Calculating species distribution change

[Acidulus Duskwalker]

Hey everyone,

I just used Maxent GUI (3.3.3k) and the package 'dismo' in R to predict the distribution of my species over the course of four major time periods using climate data from CGCM-3 (Bioclim). I do have output .asc files that Maxent gives me from both the methods, and I can import them to ArcGIS and visualize them well.

My question is, for interpreting the data, how do I calculate the number of cells that my species can occur in (maybe using a threshold for the maxent output).

Thanks in advance.

[Husam El Alqamy]

You are right you have to use thresholded output of maxent. Convert this from ascii to raster in arc (do not forget to check the integer raster option in arc) , then convert this raster to polygon (uncheck simplify polygons) and then calculate the area of all polygons with gridcode=1. This is the area of your thresholded species distribution model.
Regards

--
You received this message because you are subscribed to the Google Groups "Maxent" group.
To unsubscribe from this group and stop receiving emails from it, send an email to maxent+un...@googlegroups.com.
To post to this group, send email to max...@googlegroups.com.
Visit this group at http://groups.google.com/group/maxent.
For more options, visit https://groups.google.com/d/optout.

[romunov]

When you predict in `dismo`, a raster is created. You can find your values using regular R subsetting and counting of elements. Here's a small example

library(raster) # load package raster

r <- matrix(1:9, nrow = 3) # create a 3x3 matrix and populate it with numbers 1 through 9

r <- raster(r) # convert it into raster - this is where you are when you get a prediction

plot(r) # plot it to see how it looks

length(r[r > 5]) # find values in `r` that are larger than 5

Cheers,

Roman

--

You received this message because you are subscribed to the Google Groups "Maxent" group.
To unsubscribe from this group and stop receiving emails from it, send an email to maxent+un...@googlegroups.com.
To post to this group, send email to max...@googlegroups.com.
Visit this group at http://groups.google.com/group/maxent.
For more options, visit https://groups.google.com/d/optout.

--

In God we trust, all others bring data.

[Jamie M. Kass]

Both these strategies work, but I'd only like to add that if your predictor rasters were not projected to a datum in meters (geographic coord sys like WGS84) you'll need to first project your prediction raster to a projected coord sys like UTM before you can calculate area in sq meters. Then it's just simply counting the number of cells (i.e. with cellStats() in R or raster properties in ArcGIS), then multiplying that by the projected cell size squared to get sq m.

I also want to add that if you interpret this thresholded raster as the area your species can occur in, presumay using the logistic threshold, there are various assumptions you are making. Please read through Merow et al. 2013 in Methods in Ecology & Evolution before affirming your interpretation.

Jamie Kass
PhD Student
City College of NY