Tallgrass prairies and their obligate inhabitants once occupied a large swath of central North America, but now face the combined challenges of habitat loss and fragmentation. In Missouri, several hundred hectares of tallgrass prairie have been restored near tracts of remnant native prairie. Typically, the success of reconstructed and restored grasslands are gauged on the extent to which native prairie plants have reestablished. Invertebrates are often assumed, or at least hoped, to colonize reconstructions if native vegetation returns. However, the limited mobility of many invertebrates and the isolation of many tallgrass remnants raises serious doubts as to whether prairie invertebrate communities in reconstructed prairies actually match those in remnants. To evaluate the effectiveness of prairie reconstructions in restoring grassland invertebrate communities, we sampled two guilds of terrestrial invertebrates: native bees (superfamily Apoidea) and grasshoppers (Acrididae). Both guilds are relatively well described and include highly conservative grassland species. There are pollen specialists and rare kleptoparasitic bees found thus far only in remnant prairies in Missouri. The presence those bees, and of several species of small grasshoppers with limited mobility, on reconstructions would suggest a successful prairie reconstruction. We hypothesize that the bee and grasshopper communities in reconstructions will be characterized by common, generalist species with higher mobility that perform well in an agricultural matrix.
We sampled invertebrates from five conservation areas in Missouri containing tallgrass prairie. Three areas were located in southwestern Missouri and contained remnants adjacent to reconstructions. The remaining two areas, in central Missouri, consisted of one remnant and one reconstruction. We randomly generated sampling locations in ArcMAP 10, creating 134 locations over the five areas. We collected bees with bee bowls, which were placed in the field for 48 hours. Grasshoppers were captured with a sweep net; each sampling location was swept 40 times. Bee and grasshopper sampling occurred three times from June to September 2016. Preliminary community analyses using NMS were not significant, but they suggest that remnant and reconstructions may differ in composition, possibly due to a stark difference in abundance of sweat bees in the genus Augochlorella. We did not detect differences in species richness between remnant and restored sites. However, evenness likely differs between the two types of prairies, and rare species appear to be more common in prairie remnants. For bees in particular, pollen specialists and kleptoparasites may be less capable of colonizing and surviving in reconstructed prairies.