weta {wiqid}

Detection data for weta in gorse bushes

Description

Results of an occupancy survey to see if presence/absence of weta in 72 gorse bushes is affected by browsing of the bushes by goats. Probability of detection may be different for each observer, and observer ID is also recorded.

Usage

data(weta)

Format

A data frame with 72 observations on the following 11 variables.

D1, D2, D3, D4, D5

a numeric vector for each of 5 daily surveys, showing whether weta were detected (1) or not detected (0); NA if the bush was not surveyed on the relevant day.

Browsed

a logical vector indicating whether the bush was browsed (TRUE) or not browsed (FALSE)

ObsD1, ObsD2, ObsD3, ObsD4, ObsD5

a factor with levels A B C, indicating which observer carried out each survey.

Details

The data are provided as a data frame, such as would result from reading in data from a text file. Further formatting is needed before using these for analysis: see the examples.

Source

Discussed in MacKenzie et al (2006) p116. Data distributed with PRESENCE.

References

MacKenzie, D I; J D Nichols; A J Royle; K H Pollock; L L Bailey; J E Hines 2006. Occupancy Estimation and Modeling : Inferring Patterns and Dynamics of Species Occurrence. Elsevier Publishing.

Examples

data(weta)

DH <- weta[, 1:5]  # extract the detection history:
occSS(DH)
occSStime(DH, p~.time)
# With covariates
occSS(DH, list(psi ~ Browsed, p ~ ObsD), data=weta)
occSS(DH, list(psi ~ Browsed, p ~ Browsed), data=weta)

# Bayesian analysis

# Model with separate intercepts for occupancy in Browsed and Unbrowsed
#  bushes, and a time trend for probability of detection; specify uniform
#  priors for probability of occupancy:
Bwet <- BoccSS(DH, model=list(psi~Browsed-1, p~.Time), data=weta,
  priors=list(sigmaPsi=c(1,1)), chains=2)
Bwet
plot(Bwet)
plot(Bwet, "p_.Time")