| Titre : | Global Wetland Datasets: a Review (2017) |
| Auteurs : | Shengjie Hu, Auteur ; Zhenguo Niu, Auteur ; Yanfen Chen, Auteur |
| Type de document : | Article |
| Dans : | Wetlands (37(5) 10/2017) |
| Article en page(s) : | 807-817 |
| Langues: | Anglais |
| Mots-clés : | Zone humide ; Analyse données ; Méthodologie ; télédétection |
| Mots-clés: | Global wetland-related datasets, Wetland classification, Wetland inconsistency, Global wetland mapping |
| Résumé : | Accurate wetland datasets are indispensable for generating polices on wetland conservation and appropriate land uses, global climate change studies, and biodiversity conservation. Although increasing numbers of global wetland-related datasets have been established, the extensive disagreements among these datasets are prominent. In particular, estimates of global wetland area range from 0.54 to 21.26 million km2 and the class-specific spatial consistency of wetlands is less than 1%. The different definitions of ‘wetland’ and the intrinsic features of wetlands contribute to this extensive inconsistency. Given the various requirements of wetland-oriented data products, it is important to conduct comprehensive wetland mapping at global scale. The Ramsar wetland definition is recommended and a hierarchical and flexible structure of wetland classification system is preferred for future global wetland datasets. Time-series satellite imagery at 250–1000 m spatial resolution is preferred to characterize wetland dynamics by combination of passive/active SAR data and other ancillary data, such as topography, climate, and soil data. The classification tree method for classification of massive satellite imagery and big-data technology for sample datasets are promising for the enhancement of wetland map product accuracy. |
| En ligne : | http://link.springer.com/10.1007/s13157-017-0927-z |



