mappy Posted September 6, 2011 Share Posted September 6, 2011 Hi. It is raining. That is all. Link to comment Share on other sites More sharing options...
WxUSAF Posted September 6, 2011 Author Share Posted September 6, 2011 Thought you might enjoy a close-up of the HPC qpf forecast for the next 72 hours and 120 hours starting 8am this morning. Great stuff! Post things like that more often. I've tried a few comparisons between AccuWeather, NWS, and Weather.com, and haven't seen a lot to choose from. I will grant you that Accuweather's 15-day forecasts can be pretty funny, when on Day 1, they predict, say a high of 70 with a 10% chance of rain; and then on Day 2, revise that to a high of 60 with a 70% chance of rain. Still, a detailed study would be needed to do a proper comparison. When a service has a spectacular success or failure -- particularly with regard to snow -- that tends to be remembered, whereas less spectacular results tend to be quickly forgotten. I don't know of any rigorous studies between the 3, although I know Accuwx claims to be better (shock!). But, there have been several studies comparing straight model run forecasts to those with human input. The human input forecasts consistently outperform model forecasts through the first 7-10 days. After that, it's about a toss up. Perhaps another red-tagger who regularly reads the meteorological journals could provide a specific reference to one such study. The forecast you get on Accuwx.com is 100% model derived (and same with weather.com) as far as I know. The NWS forecast has human input, so I'd always put more value in it. I typically read the FD's, look at the zone, and fill in blanks with what I think will happen. Hard to go wrong when reading the NWS FD's and looking at your zone. The best part of the discussions is the analysis of the features when it is a low confidence forecast. It gives armchair guys like me a chance to make my own analysis and see what happens. Good post. Link to comment Share on other sites More sharing options...
H2O Posted September 6, 2011 Share Posted September 6, 2011 decent rain blob now Link to comment Share on other sites More sharing options...
yoda Posted September 6, 2011 Share Posted September 6, 2011 Hi. It is raining. That is all. pics or it didn't happen Link to comment Share on other sites More sharing options...
yoda Posted September 6, 2011 Share Posted September 6, 2011 SLGT risk for VA/DC/C MD on 1730 SPC OTLK Link to comment Share on other sites More sharing options...
Kmlwx Posted September 6, 2011 Share Posted September 6, 2011 Slight risk for severe tstms tomorrow per SPC 1730z update. Link to comment Share on other sites More sharing options...
Kmlwx Posted September 6, 2011 Share Posted September 6, 2011 SLGT risk for VA/DC/C MD on 1730 SPC OTLK I hate you. Link to comment Share on other sites More sharing options...
H2O Posted September 6, 2011 Share Posted September 6, 2011 I hate you. next time use your telescope to see it coming from farther away than yoda Link to comment Share on other sites More sharing options...
yoda Posted September 6, 2011 Share Posted September 6, 2011 I hate you. Quicker you must be. Posted that on my phone I did. Cut and paste, I did not Link to comment Share on other sites More sharing options...
yoda Posted September 6, 2011 Share Posted September 6, 2011 next time use your telescope to see it coming from farther away than yoda Link to comment Share on other sites More sharing options...
WxUSAF Posted September 6, 2011 Author Share Posted September 6, 2011 Moderate rain in Laurel. Heaviest of the day so far. Link to comment Share on other sites More sharing options...
Ian Posted September 6, 2011 Share Posted September 6, 2011 Over under on 5 torn warns tomorrow for LWX? Link to comment Share on other sites More sharing options...
Ian Posted September 6, 2011 Share Posted September 6, 2011 Problem with HPC maps in a convective event is they will be wrong. Link to comment Share on other sites More sharing options...
North Balti Zen Posted September 6, 2011 Share Posted September 6, 2011 Hi. It is raining. That is all. Meh watch cancel? Link to comment Share on other sites More sharing options...
TradeWinds Posted September 6, 2011 Share Posted September 6, 2011 Thought you might enjoy a close-up of the HPC qpf forecast for the next 72 hours and 120 hours starting 8am this morning. Nice maps but seems the bullseye is way off. The same train as last night over MoCo, Howard, Carroll, and areas west of the Bay seem to have set up in that area again today. It shows well on the storm totals map. Link to comment Share on other sites More sharing options...
mappy Posted September 6, 2011 Share Posted September 6, 2011 pics or it didn't happen It was raining, and will be raining again soon. Link to comment Share on other sites More sharing options...
Kmlwx Posted September 6, 2011 Share Posted September 6, 2011 next time use your telescope to see it coming from farther away than yoda Over under on 5 torn warns tomorrow for LWX? Zero. Problem with HPC maps in a convective event is they will be wrong. Yeah it should look more like chicken pox with little peaks here and there. Link to comment Share on other sites More sharing options...
yoda Posted September 6, 2011 Share Posted September 6, 2011 Over under on 5 torn warns tomorrow for LWX? Under... I will say 3. Over under on 60% chance of a tornado watch for LWX tomorrow? Link to comment Share on other sites More sharing options...
chris87 Posted September 6, 2011 Share Posted September 6, 2011 Under... I will say 3. Over under on 60% chance of a tornado watch for LWX tomorrow? I'll go under too... can you really have an over/under on a probability of an event? it's either yes or no... Link to comment Share on other sites More sharing options...
Grothar of Herndon Posted September 6, 2011 Share Posted September 6, 2011 Radar filling in nicely down in Lexington, Lynchburg to Farmville. This has been a good rain producer here and looks like more coming! Event total: 2.71" Link to comment Share on other sites More sharing options...
Ian Posted September 6, 2011 Share Posted September 6, 2011 I'm tempted to take the over based on history of the event but getting clearing is a pain around here. Link to comment Share on other sites More sharing options...
WxUSAF Posted September 6, 2011 Author Share Posted September 6, 2011 Problem with HPC maps in a convective event is they will be wrong. I think if you had high-density accurate observations and then smoothed the totals with nearby stations, you'd end up getting something that would look much like a HPC map. Especially since today (at least) it's more like embedded convection, so you don't have locations with 0" and nearby locations with several inches. Link to comment Share on other sites More sharing options...
Ian Posted September 6, 2011 Share Posted September 6, 2011 I think if you had high-density accurate observations and then smoothed the totals with nearby stations, you'd end up getting something that would look much like a HPC map. Especially since today (at least) it's more like embedded convection, so you don't have locations with 0" and nearby locations with several inches. I guess tho so far the pattern doesn't match as we keep getting convective blobs rotating through. I think our knowledge of where those track is pretty poor beforehand. Of course you can make educated guesses based on fronts and track etc. It's just not going to be ass accurate as say it was with Irene or a nor'easter. But yes moreso than a day with pop up storms. Link to comment Share on other sites More sharing options...
Jns2183 Posted September 6, 2011 Share Posted September 6, 2011 Just did a screenshot clip of storm totals from CCX and LWX. The HPC maps definitely have issues with convective events, just compare the HPC maps to the latest 15-hour HRRR maps and the banding nature is very clear. I am from Cumberland county up by Harrisburg and for our county alone the HPC totals range from 4 to 7 inches. CTP has been very clear in there AFD about not being sure where the heavy axis of precipitation will set-up this event, especially with areas that see training convection. Here are the current radar estimates, enjoy. Link to comment Share on other sites More sharing options...
Jns2183 Posted September 6, 2011 Share Posted September 6, 2011 Well the 12z GFS has lost its marbles since its printing out 16.81" as a max in 72 hours, and 17.18" in 120 hours. I can only guess it is having some feedback issues. Link to comment Share on other sites More sharing options...
WVclimo Posted September 6, 2011 Share Posted September 6, 2011 Just did a screenshot clip of storm totals from CCX and LWX. The HPC maps definitely have issues with convective events, just compare the HPC maps to the latest 15-hour HRRR maps and the banding nature is very clear. I am from Cumberland county up by Harrisburg and for our county alone the HPC totals range from 4 to 7 inches. CTP has been very clear in there AFD about not being sure where the heavy axis of precipitation will set-up this event, especially with areas that see training convection. Here are the current radar estimates, enjoy. Interesting that the LWX map shows my area in the 0.30-0.60" shading, while the CCX radar shows correctly the 2.00-2.50" range. The 24-hr lag by LWX on totals is a factor certainly, but even so, I had well over an inch-and-a-half after LWX started the loop yesterday afternoon. Link to comment Share on other sites More sharing options...
BethesdaWX Posted September 6, 2011 Share Posted September 6, 2011 just 0.63 so far but it feels like much more. Watching the convection with lightning on Grlevel3 slowly working it's way northward in the means, would be nice to get some good lightning video, been like 3 weeks since anything has hit nearby. Link to comment Share on other sites More sharing options...
Ian Posted September 6, 2011 Share Posted September 6, 2011 The maps are good for what they are. I think in most cases a probability map is better tho you can sort of read them the same way (IE, heavier totals more likely in the heavier total area.). They had my current location in Cape Cod at less than .5" for the 5 day period an we about doubled that this am with more inbound. Link to comment Share on other sites More sharing options...
WxUSAF Posted September 6, 2011 Author Share Posted September 6, 2011 Yeah, certainly a probability map would make more sense in a convective event, but I think the public has a hard enough time understanding forecast probabilities as is, rather than compounding probabilities with "30% chance of thunderstorms today with a 25% chance of 1" of rain". That's a big bag of WTF... In a convective event, a forecast range would make more sense, IMO. I think the synoptic maps are fine as is for stratiform precip. Still even then you get lollipops, that's just the crap chute that is weather. Link to comment Share on other sites More sharing options...
mappy Posted September 6, 2011 Share Posted September 6, 2011 Yeah, certainly a probability map would make more sense in a convective event, but I think the public has a hard enough time understanding forecast probabilities as is, rather than compounding probabilities with "30% chance of thunderstorms today with a 25% chance of 1" of rain". That's a big bag of WTF... In a convective event, a forecast range would make more sense, IMO. I think the synoptic maps are fine as is for stratiform precip. Still even then you get lollipops, that's just the crap chute that is weather. i give it 2, was close to a 3 but due to region posts, i rounded down Link to comment Share on other sites More sharing options...
Recommended Posts
Archived
This topic is now archived and is closed to further replies.