CBI-TCOON: An Integrated Observations and Operational Predictions System for Gulf of Mexico Coastal Waters

Starting in the late eighties the Conrad Blucher Institute (CBI) with local, state and federal partners has built a large coastal ocean observation network. CBI manages the 30 stations of the Texas Coastal Ocean Observation Network (TCOON) as well a number of other stations including the Texas stations of the National Long (NWLON), the stations of the Texas PORTS systems and stations managed for the National Parks Service, River authorities and other local entities. The overall DNR provides real-time or near real-time coastal measurements including water levels, wind speeds and wind directions, barometric pressures as well as other variables such as dissolved oxygen, salinity, water currents and wave climates depending on the station. The primary uses of the data has been to establish tidal datums, provide information during emergencies such as the approach of a hurricane. Other uses include navigation and dredging, the and planning and conduct of a variety of other professional and recreational coastal activities.

Starting in the mid-nineties CBI developed highly automated software to manage all aspects of the observation network The software and procedures were developed on the principle that all user interaction with the data management system takes place via web-based interfaces. The software is based on open source technologies such as Linux and Perl such that CBI is not subject to changes in proprietary systems and has the flexibility to replace software components as new technologies become available or as the needs of CBI evolve. Most software is available under the General Public License. During the past 10 years CBI systems have been expanded to take advantage of flow of real-time data from the entire Texas coast and design and implement real time models based on Artificial Intelligence (AI) Techniques such as Artificial Neural Networks and Random Forests. When time series encompass most encountered conditions such models can be trained to quantify relationships between past observations and future outcomes for the variable of interest. Advantages of this approach include the ability to model non-linear relationships and the implicit inclusion of the boundary conditions and forcing functions. Also once the models are trained AI based predictions are computed virtually instantly. CBI has developed and implemented models to predict water levels including storm surges, water temperatures and currents. Predictive models also include atmospheric predictions provided twice daily by the National Weather Service. The design of the models will be briefly presented and their use in situations such as low water level events or providing guidance for the closure of waterways will be discussed. Lessons learned while working with and interacting with CBI user based will also be discussed.