The acquisition of information about growing stock is a fundamental step in the framework of forestmanagement planning and scenario modeling, besides being essential for assessing the amount of carbonstored within forest ecosystems. Gallaun et al. (2010) produced a pan-European map of forest growingstock by the combination of ground and remotely sensed data. The first objective of the current paper is toassess the accuracy of this map versus the ground data collected during the latest Italian National ForestInventory (INFC). Next, a new wall-to-wall estimation of growing stock is obtained by combining groundmeasurements of four regional forest inventories with the CORINE land cover map of Italy and the globalcanopy height map derived from Geoscience Laser Altimeter System (GLAS) and Moderate ResolutionImaging Spectroradiometer (MODIS) data. More particularly, the growing stock measurements of thefour inventories are stratified by ecosystem type and extended over all Italian forest areas through theapplication of locally weighted regressions to the GLAS/MODIS canopy height map. When compared tothe INFC measurements, the new map shows higher accuracy than that by Gallaun et al., particularly forhigh growing stock values. The coefficient of determination between estimated and INFC growing stocksis improved by about 0.5, whilst the mean square error is reduced from 90 to 48 m3ha−1.
|Autori:||Maselli, F.;Chiesi, M.;Mura, M.;Marchetti, M.;Corona, P.;Chirici, G.|
|Data di pubblicazione:||2014|
|Titolo:||Combination of optical and lidar satellite imagery with forest inventory data to improve wall-to-wall assessment of growing stock in Italy|
|Rivista:||INTERNATIONAL JOURNAL OF APPLIED EARTH OBSERVATION AND GEOINFORMATION|
|Appare nelle tipologie:||1.1 Articolo in rivista|