Interactions between environmental variables

106 views
Skip to first unread message

Jessica Junker

unread,
Dec 10, 2009, 7:42:00 AM12/10/09
to Maxent
Dear Maxent users!

I am trying to model great ape habitat distribution across range
countries in Africa. As one of my environmental variables, I would like
to include distance to rivers. I expect that in areas where human
densities are high (human density is another one of my environmental
variables), distance to rivers is negatively correlated with ape
occurrence probability as rivers are used as the main transport pathways
in some countries (especially in central Africa) and I expect poaching
intensity close to rivers to be higher than further away from rivers.
However, in drier range countries, such as in e.g. Senegal, Mali or
Guinea, I expect ape occurrence probability to be higher closer to
rivers than further away from rivers, or perhaps not to change with
distance to rivers.

My question is whether Maxent incorporates these (two-way) interactions
between different environmental variables included in the model or
whether, rather than including e.g. human density, rainfall (or
vegetative cover) and distance to rivers as separate variables, I should
include one variable that already captures these interactions between
human densities, distance to rivers and rainfall/ vegetative cover?

Another example: as a proxy for poaching intensity, I thought of
including religious belief (on a sub-country level), because for
example, consuming ape meat is a taboo in Muslim communities, but not in
Christian communities. So I expect ape occurrence probability to be
higher in the former than the latter communities. Again, this
relationship may hold for some countries, but not for others (e.g. in
areas with relatively low human densities, there may be no effect, such
as in large areas in central Africa). In Sierra Leone for example,
poaching intensities seem relatively low in some areas, which may be due
to the weapon ban imposed on the country - I thought of thus also
including national boundaries as a separate environmental variables, as
politic and socio-economic factors may play a role in this context.

Does anyone know whether just including these variables separately into
the model will be sufficient to capture these interactions?
I am looking forward to your response and I thank you for your help!

Jessi

--
Save the great apes!
Sign the mAn: http://www.apesmanifesto.org

Jessica Junker
Max-Planck Institute for Evolutionary Anthropology
Department of Primatology
Deutscher Platz 6
04103 Leipzig
Germany

tel: +49 341 3550 217
fax: +49 341 3550 299
cell: +49 0160 90594996
email: jessica...@eva.mpg.de

A.P.E.S. Database
http://apes.eva.mpg.de/

francis mwambo

unread,
Dec 10, 2009, 10:37:36 AM12/10/09
to max...@googlegroups.com
Dear Jessica,

Your research sounds very interesting. I happen to be also doing a modelling on human and cliamtic change impacts on Chimapnzees in Cameroon using Maxent. Your choice for variables sounds interesting. But my view is how would you input these variables for instance religion, politics, ban of of weapons in to maxent? I may suggest that if you wish to use religion as a variable you may better use culture beliefs as a variable instead. My opinion on this is that in some densely populated communities you may find muslims and christians mixed. Evident from may be the number of musks or churches one may find around especially in big cities. However, culture may be input as a categorical variable where the map of the country in question is subdivided into smaller regions based on the culture. This kind of subdivision may be similar to the kind one get from a landcover classification. In this case, the culture is input as a categorical variable with different ethic groupings.  I wish you the best in your research.


Francis Mwambo.

email:mwa...@gmail.com
ISEGI-UNL
Lisbon-Portugal

Martin Damus

unread,
Dec 10, 2009, 2:08:33 PM12/10/09
to Maxent, max...@googlegroups.com
Hello Maxent architects,

Below is the error log from my run. Maxent says that a file is missing when it is not. I have no idea what to do about this. Tried running twice after fresh start-up and got the same error. I've checked the directory and the file is there, with the exact name that it should have and with which it ought to be found.

Out of curiosity I deselected the supposed "missing variable" from another training run and Maxent then told me the next variable was missing from the projection data folder. It seems to reject the second variable, not the first. I've checked the file names and there are no leading blanks or anything like that.

Hopefully this can be solved!

Cheers,
Martin Damus

Thu Dec 10 13:44:14 EST 2009
MaxEnt version 3.3.0
Checking header of C:\documents\Maxent\CurrentClimateRhagoletisMendax\alt_Rmen.grd
Checking header of C:\documents\Maxent\CurrentClimateRhagoletisMendax\prec10_Rmen.grd
Checking header of C:\documents\Maxent\CurrentClimateRhagoletisMendax\prec11_Rmen.grd
Checking header of C:\documents\Maxent\CurrentClimateRhagoletisMendax\prec12_Rmen.grd
Checking header of C:\documents\Maxent\CurrentClimateRhagoletisMendax\prec1_Rmen.grd
Checking header of C:\documents\Maxent\CurrentClimateRhagoletisMendax\prec2_Rmen.grd
Checking header of C:\documents\Maxent\CurrentClimateRhagoletisMendax\prec3_Rmen.grd
Checking header of C:\documents\Maxent\CurrentClimateRhagoletisMendax\prec4_Rmen.grd
Checking header of C:\documents\Maxent\CurrentClimateRhagoletisMendax\prec5_Rmen.grd
Checking header of C:\documents\Maxent\CurrentClimateRhagoletisMendax\prec6_Rmen.grd
Checking header of C:\documents\Maxent\CurrentClimateRhagoletisMendax\prec7_Rmen.grd
Checking header of C:\documents\Maxent\CurrentClimateRhagoletisMendax\prec8_Rmen.grd
Checking header of C:\documents\Maxent\CurrentClimateRhagoletisMendax\prec9_Rmen.grd
Checking header of C:\documents\Maxent\CurrentClimateRhagoletisMendax\tmax10_Rmen.grd
Checking header of C:\documents\Maxent\CurrentClimateRhagoletisMendax\tmax11_Rmen.grd
Checking header of C:\documents\Maxent\CurrentClimateRhagoletisMendax\tmax12_Rmen.grd
Checking header of C:\documents\Maxent\CurrentClimateRhagoletisMendax\tmax1_Rmen.grd
Checking header of C:\documents\Maxent\CurrentClimateRhagoletisMendax\tmax2_Rmen.grd
Checking header of C:\documents\Maxent\CurrentClimateRhagoletisMendax\tmax3_Rmen.grd
Checking header of C:\documents\Maxent\CurrentClimateRhagoletisMendax\tmax4_Rmen.grd
Checking header of C:\documents\Maxent\CurrentClimateRhagoletisMendax\tmax5_Rmen.grd
Checking header of C:\documents\Maxent\CurrentClimateRhagoletisMendax\tmax6_Rmen.grd
Checking header of C:\documents\Maxent\CurrentClimateRhagoletisMendax\tmax7_Rmen.grd
Checking header of C:\documents\Maxent\CurrentClimateRhagoletisMendax\tmax8_Rmen.grd
Checking header of C:\documents\Maxent\CurrentClimateRhagoletisMendax\tmax9_Rmen.grd
Checking header of C:\documents\Maxent\CurrentClimateRhagoletisMendax\tmin10_Rmen.grd
Checking header of C:\documents\Maxent\CurrentClimateRhagoletisMendax\tmin11_Rmen.grd
Checking header of C:\documents\Maxent\CurrentClimateRhagoletisMendax\tmin12_Rmen.grd
Checking header of C:\documents\Maxent\CurrentClimateRhagoletisMendax\tmin1_Rmen.grd
Checking header of C:\documents\Maxent\CurrentClimateRhagoletisMendax\tmin2_Rmen.grd
Checking header of C:\documents\Maxent\CurrentClimateRhagoletisMendax\tmin3_Rmen.grd
Checking header of C:\documents\Maxent\CurrentClimateRhagoletisMendax\tmin4_Rmen.grd
Checking header of C:\documents\Maxent\CurrentClimateRhagoletisMendax\tmin5_Rmen.grd
Checking header of C:\documents\Maxent\CurrentClimateRhagoletisMendax\tmin6_Rmen.grd
Checking header of C:\documents\Maxent\CurrentClimateRhagoletisMendax\tmin7_Rmen.grd
Checking header of C:\documents\Maxent\CurrentClimateRhagoletisMendax\tmin8_Rmen.grd
Checking header of C:\documents\Maxent\CurrentClimateRhagoletisMendax\tmin9_Rmen.grd
Reading samples from Rhagoletis_mendax_for_Maxent.csv
Read samples: max memory 1352663040, total allocated 5177344, free 1613736, used 3563608, increment 3563608
Extractor: max memory 1352663040, total allocated 5177344, free 1536640, used 3640704, increment 77096
Extracting random background and sample data
Time since start: 48.716
344895 points with values for all grids
Adding samples to background in feature space
Command line used:
Command line to repeat this species model: java density.MaxEnt -r -a nowarnings noprefixes -E "" -E null outputformat=logistic outputdirectory=C:\documents\Maxent\RMendaxCCCMAa2a projectionlayers=C:\documents\Maxent\A2a2020RhagoletisMendax samplesfile=C:\documents\Maxent\Rhagoletis_mendax_for_Maxent.csv environmentallayers=C:\documents\Maxent\CurrentClimateRhagoletisMendax replicatetype=crossvalidate
Species: R.mendax
Layers: alt_Rmen prec10_Rmen prec11_Rmen prec12_Rmen prec1_Rmen prec2_Rmen prec3_Rmen prec4_Rmen prec5_Rmen prec6_Rmen prec7_Rmen prec8_Rmen prec9_Rmen tmax10_Rmen tmax11_Rmen tmax12_Rmen tmax1_Rmen tmax2_Rmen tmax3_Rmen tmax4_Rmen tmax5_Rmen tmax6_Rmen tmax7_Rmen tmax8_Rmen tmax9_Rmen tmin10_Rmen tmin11_Rmen tmin12_Rmen tmin1_Rmen tmin2_Rmen tmin3_Rmen tmin4_Rmen tmin5_Rmen tmin6_Rmen tmin7_Rmen tmin8_Rmen tmin9_Rmen
Layertypes: Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous
outputformat: logistic
outputdirectory: C:\documents\Maxent\RMendaxCCCMAa2a
projectionlayers: C:\documents\Maxent\A2a2020RhagoletisMendax
samplesfile: C:\documents\Maxent\Rhagoletis_mendax_for_Maxent.csv
environmentallayers: C:\documents\Maxent\CurrentClimateRhagoletisMendax
replicatetype: crossvalidate
Command line:

getSamples: max memory 1352663040, total allocated 10080256, free 3277688, used 6802568, increment 3161864
Making features
Time since start: 51.366
makeFeatures: max memory 1352663040, total allocated 74948608, free 9692440, used 65256168, increment 58453600
R.mendax:
Regularization values: linear/quadratic/product: 0.050, categorical: 0.250, threshold: 1.000, hinge: 0.500
Time since start: 53.076
152 samples
Density: max memory 1352663040, total allocated 125132800, free 39044880, used 86087920, increment 20831752
linearPredictor: max memory 1352663040, total allocated 125132800, free 38963680, used 86169120, increment 81200
FeaturedSpace: max memory 1352663040, total allocated 125132800, free 38963680, used 86169120, increment 0
Sequential: max memory 1352663040, total allocated 125132800, free 38882744, used 86250056, increment 80936
Initial loss: 9.225031920719173
Time since start: 87.832
500: time = 34.144000 loss = 7.573372
Time since start: 90.718
Projecting...
Writing file C:\documents\Maxent\RMendaxCCCMAa2a\R.mendax.asc
Writing file C:\documents\Maxent\RMendaxCCCMAa2a\R.mendax.asc
Time since start: 147.559
Writing C:\documents\Maxent\RMendaxCCCMAa2a\plots\R.mendax.png
Time since start: 155.73
Projecting...
Layer prec10_Rmen is missing from C:\documents\Maxent\A2a2020RhagoletisMendax
java.lang.Exception
at density.Utils.popupError(Utils.java:155)
at density.Utils.getGridAbsolutePath(Utils.java:62)
at density.Project.newGrid(Project.java:72)
at density.Project.getGrid(Project.java:41)
at density.Project.projectGrid(Project.java:178)
at density.Project.doProject(Project.java:87)
at density.Runner.start(Runner.java:508)
at density.GUI$7.construct(GUI.java:317)
at density.SwingWorker$2.run(SwingWorker.java:108)
at java.lang.Thread.run(Unknown Source)
Time since start: 210.265
Fatal Error: java.lang.NullPointerException
java.lang.NullPointerException
at density.Extractor.CachedGridFilename(Extractor.java:51)
at density.Project.newGrid(Project.java:73)
at density.Project.getGrid(Project.java:41)
at density.Project.projectGrid(Project.java:178)
at density.Project.doProject(Project.java:87)
at density.Runner.start(Runner.java:508)
at density.GUI$7.construct(GUI.java:317)
at density.SwingWorker$2.run(SwingWorker.java:108)
at java.lang.Thread.run(Unknown Source)

__________________________________________________________________
The new Internet Explorer® 8 - Faster, safer, easier. Optimized for Yahoo! Get it Now for Free! at http://downloads.yahoo.com/ca/internetexplorer/

Martin Damus

unread,
Dec 10, 2009, 2:30:47 PM12/10/09
to max...@googlegroups.com
Thought I should try the newest version 3.3.1 -- but, the same error occurs there too. Even if I pare the number of variables to 5, it does not "find" the second variable in the projection data file.

Martin

--- On Thu, 12/10/09, Martin Damus <dam...@yahoo.com> wrote:
> --
>
> You received this message because you are subscribed to the
> Google Groups "Maxent" group.
> To post to this group, send email to max...@googlegroups.com.
> To unsubscribe from this group, send email to maxent+un...@googlegroups.com.
> For more options, visit this group at http://groups.google.com/group/maxent?hl=en.
>
>
>


__________________________________________________________________
Looking for the perfect gift? Give the gift of Flickr!

http://www.flickr.com/gift/

Martin Damus

unread,
Dec 10, 2009, 2:38:10 PM12/10/09
to max...@googlegroups.com
Ah nuts. Mea culpa. Sorry folks for clogging up the emails. One directory of files had an additional underscore in every filename that I did not notice. I figured it out finally after trying to project back on the original training environmental layers, which worked just fine.

Sorry.

(sheepishly bowing his head) Martin

--- On Thu, 12/10/09, Martin Damus <dam...@yahoo.com> wrote:

> From: Martin Damus <dam...@yahoo.com>
> Subject: Illogical error in Maxent 3.3.0
> To: "Maxent" <Max...@googlegroups.com>, max...@googlegroups.com
> Received: Thursday, December 10, 2009, 2:08 PM
Reply all
Reply to author
Forward
0 new messages