Analysis
Open the spreadsheet and create a scattergraph of the number of species (y) against the area sampled (x). Now add a line of best fit (trendline function under chart), choosing the option to include the equation for the line on the figure.
Your choice of trendline needs to provide the terms for the species-area relationship:
S = cAz
that is
S = predicted number of species
c = the number of species found in one unit
(here this is 1m because of the scale)
A = Area
z = the coefficient relating A to S, the rate at which S increases with A.
From this, you will know how many species are expected in a single square metre and the value of z. You may then compare the value of z with that of previous studies cited in Chapter 9, which gives the expected range.
Calculate the expected value of S in an equivalent habitat of 100m2 and also for 1 hectare.
The R-squared value option gives an estimate of the amount of variation in S that can be attributed to A. Take the square root of this to derive the correlation coefficient (r), which gives a measure of the closeness (and significance) of the line - values close to 1.0 show a high measure of correlation.
Further analysis
Consider how you may assess the effect of isolation of the sampling area on the value of z, given there are similar habitat patches scattered through the Corbières.
How do you expect S to change if the rocky outcrops and patches of bushes had been part of the samples? Suggest what abiotic changes might be associated with areas of bushes or rocky outcrops which might explain the very different plant communities in these locations.
The same sort of survey could be done with any group of organisms - lichens, birds, fish, insects - and you may be able to devise equivalent surveys to test the species-area relationship for one of these groups in habitat you can survey yourself.
