The National Research and Innovation Agency's Ejournal Portal
Not a member yet
7815 research outputs found
Sort by
ANALYSIS OF LAND USE SPATIAL PATTERN CHANGE OF TOWN DEVELOPMENT USING REMOTE SENSING
No full textThe Assessment of the physical character of a city is considered relatively easier than the social-cultural aspects. It is important to recognize the type of city form and to predict the behavior of people in the city and its surrounding. Due to those characteristics, the study of the pattern of physical development of the city is required. The objective of research is to analyze the change of spatial pattern of the city due to the city growing by remote sensing. The multitemporal data of Landsat 5/7/8 year 2000, 2006 and 2015 in Jabodetabek area were used. The classification technique had been done and it produced five classes of land uses. Those are water, built-up area, vegetation, other land use and no data. The results of the analysis in Jabodetabek area (Jakarta, Bogor, Depok, Tangerang and Bekasi) show that there was land use changes from vegetation and other land use area to built-up area with an average accuracy of 78% in each year. The pattern of physical development of the city looks linear from year 2000 until year 2006, which is confirmed as concentric pattern from year 2006 to 2015. Based on those analysis, it confirmed that the city development in Jakarta as the center was influenced by the spatial land development of the surrounding cities of Depok, Bogor, Bekasi and Tangerang. The pattern of spatial development from 2000 to 2006 in Bogor, Bekasi and Depok areas is Linear pattern, whereas from 2006 - 2015 the pattern of spatial development shows Propagation Concentric pattern. For Tangerang Region in 2000-2015 its development is patterned Propagation Concentric
DEVELOPMENT OF ANNUAL LANDSAT 8 COMPOSITE OVER CENTRAL KALIMANTAN, INDONESIA USING AUTOMATIC ALGORITHM TO MINIMIZE CLOUD
No full textSince January 2013, Landsat 8 data can be freely accessed from LAPAN, making it possible to use the all available Landsat 8 data to produce the cloud-free Landsat 8 composite images. This study used Landsat 8 archive images in 2015, Operational Land Imager (OLI) sensor in 30 meter resolution, geometric correction level of L1T. The eight data in L1T of 118-062, southern part of Central Kalimantanwere used to produce a cloud-free composite image. Radiometric correction using Top of Atmosphere (TOA) and Bidirectional Reflectance Distribution Function (BRDF) algorithm to produce reflectance images have been applied, and then the most cloud-free pixels were selected in composite result. Six composite methods base on greens, open area and haze indices were compared, and the best one was selected using visual analysis. The analysis shows that the composite algorithm using Max (Max (NIR, SWIR1)/Green) produces the best image composite
MODIS STANDARD (OC3) CHLOROPHYLL-A ALGORITHM EVALUATION IN INDONESIAN SEAS
No full textThe MODIS-estimated chlorophyll-a information was widely used in some operational application in Indonesia. However, there is no information about the performance of MODIS chlorophyll-a in Indonesian seas and there is no data used in development of algorithm was taken in Indonesian seas. Even the algorithm was validated in other area, it is important to know the performance of the algorithm work in Indonesian seas. Performance of MODIS Standard (OC3) algorithm at Indonesian seas was analyzed in this paper. The in-situ chlorophyll-a concentration data was collected during MOMSEI (Monsoon Offset Monitoring and Its Social and Ecosystem Impact) 2012 Cruise 25th April – 12th  May 2012 and also from archived data of the Research and Development Center for Marine Coastal Resources, Agency of Marine and Fisheries Research and Development, Indonesian Ministry of  Marine Affairs and Fisheries. The in-situ data used in this research is located in Indian Ocean the west of Sumatera part and Pacific Ocean the north of Papua Province part. Satellite data which is used is Ocean Color MODIS Level-2 Product that downloaded from NASA and MODIS L-0 from LAPAN Ground Station. MODIS Level 0 from LAPAN then processed to Level-2 using latest SeaDAS Software. The match-up resulted the MNB(%) is -4.8% that means satellite-estimated was underestimate in 4.8 % and RMSE is 0.058. When the data was separated following to the data source, the correlation and trend line equation became better. From MOMSEI Cruise data, the MNB(%) was -18.8% and RMSE 0.05. From Pacific Ocean Data, MNB (%) was -27 % and RMSE 0.049. From SONNE Cruise 2005, MNB (%) was -27 % and RMSE 0.049. MODIS standard algorithm is work well in Indonesia case-1 seawaters, which contain chlorophyll-a only, and derived that influence to the electromagnetic wave
DETECTING THE AFFECTED AREAS OF MOUNT SINABUNG ERUPTION USING LANDSAT 8 IMAGERIES BASED ON REFLECTANCE CHANGE
No full textThe position of Indonesia as part of a "ring of fire" bringing the consequence that the life of the nation and the state will also be influenced by volcanism. Therefore, it is necessary to map rapidly the affected areas of a volcano eruption. Objective of the research is to detect the affected areas of Mount Sinabung eruption recently in North Sumatera by using optical images Landsat 8 Operational Land Imager (OLI). A pair of Landsat 8 images in 2013 and 2014, period before and after eruption, was used to analysis the reflectance change from that period. Affected areas of eruption was separated based on threshold value of reflectance change. The research showed that the affected areas of Mount Sinabung eruption can be detected and separated by using Landsat 8 OLI images based on the change of reflectance value band 4, 5 and NDVI. Band 5 showed the highest values of decreasing and band 4 showed the highest values of increasing. Compared with another uses of single band, the combination of both bands (NDVI) give the best result for detecting the affected areas of volcanic eruption
MACHINE LEARNING-BASED MANGROVE LAND CLASSIFICATION ON WORLDVIEW-2 SATELLITE IMAGE IN NUSA LEMBONGAN ISLAND
No full textMachine 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%
MANGROVE ABOVE GROUND BIOMASS ESTIMATION USING COMBINATION OF LANDSAT 8 AND ALOS PALSAR DATA
No full textMangrove ecosystem is important coastal ecosystem, both ecologically and economically. Mangrove provides rich-carbon stock, most carbon-rich forest among ecosystems of tropical forest. It is very important for the country to have a large mangrove area in the context of global community of climate change policy related to emission trading in the Kyoto Protocol. Estimation of mangrove carbon-stock using remote sensing data plays an important role in emission trading in the future. Estimation models of above ground mangrove biomass are still limited and based on common forest biomass estimation models that already have been developed. Vegetation indices are commonly used in the biomass estimation models, but they have low correlation results according to several studies. Synthetic Aperture Radar (SAR) data with capability in detecting volume scattering has potential applications for biomass estimation with better correlation. This paper describes a new model which was developed using a combination of optical and SAR data. Biomass is volume dimension related to canopy and height of the trees. Vegetation indices could provide two dimensional information on biomass by recording the vegetation canopy density and could be well estimated using optical remote sensing data. One more dimension to be 3 dimensional feature is height of three which could be provided from SAR data. Vegetation Indices used in this research was NDVI extracted from Landsat 8 data and height of tree estimated from ALOS PALSAR data. Calculation of field biomass data was done using non-decstructive allometric based on biomass estimation at 2 different locations that are Segara Anakan Cilacap and Alas Purwo Banyuwangi, Indonesia. Correlation between vegetation indices and field biomass with ALOS PALSAR-based biomass estimation was low. However, multiplication of NDVI and tree height with field biomass correlation resulted R2 0.815 at Alas Purwo and R2 0.081 at Segara Anakan. Low correlation at Segara anakan was due to failed estimation of tree height. It seems that ALOS PALSAR height was not accurate for determination of areas dominated by relative short trees as we found at Segara Anakan Cilacap, but the result was quite good for areas dominated by high trees. To improve the accuracy of tree height estimation, this method still needs validation using more data
ANALYSIS ON THE QUALITY OF AEROSOL OPTICAL THICKNESS DATA DERIVED FROM NPP VIIRS AND AQUA MODIS OVER WESTERN REGION OF INDONESIA
No full textPreliminary analysis on quality data of Aerosol Optical Thickness/Depth or AOT/AOD derived from NPP VIIRS EDR (Environmental Data Record) has been done in previous work. Qualitative analysis of the previous work revealed that AOT data of VIIRS had insufficient quality due to some factors such as sun glint and cloud cover. However the accuracy of AOT VIIRS data over western area of Indonesia has not been investigated. Therefore this paper describes further analysis on AOT VIIRS data quality and accuracy. Comparison with AOT derived from Aqua MODIS data was implemented since AOT of MODIS has verified well with AOT data from field observation. Examination on cloud masking intermediate product of VIIRS was done for its importance in AOT data processing and persistent cloud cover obstacle over Indonesia. We used VIIRS and MODIS data archieved by LAPAN ground station. Further analysis on sun glint and cloud masking images indicates that these two intermediate products predominantly affect the quality of AOT from VIIRS and MODIS over the study areas. Compared with AOT of MODIS, AOT of VIIRS seems to result more pixels consisting AOT information over the same area and date. The statistical results showed that AOT values of VIIRS highly correlated with AOT values of MODIS with R2 of 78%. The accuracy of AOT derived from VIIRS was adequate as indicated by RMSE of 0.0977 or less than 0.5 for the samples over Sumatra, Borneo, and Java islands. Visual comparison of AOT images indicates that VIIRS data could result more detailed AOT values than MODIS data. Therefore the AOT of VIIRS data could be recommended for further applications in western area of Indonesia.  
THE USE OF HIGH RESOLUTION IMAGES TO EVALUATE THE EVENT OF FLOODS AND TO ANALYSIS THE RISK REDUCTION CASE STUDY: KAMPUNG PULO, JAKARTA
No full textThe flood hit Kampung Pulo region in almost every year. This disaster has caused the evacuation of some residents in weeks. Given the frequency of occurrence is quite high in the region it is necessary to do a study to support disaster risk reduction. This study aimed to evaluate the incidence of flooding that occurred in Kampung Pulo in terms of topography, river conditions, characteristics of the building, and socioeconomic conditions. Methods of study include geomorphology analysis, identification of areas of stagnant, the estimated number of people exposed, the estimation of socio-economic conditions of the population, as well as determining the location of an evacuation. The data used is high-resolution remote sensing imagery is QuickBird and SPOT-6. It also used the results of aerial photography using Unmanned Aerial Vehicle (UAV). Aerial photography was conducted on January 18, 2013, which is when the serious flooding that inundated almost the entire region of Kampung Pulo. Information risk level of buildings and population resulting from this study were obtained by using GIS. The results obtained from this study can be used to develop recommendations and strategies for flood mitigation in Kampung Pulo, Jakarta. One of them is the determination of the location for vertical evacuation plan in the affected areas.
 
COMPARISON OF THE VEGETATION INDICES TO DETECT THE TROPICAL RAIN FOREST CHANGES USING BREAKS FOR ADDITIVE SEASONAL AND TREND (BFAST) MODEL
No full textRemotely sensed vegetation indices (VI) such as the Normalized Difference Vegetation Index (NDVI) are increasingly used as a proxy indicator of the state and condition of the land cover/vegetation, including forest. However, the Enhanced Vegetation Index (EVI) on the outcome of forest change detection has not been widely investigated. We compared the influence of using EVI and NDVI on the number and time of detected changes by applying Breaks for Additive Seasonal and Trend (BFAST), a change detection algorithm. We used MODIS 16-day NDVI and EVI composite images (April 2000-April 2012) of three pixels (pixels 352, 378, and 380) in the tropical peat swamp forest area around the flux tower of Palangka Raya, Central Kalimantan. The results of BFAST method were compared to the Normalized Difference Fraction Index (NDFI) maps and the maps were validated by the hotspot of the Infrastructure and Operational MODIS-Based Near Real-Time Fire(INDOFIRE). Overall, the number and time of changes detected in the three pixels differed with both time series data because of the data quality due to the cloud cover. Nonetheless, we found that EVI is more sensitive than NDVI for detecting abrupt changes such as the forest fires of August 2009-October 2009 that occurred in our study area and it was verified by the NDFI and the hotspot data. Our results demonstrated that the EVI for forest monitoring in the tropical peat swamp forest area which is covered by intense cloud cover is better than that NDVI. Nonetheless, further research with improving spatial resolution of satellite images for application of NDFI is highly recommended. 
ORTORECTIFICATION OF SPOT-4 DATA USING RATIONAL POLYNOMIAL COEFFICIENTS
No full textOrthorectification of satellite imagery can be done in two ways i.e., rigorous sensor model and the approximation model of the satellite’s orbit. Dependence on physicalparameters, to make rigorous sensor model is more complicated and difficult to apply. The approximation model can be either Rational Polynomial Coefficients (RPC) model or parallel projection system. RPC is a mathematical model which is not depends on the sensor. It is used to improve the positioning accuracy when the parameter of the physical sensor model is unknown. This study assessed orthorectification of SPOT-4 using the RPC model with 7 coefficients. Root Mean Square Error (RMSE) of GCPs obtained from the study was less than 1 pixel. RPC did not depend on physical and satellite orbit parameters. Thus the RPC was simpler and easier to apply