Vol. 14 N0. 2 December 2017 P-ISSN 0216-6739; E-ISSN 2549-516X No. 774/AU3/P2MI-LIPI/08/2017 International Journal of Remote Sensing and Earth Sciences Published by Indonesian National Institute of Aeronautics and Space ( LAPAN ) Vol. 14 N0. 2 December 2017 P-ISSN 0216-6739; E-ISSN 2549-516X No. 774/AU3/P2MI-LIPI/08/2017 International Journal of Remote Sensing and Earth Sciences Published by Indonesian National Institute of Aeronautics and Space (LAPAN) Editorial Committee Preface Dear IJReSES Readers, We sincerely thank you for reading the International Journal of Remote Sensing and Earth Sciences Vol. 14 No 2, December 2017. In general, this journal is expected to enrich the serial publications on earth sciences. In particular this journal is aimed to present improvement in remote sensing studies and its applications on earth sciences. This journal also serves as the enrichment on earth sciences publication, not only in Indonesia and Asia but also worldwide. This journal consists of papers discussing the particular interest in remote sensing field. Those papers are having remote sensing data for image processing, geosciences, oceanography, environment, disaster, mining activities, etc. A variety of topics are discussed in this fourteenth edition. Briefly, the topics discussed in this edition are the studies of remote sensing data processing issues such as the peat thickness classes estimated from land cover, spatial projection of land use and its connection with urban ecology spatial planning, compression on remote sensing data, and preliminary study of LSU-02 photo data application to support 3D modeling. Meanwhile the topics on remote sensing applications and validation are also discussed such as determination of the best methodology for bathymetry mapping, carbon stock estimation of mangrove vegetation, detecting the area damage due to coal mining activities, and information criterion-based mangrove land classification. The publication of IJReSES is intended to supply the demands regarding the information on the Remote Sensing and Earth Sciences. This journal is also intended to motivate Indonesian as well as Asian scientists to submit their research results. Thus, by their submitted research results, it will contribute to the development and strengthening in remote sensing field particularly in Asia. To that end, we invite scientists to play their parts in this journal by submitting their scientific research papers. We look forward to receiving your research works for the next edition of this journal. Welcome to the sixth issue of the International Journal of Remote Sensing and Earth Sciences. This journal is expected to Editor-in-Chief, Dr M. Rokhis Khomarudin Dr. Orbita Roswintiarti Editorial Committee Members INTERNATIONAL JOURNAL OF REMOTE SENSING AND EARTH SCIENCES Vol. 14 N0. 2 December 2017 P-ISSN 0216-6739; E-ISSN 2549-516X Editor-in-Chief Co Editor-in-Chief : : Dr. M. Rokhis Khomarudin Prof. Dr. Erna Sri Adiningsih Dr. Rahmat Arief, M.Sc. Peer Reviewers : Prof. Dr. Ir. I Nengah Surati Jaya, M.Agr Prof. Dr. Erna Sri Adiningsih Dr. Ir. Dony Kushardono, M.Eng. Dr. Syarif Budhiman Dr. Ing. Widodo Setyo Pranowo Dr. Jonson Lumban Gaol Dr. Rahmat Arief, M.Sc. Dr. Baba Barus Dr. Indah Prasasti Dr. Sidik Mulyono, M. Eng Secretariat : Mr. Christianus R. Dewanto Mr. Jasyanto Ms. Mega Mardita Mr. Suwarsono Ms. Sayidah Sulma Ms. Fajar Yulianto Ms. Emiyati Mr. Zylshal Mr. Yudho Dewanto Mr. M. Luthfi Mr. Irianto Mr. Dwi Haryanto Mr. Aulia Pradipta Contribution Paper to: IJReSES Secretariat National Institute of Aeronautics and Space of Indonesia (LAPAN) Jl. Pemuda Persil No. 1, Rawamangun, Jakarta 13220, INDONESIA Phone. (021) 4892802 ext. 144 – 145 (Hunting) Fax. (021) 47882726 Pukasi.lapan@gmail.com; publikasi@lapan.go.id Published by: National Institute of Aeronautics and Space of Indonesia (LAPAN) iv iii INTERNATIONAL JOURNAL OF REMOTE SENSING AND EARTH SCIENCES Vol. 14 No. 2 December 2017 P-ISSN 0216-6739; E- ISSN 2549-516X No. 774/AU3/P2MI-LIPI/08/2017 Contents Editorial Committee Preface .……………………………...………………………................ Editorial Committee Members ............………………...………………..………...…............ ii iii CAN THE PEAT THICKNESS CLASSES BE ESTIMATED FROM LAND COVER TYPE APPROACH? Bambang Trisakti, Atriyon Julzarika, Udhi C. Nugroho, Dipo Yudhatama, and Yudi Lasmana…………………………………………………………………………………………………. 83 SPATIAL PROJECTION OF LAND USE AND ITS CONNECTION WITH URBAN ECOLOGY SPATIAL PLANNING IN THE COASTAL CITY, CASE STUDY IN MAKASSAR CITY, INDONESIA Syahrial Nur Amri, Luky Adrianto, Dietriech Geoffrey Bengen, Rahmat Kurnia……………………………….…………………………………………………………………… 95 THE EFFECT OF JPEG2000 COMPRESSION ON REMOTE SENSING DATA OF DIFFERENT SPATIAL RESOLUTIONS Anis Kamilah Hayati, Haris Suka Dyatmika ………………………………………………………... 111 PRELIMINARY STUDY OF LSU-02 PHOTO DATA APPLICATION TO SUPPORT 3D MODELING OF TSUNAMI DISASTER EVACUATION MAP Linda Yunita, Nurwita Mustika Sari, and Dony Kushardono .......................................................... 119 DETERMINATION OF THE BEST METHODOLOGY FOR BATHYMETRY MAPPING USING SPOT 6 IMAGERY: A STUDY OF 12 EMPIRICAL ALGORITHMS Masita Dwi Mandini Manessa, Muhammad Haidar, Maryani Hastuti, Diah Kirana Kresnawati….…………………………………………………………………………………………… 127 CARBON STOCK ESTIMATION OF MANGROVE VEGETATION USING REMOTE SENSING IN PERANCAK ESTUARY, JEMBRANA DISTRICT, BALI Amandangi Wahyuning Hastuti, Komang Iwan Suniada, Fikrul Islamy………………………… 137 DETECTING THE AREA DAMAGE DUE TO COAL MINING ACTIVITIES USING LANDSAT MULTITEMPORAL (Case Study: Kutai Kartanegara, East Kalimantan) Suwarsono, Nanik Suryo Haryani, Indah Prasasti, Hana Listi Fitriana M. Rokhis Khomarudin…………………………………………………………………………………………….. 151 MACHINE LEARNING-BASED MANGROVE LAND CLASSIFICATION WORLDVIEW-2 SATELLITE IMAGE IN NUSA LEMBONGAN ISLAND Aulia Ilham and Marza Ihsan Marzuki …………………………………………………….. 159 ON Instruction for Authors ............................................................................................................ 167 Index........................................................................................................................ .................... 168 Published by: National Institute of Aeronautics and Space of Indonesia (LAPAN) i International Journal of Remote Sensing and Earth Sciences P-ISSN 0216 – 6739; E- ISSN 2549-516X No. 774/AU3/P2MI-LIPI/08/2017 The abstract may be copied without permission or charge ABSTRACT HARMFUL ALGAL BLOOM 2012 EVENT VERIFICATION IN LAMPUNG BAY USING RED TIDE DETECTION ON SPOT 4 IMAGE / Emiyati1, Ety Parwati, and Syarif Budhiman IJRESES, 14 (1) 2017 : 1- 8 In mid-December 2012, harmful algal bloom phenomenon occurred in Lampung Bay. Harmful Algal Bloom (HAB) is blooming of algae in aquatic ecosystems. It has negative impact on living organism, due to its toxic. This study was applied Red Tide (RT) detection algorithm on SPOT 4 images and verified the distribution of HAB 2012 event in Lampung Bay. The HAB event in 2012 in Lampung Bay can be detected by using RT algorithm on SPOT 4 images quantitatively and qualitatively. According to field measurement, the phytoplankton blooming which happen at Lampung Bay in 2012 were Cochlodinium sp. Image analysis showed that Cochlodinium sp has specific pattern of RT with values, digitally, were 13 to 41 and threshold value of red band SPOT 4 image was 57. The total area of RT distribution, which are found in Lampung Bay, was 11,545.3 Ha. Based on the RT classification of RT images and field data measurement, the RT which is caused many fishes died on the western coastal of Lampung Bay spread out from Bandar Lampung City to Batumenyan village. By using confusion matrix, the accuracy of this this method was 74.05 %. This method was expected to be used as early warning system for HAB monitoring in Lampung Bay and perhaps in another coastal region of Indonesia. Keywords: harmful algal bloom, Lampung Bay, SPOT 4 image, red tide algorithm Vol. 14 No.1, June 2017 A PARTIAL ACQUISITION TECHNIQUE OF SAR SYSTEM USING COMPRESSIVE SAMPLING METHOD / Rahmat Arief IJRESES, 14 (1) 2017 : 9-18 In line with the development of Synthetic Aperture Radar (SAR) technology, there is a serious problem when the SAR signal is acquired using high rate analog digital converter (ADC), that require large volumes data storage. The other problem on compressive sensing method, which frequently occurs, is a large measurement matrix that may cause intensive calculation. In this paper, a new approach was proposed, particularly on the partial acquisition technique of SAR system using compressive sampling method in both the azimuth and range direction. The main objectives of the study are to reduce the radar raw data by decreasing the sampling rate of ADC and to reduce the computational load by decreasing the dimension of the measurement matrix. The simulation results found that the reconstruction of SAR image using partial acquisition model has better resolution compared to the conventional method (Range Doppler Algorithm/RDA). On a target of a ship, that represents a low-level sparsity, a good reconstruction image could be achieved from a fewer number measurement. The study concludes that the method may speed up the computation time by a factor 4.49 times faster than with a full acquisition matrix. Keywords: partial acquisition technique, aperture radar, compressive sampling synthetic International Journal of Remote Sensing and Earth Sciences P-ISSN 0216 – 6739; E- ISSN 2549-516X No. 774/AU3/P2MI-LIPI/08/2017 The abstract may be copied without permission or charge ABSTRACT Vol. 14 No.1, June 2017 VALIDATION OF COCHLODINIUM POLYKRIKOIDES RED TIDE DETECTION USING SEAWIFS-DERIVED CHLOROPHYLL-A DATA WITH NFRDI RED TIDE MAP IN SOUTH EAST KOREAN WATERS / Gathot Winarso and Joji Ishizaka IJRESES, 14 (1) 2017 : 19-26 A COMPARISON OF OBJECT-BASED AND PIXELBASED APPROACHES FOR LAND USE/LAND COVER CLASSIFICATION USING LAPAN-A2 MICROSATELLITE DATA / Jalu Tejo Nugroho1, Zylshal, Nurwita Mustika Sari, and Dony Kushardono IJRESES, 14 (1) 2017: 27-36 Annual summer red tides of Cochlodinium polykrikoides have happenned at southern coastal of the South Korea, accounted economic losses of 76.4 billion won in 1995 on fisheries and other economic substantial losses. Therefore, it is important to eliminate the damage and losses by monitoring the bloom and to forecast their development and movement. On previous study, ocean color satellite, SeaWiFS, standard chlorophyll-a data was used to detect the red tide, using threshold value of chlorophyll-a concentration ≥ 5 mg/m3, resulted a good correlation using visual comparison. However, statistic based accuracy analysis has not be done yet. In this study, the accuracy of detection method was analyzed using spatial statistic. Spatial statistical match up analysis resulted 68% of red tide area was not presented in satellite data due to masking. Within red tide area where data existed, 36% was in high chlorophyll-a area and 64% was in low chlorophyll-a area. Within the high chlorophyll-a area 13% and 87% was in and out of the red tide area. It was found that the accuracy of this detection is low. However if the accuracy was yearly splitted, its found that 75% accuracy on 2002 where visually red tide detected spead out to the off-shore area. The fail and false detection are not due to the failure of the detection method but caused by limitation of the technology due to the natural condition i.e. type of red tide spreading, cloud cover and other flags such as turbid water, stray light etc. In recent years, small satellite industry has been a rapid trend and become important especially when associated with operational cost, technology adaptation and the missions. One mission of LAPANA2, the 2nd generation of microsatellite that developed by Indonesian National Institute of Aeronautics and Space (LAPAN), is Earth observation using digital camera that provides imagery with 3.5 m spatial resolution. The aim of this research is to compare between object-based and pixel-based classification of land use/land cover (LU/LC) in order to determine the appropriate classification method in LAPAN-A2 data processing (case study Semarang, Central Java).The LU/LC were classified into eleven classes, as follows: sea, river, fish pond, tree, grass, road, building 1, building 2, building 3, building 4 and rice field. The accuracy of classification outputs were assessed using confusion matrix. The object-based and pixel-based classification methods result for overall accuracy are 31.63% and 61.61%, respectively. According to accuracy result, it was thought that blurring effect on LAPAN-A2 data may be the main cause of accuracy decrease. Furthermore, the result is suggested to use pixel-based classification to be applied in LAPAN-A2 data processing. Keywords: cochlodinium polykrikoides, chlorophyll-a, SeaWiFS, red tide Keywords: LAPAN-A2 microsatellite, LU/LC, objectbased, pixel-based International Journal of Remote Sensing and Earth Sciences P-ISSN 0216 – 6739; E- ISSN 2549-516X No. 774/AU3/P2MI-LIPI/08/2017 The abstract may be copied without permission or charge ABSTRACT VERIFICATION OF PISCES DISSOLVED OXYGEN MODEL USING IN SITU MEASUREMENT IN BIAK, ROTE, AND TANIMBAR SEAS, INDONESIA / Armyanda Tussadiah, Joko Subandriyo, Sari Novita, Widodo S. Pranowo IJRESES, 14 (1) 2017: 37-46 Dissolved oxygen (DO) is one of the most chemical primary data in supported life for marine organisms. Ministry of Marine Affairs and Fisheries Republic of Indonesia through Infrastructure Development for Space Oceanography (INDESO) Project provides dissolved oxygen data services in Indonesian Seas for 7 days backward and 10 days ahead (9,25 km x 9.25 km, 1 daily). The data based on Biogeochemical model (PISCES) coupled with hydrodynamic model (NEMO), with input data from satellite acquisition. This study investigated the performance and accuracy of dissolved oxygen from PISCES model, by comparing with the measurement in situ data in Indonesian Seas specifically in three outermost islands of Indonesia (Biak Island, Rote Island, and Tanimbar Island). Results of standard deviation values between in situ DO and model are around two (St.dev ± 2). Based on the calculation of linear regression between in situ DO with the standard deviation obtained a high determinant coefficient, greater than 0.9 (R2 ≥ 0.9). Furthermore, RMSE calculation showed a minor error, less than 0.05. These results showed that the equation of the linear regression might be used as a correction equation to gain the verified dissolved oxygen. Keywords: verification, PISCES model, dissolved oxygen, in situ measurement, indonesia, linear regression Vol. 14 No.1, June 2017 IN-SITU MEASUREMENT OF DIFFUSE ATTENUATION COEFFICIENT AND ITS RELATIONSHIP WITH WATER CONSTITUENT AND DEPTH ESTIMATION OF SHALLOW WATERS BY REMOTE SENSING TECHNIQUE / Budhi Agung Prasetyo, Vincentius Paulus Siregar, Syamsul Bahri Agus, Wikanti Asriningrum IJRESES, 14 (1) 2017: 47-60 Diffuse attenuation coefficient, Kd(λ), has an empirical relationship with water depth, thus potentially to be used to estimate the depth of the water based on the light penetration in the water column. The aim of this research is to assess the relationship of diffuse attenuation coefficient with the water constituent and its relationship to estimate the depth of shallow waters of Air Island, Panggang Island and Karang Lebar lagoons and to compare the result of depth estimation from Kd model and derived from Landsat 8 imagery. The measurement of Kd(λ) was carried out using hyperspectral spectroradiometer TriOS-RAMSES with range 320 – 950 nm. The relationship between measurement Kd(λ) on study site with the water constituent was the occurrence of absorption by chlorophyll-a concentration at the blue and green spectral wavelength. Depth estimation using band ratio from Kd(λ) occurred at 442,96 nm and 654,59 nm, which had better relationship with the depth from insitu measurement compared to the estimation based on Landsat 8 band ratio. Depth estimated based on Kd(λ) ratio and in-situ measurement are not significantly different statistically. Depth estimated based on Kd(λ) ratio and in-situ measurement are not significantly different statistically. However, depth estimation based on Kd(λ) ratio was inconsistent due to the bottom albedo reflection because the Kd(λ) measurement was carried out in shallow waters. Estimation of water depth based on Kd(λ) ratio had better results compared to the Landsat 8 band ratio. Keywords: in-situ measurement, diffuse attenuation coefficient, relationship with water constituent, depth estimation, shallow water, remote sensing International Journal of Remote Sensing and Earth Sciences P-ISSN 0216 – 6739; E- ISSN 2549-516X No. 774/AU3/P2MI-LIPI/08/2017 The abstract may be copied without permission or charge ABSTRACT Vol. 14 No.1, June 2017 TIME SERIES ANALYSIS OF TOTAL SUSPENDED SOLID (TSS) USING LANDSAT DATA IN BERAU COASTAL AREA, INDONESIA / Ety Parwati1 and Anang Dwi Purwanto IJRESES, 14 (1) 2017: 61-70 SIMULATION OF DIRECT GEOREFERENCING FOR GEOMETRIC SYSTEMATIC CORRECTION ON LSA PUSHBROOM IMAGER / Muchammad Soleh1, Wismu Sunarmodo, and Ahmad Maryanto IJRESES, 14 (1) 2017: 71-82 Water quality information is usually used for the first examination of the pollution. One of the parameters of water quality is Total Suspended Solid (TSS), which describes the amount of matter of particles suspended in the water. TSS information is also used as initial information about waters condition of a region. TSS could be derive from Landsat data with several combinations of spectral channels to evaluate the condition of the observation area for both the waters and the surrounding land. The study aimed to evaluate Berau waters condition in Kalimantan, Indonesia, by utilizing TSS dynamics extracted from Landsat data. Validated TSS extraction algorithm was obtained by choosing the best correlation between field data and image data. Sixty pairs of points had been used to build validated TSS algorithms for the Berau Coastal area. The algorithm was TSS = 3.3238 * exp (34 099 * Red Band Reflectance). The data used for this study were Landsat 5 TM, Landsat 7 ETM and Landsat 8 data acquisition in 1994, 1996, 1998, 2002, 2004, 2006, 2008 and 2013. For detailed evaluation, 20 regions were created along the watershed up to the coast. The results showed the fluctuation of TSS values in each selected region. TSS value increased if there was a change of any kind of land cover/land used into bareland, ponds, settlements or shrubs. Conversely, TSS value decreased if there was a wide increase of mangrove area or its position was very closed to the ocean. LAPAN has developed remote sensing data collection by using a pushbroom linescan imager camera sensor mounted on LSA (Lapan Surveillance Aircraft). The position accuracy and orientation system for LSA applications are required for Direct Georeferencing and depend on the accuracy of off-the-shelf integrated GPS/inertial system, which used on the camera sensor. This research aims to give the accuracy requirement of Inertial Measurement Unit (IMU) sensor and GPS to improve the accuracy of the measurement results using direct georeferencing technique. Simulations were performed to produce geodetic coordinates of longitude, latitude and altitude for each image pixel in the imager pushbroom one array detector, which has been geometrically corrected. The simulation results achieved measurement accuracies for mapping applications with Ground Sample Distance (GSD) or spatial resolution of 0,6 m of the IMU parameter (pitch, roll and yaw) errors about 0.1; 0.1; and 0.1 degree respectively, and the error of GPS parameters (longitude and latitude) about 0.00002 and 0.2 degree. The results are expected to be a reference for a systematic geometric correction to image data pushbroom linescan imager that would be obtained by using LSA spacecraft. Keywords: TSS, Landsat 5 TM, Landsat 7 ETM +, Landsat 8, watershed, mangrove Keywords: direct georeferencing, pushbroom imager, systematic geometric correction, LSA International Journal of Remote Sensing and Earth Sciences P-ISSN 0216 – 6739; E- ISSN 2549-516X No. 774/AU3/P2MI-LIPI/08/2017 The abstract may be copied without permission or charge ABSTRACT CAN THE PEAT THICKNESS CLASSES BE ESTIMATED FROM LAND COVER TYPE APPROACH?/Bambang Trisakti Bambang, Atriyon Julzarika, Udhi C. Nugroho, Dipo Yudhatama, and Yudi Lasmana IJRESES, 14 (2) 2017: 83-94 Indonesia has been known as a home of the tropical peatlands. The peatlands are mainly in Sumatera, Kalimantan and Papua Islands. Spatial information on peatland depth is needed for the planning of agricultural land extensification. The research objective was to develop a preliminary estimation model of peat thickness classes based on land cover approach and analyse its applicability using Landsat 8 image. Ground data, including land cover, location and thickness of peat, were obtained from various surveys and peatlands potential map (Geology Map and Wetlands Peat Map). The land cover types were derived from Landsat 8 image. All data were used to build an initial model for estimating peat thickness classes in Merauke Regency. A table of relationships among land cover types, peat potential areas and peat thickness classes were made using ground survey data and peatlands potential maps of that were best suited to ground survey data. Furthermore, the table was used to determine peat thickness classes using land cover information produced from Landsat 8 image. The results showed that the estimated peat thickness classes in Merauke Regency consist of two classes, i.e., very shallow peatlands and shallow peatlands. Shallow peatlands were distributed at the upper part of Merauke Regency with mainly covered by forest. In comparison with Indonesia Peatlands Map, the number of classes was the two classes. The spatial distribution of shallow peatlands was relatively similar for its precision and accuracy, but the estimated area of shallow peatlands was greater than the area of shallow peatlands from Indonesia Peatlands Map. This research answered the question that peat thickness classes could be estimated by the land cover approach qualitatively. The precise estimation of peat thickness could not be done due to the limitation of insitu data. Keywords: Peat thickness, Landsat 8 image, land cover, Merauke Regency, shallow peatlands Vol. 14 No.2, December 2017 SPATIAL PROJECTION OF LAND USE AND ITS CONNECTION WITH URBAN ECOLOGY SPATIAL PLANNING IN THE COASTAL CITY, CASE STUDY IN MAKASSAR CITY, INDONESIA/Syahrial Nur Amri, Luky Adrianto, Dietriech Geoffrey Bengen, and Rahmat Kurnia IJRESES, 14 (2) 2017: 95-110 The arrangement of coastal ecological space in the coastal city area aims to ensure the sustainability of the system, the availability of local natural resources, environmental health and the presence of the coastal ecosystems. The lack of discipline in the supervision and implementation of spatial regulations resulted in inconsistencies between urban spatial planning and land use facts. This study aims to see the inconsistency between spatial planning of the city with the real conditions in the field so it can be used as an evaluation material to optimize the planning of the urban space in the future. This study used satellite image interpretation, spatial analysis, and projection analysis using markov cellular automata, as well as consistency evaluation for spatial planning policy. The results show that there has been a significant increase of open spaces during 2001-2015 and physical development was relatively spreading irregularly and indicated the urban sprawl phenomenon. There has been an open area deficits for the green open space in 2015-2031, such as integrated maritime, ports, and warehousing zones. Several islands in Makassar City are predicted to have their built-up areas decreased, especially in Lanjukang Island, Langkai Island, Kodingareng Lompo Island, Bone Tambung Island, Kodingareng Keke Island and Samalona Island. Meanwhile, the increase of the built up area is predicted to occur in Lumu Island, Barrang Caddi Island, Barrang Lompo Island, Lae-lae Island, and Kayangan Island. The land cover is caused by the human activities. Many land conversions do not comply with the provision of percentage of green open space allocation in the integrated strategic areas, established in the spatial plan. Thus, have the potential of conflict in the spatial plan of marine and small islands in Makassar City. Keywords: spatial projection, land use, spatial planning, remote sensing, coastal city International Journal of Remote Sensing and Earth Sciences P-ISSN 0216 – 6739; E- ISSN 2549-516X No. 774/AU3/P2MI-LIPI/08/2017 The abstract may be copied without permission or charge ABSTRACT THE EFFECT OF JPEG2000 COMPRESSION ON REMOTE SENSING DATA OF DIFFERENT SPATIAL RESOLUTIONS/ Anis Kamilah Hayati and Haris Suka Dyatmika IJRESES, 14 (2) 2017: 111-118 The huge size of remote sensing data implies the information technology infrastructure to store, manage, deliver and process the data itself. To compensate these disadvantages, compressing technique is a possible solution. JPEG2000 compression provide lossless and lossy compression with scalability for lossy compression. As the ratio of lossy compression getshigher, the size of the file reduced but the information loss increased. This paper tries to investigate the JPEG2000 compression effect on remote sensing data of different spatial resolution. Three set of data (Landsat 8, SPOT 6 and Pleiades) processed with five different level of JPEG2000 compression. Each set of data then cropped at a certain area and analyzed using unsupervised classification. To estimate the accuracy, this paper utilized the Mean Square Error (MSE) and the Kappa coefficient agreement. The study shows that compressed scenes using lossless compression have no difference with uncompressed scenes. Furthermore, compressed scenes using lossy compression with the compression ratioless than 1:10 have no significant difference with uncompressed data with Kappa coefficient higher than 0.8. Keywords: compression, effect, spatial resolution, remote sensing, JPEG2000 Vol. 14 No.2, December 2017 PRELIMINARY STUDY OF LSU-02 PHOTO DATA APPLICATION TO SUPPORT 3D MODELING OF TSUNAMI DISASTER EVACUATION MAP/Linda Yunita, Nurwita Mustika Sari, and Dony Kushardono IJRESES, 14 (2) 2017: 119-126 The southern coast of Pacitan Regency is one of the vulnerable areas to the tsunami. Therefore, the map of the vulnerable and safe area from the tsunami disaster is required. Currently, there are many mapping technologies with UAVs used for spatial analysis. One of the UAV technologies which used in this research is LAPAN Surveillance UAV 02 (LSU-02). This study aims to map the evacuation plan area from LSU-02 aerial imagery. Tsunami evacuation area was identified by processing the aerial photo data into orthomosaic and Digital Elevation Model (DEM). The result shows that there are four points identified as the tsunami evacuation plan area. These points are located higher than the surrounding area and are easily accessible. Keywords: Aerial remote sensing, photo data of LSU-02, 3D modelling, tsunami International Journal of Remote Sensing and Earth Sciences P-ISSN 0216 – 6739; E- ISSN 2549-516X No. 774/AU3/P2MI-LIPI/08/2017 The abstract may be copied without permission or charge ABSTRACT DETERMINATION OF THE BEST METHODOLOGY FOR BATHYMETRY MAPPING USING SPOT 6 IMAGERY: A STUDY OF 12 EMPIRICAL ALGORITHMS/Masita Dwi Mandini Manessa, Muhammad Haidar, Maryani Hastuti, and Diah Kirana IJRESES, 14 (2) 2017: 127-136 For the past four decades, many researchers have published a novel empirical methodology for bathymetry extraction using remote sensing data. However, a comparative analysis of each method has not yet been done. Which is important to determine the best method that gives a good accuracy prediction. This study focuses on empirical bathymetry extraction methodology for multispectral data with three visible band, specifically SPOT 6 Image. Twelve algorithms have been chosen intentionally, namely, 1) Ratio transform (RT); 2) Multiple linear regression (MLR); 3) Multiple nonlinear regression (RF); 4) Second-order polynomial of ratio transform (SPR); 5) Principle component (PC); 6) Multiple linear regression using relaxing uniformity assumption on water and atmosphere (KNW); 7) Semiparametric regression using depth-independent variables (SMP); 8) Semiparametric regression using spatial coordinates (STR); 9) Semiparametric regression using depth-independent variables and spatial coordinates (TNP), 10) bagging fitting ensemble (BAG); 11) least squares boosting fitting ensemble (LSB); and 12) support vector regression (SVR). This study assesses the performance of 12 empirical models for bathymetry calculations in two different areas: Gili Mantra Islands, West Nusa Tenggara and Menjangan Island, Bali. The estimated depth from each method was compared with echosounder data; RF, STR, and TNP results demonstrate higher accuracy ranges from 0.02 to 0.63 m more than other nine methods. The TNP algorithm, producing the most accurate results (Gili Mantra Island RMSE = 1.01 m and R2=0.82, Menjangan Island RMSE = 1.09 m and R2=0.45), proved to be the preferred algorithm for bathymetry mapping. Keywords: bathymetry; SPOT methodology; multispectral image 6; empirical Vol. 14 No.2, December 2017 CARBON STOCK ESTIMATION OF MANGROVE VEGETATION USING REMOTE SENSING IN PERANCAK ESTUARY, JEMBRANA DISTRICT, BALI/Amandangi Wahyuning Hastuti, Komang Iwan Suniada, and Fikrul Islamy IJRESES, 14 (2) 2017: 137-150 Mangrove vegetation is one of the forest ecosystems that offers a potential of substantial greenhouse gases (GHG) emission mitigation, due to its ability to sink the amount of CO2 in the atmosphere through the photosynthesis process. Mangroves have been providing multiple benefits either as the source of food, the habitat of wildlife, the coastline protectors as well as the CO2 absorber, higher than other forest types. To explore the role of mangrove vegetation in sequestering the carbon stock, the study on the use of remotely sensed data in estimating carbon stock was applied. This paper describes an examination of the use of remote sensing data particularly Landsat-data with the main objective to estimate carbon stock of mangrove vegetation in Perancak Estuary, Jembrana, Bali. The carbon stock was estimated by analyzing the relationship between NDVI, Above Ground Biomass (AGB) and Below Ground Biomass (BGB). The total carbon stock was obtained by multiplying the total biomass with the carbon organic value of 0.47. The study results show that the total accumulated biomass obtained from remote sensing data in Perancak Estuary in 2015 is about 47.20±25.03 ton ha-1 with total carbon stock of about 22.18±11.76 tonC ha-1and CO2 sequestration 81.41±43.18 tonC ha-1. Keywords: Perancak Estuary, carbon stock estimation, mangrove, CO2 sequestration, NDVI International Journal of Remote Sensing and Earth Sciences P-ISSN 0216 – 6739; E- ISSN 2549-516X No. 774/AU3/P2MI-LIPI/08/2017 The abstract may be copied without permission or charge ABSTRACT DETECTING THE AREA DAMAGE DUE TO COAL MINING ACTIVITIES USING LANDSAT MULTITEMPORAL (CASE STUDY: KUTAI KARTANEGARA, EAST KALIMANTAN) /Suwarsono, Nanik Suryo Haryani, Indah Prasasti, Hana Listi Fitriana M. Priyatna, and M. Rokhis Khomarudin IJRESES, 14 (2) 2017: 151-158 Coal is one of the most mining commodities to date, especially to supply both national and international energy needs. Coal mining activities that are not well managed will have an impact on the occurrence of environmental damage. This research tried to utilize the multitemporal Landsat data to analyze the land damage caused by coal mining activities. The research took place at several coal mine sites in East Kalimantan Province. The method developed in this research is the method of change detection. The study tried to know the land damage caused by mining activities using NDVI (Normalized Difference Vegetation Index), NDSI (Normalized Difference Soil Index), NDWI (Normalized Difference Water Index) and GEMI (Global Environment Monitoring Index) parameter based change detection method. The results showed that coal mine area along with the damage that occurred in it can be detected from multitemporal Landsat data using NDSI value-based change detection method. The area damage due to coal mining activities can be classified into high, moderate, and low classes based on the mean and standard deviation of NDSI changes (ΔNDSI). The results of this study are expected to be used to support government efforts and mining managers in post-mining land reclamation activities. Keywords: damage area, coal mining, landsat multitemporal Vol. 14 No.2, December 2017 AKAIKE INFORMATION CRITERION BASED MANGROVE LAND CLASSIFICATION USING WORLDVIEW-2 SATELLITE IMAGES IN NUSA LEMBONGAN ISLAND/Aulia Ilham and Marza Ihsan Marzuki IJRESES, 14 (2) 2017: 159-166 Machine learning is an empirical approach for regressions, clustering and/or classifying (supervised or unsupervised) on a non-linear system. This method is mainly used to analyze a complex system for wide data observation. In remote sensing, machine learning method could be used for image data classification with software tools independence. This research aims to classify the distribution, type, and area of mangroves using Akaike Information Criterion approach for case study in Nusa Lembongan Island. This study is important because mangrove forests have an important role ecologically, economically, and socially. For example is as a green belt for protection of coastline from storm and tsunami wave. Using satellite images Worldview-2 with data resolution of 0.46 meters, this method could identify automatically land class, sea class/water, and mangroves class. Three types of mangrove have been identified namely: Rhizophora apiculata, Sonnetaria alba, and other mangrove species. The result showed that the accuracy of classification was about 68.32%. Keywords: clustering, machine learning, remote sensing data International Journal of Remote Sensing and Earth Sciences Vol.14 No. 2 December 2017: 137 – 150 CARBON STOCK ESTIMATION OF MANGROVE VEGETATION USING REMOTE SENSING IN PERANCAK ESTUARY, JEMBRANA DISTRICT, BALI Amandangi Wahyuning Hastuti1*, Komang Iwan Suniada, Fikrul Islamy 1Institute for Marine Research and Observation – Perancak, Bali *e-mail: amandangi.wahyuning@gmail.com Received: 7 November 2017; Revised: 20 November 2017; Approved: 22 December 2017 Abstract. Mangrove vegetation is one of the forest ecosystems that offers a potential of substantial greenhouse gases (GHG) emission mitigation, due to its ability to sink the amount of CO 2 in the atmosphere through the photosynthesis process. Mangroves have been providing multiple benefits either as the source of food, the habitat of wildlife, the coastline protectors as well as the CO2 absorber, higher than other forest types. To explore the role of mangrove vegetation in sequestering the carbon stock, the study on the use of remotely sensed data in estimating carbon stock was applied. This paper describes an examination of the use of remote sensing data particularly Landsatdata with the main objective to estimate carbon stock of mangrove vegetation in Perancak Estuary, Jembrana, Bali. The carbon stock was estimated by analyzing the relationship between NDVI, Above Ground Biomass (AGB) and Below Ground Biomass (BGB). The total carbon stock was obtained by multiplying the total biomass with the carbon organic value of 0.47. The study results show that the total accumulated biomass obtained from remote sensing data in Perancak Estuary in 2015 is about 47.20±25.03 ton ha-1 with total carbon stock of about 22.18±11.76 tonC ha -1and CO2 sequestration 81.41±43.18 tonC ha-1. Keywords: Perancak Estuary, carbon stock estimation, mangrove, CO2 sequestration, NDVI 1 INTRODUCTION Global warming is one of the strategic issues in the world today, as marked by the incidence of rising earth temperatures related to greenhouse gases. Several researchers noted that the major contributors to global warmings, such as carbon dioxide (CO2), and methane (CH) gases are anthropogenic, mainly produced from the human activities like fossil fuels burning, industry, deforestation, forest degradation and other forest conversion through combustion (Giri and Mandla 2017; Vicharnakorn et al. 2014). The accumulation of these gases causes the earth's temperature to rise, triggering climate change on Earth (Manuri et al. 2011). Sutaryo (2009) describes the forest biomass as highly relevant to climate change issues. Forest biomass has an important role in the biogeochemical cycle, especially in the carbon cycle. Of the total forest carbon, about 50% is stored in forest vegetation. As a consequence, if there is forest damage, forest fire, logging and etc., it will increase the possibility to have larger -amount of carbon in the atmosphere. The dynamics of carbon in nature can be explained simply by the carbon cycle. The carbon cycle is a biogeochemical cycle that includes the exchanged/transferred of carbon between the biosphere, the @National Institute of Aeronautics and Space of Indonesia (LAPAN) 137 Amandangi Wahyuning Hastuti et al. pedosphere, the geosphere, the hydrosphere and the earth's atmosphere. The carbon cycle is actually a complex process and every process related to other processes. The global carbon describes the exchanges of carbon between the Earth’s atmosphere, oceans, land and fossil fuels, which are both sources of emissions and sinks that contain carbon. One important function of the carbon cycle is the regulation of earth’s climate (Bennington 2009). Mangrove ecosystems, like other forest ecosystems, have a potential -ability to absorb carbon dioxide better than other forest ecosystems due to its ability to grow faster than any other forest vegetation. It is noted that mangrove forests have an important role in reducing the concentration of carbon dioxide in the air. Mangrove forest is one of the highest carbon-storage forests in the tropics and it is very high compared to the average carbon storages in other kinds of forest in the world (Donato et al. 2012). Although mangroves are known to have good assimilation capabilities with environmental components and have a high rate of carbon sequestration, data and information on carbon storage for some components, especially for tree biomass are very limited (Komiyama et al. 2008), so it is important to know that biomass information in the mangroves for sustainable forest management. In forest carbon inventories, carbon pools are accounted for by at least 4 carbon bags: surface biomass, subsurface biomass, dead organic matter and soil organic carbon. According to Donato et al. (2011), carbon is mostly stored in the sediments, aerial vegetal biomass, and below-ground biomass in the descending order. However recent studies suggest the importance of the carbon stock in the below-ground biomass of mangrove forests (Abohassan et al. 2012), there a few estimates 138 regarding this compartment (Komiyama et al. 2008). According to Lu (2006), field or terrestrial measurement is the most accurate way to collect biomass data, but this method is generally very expensive, time-consuming, labor-intensive and difficult to apply into remote and broad areas. Therefore there is another alternative solution in knowing the potential information of biomass that is by using aerial approach through remote sensing technology. The advantage of the remote sensing technology is to provide information needed quickly and completely at a relatively cheaper cost. In addition, the use of remote sensing technology in finding information on potential estimation of mangrove biomass as CO2 absorber can be monitored effectively and efficiently every year. One of the remote sensing data that can be utilized is Landsat satellite data. Situmorang et al. (2016) found that there was a high correlation (R²=0.729) between vegetation index resulted from satellite data and carbon stock estimation calculated using allometric equation. This high determination coefficient indicates that the satellite data is feasible to use to estimate carbon stock. Many studies on carbon stocks in mangrove vegetation by using remote sensing techniques have been conducted. In mangrove forest carbon stock (Mariana et al. 2015; Alemayehu et al. 2014; Sitoe et al. 2014; Hamdan et al. 2013; Murdiyarso et al. 2009) carbon sequestration (e.g. Bouillon et al. 2008; Khan et al. 2007) and organic carbon dynamics (Kristensen et al. 2008; Machiwa and Hallberg 2002) have been studied much. Carbon stock in mangrove ecosystem varies with species (Fu and Wu 2011; Laffoley and Grimsditch 2009), vegetation type (Sahu et al. 2016; CerónBretón et al. 2011; Mitra et al. 2011; Sapit International Journal of Remote Sensing and Earth Science Vol. 14 No. 2 December 2017 Carbon Stock Estimation using Remote Sensing ..... et al. 2011) and salinity (Adame et al. 2013). Perancak Estuary is one of the four main mangrove ecosystems in Bali Island besides West Bali National Park, Benoa Bay and Nusa Lembongan. In addition to feeding, spawning and nursery ground, information on the ability of mangrove forest in Perancak Estuary to store carbon utilizing of remote sensing technology is still very low, so this research becomes very important to do. The objective of this research is to identify the biomass and potential carbon stock of mangroves vegetation in Perancak Estuary by using remote sensing approach. This information is very useful for Jembrana district government to support sustainable development planning based on low carbon, especially for the coastal area. digital value into following formula: Radian ρλ‘ =MρQcal + Aρ using the (2-1) Figure 2-1: Location of the research area 2 2.1 MATERIALS AND METHODOLOGY Location and Data The research is located in Perancak Estuary, Jembrana District, Bali as shown in Figure 2-1. Geographically, Perancak Estuary is located between 8°22'30"S to 8°24'18"S and 114°36'18"E to 114°38'31.2"E. Perancak Estuary has an area of 2512.69 ha, with land use in the form of fishponds and mangroves. Data used in this research is Landsat 8 OLI/TIRS with acquisition date 13 September 2015 and path/ row 117/ 66 which obtained from United States Geological Survey (USGS) through website https://earthexplorer.usgs.gov. 2.2 Data Analysis 2.2.1 Converting digital values into reflectance The Landsat sensor is converted into reflectance value by using the variable factors that provided in the metadata. Landsat suggests to converting the ρλ' = a reflectance value without correction to the sun's elevation, Mρ = band specific multiplicative rescaling factor (wherex is Band (REFLECTANCE_MULT_ BAND_x), Aρ = Band-specific additive rescaling factor from the metadata (where x is Band (REFLECTANCE_ADD_BAND_x,), Qcal = quantized and calibrated standard product pixel values(DN). Conversion of reflectance value to the sun elevation follows equation (2-2). (2-2) Ρλ' = reflectance, θSE = a local sun elevation. The scene center sun elevation in degrees (SUN_ELEVATION); θSZ = local solar zenith angle, θSZ=90°-θSE. 2.2.2 Vegetation index Calculation of land covers vegetation index using Normalized Difference Vegetation Index (NDVI). NDVI is a International Journal of Remote Sensing and Earth Science Vol. 14 No. 2 December 2017 139 Amandangi Wahyuning Hastuti et al. calculation of visible light and near infrared which is reflected by vegetation. The classification of pixel values for NDVI ranges from -1 to 1. The low (negative) NDVI values identify areas of water bodies, rocks, sand, and snow. High NDVI values (positive) identify areas of vegetation in the form of savanna, bush, and forests, whereas the NDVI value near 0 generally identifies bare land (Saputra 2007). This value of NDVI can be calculated using the equation 2-3. is cloud-covered and the absence of other data available as the closing or graph fill data so that the edge of limitation digit to classify using the 563 composite band approach with mangrove-looking conditions as seen in Figure 2-2. (2-3) where NIR = near infrared band, RED = red band. The NIR reflectance is affected by leaf internal structure and leaf dry matter content. RED is the reflectance or radiance in a visible wavelength channel (0.63 - 0.69 μm) and corresponds to band 3 for ETM+ images. In this paper, NDVI images were generated to enhance mangrove forest that has higher NIR reflectance, and lower red light reflectance. Also, NDVI images were produced to eliminate water bodies, those of low red light reflectance, and those of very low NIR reflectance. 2.2.3 Image classification of mangrove land Image classification is performed to separate the spectral values contained in the pixel image unit. The unit of the pixel value is explained into several classes of land cover. The method of satellite image classification guided using imagery classification method to distinguish between mangrove area and non-mangrove area. Guided classification uses the maximum likelihood method which assumes that the class statistics in each band are normally distributed. The manual visual analysis classification is performed when the data 140 Figure 2-2: The mangrove vegetation shown in the composite image of 5-6-3 (NIRWSIR 1-Green) The pixel class is determined by the highest probability level. The results of guided data and manual visual analysis can then be converted to measure the area of land cover. 2.2.4 Above Ground Biomass (AGB) estimation Estimation of above the ground surface biomass value was done using the approach of NDVI result of equation correlation with Above Ground Biomass (AGB) of mangrove that is equal to 0.787 by Jha et al. (2015) as follows: (2-4) NDVI = the value of Vegetation Index, AGB = the Above Ground Biomass Value (ton ha-1). 2.2.5 Below Ground Biomass (BGB) estimation The estimated value of Below Ground Biomass (BGB) is obtained from the estimation of AGB which is International Journal of Remote Sensing and Earth Science Vol. 14 No. 2 December 2017 Carbon Stock Estimation using Remote Sensing ..... formulated using the equation compiled by Cairns, et al (1997) as follows: (2-5) The biomass or carbon stock unit can be converted from unit ton ha-1 into kg/Landsat Pixel (kg/900m2) by using the equation (9). AGB = the value of Above Ground Biomass (ton ha-1), BGB = the Below Ground Biomass value (ton ha-1). 2.2.6 Total Accumulated Biomass (TAC) Calculation Total Accumulation Biomass (TAB) is formulated by using: TAB = AGB + BGB (2-6) TAB = Total Accumulated Biomass (ton ha-1). 2.2.7 Total Carbon Stock (TCS) calculation Calculation of total carbon stock based on Westlake (1963) using the following formula: (2-7) TCS = the value of Total Carbon Stock (ton C ha-1), TAB = the value of Total Accumulated Biomass (ton ha-1), %C organic = the percentage value of carbon stock (0.47) or using the value of carbon emitted from the measurement results in the laboratory. 2.2.8 Amount of CO2 Sequestration (ACS) calculation IPCC (2001) suggests converting carbon stock from biomass to carbon dioxide uptake using the following conversion: (2-8) ACS = the Amount of CO2 Sequestration (ton C ha-1), TCS = the value of Total Carbon Stock (ton C ha-1). (2-9) 2.2.9 Estimation of biomass, carbon stock and CO2sequestration The biomass, carbon stocks, and carbon sequestration are calculated by constructing the equation above using ArcGIS Geoprocessing toolbox application for Landsat Image 7 and 8. 3 RESULTS AND DISCUSSION From the image analysis, we found that the extent of mangrove vegetation within the research area is approximately 101.16 ha. Green, et al (1998) did the assessment of mangrove area using NDVI to estimate of percent canopy, the accuracy of the percent canopy closure image was 80%. High accuracy to assess mangrove using NDVI also reported by Otero, et al (2016) with the overall accuracy of 87% ± 2%. The ability of the NDVI method to detect mangrove vegetation conducted by Guha (2016) has an overall accuracy of 88.75% for 1989 image and 86.25% for 2010 images, and the overall Kappa coefficient of 0.81 and 0.76. Based on the area of mangrove forest, the range of mangrove vegetation index in Perancak Estuary is 0.0025 – 0.78 (Table 3-1). This range of NDVI values differs from Prameswari et al. (2015), where the minimum value of NDVI obtained from the measurement using ALOS AVNIR-2 image data is -0.723 and maximum of 0.530 with the standard deviation 0.127. A variety of vegetation index has been developed by retrieving vegetation International Journal of Remote Sensing and Earth Science Vol. 14 No. 2 December 2017 141 Amandangi Wahyuning Hastuti et al. density from optical remote sensing. Li et al. (2007) are used the most common one method is NDVI to predict the biomass of trees. NDVI is based on the characteristics that vegetation has noticeable absorption in the near read infrared spectrum. In addition to NDVI, there are also several image data processing methods for determining vegetation index such as Simple Ratio (SR), Triangular Vegetation Index (TVI), Enhanced Vegetation Index (EVI), Ratio Vegetation Index (RVI), and Soil Adjusted Vegetation Index (SAVI) (Frananda et al. 2015). Furthermore, the vegetation index values were used for the determination value of AGB, BGB, TAB, TCS and ACS using equation (4) to (9) above. Based on Table 3-1, it can be seen that the AGB value is 38.60±20.79 ton ha1 and the value of BGB is 8.60±4.24 ton ha-1. Based on the value of AGB and BGB it can be said that the AGB is bigger than BGB. This is consistent with the results of a study conducted by (BPOL 2015) which states that by conducting field measurements in Perancak Estuary, the average value of AGB is higher than the BGB’s. The value of AGB and BGB on this research still representative with the research about assessment of mangrove forest carbon stock monitoring in Indonesia conducted by Yenni, et al (2014), which the average of AGB in Subang, West Jawa 1.65 ton ha-1; Cilacap, Central Java 4.62 ton ha-1; Badung, Bali 12.87 ton ha-1; and Merauke, Papua 3.97 ton ha-1. AGB will give the best estimation using diameter breast height (DBH) as a parameter (Alemayehu et al. 2014). The determination of the AGB value is an important step in the planning of the protection and utilization of natural mangrove resources (Meideros and Sampaio 2008). The differences in AGB and BGB valuescan also be seen among mangrove species, depending on geographical location, tree density and ecology (Sahu et al. 2016; Alongi 2012). Total accumulated biomass (TAB) is the total amount of biomass on above and below the soil surface. The value TAB in Perancak Estuary is 18.67 ton ha-1. If the ratio between BGB and AGB is bigger, then the plant undergoes substantial root growth (below ground) which is quite dominant rather than trunk growth (above ground). Plant biomass is closely related to photosynthesis, biomass increases as plants absorb CO2 from the air and convert it into organic compounds through photosynthesis. Biomass in each part of the plant increases proportionately with the larger diameter of the tree. The high ability of trees to store carbon free from air depends on the diameter of trees (Imani et al. 2017) and tree height (Fu and Wu 2011). Table 3-1: Average values of AGB, BGB, TAB, ACS and ACS in Perancak Estuary Average Values 142 Statistics (t ha-1) Mean±SD Max Min NDVI 0.63±0.11 0.78 0.0025 AGB 38.60±20.79 93.43 0 BGB 8.60±4.24 19.11 0 TAB 47.20±25.03 112.54 0 TCS 22.18±11.76 52.89 0 ACS 81.41±43.18 194.12 0 International Journal of Remote Sensing and Earth Science Vol. 14 No. 2 December 2017 Carbon Stock Estimation using Remote Sensing ..... The estimated of total carbon stock value in the Perancak Estuary is 22.18±11.76 tonC ha-1 and the amount of CO2 sequestration estimation is 81.41±43.18 tonC ha-1. The estimated total carbon stock value obtained from the remote sensing measurements is much higher than the field measurements conducted by (Sidik et al. 2014) in Perancak Estuary. Regarding of (Sidik et al. 2014), there is a difference in carbon stock value between natural mangrove forest and replantation mangroves in ex-ponds. Carbon stock produced by natural mangroves is higher than re-plantation mangroves in ex-ponds. The carbon stock value in natural mangrove forest is 171±43 MgC ha-1 while the carbon stock from mangrove that grows in ex-pond is 52±15 MgC ha-1. The estimated value of carbon stocks in Perancak Estuary can be done by remote sensing approach with a good result, even the value is still low compared with other studies (Table 3-2). The average biomass of live trees found from (Sitoe et al. 2014) is below the lower limit 58.38±19.1Mg ha-1 and the average carbon 28.02±9.2 Mg ha-1. Table 3-2 shows, the estimation of carbon stock using remote sensing data at Perancak estuary is still reasonable. Existing mangrove forest is usually producing a higher value of carbon stock comparing to the replanted mangrove. Table 3-2: Carbon stock estimation of mangrove vegetation found in the literature Reference This study Estrada and Soares (2016) Hutchinson et al. (2014) Hamdan et al. (2013) Komiyama et al. (2008) Carbon Stock (tonC ha-1) 22.18±11.76 78.0±64.5 74.5±54.6 1.01 – 259.68 78.3±51.0 Variations of carbon stock value depend on several physical factors of environmental chemistry, the diversity and density of existing plants, soil types and how they are managed. Besides on those factors, mangroves in Perancak Estuary are from rehabilitated mangroves in 2001 and 2009. The dominant mangroves found in Perancak Estuary are Rhizopora mucronata, Rhizopora apiculata, Sonneratia alba, Avicennia alba and Avicennia marina (Proisy et al. 2015). Mangroves in Perancak Estuary area grow on the mud-soil type substrate mixed with organic material (Kartikasari and Sukojo 2015). The size of the carbon stored in vegetation depends on the amount of biomass contained in the tree, soil fertility and the absorption of the vegetation (Ati et al. 2014). Figure 3-1 shows the location of natural and rehabilitation (re-plantation) mangroves in Perancak Estuary. In the natural mangroves dominated by Avicennia sp. and Sonneratia alba. While the dominant mangrove grown in the former location of ponds (rehabilitation mangroves) estimated to be around 8-10 years old is Rhizopora sp. (BPOL 2015). Based on the result of this research, the variation values of carbon stock in Perancak Estuary due to the age of the relatively young mangrove trees. Almost 70% of carbon stock variability is explained by age (Estrada and Soares 2017), species, management regime, as well as the climate (Kairo et al. 2008). It has been reported that the highest carbon stock for > 80-year-old R. apiculata dominated mangrove forest was 230.0 t C ha-1 (Putz and Chan 1986) while those of 20- and 28-year-old Rhizopora forests were 114 and 105.9 t C ha-1 respectively (Ong et al. 1995). The standing biomass International Journal of Remote Sensing and Earth Science Vol. 14 No. 2 December 2017 143 Amandangi Wahyuning Hastuti et al. for the 12-year-old Rhizopora mucronata plantation was 106.7±24.0 t/ ha, giving a biomass accumulation rate of 8.9 t/(ha year) (Kairo et al. 2008). The condition of mangrove vegetation in Perancak Estuary shown in Figure 3-2. Figure 3-1: Fine-scale maps of changes in mangrove cover between 2001 (left) and 2014 (right) over the whole Perancak Estuary (Rahmania et al. 2014) Figure 3-2: Mangrove vegetation condition in Perancak Estuary which surrounded by river and ponds (a) nypa sp. which grow in ex-ponds (b) active ponds (c) avicennia sp. (d) sonneratia sp. (e) rhizopora sp. (f) rhizopora sp. re-plantation in the ex-ponds (g) natural mangrove vegetation that grows along the river (h) natural mangrove vegetation Figure 3-3: Map of value and distribution of NDVI, AGB, BGB, TAB, TCS and (ACS) in Perancak Estuary in 2015 144 International Journal of Remote Sensing and Earth Science Vol. 14 No. 2 December 2017 Carbon Stock Estimation using Remote Sensing ..... Figure 3-3 shows the values and distributions of NDVI, AGB, BGB, TAB, TCD and ACS in the Perancak Estuary. Spatially viewed of the distribution of NDVI value, the range value of AGB, BGB, TAB, TCD and ACS are different. Variations of the distribution of these values are speculated because of the different types of mangroves, namely mangroves that grow naturally and rehabilitated mangroves. Value of NDVI, the average value of AGB, BGB, TAB, TCD and ACS in natural mangroves is higher than mangroves that grow in the former location of ex-fishponds and rehabilitation mangrove (Figure 3-3). NDVI value generated for each pixel on the images was converted to carbon stock will give a different result. The higher NDVI will produce a high biomass value as well. The relationship between NDVI and biomass is also reported (Hamdan et al. 2013), which states that there are different relationships between AGB and NDVI. Linear regression produced higher correlation coefficients but not represent the real distribution, especially when the NDVI value approaches 0. Optic data approach commonly used vegetation indices for mangrove biomass estimation (Sahu et al. 2016; Hamdan et al. 2013; Wicaksono et al. 2011; Li et al. 2007) and for the common forest (Laurin et al. 2016; Jha et al. 2015). Vegetation indices are highly related to net primary productivity (Li et al. 2007). Given that optical imagery cannot obtain tree height as a crucial parameter in biomass estimation, detailed and accurate estimation of mangrove forest AGB still presents a challenge when parameters derived from optical imagery are applied to biomass estimation. Studies of plant allometry indicated that biomass is determined not only by canopy parameters but also by other factors such as wood density, trunk taper and tree height (Komiyama 2008; Chave et al. 2006; Niklas 1995) which are closely relevant to the floristic characteristics of the species. The results of research on estimation of carbon stock by using remote sensing method still require the accuracy and field test to mangrove type and density. The NDVI method is not the best method for estimation of carbon stocks, but the method has relatively consistent accuracy at various levels of radiometric correction (Wicaksono et al. 2011). According to Frananda (2015), measuring the vegetation index using TVI has the best accuracy. In addition, the use of highresolution image data is necessary to be applied in assessing the condition and dynamics of mangroves properly (Rodriguez and Feller 2004), classification of tree species based on their reflectance value (Wang et al. 2004; Dahdouh-Guebas et al. 2005) and the other necessities. 4 CONCLUSION Estimated carbon stocks in Perancak Estuary can be done using remote sensing data with the good enough result, which is 22.18±11.76 tonC ha-1. The use of NDVI is still relevant for biomass and carbon stock estimation in a mangrove ecosystem. Moreover, the difficulty of distinguishing mangrove vegetation is due to relatively small research areas. Field measurement and high-resolution satellite data in carbon stock estimation, especially in Perancak Estuary still need further study to improve the accuracy of carbon stock estimation results. International Journal of Remote Sensing and Earth Science Vol. 14 No. 2 December 2017 145 Amandangi Wahyuning Hastuti et al. ACKNOWLEDGEMENTS The authors are thankful to Institute of Marine Research and Observation. Acknowledge to I Nyoman Radiarta, Nuryani Widagti and Frida Sidik for advice and support. We also thank to the reviewers, Prof. Dr. I Nengah Surati Jaya, M.Agr. and Dr. Indah Prasasti for their constructive comments and suggestions. Cerón-Bretón RM, Cerón-Bretón JG, SánchezJunco RC, et al., (2011), Evaluation of Carbon Sequestration Potential in Mangrove Forest at Three Estuarine Sites in Campeche Mexico. Int J Energy Environ 5(4):487-494. Chave J., Muller-Landau HC, Baker TR, et al., (2006), Regional and Phylogenetic Variation of Wood Density Across 2456 Neotropical Tree Species. Ecol. Appl. 16, REFERENCES 2356–2367. Abohassan RAA, Okia CA, Agea JG, et al., (2012), Perennial Biomass Production in Arid Dahdouh-Guebas F., Hiel EV, Chen JCW, et al., (2005), Qualitative Distinction of Mangrove Systems on the Red Sea Coast of Congeneric and Introgressive Mangrove Saudi Arab. Environ Res J 6(1): 22-31. Species Adame MF, Kauffman JB, Medina I., et al., in Mixed Patchy Forest Assemblages using High Spatial Resolution (2013), Carbon Stocks of Tropical Coastal Remotely Wetlands within the Karstic Landscape of Systematics and Biodiversity, 2, 113- 119. the Mexican 8(2):e56569. Caribbean. PLoS doi.10.1371/ One journal. pone.0056569. Sensed Imagery (IKONOS). Deng S., Shi Y., Jin Y., et al., (2011), A GISBased Approach for Quantifying and Mapping Carbon Sink and Stock Values of Alemayehu F., Richard O., James KM, et al., Forest Ecosystem: A case study. Energy (2014), Assessment of Mangrove Covers Procedia, Change and j.egypro.2011.03.263. Biomass in Mida Creek, Kenya. Open Journal of Forestry, 4:398413. http://dx.doi.org/ 10.4236/ojf.2014.44045. 5:1535-1545.doi:10.1016/ Donato D., Kauffman JB, Murdiyarso D., et al., (2012), Mangrove adalah Salah Satu Hutan Terkaya Karbon di Kawasan Tropis (No. Alongi DM, (2012), Carbon Sequestration in CIFOR Infobrief no. 12, p. 12p). Center for Mangrove Forests. Carbon Management International Forestry Research (CIFOR), 3(3), 313-322. Bogor, Indonesia. Ati RNA, Rustam A., Kepel TL, et al., (2014), Stok Karbon dan Stuktur Komunitas Mangrove (2011), Sebagai Blue Carbon di Tanjung Lesung, Carbon-Rich Forests in the Tropics. Nature Banten. Jurnal Segara. Vol. 10 No. 2, Geosci 4: 293-297. Desember 2014: 119-127. Climate Change. Hofstra University – USA. Bouillon S., Borges AV, Eda-Moya EC, et al., (2008), Mangrove Production and Carbon Sink: a Revision of Mangroves Among the Most Estrada GCD, and Soares MLG, (2017), Global Bennington JB., (2009), The Carbon Cycle and Global Patterns of Abgove Ground Carbon Stock and Sequestration in Mangroves. Anais de Academia Brasileira de Ciencias 89(2): 973-989. Budget Frananda H., Hartono, Jatmiko, RH, (2015), Estimates. Global Biogeochem Cycles 22:1- Komparasi Indeks Vegetasi untuk Estimasi 12. doi:10.1029/ 2007GB003052. Stok Karbon Hutan Mangrove Kawasan Cairns MA, Brown S., Helmer EH, et al., (1997), Rootbiomass Allocation 146 Donato DC, Kauffman JB, Murdiyarso D., et al., in Segoro Anak pada Kawasan Taman the World's Nasional Alas Purwo Banyuwangi, Jawa Upland Forests. Oecologia (1997) 111:1 - Timur. Majalah Ilmiah Globe. Vol. 17 No 2. 11. Desember 2015: 113 – 123. International Journal of Remote Sensing and Earth Science Vol. 14 No. 2 December 2017 Carbon Stock Estimation using Remote Sensing ..... Fu W., Wu Y., (2011), Estimation of Above (2001), Climate Change 2001. Intergovernmental Panel on Climate Change Trees Based on Canopy Diameter and Tree National Height. 3rd International Conference on Greenhouse Gas Inventories Programme downloaded on 15 June 2017 Environmental Science and Information from Application ipccreports/tar/wg3/index.php?idp=477S. Procedia Technology (ESIAT Environmental 2011). Sciences, 10 RKKV, Madla http://www.ipcc.ch/ Jha CS, Fararoda R., Rajashekar, et al., (2015), Spatial Distribution of Biomass in Indian (2011) 2189 - 2194. Giri IPCC., Ground Biomass of Different Mangrove VR., (2017), Study and Forests using Spectral Modelling Evaluation of Carbon Sequestration using (No.Technology Trends: Multi-Scale Remote Remote Sending and GIS: A Review on Sensing Using Optical Sensorsno. 3, 138p). Various Techniques. International Journal The International Centre for Integrated of Civil Engineering and Technology, 8(4), Mountain Development (ICIMOD), Nepal. Kairo JL, Lang’at JKS, Dahdouh-Guebas F., et 287-300. Green EP, Mumby PJ, Edwards AJ, et al., (1998), al., (2008), Structural Development and The Assessment of Mangrove Areas using Productivity High Plantations in Kenya. For. Ecol. Manag. Resolution Multispectral Airborne Imagery. Journal of Coastal Research, 14(2), 433-443. Royal Palm Beach (Florida), ISSN 0749-0208. Replanted Mangrove 255, 2670-2677. Kartikasari AD, Sukojo BM, (2015), Analisis Persebaran Ekosistem Hutan Mangrove Guha S., (2016), Capability of NDVI Technique in Detecting Mangrove Vegetation. International Journal of Advanced Biological Research. Vol. 6(2): 253-258. Manggunakan Citra Landsat-8 di Estuari Perancak Bali. GEOID. Vol. 11 No. 01. Khan MNI, Suwa R., Hagihara A., (2007), Carbon and Nitrogen Pools in a Mangrove Stand of Hamdan O., Khairunnisa MR, Ammar AA, et al., (2013), of Mangrove Carbon Kandelia obovate (S., L.) Yong: vertical Stock distribution in the soil-vegetation system. Assessment by Optical Satellite Imagery. Wetl. Ecol. Manag 15(2):141-153. doi:10. Journal of Tropical Forest Science 25(4): 1007/s11273-006-9020-8. 554-565. Komiyama A., Ong JE, Poungparn S., (2008), Huete A., Didan K., Leeuwen WV, et al., (2011), Allometry, Biomass, and Productivity of MODIS Vegetation Indices. Land Remote Mangrove Forests. Aquatic Botany. Vol. 89: Sensing 128–137. and Global Environmental Change. Springer. New York. Kristensen E., Bouillon S., Dittmar T., et al., Hutchison J., Manica A., Setnam R., et al., (2014), Predicting Global Patterns in Mangrove Forest Biomass. Conserv Let 7:233-240. Height-Diameter Allometry Biomass in and Tropical Above Montane Forests: Insights from the Albertine Rift in Africa. PLOS Organic Carbon Dynamics in Mangrove Ecosystem: a review. Aquat Bot 89:201-219. doi:10.1016/j.aquabot.2007.12.005. Imani G., Boyemba F., Lewis S., et al., (2017), Ground (2008), ONE 12(6): e0179653. Laffoley DDA, Grimsdicth G., (2009), The Management of Natural Coastal Carbon Sink. IUCN Gland, Switzerland. Li XA, Yeh GO, Wang S., et al., (2007), Regression and Analytical Models for Estimating https://doi.org/10.1371/ Mangrove Wetland Biomass in South China journal.pone.0179653. Using Radarsat Images. International Journal of Remote Sensing 28:5567-5582. International Journal of Remote Sensing and Earth Science Vol. 14 No. 2 December 2017 147 Amandangi Wahyuning Hastuti et al. Lu D., (2006), The Potential and Challenge of Detection Area Estimation of Mangroves Along the Sahara Desert Coast. International Journal of Remote Sensing. Remote Vol. 27(7), 1297-1328. 10.3390/rs8060512. Machiwa JF, Hallberg RO., (2002), An Empirical Sensing. 8: 512; doi: Parmeswari AASG, Hariyanto T., Sidik F., (2014), Model of the Fate of Organic Carbon in a Analisis Mangrove by Menggunakan Citra Satelit ALOS AVNIR-1 Anthropogenic Activity. Ecol Model 147:69- (Studi Kasus: Estuari Perancak, Bali). Forest Partly Affected 83. Doi:10.1016/S0304-3800(01)00407-0. Manuri S., Putra CAS, Saputra AD., (2011), Teknik Pendugaan Hutan. Merang Cadangan REDD Pilot Indeks Vegetasi Mangrove Geiod Vol. 11, No. 1 Agustus 2015. Proisy C., Rahmania R., Viennois G., et al., Karbon (2015), Monitoring Changes on Mangroves Project, Coats using High Resolution Satellite German International Cooperation – GIZ. Images. A Case Study in The Perancak Palembang. Estuary, Bali. 12th Biennial Conference of Mariana, Felix F., Sukendi, et al., (2015), Pan Ocean Remote Sensing Conference Estimation of Mangrove Forest’s Carbon (PORSEC 2014). 04 - 07 November 2014. Stock in Kuala Indragiri Coastal Riau Province – Indonesia. International Journal Bali - Indonesia. Putz F., Chan HT, (1986), Tree Growth, of Oceans and Oceanography. 9(2), 117- Dynamics, and Productivity in a Mature 126. ISSN 0973-2667. Mangrove Meideros TCC, Sampaio E., (2008), Allometry of Forest in Malaysia. Forest Ecology and Management 17: 211-230. Above Ground Biomasses in Mangrove Rahmania R., Proisy C., Viennois G., et al., Species in Itamaraca, Pernambuco, Brazil. (2015), 13 Years of Changes in the Extent Wetlands Ecology and Management 16 (4): and 323-330. Shrimp Farming Abandonment, Bali. 2015 Mitra A., Sengupta K., Banerjee K., (2011), Physiognomy Multitemporal Above-Ground (Multi-Temporal). Structures in Dominant Ecol Manag 261:1325-1335. doi:10.1016/j.foreco. 2011.01.012. (2009), Carbon Storage in Mangrove and Ecosystems Account from Plots CIFOR-Center for Research (CIFOR), – in A Preliminary Indonesia. After Remote Sensing IEEE Images Explore. doi:10.1109/Multi-Temp.2015. 7245801. Rodriguez W., Feller IC, (2004), Mangrove Twin Cays, Belize Using Aerial Photography and IKONOS Satellite Data. Atoll Research Bulletin. 513: 1-22. No. Sahu SC, Kumar M., Ravindranath NH, (2016), Forestry Carbon Stocks in Natural and Planted Indonesia. Mangroves Forests of Mahanadi Mangrove International Bogor, Mangroves Landscape Characterization and Change in Mudiyarso D., Donato D., Kauffman JBD, et al., Peatland of 8th International Workshop on the Analysis Standing Biomass and Carbon Storage of Mangrove Tress in the Sundarbans. For Working paper 48. Wetland, East Coast of India. Current Niklas KJ, (1995), Size-Dependent Allometry of Tree Height, diameter and trunk-taper. Ann. Bot. 75, 217–227. Science, Vol. 110, No. 12, 25 June 2016. Sapit D., Damrong S., Ladawan P., et al., (2011), An Assessment of Stand Structure and Ong JE, Gong WK, Clough BF, (1995), Structure Carbon Storage of a Mangrove Forest in and Productivity of a 20-year-old Stand of Thailand. IUFRO World Ser 29:28-30. Rhizopora Apiculata BI. Mangrove forest. Saputra GR, (2007), Model Penduga Potensi Journal of Biogeography 22: 417-424. Otero V., Quisthoudt K.. Koedam N., et al., (2016), 148 and Remote Sensing-Based Biomass Estimation. Mangroves at Their Limits: Hutan Rakyat Menggunakan Citra Aster dan Sistem Beberapa Informasi Wilayah Geografis Kabupaten di Bogor International Journal of Remote Sensing and Earth Science Vol. 14 No. 2 December 2017 Carbon Stock Estimation using Remote Sensing ..... Bagian Barat. (Skripsi). Bogor (ID): Departemen Manajemen Hutan Fakultas Kehutanan, IPB (Bogor Agricultural University), Bogor. International Indonesia Programme. Vicharnakorn P., Shrestha RP, Nagai M., et al., (2014), Carbon Stock Assessment using Sidik F., Widagti N., Kadarusman HP, et al., (2015), Laporan Wetlands Teknis Penelitian dan Remote Sensing and Forest Inventory Data in Savannakhet, Lao PDR. Remote Pengambangan: Aplikasi Sistem Observasi Sensing. 6: 5452-5479. doi: 10.3390/ Adaptasi Mangrove Terhadap Perubahan rs6065452. Iklim (Technical Report: Enhancement of Wang L., Sousa WP, Gong P., (2004), Integration of Research for Adaption of Wetlands in Object-Based Indonesia to Projected Impacts of Sea Level Classification for Mapping Mangroves with Rise). Balai Penelitian dan Observasi Laut. IKONOS imagery. Int. J. Remote Sens. 25: Pusat Pengkajian dan Perkeyasaan Teknologi Kelautan dan Perikanan. Badan Pixel-Based 5655-5668. Westlake DF, (1963), Penelitian dan Pengembangan Kelautan Productivity. dan Perikanan. Kementerian Kelautan dan 385-425. Perikanan. and Comparison Biological of Plant Reviews. 38(3): Wicaksono P., Danoedoro P., Hartono H., et al., Sitoe AA, Mandlate LJC, Guedes BS, (2014), (2011), Preliminary Work of Mangrove Biomass and Carbon Stocks of Sofala Bay Ecosystem Carbon Stock Mapping in Small Mangrove Forests. Forest. 5, 1967-1981; Island using Remote Sensing: Above and 10.3390/f5081967. ISSN 1999-4907. Below Ground Carbon Stock Mapping on Situmorang JP, Sugiatnto S., Darusman, (2016), Medium Resolution Satellite Image. Estimation of Carbon Stock Stands using Proceedings of SPIE: Vol. 8174. Remote EVI Sensing for Agriculture, Ecosystems, and and NDVI Vegetation Index in Production Forest of Lembah Seulawah Hydrology XIII. International Society Sub-District, Aceh Indonesia. Aceh Int. Optics and Photonics. Sci. Technol, 5(3):126-139. doi: Sutaryo D., (2009), Perhitungan Biomassa, Sebuah untuk Perdagangan Studi Karbon. Yenni V., Parwati E., Winarso G., et al., (2014), Assessment of Mangrove Forest Carbon 10.13170/aijst. 5.3.5836. Pengantar for Karbon Bogor dan (ID) : Stock Remote Monitoring Sensing of Indonesia Approach. using SAFE Workshop. Tokyo. 1st December 2014. International Journal of Remote Sensing and Earth Science Vol. 14 No. 2 December 2017 149 Amandangi Wahyuning Hastuti et al. 150 International Journal of Remote Sensing and Earth Science Vol. 14 No. 2 December 2017