National mapping and estimation of forest area by dominant tree species using Sentinel-2 data

• Published in: Canadian journal of forest research Vol. 51; no. 3; pp. 365 - 379
• Main Authors: Breidenbach, Johannes, Waser, Lars T, Debella-Gilo, Misganu, Schumacher, Johannes, Rahlf, Johannes, Hauglin, Marius, Puliti, Stefano, Astrup, Rasmus
• Format: Journal Article
• Language: English
• Published: 1840 Woodward Drive, Suite 1, Ottawa, ON K2C 0P7 NRC Research Press 01.03.2021; Canadian Science Publishing NRC Research Press
• Subjects: apprentissage automatique; biomass; Cartography; Deciduous forests; Dominant species; Estimates; estimation assistée par modèle; estimation sur petits domaines; forest inventory; Forest resources; Forests; Forests and forestry; Interferon; inventaire forestier national; Machine learning; Mapping; Methods; Model accuracy; model-assisted estimation; Mosaics; Municipalities; National Forest Inventory; National forests; Norway; Nuclear factor I; Picea; Plant species; poststratification; prediction; Satellite imaging; small area estimation; Sustainable forestry; Timber inventory; trees; Norway
• ISSN: 0045-5067; 1208-6037; 1208-6037
• DOI: 10.1139/cjfr-2020-0170

Nation-wide Sentinel-2 mosaics were used with National Forest Inventory (NFI) plot data for modelling and subsequent mapping of spruce-, pine-, and deciduous-dominated forest in Norway at a 16 m × 16 m resolution. The accuracies of the best model ranged between 74% for spruce and 87% for deciduous forest. An overall accuracy of 90% was found on stand level using independent data from more than 42 000 stands. Errors mostly resulting from a forest mask reduced the model accuracies by ∼10%. The produced map was subsequently used to generate model-assisted (MA) and poststratified (PS) estimates of species-specific forest area. At the national level, efficiencies of the estimates increased by 20% to 50% for MA and up to 90% for PS. Greater minimum numbers of observations constrained the use of PS. For MA estimates of municipalities, efficiencies improved by up to a factor of 8 but were sometimes also less than 1. PS estimates were always equally as or more precise than direct and MA estimates but were applicable in fewer municipalities. The tree species prediction map is part of the Norwegian forest resource map and is used, among others, to improve maps of other variables of interest such as timber volume and biomass.