Identifying saline wetlands in an arid desert climate using Landsat remote sensing imagery. Application on Ouargla Basin, southeastern Algeria

  • Fethi Medjani
  • Belkacem Aissani
  • Sofiane Labar
  • Mohamed Djidel
  • Danielle Ducrot
  • Antoine Masse
  • C. Mei-Ling Hamilton
Original Paper

Abstract

Supervised and unsupervised satellite image classifications have progressed greatly in recent years. However, discrimination difficulties still remain among classes that directly affecting data extraction and surface mapping accuracy. The Ouargla region in southeastern Algeria is intersected by wadis, where direct communication between the shallow groundwater table and these dry, overlying ephemeral stream beds exists. Underflowing groundwater exfiltrates into low-lying aeolian blowouts or endorheic basins forming oases, chotts, and sebkhas, commonly known as saline wetlands. These wetlands are becoming increasingly vulnerable to anthropogenic stress, resulting in significant water degradation. Wetland microclimates are very important to arid regions, as they promote oasis ecosystem sustainability and preservation. High water salinity in these ecosystems, however, directly affects flourishing habitat and undermines successful desert oasis development. The objective of this work is to choose the best classification method to identify saline wetlands by comparison between the different results of land use mapping within the Ouargla basin. Landsat ETM+ (2000) satellite imagery, using visual analysis with colored compositions, has identified various forms of saline wetlands in the Ouargla region desert environment in southeast Algeria. The results show that supervised classification is validated in the identification of Saharan saline wetlands, and that support vector machine (SVM) algorithm presents the best overall accuracy.

Keywords

Remote sensing Classification Mapping Sebkha Wet saline soils Sahara 

References

  1. Acevedo S. and Jones S. (2012) An evaluation of the utility of two classifiers for mapping woody vegetation using remote sensing. School of Mathematical and Geospatial Sciences RMIT University, Melbourne, 15p. ISBN: 978–0–9872527-1-5Google Scholar
  2. Aumassip G, Dagorne A, Estorges P, Lefevbre-Witier P, Mahrour M, Marmier F, Nesson C, Rouvilloisbri-gol M, Trecolle G (1972) Brief review on the evolution of a quaternary landscape and the population of the Ouargla region/Aperçu sur l’évolution du paysage quaternaire et le peuplement de la région d’Ouargla. Li- byca XX:205–257Google Scholar
  3. Auria L. and Moro R. A. (2008) Support vector machines (SVM) as a technique for solvency analysis, DIW Berlin, Discussion Paper No. 811, German Institute for Economic Research, 16p. doi: 10.2139/ssrn.1424949
  4. Ballais, J.L. (2010) From mythical wadis to artificial rivers: the hydrography of the Lower Algerian Sahara/Des oueds mythiques aux rivières artificielles: l'hydrographie du Bas-Sahara algérien. Physio-Geo, 4. URL: http://physio-geo.revues.org/1173; doi: 10.4000/physio-geo.1173
  5. Benchallal, A., Oukil, A. and Belhadj-Aissa, A. (2009) Identification and detection using satellite imagery from soil degradation by salinity in Ouargla basin, southern Algeria/Identification et détection, par imagerie satellitaire, de la dégradation des sols par la salinité dans la cuvette de Ouargla, sud de l'Algérie. Journées d’animation scientifique (JAS09) of the AUF, Algiers. 6 pGoogle Scholar
  6. Caloz R, Pointet A (2003) Comparative analysis of contextual classification and maximum likelihood: synthesis and case studies/analyse comparative de la classification contextuelle et du maximum de vraisemblance : synthèse et cas d'étude. Télédétection 3(2–4):311–322Google Scholar
  7. Congalton RG (1991) A review of assessing the accuracy of classifications of remotely sensed data. Remote Sens Environ 37:35–46CrossRefGoogle Scholar
  8. Costa, L.T., Farinha, J.C., Hecker, N. and Tomàs Vives, P. (1996) Mediterranean wetland inventory: a reference manual. MedWet/Instituto da Conservacão da Natureza/Wetlands International Publication, Volume I., 102 pGoogle Scholar
  9. Djidel M, Labar S, Medjani F, Bouafia I (2013) Study of wetland ecological changes in desert environments using Landsat imagery and GIS/Etude des changements écologiques des zones humides en milieux désertiques en utilisant l’imagerie Landsat et le SIG. Int J Environ Water 2(5):81–87 ISSN 2052-3408Google Scholar
  10. Escadafal, R. and Pouget, J. (1989) Comparison of Landsat MSS and TM data for the mapping of surface formations in arid areas (Southern Tunisia)/Comparaison des données Landsat MSS et TM pour la cartographie des formations superficielles en zone aride (Tunisie méridionale). Proceedings of a Workshop on ‹Earthnet Pilot; December 1987, Frascati (Italie), ESA Publishers, SP-Google Scholar
  11. Girard, M.C. et Girard, C.M. (1999) Remote sensing data processing/Traitement des données de télédétection. DU- NOD Ed. Paris, pp. 59–73. ISBN: 2 10 004185Google Scholar
  12. Gunn, S.R. (1998) Support vector machines for classification and regression. Technical report, University of Southampton, 54 pGoogle Scholar
  13. Hanifi, M. (2009) Extraction of texture features for classification of satellite images./Extraction de caractéristiques de texture pour la classification d’images satellites. Doctoral Dissertation, University of Toulouse III, Paul Sabatier, pp. 49–83Google Scholar
  14. Henry, J. B. (2004) Spatial information systems for lowland flood risk management/Systèmes d’information spatiaux pour la gestion du risque d’inondation de plaine. Doctoral Dissertation, University of Louis Pasteur, Strasbourg I, pp. 29–55Google Scholar
  15. Hoang KH, Bernier M, Villeneuve J-P (2009) Changes in land use in the catchment area of the Câu river (Vietnam)—a diachronical approach/Les changements de l’occupation du sol dans le bassin versant de la rivière Câu (Viêt-nam). Essai sur une approche diachronique. Revue Télédétection 8(4):227–236Google Scholar
  16. Idder, T., Idder, A. et Mensous, M. (2011) Ecological consequences of a non-rational agricultural water management of in Algerian oases of the Sahara (Case of an Oasis in Ouargla)/Les conséquences écologiques d’une gestion non raisonnée des eaux agricoles dans les oasis du Sahara algérien (Cas de l’oasis de Ouargla). International Symposium on ecological, economic and social uses of agricultural water in Mediterranean, University of Provence, Marseille, 20–21 January 2011, 12 pGoogle Scholar
  17. Jawak SD, Luis AJ (2013) Very high-resolution satellite data for improved land cover extraction of Larsemann Hills, Eastern Antarctica. J Appl Remote Sens 7:28Google Scholar
  18. Kuncheva L. I. (2004) Combining pattern classifiers: methods and algorithms, John Wiley & Sons, 300p, doi: 10.1002/9781118914564
  19. Metternicht GI, Zinck JA (2003) Remote sensing of soil salinity: potentials and constraints. Remote Sens Environ 85(1):1–20 ISSN 0034-4257CrossRefGoogle Scholar
  20. Pony, O., Descombes, X. and Zerubia, J. (2000) Hyperspectral satellite image classification in rural and suburban areas/Classification d’images satellitaires hyperspectrales en zone rurale et périurbaine. Institut National de Recherche en Informatique et en Automatique, N 4008/ISRN INRIA/RR-4008-FR, ISSN 0249–6399Google Scholar
  21. Pozzi, F. and Small, C. (2002) Vegetation and population density in urban and suburban areas in the U.S.A. Proceedings of the IEEE Joint Workshop on Remote Sensing and Data Fusion over Urban Areas, 8-9 November 2001, Rome, Italy, pp. 250-254,Google Scholar
  22. Shafri HZM, Suhaili A, Mansor S (2007) The performance of maximum likelihood, spectral angle mapper, neural network and decision tree classifiers in hyperspectral image analysis. J Comput Sci 3(6):419–423 ISSN 1549-3636CrossRefGoogle Scholar
  23. Soyer J, Wilmet J (1983) Etude de l'environnement de Lubumbashi de 1973 à 1981 à l'aide de la télédétection par satellite : croissance urbaine et déboisement. Geo-EcoTrop 7:67–81Google Scholar
  24. Ünsalan C. and Boyer K. L. (2011) Multispectral satellite image understanding: from land classification to building and road detection, Series: Advances in Computer Vision and Pattern Recognition, Springer, ISBN: 978–0–85729-666-5Google Scholar
  25. Yang C, Everitt JH, Fletcher RS, Jensen RR, Mausel PW (2009) Evaluating AISA + hyperspectral imagery for mapping black mangrove along the South Texas Gulf Coast. Photogramm Eng Remote Sens 75(4):425–435CrossRefGoogle Scholar

Copyright information

© Saudi Society for Geosciences 2017

Authors and Affiliations

  • Fethi Medjani
    • 1
  • Belkacem Aissani
    • 1
  • Sofiane Labar
    • 2
  • Mohamed Djidel
    • 1
  • Danielle Ducrot
    • 3
  • Antoine Masse
    • 3
  • C. Mei-Ling Hamilton
    • 4
  1. 1.Laboratory of Geology of the SaharaUniversity Kasdi Merbah OuarglaOuarglaAlgeria
  2. 2.Faculty of Natural and Life SciencesUniversity of Chadli Bendjedid El-TarfAnnabaAlgeria
  3. 3.Biosphere Space Studies Centre “CESBIO”University of Paul SabatierToulouse Cedex 9France
  4. 4.BakersfieldUSA

Personalised recommendations