project_maxent: Project a fitted Maxent model
In wkmor1/voiConsPlan: Value of information for conservation planning

Description Usage Arguments Details Value Warning Author(s) References

View source: R/project_maxent.r

Project a fitted Maxent model by predicting to new environmental data.

1	project_maxent(lambdas, newdata, mask)

`lambdas`	Either a `MaxEnt` fitted model object (fitted with the `maxent` function in the `dismo` package), or a file path to a Maxent .lambdas file.
`newdata`	A `RasterStack`, `RasterBrick`, `list`, `data.frame`, `data.table`, or `matrix` that has layers/elements/columns whose names correspond to the names of predictors used to fit the model. These layers/elements/columns must all have the same length.
`mask`	(Optional; requires that `newdata` is a `Raster` object.) A `Raster` object with `NA` values in cells for which the model should not* be projected. These cells will be assigned `NA` in the returned output.

project_maxent uses feature weights described in a .lambas file or MaxEnt object to predict a Maxent model to environmental data. This function performs the projection entirely in R, without the need for the Maxent Java software. For tested datasets, it performs the projection in roughly one third of the time taken for the same projection by maxent.jar.

If newdata is a RasterStack or RasterBrick, a list with two elements:

prediction_raw: a Raster layer giving the raw Maxent prediction; and
prediction_logistic: a Raster layer giving the logistic Maxent prediction.

If newdata is not a RasterStack or RasterBrick, the raster layers will be replaced with data.tables in the returned list.

This function is still in development, and no guarantee is made for the accuracy of its projections.

John B. Baumgartner, johnbaums@gmail.com

Wilson, P. W. (2009)Guidelines for computing MaxEnt model output values from a lambdas file
Maxent software for species habitat modeling, version 3.3.3k help file (software freely available here)