Effects of Acacia invasion on water quality, litterfall, aquatic decomposers, and leaf litter decomposition in streams

Abstract

Abstract Small streams and their riparian vegetation are closely linked ecosystems. Thus, the invasion of native riparian forests with non‐native species can impact stream ecosystems. We assessed the effects of the invasion of broadleaf deciduous forests by evergreen, nitrogen‐fixing Acacia species on seasonal variation of relevant instream environmental variables, litterfall in the riparian area, aquatic decomposers, and leaf litter decomposition, by comparing three streams flowing through native forests (native streams) and three streams flowing through invaded forests (invaded streams) in central Portugal. Invaded streams flow through forests composed (almost) of monospecific stands of Acacia trees. Litterfall in the riparian area was sampled with fabric traps and sorted into five categories: leaf (including phyllodes), flower, fruit and seed, wood litter, and other materials. Aquatic hyphomycete conidia suspended in water were sampled to assess conidia concentration and community composition. Leaf litter of Quercus robur was enclosed in coarse‐mesh bags and incubated in streams to assess decomposition rates and associated macroinvertebrate density and community composition. Samples from each variable were collected monthly from streams over 1 year. Aquatic hyphomycete conidia concentration was higher in invaded streams in spring/summer when litter inputs, water temperature, and aquatic nutrient concentrations were higher. In contrast, conidia concentration was lower in invaded streams in autumn/winter as they received less native deciduous leaf litter in autumn than native streams. Aquatic hyphomycete community structure changed, and species richness was lower in invaded streams because aquatic nutrient concentrations were higher and leaf litter species richness was lower. Macroinvertebrate and shredder density in decomposing leaf litter did not differ between native and invaded streams, but litter bags may have artificially increased densities by providing high quality food and/or refuges in streams with poor‐quality resources. Nevertheless, macroinvertebrate community structure changed, and family richness was lower in invaded streams. Finally, decomposition rates of Q. robur leaf litter in coarse‐mesh bags were similar between stream types, despite differences in aquatic decomposer communities. Overall, Acacia invasion changed water quality, litterfall seasonality and composition, and aquatic decomposer communities (especially aquatic hyphomycetes). However, Acacia effects on macroinvertebrate density and leaf litter decomposition rates were less pronounced, suggesting that higher trophic levels may be more resilient to invasion than basal levels, or the invasion time/extent in our invaded streams was not strong enough to affect macroinvertebrates and associated processes. Instream invasion effects on aquatic hyphomycete communities were more strongly mediated by changes in litter inputs rather than increases in aquatic nutrient concentrations because they remained oligotrophic in invaded streams. Simplification of riparian and aquatic communities may render them less efficient in coping with additional environmental changes. Acacia effects might be mitigated by the maintenance of a riparian corridor composed of native vegetation. The protection of non‐invaded riparian galleries and restoration of invaded ones could protect and restore stream ecosystems.