MegaMike
-
Posts
432 -
Joined
-
Last visited
Content Type
Profiles
Blogs
Forums
American Weather
Media Demo
Store
Gallery
Posts posted by MegaMike
-
-
Just to be clear, the HREF and NBM are great tools (I specifically use them for precipitation forecasting), however, we can do better.
-
3
-
-
19 minutes ago, AstronomyEnjoyer said:
Appreciate someone chiming in with an actual answer, thanks!
No problem, friend!
I run simulations fairly regularly so I thought I'd chime in
-
1
-
-
23 minutes ago, RedSky said:
Omg they have created the Frankenstein model do they know what they have done
The HREF is a Frankenstein model since it requires post-processing to obtain fields on a constant grid. It's not a standard modeling system like the HRRR, NAM, GFS, etc...
NWP requires a lot of static/time varying fields to run atmospheric simulations: vegetation type, elevation height, ice coverage, etc... By interpolating data onto a grid (or even transforming a horizontal datum), you degrade model accuracy since this information is lost.
The HREF exists simply because we created a high-resolution ensemble with what we currently have.
-
4
-
-
Consider the RRFS an ensemble consisting of multiple (9 members + 1 deterministic) high-resolution (3km) simulations on a constant grid (unlike the HREF and NBM).
In other words, it's similar to the HREF, but without any post-processing... The HREF requires post-processing since its ensemble members have various domain configurations (which is a bit taboo). Based on < https://gsl.noaa.gov/focus-areas/unified_forecast_system/rrfs >, it's set to replace the NAMnest, HRRR, HiResWindows, and HREF modeling systems.
Besides what I mentioned above, the lateral boundary conditions come from the GFS (control) and the GEFS (members). All share the same core (FV3). Currently, they're testing a bunch of different options to improve the initial conditions (data assimilation) of the RRFS. Overall, it sounds like it'll be a significant improvement over the HREF and NBM once the RRFS becomes operational (and well tested).
Note: I can't find much on the RRFS' performance... Based on what I've seen so far, It performs better in terms of reflectivity detection < https://www.spc.noaa.gov/publications/vancil/rrfs-hwt.pdf >
-
3
-
4
-
-
20 minutes ago, Dark Star said:
Trusted models? AI could be a good thing, especially when you can adjust for errors in the program say at day 3 and correct for biases. Might improve long range patterns? I would think the AI would correct for known biases, otherwise, why use AI? I would think no model would handle an anomalous event well? Or perhaps instead of AI, have human intervention adjusting at say, day 3 in a long range model to correct for a suspected error?
You don't think our modeling systems are trusted? Most high-resolution modeling systems perform well within 12 hours (especially once a disturbance is properly assimilated). Additionally, most global models perform reasonably well within 4 days at the synoptic level.
If you expect complete accuracy for moisture/precipitation fields, you (not you specifically, just in general) don't understand the limitations of NWP... Our initial conditions/data assimilation, boundary conditions, parameterizations, and truncations (dx,dy,dz increments) leads to significant error over time which aren't necessarily related to a modeling system itself. If we could perfectly initialize a modeling system, theoretically, there would be little to no errors post-initialization. You can't say the same thing about an AI model since it's likely trained on forcing variables such as temperature/moisture (at the surface and aloft) and is not simulated using governing equations and fundamental laws which the atmosphere adheres to.
I wrote, 'AI can be used to improve the accuracy of NWP output that have a known, and predictable bias...' so if we used AI to correct singular fields prior to initialization and while a modeling system is running, sure... It will likely improve the accuracy of NWP.
Bottom line; use AI to assist NWP or to correct fields with known biases. At this time, I don't trust atmospheric, AI models.
-
1
-
1
-
-
20 minutes ago, wdrag said:
I can only presume AI assisted modeling was tested for improved guidance prior to its implementation. Recommend staying with WiC and check their afternoon EPD when it comes out
I'm not sure why anyone would use AI to forecast the weather.
1) AI is a glorified analog so it won't perform well for anomalous events.
2) We don't know what's incorporated in AI models so we can't correct for known biases.
3) Why use AI when we already have trusted NWP models?
AI can be used to improve the accuracy of NWP output that have a known, and predictable bias though. For example, improving wind speed: < https://agupubs.onlinelibrary.wiley.com/doi/full/10.1029/2018JD029590 >.
-
2
-
2
-
-
1 hour ago, Lava Rock said:
10" rain.
okie dokie
The units are mm/hr, not inches/hr.
-
1
-
1
-
-
30 minutes ago, weatherwiz said:
IMO, the snowfall maps have been a detriment to the field. These things get tossed around left and right and literally all they do is create unrealistic expectations and create hype for the public. There is nothing worse then seeing 384-HR snow maps being thrown around all over social media or snow maps for a "D10 threat". There is absolutely ZERO reason.
When an event happens the event is then measured solely on how the f****** snow maps "performed". WHO GIVES A SHIT HOW THE HELL THEY PERFORMED.
Speaking of QPF though, I think models have come along way with QPF forecasting. But I think mesos have really helped with this.
I totally agree with this.
QPF has definitely improved over the past several decades. I just wish people would stop using mesos past 1pbl cycle. Thereafter, their reliability diminishes. It's the equivalent of using the GFS past day 7.
-
1
-
1
-
-
Snow continues to melt on contact in Albany. I'm not expecting much until after sunset. It's been like this since ~9am.
-
1
-
-
6 minutes ago, KoalaBeer said:
Has it really been that bad though? Honestly I haven’t paid a ton of attention but this is the 12z run from this morning valid 21z today showing 1-2 inches max for Boston at 10:1 which we knew to toss those ratios anyways…
People trash a model if snowfall doesn't produce in their back yard. If someone says 'x' model sucks, I want them to cite an evaluation (post-hoc or operational) of meteorological forcing fields related to heat and moisture both at the surface and aloft. Those fields contributes significantly to the accuracy of NWP... Not snowfall at one location (which is typically calculated from a private website, excluding a couple NWP models).
If someone were to run an evaluation of this event, you'll likely find that temperature performed well (sfc + aloft), moisture performed okay (Td or specific humidity at the sfc and aloft), and wind speed performed okay (directional and magnitudinal at the sfc and aloft - worse as you approach the surface). All NWP models struggle with QPF (for a number of reasons) so it's not solely used to determine a model's overall performance. Through my experience, it usually performs reasonable as a bulk statistic.
IMO, ensembles should always be used for QPF forecasting. Diagnostic models should be used to determine trends.
-
3
-
1
-
-
Although the rates haven't been great, we've been getting fairly large cat paws (looks like conglomerates of multiple dendrites) out west near Albany, NY. Unfortunately, not much accumulation since ~9am. Treated surfaces appear wet, not white ATM.
-
19 minutes ago, Lava Rock said:
wow on the windsor 25". They should get over 30" easily, no?
I'd be surprised if they didn't. Plenty of time to go for them to accumulate 5''.
-
The most recent public information statement from KBTV, KBOX, KGYX, and KALY. Some impressive totals already.
------------------------------------------------------------
***KBOX***Location Amount Time/Date Provider
...Connecticut...
...Hartford County...
7 WNW Granby 11.3 in 0815 AM 03/14
Windsor Locks 4.4 in 0814 AM 03/14 ASOS...Massachusetts...
...Franklin County...
Hawley 24.0 in 0802 AM 03/14 CO-OP Observer
Greenfield 5.8 in 0725 AM 03/14 CO-OP Observer
1 W Orange AP 3.0 in 0858 AM 03/14...Hampden County...
Blandford 16.0 in 0845 AM 03/14
Granville 14.5 in 0845 AM 03/14
2 NW Montgomery 8.0 in 0705 AM 03/14
Southwick 6.0 in 0845 AM 03/14 Emergency Mngr
2 SW Chicopee 4.5 in 0711 AM 03/14...Hampshire County...
1 ESE Plainfield 24.0 in 0859 AM 03/14
1 ESE Plainfield 20.0 in 0741 AM 03/14...Middlesex County...
Ashby 16.5 in 0905 AM 03/14
1 WNW Pepperell 4.0 in 0704 AM 03/14
2 SSE Dunstable 3.5 in 0832 AM 03/14...Worcester County...
Templeton 14.0 in 0910 AM 03/14
2 NNW Sterling 13.3 in 0901 AM 03/14
1 WNW Holden 13.0 in 0838 AM 03/14
Ashburnham 12.7 in 0745 AM 03/14 CO-OP Observer
2 NNW Sterling 11.5 in 0729 AM 03/14
1 NE Lunenburg 10.1 in 0821 AM 03/14
Lunenburg 10.0 in 0905 AM 03/14
1 N Rutland 8.0 in 0717 AM 03/14
Worcester AP 4.5 in 0815 AM 03/14 ASOS
1 NNW Lancaster 4.5 in 0755 AM 03/14
------------------------------------------------------------
***KGYX***Location Amount Time/Date Provider
...Maine...
...Franklin County...
Farmington 4.8 NNW T in 0900 AM 03/13 COCORAHS...New Hampshire...
...Belknap County...
2 WSW Meredith 2.0 in 0213 AM 03/14 Trained Spotter...Carroll County...
1 N Madison 5.5 in 0619 AM 03/14 Trained Spotter...Cheshire County...
Walpole 5.5 in 0618 AM 03/14 Public
2 SW Keene 2.0 in 0707 AM 03/14 Trained Spotter...Coos County...
4 NNW Whitefield 2.6 in 0626 AM 03/14 Trained Spotter...Hillsborough County...
2 WNW Windsor 11.0 in 0641 AM 03/14 Trained Spotter
2 N Francestown 10.0 in 0728 AM 03/14 Trained Spotter
2 N New Boston 8.0 in 0734 AM 03/14 Trained Spotter
2 N Brookline 6.2 in 0711 AM 03/14 Public
3 N Hillsborough 6.0 in 0107 AM 03/14 Trained Spotter
2 S New Boston 5.6 in 0729 AM 03/14 Trained Spotter...Merrimack County...
3 E Tilton 2.5 in 0745 AM 03/14 Public
Concord Municipal Airport 0.7 in 0717 AM 03/14 ASOS...Rockingham County...
3 ENE Derry 5.0 in 0720 AM 03/14 Trained Spotter
Nottingham 1.2 S T in 0830 AM 03/13 COCORAHS...Strafford County...
2 ENE Northwood 2.0 in 0616 AM 03/14 Trained Spotter
------------------------------------------------------------
***KALY***LOCATION TOTAL TIME/DATE COMMENTS
SNOWFALL MEASURED
(inches)CONNECTICUT
...Litchfield County...
Lakeville 4.0 730 AM 3/14 Amateur RadioMASSACHUSETTS
...Berkshire County...
Windsor 25.0 857 AM 3/14 Twitter
3 S Sandisfield 15.6 902 AM 3/14 Trained Spotter
2 ENE Pittsfield 11.4 800 AM 3/14 Amateur Radio
Cheshire 7.5 800 AM 3/14 Amateur Radio
Lenox 7.0 630 AM 3/14 Social Media
Adams 6.5 800 AM 3/14 Amateur Radio
Lenoxdale 6.5 800 AM 3/14 Amateur Radio
Williamstown 3.5 728 AM 3/14 TwitterNEW YORK
...Albany County...
CLARKSVILLE 10.0 800 AM 3/14 Elevation 1290 ft
1 NE Guilderland 8.0 700 AM 3/14 Amateur Radio
Albany Intl AP 6.0 800 AM 3/14 ASOS
Albany 5.8 905 AM 3/14 Meteorologist
Boght Corners 5.8 755 AM 3/14 NWS Employee
Colonie 5.5 735 AM 3/14 Trained Spotter
New Scotland 5.0 640 AM 3/14 Amateur Radio...Columbia County...
Austerlitz 10.0 842 AM 3/14 WeatherNet6
Taghkanic 7.0 831 AM 3/14 WeatherNet6
Ancramdale 4.0 649 AM 3/14 WeatherNet6
Germantown 4.0 812 AM 3/14 WeatherNet6...Dutchess County...
Rhinebeck 6.0 733 AM 3/14 Facebook
Red Hook 4.0 703 AM 3/14 WeatherNet6
Poughquag 4.0 745 AM 3/14 Public...Greene County...
5 SW Palenville 18.0 717 AM 3/14 Facebook
3 SSW Tannersville 16.0 630 AM 3/14 Facebook
Greenville Center 11.0 901 AM 3/14 WeatherNet6
West Kill 10.0 822 AM 3/14 WeatherNet6
Cairo 6.0 632 AM 3/14 WeatherNet6
Halcott Center 6.0 826 AM 3/14 WeatherNet6
Catskill 4.0 725 AM 3/14 WeatherNet6...Hamilton County...
Wells 10.0 700 AM 3/14 Amateur Radio
7 ENE Piseco 7.6 700 AM 3/14 Amateur Radio
Piseco 7.0 816 AM 3/14 WeatherNet6
Long Lake 3.5 700 AM 3/14 Amateur Radio
Inlet 3.0 705 AM 3/14 Trained Spotter...Herkimer County...
Ohio 6.4 855 AM 3/14 Amateur Radio
3 E Middleville 5.0 703 AM 3/14 Trained Spotter
Dolgeville 4.5 835 AM 3/14 Amateur Radio
Herkimer 4.0 850 AM 3/14 Amateur Radio
Ilion 4.0 700 AM 3/14 Amateur Radio...Montgomery County...
Hessville 8.3 857 AM 3/14 WeatherNet6
Glen 8.0 801 AM 3/14 WeatherNet6
Amsterdam 6.8 759 AM 3/14 WeatherNet6
Canajoharie 6.0 830 AM 3/14 Facebook
3 NNE Sprakers 3.5 740 AM 3/14 Amateur Radio
Fonda 3.1 723 AM 3/14 WeatherNet6...Rensselaer County...
1 NW Wyantskill 12.0 815 AM 3/14 Amateur Radio
Brunswick 11.5 855 AM 3/14 Amateur Radio
1 N Sycaway 9.1 800 AM 3/14 0.96 inches liquid
1 N North Greenbush 8.5 726 AM 3/14 Trained Spotter
Petersburg 8.0 826 AM 3/14 WeatherNet6
Berlin 7.5 703 AM 3/14 Facebook
East Greenbush 7.5 630 AM 3/14 Trained Spotter
Speigletown 4.0 721 AM 3/14 WeatherNet6
Hoosick Falls 3.2 806 AM 3/14 WeatherNet6...Saratoga County...
Providence 12.5 813 AM 3/14 Elevation 1300 ft
Clifton Park 6.5 710 AM 3/14 WeatherNet6
Galway 6.2 754 AM 3/14 WeatherNet6
Ballston Spa 4.5 632 AM 3/14 WeatherNet6
Ballston Lake 4.5 728 AM 3/14 WeatherNet6
Saratoga Springs 4.0 813 AM 3/14 WeatherNet6...Schenectady County...
Delanson 14.0 919 AM 3/14 WeatherNet6
Schenectady 6.0 650 AM 3/14 WeatherNet6
1 SSW Aqueduct 6.0 720 AM 3/14 NWS Employee
1 NNE Hawthorne Hill 6.0 830 AM 3/14 NWS Employee
Niskayuna 5.5 820 AM 3/14 Amateur Radio
Duanesburg 5.0 728 AM 3/14 WeatherNet6
Pattersonville 5.0 732 AM 3/14 Meteorologist...Schoharie County...
Jefferson 13.0 800 AM 3/14 Elevation 1900 ft
Middleburgh 8.2 630 AM 3/14 WeatherNet6
Richmondville 8.0 751 AM 3/14 WeatherNet6...Ulster County...
Plattekill 6.1 730 AM 3/14 Trained Spotter
Rosendale 3.5 815 AM 3/14 Public...Warren County...
3 WSW North Creek 10.5 804 AM 3/14 Twitter
Lake Luzerne 6.2 636 AM 3/14 WeatherNet6
Queensbury 5.5 841 AM 3/14 WeatherNet6...Washington County...
Hartford 11.5 910 AM 3/14 Trained Spotter
Hebron 6.0 813 AM 3/14 WeatherNet6
Granville 4.0 713 AM 3/14 WeatherNet6
Whitehall 1.0 731 AM 3/14 WeatherNet6VERMONT
...Bennington County...
Pownal 11.0 741 AM 3/14 Facebook
Landgrove 9.0 710 AM 3/14 WeatherNet6
3 NNE Arlington 5.5 732 AM 3/14 Trained Spotter
West Arlington 2.0 628 AM 3/14 WeatherNet6...Windham County...
Windham 21.4 900 AM 3/14 CoCoRaHS
Brattleboro 8.0 730 AM 3/14 Facebook
**********************24 HOUR SNOWFALL**********************LOCATION 24 HOUR TIME/DATE COMMENTS
SNOWFALL MEASURED
(inches)CONNECTICUT
...Litchfield County...
2 WNW Warren 4.0 700 AM 3/14 CoCoRaHS
3 WNW New Milford 1.0 700 AM 3/14 CoCoRaHS
2 SSW Woodbury Cente 0.5 700 AM 3/14 CoCoRaHS
5 SSW New Milford 0.5 730 AM 3/14 CoCoRaHSMASSACHUSETTS
...Berkshire County...
1 NNW Cheshire 7.2 700 AM 3/14 CoCoRaHS
2 NW Sheffield 2.8 700 AM 3/14 CoCoRaHSNEW YORK
...Albany County...
3 SSW Altamont 14.8 808 AM 3/14 CoCoRaHS
4 NW Altamont 10.5 700 AM 3/14 CoCoRaHS
3 SW Glenmont 5.5 700 AM 3/14 CoCoRaHS
1 SW Albany 5.5 700 AM 3/14 CoCoRaHS
NWS Albany 4.1 800 AM 3/14 CoCoRaHS...Columbia County...
1 SW Chatham 8.5 700 AM 3/14 CoCoRaHS
3 N Valatie 7.2 700 AM 3/14 CoCoRaHS
N Ancramdale 6.5 700 AM 3/14 CoCoRaHS
1 S Valatie 5.5 700 AM 3/14 CoCoRaHS
1 E Germantown 3.7 700 AM 3/14 CoCoRaHS...Dutchess County...
3 NE Lagrangeville 5.8 700 AM 3/14 CoCoRaHS
3 SE Millbrook 5.6 700 AM 3/14 CoCoRaHS
1 NNE Red Hook 4.0 700 AM 3/14 CoCoRaHS
3 N Rhinebeck 3.6 700 AM 3/14 CoCoRaHS...Fulton County...
Gloversville 7nw 8.1 700 AM 3/14 Co-Op Observer
1 NE Northville 7.0 645 AM 3/14 CoCoRaHS
1 N Northville 5.2 635 AM 3/14 CoCoRaHS...Greene County...
1 E Greenville 11.0 700 AM 3/14 CoCoRaHS
3 E Freehold 9.5 700 AM 3/14 CoCoRaHS
2 N Lexington 9.0 700 AM 3/14 CoCoRaHS
WNW Athens 2.6 730 AM 3/14 CoCoRaHS...Hamilton County...
Indian Lake 9.3 750 AM 3/14 Co-Op Observer
10 SSW Speculator 8.5 700 AM 3/14 CoCoRaHS
2 WSW Lake Pleasant 7.6 655 AM 3/14 CoCoRaHS
1 N Long Lake 4.4 700 AM 3/14 CoCoRaHS...Herkimer County...
3 NE Schuyler 4.5 745 AM 3/14 CoCoRaHS...Montgomery County...
2 WSW Hessville 8.3 800 AM 3/14 CoCoRaHS
1 NNW Amsterdam 6.0 704 AM 3/14 CoCoRaHS
SE Fort Plain 2.3 700 AM 3/14 CoCoRaHS...Rensselaer County...
7 NE Troy 10.4 700 AM 3/14 CoCoRaHS
1 E Troy 10.0 700 AM 3/14 CoCoRaHS
2 NNE Troy 9.1 800 AM 3/14 CoCoRaHS
3 SW Stephentown 5.0 700 AM 3/14 CoCoRaHS...Saratoga County...
4 WNW Rock City Fall 12.5 740 AM 3/14 CoCoRaHS
3 ESE Ballston Spa 8.0 845 AM 3/14 CoCoRaHS
2 W Greenfield Cente 7.0 800 AM 3/14 CoCoRaHS...Schenectady County...
1 NE Niskayuna 5.9 700 AM 3/14 CoCoRaHS
3 SSE Niskayuna 5.0 700 AM 3/14 CoCoRaHS
2 ENE Schenectady 4.5 700 AM 3/14 CoCoRaHS
1 NW Scotia 3.5 700 AM 3/14 CoCoRaHS...Ulster County...
4 E Wallkill 4.8 700 AM 3/14 CoCoRaHS
4 N Kerhonkson 4.3 700 AM 3/14 CoCoRaHS
SW Highland 4.1 700 AM 3/14 CoCoRaHS
2 S Hurley 3.9 700 AM 3/14 CoCoRaHS
2 WNW Rosendale Vill 3.8 815 AM 3/14 CoCoRaHS
1 ENE Woodstock 1.5 700 AM 3/14 CoCoRaHS
5 NW Saugerties 0.1 900 AM 3/14 CoCoRaHS...Warren County...
5 ESE Lake George 6.0 700 AM 3/14 CoCoRaHS
1 N Lake George 3.9 700 AM 3/14 CoCoRaHSVERMONT
...Bennington County...
Peru 11.6 700 AM 3/14 Co-Op Observer
3 ENE Manchester 10.5 730 AM 3/14 CoCoRaHS
W South Shaftsbury 2.5 720 AM 3/14 CoCoRaHS...Windham County...
1 S Brattleboro 6.4 700 AM 3/14 CoCoRaHS
1 NNE Rockingham 3.9 700 AM 3/14 CoCoRaHS
------------------------------------------------------------
***KBTV***NEW YORK
...Clinton County...
4 E Dannemora 5.5 907 AM 3/14 1,200' elevation
Chazy 1.2 654 AM 3/14 Mesonet...Essex County...
1 E Newcomb 6.7 650 AM 3/14 Mesonet
2 W Wilmington 3.9 650 AM 3/14 Mesonet
3 S Crown Point 2.8 654 AM 3/14 Mesonet
4 NNE North Hudson 2.0 651 AM 3/14 Mesonet
Moriah 2.0 710 AM 3/14 Public...Franklin County...
Lake Clear 6.0 815 AM 3/14 Public
Saranac Lake 4.8 650 AM 3/14 Public
1 S Gabriels 1.2 648 AM 3/14 Mesonet...St. Lawrence County...
9 NE Hammond 1.6 647 AM 3/14 MesonetVERMONT
...Caledonia County...
Danville 7.0 828 AM 3/14 Public
Kirby 7.0 800 AM 3/14 Public
West Danville 7.0 900 AM 3/14 1,600' elevation
Walden 7.0 813 AM 3/14 1,640' elevation
St. Johnsbury 6.0 901 AM 3/14 Public
West Burke 4.0 738 AM 3/14 Public...Chittenden County...
2 NW Westford 2.9 602 AM 3/14 NWS Employee
Underhill 0.8 845 AM 3/14 1,100' elevation...Essex County...
4 WNW Maidstone 7.5 841 AM 3/14 Public...Franklin County...
St. Albans 2.0 815 AM 3/14 Public...Lamoille County...
Morrisville 3.0 733 AM 3/14 Public...Orange County...
Randolph Center 11.5 800 AM 3/14 1,500' elevation
Chelsea 9.0 730 AM 3/14 1,400' elevation
East Randolph 8.0 914 AM 3/14 1300' elevation
West Newbury 5.8 630 AM 3/14 Public
Williamstown 5.0 700 AM 3/14 Public...Orleans County...
Albany 3.8 652 AM 3/14 Public...Rutland County...
Shrewsbury 11.0 659 AM 3/14 Public
Mt. Holly 10.0 630 AM 3/14 Public...Washington County...
Marshfield 8.0 918 AM 3/14 1,600' elevation
Fayston 7.5 735 AM 3/14 1,450' elevation
Cabot 4.0 715 AM 3/14 1900 ft elev
Barre 4.0 752 AM 3/14 Public
Montpelier 4.0 900 AM 3/14 Public
2 NNE Waterbury Cent 3.0 543 AM 3/14 Trained Spotter...Windsor County...
3 S Ludlow 16.0 900 AM 3/14 1.32 inch liquid
Andover 14.0 715 AM 3/14 2000 ft elev
1 SE West Norwich 5.1 548 AM 3/14 Public
Hartland 2.0 927 AM 3/14 Public-
1
-
-
1 hour ago, jbenedet said:
Thanks for the balanced convo but they do suck. I've been burned before.
I never said they were great as operational forecast models. I seldom use them, if at all. At this range, I'd prefer the NBM, HREF, and HRRR over almost anything else.
My point was, if you're going to disregard the WRF-ARW/2 because they're WRF models, you may as well disregard the WRF-NMM model (NAM), as well.
Regardless, WRF-ARW is fantastic for post-hoc analysis using reanalysis data.
-
1
-
-
Let's be nice to the big, beautiful WRF models (which do have a cold bias).
The NAM is WRF w/a Non-hydrostatic Mesoscale Model dynamic core (WRF-NMM).
https://www.ncei.noaa.gov/products/weather-climate-models/north-american-mesoscale
-
1
-
-
1 minute ago, MuddyWx said:
and the weenie deflator map
I've mentioned this in the past, but 1) snow depth fields from NWP include gauge losses due to melting/compaction/sublimation/etc... and 2) the ratio algorithms are underestimated (from the land surface models I'm aware of, they only incorporate near surface temperature and wind speed). They basically assume immediate compaction.
I agreed with another met on here that this product is helpful along the boundary b/n rain and snow. Further inland, it's best to use SWE (use 10:1 ratios if you must).
-
3
-
1
-
-
32 minutes ago, weatherwiz said:
Guidance has been dreadful with temperatures pretty much throughout the country for weeks. I'm sure a large part of it has to do with the pattern but I mean they've been terrible. MOS, NBM, etc. I don't know much (well anything) about model physics, but I wonder how much of an impact something like this has on the models.
In my quick opinion, the biggest differences spawn from:
1) Data assimilation methodology - better ic/bcs lead to more accurate forecasts. You really need to initialize the atmosphere as realistically as possible. I'm not only referring to met. forcing data (heat and moisture fields at the surface, beneath the surface, and aloft), but static data as well... Such as elevation, land cover, snow depth, ice cover, vegetation type, urban canopy, etc...
This stage of initialization is different pending the agency.
2) Horizontal and vertical configuration of a modeling system (resolution). Finer resolution leads to better results short-term, but worsen, spatially, wrt time. Courser models don't perform well initially, but spatial error vs. time isn't as significant. Unfortunately, courser models struggle with fine-scale phenomenon such as moisture flux which significantly contributes to the development/decay of a disturbance.
3) Microphysics, cumulus, etc... parameterization options followed by other physics/dynamics options. Pending the resolution of a modeling system, these options can lead to sizable differences.
There's an infinite amount of options an agency can create a modeling system with. All of these differences leads to incremental error (beneath the surface, at the surface, aloft, and between grid-cells in all dimensions - dxdydz) that develops into large scale discrepancies vs. time.
-
1
-
1
-
-
If the JMA is correct for the NE USA, it's for the wrong reason. It's hard to trust a global modeling system that has an extremely course resolution... Based on TT's, the global JMA (the JMA has regional/local nests too!) looks like it has a resolution of ~1 degree, but according to < https://www.wis-jma.go.jp/cms/gsm/#:~:text=The Japan Meteorological Agency's Global,layers) on an operational basis. >, the JMA has a resolution of 0.1875 degrees (20km). If the resolution is 20km, and for whatever reason, Japan truncates their output to ~1 degree, I'd (try to) trust it a bit more. Until I know for sure, in my opinion, the JMA is as useful as the NOGAPS.
It's better to trust a modeling system that's developed by a country on the same continent as you. Domains are set and configured to emphasize a particular region. US weather isn't so crucial to Japan (short-term) so they can afford to truncate it in order to optimize simulation runtime or allocate additional resources somewhere else (say, add a nested domain in Asia).
-
4
-
1
-
-
48 minutes ago, 78Blizzard said:
I don't get that. If the temps and qpf are means of all members, then so is the snow a mean of all members. If the mean temps for an area are low to mid 20's, and upper level means are well below freezing for the same area, shouldn't the mean snow for that area correspond with those indicators?
Otherwise, it would appear that the cutting members are skewing the snow totals. The placement of the low on the GEFS is also a mean, and the resulting metrics should be based on that.
I think the problem with most of the ensemble models is that they take a mean of the snow totals of all members (including the cutters showing little or zero) and plug that into their maps, when in reality the maps should be reflecting the snow resulting from the placing of the mean slp and resultant lower and upper level means. The storm vista maps appear to take the latter approach with the GEFS.
There's no way they have the ability to code that.
I write/say this all the time, but they should mention exactly what they used to produce their plots. Snowfall typically doesn't come straight from a model so post-processing is usually necessary. Post-processing is solely reliant on the vendor/user so don't be surprised if you see discrepancies from vendor-to-vendor.
Some questions that come to mind:
1) Which members did they use to calculate mean snowfall.
2) Which fields did they use to post-process snowfall using 10:1 ratios.
Without looking at the GEFS' output, there are a couple ways they can get snowfall using the 10:1 algorithm ->
2.a) if ptype == snow, SF = LWE*10.
2.b) Use SWE output from an ensemble member then simply multiply it by 10.
I'm sure there are other options too.
2.a wouldn't make sense to me. Determining ptype for global ensemble members would be a waste of resources. My guess is that they use different ensemble members (or SWE fields) to calculate their mean. Unfortunately, you'll never know unless you see their code.
-
2
-
-
3 minutes ago, tavwtby said:
Just east of Aly down to ginx jack?
It's been a splendid, little storm just north of Albany. About 4.5'' where I am (Latham) with fluctuating rates (currently). I don't expect more than additional inch here, but I'd agree that east of Albany will do best. Probably along the NY/MA border just east of the Taconic's if I had to guess.
-
That storm in Dec. 05' was the most impressive weather related event I've experienced (albeit, brief). I couldn't see ~100' in front of me for about an hour in Attleboro (with plenty of thundersnow). If I remember correctly, someone observed ~6-8'' of snow in 1 hour.
I was a paper boy at the time and had to deliver papers that afternoon - My mother drove me along Read St. and she ended up knocking down someone's mailbox because of "the whiteout"
I ran an analysis simulation (WRF) of that event using the ECMWF20c reanalysis dataset for ic/bcs. I also created a radar animation for that event too superimposed with observations.
-
2
-
2
-
-
59 minutes ago, The 4 Seasons said:
Love it! It's missing a few events though like Nov 7th 2012 and Apr 2018, theres also a few out of chronological order. But i really enjoyed it, looks like a lot of work.
Thanks, dude!
There's an archive (If I remember correctly: https://www.ncei.noaa.gov/data/snowstorm-database/archive/) that contains snow event observations. In a nutshell, I wrote script to read each shapefile, populate the grid with '0' observations, create an interpolation, plot the highest snowfall observation within multiple grid boxes, then save the image. This was done iteratively so it wasn't too time consuming.
I didn't notice some were out of order - can't win em' all, I guess.
From here on out, I'm leaving the snowfall interpolations to you lol
-
1 hour ago, BeauDodson said:
Hoping I can come in from the south tomorrow afternoon.
Sounds like they will close it.
Unsure another way in.
I think I might be 'out' on this one because of that decision... which I support.
I'm only 4 hours away so it'll hurt a little bit.
Some thoughts related to the discussion that I've seen in this thread:
1) I wouldn't trust the HRDPS on this one or for anything in the US - it's just a bit too close to the boundaries for my liking. That can cause some weird output along the outer edges of the modeling system.
2) I wouldn't trust the NAM12km on this either (or anything >3km). I mentioned this before, but you really need higher resolution models to capture moisture flux for LES. It's basically a topographically induced, semi-stationary thunderstorm. Would you trust QPF from courser models for any convective event? I hope not. Regardless, QPF is still a challenge for NWP for a number of reasons.
3) Use high-res ensembles for this one. I'd recommend using the High Resolution Ensemble Forecast system (HREF) for QPF.
4) Once the band forms, I'd lean towards the HRRR (only for init. +~6 hours) since I trust its data assimilation.
5) I expect crazy totals from this.
I hope you guys live stream this!
-
6
-
-
Oh, I know
. I lived in the Providence area which kinda' sucked too.
I'm in Latham if I had to be exact. Just by the airport. Any recommendations on what to do? I'm still exploring the area.
Originally, I'm from MA, but I've lived in RI, CT, VT, NY, and NC over the past 10 years. It's been a crazy ride!
-
1
-
Monitoring the 29th/30th for significant impact coastal redevelopment - confidence only medium for now but is trending favorably.
in New England
Posted
I published a paper for the EPA regarding an air quality modeling system/data truncation, but I haven't collaborated with NCEP on atmospheric models/simulations.
I do have experience with a bunch of different modeling systems though (in order): ADCIRC (storm surge), SWAN (wave height), WRF-ARW (atmospheric and air quality), UPP (atmospheric post-processor), CMAQ (air quality), ISAM (air quality partitioning), RAMS ( atmospheric), ICLAMS (legacy atmospheric model).