Jump to content

Ginx snewx

Members
  • Posts

    103,802
  • Joined

  • Last visited

Posts posted by Ginx snewx

  1. 27 minutes ago, J.Spin said:

    The numbers are probably too borderline for a Winter Storm Warning down here in the valley (the most recent update is more like accumulations of 5-8" without yesterday afternoon's snow included), but this is often the way they do it when the more substantial accumulations are localized right along the spine – it’s too small a geographic area to focus the alert.

    Thanks enjoy Happy Holidays 

  2. 4 minutes ago, dryslot said:

    I thought that all along with this that’s what makes me hesitant until I really see where that band is going to set up

    This process has interested me since college when we 3 day tripped skiing Maine and were blasted by an unforeseen Norlun. The climo favored spot in the entire country for Norlun is your area. I bet you see near 10

  3. 10 hours ago, J.Spin said:

    At some point during the day today Alexa alerted me that we’d been put under a Winter Weather Advisory, and it looks like the BTV NWS advisories start up at 7:00 A.M. tomorrow morning and cover a good chunk of the state outside the Champlain Valley. The most recent Event Total Snow Accumulation map shows totals in the 8-12” shading for the higher elevations of the spine around here. Down in the valley our point forecast suggests something in the 5-10” range through tomorrow evening, and the map has us in that next tier of 6-8” shading, so those projections seem to correlate fairly well.

    22DEC25A.jpg.e57486171e7a58236668036d8370ede0.jpg

    22DEC25B.thumb.jpg.6fc2680c9bc679d356dbe723e6b48e00.jpg

    Advisory for 8 to 10 no WSW?

  4. 1 hour ago, WxWatcher007 said:

    I know we’re all beaten down by the last few years and not getting as much snow as we’d like during the frigid pattern, but this relaxation is exactly the time where a shortwave could produce a solid overrunning event. With the trends we’ve seen I almost think we have to produce otherwise we might be in trouble heading into January.

     

    Thats a overunning prolific snow storm on the 6z Euro.  Ongoing at day 6

    • Like 2
  5. 5 hours ago, Damage In Tolland said:

    The big winds were in with the rains . In CT anyway. All those 60-75mph gusts were where it was raining 

    You can see all the personal wx stations went down at the same time here. Right when gusts were starting to crank and it was raining hard.

  6. 13 minutes ago, CoastalWx said:

    Well now you have me wanting to look again, but as I understand it, there are no physical equations involved like our current computer models. It’s learning from ERA data sets from the past and final outcomes. 

    Really easy to find.

    https://weathersats.com/the-evolution-and-future-of-ai-in-weather-forecasting/#:~:text=AI Models Enhancing Numerical Weather,forecasts more frequently and accurately.

    Normalization and Formatting: Data, often in GRIB or NetCDF formats, is cleaned and structured into consistent grids, frequently at 0.25-degree resolution.

    Feature Engineering: Raw data is converted into actionable features, such as calculating "feels like" temperature or creating precipitation intensity categories. 

    Data Assimilation for "Real-Time" Updates

    AI models use advanced techniques to ingest the current state of the atmosphere into the model to begin a forecast: 

    AI-Driven Assimilation: Instead of traditional, slow numerical methods, AI accelerates the integration of new observations by filling in data gaps in regions where measurements are sparse.

    Hybrid Approaches: Some models, such as NeuralGCM, blend AI with traditional physical simulations to ensure the input data respects atmospheric dynamics.

    Specialized Adapters: To shift from historical training data to live data, models may use specialized neural network architectures (like U-Net) to map real-time, "messy" data into the same format used for training. 

    Key Technologies Used

    Neural Networks (CNNs/RNNs): Convolutional Neural Networks analyze spatial satellite imagery, while Recurrent Neural Networks process time-series data.

    Cloud Platforms: Models are often hosted on AWS, Google Cloud, or Microsoft Azure to manage the high volume of data.

    Transformers: Used to process long-range dependencies in atmospheric data. 

    • Like 1
    • Confused 1
    • Crap 1
  7. 4 minutes ago, WinterWolf said:

    You have “If bringing my juju back”…I think you mean “I’m” bringing my juju back…right? Correct that so the snow gods don’t punish us for poor grammar..please:lol:

    I remember when people complained about the Juju word back on Eastern. Lol

    There was no Grinch storm instead its a Christmas miracle. Festivus for the rest of us.

    • Like 1
    • Haha 1
×
×
  • Create New...