Enhancing the accuracy of automatic eddy detection and the capability of recognizing the multi-core structures from maps of sea level anomaly

Jianqiang Yi, Yao Du, Zhonghua He, Chenghu ZhouOcean 2014 · DOI

Abstract

Automated methods are essential for detecting mesoscale eddies in large volumes of altimeter data. This paper presents a new "hybrid detection" (HD) method designed to improve detection accuracy and recognize multi-core structures from maps of sea level anomaly (SLA). By integrating the Okubo–Weiss (OW) method with sea surface height-based (SSH-based) criteria, the HD method achieves a successful detection rate of ~96.6% and an excessive detection rate of ~14.2%. This approach significantly outperforms traditional methods and provides a systematic way to track complex eddy splitting and merging events.

Introduction

Mesoscale eddies are the "weather" of the ocean—massive, rotating bodies of water that transport heat, salt, and nutrients globally. While satellite altimetry provides the necessary data to observe these phenomena, automated detection remains challenging. Existing algorithms typically rely on either physical parameters (like the Okubo–Weiss parameter), which are sensitive but noisy, or geometric height-based methods, which are stable but often fail to capture complex interactions like multi-core structures.

The inability to handle multi-core structures—where multiple eddy centers exist within a single boundary—often leads to tracking failures during splitting or merging events. This paper introduces the Hybrid Detection (HD) algorithm to bridge this gap, offering high precision and the novel capability to identify these complex structures.

Methodology

The HD approach follows a three-step process to ensure robust detection:

1. Center Identification

The method requires a potential eddy to meet two simultaneous criteria: a local extremum in the Sea Level Anomaly (SLA) and a location within a region of strong rotation as defined by the Okubo–Weiss (OW) parameter. This dual-check mechanism filters out the high levels of noise typically associated with pure OW-based detection.

Workflow for identifying eddy centers by combining SLA extrema with the Okubo–Weiss parameter.

2. Boundary Extraction and Multi-core Recognition

The algorithm generates sea-level contours at fine 0.5 cm increments. It identifies the smallest closed contour encompassing the rotating core as the qualified boundary. If a single boundary contains multiple eddy centers of the same polarity, the HD algorithm identifies it as a "composite" structure. This allows for the definition of individual "footprint" borders and a collective "composite" border, enabling the tracking of merging and splitting events without losing the eddy's trajectory.

Conceptual diagrams showing the selection of eddy boundaries and the identification of multi-core structures (d).

Data and Validation

The study utilized AVISO Reference Series delayed-time altimeter products (October 1992 to April 2012) resampled on a 1/3° × 1/3° Mercator grid. Validation was performed against a "ground truth" benchmark created by five independent experts manually identifying eddies in the South China Sea.

Results and Discussion

The HD method demonstrates significant improvements over traditional techniques:

  • Accuracy: It maintains a high Success Detection Rate (SDR) of 96.6%, matching the sensitivity of the OW method.
  • Precision: It drastically reduces the Excessive Detection Rate (EDR) from 70.3% (OW method) to just 14.2%, effectively eliminating false positives.
  • Tracking: The method successfully captured complex merging events in the South China Sea (e.g., eddies ACE2 and ACE3 in 2007) that previous algorithms misclassified.

By distinguishing between the "eddy center" (maximum height/depth) and the "eddy core" (area of physical rotation), the HD method provides a more physically consistent representation of ocean dynamics. The authors found that the 0.5 cm increments in sea-level contours provided the best balance between accuracy and computational speed, whereas standard 1 cm increments often underestimated eddy dimensions.

Conclusion and Future Work

While the HD method significantly advances surface-based eddy detection, it remains limited to features with strong surface signatures. Future work aims to integrate this detection logic into a "companion tracking" system to automatically manage the complex logic of eddy lifespans, including splitting and merging. This will provide deeper insights into the energy and mass transfer processes critical to global ocean circulation.

Summary updated

Abstract

Automated methods are important for automatically detecting mesoscale eddies in large volumes of altimeter data. While many algorithms have been proposed in the past, this paper presents a new method, called hybrid detection (HD), to enhance the eddy detection accuracy and the capability of recognizing eddy multi-core structures from maps of sea level anomaly (SLA). The HD method has integrated the criteria of the Okubo–Weiss (OW) method and the sea surface height-based (SSH-based) method, two commonly used eddy detection algorithms. Evaluation of the detection accuracy shows that the successful detection rate of HD is ~ 96.6% and the excessive detection rate is ~ 14.2%, which outperforms the OW and those methods using SLA extrema to identify eddies. The capability of recognizing multi-core structures and its significance in tracking eddy splitting or merging events have been illustrated by comparing with the detection results of different algorithms and observations in previous literature.
BibTeX
@inproceedings{yi2014enhancing,
  author = {Jianqiang Yi and Yao Du and Zhonghua He and Chenghu Zhou},
  title = {{Enhancing the accuracy of automatic eddy detection and the capability of recognizing the multi-core structures from maps of sea level anomaly}},
  booktitle = {Ocean},
  year = {2014},
  doi = {10.5194/os-10-39-2014},
}

Loading related content...

Figure