Linking faunal seed dispersal to tree dispersal traits and carbon storage in southern African woodlands

Abstract

Within climate change mitigation, forests play an important role as natural carbon pools. Human-induced defaunation, however, threatens carbon storage by putting successful seed dispersal by faunal activity at risk. Furthermore, in previous studies, the loss of large-bodied seed dispersers has been linked to a decrease of carbon storage in tropical forests. It remains unclear whether the same applies for savanna woodlands, in which a large share of carbon is stored by small number of fruit-producing tree species with large seeds, relying on large-bodied faunal seed dispersers. Savannas have largely been misunderstood as degraded forests and targeted by reforestation projects, threatening biodiversity, and affecting ecosystem processes such as fire regimes. To find synergies between climate change mitigation and biodiversity, this research aims to detangle the relationship between faunal seed dispersal, tree dispersal traits, and carbon storage in southern African woodlands. Subsequently, it investigated the effect of human disturbance on this relationship. Five new SEOSAW plots were set-up and sampled in Madagascar. Subsequently, tree inventory, carbon, and tree dispersal traits of the four southern African countries Angola, Madagascar, Mozambique, and South Africa were analyzed and compared. The analysis included a description of dispersal modes and fruit characteristics on country level and a statistical analysis including Spearman rank correlations, principal component analysis and linear regressions on species- and plot-level using community-weighted means. Results showed significant positive relationships between above-ground biomass, seed size and fruit size in tree communities. The effect of human disturbance on carbon storage and seed dispersal traits was negative. Finally, no distinct patterns between carbon storage and seed dispersal traits of faunal dispersed species on species level were found, while large gaps in trait data and dispersal mechanisms highlight the importance of further research.