run.power: Run power analysis

In dsouthwell/TestPackagev2: Spatially explicit power analysis in dynamic landscapes

Description

This function runs the main power analysis. It starts by creating empty vectors to record the number of times a significant trend is detected in the simulated detection histories It then simulates fire at sites if model.fire = TRUE, and models a decline in occupancy over time depending on the specified effect size. Detection histories are simulated at each site given the occupancy status at each site, the number of survey methods used to detect a species and the detection probability of each relevent method. Simulated detection histories are loaded into the package unmarked, and a trend in occupancy is modelled over time. The proportion of times a significant trend is detected is recorded. This process is repeated for each effect size.

Usage

run.power(effect.size, nsims, alpha, Tmax, s.years, trend, sites,
  model.disturbance, disturbance.type, species.list, dist.occ.time,
  dist.occ.freq, park.level, xy.sites, R, two.tailed, plot.sites, n.sites,
  n.species, n.park, all.loaded.sites, load.sites, new.site.selection, occ.time,
  det.method1.time, det.method2.time, det.method3.time, det.method4.time,
  n.method, occ, det.method1, det.method2, det.method3, det.method4, stratify,
  remote, parks, Pr.dist, disturbance)

Arguments

effect.size	An integer specifying the proportional reduction in occupancy between the start and end of the monitoring program
nsims	An integer specifying the number of simulations
alpha	The Type I error rate. Must be either 0.1, 0.05 or 0.01
Tmax	An integer specifying the length of the monitoring program
s.years	A vector specifying the years in which monitoring occurs. Note, the final year or monitoring must be equal to Tmax
trend	Set to 'increasing' to model an increasing trend in occupancy or 'decreasing' to model a decreasing trend
sites	An array containing the XY coordinates of pre-specified monitoring sites
model.disturbance	Set to TRUE to model a disturbance, FALSE otherwise
disturbance.type	Set to 'deterministic' or 'stochastic' depending on the type of disturbance
species.list	An array containing the name of each species in the first column, and a zero or one describing how many methods are used to detect the species
dist.occ.time	An integer specifying which layer in the raster covariate stack represents time since the last disturbance
dist.occ.freq	An integer specifying which layer in the raster covariate stack represents the number of disturbances
park.level	Set to TRUE if power is to be estimated within sub-level management units
xy.sites	The XY coordinates of monitoring sites
R	Ratio of remote to non-remote sites
two.tailed	Set to TRUE to conduct a two-tailed significance test, FALSE if conducting a one-tailed test
plot.sites	Set to TRUE to plot the location of sites at the start of each simulation, FALSE otherwise
n.sites	The number of sites to be monitored.
n.species	The number of species for which monitoring is simulated. Is equal to the number of layers in the occ raster stack
n.park	The number of nested management units in which to estimate power
all.loaded.sites	Set to TRUE if all of the loaded sites are to be monitored, FALSE otherwise
new.site.selection	Set to 'random' to select new sites randomly, 'stratified' to randomly position sites within environmental strata, or 'maxocc' to position sites on cells with the highest relative species richness
occ.time	An array containing the occupancy value for each species at monitoring sites over time
det.method1.time	An array containing the detectability values of method 1 for each species at monitoring sites over time
det.method2.time	An array containing the detectability values of method 2 for each species at monitoring sites over time
det.method3.time	An array containing the detectability values of method 3 for each species at monitoring sites over time
det.method4.time	An array containing the detectability values of method 4 for each species at monitoring sites over time
n.method	The number of repeat visits to a site in a given survey year. Each element corresponds to each of the four detection methods
occ	Occupancy raster stack with the number of layers equal to the number of species
stratify	A raster layer of the environmental strata in which to randomly stratify monitoring sites. Each strata should have its own unique integer
remote	A raster layer identifying remote and non-remote areas. Remote areas should be identified with a 1, non-remote a 0
parks	A raster layer identifying nested areas within the landscape in which to estimate power. Each sub-unit should be identified with a separate integer
Pr.dist	A vector specifying the probability of a disturbance given the time since the last disturbance
disturbance	A raster stack of the disturbance history durign a proceeding time period. Disturbned cells are givena value of 1, undisturbed cells a value of 0.
loaded.sites	Set to TRUE if monitoring is simulated at XY-coordinates provided, FALSE otherwise
det1.method1	An array that records simulated detection histories at sites using method 1
det1.method2	An array that records simulated detection histories at sites using method 2
det1.method3	An array that records simulated detection histories at sites using method 3
det1.method4	An array that records simulated detection histories at sites using method 4

dsouthwell/TestPackagev2 documentation built on May 28, 2019, 9:38 a.m.