WXforum.net

General Weather/Earth Sciences Topics => Earth Sciences => Topic started by: W3DRM on November 28, 2015, 07:32:12 PM

Title: IBM Buying Technology Side of The Weather Channel...
Post by: W3DRM on November 28, 2015, 07:32:12 PM
An interesting article I came across today in my email. I wonder what "weather sensors" they are referring to?
Don - W3DRM



IBM wants to predict earthquakes and volcanoes with Watson - No more guessing?
Written by Mike Murphy
November 23, 2015
 
We may soon have categorical evidence that living in San Francisco is a terrible idea.
 
IBM announced on Nov. 20 that it had created an award-winning simulation of the Earth’s tectonic plates that could soon be used to make predictions about when the next great earthquakes and volcanic eruptions will occur. And its artificial intelligence system, Watson, may prove to be the computer brain that can tell us when it’s time to get out of the Bay Area.

 A team of computer scientists at IBM, in partnership with researchers from the University of Texas at Austin, New York University and the California Institute for Technology, created a model that simulated the entire flow of mantle under the Earth’s surface. The model is so complex that it had to run on the Sequoia supercomputer (the third-fastest computer in the world), which the company built for the US government. The team’s model last week won
http://www.hpcwire.com/off-the-wire/gordon-bell-prize-awarded-to-ibm-and-leading-university-researchers/ (http://www.hpcwire.com/off-the-wire/gordon-bell-prize-awarded-to-ibm-and-leading-university-researchers/) the Gordon Bell Prize, the top award in supercomputing. Costas Bekas, an IBM researcher on the team, told Quartz that while we understand how the Earth’s crust moves, our fundamental knowledge of what’s going on under the surface is still quite limited. “We much more about outer space than the interior of our planet,” he said. That makes it hard to predict and prepare for disasters caused by tectonic shifts—be they earthquakes, volcanoes, or tsunamis. “How can you predict a phenomenon that you don’t understand?” Bekas said. The ever-moving molten mess below the Earth’s surface does not act uniformly, nor are the gaps between plates regularly spaced. This makes the mathematics “very, very difficult,” Bekas said. The research team took readings from seismic sensors across the planet, and paired them with the pervading geological theories for tectonic movements. Using the data, they created what they believe to be far more accurate mathematical models for the Earth’s interior and ran those models on the Sequoia supercomputer. Were you somehow able to run this model on a regular home computer, it’d take you three years to get a sense of what’s going on under the Earth’s surface. The Sequoia can do it in a day. The result does not yet predict when natural disasters are likely to happen, but it does potentially give geologists and seismologists a tool for creating such a system in the near future, Bekas said. Now that there’s a model in place, the next steps will be to feed it more information to improve it. The increasing proliferation of internet-connected sensors will provide more seismic data points in the coming years. Watson has the ability to find patterns in massive data sets, as well as analyze human writing. If you take every geological research paper ever written, throw in every data point from every seismograph in the world, give them to Watson, and combine those with advanced simulations such as the one performed on Sequoia, the result, Bekas said, would be a system that “will profoundly increase our understanding of the geological processes that drive natural disasters.” And the more data points there are, the more accurately Watson should be able to understand big tectonic movements. IBM is working with the US Department of Energy on this model, as well as its work on attempting to predict the weather http://qz.com/535345/ (http://qz.com/535345/). The company also recently bought most of the assets of Weather Company—the owners of the Weather Channel—including its myriad weather sensors around the world. IBM previously told Quartz that the acquisition “will coalesce around cloud, Internet of Things, and Watson.”

Why does IBM care about being able to predict the weather, let alone when the next earthquake is happening? Because it is looking for impressive AI feats to show off to perspective clients of Watson, on which it has already bet about $1 billion. Being able to show multinational corporations that it knows if Mauna Loa is about to blow might well help it win their data analytics business. That could truly result in some explosive growth.

http://qz.com/556172/ibm-wants-to-predict-earthquakes-and-volcanoes-with-watson/ (http://qz.com/556172/ibm-wants-to-predict-earthquakes-and-volcanoes-with-watson/)
Title: Re: IBM Buying Technology Side of The Weather Channel...
Post by: Jáchym on November 28, 2015, 07:42:45 PM
Hmm... interesting, but I am still a bit sceptic about this...

Despite the fact Im sure this supercomputer is really fast and the model quite sophisticated, IMHO it is impossible to ever completely simulate the processes in the planet core. And this also applies for forecasting weather - there will never be a computer that would be capable of simulating all the processes in the atmosphere. The major problem is that both the atmosphere and the inner core are affected by millions of factors and most importantly, there is this so-called "butterfly effect" (https://en.wikipedia.org/wiki/Butterfly_effect). It could therefore take into account 999 things and not that one single one and that could be the one that would determine the whole subsequent progress of things.

And in addition, I saw a documentary a few years ago, where they also mentioned the fact that some processes in the atmosphere are actually really random and can therefore never be predicted. They said that although our forecasts and their accuracy are getting better each year and we are also able to predict more and more distant future (they showed graphs of forecast accuracy in the last 50 years for various time periods), but... they also said that due to the laws of physics and the randomness factor, again in combination with the butterfly effect, there is a point beyond which we can never go.... we will probably one day reach a point where we can no longer make our forecasts more accurate (we are not there yet ;)

Same with the earthquakes... you can make a model , which can be very sophisticated, maybe even predict some, but the problem is that you can never rely on it 100% and then this obviously means that you will still have to be prepared all the time...
Title: Re: IBM Buying Technology Side of The Weather Channel...
Post by: Old Tele man on November 28, 2015, 08:09:40 PM
...making mistakes at the speed of light!
Title: Re: IBM Buying Technology Side of The Weather Channel...
Post by: ocala on November 29, 2015, 07:36:35 AM
About the only way to correctly forecast weather is to launch millions of nano sensors at different heights and then combine that with with satellite, radar, and ground sensor data and crunch all that  into a program.
Title: Re: IBM Buying Technology Side of The Weather Channel...
Post by: gwwilk on November 29, 2015, 01:08:24 PM
About the only way to correctly forecast weather is to launch millions of nano sensors at different heights and then combine that with with satellite, radar, and ground sensor data and crunch all that  into a program.
The ONLY 'computer' that can process all of the data and produce an infallible output is the 'real-time-computer' we call 'the weather'*.  The inevitable flaws in any attempted prognostic endeavors will manifest themselves sooner rather than later.  As a physician, now retired and out of the prognostication business, I dealt with 'forecasting' regarding the human body's responses to insults, injuries, illnesses, age, etc.  The best predictors were always based on applying past similar outcomes to the present situation.  That's also what experienced weather forecasters do while using all of the technological tools at their disposal.  Nobody can predict the future with 100% accuracy.  Forecasts come without warranties for a very good reason, i.e. they're fallible and only probabilities.  Given these limitations I think the NWS does an excellent job.

*You could consume all of the computing power in the solar system in an attempt to produce an infallible model of Earth's weather, but you wouldn't achieve that goal.  You would only demonstrate that the act of intricately observing a highly complex system changes it unpredictably.
Title: Re: IBM Buying Technology Side of The Weather Channel...
Post by: Old Tele man on November 29, 2015, 02:29:59 PM
[above]...Mother Earth's "Butterfly Effect" is *why* weather forecasters are wary of "butterflys" (wink,wink)!