Camouflage and the light environment

Lead Research Organisation: University of Exeter
Department Name: Biosciences

Abstract

Being spotted by a predator can spell the end for an animal. Camouflage is therefore critical to the survival of many species and is one of the most common anti-predator strategies in nature. An animal's camouflage often depends on matching the colours and patterns of its surroundings, but how can it achieve this when the lighting conditions continually alter the background's appearance? The vegetation and textures in a scene create complex shadows and patterns that rapidly disappear when a cloud passes over and diffuses the lighting. Likewise the colours in a scene depend on atmospheric conditions; on a clear day shadows are blue-tinted, and when overcast the lighting colour becomes uniform. Interactions between lighting conditions and three-dimensional habitat structure could therefore be essential for the success of an animal's camouflage and affect its very survival.

A number of camouflage strategies have been identified that help animals stay unseen, and they often depend on specific lighting conditions. For example, counter-shading is a strategy found in many species that have dark backs and pale underbellies. This coloration helps to cancel out the shadows on the animal's underside, making its tell-tale shape less easily recognised. Coloration that creates false textures and depth cues is also found in many species; tactics that mimic illumination by direct sunlight. Background pattern matching can protect animals from predators, but the background patterns will shift and change throughout the day. It is therefore possible that these strategies become ineffective or even counterproductive under diffuse, overcast conditions. Beyond basic colour differences it remains unknown how these camouflage strategies perform in different light environments and habitat structures; understanding this interaction - which could affect the survival of any visually hunted species - will be the focus of my proposed research.

I plan to use an innovative range of experiments to investigate the importance of three-dimensional habitat structure, colour and lighting conditions on camouflage. First, I will determine whether reed warblers can select nest sites with lighting conditions that optimise the camouflage of their eggs, and test whether a mismatch between egg coloration and lighting increases their likelihood of being spotted by predators. Next, I will create artificial nests in the wild, testing how different camouflage strategies perform given the habitat structure and lighting conditions at the time of predation. Finally, I will use an online evolution game that allows computer-generated 'eggs' to evolve under different lighting conditions and against different backgrounds. I predict that the optimal camouflage strategy should switch depending on how light interacts with the three-dimensional structure of a scene. Flat habitats should favour strategies that create false textures and depth cues, and direct lighting should promote counter-shading and red-green colour matching over blue-yellow.

My research will provide fundamental insights into the importance of lighting conditions and the shape of a habitat on the survival of individual animals. These findings would be of immediate value to conservation projects to develop land-use strategies and farming practices that help protect vulnerable species from predation. Humans are changing the lighting conditions of habitats across the world as a consequence of climate change, potentially altering the efficacy of camouflage strategies that had evolved under different lighting regimes. My research therefore has the potential to open up new areas of research in camouflage, environmental change and conservation.

Planned Impact

Conservation and Land Management:
Few habitats in the UK are not visually altered by human behaviour and land use practices. Understanding how vegetation structure and colour affects individual survival therefore has the potential to impact conservation via land-use policy. I will use my links with the British Trust for Ornithology and the Game and Wildlife Conservation Trust, and will contact other relevant organisations (such as Natural England, the National Trust, RSPB, BirdLife International, BugLife, The Woodland Trust, The Forestry Commission, and The Wildlife Trusts), providing them with reports and inviting delegates to a workshop that will disseminate the findings of the proposed research. A working group will then be formed for exploring the practical applications of the findings in land-use strategies and policy.

Industry:
Visual ecology affects animal behaviour, and the ability to hide from view is essential in a number of sectors. The proposed research will develop new models for determining how coloration affects the conspicuousness of a given target, and how the light environment or surrounding features can affect the appearance or behaviour of an animal. These findings will be of immediate relevance to the defence sector, and I will discuss these with representatives from the Defence Science and Technology Laboratory, QinetiQ and the Ministry of Defence by creating contacts and meeting delegates at camouflage conferences. Animal welfare in farming and pest control are other areas that could benefit from the project's output, where, for example the light environment could affect the behaviour of prey, or their camouflaged predators (e.g. creating microhabitats where birds are more comfortable to forage for insect pests, or light environments that reduce stress in livestock).

Public:
Humans are highly visual animals and research into animal coloration is often covered extensively by the popular press and wildlife/nature production companies. The public interest in camouflage and willingness to participate in online 'citizen science' games is pivotal to the proposed camouflage evolution experiment. The game will gather valuable empirical data for the project while simultaneously engaging the public interactively, and offering them further information about the research. The game will attract over ten thousand participants online, and a number of stands will be set up in public spaces to interact with the local community. I have considerable experience in disseminating my research through television, radio and printed/online media outlets, and have attended a course on media communication. I will use my links with production companies to investigate opportunities for disseminating research output, and will further publicise my research through press releases, blog posts, social media and school visits (continuing in my role as STEM ambassador). I will also continue to develop my open source image analysis toolbox which allows anyone to turn a digital camera into a powerful tool for measuring the visual world. Creating an online objective image analysis forum and filming additional 'how-to' videos will continue to allow members of the public to engage with science or pursue their own science projects.

Artists:
Impressionist painters of the nineteenth century developed a new appreciation of how light, shadow and surrounding colours can influence the appearance of a scene long before the importance of these factors were investigated by science and industry. The artist Abbott Thayer later developed many theories of camouflage based his observations of nature that we are continuing to study today. Camouflage is a field where artists and scientists can easily learn from one another, and I have links with a number of artists interested in developing work based on animal vision, and software developers interested in augmented reality and seeing the world through the eyes of another species.
 
Description The new "QCPA" toolbox has been released, allowing researchers to use calibrated digital imaging to model various aspects of animal vision, and to use this for spatiochromatic analyses (i.e. measuring colour and pattern simultaneously).

We/I have set up an entire website and online community to host the software and provide detailed user guides.
Exploitation Route This software is already in use by many researchers, and given its accessibility and scope is likely to be used by many more in the future. The software is primarily used by biologists, but has also been used in biomedicine, dentistry and industry (e.g. fashion industry)
Sectors Aerospace, Defence and Marine,Creative Economy,Digital/Communication/Information Technologies (including Software),Education,Culture, Heritage, Museums and Collections

URL http://www.empiricalimaging.com/
 
Description I have developed an animal vision image processing framework. This framework synthesises the past 40 years of theoretical work into combining spatial, chromatic and achromatic information into a range of meaningful parameters. This framework will be essential for the image processing required for my fellowship, but I am also releasing it as an open-source software add-on for ImageJ. This framework will be invaluable for researchers in a wide range of fields, from evolutionary and sensory biology, to physics, chemistry, paleontology, forensics, advertising, military camouflage development, biomedical sciences, and many more fields which require objective image analysis and colour measurement.
First Year Of Impact 2019
Sector Aerospace, Defence and Marine,Agriculture, Food and Drink,Chemicals,Creative Economy,Digital/Communication/Information Technologies (including Software),Environment,Pharmaceuticals and Medical Biotechnology,Other
Impact Types Cultural,Societal,Economic,Policy & public services

 
Title Quantitative colour and pattern analysis framework 
Description The QCPA framework brings together tools for analysing spatial, chromatic and achromatic aspects of any scene, as perceived by a given visual system. The framework uses models of animal vision throughout to provide a large number of parameters which describe important visual properties of any sample/scene. The framework is built using open-source code and user-friendly workflows. 
Type Of Material Biological samples 
Year Produced 2019 
Provided To Others? No  
Impact Major components of this framework have already been used by one of my collaborators for analysing the evolution of birds of paradise (paper in PLoS Biology). The tools should be released in a couple of weeks, and will hopefully be used in a wide range of fields by a large number of researchers. 
URL http://www.empiricalimaging.com
 
Description Butterfly wing dazzle 
Organisation University of Exeter
Country United Kingdom 
Sector Academic/University 
PI Contribution I discovered a unique function of the patterns of butterfly wings when taking off, essentially tricking visual motion detectors into thinking the butterfly is moving in different directions to its actual direction of travel. I conceived the study, and wrote the code for analysing motion from videos.
Collaborator Contribution My collaborators are providing general expertise on the function of "dazzle" motion camouflage (Laura Kelley), and expertise in human psychophysics analysis (Anna Hughes).
Impact Motion detection algorithms for processing high-speed videos, controlling for the receiver's temporal and spatial acuity.
Start Year 2018
 
Description Coevolution between brood parasitic cuckoos and their hosts 
Organisation Griffith University
Country Australia 
Sector Academic/University 
PI Contribution I am providing expertise in analysing the appearance of parasitic cuckoo eggs, and producing colour-matched artificial eggs which let us explore the evolutionary function of egg phenotype.
Collaborator Contribution This project is being led by a PhD student of Andrea Manica and Will Feeney. They run the field site, and are the experts in brood parasitism.
Impact Methods for creating patterned, colour-calibrated artificial eggs, with varied shapes and volumes.
Start Year 2018
 
Description Coevolution between brood parasitic cuckoos and their hosts 
Organisation University of Cambridge
Country United Kingdom 
Sector Academic/University 
PI Contribution I am providing expertise in analysing the appearance of parasitic cuckoo eggs, and producing colour-matched artificial eggs which let us explore the evolutionary function of egg phenotype.
Collaborator Contribution This project is being led by a PhD student of Andrea Manica and Will Feeney. They run the field site, and are the experts in brood parasitism.
Impact Methods for creating patterned, colour-calibrated artificial eggs, with varied shapes and volumes.
Start Year 2018
 
Description Quantitative Colour and Pattern Analysis Framework 
Organisation Deakin University
Country Australia 
Sector Academic/University 
PI Contribution I have helped develop an image analysis framework which combines spatial, chromatic and achromatic visual processing. I have written all of the code 9many thousands of lines), and conceived key elements of the image-processing framework, including noise reduction/edge reconstruction, and image segmentation steps which are all based on models of colour and luminance detection in animals.
Collaborator Contribution The conceptual framework behind this analysis has been in development for over 30 years (primarily by Prof. John Endler), and John has continued to provide ideas, clarifications and fixes. Together with the labs of Karen Cheney, Justin Marshall and John Endler (brought together by a shared PhD student, Cedric van den Berg) we have further developed the framework, and performed debugging and testing of the code.
Impact The framework we have developed in beta testing phase (ready for advanced release). This framework is a huge undertaking, involving many thousands of lines of code. It provides a comprehensive suite image processing, analysis and visualisation tools. We currently have a manuscript nearly ready for pre-print release (we are on the final edits). We have built a website which hosts a knowledge base (approximately 50 articles spanning >100 pages of text at the moment), a user forum, and video walk-through guides.
Start Year 2018
 
Description Quantitative Colour and Pattern Analysis Framework 
Organisation University of Queensland
Country Australia 
Sector Academic/University 
PI Contribution I have helped develop an image analysis framework which combines spatial, chromatic and achromatic visual processing. I have written all of the code 9many thousands of lines), and conceived key elements of the image-processing framework, including noise reduction/edge reconstruction, and image segmentation steps which are all based on models of colour and luminance detection in animals.
Collaborator Contribution The conceptual framework behind this analysis has been in development for over 30 years (primarily by Prof. John Endler), and John has continued to provide ideas, clarifications and fixes. Together with the labs of Karen Cheney, Justin Marshall and John Endler (brought together by a shared PhD student, Cedric van den Berg) we have further developed the framework, and performed debugging and testing of the code.
Impact The framework we have developed in beta testing phase (ready for advanced release). This framework is a huge undertaking, involving many thousands of lines of code. It provides a comprehensive suite image processing, analysis and visualisation tools. We currently have a manuscript nearly ready for pre-print release (we are on the final edits). We have built a website which hosts a knowledge base (approximately 50 articles spanning >100 pages of text at the moment), a user forum, and video walk-through guides.
Start Year 2018
 
Description Signal evolution in birds of paradise 
Organisation Cornell University
Country United States 
Sector Academic/University 
PI Contribution I contributed to the analysis of bird of paradise visual displays, specifically providing tools for segmenting the displays into discrete colours for subsequent analysis.
Collaborator Contribution My collaborators ran all other parts of the study (data collection, analysis etc...)
Impact We have published a paper in PloS Biology.
Start Year 2017
 
Description Signalling and anti-predator function of guppy eye colour changes 
Organisation University of Exeter
Country United Kingdom 
Sector Academic/University 
PI Contribution I have provided methodological expertise for making colour-matched artificial 3D guppies for behavioural experiments, and novel methods for tracking the movement of guppies. These are essential components for investigating the importance of eye colour changes in guppies.
Collaborator Contribution My collaborators run their guppy field and lab study systems, and have contributed all other aspects of the research.
Impact One paper in Current Biology to date, and another paper in preparation. We have pioneered new methods for performing predation experiments without killing the prey (by making colour-calibrated robotic prey)
Start Year 2018
 
Description The impact of artificial lighting on hawkmoth visual ecology 
Organisation University of Exeter
Country United Kingdom 
Sector Academic/University 
PI Contribution This project is investigating the effect of different artificial light sources on the visual ecology of hawkmoths, and their predators. We are building up a picture of how artificial light could affect nocturnal pollination, sexual signalling, and visual anti-predator defences. We are providing the conceptual analysis framework for this study and collecting or collating the data.
Collaborator Contribution This collaboration is partnered with Kevin Gaston's lab, and Jon Bennie's lab. Kevin and Jon are providing expertise on ecology and artificial lighting.
Impact Currently we have assembled large datasets of hawkmoth reflectance spectra, and relevant background reflectance spectra.
Start Year 2018
 
Description Using imaging techniques to classify cuckoo eggs and chicks 
Organisation Academy of Sciences of the Czech Republic
Country Czech Republic 
Sector Academic/University 
PI Contribution I developed techniques for analysing the colours and patterns of cuckoo eggs, and the colours of cuckoo chicks in order to derive biologically useful information, such as the mother of a given egg, or the sex of a chick. The methods involve colour quantisation and egg-specific pattern analysis.
Collaborator Contribution My collaborators run the study system being used in these studies, and provide system-specific knowledge.
Impact One paper in the Journal of Ornithology, and new methods for investigating egg appearance based on pattern.
Start Year 2018
 
Title Quantitative colour and pattern analysis framework 
Description The QCPA framework brings together tools for analysing spatial, chromatic and achromatic aspects of any scene, as perceived by a given visual system. The framework uses models of animal vision throughout to provide a large number of parameters which describe important visual properties of any sample/scene. The framework is built using open-source code and user-friendly workflows. 
Type Of Technology Software 
Year Produced 2019 
Open Source License? Yes  
Impact Major components of this framework have already been used by one of my collaborators for analysing the evolution of birds of paradise (paper in PLoS Biology). The tools should be released in a couple of weeks, and will hopefully be used in a wide range of fields by a large number of researchers. 
URL http://www.empiricalimaging.com