Utilization of Jason-3 Satellite Altimetry Data for Observation of TC Seroja

The Jason-3 Satellite Altimetry has two main instruments: the Poseidon-3B Altimeter and the Advanced Microwave Radiometer. The two sensors produce many output data. Satellite altimetry data commonly used to observe TC are Sea Level Anomaly (SLA) and Significant Wave Height (SWH). This research tries something new by adding three other data, namely Precipitable Water Vapor (PWV), Liquid Water Content (LWC), and Wind Speed (WS). The object of this research is TC Seroja which occurred from 2 to 12 April 2021. This research was conducted to obtain all variable’s spatial and temporal profiles. The results of this study indicate that TC Seroja causes an increase in SLA, PWV, SWH, LWC, and WS. The highest SLA was 0.495 m, SWH was 4.649 m, PWV was 0.10760 m, LWC was 2.08 kg/m2, and WS was 18.42 m/s. In addition, the spatial model for the five variables in April 2021 obtained using IDW (Inverse Distance Weighting) can also provide information on the influence of TC in various places.


Introduction
A tropical cyclone (TC) is a form of extreme weather disturbance that begins with a tropical depression or an intensive center of low pressure over the ocean, thus triggering intensive convection processes and cloud formation [1].TC can occur in warm waters with temperatures over 26.5 o C [2].Most TC form at latitude 10 o -20 o from the equator [3].TC that form below 10 o from the equator are few due to the small Coriolis force, even though the temperature in that area is high.Conversely, TCs that form above latitude 20 o from the equator are few due to the low temperature in that area, even though the Coriolis force in that area is more significant.
Indonesia is located at latitude 6 o N to 11 o S and longitude 95 o E to 141 o E.Even though they are located at low latitudes, TCs rarely occur in Indonesia [4], [5].The characteristics of TC in Indonesia are divided by region, namely TC in southern Indonesia and TC in northern Indonesia.Most TC in southern Indonesia form from December to April.Meanwhile, most TCs in northern Indonesia are formed from 2 July to October [6].Several TCs that have occurred in Indonesia are TC Vamei in 2001, Rosie in 2008, Cempaka in 2017, Dahlia in 2017, Flamboyan in 2018, Kenanga in 2018, and Seroja in 2021 [7].TC can be observed using data from satellite altimetry [8].The satellite altimetry data commonly used in TC observations is Sea Level Anomaly (SLA) and Significant Wave Height (SWH).Several studies use this data to observe TC.One study showed that the TC Cempaka and Dahlia caused an increase in SWH in the south, north, and strait areas from Lampung to Central Java with the highest value was 3.75 m and the largest percentage of SWH increase was 1028.31%[9].Another study used SLA data from satellite altimetry to observe sea level height in southern Java when TC Cempaka occurred.This study's results indicate a decrease in SLA reaching 0.2 m at the center of the cyclone and an increase in SLA reaching 0.4 m in the southern coastal areas of Java [10].In addition, there is also research to examine SLA changes on the Lampung Coast when TC Dahlia occurs.This study uses SLA data from satellite altimetry from 30 November to 2 December 2017.As a result, TC Dahlia increased SLA in Kota Agung and Krui waters [11].
One of the huge TCs that recently happened was TC Seroja which occurred from 2-12 April 2021.On 31 March 2021, wind speeds over the island of Timor began to increase, followed by the formation of a depression zone in the Savu Sea on 1 April 2021.The depression zone continued to strengthen to form a low-pressure system.On 2 April 2021, a low-pressure system formed a TC [12].Madden-Julian Oscillation (MJO) at that time which was at the turn of phase 4 and phase 5 formed a supportive environment for cyclogenesis [13].The TC moves southwards until it ends in southern Australia on 12 April 2021.The movement of TC Seroja's center from its beginning to its end can be seen in Figure 1.Other research found that TC Seroja influenced the intensity of extreme rainfall and wind speed in Nusa Tenggara Timur and western Australia [13], [14].This extreme rainfall triggered flooding in several areas [15], [16].
Not only to analyze TC Seroja's impact, but this research also aims to analyze the potential of satellite altimetry data to observe TC.Other research said that TC Seroja caused extreme rainfall and wind speed based on rainfall and wind speed data from other sources.This study tries to observe the impact of TC on rainfall and wind speed using satellite altimetry data.Satellite altimetry has two main instruments, namely radar altimeter and radiometer.To study wind speed, wind speed data from altimeter radar sensor measurements are used.While rainfall was studied indirectly using Liquid Water Content (LWC) and Wet Tropospheric Correction (WTC) data from radiometer measurements.In addition, this study also adds SLA and SWH data from radar altimeter measurements to observe the impact of TC Seroja on these two ocean parameters.
This study describes the feasibility of satellite altimetry data to observe TC.If proven usable, it is hoped that satellite altimetry data can be combined with other data to conduct more comprehensive research.

Data
The data used in this study are SLA, SWH, WS, WTC, and LWC from the Jason-3 satellite altimetry.Data were obtained using the Radar Altimetry Database System (RADS) in the ITB Geodesy Research Group room.The Jason-3 satellite altimetry has 9,9156 days repeat cycle period and 254 passes within a cycle, ascending (asc) passes are odd-numbered and descending (dsc) passes are even-numbered [17].
SLA is the height of the sea level relative to the Mean Sea Surface (MSS).SLA can be obtained by reducing satellite altitude data by satellite distance to sea level, propagation correction, bias, and MSS models [18].The measurement of the distance between the satellite and the sea surface is carried out by an active radar altimeter sensor.The signal emitted by the altimeter radar will be reflected by the sea surface and again captured by the altimeter sensor.The time difference between signal transmission and signal capture is again recorded and then processed into data measuring the distance between the satellite and the sea surface [19].
SWH is the average height of 1/3 of the highest ocean waves [20].The presence of waves on the surface of the sea changes the shape of the return wave which is measured by the radar altimeter.The leading edge of the returned waveform is stretched as a result of the previous return from the crest of the wave and then the return from the trough of the wave.This stretching increases with increasing wave height.Therefore, wave conditions can be estimated from the slope of the leading edge of the returned waveform [19].The Jason-3 satellite mission measures SWH using two frequencies, namely Ku-Band and C-Band [18].TC cause rain in the affected areas.Signals at the Ku-Band frequency get more significant interference from rain when compared to signals at the C-Band frequency [21].Therefore, in this study, SWH measurements were used at the C-band frequency to observe the influence of TC Seroja on the characteristics of sea waves.
WTC is a correction to signal propagation disturbances by water vapor and cloud grains in the tropospheric medium [19].And LWC is the amount of water content in the atmosphere in liquid form.WTC is obtained using brightness temperature (BT) measurements measured at a multi-frequency radiometer.Channels with a frequency of 23.8 GHz are very sensitive to water vapor which captures most of the WTC information.Higher frequency channels are also used to capture LWC contributions.This second channel has a frequency of 34 GHz.The radiometer used by the Jason mission also uses a third channel at 18.7 GHz, in addition to good sensitivity to water vapor, providing additional information about the sea surface roughness [23].
The last data used is WS.In addition to measuring the waveform of the received signal to estimate SWH, the altimeter also measures the energy of the received signal (backscatter coefficient) which can be used to estimate WS at sea level [19], [22].The wind will reduce the energy of the signal as it travels through the atmosphere.The smaller the energy of the received signal, the greater the wind speed in the area traversed [18].Two kinds of WS data can be obtained from the Jason-3 satellite altimetry, namely altimeter wind speed and water vapor radiometer wind speed.In rainy conditions, the wind speed measurement by the water vapor radiometer sensor has worse accuracy than the wind speed by the radar altimeter sensor [23], [24].Therefore, the wind speed measurement results from the altimeter sensor were used in this study.

Convert WTC to PWV (Precipitable Water Vapor)
Altimetry satellites have an on-board microwave radiometer whose original function is to measure the WTC for their measurement using radar altimeter sensor.The WTC, as it reflects the amount of water vapor in the atmosphere, could be used to derive the PWV [25].This section will describe an algorithm for obtaining the PWV from the WTC generated by satellite altimetry.
The WTC obtained by the satellite altimetry is negative of the Zenith Wet Delay (ZWD) (Equation 1).To get PWV from ZWD can use equation 2, with an approximate value for Π being 0.16 [26].

Inverse Distance Weighting (IDW) Interpolation
The first law of geography states, "Objects that are close together in space are always more similar than those that are far apart" [27].IDW is an interpolation method that fulfills the first law of geography and is one of the most frequently used interpolation methods in geographic information systems [28].IDW performs interpolation by assigning a weight to the sample data based on the distance.Closer sample points will have greater weight [29].

Result and Discussion
SLA, SWH, WTC, LWC, and WS data were obtained from RADS in areas ranging from latitude 4 o S to 40 o S and longitude 90 o E to 135 o E. This area was taken to determine the extent of the influence of TC Seroja on SLA, SWH, PWV, LWC, and WS.In this area, there are 44 satellite altimetry passes consisting of 22 ascending and 22 descending passes, as seen in Figure 2. It should be noted that satellite altimetry does not pass through all of these passes at the same time as the TC Seroja.Therefore, it is necessary to select the pass that the satellite passes at the same time as TC Seroja occurs.Data processing is divided into 3, namely spatial profiles, temporal profiles, and spatial models.

Spatial Profile
The Jason-3 satellite altimetry orbits at a speed of 7.2 km/s in outer space or the equivalent of 5.8 km/s on the earth's surface.During operation, satellite altimetry retrieves data every 1.0187 seconds.Thus, the distance between the nearest observation points on the same pass is 5.8636 km.This sampling distance is close enough to observe TC, which generally has a radius of up to hundreds of kilometers [3].In this area, there are 44 satellite altimetry passes.But only pass 64 (dsc) that close to TC Seroja's center (< 150 km).This pass was passed by satellite altimetry on April 7, 23:26 UTC.This pass is the closest to the center of TC Seroja compared to the other 43 passes.The distance between pass 64 and the center of TC Seroja is 23,231 km.
The profiles of the five variables in pass 64 can be seen in Figure 3.The figure shows data gaps in the SLA profile at -14 o to -16 o latitude.This deletion can occur because the SLA value in that range is above the threshold set by RADS, so it was considered as outlier and deleted automatically.However, the SLA profile still shows the influence of TC.The figure shows that the value of SLA decreases near the center of the TC and increases in value when it is away from the cyclone's center.The largest SLA value occurs near Java Island, with a value of 0.495 m at 943.610 m from the TC's center.This indicates constructive interference between storm surge waves and daily tides, causing higher tides.
The PWV and LWC profiles show a significant increase at -15 o latitude.This shows a huge increase in water vapor and liquid water content.The most significant water evaporation occurs at the cyclone walls [30].The PWV value reaches 10.760 cm with 143.165 m from the TC's center.At the same time, the largest LWC value is 2.08 kg/m2.The figure also shows that the increase and decrease in PWV are slower and gentler than LWC.
The SWH and WS profiles have the same characteristics: they generally increase as they approach the center of a TC.This shows the magnitude of the influence of the wind on the formation of sea waves when a TC occurs.The wind that blows on the water's surface will form waves in the water [31].When the wind speed increases, the waves will grow [32].The highest SWH value reached 4.649 m at a distance of 129.532 km from the TC's center.At the same time, the highest WS value occurs at a distance of 63.104 km at 18.42 m/s or 66.321 km/hour.
The WS profile shows fluctuations in value when approaching -15 o latitude.The WS of the altimeter is obtained using backscatter coefficient (sigma 0) data.Fluctuations in WS values around -15 o latitude indicate disturbances from the amount of rainfall in the area.Measurements on the Ku-band are greatly affected by rain, while measurements on the C-band are not so affected [33].The effect of rain on altimeter measurements can be explained in more detail by comparing the backscatter coefficient values of the Ku-band and C-band measurements, as seen in Figure 4.The backscatter coefficient on Ku-band will experience a significant decrease in value when measuring in cloudy areas and rainy conditions.This can also be seen from the Ku-band's backscatter coefficient, which generally has a smaller value than the backscatter coefficient of the C-band.The figure also shows a significant difference between the Ku-band's backscatter coefficient and the C-band's backscatter coefficient at latitude 14S o to 19S o .From this, it can be concluded that TC Seroja resulted in high rainfall in the area.

Temporal Profile
Processing of the temporal profile for the five variables was carried out at one sample point (-16.117o , 112.970 o ).This processing was carried out to compare the values of each variable from March to May 2021.Since the period for satellite altimetry to pass through the same point for the second time is 9.9156 days (10 days less 2 hours), there are 9 times to be compared.The temporal profile can be seen in Figure 5.This figure shows that the highest values of SWH, LWC, PWV, and WS occurred during TC Seroja.

Spatial Model
The spatial modeling of SLA, SWH, PWV, and WS during April 2021 was carried out by visualizing the five variables, which were divided into five timeframes, namely week-0, week-1, week-2, week-3, and week-4.Week-0 is used as a visual reference for week-1 to week-4.The data used in week-0 is data from 22 to 31 March 2021.The data used in week-1 is data from 1 to 7 April 2021.The data used in week-2 is data from 8 to 14 April 2021.The data used in week 3 is data from 15 to 21 April 2021.The data used in week 4 is data from 22 to 30 April 2021.The difference in the period results in a difference in the distribution of pass used in data processing each week, so the reduction process by week-0 data cannot be done directly.Therefore, an interpolation process is carried out to change the data distribution into a grid measuring 15' x 15'.The interpolation process is carried out using cubic interpolation or thirddegree polynomials.After the data has been spread evenly on a 15' x 15' grid, the process of reducing week-1 to week-4 data by week-0 can be carried out.center of the TC Seroja is dominated by class 4 with an SLA difference ranging from -0.13 m to 0.05 m.This shows a tiny change in SLA value in that area.Figure 5.b. shows the weekly model of SWH in April.The figure shows an increase in SWH values around the cyclone's center during week-1 and week-2.The rise in SWH in this area was not seen in week-3 and week-4.Thus, it can be concluded that the increase in SWH in the region during week-1 and week-2 is the result of TC Seroja.SWH values at week-1 and week-2 ranged from 1 to 3 meters higher than SWH at week-0.In week 2, it can be seen that the increase in SWH is not right at the center of TC Seroja but is to the west of the center of TC Seroja.This can happen because the satellite altimetry data obtained in the area is from pass 127 (asc) observed on 10 April 2021 and pass 166 (dsc) observed on 11 April 2021.Both passes are located to the west of the center of TC Seroja.The high SWH in this area is also caused by another TC, namely TC Odette.
The weekly PWV model is shown in Figure 5.c.Basically, the PWV value will be greater as it gets closer to the equator.This is due to the higher water temperature causing greater evaporation.The PWV value at the equator during week-0 ranged from 4.7 cm to 7.5 cm.Like SWH, the PWV value around the center of TC Seroja also increased during week-1 and week-2.In week-1, the PWV value increased to 2.3 cm higher than week-0.While the increase in PWV at week-2 has a value of more than 2.3 cm higher than week-0.This is indicated by areas with the highest class (red).
From the distribution of PWV in week-1 and week-2, it can be seen that the PWV value decreased significantly in the waters south of NTT.This can be seen from the color of the area.During week-1, that aresa has an orange color which indicates the PWV value of week-1 is higher than the PWV value of week-0.However, during week-2, that area is blue, which indicates the PWV value during week-2 in that region becomes smaller than the PWV value during week-0.This shows that the amount of water vapor content has decreased quite a lot after the occurrence of a TC.The decrease in water vapor can occur due to moisture divergence or because there has been a huge amount of precipitation before.
Similar to SWH, the magnitude of the PWV value in week 2 was also influenced by the presence of TC Odette.However, the effect of using data from pass 127 and pass 166 is not as big as that found in the SWH week-2.
The LWC model for each week in April 2021 can be seen in Figure 5.d.The image shows rain when TC Seroja occurs.However, the phenomenon of rain when a TC occurs cannot be observed as a whole.This is because the spatial and temporal resolution of the Jason-3 satellite is not good for rain observations.The phenomenon of rain only occurs in a small area.
The weekly WS model in April 2021 can be seen in Figure 5.e.Just like SWH and PWV, the figure shows an increase in WS values around the center of TC Seroja during week-1 and week-2.The decrease in WS values from week-1 to week-2 was also seen in the waters south of NTT.In addition, a decrease in the value of WS in the waters southwest of Australia was also seen from week 2 to week 3.This decrease in the value of WS again confirms the findings on the temporal profile.The WS distribution in the 2nd week shows an inappropriate increase in the center of TC Seroja.This is the same as the SWH distribution in week-2.One reason is the use of data from pass 127 and pass 166 which are west of the center of TC Seroja week-2.
The spatial model created can be used to determine the distribution of areas affected by TC Seroja.Even so, there are still things lacking in using satellite altimetry data in modeling phenomena with a short period and narrow coverage.For example, as seen in the SWH and WS week-2 models.The increase in the values of these two variables did not occur exactly at the center of TC Seroja, but to the west.This can happen due to limited data availability.In week 2, sample data when the TC Seroja occurred was obtained from pass 127, passed by the satellite on 10 April 2021 at 10.14 UTC and pass 166, passed by the satellite on 11 April 2021 at 23.05 UTC.The positions of the two passes can be seen in Figure 5.f.
The SWH and WS values in the two passes have a higher value than the other passes used in the interpolation.Because the interpolation method makes it impossible to make the maximum value at a point other than the sample point, the point with the highest value remains in the two passes.This is why the increase in SWH and WS in the 2nd week is not right at the center of TC Seroja.One effort that can IOP Publishing doi:10.1088/1755-1315/1245/1/0120369 be made to overcome this is to use satellite altimetry data with different missions.Different satellite missions have different orbits.

Conclusion
From the research, satellite altimetry can be used to observe TC.The spatial profiles of SLA, SWH, PWV, LWC, and WS that have been made can show that TC Seroja increases SWH, PWV, LWC, and WS in areas with latitudes close to the cyclone's center.Meanwhile, SLA decreased when it was close to the cyclone's center and increased when it was away from the cyclone's center up to a certain distance.The highest SLA was 0.495 m, SWH was 4.649 m, PWV was 0.10760 m, LWC was 2.08 kg/m2, and WS was 18.42 m/s.
The temporal profile from March to May showed that the SWH, PWV, LWC, and WS values when TC Seroja occurred were the highest.In addition, PWV, LWC, and WS also show a significant decrease in value up to 10 days after TC Seroja.This shows that the amount of water vapor content has decreased quite a lot after the occurrence of a TC.The decrease in water vapor can occur due to moisture divergence or because there has been a huge amount of precipitation before.
The obtained SLA spatial model does not show the influence of TC Seroja.The SWH, PWV, LWC, and WS spatial models show the influence of TC Seroja on these variables.This effect can be seen by increasing the values of the four variables near the center of TC Seroja in week-1 and week-2.However, the spatial and temporal resolution of the Jason-3 satellite in carrying out data collection causes several deficiencies in the resulting spatial model so that it does not represent the actual conditions.In order for TC observations to provide better results, a combination of satellite altimetry data can be performed with other data that have a higher spatial and temporal resolution.

Figure 1
Figure 1 Movement of TC Seroja's center

Figure 2
Figure 2 Ascending (black) and descending (blue) passes of Jason-3 in the research area

Figure 3
Figure 3 Spatial profile of SLA, SWH, PWV, LWC, and WS (from left to right) when TC Seroja occur on April 7 th , 2021

Figure 4
Figure 4 Backscatter coefficient of Ku-band and C-band

Figure 5
Figure 5 Temporal profile

Figure 6
Figure 6 Spatial model a) SLA, b) SWH, c) PWV, d) LWC, e) WS, f) Pass 127 and 166 Data processing is continued by performing IDW interpolation to create a spatial model for each variable.The SLA model every week can be seen in Figure 5.a.The figure shows that TC Seroja only occurs in week-1 and week-2.The figure shows a visualization of the difference in SLA from week 1 to 4 to week 0. From this figure, the influence of TC Seroja on the SLA is unclear.The area around the