Environmental Engineering Reference
In-Depth Information
strategically important because it is coarse enough for a global
coverage, yet fine enough to capture human-dimension processes
such as deforestation and urban growth (Paulsson, 1992). Over a
period of nearly four decades, the Landsat program has acquired a
scientifically valuable image archive unmatched in quality, details,
coverage, and length, which has become freely available since 9
January 2009. This visually stunning data set has supported a
wide variety of applications such as natural resource assessment
and management, urban and regional planning, surveying and
mapping, global change studies, among others.
As the first generation space-borne multispectral imaging
system, the Landsat MSS sensor began to acquire data in 1972,
with a spatial resolution approaching that of medium-scale aerial
photographs. The MSS system acquires images over 4 spectral
bands with the first two (green and red bands) suitable for
detecting cultural features such as urban areas, roads, new sub-
divisions, gravel pits, and quarries. The image size is 185
to monitor the growth of Greater Athens, Greece by using the
spectral mixing technique. Masek, Lindsay and Goward (2000)
used three TM images, together with one MSS scene, to detect
urban growth in the Washington DC area and found that the
urban physical growth can be reasonably correlated with regional
and national economic patterns. Seto and Fragkias (2005) used
two TM images to quantify spatio-temporal patterns of urban
land-use changes in four Chinese cities through the landscape
metrics technique.
Landsat-7 was successfully launched on 15 April 1999, with
the ETM
+
sensor onboard. This latest Landsat multispectral
sensor includes two important updates when comparing to
the TM sensor: a new panchromatic band with 15 m spatial
resolution and an improved thermal band with 60 m resolution.
These improvements allow more details in urban land use/cover
changes to be detected. On the other hand, there was a significant
change in Landsat-7 ETM
185
km that allows some large natural or human-dimensional pat-
terns or processes to be examined within a single scene. In
addition, the MSS data set has been well archived and main-
tained at the USGS EROS Data Center and more than one
dozen of international Landsat ground stations (Draeger
et al
.,
1997). It is the only digital data set from operational satellites
available for the period of 1972 to early 1982. The Landsat data
were made available at an affordable price before the entire
data set is open at no charge since early 2009. Given the above
considerations, the MSS data have the unique values and thus
have been widely used in connection with urban and land-
scape change analysis. Some examples include: mapping urban
change in Atlanta (Todd, 1977), Denver and Richmond (Toll
et al
., 1980); monitoring anthropogenic land use as well as
albedo changes in the Montreal, Ottawa, and Quebec regions
(Royer, Charbonneau and Bonn, 1988); and mapping urban
spatial extent in the Bombay metropolitan region (Pathan
et al
.,
1993).
The TM data have become available since 1982 when Landsat-
4 was successfully launched, prompting a surge of interests
to evaluate the utilities of this type of data for urban and
landscape change analysis (Jensen
et al
., 1989). Comparing to
the MSS, the TM sensor acquires data over seven carefully
designed bands, including new bands in the visible (blue), mid-
infrared, and thermal portions of the spectrum, which have
helped improve the spectral differentiability of major land surface
features, particularly vegetation whose proportion is a criterion
for the discrimination of different types of urban land use such
as commercial, industrial, or residential use. With the improved
spatial, spectral, and radiometric resolutions, the Landsat-5 TM
sensor has become the Landsat workhorse imager, providing
invaluable data for urban land use/cover mapping with Anderson
Level I&II classification (Anderson
et al
., 1976) and for detecting
finer urban changes with an improved accuracy with some
advanced algorithms or techniques (see the extended discussions
in Part III of this topic). For example, Gomarasca
et al
. (1993)
used two TM images in conjunction with other digitized data
to assess one century of land use modifications driven by rapid
agricultural transformation and urban development in the Milan
metropolitan area, Italy. Green, Kempka and Lackey (1994) used
two TM images to detect land cover/use change in the Portland
metropolitan area through the image differencing technique.
Hill and Hostert (1996) employed a multitemporal TM data set
×
data delivery and distribution policies
according to the 1992 Land Remote Sensing Policy Act. With a free
license, the initial US price for a single ETM
+
scene was at $600
from the USGS EROS Data Center, compared to $4400 for one
TM scene ordered from the Earth Observation Satellite Company
(EOSAT), a commercial company that was contracted to run the
Landsat program during 1985 - 2001 according to the 1984 Land
Remote Sensing Commercialization Act. This substantially low
price and free licensing allowed the ETM
+
data to be widely used
for urban and landscape change analysis (e.g., Yang, 2002; Lo and
Choi, 2004; Doygun and Alphan, 2006; Millward, Piwowar and
Howarth, 2006). Unfortunately, the Scan Line Corrector (SLC)
in the ETM
+
instrument, which compensates for the along-track
forward motion of the spacecraft, failed permanently on 31 May
2003. This failure forces the ETM
+
instrument to image the
Earth in a zigzag fashion, resulting in some areas that are imaged
twice and approximately one-fourth of an ETM
+
scene that is
not imaged at all. Although the Landsat-7 data acquired after
this failure are available with the data gaps optionally filled in by
using other data, recent studies indicate that the SLC failure has
underminedtheroleofETM
+
data as a complete land use/cover
inventory dataset (Trigg, Curran and McDonald, 2006; Wulder
et al
., 2008).
The technical problems with the Landsat-7 instrument, the
operation of Landsat-5 substantially beyond its designed life,
and delays in the development and launch of a successor have
increased the likelihood that a gap in the Landsat data archive
may occur. And a Landsat Data Gap Study team formed by
USGS and NASA in 2005 found that there are no other sys-
tems in orbit or planned for launch in the short term that
can supply data to fill the Landsat gap should system failures
occur to Landsat-5 and -7 (Wulder
et al
., 2008). Considering the
scientific, environmental, economic, and social benefits offered
by the Landsat program, the Bush Administration decided in
December 2005 to continue Landsat instruments in the form
of a free-flyer spacecraft, which has resulted in a NASA and
USGS joint initiative called the Landsat Data Continuity Mission
(LDCM). Under this initiative, a new satellite named LDCM-
1 with the Operational Land Imager (OLI) and the Thermal
Infrared Sensor (TIRS) onboard is scheduled to launch in
December 2012. This plan, once implemented, would ensure
the continuity of the Landsat data acquisition that has become
unique and indispensable for monitoring, management, and
scientific activities.
+
Search WWH ::
Custom Search