ability to discern differ-
ences in abundance. However, M. femurrubrum was consistently
1 1
P < 0.001). Grasshoppers with mixed and grass diets were more abundant on more recent reconstructions compared to older reconstructions (Supporting Information Fig. S5).
Are grasshopper communities on older reconstructions converging with remnants?
Species that differentiated reconstructions from remnants, specifically long-winged generalists such as M. femurrubrum and M. bivittatus, were more common on younger reconstruc- tions than older reconstructions; therefore, it was important to determine whether older reconstructions were distinct from
more abundant on reconstructions than remnants in any age group. S. admirabilis appears to be a species that differentiates reconstructions of 5 or less years from remnants, whereas abun- dance of P. nebrascensis was significantly greater on remnants than on older reconstructions.
Grasshopper abundance on young reconstructions was greater than on remnants, but decreased with age. Reconstructions older than 9 years had fewer grasshoppers than remnants. Decreased grasshopper abundance on older reconstructions was partly due to fewer common generalists, specifically M. femurrubrum and
M. bivittatus. However, there must be other taxa contributing to this trend, as there were still more of those two species on older reconstructions than on remnants. Models of wing length
and diet, which contained age group (including remnant prairies) (χ2 = 30.18, P < 0.001; χ2 = 53.63, P < 0.001) and location (χ2 =
our study. Three of the locations offered convincing evidence that grasshopper communities were more diverse on remnants
1 1 1
73.18, P < 0.001; χ2 = 105.6 P < 0.001). The diet model did not detect a difference in abundance of generalist grasshoppers on older reconstructions and remnants, although the abundance on younger reconstructions did differ from remnants. Grasshoppers that preferred grass were more abundant on remnants than on older reconstructions, whereas forb-preferring grasshoppers were less common on remnants than on younger reconstructions.
Did vegetation density or the forb:grass ratio affect grasshopper communities?
There was no overall effect of status or location on forb per-
than reconstructions. The North location was the exception to a trend of greater Shannon diversity and raw species richness on remnants, which could be explained by landscape differences. Wah’Kon-Tah, Linscomb, and Schell prairies have a remnant adjacent to a reconstruction, and they are similar in terms of land- scape heterogeneity. The North location consisted of a square patch of remnant prairie bordered by an interstate highway and corn fields, and a reconstruction 32 km away consisting of alter- nating patches of reconstruction and forest, bordered by roads, cornfields, and drainages. The discrepancy in habitat heterogene- ity between remnant and reconstruction at the North location, absent at the other prairie locations, may be responsible for greater grasshopper diversity on the reconstruction. Further sup-
centage, although there was a significant interaction (F
3,125 =
port for this hypothesis comes from the two species unique to
reconstructions, both found at the North location; Hippiscus oce-
3.1881, P = 0.027) due to the reconstruction at Schell having a greater forb percentage compared to the remnant. Reconstruc- tions had greater vegetation density than remnants (t65 = 4.831, P < 0.0001), and density differed by location as well (F3,125 = 5.3377, P = 0.0017). Vegetation density and forb per- centage were not significantly associated with total abundance of grasshoppers (χ22 = 0.433, P = 0.80). Vegetation measure- ments were not related to the abundance of grasshoppers in diet groups or wing length categories.
Discussion
Tallgrass prairies have been degraded more than any other eco- system in North America (Samson & Knopf, 1994). Those attempting to restore tallgrass prairies can plant native vegetation and monitor the plant community over time, adjusting seed mixes and management practices (Rowe, 2010); however, ani- mal communities must independently colonise restored habitat from other areas, and therefore must be monitored in order to determine if reconstructions provide the appropriate habitat. Invertebrates make up a large portion of the tallgrass prairie com- munity and may require very specific habitat characteristics typ- ically only found on remnant prairies (Opler, 1981).
Our first objective was to compare the grasshopper communi- ties of reconstructions and remnants and identify species or func- tional groups that can be used to evaluate prairie reconstruction progress. Grasshopper communities on remnant and reconstructed prairies were distinct, and remnants appeared to contain more spe- cies than reconstructions. Of the ten species only found on rem- nants, eight of them are known to be associated with prairie habitat (Pfadt, 1994; Reed, 1996; Kirk & Bomar, 2003). The two other species, Paratylotropidia brunneri (Scudder) and Schis- tocerca obscura (Fabricius), are associated to some degree with field edges or woodlands. The presence of prairie-associated spe- cies found on the remnants, but not on the reconstructions, pro- vides some evidence that remnant prairies host a more diverse community of grasshoppers because reconstructions lack certain prairie habitat characteristics which deserve further investigation. Landscape heterogeneity may be responsible for the one example of greater grasshopper diversity on reconstructions in
lote (Saussure) prefers open patches within woodlands (Brust et al., 2014), and Dissosteira. carolina (Linnaeus) prefers crop field edges and disturbed sites (Pfadt, 1994). Those habitats were much more common at the North reconstruction than the remnant.
Grasshopper communities on reconstructions and remnants differed in the abundance of functional groups. Three long- winged generalists (mixed diet) were more abundant on recon- structions than on remnants (M. femurrubrum, M. bivittatus, and M. differentialis), which supports our hypothesis that reconstructions would be characterised by highly mobile gen- eralists. Two short-winged species (P. nebrascensis,
M. gracilis) with a diet preference for grass or forbs were more abundant on remnants. This supports our hypothesis that rem- nants would be characterised by more sedentary specialists, which aligns with two studies that showed that generalist grasshoppers tend to dominate newly created grassland habi- tats (Bomar, 2001; Picaud & Petit, 2008). A comparison of reconstructed and remnant prairies in Wisconsin also found that M. femurrubrum was a dominant species on reconstruc- tions (Bomar, 2001).
Our second objective was to evaluate the effect of reconstruc- tion age on grasshopper communities. Older reconstructions, although still distinct from remnants, were closer in community composition to remnants than young reconstructions and many of the functional groups and taxa that identified recon- structions decreased in abundance with reconstruction age.
M. femurrubrum decreased in abundance with reconstruction age but remained more common on older reconstructions than remnants. Bomar (2001) sampled from reconstructions as old as 50 years and found that M. femurrubrum remained the domi- nant grasshopper on reconstructions. We are cautious in inter- preting the effects of reconstruction age because reconstruction and management practices have changed over the last 20 years, including seeding a greater diversity and ratio of forb seeds in comparison to grasses, which could affect how grasshopper communities changed over time on reconstructions of different ages. Nevertheless, it appears that grasshopper communities on reconstructed prairies are initially dominated by long-winged generalists and that over time those taxa diminish in numbers until communities are similar to remnants. However, at least on
reconstructions <15 years of age, grasshoppers associated with remnants do not replace the early colonisers.
Differences in the diversity and community composition of prairie grasshoppers between remnants and reconstructions have important conservation implications. Eight prairie grasshopper species associated with prairies were not found on reconstruc- tions, and thus reconstructions may not be able to bolster the small grasshopper populations of nearby remnants. There are two, non-mutually exclusive explanations for this finding: grass- hopper species on remnants cannot disperse to the new habitats or they do not survive and reproduce on reconstructions if they are able to reach them (Peterson & Denno, 1998; Beck & Kitching, 2007). Seven of the 10 grasshoppers unique to rem- nants were long-winged and probably capable of dispersing the short distances between remnants and reconstructions at the three locations in the Osage Plains. This suggests that dispersal ability is not the sole reason that reconstructions were less diverse than remnants.
Habitat suitability differences between remnants and recon- structions likely contributed to the differing grasshopper com- munities. We investigated some possible habitat characteristics responsible for the difference in communities by measuring forb:grass ratio and vegetation density. However, neither vegeta- tion metric was a significant predictor of grasshopper abundance. Other studies have also failed to find consistent links between grasshopper populations and environmental variables (Anderson, 1964; Hastings & Pepper, 1964; Evans, 1988). Other characteristics, including plant community composition and soil properties, could impact grasshopper community composition on reconstructions and remnants. Plant community composition is likely a contributing factor, as grasshopper species’ diets vary and reconstructions are known to differ from remnants in plant composition and structure (Kindscher & Tieszen, 1998; Olech- nowski et al., 2009). Soil characteristics could also have impacted grasshopper community composition. Grasshoppers deposit their eggs in the soil, and many require fairly specific soil conditions or temperatures for egg deposition (Uvarov, 1966). Plant and soil characteristics may also influence interspecific competition among grasshoppers, which is commonly size- dependent (Belovsky, 1986; Whitman, 2008). Reconstructions that support large populations of generalist, long-winged grass- hoppers, which in this study consisted of two large species,
M. differentialis and M. bivittatus, and one midsize species,
M. femurrubrum, may not be supportive habitat for midsize grasshoppers due to direct competition. From a conservation per- spective, making reconstructions hospitable to rare grasshoppers may be more important than attempting to aid species’ dispersal from isolated habitat patches.
Our study produced results that can aid future reconstruction projects and grasshopper conservation. Monitoring can be more effective by focusing on certain taxa, such as the 10 prairie rem- nant species not found on reconstructions, as well as species that were simply less abundant on reconstructions, such as
P. nebrascensis. All of those species warrant further monitoring and investigation into their habitat requirements to determine why they were not abundant on reconstructions. The long- winged generalist species that were very abundant on reconstruc- tions could be influencing the reconstruction process itself and
their impact should be investigated. Additionally, it is important that grasshopper monitoring efforts continue to incorporate a community level perspective. Monitoring that only focuses on certain species suggested by previous studies as particularly important can lead to inconsistent findings because grasshopper populations are dynamic and exhibit frequent shifts in the dom- inant species (Campbell et al., 1974; Capinera & Thompson, 1987; Evans, 1988). Continued monitoring will be necessary to determine if grasshopper communities on reconstructions even- tually align with those on remnants.
Acknowledgements
The Missouri Department of Conservation, Wildlife Division was the primary funder of this project. Prairie Fork Charitable Endowment Trust and Prairie Biotic also provided grant support. We thank Paul Lenhart for assistance in identification and Rachel Blakey for advice on analyses. The authors have no conflicts of interest. The Missouri Cooperative Fish and Wildlife Research Unit is jointly sponsored by the MDC, the University of Missouri, the U.S. Fish and Wildlife Service, the U.S. Geological Survey, and the Wildlife Management Insti- tute. Any use of trade, firm, or product names is for descriptive purposes only and does not imply endorsement by the
U.S. Government.
Supporting information
Additional supporting information may be found online in the Supporting Information section at the end of the article.
Appendix S1. Supporting Information.
References
Alignan, J.-F., Debras, J.-F. & Dutoit, T. (2018) Orthoptera prove good indicators of grassland rehabilitation success in the first French Natu- ral Asset Reserve. Journal for Nature Conservation, 44, 1–11.
Anderson, N.L. (1964) Some relationships between grasshoppers and vegetation. Annals of the Entomological Society of America, 57, 736–742.
Ballard, H.E. (1992) Keys to known and potential Missouri Orthoptera. The Nature Conservancy.
Bazelet, C.S. & Samways, M.J.J.E.I. (2011) Identifying grasshopper bioindicators for habitat quality assessment of ecological networks. Ecological Indicators, 11, 1259–1269.
Beck, J. & Kitching, I.J. (2007) Correlates of range size and dispersal ability: a comparative analysis of sphingid moths from the Indo-Australian tropics. Global Ecology and Biogeography, 16, 341–349.
Belovsky, G.E. (1986) Generalist herbivore foraging and its role in com- petitive interactions. American Zoologist, 26, 51–69.
Benayas, J.M.R., Newton, A.C., Diaz, A. & Bullock, J.M. (2009) Enhancement of biodiversity and ecosystem services by ecological restoration: a meta-analysis. Science, 325, 1121–1124.
Benkobi, L., Uresk, D.W., Schenbeck, G. & King, R.M. (2000) Protocol for monitoring standing crop in grasslands using visual obstruction. Journal of Range Management, 53, 627–633.
Blakey, R.V., Law, B.S., Kingsford, R.T., Stoklosa, J., Tap, P., Williamson, K. & Minderman, J. (2016) Bat communities respond positively to large-scale thinning of forest regrowth. Journal of Applied Ecology, 53, 1694–1703.
Block, W.M., Franklin, A.B., Ward, J.P., Ganey, J.L. & White, G.C. (2001) Design and implementation of monitoring studies to evaluate the success of ecological restoration on wildlife. Restoration Ecology, 9, 293–303.
Bomar, C.R. (2001) Comparison of grasshopper (Orthoptera: Acrididae) communities on remnant and reconstructed prairies in western Wis- consin. Journal of Orthoptera Research, 10, 105–112.
Brennan, L.A. & Kuvlesky, W.P., Jr. (2005) North American grassland birds: an unfolding conservation crisis? Journal of Wildlife Manage- ment, 69, 1–13.
Brückmann, S.V., Krauss, J. & Steffan-Dewenter, I. (2010) Butterfly and plant specialists suffer from reduced connectivity in fragmented land- scapes. Journal of Applied Ecology, 47, 799–809.
Brust, M., Thurman, J., Reuter, C., Black, L., Quartarone, R., & Redford, A.J. (2014) Grasshoppers of the Western U.S. USDA APHIS Identfication Technology Program (ITP).
Campbell, J.B., Arnett, W.H., Lambley, J., Jantz, O.K. & Knutson, H. (1974) Grasshoppers (Acrididae) of the Flint Hills native tallgrass prairie in Kansas. Kansas State Agricultural Experiment Station Research Paper, 19(147).
Capinera, J.L., & Sechrist, T.S. (1982) Grasshoppers (Acrididae) of Col- orado: Identification, Biology and Management. Colorado Agricul- tural Experiment Station. Fort Collins, CO.
Capinera, J.L. & Thompson, D.C. (1987) Dynamics and structure of grasshopper assemblages in shortgrass prairie. The Canadian Ento- mologist, 119, 567–575.
Ceballos, G., Davidson, A., List, R., Pacheco, J., Manzano-Fischer, P., Santos-Barrera, G. & Cruzado, J. (2010) Rapid decline of a grassland system and its ecological and conservation implications. PLoS One, 5, e8562.
Chao, A., Gotelli, N.J., Hsieh, T.C., Sander, E.L., Ma, K.H., Colwell, R.
K. & Ellison, A.M. (2014) Rarefaction and extrapolation with Hill numbers: a framework for sampling and estimation in species diversity studies. Ecological Monographs, 84, 45–67.
Christisen, D.M. (1972) Prairie preservation in Missouri. Proceedings of the Third Midwest Prairie Conference (ed. by (ed. by L.C. Hulbert), pp. p. 42–46. Division of Biology, Kansas State University, Manhattan, KS.
Clarke-Wood, B.K., Jenkins, K.M., Law, B.S. & Blakey, R.V. (2016) The ecological response of insectivorous bats to coastal lagoon degra- dation. Biological Conservation, 202, 10–19.
Colwell, R.K., Chao, A., Gotelli, N.J., Lin, S.Y., Mao, C.X., Chazdon, R.
L. & Longino, J.T. (2012) Models and estimators linking individual- based and sample-based rarefaction, extrapolation and comparison of assemblages. Journal of Plant Ecology, 5, 3–21.
Davis, J.D., Hendrix, S.D., Debinski, D.M. & Hemsley, C.J. (2007) But- terfly, bee and forb community composition and cross-taxon incongru- ence in tallgrass prairie fragments. Journal of Insect Conservation, 12, 69–79.
Dennis, R.L., Dapporto, L., Fattorini, S. & Cook, L.M. (2011) The generalism–specialism debate: the role of generalists in the life and death of species. Biological Journal of the Linnean Society, 104, 725–737.
Diepenbrock, L.M., Finke, D.L., Stewart, A. & Littlewood, N. (2013) Refuge for native lady beetles (Coccinellidae) in perennial grassland habitats. Insect Conservation and Diversity, 6, 671–679.
Dobson, A.P., Bradshaw, A. & Baker, A.á. (1997) Hopes for the future: restoration ecology and conservation biology. Science, 277, 515–522. Evans, E.W. (1988) Community dynamics of prairie grasshoppers sub- jected to periodic fire: predictable trajectories or random walks in
time? Oikos, 52, 283–292.
Evans, E.W., Rogers, R.A. & Opfermann, D.J. (1983) Sampling grass- hoppers on burned and unbruned tallgrass prairie: night trappin vs. sweeping. Environmental Entomology, 12, 1449–1454.
Gotelli, N.J. & Colwell, R.K. (2011) Estimating species richness. Biolog- ical Diversity: Frontiers in Measurement and Assessment (ed. by (ed. by A.E. Magurran and B. McGill) , pp. p. 39–54. Oxford Univer- sity Press, New York.
Griffin, S.R., Bruninga-Socolar, B., Kerr, M.A., Gibbs, J. & Winfree, R. (2017) Wild bee community change over a 26-year chronosequence of restored tallgrass prairie. Restoration Ecology, 25, 650–660.
Harmon-Threatt, A.N. & Hendrix, S.D. (2015) Prairie restorations and bees: the potential ability of seed mixes to foster native bee communi- ties. Basic and Applied Ecology, 16, 64–72.
Harrison, X.A., Donaldson, L., Correa-Cano, M.E., Evans, J., Fisher, D. N., Goodwin, C.E., Robinson, B.S., Hodgson, D.J. & Inger, R.J.P. (2018) A brief introduction to mixed effects modelling and multi- model inference in ecology. PeerJ, 6, e4794.
Hastings, E. & Pepper, J. (1964) Population studies on the big-headed grasshopper Aulocara elliotti. Annals of the Entomological Society of America, 57, 216–220.
Heidinger, I.M.M., Hein, S. & Bonte, D. (2010) Patch connectivity and sand dynamics affect dispersal-related morphology of the blue-winged grasshopper Oedipoda caerulescensin coastal grey dunes. Insect Con- servation and Diversity, 3, 205–212.
Hilliard, J.R., Jr. (2001) Two new grasshopper species in the Texanus group of the genus Melanoplus (Orthoptera: Acrididae: Melanoplinae) with biological notes on the group. Transactions of the American Entomological Society, 127, 31–68.
Hjermann, D.O. & Ims, R.A. (1996) Landscape ecology of the wart-biter Decticus verrucivorus in a patchy landscape. Journal of Animal Ecol- ogy, 65, 768–780.
Hsieh, T.C., Ma, K.H., Chao, A. & McInerny, G. (2016) iNEXT: an R package for rarefaction and extrapolation of species diversity (Hill numbers). Methods in Ecology and Evolution, 7, 1451–1456.
Humbert, J.Y., Ghazoul, J., Sauter, G.J. & Walter, T. (2010) Impact of different meadow mowing techniques on field invertebrates. Journal of Applied Entomology, 134, 592–599.
Joern, A. (1979) Feeding patterns in grasshoppers (Orthoptera: Acridi- dae): factors influencing diet specialization. Oecologia, 38, 325–347.
Keylock, C. (2005) Simpson diversity and the Shannon–Wiener index as special cases of a generalized entropy. Oikos, 109, 203–207.
Kindscher, K. & Tieszen, L.L. (1998) Floristic and soil organic matter changes after five and thirty-five years of native tallgrass prairie resto- ration. Restoration Ecology, 6, 181–196.
Kirk, K. & Bomar, C.R. (2003) Guide to the Grasshoppers of Wisconsin. Wisconsin Department of Natural Resources, Madison, WI.
Knowles, L.L. (2001) Genealogical portraits of speciation in montane grasshoppers (genus Melanoplus) from the sky islands of the Rocky Mountains. Proceedings of the Royal Society of London B: Biological Sciences, 268, 319–324.
Kremen, C., Merenlender, A.M. & Murphy, D.D. (1994) Ecological monitoring: a vital need for integrated conservation and development programs in the tropics. Conservation Biology, 8, 388–397.
Kuefler, D., Hudgens, B., Haddad, N.M., Morris, W.F. & Thurgate, N. (2010) The conflicting role of matrix habitats as conduits and barriers for dispersal. Ecology, 91, 944–950.
Kurtz, C. (2013) A Practical Guide to Prairie Reconstruction. University of Iowa Press, Iowa City, IA.
Larson, D.P., O’Neill, K.M. & Kemp, W.P. (1999) Evaluation of the accuracy of sweep sampling in determining grasshopper (Orthoptera
: Acrididae) community composition. Journal of Agricultural and Urban Entomology, 16, 207–214.
Lester, S.E., Ruttenberg, B.I., Gaines, S.D. & Kinlan, B.P. (2007) The relationship between dispersal ability and geographic range size. Ecol- ogy Letters, 10, 745–758.
Lockwood, J.A. (2004) Locust: The Devastating Rise and Mysterious Disappearance of the Insect that Shaped the American Frontier. Basic Books, New York, NY.
Lockwood, J.A., Kemp, W.P. & Onsager, J.A. (1988) Long-term, large-scale effects of insecticidal control on rangeland grasshopper populations (Orthoptera: Acrididae). Journal of Economic Entomology, 81, 1258–1264. Marini, L., Bommarco, R., Fontana, P. & Battisti, A. (2010) Disentangling effects of habitat diversity and area on orthopteran species with contrast-
ing mobility. Biological Conservation, 143, 2164–2171. Marini, L., Ockinger, E., Battisti, A. & Bommarco, R. (2012) High mobil-
ity reduces beta-diversity among orthopteran communities - implica- tions for conservation. Insect Conservation and Diversity, 5, 37–45.
Missouri Natural Heritage Program (2018) Missouri Species and Com- munities of Conservation Concern Checklist. Missouri Department of Conservation, Jefferson City, MO.
Oksanen, J., Blanchet, F., Kindt, R., & Legendre, P. (2016) Vegan: Com- munity Ecology Package. R Package Version 2.0-5.
Olechnowski, B.F., Debinski, D.M., Drobney, P.B., Viste-Sparkman, K. & Reed, W.T. (2009) Changes in vegetation structure through rime in a restored tallgrass prairie ecosystem and implications for avian diversity and community composition. Ecological Restoration, 27, 449–457.
O’Neill, K.M., Larson, D.P. & Kemp, W.P. (2002) Sweep sampling technique affects estimates of the relative abundance and community composition of grasshoppers (Orthoptera : Acrididae). Journal of Agricultural and Urban Entomology, 19, 125–131.
Opler, P.A. (1981) Management of prairie habitats for insect conserva- tion. Journal of the Natural Areas Association, 1, 3–6.
Ortego, J., Aguirre, M.P., Noguerales, V. & Cordero, P.J. (2015) Conse- quences of extensive habitat fragmentation in landscape-level patterns of genetic diversity and structure in the Mediterranean esparto grass- hopper. Evolutionary Applications, 8, 621–632.
Otte, D. (1981) The North American Grasshoppers: Acrididae: Oedipo- dinae. Harvard University Press, Cambridge, MA.
Otte, D. (2012) Eighty new melanoplus species from the United States (Acrididae: Melanoplinae). Transactions of the American Entomolog- ical Society, 138, 73–167.
Panzer, R. (2002) Compatibility of prescribed burning with the conserva- tion of insects in small, isolated prairie reserves. Conservation Biol- ogy, 16, 1296–1307.
Paton, D., Rogers, D., Hill, B., Bailey, C. & Ziembicki, M. (2009) Tem- poral changes to spatially stratified waterbird communities of the Coorong, South Australia: implications for the management of heter- ogenous wetlands. Animal Conservation, 12, 408–417.
Peet, R.K. (1975) Relative diversity indices. Ecology, 56, 496–498.
Peterson, M.A. & Denno, R.F. (1998) The influence of dispersal and diet breadth on patterns of genetic isolation by distance in phytophagous insects. The American Naturalist, 152, 428–446.
Pfadt, R.E. (1994) Field Guide to Common Western Grasshoppers. Uni- versity of Wyoming, Laramie, WY.
Picaud, F. & Petit, D.P. (2008) Body size, sexual dimorphism and eco- logical succession in grasshoppers. Journal of Orthoptera Research, 17, 177–181.
Piechnik, D.A., Lawler, S.P. & Martinez, N.D. (2008) Food-web assem- bly during a classic biogeographic study: species’ “trophic breadth” corresponds to colonization order. Oikos, 117, 665–674.
Pleasants, J.M. & Oberhauser, K.S. (2013) Milkweed loss in agricultural fields because of herbicide use: effect on the monarch butterfly popu- lation. Insect Conservation and Diversity, 6, 135–144.
R Core Team (2017) R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria.
Final Report to the USFWS Cooperating Agencies. Minnesota Depart- ment of Natural Resources, St. Paul, MN.
Reinhardt, K., Kohler, G., Maas, S. & Detzel, P. (2005) Low dispersal ability and habitat specificity promote extinctions in rare but not in widespread species: the Orthoptera of Germany. Ecography, 28, 593–602.
Ries, L., Debinski, D.M. & Wieland, M.L. (2001) Conservation value of roadside prairie restoration to butterfly communities. Conservation Biology, 15, 401–411.
Routledge, R. (1979) Diversity indices: which ones are admissible?
Journal of Theoretical Biology, 76, 503–515.
Rowe, H.I. (2010) Tricks of the trade: techniques and opinions from 38 experts in tallgrass prairie restoration. Restoration Ecology, 18, 253–262.
Samson, F. & Knopf, F. (1994) Prairie conservation in north america.
BioScience, 44, 418–421.
Sauberer, N., Zulka, K.P., Abensperg-Traun, M., Berg, H.-M., Bieringer, G., Milasowszky, N., Moser, D., Plutzar, C., Pollheimer, M. & Storch, C. (2004) Surrogate taxa for biodiversity in agricultural landscapes of eastern Austria. Biological Conservation, 117, 181–190.
Schultz, C.B., Russell, C. & Wynn, L. (2008) Restoration, reintroduc- tion, and captive propagation for at-risk butterflies: a review of British and American conservation efforts. Israel Journal of Ecology and Evolution, 54, 41–61.
Smith, D. (2010) The Tallgrass Prairie Center Guide to Prairie Restoration in the Upper Midwest. University of Iowa Press, Iowa City, USA.
Solecki, M. & Toney, T. (1986) Characteristics and management of Mis- souri’s public prairies. Proceedings of the North American Prairie Conference, 9, 168–171.
Song, H. (2009) Taxonomic Identification Key to Schistocerca Species.
pared to other conservation managements of open habitat. Biodiver- sity & Conservation, 10, 1141–1169.
Thom, R.M. (2000) Adaptive management of coastal ecosystem restora- tion projects. Ecological Engineering, 15, 365–372.
Thompson, J.D. (1999) Population differentiation in Mediterranean plants: insights into colonization history and the evolution and conser- vation of endemic species. Heredity, 82, 229–236.
Tonietto, R.K., Ascher, J.S. & Larkin, D.J. (2017) Bee communities along a prairie restoration chronosequence: similar abundance and diversity, distinct composition. Ecological Applications, 27, 705–717.
Török, P. & Helm, A. (2017) Ecological theory provides strong support for habitat restoration. Biological Conservation, 206, 85–91.
Uresk, D.W. & Benzon, T.A. (2007) Monitoring with a modified Robel pole on meadows in the central Black Hills of South Dakota. Western North American Naturalist, 67, 46–50.
Uvarov, B.P. (1966) Grasshoppers and Locusts: A Handbook of General Acridology , Vol. 1. Cambridge University Anti-Locust Research Cen- tre, Cambridge, UK.
Venables, W.N. & Ripley, B.D. (2002) Modern Applied Statistics with S. Springer, New York.
Vogel, J.A., Debinski, D.M., Koford, R.R. & Miller, J.R. (2007) Butter- fly responses to prairie restoration through fire and grazing. Biological Conservation, 140, 78–90.
Wang, Y., Naumann, U., Wright, S.T., Warton, D.I. & J.M.i.E., & Evolu- tion (2012) mvabund–an R package for model-based analysis of multi- variate abundance data. Methods in Ecology and Evolution, 3, 471–474. Weibull, A.-C., Östman, Ö. & Granqvist, Å. (2003) Species richness in agroecosystems: the effect of landscape, habitat and farm manage-
ment. Biodiversity & Conservation, 12, 1335–1355.
Whitman, D.W. (2008) The significance of body size in the Orthoptera: a review. Journal of Orthoptera Research, 17, 117–134.
Williams, N.M. (2011) Restoration of nontarget species: bee communi- ties and pollination function in riparian forests. Restoration Ecology, 19, 450–459.