High species turnover but functional stability in tropical ground‐dwelling ant assemblages over 12 years of monitoring in Central Amazonia

Abstract Understanding how assemblages change over time improves our ability to predict responses to local disturbances and global threats. Based on an extensive dataset collected across 25 km 2 in Reserva Ducke, Central Amazonia, we investigated the structure of taxonomical and functional‐group ant assemblages over 12 years. Ants were sampled in 30 permanent plots (250 m in length) spaced 1 km apart, using pitfall traps during four sampling events in 2006, 2012, 2014, and 2018. Each sampling event deployed 300 traps, totaling 1200 samples across the study. Our results revealed significant taxonomic shifts over time, with a sharp decrease in species richness and abundance in 2014, followed by a recovery in 2018. Approximately 50% of species were re‐sampled in subsequent years, suggesting a directional shift with no clear evidence of convergence or return to the composition observed in 2006. Species turnover was influenced by topographic factors, with flatter plots and valleys exhibiting higher turnover compared to steeper or well‐drained higher areas. Changes in functional groups at the plot level were also very high between sampling events, showing no clear pattern. However, all functional groups remained relatively stable at the site scale, with no clear sign of directional changes over time. Despite climate change and the increase in the frequency of extreme weather events in the region, our results are consistent with the scarce information that indicates ant assemblages are relatively stable in tropical forests in the long term. These findings suggest that although ant species composition in tropical forests may fluctuate due to environmental changes, functional redundancy within the community may buffer ecosystem processes, ensuring long‐term functional stability. However, our results also show large variations in univariate ant metrics (richness and abundance) between years that may obscure long‐term patterns.