Differences between primary and secondary plant succession among biomes of the world

Abstract

Successional theory lacks an explicit, conceptual integration across types of disturbances and biomes. Most successional research addresses site- or process-specific questions, but extrapolation of the findings to broad scales is limited. Studies of plant succession are often distinguished by the severity of the disturbance that triggers them (severely disturbed: primary; less severely disturbed: secondary). Here, we examine the common assumption that primary and secondary succession differ. We use two anthropogenic disturbances for this comparison, mining and ploughing. Successional generalizations are confounded by a complexity of environmental factors including climatic and geographic variation. We address these issues by contrasting succession across multiple biomes. We selected 166 studies of succession on mining sites (n = 73) or abandoned fields (n = 93) that allowed us to compare successional trajectories in terms of the likelihood of a return to target vegetation, changes in species richness, type of trajectories, and the importance of alien species. Success of both types of succession (a return to target vegetation) differed significantly among biomes, with more likely success in cold than warm biomes (for both primary and secondary seres) and in humid than arid biomes (primary seres only). Primary seres were also more likely than secondary seres to have increases in species richness, a predominance of divergent trajectories, and a lower likelihood that alien species influenced succession. Synthesis. The probability of reaching target vegetation by spontaneous succession is generally greater in biomes at higher versus lower latitudes. Primary and secondary seres differed regarding species richness, trajectories, and the role of alien species. Our results highlight that broad generalizations about succession are possible, despite a paucity of comparable data and our use of only two types of anthropogenic disturbances. Any generalizations that might be found among seres across disturbances and biomes will provide a practical framework for land managers to guide restoration efforts, particularly in poorly studied areas.