Excellent document, covering widely the different aspects of establishing a national forest monitoring system. Few hints to consider:
1. Chapter 3.1 deals with the mandate. Would it be necessary to discuss here, or somewhere, the different roles of implementing organizations. I mean, if one of the goals of the NFMS is to monitor quality of forest management, the NFMS should be independent from organizations responsible for forest management. Perhaps this was discussed in the earlier Chapters, I did not check.
2. Chapter 3.2 Identification of information Needs. I think that in one WFC2015 sessions, Andrew Lister pointed out that it would be useful to document INA process and the decisions made on the information content of NFMS. This would mean some kind of listing of reasons why some themes were rejected and why some themes where accepted in the NFMS information content. It was an excellent proposal, that could be added here?
3. Chapter 4.2.2 Sampling Design, Box 7. Please add a statement that there are also well established methods for estimating variance from systematic samples, reference to the work of Matern, for example.
4. Chapter 4.2.3 Sampling Design, page 24. Very well discussion about the need to consider the long term monitoring aspect when planning stratifications. Could add here some warnings about the need to consider all the key variables when planning the stratification, because a stratfication that may improve estimation of one key variable may be very ineffective for another key variable?
5. Chapter 4.2.6. Here or somewhere else(4.4?) add a recommendation that data processing, calculations, should be started as soon as you have some data. This may reveal some unexpected errors in the measurement guidelines or practices.
6. Chapter 4.3.4. Would this be good place to remind about work safety? One useful tool to improve the safety of field work is mapping of risks and sharing the results of this mapping in the training.
7. Chapter 4.4.1: Add about the need to make responsibilities very clear: who is allowed to clean the data, who is responsible to back up the data etc. Or, perhaps this was explained in some other Chapter?
8. Finally, a more general comment: I think you should consider possibility to make this document more practical by showing some good practices. For example, how the efficiency of monitoring has been improved. And estimators for means and totals, and their variances in the most common sampling designs should be shown in a compact and practical manner. I think that for example the German and US NFI's have made such details available, but it would be very useful to collect this information under this document, as well.