Relationships between acoustic scattering layers and ocean stratification: An acoustic approach to estimate mixed layer depth

Abstract Ocean stratification, measured as mixed layer depth (MLD), is an important feature that influences many processes in the ocean; for example, by mediating the fluxes of momentum, heat and crucially key gases such as CO 2 and oxygen from the atmosphere into the water column and vice versa. The conventional methods to measure stratification (e.g., Conductivity, Temperature and Depth profiling from ARGO floats) are limited in terms of spatial and temporal coverage especially in shelf sea areas and can be expensive to conduct. Several studies have highlighted the possibility of detecting water stratification using active acoustic measurements by scientific echosounders that are able to provide high spatial and temporal resolution data, which can be collected continuously, including on surveys carrying out other tasks. In this work we apply a method to detect and map the distribution of stratification by using narrowband echosounder data collected over an area that included the Western English Channel and Celtic Sea from 2012 to 2019. The approach is based on the identification of the scattering layers located at the depth of the thermocline that can be attributed to biological scattering or scattering from the oceanic microstructure. The relationship between the thermocline and the scattering layers was modeled using machine learning methods to predict the intensity and extent of the stratification. The methodology developed within this project can offer the opportunity for wide‐scale routine application on surveys where acoustic data are collected for other purposes (e.g., fisheries survey) and it can also be applied to datasets already collected.