Kleunen, M. v. | bioRxiv Science

bioRxiv2024

Diversity and productivity of a natural grassland decline with the number of global change factors

Grasslands are highly diverse ecosystems providing important ecosystem services, but they currently face a variety of anthropogenic stressors simultaneously. Quantifying grassland responses to global change factors (GCFs) is crucial for developing effective strategies to mitigate the negative impacts of global change on grassland communities and to promote their resilience in the face of future environmental challenges. We conducted a field experiment in the Songnen grassland, northeastern China, to test the combined effects of 0, 1, 2, 4, 6, and 8 GCFs, including fungicide, herbicide, insecticide, antibiotic stress, heavy metal pollution, light pollution, microplastic pollution, nitrogen deposition, tillage disturbance, and increased precipitation. We found that within one year, the increasing number of GCFs negatively impacts on both the productivity and diversity of grassland communities. In comparison to exposure to a single GCF, exposure to 8 GCFs led to a reduction in productivity and species richness by 42.8% and 42.9%, respectively. Furthermore, these negative effects seem to be linked to the reduction of dominant species and the concurrent increase in neonative species. The results of hierarchical diversity-interaction modeling suggested that the adverse impacts of an increasing number of GCFs on community productivity and diversity are attributable to both the specific identities of GCFs involved and their unique pairwise interactions. The results suggest that grasslands may quickly loose stability and degrade more rapidly in response to multiple co-occurring GCFs. Greater efforts should be made to conserve the functions and services of grassland ecosystems by reducing the impacts of human activities.

ecology↗

bioRxiv2022

Machine learning improves global models of plant diversity

Despite the paramount role of plant diversity for ecosystem functioning, biogeochemical cycles, and human welfare, knowledge of its global distribution is incomplete, hampering basic research and biodiversity conservation. Here, we used machine learning (random forests, extreme gradient boosting, neural networks) and conventional statistical methods (generalised linear models, generalised additive models) to model species richness and phylogenetic richness of vascular plants worldwide based on 830 regional plant inventories including c. 300,000 species and predictors of past and present environmental conditions. Machine learning showed an outstanding performance, explaining up to 80.9% of species richness and 83.3% of phylogenetic richness. Current climate and environmental heterogeneity emerged as the primary drivers, while past environmental conditions left only small but detectable imprints on plant diversity. Finally, we combined predictions from multiple modelling techniques (ensemble predictions) to reveal global patterns and centres of plant diversity at multiple resolutions down to 7,774 km2. Our predictive maps provide the most accurate estimates of global plant diversity available to date at grain sizes relevant for conservation and macroecology.

ecology↗