An updated isoerodent map of the conterminous United States

Authors: MCGEHEE, RYAN - Purdue University; Flanagan, Dennis; SRIVASTAVA, PUNEET - University Of Maryland; ENGEL, BERNARD - Purdue University; Huang, Chi Hua; Nearing, Mark

Submitted to: International Soil and Water Conservation Research
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 6/7/2021
Publication Date: 6/10/2021
Citation: McGehee, R.P., Flanagan, D.C., Srivastava, P., Engel, B.A., Huang, C., Nearing, M.A. 2021. An updated isoerodent map of the conterminous United States. International Soil and Water Conservation Research. https://doi.org/10.1016/j.iswcr.2021.06.004.
DOI: https://doi.org/10.1016/j.iswcr.2021.06.004

Interpretive Summary: Soil erosion by water on cropland is largely driven by rainfall intensity and the energy of raindrops hitting the soil surface. Some of the most fundamental equations used to predict the amount of soil loss are based upon the product of the raindrop energy (E) and the maximum 30-minute intensity (I30) of a rainstorm. These storm “erosivity” values can be summed over a year or multiple years to come up with annual or average annual erosivity values, also known as “R” in the Universal Soil Loss Equation (USLE). When the USLE was first developed back in the 1950s, very detailed analog rainfall intensity data were available to determine the storm and R-factor values. However, over time, weather stations switched to digital discrete interval rainfall reporting, with 15-minute rainfall data commonly reported today. While information for thousands of locations are widely available in the U.S., the 15-minute data can result in errors and underprediction of erosivity and R factors, because actual peak rainfall intensities that might occur are depressed when rainfall depths are only reported for 15 minutes. In this research we utilized modern data from 1970-2013 from more than 3000 weather stations and advanced gap-filling and interpolation techniques to produce new high quality nationwide erosivity maps. Results were compared to existing R-factor maps, and disparities discussed. Substantial differences are apparent when comparing the new maps to the existing ones. These results are critically important when applying USLE technologies, and impact field soil conservation personnel, scientists, university faculty, students, extension agents, farmers, and others. Further research and additional refinement of erosivity maps within the U.S. are warranted.

Technical Abstract: Maps of erosivity, which are also commonly referred to as isoerodent maps, have played a critical role in soil conservation efforts in the United States and around the world. Currently available erosivity maps are either outdated, conflict with erosivity values from benchmarking studies, or used less advanced spatial mapping methods. In this study, we used more than 3,400 15-minute, fixed-interval precipitation gauges to update the isoerodent map of the conterminous United States. Erosivity values were interpolated using universal kriging under several spatial model configurations and resolutions. The optimal spatial model was selected based on which result had the lowest sample variogram error. Rainfall, erosivity, and erosivity density maps were compared to existing products. Some average annual and annual erosivity results were compared to high-quality erosivity benchmarking publications. Erosivity values from both RUSLE2 and Panagos et al. (2017) were generally lower in the eastern United States and mixed in the western United States as compared to results from this study. Topographic effects resulted in much greater erosivity differences in this study as compared to prior maps in the West United States. Benchmark comparisons revealed that erosivity maps from this study and others were lower than the benchmark by 14% or more (up to 38%). These findings suggest that current practices of storm omission and intensity dampening correction may need to be revisited, especially in locations with relatively moderate or low rainfall erosivity such as the Midwest or Northeast United States.