Seasonal environmental variation drives host and symbiont physiological state of three important reef-building coral species in Moorea, French Polynesia
Seasonal cycles in marine ecosystems generate fluctuations in important environmental factors key to triggering or driving life history states (i.e., gametogenesis, spawning, flowering, migration, etc.). Importantly, the seasonally driven change in environmental conditions influences the energetic and, therefore, physiological state of an organism, which is critical to its capacity to respond to environmental stress and perturbations. For reef-building corals, the nutritional symbiosis of the cnidarian host with intracellular single celled dinoflagellates in the Family Symbiodiniaceae, results in a combined responsiveness to fluctuations in temperature, light, and nutrients, all of which change seasonally in tropical oceans. Deviations in these three factors outside of normal ranges are major drivers of dysbiosis of the coral-Symbiodiniaceae relationship that can result in coral bleaching and mass mortality. However, many studies of the effects of these variables take place in the absence of a consideration of the current physiological state of the coral holobiont due to seasonal timing. Here, we used the well-described site of Moorea, French Polynesia to test the effect of seasonal variation from environmentally distinct sites on three dominant and ecologically important genera, Acropora, Pocillopora, and Porites. We collected samples from coral colonies in January, March, September, and November of 2020 and quantified a suite of physiological variables (13 characteristics measured) in the coral host and algal symbiont, as well as molecular identification of the Symbiodiniaceae community and host species. Physiology of the three genera differed significantly. Within each genus, variance partitioning analyses identified seasonal timepoint as the dominant explanatory factor, with a lesser influence of site. Porites showed the highest level of variance explained by site due to variability in symbiont population growth and biomass. Seasonal time point was the major driver of shifts in holobiont energetic state, characterized by strong shifts in tissue biomass, host protein, and symbiont photosynthesis and biomass characteristics, with site-specific physiological optimization in each genus. Despite Porites hosting high fidelity symbionts, we observed the strongest seasonal acclimation of the symbiont population physiology compared to that in Acropora (stable symbiont populations) and Pocillopora (exhibiting the highest diversity in symbiont populations). These data provide an essential picture of the need for considering physiological state and its seasonality, particularly considering the plethora of climate change related stress test assays taking place throughout the year.
This repository is organized such that each sampling time point (1 = January 2020, 2 = March 2020, 3 = September 2020, 4 = November 2020) has a dedicated folder with all metadata, data files, and scripts necessary for calculating physiological responses. All data output in these individual time points is then read in, assembled, and analyzed in the time_series_analysis
directory.
scripts
, data
, and output
for physiological responses collected in January 2020. The scripts in this folder read in raw data files and conduct normalization and calculations for each response. Calculated responses are output in the output
folder and analyzed as described in the time_series_analysis
folder. Open this project with the timepoint_1.Rproj
file. scripts
, data
, and output
for physiological responses collected in March 2020. The scripts in this folder read in raw data files and conduct normalization and calculations for each response. Calculated responses are output in the output
folder and analyzed as described in the time_series_analysis
folder. Open this project with the timepoint_2.Rproj
file. scripts
, data
, and output
for physiological responses collected in September 2020. The scripts in this folder read in raw data files and conduct normalization and calculations for each response. Calculated responses are output in the output
folder and analyzed as described in the time_series_analysis
folder. Open this project with the timepoint_3.Rproj
file. scripts
, data
, and output
for physiological responses collected in November 2020. The scripts in this folder read in raw data files and conduct normalization and calculations for each response. Calculated responses are output in the output
folder and analyzed as described in the time_series_analysis
folder. Open this project with the timepoint_4.Rproj
file. time_series_analysis
folder described below. Open this project with the time_series_analysis.Rproj
file. This directory contains scripts, a Figures
directory, and a Output
directory and subfolders within these for each data type.
environmental_data_analyses.Rmd
5_plasticity_analysis_travel_distance.Rmd
8_plasticity_analysis_centroid_distances.Rmd
9_multivariate_analysis_betadispersion.Rmd
We measured the following physiological responses:
To run .Rmd scripts, set Knit directory to "Project Directory" prior to running full scripts. Restart your R environment prior to running scripts.
To run all data from start to finish or re run all multivariate and univariate analyses after revising individual time point analyses:
Open the R project for the time points individually (e.g., timepoint_1.Rproj
within each time point folder). Run all scripts in the scripts
folders in individual time point folders. First run the surface_area.Rmd
and then the protein.Rmd
to generate output for normalizers needed in other data sets. Note that output of calculated metrics will be stored in the output
folder within each time opint with the timepoint number in the file name: 1_surface_area.csv
. Run these scripts for all four time points in turn (timepoint 1, 2, 3, and 4).
Open the time_series_analysis.Rproj
project in the time series analysis folder. This project contains all scripts to read in data from each time point, conduct QC and metadata merging, and perform univariate and multivariate analyses. Run the scripts in this project in the numerical order they are named (e.g., 1_assemble_data.Rmd
then 2_univariate_analyses.Rmd
.
If you are looking to run individual steps of the analysis without changing any inputs, you can open and run scripts at any point in the workflow.
QC'd colony metadata is available in the time series analysis repository here. It is critical to use this file for colony metadata. Colony metadata in individual time point folders includes errors and typos that are corrected in the time series analysis.
Use the colony_id_corr
column for the corrected colony_id.
'site' (site where sample was collected)
'lon' (longitude coordinate)
'lat' (latitude coordinate)
'date'
'time'
'species'. (species of coral sampled)
'colony_id' (coral tag number)
If you have any questions, contact Ariana Huffmyer at ashuffmyer (at) uri.edu.