As the saying goes, “The trouble with weather forecasting is that it’s right too often for us to ignore it and wrong too often for us to rely on it.” It’s easy to make jokes about the fallibility of predicting the weather, but we’ve actually gotten much better at it–especially when there is a great deal at stake.
An article in our June issue, for example, looked at some of the different ways cities and project managers are predicting weather conditions–and the likelihood of flooding–to save money, protect property, and evacuate an area in time to save lives. The city of Victoria, TX, recently used XP Solutions’s xpStorm to model flooding of streets and structures and compared the model to citizens’ reports of floods; engineers were able to analyze how potential improvements to the storm sewer system would change a neighborhood’s hydrology and reduce flooding–or not–before they actually started digging.
In another part of Texas, the Trinity River Authority was working in a floodplain to rehab miles of storm sewer pipe and needed to be sure that no water would flow through the pipes before newly placed linings had time to cure. It was necessary to have at least five dry days at a stretch, which is notoriously difficult to predict. Project managers used a rainfall monitoring system from Vieux and Associates to determine the chance of precipitation, as well as the chance of excessively high water levels in the local reservoir–both of which could spell disaster for the project–and to notify project personnel via e-mail of the current conditions so they could make informed decisions about scheduling the work.
Another use of prediction systems, of course, is to alert people to get out of the way of a hurricane or typhoon. That’s just what scientists are doing with some extremely sophisticated models, and they’re pinpointing the results to ever-smaller areas, as the article on page 20 of this issue illustrates. The author recently traveled to Taiwan, where he spent time with scientists from various research institutions and the National Science and Technology Center for Disaster Reduction. He also interviewed Bill Kuo, the director of COSMIC, the Constellation Observing System for Meteorology, Ionosphere, and Climate, in Colorado, who is working on forecasting heavy rain at an extremely localized scale. Born in Taiwan, Kuo was six years old when a typhoon destroyed his family’s farm–and coincidentally set him on his life’s work of predicting the weather.
Using a number of different techniques, scientists in Taiwan and the US are calibrating their models in part by looking at Typhoon Morakot, which struck Taiwan in 2009 and killed more than 400 people in a single village. Using data from Morakot as a test case and running repeated simulations, they are now able to predict where individual clouds might form, how much moisture they’re likely to carry, and most importantly where they will release that moisture. In mountainous terrain like Taiwan’s this becomes especially important for knowing where to plan evacuation routes, where to locate pump stations, and other activities–and it will have increasing importance around the world as well as we experience more Hurricane Sandy-like storms.
What has been your experience with the newest generation of modeling or prediction systems? Let us know at [email protected], or leave a comment on our website at www.stormh2o.com.
