sampling design

Richard B. Aronson raronson at
Fri Jul 16 12:33:45 EDT 1999

Dear Bridget,

As always, the approach you take will depend on the hypothesis you wish 
to test.  Articulating the specific hypothesis is the only way to answer 
your questions about sampling design.


Rich Aronson

Richard B. Aronson
Senior Marine Scientist
Dauphin Island Sea Lab
101 Bienville Boulevard
Dauphin Island, AL 36528

Voice: (334) 861-7567
Fax: (334) 861-7540
email: raronson at

On Thu, 15 Jul 1999, Bridget Elliott wrote:

> Hullo
> Is anyone out there sampling a representation of patchily distributed reefs along a stretch of coast along several hundred kilometres ?
> I am looking for suggestions for the sampling design of a monitoring program, when reefs are of a wide range of sizes and depths, and patchily distributed along the coast. One source has suggested dividing the coast into equal sectors of say, 50 km, and doing a fixed number of transects for a range of depths regardless of the amount of reef in that sector. This would allow for direct comparisons between the reefs (sectors) for variables such as species abundance and composition, with equal replications.
> However, surely if one sector only has a few square metres of scrappy reef, it does not justify the same intensity of sampling as a sector with a lot of reef ?
> In the southern part of our province the reefs are typical temperate rocky substrate, graduating to a layer of sub-tropical corals growing on top of submerged Pleistocene dunes in the north. I hope to use line-transects (GCRM method) to establish a long-term monitoring program of the benthos.
> Are there any biologists or statisticians out there who can help ?
> Many thanks
> Bridget
> Bridget Elliott
> Marine Ecologist
> KwaZulu-Natal Nature Conservation Service
> Private Bag X3
> Congella, Durban
> 4013
> South Africa
> phone: +27 - 31 - 2051271
> fax: +27 - 31 - 2051547
> email: b.elliott at
-------------- next part --------------
An HTML attachment was scrubbed...

More information about the Coral-list-old mailing list