Hurricane track prediction

[2hats]

One problem with working out where hurricanes will end up is that small to intermediate scale fluctuations in the turbulent wall, that are not resolved by numerical weather modelling codes, can have a substantial influence on storm progression. See for example the variation in predicted tracks of recent Hurricane Matthew for 0-5 day forecasts:

A team at UC San Diego have developed efficient control methods for managing (keeping evenly distributed with minimal energy expenditure by taking advantage of differing wind levels in the storm) swarms of low cost, lightweight, sensor laden balloons that they propose to deploy into developing hurricanes so that they may be monitored in situ, 24/7 at higher resolution than at present.

The development of the small (~3kg) robust, energetically-efficient, buoyancy-controlled balloons that can survive, without significant accumulation of ice, in the cold, wet, turbulent, electrically active environment of a hurricane, is underway with a view to testing them in due course. They will use onboard GPS and cellphone-grade tech, drifting as a swarm and periodically reporting position, temperature, pressure, humidity and wind velocity via a satellite link.

DOI:10.1103/PhysRevFluids.1.060507

 

[StoneRoad]

Hope they can get the info from those balloons, but designing / developing them should be an interesting challenge - collecting that data should mean more accurate track predictions, which would be a good thing. More good quality warnings should assist infrastructure resilience and increase survivability for those people and areas subjected to hurricanes.