April 14-15 Snow Potential

[michsnowfreak]

6 hours ago, Hoosier said:

A look back at that April 1961 storm:

That one had a bigger area of snow.  For the city of Chicago/immediate suburbs, the 2 storms appear to be pretty similar.  Farther west of the city, the 2019 storm is more impressive. 

The Winter of 1960-61 was an absolute train wreck for snow in Detroit. The April 16/17th storm was the biggest of the entire season! 

[Hoosier]

Really did ride the edge

[SchaumburgStormer]

3 minutes ago, Hoosier said:

Really did ride the edge

I think the HRRR from the night before was about spot on 

[RCNYILWX]

I have not posted in a while, especially since going back to work in early March after paternity leave for birth of my 2nd child, but I've been lurking. Worked leading up to the event and saw that a few of my AFDs were pasted here. Several posters criticized my office's (LOT) handling of the event, and admittedly, when amounts are that much higher than we were forecasting, some criticism is warranted. Wanted to make a few points from my perspective as an explanation but certainly not an excuse for what wasn't a good forecast. Apologies if this is a bit meandering and a lot long.

 

We were well aware of the models that showed the heavier amounts; climo and time of day definitely played a role in not really fully buying into the extreme amounts. At the same time we were also well aware of the potential ingredients it had going for it toward producing heavy snowfall rates, many of which I included in my AFDs so I won't list them here.

 

I saw the points made about snowfall rates and run angle w.r.t. road impacts and the point is well taken that extreme rates can and will overcome the hostile sun angle this time of year. On the day before the event, I wasn't doubting that there would be heavy snow, mentioned it and the TSSN potential. The main questions in my mind were exactly how heavy it would be, how long/persistent the heavy rates would last and how narrow the area affected by heavier rates would be.

 

These are important because I think the sun angle is still part of the equation. Once the rates eased on Sunday, the accums basically stopped until toward and after sunset, which showed the how important it was to get that very persistent death band with 30-45+ dbz echoes to get the sig accums, accums on roads and magnitude of travel impacts that occurred. Ultimately I think it's hard to fully buy into the idea that we could be staring at a historic event, despite there being support from the Euro/NAM12, SREF and RAP/HRRR, when the bulk of the snow qpf would occur during daylight hours. I believe it's a natural thing for a forecaster to gravitate toward the idea that some failure modes would help keep things relatively in check, given how rare heavy snow events are at our latitude in mid April. This appeared to be the case not just at my office but also at surrounding offices in the lead up to the event. I do believe that if the heaviest rates were forecast to be at night, we would've been more bullish with our amounts.

 

I personally am leery of events that are very dependent upon dynamic cooling, an issue any time of year but especially in April because so much needs to go right. One of my most memorable busted forecasts was Christmas Eve 2014, a relatively similar setup modeled to what occurred on Sunday. I had made mention in one of my AFDs of the potential for widespread intense warm sector convection to *possibly* mess with cold sector dynamics, with that 2014 event in mind, something that obviously didn't come to pass this time.

 

With all of the above being said, I definitely was quite concerned for the potential for heavy snow rates Sunday AM-early PM, and hit my Saturday afternoon AFD and graphic much harder than the Friday PM ones and had a pretty good idea of what would end up being the hardest hit swath. In hindsight, I wish I issued a WWA instead of only a SPS with that Sat. PM forecast package considering what played out even though my amounts would've been too low. I was open to the idea, but DVN/MKX/ILX were not leaning that way, so ultimately felt I was lacking confidence to do one. I can't speak for the other forecasters at my office that worked the event, but I can assure you we all worked hard like we always do, but end result wasn't good.

 

As to the slower response to issuing the headline on Sunday, I'd like to say I'd have been a bit quicker to do so given radar and report trends and that DVN and ILX had issued, and that it is a valid point of criticism. However, it's never an easy thing to be in the driver's seat at a NWS WFO with those decisions to be made. From asking someone who was there, I think it came down to exactly how persistent that extremely intense would be and how roads would respond, and how widespread impacts would get and wait more for trends to unfold. How it exactly transpired made it seem like we were ignoring the obvious but again I can assure everyone that it's not something that was taken lightly. We will certainly have this as important case to learn from moving forward, and I know I'll never forget it. Sorry again for this being long, but if you have any other questions, I'd be happy to at least attempt to answer.

 

 

 

 

 

 

 

 

[Hoosier]

1 hour ago, RCNYILWX said:

I have not posted in a while, especially since going back to work in early March after paternity leave for birth of my 2nd child, but I've been lurking. Worked leading up to the event and saw that a few of my AFDs were pasted here. Several posters criticized my office's (LOT) handling of the event, and admittedly, when amounts are that much higher than we were forecasting, some criticism is warranted. Wanted to make a few points from my perspective as an explanation but certainly not an excuse for what wasn't a good forecast. Apologies if this is a bit meandering and a lot long.

 

We were well aware of the models that showed the heavier amounts; climo and time of day definitely played a role in not really fully buying into the extreme amounts. At the same time we were also well aware of the potential ingredients it had going for it toward producing heavy snowfall rates, many of which I included in my AFDs so I won't list them here.

 

I saw the points made about snowfall rates and run angle w.r.t. road impacts and the point is well taken that extreme rates can and will overcome the hostile sun angle this time of year. On the day before the event, I wasn't doubting that there would be heavy snow, mentioned it and the TSSN potential. The main questions in my mind were exactly how heavy it would be, how long/persistent the heavy rates would last and how narrow the area affected by heavier rates would be.

 

These are important because I think the sun angle is still part of the equation. Once the rates eased on Sunday, the accums basically stopped until toward and after sunset, which showed the how important it was to get that very persistent death band with 30-45+ dbz echoes to get the sig accums, accums on roads and magnitude of travel impacts that occurred. Ultimately I think it's hard to fully buy into the idea that we could be staring at a historic event, despite there being support from the Euro/NAM12, SREF and RAP/HRRR, when the bulk of the snow qpf would occur during daylight hours. I believe it's a natural thing for a forecaster to gravitate toward the idea that some failure modes would help keep things relatively in check, given how rare heavy snow events are at our latitude in mid April. This appeared to be the case not just at my office but also at surrounding offices in the lead up to the event. I do believe that if the heaviest rates were forecast to be at night, we would've been more bullish with our amounts.

 

I personally am leery of events that are very dependent upon dynamic cooling, an issue any time of year but especially in April because so much needs to go right. One of most memorable busted forecasts was Christmas Eve 2014, a relatively similar setup modeled to what occurred on Sunday. I had made mention in one of my AFDs of the potential for widespread intense warm sector convection to *possibly* mess with cold sector dynamics, with that 2014 event in mind, something that obviously didn't come to pass this time.

 

With all of the above being said, I definitely was quite concerned for the potential for heavy snow rates Sunday AM-early PM, and hit my Saturday afternoon AFD and graphic much harder than the Friday PM ones and had a pretty good idea of what would end up being the hardest hit swath. In hindsight, I wish I issued a WWA instead of only a SPS with that Sat. PM forecast package considering what played out even though my amounts would've been too low. I was open to the idea, but DVN/MKX/ILX were not leaning that way, so ultimately felt I was lacking confidence to do one. I can't speak for the other forecasters at my office that worked the event, but I can assure you we all worked hard like we always do, but end result wasn't good.

 

As to the slower response to issuing the headline on Sunday, I'd like to say I'd have been a bit quicker to do so given radar and report trends and that DVN and ILX had issued, and that it is a valid point of criticism. However, it's never an easy thing to be in the driver's seat at a NWS WFO with those decisions to be made. From asking someone who was there, I think it came down to exactly how persistent that extremely intense would be and how roads would respond, and how widespread impacts would get and wait more for trends to unfold. How that occurred made it seem like we were ignoring the obvious but again I can assure everyone that it's not something that was taken lightly. We will certainly have this as important case to learn from moving forward, and I know I'll never forget it. Sorry again for this being long, but if you have any other questions, I'd be happy to at least attempt to answer.

 

 

 

 

The bold is something I alluded to on Sunday.  When something is modeled that you've rarely or never seen before, it's human nature to have some difficulty in processing that.  Another somewhat recent example from outside the region is Hurricane Michael... probably even more of an outlier.  Models suggested a very intense hurricane hitting the FL panhandle, but did you (the global you, not you specifically) really think it was going to landfall there as a high end cat 4 (or cat 5 if rumors are true) until it strengthened toward that level?  Forecasting is prone to error anyway but outlier events just add another layer of difficulty.   

[SchaumburgStormer]

8 hours ago, RCNYILWX said:

I have not posted in a while, especially since going back to work in early March after paternity leave for birth of my 2nd child, but I've been lurking. Worked leading up to the event and saw that a few of my AFDs were pasted here. Several posters criticized my office's (LOT) handling of the event, and admittedly, when amounts are that much higher than we were forecasting, some criticism is warranted. Wanted to make a few points from my perspective as an explanation but certainly not an excuse for what wasn't a good forecast. Apologies if this is a bit meandering and a lot long.

 

We were well aware of the models that showed the heavier amounts; climo and time of day definitely played a role in not really fully buying into the extreme amounts. At the same time we were also well aware of the potential ingredients it had going for it toward producing heavy snowfall rates, many of which I included in my AFDs so I won't list them here.

 

I saw the points made about snowfall rates and run angle w.r.t. road impacts and the point is well taken that extreme rates can and will overcome the hostile sun angle this time of year. On the day before the event, I wasn't doubting that there would be heavy snow, mentioned it and the TSSN potential. The main questions in my mind were exactly how heavy it would be, how long/persistent the heavy rates would last and how narrow the area affected by heavier rates would be.

 

These are important because I think the sun angle is still part of the equation. Once the rates eased on Sunday, the accums basically stopped until toward and after sunset, which showed the how important it was to get that very persistent death band with 30-45+ dbz echoes to get the sig accums, accums on roads and magnitude of travel impacts that occurred. Ultimately I think it's hard to fully buy into the idea that we could be staring at a historic event, despite there being support from the Euro/NAM12, SREF and RAP/HRRR, when the bulk of the snow qpf would occur during daylight hours. I believe it's a natural thing for a forecaster to gravitate toward the idea that some failure modes would help keep things relatively in check, given how rare heavy snow events are at our latitude in mid April. This appeared to be the case not just at my office but also at surrounding offices in the lead up to the event. I do believe that if the heaviest rates were forecast to be at night, we would've been more bullish with our amounts.

 

I personally am leery of events that are very dependent upon dynamic cooling, an issue any time of year but especially in April because so much needs to go right. One of my most memorable busted forecasts was Christmas Eve 2014, a relatively similar setup modeled to what occurred on Sunday. I had made mention in one of my AFDs of the potential for widespread intense warm sector convection to *possibly* mess with cold sector dynamics, with that 2014 event in mind, something that obviously didn't come to pass this time.

 

With all of the above being said, I definitely was quite concerned for the potential for heavy snow rates Sunday AM-early PM, and hit my Saturday afternoon AFD and graphic much harder than the Friday PM ones and had a pretty good idea of what would end up being the hardest hit swath. In hindsight, I wish I issued a WWA instead of only a SPS with that Sat. PM forecast package considering what played out even though my amounts would've been too low. I was open to the idea, but DVN/MKX/ILX were not leaning that way, so ultimately felt I was lacking confidence to do one. I can't speak for the other forecasters at my office that worked the event, but I can assure you we all worked hard like we always do, but end result wasn't good.

 

As to the slower response to issuing the headline on Sunday, I'd like to say I'd have been a bit quicker to do so given radar and report trends and that DVN and ILX had issued, and that it is a valid point of criticism. However, it's never an easy thing to be in the driver's seat at a NWS WFO with those decisions to be made. From asking someone who was there, I think it came down to exactly how persistent that extremely intense would be and how roads would respond, and how widespread impacts would get and wait more for trends to unfold. How it exactly transpired made it seem like we were ignoring the obvious but again I can assure everyone that it's not something that was taken lightly. We will certainly have this as important case to learn from moving forward, and I know I'll never forget it. Sorry again for this being long, but if you have any other questions, I'd be happy to at least attempt to answer.

 

 

 

 

 

 

 

 

I certainly understand and don’t necessarily disagree with the handling of headlines/forecast ahead of the storms arrival. I think prudent forecasting would call for not leaning on a solution that is historic and/or well on the fringes of probability. 

With that being said, you also cannot sit on your hands when it becomes apparent that this historic event is unfolding. Upstream obs made it fairly clear early in the morning that the low probability event was unfolding. I was getting 2+ inch per hour rates with thundersnow and near zero visibility with no headline. Roads went to shit very quickly, with the general public mostly unaware that this kind of impact was even on the table. After the death band had shifted north a bit (but still under mod-heavy snow) we get a WWA for 2-6” when central Dekalb county and areas NE already have 6” already on the ground. 

While I don’t envy your position, the messages seemed to vary wildly in the AFD’s leading to the event. From informative of the issues at play, to a complete downplaying of the models, it seemed like the shifts could not get on the same page. 

A good event to look back on and study. I would assume that the NWS is having some debriefs and discussing all sides of this and determining what can be done better next time. Your office does a great job, and we appreciate your inside info. 

[michsnowfreak]

On April 23-25, 2005 a historic late season snowstorm hit southeast MI with 4-16" of snow. It was SO late in the season and coming on very warm ground, that the same doubts occurred. What as crazy is that while snow fell all night Apr 23-24, it was on the light side, so accumulation was just a bit of slush, despite coming during the preferred overnight hours. Rates picked up to heavy snow during the late morning and very heavy snow fell most of the afternoon, which was when a huge majority of the accumulation occurred. During daylight hours. On April 24. Lesser rates during the overnight produced little if any accum, so despite snowing all night Apr 23-24, all day Apr 24, and all night Apr 24-25, about 80-90% or accum occurred during the daytime hours because of rates. If heavy rates are realizes, it will accumulate, period!

[kevlon62]

One for the books. Last shot from the end of our day.

Sent from my SM-N960U using Tapatalk

[RobertSul]

On 4/18/2019 at 2:11 PM, michsnowfreak said:

On April 23-25, 2005 a historic late season snowstorm hit southeast MI with 4-16" of snow. It was SO late in the season and coming on very warm ground, that the same doubts occurred. What as crazy is that while snow fell all night Apr 23-24, it was on the light side, so accumulation was just a bit of slush, despite coming during the preferred overnight hours. Rates picked up to heavy snow during the late morning and very heavy snow fell most of the afternoon, which was when a huge majority of the accumulation occurred. During daylight hours. On April 24. Lesser rates during the overnight produced little if any accum, so despite snowing all night Apr 23-24, all day Apr 24, and all night Apr 24-25, about 80-90% or accum occurred during the daytime hours because of rates. If heavy rates are realizes, it will accumulate, period!

This was one week before I moved to Michigan - upset that I missed it!

[michsnowfreak]

2 hours ago, RobertSul said:

This was one week before I moved to Michigan - upset that I missed it!

2004-05 was a great Winter, then the next 2 were duds before a string of great winters commenced. Apr 23-25 was certainly not the biggest storm of the season here, that was a 12 incher in January, but parts of the thumb had around 16" of snow with 3' drifts.  I can tell you my time line for the storm. Despite starting to snow around mid afternoon April 23rd, by 8:00 a.m. April 24th, the snow at this point having been non stop, we probably had a total of about 1.5" of slush. Then from about 11:00 a.m. To 4:00 p.m. April 24th, we picked up 5" of great packing snow before rates tailed off and we picked up minimal additional accumulation, though snow continued unabated til early morning the 25th. My storm told snowfall was 6.7", with a peak depth of around 6" which occurred mid afternoon the 24th. By 8:00 a.m. April 25th the snow depth was down to 2" and it was gone by noon the 25th.