ABSTRACT. Internal solitary waves (ISW) in the Lombok Strait have been widely known to be generated intensively, energetic, and have high amplitudes. The effect of ISW in the Lombok strait has never been studied for its effect on coastal interactions. This research reveals the phenomena affected by ISW activity using satellite imagery. ISW pattern detected in the chlorophyll-a concentration (Chl-a) image from the GCOM-C/SGLI sensor is accompanied by a high Chl-a around the observation area. Chl-a blooms pattern detected by SGLI sensors was then confirmed with Sea Surface Temperature (SST) images which showed no significant temperature difference to show upwelling phenomena. Investigation results using optical imagery on Sentinel-2 showed river discharge due to floods on Lombok Island. Three different ocean color sensors (SGLI, VIIRS, and MODIS) were used in this study to investigate ISW patterns and high Chl-a due to the suspended matter bias. However, this bias in the Chl-a is confirmed by the high value of the total suspended matter (TSM) data from SGLI sensor. River discharge transport induced by ISW causes suspended matter to spread away from its source. It indicates that ISW can be an important medium influencing suspended matter distribution from coastal areas.

ABSTRACT. Of the many oceanic ecosystems experiencing seasonal and climate-driven changes worldwide, the Arabian Sea (AS) stands out as the one with the most dramatic and extreme changes of all. However, the phytoplankton community distribution pattern and environmental adaptation to the extreme events in the Arabian Sea are less explored. Hence these features of the Arabian Sea and the fact that no comprehensive studies have been attempted to investigate the phytoplankton structure based on their size (micro- >20 μm, nano- 2 – 20 μm and picoplankton- < 2 μm) in this regional basin provided an opportunity to explore and understand the tropical cyclone influence on the phytoplankton size classes (PSC) distribution in the AS. The aim of this study is to investigate the phytoplankton size classes (PSC) distribution pattern in the AS during tropical cyclone events through a three-component approach and reconstructed chlorophyll-a datasets. The study revealed interesting results not reported earlier. The cyclone influences PSC variations observed in the AS revealed that the short extreme events regulate the PSC changes in the Arabian sea by controlling the physical drivers.

ABSTRACT. Remote sensing has proven to be a powerful tool for gaining an overall understanding of natural and anthropogenic phenomena. Satellites have become useful for characterizing and monitoring the continuous prospecting of natural resources. This study aimed to analyse the spatial dynamics and destruction of coral reef areas and the correlation between the reduction in live coral and the population of small islands. Landsat TM, ETM, and OLI-TIRS were used to analyse spatial coral reef dynamics from 1990 to 2019. Image processing included gap-filling, atmospheric correction, geometric correction, creating image composite (true colour), water column correction, unsupervised classification, reclassification and an accuracy assessment. Statistical analysis methods, principal component analysis (PCA) and multiple regression analysis, were conducted to determine the relationship between dynamic population data and the reduction in live coral areas. The effect of population has a positive correlation with the reduction in this area, although it is significant. However, the practice of coral destruction on islands is not related to the practices of residents from a particular island, but it is also conducted by residents from different islands

ABSTRACT. This present study employed a generalized additive model (GAM) to understand and predict the spatial distribution of escolar fish in the Southwestern Indian Ocean. The fishery data was from Taiwan’s longline fishery from 2010 to 2014. The catch per unit effort (CPUE) was calculated and standardized using the Generalized Linear Model (GLM). Five remotely sensed oceanographic data were collected to evaluate the relationship between escolar catch and the environment. The five parameters were oxygen (O), sea surface chlorophyll (SSC), sea surface height (SSH), sea surface salinity (SSS), and sea surface temperature (SST). A step-wise approach was applied to construct the environmental variables. The model showing the least Akaike Information Criterion (AIC) was determined as the best model. The model with the combination of all variables showed the lowest AIC. The optimum environmental ranges for all parameters were as follows: oxygen (205-245 mmol m-3), SSC (0.1-0.85 mg m-3), SSH (0.2-1.5 m), SSS (35.3-35.8 PSU), and SST (17-26.3 oC). Six analyzed months (April, May, June, July, August, and September) showed that escolar fish move northward. The catch distribution of escolar fish showed consistently high catch areas with high predicted CPUE areas.

ABSTRACT. Seaweeds are the one of commercially important marine living renewable resource. Pakistan has appreciable seaweed resources that are still unmapped. The need to preserve and map seaweed sites in Pakistan can be appreciated by realizing the importance of these precious coastal resources. The coast of Pakistan is 1,050 km long and seaweed resources along the coast have great economic potential due to their utilization in food, cosmetics, and industrial uses. They also play a significant role in aquaculture and fish breeding. Many sea species rely on seaweeds for their shelter and food requirements. To protect marine biodiversity and to regulate growth, systematic monitoring and mapping of seaweeds are of vital importance so that the dependent species are able to maintain their biological associations. The main objective of this study was to map and develop indices to extract seaweed resources along Karachi coast using high resolution WorldView-2 satellite data. Traditionally, marine scientists have used sampling methods to describe shallow intertidal water marine habitats but these methods are time intensive. Remote sensing data and GIS tools are efficient and less time consuming for mapping and classifying marine resources. This study mapped most of the seaweed resources in the study area that were further verified through site observations. Keywords: Coastal Resources, Remote Sensing, Seaweed, World View-2

*Corresponding E-Mail: danish769@yahoo.com

ABSTRACT. Multi-satellite remote sensing sources coupled with habitat models are widely being used to obtain comprehensive descriptions of the oceanographic features that structure species’ habitats and to determine their habitat selection patterns. In this study, generalised additive models (GAM) were used to analyse the power of seven remotely sensed oceanographic parameters as predictor variables together with generalised linear model-standardised (GLM) catch per unit effort (CPUE) data as response variable to explain the spawning habitat distribution of mature South Atlantic albacore tuna (ALB) (Thunnus alalunga). Taiwanese longline fishery data were collected for mature fish weighing >14 Kg from 2005 to 2016 between October to March. Eight oceanographic remotely sensed parameters were used: sea surface salinity (SSS), sea surface height (SSH), sea surface chlorophyll (SSC), eddy kinetic energy (EKE), pH, sea surface temperature (SST), oxygen (O), and mixed layer depth (MLD). Based on GAM, boosted regression tree (BRT), and random forest (RF) models, those parameters with R2>10% were selected for further analysis. Model selection was based on Akaike’s Information Criterion (AIC), R2, deviance explained, and generalised cross-validation statistic (GCV). The selected GAM model (R2=0.85) was used to predict ALB CPUE using monthly-binned environmental data to evaluate model performance and to understand their spawning habitat distribution. The 7-parameter model demonstrated that ALB spawning habitat is highly influenced by SSS (36-37 PSU), SSH (0.02-0.1 m), SSC (0.05-0.1 mg/m3), pH (8.04-8.06), EKE (0-0.03 m2/s2), SST (24-26°C), and O (210-218 mmol/m3). Predicted CPUE showed a consistent southward migration of ALB from February and was consistently high between 10° and 30°S and 8° to 40°W, marking it as the main spawning location. The major spawning season for ALB is from November to January. These findings signify the relevance of multi-satellite data-driven habitat models for both habitat prediction and ecological knowledge comprehension which are pivotal for ecosystem-based fishery management for tuna fisheries.

ABSTRACT. This study caries out by implementing the regionally tuned three component abundance-based model to MODIS-Aqua chlorophyll-a data for determining the three size classes i.e., picoplankton, nanoplankton and microplankton in the northern Bay of Bengal during 2015-2019.Small size fraction dominates in oligotrophic waters, whereas large size fraction dominates in nutrient rich waters. In the coastal water microplankton domination observes in southwest monsoon of 2016, 2018 and post monsoon of 2019, whereas the picoplankton dominates in three stations (Inner, Outer, Central) during the study period. Micro-plankton starts dominating nano-plankton and picoplankton when the total chlorophyll-a concentration is 0.38 mg/m3 and >0.46 mg/m3 respectively. Microplankton is low (<20%), nanoplankton are (<40%) and the picoplankton are high (>40%) when the total chlorophyll-a concentrations is low. Pico fraction dominated by 44.65% share where nano and micro fraction are 32.65% and 22.34% respectively in the whole study period are observes.

ABSTRACT. The El Niño Southern Oscillation (ENSO) is an important oceanographic phenomenon that influence the interannual fluctuations of climate and ecosystem productivity in the tropical and subtropical areas, which has an impact on the variability of habitat suitability for many commercial fish species. However, there is a lack of detailed information regarding this major phenomenon and its associated environmental variations impacting the catch and habitat of the economically and ecologically important species greater amberjack (Seriola dumerili) in the Taiwan Strait (TS). Based on a weighted habitat suitability index (HSI) modeling approach using remotely sensed marine environmental data as well as voyage data recorders from Taiwanese fishing vessels (2014–2019), is the primary study aims to better understand the effects of ENSO episodes on catch rates and habitat suitability for greater amberjack in the TS. Using ANOVA, environmental factors had a substantial impact on the greater amberjack habitat and catch rates during the ENSO episodes throughout the seasons. A high catch rates during spring and summer was observed across the southern and northern TS, whereas as in autumn and winter were recorded in the southern TS. Comparing the catch rates, for spring, summer, and autumn were higher (>9.0 kg/h) during El Niño years, whereas in winter the higher catch rates (>12.0 kg/h) during normal years, noticeably the lower catch rates under La Niña years. However, the predicted HSI along the southern and northern TS revealed that the greater amberjack was predominantly distributed in 20–24°N and 24–28°N, respectively. Synchronous opposite habitat suitability fluctuations were shown between spring and summer under different ENSO episodes, with higher HSI seen in El Niño and normal years in spring and higher HSI seen in La Niña years in summer. While wintertime HSI is higher in the southern and northern TS during El Niño and normal years, it is substantially lower in La Niña years, despite the habitat suitability being extremely lower in autumn than other seasons. These results suggested that the ENSO episodes play an important role in regulating environmental conditions and affected the catch and habitat suitability of greater amberjack in the TS.

ABSTRACT. Wind and wave energies are one of the major renewable ocean energy that are widely used for electricity generation. Renewable energies from ocean have a significant potential for future power generation in Malaysia since this country is vastly surrounded by oceans. Thus, better understanding of wind and wave patterns is essential before these energies can be harvested. The major characteristic of wind and wave is they are highly changeable over location and time. This fact anticipates the existence of considerable spatio-temporal aspects in estimating ocean energy since they can alter the ocean energy patterns in various ways. Therefore, a comprehensive review from various techniques has been conducted in order to determine the important of space and time for exploring wind-wave patterns. From this review, we have discovered a very limited number of studies that have considered spatial and temporal aspect simultaneously for analyzing wind and wave patterns. Discovering spatio-temporal patterns is essential because these patterns can represent the natural behavior of wind and wave throughout the year. Ultimately, this study has proposed a new approach that able to mine spatial and temporal aspect simultaneously for determining wind-wave patterns that can contribute to energy harvesting in future.

ABSTRACT. This research critically assesses the spatiotemporal changes in morphology of the Nijhum Dwip Island caused by the dynamic wave action of the Bay of Bengal. Using image analysis techniques on Landsat Satellite Data (ETM+, OLI), the nature of shoreline alterations, erosion-accretion, and lob shifting from 2003 to 2022 has been identified. The findings reveal that since 2013, the island's morphology and geographical extent have undergone considerable changes, and it has gradually reached its current stable state in the Bay of Bengal. The fluvial sediments brought by the Ganges-Brahmaputra-Meghna (GBM) river system influenced by tide is mainly responsible for the sedimentary accretion in and around the island. Due to the complex wave action, the island has experienced alternating erosion and accretion domination scenarios. Compared with its extent in 2003, up until 2021, it has lost approximately 2.57 km2 area, erosion and accretion mostly occurring in the southern part, geometrically transforming the island from a rectangular to a subrounded shape. The depocenter shifting of the island was predominantly in the south-eastern direction during the whole period but has shifted to reverse direction in 2021. The minimum and maximum lob shifting occurred in 2006 and 2015, respectively, at approximately 600 m and 1800 m. ALOS PALSAR DEM (12.5m) in combination with secondary data from the Bangladesh Meteorological Department has been used to develop flood inundation scenarios under different surge heights. The projections illustrate that in the case of 2 m, 3 m, and 4 m surge heights, approximately 46.45%, 56.51%, and 61.42% of the total land will go underwater, respectively. In all scenarios, cropland and bare ground are the most affected classes.

ABSTRACT. Atmospheric rivers and extratropical cyclones (ETC) are key elements of atmospheric circulation over the polar oceans. Atmospheric rivers (AR), long narrow bands characterized by increased values of total water vapor content V, total cloud liquid water content Q, precipitation and strong winds, can deliver a significant portion of precipitation to coastal areas, including Antarctica, for several years in just a few winter storms. Numerous comprehensive ETC and AR studies based on satellite measurements and re-analysis data were performed mainly for the Northern Hemisphere. In our study, AR and ETC over the Southern Ocean were considered using microwave data obtained in 2014-2020 by multichannel radiometers GCOM-W1 AMSR2, GPM GMI and Russian Meteor-M N 2 /and 2-2 MTVZA-GY at frequencies in the range of 6-190 GHz and GPM DPR and CloudSat CPR radars at frequencies ν = 13.6, 35.55 and 94 GHz, as well as visible and infrared satellite images and radiosonde profiles from coastal stations in Antarctica. The fields V, Q, precipitation, total atmospheric absorption, sea surface wind W, and sea ice concentration were obtained over the waters around Antarctica from measured brightness temperatures using original algorithms. Using these data, we studied the life cycle of individual intensive atmospheric rivers and the spatial variability of their parameters depending on latitude.

ABSTRACT. Southern Indonesia is the region that location in Southern Part of Indonesian Maritime Continent. This location has a high intensity of coastal upwelling caused by Monsoon Winds. The coastal upwelling phenomenon show an impact on oceanographic feature, such as Sea Surface Temperature (SST) and Chlorophyll-a (Chl-a) concentration and also high impacted by El Nino Southern Oscillation (ENSO) and Indian Ocean Dipole (IOD). There are currently a lot of remote sensing data with long time series that can be used to calculate the variability of difference oceanographic phenomena. This research will look at how the intensity of coastal upwelling is affected by the strength of these two occurrences based on remote sensing data. The coastal upwelling was calculated using the gradient of SST at coastal and offshore location. The annual variation show that the upwelling have high intensity during May to November along the Southern Indonesia. When in the Southern Bali-Nusa Tenggara the upwelling begin at May and South Java at July with finished same time at November. The interesting result also found in the Indonesian Throughflow (ITF) where the upwelling intensity shows the lower condition than other location. Anomaly analysis shows that South Java has an higher anomaly value of upwelling intensity than South of Bali-Nusa Tenggara. This is because the influence of the ENSO and IOD in the South Java tend to be stronger than South Bali-Nusa Tenggara. Based on this result show the remote sensing describes good result regarding the upwelling variability in Southern Indonesia region.

ABSTRACT. Wave information on the coast is very important for analyzing shoreline changes due to sediment transport as well as for the design of coastal structures. However, most of the wave observations in the coast are performed locally as a point measurement using hydraulic pressure gauges or ADCP equipment, and temporal/spatial wave observation over the entire coast is a very difficult task. Therefore, to secure the spatial wave information such as wave propagation, refraction, and diffraction, waves are numerically simulated over the interesting coastal area with verification with only a few local in-situ wave measurements. Recent advances in image analysis technology and a widespread of convenient aerial imaging equipment, such as drones, have made it possible to observe wave propagation in coastal areas temporally/spatially. In particular, the hovering technology of drones and the technique of stabilization of images taken during hovering with minute movement can produce orthoimages of wave propagation which also can generate movies for wave patterns similar to the numerically simulated wave movies. In this study, we introduce the process of making a wave propagation movie using orthoimages taken by drones on the coast like numerical wave model results and also show the technique of analyzing the wave characteristics from the spatiotemporal wave propagation images.

ABSTRACT. West water of Sumatra is kind of vital area due to its role as convection cloud formation with intensive irradiation throughout the year and it influenced to the abundance of SST front area. This study examined sea surface temperature (SST) in territorial waters which are covered West Sumatra waters to Andaman Sea. This research analyzed mechanism of SST front anomalies which is indicated due to relationship between extreme positive IOD 2019 event and the presence of westerly wind which is stronger towards Indonesia in late 2019 to early 2020. Methods that used in processing data are based on remote sensing approaching, moreover, SST-front analysis utilized gradient magnitude (GM). During the end evolution of extreme IOD in boreal winter, a widespread area of SST front occurred at the tip of Sumatra water, especially in January 2020. In contrast, the phenomenon was not found in previous years in the area, including similar events such as positive IOD 2016 event. The area showed a high value of GM with more than 0.2 0C/km, where SST was lower reaching around 26.0 0C than the surrounding waters in the northern channel of Banda Aceh. Its well-shaped lengthened high GM towards Indian Ocean corresponded to cold fronts. Lower SST is associated with upwelling process which is influenced by strong westerly wind with a speed of around 8.0 m/s and downwelling (reflective downwelling coastal Kelvin wave) process in off-water of Java-Sumatra by the end of extreme IOD event 2019 where speed of its water movement illustrated by surface current direction in long vector. The contrast of sea surface height illustrated upwelling and downwelling process at the study area with different values of around 0.3 m. It can be concluded, that the convergent areas of vigorous upwelling-downwelling as the impact of the end extreme IOD 2019 event are the cause of the abrupt alteration water-mass characteristics in study area which is lead to massive SST front.

ABSTRACT. The Gulf of Mannar is a large shallow bay forming part of the Laccadive Sea in the Indian Ocean. Long-term temporal and spatial observations of monthly, inter-annual and decadal variation of sea surface temperature (SST) of Gulf of Mannar (GOM) were studied for the period 1870-2018; climatology data was obtained from the Met Office Hadley Centre, UK. The mean highest and lowest SST were 29.850C in April and 27.150C in August respectively during the study period. Monthly time series showed a cooling trend of SST by -0.050C from January to December during 1870-2018. The mean annual highest and lowest SST were observed in 2015 and 1890 with SST of 28.930C and 27.450C respectively. Annual time series also showed a warming trend of SST by 0.0040C from 1870 to 2018. Decadal time series also showed a warming trend of SST by 0.040C from 1870 to 2018 in which mean highest and lowest mean decadal SST were 28.560C in 2010-2018 and 27.780C in 1890-1889 respectively. The spatial distribution of climate trends through the decades across the GoM revealed a strong spatial gradient during 1870-2018. The area between 6-80N and 77-780E was warmer than all other places of GoM throughout the study period.

ABSTRACT. Classification of mesoscale cloud morphology in satellite images is an application where deep learning has excelled. These classifications rely heavily on interpreting cloud spatial structure and are difficult, or impossible, to do with conventional satellite retrieval algorithms. Mesoscale cloud classifications can ultimately be used to enable targeted cloud studies that move beyond bulk cloud statistics and can differentiate the many diverse structures of marine low cloud. A major limitation of deep learning, however, is the large number of hand-labeled samples that are required for training. This work applies recently developed contrastive learning schemes to train a deep convolutional neural network (CNN), in a self-supervised manner, that can be used to generate vector embeddings that capture information about mesoscale cloud morphology and are useful for satellite image classification. The model is evaluated against existing cloud classification datasets, and we briefly discuss and demonstrate several applications of the self-supervised CNNs, including: building cloud classifiers with very few training samples, uses of the CNN’s learned internal feature representations, cross-instrument application, and resilience against sensor calibration drift.

ABSTRACT. This research is based on the human population on a small island which is increasing every time in line with the amount of waste production and considering the very limited land area of the island. Indonesia itself is the world's second largest plastic marine dumper after China, based on a 2010 study conducted in 192 coastal countries by Jambeck et al. (2015). This study aims to: (1) Identify the existing condition of the waste management system on Balang Lompo Island; (2) Identify the behavior of the community towards the way of managing their waste and; (3) Prepare directions for handling and managing island waste (shipping waste and local waste). The data collection method used is the primary method through interviews, observations, and UAV image retrieval and secondary methods through literature studies. The analysis used is spatial analysis, projection analysis (Forecasting), analysis of the availability of solid waste infrastructure, descriptive analysis, and gap analysis. The results of this study indicate that the amount of household waste on the island of Balang Lompo produced per day has exceeded the standard and on every coast of the island, especially in the north and east there is abundant waste generation. The right direction to overcome this waste problem is to apply the use of local waste and marine waste, both organic and inorganic by: recycling; increase reuse (reuse) and; increase energy conversion (energy recovery).

ABSTRACT. Tides have a substantial impact on coastal modelling and ocean circulation, and it plays a vital role in climate. Usually, tide gauges served as the primary data sources for tidal measurement prior to the launch of satellite altimeters. Due to their sparseness and predominantly located in the coastal region, the tidal measurements from the tide gauges provide limited tidal information, particularly in the open seas. With the emergence of the satellite altimeter, a global perspective of ocean surfaces was made possible. However, the sun-synchronous sampling severely restricts the solar tide observations, primarily the principal semidiurnal S2 tidal constituent at the certain satellite mission that provided the much denser track pattern, such as ERS mission. Therefore, this study is conducted to derive the amplitude and phase of S2 tidal constituents from the ERS-2 and SARAL missions by using the S2 model from the multi-mission of TOPEX, Jason-1, Jason-2, Jason-3 and GFO-1. Over 29 years of sea surface height data is used and analysed using the harmonic method to estimate the amplitude and phase of S2 constituents from the TOPEX, Jason-1, Jason-2, Jason-3 and GFO-1 missions. The results indicate reliability to support, enhance, and complement the current tidal measurements to overcome the weakness in the deep ocean without data.

ABSTRACT. This paper presents the comparison of Digital Model (DEM) accuracy derived from the varying flying heights of Unmanned Aerial Vehicle (UAV). The UAV is flew on the heights of 120m and 80m over coastline of Setiu Terengganu in Malaysia. The study are selected to see the appropriate height to produce an accurate DEM over the coastal region. The Global Positioning System (GPS) is established using real-time kinematic techniques for ground control points and for validating the UAV-derived DEM. In total, 64 GPS points are collected. The positioning accuracy of the UAV-derived DEM with respect to the GPS data are expected to be three time that of the Ground Sampling Distance (GSD). The GSD value for 80m flight is 0.219 and for 120m flight is 0.329m. The results of this study show that accuracy of 80m flight altitude is more accurate than 120m flight altitude. That is, a flight height of 80m has an average root square error of 0.500m compared to 4.554m for a flight height of 120m. The result of this study is beneficial to coastal and ocean management as for assessing the Coastal Vulnerability Index, where accurate DEM become one of important parameter.

ABSTRACT. More than twelve satellite scatterometers have operated since 1992 through the present providing the main source of surface wind vector observations over global oceans. In this study, these scatterometer winds are used in combination with radiometers and synthetic aperture radars (SAR) for better determination and characterization of high spatial and temporal resolution of regional surface wind parameters, including wind speed and direction, wind stress components, wind stress curl and divergence. 27-year long (1992 – 2018) 6-hour satellite wind analysis with a spatial resolution of 0.125° in latitude and longitude is calculated using spatial structure functions derived from high-resolution SAR data. The main objective is to improve regional winds over three major upwelling regions (the Canary, Benguela, and California regions) through the use of accurate and homogenized wind observations and region-specific spatial and temporal wind variation structure functions derived from buoy and SAR data. The long time series of satellite wind analysis over the California upwelling, where a significant number of moorings is available, are used for assessing the accuracy of the analysis. The latter is close to scatterometer wind retrieval accuracy. This assessment shows that the root-mean-square difference between collocated 6-hour satellite wind analysis and buoys is lower than 1.50 m s-1 and 1.80 m s-1 for offshore and nearshore locations, respectively. The temporal correlation between buoy and satellite analysis winds exceeds 0.90. The analysis accuracy is lower for 1992 – 1999 when satellite winds were mostly retrieved from ERS-1 and/or ERS-2 scatterometers. To further assess the improvement brought by this new wind analysis, its data and data from three independent products (ERA5, CMEMS, and CCMP) are compared with purely scatterometer winds over the Canary and Benguela regions. Even though the four products are generally similar, the new satellite analysis shows significant improvements, particularly in the upwelling areas.

ABSTRACT. The sea surface height (SSH) derived satellite altimeter in the last 25-years shows the acceleration of the global sea level rise driven by climate change. This study summarizes our current understanding of SSH variability and trends based on the 25-years satellite altimeter record in the Southeast Asian Seas (SEAS). The SSH derived altimetry show good agreement with tide gauge records in the SEAS. The SSH shows significant variability at annual and interannual time scale related to the seasonally reversing monsoons and El Niño Southern Oscillation (ENSO) and the Indian Ocean Dipole (IOD) mode. Sea level trends show spatial variation in the SEAS. Some regions, such as in the internal waters of the SEAS, are experiencing a rise of up to 3 mm/year. In the Eastern Indian Ocean off Sumatra, Java, and Ambon, sea level trends are positive, with most areas showing increases of 4-5 mm/year.