MSDM

Methods to deal with overprediction of species distribution models

Installation 

require(devtools)  
install_github("sjevelazco/MSDM")  
require(MSDM)

# See more details of funcions and examples
?MSDM_Priori
?MSDM_Posteriori

Description

MSDM provides tools to correct or reduce overprediction of species distribution models. There are two groups of methods compiled in two main functions MSDM_Priori() and MSDM_Posteriori(). The main difference between both methods is that MSDM_Priori generates predictive spatial variables that will be used (along with the environmental variables) to construct species distribution models, whereas MSDM_Posteriori works on the already fitted models. For more information on MSDM see Mendes et al. (2020).

. MSDM_Priori: offers four methods, named XY, MIN, CML, and KER. All these methods consist of creating spatial predictor variables to be used with environmental variable to construct species distribution models. These approaches constrain the suitability predicted by species distribution models to species occurrences.

. MSDM_Posteriori: provides four methods, named OBR, PRES, LQ, MCP, and BMCP. These methods reduce overprediction of species distribution models already fitted based on the occurrences and suitability patterns of species.

( a) Example of suitability pattern of an SDM based on a generalized linear model (GLM) constructed without spatial restriction and (b) different A priori and A posteriori MSDM methods

CITATION:

Mendes, P., Velazco, S. J. E., Andrade, A. F. A. de, & De Marco, P. (2020). Dealing with overprediction in species distribution models: How adding distance constraints can improve model accuracy. Ecological Modelling, 431, 109180. https://doi.org/10.1016/j.ecolmodel.2020.109180

NEWS

MSDM functions are now available in the new R package flexsdm

Package website

See MSDM package website (https://sjevelazco.github.io/MSDM) for further details of functions and examples

Test the package and give us feedback here or send an e-mail to .

MSDM package is integrated to ENMTML R package