Coastline Change Analysis using Landsat-8 on Rebana Metropolitan Area, West Java

The Rebana metropolitan area is an industrial area development covering Cirebon-Patimban-Kertajati to support accelerated economic growth, especially in the West Java region. In order to minimize failures in supporting the development targets of the Rebana metropolitan area, it is necessary to carry out spatial planning with consideration of the hazard aspects in the development area. One of the aspects that need to be studied further is the dynamics of the coastal area such as coastline changes that occur due to erosion or sedimentation. Analysis of erosion and sedimentation is essential for further studies to be carried out regarding the factors that cause erosion and sedimentation and how to protect and prevent them in the future. This research utilizes remote sensing techniques and integrated geographic information systems (GIS) to identify coastline changes from 2017-2022 in the coastal areas of Subang, Indramayu, and Cirebon districts which are included in the Rebana metropolitan area. In this study, Landsat-8 satellite imagery is used to identify and quantify Spatio-temporal changes that occur in the coastal area of the Rebana metropolitan area in 6 years. The Normalized Difference Wetness Index (NDWI) algorithm is used to separate wet and dry objects which are interpreted as bodies of water and land. Thresholding the results of the NDWI algorithm along with the on-screen digitization approach was carried out for coastline extraction. The calculation of coastline changes is carried out using the Digital Shoreline Analysis System (DSAS) so that the rate of change of coastline in the study area is obtained. During the study period, the results showed that the average erosion rate in the Rebana metropolitan area was -3.66 m/year and the average sedimentation rate was +6.28 m/year, where the greatest erosion rate occurred in Kapetakan sub-district, Cirebon Regency with an average rate -7.63 m/year, and the greatest sedimentation rate occurs in Pusakanagara sub-district, Subang Regency with an average rate +24.64 m/year. Karangampel Sub-district in Indramayu Regency has 79.89%, the widest erosion area of its total coastline length, and Balongan Subdistrict, Indramayu Regency has the widest sedimentation, 91.37% of the total coastline length. The land use that is most affected by erosion and accretion in the coastal area of the Rebana metropolitan area is fishpond.


Introduction
Every year there is an increase in population in Indonesia [1].Demand will increase as the population rise, makes growth in the industrial sector must be accompanied by population growth to meet supply [2].The industrial sector also plays a vital role as the prime mover and supporter of the national economy [3].
However, industrial planning requires careful planning to carry out industrial development strategically.Several parameters must be considered in industrial development, including socio-economic, physicalenvironmental, infrastructure and urban development, and potential hazards [4][5][6].Some hazard parameters that need to be reviewed from industrial development are erosion and accretion, especially for industrial planning in coastal area [7].
One area of Indonesia that has the potential for a large increase in industrial in the coastal area is Rebana Metropolitan.Rebana Metropolitan is the north/northeast region of West Java Province which consists of seven cities in which there are Patimban Port, Cirebon Port and Kertajati West Java International Airport (BIJB), which functions as a connectivity and logistics center.The strategic area of the Rebana Metropolite can support the increase in the industrial economy [8].
However, the Rebana metropolitan area, which is located in the lowlands in northern Java, has considerable potential for coastline dynamics.Areas dominated by sedimentary mud are very easy to change position and cause changes in the shape of the beach [9], [10].Changes in the coastal area can be seen from the description of the differences in the coastline.This change is indicated to have occurred due to erosion and accretion along the coast which can be known through remote sensing imagery in the form of satellite imagery, namely in 2017 and 2022 [11], [12] .The results of remote sensing data analysis are then processed with a geographic information system to analyze coastal environmental conditions.
Aim of this research is to determine changes in the coastline in the coastal Rebana Metropolitan area which is estimated to become an industrial center.Coastal land management is analyzed to see the impact and causes of changes in the coastline that occur.It is hoped that this study will identify the factors that cause erosion and sedimentation as well as ways to protect and prevent them so that economic development in Metropolitan Rebana can be sustainable.

Study Area
The area study in this research focuses on the coastal area of the Rebana Metropolitan area, which is in the province of West Java.The case study area comprises 20 sub-districts in three districts adjacent to the coastline along the Rebana Metropolitan Area [8].The study area can be seen in Figure 1.

Data
Data used in this study can be seen in Table 1.In general, this study uses five data consisting of administrative data, Ocean tide data, Landsat-8, Digital Elevation Model (DEM), and Landuse.

Methods
The general methods in this study can be seen in Figure 2. The method can be divided into four sections that will be explained below.The radiometric correction process can be carried out using various applications, one of which was carried out in this study using QGIS software.The radiometric correction equation used for Landsat is used in this study [16].the equation for radiometric correction is in equation 1 below.
= Reflectance value (which has been corrected for the angle of the sun).= Rescaling constant.= Pixel Value (Digital Number).A= Addition Constant.
se= The sun's elevation angle during image recording.

Coastline Extractions a) Normalized Difference Water Index (NDWI)
The Normalized Difference Water Index (NDWI) method is a method which is used to compare humidity degrees on satellite imagery.The NDWI method uses Band 3 (green) to assess the strength of plants and water bodies, and Band 5 (NIR) emphasizes the amount of biomass The NDWI index results can range from -1 to +1.The equation for calculating NDWI is in equation 2 below.The selection of this wavelength is done for: 1) maximizing the reflection of water features by using green wavelengths; 2) minimizing low NIR reflectance by water features; And 3) take advantage of the high NIR reflectance by terrestrial vegetation and soil features.

Green+NIR
(2) When equation ( 2) is used to process multispectral satellite imagery, the water feature has a positive value.While ground and terrestrial vegetation features have a zero or negative value, because the NIR reflectance is usually higher than the green channel [17].The equation used for using the NDWI algorithm can be seen in equation 2 [18].the results of the NDWI algorithm can be seen in Figure 3b.

b) Thresholding
The results of processing using the NDWI algorithm produce spectral values between -1 to 1, which indicate the degree of wetting in the image processing results.Wetness classification using NDWI can be seen in Tabel 3 below [19].Thresholding aims to make boundaries based on the classification in the table above so that boundaries between dry land and wetlands which in this study are called land and sea will be obtained.land classification is at a value of -1 to 0, while waters is a value of 0 to 1 [19].The thresholding results for NDWI processing based on the classification in table 2 can be seen in Figure 4 below.

Coastline Change Analysis using the Digital Shoreline Analysis System (DSAS)
The stages of analysis using DSAS are establishment of a coastline, making baselines, transect analysis, and calculate the coastline change.The coastline change rate was analyzed using the End Point Rate (EPR) statistical approach using the Digital Shoreline Analysis System (DSAS) tool.The EPR method calculates the rate of change of coastline by dividing the distance between the oldest and the most recent coastline by the time.EPR calculations are performed on each transect that is formed with a distance of 15 m between transects.the distance between transects is chosen based on half of the pixel size of the Landsat 8 image.Figure 6 is an illustration of the results of processing using DSAS.

Erosion Hazard Classification
The classification of coastal erosion hazard is determined based on the Coastal Vulnerability Index (CVI), where erosion and accretion hazards are divided into five classes that can be seen in Table 4.The danger of coastal erosion will be greater if the coastline changes towards the land and lower if there is a change in the coastline towards the sea (accretion).

Result and Discussion
The results of this study are divided into two, an analysis of coastline changes in the Rebana metropolitan area and the relationship between land cover in the Rebana metropolitan area and phenomena that occur along the coast.Clearly, the analysis of the results of this study is explained as follows.

coastline dynamics
The dynamics of the coastline in the Rebana metropolitan area, divided into 20 sub-districts, is obtained using DSAS.The baseline is used as a reference in calculating the length of the coastline, which is taken based on the coastline on the results of the 2017 coastline extraction as the first data in this study.The distance between transects along the coastline is 15 m, with the number of transects attached in the Table 5.
Table 5.The results of making transects along the coastline From the results of coastline extraction, the sub-district with the longest coastline was obtained, namely the Pasekan sub-district, Indramayu Regency, which is 12.69% of the total coastline length in the Rebana metropolitan area.Pasekan sub-district obtained longest coastline affected by erosion, which is 931 m long.Meanwhile, Cantigi sub-district obtained coastline most affected by accretion.From the results of processing using DSAS, the average coastline change rate is obtained from the calculation of the EPR, which can be seen in the Figure 7.The results of the average EPR calculation show that the district with the largest average erosion rate is experienced by Pasekan District, Indramayu Regency, which is equal to 8.97 m/year.Meanwhile, the district with the highest average accretion rate was Pusakanagara District, Subang Regency with a maximum value of 24.64 m3/year.Pasekan District experienced the greatest erosion phenomenon with the majority of the coastal area being used as fish ponds.The lack of mangroves and water barriers, coupled with the sloping topography with sandy mud, causes continous erosion.Meanwhile, the Pusakanagara area experienced the largest average accretion because it has many upstream rivers in its area that carry sediment particles from the river to its coastal areas.The existence of mangroves in the coastal area also plays a role in binding sediments, causing an increase in land area in the Pusakanagara sub-district.

relationship between landuse and erosion hazardcoastline dynamics
The results of the calculation of coastline changes using DSAS are then carried out to classify the erosion hazard based on the Coastal Vulnerability Index (CVI), which is divided into five classes.The results of the erosion hazard classification in the coastal area of the Rebana metropolitan area can be seen based on the Figure 9 below.From Table 6, the results of land cover with a high erosion hazard impact (class 5) have an impact on the fish pond area of 28.37% or 1164252.77m along the coastline.While land use with high accretion hazard impacts (class 1) also has an impact on the fish pond area of 35.40% or 4104386.87m along the coastline.This is probably caused by the slope of the land in which the land use is relatively flat.Not only is there a natural accretion phenomenon in the downstream area of the river and fish ponds, human activities that take advantage of the accretion phenomenon in the fish pond area, such as adding new fish ponds from the results of sedimentation in the surrounding area are also one of the causes of rapid coastline changes in fish pond area.

Conclussion
The conclusions obtained from the results of this study are as follows.
1.During the study period, the results showed that the average erosion rate in the Rebana metropolitan area was -3.66 m/year and the average sedimentation rate was +6.28 m/year, where the greatest erosion rate occurred in Pasekan District, Indramayu Regency with an average rate 8.97 m/year m/year, and the greatest sedimentation rate occurs in Pusakanagara sub-district, Subang Regency with an average rate +24.64 m/year.Karangampel Subdistrict in Indramayu Regency has 79.89%, the widest erosion area of its total subdistrict coastline length, and Balongan Subdistrict, Indramayu Regency has the widest sedimentation, 91.37% of the total sub district coastline length.2. The land use most affected by erosion and accretion in the coastal area of the Rebana metropolitan area is ponds, which occur due to natural phenomena and the result of human activities.3. The phenomenon of erosion and accretion that occurs in this study cannot be explained with certainty whether caused by natural factors or human factors, so additional parameters are needed to ascertain the causes of erosion and accretion such as hydrological morphodynamic models.4. Detected coastline changes of less than 15 m are still ambiguous whether they occur due to errors in digitizing or are indeed shoreline changes, so for further research it is necessary to study shoreline changes using high-resolution imagery.

Figure 2 .
Figure 2. Methodology Flowchart2.3.1.Radiometric CorrectionRadiometric correction is intended to eliminate or minimize radiometric errors due to external aspects in the form of atmospheric disturbances during the recording process.[15].This process also converts

Figure 6 .
Figure 6.Example of coastline dynamics in Subang Regency hazard classes are divided into 5 classes, namely very low with a rate of above 2 m/year, low class with a rate of 1.0 to 2.0 m/year, moderate class with a rate of 1.0 to -1.0, high class with a rate of -1.1 to 2.0 m/year and very high class with a rate of more than -2.0 m/year.A positive value indicates a change in the coastline towards the sea (increasing land area) while a negative value indicates a change in the coastline towards land (decreasing land area).

Figure 8 .
Figure 8. Erosion Hazard Classification in Rebana Metropolitan Area

Table 1 . Data used in this study No Data Product Spatial Resolution Referensce 1
The choice of data acquisition date in Table2is based on the availability of data each year and the quality of satellite data as seen from the percentage of cloud cover in coastal areas.

Table 2 .
Data and quality of satellite imagery selected in the

Table 3 .
Wetness level based on NDWI value

Table 6 below . Table 6 .
Hazard classification in percent