Maxent

A map of South America with pixels and colors representing areas of high density species. "Maxent software for modeling species niches and distributions."

Maxent software is used for modeling species niches and distributions by applying a machine-learning technique called maximum entropy modeling. From a set of environmental (e.g., climatic) grids and georeferenced occurrence localities, the model expresses a probability distribution where each grid cell has a predicted suitability of conditions for the species. Under particular assumptions about the input data and biological sampling efforts that led to occurrence records, the output can be interpreted as predicted probability of presence (cloglog transform), or as predicted local abundance (raw exponential output).

The idea for Maxent was first conceived of at the Center for Biodiversity and Conservation at the American Museum of Natural History (AMNH) through a public-private partnership between the AMNH and AT&T-Research. Steven Phillips and the other developers of Maxent are still engaged in its development and maintenance, and the Google group will remain the main mechanism for user questions. Much additional information can be found in the Google group, software tutorials, and other resources listed below.

Maxent is Now Open Source!

Download Current version 3.4.4

Tutorials

If you would like to reference the tutorial in a publication, report, or online posting, an appropriate citation is:
Steven J. Phillips. 2017. A Brief Tutorial on Maxent. Accessed on.

Datasets used in the Ecological Modelling paper

  • coverages.zip (3.8 MB) - the coverages used in modeling.
  • ipcc.zip (10.8 MB) - the raw ipcc data, as dowloaded from the IPCC Data Distribution Centre.  Also the annual and monthly variables used for the study, extracted from the raw ipcc data and converted into world-wide coverages in .asc format.
  • samples.zip (9.5 kB) - the training and testing sample localities used.

Source Code

Citation

If you use the application for analyses that result in a publication, report, or online posting, the following represents a proper citation of the software itself:

Steven J. Phillips, Miroslav Dudík, Robert E. Schapire. [Internet] Maxent software for modeling species niches and distributions (Version 3.4.1). Available from url: https://hdl.handle.net/2246/7578. Accessed on .

**For information about earlier versions, please refer to the readme file on github or contact the developers [email protected]

Updates

Discussion Group

There is a long-standing google discussion group for users of this software at http://groups.google.com/group/Maxent.  Sign up for this group for discussions and questions about the software, and also to get information about updates, bugs, etc.

Maxent Publications

  • Steven J. Phillips, Robert P. Anderson, Miroslav Dudík, Robert E. Schapire, Mary Blair. 2017. Opening the black box: an open-source release of Maxent. In Ecography. PDF
  • Steven J. Phillips, Robert P. Anderson, Robert E. Schapire. 2006. Maximum entropy modeling of species geographic distributions. Ecological Modelling, 190:231-259, 2006. PDF *Datasets used in this paper are available in the download section
  • 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. PDF
     
  • Additional key references regarding Maxent
    • Jane Elith, Steven J. Phillips, Trevor Hastie, Miroslav Dudík, Yung En Chee, and Colin J. Yates. 2011. A statistical explanation of MaxEnt for ecologists. Diversity and Distributions, 17:43-57. PDF
    • Corey Merow, Matthew J. Smith, and John A. Silander, Jr. 2013. A practical guide to MaxEnt for modeling species’ distributions: what it does, and why inputs and settings matter. Ecography, 36: 1058–1069. PDF
    • Aleksandar Radosavljevic and Robert P. Anderson. 2014. Making better Maxent models of species distributions: complexity, overfitting, and evaluation. Journal of Biogeography, 41: 629–643. PDF