[Coral-List] Habitat condition metrics for coral reefs - is there a convention?

John McManus jmcmanus at rsmas.miami.edu
Sat Jun 9 16:52:14 EDT 2012

This is a challenge that I believe more people think they have solved than
actually have.

Here are some thoughts:
1. Indices based on change over time are going to be much better than those
based on static one-time data set analyses. Identifying what are the most
important things to measure over time is not always clear, but
statistically-valid periodic monitoring of even the most obvious indices is
immensely more reliable than making static inferences.  

2. In static analyses, for any given level of coral cover, algae cover, silt
load, etc. that someone thinks is an indication of anthropogenic stress, we
can probably find a part of, or a whole, coral community that is that way

3. Fish abundance is probably the better measure, as natural fish
populations around coral communities (in the few areas where they are not
fished at a scale of hundreds of miles) are usually very abundant. However,
fish populations, especially when fished, can vary radically throughout the
year and between years. So, I doubt anyone knows what the abundances by
major species are in any reef unless they have done surveys 4 to 6 times per
year for 3 or more years -- or the annual surveys for three years running
show no appreciable change .  Still, if you go to a coral community with 20%
or more coral cover and variable relief , and do not see substantial numbers
of adult fish within 100 ft of a fixed point, you can be sure there is a
problem. If such a place has reasonably abundant small fish but large fish
are extremely rare, and has no recent history of obvious fish kills, that is
extremely likely to be a problem with overfishing.

4. AGRRA was designed to support static ecosystem health and integrity
assessments, based on the establishment of norms. These analyses are
particularly good for identifying overfishing. With a large enough sample
size over a large area (many 10's of reefs), one can use these for
reasonable health assessments. However, to say that a single coral reef that
has never been studied before is in trouble based on coral-algal ratios or
the like is extremely shaky. We don't yet understand the relationship of
nutrient loading to herbivory in predicting what algal cover to expect. One
can make inferences from comparing living coral cover to recently (within a
few years to a few decades) dead coral, but a knowledge of prior storms,
cold spells, warm spells, phytoplankton blooms, freshwater flooding, etc. is
necessary. With lots of local knowledge, one can do this. For example, one
need only look at the massive dead elkhorn colonies (often more than 10 ft.
long) at North Sound, Antigua, and compare those with the small living
colonies to identify a problem -- albeit probably one due mostly to prior
disease. Note that despite this, that Sound has outstanding and truly
fascinating coral communities, although they are severely lacking (as of
five years ago) in adult fish. 

5. For your purposes (Saipan reefs), the norms established under AGRRA are
not useful (though the example is certainly worth following). For example,
coral cover tends to be much higher across the Indo-Pacific than in the

6. The Reef Check data set can be used for comparisons over very large
areas. There has often been a tendency in most broad area surveys aimed at
involving volunteers to bias the sampling toward areas of high coral cover.
Because of natural variability over time, high coral cover areas have a much
greater chance of seeing natural reductions than low-coral cover areas. One
needs to deal with average conditions, not extreme ones. Regardless, Reef
Check is a superb system for identifying potential problems over large
areas, and nothing else exists with that broad a level of coverage. That
bias might even be less important than the biases we have in meta-analyses
of the published literature because of the tendency to do short, summer-only
studies of colorful, very convenient coral communities in (mostly) shallow,
clear waters with low wave action. 

7. For your purposes, the surveys of the Hawaii-based NOAA folks  probably
come the closest to providing reasonable norms of comparative analysis.
However, it is still important not to infer dynamics that one knows little
about, except at the scale of hundreds to thousands of miles and 10's to
hundreds of reefs. The exceptions would be obvious, nearly unmistakable
conclusions based on comparing living to dead corals, or small to large fish

So, is the future of static analyses equally bleak? Not necessarily. Some
species and/or growth forms are more susceptible to certain stressors than
others. For example, natural silty water coral communities have particular
coral communities with species and growth forms that can easily expel
sediments. Some corals are clearly more susceptible to bleaching than
others, as are the Symbiodinium that they support. Once we know more about
these and many other types of differential responses in corals, forams,
algae, sponges, etc. we should be able to develop stress indices based on
recent reductions in particular species or growth forms, and/or on analyses
of Symbiodinium. Similarly, the in-situ, non-invasive analysis of coral
tissue could be used to identify effects of some stressors, though large
samples in space and time may be necessary to rule out temporary, very local
issues. We are not quite there yet.  

Finally, the most useful thing to do currently in general assessment surveys
is to apply a critical, CSI-like assessment. There are recent books and
reports available on this, and old papers on general assessment surveying.
One must look for clues as to what is going on and has happened recently,
and rule out alternative explanations one-by-one. That may include a wide
range of analytical approaches as needed, such as total nutrient accounting
(in-water measurements are not always enough), studies of breakage patterns,
interviews with local people, etc. One may miss the subtle issues, but
sometimes the major problems become obvious.



John W. McManus, PhD
Director, National Center for Coral Reef Research (NCORE)
Professor, Marine Biology and Fisheries
Coral Reef Ecology and Management Lab (CREM Lab)
Rosenstiel School of Marine and Atmospheric Science (RSMAS)
University of Miami, 4600 Rickenbacker Causeway, Miami, 33149
jmcmanus at rsmas.miami.edu      http://ncore.rsmas.miami.edu/
Phone: 305-421-4814   

"Far better an approximate answer to the right question, which is often
   than an exact answer to the wrong question, which can always be made
     --John Tukey, Statistician, National Medal of Science and IEEE Medal of



So, is it impossible to make static assessments without

-----Original Message-----
From: coral-list-bounces at coral.aoml.noaa.gov
[mailto:coral-list-bounces at coral.aoml.noaa.gov] On Behalf Of Jeffrey Maynard
Sent: Friday, June 08, 2012 11:46 AM
To: coral-list at coral.aoml.noaa.gov
Subject: [Coral-List] Habitat condition metrics for coral reefs - is there a


Colleagues and I have recently conducted fieldwork in Saipan, CNMI.  We are
now interested in calculating a habitat condition metric for the reefs we
surveyed that can be compared among our survey sites.  Lit and web-based
reviews and research strongly suggest most published information on coral
reef 'habitat condition metrics' comes from the Caribbean.  Some of those
could be useful to us but we can't tell whether there's a 'convention' for
this right now, either globally or for the Indo-Pacific.

Please get in touch with me if you know of a formula to calculate a habitat
condition metric for coral reefs that you think represents 'best practice'.

Thank you.

Looking forward to responses,

-Jeff Maynard.

Coral-List mailing list
Coral-List at coral.aoml.noaa.gov

More information about the Coral-List mailing list