| Popis: |
In Italy, reliable updated data on forest expansion on a national or regional scale are not available. The aim of this note is to present a probabilistic estimation approach based on land cover classification repeated on the same sample points for two successive occasions. The approach was experimented in the Abruzzo region (Central Italy), where the rate of forest expansion was assessed by multitemporal classification of permanent sample points on orthocorrected aerial photos. The forest area at the year 1990, a figure of high relevance for reporting under the Kyoto Protocol, was also assessed. The proposed procedure proved to be relatively easy to implement and able to provide precise estimates, given the adopted sampling intensity (the same of the Italian national forest inventory). It is stressed that statistical figures of forest areas on wide territories obtained by estimation approaches like the proposed one are scientifically more defensible than those traditionally based on the complete mapping, being more objectively repeatable. |
