Abstract's details
Best practices for intercomparing ADCP and altimeter statistics for submesoscale analyses: Lessons learned from the California Current System, the Gulf Stream, and the tropics
CoAuthors
Event: 2023 Ocean Surface Topography Science Team Meeting
Session: Science III: Mesoscale and sub-mesoscale oceanography
Presentation type: Type Poster
Contribution: PDF file
Abstract:
Underway surveys using hull-mounted shipboard acoustic Doppler current profilers (ADCPs) offer the most comprehensive in-situ measurements of the spatial structure of upper ocean oceanic currents. ADCP measurements, which started in the late 1980s on U.S. oceanographic research ships, are now collected routinely by research and offshore industry vessels and even from a few commercial shipping lines, providing nearly global coverage. ADCP data offer an invaluable point of reference to guide interpretation of altimeter data. By analyzing ADCP data collected from multiple ships and with different sonar frequencies, we have compiled an expanded data base of ADCP records extending back to the early 1990s and have assessed their utility, particularly in the context of altimeter comparisons.
Our principal conclusions provide guidance on the usage of high-quality vector velocity data to infer ocean dynamics:
(i) Although ships travel more slowly than satellites, ADCP velocity data are sufficiently synoptic to capture dominant modes of variability, particularly for statistical comparison with altimetry or for use in a Helmholtz decomposition to assess rotational and divergent components of the flow.
(ii) In general, ADCP data need to be post-processed using high-accuracy navigation systems (e.g. GPS and DGPS-inertial systems) in order to obtain velocity data with accuracies better than 5 cm/s. However, for many altimeter applications, this condition can be relaxed, particularly for transiting vessels that make very few stops or directional changes.
(iii) The existing archive of geographically distributed, multi-directional ADCP data is effective at sampling statistical properties of the ocean over key regimes and in environments that may show anisotropic variability.
(iv) Given a choice of ADCP sonar frequency, high-frequency systems are preferable to low-frequency since they yield better signal-to-noise ratios at high wavenumbers. Random measurement noise is particularly challenging in upper ocean bins, near the ship hull. Error velocity statistics reported by the ADCP frequently underestimate the noise level in many data sets, in particular those with low-frequency sonars operating in rough weather. Both vertical and horizontal averaging is recommended to improve signal-to-noise ratios, though averaging acts as a dynamical filter of small-scale phenomena.
Analyses following these guidelines show that ADCP and altimeter measurements provide a consistent depiction of low-wavenumber geostrophic variability. We have released the software tools and data products developed in the context of this project online.
Our principal conclusions provide guidance on the usage of high-quality vector velocity data to infer ocean dynamics:
(i) Although ships travel more slowly than satellites, ADCP velocity data are sufficiently synoptic to capture dominant modes of variability, particularly for statistical comparison with altimetry or for use in a Helmholtz decomposition to assess rotational and divergent components of the flow.
(ii) In general, ADCP data need to be post-processed using high-accuracy navigation systems (e.g. GPS and DGPS-inertial systems) in order to obtain velocity data with accuracies better than 5 cm/s. However, for many altimeter applications, this condition can be relaxed, particularly for transiting vessels that make very few stops or directional changes.
(iii) The existing archive of geographically distributed, multi-directional ADCP data is effective at sampling statistical properties of the ocean over key regimes and in environments that may show anisotropic variability.
(iv) Given a choice of ADCP sonar frequency, high-frequency systems are preferable to low-frequency since they yield better signal-to-noise ratios at high wavenumbers. Random measurement noise is particularly challenging in upper ocean bins, near the ship hull. Error velocity statistics reported by the ADCP frequently underestimate the noise level in many data sets, in particular those with low-frequency sonars operating in rough weather. Both vertical and horizontal averaging is recommended to improve signal-to-noise ratios, though averaging acts as a dynamical filter of small-scale phenomena.
Analyses following these guidelines show that ADCP and altimeter measurements provide a consistent depiction of low-wavenumber geostrophic variability. We have released the software tools and data products developed in the context of this project online.