r.maxent.lambdas - GRASS GIS manual

NAME

r.maxent.lambdas - Computes raw or logistic prediction maps from MaxEnt lambdas files

KEYWORDS

raster, maxent, ecology, niche, parallel

SYNOPSIS

r.maxent.lambdas

r.maxent.lambdas --help

r.maxent.lambdas [-pnNc] lambdas_file=name [alias_file=name] [logistic=name] [raw=name] [ndigits=integer] [nprocs=integer] width=integer height=integer [--overwrite] [--help] [--verbose] [--quiet] [--ui]

Flags:

-p: Print only; Print mapcalculator expressions and exit
-n: Do not include cells where any variabel contains no data
-N: Do not include cells where all variabels contain no data
-c: Clamp values in raster maps to value range seen by the MaxEnt model
--overwrite: Allow output files to overwrite existing files
--help: Print usage summary
--verbose: Verbose module output
--quiet: Quiet module output
--ui: Force launching GUI dialog

Parameters:

lambdas_file=name [required]: MaxEnt lambdas-file to compute distribution-model from
alias_file=name: CSV-file to replace alias names from MaxEnt by GRASS map names
logistic=name: Raster map with logistic output
raw=name: Raster map with raw output
ndigits=integer: Produce logistic output as integer map with this number of digits preserved; Default: 0
nprocs=integer: Number of r.mapcalc processes to run in parallel (requires r.mapcalc.tiled addon); Options: 1-; Default: 1
width=integer [required]: Width of tiles (columns) (requires r.mapcalc.tiled addon and nprocs > 1); Default: 1000
height=integer [required]: Height of tiles (requires r.mapcalc.tiled addon and nprocs > 1); Default: 1000

DESCRIPTION
NOTES
REFERENCES
SEE ALSO
AUTHOR

DESCRIPTION

The script is intended to compute (raw) or (logistic) prediction maps from a lambdas file produced with MaxEnt >= 3.3.3e.

It will parse the specified lambdas_file from MaxEnt >= 3.3.3e and translate it into an r.mapcalc-expression. If alias names had been used in MaxEnt, these alias names can automatically be replaced according to a CSV-like file (alias_file) provided by the user, as it can be produced with r.out.maxent_swd. This file should contain alias names in the first column and map names in the second column, separated by comma, without header. It should look e.g. like this:

alias_1,map_1
alias_2,map_2
...,...

If such a CSV-file with alias names used in MaxEnt is provided, the alias names from MaxEnt are replaced by raster map names.

The logistic map can be produced as an integer map. To do so the user has to specify the number of decimal places, that should be preserved in integer output in the ndigits option.

Optionally the map calculator expressions can be printed to stdout with the p-flag for inspection or documentation as they likely exceed the space in the map history.

By default, NoData for each function in the lambdas file is set to zero. The user can however choose to set pixels to null where a single variable contains NoData (n-flag) or where all variables produce NoData (N-flag).

Extraction of random points in MaxEnt can be a reason why values in raster maps exceed values seen by the MaxEnt model. To limit raster map values to the valid range for the model, raster map values can be clamped to the value range in the model with the c-flag.

Complex models (and thus mapcalculator expressions) can become CPU intensive to process. On multicore computers, processing such large models can benefit from tiled, parallel processing (nprocs larger than 1). This requires that the r.mapcalc.tiled addon is installed. The size of tiles can be controlled by the height and width options.

NOTES

This script works only if the maps containing the input data to MaxEnt are accessible from the current region.

Due to conversion from double to floating-point in exp()-function, a loss of precision from the 7th decimal place onwards may occur in the logistic output. Differences to logistic predictions from MaxEnt are supposed to be below 0.0001. This can be checked by importing sample or background predictions from the MaxEnt output (e.g. with r.in.xyz).

REFERENCES

Wilson, Peter D. 2009: Guidelines for computing MaxEnt model output values from a lambdas file. (Available at https://gsp.humboldt.edu/OLM/Courses/GSP_570/Learning%20Modules/10%20BlueSpray_Maxent_Uncertinaty/MaxEnt%20lambda%20files.pdf)
Steven J. Phillips, Miroslav Dudík, Robert E. Schapire. 2020: Maxent software for modeling species niches and distributions (Version 3.4.1). Available from url: https://biodiversityinformatics.amnh.org/open_source/maxent and https://github.com/mrmaxent/Maxent
Steven J. Phillips, Miroslav Dudík, Robert E. Schapire. 2004: A maximum entropy approach to species distribution modeling. In Proceedings of the Twenty-First International Conference on Machine Learning, pages 655-662, 2004.
Steven J. Phillips, Robert P. Anderson, Robert E. Schapire. 2006: Maximum entropy modeling of species geographic distributions. Ecological Modelling, 190:231-259, 2006.
Jane Elith, Steven J. Phillips, Trevor Hastie, Miroslav Dudík, Yung En Chee, Colin J. Yates. 2011: A statistical explanation of MaxEnt for ecologists. Diversity and Distributions, 17:43-57, 2011.

AUTHOR

Stefan Blumentrath, Norwegian Institute for Nature Research (NINA), https://www.nina.no

SOURCE CODE

Available at: r.maxent.lambdas source code (history)

Latest change: Monday Nov 11 18:04:48 2024 in commit: 59e289fdb093de6dd98d5827973e41128196887d