Resilience Potential of Coral Reefs in the Mariana Islands

Video Transcript
Download Video
Right-click and save to download

Detailed Description

This webinar was conducted as part of the "Climate Change Science and Management Webinar Series" held in partnership between the USGS National Climate Change and Wildlife Science Center and the FWS National Conservation Training Center. Webinar Summary: Reducing coral reef vulnerability to climate change requires that managers understand and support the natural resilience of coral reefs. To assist these managers, a team of researchers, supported by the Pacific Islands Climate Science Center (PI CSC) undertook a project to: 1) assess ecological resilience in the Commonwealth of the Northern Mariana Islands (CNMI), which is in the west Pacific near Guam, and 2) collaboratively develop a decision-support framework with local management partners for resilience-based management. The team used an approach that included surveys of 78 sites along reefs surrounding the most populated islands in CNMI (Saipan, Tinian/Aguijan, and Rota), data from environmental monitoring satellites and computer models, an analysis of ‘indicators’ of the processes that underlie reef resilience (e.g., recruitment of new corals), and an assessment of proxies of anthropogenic stress (land-based sources of pollution and accessibility due to wave exposure). The project resulted in a set of scores for relative resilience potential and allowed the team to rank the survey sites within and among the islands from high to low resilience and develop a set of custom criteria for a decision-support framework that can identify sites that warrant management attention.


Date Taken:

Length: 00:43:50

Location Taken: US


Ashley Isham:  Good afternoon, or good morning
from the U.S. Fish and Wildlife Service's

National Conservation Training Center in Shepherdstown,
West Virginia.

My name is Ashley Fortune Isham.

I would like to welcome you to our webinar
series, held in partnership with the U.S.

Geological Survey's National Climate Change
and Wildlife Science Center in Reston, Virginia.

The NCCWSC climate change science and management
webinar series highlights their sponsored

science projects related to climate change
impact and adaptation, and aims to increase

awareness and inform participants like you
about potential and predicted climate change

impact on fish and on wildlife.

We appreciate you joining us today.

I'd like to introduce Laurie Raymundo.

She's at the University of Guam, and was one
of the principal investigators with our speaker,

Jeff Maynard.

Laurie, welcome.

Laurie Raymundo:  Good morning, Ashley.

Ashley:  Good morning.

Laurie:  [laughs] Yes, I have the honor of
introducing a young man with whom I have enjoyed

working, Dr. Jeff Maynard.

He's an applied scientist.

He works as a coral reef ecologist and he
focuses on structured decisionmaking, risk

analysis and climate change.

He uses climate and ecological modeling to
advance research into exploring and forecasting

the impacts of climate change on coral reefs.

He also applies these advances with coral
reef managers to help address the threats

posed to reefs by climate change.

He's especially interested in assessing the
relative resilience potential of coral reefs

and using the results of these assessments
to target different types of management action.

This webinar is going to share the results
of one of these ecological resilience assessments

from a USGS PI CSC funded project that took
place in the Commonwealth of the Northern

Mariana Islands in the West Pacific last year.

This is post bleaching.

I'm very happy to welcome Jeff.

Ashley:  Jeff.

Jeff Maynard:  Good morning, afternoon, evening,
and thank you, everyone, from me, for listening


Laurie Raymundo, who you've just heard from,
and I have had the great pleasure these last

couple of years of coleading a team of scientists
and managers that have been working in CNMI

to better understand spatial variation and
resilience potential.

We have been applying what we've learned to
the making of resiliencebased management suggestions

that are aiding our manager partners with

This is a highly collaborative project that
was funded by a grant at the University of

Guam's marine laboratory from the Pacific
Islands Climate Science Center, which is based

in Hawaii and led by Dr. David Helweg.

The PI CSC is one of the Climate Science Centers
of the U.S. Geological Survey.

We've enjoyed working with the PI CSC through
the course of this project.

PI CSC staff have helped with aspects of the
presentation of our results and the project

summaries we have prepared for the public.

As is usual to these large projects, many
have contributed their time, expertise and


I'm speaking today on behalf of everyone who
has contributed, especially the main contributors

listed and the project coleaders, Laurie Raymundo
of the University of Guam and Steven McKagan,

who serves as the local NOAA Fisheries liaison
in CNMI.

We'll cover 10 topics during today's talk.

I'll provide the background and history of
resilience assessments in reef areas, our

study objectives, the steps of the resilience
assessment process, highlights of our methods,

how we analyzed our data, the ways we assessed
anthropogenic stressors, the ways our results

can inform management, and how we used connectivity
to interpret our results.

I'll conclude by reviewing our main results,
describing resources you can access and describing

some of what we see as future directions for
this research area.

I want to first talk about how undertaking
ecological resilience assessments got its

start and how this idea and approach has evolved
to what is being used and recommended today.

Much resilience theory, as it pertains to
coral reefs, started in the wake of the global

scale bleaching event of 1998, associated
with the El Nino that occurred that year.

Prior to the early to mid1980s, bleaching
tended to be rare and localized and corals

generally recovered.

There were even minor global scale events
in 1987 and 1990.

The 1998 event, however, was something altogether

Coral reefs in 60 countries were affected
by bleaching, and up to 70 percent mortality

was documented in severely affected areas.

Overall, the widely reported statistic about
the '98 coral bleaching event is that as much

as 16 percent of the world's coral may have
died that year.

The 1998 event raised awareness of the implications
of global warming and climate change, at least

among the coral reef community.

The event also got people thinking since the
impacts, so widespread, were clearly not spatially


Jordan West of the US EPA and Rob Salm of
the Nature Conservancy, who is shown here

skin diving in the West Pacific, first proposed
that spatial variation in factors that increase

bleaching resistance and support recovery
can be used as an assessment framework that

can inform conservation.

We refer to these factors as resilience indicators.

Areas with characteristics that reduce stress
or confer resistance or support recovery processes

may be more robust in the face of continuing
climate change, and thus, are priority areas

of target management actions to reduce stressors
related to human activities.

The Nature Conservancy then included the concept
of identifying areas with greater resilience

potential, be it from resistance or recovery
potential, in their conceptual resilience

model, which was designed to assist with designing
resilient networks of marine protected areas.

Sites with greater relative resilience potential
are critical areas, and preferentially investing

management effort in these areas is one of
the guiding principles that can help ensure

we support the natural resilience of coral
reef systems.

This is a critical point I want to highlight
as I work through this simple summary of why

ecological resilience assessments are useful.

The results can be used to target management
actions that benefit site and system resilience,

and thus, can help optimize application of
our limited conservation and management resources.

Importantly, up to 2009, none of these ideas
had been formalized into guidance people could

follow to undertake a resilience assessment.

In 2009, scientists at the IUCN, TNC, and
Great Barrier Reef Marine Park Authority published

a guide for resilience assessment of coral

Features of the guidance within this document
include that it was recommended that 61 indicators

be assessed or measured, which meant implementing
the framework was highly resource intensive

and included many subjective assessments.

For our work, the most recent and most important
step in the evolution of the methods for resilience

assessment in reef areas happened in 2012
at the Marine Conservation Congress in Vancouver.

A group of us designed a survey, 30 scientists
and managers participated in, that examined

what we considered to be the best subset of
30 of the previously proposed resilience indicators.

Indicators were scored for perceived importance,
scientific evidence and the feasibility of

assessment and measurement.

In the end, 11 indicators were recommended
for resilience assessment.

These were in either the top 10 for perceived
importance or for scientific evidence and

were considered feasible to assess or measure.

The recommended indicators are: resistant
coral species, coral diversity, coral recruitment,

coral disease, macro algae cover, herbivore
biomass and temperature variability and the

anthropogenic stressors, nutrients, sediments,
visible human impacts and fishing pressure.

Our goal was to greatly reduce the number
of indicators being assessed, as including

weak indicators actually dilutes the importance
of each indicator, and because evidence is

increasing that, though complex, resilience
processes in coral reefs are likely controlled

by only a few factors.

We found no relationship between results using
our new method and results using the methods

proposed with the IUCN 2009, and found that
using 11 indicators resulted in much greater

separation among sites in assessed resilience.

The research our team has undertaken in these
last few years in CNMI represents the first

fieldbased implementation of the ideas presented
within the paper on prioritizing indicators

that I just described.

Our collaborative work has three primary objectives
to assess the resilience potential of areas

where management activities have already been
implemented, to identify priority conservation

areas and better understand the primary drivers
of resilience potential at the island, and

CNMI wide scale.

We also had a secondary objective.

This was to support the coral reef conservation
community by developing a detailed, adaptable

process that guides the implementation of
ecological resilience assessments.

There are six steps in the process we used.

These are: deciding whether to undertake an
assessment, selecting indicators, collecting

and compiling the data, analyzing the data,
then identifying sites that warrant management

attention and presenting and communicating
the results.

We're going to review these steps during this
presentation, and I've listed the relevant

steps on some of the slides.

I'll rereview these once I've shared all the
results, as I think it's easier to understand

all the steps once you've seen our example.

I'll start describing our research by providing
some highlights of the methods we used.

We measured or assessed all 11 of the indicators
recommended within the McClanahan et al. 2012


I want to be really clear here that the stressors
related to human activity, listed at the end,

are considered resilience indicators within
the list of 11 recommended in the review.

However, these challenge resilience, so are
unlike the others, which are all indicators

of resilience processes.

Keep in mind through the coming slides that
the anthropogenic stressors are assessed separately

in our assessment.

They're not included in the assessment of
relative resilience potential.

I'll review how they fit into our decision
support framework a bit later in the presentation.

Moving on to the indicators included in our
resilience assessment and our field work.

For the coral community, 12 to 16 quarter
meter quadrants were used and all corals were

identified to species, and the longest and
perpendicular diameter were both estimated.

In all, approximately 160 coral species were
identified during our surveys.

Stationary point counts were used to assess
the fish community.

One of our project coleaders, Steve McKagan,
conducted a minimum of nine three minute long

stationary point counts, identified all fish
to species and estimated their lengths.

In all, Steve estimated the lengths of tens
of thousands of reef fish and identified 250


You'll see, we put a small step two up, top
right of this slide, as this is the step when

indicators are selected and methods decided

We surveyed 78 sites along the 30 foot contour
of the four reefs of four islands of the CNMI,

including Saipan, Tinian, Guguan and Rota.

This map of our survey sites in Saipan shows
that we had good spatial coverage around these

islands, with our sites roughly a mile apart
all the way around the island.

This slide reviews the six indicators that
were actually included in our assessment.

Aside from the coral and fish communities,
which I described on the previous slide, warm

season temperature variability is also included.

As sites where warm season temperatures are
more variable, they'd be better acclimated

to the temperature extremes that cause coral

The units are listed here, and as will be
obvious to everyone, the units for all these

indicators are necessarily very different.

This is a key point as it speaks to the first
required specimen data analysis, which I'll

describe in an upcoming data analysis slide.

One point, though, about our resilience indicators.

We consulted colleagues working within the
Nature Conservancy and NOAA's Coral Reef Ecosystems

Division in developing our herbivore biomass

Our method is inclusive of three herbivore
functional groups.

We calculated that average biomass in kilograms
per hectare of these three groups.

Consequently, our herbivore biomass metric
is inclusive of herbivore diversity, which

much recent research suggests is just as important
as herbivore biomass.

This means that our average biomass values
are not directly comparable with total herbivore

biomass values from elsewhere.

Now, over just a couple of minutes, I'll share
a little of what represented many hundreds

of person hours for our team.

The reefs in CNMI are really, really beautiful.

There are well over 400 reef fish species
in CNMI, and at least 200 coral species, meaning

CNMI definitely has among the greatest reef
biodiversity among US coral reef locations.

We learned, remarkably, the waters in CNMI
are very clear.

We set out three 50 meter transects, and people
that were serving on snorkel safety support

could frequently see our entire dive team
in transects across a 100 plus meters of reefs.

Here is Steve McKagan undertaking a fish species
census to end his dive.

He's diving in the coral gardens near Rota,
which is in one of the established marine

protected areas in CNMI.

These photos will give you a bit of virtual

Another of our coleaders, Laurie Raymundo,
is shown here assessing the coral community

and coral disease prevalence at a site near
Tinian Island.

Steven Johnson is a reef ecologist with the
marine monitoring team in CNMI.

He's one of our two coral biologists and is
a new Masters of Science student at the University

of Guam's marine laboratory.

Trust me, I would have liked to have shown
him wearing a little more than he is here,

but he only ever dives in board shorts.

Here's Lyza Johnston.

Liza is the science and team lead for the
CNMI marine monitoring team.

She is assessing the coral community at Bird
Island, which is in northeast Saipan, and

is another of the established marine protected
areas in CNMI.

Here are a few photos from other sites we
surveyed to help you visualize what the coral

reefs in CNMI are like.

I'll now talk everyone through the basics
of step four analyzing the data you collect

and compile the resilient indicators.

I'm calling this a look under the hood.

You can see all of you in the audience depicted
there in the top right.

I thought if I showed a sideon view of this
bloke working on a car, it would help to get

your attention for the only slide I'll share
that has a pretty detailed description of

how the math for these analyses works.

I mentioned before that each of the resilience
indicators that we include in our assessment

has different units.

This means that all of the indicators have
different scales.

The data have to be normalized and converted
to a unidirectional scale prior to calculating

a composite score for resilience potential.

To normalize the data, all values for each
indicator are divided by the maximum value.

This expresses all values as a decimal percentage
of the site with the maximum value and ensures

all indicators have a scale ranging from zero
to one.

The scale is inverted for macro algae covers,
such that a high score always means higher

relative resilience potential, which is why
I call it a unidirectional scale.

High score's always a good score.

These normalized scores are then scaled or

Part of Table Two from McClanahan et al. 2012,
is reshown here.

We compared the perceived importance scores
by dividing these scores by the lowest important

score for our indicators.

This results in a multiplying factor you can
see it on the right there, ranging from 1.36

to one, up top.

The normalized scores for the indicators are
weighted using these multiplying factors because

we intuitively know that some of the indicators
are more important than others.

The normalized scores are multiplied by the
scaling factors we calculated.

These converted scores are then averaged to
produce the raw score for resilience potential.

These values are then renormalized, which
expresses resilience potential for each site

as a decimal percentage of the site with the
maximum score.

We call it relative resilience potential.

We then rank the sites from high to lowest
score and use four relative classes based

on where the final score fit into the distribution
of scores.

Sites with low relative resilience had scores
less than the average minus one standard deviation.

Sites with high scores had scores greater
than the average plus one standard deviation.

On this map, we show the results for our analysis
that compared all sites against all other


On the top right, you can see the distribution
of resilience scores with the average near


The assessment results suggest the resilience
of 17 of the sites is distinctly different

and either greater or lower than the distribution
defined by the average plus and minus one

standard deviation.

Seven of the sites have high relative resilience
potential, and 10 have low relative resilience.

37 of the sites have mediumhigh, and 24 of
the sites have mediumlow relative resilience


All but one of the established MPAs has high
or mediumhigh relative resilience.

We had no preconceived notions as to where
exactly the sites with highest and lowest

relative resilience potential would be in
CNMI, but suspected the sites most remote

and least exposed to anthropogenic stressors
would be among those with the highest resilience


We found the exact opposite to be the case.

The majority of the high resilience locations
among the surveyed islands are in Saipan,

where greater than 90 percent of the 50,000
people residing in CNMI live.

There are no high resilience sites in Rota,
which is 50 kilometers south of Tinian and

Aguijan, and roughly 50 kilometers north of

That island has only 2,000 residents.

Indeed, seven of the ten low resilience sites
are in Rota.

Our connectivity simulations help to explain
this result, and I'll review those simulations

in upcoming slides.

Hopefully it's not too bad a memory, but I
want to quickly show the looks under the hood

and data analysis we'll have to distribute.

This is just so that I can share that those
steps I described can be undertaken for multiple

spatial scales.

On the slide I just presented, I mentioned
that all of our survey sites were compared

against all other survey sites.

For the analysis shown on this slide, sites
were only compared against sites surveyed

on the same island, with Tinian and Aguijan

Conducted two analyses so that our results
could inform decisionmaking, both at the region

wide and island level scale.

We found that there are at least two locations
with low relative resilience potential and

two sites with high relative resilience potential
at each island.

Generally, sites on more exposed sides of
the island, east for Saipan and Tinian, and

south for Rota have higher relative resilience

Understanding which variables most influenced
differences in resilience potential is another

valuable product of resilience investment.

This is because the indicators most influencing
rankings are the most important to include

in monitoring programs, and they reveal the
types of management actions that would benefit

the greatest number of sites.

We used two different analyses to examine
which of the indicators are most driving differences

in resilience potential among our survey sites,
which was our project objective too.

The scaling factors we used are pretty small.

They're shown again here on the top right.

Resulting in increases in the scores of greater
than ten percent for only two of our indicators.

Consequently, variation in the scores for
the variables is indicative of which indicators

are most driving rankings.

Indicators with greater variability are most
distinguishing sites from one another.

You can clearly see for the interisland and
all three intraisland analyses that coral

recruitment and herbivore biomass are the
most variable indicators and have the greatest

range in value.

We also used a canonical analysis of principle
coordinates, which is a type of ordination

analysis conducted in collaboration with our
group by Gareth Williams, who works at Scripps

Institution of Oceanography in San Diego.

You can see that the relative classes we set
are very different from one another, and clearly

line up along the horizontal axis.

The length of the line for each of the variables
is indicative of the importance of the variable

in distinguishing sites.

Herbivore biomass, coral diversity, coral
recruitment and macro algae cover are most

driving differences among sites.

We conducted the same analysis for the three
islands and island groups.

As is shown here for Rota, herbivore biomass
and coral recruitment are driving differences

in resilience potential as assessed here and
this was the result for all of the surveyed


I mentioned earlier that the anthropogenic
stressors were assessed separately to the

resilience indicators.

Anthropogenic physical impacts, such as from
anchoring, were excluded as we observed almost

none of these kinds of impacts during our

However, we assessed landbased sources of
pollution, which is inclusive of both nutrients

and sediments as well as fishing access using
GIS software and existing spatial data layers.

For LBSP, we used land use spatial data layers
from the forest service, and worked out which

drainages affected each of the sites that
we surveyed.

Our LBSP metric included the proportion of
the relevant drainages made up by urban and

cleared land and the local human population

The color scheme is the same here as is shown
on the other maps in that green is good.

So green scores would be low here.

We're still using the relative scale.

There were no sites with LBSP values in CNMI
lower than the average minus one standard


You can see, though, that there are sites
with above average LBSP values and even sites

with values greater than the average plus
one standard deviation.

LBSP values are, of course, highest near human
populations and near cleared lands.

For fishing access, our assumptions are that
access to the fishery in these islands is

primarily determined by the average wave height
at a site, the distance between that site

and an access point, such as a marina or boat
ramp, and the human population density near

the closest access point.

There's good evidence that this is the case,
given the exposed sides of these islands are

difficult to impossible for small craft to
access for much of the year.

The wave symbols show prevailing wind exposure
at the islands we surveyed.

You can see access is low in these locations,
which we assumed to be of benefit to the coral

reef fish community.

i.e., fishing pressure is probably lower in
locations with high wave exposure and greater

in locations with low wave exposure.

We've now reviewed the first four of the six
steps I described, and some of the six, given

I've shown how we presented our main results
in maps and tables.

Step five is as or even more important than
the others.

It's where you could say "the rubber hits
the road."

This is where we maximize the value of the
assessments and analyses for informing management


We set up a total of six custom queries of
our data.

And set criteria for these queries given there
are different reasons sites may warrant management

attention and actions to support resilience

Our queries identified targets for conservation,
LBSP reduction, fishery regulations and enforcement,

bleaching monitoring and supporting recovery,
reef restoration and coral translocation,

and tourism outreach and stewardship.

Our first three queries are based on targeting
actions to sites with greater relative resilience


This thinking is based on results presented
within a "Conservation Biology" paper written

in 2008 by Ed Game, who now works with TNC,
and a few of his colleagues.

The long and short of the findings from the
modeling is that we should protect strong

or high resilience sites if we are expecting
sites to spend most of their time in a degraded


The benefits of many types of management actions
take a long time to manifest and disturbance

frequencies are expected to increase in the
coming decades as our climate changes.

For these reasons, high resilience sites have
greater conservation priorities.

You may remember that I mentioned on one of
the introductory slides that high resilience

sites are among the critical areas we need
to manage to support site and system resilience.

A whole presentation could be prepared to
explain that line of thinking.

I'm sorry I had to review that so quickly
for this webinar.

I'll bring this up again really briefly when
I review vulnerability assessments on one

of the future direction slides.

I'll show two examples of the results of the
queries I just described.

We identified high and low resilience sites
that are currently outside established notake


As I was saying, the high resilience sites
that aren't currently being protected can

be considered conservation priorities.

New MPAs or temporary closures or similar
are not planned for CNMI right now, importantly.

There are a range of other management initiatives
that can be considered, though, including

other types of fishing regulations along with
increased enforcement and debris removal.

This example is shared because many managers
in coral reef areas will want to undertake

an ecological resilience assessment to identify
high resilience conservation priority sites.

This is how we went about that.

You can see the results, that there are high
resilience sites not currently protected or

rather not currently within the MPAtype of
protection in both Saipan and Tinian.

With this query, we show the locations that
have high or mediumhigh relative resilience

and above average scores for land based sources
of pollution.

These are targets for LBSP reduction.

13 of the 78 sites meet the criteria set for
this query.

This summary graphic has the first letter
of the query name within purple circles for

all of the sites to which the criteria for
at least one of the queries applied.

In total, 55 of the 78 survey sites meet at
least one of the six sets of query criteria.

I want to make two important closing points
about the queries we used to identify targets

for different types of management actions.

Firstly, the list of queries we set is not
exhaustive of all the possible options.

This is one of many reasons we always stress
that the process we used in CNMI can be replicated

or adapted.

There are likely to be other kinds of queries
that will make sense in other areas depending

on the local context and the type of stressors
related to human activity that are most likely

to be challenging the resilience of local

Secondly, we know that none of the management
action options I've just described are new.

The innovation is in using resilience explicitly
as an information layer such that actions

are targeted to maximize site and system resilience.

Within our project, we also examined connectivity
at the island scale.

This part of the broader study involved collaboration
with Matt Kendall, who works with NOAA's biogeography

branch, and happened to be concurrently leading
a project examining connectivity in CNMI while

we were conducting our resilience assessments.

Understanding connectivity, even at the wholeisland
scale, can help us better understand the resilience

assessment results and decide where to implement
management actions.

This second point has two parts.

We can identify where actions are required
to maintain larvae supply, and where actions

may be ineffective, due to the larvae supply
being really limited.

The question we wanted to answer was, "What
is the relative extent to which each of our

survey islands is a larvae source and destination?"

All I'm going to share about the methods for
the connectivity simulations is that they're

cool and really complicated.

As you can see from this animation on the
right, which shows a 4D hybrid coordinate

ocean model with a one day timestep, which
was used to examine connectivity by resolving

island scale current patterns.

We used two simulations.

One that assumed larvae had no swimming ability,
which is the case for coral larvae.

And another that assumed larvae could swim
with sensory capacity, which is often the

case with reef fish larvae.

We also included four pelagic larval durations
to capture the range of days coral and fish

larvae spend in the pelagic environment before

Our results took the form of eight matrices
set out as you see here, with sources as columns

and destinations as rows.

Numbers in the table cells are virtual larvae.

And we examined connectivity among our survey
islands within the CNMI portion of the Marianas

chain, Guam and other archipelagos.

Here's the really simple summary.

Considering both simulations and all four
of the pelagic larval durations we used, Saipan

is roughly twice the source as Tinian and
Aguijan is and ten times the source that Rota


Saipan and Tinian and Aguijan are comparable
destinations and roughly twice the destination

that Rota is.

Here's what those results mean for the two
reasons I described that summarize our interest

in the connectivity information.

Firstly, the lower connectivity between Rota
and the other islands may be why seven of

the ten sites with low relative resilience
potential are in Rota.

Secondly, management actions to reduce stress
and support resilience in Saipan and Tinian/Aguijan

can help to maintain larvae supply.

Also, actions to support resilience in Rota
may be insufficient to support recovery there,

given the limited supply of larvae.

We've covered a fair bit of ground, so I want
to offer four highlights of our results that

work as takehome messages.

The first is that resilience potential varied
greatly within and among islands for our analyses,

and some sites have high and some have low
relative resilience potential.

Secondly, herbivore biomass and coral recruitment
are key drivers in CNMI of differences in

relative resilience potential as assessed

The majority of sites were identified as warranting
management attention for at least one reason

we can relate to an action that will support

Lastly, connectivity information really helps
explain assessment results and prioritize

from among the sites that warrant management

Here are our six steps again to review.

First, was deciding whether to undertake an

The second was selecting indicators.

Third was collecting and compiling data.

The fourth is analyzing the data.

The fifth, identifying sites that warrant
management attention.

The sixth, presenting and communicating the

I want to emphasize to you that our team considers
scientist and manager collaboration to be

essential for all of these steps.

I actually had to remake this graphic because
the first only had a sign on one side, which

is actually indicative of the problem rather
than the solution.

The solutions we need and the building of
stronger bridges between science and management

requires scientists make suggestions to managers
and viceversa.

Our team believes our work to be a great example
in the realm of operationalizing resilience,

of scientists and managers collaborating.

More collaboration like this are required
both to undertake work like what we described,

and so that the work can evolve and be refined
and improved.

As detailed as this presentation has been,
this is really a highlightlike summary of

what has been a really large body of work
that has produced more results and local management

suggestions than can be summarized here.

The main aspect of our work and some important
theoretical background hasn't been covered.

However, we have produced a range of resources
that share our process and our project results

so that people can learn more about our work.

These include our 2012 project report available
on a NOAA CORIS website.

A howto guide for undertaking resilience assessment
is available on the TNC's reef resilience

web page.

Summaries of our guidance undertaking resilience
assessments, which can also be found on TNC's

reef resilience web page.

A workshop report on resiliencebased management
from a resiliencebased management workshop

held in Honolulu late last year.

Our USGS Pacific Islands Climate Science Center
project report, which is available on a USGS

web page that describes our project.

And an 84page site summary appendix we are
currently finishing to share results and management

suggestion for each of the sites we surveyed.

Lastly, we just submitted a manuscript for
review that will be published open access

later this year.

All of these materials are either publicly
accessible now or available upon request by

sending me or one of the other project leaders
an email.

There are two different future directions
for the applied research presented here that

I want to quickly review before I conclude.

Firstly, I want to make clear that the relative
importance of resilience indicators will very

spatially, especially among reef regions.

For this reason, those interested in undertaking
an assessment can start with recommended lists,

and then include and exclude indicators as
is appropriate for local context.

We're going to need to develop recommended
lists for the indicators for different reef


Those aren't available now.

For example, we know the drivers of resilience
processes are different in the Caribbean than

they are in the Pacific.

This is visually exemplified here using two
photos from sites that were rated as having

the greatest relative resilience potential
from an assessment undertaking in the Cayman

Islands in the Caribbean and for our study
in CNMI.

Secondly, we can undertake vulnerability assessments
that combine resilience assessments with remote

sensing and climate model based maps and projections
of spatial variations and exposure to disturbances.

In the IPCC’s framework for assessing vulnerability,
exposure and sensitivity combined to produce

potential impact which is moderated by adaptive
capacity to yield vulnerability.

The sensitivity and adaptive capacity terms
can be seen as resilience, so by combining

resilience assessments with exposure information,
we can both assess vulnerability and target

actions to the site with the lowest vulnerability.

These are the high resilient sites with lower
projected exposure.

For example, we recently produced downscaled
projections of coral bleaching conditions

for the Caribbean.

These projections are 4-km resolution and
we identified countries where the variation

in the projected timing of the onset of annual
severe bleaching conditions is greater than

10 years.

In doing so, we're identifying locations that
maybe temporarily refugia.

It'll be interesting in coming years to identify
locations that meet that criteria and have

greater relative resilience.

I’m just sort of scratching the surface
of the kind of mapping our and other teams

are doing in this working area, but, hopefully,
you can see the potential.

This and next year, Laurie Raymundo and I
will lead another PI CSC project related to

this modeling capability in collaboration
with Ruben van Hooidonk.

We will be producing downscale climate model
projections from Micronesia.

We'll be combining resilience assessment results
with our downscale projections to collaboratively

develop sustainability forecast with managers
and reef stakeholders.

I'm going to leave this slide up so that the
recording shows the details of references

cited in the presentation.

Jeff:  I want to conclude the way I started,

to reiterate that this project includes numerous
contributors and has been made possible by

funding provided by the USGS Pacific Island
Climate Science Center along with grants to

the project leaders from the other agencies
listed here.

Contact details of the project leaders are
on the bottom right.

While I'm likely to have enough time for all
of the questions people have, I really encourage

people to get in touch via email to provide
comments or ask questions, even to set up

the time to discuss the project results or
assessment process.

Thanks once again for everyone's attention
and for your interest in our project.

Ashley:  Excellent.

Thank you very much, Jeff.

All right.

I see that some are coming in.

I'd like to take the first question.

It's going to come from Carl.

It says, "Is there any concern relating to
sea currents and ecological resiliencies of

coral reefs?"

Jeff:  I see that, I guess in the first instance,
as a really broad question.

If you write me, then we can put you in touch
with Matt Kendall who was really the specialist

that worked with us that did all the connectivity
simulation that we summarized in the matrices

that we used to make our various management
recommendations and to interpret our results.

The short answer, sort of broadly is that,
yes, certainly there's uncertainty in that

kind of modeling when we think about the future
and how future changes in regional and global

climate may affect currents because there's
future uncertainty in that modeling.

For us, it's still uncertain at the island

We saw it as being the best available information
on a really important aspect of resilience

at the best possible scale.

Rather than ignore it, we build it in to the
extent possible.

I think it's a strength of what we've been
able to do because it would have been confusing,

I think, to a lot of people that worked in
CNMI with us that suspected even more strongly

than we did that the sites in Rota that are
very far from where most of the people live

didn't fare better in the assessment.

Having the connectivity results helps us to
explain that and, as you saw, really create

a lens for us for how we can prioritize management
actions among the island.

There's definitely concerns about the uncertainty
related to the work, but that's the best available

information at the best available scale.

As it's improved and refined, it can be continually
included in these kinds of assessments and

in future assessments.

Hopefully, that touches on what you're asking.

Ashley:  Thank you.

The next question is from Bob Glazer.

Hi, Bob.

"Jeff, I'm particularly intrigue by the connectivity

You presented it well with respect to resilience
and biodiversity conservation.

I'm wondering if you would care to speculate
on what it means for regulation, with respect

to fisheries management and fisheries sustainability."

Jeff:  Two questions on connectivity which
really wasn't the part I was leading.

This is particularly a tough one, so I don't
want to necessarily just table it because

I feel that question for the role that we
had in this work needs to be answered by local

managers that were working with us in CNMI.

I can say though that the report that Matt
Kendall and his colleagues produced is becoming

available around the same time that we were
developing this webinar.

It's only in this last couple of months that
people in Guam, where they're having Coral

Reef Symposium this week actually, and in
CNMI became exposed to it.

It's definitely raising a lot of eyebrows,
perking a lot of ears or whatever expression

you want to use.

They're looking into it.

I think it will be built in the future fisheries

How exactly, you'll need to follow up with
us on a bit later.

We could put you in contact with the managers
that really are making those decisions.

Ashley:  I'm not seeing any more hands or
questions coming into the chat box.

Holly, are you still on?

Did you want to make any closing remarks?

Holly:  I am on.

This is Holly at the USGS and NCCWSC.

We just like to say thank you to Jeff.

That was excellent.

It's always good to see what's going on out
in the field.

Thanks again.

Ashley:  Thanks, Holly.

I saw Dave on who was key in this as well
in supporting it.

I just want to give him the opportunity to
make any remarks as well.

Dave:  I'm very appreciative of both Laurie
and Jeff and the team's hard work both in

the field, the blitz they did on data collection,
and they're windows to good weather, all of

this is very sophisticated analytic work that
went into this presentation.

Thank you also to Ashley and Holly for helping
us set this up.

I hope this is just the first of a long series
of collaborations between the Climate Science

Center and the teams out in the Western Pacific.

Ashley:  Thanks, Dave.

All right.

Well, I'd like to say one more time.

Thank you very much, Jeff.

That was a wonderful presentation.