vuln: Multidimensional vulnerability index

In phytomosaic/vuln: Niche-based Vulnerability of Species and Communities

Description

Multidimensional vulnerability index, using truncated or extended niches.

Usage

vuln(spe, y, ybin, yinv, na.rm = TRUE, fullinfo = TRUE, ...)

gv(obj, pick = 1, ...)

Arguments

spe	species abundance or occurrence matrix (response variables).
y	vector, matrix, or list of explanatory variables (climate, fire, etc), where rows correspond to rows in spe.
ybin	optional, data.frame or list of data.frames from bingrid, with binned explanatory variables from rasterizing. Function gives extended niches if supplied, possibly truncated niches if not supplied.
yinv	vector of logicals, in order of y, defining which explanatory variables y should be 'inverted' (e.g., low moisture = high vulnerability). Defaults to rep(FALSE, NCOL(y)).
na.rm	logical, indicating whether to strip NA values before computation. Default is na.rm = TRUE
fullinfo	logical, default fullinfo=TRUE propagates any NA cases that occur in any y variable; when fullinfo=TRUE, percentiles will be based only on non-NA y cases.
...	further arguments passed to other functions.
obj	object of class 'vuln'.
pick	numeric, indicating which vulnerability metric (1,2,3).

Value

List object of class 'vuln', containing:

v

y values at all sites in spe.

ve

y values at all sites in ybin.

vo

vulnerable occurrences matrix.

va

vulnerable abundances matrix.

t1

values for 1st vulnerability index, describing the proportion of observed species which are vulnerable.

t2

values for 2nd vulnerability index, describing the community-mean percentile at the observed site.

t3

values for 3rd vulnerability index, describing the safety margin as the deviation of community-mean upper y limits from the observed y value.

Examples

## Load data
## Not run: 
data(braun, package='ecole')
spe <- braun$spe
env <- braun$env
## Vulnerability estimates
v <- vuln(spe, y=env$bio1) # w/o bingrid, may be 'truncated' niche
## Mapping
d <- data.frame(env, t1 = gv(v,1), t2 = gv(v,2), t3 = gv(v,3))
f <- function(d, j='bio1', ...){
     cc <- c("#420A6899","#65156E99","#89226A99",
             "#AD305D99","#CC424899","#E75D2E99",
             "#F7821299","#FCAA0F99","#F6D64599")
     plot(d$lat, d$lon, col=cc[cut(d[,j],breaks=9)],pch=16,main=j)
}
par(mfrow=c(2,2), bty='L', las=1)
f(d, 'bio1')
f(d, 't1')
f(d, 't2')
f(d, 't3')

## End(Not run)

phytomosaic/vuln documentation built on Sept. 21, 2019, 8:23 a.m.