IEEE Geoscience and Remote Sensing Magazine - September 2019 - 113

of the savanna communities in Emas
National Park [20].

DAM

User

EUCALYPTUS SITES
BE
TSR
To exemplify the use of the GPC
GPC
module, we used as input several
ROI
Requests Download Module
Definition
points sampled from the region deData
Map Tab
limited by the black-colored outline
in Figure 7. The region is distributed
throughout the state of São Paulo,
Brazil. Those points were previously
Data Exploration
sampled [21] and labeled [22] either
Provides Time
and Time-Series
Series Data
as eucalyptus (red color) or noneucalypChange Detection
Analysis Tab
tus (green color).
We selected a single region/point
in the Amazon Basin (see Figure 8) to FIGURE 5. A functional overview of the BE module.
illustrate the results obtained from a
structural-break-detection technique through the BE module. To choose that specific point, we selected the region
most exposed to an extreme drought event [23] quantified
by the Standardized Precipitation Index (SPI) [24].
The Amazon Basin is the largest drainage basin in the
world, covering more than six million km2 across nine different countries: Bolivia, Brazil, Colombia, Ecuador, French
Guyana (an overseas French territory), Guyana, Peru, Suriname, and Venezuela [25]. The prevailing vegetation type
in this region is dense, moist tropical forests. Other types
of vegetation, such as savannas, grasslands, swamps,
and floodplain forests, are also present, although less
frequently [40]. Rainfall in the basin area ranges from
1,500 to 3,000 mm/year, with averages of 2,000 mm/year
in the central region [26].
The Amazon Basin has already been exposed to
several extreme climatic events, such as the El Niño
FIGURE 6. Google Earth satellite imagery and a delimitation of the
of 1982/1983 [27], [28]; the El Niño of 1991/1992
ROI (outlined in yellow) inside Emas National Park.
[28]; the El Niño of 1997/1998 [28], [29]; the Atlantic drought of 2010 [30]; and the El Niño of 2015/2016
[31], which may have directly impacted the vegetation
Eucalyptus
behavior [23], [32].

Noneucalyptus

DATA
EMAS NATIONAL PARK
Using the DAM module, we acquired a total of 346 2-band
Enhanced Vegetation Index (EVI2) images from the Landsat
5 Surface Reflectance Tier 1 data [33], ranging from April
1984 to November 2011.
EUCALYPTUS SITES
Experiments were carried out on the data collected by the
Moderate Resolution Imaging Spectroradiometer (MODIS) sensor (Terra and Aqua satellites). The working data
set consists of 715 normalized difference vegetation index
(NDVI) images, ranging from February 2000 to November 2016. Preprocessing procedures, such as atmospheric
corrections, cloud detection, and georeferencing, have
SEPTEMBER 2019

IEEE GEOSCIENCE AND REMOTE SENSING MAGAZINE

B
A

FIGURE 7. The spatial distribution of eucalyptus (totaling 250

points, in red) and noneucalyptus (totaling 1,000 points, in green)
samples. Point A shows details specific to eucalyptus samples, and
point B shows details specific to noneucalyptus samples.
113
IEEE Geoscience and Remote Sensing Magazine - September 2019

Table of Contents for the Digital Edition of IEEE Geoscience and Remote Sensing Magazine - September 2019
