ESA Earth Home Missions Data Products Resources Applications
   
EO Data Access
How to Apply
How to Access
Services
Site Map
Frequently asked questions
Glossary
Credits
Terms of use
Contact us
Search


 
 
 

 

Testing MERIS FR capabilities and limits to map forest classes in Spain

Donata Pedrazzani(1), Yagüe Ballester Julia(1), Roberto Vallejo Bombín(2) and Guillermo Fernández Centeno(2)

(1) GMV Aerospace and Defence, S.A., Isaac Newton, 11, Tres Cantos E-28760 Madrid., Spain
(2) Ministerio de Medio Ambiente., Rios Rosas, 24., E-28003 Madrid., Spain

Abstract

Envisat MERIS Full Resolution Level 2 (FR-L2) imagery has been processed to obtain medium spatial resolution (300m) forest type maps over Navarra and La Rioja, in northern Spain; the classification method was designed and verified in close work with the State Biodiversity Office (Spanish Ministry of Environment), within the framework and standards defined by the European Global Monitoring for Environment and Security (GMES) Service Element Forest Monitoring (GSE-FM) context. Following ESA-GSE FM agreements on data supply under Category 1 (proposal no. 3810), images processed were of September 19th 2003, June 4th 2006, September 9th 2006, July 26th 2007. Original MERIS FR L2 products were georeferenced to UTM, Zone 30, European Datum 1950; channels were restructured and subset into a new dataset comprising reflectances, save blue bands, and vegetation products. Results are provided, maintaining the criteria of 81 ha. as minimum mapping unit. Thematic classification accuracy values are provided for generic forest and non-forest classes as well as for four forest types: (i) broad leaved, (ii) needle leaved, (iii) mixed forests and (iv) wooded shrubs and shrubs. National Forest Map (NFM) data were used to train and verify class categories, operating with the structural type, the dominant species (first, second and third specie) and the density of occupation or canopy cover fraction. Overall result values must be understood within the complex bio-geographic Mediterranean environment. Classification results yield an 82% accuracy of MERIS reflectances to rightly classify generic forested areas; the count is even higher (83%) when classifying broad leaved forests. Yet, limitations were found, particularly in the classification of mixed forests (16%) or in a more acute discrimination of needle leaved forests (51%). Methodological and structural arguments are given to support these results, while new processing procedures are proposed in the discussion to correct low results and enhance the future use of MERIS FR products for land applications.

 

Workshop presentation

Keywords: ESA European Space Agency - Agence spatiale europeenne, observation de la terre, earth observation, satellite remote sensing, teledetection, geophysique, altimetrie, radar, chimique atmospherique, geophysics, altimetry, radar, atmospheric chemistry