5th – 11th January 2025 Altnaharra 3 hour forecast temperature verification Data from the UKMOUKMOThe Meteorological Office is the United Kingdom's national weather service. It is an executive agency and trading fund of the Department for Business, Energy and Industrial Strategy Weather App
I doubt if you’ll ever see a temperature forecast as way out as this one got. It was issued by the Met OfficeUKMOThe Meteorological Office is the United Kingdom's national weather service. It is an executive agency and trading fund of the Department for Business, Energy and Industrial Strategy in their weather app on the 5th of January for Altnaharra in Sutherland. The blue line in the top chart shows three hourly forecast temperatures for the next seven days. The red filled area shows air temperatures taken from AWSAWSAutomatic Weather Station observations for Altnaharra. The lower bar chart, shows the difference between the forecast and the observed. Although the forecast for the 7th to the 9th was reasonably accurate, the 10th and 11th got progressively worse, with forecast temperatures being as much as 20°C too high at one point. I’m not sure what NWPNWPNumerical weather prediction uses mathematical models of the atmosphere and oceans to predict the weather based on current weather conditions. model the Met Office use for these site specific forecasts, but it has got some severe problems, and not just with record cold temperatures I might add.
I first came across the term ‘prebaratic’ or ‘prebar’ when I joined the Met OfficeUKMOThe Meteorological Office is the United Kingdom's national weather service. It is an executive agency and trading fund of the Department for Business, Energy and Industrial Strategy in 1970 as an assistant. To me the term was synonymous with the ‘T+24 forecast chart’ which we received at outstations by MOLFAXMOLFAXAn early type of fax machine using wet paper.* from Bracknell every six hours. I never realised until now that ‘prebaratic’ was a rather crude acronym for “barometric pressure distribution over the Atlantic” until I read this in the 1956 book ‘Weather Map’. It’s hard to believe that these prebaratic charts were the very forecast charts which we take for granted these days.
* This is a MOLFAX recorder which was an essential piece of equipment at outstations. This is an image of a MUFAX machine made by MUirehead. I’m sure we didn’t call it MUFAX back then, but MOLFAX, don’t ask me what that acronym stood for, but you can be sure Met Office was in there somewhere. The Met Office have always been good at inventing acronyms. 😉
The Met OfficeUKMOThe Meteorological Office is the United Kingdom's national weather service. It is an executive agency and trading fund of the Department for Business, Energy and Industrial Strategy have just introduced a new feature to their forecasts you find on their weather app or on their website. As well as including the probable daily maximum and minimum temperatures for the next seven days, they also now include values for the highest and lowest possible temperatures for that day too. The BBCBBCThe British Broadcasting Corporation (BBC) is the national broadcaster of the United Kingdom, based at Broadcasting House in London. It is the world's oldest national broadcaster, and the largest broadcaster in the world by number of employees, employing over 22,000 staff in total, of whom approximately 19,000 are in public-sector broadcasting. do the same thing in their forecasts, but as far as I know, it remains hidden and unused in the HTML. Personally, I can’t see what benefit this ‘fudge factor’ is to any forecast, other than adding a degree (pardon the pun) of uncertainty to the end user – which temperature is it going to be today or tonight, probable or possible? It also makes a bit of a nonsense of having fixed hourly temperatures, which remain presumably of the ‘probable’ variety. I’ve been doing some verification work on forecast three hourly temperatures over the past year, and I’ve found their forecast temperatures vary dramatically from the actual reported values beyond T+36. It’s a little bit early but I’ve not noticed any significant improvement since I started using the new beta data. All I can say is if that people do complain to them about the accuracy of their temperature forecasts in the future, this will be a very neat way of getting them out of a fix. 🤨
The UKMOUKMOThe Meteorological Office is the United Kingdom's national weather service. It is an executive agency and trading fund of the Department for Business, Energy and Industrial Strategy are going overboard at the moment about high UVUVUltraviolet (UV) light is electromagnetic radiation of wavelengths of 10–400 nanometers, shorter than that of visible light, but longer than X-rays. levels and the dangers of getting sunburnt. It’s a bit ironic, because so far (June 14th) 2024 has seen the coolest and cloudiest start to a meteorological summer in at least 23 years. What’s puzzling is that they never mention in their forecast that UV levels can be considerably higher on a mountain than on lower ground. I used AIAIArtificial intelligence is intelligence - perceiving, synthesizing, and inferring information - demonstrated by machines, as opposed to intelligence displayed by animals and humans. in the form of ChatGPT to confirm what I already suspected. I well remember two colleagues at RAF Kinloss coming to work for a night shift, both as red as a beetroots, after a days hillwalking on Ben Wyvis on a sunny day. Back in the 1980’s people weren’t so concerned about the risk of skin cancer from getting sunburnt, but today skin cancer are much more common.😮
I download the site specific NWPNWPNumerical weather prediction uses mathematical models of the atmosphere and oceans to predict the weather based on current weather conditions. data that resides in the HTML the Met OfficeUKMOThe Meteorological Office is the United Kingdom's national weather service. It is an executive agency and trading fund of the Department for Business, Energy and Industrial Strategy weather application requests whenever anyone looks for a forecast for a location for a number of my applications I’ve written to visualise the forecast data in a table, graph or on a map. Parsing the data was a tricky business, but I persevered, and can now grab a week of one and the three hourly data for any number of elements including temperature. As well as visualising the data, I thought it might be interesting to do a spot of forecast verification by comparing the three hourly forecast data with the actual values from SYNOPSYNOPSYNOP (surface synoptic observations) is a numerical code (called FM-12 by WMO) used for reporting weather observations made by manned and automated weather stations. SYNOP reports are typically mad hourly and consist of groups of numbers (and slashes where data is not available) describing general weather information, such as the temperature, barometric pressure and visibility at a weather station. observation for any location in the world. I know that the forecast values although quite accurate, are far from being spot on. The question I was intrigued to find out was just how accurate they are. Here are some recent preliminary results I have produced from the add-on to my UKMOUKMOThe Meteorological Office is the United Kingdom's national weather service. It is an executive agency and trading fund of the Department for Business, Energy and Industrial Strategy NWP application I wrote a number of years ago.
Notice the very warm day on the 6th of April associated with storm Kathleen, and how underestimated temperatures at Kinloss were because of a slight foehnFoehnA foehn, is a type of dry, relatively warm, downslope wind that occurs in the lee (downwind side) of a mountain range. It is a rain shadow wind that results from the subsequent adiabatic warming of air that has dropped most of its moisture on windward slopes (see orographic lift). As a consequence of the different adiabatic lapse rates of moist and dry air, the air on the leeward slopes becomes warmer than equivalent elevations on the windward slopes. effect.
Generally a pretty good result with temperatures +/- 2°C at Heathrow.
Again at Exeter temperatures within +/- 2°C of the forecast, although it didn’t do well with some of the minima, and the 8th of April was a bit of a disaster because heavy rain suppressed temperatures.
I’ve noticed that recently the Met Office are in the process of updating the NWP data their app uses, so they must have some concerns themselves about its accuracy, although the changes in the NWP might have more to do with forecast weather, rather than forecast temperatures. At the moment I am still using the old data and haven’t switched to the new trial data. In the meantime, let me know about what you think about the accuracy of the forecasts the Weather App produces in your area. I’ve still got a bit more testing, tweaking and bug fixes to do to my verification application, but I’ll keep you posted.😉
So a daytime maximum on Friday the 11 November 2022 of 18°C in the north of Scotland is termed Mild, whilst at the same time a maximum of 17°C in the southeast of England is termed Exceptionally Mild. Why the disparity? Also notice the difference in wordcount which also reflects the UKMOUKMOThe Meteorological Office is the United Kingdom's national weather service. It is an executive agency and trading fund of the Department for Business, Energy and Industrial Strategy preoccupation with regions outwith London and the south east.
