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. 

Input Data:

There are four files in this dataset. Two file are original measurments of ocean surface velocity obtained from High Frequency radars.

1. HFR_MarthaVineyard_2014_original.nc

2. HFR_MontereyBay_2017_original.nc

The above NetCDF files contain east and north velocity fields at two sites.The dataset at Martha's Vineyard, Massachussetts, is collected by HF radar operated by Woods Hole Oceanographic Institue (WHOI). The dataset has 400m resolution and covers July to September of 2014. The dataset at Monterey Bay, California has 2km resolution spanning a larger domain and covers January of 2017. Both datasets are publically availble 

Output Data:

These data have incomplete coverage. With our method, we have resotored the missing coverage. The output files of our method are:

1. HFR_MarthaVineyard_2014_restored.nc

2. HFR_MontereyBay_2017_restored.nc

Visualization of data:

The above datasets can be interactively visualized online at the following links:

• Original Martha's Vineyard data: http://transport.me.berkeley.edu/terriajs/#clean&WHOI_HFR_2014_original
• Restored Martha's Vineyard data: http://transport.me.berkeley.edu/terriajs/#clean&WHOI_HFR_2014_restored
• Original Monterey Bay data: http://transport.me.berkeley.edu/terriajs/#clean&MontereyBay_2km_original
• Restored Monterey Bay data: http://transport.me.berkeley.edu/terriajs/#clean&MontereyBay_2km_restored

Online software:

We have developed a web-based gateway to accompany our manuscript available at http://transport.me.berkeley.edu/restore/ and has been designed to be a community tool to process incomplete oceanographic datasets. The gateway contains a user guide, documentation and sample data, including those that has been used in this manuscript.

Demo video:

A brief video of the tool is demonstrated at https://vimeo.com/274810038. All results of this manuscript can be reproduced using our online tool.