Vegetation mapping in an alkali landscape - application of airborne hyperspectral data
DOI:
https://doi.org/10.56617/tl.3635Keywords:
vegetation mapping, Maximum Likelihood Classifier (MLC), Random Forest (RF), Support Vector Machine (SVM), alkali grassland, alkali meadow, alkali marshAbstract
We mapped the vegetation of an alkali landscape using airborne hyperspectral data. The aim of our study was to test the applicability of hyperspectral data in mapping of these complex habitats. We tested the performance of three frequently applied classifiers (Maximum Likelihood Classifier – MLC, Random Forest – RF and Support Vector Machine – SVM) using 10 and 30 training pixels and MNF transformed bands. For data collection we used an AISA EAGLE II sensor, which produced 1 m ground pixel size. Based on the coenological categories, land cover types and the experiences of the preliminary field survey we assigned the training polygons to 20 classes. Classes were aggregated to four main habitat types: steppes, open alkali swards, alkali meadows and alkali and non-alkali marshes. We found that the SVM and the RF classifiers provided a high overall accuracy for most of the classes independently from the number of training pixels. Even though the MLC classifier provided a high overall accuracy when using 30 training pixels, its efficiency was low when using 10 training pixels. Overall accuracies increased considerably in case of all classifiers when using aggregated habitat types. Based on our results, in complex, open habitats the SVM is the most effective classifier, it provided the highest accuracy. Furthermore it was the less sensitive for the low number of training pixels, thus can be effective in those cases when the number of training pixels is low for some classes.
References
Adam, E., Mutanga, O., Rugege, D. 2010: Multispectral and hyperspectral remote sensing for identification and mapping of wetland vegetation: A review. Wetlands Ecology Management 18: 281−296. https://doi.org/10.1007/s11273-009-9169-z
Alexander, C., Deák, B., Kania, A., Mücke, W., Heilmeier, H. 2015: Classification of vegetation in an open landscape using full-waveform airborne laser scanner data. International Journal of Applied Earth Observations and Geoinformation 41: 76−87. https://doi.org/10.1016/j.jag.2015.04.014
Alexander, C., Deák, B., Heilmeier, H. 2016: Micro-topography driven vegetation patterns in open mosaic landscapes. Ecological Indicators 60: 906−920. https://doi.org/10.1016/j.ecolind.2015.08.030
Ambrus, A., Burai, P., Lénárt, Cs., Enyedi, P., Kovács, Z. 2015: Estimating biomass of winter wheat using narrowband vegetation indices for precision agriculture. Journal of Central European Green Innovation 3: 13−22.
Belluco, E., Camuffo, M., Ferrari, S., Modenese, L., Silvestri, S., Marani, A., Marani, M. 2006: Mapping salt-marsh vegetation by multispectral and hyperspectral remote sensing. Remote Sensing of Environment 105: 54–67. https://doi.org/10.1016/j.rse.2006.06.006
Borhidi A., Kevey B., Lendvai G. 2012: Plant Communities of Hungary. Akadémiai Kiadó, Budapest.
Borre, J.V., Paelinckx, D., Mücher, C.A., Kooistra, L., Haest, B., Blust, G.D., Schmidt, A.M. 2011: Integrating remote sensing in Natura 2000 habitat monitoring: Prospects on the way forward. Journal for Nature Conservation 19: 116–125. https://doi.org/10.1016/j.jnc.2010.07.003
Burai, P., Deák, B., Valkó, O., Tomor, T. 2015: Classification of herbaceous vegetation using airborne hyperspectral imagery. Remote Sensing 7: 2046–2066. https://doi.org/10.3390/rs70202046
Deák B, Tóthmérész B. 2006: Kaszálás hatása a növényzetre a Nyírőlapos (Hortobágy) három növénytársulásában. In: Molnár E. (szerk.) Kutatás, oktatás értékteremtés. A 80 éves Précsényi István köszöntése. MTA Ökológiai és Botanikai Kutatóintézet, Vácrátót, pp. 169-180.
Deák, B., Valkó, O., Alexander, C., Mücke, W., Kania, A., Tamás, J., Heilmeier, H. 2014a: Fine-scale vertical position as an indicator of vegetation in alkali grasslands - case study based on remotely sensed data. Flora 209: 693–697. https://doi.org/10.1016/j.flora.2014.09.005
Deák, B., Valkó, O., Török, P., Tóthmérész, B. 2014b: Solonetz meadow vegetation (Beckmannion eruciformis) in East-Hungary – an alliance driven by moisture and salinity. Tuexenia 34:187–203.
Deák, B., Valkó, O., Török, P., Kelemen, A., Miglécz, T., Szabó, Sz., Szabó, G., Tóthmérész, B. 2015a: Micro-topographic heterogeneity increases plant diversity in old stages of restored grasslands. Basic and Applied Ecology 16: 291−299. https://doi.org/10.1016/j.baae.2015.02.008
Deák, B., Valkó, O., Török, P., Kelemen, A., Tóth, K., Miglécz, T., Tóthmérész, B. 2015b: Reed cut, habitat diversity and productivity in wetlands. Ecological Complexity 22: 121–125. https://doi.org/10.1016/j.ecocom.2015.02.010
Harris, A., Charnock, R., Lucas, R.M. 2015: Hyperspectral remote sensing of peatland floristic gradients. Remote Sensing of Environment 162: 99–111. https://doi.org/10.1016/j.rse.2015.01.029
Hestir, E., Khanna, S., Andrew, M.E., Santos, M.J., Viers, J.H., Greenberg, J.A., Rajapakse, S.S., Ustin, S.L. 2008: Identification of invasive vegetation using hyperspectral remote sensing in the California delta ecosystem. Remote Sensing of Environment 112: 4034–4047. https://doi.org/10.1016/j.rse.2008.01.022
Huang, C., Asner, G.P. 2009: Applications of remote sensing to alien invasive plant studies. Sensors 9: 4869– 4889. https://doi.org/10.3390/s90604869
Hurcom, S.J.; Harrison, A. R. 1998: The NDVI and spectral decomposition for semi-arid vegetation abundance estimation. International Journal of Remote Sensing 19: 3109–3126. https://doi.org/10.1080/014311698214217
Kelemen, A., Török, P., Valkó, O., Miglécz, T., Tóthmérész, B. 2013: Mechanisms shaping plant biomass and species richness: plant strategies and litter effect in alkali and loess grasslands. Journal of Vegetation Science 24: 1195–1203. https://doi.org/10.1111/jvs.12027
Kelemen, A., Török, P., Valkó, O., Deák, B., Tóth, K., Tóthmérész, B. 2015: Both facilitation and limiting similarity shape the species coexistence in dry alkali grasslands. Ecological Complexity 21: 34–38. https://doi.org/10.1016/j.ecocom.2014.11.004
Lawrence, R., Wood, S., Sheley, R. 2006: Mapping invasive plants using hyperspectral imagery and Breiman Cutler classifications (randomForest). Remote Sensing of Environment 100: 356–362. https://doi.org/10.1016/j.rse.2005.10.014
Lukács, B.A, Török, P, Kelemen, A., Várbíró, G., Radócz, Sz., Miglécz, T., Tóthmérész, B., Valkó, O. 2015: Rainfall fluctuations and vegetation patterns in alkali grasslands – Self-organizing maps in vegetation analysis. Tuexenia 35: 381–397.
Mirik, M., Ansley, R.J., Steddom, K., Jones, D.C., Rush, C.M., Michels, G.J. Elliott, N.C. 2013: Remote distinction of a noxious weed (musk thistle: Carduus nutans) using airborne hyperspectral imagery and the Support Vector Machine Classifier. Remote Sensing 5: 612–630. https://doi.org/10.3390/rs5020612
Molnár, Z.; Bölöni, J., Biró, M., Horváth, F. 2008: Distribution of the Hungarian (semi-)natural habitats I. Marshes and grasslands. Acta Botanica Hungarica 50: 59–105. https://doi.org/10.1556/ABot.50.2008.Suppl.5
Mücke, W., Deák, B., Schroiff, A., Hollaus, M., Pfeifer, N. 2013: Estimation of dead wood using small footprint airborne laser scanning data. Canadian Journal of Remote Sensing 39: 32–40.
Paruelo, J. M., Epstein, H. E., Lauenroth, W. K., Burke, I. C. 1997: ANPP estimates from NDVI for the central grassland region of the United States. Ecology 78: 953−958. https://doi.org/10.1890/0012-9658(1997)078[0953:AEFNFT]2.0.CO;2
Pettorelli, N., Vik, J. O., Mysterud, A., Gaillard, J. M., Tucker, C. J., Stenseth, N. C. 2005: Using the satellite-derived NDVI to assess ecological responses to environmental change. Trends in Ecology & Evolution 20: 503−510. https://doi.org/10.1016/j.tree.2005.05.011
Pettorelli, N. 2013: The Normalised Difference Vegetation Index; Oxford University Press, Oxford. https://doi.org/10.1093/acprof:osobl/9780199693160.001.0001
Rabe, A., Jakimow, B., Held, M., Van Der Linden, S., Hostert, P. 2014: EnMAP-Box, Version 2.0. (http://www.enmap.org).
Sokal, R.R., Rohlf, F.J. 1987: Introduction to Biostatistics. W.H. Freeman and Company, 2nd edition, New York. p. 368.
Szabó, Sz., Szilassi, P., Csorba, P. 2012: Tools for landscape ecological planning– Scale, and aggregation sensitivity of the contagion type landscape metric indices. Carpathian Journal of Earth and Environmental Sciences 7: 127−136.
Thenkabail, P.S. 2011: Hyperspectral Remote Sensing of Vegetation. Taylor & Francis, New York. p. 782. TÓTH, K., Hüse, B. 2014: Soil seed banks in loess grasslands and their role in grassland recovery. Applied Ecology and Environmental Research 12: 537–547. https://doi.org/10.15666/aeer/1202_537547
Tóth, T., Kertész, M. 1996: Application of soil–vegetation correlation to optimal resolution mapping of solonetzic rangeland. Arid Soil Research and Rehabilitation 10: 1–12. https://doi.org/10.1080/15324989609381415
Valkó, O., Tóthmérész, B., Kelemen, A., Simon, E., Miglécz, T., Lukács, B., Török, P. 2014: Environmental factors driving vegetation and seed bank diversity in alkali grasslands. Agriculture, Ecosystems & Environment 182: 80–87. https://doi.org/10.1016/j.agee.2013.06.012
Zlinszky, A., Deák, B., Kania, A., Schroiff, A., Pfeifer, N. 2015: Mapping Natura 2000 habitat conservation status in a pannonic salt steppe with airborne laser scanning. Remote Sensing 7: 2991–3019. https://doi.org/10.3390/rs70302991
Zlinszky, A., Schroiff, A., Kania, A., Deák, B., Mücke, W., Vári, Á., Székely, B., Pfeifer, N. 2014: Categorizing grassland vegetation with full-waveform airborne laser scanning: a feasibility study for detecting Natura 2000 habitat types. Remote Sensing 6: 8056–8087. https://doi.org/10.3390/rs6098056
Downloads
Published
Issue
Section
License
Copyright (c) 2016 Burai Péter, Lénárt Csaba, Valkó Orsolya, Bekő László, Szabó Zsuzsanna, Deák Balázs
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
A folyóirat Open Access (Gold). Cikkeire a Creative Commons 4.0 standard licenc alábbi típusa vonatkozik: CC-BY-NC-ND-4.0. Ennek értelmében a mű szabadon másolható, terjeszthető, bemutatható és előadható, azonban nem használható fel kereskedelmi célokra (NC), továbbá nem módosítható és nem készíthető belőle átdolgozás, származékos mű (ND). A licenc alapján a szerző vagy a jogosult által meghatározott módon fel kell tüntetni a szerző nevét és a szerzői mű címét (BY).
