Re-analysis of land used for energy: Comparison of spatially explicit observations and literature sources

• Literature was compiled using a systematic methodology. • Land transformed for energy is compared between literature and geospatial data. • Geospatial land transformation values are higher than literature for total life cycle of natural gas, oil, nuclear, coal, solar, and wind. • Literature recycles common data points, which could lead to error propagation. Energy production has many life cycles, each requiring expansive infrastructure and a significant spatial footprint in the landscape. As energy systems expand and technologies transition from non-renewable to renewable energy sources, it is imperative to accurately quantify the amount of land needed. Since life cycles of different technologies may require very different conversion of land surfaces, land transformation estimates can provide a standardized measure of the efficiency of an energy technology. Although there is an abundance of existing literature, spatial footprint estimates vary substantially among technologies and life-cycle stages. These varied sources could benefit from a standardized comparison and validation using a comprehensive and consistent ground-truth assessment. The National Water Energy Land Dataset (NWELD) provides comprehensive and spatially explicit mapping of land used for energy technology. Therefore, we present a methodological re-analysis of land used for energy by comparing spatially explicit observations from NWELD to coefficients found in literature for specific fuels and life cycles. Literature was compiled using a systematic methodology, filtered to collect pertinent data values, and summarized. NWELD land requirements were calculated and coupled with U.S. Energy Information Administration (EIA) data to determine the energy production per technology. Our results suggest that the total life cycle of NWELD’s natural gas, oil, nuclear, and coal have higher median land footprints than what is reported in literature, except for biomass. Furthermore, we find that literature resources recycle common data points, which if inaccurate, could lead to error propagation in estimating land used for energy.