Hello there,

I am estimating the amount of plant litter on the soil in forests using quadrats. I have already literally 1000's of data points.

What I want to do, merely as an exercise, is determine the sample size needed to cover the variability I've seen within defined confidence intervals (e.g. 95%). For instance, what would be the minimum sampling effort required to get the current estimate I have of the mean, using this method? As part of this I want to know how many samples within this population are needed to see the variance stabilize/reduce to an acceptable level.

Can anyone outline an accepted way of doing/visualizing this. I do use R programing to a limited extent.

Many thanks! :)