SONG Chuangye, HUANG Chong, LIU Huiming. Predictive Vegetation Mapping Approach Based on Spectral Data, DEM and Generalized Additive Models[J]. Chinese Geographical Science, 2013, 23(3): 331-343. doi: 10.1007/s11769-013-0590-0
Citation: SONG Chuangye, HUANG Chong, LIU Huiming. Predictive Vegetation Mapping Approach Based on Spectral Data, DEM and Generalized Additive Models[J]. Chinese Geographical Science, 2013, 23(3): 331-343. doi: 10.1007/s11769-013-0590-0

Predictive Vegetation Mapping Approach Based on Spectral Data, DEM and Generalized Additive Models

doi: 10.1007/s11769-013-0590-0
Funds:  Under the auspices of National Natural Science Foundation of China (No. 41001363)
More Information
  • Corresponding author: SONG Chuangye. E-mail: songcy@ibcas.ac.cn
  • Received Date: 2012-04-18
  • Rev Recd Date: 2012-09-21
  • Publish Date: 2013-05-29
  • This study aims to provide a predictive vegetation mapping approach based on the spectral data, DEM and Generalized Additive Models (GAMs). GAMs were used as a prediction tool to describe the relationship between vegetation and environmental variables, as well as spectral variables. Based on the fitted GAMs model, probability map of species occurrence was generated and then vegetation type of each grid was defined according to the probability of species occurrence. Deviance analysis was employed to test the goodness of curve fitting and drop contribution calculation was used to evaluate the contribution of each predictor in the fitted GAMs models. Area under curve (AUC) of Receiver Operating Characteristic (ROC) curve was employed to assess the results maps of probability. The results showed that: 1) AUC values of the fitted GAMs models are very high which proves that integrating spectral data and environmental variables based on the GAMs is a feasible way to map the vegetation. 2) Prediction accuracy varies with plant community, and community with dense cover is better predicted than sparse plant community. 3) Both spectral variables and environmental variables play an important role in mapping the vegetation. However, the contribution of the same predictor in the GAMs models for different plant communities is different. 4) Insufficient resolution of spectral data, environmental data and confounding effects of land use and other variables which are not closely related to the environmental conditions are the major causes of imprecision.
  • [1] Austin M P, 2002. Spatial prediction of species distribution: An interface between ecological theory and statistical modeling. Ecological Modeling, 157(2-3): 101-118. doi:  org/10.1016/S0304-3800(02)00205-3
    [2] Braun-Blanquet J, 1933. Phytosociological nomenclature. Ecology, 14(3): 315-317. doi:  org/10.2307/1932802
    [3] Dirnböck T, Dullinger S, Gottfried M et al., 2003. Mapping alpine vegetation based on image analysis, topographic variables and Canonocal Correspondence Analysis. Applied Vegetation Sciences, 6(1): 85-96. doi:  10.1111/j.1654-109X.2003.tb00567.x
    [4] Dirnböck T, Hobbs R J, Lambeck R J et al., 2002. Vegetation distribution in relation to topographically driven processes in south-western Australia. Applied Vegetation Science, 5(1): 147-158. doi:  10.1111/j.1654-109X.2002.tb00544.x
    [5] Dirnbök T, Dullinger S,Grabherr G, 2001. A new grassland community in the Eastern Alps (Austria): Evidence of envi-ronmental distribution limits of endemic plant communities. Phytocoenologia, 31(4): 521-536.
    [6] Dullinger S, Dirnbök T, Grabherr G, 2003. A resampling approach for evaluating effects of pasture abandonment on subalpine plant species diversity. Journal of Vegetation Science, 14(2): 243-252. doi:  10.1111/j.1654-1103.2003.tb02149.x
    [7] Fawcett T, 2006. An introduction to ROC analysis. Pattern Rec-ognition Letter, 27(8): 861-874. doi: 10.1016/j.patrec.2005.10. 010
    [8] Ferrier S, Guisan A, 2006. Spatial modeling of biodiversity at the community level. Journal of Applied Ecology, 43(3): 393-404. doi:  10.1111/j.1365-2664.2006.01149.x
    [9] Ferrier S, Watson G, Pearce J et al., 2002. Extended statistical approaches to modelling spatial pattern in biodiversity: The north-east New South Wales experience. I. Species-level mod-elling. Biodiversity Conservation, 11(12): 2275-2307. doi:  10.1023/A:1021302930424
    [10] Fielding A H, Bell J F, 1997. A review of methods for the as-sessment of prediction errors in conservation presence/absence models. Environment Conservation, 24(1): 38-49. doi:  10.1017/S0376892997000088
    [11] Goward S N, Huemmerich K F, Waring R H, 1994. Visible-near infrared spectral reflectance of landscape components in West Oregon. Remote Sensing of Environment, 47(2): 190-203. doi:  10.1016/0034-4257(94)90155-4
    [12] Guisan A, Zimmermann N E, 2000. Predictive habitat distribution models in ecology. Ecological Modelling, 135(2-3): 147-186. doi:  10.1016/S0304-3800(00)00354-9
    [13] Hastie T J, Tibshirani R, 1990. Generalized Additive Models. London: Chapman and Hall.
    [14] He Xiaohui, Wen Zhongming, Wang Jinxin, 2008. Spatial distri-bution of major grassland species and its relations to environ-ment in Yanhe River catchment based on generalized additive model. Chinese Journal of Ecology, 27(10): 1718-1724. (in Chinese)
    [15] Hosmer D W, Lemeshow S, 2000. Applied Logistic Regression. New York: Wiley & Sons.
    [16] Joseph C G, Gary D R, 2004. Expert System: Principles and Pro-gramming. Stamford: Thomson Learning.
    [17] Karl J W, Maurer B A, 2010. Spatial dependence of predictions from image segmentation: A variogram-based method to de-termine appropriate scales for producing land-management information. Ecological Informatics, 5(3): 194-202. doi: 10. 1016/j.ecoinf.2010.02.004
    [18] Lehmann A, Overton J M, Leathwick J R, 2002. GRASP: Gene-ralized regression analysis and spatial prediction. Ecological modelling, 157(2-3): 189-207. doi: 10.1016/S0304-3800(02) 00354-X
    [19] Liu Guangsong, 1996. Soil Physical and Chemical Analysis & Description of Soil Profiles. Beijing: China Standards Press. (in Chinese)
    [20] Miller J, Franklin J, Aspinall R, 2007. Incorporating spatial de-pendence in predictive vegetation models. Ecological Modelling, 202(3-4): 225-242. doi: 10.1016/j.ecolmodel.2006.12. 012
    [21] Obuchowski N A, 2003. Receiver operating characteristic curves and their use in radiology. Radiology, 229(1): 3-8. doi:  10.1148/radiol.2291010898
    [22] Ohmann J L, Gregory M J, 2002. Predictive mapping of forest composition and structure with direct gradient analysis and nearest neighbor imputation in coastal Oregon, USA. Canadian Journal of Forest Research, 32(4): 725-741. doi:  10.1139/x02-011
    [23] Peleg K, Anderson G L, 2002. FFT regression and cross-noise reduction for comparing images in remote sensing. Interna-tional Journal of Remote sensing, 23(10): 2097-2124. doi:  10.1080/01431160110075910
    [24] Pfeffer K, Pebesma E J, Burrough P A, 2003. Mapping alpine vegetation using vegetation observation and topographic attributes. Landscape Ecology, 18(8): 759-776. doi: 10.1023/B: LAND.0000014471.78787.d0
    [25] Sanders M E, Dirkse G M, Slim P A, 2004. Objecting thematic, spatial and temporal aspects of vegetation mapping for moni-toring. Community Ecology, 5(1): 81-91. doi: 10.1556/ComEc. 5.2004.1.8
    [26] Schmidtlein S, Zimmermann P, Schupferling R et al., 2007. Mapping the floristic continuum: Ordination space position estimated from imaging spectroscopy. Journal of Vegetation Science, 18(1): 131-140. doi: 10.1111/j.1654-1103.2007. tb02523.x
    [27] Shen Zehao, Zhao Jun, 2007. Prediction of the spatial patterns of species richness based on the plant topography relationship: An application of GAMs approach. Acta Ecologica Sinica, 27(3): 953-962. (in Chinese)
    [28] Song C H, Woodcock C E, Seto K et al., 2000. Classification and change detection using Landsat TM data: When and how to correct atmospheric effects? Remote Sensing of Environment, 75(2): 230-244. doi:  10.1016/S0034-4257(00)00169-3
    [29] Song C Y, Liu G H, Liu Q S, 2009. Spatial and environmental effects on plant communities in the Yellow River Delta, Eastern China. Journal of Forest Research, 20(2): 117-122. doi:  10.1007/s11676-009-0021-3
    [30] Song Yongchang, 2001. Vegetation Ecology. Shanghai: East China Normal University Press. (in Chinese)
    [31] Treitz P M, Howarth P J, Suffling R C, 1992. Application of de-tailed ground information to vegetation mapping with high resolution digital imagery. Remote Sensing of Environment, 42(1): 65-82. doi:  10.1016/0034-4257(92)90068-U
    [32] Weber K T, 2006. Challenges of integrating geospatial technolo?gies into rangeland research and management. Rangeland Ecology and Management, 59(1): 38-43. doi:  10.2111/05-010R.1
    [33] Wehrden H V, Zimmermann H, Hanspach J et al., 2009. Predictive mapping of plant species and communities using GIS and Landsat data in a southern Mongolian mountain range. Folia Geobotanica, 44(3): 211-225. doi:  10.1007/s12224-009-9042-0
    [34] Wen Zhongming, Jiao Feng, Jiao Juying, 2008. Prediction and mapping of potential vegetation distribution in Yanhe River catchment in hilly area of Loess Plateau. Chinese Journal of Applied Ecology, 19(9): 1897-1904. (in Chinese)
    [35] White P S, Wilds S P, Stratton DA, 2001. The distribution of heath balds in the Great Smoky Mountains, North Carolina and Tennessee. Journal of Vegetation Science, 12(4): 453-466. doi:  10.2307/3236997
    [36] Yang Cunjian, Zhou Chenghu, 2001. Investigation on classifica-tion of remote sensing image on basis of knowledge. Geogra-phy and Territorial Research, 17(1): 72-77. (in Chinese)
    [37] Yue T X, Liu J Y, Jorgensen S E et al., 2003. Landscape change detection of the newly created wetland in the Yellow River Delta. Ecological Modeling, 164(1): 21-31. doi:  10.1016/S0304-3800(02)00391-5
    [38] Zang Qiyun, 1996. Near Shore Sediment along the Yellow River Delta. Beijing: Ocean Press. (in Chinese)
    [39] Zhang Jintun, 1995. Quantitative Method of Vegetation Ecology. Beijing: China Science and Technology Press. (in Chinese)
    [40] Zhang R Q, Zhu D L, 2011. Study of land cover classification based on knowledge rules using high-resolution remote sensing images. Expert System with Application, 38(4): 3647-3652. doi:  10.1016/j.eswa.2010.09.019
    [41] Zhao Yingshi, 2003. The Principle and Method of Analysis of Remote Sensing Application. Beijing: Science Press. (in Chin-ese)
  • 加载中
通讯作者: 陈斌, bchen63@163.com
  • 1. 

    沈阳化工大学材料科学与工程学院 沈阳 110142

  1. 本站搜索
  2. 百度学术搜索
  3. 万方数据库搜索
  4. CNKI搜索

Article Metrics

Article views(707) PDF downloads(1103) Cited by()

Proportional views
Related

Predictive Vegetation Mapping Approach Based on Spectral Data, DEM and Generalized Additive Models

doi: 10.1007/s11769-013-0590-0
Funds:  Under the auspices of National Natural Science Foundation of China (No. 41001363)
    Corresponding author: SONG Chuangye. E-mail: songcy@ibcas.ac.cn

Abstract: This study aims to provide a predictive vegetation mapping approach based on the spectral data, DEM and Generalized Additive Models (GAMs). GAMs were used as a prediction tool to describe the relationship between vegetation and environmental variables, as well as spectral variables. Based on the fitted GAMs model, probability map of species occurrence was generated and then vegetation type of each grid was defined according to the probability of species occurrence. Deviance analysis was employed to test the goodness of curve fitting and drop contribution calculation was used to evaluate the contribution of each predictor in the fitted GAMs models. Area under curve (AUC) of Receiver Operating Characteristic (ROC) curve was employed to assess the results maps of probability. The results showed that: 1) AUC values of the fitted GAMs models are very high which proves that integrating spectral data and environmental variables based on the GAMs is a feasible way to map the vegetation. 2) Prediction accuracy varies with plant community, and community with dense cover is better predicted than sparse plant community. 3) Both spectral variables and environmental variables play an important role in mapping the vegetation. However, the contribution of the same predictor in the GAMs models for different plant communities is different. 4) Insufficient resolution of spectral data, environmental data and confounding effects of land use and other variables which are not closely related to the environmental conditions are the major causes of imprecision.

SONG Chuangye, HUANG Chong, LIU Huiming. Predictive Vegetation Mapping Approach Based on Spectral Data, DEM and Generalized Additive Models[J]. Chinese Geographical Science, 2013, 23(3): 331-343. doi: 10.1007/s11769-013-0590-0
Citation: SONG Chuangye, HUANG Chong, LIU Huiming. Predictive Vegetation Mapping Approach Based on Spectral Data, DEM and Generalized Additive Models[J]. Chinese Geographical Science, 2013, 23(3): 331-343. doi: 10.1007/s11769-013-0590-0
Reference (41)

Catalog

    /

    DownLoad:  Full-Size Img  PowerPoint
    Return
    Return