Design concepts adopted in long-term forest monitoring programs in Europe--problems for the future?

Long-term intensive monitoring in Europe is presently proceeding in more than 800 plots, where a number of investigations are carried out according to allegedly standardized protocols. While the potential of the program cannot be denied, certain aspects that are binding for data analysis may be a source of problems for future evaluation of program results. Here it is argued that: (i) current biological response indicators adopted by the program will not permit air pollution effects to be distinguished from effects due to other stressors and/or natural variation; (ii) the sampling strategy adopted to select monitoring sites does not enable European scale estimates of the status of attributes of interest or their changes; and that (iii) the sampling tactic suggested at plot level is ambiguous and cannot provide representative, unbiased estimates at plot scale. This latter point implies consequences when plot-level data are used in models, correlative studies and/or to infer cause-effect relationships.