DATA PROCESSING SATELLITE NOAA-AVHRR SEA SURFACE TEMPERATURE FOR THE MEASUREMENT OF AVERAGE DAILY

In 2000, BRKP establish Earth Station Satellite Data Receiver NOAAAVHRR in 3 (three) locations in Indonesia in the interests of fishing activities by
the fishermen. In 2004, the French Government to provide grants to BRKP
enhance the ability of these earth stations are located in Perancak, Jembrana, Bali and
Bitung, North Sulawesi.
Increased capacity of the earth station will allow the receiver to
receives satellite data throughout Indonesia. There currently are 6 series NOAAAVHRR satellites that orbit the NOAA 12, NOAA 14, NOAA 15, NOAA 16, NOAA 17 and
NOAA 18, NOAA 18 which is the most recent satellite, launched in
dated May 20, 2005.
Each series of NOAA satellites across the territory of Indonesia about two times a day.
With the 6 series which is still orbiting NOAA satellites today, then in one day less
more will receive 12 satellite images that can be used for various purposes.
*) Research and Observation of Marine, Marine Technology Research Center, BRKP-
Ministry of Maritime Affairs and Fisheries 2
Particular interpretation of satellite data on the reflection of infrared wavelengths to
measurements of sea surface temperature requires a homogeneous target with emissivity
are known. Previous studies on sea surface temperature using satellite
Nimbus clouds using the photo as a comparison in which at a certain temperature will
detected as a cloud. The presence of clouds is of course reduces the area that could be detected
sea ​​surface temperature.
Subsequent temperature patterns can be mapped, not just on individual data daily to
small parts of the earth's surface, but also replicates the data in a particular area
to see changes. (EC Berets and LFCurtis, 1992)
Sea surface temperatures are very closely related to primary productivity and
currents. Changes in sea surface temperatures caused by wind currents, turbidity and
waves commonly called ocean dynamics. Differences in sea surface temperatures can also be
observed by remote sensing technology (F. Srihadiyanti Purwadhi, 1986)
Is an opportunity for monitoring sea surface temperatures in
study and exploitation of marine and coastal areas, where the application of remote sensing technology
for oceanography has been introduced since the last two decades. Visible satellite data
and infrared radiometer in oceanography was launched aimed at first time
in 1978 using satellite TIROS spacecraft. In addition, there are many
efforts and research to determine the parameters which allow for
used by the oceanographic community is provided with a vehicle or aircraft
satellite. (K.Beier, 1980)
Knowledge of sea surface temperature is very useful for many things
associated with other research and application utilization. Sea surface temperature
is one of the main factors driving cycle of the seasons in both the regions and sub tripis
tropical sea surface temperatures which will affect the atmospheric conditions, weather and seasons,
even the emergence of the phenomenon of El Nino and Lanina can be learned through the surface temperature
the sea. Many other things that related to the applications that can be affected by temperature
sea ​​level, including the fertility waters / marine and fisheries.
One of the major constraints the use of NOAA satellite data for monitoring the temperature
sea ​​level is limited by the wavelength used satellites where
wavelength used is sensitive to temperature changes or differences
sea ​​level but can not penetrate clouds.
Limitations in this case clearly does not penetrate clouds reduce the availability of data
continuously and complete because of the general atmosphere in the area of ​​Indonesia has
cloud cover is so high that often times with data from six satellites NOAA
received was not one that can be used to obtain the temperature information
sea ​​level.
This paper aims to provide sea surface temperature of the average
Daily sea surface temperature and distribution of marine territory of Indonesia.
Theory
NOAA satellite data analysis is performed to extract phase
sea ​​of ​​information through a series of processes according to the characteristics of satellite data
and outputs to be generated.
Satellite data received at the earth station will be stored in the format. L1b and. N10
where format is a raw data to be processed further into information
Another sea surface temperatures in particular. Stages of data processing to be done
to obtain results that can be analyzed as follows: 3
Format Conversion
NOAA satellite data processing is done using the software HRPT
Reader and ER Mapper where the processing of the original raw satellite data have
noaa.L1b will convert format in a format noaa.ers so that raw data can be read and
further processed using the software.
Conversion of NOAA satellite data into the format. Ers will generate satellite data
with 5 channel / channels / bands, where Channel 3, 4 & 5 are used to monitor temperature
sea ​​level. NOAA satellite spatial resolution is 1.1 km, or in other words each
pixels in the image represents an area of ​​1.1 km2 of data on the earth's surface.
Invert Value
Raw NOAA satellite data must be processed further so that the pixel values ​​have
accordance with the existing sea surface temperature at the time (when recording data). In
common value of pixels in the raw satellite data must be in Invert (reverse value) where value
pixels in the range of 0-255 behind a 255-0. which means for instance pixel value
object representing the cloud initially has a value 255 or white in greyscale
color mode was changed to 0 or black, and vice versa colored objects
black or pixel value 0 will be changed to be 255 or white. To object to the value
50 pixels will be changed to 205 and beyond.
Rectification
Apart from having the convert and invert, raw data should also be in rectification, where
previous raw data only has the position information matrix of pixels. With
perform rectification then each pixel will have a geographical position as
the earth's surface.
Rectification is done by using another map as a guide. Usually
in a one-time recording it will get the image to the sweep width of ± 2000 km or
about one-third of Indonesia. By using the map of Indonesia as
guide the NOAA satellite data can be plotted according to the geometry of the
Indonesia. One common problem that often occurs is the amount of cloud cover in
the islands of Indonesia that although the overall cloud cover below
20% but will be very difficult to be processed in the absence of a control point in time will
carried out rectification.
Calculation of sea surface temperature
The process is performed to obtain the Sea Surface Temperature is the
using the algorithm of sea surface temperature (in this study using the software
ER Mapper), which was introduced by Crosby & Mc Millin, where to get
sea ​​surface temperature information is used bands 4 and 5 as a band that is sensitive to
changes in sea surface temperature.
Sea Surface temperature algorithms:
SST: input1 +2702 * (input1-INPUT2) -0582
Input1: Value Band 4
Input2: Value Band 5 4
Mozaicking
Data is already in the process of rectification then the algorithm to be changed
a temperature value in units of ° C. The data obtained is still a single data
for the recording just about one-third of Indonesia and the region has not
free from clouds.
To get the data cloud coverage and better coverage area
all parts of Indonesia it is necessary to the process of merging or mozaicking
a single data with the different coverage.
Done with the process of merging with an average value of sea surface temperature
so there is no apparent alignment linement or as a result of temperature differences
contrast between two single data or lebih.proses average temperature value can be explained
simply by summing the values ​​of sea surface temperature of all single data
and divided by the amount of data used. But with the way it turned out to be simple
result in less accurate data as a result of having a single data
clouds and in total the land and divided by the amount of a single data used. Sure
The resulting values ​​are no longer pure as the average sea surface temperature due to
presence of clouds and land value in it.
Making Value Index
To avoid the error value as a result of clouds and land value
then the required value index where the data that has clouds and land value will
given the index 0 while the value given the index of sea surface temperature 1.
In the areas of cloud cover so small that the value will indeknya
average divider factor is also small, and thus the resulting average is
affected by cloud cover.
Calculating the average
By using the value index and sea surface temperature data from several
single dat then combined and then on the average to produce the data
Daily sea surface temperature with a more refined and easier to
diinterpetasi.
Average calculation is done by adding the value of sea surface temperature and
divided by the index. Value index was obtained by changing the value of land and clouds
to 0 (null) and sea surface temperature value to 1 (one). In this way,
the average value obtained pure value of sea surface temperature.
Average sea surface temperatures: Number of Total Value / Total Total Index
Number of Total Value: the sum value of all the surface temperature data
sea ​​in the mozaic
Total number of indexes: the sum of value index of the data cloud and land
the dimozaic 5
Method
The study was conducted by using remote sensing methods in a way
processing of satellite data for sea surface temperature through the stages
processing: Convert format, Invert Value, rectification, Calculation of Surface Temperature
Sea, Mozaicking, Making Value Index and average calculation.
Satellite data obtained from NOAA-AVHRR ground stations Perancak-Bali
operated by the Center for Marine Research and Observation, Department of Marine. Software that
ER Mapper is used which has the capability to process satellite data with
applying satellite data pixel calculation process.
Sea surface temperature derived from the processing of NOAA-AVHRR satellite data that
has 5 channels in the vehicle that carries it. To obtain the surface temperature
the sea lines used were 4 and 5 channels where the channel is
thermal infrared channels.
Calculation using the average index value as the divisor factor
performed to obtain daily sea surface temperature for the entire territory of Indonesia,
this is done to overcome the limitations of satellite data which can not penetrate clouds,
so that this process will be obtained with the distribution of sea surface temperatures are more subtle
Maps and eliminate cloud cover.
Data
NOAA-17 satellite data and NOAA-18 recording on 1 January 2006 was chosen
as raw data. On that date from NOAA satellite data seen the object of the island
and a coastline that could be used as a fastening point at the time of the data processing
them. Indonesia region covered by three satellite data. Indonesia recorded the western region
by NOAA-18 satellite as shown in Fig.1
Fig.1. NOAA satellite data 18 to January 1, 2006 at 14:26 AM

While the central region of Indonesia recorded by NOAA-17 satellite which covers
Kalimantan, Sulawesi, Java, Bali and Nusa east as shown on
gambar.2.
Gambar.2. NOAA satellite data 17 to January 1, 2006 at 10:12 AM

The next area of ​​eastern Indonesia recorded by NOAA-18 satellite which covers
Papua, Ambon and East Nusa Tenggara as shown in gambar.3
Gambar.3. NOAA satellite data 18 to January 1, 2006 at 12:48 AM

Results and Discussion
Gambar.4. Sea Surface Temperature data is covering the eastern part of Indonesia,
obtained by processing the satellite data NOAA-18 (gambar.3). Gambar.4. show
a high value close to 28 ° C are shown as green to yellow,
with uneven distribution of high temperatures where only scattered at the top of
Papua, which is part of the Pacific Ocean.
Processing of satellite data NOAA-17 (gambar.2) showed that high temperature
represented by the color yellow to red, visible in the Makassar Strait and the Java sea sea
range of temperatures near 30 ° C which covers the central part of the territory of Indonesia, such as
shown in gambar.5.

Gambar.5. Sea surface temperature satellite data from NOAA-17 dated 1
January 2006 at 10:12 AM
Indonesia's western region recorded by the satellite NOAA-18 (Fig.1) after
processing carried out showed that sea surface temperature high enough in the south of Java
until the temperature approached 29 ° C shown in yellow to reddish, but
Thus because of the high cloud cover caused sea surface temperature data in the
others are not detected, as shown in gambar.6.
Gambar.6. Sea surface temperature satellite data from NOAA-18 dated January 1
2006 at 14:26 AM
Of gambar.4, gambar.5, and gambar.6 looks different directions of tilt data,
where gambar.4 and gambar.6 has a leaning to the left while gambar.5
has a tilt to the right. This difference is due to the direction of the satellite across the
simultaneously in different parts of Indonesia where the past and gambar.6 gambar.4 (NOAA-18)
across Indonesia from south to north (ascending), at the time the satellite passed
simultaneously with the rotation of the earth to the right so that the earth's surface is recorded
left shift and produce a skewed left. In contrast to the gambar.5
(NOAA-17) across Indonesia is the region from the north to the south (descending)
resulting in an oblique view to the right. However, differences in appearance
does not affect the value of sea surface temperature resulting from the analysis
the data.
Fig.7. shows the distribution of sea surface temperature as if there is a temperature
sea ​​surface height in the middle like in Java, Kalimantan and Sulawesi are
shown in red and yellow colors. In fact the temperature is measured in
telalu area is not high. This occurs because of overlapping stacking (overlay)
in the region, especially between the data with the data of NOAA-17 NOAA-18 where both
Gambar.6. Sea surface temperature satellite data from NOAA-18 dated January 1
2006 at 14:26 AM
30 ° C
20 ° C 9
these data have different sea surface temperature, because it does not do the calculation
the average of the boundary between the two data lines are ultimately pseudo
make these data difficult to interpret.
Fig.7. Sea surface temperature is the average area of ​​Indonesia on 1 January 2006
without using the index value as the divisor factor

The results showed an average daily sea surface temperature on the 1
January 2006 and its distribution as shown in Fig.8. Data on
that after the merger process and the average use value index
as a divisor shows the temperature variation in the waters of Indonesia.
Fig.8. Sea surface temperature is the average area of ​​Indonesia on 1 January 2006

Sea surface temperature distribution as in gambar.4. shows the temperature variation in
Indonesia region with a temperature range between 20 ° C with up to 30 ° C in the show
in blue to low temperature and the red color for high temperatures seen
smoother and easier on the interpretation.
By entering a value index as the divisor factor as in Fig.8. then
sea ​​surface temperature daily average will be obtained more accurately where the value in
each point is not biased by cloud cover. If you do not use the index value as a factor
divider then when there is a point on sati cloud cover is relatively lower temperature
compared to sea surface temperature, the average daily temperature into this rendah.hal
of course reduces the accuracy of sea surface temperature data daily average.
conclusion
From this study it can be concluded that the surface temperature data on average daily
will be much smoother and more easily interpretable when using data on more than one
to cover the whole territory of Indonesia and the analysis using the calculation
the average index value as the divisor. In this way the data will be obtained
sea ​​surface temperature daily average for the area of ​​Indonesia with more distribution
clear and close to the real situation on the ground.