There are a couple ways in which WeatherBug is working to help improve forecasts.
•1) WeatherBug data is used by NOAA meteorologists to compare the output of their forecast models to what is currently happening. The 120 or so forecast offices around the USA are also looking at WeatherBug data on a continuous basis so that they can adjust forecasts accordingly as they conditions changing in real time.
•2) NWS forecasters use WeatherBug real-time data verify the need to issue alerts during severe weather situations.
•3) WeatherBug is implementing the National Digital Forecast Database (NDFD) into all of their products. The NDFD is a forecast produced by meteorologists at the 120+ National Weather Service forecast offices. The NDFD is a forecast which is based on a 5km grid that has a time resolution down to 3 hour intervals. Having spatial and temporal resolutions on that low scale allows WeatherBug to give more pertinent and accurate forecast information for the individual’s location and time.
•4) Addition of more now-casting capabilities into our products. Our CEO has done some presentations on what he calls the “surprise factor”. The closer you get to an event, the bigger the surprise factor if the forecast you are expecting is wrong. I would say that as the time draws near to the event or time-frame you are interested in, the more you are using things like radar animations and looking at the current temperatures across your region to make your own prediction as to whether or not it is going to rain, or how hot it will be.
The map below shows how the NDFD forecast is different from actual observations. In the southeast part of Texas, it is actually much colder than what the NDFD forecast is predicting. This is likely because the forecast was made many hours ahead of time. That is why real-time data and information is so much more important as you get nearer to the event.
The NWS also has a NOWCAST product that we display along with the current forecast which gives you an idea of what will be happening in the next hour or 2. These update on a regular basis, especially when severe weather is in the area. Below is an example of a NOW-cast.
This is the first part of the article. I’ll be adding part two as I get all the formatting straightened out. Transferring a Microsoft Word document to WordPress isn’t fun by any stretch of the imagination. Not to mention the fact that several photos aren’t displaying properly. Most likely do to the fact that they were transferred from an Apple machine. So for now the second part of the email interview will not include these displays.
Do you feel popular televisions programs such as the Discovery Channel’s Storm Chasers reinforce the notion that forecasting weather events is still a best guess approach? When watching these types of programs viewers see how big a part technology plays in determining weather patterns. Yet in the end we see that forecasting still seems to have a lot of prediction left in it.
Predicting the type of weather you see on Discovery Channel’s “Storm Chasers” is a whole new realm of weather forecasting and prediction. The forecast models are generally good at predicting large scale weather…IE, it will be stormy or rainy in this general location today…but predicting exactly where a big storm or tornado is going to hit is something that we are not able to predict with certainty days or even hours ahead of time. In these cases, meteorologists are relying more on the live weather tools like Doppler radar and ground observations from the WeatherBug Network that can help alert them of a very severe storm or tornado. The current state of the art is giving *maybe* a 5-15 minute window of warning to people ahead of an approaching tornado and that is usually ONLY after a tornado has already formed.