Centre dedocumentation
scientifique

1 recherche sur le mot-clé libre 'Weibull distribution'

Airborne laser scanning for tree diameter distribution modelling: a comparison of different modelling alternatives in a tropical single-species plantation / Matti Maltamo in Forestry, an international journal of forest research, vol 91 n° 1 (January 2018)  

Public [article]  inForestry, an international journal of forest research > vol 91 n° 1 (January 2018) . - pp 121 - 131 Titre : Airborne laser scanning for tree diameter distribution modelling: a comparison of different modelling alternatives in a tropical single-species plantation Type de document : Article/Communication Auteurs : Matti Maltamo, Auteur ; Lauri Mehtätalo, Auteur ; R. Abdelfattah, Auteur ; Jari Vauhkonen, Auteur ; Petteri Packalen, Auteur Année de publication : 2018 Article en page(s) : pp 121 - 131 Note générale : Bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Lasergrammétrie [Termes IGN] Brésil [Termes IGN] diamètre à hauteur de poitrine [Termes IGN] diamètre des arbres [Termes IGN] données lidar [Termes IGN] données localisées 3D [Termes IGN] Eucalyptus grandis [Termes IGN] modèle numérique [Termes IGN] plantation forestière [Termes IGN] positionnement par GNSS Mots-clés libres : Weibull distribution Résumé : (Auteur) This study examines the potential of airborne laser scanning (ALS) to predict diameter distributions in an even-aged plantation of Eucalyptus urograndis in Brazil. The single-species plantation conditions allow different modelling alternatives to be compared without the presence of minor tree species or an understory layer affecting the results. Three modelling alternatives based on the two-parametric Weibull function form; parameter prediction, parameter recovery and distribution matching were tested with a k-nearest neighbour prediction (k-nn) method. We also evaluated a parameter prediction alternative, in which the Weibull distribution was predicted using field attributes. The results showed that ALS information can predict diameter distributions with an error margin of slightly more than 10 per cent of the RMSE of the mean of the third power of diameter, and with error index values between 50 and 60. The degree of accuracy was only slightly improved when the Weibull distribution was predicted using field attributes. According to the accuracy metrics, the differences between modelling alternatives were minor but parameter recovery and k-nn seem to be the most favourable ALS-based prediction methods. To conclude, the results showed a strong relationship between ALS information and diameter distributions in a tropical single-species plantation and we discuss how these results could be applied in other types of forests. Numéro de notice : A2018-633 Affiliation des auteurs : non IGN Thématique : FORET/IMAGERIE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1093/forestry/cpx041 En ligne : https://doi.org/10.1093/forestry/cpx041 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=93233 [article]