Big Ideas

RESTORING ISLANDS TO RESTORE REEFS

Demonstrating the ecological benefits of island restoration from the land to the sea.

Contributors

Rebecca Vega Thurber, Oregon State University
Deron Burkepile, University of California, Santa Barbara
Richard Griffiths, Island Conservation

f

CONTEXT

Coral reefs may disappear within the next century as climate change intensifies. Conservation of reefs in the 21st century needs to focus on promoting natural ecological processes that restore the resistance and resilience of reefs to climate-driven disturbances. Restoring seabirds to islands appear to be one of these key natural processes as seabirds provide the nutrient subsidies to islands and near shore marine habitats that promote the growth of both corals and fishes. Coastal zones are rich in natural resources, support one third of the world’s human population, and are cradles of biodiversity. Tropical islands and their adjacent coral reefs make up only ~0.2% of Earth’s surface, but they are home ~25% of all marine species. The cultures and identity of island communities are intrinsically linked to the sea and are dependent on healthy coral reefs. The fisheries on these reefs supply livelihoods and protein to roughly 1 billion people on the planet. Yet local and global anthropogenic disturbances, such as climate change, coastal development, pollution, and invasive species threaten the health of these island ecosystems and their reefs worldwide.

The eradication or complete removal of invasive alien species (invasives) from islands has the potential to be a game-changing conservation intervention for reviving and restoring coral reefs. Invasive species are one of the greatest threats to island ecosystems, often leading to extinction of endemic island flora and fauna, loss of habitat, and disruption of critical ecosystem services (Russell and Holmes 2015). Critically, the effect of invasives removal can also extend beyond the terrestrial environment, potentially helping restore critical functions on coral reefs.

Invasive rats often drive extirpations of seabird species on many oceanic atolls and islands globally (Butchart et al. 2018). The loss of seabirds disrupts the transfer and concentration of marine-derived nutrients between terrestrial and marine island habitats (Polis and Hurd 1996). This disruption of nutrient subsidies can dramatically alter ecosystem functions and resilience of islands (Graham et al. 2018). Recent rat eradication on the coral atoll of Palmyra, showed that removing the invasive species quickly alters the food webs of these islands (Young et al in press) and reduces the abundance of other non-native species such as mosquitos (Lafferty et al. 2018).

Eradicating rats is also known to promote seabird recovery on islands and may be a critical link to helping conserve coral reefs. A recent study in the Seychelles showed that coral reefs adjacent to atolls with intact seabird populations (without invasive rats) had healthier reefs (higher fish and coral abundance) than on atolls where invasive rats had virtually eliminated seabirds (Graham et al. 2018). The specific nutrients that seabirds deposit on islands increase coastal phytoplankton abundance as compared to islands with rats and without seabirds (Graham et al. 2018). This increase in phytoplankton can be important for corals by allowing them to feed more from the water column (which provides energy in addition to that gained from their algal symbionts) and grow faster, to store more energy, and to resist and recover from thermal stress (Shantz and Burkepile 2014, Ezzat et al. 2016). The higher nutrients in the water around seabird islands also increases benthic algal production leading to higher fish biomass (Graham et al. 2018). Herbivorous fishes in particular are beneficial members of the reef community since they remove harmful algae and facilitate coral recovery after disturbances. The nutrients from seabirds also directly enhance coral growth (Savage 2019). Thus, the presence of seabird breeding colonies on atolls is likely critical for the facilitation and growth of corals and the overall resilience of reefs. Despite this growing body of correlative evidence, there has yet to be a rigorous study examining the causal linkages of exactly how the removal of rats and recovery of seabirds impacts both the terrestrial biome and the coral reefs around these atolls.

This project leverages a pending proposal to the Schmidt Ocean Institute (SOI) that would create an ecosystem-level restoration experiment where invasive rats would be removed from a series of South Pacific atolls. Rat eradication will allow recovery of endangered seabird populations and their key nutrient subsidies to these islands. If successful, the funding from the Schmidt Ocean Institute would only cover the costs of the eradication and the ship time for the eradication on the R/V Falkor. This project directly complements and leverages the SOI proposal (intended to only fund the eradication expeditions and no follow-up science) to document the effects of the eradication. The research would focus on how the recovery of seabirds and their nutrients enhances the resistance and resilience of coral reefs to climate change. The collection and analysis of environmental DNA (eDNA) will enable the exploration of how the recovery of seabirds impacts coral reefs across scales of biological organization from microbes to manta rays. A variety of ‘omics methods will be employed to examine how the restoration of key nutrient subsidies from seabirds impacts the health, physiology, and resilience of corals. Ultimately, the removal of rats from islands may turn out to be one of the most effective local interventions we can undertake to protect the future of coral reefs. Ultimately, removal of invasive rats from islands will conserve bird species on Pacific atolls and, by returning guano, also represents a strategic means to enhance reef resilience in the face of climate-induced threats to reef ecosystems.

f

BIG IDEA

Demonstrate exactly how, and to what degree, restoring seabird islands (by eradicating rats) can enhance resilience of coral reefs to the effects of climate change.

This project is an unprecedented opportunity to reveal how the restoration of islands and their seabird populations can spill over to generate resilient coral reefs. Importantly, our work will provide critical evidence to managers and policy makers that the removal of rats from islands and seabird population restoration is one of the most effective local interventions we can undertake to future proof coral reefs and reef ecosystems. We propose to use environmental DNA (eDNA) to track how rat removal and the associated recovery of seabird populations affects this ecosystem across several levels of biological complexity, from the microbiomes of corals, soil, and water to the manta rays and whale sharks that visit seabird islands.

The power of our research is that it will provide scientific rationales to further illuminate the benefits of rat eradication as a strategy conservation of island atolls and coral reefs. The benefits and mechanisms we uncover, for linking the removal of rats to the recovery of seabirds and the climate-proofing of coral reefs, will inform government policies, conservation management actions, and catalyze new, public and private funding streams that target increasingly ambitious and innovative strategies for the removal of invasive species from islands. Ultimately, the restoration of the functional nutrient cycling benefits of restored island ecosystems could be a critically important component of enhancing the resilience of highly biodiverse nearshore ecosystems like coral reefs.

The reefs of French Polynesia are an ideal target for this approach. These islands are priority conservation targets as potential ‘climate refugia’ as they are less impacted by rising ocean temperatures than other reefs worldwide (Beyer et al. 2018). There are 118 atolls and islands in French Polynesia, but only 4 atolls known to be rat free. Thus, there is immense potential for scaling coral reef conservation via rat removal while also conducting transformative ecosystem-level science on linkages between land and sea conservation. In addition to the restoration outcomes featured in this project, rat eradication will benefit two critically endangered birds in French Polynesia, the Tuamotu sandpiper and the Polynesian ground dove.

Recent advances in eDNA methods make it possible to investigate and document potential causal factors between the removal of rats and the facilitation of healthier coral reefs. Lab experiments suggest that naturally-derived nutrients such as urea and ammonium, forms of nitrogen in bird  guano, can benefit corals, especially when under thermal stress, even promoting the symbiotic algae to give more carbon back to their coral hosts (Ezzat et al. 2015). Further, recent work from our group and others shows that naturally-derived nutrients, such as those from fishes, can help corals recover faster from coral bleaching events (Chase et al. 2018)(Shantz and Burkepile, unpublished). Importantly, the abundance of phosphorus in the water, of which seabird guano is an important source, appears to increase coral growth (Shantz and Burkepile 2014) and help corals to withstand thermal stress (Ezzat et al).

Innovations in eDNA also now make it possible to examine how seabird nutrients may impact coral microbiomes. The members of the coral microbiome (eukaryotes, bacteria, archaea, and viruses) are key to the metabolism and health of the coral animal (Thompson et al. 2014, Thurber et al. 2017). This partnership is so fundamentally important in corals that it was given a name, the holobiont (Knowlton and Rohwer 2003). Corals exposed to various stressors (e.g., climate change-mediated temperature effects, coastal pollution, overfishing), exhibit disruption (‘dysbiosis’) of their microbiomes manifested as increases in both the abundance of pathogens (bacterial and viral) and variability of the microbiome assemblage (Vega Thurber et al. 2008, 2009, McDevitt-Irwin et al. 2017, Zaneveld et al. 2017). This dysbiosis compromises coral health and increases coral mortality (Vega Thurber et al. 2014, Zaneveld et al. 2016). Thus, ensuring the health of the coral microbiome may increase the ability of reefs to resist or recover from disturbances (Rohwer and Youle 2010, Dinsdale and Rohwer 2011).

Seabird guano is a key source of good nutrients to reefs. The recovery of seabirds may both help corals grow faster and help desensitize them to thermal stress and coral bleaching as well as protect them from shifts in their microbiome. Yet, although we know mechanisms exist whereby naturally-derived nutrients can help corals thrive and withstand and recover from thermal stress, a critical missing link is to explore these relationships with seabird-derived nutrients at meaningful ecological scales. This project therefore aims to study the effects of invasive species removal on both the macroscopic and the microscopic dynamics of island ecosystems and at a grand spatial scale – spanning six independent atolls in French Polynesia.

Project Plan

Project Goals and Plan
eDNA as a critical tool to unlock the dynamics of microbiomes and megafauna

We will use modern approaches in DNA sequencing technology (metagenomics and eDNA; or ‘omics) and data analytics to track ecological dynamics, before and after the removal of rats on three Pacific Atolls. These data will be compared to three different atolls where rats are not removed, providing a replicated experimental design that can demonstrate a causative link between the removal of rats, recovery of seabirds, and changes in ecosystem (i.e., both land and sea) health. This will be one of the largest replicated conservation initiatives of its kind and will establish a baseline for long term ecological research. Further, revealing causative linkages among rats, seabirds, corals, and microbes can provide key evidence to managers, and public policy makers and funders that such similar eradication methods provide benefits for both land and sea.

This international and interdisciplinary scientific team will implement integrative programs to simultaneously document changes to both the marine and terrestrial ecosystems following invasive rat removal. A key innovation will come from the collection and analysis of eDNA that offers potentially profound new insights into the health and resilience of reef ecosystems. eDNA can provide a much more systematic documentation of rare or cryptic species as well as the dynamics of the microbial communities that have profound effects on ecosystem functions. Surveys using both in situ methods and eDNA will generate parallel data streams on coral reefs (population dynamics of fish, algae, & corals, coral physiology, health & coral microbiomes), oceanic ecosystems (plankton & nutrient dynamics), and terrestrial ecosystems (seabird populations, native vegetation recovery, soil/plant nutrient & ecosystem dynamics, invertebrate diversity). Having detailed surveys of both marine and terrestrial ecosystems will be critical as integrating these data streams will allow us to link changes in the terrestrial ecosystem (e.g. increases in birds and soil nutrients) to changes in the marine ecosystem (increases in coral growth, changes in the coral microbiome, increases in fish production).

In another key use of eDNA, the project will examine how the removal of rats and recovery of seabirds impacts the abundance of megafauna such as manta rays, sea turtles, and sharks. Hotspots of seabird-derived nutrients and phytoplankton production can attract these large pelagic consumers (McCauley et al. 2012). Additionally, the higher biomass of fishes that can develop around islands with abundant seabirds, as seen in the Seychelles (Graham et al. 2018), could attract more predators such as sharks to islands without rats. Consequently, the removal of rats and recolonization of seabirds is expected to attract an increased number of these charismatic megafauna. However, these rare and highly mobile animals are often difficult to sample using traditional ecological methods. Thus, eDNA from water samples can be used to detect the presence and abundance of large vertebrates around rat-free and rat-infested islands. The use of eDNA will be a key tool that will allow us to detect differences in megafauna abundance over space and time.

f

PARTNERS

Rebecca Vega Thurber, Oregon State University
Expertise: Coral Reef Omics & Microbiology

Deron Burkepile, University of California
Expertise: Coral Reef Ecology

PIs Vega Thurber and Burkepile have been collaborating on the ecology and microbiology of coral reefs for over a decade. They have collectively published over 120 papers and received over $10 million in grants including money from the National Science Foundation, the National Institutes of Health, the Moore Foundation, NOAA Coral Reef Conservation Program, the Belmont Forum, and the Zegar Family Foundation. Importantly, their work has been important for understanding how anthropogenic stressors impact coral reefs across multiple scales of ecological organization, with a particular focus on coral microbiomes.

Richard Griffiths, Island Conservation
Expertise: Island Conservation Biology

Island Conservation (IC) is the only global, not-for-profit conservation organization whose mission is to prevent extinctions by removing invasive species from islands. The Island Conservation team will support the terrestrial monitoring design, collection, synthesis, and analysis of the terrestrial seabird and vegetation pre- and post-monitoring. In the past 25 years, IC and partners have successfully restored 63 islands, benefitting 468 species and subspecies and 1173 populations. IC works where the concentration of both biodiversity and species extinction is greatest – islands. Removing invasive vertebrates is one of the most critical interventions for restoring island island ecosystems. PI Richard Griffiths and his island conservation biologist teammates at Island Conservation have documented and published these restoration findings in more than 100 papers (www.islandconservation.org/publications/).

f

BUDGET

ITEMUSDDETAILS
Dr. Vega Thurber's Lab
(Oregon State University)
$1.29 million– 5-year salary and benefits for lab and field technician
– 3-year salary and benefits for postdoctoral researcher to conduct data analysis
– 1-month (summer) salary annually for Dr. Vega Thurber
– 5-year stipend and tuition for Ph.D student
– Supplies for sample collection
– Reagents for eDNA and micro biome high throughput libraries
– Sequencing
– Sample storage
– Travel
– Indirect costs set at 10%
Dr. Burkepile's Lab
(U.C. Santa Barbara)
$1.27 million– 5-year salary and benefits for lab and field technician
– 1-month (summer) salary annually for Dr. Burkepile
– 5-year stipend and tuition for Ph.D student
– Supplies for sample collection and analytical assays
– Travel
– Publication
– Indirect costs set at 10%
Island Conservation$0.58 million– 1-month salary annually for Richard Griffiths (SW Pacific Regional Director)
– 3-month salary annually for conservation measures manager
– 3-month salary annually for island restoration specialist
– 1-month salary annually for global and external affairs director
– 1 month salary annually for communications director and manager
– 2-week salary annually for conservation director
– Supplies for sample collection
– Social attraction equipment (decoys, audio recording, nesting facilitation)
– Data analysis
–Travel
– Indirect costs set at 10%
Revive & Restore $80,000– Grant administration ($30,000)
– Outreach and education ($50,000)
TOTAL$3.06 million

f

RISKS & CHALLENGES

Potential for Rat Eradication Failure

There have been more than 1200 island invasive species eradication attempts globally, with about half of these being rodent eradications. The success rate in the tropics is about 85%. Island Conservation’s success rate is significantly above average. However, there is always a small risk that rats will not be eradicated from one of the three islands.  This risk will be mitigated by following best practice. Undertaking eradication  on three islands maximizes the likelihood that we will meet our objectives.

Uncertainty in the Time Horizon of Seabird Response

A central premise of the study design is that seabird colonies will recover in sufficient numbers to the atolls within the five year proposed timeline of this project. This assumption is based on research by Island Conservation and others that demonstrated that island systems do recover rapidly within 5-10 years (Jones 2010; Newton et al. 2016). However, few studies have examined shorter term effects and none have looked at the effects on microbiology. We predict that even small changes in rat and bird populations on small atolls will show biogeochemical and food web changes within 2-3 years, well within the timeline of this experiment. To facilitate recolonization by seabirds quickly, we will deploy proven social attraction methods including the use of visual and acoustic decoys that have been shown to mimic seabird breeding activity and accelerate recolonization by the targeted nesting species. Further, we can use data from rat free atolls Tenararo and Morane (historically rat free), and previous eradication experiments on Vahanga and Tenarunga (rat free since 2015) to compare with rat-infested islands to increase our power to detect the effect of rats on seabirds and corals.

The positive effects of this experiment will continue to increase well beyond the 5 year window as seabird populations continue to rise and the long-term effects on the reefs accumulate. Accordingly, we are confident that additional federal and private funding will be attainable to record longer term benefits that result from rat eradication. At its heart, this effort represents an unprecedented experiment in terms of scale and scope that will document long-term ecosystem gains to conservation efforts and the study of methods to improve conservation outcomes across systems. As such, the science behind the effects will become similar to National Science Foundation Long-term Ecological Research (LTER) sites, of which PIs Vega Thurber and Burkepile are Associate Investigators of in Moorea, French Polynesia,  that can provide extensive and foundational time series data linking terrestrial and marine ecosystem health.

f

TIMELINE

Year 1

  • Conduct baseline surveys of the reefs and terrestrial ecosystems of 6 rat-infested atolls in the remote Tuamotu archipelago of FP.
  • Collect baseline data focused on:
  1. Seabird populations and terrestrial vegetation,
  2. Terrestrial soil nutrients,
  3. Reef benthic communities,
  4. Reef fish communities,
  5. Abiotic oceanographic data (temperature and pH),
  6. Microbes associated with important reef species (corals, algae, fishes) via eDNA, and
  7. Large vertebrate megafauna (e.g., abundances of top predators, fishes, manta rays, whale sharks, etc) also via eDNA.

Importantly, we will collect samples for nitrogen stable isotopes from terrestrial soils, ocean water, seaweeds, corals, and fishes to track how seabird-derived nutrients permeate both terrestrial and marine ecosystems as seabirds recover.

Year 2

  • Analyze the ecological and omics data collected in Year 1.
  • Use the R/V Falkor (supported by the Schmidt Ocean Institute–complimentary pre-proposal currently under review) and her oceanographic and technological capabilities to conduct the eradication on 3 atolls while leaving the other 3 atolls as rat-infested controls.

Year 3

  • Resurvey the reefs and atolls post-eradication using live aboard dive vessels.
  • Continue processing and analyzing the collected data.

Year 4

Analyze and synthesize the ecological and ‘omics data from the past years.

Year 5

  • Resurvey the reefs and atolls post-eradication using live aboard dive vessels and process the resulting samples.
  • Analyze and synthesize the data collected in all years, and draft manuscripts to publicize findings, and support outreach to catalyze replication, scaling, and mainstreaming.

Analytical Approaches and Team Expertise 

At each collection time point we will collect eDNA and extensive ecological and oceanographic metadata.  

  • High throughput DNA sequencing of eDNA will be used to quantify both bacterial and viral dynamics and the abundances of large marine animals and plants. Dr. Vega Thurber has over 15 years of expertise in the analysis of metagenomes, microbiomes, and viromes from coral reefs. Her genomics work has been referenced in over 7000 citations.  
  • All marine benthic and oceanographic data will be processed and analyzed in PI Burkepile’s lab. Dr. Burkepile is an expert in both marine and terrestrial ecology and coral reef biology, particularly in regards to the effects of nutrients on host and ecosystem function.  
  • Two research technicians (one in each lab) will be hired to manage, collect, store, and process the samples. One postdoctoral researcher based at OSU will be hired in Year 3 to focus on assembling the eDNA data, conducting bioinformatic analyses, and interpreting the findings. Two Ph.D. students (one in each lab) will be key scientists that take on specific portions of the field and laboratory science as part of their graduate work and will participate in all field work. 

REFERENCES

Beyer, H. L., E. V. Kennedy, M. Beger, C. A. Chen, J. E. Cinner, E. S. Darling, C. Mark Eakin, R. D. Gates, S. F. Heron, N. Knowlton, D. O. Obura, S. R. Palumbi, H. P. Possingham, M. Puotinen, R. K. Runting, W. J. Skirving, M. Spalding, K. A. Wilson, S. Wood, J. E. Veron, and O. Hoegh‐Guldberg. 2018. Risk‐sensitive planning for conserving coral reefs under rapid climate change.

Butchart, S. H. M., S. Lowe, R. W. Martin, A. Symes, J. R. S. Westrip, and H. Wheatley. 2018. Which bird species have gone extinct? A novel quantitative classification approach.

Chase, T. J., M. S. Pratchett, G. E. Frank, and M. O. Hoogenboom. 2018. Coral-dwelling fish moderate bleaching susceptibility of coral hosts. PloS one 13:e0208545.

Dinsdale, E. A., and F. Rohwer. 2011. Fish or Germs? Microbial Dynamics Associated with Changing Trophic Structures on Coral Reefs.

Ezzat, L., J.-F. Maguer, R. Grover, and C. Ferrier-Pagès. 2015. New insights into carbon acquisition and exchanges within the coral-dinoflagellate symbiosis under NH4+ and NO3- supply. Proceedings. Biological sciences / The Royal Society 282:20150610.

Ezzat, L., J.-F. Maguer, R. Grover, and C. Ferrier-Pagès. 2016. Limited phosphorus availability is the Achilles heel of tropical reef corals in a warming ocean. Scientific reports 6:31768.

Graham, N. A. J., S. K. Wilson, P. Carr, A. S. Hoey, S. Jennings, and M. A. MacNeil. 2018. Seabirds enhance coral reef productivity and functioning in the absence of invasive rats. Nature 559:250–253.

Jones, H. P. 2010. Seabird islands take mere decades to recover following rat eradication. Ecological applications: a publication of the Ecological Society of America 20:2075–2080.

Knowlton, N., and F. Rohwer. 2003. Multispecies microbial mutualisms on coral reefs: the host as a habitat. The American naturalist 162:S51–62.

Lafferty, K. D., J. P. McLaughlin, D. S. Gruner, T. A. Bogar, A. Bui, J. N. Childress, M. Espinoza, E. S. Forbes, C. A. Johnston, M. Klope, A. Miller-Ter Kuile, M. Lee, K. A. Plummer, D. A. Weber, R. T. Young, and H. S. Young. 2018. Local extinction of the Asian tiger mosquito () following rat eradication on Palmyra Atoll. Biology letters 14.

McCauley, D. J., P. A. DeSalles, H. S. Young, R. B. Dunbar, R. Dirzo, M. M. Mills, and F. Micheli. 2012. From wing to wing: the persistence of long ecological interaction chains in less-disturbed ecosystems.

McDevitt-Irwin, J. M., J. K. Baum, M. Garren, and R. L. Vega Thurber. 2017. Responses of Coral-Associated Bacterial Communities to Local and Global Stressors.

Newton, K. M., M. McKown, C. Wolf, H. Gellerman, T. Coonan, D. Richards, A. Laurie Harvey, N. Holmes, G. Howald, K. Faulkner, B. R. Tershy, and D. A. Croll. 2016. Response of Native Species 10 Years After Rat Eradication on Anacapa Island, California.

Polis, G. A., and S. D. Hurd. 1996. Linking Marine and Terrestrial Food Webs: Allochthonous Input from the Ocean Supports High Secondary Productivity on Small Islands and Coastal Land Communities.

Rohwer, F., and M. Youle. 2010. Coral Reefs in the Microbial Seas. Plaid Pub.

Russell, J. C., and N. D. Holmes. 2015. Tropical island conservation: Rat eradication for species recovery.

Savage, C. 2019. Seabird nutrients are assimilated by corals and enhance coral growth rates.

Shantz, A. A., and D. E. Burkepile. 2014. Context-dependent effects of nutrient loading on the coral-algal mutualism. Ecology 95:1995–2005.

Thompson, J. R., H. E. Rivera, C. J. Closek, and M. Medina. 2014. Microbes in the coral holobiont: partners through evolution, development, and ecological interactions. Frontiers in cellular and infection microbiology 4:176.

Thurber, R. V., J. P. Payet, A. R. Thurber, and A. M. S. Correa. 2017. Virus–host interactions and their roles in coral reef health and disease.

Vega Thurber, R. L., K. L. Barott, D. Hall, H. Liu, B. Rodriguez-Mueller, C. Desnues, R. A. Edwards, M. Haynes, F. E. Angly, L. Wegley, and F. L. Rohwer. 2008. Metagenomic analysis indicates that stressors induce production of herpes-like viruses in the coral Porites compressa. Proceedings of the National Academy of Sciences of the United States of America 105:18413–18418.

Vega Thurber, R. L., D. E. Burkepile, C. Fuchs, A. A. Shantz, R. McMinds, and J. R. Zaneveld. 2014. Chronic nutrient enrichment increases prevalence and severity of coral disease and bleaching. Global change biology 20:544–554.

Vega Thurber, R., D. Willner-Hall, B. Rodriguez-Mueller, C. Desnues, R. A. Edwards, F. Angly, E. Dinsdale, L. Kelly, and F. Rohwer. 2009. Metagenomic analysis of stressed coral holobionts. Environmental microbiology 11:2148–2163.

Zaneveld, J. R., D. E. Burkepile, A. A. Shantz, C. E. Pritchard, R. McMinds, J. P. Payet, R. Welsh, A. M. S. Correa, N. P. Lemoine, S. Rosales, C. Fuchs, J. A. Maynard, and R. V. Thurber. 2016. Overfishing and nutrient pollution interact with temperature to disrupt coral reefs down to microbial scales. Nature communications 7:11833.

Zaneveld, J. R., R. McMinds, and R. V. Thurber. 2017. Stress and stability: applying the Anna Karenina principle to animal microbiomes.