EXPLORING THE POTENTIAL OF MODIS VISIBLE AND THERMAL
CHANNELS IN MONITORING AND ASSESSING THE IMPACT OF
DESALINATION PLANT DISCHARGES IN THE ARABIAN GULF
Ammar Al Muhairi1, Hosni Ghedira2, Hussain Al-Ahmad3, Ali Dawood3, and Mohammed Al-Mualla3
1Associate Research Engineer, DubaiSat-1 Program
Emirates Institution for Advanced Science & Technology (EIAST), Dubai, UAE.
2American University in Dubai, Dubai UAE
3Khalifa University of Science, Technology and Research (KUSTAR), Sharjah, UAE
Sea water desalination has experienced an unprecedented
growth in the GCC countries to meet the ever growing
demand of water for household consumption as well as for
industrial and agricultural purposes. However, the current
technologies used in water desalination are also
accompanied by negative environmental impacts especially
on the surrounding marine ecosystems. Since major seawater
desalination plants are located by the shoreline, the main
environmental considerations in desalination are water
intakes and sea outfall discharges. We intent through this
study to evaluate the potential of current polar orbiting
satellites in evaluating the impact of desalination plant
discharges, usually used to dispose of brine waste stream, on
surrounding ecosystems and water quality. The objective of
this project is to develop an automated approach for
monitoring water quality and temperature (thermal
properties) surrounding the discharges of desalination plants
in the UAE coastal areas. Visible and thermal measurements
provided by MODIS sensors on board of Terra and Aqua
satellites are used in this project. The first four bands
(visible) and band 31 & 32 (thermal) were selected. Future
multi-spectral data from DubaiSat-1 (5-m resolution) will be
also used to detect small changes in water color that cannot
be detected with the MODIS data (250 m).
Index Terms— environmental impact, remote sensing,
desalination plants, MODIS, water quality.
The four major desalinations plants in the UAE have
been selected in this study with total capacity of around 750
million gallon per day (MGD): Jebel Ali, Al Marfa, Umm
Al-Nar, and Shuwayhat (figure 1).
Umm Al Nar
Shuwayhat Al Marfa
Fig. 1: Monitored desalination stations
III - 357 978-1-4244-3395-7/09/$25.00 ©2009 IEEEIGARSS 2009
The geographic coordinates and current production
capacities of these stations are presented in table 1. This
paper will cover the first stage of this project that focus on
the evaluation of the impact of Jebel Ali station in Dubai
which has the highest production capacity in the UAE.
Table 1: location and capacity of monitored stations
*As of March 2008
The arid climate of the Arabian Gulf contributes to
sedimentation through increased wind action and the
infrequent but heavy rainfalls which cause flash floods.
Additionally, high water temperatures with high evaporation
rate and low water inflow through rivers and precipitation
cause a circulation pattern that favor inorganic carbonate
development of sediments, as shown in MODIS image
presented in figure 1.
2. LITERATURE REVIEW
Several studies related to environmental impacts of
desalinated plants in the Arabian Gulf have been carried out
in the past 10 years. In 2002, Abdul Azis et al.  presented
a set of environmental data collected in the Saudi coastal
waters near Al-Jubail desalination plant. Water samples
from six different sites covering the intake and discharge
zones were collected between 1997 and 1998. Twenty-eight
species of phytoplankton were identified and analyzed. This
study found that the desalination plant discharge has no
significant effect on phytoplankton and chlorophyll pigments
. Another study was also performed in 2005 by Abu
Dhabi Water & Elec. Authority  to evaluate the
environmental impact of Umm Al-Nar desalination station in
the UAE. This station is surrounded by a sensitive
ecosystem with extensive areas of mangroves and seagrass
meadows. The spatial distribution of high temperature and
salinity as well as water flow properties surrounding the
desalination plant has been observed and studied. Their
observation had shown no significant deterioration of water
quality surrounding the plant.
Other type of environmental study was also performed
to evaluate and mitigate the potential damage of an eventual
oil spill accidents. Elshorbagy and Elhakeem (2007) have
produced a set of 10 hazard contour maps for the prediction
of oil spill travel time and critical wind direction in
association with five selected mega-desalination plants along
the UAE coast: Al-Shuwayhat, Al-Marfa, Umm AlNar,
Taweelah, Jebel Ali and Al-Layah. The author produced
to frequent and visible
hazard maps using a three-dimensional coupled set-up of a
hydrodynamic model (Mike3-HD) and oil spill model
Satellite measurements have been widely used for water
quality monitoring in the last three decades. CZCS and
SeaWiFS, which were launched in 1978 and 1997
respectively, were the first two earth observation satellites
devoted to water quality monitoring and measurement. The
long gap between launching the two instruments proofs that
the lack of suitable sensors has limited the use of remote
sensing in the past . Water quality monitoring from space
has been extended by the launch of MODIS sensors on
board of Terra and Aqua satellites. Satellite data has been
used to measure different water quality parameters including
color, phytoplankton (chlorophyll a), total suspended matter
(TSM), colored dissolved organic matter (CDOM), turbidity
Hu et al. (2004) have successfully used MODIS data to
map chl a, aCDOM(400) and TSM values in Tampa Bay
(Florida, USA). For TSM estimation, they used a channel
difference algorithm: R(645)−R(859).
algorithm R(469)/R(555) was used to retrieve aCDOM(400)
and chl a. For the combined data set, the authors used a
multi-band ratio which is based on complex ratio of channel
differences. A high correlation coefficients (up to 0.96)
between satellite data and water parameters were obtained
250-m resolution data of MODIS has been also used by
Miller and McKee (2004) to map the concentration of TSM
in the Northern Gulf of Mexico. The authors found more
than 90% correlation between the first band of MODIS (645
μm) and in situ measurements of TSM .
MODIS data were also compared with SeaWiFS data by
Dall’Olmo et al. (2005) in estimating chl a in turbid water
using Red and NIR bands. MODIS (667,748) appeared to
provide the most accurate prediction of chl a .
MODIS data has also been used in monitoring water
quality in the Middle East region (Red Sea and Arabian
Gulf). Nasr et al. (2007) estimated chl a, TSM and Sea
Surface Temperature (SST) in the Red Sea. OC4 algorithm
was used to estimate chl a and other two empirical
algorithms were used for estimating TSM and SST. The
derived and the measured chl a and SST were very well
correlated (RMSE= 0.13 and 0.37) respectively. While the
concentration of TSM needed an offset for satisfactory
correspondence (RMSE= 4.86) .
Reza (2008) used MODIS data to develop an algorithm
for retrieving Suspended Sediment Concentration (SSC)
from MODIS spectral radiance over Bahmansheer River
Estuary at the North-West of Arabian Gulf. Algorithm 1 was
obtained from the relationship between the measured
concentrations and MODIS channel 1 reflectance difference
(Δρch1). The resulting r2 was poor (0.59). As for algorithm 1,
Algorithm 2 was obtained but with channel 4 reflectance
difference (Δρch4). A better r2 of 0.77 was found. Algorithm
III - 358
2 has shown more sensitivity to SSC variation and was
recommended for turbid estuaries and coastal waters
In this project, different existing tools such as the Ocean
Color Chlorophyll (OC4) algorithm will be also tested. This
algorithm, based on a four-band maximum band ratio
formulation, has been successfully used to estimate chl-a
concentration. Similar MODIS-based algorithms have also
been developed to map the Total Suspended Matter (TSM)
distribution and Sea Surface Temperature (SST).
3. DATA ANALYSIS
Visible reflectances and sea surface temperatures have
been derived from MODIS data. An average of six scenes
has been used each month. The images presented in figure 2
show the sensitivity of the three visible channels of MODIS
to the variation of water properties. High sensitivity was
observed with the blue channel where the reflectance has the
better response to the variation in water turbidity.
10 2030 40 50
Green (545-565 nm)
Fig. 2. Three visible channels and one temperature channel derived from MODIS on January 10th 2009
In addition to its sensitivity to water properties, the
measured reflectances can be also affected by the satellite
acquisition angles which vary from day to day. Figure 3
shows how reflectance changes with the acquisition angle
within a specific area. The green channel shown mor
sensitivity to satellite acquisition angle. An angle of 45º
results in highest reflectance and reflectance decreases with
a higher or lower angle. In the NIR band, the reflectances
are much less sensitive to the acquisition geometry.
The in situ measurements were provided by Jebel Ali
desalination plant in a monthly basis. The difference in
temperature of intake and brine discharge was about 10˚C in
average. Moreover, there was a difference of about 1.1˚C in
average between the temperature of the intake seawater and
the one measured by MODIS. Figure 4 illustrate the
difference between field-measured and satellite-measured
Fig. 3. Reflectance vs. Acquisition angle
III - 359
Fig. 4. Comparison between field-measured and satellite- Download full-text
The preliminary results show that satellite-based
temperature is closer to the intake temperature rather than
brine temperature. This observation shows that the high
temperature of the brine has a local impact on surrounding
water. However, it is important to mention that the low
spatial resolution of MODIS thermal channel (~1 km) make
it less sensitive to local variations of the temperature.
Additionally, the short distance between the intake and the
outfall to the shoreline adds more noise to the measured
temperature by increasing the contribution of the land
radiation to the temperature measured by the satellite.
For the next stage, the spatial and temporal variation of
water reflectance will be analyzed in two other stations in
the UAE. The correlation level between satellite reflectances
and concentrations of some water quality parameters will be
 P.K. AbdulAzis, I.A. Al-Tisan, M.A. Daili, T.N. Green, A.G.I.
Dalvi and M.A. Javeed, “Chlorophyll and plankton of the Gulf
coastal waters of Saudi Arabia bordering a desalination plant,”
Desalination, vol. 154, pp. 291-302, 2002.
 A. Areiqat and K. A. Mohamed, "Optimization of the negative
impact of power and desalination plants on the ecosystem,"
Desalination, vol. 185, pp. 95-103, 2005.
 W. Elshorbagy and A. Elhakeem, "Risk assessment maps of oil
spill for major desalination plants in the United Arab Emirates,"
Desalination, vol. 228, pp. 200-216, 2007.
 S. Koponen, “Remote Sensing of Water Quality for Finnish
Lakes and Coastal Areas,” Ph.D. dissertation, Helsinki University
of Technology, Espoo, 2006.
 C. Hu, Z. Chen, T. D. Clayton, P. Swarzenski, J. C. Brock and
F. E. Muller–Karger, “Assessment of estuarine water-quality
indicators using MODIS medium-resolution bands: Initial results
from Tampa Bay, FL,” Remote Sens. Environ., vol. 93, no. 3, pp.
 R. L. Miller, and B. A. McKee, “Using MODIS Terra 250 m
imagery to map concentrations of total suspended matter in coastal
waters,” Remote Sens. Environ., vol. 93, pp. 259-266, 2004.
 G. Dall’Olmo, A. A. Gitelson, D. C. Rundquist, B. Leavitt, T.
Barrowb, and J. C. Holz, “Assessing the potential of SeaWiFS and
MODIS for estimating chlorophyll concentration in turbid
productive waters using red and near-infrared bands," Remote
Sens. Environ., vol. 96, pp. 176-187, 2005.
 A. H. Nasr, B. M. El Leithy, and A. K. Helmy “Assessment of
Some Water Quality Parameters Using MODIS Data along the Red
Sea Coast, Egypt,” ICGST-GVIP Journal, vol. 7, no. 3, 2007.
 M. M. Reza, “Assessment of Suspended Sediments
Concentration in Surface Waters, Using MODIS Images,” Am. J.
of Applied Sci., vol. 5, pp. 798-804, 2008.
III - 360