Clouds have profound consequences for ecosystem structure and function. Yet, the direct monitoring of clouds and their effects on biota is challenging especially in remote and topographically complex tropical cloud forests. We argue that known relationships between climate and the taxonomic and functional composition of plant communities may provide a fingerprint of cloud base height, thus providing a rapid and cost-effective assessment in remote tropical cloud forests. To detect cloud base height, we compared species turnover and functional trait values among herbaceous and woody plant communities in an ecosystem dominated by cloud formation. We measured soil and air temperature, soil nutrient concentrations, and extracellular enzyme activity. We hypothesized that woody and herbaceous plants would provide signatures of cloud base height, as evidenced by abrupt shifts in both taxonomic composition and plant function. We demonstrated abrupt changes in taxonomic composition and the community-weighted mean of a key functional trait, specific leaf area, across elevation for both woody and herbaceous species, consistent with our predictions. However, abrupt taxonomic and functional changes occurred 100 m higher in elevation for herbaceous plants compared to woody ones. Soil temperature abruptly decreased where herbaceous taxonomic and functional turnover was high. Other environmental variables including soil biogeochemistry did not explain the abrupt change observed for woody plant communities. We provide evidence that a trait-based approach can be used to estimate cloud base height. We outline how rises in cloud base height and differential environmental requirements between growth forms can be distinguished using this approach.