Methods of soil seed bank estimation: a literature review proposing further work in Africa

Abstract

A number of methods are used to assess the soil seed banks of a range of plant species in various habitats around the world, with approaches that differ between countries and continents. An understanding of the differing techniques emphasises the need for further research, especially in Africa. We reviewed 97 articles on soil seed bank estimation, published between 2010 and 2020, and only 13.41% of these were from Africa. Soil sample collection in Africa was based mainly on stratified random sampling, systematic sampling, random sampling or cluster sampling carried out at the end of each region’s rainy season. Random and cluster sampling were more widely used in savannas, while stratified random and systematic samplings were more common in forests. The shape of the samples was either circular or quadrilateral (square and rectangular) or they were measured by soil mass or volume. The soil sampler cores most often applied were: circular diameter of 5 cm; square sizes of 10 × 10 cm, 20 × 20 cm and 25 × 25 cm; and rectangular sizes of 20 × 25 cm and 20 × 10 cm. The most-used soil core depths were 5 cm and 10 cm. No specific sample shape was linked with either forest or savanna ecosystems, although the number of samples depended on the land use and land cover. Soil seed bank densities and species composition were mainly assessed with direct greenhouse germination over trial duration depending on the plant species’ functional traits. In analysing soil seed bank data, non-parametric statistics were more frequently used than parametric statistics because of the skews in the data. This review will contribute to future soil seed bank studies in Africa.