Ocean Front Detection Research Articles

Southwestern Atlantic ocean fronts detected from the fusion of multi-source remote sensing data by a deep learning model

In the Southwestern Atlantic, the Falkland Current intrudes onto the South American shelf, resulting in the meeting of two water masses which are completely different in temperature and dynamic characteristics, thus generating the Southwestern Atlantic Front (SAF). Therefore, the SAF has prominent characteristics of thermal and dynamics. The current ocean front detection is mainly by performing gradient operations on sea surface temperature (SST) data, where regions with large temperature gradients are considered as ocean fronts. The thermal gradient method largely ignores the dynamical features, leading to inaccurate manifestation of SAF. This study develops a deep learning model, SAFNet, to detect the SAF through the synergy of 10-year (2010-2019) satellite-derived SST and sea surface height (SSH) observations to achieve high accuracy detection of SAF with fused thermal and dynamic characteristics. The comparative experimental results show that the detection accuracy of SAFNet reaches 99.45%, which is significantly better than other models. By comparing the frontal probability (FP) obtained by SST, SSH and SST-SSH fusion data respectively, it is proved that the necessity of fusion multi-source remote sensing data for SAF detection. The detection results of fusion data can reflect the spatial distribution of SAF more comprehensively and accurately. According to the meridional variation of FP, the main reason for the seasonal variation of the SAF is the change in its thermal characteristics, and the SAF has stable dynamic characteristics.

Ocean Front Detection With Bi-Directional Progressive Fusion Attention Network

Ocean fronts are a mesoscale phenomenon in the ocean. It is important for fisheries, environmental protection, and military activities. Therefore, more and more attention has been attracted to ocean front detection. However, the distribution of front and non-front pixels is highly unbalanced in remote sensing images, and it is not easy to establish an effective ocean front detection algorithm with high accuracy. To alleviate these problems, we model the problem of detecting ocean fronts as an edge detection task and design a new end-to-end bi-directional progressive fusion attention network (BPFANet). Specifically, BPFANet consists of an effective backbone and a bi-directional path. The whole backbone has four stage detection blocks (SD blocks), which capture the ocean front features at different scales. Each SD block contains a side branch structure, which includes a deep residual dilated convolution (DRDC) module to enrich multi-scale edge information and an attention module (AM) to enhance the feature representation in both the channel and spatial dimensions. In addition, the bi-directional path can progressively fuse the four SD blocks of ocean front information. To evaluate BPFANet, we perform experiments on the OFDS365 dataset and show its advantages over existing ocean front detection methods.

Oceanic Front Detection Model Based on U-Net Network

Oceanic Front Detection Model Based on U-Net Network

Oceanic Front Detection Model Based on U-Net Network

Oceanic Front Detection Model Based on U-Net Network

Fast Ocean Front Detection Using Deep Learning Edge Detection Models

Small-scale ocean fronts play a significant role in absorbing the excess heat and CO2 generated by climate change, yet their dynamics are not well understood. Existing in-situ and remote sensing measurements of the ocean have inadequate spatial and temporal coverage to map small-scale ocean fronts globally. Additionally, conventional algorithms to generate ocean front maps are computationally intensive and require data with long lead times. We propose machine learning (ML) models to detect temperature and chlorophyll ocean fronts from unprocessed and radiometrically uncorrected satellite imagery by transfer learning from existing models for edge detection. We use two separate datasets: one based on conventional approaches to ocean front detection, and a second based on human annotated ground truth <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">1</sup> . The deep learning front detection approach significantly reduces the resources and overall lead times needed for detecting ocean fronts. The deep learning models are developed with resource-constrained edge compute platforms like CubeSats in mind, as such platforms can address the spatial and temporal coverage challenges. The highest performing models achieve accuracies of 96% and make predictions in milliseconds using unoptimized desktop CPUs and less than 100 MB of storage; these capabilities are well-suited for CubeSat deployment.

LSENet: Location and Seasonality Enhanced Network for Multiclass Ocean Front Detection

Ocean fronts can cause the accumulation of nutrients and affect the propagation of underwater sound, so high-precision ocean front detection is of great significance to the marine fishery and national defense fields. However, the current ocean front detection methods either have low detection accuracy or most can only detect the occurrence of ocean front by binary classification, rarely considering the differences of the characteristics of multiple ocean fronts in different sea areas. In order to solve the above problems, we propose a semantic segmentation network called location and seasonality enhanced network (LSENet) for multi-class ocean fronts detection at pixel level. In this network, we first design a channel supervision unit structure, which integrates the seasonal characteristics of the ocean front itself and the contextual information to improve the detection accuracy. We also introduce a location attention mechanism to adaptively assign attention weights to the fronts according to their frequently occurred sea area, which can further improve the accuracy of multi-class ocean front detection. Compared with other semantic segmentation methods and current representative ocean front detection method, the experimental results demonstrate convincingly that our method is more effective.

Ocean Front Detection with Glider and Satellite-Derived SST Data in the Southern California Current System

This study proposes a method to detect ocean fronts from in situ temperature and density glider measurements. This method is applied to data collected along the CalCOFI Line 90, south of the California Current System (CCS), over the 2006–2013 period. It is based on image-processing techniques commonly applied to sea surface temperature (SST) satellite data. Front detection results using glider data are consistent with those obtained in other studies carried out in the CCS. SST images of the Multi-scale Ultra-high Resolution (MUR) dataset were also used to compare the probability of occurrence or front frequency (FF) obtained with the two datasets. Glider and MUR temperatures are highly correlated. Along Line 90, frontal frequency exhibited the same maxima near the transition zone (~130 km offshore) as derived from MUR and glider datasets. However, marked differences were found in the bimonthly FF probability with high (low) front frequency in spring-summer for glider (MUR) data. Methodological differences explaining these contrasting results are investigated. Thermohaline-compensated fronts are more abundant towards the oceanic zone, although most fronts are detected using both temperature and density criteria, indicating a significant contribution of temperature to density in this region.

Ocean front detection and tracking using a team of heterogeneous marine vehicles

AbstractOcean monitoring is an expensive and time consuming endeavor, but it can be made more efficient through the use of teams of autonomous robots. In this paper, we present a system for the autonomous identification and tracking of ocean fronts by coordinating the sampling efforts of a heterogeneous team of autonomous surface vehicles (ASVs) and autonomous underwater vehicles (AUVs). The primary contributions of this study are (1) our algorithm for performing autonomous coordination using general autonomy principles: Sequential Allocation Monte Carlo Tree Search (SA‐MCTS) which incorporates domain knowledge into the environmental estimation through both augmenting a standard Gaussian process with a nearest neighbors prior and planning in a drifting reference frame, (2) our decision support user interface to help human operators oversee the autonomous system, and (3) the demonstration of the system's operation in a 2‐week long deployment in the Gulf of Mexico using a heterogeneous team of four Slocum gliders and two robotic ocean surface samplers. With these contributions, we aim to bridge the gap between state of the art autonomy algorithms and marine vehicle planning methods that have been tested in large‐scale field trials. This paper presents the first deployment of a general, heuristic‐based, multi‐robot coordination algorithm for an extended sampling mission.

Weak Edge Identification Network for Ocean Front Detection

Ocean fronts have an important influence on global ocean-atmosphere interactions and marine fishery. Hence, it is of great significance to obtain the positions of the ocean fronts. However, current ocean front detection research confronts two challenges: scarcity of labeled data and limitations of ocean front detection algorithms. To address these two problems, we have collected and labeled an ocean front data set and proposed a new deep learning model for ocean front detection. For concreteness, due to the weak edge property of the ocean fronts, we formulate ocean front detection as a weak edge identification problem and propose the weak edge identification network (WEIN) for ocean front detection. WEIN consists of four convolutional blocks. Each block has a side output layer used to detect front edges at a specific image representation level. The side outputs are then fused to predict (detect) the locations of the ocean fronts. In this work, we adopt two metrics to measure the experimental results, i.e., the F₁-score and intersection over union (IoU). The experimental results with comparison to traditional and deep learning approaches demonstrate the superiority of WEIN for ocean front detection.

Ocean front detection method based on improved Mask R-CNN

Ocean front detection method based on improved Mask R-CNN

Simulation Study on SAR Signatures of ocean thermal fronts

Thermal fronts are common ocean phenomena with obvious characteristics of strong temperature gradient. The strong gradient causes the change of the air-sea temperature differences at both sides of the front, leading to the local circulation and wind field change in the thermal front regions. Based on the thermal unstable stratification, friction velocity model is established, according to which the influence of air-sea temperature differences on friction velocity is analyzed. Friction velocity is also selected to be the intermediate variable to simulate the backscattering coefficient in the front regions, using the classical Bragg backscattering model and E_H ocean wave spectrum, and the influences of wave band, polarizations and background wind speed on NRCS in the front regions are discussed. Simulation results indicate that VV polarization, higher frequency and strong wind are most favorable to ocean front detection. The simulation of NRCS in the front regions helps to observe and understand the ocean fronts on SAR images and lays the theoretic foundation for ocean front detection by active radar.

A Multiscale Deep Framework for Ocean Fronts Detection and Fine-Grained Location

Ocean front plays an important role in marine fishery production and biogeochemical cycling. This letter proposes a multiscale deep framework to meet the need for automatic ocean front detection and fine-grained location. The framework mainly focuses on bringing a well-trained deep learning model into front detection and location on the global satellite sea surface temperature image. First, a multiscale scanner is designed to divide the ocean into small areas of different scales. Then, we introduce the deep model to determine that a front has occurred, and translate the global image into binary ones of various grained. Here, an overlapping scanning way is suggested to locate the front in a small region. Finally, all the binary images are scale-weighted fused into one image, which presents the center and periphery with different brightness levels. Experimental illustrations on three typical areas of the ocean are featured with six scanning scales to show the effectiveness and practical use of the proposed framework. Moreover, the comparison experiments with the traditional method also show its advantages.

Global observations of quasi‐zonal bands in microwave sea surface temperature

AbstractGlobal observations of quasi‐zonal jet‐like structures have recently been reported in estimates of upper ocean circulation. To date, these observations have come primarily from float‐derived and altimeter‐derived estimates of zonal velocity. Here, we explore the existence of similar structures in the ocean using satellite‐derived estimates of sea surface temperature (SST) from the Advanced Microwave Scanning Radiometer for the Earth Observing System (AMSR‐E). Applying an ocean front detection algorithm globally to microwave measurements of SST, we find that repeated ocean fronts occur along quasi‐zonal bands in a multiyear (2002–2011) average of detections. Such a pattern is also observed in SST gradient magnitude. Composite analyses of SST, sea surface height (SSH), and upper ocean temperatures from Argo profiling floats suggest repeated fronts in the subtropics occur as a result of neighboring anticyclonic and cyclonic eddies. Horizontal advection in the presence of a background temperature gradient likely plays a role as evidenced by the tilt of temperature anomalies with depth. High gradient events found within the bands are observed to propagate westward with speed comparable to mesoscale eddies and we estimate these events explain 20% of the observed variance in SST gradient magnitude (2002–2011). In a final analysis, we regress the decay of the bands with averaging period and observe mild‐to‐strong persistence throughout much of the World Ocean. These findings support the view that propagating eddies help give rise to the bands. Whether or not eddies follow preferred paths remain unanswered.

A model of sea surface temperature front detection based on a threshold interval

A model (Bayesian oceanic front detection, BOFD) of sea surface temperature (SST) front detection in satellite-derived SST images based on a threshold interval is presented, to be used in different applications such as climatic and environmental studies or fisheries. The model first computes the SST gradient by using a Sobel algorithm template. On the basis of the gradient value, the threshold interval is determined by a gradient cumulative histogram. According to this threshold interval, front candidates can be acquired and prior probability and likelihood can be calculated. Whether or not the candidates are front points can be determined by using the Bayesian decision theory. The model is evaluated on the Advanced Very High-Resolution Radiometer images of part of the Kuroshio front region. Results are compared with those obtained by using several SST front detection methods proposed in the literature. This comparison shows that the BOFD not only suppresses noise and small-scale fronts, but also retains continuous fronts.

Application of the Model of Universal Gravity to Oceanic Front Detection Near the Kuroshio Front

Oceanic front is a narrow transitional zone that the penetration of sea is obviously different between two or more waters there,and it plays an important role in the national production,national defense,marine and weather.Based on the modified theory of universal gravity,sea surface temperature(SST) data near the Kuroshio front are used for front detection.The theory of universal gravity assumes that each image pixel is a celestial body with a mass represented by its value.According to the law of universal gravity,the forces of the pixels in the 3×3 neighbourhood exerted on the central pixels can be calculated.Because fronts are susceptible to the noise and intense of fronts are commonly low,a modified method are proposed to solve these problems in this article.This method firstly eliminates the pixels that values equal to 0.Then in order to decrease the reliance on the brightness level of original data,a normalization step is applied to each 3×3 neighbourhood and next based on image enhancement function,each normalized 3×3 area can be enhanced.Finally,the theory of universal gravity is applied to enhanced data for front detection.The algorithm was tested and compared with conventional methods using in the fronts detection such as Sobel,Jensen-Shannon.The results show that compared to conventional methods in some areas,the proposed algorithm can decrease noise while not cause fronts discontinuous.

A New Statistical Modeling Approach to Ocean Front Detection from SST Satellite Images

Abstract Ocean fronts are narrow zones of intense dynamic activity that play an important role in global ocean–atmosphere interactions. Owing to their highly variable nature, both in space and time, they are notoriously difficult features to adequately sample using traditional in situ techniques. In this paper, the authors propose a new statistical modeling approach for detecting and monitoring ocean fronts from Advanced Very High Resolution Radiometer (AVHRR) SST satellite images that builds on a previous “front following” algorithm. Weighted local likelihood is used to provide a smooth, nonparametric description of spatial variations in the position, mean temperature, width, and temperature change of an individual front within an image. Weightings are provided by a Gaussian kernel function whose width is automatically determined by likelihood cross-validation. The statistical model fitting approach allows estimation of the uncertainty of each parameter to be quantified, a capability not possessed by other techniques. The algorithm is shown to be robust to noise and missing data in an image, problems that hamper many of the existing front-detection schemes. The approach is general and could be used with other remotely sensed datasets, model output, or data assimilation products.

An algorithm for oceanic front detection in chlorophyll and SST satellite imagery

An algorithm is described for oceanic front detection in chlorophyll (Chl) and sea surface temperature (SST) satellite imagery. The algorithm is based on a gradient approach: the main novelty is a shape-preserving, scale-sensitive, contextual median filter applied selectively and iteratively until convergence. This filter has been developed specifically for Chl since these fields have spatial patterns such as chlorophyll enhancement at thermohaline fronts and small- and meso-scale chlorophyll blooms that are not present in SST fields. Linear Chl enhancements and localized (point-wise) blooms are modeled as ridges and peaks respectively, whereas conventional fronts in Chl and SST fields are modeled as steps or ramps. Examples are presented of the algorithm performance using modeled (synthetic) images as well as synoptic Chl and SST imagery. After testing, the algorithm was used on > 6000 synoptic images, 1999–2007, to produce climatologies of Chl and SST fronts off the U.S. Northeast.

A New Statistical Modelling Approach to Ocean Front Detection from SST Satellite Images

A New Statistical Modelling Approach to Ocean Front Detection from SST Satellite Images