Mapping macrophytic vegetation in shallow lakes using the Compact Airborne Spectrographic Imager (CASI)

作者：Hunter P D^*; Gilvear D J; Tyler A N; Willby N J; Kelly A

来源：Aquatic Conservation: Marine and Freshwater Ecosystems , 2010, 20(7): 717-727.

DOI：10.1002/aqc.1144

• aquatic plants
• Habitats Directive
• lakes
• remote sensing
• support vector machines
• Water Framework Directive

1. The ecological status of shallow lakes is highly dependent on the abundance and composition of macrophytes. However, large-scale surveys are often confined to a small number of water bodies and undertaken only infrequently owing to logistical and financial constraints.
2. Data acquired by the Compact Airborne Spectrographic Imager-2 (CASI-2) was used to map the distribution of macrophytes in the Upper Thurne region of the Norfolk Broads, UK. Three different approaches to image classification were evaluated: (i) Euclidean minimum distance, (ii) Gaussian maximum likelihood, and (iii) support vector machines.
3. The results show macrophyte growth-habits (i. e. submerged, floating-leaved, partially-emergent, emergent) and submerged species could be mapped with a maximum overall classification accuracy of 78% and 87%, respectively. The Gaussian maximum likelihood algorithm and support vector machine returned the highest classification accuracies in each instance.
4. This study suggests that remote sensing is a potentially powerful tool for large-scale assessment of the cover and distribution of aquatic vegetation in clear water shallow lakes, particularly with respect to upscaling field survey data to a functionally relevant form, and supporting site-condition monitoring under the European Union Habitats (92/43/EEC) and Water Framework (2000/60/EC) directives.

• 出版日期2010-12