Seed Reserves Diluted During Surface Soil Reclamation in Eastern Mojave Desert

Abstract

Abstract Surface soil reclamation is used to increase the re-establishment of native vegetation following disturbance through preservation and eventual replacement of the indigenous seed reserves. Employed widely in the mining industry, soil reclamation has had variable success in re-establishing native vegetation in arid and semi-arid regions. We tested whether variable success could be due in part to a decrease of seed reserves during the reclamation process by measuring the change in abundance of germinable seed when surface soil was mechanically collected, stored in a soil pile for 4 months, and reapplied upon completion of a roadway. Overall seed reserve declines amounted to 86% of the original germinable seed in the soil. The greatest decrease in seed reserves occurred during soil collection (79% of original reserves), compared to the storage and reapplication stages. At nearby sites where stored surface soil had been reapplied, no perennial plant cover occurred from 0.5 to 5 years after application and < 1% cover after 7 years compared to 5% cover in nearby undisturbed areas. The reduction in abundance of germinable seed during reclamation was primarily due to dilution of seed reserves when deeper soil fractions without seed were mixed with the surface soil during collection. Unless more precise techniques of surface soil collection are utilized, soil reclamation alone as a means for preserving native seed reserves is a method ill-suited for revegetating disturbed soils with a shallow seed bank, such as those found in the Mojave Desert. Keywords: lake meadrestorationrevegetationseed bank Funding was provided by the United States Geological Survey (USGS), Western Ecological Research Center, and in-kind contributions from Lake Mead National Recreation Area (LMNRA), including all costs and logistics associated with the removal, storage, and application of surface soil. We thank Alice Newton, Dona LeNoue, Shirley Yoon, and Sarah Stevens with the National Park Service for their assistance during all phases of this research project. We also thank Sara Eckert, Todd Esque, Kim Goodwin, Dustin Haines, and Jason Williams of the USGS for their help in the field and greenhouse. Scott Wolfert, project engineer from the Central Federal Lands Highway Division of the Federal Highway Administration, worked to accommodate our sampling and design needs. Catherine Bukowski, Elizabeth Burgieres, and Nira Salant from the Student Conservation Association (SCA) assisted with data collection. Julie Yee provided valuable statistical advice, and the manuscript was strengthened by the comments of two anonymous reviewers. Notes The significance of the rank sum multiple comparisons are based on a Bonferroni's adjustment of α = 0.05. Ranks that do not share the same lowercase superscripts are significantly different from each other.