R-Forge Logo

Welcome to modEvA project!

modEvA is an R package for species distribution model evaluation and analysis.

It includes functions for variation partitioning, calculating several measures of model discrimination and calibration, optimizing prediction thresholds based on a number of criteria, performing multivariate environmental similarity surface (MESS) analysis, and displaying various analytical plots. Includes also a sample data set with some species distribution models.

Package modEvA (v1.3.2) is now on CRAN!

Install and load

To install the currently official version of modEvA, paste the following command in the R console (while connected to the internet):


To install a developing version (with possibly new additions, but probably also new bugs) of modEvA from R-Forge, paste the following command in the R console (while connected to the internet):

install.packages("modEvA", repos="http://R-Forge.R-project.org")

This should work if you have the latest version of R; otherwise, it may either fail (producing a message like "package 'modEvA' is not available for your R version") or install an older version of modEvA. To check the modEvA version that you have actually installed, type citation(package="modEvA"). To install the latest modEvA version, you can either upgrade R or download the compressed modEvA package source files to your disk (.tar.gz for Linux/Mac or .zip for Windows, available here or here) and then install the package from there, e.g. with R menu "Packages - Install packages from local zip files" (Windows), or "Packages & Data - Package installer, Packages repository - Local source package" (Mac), or "Tools - Install packages - Install from: Package Archive File" (RStudio).

You only need to install (each version of) the package once, but then every time you re-open R you need to load the package by typing:


You can then check out the package help files and try some of the provided examples:



Papers introducing modEvA:

Barbosa A.M., Real R., Muñoz A.R. & Brown J.A. (2013) New measures for assessing model equilibrium and prediction mismatch in species distribution models. Diversity and Distributions 19: 1333-1338 (DOI: 10.1111/ddi.12100)

Barbosa A.M., Brown J.A., Acevedo P., Lobo J.M. & Real R. (in prep.) The ABC of model evaluation: a visual method for a clearer assessment of model accuracy

Other papers using modEvA:

Dias A., Palma L., Carvalho F., Neto D., Real J. & Beja P. (2017) The role of conservative versus innovative nesting behavior on the 25-year population expansion of an avian predator. Ecology and Evolution, DOI: 10.1002/ece3.3007

Reino L., Ferreira M., Martínez-Solano I., Segurado P., Xu C. & Barbosa A.M. (2017) Favourable areas for co-occurrence of parapatric species: niche conservatism and niche divergence in Iberian tree frogs and midwife toads. Journal of Biogeography, 44: 88-98

Areias-Guerreiro J., Mira A. & Barbosa A.M. (2016) How well can models predict changes in species distributions? A 13-year-old otter model revisited. Hystrix – Italian Journal of Mammalogy, DOI: http://dx.doi.org/10.4404/hystrix-27.1-11867

De Araújo, C.B., Marcondes-Machado, L.O. & Costa, G.C. (2013) The importance of biotic interactions in species distribution models: a test of the Eltonian noise hypothesis using parrots. Journal of Biogeography, 41: 513-523

Find out more

Here's a quick illustrated tutorial, also available in PDF (updated 7 May 2015); and a reference manual based on the package help files. There's also a course manual on model building with fuzzySim and model evaluation with modEvA (in Spanish)

Click here for further info on the package and its origins.

The R-Forge project summary page you can find here.