maxent_to_resNet is a C++ program developed at the Biodiversity and
Biocultural Conservation Laboratory that constructs a ResNet input
file from the output of the Maxent software package. The latter
consists of one ASCII raster file for each species. Maxent can
generate three types of output: "cumulative", "logistic", and "raw".
maxent_to_resnet is intended for use with the "raw" output format.
Maxent assigns to each pixel of each ASCII raster in "raw" format a
real number between 0 and 1 that represents the pixel's suitability as
habitat for the species. Typically these values are extremely small.
maxent_to_resnet finds the largest value for each species and divides
the value of every other pixel by the largest value. As a result,
there is at least one pixel in each raster with the value of one. Each
ASCII raster becomes one column of the ResNet input file. Following
the division, maxent_to_resnet sets all pixel values that are less
than 0.01 to zero.
As I understand it, Nicolas' question is: "does maxent_to_resnet
set all pixel values greater than the 'minimum training presence'
equal to one?". The minimum training presence is a value computed by
Maxent. The answer to the question is no. I think that the best way to
accomplish this would be to modify the raster for each species using a
GIS then use the modified rasters as input for maxent_to_resnet. In
ArcMap 9.2, the ASCII rasters can be imported via
ArcToolBox>Conversion Tools>To Raster>ASCII to Raster. The rasters can
then be reclassified so that all pixels greater than or equal to the
minimum training presence are set equal to one via Spatial Analyst
Tools>Reclass>Reclassify. The last step would be to export each
modified raster to ASCII via Conversion Tools>From Raster>Raster to
ASCII. Following these operations in ArcMap, each raster will consist
of zeroes and ones only. Thus, the division and thresholding
operations in maxent_to_resnet will not alter the data.
Trevon