Grassland restoration is largely focused on creating plant communities that match reference conditions. However, these communities reflect only a subset of the biodiversity of grassland systems. We conducted a multi-trophic study to assess ecosystem recovery following energy development for oil and gas extraction in northern U.S. Great Plains rangelands. We compared soil factors, plant species composition and cover, and nematode trophic structuring between reclaimed oil and gas well sites ("reclaims") that comprise a chronosequence of two – 33 years since reclamation and adjacent, undeveloped rangeland at distances of 50 m and 150 m from reclaim edges. Soils and plant communities in reclaims did not match those on undeveloped rangeland even after 33 years. Reclaimed soils had higher salt concentrations and pH than undeveloped soils. Reclaims had lower overall plant cover, a greater proportion of exotic and ruderal plant cover and lower native plant species richness than undeveloped rangeland. However, nematode communities appear to have recovered following reclamation. Although total and omni-carnivorous nematode abundances differed between reclaimed well sites and undeveloped rangeland, community composition and structure did not. These findings suggest that current reclamation practices recover the functional composition of nematode communities, but not soil conditions or plant communities. Our results show that plant communities have failed to recover through reclamation: high soil salinity may create a persistent impediment to native plant growth and ecosystem recovery.

Description: This dataset consists of field data (arthropods, nematodes and NDVI) collected over the course of 6 field excursions in 2015 and 2016 near TyTy, GA, in a field used for growing Miscanthus x giganteus. It also includes interpolated values of soil measurements collected in 2015 and meteorological data collected on an adjacent farm. Point-in-time measurements include all meteorological, NDVI, arthropod and nematode measurements and their derivatives. Fixed values were measurements that were held constant across all sampling dates, including location, terrain and soils measurements and their derivatives. Dawn Olson and Jason Schmidt collected and processed arthropod count data. Jason Schmidt collected and processed spider count data and computed spider diversity. Richard Davis collected and processed nematode count data. Alisa Coffin collected and processed NDVI data and positional locations. Tim Strickland collected and processed soils data and Alisa Coffin interpolated soils values using kriging to derive values at arthropod sample locations. David Bosch collected and processed meteorological data. Lynne Seymour provided statistical expertise in deriving any estimated values (phloem feeders, parasitoids, spiders, and natural enemies). Alisa Coffin derived terrain data (elevation, slope, aspect, and distances) from publicly available datasets, transformed values (SI, WI, etc), carried out the geographically weighted regression analysis and calculated C:SE values, harmonized the full dataset, and compiled it using Esri's ArcGIS Pro 2.5. Methods for most data are published in the accompanying paper and associated supplements. Questions about dataset development and management should be directed to Alisa Coffin (alisa.coffin@usda.gov). This work was accomplished as a joint USDA and University of Georgia project funded by a cooperative agreement (#6048-13000-026-21S). This research was a contribution from the Long-Term Agroecosystem Research (LTAR) network. LTAR is supported by the United States Department of Agriculture. At request of the author, the data resources are under embargo. The embargo will expire on Fri, Jan 01, 2021.