Predicting subsurface sonar observations with satellite-derived ocean surface data in the California Current...

Area of Science:

• Marine ecology
• Oceanography
• Remote sensing

Background:

• Vessel-based sonar systems offer insights into marine organism distribution but are costly and limited in scope.
• Satellite data provide extensive surface ocean information, offering potential for broader marine ecosystem monitoring.

Purpose of the Study:

• To assess the feasibility of using satellite imagery to predict subsurface biomass, using sonar data as a proxy.
• To evaluate the predictive capability of satellite data for marine life distribution in the California Current Ecosystem.

Main Methods:

• Random forest models were employed to correlate satellite-derived variables with sonar-measured biomass proxies.
• The study analyzed data across various sonar frequencies and depths within the California Current Ecosystem.

Main Results:

• Satellite data demonstrated limited utility in predicting subsurface biomass.
• Model performance was generally low across different sonar frequencies and depths.
• Spatial interpolation showed better results than spatial or temporal extrapolation, indicating limitations in predictive power.

Conclusions:

• While satellite data show some potential, current models are not yet reliable for forecasting marine life distribution or for broad spatial extrapolation.
• Further research is needed to overcome limitations and improve the integration of satellite and sonar data for marine ecosystem assessment.