Or: Why is the weather forecast wrong again?
Ring, ring… ring, ring. The phone is ringing. I sit at my desk in the office and pick up the phone: “german weather service, Lisa Brunnbauer“. A male, somewhat rough, deep voice announces itself: “Yes, hello. I have a question about the weather fax, am I right with you?” The man at the other end of the line sounds a bit agitated and hectic. I answer in a calm tone: “Yes, you can ask me questions about the weather fax. What is it all about?” The man explains to me that he is a farmer near Offenburg am Rhein and has been a customer of the weather fax for many years. But this time his concern had nothing to do with agriculture. At first he intended to take photos on the same day, in bright sunshine and a blue sky. But when he now takes a look out of the window, the sky is covered with numerous clouds and the sun only peeps out from time to time. He asks me with indignation: “Where is the sun that you promised me yesterday in the weather report? The weather forecast was not right!“
In shops, coffees, buses and trains I listen more or less involuntarily to the discussions about the weather and the weather forecast. Sometimes I can only smile with difficulty. Often I have to bite my tongue in order not to interfere in the conversations of other people. But as a meteorologist one would like to explain to the people why the weather forecast was wrong again.
Until the finished weather report is published in the newspaper, on the radio, internet or television, numerous steps are necessary. In this longer development process there are numerous “stumbling blocks”, i.e. a quite large number of possible sources of error. Whether a report is apparently correct or not depends not only on the report itself, but also on the person who reads or hears it. In order not to complicate matters, we shall confine ourselves in the following to reports in text form.
Let’s start all over again, with the weather forecast. We do a thought experiment. Imagine you would have to write the weather of the last three days as text. Or perhaps better: at first only yesterday. Everyone hears or reads weather reports almost daily, so this exercise should be easy. One or the other will quickly encounter difficulties. If it wasn’t a steel-blue sky under high pressure, the description, for example the clouds, will be more difficult. Give it a try! How many clouds were there? How long did the sun shine? Was there rain? All day long? Has it poured?
As a meteorologist in the forecasting service, at the beginning of one’s working career one has to swot vocabulary: “weather vocabulary”. In order to describe the weather, its condition and future course correctly, you need a number of different “filler words”, usually adjectives and adverbs. On the one hand because the weather is very varied, on the other hand because for reasons of better legibility, the same word should not be used in every second sentence. However, since a weather forecast is not primarily intended to win literary prizes, but rather to ensure that the content is as accurate as possible, word repetitions cannot be completely avoided, depending on the weather conditions.
Official classifications can be used to describe a weather condition. These are used, for example, by the aviation weather service. This allows them to be used uniformly on an international basis. Let’s take the cloudiness as an example. There are terms such as “bright”, “slightly cloudy”, “heavily cloudy” or “cloudy” which reflect the proportion of total sky coverage. However, the meteorologist often also uses the common colloquial language or refers to cloud types: “many clouds”, “dense clouds”, “veil clouds”, “spring clouds”. However, the individual readers do not know definitions or imagine a different sky image.
However, these expressions are only used to describe the weather element at a certain point in time. The spatial differences as well as the temporal course are not yet seized. Often adjectives and adverbs apply to both spatial and temporal changes simultaneously. They often differ in small but decisive details! Many readers of a report are not even aware of these differences. A meteorologist, on the other hand, may worry about it. For him, it makes a difference whether showers occur sporadically, in places, locally, regionally or in a widespread manner, whether they occur frequently, rarely, hardly or repeatedly.
When evaluating a report, a personal expectation is added. The report says: “Today there are hardly any showers, in most regions it remains dry”. The accuracy of the report is often determined by subjective impressions. For clarification, here are a few examples. The person who takes an umbrella to work in the morning and unpacks it in the lunch break while going to the canteen will be satisfied and say to his colleagues at work: “Rain has already been reported on the radio this morning”. The one who doesn’t carry an umbrella might complain that the weather forecast was not quite right because it should stay “mostly dry”. So one and the same report can be regarded at the same time – depending upon subjective impression – as right or wrong. Psychologists deal with the quite complex topic of this “subjective perception”. Thus the “objective” content of the weather report is subjectively filtered by our expectations. Sometimes, when reading a report, we interpret things into it or hide what we have read in order to obtain the “desired” weather. If a report reads: “Over the next few days, partly cloudy, with occasional sunny spells and occasional showers,” a farmer eagerly awaiting hay weather will focus on the “sunny spells” and hope for dry, friendly weather. Another farmer, whose grain is in the grain filling phase on the field at the same time, will focus on rain, i.e. the “individual showers”.
So much for theory. Well, there I am sitting in my office. On the phone is the somewhat angry man, who would have liked to take photos today in the glistening sunlight. As always with telephone enquiries, I first take a look at the current satellite picture and see a row of cloud fields moving across the Rhine Valley. Cloud fields that occasionally allow a ray of sunshine: as predicted yesterday! The farmer asks me again: “What happened to the sun you predicted yesterday? Can we still get it today?” I answer him calmly that yesterday’s forecast for his region is absolutely correct. He replies somewhat indignantly: “But you wrote yesterday that it will be sunny“. I deny: “Yesterday’s weather forecast is correct. That it would be sunny was not predicted“. But the man insists vehemently on it: “It should become sunny!” Puh, I know exactly what I had written, there was no mention of “sunny”. I ask the man to get the weather fax from the day before and to read me on the phone what is written there. A rustling in the receiver, silence and another rustling. The man fetched the report. He begins to read: “After the local early fog fields have disintegrated, there are some denser clouds and some longer friendly sections with sunshine. It remains mostly dry at the same time.” Silence on the phone. I hear nothing more. The man is silent. He hesitantly begins to speak again: “Hm, there is actually nothing about sunny. I am sorry. Somehow I remembered it differently. Strange.” Yes, at that moment it was strange for him. For you, after our digression to the “subjective evaluation of weather reports”, it is no longer that. And also the man on the phone quickly understood it after a short explanation on my part and we both laughed heartily.
