Texas Emergency Management ONLINE2010 Vol. 57 No. 1
HURRICANE ALEX AND FLOOD PREDICTION

Hurricane Alex was the most powerful hurricane to emerge in the Atlantic during the month of June in 44 years. The storm made landfall as a Category 2 some 110 miles south of Brownsville on June 30. Days later, a flood of historic proportions rolled down the Rio Grande and poured into Texas and Mexican communities along its banks. What made the impact of this flood difficult to predict is a complex story.

The storm initially spared Texas while devastating areas of Mexico, including Monterrey, where some areas recorded 30 inches of rain. Then, with something like a boomerang effect, the waters of Alex sloshed back down the mountains to fill watersheds and reservoirs to overflowing. The storm also caused damages to infrastructure in the Panhandle. On Aug. 3, a presidential disaster declaration was issued for 10 counties. Soon an additional 10 Panhandle counties were declared eligible for public assistance – proving, yet again, that hurricanes can cause floods anywhere in Texas, even if they strike Mexico.

Hurricane modeling and prediction has steadily improved through the years. But many challenges remain in predicting the magnitude of floods – especially along a river fed by tributaries in two nations. Dr. Gordon Wells, program manager for the University of Texas Center for Space Research said: “The really unusual thing is that this involved two major [river] sub-basins in Mexico along with the main stem of the Rio Grande. I do not believe that there has been a comparable event since regular measurements began along the Rio Grande in the early 1900s.”

Greg Shelton, the Service Coordination Hydrologist with the West Gulf River Forecast Center, explained that several things go into predicting river flooding. To simplify, this includes data on the amount of rain that falls and models that convert rainfall into estimates of runoff across a particular watershed. Information on the height of the flood (stage), speed of the water (flow) and the volume of water all go into the models used to predict flooding.

National Weather Service (NWS) experts also use historical records of rainfall and stream flow to build their model, double-checking this against actual observations. “River observations are critical in verifying the model simulations that NWS uses to develop river forecasts,” Shelton said.

Shelton said that all river forecasts are subject to uncertainty for various reasons, including inaccurate estimates of past rainfall or inaccurate forecasts of future precipitation – or both. An additional complication is the fact record high floodwaters are simply not going to match up with historical data. “A new record flood is just that – something that has never been recorded before,” Shelton said.

However, one of the challenges in predicting effects of floods along the Rio Grande is a lack of river gauges and personnel to monitor them in sparsely populated areas in Mexico. The same is true to some extent for sparsely populated areas of West Texas. Automated stream gauges only measure one piece of the model – that is the height (or stage) – of floodwater. But “the more stage data that is available [both number of locations and frequency of observations at a location] the more accurate forecasts can be,” Shelton said.

Sally Spener, of the International Boundary and Water Commission (IBWC), said flood management from the U.S. perspective has improved in recent years. “Our Rio Grande gauging stations are now equipped with telemetry so we can receive real-time data about Rio Grande flows via satellite. This has made it easier for the agency to manage flood operations, and also provide instant access to data to other emergency managers or members of the public.” In addition, Spener said, the IBWC sends personnel to the field to do flow measurements “and we have new technology, the Acoustic Doppler Current Profiler (ADCP) measuring device that has facilitated accurate readings.”

Wells explained that Hurricane Alex rolled down the river in three phases. The first flood wave came down the main channel of the Rio Grande, triggered by flooding in the Lake Amistad area and minor tributaries in Mexico. The second wave came down the Rio San Juan from Monterrey and mountains to the east. The third came down the Rio Salado as Alex dissipated three to seven days after landfall in the interior of the state of Coahuila. “The flooding in Coahuila and parts of Nuevo Leon was record breaking,” Wells said.

“Each flood wave converged on the Lower Rio Grande Valley. Fortunately, they did not arrive at the same time,” Wells said. Previous hurricanes (Beulah in 1967 and Gilbert in 1998) produced river flooding along the main branch of the Rio Grande and the Rio San Juan.

Hurricanes can rearrange coastlines, eroding beach sand and cutting new channels through barrier islands. Similarly, the sheer force of a flood can change a river channel while the flood is happening. “Water can scour the river channel, changing the shape of the channel, and changing the relationship between stage and flow,” Shelton said. When the shape of the channel changes, each of the puzzle pieces on which predictions are based will also change.

When information is coming from another country, like Mexico, delays may occur with data communications and intelligence regarding that nation’s reservoirs. “In general, the delays we have sometimes seen in getting data from Mexico greatly improved during the flooding associated with Hurricane Alex,” Shelton said.

NWS is exploring ways to work with Mexican officials “to increase observed data, both rainfall and river data. Securing the resources to deploy new equipment is a very large challenge. Due to sparse population, automated gauging is the more viable solution, provided there is a long-term plan and resources for maintenance,” Shelton said.

Hurricane Alex

Hurricane Alex on June 30, 2010. Here Alex was a Category 1 hurricane on course for landfall in Mexico
and reached Category 2 before landfall.
(Photo courtesy of NOAA)