Comparison of methods to estimate soil seed banks: The role of seed size and mass

Abstract

There are two main methods for estimating seed density and species composition of soil seed banks: manual seed extraction and seedling emergence. These methods were used to determine and compare seed density and species composition in the soil of a natural grassland in Patagonia. Additionally, known amounts of seeds of different sizes from Patagonian grassland species were mixed with soil to evaluate the efficiency of the seed extraction method, and determine their recovery percentage. Seed density found in the grassland soil with the extraction method was four times higher than that found with the seedling emergence method. Through the use of these two methods, there was very little overlap found in species composition. Small seeds (< 1 mm) were only found with the seedling emergence method, whereas the seeds of species with specific germination requirements were found with the seed extraction method. Seed recovery of grassland species varied from 2.5% for smaller seeds (Erophila verna) to 100% for larger seeds (Rumex acetosella) with the seed extraction method. This method was more effective in detecting seeds of large-seeded species. Discrepancies in seed detection between both methods may be related to seed dormancy, specific germination requirements, seed size and mass. These two methods are necessary to describe seed density and seed bank composition.