Weather and forecast. Trend of the time or something that has been happening for thousands of years and just now we have all discovered mobile applications and we "play" it meteorologists?
The tools that meteorologists have in their hands have obviously evolved, in terms of monitoring, but also the processing of (innumerable) data.
Always at the end of the day the prediction has to do with the human factor - and its own, separate judgment. This is one of the reasons why one app informs you that it has 12 degrees Celsius in your city (while you are looking for information) and the other that it has 10. Disagreement between applications is not the exception, but the rule. What are they hiding from us?
If you search on your mobile, the word weather and do a research you will find that there are thousands of application results. There are more than 8000 on Android and over 2000 on iOS. This in itself explains how huge this concern is. The reasons are specific. At the moment it holds that no application is good at all. They are all great in one element - what they have chosen to stand out.
There are separate applications for each city, for all cities, live, for the next hours, for the following days, for every sport, even for pets. Their creators are trying to find this one factor that will help them stand out from the competition and gain their own -different- audience.
You would say that they all deal with the same thing (the weather), using the same data (about the state of the atmosphere and the surface of the earth - temperature, humidity, wind), which are powered by super computers, with the ability to do one thousand trillion calculations per second. So many are required, due to the complexity of the forecasting models - which we will talk about later. A study published in 2015 in Nature reported that meteorologists' ability to predict atmospheric pressures three to 10 days earlier is improving at a rate of about one day per decade since 1981.
Why do applications not agree?
Weather forecasting is the No. 1 factor in the "decision making" of our daily lives, as it affects every aspect of our lives: from deciding what to wear and when it is useful to stay in our homes, to predicting natural disasters. The economist has estimated that US GDP could vary by up to $ 485 billion, depending on the weather.
However, as the Guardian reports, "when we try to find out from the relevant applications, we often find that they do not agree on the temperatures in the same places, at the same time - even though they use the same data. What is the reason; Journalist Josh Toussaint-Strauss undertook to investigate the matter. " Toussaint-Strauss felt that he first had to explain how the weather was forecast.
"Think of the earth divided into many different vertical 'boxes' that 'stand' on columns, which extend to the stratosphere. Meteorologists monitor a lot of different data to see what happens - at any given time - in each of these 'boxes'. For example, wind speed, temperature, air pressure, etc. Once they get the numbers for each of these 'boxes', they calculate the differences between them. "With the data they get (obviously not everyone has the same access to everything), they make weather models to predict what will happen in the future."
You realize that no meteorologist has the ability to know everything, for every inch of the world, at any time. The knowledge gap is filled by estimates concerning historical forecasts, in the same area and time. Prediction models - consisting of sets of equations - use as a starting point the older predictions and the extrapolation of available observations, for the calculation of future conditions. "Obviously, this little system also leaves room for mistakes to be made."
The explanation for the lack of accuracy in the forecasts does not apply to the different data offered by each app. Otherwise, everyone would make the same mistakes at the same time. Correctly; Error.
Whoever has the smallest average total error wins
It all depends on the meteorological sources that meteorologists choose to use and how they translate this data. Applications use different algorithms, involving different prediction models, with different levels of detail.
Others trust what the computer does, while some have human resources who put their hand up to supervise and correct the supercomputer, especially in extreme and unusual weather events. Confirmations or corrections are made based on the skills of each meteorologist who is paid for this work, in the interpretation of whatever data is in front of him. Clearly not everyone has the same skills.
Meteorologists give guidance - they do not predict the future
In general, you would not say that weather forecasting is a field in which no mistakes are made. You would say, however, that the percentage of errors judges the effectiveness. In a related question, Weather Underground creator Jeff Masters told Virginian Pilot that "something that happens because each app or station uses its own systems and types to make the prediction. "They all start with the same fundamental mathematical equations that govern the motion of the atmosphere, yet the coding is different."
He added that every forecaster starts with the same - raw - data, a torrent of information coming from satellites, radar and meteorological stations around the world. Everyone decides how the data will be processed and which variables will be relevant to each given situation. "There is room for a small mistake everywhere" and that is why meteorologists clarify that their forecasts are for guidance and not what will actually happen. Some apps achieve reality more often, but others. Which is the best; What gives you the information you want - and it's different from what I want, this or that. I hate to repeat, but the different 'I want' is also the reason for the existence of thousands of different applications with weather forecasts.
The difficulty of Athens - the tornadoes, the storms and the height of the snow
MIT meteorologist Angela Zalucha, who holds a doctorate in atmospheric science from MIT and a lecturer at Iowa State University, explained that "assuming we could place meteorological equipment anywhere on earth, the most difficult places would be where weather changes at short intervals and distances. Some examples include mountains (they are mechanical barriers to flow - they block the rain that is predicted to come) and the limits of dry water (where the thermal capacity of the surface changes abruptly). That is, places that have both sea and mountain, that is, Athens.
"Other places that are difficult to predict are in the middle back, where the weather is characterized by high and low pressure systems and frontal limits (sharp jumps in temperature, wind direction and rainfall). These are very chaotic. There is a mathematical limit to how well we can predict a chaotic system.
Storms and tornadoes are also difficult to predict because a) they are small-scale phenomena and b) we do not fully understand their development and evolution.
For the same reason, snow accumulation is difficult to predict, as the amount of snow that falls to the ground depends on the temperature, humidity and shape of the snowflakes (which in turn depend on temperature and humidity, vertically in the atmosphere). "The physics of snowflake formation is another part of science that we do not fully understand."
These are the reasons that especially in the Attica basin - which is surrounded by both mountains and the sea - the weather forecasts for the coming days often deviate more or less from what we finally see happening in reality. However, this is not a phenomenon only for Attica, but in general in areas surrounded by mountains and sea.
Source: news247.gr