Abstract

--- - Lay Summary & BULL; Camera traps are often used for biodiversity inventories and population monitoring of terrestrial and understory birds in tropical forests. Their use has considerably increased in the last couple of decades. & BULL; Assessing camera trap performance for bird sampling is key to improve bird monitoring in the tropics. & BULL; Mist nets recorded more bird species than camera traps, as small birds that fly in the sub-canopy were captured by those nets but not detected by cameras. & BULL; Conversely, camera traps recorded large ground-dwelling birds not captured in the mist nests, providing complementary information for bird surveys. & BULL; Mist nets and camera traps detect different species, but estimated species richness patterns were similar for different categories of body size, detection difficulty, population trend, and behavior. Therefore, it is important to use them together. & BULL; We encourage future research to investigate canopy ecology in tropical forests using camera traps, which may expand our understanding of bird community spatial and temporal variability. - "Different sampling methods are used to study bird diversity in the tropics, mist nets being one of the most common approaches. However, camera traps have been used for this purpose in the last 20 years. We compared the performance of mist nets and camera traps in sampling the diversity of understory and sub-canopy birds in a Neotropical rainforest. Given its high biodiversity, obtaining accurate bird diversity estimates in these forests is a challenging task. We sampled secondary forest patches in the Colombian central Andes between 2019 and 2021. We compared the effectiveness of both methods based on six comparison criteria (body size, foraging stratum, detection difficulty, habitat specialization, population trend, and migratory status). We recorded a total of 99 bird species: 92 species were captured using mist nets and 37 species using camera traps; 30 species were detected using both methods (mainly generalist and abundant birds). Mist net's effectiveness was 83%, while camera trap effectiveness was 68%. Differences in sampling effectiveness were mainly determined by body size, as small birds were captured using mist nets, while camera traps often recorded large birds. However, detection difficulty and habitat specialization were also relevant. Camera traps effectively recorded ground-dwelling species that were not captured in mist nets. In contrast, those birds that forage in the sub-canopy stratum were mainly captured in the mist nets but not detected in camera traps. Although both methods detected different species, we found similar patterns regarding estimated species richness among different categories of body size, detection difficulty, population trend, and behavior, but not for habitat specialization. As both methods have advantages and limitations, using mist nets and camera traps together would improve bird diversity estimations. We urge investigators to explore canopy ecology with camera traps, as they provide long-term information that cannot be obtained with other methods."