Floods have been devastating recently. Especially the ones last year in Pakistan that flooded an area that was larger than many European countries combined. Many lives were lost that could have been saved, had there been a flood forecast. Early warning systems can prove to be absolutely essential, especially in densely populated regions where whatever time that can be bought saves as many lives.
IBM is apparently testing a new system that utilizes weather patterns and the maps to accurately predict the river behavior as much as 100 hours in advance. Over four days in advance before a water catastrophe is about to strike you. This time is ample enough to initiate evacuation process as well as plan out rescue operations. Well at least you can get as many people far from the high impact regions.
The researchers are are busy testing the same in a 230 mile long Guadalupe river in Austin across 9,000 miles of its tributaries. The system is able to predict the next 100 hours of river activity every hour. IBM researcher, Fadi Gebara stated:
We’re taking in a series of data, running it through a simulation and analytics engine, and getting an output of the flow and depth of the river. If you compare the flows and depths to the topography of the [surrounding] land, you can make a prediction of where flooding will occur.
The system is based on IBM’s already powerful weather modeling system call Deep Thunder as well as the ultra fact POWER 7 Microprocessor.
The system is readily usable but it relies heavily on the data input. The better the data, the more accurate are the predictions, perhaps the lack of this is exactly the reason why Pakistan might just have to wait before it sees the implementation.
This is another big one from IBM as it is also in line to produce chips that are capable of mimicking the Human Brain.