TL;DR
This paper investigates how ocean eddies, specifically an anticyclonic eddy in the Sea of Japan, affect long-range underwater sound propagation. By using oceanographic reanalysis data and normal mode theory, the researchers quantify the impact of horizontal refraction on acoustic signal arrival times, crucial for accurate underwater navigation. They found that while eddies do cause horizontal refraction, their effect on path elongation is relatively small for this specific experiment, suggesting robustness in acoustic navigation systems.
What Problem Does This Paper Address?
Accurate underwater navigation, especially for autonomous underwater vehicles (AUVs), relies heavily on precise acoustic ranging. This precision, in turn, depends on a detailed understanding of the sound speed field in the ocean. However, the ocean is a dynamic environment with various inhomogeneities, such as eddies, that can significantly alter sound propagation paths and arrival times. These deviations from simple straight-line propagation (geodesics) can lead to errors in ranging and navigation. This paper specifically addresses the challenge of quantifying the impact of a mesoscale anticyclonic eddy on long-range sound propagation, aiming to understand how it affects the delay times of different acoustic signal components and the overall accuracy of acoustic ranging.
Methods
The study employs a two-step approach rooted in normal mode theory.
- Sound Speed Field Modeling: The researchers utilize data from high-resolution oceanographic reanalysis models (GOFS 3.1 and GLORYS12V1) to obtain a three-dimensional sound speed field ($C_w(x, y, z)$). This is a key advantage over sparse experimental measurements, providing a more complete picture of the ocean environment.
- Horizontal Ray Tracing and Arrival Time Estimation:
- Normal Mode Decomposition: The acoustic field is decomposed into its constituent modes, each characterized by a specific wavenumber ($k_j$) and group velocity ($v_g$).
- Horizontal Eigenray Calculation: Using the computed sound speed field, the study traces "horizontal eigenrays" – the actual paths taken by each mode component, accounting for horizontal refraction. This is done by solving a Hamiltonian system derived from ray theory applied to mode amplitudes.
- Arrival Time Calculation: The arrival time for each modal component ($t_j$) is calculated by integrating the inverse of the group velocity along its respective horizontal eigenray. This accounts for both the speed of sound and the actual path length, including deviations caused by refraction.
- Elongation Quantification: The difference between the length of the horizontal eigenray ($S_j$) and the geodesic distance ($X_{max}$) is calculated as $\Delta S_j = S_j - X_{max}$. This value quantifies the effect of horizontal refraction on path elongation.
The study also compares these theoretical estimates with experimental impulse response functions (IRFs) to validate their approach.
Datasets
- Oceanographic Reanalysis Data:
- GOFS 3.1 (Global Ocean Forecasting System 3.1): Based on the HYCOM model.
-
GLORYS12V1 (Copernicus Global 1/12° Oceanic and Sea Ice Reanalysis): Based on the NEMO model. These datasets provide high-resolution, three-dimensional fields of temperature, salinity, and derived sound speed, covering the experimental area in the Sea of Japan.
-
Experimental Data:
- Acoustic Signals: Broadband acoustic signals transmitted from a Sound Navigation and Ranging (SNR) source and received by a hydrophone array.
- CTD Measurements: Vertical profiles of temperature, salinity, and sound speed obtained at several points along the acoustic track. These were used for validation and comparison with model data.
- Satellite Altimetry Data (AVISO): Used for identifying and validating the presence and location of mesoscale eddies.
Key Insights and Ideas
- Eddies as Refractors: Anticyclonic eddies, characterized by temperature perturbations, can indeed influence sound propagation by deforming the paths of acoustic signals (horizontal eigenrays).
- Competing Effects: The presence of an eddy can have two competing effects: increased sound speed due to warmer water (potentially shortening travel time) and horizontal refraction (potentially lengthening the path).
- Model Data is Crucial: Sparse experimental measurements are insufficient for accurate long-range propagation modeling. High-resolution oceanographic reanalysis data is essential for capturing mesoscale features like eddies.
- Horizontal Refraction is Subtle Here: For the specific experimental setup (a ~1000 km track in the Sea of Japan with a relatively short shallow-water segment), the horizontal refraction caused by the anticyclonic eddy resulted in relatively small elongations of the acoustic path (up to 50 m) compared to the total path length.
- Robustness of Acoustic Navigation: The study suggests that acoustic navigation systems might be relatively robust to horizontal refraction caused by eddies in certain scenarios, as the effect is small compared to the overall ranging error.
Key Results
- The study successfully modeled the sound speed field using GOFS 3.1 and GLORYS12V1 reanalyses, showing good agreement with satellite altimetry data for eddy identification.
- Horizontal eigenrays were computed for various modal components, revealing deviations from geodesic paths, particularly in the shallow-water region.
- The elongation of horizontal eigenrays due to the anticyclonic eddy was quantified, showing a maximum increase of approximately 50 meters over a 1000 km path for the modes analyzed.
- Theoretical estimates of modal component arrival times, incorporating horizontal refraction, were compared to experimental impulse response functions. The discrepancy between predicted and observed arrival times for low-order modes was found to be around 0.5-0.7 seconds, translating to a range estimation error of approximately 800 meters. This error is attributed to limitations in the oceanographic data used by the models.
Limitations and Future Work
- Data Sparsity: Despite using reanalysis data, the study acknowledges that the area of interest (Sea of Japan) has a relatively low concentration of ARGO buoys, potentially leading to incomplete information in the ocean circulation models. This is cited as a reason for the observed discrepancy between theoretical and experimental arrival times.
- Eddy Contribution Separation: The study notes that it's difficult to completely separate the eddy's contribution to horizontal refraction from the influence of bottom relief, although it suggests the eddy's effect is smaller in this case.
- Future Work: The authors suggest that continued improvements in ocean circulation modeling will lead to more accurate sound speed fields and, consequently, more precise acoustic ranging and navigation. Further investigation into the robustness of acoustic navigation systems to various oceanographic features is also implied. The practical application of pre-computed wavenumber fields ($k_j(x, y)$) for real-time AUV navigation is also highlighted.
Summary updated