Jump to content


  • Content Count

  • Joined

  • Last visited

Everything posted by 03jtrickey

  1. From my understanding the deterministic forecast (i.e. operational run) calculates one particular outcome/forecast at high resolution based on one set of initial weather conditions (which are estimated based on observations). However, the initial weather conditions can never be perfectly known due to incomplete geographic coverage of weather observation networks and uncertainty in measurements etc. To account for this uncertainty, ensemble predictions are created by running the same model multiple times at a lower resolution using a range of slightly perturbed/adjusted (but still plausible) initial weather conditions as input data. Examining the range of different forecasts that result from the set of slightly different weather conditions gives an idea of the uncertainty in the forecast and allows the deterministic run to be considered in the context of these possible outcomes (e.g. mild outlier etc.). However the deterministic run is still valuable because its higher resolution resolves weather features better, particularly small scale features that may be critical for how a forecast develops.
  2. Very loud thunder at work in Pboro earlier. Storm right overhead.
  3. Could have been me! I've just got back from a cycle ride wih a friend from Oxford. We sheltered in Church Hanborough Church during the thunderstorm - heavy rain, regular thunder and a bit of lightning.
  4. I wonder whether it might backbuild and join up with echoes appearing just west of Winchester
  5. Just seen the Estofex forecast for tomorrow - it'll be just my luck if all hell breaks loose in Peterborough now I'm down in Oxford for the weekend! Level 2 warning out there...
  6. Very impressive and prolonged storm here in Peterborough today, on and off from about 3pm to 445pm. Several very close strikes and some gunshot type thunder. Mostly intercloud lightning. Intense rainfall and hail thrown in too.
  7. I'm currently in Northern France (Arras) and will be crossing the channel by rail tomorrow. Very much looking forward to seeing the effects of this storm and it looks like I may be well placed to witness some intense wind gusts!
  8. Went out cycling from Peterborough this evening - pretty sure I saw a few flashes of lighting in with the heavy rain.
  9. Thunder and lightning in Peterborough (just moved here with a new job)
  10. I remember this event well...I was a chorister at Gloucester Cathedral, and we had just begun our evening rehearsal when the choirmaster told us that the heavy rain that had initially been falling had started to turn to sleet, and soon afterwards there were blizzard-like conditions outside. Frustratingly, I missed seeing most of the snow falling due to being indoors singing the service, which was a shame. Later, outside the cathedral, paved surfaces were completely glazed with sheet ice where the rain had frozen, beneath 1-2 inches of fresh snow. The snow was slightly deeper at home in Cheltenham. It was probably one of the best snow events from my somewhat snow-starved childhood.
  11. Good point - I will add that to my list of things to test out. Thanks As I went to test that suggestion out I noticed a mistake in my AO/NAO/CET correlations (I'll blame it on copying formulas in Excel late at night). The updated values are as follows and affect the correlations between the CET and AO 60 days after displacement: Using these corrected values, the correlation between CET and AO increases for both displacements and split vortex events after SSW onset, and in both cases the AO is more strongly correlated with CET than the NAO after onset. Sorry about that.
  12. Just to add a caveat to the above mean CET anomaly values and emphasise the variability in the data...here are graphs for displacements and splits with the upper and lower quartiles of the CET anomalies plotted for each day. The interquartile range has a tendency to increase every time there is is a strongly negative CET anomaly. This might indicate that the mean is being forced down at these points by just a few of the events, so while there could be strongly negative values at these points, it is by no means a certainty.
  13. I have been doing some analysis today of the relationships between Central England Temperature (CET), NAO and AO in the 60 days before and after vortex displacements and splits. Here are some results: Some initial comments: Mean CET anomalies are relatively lower in days 60-30 prior to splits compared to displacements. There is a large peak in the CET anomaly prior to both displacement and split vortex events, but for displacement events this occurs 10 days earlier and is followed ~10 days later by a negative anomaly just before SSW onset (10hPa wind reversal). Both displacements and splits are, in the mean, followed by swings between positive and negative CET anomalies. The displacement event mean CET anomalies are almost the inverse of those for the split vortex events from ~days -5 to day 25 (i.e. the alternating pattern is shifted slightly earlier for displacements), but both splits and displacements have their strongest negative CET anomaly around day 40. I'm not sure what the exact onset of the previous SSW was (anyone?) but first few days of Jan +40 days would take us up to ~10th+ of February which would tie in well with the expectations of others on this forum. With regard to the AO and NAO, people have been commenting recently that the NAO in particular hasn't correlated well with the cold we have been getting. I did some simple r^2 correlation calculations for mean values of NAO against AO, CET against NAO and CET against AO in the periods before and after both types of SSW. In the case of vortex displacements, the daily AO values appear to become much less correlated with the NAO after displacement, and neither index is strongly correlated with CET temperature. In the case of vortex splits, AO also becomes less correlated with NAO after displacement, but the AO correlates quite a lot more highly with CET than the NAO does. Could this suggest that the AO is more important for post-split-SSW cold? (NB the correlations might increase if they were taken over longer than daily time periods...) Investigating the cause of these patterns would be interesting & I have yet to think about how they may relate to the precursors identified by Martius/Cohen and Jones. I will also probably do some more detailed analysis at some stage, to look at individual years, test correlations for different time periods, and look at the relative frequency of different anomaly values etc. However I would be interested to hear your thoughts (and please let me know if you see anything that is obviously erroneous) Methods: Dates of SSWs analysed are from Cohen and Jones (2012) and only events with no overlap between the onset +/-60 day periods have been included. The dates used are therefore as follows: Displacements: 30/11/1958, 16/01/1960, 08/12/1965, 02/01/1970, 29/02/1980, 04/12/1981, 24/02/1984, 23/01/1987, 16/12/2000, 02/01/2002, 07/01/2004, 24/02/2007, 22/02/2008 Splits: 30/01/1958, 23/03/1965, 24/02/1966, 08/01/1968, 02/02/1973, 22/02/1979, 02/01/1985, 22/02/1989, 18/01/2003, 21/01/2006, 24/01/2009, 09/02/2010 The Daily CET anomalies were calculated relative to the 11-point binomially smoothed climate normals for 1961-1990 from Parker (2009) "Anomalies of Central England Temperature Classified by Air Source". NAO/AO data is from the NOAA website (ftp://ftp.cpc.ncep.noaa.gov/cwlinks/norm.daily.ao.index.b500101.current.ascii and ftp://ftp.cpc.ncep.noaa.gov/cwlinks/norm.daily.nao.index.b500101.current.ascii).
  14. Are these accessible online? If so, where can I find these charts? Thank you.
  15. I can't find reference to it using a forum search, although it may well have been mentioned.
  16. I've got to write three extended essays for my Geography degree...and I can write about pretty much anything. One of my modules was on seasonal prediction so I'm going to attempt something along the lines of "Will improved knowledge of stratosphere-troposphere interactions increase the skill of seasonal forecasts for UK winters?". I've really enjoyed following the stratosphere thread this winter (many thanks to chionomaniac & GP) so I thought I'd do something that's topical and that I'm interested in! I've just come across a very recent paper published online on 13 January 2013 which may be of interest. http://www.nature.co...l/ngeo1698.html "Enhanced seasonal forecast skill following stratospheric sudden warmings Advances in seasonal forecasting have brought widespread socio-economic benefits. However, seasonal forecast skill in the extratropics is relatively modest1, prompting the seasonal forecasting community to search for additional sources of predictability2, 3. For over a decade it has been suggested that knowledge of the state of the stratosphere can act as a source of enhanced seasonal predictability; long-lived circulation anomalies in the lower stratosphere that follow stratospheric sudden warmings are associated with circulation anomalies in the troposphere that can last up to two months4, 5. Here, we show by performing retrospective ensemble model forecasts that such enhanced predictability can be realized in a dynamical seasonal forecast system with a good representation of the stratosphere. When initialized at the onset date of stratospheric sudden warmings, the model forecasts faithfully reproduce the observed mean tropospheric conditions in the months following the stratospheric sudden warmings. Compared with an equivalent set of forecasts that are not initialized during stratospheric sudden warmings, we document enhanced forecast skill for atmospheric circulation patterns, surface temperatures over northern Russia and eastern Canada and North Atlantic precipitation. We suggest that seasonal forecast systems initialized during stratospheric sudden warmings are likely to yield significantly greater forecast skill in some regions."
  17. Fantastic news - should get some decent storms in Oklahoma! Celebrating last night? I'm glad I went out, although at one point I thought I'd be too tired - I hardly slept either! Such a relief to have the results and know what's happening with the next three years of my life.
  • Create New...