SWD projecting

[mike]

I'm trying to project the current land cover of the American southwest
into the future under different climate change scenarios. However,
the Maxent manual doesn't go into enough detail on how to create SWD
files for the occurrence points and environmental layers. Could
somebody please help me by going into detail on how to create the SWD
files and projecting using Maxent?


Thanks,

Mike

[Julian Ramirez]

Hi Mike,

Not sure if this confuses you more, or if i didnt get your point. You may not use MaxEnt to create an SWD file because the software is only for species distribution modeling, not for creating input data. You have to provide all input data for running MaxEnt (asciis, grds, csv, etc). What you need to do to create an SWD file is to extract the climate data for each one of the sample points with which you're modeling (you can do this using GIS software) and then create a .csv file using either a spreadsheet or an advanced textfile editor (i.e. UEStudio, UltraEdit, etc). Now you can use this file as input in your MaxEnt modeling step.

An SWD file is however not indispensable for running MaxEnt: you can only use (1) asciis (ESRI ascii grids) or grd (DIVA-GIS grids) with climatic data (for projecting you need two different climatic datasets: current and future), (2) a file containing occurrence data (species, longitude, latitude). You must load current conditions folder in "Environmental layers", occurrence data points in "Samples" and future climatic data in "Projection layers directory". Current and future conditions need to have the same grid characteristics (extent, cellsize, etc) and need to be the same variables. You can play around with the modeling options in the software to test results. 

If you have climatic gridded data then you can create asciis easily by using either ArcGIS or the Arc/Info console. You need to perform a new maxent run if you want to process a different climate change scenario (changing only projection layers, and leaving environmental layers as current conditions)

Hope this helps,

Julián R. V.

Research Assistant

International Centre for Tropical Agriculture

Cali, Colombia

[wilbersa]

Hi All,
Mike, a good alternative is to use DIVA GIS to extract the values of
environmental variables for you samples. Diva gis is not just free,
but it's also very friendly to use. Then, you could use a text editor
or even Excel to save your files as *.cvs. Cvs is the format MAXENT
uses to read docs.
Now I got a question for you (actually everyone). According to
MAXENT's docs, it is possible to use swd files instead of the normal
samples files and environmental layers. If the researcher does it,
runs are like light speed way faster. Is it also possible to use swd
files to project the species distribution into other climate
scenarios?. I mean, to use also a swd background data (random points
associated to values from environmental layers from the new climate
scenario) to project species distributions instead of grid files
representing the new climate scenario.

Cheers
R.

On Nov 6, 2:18 am, "Julian Ramirez" <dawnpatrolmusta...@gmail.com>
wrote:

[Sam Veloz]

Yes you can project to future( past) climate data in swd format. Just
put the file you want to project to in the test sample spot. If you want
a map though, you need the ascii's of the climate layers.
Sam

[Steven Phillips]

Sam and R.,

You can also use a .swd format file in the "Projection layers
directory/file" entry.

-- Steven

Date: Thu, 06 Nov 2008 07:18:58 -0800
From: Sam Veloz <swam...@sbcglobal.net>