Belgian Ardenne ecoregion; per-pixel classification; random forest; remote sensing; satellites; tree species; télédétection; écorégion de l'Ardenne belge; espèces d'arbre; satellite; classification par pixel; forêt aléatoire
Abstract :
[en] Description of the subject.Understanding the current situation and evolution of forests is essential for a sustainable management plan that maintains forests’ ecological and socio-economic functions. Remote sensing is a helpful tool in developing this knowledge.
Objectives. This paper investigates the new opportunities offered by using Sentinel-2 (S2) imagery for forest mapping in Belgian Ardenne ecoregion. The first classification objective was to create a forest map at the regional scale. The second objective was the discrimination of 11 forest classes (Fagus sylvatica L., Betula sp., Quercus sp., other broad-leaved stands, Pseudotsuga menziesii (Mirb.) Franco, Larix sp., Pinus sylvestris L., Picea abies (L.) H.Karst., young needle-leaved stands, other needle-leaved stands, and recent clear-cuts).
Method. Two S2 scenes were used and a series of spectral indices were computed for each. We applied supervised pixel based classifications with a Random Forest classifier. The classification models were processed with a pure S2 dataset and with additional 3D data to compare obtained precisions.
Results. 3D data slightly improved the precision of each objective, but the overall improvement in accuracy was only significant for objective 1. The produced forest map had an overall accuracy of 93.3%. However, the model testing tree species discrimination was also encouraging, with an overall accuracy of 88.9%.
Conclusions. Because of the simple analyses done in this study, results need to be interpreted with caution. However, this paper confirms the great potential of S2 imagery, particularly SWIR and red-edge bands, which are the most important S2 bands in our study. [fr] Description du sujet. Étudier l’état et l’évolution des forêts est essentiel pour assurer une gestion durable maintenant leurs fonctions écologiques et socio-économiques. La télédétection est un outil précieux pour le développement de ces connaissances.
Objectifs. Cette étude analyse l’opportunité offerte par l’imagerie Sentinel-2 (S2) pour cartographier les forêts de l’écorégion de l’Ardenne belge. Le premier objectif de classification était la création d’une carte forestière à l’échelle régionale. Le second objectif était la discrimination de 11 classes forestières (Fagus sylvatica L., Betula sp., Quercus sp., other broad-leaved stands, Pseudotsuga menziesii (Mirb.) Franco, Larix sp., Pinus sylvestris L., Picea abies (L.) H.Karst., young needle-leaved stands, other needle-leaved stands, and recent clear-cuts).
Méthode. Deux scènes S2 ont été utilisées et une série d’indices spectraux ont été générés pour chacune d’entre elles. Nous avons réalisé une classification supervisée par pixel avec l’algorithme de classification Random Forest. Les modèles de classification ont été générés avec un jeu de données S2 pur et avec des données 3D supplémentaires pour comparer les précisions obtenues.
Résultats. Les données 3D ont légèrement amélioré la précision de chaque objectif, mais l’amélioration globale de précision fut uniquement significative pour l’objectif 1. La carte forestière produite avait une précision globale de 93,3 %. Le modèle testant la discrimination des espèces d’arbre fut encourageant également, avec une précision globale de 88,9 %.
Conclusions. Tenant compte des simples analyses réalisées dans cette étude, les résultats doivent être interprétés avec prudence. Cependant, ce travail confirme le grand potentiel de l’imagerie S2, particulièrement les bandes SWIR et red-edge, qui jouèrent un rôle essentiel dans ce travail.
Pour citer cet article
Corentin Bolyn, Adrien Michez, Peter Gaucher, Philippe Lejeune & Stéphanie Bonnet, «Forest mapping and species composition using supervised per pixel classification of Sentinel-2 imagery», BASE [En ligne], Volume 22 (2018), Numéro 3, URL : https://popups.uliege.be:443/1780-4507/index.php?id=16524.
Agapiou A. et al., 2011. The importance of accounting for atmospheric effects in the application of NDVI and interpretation of satellite imagery supporting archaeological research: the case studies of Palaepaphos and Nea Paphos sites in Cyprus. Remote Sens., 3(12), 2605-2629.
Alderweireld M., Burnay F., Pitchugin M. & Lecomte H., 2015. Inventaire forestier wallon. Résultats 1994-2012. Namur, Belgique: Service Public de Wallonie.
Blackburn G.A., 1998. Quantifying chlorophylls and caroteniods at leaf and canopy scales: an evaluation of some hyperspectral approaches. Remote Sens. Environ., 66(3), 273-285.
Breiman L., 2001. Random forests. Mach. Learn., 45(1), 5-32.
Chen P.-F. et al., 2010. New index for crop canopy fresh biomass estimation. Spectrosc. Spectral Anal., 30(2), 512-517.
Datt B., 1999. Visible/near infrared reflectance and chlorophyll content in Eucalyptus leaves. Int. J. Remote Sens., 20(14), 2741-2759.
Domenech E. & Mallet C., 2014. Change detection in high-resolution land use/land cover geodatabases (at object level). EuroSDR, 64.
European Commission, 2015. Copernicus: Europe’s Eyes on Earth. Luxembourg: Publications Office.
European Space Agency, 2015. Sentinel-2a MSI Spectral Responses.xlsx, https://earth.esa.int/web/sentinel/document-library/latest-documents/-/asset_publisher/EgUy8pfXboLO/content/sentinel-2a-spectral-respo nses;jsessionid=AA5AEEAE5B3515EFB534D44F 239D5FD1.jvm2?redirect=https%3A%2F%2Fearth. esa.int%2Fweb%2Fsentinel%2Fdocument-library%2Flatest-documents%3Bjsessionid%3DAA5A EEAE5B3515EFB534D44F239D5FD1.jvm2%3Fp_p_ id%3D101_INSTANCE_EgUy8pfXboLO%26p_p_ lifecycle%3D0%26p_p_state%3Dnormal%26p_p_ mode%3Dview%26p_p_col_id%3Dcolumn-1%26p_p_ col_pos%3D1%26p_p_col_count%3D2, (18/12/2017).
Fassnacht F.E. et al., 2016. Review of studies on tree species classification from remotely sensed data. Remote Sens. Environ., 186, 64-87.
Filella I. & Penuelas J., 1994. The red edge position and shape as indicators of plant chlorophyll content, biomass and hydric status. Int. J. Remote Sens., 15(7), 1459-1470.
Forster M. & Kleinschmit B., 2014. Significance analysis of different types of ancillary geodata utilized in a multisource classification process for forest identification in Germany. IEEE Trans. Geosci. Remote Sens., 52(6), 3453-3463.
Führer E., 2000. Forest functions, ecosystem stability and management. For. Ecol. Manage., 132(1), 29-38.
Gao B.-C., 1996. NDWI. A normalized difference water index for remote sensing of vegetation liquid water from space. Remote Sens. Environ., 58(3), 257-266.
Genuer R., Poggi J.-M. & Tuleau-Malot C., 2015. VSURF: An R package for variable selection using random forests. R Journal, 7(2), 19-33.
Genuer R., Poggi J.-M. & Tuleau-Malot C., 2016. VSURF: variable selection using random forests. R package version 1.0.3.
Gitelson A.A., Kaufman Y.J. & Merzlyak M.N., 1996. Use of a green channel in remote sensing of global vegetation from EOS-MODIS. Remote Sens. Environ., 58(3), 289-298.
Guanter L., Gomez-Chova L. & Moreno J., 2008. Coupled retrieval of aerosol optical thickness, columnar water vapor and surface reflectance maps from ENVISAT/ MERIS data over land. Remote Sens. Environ., 112(6), 2898-2913.
Hagolle O., Huc M., Pascual D. & Dedieu G., 2015. A multi-temporal and multi-spectral method to estimate aerosol optical thickness over land, for the atmospheric correction of FormoSat-2, LandSat, VENS and Sentinel-2 Images. Remote Sens., 7(3), 2668-2691.
Huete A.R., 1988. A soil-adjusted vegetation index (SAVI). Remote Sens. Environ., 25(3), 295-309.
Immitzer M., Atzberger C. & Koukal T., 2012. Tree species classification with random forest using very high spatial resolution 8-band worldview-2 satellite data. Remote Sens., 4(9), 2661-2693.
Immitzer M., Vuolo F. & Atzberger C., 2016. First experience with sentinel-2 data for crop and tree species classifications in central Europe. Remote Sens., 8(3), 166.
Inglada J. et al., 2017. Operational high resolution land cover map production at the country scale using satellite image time series. Remote Sens., 9(1), 95.
Jacques D.C. et al., 2014. Monitoring dry vegetation masses in semi-arid areas with MODIS SWIR bands. Remote Sens. Environ., 153, 40-49.
Jensen J.R., 2005. Introductory digital image processing: a remote sensing perspective. Upper Saddle River, NJ, USA: Prentice Hall.
Kaufman Y. et al., 1997. Passive remote sensing of tropospheric aerosol and atmospheric correction for the aerosol effect. J. Geophys. Res., 102(D14), 16815-16830.
Kumar N., 2006. Multispectral image analysis using the object-oriented paradigm. Boca Raton, FL, USA: CRC Press.
Lacaux J. et al., 2007. Classification of ponds from high-spatial resolution remote sensing: application to Rift Valley fever epidemics in Senegal. Remote Sens. Environ., 106(1), 66-74.
Le Maire G., François C. & Dufrêne E., 2004. Towards universal broad leaf chlorophyll indices using PROSPECT simulated database and hyperspectral reflectance measurements. Remote Sens. Environ., 89(1), 1-28.
Liaw A. & Wiener M., 2002. Classification and regression by randomForest. R News, 2(3), 18-22.
Lichtenthaler H.K. et al., 1996. Detection of vegetation stress via a new high resolution fluorescence imaging system. J. Plant Physiol., 148(5), 599-612.
Lillesand T.M., Kiefer R.W. & Chipman J.W., 2008. Remote sensing and image interpretation. Hoboken, NJ, USA: John Wiley & Sons.
Lindenmayer D.B., Margules C.R. & Botkin D.B., 2000. Indicators of biodiversity for ecologically sustainable forest management. Conserv. Biol., 14(4), 941-950.
McDermid G. et al., 2009. Remote sensing and forest inventory for wildlife habitat assessment. For. Ecol. Manage., 257(11), 2262-2269.
McRoberts R. & Tomppo E., 2007. Remote sensing support for national forest inventories. Remote Sens. Environ., 110(4), 412-419.
Michez A., Piégay H., Lejeune P. & Claessens H., 2017. Multi-temporal monitoring of a regional riparian buffer network (> 12,000 km) with LiDAR and photogrammetric point clouds. J. Environ. Manage., 202(2), 424-436.
Müller-Wilm U., 2016. Sentinel-2 MSI. Level-2a prototype processor installation and user manual. Darmstadt, Germany: Telespazio Vega.
Olofsson P., Foody G.M., Stehman S.V. & Woodcock C.E., 2013. Making better use of accuracy data in land change studies: estimating accuracy and area and quantifying uncertainty using stratified estimation. Remote Sens. Environ., 129, 122-131.
Pinty B. & Verstraete M.M., 1992. GEMI: a non-linear index to monitor global vegetation from satellites. Vegetatio, 101(1), 15-20.
R Core Team, 2016. R: a language and environment for statistical computing. Vienna: R Foundation for Statistical Computing.
Radoux J. et al., 2016. Sentinel-2’s potential for sub-pixel landscape feature detection. Remote Sens., 8(6), 488.
Schuster C., Förster M. & Kleinschmit B., 2012. Testing the red edge channel for improving land-use classifications based on high-resolution multi-spectral satellite data. Int. J. Remote Sens., 33(17), 5583-5599.
Shahi K. et al., 2015. A novel spectral index to automatically extract road networks from WorldView-2 satellite imagery. Egypt. J. Remote Sens. Space Sci., 18(1), 27-33.
Song C. et al., 2001. Classification and change detection using Landsat TM data: when and how to correct atmospheric effects? Remote Sens. Environ., 75(2), 230-244.
Sripada R.P., Heiniger R.W., White J.G. & Meijer A.D., 2006. Aerial color infrared photography for determining early in-season nitrogen requirements in corn. Agron. J., 98(4), 968.
Stratoulias D. et al., 2015. Evaluating Sentinel-2 for lakeshore habitat mapping based on airborne hyperspectral data. Sensors, 15(9), 22956-22969.
Suhet & Hoersch B., 2015. Sentinel-2 user handbook. European Space Agency.
Tomppo E., Gschwantner T., Lawrence M. & McRoberts R.E., eds, 2010. National forest inventories. Dordrecht, The Netherlands: Springer Netherlands.
Tucker C.J., 1979. Red and photographic infrared linear combinations for monitoring vegetation. Remote Sens. Environ., 8(2), 127-150.
Van Deventer A.P., Ward A.D., Gowda P.H. & Lyon J.G., 1997. Using thematic mapper data to identify contrasting soil plains and tillage practices. Photogramm. Eng. Remote Sens., 63, 87-93.
Vogelmann J.E. & Rock B.N., 1985. Spectral characterization of suspected acid deposition damage in red spruce (Picea Rubens) stands from Vermont. In: Proceedings of the Airborne Imaging Spectrometer Data Analysis Workshop, April 8-10, 1985, Jet Propulsion Laboratory, California Institute of Technology, Pasadena, California, United States, 51-55.
Waser L. et al., 2011. Semi-automatic classification of tree species in different forest ecosystems by spectral and geometric variables derived from Airborne Digital Sensor (ADS40) and RC30 data. Remote Sens. Environ., 115(1), 76-85.
Wulf H. & Stuhler S., 2015. Sentinel-2: land cover, preliminary user feedback on Sentinel-2a data. In: Proceedings of the Sentinel-2a expert users technical meeting, 29-30 September 2015, Frascati, Italy.
