In collaboration with faculty members of the College of Science and Technology, the TAMUCC Center for Water Supply Studies (CWSS) and the Weather Forecasting Offices of Corpus Christi and Brownsville, DNR develops and implements forecasting techniques for coastal variables. The forecasting methodologies are based on Artificial Neural Networks (ANNs^?), statistical techniques, harmonic analysis and other techniques such as rule based decision systems. The main projects are listed below.

Water Level Forecasts

A main focus of DNR is the prediction of water levels as an aid to navigation in the coastal waters of the Gulf of Mexico and as a tool for emergency management. Several models have been developed including an Artifical Neural Network model which provides real-time water level forecasts and improves substantially over harmonic forecasts (tide tables). The different water level projects are listed below.

Water Levels & Wind Predictions at Bob Hall Pier

Artificial Neural Network and Persistence Water Level Forecasts

Test page for ANN water level forecasts at Bird Island Basin

For present water levels and historical water levels during hurricane, consult this other DNR webpage: http://lighthouse.tamucc.edu/Main/HurricaneAwareness

Statistically based water level forecasts and gap filling

Harmonic analysis and prediction of water levels

Wiki testing page for ANN models

Atmospheric Forecasts

As part of a collaboration with the Weather Forecasting Offices of Corpus Christi (CCWFO) and Brownsville (BWFO) DNR is archiving historical atmospheric predictions for various locations along the coast of Texas. The predictions are extracted from several NCEP models and are sent four times a day to DNR from CCWFO. These atmospheric predictions are used to compare the different model predictions with measurements and as input for other prediction models such as the computation of water level predictions.

Water Temperature Forecasts

A joint project with the Coastal Conservation Association (CCA), Texas Parks and Wildlife and the Gulf Intracoastal Canal Association (GICA) to predict water temperatures in the Laguna Madre. Unusually low temperatures can result in substantial fish kills. A model predicting low water temperatures below about 45 F with about 24 hours notice would allow coastal stakeholders to take some measures to try to minimize the fish kills. A prediction model was developed to predict water temperatures in the middle and upper Laguna Madre. The Model can be accessed at the following link:

http://lighthouse.tamucc.edu/Forecasts/WTPTests

Other Forecasting Projects

DNR and its partners have worked on prediction models using Artificial Neural Network and Rule Based systems for other coastal and environmental variables such as prediction of recreational water quality (enterococci colony forming units) and spring flows in a karst aquifer. If you would like more information on these models we woudl be glad to discuss our research.

For more information on DNR forecasting projects, E-mail Philippe Tissot : ptissot@lighthouse.tamucc.edu