Losing Nemo? Investigating the relationship between coral reef health and the presence of colorful fish

Methods

Summary

We will catch fish on scuba and quantify the color diversity of fishes using hyper-spectral cameras that can capture the entire spectrum of color visible to fishes. To fully capture the array of color diversity within a fish, we will make a grid of 20 measurements, allowing us to interpolate the spectrum of color within each species. 

To quantify the distribution and abundance of fish across reef sites, we will  conduct replicate 50 meter video transects across each site following standard methods.

We have previously worked in this reef system and know it to contain a broad diversity of abiotic conditions (e.g., various levels of environmental perturbation within each). We will quantify ambient light at the site of capture for each target fish using spectrophotometers. As light environments change at small local scales, this approach allows us to quantify light experienced by species using different portions of the water column (e.g., midwater versus crevice dwelling). 

We will also deploy several kdUINO buoys that span a customizable portion of the water column, enabling quantification of the diffuse attentuation coefficient (kd) at different depths for extended periods of time. See our lab note on Buoys!

We already have a phylogenetic tree based on genomic scale sequence data for all known reef fish species. Using this as a framework, we can test how the community dynamics of color change across sites with varying levels of disturbance. 

In tandem, we have also developed web based software that enables quantification of colors on fish photos. By photographing each fish, we can allow people to first-hand compare the color both on individual species of fish and across a reef community (see lab note on our software!).

Challenges

Color diversity is tricky to quantify for fish that perceive color differently from us. For example. some reef fish use different parts of the light spectrum. Hyperspectral cameras allow us to capture the entire range of spectral values and see colors the way different species do. The catch is that we have to obtain these values in the lab (there is no underwater housing for these units yet). 

Our previous work in catching and acclimating fish for physiology makes us confident we can do the measurement, but what is lost is the effect of depth. To overcome this, we will create models of depth using the bouy data AND integrate light source data from the spectometer. The buoy allows us to get good long-term data, but remember that fishes may be in parts of the reef that have different light regimes than where the buoy sensor is (e.g., a crevice dwelling fish versus a mid water fish). Although more complicated, this two tiered approach will insure a much higher level of ecological realism. 

Pre Analysis Plan

The primary purpose of this experiment is to get the first ever quantification of how color is distributed in a reef community. Using our transect and color data and a suite of commonly used phylogenetic comparative and ecological methods available in R, we can test how abiotic and biotic factors control or enhance the diversification of color.  Further, we can test how changing abiotic factors (turbidity) shifts the colorful part of a reef system away from the distribution in undisturbed sites. 

We can also test more detailed hypotheses. For example, by keeping track of male, females, and juveniles within each species, we can assess more nuanced changes in reef community composition (e.g., are brightly colored males duller at intermediate levels of disturbance).