[Coral-List] Best practice for LIT surveys

[Michael Risk]

Dear Bill, colleagues:

I really was going to stay out of this whole debate, until yesterday I
received a message from a dear friend, a senior and very
well-established coral reef scientist, saying (in part):

“I read this and thought of you immediately. It's Jan. 2009, and people
are still chasing their tails on this one. Oh, well - what else is new,
anyway?”

As far as I can make out, I was the first to use this technique (LIT)
on coral reefs (work done in 1967, published as Risk, 1972, Atoll Res
Bull 153). My logic was, a technique that was used to assess grade in
Gold mines would probably work on corals...

In 1977, Terry Done published a study of survey techniques (Proceedings
of the 3rd International Coral Reef Symposium, 9.14) that showed all
methods gave basically the same answer.

In 1997, my daughter and I published a study (Proceedings of the 8th
International Coral Reef Symposium 2, 1471.4) that showed:
-Terry was right, in 1977-all the methods currently in use, applied
properly, give the same answers.
-for management purposes, survey results need only have a precision of
± 20%, and that efforts to achieve higher precision were "a waste of
time and money." (p. 1472).
(For those who are interested-Andrea at the time was a student of Peter
Sale’s. She has now seen the light, and practises law in Kingston.)

In 1999, fed up with some of the things I was observing, I wrote a
little paper called “Paradise Lost” (Mar. Freshwater Res., 1999, 50,
831-837), which received more attention than much of my serious
science. Some parts of that paper are worth quoting here:
“Rapid, cooperative agreement on methodology or, at least, agreement on
the results to be attained, has not been a feature of reef science. No
references will be cited here, but almost three decades after Risk
(1972) and Done (1977), papers are still published that compare survey
methods. This is resumé-padding, not problem resolution.”

I also note, with sadness, that none of the recommendations in Paradise
Lost have been adopted, including this one: “All existing and
contemplated coral reef monitoring
programmes should be abandoned and replaced with community-based
bioindicator programmes designed to identify stress. These
early-warning indicators should be tied to known geochemical techniques
that can fingerprint and date the sources of stress:  information that
feeds directly into policy, legislation, and management.” With even
more sadness, I note that all my dire predictions have come true.

So, some final comments on LIT and monitoring.

In the right hands, LIT is capable of producing superb data. Bill
Allison’s plumb-bob technique is the way to go, and his data are among
the best such anywhere in the world.

Precision in surveys used for management need not be as good as
research surveys. That being said, there is at least one well-known
research group that lies their transect lines directly on the surface
of the reef, following every bump and knob. I need only point out here
that their results are a unique combination of rugosity and diversity,
and their estimates of % cover are not to be believed.

By all means, keep on monitoring. Just don’t get hung up on
methodology, because monitoring isn’t the end-it’s the beginning of the
identification-quantification-amelioration process. And if you monitor,
here are a couple of tips:
-don’t report your results to two decimal places, as entirely too many
of you do. (There seems no general understanding that this implies
field measurements with a precision of 50 microns...)
-if your methods do not include at least a nonparametric assessment of
bioerosion, you are wasting your time-but you won't be wasting much of
mine, because I will give your MS or proposal about 30 seconds.

Mike the Grumpy


On Mon, 5 Jan 2009 09:54:25 -0500
 "William Allison" <allison.billiam at gmail.com> wrote:
> Clarification
> 
> Vera and I seem to be talking about two different aspects of LIT
> surveys.
> 
> If I understand her methods correctly, Vera's results reflect
> variation
> among contiguous but *spatially distinct samples* (ten sections of a
> 50 m
> transect).
> 
> In contrast, my results reflect variation among *repeat surveys of
> the
> "same" spatial location* (i.e., transects re-set as closely as
> possible in
> the same location, using the same stakes etc each time). My results
> demonstrate that repeat LIT surveys results can be both accurate and
> precise, with very high statistical power, if transect lines are
> re-deployed
> with care and surveyed using a plumb-line (1999 International
> Conference on
> Scientific Aspects of Coral Reef Assessment, Monitoring, and
> Restoration,
> NOVA, Fort Lauderdale, Florida, poster and oral presentations). Such
> repeat
> LIT surveys can be useful for detecting change over time, require
> about 30
> minutes per 10 m, have low capital costs and do not depend on
> expensive gear
> that can cease to function (or worse) in the field or lab.
> 
> Vera, I am interested to know what roughly was the average slope of
> the reef
> with respect to the horizontal and if your quadrats were set
> horizontal to
> or in conformity to that slope.
> 
> Sincerely,
> Bill
> 
> 
> 
> 
> 
> 
> On Sun, Jan 4, 2009 at 12:41 PM, Wera Leujak <wera.leujak at web.de>
> wrote:
> 
> > Dear Lindsay,
> >
> > we have tested six commonly used coral survey methods (line
> intercept
> > transect, line point transect, mapping of quadrats, photoquadrats
> analysed
> > in 2 different ways and video belt transects) on a Red Sea reef and
> found
> > LIT to perform poorly in terms of precision and accuracy (Leujak &
> Ormond
> > 2007, Journal of Experimental Marine Biology and Ecology 351, pp
> 168-187)..
> >
> > LIT overestimated hard and soft coral cover by 3 to 6% (live coral
> cover
> > was 34%, predominantly large colonies) and we believe that this is
> due to
> > what we termed the  contour effect . Although our transect line was
> > stretched taught and attached every 5m along the transect the tape
> still
> > tended to follow the contours of larger colonies, thereby
> estimating the
> > circumferential area of the coral rather than the smaller planar
> area (as is
> > the case for quadrats that served as our baseline). Also, using LIT
> in the
> > field was extremely time consuming (10m recorded during 1 dive of
> approx.
> > 60min) and we had difficulties determining what substrate lay
> beneath the
> > transect line, even when using a plumb line. We were often tempted
> to record
> > a coral which  might  be directly below the line in case an error
> is made by
> > omitting to record it.
> >
> > Power analysis indicated that the sample size required to yield an
> 80%
> > chance of detecting even a 20% relative difference in total hard
> coral cover
> > was 135m for LIT, with much greater sample sizes required to detect
> similar
> > differences in the cover of individual coral growth forms. The poor
> > repeatability of LIT was responsible for these high sampling sizes.
> We
> > therefore did not recommend the use of LIT in a monitoring
> programme for Red
> > Sea reefs but suggested photoquadrats instead.
> >
> > Best regards
> >
> > Wera
> >
> >
>
_______________________________________________________________________
> > Täglich 1.000.000 Euro gewinnen! Jetzt kostenlos WEB.DE
> MillionenKlick
> > spielen! https://millionenklick.web.de/?mc=mail@footer.mklick@home
> >
> > _______________________________________________
> > Coral-List mailing list
> > Coral-List at coral.aoml.noaa.gov
> > http://coral.aoml.noaa.gov/mailman/listinfo/coral-list
> >
> _______________________________________________
> Coral-List mailing list
> Coral-List at coral.aoml.noaa.gov
> http://coral.aoml.noaa.gov/mailman/listinfo/coral-list

Mike Risk
Marine Ecologist
PO Box 1195
Durham Ontario
N0G 1R0