Create Virtual Species in ModestR with Environmental Variables

MODESTR QUICK TUTORIALS HTTP://WWW.IPEZ.ES/MODESTR/
Step by step tutorial:
Creating virtual species
with ModestR

What do you need for this tutorial:
1. ModestR 6.4 or later
2. Internet connection
3. About 25 minutes

We’ll describe how to create virtual
species using ModestR.
Follow the next steps!

First of all it is important to know how the virtual species generation works. This ModestR feature actually
relies on the virtualspecies R package developed by Leroy et al. (2015) So the parameters you will be able to
setup in ModestR to generate species are explained in this package documentation.
It is important to remark that currently, virtual species generated from ModestR use the following settings of
the virtualspecies package:
• Virtual species are generated using response functions randomly selected among the included in the
virtualspecies package (see 2. First approach: generate virtual species distributions by defining response
functions)
• Probabilistic conversion and user-specified species prevalence are used to generate presence/absence
maps (see 4. Conversion of environmental suitability to presence-absence)
• Occurrence sampling is done in the whole area of presence of each species (see 7. Sampling occurrence
points)
You can fin more information in the online tutorial set by the author:
http://borisleroy.com/files/virtualspecies-tutorial.html
Also in the following paper:
Leroy B, Meynard CN, Bellard C, Courchamp F (2015). “virtualspecies, an R package to generate virtual species
distributions.” Ecography. doi: 10.1111/ecog.01388.

Run ModestR DataManager program. The virtual species generation feature can be used independently from any
database. But if you expect to import the virtual species to a database, you’d rather create a new database first.
Alternatively you may open an existing database to import the virtual species to.
To create a new
database go to menu
File/New taxonomy
database.
ModestR will ask you
for a file name then
create an empty
database.

Whether you created a new database, opened an existing one, or none of them, the first step to generate virtual species is
going to menu Mapping/Virtual species generation.

The next step is selecting the environment variables to be used to
generate the species. At least two variables are required. The more
variables used, the more complex is the generation process (thus
more time-consuming).
The variables to be used must have been integrated in ModestR
previously. An explanation about integrating environmental data in
ModestR can be found in the tutorial How to integrate and use
environmental data in ModestR.
You can choose to use the full geographical extent covered by the
selected variables (for example WorldClim variables extent includes
all land areas worldwide). Or you may select a specific extent by
selecting a shape that corresponds to a geographic area. ModestR
shape concepts are explained in this tutorial:
Working with shapes. Searching for species in any area. Assessing
environmental conditions.
Then click on “Continue”.

In the next step the selected variables are shown
with the % of common coverage. That is, how much
geographical extent they have in common. If you
selected variables that cover the same geographic
areas, this % will be typically near to 100%. A very
low coverage indicates that some of the selected
variables don’t cover the same geographic areas.
For example if you selected variables from land
areas and other ones from marine areas. Those data
are merely informative, but can allow to detect
mistakes in the variable selection.
You may optionally calculate VIF (variance inflation
factor) to detect possible collinearities between the
variables and, eventually, remove some of them or
go back to change your selection.
Then click on “Continue”.

In the next step you can set several parameters about virtual species to
be generated, such as:
• Number of virtual species to be created.
• Number of variables to be used to generate each species: for
example, if you selected 5 variables in the previous steps, here you
can indicate a value from 2 to 5 variables. If you select 3, for
example, then for each species 3 variables will be randomly picked
among the 5 initially selected, to use them to generate the species.
• Species prevalence: you can set a prevalence range. The prevalence
for each species will be a random value within this range. (see
virtualspecies package documentation for more details on this
parameter).
• Generate occurrences: virtual species generation always generate a
suitability map and a presence map. If you also want to generate
occurrences to simulate a random sampling, enable this option and
set the range of occurrences to be sampled for each species. You
can choose to use suitability as sampling bias (see virtualspecies
package documentation for more details on this parameter).
(continue on next slide)

(continuation from previous slide)
• Import options: if you have a database currently opened in
DataManager (see first steps of this tutorial) you can select to
import the virtual species to be generated. You can import:
• Presence (binary) maps: they will be imported as range maps.
• Occurrences: they will be imported as occurrences.
• Valid habitats: if species will be imported to ModestR, valid habitats
must be indicated.
• Base output folder: select a folder where detailed results will be
stored.
• CSV style: select a CSV style, between Spanish (decimal comma for
numbers and semicolon as separator) or English (decimal point for
numbers and comma as separator)
Once all parameters set, press “Start”.

The virtual species generation will start. Depending on the parameters and the number of species to generate, this may
take some time.
If you also selected to import virtual species to the current database, the importation will begin once all species created.
Finally, a dialog will indicate that the process is completed. To see a detail of the results, click on “Open folder”.

The content of the output folder typically contains the following
elements:
• InputVariables: in this folder are stored the environmental
variables you selected to create the virtual species.
• Presence: this folder contains the presence/absence map for each
virtual species in ASC raster format.
• Suitability: this folder contains the suitability map for each virtual
species in ASC raster format.
• generatespeciesscript.r : the script used to generate the virtual
species. You can run it with R, Rstudio, etc. (you may need to
install some packages in R to be able to run it)
• R_Output_Log.txt and R_Messages_Log.txt: text output, warnings
and error messages generated by R during the process. In case of
errors you may examine those files for more information.
• Report.CSV: this report contains a table with the variables used to
generate each virtual species, the response formula, etc.
• VirtualSpeciesOccurrences.CSV: if occurrence generation was
choosen, this file contains the generated occurrences.

As previously said, the Presence folder contains the presence/absence map for each virtual species in ASC raster format.
You can open those files using MapMaker or any other application to view rasters (i.e. R with raster package, or any GIS
program). As shown in this figure, those rasters have a value of 0 where the species is absent, and 1 where it is present.c

Similarly, the Suitability folder contains the suitability map for each virtual species in ASC raster format. As shown in this
figure, those rasters have values in the [0-1] range, being 0 an area not suitable at all for the species and 1 an area totally
suitable.

If you previously choose in DataManager to import the virtual species to the current database, they will appear in the
taxonomy tree under the “Virtual_[level]” taxonomy, which will be automatically added.

If you selected to import presence data for each species, opening any of the imported maps you will find a range map
with the presence areas of the species.

If you also selected to import occurrences they will be also in the species maps.

It was the Step by step tutorial:
Creating virtual species with ModestR
Thank you for your interest.
You can find this one and other tutorials in http://www.ipez.es/ModestR
By the ModestR team

Create Virtual Species in ModestR with Environmental Variables

Recommended

Recommended

More Related Content

What's hot

What's hot (12)

Similar to Create Virtual Species in ModestR with Environmental Variables

Similar to Create Virtual Species in ModestR with Environmental Variables (18)

Recently uploaded

Recently uploaded (20)

Create Virtual Species in ModestR with Environmental Variables