Bod

Morning all Still clinging onto some slush here. Visited a few customers last night and the roads weren't too bad, so long as you were careful. Interestingly, where there were junctions from ungritted side roads the slush and ice tended to get dragged out onto the main road making it very slippy but only in sparse patches. For instance, for anybody that knows the area, Queen Anne Drive/Magna road had a patch coming out of the Hamworthy Club. Hit it and immediately I felt a small lateral slide. It leaves you in a quandry because the speed that I judge to be safe is slower than the speed that about 60% of the idiots round here seem to want to go. The roads aren't particularly dangerous but the clods driving other cars clearly are. Nearly all of the cars I've seen coming to the exit of ungritted side roads over the last 24 hours have had their front wheels locked up and quite a few have overshot the line. I get really angry about this - I mean are they completely oblivious that they're out of control? WTF, they seem to do it junction after junction. Yet, I also saw a car slowly approaching a juction get rammed from behind by a car that was going too fast to ever stop anyway. What is wrong with people round here? Argghhhhhhh!!!!

Roughly speaking a mean of below zero for your temperature and dewpoint - i.e. add the temperature and dew point (in degrees C) together then divide by two. So, temperature 5c, dewpoint -6c = snow, temperature 5c dewpoint -4c = no snow. VERY roughly as a rule of thumb. There are many other things to consider, but you won't go far wrong with this when nowcasting. Also see here to calculate with relative humidity: http://www.sciencebits.com/SnowProbCalc Forgot to say, nice pics - he certainly beats the theoretical slush man I could probably have made if I wasn't so lazy!

Yep GTLTW. the 12z surface analsys fax from the Met O. shows what a complicated situation this became today: http://www.netweather.tv/fax/PPVA89.png Belting down with rain now. All the footprints in the garden are making loads of little ponds full of water. Most odd. Some surfaces now snow free too, including a garden table that had about 3cm on it this morning. Temperature has bounced around from 1 to 4c all afternoon, with the dew point trailing it as you'd expect with a Southerly draw.

Here's a quick look at the situation, from the 12z actuals and predictions. First off, the wider european plot: The low pressure is positioned in Sweden with a central pressure of 981 hPa. The warm and cold fronts can clearly be seen by the kinks in the isobars. Here is the corresponding satellite image: You can clearly see the differing air masses there, well specifically you are seeing the height of the cloud tops - the IR image shows the temperature of the water vapour. If we take the satellite with precipitation overlaid: I would draw on the fronts roughly like this: Now. if we zoom in on the UK surface analysis, we can see the temperature drops behind the cold front. Widely 5-7c before the front and 2-4c immediately behind with that dropping down to 0-1c by the Borders and widely below 0c in Scotland as the cold air digs in. Lastly, we can compare the actual results with predictions from the GFS and the NAE. Here is the GFS prediction: Here is the NAE prediction: Comparing these they seem pretty close. The GFS, because of its resolution looks like more of a broad-brush of where the precipitation will be - not bad. The NAE also looks a bit broad and has slightly underestimated the speed of the front and overestimated the precipitation but otherwise not bad all round. One last image - the UKMO surface pressure chart for 12z, which looks pretty close to me (as you'd expect at this range): What will be interesting to see later on is how much back edge snow (if any) there is towards the South and whether the front largely fizzles out - it doesn't look particularly vigorous to me (yet).

You have a couple of other options, other than the ones already mentioned. You can use one of the many online websites to convert word documents to PDF files for free. Pop "word to pdf conversion" into your favourite search engine. If you run Microsoft Windows, you can also install the free Word Viewer from Microsoft. This will enable you to view and print word documents. See here: http://office.microsoft.com/en-us/word-help/use-word-viewer-to-read-and-print-documents-HA010186993.aspx?CTT=1

Just following on this discussion from the model discussion thread regarding whether the EPS outputs assume a normal distribution when calculating the standard deviation (this thread seems a more fitting location). I've been looking at the super-ensemble here: http://www.meteogroup.co.uk/uk/home/weather/services/ensemble-forecast.html It strikes me that some form of ensemble density function might give a more meaningful output than a confidence interval based on a Gaussian curve. If we look at the current (0z 10/12/2010) output for 2m temperature from the 17th to the 21st-ish it appears that there is a greater ensemble density towards the bottom of the temperature range, so some form of presentation representing that would be interesting to see. The other thought I've had is that the ensemble purturbations are done in some clever way to make the output Gaussian. I'm stuggling to see how this could be the case, though. It's also possible that the real EPS output - that digested by met agencies and passed to customers is simply not presented this way and this is merely a presentation error of the chaps who set up the charts we're looking at. The more I consider it, the more l believe that I've got the wrong end of the stick somewhere along the line! This is definitely not my area of expertise (not that I really have an area of expertise lol).