Ian Posted December 6, 2013 Share Posted December 6, 2013 Probably, I thought I got around .2 before changeover when I looked at the temps off the euro. Jason has my 1" probability at 50% which I think this early is fine. Ryan's really good at the model stuff but he's a tropical guy. I'm not 100% sure he knows the snow stuff.. if he tweaked the algorithms at all. I'm still unsure if that comes packaged when it goes out or how companies change it. But by and large their snow maps seem to keep bringing snow after it should stop. They are pretty though. Link to comment Share on other sites More sharing options...
clskinsfan Posted December 6, 2013 Share Posted December 6, 2013 Wes' piece http://www.washingtonpost.com/blogs/capital-weather-gang/wp/2013/12/06/major-winter-storm-sunday-briefly-snow-then-ice-significant-ice-storm-western-areas-possible/ Nice piece. Thanks! Link to comment Share on other sites More sharing options...
mappy Posted December 6, 2013 Share Posted December 6, 2013 more conservative. Seems better in this instance I think thats Ellinwoods forecast? Link to comment Share on other sites More sharing options...
lsukev Posted December 6, 2013 Share Posted December 6, 2013 This is overdone but here: ecmwf_snow_24_washdc_13.pn This map is weird because we lose the 850's around 57hrs but the snow keeps accumulating until 72hrs. At 60hrs, if you take those accumulations and lower the ratio to 5 or 7-1, it's not that crazy. 2-3 for IAD, 3-4 for Reagan? Link to comment Share on other sites More sharing options...
chris87 Posted December 6, 2013 Share Posted December 6, 2013 Ryan's really good at the model stuff but he's a tropical guy. I'm not 100% sure he knows the snow stuff.. if he tweaked the algorithms at all. I'm still unsure if that comes packaged when it goes out or how companies change it. But by and large their snow maps seem to keep bringing snow after it should stop. They are pretty though. What I assume he's doing is just taking the categorical yes/no for snow and then summing all the liquid equivalent when a given model says it should be snow (as a function of their thermal profiles) and then applying the standard 10:1 ratio -- that's the path of least resistance with trying to "estimate" liquid equivalent that falls as snow in model space (not necessarily what accumulates -- at that point their is no interaction with the land surface scheme of the model). In other words -- it will always be the perfect case scenario (assuming a perfect model) for accumulation but generally always an over-prediction and at times a gross over-prediction. Another thing to consider is that the ECMWF output may be poorly classifying precipitation type -- so that might not be his fault but if he was doing a more sophisticated method to retrieve snow he wouldn't care what the raw output says. Link to comment Share on other sites More sharing options...
usedtobe Posted December 6, 2013 Share Posted December 6, 2013 What I assume he's doing is just taking the categorical yes/no for snow and then summing all the liquid equivalent when a given model says it should be snow (as a function of their thermal profiles) and then applying the standard 10:1 ratio -- that's the path of least resistance with trying to "estimate" liquid equivalent that falls as snow in model space (not necessarily what accumulates -- at that point their is no interaction with the land surface scheme of the model). In other words -- it will always be the perfect case scenario (assuming a perfect model) for accumulation but generally always an over-prediction and at times a gross over-prediction. Another thing to consider is that the ECMWF output may be poorly classifying precipitation type -- so that might not be his fault but if he was doing a more sophisticated method to retrieve snow he wouldn't care what the raw output says. But those amounts don't seem to fit the 850 temps so I think something else is going one. I wonder if he really looks at the thermal profiles to look for a warm layer. Link to comment Share on other sites More sharing options...
lsukev Posted December 6, 2013 Share Posted December 6, 2013 Ryan's really good at the model stuff but he's a tropical guy. I'm not 100% sure he knows the snow stuff.. if he tweaked the algorithms at all. I'm still unsure if that comes packaged when it goes out or how companies change it. But by and large their snow maps seem to keep bringing snow after it should stop. They are pretty though. Just re-read this and you made my point. Thanks, Ian. Link to comment Share on other sites More sharing options...
Bob Chill Posted December 6, 2013 Share Posted December 6, 2013 As long as snow maps are used as just a general guide to the potential event but not used as a forecast then I think they are fine. The problem is their prolific use and they look much like tv forecast map. They fly all over the internet and the vast majority of people use them the wrong way. It's not that hard to look at precip and temp panels and drill down a more accurate picture of what may be snow/sleet/rain. Many of us weenies can do it. I don't have a problem with the maps themselves but the misuse is a misinformation highway. Especially on the biggest weather board in the universe...Facebook... Link to comment Share on other sites More sharing options...
chris87 Posted December 6, 2013 Share Posted December 6, 2013 But those amounts don't seem to fit the 850 temps so I think something else is going one. I wonder if he really looks at the thermal profiles to look for a warm layer. Yes, exactly, that's what I was trying to convey -- I don't believe he's looking at any thermals -- if the raw ECMWF output says yes on snow (which I can't confirm because I don't work or have access to any raw ECMWF grib files) he sums up the QPF for that period and applies a standard ratio. I think it's pretty obvious he's not looking for warm layers to try to either correct a poor classification in the raw ECMWF or if he's doing his own estimation of p-type. Link to comment Share on other sites More sharing options...
snow. Posted December 6, 2013 Share Posted December 6, 2013 Probably, I thought I got around .2 before changeover when I looked at the temps off the euro. Jason has my 1" probability at 50% which I think this early is fine. 12z GFS, DCA goes to sleet around ~19z...so get around .20" as snow....I am not sure the euro is really that different..probably 0.2" before the flip...I think 1-2" then sleet is a good call for DCA right now, Link to comment Share on other sites More sharing options...
BTRWx Posted December 6, 2013 Share Posted December 6, 2013 As long as snow maps are used as just a general guide to the potential event but not used as a forecast then I think they are fine. The problem is their prolific use and they look much like tv forecast map. They fly all over the internet and the vast majority of people use them the wrong way. It's not that hard to look at precip and temp panels and drill down a more accurate picture of what may be snow/sleet/rain. Many of us weenies can do it. I don't have a problem with the maps themselves but the misuse is a misinformation highway. Especially on the biggest weather board in the universe...Facebook... good post, I fully agree Link to comment Share on other sites More sharing options...
snow. Posted December 6, 2013 Share Posted December 6, 2013 What I assume he's doing is just taking the categorical yes/no for snow and then summing all the liquid equivalent when a given model says it should be snow (as a function of their thermal profiles) and then applying the standard 10:1 ratio -- that's the path of least resistance with trying to "estimate" liquid equivalent that falls as snow in model space (not necessarily what accumulates -- at that point their is no interaction with the land surface scheme of the model). In other words -- it will always be the perfect case scenario (assuming a perfect model) for accumulation but generally always an over-prediction and at times a gross over-prediction. Another thing to consider is that the ECMWF output may be poorly classifying precipitation type -- so that might not be his fault but if he was doing a more sophisticated method to retrieve snow he wouldn't care what the raw output says. as you probably agree It is pretty apparent that DCA at least flips shortly after 18z even on the euro...most of the thump between 18z and 00z would not be snow....on the 12z GFS, at 21z, DCA has 850mb temp of -1.2 and a 800mb temp of +0.8..and the euro orientation is similar...so these maps are kind of for entertainment only...I don't think they have any value in issuing a forecast...like not even minor value Link to comment Share on other sites More sharing options...
Ji Posted December 6, 2013 Author Share Posted December 6, 2013 Ryan has big arms Link to comment Share on other sites More sharing options...
snow. Posted December 6, 2013 Share Posted December 6, 2013 this is JYO at 21z...any big snow totals are kind of dangerous, Link to comment Share on other sites More sharing options...
Stormpc Posted December 6, 2013 Share Posted December 6, 2013 Watches hoisted for everyone it looks like Link to comment Share on other sites More sharing options...
Ian Posted December 6, 2013 Share Posted December 6, 2013 What I assume he's doing is just taking the categorical yes/no for snow and then summing all the liquid equivalent when a given model says it should be snow (as a function of their thermal profiles) and then applying the standard 10:1 ratio -- that's the path of least resistance with trying to "estimate" liquid equivalent that falls as snow in model space (not necessarily what accumulates -- at that point their is no interaction with the land surface scheme of the model). In other words -- it will always be the perfect case scenario (assuming a perfect model) for accumulation but generally always an over-prediction and at times a gross over-prediction. Another thing to consider is that the ECMWF output may be poorly classifying precipitation type -- so that might not be his fault but if he was doing a more sophisticated method to retrieve snow he wouldn't care what the raw output says. Interesting, thanks. I've tried to figure out where the issue is but haven't been able to yet. It does seem that sometimes the snow output tracks with the 32 surface line at wxbell but maybe I'm imagining it. Link to comment Share on other sites More sharing options...
chris87 Posted December 6, 2013 Share Posted December 6, 2013 as you probably agree It is pretty apparent that DCA at least flips shortly after 18z even on the euro...most of the thump between 18z and 00z would not be snow....on the 12z GFS, at 21z, DCA has 850mb temp of -1.2 and a 800mb temp of +0.8..and the euro orientation is similar...so these maps are kind of for entertainment only...I don't think they have any value in issuing a forecast...like not even minor value Yeah -- I've only looked at the GFS -- but i'd agree. Another thing to consider about snow maps (and could be why the WxBell maps may be so poor) -- I assume he's working with 6-hr ECMWF output -- so if he is just assuming that all QPF that falls in a 6-hour window is snow based on the categorical field in the raw output -- I believe that is an average over the 6-hr period rather than an instantaneous so if 3.01 hours of the 6 hour window support snow but 75% of the QPF falls in the second half of the 6-hour window -- the categorical snow would be yes while most of the QPF actually fell as rain. so if you're assigning where that QPF goes based on the categorical field you just added a bunch of extra snow that wasn't support by the thermal profiles -- i'm sure if he had hourly ECMWF output his snow maps would be greatly improved. It's the only explanation I have -- I can't believe the ECMWF p-type algorithm would be so poor it's much more likely he doesn't have the adequate temporal resolution to break down 6-hr QPF totals into the correct p-type bins. Link to comment Share on other sites More sharing options...
Ian Posted December 6, 2013 Share Posted December 6, 2013 Watches hoisted for everyone it looks like URGENT - WINTER WEATHER MESSAGE NATIONAL WEATHER SERVICE BALTIMORE MD/WASHINGTON DC 233 PM EST FRI DEC 6 2013 DCZ001-MDZ004>007-009>011-013-014-016-VAZ039-040-042-050>057-501- 502-070345- /O.EXB.KLWX.WS.A.0008.131208T1400Z-131209T1200Z/ DISTRICT OF COLUMBIA-FREDERICK MD-CARROLL-NORTHERN BALTIMORE- HARFORD-MONTGOMERY-HOWARD-SOUTHERN BALTIMORE-PRINCE GEORGES- ANNE ARUNDEL-CHARLES-MADISON-RAPPAHANNOCK-LOUDOUN-ORANGE-CULPEPER- PRINCE WILLIAM/MANASSAS/MANASSAS PARK-FAIRFAX- ARLINGTON/FALLS CHURCH/ALEXANDRIA-STAFFORD-SPOTSYLVANIA- KING GEORGE-NORTHERN FAUQUIER-SOUTHERN FAUQUIER- INCLUDING THE CITIES OF...WASHINGTON...FREDERICK...WESTMINSTER... GAITHERSBURG...COLUMBIA...BALTIMORE...ANNAPOLIS...WALDORF... LEESBURG...CULPEPER...MANASSAS...MANASSAS PARK...FAIRFAX... ALEXANDRIA...FALLS CHURCH...FREDERICKSBURG...WARRENTON 233 PM EST FRI DEC 6 2013 ...WINTER STORM WATCH IN EFFECT FROM SUNDAY MORNING THROUGH MONDAY MORNING... THE NATIONAL WEATHER SERVICE IN BALTIMORE MD/WASHINGTON HAS ISSUED A WINTER STORM WATCH...WHICH IS IN EFFECT FROM SUNDAY MORNING THROUGH MONDAY MORNING. * PRECIPITATION TYPE...SNOW AND SLEET WILL CHANGE OVER TO FREEZING RAIN SUNDAY AFTERNOON. FREEZING RAIN IS EXPECTED TO LAST UNTIL LATE SUNDAY NIGHT OR MONDAY MORNING BEFORE CHANGING TO RAIN. * ACCUMULATIONS...1 TO 2 INCHES OF SNOW AND SLEET AND A QUARTER INCH OR MORE OF ICE ACCUMULATION IS POSSIBLE. * TIMING...SNOW AND SLEET ARE EXPECTED TO BEGIN SUNDAY MORNING. A CHANGEOVER TO FREEZING RAIN IS EXPECTED MIDDAY SUNDAY AND WILL CONTINUE INTO THE NIGHT BEFORE CHANGING OVER TO RAIN LATE SUNDAY NIGHT OR MONDAY MORNING. * TEMPERATURES...AROUND 30 DEGREES. * WINDS...LIGHT NORTHERLY WINDS. * IMPACTS...HAZARDOUS TRAVEL CONDITIONS ARE PROBABLE DUE TO SLICK ROADS AND ICY CONDITIONS SUNDAY AND POSSIBLY INTO MONDAY. INCREASED POWER OUTAGES ARE POSSIBLE. PRECAUTIONARY/PREPAREDNESS ACTIONS... A WINTER STORM WATCH MEANS THERE IS A POTENTIAL FOR SIGNIFICANT SNOW...SLEET...AND ICE ACCUMULATIONS THAT MAY IMPACT TRAVEL. CONTINUE TO MONITOR THE LATEST FORECASTS. Link to comment Share on other sites More sharing options...
chris87 Posted December 6, 2013 Share Posted December 6, 2013 Interesting, thanks. I've tried to figure out where the issue is but haven't been able to yet. It does seem that sometimes the snow output tracks with the 32 surface line at wxbell but maybe I'm imagining it. That's unfortunately my other thought -- that he's using some poor surface temp threshold for snow but given his education level he should know better. Link to comment Share on other sites More sharing options...
usedtobe Posted December 6, 2013 Share Posted December 6, 2013 Yeah -- I've only looked at the GFS -- but i'd agree. Another thing to consider about snow maps (and could be why the WxBell maps may be so poor) -- I assume he's working with 6-hr ECMWF output -- so if he is just assuming that all QPF that falls in a 6-hour window is snow based on the categorical field in the raw output -- I believe that is an average over the 6-hr period rather than an instantaneous so if 3.01 hours of the 6 hour window support snow but 75% of the QPF falls in the second half of the 6-hour window -- the categorical snow would be yes while most of the QPF actually fell as rain. so if you're assigning where that QPF goes based on the categorical field you just added a bunch of extra snow that wasn't support by the thermal profiles -- i'm sure if he had hourly ECMWF output his snow maps would be greatly improved. It's the only explanation I have -- I can't believe the ECMWF p-type algorithm would be so poor it's much more likely he doesn't have the adequate temporal resolution to break down 6-hr QPF totals into the correct p-type bins. That make a lot of sense. We have those for the GFS but not for the Euro and it doesn't look like there is an interprolation scheme to weight how much fell as rain versus snow based on the change in temp. Link to comment Share on other sites More sharing options...
usedtobe Posted December 6, 2013 Share Posted December 6, 2013 URGENT - WINTER WEATHER MESSAGE NATIONAL WEATHER SERVICE BALTIMORE MD/WASHINGTON DC 233 PM EST FRI DEC 6 2013 DCZ001-MDZ004>007-009>011-013-014-016-VAZ039-040-042-050>057-501- 502-070345- /O.EXB.KLWX.WS.A.0008.131208T1400Z-131209T1200Z/ DISTRICT OF COLUMBIA-FREDERICK MD-CARROLL-NORTHERN BALTIMORE- HARFORD-MONTGOMERY-HOWARD-SOUTHERN BALTIMORE-PRINCE GEORGES- ANNE ARUNDEL-CHARLES-MADISON-RAPPAHANNOCK-LOUDOUN-ORANGE-CULPEPER- PRINCE WILLIAM/MANASSAS/MANASSAS PARK-FAIRFAX- ARLINGTON/FALLS CHURCH/ALEXANDRIA-STAFFORD-SPOTSYLVANIA- KING GEORGE-NORTHERN FAUQUIER-SOUTHERN FAUQUIER- INCLUDING THE CITIES OF...WASHINGTON...FREDERICK...WESTMINSTER... GAITHERSBURG...COLUMBIA...BALTIMORE...ANNAPOLIS...WALDORF... LEESBURG...CULPEPER...MANASSAS...MANASSAS PARK...FAIRFAX... ALEXANDRIA...FALLS CHURCH...FREDERICKSBURG...WARRENTON 233 PM EST FRI DEC 6 2013 ...WINTER STORM WATCH IN EFFECT FROM SUNDAY MORNING THROUGH MONDAY MORNING... THE NATIONAL WEATHER SERVICE IN BALTIMORE MD/WASHINGTON HAS ISSUED A WINTER STORM WATCH...WHICH IS IN EFFECT FROM SUNDAY MORNING THROUGH MONDAY MORNING. * PRECIPITATION TYPE...SNOW AND SLEET WILL CHANGE OVER TO FREEZING RAIN SUNDAY AFTERNOON. FREEZING RAIN IS EXPECTED TO LAST UNTIL LATE SUNDAY NIGHT OR MONDAY MORNING BEFORE CHANGING TO RAIN. * ACCUMULATIONS...1 TO 2 INCHES OF SNOW AND SLEET AND A QUARTER INCH OR MORE OF ICE ACCUMULATION IS POSSIBLE. * TIMING...SNOW AND SLEET ARE EXPECTED TO BEGIN SUNDAY MORNING. A CHANGEOVER TO FREEZING RAIN IS EXPECTED MIDDAY SUNDAY AND WILL CONTINUE INTO THE NIGHT BEFORE CHANGING OVER TO RAIN LATE SUNDAY NIGHT OR MONDAY MORNING. * TEMPERATURES...AROUND 30 DEGREES. * WINDS...LIGHT NORTHERLY WINDS. * IMPACTS...HAZARDOUS TRAVEL CONDITIONS ARE PROBABLE DUE TO SLICK ROADS AND ICY CONDITIONS SUNDAY AND POSSIBLY INTO MONDAY. INCREASED POWER OUTAGES ARE POSSIBLE. PRECAUTIONARY/PREPAREDNESS ACTIONS... A WINTER STORM WATCH MEANS THERE IS A POTENTIAL FOR SIGNIFICANT SNOW...SLEET...AND ICE ACCUMULATIONS THAT MAY IMPACT TRAVEL. CONTINUE TO MONITOR THE LATEST FORECASTS. It left me out. Link to comment Share on other sites More sharing options...
Ian Posted December 6, 2013 Share Posted December 6, 2013 It left me out. so close Link to comment Share on other sites More sharing options...
AvantHiatus Posted December 6, 2013 Share Posted December 6, 2013 It left me out. The timing seems to disagree somewhat with the Euro and GFS. Changeover by midday Sunday? Link to comment Share on other sites More sharing options...
chris87 Posted December 6, 2013 Share Posted December 6, 2013 That make a lot of sense. We have those for the GFS but not for the Euro and it doesn't look like there is an interprolation scheme to weight how much fell as rain versus snow based on the change in temp. Yes, so in that case it's maybe not his fault that the maps don't seem to make sense but it does make a compelling case that unless outside of any margin for warm layers, the ECMWF snow maps are pretty much worthless. All an educated guess on my end -- if anyone has worked with raw Euro data, they might have some additional thoughts about what could be going wrong. Link to comment Share on other sites More sharing options...
Ji Posted December 6, 2013 Author Share Posted December 6, 2013 It left me out. why dont you move to loudoun county....we have fish here too Link to comment Share on other sites More sharing options...
usedtobe Posted December 6, 2013 Share Posted December 6, 2013 why dont you move to loudoun county....we have fish here too My wife likes it here and I like her to be happy. What's funny about that watch is that I usually stay frozen longer than Waldorf because I'm farther north. Link to comment Share on other sites More sharing options...
clskinsfan Posted December 6, 2013 Share Posted December 6, 2013 NAM is still showing a burst of snow tomorrow morning. Not a big deal but its still there. Link to comment Share on other sites More sharing options...
Jebman Posted December 6, 2013 Share Posted December 6, 2013 Quick question. Why does the NWS have us under a winter storm watch with 1 to 2 inches of snow expected? That merits a winter weather advisory, not a WSW. Don't tell me the NWS has just gone all weenie on us. Link to comment Share on other sites More sharing options...
BTRWx Posted December 6, 2013 Share Posted December 6, 2013 18Z NAM looks colder through hr15. Link to comment Share on other sites More sharing options...
snow. Posted December 6, 2013 Share Posted December 6, 2013 NAM is still showing a burst of snow tomorrow morning. Not a big deal but its still there. I'm not seeing that at all Link to comment Share on other sites More sharing options...
Recommended Posts
Archived
This topic is now archived and is closed to further replies.