A Transport Method for Restoring Incomplete Ocean Current Measurements

Abstract

Remote sensing of oceanographic data often yields incomplete coverage of the measurement domain. This can limit interpretability of the data and identification of coherent features informative of ocean dynamics. Several methods exist to fill gaps of missing oceanographic data and are often based on projecting the measurements onto basis functions or a statistical model. Herein, we use an information transport approach inspired from an image processing algorithm. This approach aims to restore gaps in data by advecting and diffusing information of features as opposed to the field itself. Since this method does not involve fitting or projection, the portions of the domain containing measurements can remain unaltered, and the method offers control over the extent of local information transfer. This method is applied to measurements of ocean surface currents by high frequency radars. This is a relevant application because data coverage can be sporadic, and filling data gaps can be essential to data usability. Application to two regions with differing spatial scale is considered. The accuracy and robustness of the method is tested by systematically blinding measurements and comparing the restored data at these locations to the actual measurements. These results demonstrate that even for locally large percentages of missing data points, the restored velocities have errors within the native error of the original data (e.g., <10% for velocity magnitude and <3% for velocity direction). Results were relatively insensitive to model parameters, facilitating a priori selection of default parameters for de novo applications. Plain Language SummaryResearchers measure many geophysical phenomena by remote sensing. One example is the measurement of the ocean surface currents by land-based radar stations. The data are useful for understanding coastal dynamics and how material is transported near the ocean surface. Based on various factors, these measurements are prone to incomplete coverage. The measured data field on the map is like an incomplete image with spatial gaps where data are missing. The incompleteness in the data field reduces its utility, especially for analyses that rely on continuous coverage such as tracking the movement of objects or identifying coherent flow patterns. Several methods have been proposed to restore lost data. Inspired by a technique from image processing, we developed a method to restore incomplete field data. This method uses equations that aim to transport features in the data field into missing parts, as opposed to directly transporting the field data itself. We present this method applied to remote measurements of ocean surface currents and evaluate its ability to restore missing information. However, this approach can be applied to restore other types of incomplete field measurements to improve usability and interpretation of such data.

Advancing the theory and applications of Lagrangian Coherent Structures methods for oceanic surface flows

“…The FTLE approach is very reliable as it always outputs a visualization of the separation rates between particles in the flow and clearly indicates where the high rates versus low rates of mixing are located. It is widely established and there exist several tools to make these computations very fast, such as the TRACE interface by Ameli & Shadden [2019] that was described in previous chapters. In the examples presented in thesis, the FTLE was sensitive to and revealing of three-dimensional effects within the domain, as well as the periodicity of the underlying dynamics.…”

“…For ease and speed of calculations, the model data was subsequently uploaded in netCFD format to a web-based gateway by Ameli & Shadden [2019], TRACE, which stands for the 2016 dataset, the numerical trajectories were all generated in TRACE. South as well, but these features lacked the periodicity of the ridges in the channel, which were recurring at high tide.…”

“…All model ensembles were run by the MSEAS team, who also output the model velocity fields. The data was uploaded to the TRACE web-based gateway, described in the previous chapter [Ameli & Shadden, 2019], and can be found at http://transport.me.berkeley.edu/thredds/catalog/public/MIT-MSEAS/ catalog.html. In this chapter, I used the TRACE web-based gateway for the advection of numerical trajectories, for ease and speed of calculations.…”

“…For ease and speed of calculations of both tracer trajectories and FTLE, the model data was uploaded in NetCDF format to a web-based gateway by Ameli & Shadden [2019], TRACE, which stands for Trajectory Reconstruction and Analysis for Coherent Structure Evaluation. TRACE uses parallel processing for the Lagrangian computations and can handle and process very large datasets that are hosted remotely.…”

A Parameter-Free Spectral Clustering Approach to Coherent Structure Detection in Geophysical Flows

“…For brevity, in this paper we only present the analyses of the MSEAS hindcasts, but the forecast-based results obtained in real time during the field experiments are qualitatively similar. For ease and speed of Lagrangian calculations, the TRACE web-based gateway [3] (http:// transport.me.berkeley.edu/thredds/catalog/public/MIT-MSEAS/catalog.html) was used to compute trajectories of simulated drifters advected by MSEAS currents, as well as the FTLE fields. The trajectory integration time in this study was 6 hours, which is an important timescale for a flow with a strong M2 tidal component, and also represents the upper time limit in the real person-lost-at-sea search-and-rescue operations.…”