If you have been around our website, you are already familiar with satellites and how they capture information given by our planet, in the format of waves. Every part of Earth’s surface emits and/or absorbs waves, and different sensors are used to interpret that information and transform it into images – that’s what we call remote sensing. Now that we reviewed that, it’s just natural to ask an important question: how do we use images and data from the present and past to guess what’s going to happen in the future? The articles below will hopefully help you to understand important topics involving forecast – and maybe in the end you will even want to explore our additional resources to become a weather wiz yourself!
First and foremost, it’s important to understand that weather forecast is never 100% accurate; it’s not like meteorologists and other scientists have a time machine that allows them to observe if a blizzard will hit your neighborhood at an exact time and location – at least not yet! Instead, scientists use very complex computer models to go through a lot of data in a very short period of time. That data comes not just from polar and geostationary satellites, but also deep space satellites placed even farther away. Even simple instruments like buoys in ocean’s surfaces and anemometers – which is somewhat like a windmill – help scientists to tell us when to take that extra coat or an umbrella when we leave for the day!
All that data is collected by and/or arrives at supercomputers in weather stations – big or small building strategically mounted in places around the globe. Those computers are like your brain – the computer of computers! If you are into fantasy football, for example, your brain will use information from past seasons and the present time – such as players statistics, team’s history, and etc. – to try and foresee who is the MVP of the season or which two teams will be playing at the Super Bowl.
Maybe you are not into sports, but perhaps like mystery TV shows or books; you know when you try to decipher the big mystery based on small clues that may be given to you throughout the whole story? That’s similar! Our brain tries to use information such as a character’s behavior and history to guess if they are guilty of a crime, for example, just like computers read atmospheric information to try and guess the weather for the next days. The more data you have, the higher are the chances the guess is correct. And our computers can receive and interpret BILLIONS of numbers and data, and the more we advance technologically, the better we become at foreseeing the weather. Of course, very trained personnel are also needed to fix any errors or help to explain any anomaly, or unexpected trends, brought up by the models – humans are still very needed in the weather forecast!
Another important aspect of weather forecast precision is how far ahead you are trying to predict the weather. You probably guessed the trend: the closer you are to the day of your forecast, the most accurate that will be. This is because atmospheric conditions might change rapidly, so the models the computers use are more accurate for real-time weather, and the farthest you are from real-time, more error accumulates from a computer’s calculations. The National Oceanic and Atmospheric Agency (NOAA) shows a 90% chance of accuracy for a 5-day forecast; two extra days for a forecast will lower that accuracy to 80%. If you are trying to foresee the weather 10 days in advance, that drop is even larger: you will only have a 50% chance of getting accurate predictions! That’s why any longer than that, and we will be really out of luck.
The short answer is YES! As you might have learned from previous articles, a TV forecast – which is the result of predictions made by supercomputers and experts in weather stations – can be up to 90% accurate. Yeah, that means there’s still a 10% chance of failed predictions – and the farthest in the future you try to predict the weather, the higher are the chances of having no accuracy at all. However, 90% accuracy is very high when it comes to “predicting the future”, don’t you think?
Another important point that needs to be explored is the difference between deterministic and probabilistic forecasts. Those are big words, but not so hard to understand: deterministic forecast relates cause and effect to predict exactly what the weather will be. As you can imagine, that is not just a great way to give a forecast for the simple reason we don’t have that level of accuracy yet, unless the weather person is predicting the weather for the next seconds. That’s the reason why, now, we hardly ever have forecasts saying we will see rain tomorrow. Instead, they use the probabilistic forecast to say there is a 90% chance of rain, for example. That way, the receiver of that forecast is aware of the chance of rain.
Many times, a TV meteorologist gets your weather forecast wrong – and the reason for that is not just because their models have at least a 10% chance to fail. In fact, there are many other variables that affect TV predictions. For example, you might have heard on TV we would have very high chances of raining tomorrow in Denver, Colorado. When tomorrow comes, the sun might be shining bright all day, and you get mad at the TV forecast because you got your umbrella out of the closet for nothing! While that might be frustrating for some, we need to understand that Denver – and any other city you live – is probably very large, and it probably did rain somewhere in that whole area that we call Denver. So, next time you are watching TV for the forecast, you might have to be a little forgiving with the local weather person; predicting the exact weather at your exact location is not something we normally see in TV forecast, because those professionals normally cover a very large area when reporting their predictions.
Because weather forecast heavily depends on the technology available, it is natural that with advances in satellite technology and computational methods, we become better to predict atmospheric changes. If you are wondering if one day our technology will be so advanced that our forecast will be 100% accurate, it might disappoint you to learn that, no, that probably won’t happen anytime soon – unless, of course, we invent that time machine that allows us to peek in the future to see if it’s raining tomorrow!
When delivering a forecast on a communication outlet, we hear a lot about the chances of precipitation – that is, the chances of water falling from the sky. But did you know that we have many different types of forecast? The list below is some of the important ways different people and industries use the forecast to be better prepared for future weather.
Unless otherwise noted, all content on the CIRA RAMMB: VISIT, SHyMet and VLab webpages are released under a Creative Commons Attribution 3.0 License.