首页    期刊浏览 2024年12月04日 星期三
登录注册

文章基本信息

  • 标题:ACCURACY OF GLOBAL LAND COVER DATASETS: EXAMPLE OF GLOBAL LAND COVER CHARACTERIZATION DATA BASE
  • 其他标题:DOKŁADNOŚĆ GLOBALNYCH MAP UŻYTKOWANIA ZIEMI I POKRYCIA TERENU NA PRZYKŁADZIE GLOBAL LAND COVER CHARACTERIZATION DATA BASE
  • 本地全文:下载
  • 作者:Jacek Kozak
  • 期刊名称:Archiwum Fotogrametrii, Kartografii i Teledetekcji
  • 印刷版ISSN:2083-2214
  • 出版年度:2003
  • 卷号:13
  • 语种:English
  • 出版社:Main Board of Association of Polish Surveyors
  • 摘要:The study attempts to estimate the accuracy of the delimitation of European forests within the Global Land Cover Characterization (GLCC) Data Base. As a reference,CORINE Major Land Cover Types of Europe data set was used. The analysis was carried out for data aggregated to two classes: forests and non-forest areas. Matrix overlays allowed computation of accuracy indices. Kappa coefficient and the difference of actual (CORINE) and predicted (GLCC) forest proportion were mapped for spatial units 100 × 100 km. The area of forests derived from the CORINE data equals 1.18·106 km2 ,and the value predicted is almost exactly the same,1.17·106 km2 . User’s and producer’s accuracies of forest delimitation equal 57%. Local differences between actual and predicted forest proportion may exceed 25 percentage points. Overestimation of the forest area occurs in Scandinavia,while underestimation in lowland Western and Central Europe. Kappa coefficient varies from 0 up to almost 70%. Low values were found in lowland Central and Western Europe,Ireland,the Mediterranean and Scandinavia. Values exceeding 40% were observed in a belt stretching from northern Spain to Bulgaria. Differences between actual and predicted forest proportion,as well as Kappa coefficient depend on the actual forest proportion. Errors of forest delimitation within the GLCC data set are related mainly to the classification of coarse resolution data in areas of highly diverse land cover pattern. The overestimation of dominant land cover classes in the GLCC leads to locally inaccurate predictions,which should be calibrated for the purposes of regional research.
  • 其他摘要:W badaniach oceniono dokładność wyznaczenia powierzchni europejskich lasów w bazie danych Global Land Cover Characterization (GLCC). Jako dane referencyjne wykorzystana została mapa CORINE Major Land Cover Types of Europe. Dla danych zagregowanych do dwóc
  • 关键词:global data;forests;Europe;accuracy
  • 其他关键词:dane globalne;lasy;Europa;dokładność
国家哲学社会科学文献中心版权所有