Habitat integrity and canopy cover shape small-scale variation in coloration among odonate assemblages in Neotropical...

Small aquatic ecosystems, such as Amazonian igarapés, harbor high biodiversity composed of organisms essential for ecosystem functioning. Dragonflies and damselflies (Odonata) are abundant and sensitive to habitat alterations, making them useful indicators of environmental conditions. Their conspicuous diversity and coloration patterns support behavioral strategies and thermoregulation, communication, and protection against solar radiation and water loss. In this study, we evaluated whether environmental gradients influence Odonata species distribution in Eastern Amazonian streams through selection on coloration traits. Specifically, we tested (i) whether local environmental conditions are associated with assemblage-level variation in non-circular coloration traits (Saturation, Value, melanin coverage, and wax-derived pruinescence coverage); and (ii) whether species composition varies along gradients of habitat integrity and canopy cover. We sampled 25 streams and analyzed 213 individuals from 32 species/morphospecies. Using standardized photographs, we quantified body coloration in the HSV colour space (Hue, Saturation, and Value) and estimated melanin and pruinescence coverage. We analyzed community-weighted means of Saturation, Value, melanin coverage, and pruinescence coverage; we used Hue descriptively. Saturation and Value declined with habitat integrity; Value increased with canopy cover, whereas pruinescence coverage decreased. Melanin coverage showed no clear environmental association. More shaded streams unexpectedly supported assemblages with higher Value, and species composition varied with habitat integrity. These findings suggest that distinct coloration components, together with other physiological traits, may help shape fine-scale Odonata community composition along microclimatic gradients. Coloration may complement occurrence and abundance data in environmental assessment, pending experimental validation of the mechanisms involved.