Three options for the sampling design of the field plot clusters of NAFORMA II biophysical survey are compared in this report. Option 1 consists of re-measuring all NAFORMA I field sample plots (3 205 clusters) and Option 2 of re-measuring only those that were established as permanent (848 clusters). The recommended Option 3 is a compromise between these two “extreme” options: Re-measure a subset (1 405 clusters) of NAFORMA I field sample plots including (almost) all permanent clusters and a carefully selected set of other NAFORMA I field plot clusters to obtain a uniform sample within each TFS zone.

Design Option 3 has the following features:

• Sampling intensity is uniform within each TFS zone. This makes it simple to use the data. For example, mean volumes can be estimated by averages over the plots.

• The selected clusters are well-spread over the target population.

• The anticipated precision of land-class area and mean wood volume relative to sample size is nearly as good as that of NAFORMA I.

• All proposed clusters were measured in NAFORMA I, which enables precise estimation of change based on repeated measurements.

The costs and precision were anticipated by utilizing NAFORMA I field data, information about subsequent improvements in the road network, and changes in land-use using satellite imaging derived land-class maps.