Bounding Box Error

Erin

unread,

Nov 8, 2009, 8:53:51 PM11/8/09

to Maxent

I'm trying to run Maxent with about 7000 presence points and 8
environmental variables. I've extracted all rasters to the same
extent, used snap raster, and projected all of my data several times
using Arc9.3.1. Every time I try to run Maxent, it gives the same
error: '[lat], [long], is outside the bounding box of environmental
data, skipping'. It does this for every sample point, it will then
run, and give me an error message at the end about there being no
training samples. Then, there are no outputs.

All of the ascii headers in the environmental variables match exactly.
All projections are exactly the same. I'm not sure if this is a
problem with my sample point data or the environmental variables. Any
suggestions?

Thanks,
Erin

Husam El Alqamy

unread,

Nov 10, 2009, 8:13:25 AM11/10/09

to max...@googlegroups.com

Dear Erin
You should examin your raster data with a text editor to see whats wrong. I
think that you most probably doing a very simple mistake in your presence
.csv file. Make sure that your data starts by the species name, x
coordinate(longitude), and last the Y coordinate(latitude). It might be that
you are putting the Y coor ahead of the X coor and MAXENT is assuming the
opposite which makes your points out of the bounding box. Hope this helps
Regards

Husam El Alqamy, B.Sc., M.Phil.
Protected Area Coordinator,
Terrestrial Environmental Research Center, TERC
Environmental Agency Abu Dhabi
Antelope Specialist Group - ASG, IUCN.

Erin

unread,

Nov 11, 2009, 3:29:13 PM11/11/09

to Maxent

I found the problem was that when I was exporting my presence file
from Arc, the coordinates were coverted to decimal degrees, even
though the projection was in UTM. I converted these to meters, and it
fixed the problem.

Erin

tmlehman

unread,

Dec 1, 2015, 4:24:09 PM12/1/15

to Maxent

I am having this same problem. My environmental data is in UTM and so were my sample points. After I received this error the first time, I converted them to lat/long based on some info I read in the help section of the program. I am still getting the same error. Any suggestions?

Idoia

unread,

Oct 12, 2016, 7:55:00 PM10/12/16

to Maxent

HI!! I'm having the same problem, I also changed them into lat/long and introduce the data like this "species, longitude, latitude". Did you solve it?

tmlehman

unread,

Oct 13, 2016, 2:30:17 PM10/13/16

to Maxent

If my memory is correct, I was getting this particular error because the Extract by Mask Tool did not run correctly, even though ArcMap said it did. Import all your environmental ASCII files into ArcMap and double check that the extent and coordinate systems all match. You should not have to convert your sample or environmental layers to lat/long. I have mine set up as Species (Ex. T. carolina), UTMX, UTMY and it worked. Another issue that may cause this error or a similar error is if your environmental layer does not have data for the entire spatial extent. An example is the National Wetland Inventory layer; only wetlands are shown and everything else is NoData. You would need to convert these layers to distance layers using the Distance Toolset under the Spatial Analyst Toolset.

Hope this helps

Rebekah Persad

unread,

Feb 13, 2019, 7:30:38 PM2/13/19

to Maxent

Hi Erin,

How exactly did you convert the coordinates from decimal degrees to meters? If you can recall from 2009.. thank you!