Model Formulation

There is considerable theoretical interest in the existence and magnitude of heterogeneity in survival and other vital rates among individuals (Cam et al., 2002a, 2004). In addition, heterogeneity masks other effects that might be the focus of model {

phi0~dunif(0,1)

theta~dunif(0,1)

for(j in 1:sumNumNeigh) { weights[j] <- 1} spacetau~dgamma(.1,.1)

alpha[1:ngrid] ~ car.normal(adj[], weights[], num[], spacetau)

tmp[i]<- pow(1,resident[i])*pow(theta,1-resident[i])

for(j in (first[i]+1):nyear){ mu1[i,j]<-p[j]*z[i,j] x[i,j]~dbern(mu1[i,j]) mu2[i,j]<-z[i,j-1]*phi[i]*R[i] z[i,j]~dbern(mu2[i,j])

Panel 11.6. WinBUGS model specification of a CJS model allowing for transients, and with spatially-correlated survival probability parameters. For illustration, this model uses the alternative parameterization for transients having parameter 6 that is related to residency probability by Eq. (11.4.1).

Figure 11.2. Map depicting spatial variation in yellow warbler survival. Values are point-wise posterior means. The shading reflects the magnitude of predicted survival probability whereas the size of the plot symbol is proportional to the inverse of the posterior standard deviation. Larger circles indicate smaller posterior standard deviations. The 139 Monitoring Avian Productivity and Survivorship (MAPS) stations used in the analysis are indicated with square symbols.

Was this article helpful?

0 0

Post a comment