Mapping tropical forest fractional cover from coarse spatial resolution remote sensing imagery

At regional to global scales the only feasible approach to mapping and monitoring forests is through the use of coarse spatial resolution remotely sensed imagery. Significant errors in mapping may arise as such imagery may be dominated by pixels of mixed land cover composition which cannot be accommodated by conventional mapping approaches. This may lead to incorrect assessments of forest extent and thereby processes such as deforestation which may propagate into studies of environmental change. A method to unmix the class composition of image pixels is presented and used to map tropical forest cover in part of the Mato Grosso, Brazil. This method is based on an artificial neural network and has advantages over other techniques used in remote sensing. Fraction images depicting the proportional class coverage in each pixel were produced and shown to correspond closely to the actual land cover. The predicted and actual forest cover were, for instance, strongly correlated (up to r = 0.85, significant at the 99% level of confidence) and the predicted extent of forest over the test site much closer to the actual extent than that derived from a conventional approach to mapping from remotely sensed imagery.

[1]  Peter D. Moore,et al.  Biogeography: An Ecological and Evolutionary Approach , 1974 .

[2]  Janet Franklin.  Discrimination of tropical vegetation types using SPOT multispectral data , 1993 .

[3]  J. Downing,et al.  Workshop Summary Statement: Terrestrial Bioshperic Carbon Fluxes Quantification of Sinks and Sources of CO2 , 1993 .

[4]  Brent N. Holben,et al.  Fraction images derived from NOAA AVHRR data for studying the deforestation in the Brazilian Amazon , 1994 .

[5]  T. Richards,et al.  NOAA-11 AVHRR/2— Thermal channel calibration update , 1995 .

[6]  R. Nelson.  Regression and ratio estimators to integrate AVHRR and MSS data , 1989 .

[7]  H. Kerdiles,et al.  NOAA-AVHRR NDVI decomposition and subpixel classification using linear mixing in the Argentinean Pampa , 1995 .

[8]  Igor Aleksander,et al.  Introduction to Neural Computing , 1990 .

[9]  D. Skole.  Data on global land-cover change: acquisition, assessment, and analysis , 1994 .

[10]  G. Green,et al.  Deforestation History of the Eastern Rain Forests of Madagascar from Satellite Images , 1990, Science.

[11]  Michael F. Goodchild,et al.  Integrating GIS and remote sensing for vegetation analysis and modeling: methodological issues , 1994 .

[12]  Geoffrey E. Hinton,et al.  Learning internal representations by error propagation , 1986 .

[13]  The Carbon Cycle and Global Forest Ecosystem , 1993 .

[14]  Giles M. Foody,et al.  Estimation of tropical forest extent and regenerative stage using remotely sensed data , 1994 .

[15]  Raymond L. Czaplewski,et al.  Calibration of Remotely Sensed Proportion or Area Estimates for Misclassification Error , 1992 .

[16]  David J. Martin,et al.  Unmixing Aggregate Data: Estimating the Social Composition of Enumeration Districts , 1998, Environment & planning A.

[17]  Michael A. Spanner,et al.  Unmixing AVHRR imagery to assess clearcuts and forest regrowth in Oregon , 1995, IEEE Trans. Geosci. Remote. Sens..

[18]  J. Settle,et al.  Linear mixing and the estimation of ground cover proportions , 1993 .

[19]  Fabio Maselli,et al.  Use of error matrices to improve area estimates with maximum likelihood classification procedures , 1992 .

[20]  A. Fischer.  A model for the seasonal variations of vegetation indices in coarse resolution data and its inversion to extract crop parameters , 1994 .

[21]  Eduardo Bayro-Corrochano,et al.  Self-organizing neural-network-based pattern clustering method with fuzzy outputs , 1994, Pattern Recognit..

[22]  C. Tucker,et al.  North American vegetation patterns observed with the NOAA-7 advanced very high resolution radiometer , 1985, Vegetatio.

[23]  C. Tucker,et al.  Tropical Deforestation and Habitat Fragmentation in the Amazon: Satellite Data from 1978 to 1988 , 1993, Science.

[24]  Alan R. Gillespie,et al.  Vegetation in deserts. I - A regional measure of abundance from multispectral images. II - Environmental influences on regional abundance , 1990 .

[25]  N. A. Quarmby,et al.  Linear mixture modelling applied to AVHRR data for crop area estimation , 1992 .

[26]  Yoram J. Kaufman,et al.  Remote sensing of biomass burning in the tropics , 1990 .

[27]  B. Turner,et al.  Changes in land use and land cover: a global perspective , 1995 .

[28]  S. Gerstl,et al.  Nonlinear spectral mixing models for vegetative and soil surfaces , 1994 .

[29]  Robert G. Bryant,et al.  Validated linear mixture modelling of Landsat TM data for mapping evaporite minerals on a playa surface: methods and applications , 1996 .

[30]  B. Holben,et al.  Linear mixing model applied to coarse spatial resolution data from multispectral satellite sensors , 1993 .

[31]  P. Curran,et al.  Environmental Remote Sensing From Regional to Global Scales , 1995 .

[32]  Christopher Justice,et al.  The impact of misregistration on change detection , 1992, IEEE Trans. Geosci. Remote. Sens..

[33]  Roger A. Sedjo,et al.  The carbon cycle and global forest ecosystem , 1993 .

[34]  Fangju Wang,et al.  Fuzzy supervised classification of remote sensing images , 1990 .

[35]  Giles M. Foody,et al.  Fully fuzzy supervised classification of land cover from remotely sensed imagery with an artificial neural network , 1997, Neural Computing & Applications.

[36]  J. Campbell.  Introduction to remote sensing , 1987 .

[37]  W. B. Yates,et al.  Classification of remotely sensed data by an artificial neural network: issues related to training data characteristics , 1995 .

[38]  J. Estes,et al.  Applications of NOAA-AVHRR 1 km data for environmental monitoring , 1994 .

[39]  James L. McClelland,et al.  Parallel distributed processing: explorations in the microstructure of cognition, vol. 1: foundations , 1986 .

[40]  Jacques de Villiers,et al.  Backpropagation neural nets with one and two hidden layers , 1993, IEEE Trans. Neural Networks.

[41]  Eric F. Lambin,et al.  Estimation of tropical forest area from coarse spatial resolution data: A two-step correction function for proportional errors due to spatial aggregation , 1995 .

[42]  C. Justice,et al.  Global land cover classification by remote sensing: present capabilities and future possibilities , 1991 .

[43]  C. Tucker,et al.  AVHRR for Monitoring Global Tropical Deforestation , 1989 .

[44]  Paul E. Johnson,et al.  Quantitative determination of mineral types and abundances from reflectance spectra using principal components analysis , 1985 .

[45]  Russell G. Congalton,et al.  A review of assessing the accuracy of classifications of remotely sensed data , 1991 .

[46]  John E. Colwell,et al.  Coarse-resolution Satellite Data for Ecological SurveysNOAA satellite data offer ecologists new opportunities for examining large areas , 1986 .

[47]  Yves Chauvin,et al.  Generalization Performance of Overtrained Back-Propagation Networks , 1990, EURASIP Workshop.

[48]  J. Settle,et al.  Subpixel measurement of tropical forest cover using AVHRR data , 1991 .

[49]  P. Fearnside.  An ecological analysis of predominant land uses in the Brazilian Amazon , 1988 .

[50]  Estimating sub-pixel components of a semi-arid woodland , 1994 .

[51]  P. Swain,et al.  Neural Network Approaches Versus Statistical Methods In Classification Of Multisource Remote Sensing Data , 1990 .

[52]  J. Dymond.  How accurately do image classifiers estimate area , 1992 .

[53]  Minhua Wang,et al.  Modeling errors in remote sensing image classification , 1993 .

[54]  G. Foody,et al.  Sub-pixel land cover composition estimation using a linear mixture model and fuzzy membership functions , 1994 .

[55]  Robert J. Schalkoff,et al.  Pattern recognition - statistical, structural and neural approaches , 1991 .