Crown condition surveys in Italian forests: issues in reporting findings.

This study examines defoliation and discoloration findings collected throughout the Italian Level I network between 1997 and 2000. Prior to this period no Quality Assurance procedures had been implemented, so that earlier findings cannot be considered reliable. The aim of the study is to compare different indices used in reporting findings, in order to determine the full potential and limitations of each. In international surveys, findings are normally expressed in terms of individual trees presenting a defoliation level greater than 25%; this criterion, however, has been challenged since it is not based on scientific evidence. By analysing the distribution of defoliationvalues (grouped in 5% classes) relating to the main species (Fagus sylvatica, Quercus pubescens, Quercus cerris, Picea abies), the study examines the behaviour of statistical indices such as the median, the mode and the mean. Instead of using the traditional 25% threshold to determine the part of the population with the highest defoliation values, a procedure is suggested whereby a different threshold can be determined for each species. It is established based on the 90 degrees percentile of cumulative defoliation values recorded over the four-year period. Each index used to report findings possesses its own information potential, and the findings may at times appear contradictory. The use of a combined defoliation and discoloration index (Ilce = Lacking Crown Equivalent Index) is also proposed, but the results obtained do not differ in substance from the results relating to defoliation alone. The functional limitations of this index are due mainly to the way the data are collected: for this reason suggestions are made on how to improve the procedure in future surveys.