scispace - formally typeset
SciSpace - Your AI assistant to discover and understand research papers | Product Hunt

Proceedings ArticleDOI

Mapping land cover types in Amazon basin using 1 km JERS-1 mosaic

28 Jun 1999-Vol. 2, pp 934-936

AbstractThe 100 meter JERS-1 Amazon mosaic image was used in a new classifier to generate a 1 km resolution land cover map. The inputs to the classifier were 1 km resolution mean backscatter and seven first order texture measures derived from the 100 m data by using a 10/spl times/10 independent sampling window. The classification approach included two interdependent stages: 1) a supervised maximum a posteriori Baysian approach to classify the mean backscatter image into 5 general land cover categories of forest, savanna, inundated, white sand, and anthropogenic vegetation classes, and 2) a texture measure decision rule approach to further discriminate subcategory classes based on taxonomic information and biomass levels. Fourteen classes were successfully separated at 1 km scale. The results were verified by examining the accuracy of the approach by comparison with the IBGE and the AVHRR 1 km resolution land cover maps.

Topics: Land cover (58%)

Summary (1 min read)

Jump to:  and [Summary]

Summary

  • The 100 meter JERS-I Amazon mosaic image was used in a new classifier to generate a 1 km resolution land cover map.
  • The inputs to the classifier were lkm resolution mean backscatter and seven first order texture measures derived from the 100 m data by using a 10 x 10 independent sampling window.
  • The classification approach included two interdependent stages: 1) a supervised maximum a posteriori Baysian approach to classify the mean backscatter image into 5 general land cover catagories of forest, savanna, inundated, white sand, and anthropogenic vegetation classes, and 2) a texture measure decision rule approach to further discriminate subcatagory classes based on taxonimc information and biomass levels.
  • Fourteen classes were successfully separated at lkm scale.
  • The results were verified by examining the accuracy of the approach by comparison with the IBGE and the AVHRR 1 km resolution land.

Did you find this useful? Give us your feedback

...read more

Content maybe subject to copyright    Report

Mapping Land Cover Types in Amazon Basin Using l km
JERS- 1 Mosaic
Sassan S. Saatchi t, Bruce Nelson z, Erika Podest I, John Holt _
1. Jet Propulsion Laboratory, 4800 Oak Grove Drive, Pasadena, California 91109
2. Instituto Nacional de Pesquisas de Amazonia, CP 478, Ecologia, 69011-970, Manaus,
Amazonia, Brazil
Abstract
In this paper, the 100 meter JERS-I Amazon
mosaic image was used in a new classifier to
generate a 1 km resolution land cover map. The
inputs to the classifier were lkm resolution mean
backscatter and seven first order texture measures
derived from the 100 m data by using a 10 x 10
independent sampling window. The classification
approach included two interdependent stages: 1) a
supervised maximum a posteriori Baysian
approach to classify the mean backscatter image
into 5 general land cover catagories of forest,
savanna, inundated, white sand, and
anthropogenic vegetation classes, and 2) a texture
measure decision rule approach to further
discriminate subcatagory classes based on
taxonimc information and biomass levels.
Fourteen classes were successfully separated at
lkm scale. The results were verified by examining
the accuracy of the approach by comparison with
the IBGE and the AVHRR 1 km resolution land
cover maps.
Introduction
Understanding the human or climate induced
changes of a tropical landscape requires
knowledge of the current status of the ecosystem,
the extent of the land cover types susceptible to
change, and the causes and impacts of such
changes. Recent advances in remote sensing
technologies have partially contributed in
documenting and monitoring these land use
changes. There are, however, several unresolved
problems associated with mapping land cover
types and monitoring the tropics on regional and
continental scales. These derive from limitations
of current remote sensing techniques and the
methodologies used in both defining the land
cover types and identifying the parameters to be
monitored.
Optical remote sensing has been successfully used
for classification of land cover types and the study
of their changes on local to regional scales. High
resolution (30 m) Landsat Thematic Mapper (TM)
data has been the primary source for estimating
the rate of deforestation by INPE (Instituto
Nacional de Pesquisas Espaciais) and the Landsat
Pathfinder Program (Skole and Tucker, 1993;
Justice and Townshend, 1994). These studies
have used visual interpretation and classification
as their primary approach for extracting thematic
information. Most large scale maps derived from-
these show few land cover types. This is due in
part to difficulties in interpreting the spectral
information of Landsat data acquired at different
years or seasons.
During the past decade, several radar sensors have
been deployed in space such as the shuttle
imaging radar (SIR-A, SIR-B, and SIR-C/X-
SAR), ERS-1,2, JERS-1, and Radarsat. Except
the SIR-C/X-SAR system, all radar sensors have
only one channel. Though none were designed
specifically for land cover mapping, several
investigations have demonstrated that the data
provide unique information about the
characteristics of tropical landscapes (Sader,
1987; Foody and Curran, 1994). First, the radar
data can be acquired as frequently as possible due
to insensitivity to atmospheric condition and sun
angle. This allows continental scale high
resolution data for systematic assessment of
deforestation and regrowth processes. Second,
depending on the wavelength, the radar
backscatter signal carries information about forest
structure and moisture condition by penetrating
into the forest canopy. Few studies have
addressed these characteristics by using the radar
data for mapping tropical land cover and
estimating the biomass of regenerating forests
(Foody and Curran 1994; Luckman et al., 1997;
Saatchi et al., 1997; Rignot et al., 1997).
During the Global Rain Forest Mapping (GRFM)
project, JERS-I SAR (Synthetic Aperture Radar)
satellite was used to map the humid tropical

forestsof theworld. The rationalefor the project
was to demonstrate the application of the
spaceborneL-bandradarin tropicalforeststudies.
In particular,the use of data for mapping land
cover types, estimatingthe areaof floodplains,
and monitoring deforestation and forest
regenerationwereof primaryimportance. In this
paper,we examinetheinformationcontentof the
JERS-1SARdatafor mappinglandcovertypesin
theAmazonbasin.
JERS-1 Amazon Mosaic
JERS-I SAR is an L-band spacebome SAR
system launched by the National Space
Development Agency of Japan fNASDA) in
February, 1992. The system operates at 1.275
GHz with horizontal polarization for both
transmission and reception. The spatial
resolution of the system is 18 m in both azimuth
and range. The swath width is 75 km and the
incidence angle of radar at the center of swath is
38.5 °. The single-look images have 4.2 m pixel
spacing in azimuth and 12.5 m in range and the
standard three look image has 12.5 m pixel
spacing in both azimuth and range.
In late 1995, the JERS-1 satellite entered into its
Global Rain Forest Mapping (GRFM) phase to
collect high resolution SAR data over the entire
tropical rainforest. In approximately 60 days, the
satellite obtained wall-to-wall data over the
Amazon basin. Because of cloud cover, similar
coverage with high resolution optical data such as
Landsat could only be provided on annual or
decadal time frames. The JERS-1 coverage of the
Amazon basin is shown in Figure 1.
We have used I00 m resolution JERS-1 data (8
by 8 averaging of high resolution 12.5 m three-
look data) to generate a map of the entire basin
from 1500 images. The spatial mosaicking
technique is based on a mathematical wallpapering
approach which minimizes the propagation of
errors. The inter-scene overlaps both in the
along-track and cross-track directions were used
for individual scene geoiocation. The scenes were
placed on a global coordinate system with the
flexibility of having scenes float freely with
respect to one another. The locations of all scenes
were calculated simultaneously, avoiding any
directional errors. The result is an optimum
seamless mosaic (Figure 2)
Figure 1.
basin.
¢)..... ,
..[ '_" [zr,,._:i _ i
,l' '¢' /'
, / j'
Ifl_
/ i _-;'-J
/ _ ,_. j
JERS-I coverage map of the Amazon
Figure 2. JERS-1 Amazon mosaic image during
the dry season.
Methodology
In classifying the JERS-I SAR data, we
developed texture measures from the 100 m
JERS-I mosaic over a 10 x l0 window, resulting
in 1 km resolution texture images with
independent pixel intormation, as the windows
were shifted in a blockwise fashion in a 10-pixel
increments. These first order histogram statistics
characterize the frequency of occurrence of grey
levels within the window in the single channel
radar data and are sensitive to window size which
_as determined a priori. The JERS-I mosaic
image in figure 2 shows orbital stripes which are
due to slight radiometric discrepancy between the
calibration of orbital data takes. These calibration
discrepancies which are often less than 0.5 dB in

intensitydo not affect the texture measures and the
classification of the image at a resampled
resolution of 1 km.
Eight texture measures were calculated for
classification from the first order histograms
(mean, variance, energy, entropy, contrast,
kurtosis, skewness, coeffiecient of variation).
Training areas were chosen from the mosaic
image by consulting the RADAMBRASIL
vegetation map of Amazonia. The class
separability using all texture measures were
calculated by using the B-distance (Bhattacharyya
distance). The separability test helped us to
identify the significance of texture measures in
classifying each land cover types.
After choosing a list of texture measures,
we employ a two stage approach to perform a
supervised classification of the JERS-1 mosaic.
In the first stage, we use a maximum a posteriori
Baysian (MAP) classifier on the JERS-I mean
backscatter image at 1 km scale to select five
general categories of land cover types (forest,
nonforest, savanna, floodplain, open water)
In the second stage, the texture measures and the
MAP classified image were used in a hierarchical
decision rule based algorithm to further
discriminate more specific vegetation types within
each of five general land cover categories. The
decision rules were derived by using predictor
variables obtained from the multi-dimensional
separability analysis of the backscatter and texture
measures for each vegetation type. The rules for
texture measures are calculated irr a hierarchical
fashion using the normal distribution for the
statistics of training areas. This assumption was
verified when using the B-distance separability
test. Once the training data set was obtained and
the MAP classified image was produced, the
decision rules were determined in an automated
procedure in order to guarantee the repeatability of
the classifier. Note that the input texture images
to the classifier and the training data sets are
derived from the 100 m JERS-I using a 10 x 10
window. The use of 100 m backscatter image
rather than the 1 km images for collecting the
training data, helps avoiding mixed pixel
information in training data set.
Discussion
A total of twenty sites were chosen for each
vegetation type, of which ten were used for
training the classifier and ten for verifying the
accuracy of classification. To make the selection
of training and validation sites as accurate as
possible, we registered the RADAM map with the
JERS-I mosaic and extracted the backscatter and
texture measures for each land cover category as
indicated on the map. During the selection of
these, we avoided those cases where the RADAM
vegetation types did not match with the general
characteristics of the radar image suggesting a
possible error in the RADAM data.
A total of 14 classes were separated JERS-I
mosaic. The accuracy of classification were
calculated by developing a confusion matrix of
classes, comparison with the RADAM map, and
pixel-by-pixel comparison with the AVHRR land
cover map. The results and detailed discussion
are reported in Saatchi et a1.(1999).
Acknowledgment
This work is performed at the Jet Propulsion
Laboratory, California Institute of Technology,
under contract from National Aeronautic and
Space Administration.
References
Foody, G.M., and Curran, P.J., 1994, J. of
Biogeograhy. 21:223-244.
Luckrnan, A., Baker, J., Kuplich, T.M.,
Yanasse, C., and Frery, A.C., 1997, Remote
Sens. Environ., 60:1-13.
Rignot, E., Salas, W., and Skole, D.L., 1997,
Remote Sens. Environ., 59:167-179.
Saatchi, S.S., Soares, J.V., and Alves, D.S.,
1997, Remote Sens. Environ. , vol. 59: 191-202.
Saatchi, S., Nelson, B., Podest, E., Holt, J.,
1999, Int. J. Remote Sens., In press.
Sader, S.A., 1987, Phtogramm. Eng. Remote
Sens. 53(2): 193-202.
Skole, D.L. and Tucker, C.J., 1993, Science,
260:1905-1910.
Veloso, H.P., Rangel Filho, A.L.R.& Lima,
J.C.A., 1991, IBGE. Rio de Janeiro. 123 pp.
Citations
More filters

Journal ArticleDOI
Abstract: The amount and spatial distribution of forest biomass in the Amazon basin is a major source of uncertainty in estimating the flux of carbon released from land-cover and land-use change. Direct measurements of aboveground live biomass (AGLB) are limited to small areas of forest inventory plots and site-specific allometric equations that cannot be readily generalized for the entire basin. Furthermore, there is no spaceborne remote sensing instrument that can measure tropical forest biomass directly. To determine the spatial distribution of forest biomass of the Amazon basin, we report a method based on remote sensing metrics representing various forest structural parameters and environmental variables, and more than 500 plot measurements of forest biomass distributed over the basin. A decision tree approach was used to develop the spatial distribution of AGLB for seven distinct biomass classes of lowland old-growth forests with more than 80% accuracy. AGLB for other vegetation types, such as the woody and herbaceous savanna and secondary forests, was directly estimated with a regression based on satellite data. Results show that AGLB is highest in Central Amazonia and in regions to the east and north, including the Guyanas. Biomass is generally above 300Mgha(sup 1) here except in areas of intense logging or open floodplains. In Western Amazonia, from the lowlands of Peru, Ecuador, and Colombia to the Andean mountains, biomass ranges from 150 to 300Mgha(sup 1). Most transitional and seasonal forests at the southern and northwestern edges of the basin have biomass ranging from 100 to 200Mgha(sup 1). The AGLB distribution has a significant correlation with the length of the dry season. We estimate that the total carbon in forest biomass of the Amazon basin, including the dead and below ground biomass, is 86 PgC with +/- 20% uncertainty.

620 citations


Cites background or methods from "Mapping land cover types in Amazon ..."

  • ...Backscatter and texture at L-band (1.25 GHz) from this instrument are sensitive to forest structure and biomass at low densities of tree cover, such as open forests and woodland savannas (Saatchi et al., 1997, 2000; Luckman et al., 1998; Podest & Saatchi, 2002)....

    [...]

  • ...DTM has been successfully applied to remote sensing data in the past because of its simplicity, efficiency, and robustness (Hansen et al., 2000; Saatchi et al., 2000, 2005a, 1998; Simard et al., 2000, 2001)....

    [...]

  • ..., 2001), JERS-1 radar (Saatchi et al., 2000), and SRTM data (http://www2....

    [...]


Journal ArticleDOI
Abstract: Wetland extent was mapped for the central Amazon region, using mosaicked L-band synthetic aperture radar (SAR) imagery acquired by the Japanese Earth Resources Satellite-1. For the wetland portion of the 18×8° study area, dual-season radar mosaics were used to map inundation extent and vegetation under both low-water and high-water conditions at 100-m resolution, producing the first high-resolution wetlands map for the region. Thematic accuracy of the mapping was assessed using high-resolution digital videography acquired during two aerial surveys of the Brazilian Amazon. A polygon-based segmentation and clustering was used to delineate wetland extent with an accuracy of 95%. A pixel-based classifier was used to map wetland vegetation and flooding state based on backscattering coefficients of two-season class combinations. Producer's accuracy for flooded and nonflooded forest classes ranged from 78% to 91%, with lower accuracy (63–65%) for flooded herbaceous vegetation. Seventeen percent of the study quadrat was occupied by wetlands, which were 96% inundated at high water and 26% inundated at low water. Flooded forest constituted nearly 70% of the entire wetland area at high water, but there are large regional variations in the proportions of wetland habitats. The SAR-based mapping provides a basis for improved estimates of the contribution of wetlands to biogeochemical and hydrological processes in the Amazon basin, a key question in the Large-Scale Biosphere–Atmosphere Experiment in Amazonia.

509 citations


Journal ArticleDOI
Abstract: In this paper we demonstrate a new approach that uses regional/continental MODIS (MODerate Resolution Imaging Spectroradiometer) derived forest cover products to calibrate Landsat data for exhaustive high spatial resolution mapping of forest cover and clearing in the Congo River Basin. The approach employs multi-temporal Landsat acquisitions to account for cloud cover, a primary limiting factor in humid tropical forest mapping. A Basin-wide MODIS 250 m Vegetation Continuous Field (VCF) percent tree cover product is used as a regionally consistent reference data set to train Landsat imagery. The approach is automated and greatly shortens mapping time. Results for approximately one third of the Congo Basin are shown. Derived high spatial resolution forest change estimates indicate that less than 1% of the forests were cleared from 1990 to 2000. However, forest clearing is spatially pervasive and fragmented in the landscapes studied to date, with implications for sustaining the region's biodiversity. The forest cover and change data are being used by the Central African Regional Program for the Environment (CARPE) program to study deforestation and biodiversity loss in the Congo Basin forest zone. Data from this study are available at http://carpe.umd.edu.

413 citations


Cites background from "Mapping land cover types in Amazon ..."

  • ...The Landsat sensor, with its comparatively narrow field of view, is not as affected by surface anisotropic effects as wider field of view sensors, such as MODIS (Schaaf et al., 2002)....

    [...]


Journal ArticleDOI
TL;DR: The objective of this paper is to review and assess general medium spatial resolution satellite remote sensing land cover classification approaches with the goal of identifying the outstanding issues that must be overcome in order to implement a large-area, land cover classified protocol.
Abstract: Numerous large-area, multiple image-based, multiple sensor land cover mapping programs exist or have been proposed, often within the context of national forest monitoring, mapping and modelling initiatives, worldwide. Common methodological steps have been identified that include data acquisition and preprocessing, map legend development, classification approach, stratification, incorporation of ancillary data and accuracy assessment. In general, procedures used in any large-area land cover classification must be robust and repeatable; because of data acquisition parameters, it is likely that compilation of the maps based on the classification will occur with original image acquisitions of different seasonality and perhaps acquired in different years and by different sensors. This situation poses some new challenges beyond those encountered in large-area single image classifications. The objective of this paper is to review and assess general medium spatial resolution satellite remote sensing land cover cl...

296 citations


Cites background or methods from "Mapping land cover types in Amazon ..."

  • ...However, based on the large-area classification work in Congo and Amazon forests (De Grandi et al., 2000a,b; Saatchi et al., 2000), it seems likely that general land cover classes can be readily mapped (perhaps manually rather than digitally) with satellite SAR data....

    [...]

  • ...JERS-1 data were selected for this application partly because of ‘difficulties in interpreting the spectral information of Landsat data acquired at different years or seasons’ (Saatchi et al., 2000: 1202)....

    [...]


Journal ArticleDOI
Abstract: Summary 1 The floristic variation in Amazonian lowland forests is poorly understood, especially in the large areas of non-inundated (tierra firme) rain forest. Species composition may be either unpredictable as abundances fluctuate in a random walk, more-or-less uniform, or it may correspond to environmental heterogeneity. 2 We tested the three hypotheses by studying the floristic variation of two phylogenetically distant plant groups along a continuous 43-km long line transect that crossed tierra firme rain forest in northern Peru. 3 The observed floristic patterns were compared to patterns in the spectral reflectance characteristics of the forest as recorded in a Landsat TM satellite image. The topography of the transect was measured in the field, and surface soil samples were collected to document edaphic conditions. The two plant groups, pteridophytes and the Melastomataceae, were assessed in 2-m wide and 500-m long sampling units. 4 Floristic similarity (Jaccard index) between sampling units ranged from 0.01 to 0.71 (mean = 0.27), showing that some units were almost completely dissimilar while others were very alike. 5 Spatially constrained clustering produced very similar subdivisions of the transect when based separately on satellite image data, pteriophytes, and Melastomataceae, and the subdivisions were also related to topography and soil characteristics. Mantel tests showed that floristic similarity patterns of the two plant groups were highly correlated with each other and with similarities in reflectance patterns of the satellite image, and somewhat less correlated with geographical distance. 6 Our results lend no support to the uniformity hypothesis, but they partially support the random walk model, and are consistent with the hypothesis that species segregate edaphically at the landscape scale within the uniform-looking forest.

209 citations


References
More filters

Journal ArticleDOI
25 Jun 1993-Science
TL;DR: Although this rate of deforestation is lower than previous estimates, the effect on biological diversity is greater and tropical forest habitat, severely affected with respect to biological diversity, increased.
Abstract: Landsat satellite imagery covering the entire forested portion of the Brazilian Amazon Basin was used to measure, for 1978 and 1988, deforestation, fragmented forest, defined as areas less than 100 square kilometers surrounded by deforestation, and edge effects of 1 kilometer into forest from adjacent areas of deforestation. Tropical deforestation increased from 78,000 square kilometers in 1978 to 230,000 square kilometers in 1988 while tropical forest habitat, severely affected with respect to biological diversity, increased from 208,000 to 588,000 square kilometers. Although this rate of deforestation is lower than previous estimates, the effect on biological diversity is greater.

1,541 citations


"Mapping land cover types in Amazon ..." refers methods in this paper

  • ...High resolution (30 m) Landsat Thematic Mapper (TM) data has been the primary source for estimating the rate of deforestation by INPE (Instituto Nacional de Pesquisas Espaciais) and the Landsat Pathfinder Program (Skole and Tucker, 1993; Justice and Townshend, 1994)....

    [...]


Book
01 Jan 1952

1,390 citations


Book
01 Jan 1984
Abstract: The Amazon -that name was given to the biggest river on earth and is often used for the whole area of its basin too. This geographical region is currently referred to as Amazonia, thus emphasizing the peculiar character of its aquatic and terrestrial reaches. The Amazon embodied the dream of many a naturalist to explore what for a long time was a terra incognita. In recent years, however, Amazonia has emerged as a main centre for 'development' by some of the countries in which it lies and by foreign industrialized nations. The development projects and enterprises have aroused woridwide interest and have given rise to discussions on their aims and their consequences to the Amazonian nature. Limnological and ecological investigations in Amazonia started only about 40 years ago. The editor had the good fortune to partake in them from the very beginning. He spent his decisive years in Amazonia, and dedicated his life's work to that research and to that country and the Amazonian people. Nearing the end of his scicntific activities, hc is gratcful to bc ablc to summarizc in this book most of the knowledge we possess at present of Amazonian limnology and landscape ecology.

605 citations


Journal ArticleDOI
Abstract: The types of Amazonian forests subject to inundation can be organized into seven categories which are herewith named and described. This classification is intended to set in order the confusion of terminology used in the past. The types are: (1)seasonal varzea—forest flooded by regular annual cycles of white-water rivers; (2)seasonal igapo—forest flooded by regular annual cycles of black- and clear-water rivers; (3) mangrove—forests flooded twice daily by salt-water tides; (4)tidal varzea—forest flooded twice daily by fresh water backed up from tides; (5)floodplain forest—on low lying ground flooded by irregular rainfall, generally in upper reaches of rivers; (6)permanent white- water swamp forest; (7)permanent igapo—black-water forest. The first five types are periodically inundated and the last two are permanently waterlogged. This terminology is closer to that used by lim nologists by restricting the use ofigapo to forest inundated by black and clear water.

465 citations


Journal ArticleDOI
23 Mar 1984-Science
TL;DR: These values are considerably lower than those previously reported and raise questions about the role of the terrestrial biota in the global carbon budget.
Abstract: Recent assessments of areas of different tropical forest types and their corresponding stand volumes were used to calculate the biomass densities and total biomass of tropical forests. Total biomass was estimated at 205 x 10(9) tons, and weighted biomass densities for undisturbed closed and open broadleaf forests were 176 and 61 tons per hectare, respectively. These values are considerably lower than those previously reported and raise questions about the role of the terrestrial biota in the global carbon budget.

433 citations