Extractable and Germinable Seedbank Methods Provide Different Quantifications of Weed Communities

Abstract

AbstractSeedbank sampling and quantification methods vary in their ability to describe weed diversity and density, so proper method selection is critical for studying weed communities. The germinable seedbank (GSB) method is commonly preferred over extractable seedbank method (ESB), because the latter is more time-consuming. However, these two methods have only been compared using a few weed species and a relatively small number of samples. A total of 204 weed seedbank samples were used to compare both methods for weed density, richness, evenness, and Shannon-Weiner diversity using a split-sample approach. The two methods yielded dramatically different results. The ESB had 418% higher density and 35% more species per sample but 11% less evenness than the GSB. Diversity was estimated to be only 9% higher using the extractable compared with the germinable method. While the extractable method had higher density and richness overall, this was not true for every species, with only 7 of 14 common species detected in higher amounts by the extractable method. The results indicate the two methods are not strongly correlated, limiting the possibility of generating a conversion factor between methods. Nevertheless, evenness and Shannon-Weiner diversity might allow comparisons between both methods when the emphasis is on characterizing predominant weed species. The GSB is a practical approach to compare treatments or conditions; however, the ESB is more useful to accurately quantify weed species richness, diversity, and density.