Detecting and Classifying Marine Plastic Debris from high-resolution multispectral satellite data

2021 ◽  
Author(s):  
Aikaterini Kikaki ◽  
Ioannis Kakogeorgiou ◽  
Paraskevi Mikeli ◽  
Dionysios E. Raitsos ◽  
Konstantinos Karantzalos
Keyword(s):  
High Resolution ◽  
Satellite Data ◽  
Marine Pollution ◽  
Sea Surface ◽  
Marine Debris ◽  
Global Ocean ◽  
Plastic Debris ◽  
Machine Learning Classification ◽  
Geographical Regions ◽  
Marine Plastic Debris

<p>Plastic debris in the global ocean is considered an essential issue with severe implications for human health and marine ecosystems. Remote sensing is a useful tool for detecting and identifying marine pollution; however, there are still few studies and benchmark datasets for developing monitoring solutions for marine plastic debris detection from high-resolution satellite data.</p><p>Here, we present an annotated plastic debris dataset from different geographical regions, seasons, and years, including annotations for sea surface features (e.g., foam), objects (e.g., ship) and floating macroalgae species such as Sargassum. Our dataset is based on high-resolution multispectral satellite observations collected mainly for the period 2014-2020 over the Gulf of Honduras (Caribbean Sea). Over this region, large plastic debris masses and Sargassum macroalgae blooms have been frequently reported, suggesting that it is an ideal region to examine satellite sensors' effectiveness in plastic debris identification, as well as monitoring along with sea surface circulation and meteorological data.</p><p>We also present a set of machine learning classification frameworks for marine debris detection on high-resolution satellite imagery, comparing qualitatively and quantitatively their overall performance. The new algorithms were validated against different regions that have been reported as major plastic polluted areas, as well as their performance was compared to well-established FAI and new promising FDI. This benchmark study can trigger more research and developement efforts towards the systematic detection and monitoring of marine plastic pollution.</p>

scholarly journals High-Resolution Sea Surface Temperature and Salinity in the Global Ocean from Raw Satellite Data

2019 ◽  
Vol 11 (19) ◽  
pp. 2191 ◽  
Author(s):  
Encarni Medina-Lopez ◽  
Leonardo Ureña-Fuentes
Keyword(s):  
Neural Network ◽  
High Resolution ◽  
Satellite Data ◽  
Google Earth ◽  
Sea Surface ◽  
Global Ocean ◽  
The Neural Network ◽  
Wide Range ◽  
Sentinel 2

The aim of this work is to obtain high-resolution values of sea surface salinity (SSS) and temperature (SST) in the global ocean by using raw satellite data (i.e., without any band data pre-processing or atmospheric correction). Sentinel-2 Level 1-C Top of Atmosphere (TOA) reflectance data is used to obtain accurate SSS and SST information. A deep neural network is built to link the band information with in situ data from different buoys, vessels, drifters, and other platforms around the world. The neural network used in this paper includes shortcuts, providing an improved performance compared with the equivalent feed-forward architecture. The in situ information used as input for the network has been obtained from the Copernicus Marine In situ Service. Sentinel-2 platform-centred band data has been processed using Google Earth Engine in areas of 100 m × 100 m. Accurate salinity values are estimated for the first time independently of temperature. Salinity results rely only on direct satellite observations, although it presented a clear dependency on temperature ranges. Results show the neural network has good interpolation and extrapolation capabilities. Test results present correlation coefficients of 82 % and 84 % for salinity and temperature, respectively. The most common error for both SST and SSS is 0.4 ∘ C and 0 . 4 PSU. The sensitivity analysis shows that outliers are present in areas where the number of observations is very low. The network is finally applied over a complete Sentinel-2 tile, presenting sensible patterns for river-sea interaction, as well as seasonal variations. The methodology presented here is relevant for detailed coastal and oceanographic applications, reducing the time for data pre-processing, and it is applicable to a wide range of satellites, as the information is directly obtained from TOA data.

scholarly journals Integration of HF Radar Observations for an Enhanced Coastal Mean Dynamic Topography

2020 ◽  
Vol 7 ◽  
Author(s):  
Ainhoa Caballero ◽  
Sandrine Mulet ◽  
Nadia Ayoub ◽  
Ivan Manso-Narvarte ◽  
Xabier Davila ◽  
...  
Keyword(s):  
High Resolution ◽  
Sea Level ◽  
Satellite Data ◽  
Coastal Areas ◽  
Hf Radar ◽  
Global Ocean ◽  
Dynamic Topography ◽  
Mean Dynamic Topography ◽  
Radar Measurements

Satellite altimeters provide continuous information of the sea level variability and mesoscale processes for the global ocean. For estimating the sea level above the geoid and monitoring the full ocean dynamics from altimeters measurements, a key reference surface is needed: The Mean Dynamic Topography (MDT). However, in coastal areas, where, in situ measurements are sparse and the typical scales of the motion are generally smaller than in the deep ocean, the global MDT solutions are less accurate than in the open ocean, even if significant improvement has been done in the past years. An opportunity to fill in this gap has arisen with the growing availability of long time-series of high-resolution HF radar surface velocity measurements in some areas, such as the south-eastern Bay of Biscay. The prerequisite for the computation of a coastal MDT, using the newly available data of surface velocities, was to obtain a robust methodology to remove the ageostrophic signal from the HF radar measurements, in coherence with the scales resolved by the altimetry. To that end, we first filtered out the tidal and inertial motions, and then, we developed and tested a method that removed the Ekman component and the remaining divergent part of the flow. A regional high-resolution hindcast simulation was used to assess the method. Then, the processed HF radar geostrophic velocities were used in synergy with additional in situ data, altimetry, and gravimetry to compute a new coastal MDT, which shows significant improvement compared with the global MDT. This study showcases the benefit of combining satellite data with continuous, high-frequency, and synoptic in situ velocity data from coastal radar measurements; taking advantage of the different scales resolved by each of the measuring systems. The integrated analysis of in situ observations, satellite data, and numerical simulations has provided a further step in the understanding of the local ocean processes, and the new MDT a basis for more reliable monitoring of the study area. Recommendations for the replicability of the methodology in other coastal areas are also provided. Finally, the methods developed in this study and the more accurate regional MDT could benefit present and future high-resolution altimetric missions.

High-resolution sea surface salinity and temperature in coastal areas from Sentinel-2 and Copernicus Marine in situ data

2020 ◽  
Author(s):  
Encarni Medina-Lopez
Keyword(s):  
Neural Network ◽  
High Resolution ◽  
Sea Surface ◽  
Sea Surface Salinity ◽  
Global Ocean ◽  
Surface Salinity ◽  
In Situ Data ◽  
The Neural Network ◽  
Sentinel 2

<p>The aim of this work is to obtain high-resolution values of sea surface salinity (SSS) and temperature (SST) in the global ocean by using raw satellite data (i.e., without any band data pre-processing or atmospheric correction). Sentinel-2 Level 1-C Top of Atmosphere (TOA) reflectance data is used to obtain accurate SSS and SST information. A deep neural network is built to link the band information with in situ data from different buoys, vessels, drifters, and other platforms around the world. The neural network used in this paper includes shortcuts, providing an improved performance compared with the equivalent feed-forward architecture. The in situ information used as input for the network has been obtained from the Copernicus Marine In situ Service. Sentinel-2 platform-centred band data has been processed using Google Earth Engine in areas of 100 m x 100 m. Accurate salinity values are estimated for the first time independently of temperature. Salinity results rely only on direct satellite observations, although it presented a clear dependency on temperature ranges. Results show the neural network has good interpolation and extrapolation capabilities. Test results present correlation coefficients of 82% and 84% for salinity and temperature, respectively. The most common error for both SST and SSS is 0.4 C and 0.4 PSU. The sensitivity analysis shows that outliers are present in areas where the number of observations is very low. The network is finally applied over a complete Sentinel-2 tile, presenting sensible patterns for river-sea interaction, as well as seasonal variations. The methodology presented here is relevant for detailed coastal and oceanographic applications, reducing the time for data pre-processing, and it is applicable to a wide range of satellites, as the information is directly obtained from TOA data.</p>

scholarly journals Marine plastic debris in Indonesia: Baseline estimates (2010-2019) and monitoring strategies (2021-2025)

2020 ◽  
Vol 45 (2) ◽  
pp. 97-102
Author(s):  
Intan Suci Nurhati ◽  
Muhammad Reza Cordova
Keyword(s):  
Marine Debris ◽  
Plastic Debris ◽  
Debris Accumulation ◽  
In Situ Data ◽  
Monitoring Strategies ◽  
Level Model ◽  
Data Source ◽  
Baseline Values ◽  
Marine Plastic Debris

Indonesia set the mission to reduce marine plastic debris by 70% between 2018-2025 with a global significance to support the UN Sustainable Development Goal 14.1. This short communication assesses marine debris baseline estimates in Indonesia before 2020 from available contributions and provides recommendations for monitoring marine debris mitigation between 2021-2025. Widely ranging model estimates of plastic debris released into seas highlight the roles of data source, the spatial resolution of models, and in situ data to provide representative baseline values. Recognizing the strengths and uncertainties of available contributions, model outputs converge on a baseline value of 0.52 ± 0.36 million tons (Mt) per year prior to 2020 in Indonesia, therefore setting a targeted reduced number of 0.16 Mt of marine debris releases in 2025. The Indonesian Institute of Sciences showed a preliminary value of plastic debris accumulation in beaches at 113.58 ± 83.88 g/m2 monthly or equivalent to 0.40 Mt/year by assuming plastic debris is most pervasive within 3 meters from Indonesia’s 99,093 km-long coastlines. It is important to distinguish that while river monitoring data informs land-based plastic debris releases, stranded beach debris represents a fraction of debris that is not present in the water column and bottom sediments. Moving forward, monitoring initiatives to mitigate marine debris should leverage on nationwide municipality-level model estimates (e.g., the source to leakage route framework of the National Plastic Action Partnership) as well as in situ river and coastal particularly but not limited to sites co-identified in previous monitoring studies (i.e., Medan, Batam-Bintan, Padang, Jakarta-Seribu Islands, Semarang, Pontianak, Bali, Lombok, Makassar, Manado, Bitung). The latter should be conducted at least seasonally, considering evidence of monsoonal variations of marine debris release and accumulation in Indonesia. Indonesia's vastness and regional diversity require coordination among stakeholders (government agencies, research institutions, universities, NGOs, citizen scientists) to monitor progress in the environments.

scholarly journals Strategy on marine debris reduction in Indonesia: a review and recommendation

2021 ◽  
Vol 925 (1) ◽  
pp. 012027
Author(s):  
FY Prabawa ◽  
NS Adi ◽  
WS Pranowo ◽  
SS Sukoraharjo ◽  
BG Gautama ◽  
...  
Keyword(s):  
Task Force ◽  
Action Plan ◽  
Green Technology ◽  
Waste Reduction ◽  
Marine Debris ◽  
Action Plans ◽  
Plastic Debris ◽  
Use Of Technology ◽  
Integrated Methods ◽  
Marine Plastic Debris

Abstract In 2018, the Indonesian government started a program: National Action Plan on Marine Debris, with the target to reduce 70% of marine plastic debris by 2025. Based on local research’s result in 2018, there was an estimated 0,27 to 0,59 million tons (MT) of marine plastic debris on local seas. Thus, the target of 70% debris reduction would be at 0.35 MT per year, or the reduction of 29.500 Tons of debris per month. That is a huge number to deal with, considering there are only 4 years left to 2025. To achieve the program, a roadmap was developed, parallel to other supporting programs as well the regulations, a national task force TKN PSL also established to run the programs. But an intriguing question remains: how to improve the achievement of this challenging target in a limited time? This study aimed to figure out the progress of existing waste reduction programs and contribute the way to improve the program. The method is a combination of literature review to collect data, a comparative and analytical work and finally the development of concept and action plans to formulate recommendation. We concluded that to improve the achievement of the target, proper strategy and program are needed to accelerate and boost the progress of marine debris reduction programs. To strengthen the waste reduction effort, the use of technology needs to be strongly emphasized. The program is best to be imply directly on sites, using various integrated methods to reduce more marine debris. More units of waste processing TPS 3R or “Reuse, Reduce, Recycle” are in urgency to obtained. The units will be located along the water body areas covering upstream to downstream, inland as well on-water. For the on-water site works, a concept of the green technology-based system integrated with small-sized floating TPS 3R barge, called STAMSAL P2K, is recommended to be implemented in the action plans.

scholarly journals Sea Surface Temperature Daytime Climate Analyses Derived from Aerosol Bias-Corrected Satellite Data

2006 ◽  
Vol 19 (3) ◽  
pp. 410-428 ◽  
Author(s):  
Nicholas R. Nalli ◽  
Richard W. Reynolds
Keyword(s):  
High Resolution ◽  
Surface Temperature ◽  
Sea Surface Temperature ◽  
Satellite Data ◽  
Bias Correction ◽  
Tropical Rainfall Measuring Mission ◽  
Stratospheric Aerosol ◽  
Sea Surface ◽  
Diurnal Warming

Abstract This paper describes daytime sea surface temperature (SST) climate analyses derived from 16 years (1985–2000) of reprocessed Advanced Very High Resolution Radiometer (AVHRR) Pathfinder Atmospheres (PATMOS) multichannel radiometric data. Two satellite bias correction methods are employed: the first being an aerosol correction, the second being an in situ correction of satellite biases. The aerosol bias correction is derived from observed statistical relationships between the slant-path aerosol optical depth and AVHRR multichannel SST (MCSST) depressions for elevated levels of tropospheric and stratospheric aerosol. Weekly analyses of SST are produced on a 1° equal-angle grid using optimum interpolation (OI) methodology. Four separate OI analyses are derived based on 1) MCSST without satellite bias correction, 2) MCSST with aerosol satellite bias correction, 3) MCSST with in situ correction of satellite biases, and 4) MCSST with both aerosol and in situ corrections of satellite biases. These analyses are compared against the Tropical Rainfall Measuring Mission (TRMM) Microwave Imager OI SST, along with the extended reconstruction SST in situ analysis product. The OI analysis 1 exhibits significant negative and positive biases. Analysis 2, derived exclusively from satellite data, reduces globally the negative bias associated with elevated atmospheric aerosol, and subsequently reveals pronounced variations in diurnal warming consistent with recently published works. Analyses 3 and 4, derived from in situ correction of satellite biases, alleviate biases (positive and negative) associated with both aerosol and diurnal warming, and also reduce the dispersion. The PATMOS OISST 1985–2000 daytime climate analyses presented here provide a high-resolution (1° weekly) empirical database for studying seasonal and interannual climate processes.

Plastic Debris Pollution on Recreational Beaches: A Malaysian Case Study

2015 ◽  
Vol 768 ◽  
pp. 804-809
Author(s):  
Shahul Hamid Fauziah ◽  
Atikah Afifah Ibrahim Nurul
Keyword(s):  
Marine Pollution ◽  
Marine Ecosystem ◽  
The Other ◽  
Total Weight ◽  
Marine Debris ◽  
Mitigation Measures ◽  
Plastic Debris ◽  
Other Hand ◽  
Monitoring Strategies ◽  
Solid Foundation

The significance of marine debris presence particularly plastic in marine ecosystem calls for proper monitoring strategies to establish a solid foundation for mitigation measures. The objectives of this paper are to study plastic debris distribution while correlating it to the level of awareness on marine pollution among beach users. It is aimed to investigate the impacts of tourism activities and the abundance of plastic debris on the beach. To quantify plastic debris on the beach, five points with duplicates were taken for three consecutive months. These sand samples were sieved through 1.00mm, 2.80mm and 4.7mm apertures. On the other hand, questionnaires were distributed to 625 beach users to study the awareness on issues related to marine pollution. Results indicated that the most crowded site accumulated the highest number of plastic debris (59 items), ranging between 1.00mm to 2.8mm (48% of the total weight of plastic). Debris sizing 4.75mm and more only contributed 41% of the total weight. Questionnaires data revealed that 2.4% of the respondents admitted to leave waste on the beach particularly if no garbage bins are provided, while the majority (92%) collect and throw the waste elsewhere. As for the cause of polluted beaches, 56% believed it is due to the indifferent attitude of the beach users that 20% of the respondents felt that stricter law should be enforced. It can be concluded that the number of plastic debris is highly influenced by the number of beach users. On the other hand, though it is lower in weight, smaller debris makes the largest number of items on the beach. While public believe that more stringent enforcement should be in place, an efficient waste management is also vital to prevent further detrimental impacts of plastic debris to the marine ecosystem.

Marine Plastic Debris

2020 ◽  
pp. 94-121 ◽  
Author(s):  
Muhammad Reza Cordova
Keyword(s):  
Marine Pollution ◽  
Anthropogenic Activities ◽  
Plastic Debris ◽  
Outreach Programs ◽  
Global Environmental ◽  
Behavior Science ◽  
Laws And Policies ◽  
The Right ◽  
And Behavior ◽  
Marine Plastic Debris

Marine pollution due to littering from anthropogenic activities is a serious global environmental problem—the main reason accumulation of debris in the environment, including in the ocean. There is a significant hazard coming from plastic debris. Besides entanglement and ingestion, marine plastics debris has more complex problems and can release additional and by-product chemical substances. If we keep producing and not doing anything, a recent study said by 2050 there would be three times more plastic than fish in the ocean. We only have a limited understanding of marine plastic debris distribution, implication, fate, and behavior. Science is the key to getting the right alternative for processing debris. To prevent marine pollution successfully requires education and outreach programs, strong laws and policies, and law enforcement for government and private institutions. This chapter explores marine plastic debris.

A hybrid CNN-LSTM based model for the prediction of sea surface temperature using time-series satellite data.

2020 ◽  
Author(s):  
Pavan Kumar Jonnakuti ◽  
Udaya Bhaskar Tata Venkata Sai
Keyword(s):  
Time Series ◽  
Deep Learning ◽  
Surface Temperature ◽  
Sea Surface Temperature ◽  
Satellite Data ◽  
Time Series Data ◽  
Sea Surface ◽  
Machine Learning Techniques ◽  
Global Ocean ◽  
Series Data

<p>Sea surface temperature (SST) is a key variable of the global ocean, which affects air-sea interaction processes. Forecasts based on statistics and machine learning techniques did not succeed in considering the spatial and temporal relationships of the time series data. Therefore, to achieve precision in SST prediction we propose a deep learning-based model, by which we can produce a more realistic and accurate account of SST ‘behavior’ as it focuses both on space and time. Our hybrid CNN-LSTM model uses multiple processing layers to learn hierarchical representations by implementing 3D and 2D convolution neural networks as a method to better understand the spatial features and additionally we use LSTM to examine the temporal sequence of relations in SST time-series satellite data. Widespread studies, based on the historical satellite datasets spanning from 1980 - present time, in Indian Ocean region shows that our proposed deep learning-based CNN-LSTM model is extremely capable for short and mid-term daily SST prediction accurately exclusive based on the error estimates (obtained from LSTM) of the forecasted data sets.</p><p><strong>Keywords: Deep Learning, Sea Surface Temperature, CNN, LSTM, Prediction.</strong></p><p> </p>

Sign in / Sign up

Export Citation Format

Share Document