Testing the mechanism and function of avian plumage camouflage

Abstract

The overall aim of my PhD was to investigate the plumage coloration and patterns of birds to determine how their camouflage matches the biome background in which they live. Animals in a wide variety of biomes rely on phenotype-environment matching to avoid detection by predators in their specific biomes. A key prediction regarding the function of phenotype-environment matching rests on the idea that camouflage should match not only to the specific background conditions but also involve matching characteristics of the habitat. However, this has rarely been tested in animal species with fixed camouflage appearance. In this thesis, I explored the plumage attributes and the camouflage function used by specialist ground-nesting bird species of tropical rainforest, taiga forest, dry forest, grassland, desert and tundra biomes. Digital photography and image analysis were used to quantify colour patterns to models of predator vision to gain an ecologically valid assessment of camouflage. With this information I created bird models that were tested in different biomes. I found that the plumage of ground-nesting bird species have a phenotype-environment association to their respective biome, likely driven by selection for camouflage (Chapter 2). I also show that specialist bird species have a phenotype-environment matching that matches predominant background characteristics across large spatial scales of their biome backgrounds (Chapter 3). Finally, I tested how camouflage match relates to the detection of bird models in their own and different biomes (Chapter 4). In this predation experiment I found that the match for luminance, marking luminance and marking size were the main predictor of survival probability (detection times) of forest and grassland birds. However, unexpectedly, I found that the matching degree to predominant biome background structures increased bird survival probability against a different biome. This thesis chapter demonstrates that specialist birds match their phenotype to the predominant substrate composition and vegetation structure of their specific biome. However, in my experiments, these differences did not translate into a reduced detection time.