Trait patterns can give insights into how communities assemble under a functional perspective. We constructed a rotifer trait matrix related to food acquisition and predator defence and calculated Rao's quadratic entropy (Q) as an index of functional diversity to investigate trait patterns in different layers (0-2, 5-35, 0-35 m) for a 5-year dataset of Lake Tovel, Italy. Trait patterns were determined by comparing Q observed to Q from random communities. While trait patterns can be determined by species traits, richness, and abundance, in most samples, irrespective of layer, trait patterns could be solely attributed to traits indicating their importance for community assembly. Trait convergence dominated in the upper layer, while trait divergence dominated in the lower layer. Using logistic regression, we related trait patterns to environmental parameters. In the lower layer, trait divergence was linked to competition for food while trait convergence was linked to copepod predation. However, in the upper layer neither competitors nor predators influenced trait patterns, and we suggest that ultraviolet radiation and temperature were the main drivers of trait convergence. Our study indicated that environmental filtering drove rotifer trait patterns in the upper layer, whereas species interactions drove trait patterns in the lower layer.