The Weather App Trap: Can You Make Decisions Using Weather Apps?

No Gravatar

How do you check the weather? Are you a wet-finger-in-the-air guy? Do you watch the morning news and commit the forecast to memory for the rest of the day? Do you pull out your smartphone and see what your home page tells you? Or do you have several apps on your phone—maybe you have a favorite, maybe you have dismissed a few and are still searching for the right one, or maybe you take the middle ground between the various apps? After all, some apps have “this” and some apps have “that,” but no app has everything. And that includes accuracy.

Accuracy is what separates the good apps from all the rest. If one weather app lets you down more than a couple of times, chances are you’ll move it to your trash bin or ignore it. But it pays to know how your apps work—what’s the “engine” that drives it – because that will have a direct result on the overall accuracy. Many apps run off a computer model or perhaps several and some may give you the choice of what model you prefer to view. Some apps take the model data and convert it into something human friendly, while some are simply a portal to a data supermarket. Hopefully you know what you want and how to get it because it might be daunting!

Interface on tablet and smartphone screens against stormy weather by the seaWhat about other apps? You don’t have to be a meteorologist or work for the government or a weather company to publish a weather app. I just checked Google’s Play Store from my phone and searched for “weather.” I stopped counting when I reached 150 individual apps. Surely some of them must be disappointing, or worse. Could a few of the apps have a teenaged designer who fancies herself/himself as a budding meteorologist? Maybe this self-proclaimed weather expert has given his app a hi-tech-sounding name that gives the app an air of science or importance that just sounds accurate.

More than a week ago one of my apps that uses the ECMWF pegged snow and freezing temperatures moving into the Houston area (and other Gulf Coast states) for New Year’s Eve and New Year’s Day. That’s an attention-grabber, for sure. As ImpactWeather’s StormWatch team was considering the same thing, I thought, “Nice job, app!” The next day however, the app’s forecast had moved warmer—snow replaced by rain and temperatures now in the 40s. Over the past week the app indicated bouncing temperatures and the precipitation has come and gone—and then come and gone again. Can you make decisions based on that? Meanwhile, StormWatch has hung onto the cold and perhaps snowy forecast and opened an ongoing line of communication with our clients—being sure to discuss everything in the communication: the good, the bad; the known, the unknown.

Apps have other drawbacks. Your app can’t call you to say bad weather is on the way. It can’t give you a heads-up that it’s going to snow or that the bottom of the thermometer is going to fall out or that thunderstorms may shut down your operations in six hours. Something else your app can’t do? Express forecaster confidence.

Forecaster confidence is something we always express at ImpactWeather and we express it as high, medium or low. Don’t be confused by probability of precipitation (POP). Models derive POPs, so do human forecasters. Though many consider POP a form of a fudge factor, it’s defined as the probability of measurable precipitation occurring in the forecast area during a given time period. Granted, the difference between POP and confidence may be subtle but there are indeed times when a meteorologist has little faith in what the models are indicating. Combine that with missinNo Signal With This Stormy Skyg or incomplete data (technology can sometimes let you down), and it’s easy to have low confidence. However, sometimes the forecast is so tricky—like an ice storm on the Gulf Coast—that even model consensus and all the data in the world will still cause the forecaster to say, “Wait a second. Let’s take another look at this and not be so quick to pull the trigger.” An app driven solely by model data will plow through with the icy forecast, while a forecast with a human factor can present a more full picture.

And what about when your app crashes, the cell or Wi-Fi network fails or your battery dies?

Smartphone weather apps are great—at least, many of them are—but know what their limits and capabilities are and don’t rely on them as your only source of weather information.

Share:
  • Digg
  • del.icio.us
  • Facebook
  • Twitter
  • email
  • LinkedIn
  • Reddit
  • Google Bookmarks
Leave a comment

0 Comments.

Leave a Reply

You must be logged in to post a comment.