More common and more tractable, of course,

because we reduce the number of species included.

And a community can also be defined by a consistent special boundary.

So definition of community depends so how we design the studies,

and the sampling aimed as we want to understand and we want to calculate.

There are different aspects of bio diversity.

Before starting our research study, we need to ask ourselves what can we measure?

There are different possibilities.

We can measure richness of species, abundances of species, diversity that is

a combination or relationship between richness and abundances.

Guild, perfect structure, evolutionary diversity with each species diversity.

Such as genetic morphological diversity, and other features.

Then we need to ask how to summarize and describe nature,

because we got near infinite number of things to record.

So we need to simplify.

The simplification is dictated by the experimental question, the location,

the taxon, and he sample or real the sub sample from nature.

It depending on to choose an aspect of biodiversity, to choose a location,

to choose a life stage etc.

The sampling effort is just the number of samples we need to collect to insure

a correct sampling of the community.

It depends on the economic resources we have available,

the human resources, time and question to answer.

It has been suggested to kind of rule of ten.

So it means that collectedly stem samples, plot, grades, etc, for

each treatment would be enough, sufficient to provide a very good research study.

Three, of course, is the magic minimal number in statistics, but

is usually is a too low for ecology.

So how to sample the data?

We need first to keep data separately

because merger data cannot be separated later.

That's very important point.

If you look for viability, you need statistic.

That's the second very important point.

But for doing statistic,

if you merge raw data, you cannot perform any kind of statistic.

So, for instance,

if your sample builds per day, you need to take information on the hour.

Otherwise, later,

you can have this information but only together with day by day.

To sample diversity, you need to remember that less sample you get,

more biases you of course you have.

So biases is just systematical differet from the population parameter of interest.

So it's important to get increase the number of samples to approach the real

numbers of species.

If you are evaluating treatments, you need less samples, but

if you are evaluating the total diversity, you need more samples.

So how to compare communities based on samples?

You need to use abundances.

That's the basic information for each taxon for

instances if you are collecting species, genes, family, etc.

All for any operational taxonomic unit.

Then you need to show this biodiversity data.

And tere are different ways to show.

I will not explain in details now, because this is argument of next lecture.

But I was just to introduce which system we have to explain and

to show biodiversity data.

First is the rarefaction and accumulation curves.

These curves just compare different samples each other, and makes information,

and provide information how different are these samples to each other.

And which one is the richest, which one is the poorest, etc.

Another way is the log normal of distribution of data.

We can have different distribution of our abundance per species, and

these can be shifted on the left.

It means that we are missing array of species.

They can be balanced.

So they can follow a log normal curve,

it means that we have a good representation of common and rare species.

Another way to show biodiversity data is the rank abundance plot.

The rank abundance plot is a plot that only shows the abundance ranks of species,

and on the ordinate, the proportional abundance of species.

And this is another way to compare different samples.