International Archives of the
UNITED ARAB EMIRATES
FROM SPACE
©) © mie £ 4
Figure 1. satellite image showing UAE with overlay of a mosaic
SAR image used for the study area.
The Arabian Gulf region is the largest offshore oil development
area in the world. Bahrain, Iran, Iraq, Kuwait, Qatar, Saudi
Arabia, and the United Arab Emirates(UAE) produced over
27% of the world’s oil in 2000. The area also holds 65% of the
world’s oil reserves. The study area have one of the busiest and
most important tanker shipping lanes in the world; one ship
passes the strait approximately every 6 minutes, another
statistic indicates that more than 40 % of the world’s total oil
transportation passes through the region. The oil sludge,
released by tankers cruising in the Arabian Gulf is estimated to
be around 8 million metric tons per year, representing 60 % of
the total pollution in the area. Oil of 15.5 million barrels per
day is transported through the strait of Hormuz. Contamination
influx is mainly from tankers releasing ballast, tank cleaning
leakage from drilling rigs and production platforms, and ship
accidents.
3.2 Data set
Ers-1/2 C-band SAR data has been used for the great majority
of oil spill detection. However, other satellites images even
optical sensor images have shown good capabilities for this
purpose. Therefore, we searched all available image data
archives and selected more than one hundred of images derived
from different platforms that covered the most of the offshore
water of the UAE. We conducted a search of ERS-1/2 data
archive to compile a list of all images acquired over the study
area. To evaluate their suitability for slick detection, historical
wind conditions for corresponding SAR images were obtained.
For each date of acquisition, surface wind speed histories were
reconstructed using historical records. Because radar
backscatter from the sea surfaces are strongly affected by
surface wind patterns, the wind speed histories were used to rate
the suitability of each image for oil slick detection.
A total of more than 300 frames of ERS-1/2, Landsat-7 ETM,
JERS-1 SAR, Terra ASTER, RADARSAR, and Space Shuttle
photographs were used. The images were acquired within the
period between mid 1980's and early 2001. These images were
used. The images were not uniformly distributed over the study
area. The frequency of coverage ranges Was maximum 5 scenes
per frame. the following are the satellite imagery used in his
project:
ERS-1 SAR (Japan), 64 scenes
Shuttle Imaging Radar C/X-SAR (US), 37 segments
ERS-1/2 AMI (EU), 15 scenes
Radarsat SAR (Canada), 3 scenes
Envisat ASAR (EU), 7 scenes
BMC
Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol XXXV, Part B7. Istanbul 2004
Hexagon KH-9 (US), 12 scenes
Landsat-7 ETM+ (US), 15 scenes
JERS-1 OPS (Japan), 92 scenes
Terra ASTER (Japan/US), 95 scenes
0. Shuttle Handheld Camera Photographs, 50photos
== ND OU IO
3.3 Image Analysis and oil Slick detection
Briefly summarizing the characteristics of each satellite-borne
imaging sensor in oil slick detection, in the visible and near-
infrared region, the absorption and reflection of solar
illumination determines the characteristics of the oil by means
of spectral reflectance of electromagnetic energy. Crude oils
show different colours, that is spectral reflectance varies
depending upon chemical composition of crude oil. Sea surface
roughness changes the direction of solar illumination reflected
from water, due to randomly scattered sun glint from wave
facets oriented at the specular angle, as does the presence of sea
surface condition(berry, 1995). As a results, reflectance contrast
between clear water and oily water varies with sea state at any
given wavelength in the visible and near-infrared region.
By looking at the thermal infrared region, the water surface
temperature can be calculated. Brightness temperature is
calculated with respect to function of both emissivity and
kinetic temperature. A film of oil on water has the same kinetic
temperature as the water since they are in direct contact. The
difference of emissivity 0.02 between Seawater and crude oil
makes an apparent difference of 1.20C in brightness
temperature in the wavelength region of Landsat TM band 6 at
room temperature (Salisbury et al., 1993). Generally an oil
slicked surface shows a lower temperature than the surrounding
clear water surface (Figure 2.). The image data observed in
night time is more reliable to avoid the influence of solar
illumination difference between seawater and oil slicks.
Figure 2. water surface temperature of clear water and of
slicked water.
©
In the microwave region, the brightness of the sea surface is
measure of backscatter of the sea surface roughness. As smooth
sea surface appears dark and the brightness increases as the sea
surface becomes rougher. Oil films damp wind-generated
gravity capillary waves on the sea surface. Hence they appear
dark against brighter surrounding areas in a SAR image.
660
Inter.
A lai
the A
nece:
The :
funct
time
differ
oil a
unifo
raisin
if the
effec!
rougl
or ob
The i
map
image
optim
range
proce
ERS
hours
satelli
was b
consk
acqui:
per se
during
condi
detect
speed
hours,
oil sli
during
Analy
image
of po
patter
differ:
certait
Satelli
(centre
spill c
Lands
Other
the c
Horm
routes
that th
of Om
areas.
the cc
image:
region
associ;
Arabia
deliber
Oil dis
observ
ballast
silver
on the
