Sensors Mag

Better Modeling with A Bigger View

October 12, 2007 By: Melanie Martella, Sensors


E-mail Melanie Martella

One of the signal benefits of increased processing power is the ability to create better models of real-world processes. When computational power is limited, we are forced to make simplifying assumptions to model and understand the gross forces at work in a system—the infamous "assume that the object is a perfect sphere ...". Weather models, for instance, have improved radically over the last ten years and, with the addition of real-time storm data, they should get even better.

Lightning, Thunder, and Simulations

The recent article "Drone, Sensors May Open Path Into Eye of Storm" from the Washington Post highlights NASA's work, in conjunction with the U.S. Office of Naval Research, to acquire real-time, more complete data from hurricanes to predict just how powerful an oncoming storm may become.

Apparently there is a strong correlation between the rate of lightning strikes and the intensity of a hurricane. By incorporating data from ground-based, long-range lightning sensors(part of the Pacific Lightning Detection Network), researchers can explore this interaction and improve their forecasting abilities. As Kirt Squires says, in NASA's description of the project, "What's really compelling about the new sensors is their increased sensitivity to pick up lightning's electromagnetic signal over water from such a long distance. As a result, we can see thunderstorm activity over the ocean from thousands of miles away for the first time. This development is essential to improving the way meteorologists can look at a growing storm to judge just how harsh it will be." The earlier you know what to expect, the more time you have to make pre-landfall precautions.

Central to the success of this endeavor is the ability to collect and process the vast amounts of data involved. The clearer a picture of the forces at work in a storm, the better your models. The better the real-world data you collect, the greater your ability to compare your model's behavior to what actually happens. The better your models, the better your forecasting ability. The better your forecasting, the more warning you can give to the people and areas affected. Sounds like a win to me.

Let it Snow

Here in New Hampshire, I'm less worried about hurricanes than I am about snowfall. So I read about Colorado State University's snow sensor study with great interest. Last winter, 17 sites across the U.S. tested ultrasonic snow depth sensors against manually collected snow-depth data to see whether the sensors performed well enough to be included in the National Weather Service's (NWS) automated weather stations. The project continues this year. If you, too, are interested in your local snow cover, you may want to bookmark the NOAA's Recent Snowfall and Snowdepth Maps page.


Add Comment




IIoT University


Deep Learning for Vision Using CNNs and Caffe: A Hands-on Tutorial – 9/22/16 – Cambridge, Mass


IDE






Sensors 2017 Call for Speakers


Sensors Midwest


Advertise


Subscribe



Twitter Feed

Find It Fix It Forum

Sensors invites you to join the Findit-Fixit Forum, where you can get answers to your sensing questions—concerning technologies, products, methods, applications, and services--and also offer help to your fellow engineers. The Forum covers all kinds of topics, from the basics to the extraordinary.

Join the discussion!


© Copyright 2016 Questex, LLC. All Rights Reserved. Sensorsmag. Privacy Policy | Terms of Use

If you are having technical difficulties or considerations, please contact the webmaster.