BIOMOD_EnsembleForecasting: Ensemble projections of species over space and time

Description

This function use projections of ‘individual models’ and ensemble models from BIOMOD_EnsembleModeling to build an ensemble of species' projections over space and time.

Usage

BIOMOD_EnsembleForecasting( EM.output,
                            projection.output = NULL,
                            new.env = NULL,
                            xy.new.env = NULL,
                            selected.models = 'all',
                            proj.name = NULL,
                            binary.meth = NULL,
                            filtered.meth = NULL,
                            compress = TRUE,
                            ...)

Arguments

EM.output

a "BIOMOD.EnsembleModeling.out" returned by BIOMOD_EnsembleModeling

projection.output

a "BIOMOD.projection.out" returned by BIOMOD_Projection

new.env

a RasterStack or a data.frame object containing explanatory data for the studied area. Needed only if projection.output is NULL. Prefer to use "BIOMOD.projection.out" if you have ever done the calculations.

xy.new.env

the matching coordinates of new.env if new.env is defined and if it is a data.frame

selected.models

if not 'all', a character vector containing a subset of ensemble-models you want make projacion

proj.name

the projection name (results will be saved within proj_proj.name directory). Only needed if projection.output is NULL

binary.meth

vector specifying the names of evaluation metrics and associated thresholds to transform the probabilities of presence into presence and absence (binary transformation).

filtered.meth

vector specifying the names of evaluation metrics and associated thresholds to transform into 0 the probabilities of presence lower than the thresholds.

compress

boolean or character, the compression format of objects stored on your hard drive. May be one of ‘TRUE’, ‘FALSE’, ‘xz’ or ‘gzip’ (see save)

…

further arguments (see details)

Value

Nothing returned but specific ‘projection files’ are saved on the hard drive projection folder. This files are either an array or a RasterStack depending the original projections data type. Load these created files to plot and work with them.

Details

This function requires to have successfully run biomod2 modeling, ensemble-modeling and projection steps. Ensemble projections will be created in respect to projection.output projections, which are combined following EM.output ensemble-modeling rules.

The ‘total.consensus’ projection is basically the mean of all projections (for having only one output).

… may be :

on_0_1000:logical, if TRUE (default), 0 - 1 probabilities are converted into a 0 - 1000 integer scale. This implies a lot of memory saving. User that want to comeback on a 0 - 1 scale latter will just have to divide all projections by 1000

Examples

Run this code

# NOT RUN {
# 0. Load data & Selecting Data
# species occurrences
DataSpecies <- read.csv(system.file("external/species/mammals_table.csv",
                                    package="biomod2"), row.names = 1)
head(DataSpecies)

# the name of studied species
myRespName <- 'GuloGulo'

# the presence/absences data for our species 
myResp <- as.numeric(DataSpecies[,myRespName])

# the XY coordinates of species data
myRespXY <- DataSpecies[,c("X_WGS84","Y_WGS84")]


# Environmental variables extracted from BIOCLIM (bio_3, bio_4, bio_7, bio_11 & bio_12)
myExpl = stack( system.file( "external/bioclim/current/bio3.grd", 
                     package="biomod2"),
                system.file( "external/bioclim/current/bio4.grd", 
                             package="biomod2"), 
                system.file( "external/bioclim/current/bio7.grd", 
                             package="biomod2"),  
                system.file( "external/bioclim/current/bio11.grd", 
                             package="biomod2"), 
                system.file( "external/bioclim/current/bio12.grd", 
                             package="biomod2"))

# 1. Formatting Data
myBiomodData <- BIOMOD_FormatingData(resp.var = myResp,
                                     expl.var = myExpl,
                                     resp.xy = myRespXY,
                                     resp.name = myRespName)
                                                                     
# 2. Defining Models Options using default options.
myBiomodOption <- BIOMOD_ModelingOptions()

# 3. Running the models 
myBiomodModelOut <- BIOMOD_Modeling( myBiomodData, 
                                       models = c('RF'), 
                                       models.options = myBiomodOption, 
                                       NbRunEval=2, 
                                       DataSplit=60, 
                                       Yweights=NULL, 
                                       VarImport=0, 
                                       models.eval.meth = c('TSS'),
                                       SaveObj = TRUE,
                                       rescal.all.models = FALSE,
                                       do.full.models = FALSE)
                                       
# 4. Creating the ensemble models 
myBiomodEM <- BIOMOD_EnsembleModeling( 
                 modeling.output = myBiomodModelOut,
                 chosen.models = grep('_RF', get_built_models(myBiomodModelOut), 
                                  value=TRUE),
                 em.by = 'algo',
                 eval.metric = c('TSS'),
                 eval.metric.quality.threshold = c(0.7),
                 prob.mean = TRUE,
                 prob.cv = FALSE,
                 prob.ci = FALSE,
                 prob.ci.alpha = 0.05,
                 prob.median = FALSE,
                 committee.averaging = FALSE,
                 prob.mean.weight = FALSE,
                 prob.mean.weight.decay = 'proportional' )
                                       
# 5. Individual models projections on current environmental conditions
myBiomodProjection <- BIOMOD_Projection(
                        modeling.output = myBiomodModelOut,
                        new.env = myExpl,
                        proj.name = 'current',
                        selected.models = grep('_RF', get_built_models(
                                                myBiomodModelOut), value=TRUE),
                        compress = FALSE,
                        build.clamping.mask = FALSE)
                          

# 4. Creating the ensemble projections
BIOMOD_EnsembleForecasting( projection.output = myBiomodProjection,
                            EM.output = myBiomodEM)
# }