Related Experiment Video
Updated: Aug 13, 2026
09:36
Continuous-Wave Propagation Channel-Sounding Measurement System - Testing, Verification, and Measurements
Published on: June 25, 2021
Bidirectional long short-term memory-based subsurface sound speed profile inversion from satellite data with sparse
Bing Yue1,2, Cheng Chen1,2,3, Xiao Feng1,2
1School of Marine Science and Technology, Northwestern Polytechnical University, Xi'an 710000, China.
JASA Express Letters
|August 12, 2026
Summary
Integrating sparse in situ observations into satellite-derived sound speed profile (SSP) inversion significantly improves accuracy. This method enhances subsurface oceanographic mapping, even without sea surface data, guiding optimal sensor deployment.
Area of Science:
- Oceanography
- Geophysics
- Remote Sensing
Background:
- Satellite remote sensing provides large-scale sea surface data for subsurface sound speed profile (SSP) inversion.
- In situ constraints are often lacking, leading to significant inversion errors in current methods.
- Accurate SSPs are crucial for underwater acoustics, navigation, and marine research.
Purpose of the Study:
- To develop a novel framework for improving SSP inversion accuracy using sparse in situ observations.
- To assess the impact of sparse observations, with and without sea surface data, on inversion performance.
- To provide guidance for optimal deployment of underwater observation systems.
Main Methods:
- A bidirectional long short-term memory (BiLSTM) network was employed for SSP inversion.
- The framework integrates sparse in situ measurements to constrain the inversion process.
- Experiments were conducted in the Kuroshio Extension and the Philippine Sea.
Main Results:
- The proposed framework significantly improved SSP inversion accuracy when incorporating sparse observations.
- Remarkable accuracy gains were achieved even when sea surface data were excluded.
- Depth-layer sensitivity analysis identified optimal sparse sampling depths for different scenarios.
Conclusions:
- Integrating sparse in situ data with BiLSTM networks is an effective strategy for enhancing SSP inversion accuracy.
- The method offers a viable solution for improving large-scale oceanographic mapping in data-sparse regions.
- Findings guide the strategic deployment of underwater sensors for improved oceanographic data collection.
