Table of Contents

• Coverage
• Dataset Description
  • Methods & Sampling
  • Data Processing Description
  • BCO-DMO Processing Description
• Related Publications
• Related Datasets
• Parameters
• Instruments
• Project Information
• Funding

Porites rus was chosen as the focal species for this experiment as it is common on tropical coral reefs worldwide, it is one of the dominant coral species at the study site, and it is often found in direct contact with both conspecific and heterospecific coral species. To test the hypothesis that coral neighbors mediate the effect of SGD on coral host and endosymbiont physiology, we placed living P. rus corals into four neighbor treatments placed at each of the 20 experimental locations, which included: 1) no neighbors (a solitary P. rus), 2) two dead skeletal fragments of P. rus, 3) two conspecific fragments (P. rus from a different colony than the focal colony), and 4) two heterospecific fragments (Pocillopora acuta) (Figure 1 in Kerlin et al. (2025)). The dead skeletal fragments acted as a non-coral control, but were not cleaned during the experiment; thus, algal growth mimicked the natural succession on dead coral.

Fragments (3 cm in height and 2 cm in width) of P. rus and P. acuta were collected haphazardly approximately 200-650 meters up current of the SGD seep in ambient seawater conditions. Six fragments were collected from each of 20 putative P. rus colonies (i.e., colonies at least 20-m apart from each other) for center (“focal”) coral fragments (n = 80), pre-deployment metabolism sampling fragments (n = 20), and pre-deployment endosymbiont measurement fragments (n = 20). All physiological measurements described below were conducted on these 120 P. rus fragments. Neighbor fragments of P. rus (n = 80) and P. acuta (n = 40) were collected from an additional 20 colonies of each species (two fragments from each colony). All coral fragments were collected using a chisel and hammer, placed in Ziploc bags underwater, and transported to Richard B. Gump South Pacific Research Station (“Gump Research Station”). At the Gump Research Station, fragments were placed in outside flow-through seawater tables and resized to 3 cm height by 2 cm width using bone cutters, as needed. All deployment fragments were randomly assigned to an experimental location, with the four focal fragments from the same putative colonies assigned to each neighbor treatment within an experimental location. The focal P. rus fragments were hot glued with Gorilla Glue Hot Glue to a nylon bolt connected to a 5-cm2 PVC plate. The neighbor fragments were hot glued to the PVC plate as close as possible to the focal fragment, with the neighbor corals in direct contact with the focal fragment. Four 5-cm2 plates, one of each neighborhood treatment, were then attached to a larger 25-cm2 PVC plate using bolts. Each plate was deployed at its experimental location for two weeks by attaching the plate to rebar epoxied to hard benthos and then collected to measure post-deployment response variables. 

        Endosymbiont density and chlorophyll a content were measured following methods within Becker and Silbiger (2020) at the start of the experiment from the pre-deployment fragments and at the end of the two-week SGD exposure period from each of the deployed center fragments. Coral fragments were frozen at -40°C immediately after collecting the pre-deployment corals or after respirometry measurements (described below) for the center corals. The fragments were thawed and airbrushed to remove tissue using an Iwata Eclipse HP-BCS airbrush (Oregon, USA) with 0.2 μm filtered seawater collected from the lagoon offshore of Gump Research Station. Coral tissue was transferred into falcon tubes, homogenized with a PRO Scientific Bio-Gen PRO200 Homogenizer (Oxford, Connecticut), and aliquoted into two 1 mL microcentrifuge tubes for endosymbiont density and chlorophyll a content. Samples were frozen again at -40°C until processing, and final tissue blastate volume was recorded for each coral fragment prior to aliquoting.

            Samples aliquoted for chlorophyll a content were centrifuged (13,000 rpm for 3 minutes) (Labnet Spectrafuge 24D) and the supernatant was discarded to isolate the algal pellet. Acetone was added to extract the chlorophyll and the sample was vortexed and placed in 4°C in the dark for 24 hours. The samples were again vortexed and centrifuged at the same settings to separate out the debris and the extract was collected. The extract samples were processed on a Synergy HTX Multi-Mode Microplate Reader (BioTek, California, USA). Chlorophyll a content was calculated using equations from Jeffrey and Humphrey (1975) and normalized to surface area and endosymbiont density. E indicates the extinction at each wavelength (663 nm or 630 nm).

Chlorophyll a = 11.43(E663) – 0.64(E630)

            Aliquot tissue slurries for endosymbiont density were sent to the University of Hawai’i at Mānoa and measured by flow cytometry following methodology from Fox et al. 2021. For each coral fragment, one sample of 150 μL was analyzed on a flow cytometer (Beckman Coulter CytoFLEX S) at a rate of 60 μL minute-1 with excitation wavelengths of 375 nm, 405 nm, 488 nm, and 561 nm. Due to the uneven distribution of tissue blastate at the beginning of each run, the first 30 μL of each sample was removed from the analysis. Endosymbiont density was normalized to tissue blastate volume and coral surface area.

After removing the coral tissue using the airbrush methods described above, skeletal fragments were placed in a drying oven at 60°C to prepare for surface area measurements using the wax dipping method (Stimson and Kinzie 1991). First, a calibration curve (r2 > 0.9) of mass change of weight against surface area was created by using wooden dowels of known surface area. Coral fragments were then weighed, dipped in a 65°C Minerva paraffin wax bath (Georgia, USA) for two seconds, and then rotated in the air for two seconds at a constant rate. Fragments were set for 10 minutes to cool and then weighed again to obtain the mass change from wax dipping. The surface area of each coral fragment was calculated using the calibration curve obtained with wooden dowels. 

Coral metabolism

All metabolism measurements were conducted following methods within Silbiger et al (2019). We first characterized the relationship between net photosynthesis and photon flux density, commonly known as a photosynthesis-irradiance (PI) curve, to ensure photosynthesis rates in the experimental fragments were measured at saturating light conditions. For the PI curve, additional fragments from six of the donor colonies were collected and placed in flow-through seawater tables for approximately 48 hours to recover from the collection process and handling. Fragments were then placed in 650 mL acrylic chambers full of seawater (collected from the flow-through system at the Gump Research Station and filtered to 5 µm) at ambient temperature (28.4℃) with no air bubbles, a stir bar, a fiber-optic oxygen probe (Presens Oxygen Dipping Probes DP-PSt7; calibrated by Presens; Regensburg, Germany), and a temperature probe (Presens Pt1000, Regensburg, Germany, precision:  ± 0.1° C). The two probes were connected to a Presens Oxygen Meter [OXY-10 SMA (G2)], which measures oxygen percentage saturation and temperature (°C) at a frequency of 1 Hz. Oxygen concentrations (µmol L−1) were estimated from percent saturation accounting for a seawater salinity of 35 psu and standard oxygen solubility. Net photosynthesis was measured at eight light levels (µmol m−2 s−1) using an LED light (Mars Aqua 300w LED Brand Epistar, LongGang District, ShenZhen, China) for 20 minutes at each light level: 0, 57, 144, 219, 300, 435, 573, and 809 µmol m−2 s−1. Photosynthetically active radiation (PAR) was measured above each coral fragment with a cosine corrected MQ-510 Quantum Meter (error ± 2% and ± 5% at 45º and 75º from the light source, respectively; Apogee Instruments, Logan, UT, USA).

To calculate photosynthetic rates, the first two minutes of each run were removed to exclude the initial responses of the corals to changing light conditions and to ensure that the oxygen has reached equilibration within the chamber. The data were then thinned from every second to every 20 seconds to reduce noise in the data and to allow for processing of local linear regressions through the large dataset. Repeated local linear regressions were then used to calculate oxygen flux rates in the chambers using the R package LoLinR. Rates were normalized to the surface area (cm2) of each fragment after accounting for chamber seawater volume and blank control chamber rates. Saturating light (Ik) is the irradiance at which photosynthesis will no longer continue to increase.

Oxygen evolution was measured in all pre- and post-deployment coral fragments in 5 µm-filtered, ambient seawater (28℃) first in saturating light (approximately 590 µmol m−2 s−1) for 20 minutes to measure the net photosynthesis and then in complete darkness for 20 minutes in the same seawater to measure light-adapted dark respiration. Ten chambers were measured at a time, with nine chambers having coral fragments and one chamber acting as a control seawater-only chamber to account for background fluctuation in oxygen. The volume of seawater in each chamber was measured with a graduated cylinder after each respirometry measurement. Metabolic rates were calculated using the same methods outlined above for the photosynthesis-irradiance curve. Gross photosynthesis was calculated by summing net photosynthesis and respiration rates (as absolute values). 

Organism names and Life Sciences Identifiers (LSIDs):

Porites rus,urn:lsid:marinespecies.org:taxname:207231
Pocillopora acuta, Pocillopora acuta,urn:lsid:marinespecies.org:taxname:759099

All code and data processing is available on Github at https://github.com/njsilbiger/Neighborhood_effects_and_SGD (DOI for archival copy of V1.0, doi: 10.5281/ZENODO.14262764)

* Table within submitted file "Kerlin et al coral data.csv" was imported into the BCO-DMO data system for this dataset. Values "NA" imported as missing data values.   Table will appear as Data File: 960148_v1_sgd-response-conspecific-coral-physiology.csv (along with other download format options).

* BCO-DMO requires each column contain an individual Response_measurement (and units consistent per column). In order to meet this requirement, the BCO-DMO data manager transformed the data table and worked with the data contributor to review and make any additional changes needed. The original data format "Kerlin et al coral data.csv" provided to BCO-DMO was included as supplemental file 960148_v1_coral-phys-alternate-format.csv

Transformation notes (going from the originally provided file "Kerlin et al coral data.csv" to primary data table format 960148_v1_sgd-response-conspecific-coral-physiology.csv) :

* A table pivot was performed (transforming columns Response_measure, Initial, Final) to separate sets of columns per rate type (CCED_Initial,CCED_Final, etc)(See Parameters section for column descriptions).
* To do this tables were separated by filtering on rate_type (CCED,CCcm2,ED,GP,NP,R). Then joined to combine all columns using full joins on unique keys .
** "FragmentID" was unique per Rate_measure, Initial, Final so it was used as the join key.

Missing Data Identifiers:
* In the BCO-DMO data system missing data identifiers are displayed according to the format of data you access. For example, in csv files it will be blank (null) values. In Matlab .mat files it will be NaN values. When viewing data online at BCO-DMO, the missing value will be shown as blank (null) values.

* Column names adjusted to conform to BCO-DMO naming conventions designed to support broad re-use by a variety of research tools and scripting languages. [Only numbers, letters, and underscores.  Can not start with a number]

* Date converted to ISO 8601 format

* Organism names in this dataset were matched to Life Science Identifiers (LSIDs) using the World Register of Marine Species (WoRMS) on 2025-05-06.

NSF Award Abstract:
Submarine groundwater discharge (SGD) is the flow of water from land through the coastal seafloor into the nearby ocean. Approximately 13,000 cubic kilometers of groundwater is discharged into coastal environments every year, yet the effects of this fresh and often nutrient rich SGD are still poorly understood for coral reefs. This SGD input is driven by changes in precipitation, human land use, sea-level rise, tidal amplitude, and groundwater usage, many of which are rapidly changing with climate and human impacts. This project improves our understanding of SGD effects on coral reefs to better predict how both natural and human-induced changes will affect coastal ecosystem functioning in the future. Working in one of the most comprehensively studied coral reef ecosystems in the Pacific (Mo'orea, French Polynesia, home of the Mo'orea Coral Reef Ecosystem LTER); this project tests the influence of SGD on individual, community, and ecosystem-scale coral reef processes. Using mensurative studies, caging experiments, and a synthetic model, the investigators: 1) characterize SGD gradients and relate it to high resolution coral reef cover data, 2) determine how individual to ecosystem processes are influenced by SGD, and 3) develop a synthetic model to show how changes in SGD fluxes will alter reef ecosystem functioning. As SGD is a common feature on nearshore coral reefs worldwide, the results of this study have global implications for understanding the performance of coral reefs, which are essential economic, cultural, and scientific resources. This project is structured to provide training across multiple career levels, linking 13 undergraduate students, 2 graduate students, 2 senior personnel, 1 postdoctoral researcher, 1 female beginning lead investigator, and 2 senior co-investigators, with a focus on encouraging participation from underrepresented groups (e.g., through the Alaska Native and Native Hawaiian, Asian American and Native American Pacific Islander, and Hispanic-Serving Institutions of California State University Northridge, the University of Hawaiʻi at Mānoa, and California State University Long Beach). The investigators work with local K-12 students and teachers in Mo'orea and collaborate with an artist-in-residence to communicate science to the broader public through interactive and immersive art experiences in Mo'orea, Miami, and Los Angeles.

SGD is a natural and understudied feature of many nearshore coral reef ecosystems, which can contribute substantial changes to marine biogeochemistry, with impacts for coastal organisms such as reef-building corals, macroalgae, and bioeroders. SGD may play a key role in coral reef ecosystem functioning because it alters key physicochemical parameters (e.g., temperature, salinity, and nutrient and carbonate chemistry) that substantially affect both biotic and abiotic processes on coral reefs. This project (i) characterizes the spatial extent and biogeochemical signal of SGD in Mo'orea, French Polynesia, (ii) identifies how SGD influences microbial processes, benthic organism growth rates and physiology, species interactions between corals, macroalgae, and herbivores, and net ecosystem calcification and production rates, and (iii) quantitatively assesses how changes in SGD fluxes will alter reef biogeochemistry and ecosystem functioning through an integrative modelling effort. Specifically, the hydrogeological, biogeochemical, and ecological data collected in this study are synthesized in a Bayesian structural equation model. This project characterizes and quantifies how SGD directly and indirectly affects ecosystem functioning via changes in biogeochemistry and altered individual to ecosystem responses, thereby providing a better capacity to track and predict alterations in reef ecosystem function.

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.