Is floristic quality assessment reliable in human-managed ecosystems?

The Floristic Quality Analysis (FQA) is a method to assess the quality of a flora based on the assignment of scores to plant species and subsequent calculation of indices. This method is widely applied, but inadequate investigation has been devoted to test its potential problems due to human factors. This work is aimed to specifically test how the human factor can affect the calculation of the FQA indices, by addressing three questions: (i) Are the scores given to plant species consistent among different experts?; (ii) Are the floristic quality indices calculated by different experts consistent in ordering individual sites?; and (iii) Does the use of an appropriate statistics change the ordering of individual sites? To answer these questions, a list of species obtained in 136 plots in central Italy was submitted to nine experts, who scored each species. The FQA indices were then calculated from the scores of each of the experts. The results showed that: (i) the scores given to the species by the experts were not consistent and the derived floristic quality indices were statistically different; (ii) the floristic quality indices calculated for each plot were significantly different among experts, but the ranking of these plots based on their floristic quality was rather consistent; and (iii) the use of ordinal statistics, which is more adequate for this type of data, did not change the results. This study demonstrated that the Floristic Quality Analysis does not provide reliable and objective tools to assess the quality of the flora in a human-managed ecosystem. The application of these indices should be preceded with resolution of the methodological problems associated with the use of inappropriate statistics, and by procedures to reduce the degree of subjectivity in assigning the CC scores.