enirg: Ecological Niche in R-Grass
In ENiRG: Ecological Niche in R and GRASS

Description Usage Arguments Details Value Author(s) References See Also Examples

enirg performs an Ecological Niche Factor Analysis (ENFA) by coupling R and GRASS softwares and following Hirzel et al. (2002).

1
2
3

enirg(presences.table, qtegv.maps, qlegv.maps = NULL, col.w = NULL,
      scannf = TRUE, nf = 1, method = "normal", load.maps = FALSE,
      species.name = "species", map.center = NULL, res = NULL)

`qtegv.maps`	vector of strings, giving names of raster maps, containing quantitative environmental variables.
`qlegv.maps`	by default is set to NULL. Vector of strings, giving names of the raster maps, containing qualitative environmental variables (see details).
`presences.table`	data frame with species presence records in three columns, containing: 1 - X coordinate or longitude. 2 - Y coordinate or latitude. 3 - number of presences or observations.
`col.w`	vector with column weights with same length as number of quantitative and qualitative maps.
`scannf`	logical. Whether number of factors should be introduced by user (TRUE) or not (FALSE). A barplot with eigenvalues after analysis is displayed.
`nf`	integer, indicating number of kept specialization axes when `scannf=FALSE`.
`method`	string character. Selection of method to perform the calculations, "normal" for normal data sets or "large" for large data sets with huge amount of data. See also details.
`load.maps`	logical. Whether produced maps should be uploaded into R. Note that if you are working with large data sets (`method="large"`), this option should be FALSE in order to avoid memory allocation problems.
`species.name`	string, indicating name of modelled species.
`map.center`	string, indicating name of a map for centering studied area (also a mask). When NULL, first EGV map will be used.
`res`	integer, indicating resolution for modelling. Unit should be the same as in used maps. When NULL, resolution will be fixed from `map.center` or first EGV map.

Function enirg performs the Ecological Niche Factor Analysis (ENFA), following Hirzel et al. (2002). Analysis can be carried out for both quantitative variables and qualitative variables. This function requires occurrences records of the modelled species and environmental predictors of the study area.

User can choose number of factors that better describes the variance and the eigenvalue computation for factor extraction, by setting scannf=TRUE.

Depending on extension of the study area and environmental variables used to performed an analysis, user can choose one of two available methods: "normal" and "large". The first, strongly relies on the rgrass7 package and thus is limited by R storage capacity and computations; on the contrary, "large" method directly interface with GRASS, allowing calculations over large areas or high resolution maps with huge amount of data (NOTE: it is only available for Linux/Unix OS, at the moment).

Global marginality is calculated as:

M = \frac{ √{ ∑ _{i=1}^V m_i^2 } }{ 1.96 }

where m_i represents marginality of focal species on each EGV (up to V number of EGVs), in units of standards deviations of global distribution.

Global specialization is calculated as:

S = \frac{ √{ ∑ _{i=1}^V λ _i } }{ V }

where the eigenvalue λ _i associated to any factor expresses the amount of specialization it accounts for, i.e., ratio of the variance of global distribution to that of the species distribution on this axis.

If load.maps is set to TRUE, then maps of marginality and nf number of specialization maps will be loaded into R, by using raster library. Note that this is not recommended for large data sets.

enirg returns a list object of class "enirg" containing the following components:

call. Original call
nf. An integer indicating the number of kept specialization axes
cw. Column weights, a vector with p components
species. Species' name
egvs. A list of used environmental predictors
qt.egvs. A list of used quantitative environmental predictors
ql.egvs. A list of used quantitative environmental predictors
presences. A data frame containing the observations
total.marginality. Global marginality
marginalities. Vector of marginalities for axis
total.specialization. Global specialization
specializations. Vector of specializations by axis
co. Column coordinates, data frame with V rows and nf columns
mar. Coordinates of marginality vector
m. Marginality
s. Vector of eigenvalues
obs.li. Marginality and specialization axis for observations

Chiara Magliozzi chiara.magliozzi@libero.it, Fernando Canovas fcgarcia@ualg.pt, Jose Antonio Palazon-Ferrando palazon@um.es

Basille, M., Calenge, C., Marboutin, E., Andersen, R. \& Gaillard, J.M. (2008). Assessing habitat selection using multivariate statistics: Some refinements of the ecological-niche factor analysis. Ecological Modelling, 211, 233-240.

Canovas, F., Magliozzi, C., Mestre, F., Palazon-Ferrando, J.A. and Gonzalez-Wanguemert, M. (2016). ENiRG: R-GRASS interface for efficiently characterizing the ecological niche of species and predicting habitat suitability. Ecography 39: 593-598.

Hirzel, A.H., Hausser, J., Chessel, D. \& Perrin, N. (2002). Ecological-niche factor analysis: How to compute habitat-suitability maps without absence data? Ecology, 83, 2027-2036.

stdz.maps, enfa, initGRASS

## Not run: 
# starting GRASS session
initGRASS("/usr/bin/grass-7.0.0", home=tempdir())
initGRASS("C:/GRASS", home=tempdir())

data(apis.enirg)

# presences table
lina <- apis.enirg$presences

# loading the environmental information in batch
predictor.names <- c("tann", "mntcm", "rfdm", "rfwm")
predictor.maps <- paste("std_", predictor.names, sep="")
file.names <- paste(system.file(package = "ENiRG"),
                    "/ext/", predictor.names, ".asc", sep="")

import.egvs(file.names, predictor.names)

# standardization
stdz.maps(predictor.names, predictor.maps)

# setting the mask
mask.file <- paste(system.file(package = "ENiRG"),
                    "/ext/", "mask.asc", sep="")
import.egvs(mask.file, "mask")

# performing the Ecologigal Niche Factor Analysis (ENFA)
enirg(presences.table = lina, qtegv.maps = predictor.maps,
      species.name = "African", nf = 3,
      scannf = FALSE, load.maps = TRUE
      map.center = "mask", method = "normal") -> apis.enirg

summary(apis.enirg)

require(raster)
plot(apis.enirg$African_li_Mar)
plot(apis.enirg$African_li_Spec1)

## End(Not run)