BBWO_Ensemble: Classify HSIs and generate ensemble predictions for nesting...

In qureshlatif/WoodpeckerHSI: Woodpecker habitat suitability models

Description

Wrapper function that classifies model HSIs using pre-defined thresholds and sums classified values to generate ensemble predictions, which represent the number of models that characterize a given observation or pixel as suitable for nesting Black-backed Woodpecker (BBWO; Latif et al. 2013) in burned dry conifer forest of the Inland Northwest and central Rocky Mountains.

Usage

BBWO_Ensemble(Data, mods = c("BC1", "TBC1", "TB1", "SG_wlr", "TP_wlr", "TB_wlr", "Maxent_3v", "Maxent_brn"),
  thrhds = c(0.17,0.17,0.32,0.43,0.43,0.45,0.41,0.37), RAVG = F)

Arguments

Data	matrix (or data frame coercible to matrix) containing columns representing HSIs (range 0-1) named in 'mods'. Most likely, Data = output from the 'BBWO_IndivModelHSIs' function.
mods	Character vector containing names of columns containing model HSIs to be combined into ensemble. Default are the names assigned by the 'BBWO_IndivModelHSIs' function.
thrhds	Numeric vector containing thresholds used to classify individual model HSI values into more suitable (1) vs less suitable (0). Length of 'thrhds' should equal length(mods). Default are thresholds identified by Latif et al. (2013).
RAVG	If false (default), thresholds for classifying individual model HSIs are as defined by the 'thrhds' argument. If true, thresholds are re-defined within the function as needed for RAVG-based alternate versions of individual models.

Details

This the final wrapper function that classifies 8 component model HSIs and combines classified values into an ensemble prediction of habitat suitability for nesting Black-backed Woodpeckers (Latif et al. 2013). Ensemble predictions describe the number of models classifying a given observation or pixel as suitable. The 'Data' argument should contain all continuous HSIs from individual models, and will most likely be the output from the 'BBWO_IndivModelHSIs' function.

Value

Matrix of classified HSIs (8 binary columns) and the final ensemble prediction (1 integer column with range 0-8). Columns are labeled 'BC1', 'TBC1', 'TB1', 'SG_wlr', 'TP_wlr', 'TB_wlr', 'Maxent_3v', 'Maxent_brn', and 'Ensemble'.

Author(s)

Quresh S. Latif, Rocky Mountain Research Station, U.S. Forest Service

References

Latif, Q. S., V. A. Saab, J. G. Dudley, and J. P. Hollenbeck. 2013. Ensemble modeling to predict habitat suitability for a large-scale disturbance specialist. Ecology and Evolution 3:4348-4364.

See Also

Mahal_HSI, BBWO_MxWLRHSIs, BBWO_Mxnt3v, BBWO_Mxntbrn, BBWO_IndivModelHSIs

Examples

Data <- BBWO_IndivModelHSIs(lndscp) # lndscp = matrix or data frame with required input variables.
EnsPred <- BBWO_Ensemble(Data)

qureshlatif/WoodpeckerHSI documentation built on May 29, 2019, 7:51 a.m.