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

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.

Download Full-text

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

10.5194/egusphere-egu2020-3465 ◽

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

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.

Download Full-text

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

Frontiers in Marine Science ◽

10.3389/fmars.2020.588713 ◽

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.

Download Full-text

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

Journal of Climate ◽

10.1175/jcli3644.1 ◽

2006 ◽

Vol 19 (3) ◽

pp. 410-428 ◽

Cited By ~ 7

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.

Download Full-text

High resolution 3-D temperature and salinity fields derived from in situ and satellite observations

Ocean Science ◽

10.5194/os-8-845-2012 ◽

2012 ◽

Vol 8 (5) ◽

pp. 845-857 ◽

Cited By ~ 107

Author(s):

S. Guinehut ◽

A.-L. Dhomps ◽

G. Larnicol ◽

P.-Y. Le Traon

Keyword(s):

High Resolution ◽

Surface Temperature ◽

Sea Surface Temperature ◽

Large Scale ◽

Sea Surface ◽

Temperature Fields ◽

Global Ocean ◽

In Situ Observations ◽

Temperature Observations

Abstract. This paper describes an observation-based approach that efficiently combines the main components of the global ocean observing system using statistical methods. Accurate but sparse in situ temperature and salinity profiles (mainly from Argo for the last 10 yr) are merged with the lower accuracy but high-resolution synthetic data derived from satellite altimeter and sea surface temperature observations to provide global 3-D temperature and salinity fields at high temporal and spatial resolution. The first step of the method consists in deriving synthetic temperature fields from altimeter and sea surface temperature observations, and salinity fields from altimeter observations, through multiple/simple linear regression methods. The second step of the method consists in combining the synthetic fields with in situ temperature and salinity profiles using an optimal interpolation method. Results show the revolutionary nature of the Argo observing system. Argo observations now allow a global description of the statistical relationships that exist between surface and subsurface fields needed for step 1 of the method, and can constrain the large-scale temperature and mainly salinity fields during step 2 of the method. Compared to the use of climatological estimates, results indicate that up to 50% of the variance of the temperature fields can be reconstructed from altimeter and sea surface temperature observations and a statistical method. For salinity, only about 20 to 30% of the signal can be reconstructed from altimeter observations, making the in situ observing system essential for salinity estimates. The in situ observations (step 2 of the method) further reduce the differences between the gridded products and the observations by up to 20% for the temperature field in the mixed layer, and the main contribution is for salinity and the near surface layer with an improvement up to 30%. Compared to estimates derived using in situ observations only, the merged fields provide a better reconstruction of the high resolution temperature and salinity fields. This also holds for the large-scale and low-frequency fields thanks to a better reduction of the aliasing due to the mesoscale variability. Contribution of the merged fields is then illustrated to describe qualitatively the temperature variability patterns for the period from 1993 to 2009.

Download Full-text

Machine Learning and the End of Atmospheric Corrections: A Comparison between High-Resolution Sea Surface Salinity in Coastal Areas from Top and Bottom of Atmosphere Sentinel-2 Imagery

Remote Sensing ◽

10.3390/rs12182924 ◽

2020 ◽

Vol 12 (18) ◽

pp. 2924

Author(s):

Encarni Medina-Lopez

Keyword(s):

Machine Learning ◽

High Resolution ◽

Coastal Areas ◽

Sea Surface ◽

Sea Surface Salinity ◽

Surface Salinity ◽

Common Error ◽

Source Of Error ◽

Sentinel 2

This paper introduces a discussion about the need for atmospheric corrections by comparing data-driven sea surface salinity (SSS) derived from Top- and Bottom-of-Atmosphere imagery. Atmospheric corrections are used to remove the effect of the atmosphere in reflectances acquired by satellite sensors. The Sentinel-2 Level-2A product provides atmospherically corrected Bottom-of-Atmosphere (BOA) imagery, derived from Level-1C Top-of-Atmosphere (TOA) tiles using the Sen2Cor processor. SSS at high resolution in coastal areas (100m) is derived from multispectral signatures using artificial neural networks. These obtain relationships between satellite band information and in situ SSS data. Four scenarios with different input variables are tested for both TOA and BOA imagery, for interpolation (previous information on all platforms is available in the training dataset) and extrapolation (certain platforms are isolated and the network does not have any previous information on these) problems. Results show that TOA always outperforms BOA in terms of higher coefficient of determination (R2), lower mean absolute error (MAE) and lower most common error (μe). The best TOA results are R2=0.99, MAE=0.4PSU and μe=0.2PSU. Moreover, the evaluation of the neural network in all the pixels of Sentinel-2 tiles shows that BOA results are accurate only far away from the coast, while TOA data provides useful information on nearshore mixing patterns, estuarine processes and is able to estimate freshwater salinity values. This suggests that land adjacency corrections could be a relevant source of error. Sun glint corrections appear to be another source of error. TOA imagery is more accurate than BOA imagery when using machine learning algorithms and big data, as there is a clear loss of information in the atmospheric correction process that affects the multispectral–in situ relationships. Finally, the time and computational resources gained by avoiding atmospheric corrections can make the use of TOA imagery interesting in future studies, such as the estimation of chlorophyll or coloured dissolved organic matter.

Download Full-text

A Practical Method for High-Resolution Burned Area Monitoring Using Sentinel-2 and VIIRS

Remote Sensing ◽

10.3390/rs13091608 ◽

2021 ◽

Vol 13 (9) ◽

pp. 1608

Author(s):

Miguel M. Pinto ◽

Ricardo M. Trigo ◽

Isabel F. Trigo ◽

Carlos C. DaCamara

Keyword(s):

High Resolution ◽

Satellite Data ◽

Computation Time ◽

Practical Method ◽

Google Earth ◽

Burned Area ◽

High Resolution Data ◽

Burned Areas ◽

Resolution Data ◽

Sentinel 2

Mapping burned areas using satellite imagery has become a subject of extensive research over the past decades. The availability of high-resolution satellite data allows burned area maps to be produced with great detail. However, their increasing spatial resolution is usually not matched by a similar increase in the temporal domain. Moreover, high-resolution data can be a computational challenge. Existing methods usually require downloading and processing massive volumes of data in order to produce the resulting maps. In this work we propose a method to make this procedure fast and yet accurate by leveraging the use of a coarse resolution burned area product, the computation capabilities of Google Earth Engine to pre-process and download Sentinel-2 10-m resolution data, and a deep learning model trained to map the multispectral satellite data into the burned area maps. For a 1500 ha fire our method can generate a 10-m resolution map in about 5 min, using a computer with an 8-core processor and 8 GB of RAM. An analysis of six important case studies located in Portugal, southern France and Greece shows the detailed computation time for each process and how the resulting maps compare to the input satellite data as well as to independent reference maps produced by Copernicus Emergency Management System. We also analyze the feature importance of each input band to the final burned area map, giving further insight about the differences among these events.

Download Full-text

Cross-Country Assessment of H-SAF Snow Products by Sentinel-2 Imagery Validated against In-Situ Observations and Webcam Photography

Geosciences ◽

10.3390/geosciences9030129 ◽

2019 ◽

Vol 9 (3) ◽

pp. 129 ◽

Cited By ~ 6

Author(s):

Gaia Piazzi ◽

Cemal Tanis ◽

Semih Kuter ◽

Burak Simsek ◽

Silvia Puca ◽

...

Keyword(s):

High Resolution ◽

Snow Cover ◽

Large Scale ◽

Complex Topography ◽

Snow Properties ◽

High Resolution Imagery ◽

Wide Range ◽

Moderate Resolution ◽

Sentinel 2

Information on snow properties is of critical relevance for a wide range of scientific studies and operational applications, mainly for hydrological purposes. However, the ground-based monitoring of snow dynamics is a challenging task, especially over complex topography and under harsh environmental conditions. Remote sensing is a powerful resource providing snow observations at a large scale. This study addresses the potential of using Sentinel-2 high-resolution imagery to assess moderate-resolution snow products, namely H10—Snow detection (SN-OBS-1) and H12—Effective snow cover (SN-OBS-3) supplied by the Satellite Application Facility on Support to Operational Hydrology and Water Management (H-SAF) project of the European Organisation for the Exploitation of Meteorological Satellites (EUMETSAT). With the aim of investigating the reliability of reference data, the consistency of Sentinel-2 observations is evaluated against both in-situ snow measurements and webcam digital imagery. The study area encompasses three different regions, located in Finland, the Italian Alps and Turkey, to comprehensively analyze the selected satellite products over both mountainous and flat areas having different snow seasonality. The results over the winter seasons 2016/17 and 2017/18 show a satisfying agreement between Sentinel-2 data and ground-based observations, both in terms of snow extent and fractional snow cover. H-SAF products prove to be consistent with the high-resolution imagery, especially over flat areas. Indeed, while vegetation only slightly affects the detection of snow cover, the complex topography more strongly impacts product performances.

Download Full-text

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

10.5194/egusphere-egu21-15243 ◽

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

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.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.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.

Download Full-text

Robust sideslip angle observer of commercial vehicles based on cornering stiffness estimation using neural network

Proceedings of the Institution of Mechanical Engineers Part D Journal of Automobile Engineering ◽

10.1177/09544070211072401 ◽

2022 ◽

pp. 095440702110724

Author(s):

Chenyu Zhou ◽

Liangyao Yu ◽

Yong Li ◽

Jian Song

Keyword(s):

Neural Network ◽

Mathematical Model ◽

Stability Control ◽

Estimation Accuracy ◽

Accurate Estimation ◽

Commercial Vehicles ◽

Sideslip Angle ◽

The Neural Network ◽

Wide Range ◽

The Mathematical Model

Accurate estimation of sideslip angle is essential for vehicle stability control. For commercial vehicles, the estimation of sideslip angle is challenging due to severe load transfer and tire nonlinearity. This paper presents a robust sideslip angle observer of commercial vehicles based on identification of tire cornering stiffness. Since tire cornering stiffness of commercial vehicles is greatly affected by tire force and road adhesion coefficient, it cannot be treated as a constant. To estimate the cornering stiffness in real time, the neural network model constructed by Levenberg-Marquardt backpropagation (LMBP) algorithm is employed. LMBP is a fast convergent supervised learning algorithm, which combines the steepest descent method and gauss-newton method, and is widely used in system parameter estimation. LMBP does not rely on the mathematical model of the actual system when building the neural network. Therefore, when the mathematical model is difficult to establish, LMBP can play a very good role. Considering the complexity of tire modeling, this study adopted LMBP algorithm to estimate tire cornering stiffness, which have simplified the tire model and improved the estimation accuracy. Combined with neural network, A time-varying Kalman filter (TVKF) is designed to observe the sideslip angle of commercial vehicles. To validate the feasibility of the proposed estimation algorithm, multiple driving maneuvers under different road surface friction have been carried out. The test results show that the proposed method has better accuracy than the existing algorithm, and it’s robust over a wide range of driving conditions.

Download Full-text

An updated national-scale mangrove forest map in China Using Sentinel-1 and Sentinel-2 Time-Series Data with Google Earth Engine

10.5194/egusphere-egu21-5310 ◽

2021 ◽

Author(s):

Luojia Hu ◽

Wei Yao ◽

Zhitong Yu ◽

Yan Huang

Keyword(s):

Time Series ◽

High Resolution ◽

Mangrove Forest ◽

Google Earth ◽

Series Data ◽

National Scale ◽

Mangrove Area ◽

Google Earth Engine ◽

Forest Maps ◽

Sentinel 2

A high resolution mangrove map (e.g., 10-m), which can identify mangrove patches with small size (< 1 ha), is a central component to quantify ecosystem functions and help government take effective steps to protect mangroves, because the increasing small mangrove patches, due to artificial destruction and plantation of new mangrove trees, are vulnerable to climate change and sea level rise, and important for estimating mangrove habitat connectivity with adjacent coastal ecosystems as well as reducing the uncertainty of carbon storage estimation. However, latest national scale mangrove forest maps mainly derived from Landsat imagery with 30-m resolution are relatively coarse to accurately characterize the distribution of mangrove forests, especially those of small size (area < 1 ha). Sentinel imagery with 10-m resolution provide the opportunity for identifying these small mangrove patches and generating high-resolution mangrove forest maps. Here, we used spectral/backscatter-temporal variability metrics (quantiles) derived from Sentinel-1 SAR (Synthetic Aperture Radar) and sentinel-2 MSI (Multispectral Instrument) time-series imagery as input features for random forest to classify mangroves in China. We found that Sentinel-2 imagery is more effective than Sentinel-1 in mangrove extraction, and a combination of SAR and MSI imagery can get a better accuracy (F1-score of 0.94) than using them separately (F1-score of 0.88 using Sentinel-1 only and 0.895 using Sentinel-2 only). The 10-m mangrove map derived by combining SAR and MSI data identified 20,003 ha mangroves in China and the areas of small mangrove patches (< 1 ha) was 1741 ha, occupying 8.7% of the whole mangrove area. The largest area (819 ha) of small mangrove patches is located in Guangdong Province, and in Fujian the percentage of small mangrove patches in total mangrove area is the highest (11.4%). A comparison with existing 30-m mangrove products showed noticeable disagreement, indicating the necessity for generating mangrove extent product with 10-m resolution. This study demonstrates the significant potential of using Sentinel-1 and Sentinel-2 images to produce an accurate and high-resolution mangrove forest map with Google Earth Engine (GEE). The mangrove forest maps are expected to provide critical information to conservation managers, scientists, and other stakeholders in monitoring the dynamics of mangrove forest.

Download Full-text