model{
   for(i in 1:N1){
      K[i,1:2]~dmnorm(mu,prec)
   }
   # Construct the covariance matrix
   prec<-inverse(cov)
   cov[1,1]<-sigma[1]^2
   cov[2,2]<-sigma[2]^2
   cov[1,2]<-sigma[1]*sigma[2]*rho
   cov[2,1]<-sigma[1]*sigma[2]*rho
   # Priors
   for(j in 1:N2){
      mu[j]~dnorm(0,1/10000)
      sigma[j]~dunif(0.001,1000)
   }
   rho~dunif(-1,1)
}