Climate mediates long-term impacts of rodent exclusion on desert plant communities

Abstract

Determining how climate affects biotic interactions can improve understanding of drivers of context-dependence and inform predictions of how interactions may influence plants under future climates. In arid environments, the community-level impacts of seed predators may depend strongly on aridity; yet, long-term studies documenting impacts of granivores on plant communities over variable climate conditions remain scarce. We evaluated how rodent exclusion interacted with climate to influence grassland and shrubland forb communities and the community-scale distribution of seed mass over 15 years in the climatically variable northern Chihuahuan Desert. In this dynamic system, two seasonally distinct plant community phases occur annually, one in spring and the other during the summer monsoon. Rodent exclusion significantly altered the community composition of monsoon season plant communities in both grassland and shrubland, but did not affect spring plant composition. Rodents suppressed the abundance of larger-seeded forb species and promoted smaller-seeded species. As a consequence, rodent exclusion increased community seed mass (CWM) in monsoon forb communities, most strongly in grassland. The magnitude of impacts of rodents on seed mass varied substantially from year-to-year, tracking variation in climate. Specifically, rodent exclusion increased community mean seed mass the most in dry years (grassland) or in years following dry years (shrubland). Rodent exclusion had relatively weak effects on plant species diversity and richness. Our results indicate that climate interacts with the presence of rodents to structure not only the composition but also the traits of desert plant communities.