Hurricane Prediction Tools

2000/07/25 Roa Zubia, Guillermo - Elhuyar Zientzia

• Klima

Although the islands and beaches of the Caribbean are essential within the idea of paradise, do not forget the hurricanes that occur in them. Not only do they appear in stories of pirates or sailors. Hurricanes are part of reality. The most abundant natural deterioration in these areas.

One of the characteristics of these phenomena is the absence of a fixed direction in the route. Sometimes they get rid of anything else and sometimes they are less violent. At present, the National Weather Service of the United States has a system of precalculus of these factors. There are many people who need it.

A group from the University of Rhode Island (URI) has prepared a model to increase the accuracy of hurricane predictions. This model relates hurricanes to the behavior of the sea. The accuracy is estimated at 30% the improvement of the system used to date. It is the result of six years of work. During this period, many data have been collected, including those of the unexpected stations of hurricanes. Together with this work, URI researchers Isaac Ginis and Lewis Rothstein have become members of the national system of predictions. This system of predictions contemplates the tropical storms that form in the Gulf of Mexico and the Atlantic.

Escape beta

The objective is to announce the route and strength of hurricanes. In this way, the announcement will be reliable for the inhabitants of the area. "The error of our model is 90 miles (about 145 kilometers), when there are 24 hours of dry storm," says Rothstein. «With a breakthrough throughout the whole day is a great detail».

The key to the precision of this new system lies in the linkage of both theoretical models. The URI groups have developed an oceanic behavior model that has been adapted and combined with the hurricane model of the NOAA (National Oceanic Atmospheric Administration). The interaction between the atmosphere and the sea is very important in the evolution of storms. The factors related to this interaction can, therefore, be used to follow the changes that occur. One of these factors is the surface temperature of sea water.

It is a computer model. It will be in operation at one of NOAA's headquarters and will send real-time ads to the National Hurricane Center (NHC) laboratories. There, taking into account the information of other models, the last general prediction will be prepared.

NHC teams are eager to try out these new models. Isaac Ginis and Lew Rothstein have done a pioneering job in unifying both models," says Max Mayfield, president of the model. The system used so far in NOAA was developed in 1995 in its laboratories. This program is underway in a NOAA headquarters for monitoring the storms that form in the east of the Pacific and in the Atlantic zone. This system is, in general, suitable for predicting the direction of storms. However, it does not properly inform of the intensity. The vitality of a hurricane is closely related to the surface water temperature of the sea.

The current system supports an unchanged temperature. With the results obtained with this model, in most cases, predictions are made more intense than the real force of storms.

Hurricanes are the most violent storms and one of the most destructive natural disasters. The communities of the East of the United States and the Gulf of Mexico suffer its consequences every year. The model of Ginis and Rothstein analyzes the circulation of temperatures in the sea. In the results, strength and changes in temperature have been completely related.

Term of commissioning

These theoretical results have been confirmed with data from hurricanes collected in the previous two years. The model's authors have received $315,000 from the National Weather Service to launch the new system in the next three years. The sea and atmosphere data that are saved are already available to improve the model and get maximum performance.

For their part, Ginis and Rothstein have created Accurate Environmental Forecasting, Inc. to sell weather information to insurance companies. The scientific base of the new model has been published since April 1999.