https://doi.org/10.19184/geosi.v7i3.30797 Research Article The Application of Multi Temporal Satellite Data For Land Cover Mapping of Bawean Island, East Java Zainul Hidayah1,*, Haryo Dwito Armono2 , Maulinna Kusumo Wardhani1, Dwi Budi Wiyanto3 1 Department of Marine Science and Fisheries, Trunojoyo University of Madura, Jl. Raya Telang 02 Kamal Bangkalan Madura 69162 East Java, Indonesia 2 Department of Marine Engineering, Sepuluh Nopember Institute of Technology, Jl. Raya ITS Keputih Sukolilo Surabaya 60111 East Java, Indonesia 3 Department of Marine Science, Udayana University, Bukit Jimbaran Campus 80361 Bali, Indonesia *Corresponding author, Email address: zainulhidayah@trunojoyo.ac.id 1. Introduction Bawean island is located in the Java Sea, around 80 nautical miles north of the mainland of Gresik Regency, East Java Province. Due to its beauty and preservation of coastal and underwater nature, this island has been designated as one of the East Java's marine tourist destinations (Hidayah et al., 2021). As a result, the land and water conditions on Bawean Island must be thoroughly studied in order to support the regional development plan. Several studies have been conducted in recent years, in addition to the designation of Bawean Island and its surrounding areas as a marine ecotourism zone (Noor & Romadhon, 2020). Moreover, several key factors have been identified as major issues that may influence the eco-tourism development of the island, incorporating infrastructures, economics, environment, and socio-cultural issues (Wardani et al., 2017b, Sukandar et al, 2017). Meanwhile, the Geosfera Indonesia p-ISSN 2598-9723, e-ISSN 2614-8528 available online at : https://jurnal.unej.ac.id/index.php/GEOSI Vol. 7 No. 3, December 2022, 207-218 ABSTRACT Land cover dynamics in a small island can be determined using Geographic Information System (GIS) approach based on multitemporal image analysis. This study aimed to classify major land cover types and to map land cover changes of Bawean Island. Two sets of 10 meter resolution satellite data ALOS AVNIR (2010) and Sentinel-2A (2020) were used in this study. Satellite image analysis was carried out through several stages namely image pre-processing including radiometric and geometric correction, supervised image classification and accuracy test. Image classification results from 2010 to 2020 showed a significant change in land cover on Bawean Island. The forest vegetation land cover declined significantly from 13,470.5 Ha in 2010 to 8,543.4 Ha in 2020. Most of the area have been converted into paddy fields and built-up areas. The accuracy test and validation were determined by comparing the 2020 Sentinel image classification results with field observation conducted in 2021. The analysis showed good results with 82.52% overall accuracy and 79.66 Kappa coefficient. Further investigation found that changes in land cover on Bawean Island occured due to the agriculture and infrastucture development. Keywords : Geographic Information System (GIS); land cover; satellite images; small island ARTICLE INFO Article history Received : 4 July 2022 Revised : 19 November 2022 Accepted : 3 December 2022 Published : 24 December 2022 . © 2022 by Geosfera Indonesia and Department of Geography Education, University of Jember. This is an open access article under the CC-BY-SA license (https://creativecommons.org/licenses/by-sa/4.0/) 207 https://doi.org/10.19184/geosi.v7i3.%2030797 mailto:zainulhidayah@trunojoyo.ac.id https://orcid.org/0000-0002-5215-9764 https://jurnal.unej.ac.id/index.php/GEOSI https://creativecommons.org/licenses/by-sa/4.0/ 208 Zainul Hidayah et al. / Geosfera Indonesia 7 (3), 2022, 207-218 previously mentioned studies are more focused on analyzing the condition of the coastal area to support marine tourism, with another study in focusing more on the hydro-oceanographic conditions of Bawean Island (Hidayah, et al., 2021). Studies on the condition of Indonesia's small islands, particularly those with tourism and conservation potential, have been widely documented (Marasabessy et al., 2016; Santoso et al., 2020; Suryanti & Apdillah, 2022). This is integrally linked to the fact that Indonesia has thousands of small islands scattered across the country. Previous studies have explored the condition of Bawean Island particularly its coastal and marine environment, including biological diversity (Asaad et al., 2018), small island ecosystems including mangroves (Muhtadi et al., 2020), seagrass (Brodie et al., 2020), coral reefs (Hafezi et al., 2020) and vulnerability to sea-level rise as the results of global climate change (Hidayah et al., 2018; Stephenson & Jones, 2017; Jamero et al., 2017). While many studies focussed on the state of the coastal environment and ecosystems of Bawean Island, there was no studies have been reported regarding the island’s land cover dynamic. Small island's surface and bathymetry are becoming increasingly important sources of information for a variety of purposes, including spatial planning, coastal management, infrastructure development and conservation (Caballero et al., 2019). Land cover changes and dynamics are primarily caused by anthropogenic and natural factors. If it is not addressed and mitigated carefully, it will have an impact on human life and disrupt the sustainability of the environment, particularly on small islands with limited natural carrying capacity (Phiri et al., 2020). As a result, advanced and cutting-edge spatial monitoring systems are required. The most effective method for providing surface data of an area within a specific time frame is to use satellite imagery with various characteristics (Ketjulan et al., 2019; Muhammad, 2022; Wael et al., 2022). The rapid development of satellite remote sensing technology and geographic information systems for the small island is the primary reason for Indonesia's rapid development of spatial data availability (Dharmawan et al., 2020; Hidayah et al., 2019). Multiple studies on Bawean Island have been conducted in the last five years, but the majority of them have focused on the condition of the coastal environment, conservation, and suitability for eco- tourism activities (Hidayah et al., 2018; Sukandar et al., 2017; Wardani et al., 2017b). Because of the abundance of coral reefs surrounding the island, more detailed studies have been conducted to determine the state of the ecosystem (Dharmawan et al., 2020; Luthfi & Anugrah, 2017). Periodic assessment of land cover change is required to understand the extent and impact of natural and human activities factors on the island (Cahyani et al., 2018; Kim, 2016; Neksidin et al., 2021). Additionally, the endemic and critically endangered species of Bawean deer (Axis kuhlii) was also an interesting topic of investigation regarding this island (Rahman et al., 2017). Meanwhile, this study provides a new perspective on the changes in the island's condition based on satellite imagery analysis. The objectives of this paper is to observe land cover alterations using satellite imagery, particularly those related to land conversion, from 2010 to 2020. 2. Methods 2.1 Study Area This research was conducted on Bawean Island from July – October 2021. Bawean island is located roughly halfway between Java Island and Borneo Island. Because of its remote location and distance from the mainland, the island has some rare and distinctive flora and fauna. The coordinates of the island is 05o47’23’’S and 112o39’20’’E (Figure 1). The total area of Bawean island is ± 196.27 km2 divided administratively into 2 districts namely Sangkapura in the south and Tambak on the north side of the island. 209 Zainul Hidayah et al. / Geosfera Indonesia 7 (3), 2022, 207-218 Figure 1. The map of Bawean Island (Source : Modified from Google Earth) This study's ROI includes the entire land surface of Bawean Island and the surrounding small islands. The land cover category of this study was divided into 6 main classes, namely forests, sandy beaches, settlements, rice fields, inland water body/ lake and open field. The classification results were then compared to the results of field observations to obtain an accuracy value. A total of 103 observation points were randomly established to check the consistency of the land cover classification results and the actual condition in the field. 2.2 Satellite Images To identify land cover changes of Bawean island, this study utilized two sets of medium resolution satellite images namely ALOS AVNIR-2 acquisition date July 15th 2010 and Sentinel-2A acquisition date August 23rd 2020. The percentage of cloud cover for both images were less then 5%. The Advanced Land Observing Satellite (ALOS) was launched in January 2006 by JAXA (Japan Aerospace Exploration Agency). This satellite has several distinctive sensors i.e (1) PRISM (Panchromatic Remote-sensing Instrument for Stereo Mapping) high precision elevation mapping; (2) AVNIR-2 (Advanced Visible and Near Infrared Radiometer type 2) for land use and land cover observation; (3) PALSAR (Phased Array type L-band Synthetic Aperture Radar) an active sencor to monitors the land in any weather conditons. To be more spesific, the AVNIR-2 sensor has 4 visible bands i.e blue (0.42-0.50 μm), green (0.52-0.60 μm), red (0.61-0.69 μm) and near infra-red (0.76-0.89 μm). Meanwhile, the Sentinel 2A and 2B satellites were consecutively launched between June 2015 and March 2017 by the Copernicus Programme initiated by the European Space Agency (ESA). The satellites carry multispectral imaging (MSI) instruments capable of recording 13 wide-swath bands, 210 Zainul Hidayah et al. / Geosfera Indonesia 7 (3), 2022, 207-218 including the 10-meter bands (blue, green, red and visible near infra-red). Sentinel 2 satellite imagery has been widely used in recent years to monitor the condition of the earth's surface, particularly in coastal areas (Artaningh et al., 2020; Kurniadin & Fadlin, 2021; Mastu et al., 2018). With higher spatial resolution in visible bands (10 meters) compare to Landsat images (30 meters), those images have been used to produce maps not only for the land surface but also sea surface and benthic environments in higher detail (Li et al., 2019; Wicaksono et al., 2019). 2.3 Image Classification and Change Detection Land cover classes were generated from pixel values in satellite images using the supervised classification method. In supervised classification, analysts must select representative samples for each land cover class. The software then used the maximum likelihood algorithm to process these samples or training sites and applied them to the entire image. By comparing classification results based on satellite imagery from 2010 and 2020, changes in the area of each land cover class were calculated (Figure 2). 2.4 Accuracy Assesment The precision of the classification results was calculated using the two dimensional array. The confusion matrix compares the classification results and the reference data. In order to obtain appropriate results, comparisons were made using the results of image classification in 2020, while ground truth was carried out in 2021. The confusion matrix yields the overall accuracy and Kappa coefficient (Lillesand & Kiefer, 1994). The Kappa coefficient value is divided into 4 classes (Congalton and Green, 2008), namely Kappa < 0.4 (low accuracy), 0.41< Kappa < 0.61 (medium accuracy), 0.61< Kappa < 0.80 (substantial accuracy) and Kappa > 0.81 (high accuracy). Figure 2. Flow chart of image processing 211 Zainul Hidayah et al. / Geosfera Indonesia 7 (3), 2022, 207-218 3. Results and Discussion Bawean Island has a tropical climate, as do most of Indonesia's coastal areas. The mean daily temperature ranges from 28o C – 31o C with an average of 28.50 o C ± 1.05 o C. The air temperature in the dry season is slightly higher than the temperature in the rainy season (Figure 3). However, there is significant difference in the average daily temperature for each month (One Way ANOVA; Fcalculate=6.82; sig <0.05). The highest rainfall occurs at the peak of the rainy season, from December to February. In 2020, the highest rainfall reached 85.57 mm to 95.1 mm occurred in January - February. Meanwhile, during the dry season there is no rain at all. The relative humidity on Bawean Island is quite high with an average of 81.40% 5.53%. The difference in air humidity every month is fairly significant (One Way ANOVA; Fcalculate=21.7; sig <0.05). Figure 3. Temperature, Humidity and Rainfall of Bawean Island 2021 The map in Figure 4 shows the results of land cover classification based on satellite imagery of ALOS (2010) and Sentinel 2A (2020). It can be seen that most of the land cover on Bawean Island is forest/vegetation cover which is distributed mostly in the middle of the island. The forest types on this island are lowland primary and secondary tropical rainforests (Trimanto & Hapsari, 2016). Forests on Bawean Island have been designated as nature reserves and wildlife reserves with the dominance of teak forests (Tectona grandis) (Rahman et al., 2017). Moreover, a botanical survey conducted in 2015- 2016 have identified 432 species of plants from 287 genera and 103 families in the forests of Bawean Island at elevations ranging from 8 to 572 meters above sea level (Trimanto & Hapsari, 2016). The 212 Zainul Hidayah et al. / Geosfera Indonesia 7 (3), 2022, 207-218 natural forest condition on Bawean Island support the life of various plant species, therefore the island has a considerably high plant diversity. Figure 4. Land cover classification based on satellite images classification : (A) Results of ALOS- AVNIR 2010 image classification; (B) Results of Sentinel 2A 2020 image classification Results of land cover classification based on multi temporal satellite images (2010 and 2020) of Bawean Island is shown in Figure 4. Six land cover classes were identified, namely forest, built up area, paddy field (agriculture area), open field, sandy beaches and in-land water body from Lake Kastoba. In 2010 forest dominated the island’s land cover comprising approximately 56.6% (13,470.5 Ha) of the total area. Sandy beaches and paddy field were the next dominant classes, covered about 16.2 % (3,851.2 Ha) and 16.1 % (3,816.8 Ha) respectively. Meanwhile there was only about 863.9 Ha or 3.6% of the island’s classified as built up area and mostly concentrated on the coastal area. Figure 5. Land cover area change of Bawean Island 2010-2020 213 Zainul Hidayah et al. / Geosfera Indonesia 7 (3), 2022, 207-218 Significant changes occurred of land cover in 2020 compared to the previous 10 years (Figure 5). Although forest cover remained dominant on the island (35,9%), its area experienced a sharp decline. Around 4,927.1 Ha of forest has been converted into other types of land cover in the last 10 years. Rice fields and sandy beaches occupy the next position with around 30.2 % (7,186.7 Ha) and 16.5% (3,935.3 Ha) of land cover area. The development of paddy fields was rapid and has doubled in 10 years. The built-up area expanded from the coastal area to the center of the island with the percentage reached nearly 10% (2,388.1 Ha) of the total island area. Table 1. The dynamic of land cover change of Bawean Island (2010-2020) Land Cover Class Land Cover Area (%) Changes (%) 2010 2020 2010-2020 Forest Vegetation 56.59 35.89 -20.70 Paddy Field 16.03 30.19 14.16 Lake 0.14 0.16 0.03 Open Field 7.43 7.19 -0.24 Built Up Area 3.63 10.03 6.40 Sandy Beach 16.18 16.53 0.35 The temporal analysis revealed that the most significant changes occurred in the land cover classes of forest vegetation, paddy fields, and built-up areas (Table 1). In the 2010-2010 period, the area of forest vegetation decreased by 20.70%, while paddy field cover increased by 14.6%. The built-up area also experienced a significant increase in land cover area, which was around 6.40%. Further investigation revealed that 70.91% of the total lost forest was converted into paddy fields, while 20.33% was converted into developed areas. In addition, based on a rigorous spatial review the reduction in forest area occurs primarily in the lowlands. Meanwhile, primary forest in the highlands in the island's center is relatively undisturbed. As a result, forest loss can be justified as a result of agricultural and settlement needs rather than illegal logging. However, the rate of deforestation is relatively high (around 400 Ha/year) and must be controlled in order to preserve the remaining forest. Table 2. Confusion matrix of classification Classification Results (Pixel) Ground Truth Data (Pixel) Lake Open Field Sandy Beaches Built Up Area Paddy Field Forest Veg Total Lake 10 2 0 0 0 0 12 Open Field 0 14 0 2 0 0 16 Sandy Beaches 0 0 14 0 0 0 14 Built Up Area 1 2 0 15 0 0 18 Paddy Field 0 1 1 0 17 3 22 Forest Veg 0 0 1 2 3 15 21 Total 11 19 16 19 20 18 103 The confusion matrix is the most common method used to quantify classification accuracy. A confusion matrix compares the relationship between known reference data (ground truth) and the corresponding classification procedure results on a class-by-class basis. The values along the diagonal of the table represent the correct pixel classification based on image analysis and field observations. 214 Zainul Hidayah et al. / Geosfera Indonesia 7 (3), 2022, 207-218 According to the confusion matrix (Table 2), the majority of the pixels tested have a match between the image classification results and the actual land cover conditions on Bawean Island. Table 3. Statistical parameters of classification Classes User's Accuracy (%) Producer's Accuracy (%) Error of Comission (%) Error of Omission (%) Overall Accuracy (%) Kappa Coefficient Lake 83.33 90.91 16.67 9.09 82.52 79.66 Open Field 87.50 73.68 12.50 26.32 Sandy Beaches 100.00 87.50 0.00 12.50 Built Up Area 83.33 78.95 16.67 10.53 Paddy Field 77.27 85.00 22.73 15.00 Forest 71.43 83.33 28.57 16.67 Average ± Stdev 83.81 ± 9.72 83.23 ± 6.16 16.19 ± 9.72 15.02 ± 6.20 The calculation of accuracy of image classification is explained in Table 3. User’s accuracy refers how actually classified map is real on the ground. The average of user’s accuracy of this study is considerably high with percentage of 83.81% ± 9.72%. Sandy beaches land cover class has the highest invidual percentage of user’s accuracy, while paddy field and forest vegetation have the lowest percentage with 77.27% and 71.43% respectively, with error of comission reached more than 20%. These findings indicate that there are significant differences between the classification results of paddy fields and forest vegetation and the actual land cover. Furthermore, the spectral signatures of paddy field cover and vegetation are relatively similar therefore they are difficult to distinguish. This is particularly the case of pixels that are located adjacent to each other. Furthermore, the producer's accuracy represents class-wise accuracies from the map maker's perspective. As a result, it provides a probability that a specific sample of a specific land cover is corresponding as the same class in the classification map. Based on the calculation, the average of producer’s accuracy of this study is fairly high in the range of 83.23% ± 6.16% with the average of omission error around 15% (Table 3). These results confirm the analyst's ability to classify land cover classes based on the training area created according the spectral appearance on satellite imagery. The overall accuracy of 82.52% indicates that the results of land cover classification in this study are considerably good. The total accuracy value obtained in this study is not significantly different from the classification results obtained in several research studies, which are in the range of 80% or higher, despite the use of images with lower resolutions such as Landsat 8 with spatial resolution of 30 meters (Derajat et al., 2020; Rini, 2018; Siregar & Asbi, 2020; Wulansari, 2017). Moreover, Kappa coefficient value produced by the confusion matrix is 79.66 and categorized as substantial (61.00 < Kappa < 80.00). Previous research suggests combining the Kappa coefficient and the overall accuracy value to validate land cover classification results (Wulansari, 2017). In contrast to overall accuracy, which is calculated by dividing the number of diagonal values by the total number of cells in the matrix, the Kappa coefficient considers non-diagonal elements (Marlina, 2022; Putri et al., 2018). In addition, the Kappa coefficient measures the difference between the suitability of the classified data and the probability of a random classification match being compared with the reference data. Several factors influence land use/land cover change in a limited environment such as a small island. The primary factor is population growth, which raises the demand for housing and food (Kim, 2016; Purbani et al., 2020). As a result, land clearing is carried out in order to provide settlements and agricultural land. The decline in fisheries production due to overfishing or climate change may also be a driving factor in small island land changes (Pinuji et al., 2018). Bawean Island is a small island with abundant resources (Wardani et al., 2017a). Coastal communities depend on fishery products, while people who live inland depends on agriculture. If the decline of fisheries production happen for a long 215 Zainul Hidayah et al. / Geosfera Indonesia 7 (3), 2022, 207-218 period, it is possible that there will be a shift in the profession of fishermen in to farmers (Pinuji et al., 2018). As a result, more agricultural land must be provided. Furthermore, the lack of land use regulation based on regional spatial planning is a contributing factor to uncontrolled land use changes that endanger environmental sustainability. Bawean Island provide a distinct model for nature conservation that differs from the mainland in general. Economic and social development are both major concerns in areas with small economies that rely heavily on external markets, high transportation costs, and small populations. As a result, a distinctive strategy is required to balance the development of small islands without risking their natural sustainability. The presence of diverse ecosystems on Bawean Island is the main draw for visitors. However, due to the island's limited carrying capacity, tourism activities on the island must be carefully managed. Economic and population growth will result in the development of new lands. The major concern should be the balance between built-up areas to support development and conservation areas to maintain the island’s natural biodiversity. 4. Conclusion This study has demonstrated the use of satellite images to identify land cover changes on a small island. The results of image classification in 2010 - 2020 show a significant change of land cover on Bawean Island. Forest vegetation land cover experienced a significant degradation from 13,470.5 Ha in 2010 to 8,543.4 Ha in 2020. On the other hand, there is an increase of built-up area and paddy fields in the same period. Changes in land cover on Bawean Island due to the population growth and infrastuctures construction. The rapid development on the island requires larger residential and agricultural areas to be provided. The development on Bawean Island must be regulated based on regional spatial planning in order to provide optimal results for the welfare of the community while at the same time preserving its rich natural diversity and ecosystems. Conflicts of Interest The authors are not involved in a conflict of interest from funds, personal and institutional or any other relationships from this article. References Artaningh, F., Septi, T., Sihotang, E., & Dimara, A. (2020). Analisis data sentinel-2 untuk mendukung pariwisata Kawasan Wakatobi. Jurnal Penginderaan Jauh Indonesia, 02(02), 25–31. Asaad, I., Lundquist, C. J., Erdmann, M. V., & Costello, M. J. (2018). Delineating priority areas for marine biodiversity conservation in the Coral Triangle. Biological Conservation, 222, 198–211. https://doi.org/10.1016/j.biocon.2018.03.037. Brodie, G., Holland, E., N’Yeurt, A. D. R., Soapi, K., & Hills, J. (2020). Seagrasses and seagrass habitats in Pacific small island developing states: Potential loss of benefits via human disturbance and climate change. Marine Pollution Bulletin, 160(May 2019), 111573. https://doi.org/10.1016/j.marpolbul.2020.111573. Caballero, I., Stumpf, R., & Meredith, A. (2019). Preliminary assessment of turbidity and chlorophyll impact on bathymetry derived from Sentinel-2A and Sentinel-3A Satellites in South Florida. Remote Sensing, 11(6), 645. https://doi.org/10.3390/rs11060645. Cahyani, W. S., Setyobudiandi, I., & Affandy, R. (2018). Kondisi dan status keberlanjutan ekosistem terumbu karang di kawasan konservasi perairan Pulo Pasi Gusung, Selayar. Jurnal Ilmu Dan Teknologi Kelautan Tropis, 10(1), 153–166. https://doi.org/10.29244/jitkt.v10i1.21672. Congalton, R.G & Green, K. 2008. Assessing the accuracy of remotely sensed data : principles and practises. Third Edition. CRC Press. Taylor Francis Group. 216 Zainul Hidayah et al. / Geosfera Indonesia 7 (3), 2022, 207-218 Derajat, R. M., Sopariah, Y., Aprilianti, S., Candra Taruna, A., Rahmawan Tisna, H. A., Ridwana, R., & Sugandi, D. (2020). Klasifikasi tutupan lahan menggunakan Citra Landsat 8 Operational Land Imager (OLI) di Kecamatan Pangandaran. Jurnal Samudra Geografi, 3(1), 1–10. https://doi.org/10.33059/jsg.v3i1.1985. Dharmawan, A., Rahmawan, A., & Ashoffi, A. M. (2020). Identification of damage and key species of coral reefs on Gili-noko Bawean Island. AIP Conference Proceedings, 2231, 1–5. https://doi.org/10.1063/5.0002579. Hafezi, M., Sahin, O., Stewart, R. A., Connolly, R. M., Mackey, B., & Ware, D. (2020). Adaptation strategies for coral reef ecosystems in Small Island Developing States: Integrated modelling of local pressures and long-term climate changes. Journal of Cleaner Production, 253, 119864. https://doi.org/10.1016/j.jclepro.2019.119864. Hidayah, Z., Wiyanto, D.B & Romadhon, A. (2021). The assessment of benthic ecosystems and substrate profile of Bawean Island East Java Province. IOP Conference Series: Earth and Environmental Science, 860(1). https://doi.org/10.1088/1755-1315/860/1/012003. Hidayah, Z., Prayogo, L. M & Wardhani, M. K. (2018). Sea level rise impact modelling on small islands: Case study gili raja island of east Java. MATEC Web of Conferences, 177, 1–8. https://doi.org/10.1051/matecconf/201817701017. Hidayah, Z., Romadhon, A., & Witjarnoko, Y. (2019). Penilaian kerentanan wilayah pesisir selatan Pulau Bawean terhadap Kenaikan Muka Air Laut. Jurnal Perikanan Universitas Gadjah Mada, 20(2), 79. https://doi.org/10.22146/jfs.36109. Hidayah, Z., Wirayuhanto, H., Norma Sari, Z. R., & Wardhani, M. K. (2021). Modelling sea surface currents in the eastern coast of Bawean Island, East Java. IOP Conference Series: Earth and Environmental Science, 925(1). https://doi.org/10.1088/1755-1315/925/1/012006. Jamero, M. L., Onuki, M., Esteban, M., Billones-Sensano, X. K., Tan, N., Nellas, A., Takagi, H., Thao, N. D., & Valenzuela, V. P. (2017). Small-island communities in the Philippines prefer local measures to relocation in response to sea-level rise. Nature Climate Change, 7(8), 581–586. https://doi.org/10.1038/NCLIMATE3344. Ketjulan, R., Boer, M., Imran, Z., & Siregar, V. P. (2019). Daya dukung lahan untuk pemukiman penduduk dan implikasinya terhadap kualitas perairan di pulau-pulau kecil (kasus pulau-pulau kecil Selat Tiworo Kabupaten Muna Barat). Jurnal Ilmu Dan Teknologi Kelautan Tropis, 11(3), 569–582. https://doi.org/10.29244/jitkt.v11i3.25731. Kim, C. (2016). Land use classification and land use change analysis using satellite images in Lombok Island, Indonesia. Forest Science and Technology, 12(4), 183–191. https://doi.org/10.1080/21580103.2016.1147498. Kurniadin, N., & Fadlin, F. (2021). Analisis perubahan morfologi garis pantai akibat tsunami di Teluk Palu Menggunakan data Citra Sentinel-2. Geoid, 16(2), 240. https://doi.org/10.12962/j24423998.v16i2.8078. Li, J., Schill, S. R., Knapp, D. E., & Asner, G. P. (2019). Object-based mapping of coral reef habitats using planet dove satellites. Remote Sensing, 11(12). https://doi.org/10.3390/rs11121445. Lillesand, T. M & Keifer, R. W. (1994). Remote Sensing and Image Interpretation. 3rd. Edition. John Wiley and Sons, Inc. Canada. Luthfi, O. M., & Anugrah, P. T. (2017). Distribusi karang keras (Scleractinia) sebagai penyusun utama ekosistem terumbu karang di Gosong Karang Pakiman, Pulau Bawean. Depik, 6(1), 9–22. https://doi.org/10.13170/depik.6.1.5461. Marasabessy, I., Fahrudin, A., Imran, Z., Agus, S. B., Studi, P., Sumberdya, P., & Ilmu, D. (2016). Kajian kesesuaian ekowisata bahari pulau kecil (Studi Kasus : Nusa Manu dan Nusa Leun) Maluku Tengah. Jurnal Riset Kelautan Dan Perikanan, 1(2), 67–82. 217 Zainul Hidayah et al. / Geosfera Indonesia 7 (3), 2022, 207-218 Marlina, D. (2022). Klasifikasi tutupan lahan pada Citra Sentinel-2 Kabupaten Kuningan dengan NDVI dan Algoritme Random Forest. STRING (Satuan Tulisan Riset Dan Inovasi Teknologi), 7(1), 41. https://doi.org/10.30998/string.v7i1.12948. Mastu, L. O. K., Nababan, B., & Panjaitan, J. P. (2018). Pemetaan habitat bentik berbasis objek menggunakan Citra Sentinel-2 di Perairan Pulau Wangi-Wangi Kabupaten Wakatobi. Jurnal Ilmu Dan Teknologi Kelautan Tropis, 10(2), 381–396. https://doi.org/10.29244/jitkt.v10i2.21039. Muhammad A.Z., Prasita, V.P & Nirmalasari I.W. (2019). Parameter oseanografi untuk kesesuaian wisata snorkeling di Pulau Gili Dan Pulau Noko, Kepulauan Bawean. Jurnal Riset Kelautan Tropis (Journal Of Tropical Marine Research) (J-Tropimar), 1(2), 29. https://doi.org/10.30649/jrkt.v1i2.29. Muhammad, D. (2022). Kerentanan pesisir pulau kecil (Studi Kasus: Pulau Karimunjawa Dan Kemujan). Journal of Fisheries and Marine Research, 6(1). https://doi.org/10.21776/ub.jfmr.2022.006.01.11. Muhtadi, A., Harahap, Z. A., Pulungan, A., Nurmatias, N., Lubis, P., Siregar, Z., Ompusunggu, R. Y., & Aulia, F. (2020). Status dan sebaran mangrove di kawasan konservasi Taman Pulau Kecil, Kabupaten Tapanuli Tengah, Provinsi Sumatera Utara. Depik, 9(2), 200–209. https://doi.org/10.13170/depik.9.2.15065. Neksidin N., Fahrudin, A., & Krisanti, M. (2021). Keberlanjutan pengelolaan wisata bahari di Pulau Pari, Kabupaten Kepulauan Seribu. Jurnal Ilmu Pertanian Indonesia, 26(2), 284–291. https://doi.org/10.18343/jipi.26.2.284. Noor, M. A. F., & Romadhon, A. (2020). Analisis kesesuaian dan daya dukung lingkungan Pulau Gili Noko Bawean Sebagai Kawasan Ekowisata Pantai. Juvenil:Jurnal Ilmiah Kelautan Dan Perikanan, 1(1), 38–46. https://doi.org/10.21107/juvenil.v1i1.6749. Phiri, D., Simwanda, M., Salekin, S., Ryirenda, V. R., Murayama, Y., & Ranagalage, M. (2020). Sentinel-2 Data for Land Cover / Use Mapping : A Review. Remote Sensing, 2291(12), 14. Pinuji, S., Suhattanto, M. A., & Arianto, T. (2018). Dinamika dan tantangan penggunaan dan pemanfaatan tanah di wilayah pulau kecil. BHUMI: Jurnal Agraria Dan Pertanahan, 4(1). https://doi.org/10.31292/jb.v4i1.218. Purbani, D., Ramdhan, M., Salim, H. L., Daulat, A., Dwiyanti, D., Winarso, G., & Syahputri, S. (2020). Study of coastline changes and land use change in marine protected area (Case study: Derawan Island, East Kalimantan Province). IOP Conference Series: Earth and Environmental Science, 561(1). https://doi.org/10.1088/1755-1315/561/1/012056. Putri, D. R., Sukmono, A., & Sudarsono, B. (2018). Analisis kombinasi citra Sentinel-1A dan Citra Sentinel-2A untuk klasifikasi tutupan lahan (Studi Kasus: Kabupaten Demak, Jawa Tengah). Jurnal Geodesi Undip, 7(2), 85–96. Rahman, D. A., Gonzalez, G., & Aulagnier, S. (2017). Population size, distribution and status of the remote and critically endangered bawean deer Axis kuhlii. Oryx, 51(4), 665–672. https://doi.org/10.1017/S0030605316000429. Rini, M. S. (2018). Kajian kemampuan metode neural network untuk klasifikasi penutup lahan dengan menggunakan Citra Landsat-8 OLI (kasus di Kota Yogyakarta dan sekitarnya). Geomedia: Majalah Ilmiah Dan Informasi Kegeografian, 16(1), 1–12. https://doi.org/10.21831/gm.v16i1.20974. Santoso, D. H., Prasetya, J. D., & Rahman, D. (2020). Analisis daya dukung lingkungan hidup berbasis jasa ekosistem penyediaan air bersih di Pulau Karimunjawa. Jurnal Ilmu Lingkungan, 18(2), 290–296. https://doi.org/10.14710/jil.18.2.290-296. 218 Zainul Hidayah et al. / Geosfera Indonesia 7 (3), 2022, 207-218 Siregar, D. I., & Musadri Asbi, A. (2020). Pemanfaatan Citra Landsat 8 Operational Land Imager (OLI) untuk klasifikasi tutupan lahan di Taman Nasional Gunung Merbabu. Wahana Forestra: Jurnal Kehutanan, 15(2), 28–39. https://doi.org/10.31849/forestra.v15i2.4731. Stephenson, T. S., & Jones, J. J. (2017). Impacts of Climate Change on Extreme Events in the Coastal and Marine Environments of Caribbean Small Island Developing States (SIDS). Science Review, 10–22. Retrieved from https://assets.publishing.service.gov.uk. Sukandar, S., Dewi, C. S. U., & Handayani, M. (2017). Analisis kesesuaian dan daya dukung lingkungan untuk pengembangan wisata bahari di Pulau Bawean Kabupaten Gresik Provinsi Jaya Timur. Depik, 6(3), 205–2013. https://doi.org/10.13170/depik.6.3.7024. Suryanti, A., & Apdillah, D. (2022). Kesesuaian dan daya dukung kawasan untuk kegiatan ekowisata diving dan snorkeling di Perairan Gugusan Pulau Duyung. Journal of Marine Research, 11(3), 483–494. Trimanto, T. & Hapsari, L. (2016). Botanical survey in thirteen montane forests of Bawean Island Nature Reserve, East Java Indonesia: Flora diversity, conservation status, and bioprospecting. Biodiversitas, 17(2), 832–846. https://doi.org/10.13057/biodiv/d170261. Wael, K., Siahaya, W. A., Pertanian, J. B., Pertanian, F., Pattimura, U., Putuhena, J. I. M., & Poka, K. (2022). Klasifikasi tutupan lahan Pulau Kei Kecil Tahun 2019 berdasarkan analisis citra multispektral. Jurnal Budidaya Pertanian, 18(1), 18–27. https://doi.org/10.30598/jbdp.2022.18.1.18. Wardani, M. P., Fahrudin, A., & Yulianda, F. (2017a). Analysis of successful strategy to develop sustainable marine ecotourism in Gili Bawean Island, Gresik, East Java. IOP Conference Series: Earth and Environmental Science, 89(1). https://doi.org/10.1088/1755-1315/89/1/012036. Wardani, M. P., Fahrudin, A., & Yulianda, F. (2017b). Analysis of Successful Strategy to Develop Sustainable Marine Ecotourism in Gili Bawean Island, Gresik, East Java. IOP Conference Series: Earth and Environmental Science, 89(1). https://doi.org/10.1088/1755-1315/89/1/012036. Wicaksono, P., Aryaguna, P. A., & Lazuardi, W. (2019). Benthic habitat mapping model and cross validation using machine-learning classification algorithms. Remote Sensing, 11(11), 1–24. https://doi.org/10.3390/rs11111279. Wulansari, H. (2017). Uji akurasi klasifikasi penggunaan lahan dengan menggunakan metode defuzzifikasi maximum likelihood berbasis Citra Alos Avnir-2. BHUMI: Jurnal Agraria Dan Pertanahan, 3(1), 98. https://doi.org/10.31292/jb.v3i1.96.