Geostatistical models were used to estimate cockle abundance for 12 sites in the 2019–20 northern bivalve survey. Estimates often had greater precision than the sampling-based estimator, and fine-scale predictions reproduced sampling observations. Nevertheless, results were sensitive to model settings and site configuration. Spatio-temporal models accounting for earlier surveys also improved precision in population estimates, except for sites where cockle beds shifted between years.
If you are having problems opening this file, you may not have the software on your computer needed to view it. Click the link below to download and install a suitable program.
