In this Insight, Rafael Dudeque Zenni talks about his recent research looking at how co-occurring invasive species join forces for increased and persistent impact on the ecosystem and why the the Cerrado of Brazil was the perfect place for this experiment.

The Cerrado is a biodiversity hotspot facing rapid transformations due to deforestation, land conversion and habitat fragmentation. The Cerrado is also a hotspot of naturalized and invasive species, with more than 200 known widespread naturalized plant species across the biome, which represents about 2% of the flora of the Biome. Among these naturalized species there are numerous non-native grasses, such as Urochloa brizantha and Andropogon gayanus, that were initially planted for pasture and now threaten ecosystem functioning by increasing flammable biomass and altering fire regimes. Others, like Melinis minutiflora, initially arrived by accident in association with human activities and were latter picked up for human use.

Previous research has documented severe impacts of plants invasions, mainly African grasses, in different Cerrado formations (from grassland and savanna to forest formations). For instance, plant invasions in the Cerrado have caused local extinction of native species, homogenized habitats, altered fire regimes, hindered secondary succession, and altered soil properties. However, none of these studies teased apart impacts of co-occurring invasive species or evaluated the ecological consequences of their removal. Right after finishing my PhD I was awarded a post-doctoral grant to study impacts of invasive plants in the Cerrado, so I decided to tackle the question of impacts of invasions when more than one invasive species were present and also address the ecological effects of removing the invaders. Were impacts of co-occurring invasive plants the same as if a single invader was present? Did invaders compete between them alleviating their negative effect on the ecosystem? Did they facilitate each other and exacerbate their negative effect on the ecosystem? And what about management? What happens if only one of the invaders was removed? Could a management plan for a single invasive species work when multiple invaders are present? These questions had never been addressed for the Cerrado and I had the chance to tackle them.

Brasilia, where the experiment for the paper was done, is located right in the core of the Cerrado biome and the IBGE reserve was the ideal place for a manipulative field experiment because of its proximity to the university and the presence of multiple plant invaders. The Department of Ecology at the University of Brasilia also has many top-notch scientists with expertise in different aspects of the Cerrado and I had the opportunity to collaborate with some of them for this project. The initial idea was to study sites with four co-occurring invasive species (Urochloa brizantha, Andropogon gayanus, Melinis minutiflora and Hyparrhenia rufa). However, it turned out impossible to find enough replicates with this combination of invasive species for any proper experimental design and hypothesis testing, so we settled for two species. I still want to find a place to test interactions of more than two invasive species co-occurring, but that is a project for the future.

I think one important message of our paper, one that is perhaps important for management and biodiversity conservation, is that merely removing invasive species from natural habitats will not necessarily and immediately trigger the recovery of native communities and ecosystems processes. We showed that the impacts of the invasive species were persistent even after two years of removal, and that removing one invader resulted in the growth of the secondary invasive instead of the native species. These results highlight the importance of active ecological restoration after invasive species control and that is probably true for other ecosystems.