Abiotic processes are insufficient for fertile island development: A 10‐year artificial shrub experiment in a desert grassland

Abstract The relative importance of biotic and abiotic processes in the development of “fertile islands” in dryland systems has rarely been investigated. Here we approached this question by using artificial shrubs, which exclude plant litter production and soil nutrient uptake, but retain the functions of trapping windblown material, funneling of stemflow, and differential rain splash. We conducted a vegetation manipulation study more than a decade ago in the desert grassland of southern New Mexico and subsequently revisited the site in 2012 and 2015. The results show that no notable soil mounds were observed under the artificial shrubs; however, soil texture under the artificial shrubs has gradually changed to resemble the patterns of soil particle‐size distribution under natural shrubs. Our results highlight that with the exclusion of direct biotic additions, soils captured by shrub canopies are not necessarily fertile and thus do not themselves contribute to the development of fertile islands.