Impact of an invasive alien plant on litter decomposition along a latitudinal gradient

Abstract

Abstract. Invasive alien plant effects on ecosystem functions are often difficult to predict across environmental gradients due to the context-dependent interactions between the invader and the recipient communities. Adopting a functional trait-based framework could provide more mechanistic predictions for invasive species' impacts. In this study, we contrast litter decomposition rates among communities with and without the invasive plant Impatiens glandulifera in five regions along a 1600 km long latitudinal gradient in Europe. Across this gradient, four functional traits, namely leaf dry matter content (LDMC), specific leaf area (SLA), stem-specific density (SSD), and plant height, are correlated to rates of litter decomposition of standardized rooibos (labile), green tea (recalcitrant), and I. glandulifera litter. Our results show that both invaded and non-invaded plant communities had a higher expression of acquisitive traits (low LDMC and SSD, high SLA) with increasing temperature along the latitudinal gradient, partly explaining the variation in decomposition rates along the gradient. At the same time, invasion shifted community trait composition toward more acquisitive traits across the latitudinal gradient. These trait changes partly explained the increased litter decomposition rates of the labile litter fraction of rooibos and I. glandulifera litter in invaded communities, a shift that was most evident in the warmer study regions. Plant available nitrogen was lower in invaded communities, likely due to high nutrient uptake by I. glandulifera. Meanwhile, the coldest study region was characterized by a reversed effect of invasion on decomposition rates. Here, community traits related to low litter quality and potential allelopathic effects of the invader resulted in reduced litter decomposition rates, suggesting a threshold temperature at which invader effects on litter decomposition turn positive. This study therefore illustrates how functional trait changes toward acquisitive traits can help explain invader-induced changes in ecosystem functions such as increased litter decomposition.

Inter‐ and intraspecific trait variation shape multidimensional trait overlap between two plant invaders and the invaded communities

“…In an ecosystem functioning context, these results show that invasion results in a shift of the community trait space towards (leaf) traits related to increased litter decomposability for both species (Kazakou et al , Orwin et al ). A previous study on I. glandulifera did also show that these trait shifts in the native community, quantified as community weighted means, were correlated with the increasing litter decomposition rates following invasion (Helsen et al ), thus illustrating the potential use of trait hypervolume results in an invader impact framework.…”

“…Although overlap between the invader and native species in both control and invaded plots increased around 2.5‐fold for the full dataset compared to the average overlap per study region, overlap was still much larger between the invader and NAT I than between the invader and NAT C (61.3% larger for the average hypervolumes per study region, 86.7% larger for the full hypervolume). This result can likely be explained by the relatively uniform species composition of the plant communities invaded by I. glandulifera across the five study regions (Helsen et al ). For R. rugosa , however, sampling across a larger part of its invaded range should be performed to evaluate the generality of the obtained results for this species.…”

The potential for mass ratio and trait divergence effects to explain idiosyncratic impacts of non‐native invasive plants on carbon mineralization of decomposing leaf litter

“…In our study, the average differences between understorey native and non‐native invasive plants in leaf nutrient levels were often not as pronounced as in other ecosystems (Leffler, James, Monaco, & Sheley, ), suggesting that differences between plant growth habitat (i.e., understorey plants and canopy trees in this instance) are a key feature in understanding decomposition dynamics in these forests. This observation highlights that, on average, non‐native and native plants affect decomposition through similar mechanisms, and corroborates that non‐native understorey plants with overlapping green leaf and litter traits are less likely to alter forest decomposition dynamics than those that are functionally unique (Drenovsky et al, ; Finerty et al, ; Funk, Cleland, Suding, & Zavaleta, ; Helsen et al, ). Importantly, however, while the average effects of native and non‐native understorey plants were similar, non‐native understorey species tended to be more functionally unique (i.e., occupy more distant areas of trait space) than native understorey species.…”

“…Overall, we found that the mass ratio effect and trait diver- are less likely to alter forest decomposition dynamics than those that are functionally unique (Drenovsky et al, 2012;Finerty et al, 2016;Funk, Cleland, Suding, & Zavaleta, 2008;Helsen et al, 2018). Importantly, however, while the average effects of native and non-native understorey plants were similar, non-native understorey species tended to be more functionally unique (i.e., occupy more distant areas of trait space) than native understorey species.…”

“…Mass ratio and trait divergence effects on carbon mineralization rates explained more variation and were more robust to potential influential observations of single‐species in oak–hickory than in maple–poplar litter mixtures. Variability in an invasive species impact across ecosystems is not unexpected (Helsen et al, ; Hulme et al, ), and perhaps the larger effects in oak–hickory litter mixtures were because native and non‐native understorey plants were overall more distinct from the canopy tree species in terms of their functional traits (i.e., significant difference in location in trait space) for oak–hickory, but not maple‐poplar litter mixtures. Our findings support that more research should intentionally compare impacts across multiple locations, which is currently an infrequent practice in most invasive impact studies (Hulme et al, ).…”