Anyway to avoid a bad row in raster data?

[wolf]

I got the following java exception running MaxEnt which I assume is either a bug in the new version or bad data (but it was open source raster data)

Anyone have a workaround for this?

Thanks in advance!

[Jamie M. Kass]

Please include the java exception in your post, and some details about what your inputs were. Thanks.

[wolf]

Hi Jamie,

Thanks for your reply!

I have attached a new error screen shot with the corresponding java errors that I received when executing a new MaxEnt run.  I am thankful for your response and hope you could take a look at this new log file.  I realize the problem is that MaxEnt does not seem to recognize any of the raster points in my "bias" ASC data file, but I am not exactly sure why not.  I have edited the attached log file to remove a bunch of skipped records out of my boundary scope, but those skipped records are not a problem for the solution I am try to get to get to.

P.S.  My original problem was that I didn't realize in my export from ArcGIS that I was running out of disk space and it was truncating my ASC file which is why MaxEnt choked on it.  I fixed that problem.

Thanks again in advance!

[Jamie M. Kass]

The "0 points with values for all grids" in the log file looks like it's the problem. Can you verify that the input coordinate data and the predictor variables are in the same geographic projection? It seems like projection mismatch could be the issue. If the predictors have one kind of coordinates, and the occurrences another, the extraction of values won't work.

Jamie

[wolf]

Nice, I believe you are right!

The sample observation export didn't allow for a coordinate choice, so I guess I need to convert that in ArcGIS and then move it back to a CSV file...  That is unless MaxEnt will allow me to use an ASC format for the sample also.

Thanks much!