ZR Concerns

Forums (Forecast Builder)

RSS (Opens New Window)

Dennis Vancleve, modified 7 Years ago.

ZR Concerns

Youngling Posts: 4 Join Date: 11/24/12 Recent Posts

My first mixed precip event of this season has exposed some concerns. I ran through the top down grids and felt I would see rain, snow, or a rain/snow mix when FB finished running. I was surprised to see a slight chance of ZR mixed in there. The location of where it was didn’t make any sense based on the top down grids, so I investigated.

The slight chance of ZR only showed up for surface temps of 33 degrees, pops greater than 53, and MaxTwAloft greater than 0. I had ~100% prob ice for the entire period. If I lowered the T to 32, it was just snow. If I raised the T to 34, it was rain/snow. If the pop was lowered to 53 from 54, the slight chance ZR went away. If the MaxTwAloft was lowered from 0.1 to 0, the ZR went away.

-I understand the 33 plays a role for ZR based on the training, but this is out of place here. If there was a concern for ZR based on temps aloft, then it would exist for 32 degrees and less as well.

-What do pops have to do with anything? Maybe I missed it in the training, but I don’t see what bearing this would have on precip type.

-Is it meant for ZR to be introduced the moment MaxTw aloft exceeds 0?

Below are some graphics demonstrating these issues. Thanks for looking into this!

Edit area shows area where pops are greater than 53 percent:

For PotZR, area shows area where temps are equal to 33 degrees. Prob is 0 for 32 degree or lower to the nw:

Edit area shows area where MaxTwAloft is greater than 0. ZR chances start at 0.1 C:

Edward Townsend, modified 7 Years ago.

RE: ZR Concerns

Youngling Posts: 11 Join Date: 8/18/13 Recent Posts

Hello,

I think surface Tw may be the main factor. For freezing precip in PoWT you need MaxTwAloft >0, T between 32 and 35, and Tw <= 0. The latter is computed behind the scenes.

I think the PoP thing was a coincidence (or a bug but hopefully that isn't the case) as PoP doesn't play a role, I believe.

Do you have any screenshots of Td and MaxTwAloft? I only saw three images in your post, two of wx (with PoP and MaxTwAloft shown as edit areas) and PotZR. I thought my browser may be failing to load an image or two.

Thanks,

Ed T

Dennis Vancleve, modified 7 Years ago.

RE: ZR Concerns

Youngling Posts: 4 Join Date: 11/24/12 Recent Posts

Hi Ed! It's been a while...hope you're doing well!

I was able to easily recreate this situation today and have a few more images below. I do understand that there are background calculations going on when dewpoints are low enough to consider the surface Tw. That surface Tw calculation is clearly being made for temps of 33 degrees, but then ZR is only being added for the 33 degree areas.

I do wonder if ZR makes sense for TwAloft of just 0.1. Seems like a sleet situation if anything, with only partial melting.

Also, the pop >53 is not a coincidence...I can consistently recreate this. It doesn't affect the PotZR, but rather the final wx grid...only applying the PotZR for pops greater than 53.

I'd be happy to be wrong about all this (certainly not the first time!), but it really seems like something goofy is going on.

Thanks for the reply!

Denny

Area where PoP is greater than 53. Slight chance ZR cuts off to the east of this edit area:

Area where T = 33:

PotFreezingRain:

Temps:

Dewpoints:

Edward Townsend, modified 7 Years ago.

RE: ZR Concerns

Youngling Posts: 11 Join Date: 8/18/13 Recent Posts

Hey Denny,

I'm good! It has been a while, we will have to catch up :).

Anyway, it does look like you're right. I was able to reproduce your behavior. A pdf is attached showing my reproduction.

In my testing the following were constant: Td at 30 F, MaxTwAloft was above 0 (at 0.1 F), ProbIce was 100, ProbRefreezeSleet was 100. I did three cases total with T alternating between 31, 32, and 33 F. Lastly, I had PoPs alternating between 53 and 54. I only got ZR in the case with PoP set to 54 and T set to 33 F. Thus I didn't get it for the combinations of T = 31 (with PoP = 53 or PoP = 54) or T = 32 (with PoP = 53 or PoP = 54). So it does look like PoP is playing some role.

Unfortunately, I'm a little confused in what is happening and will need to dig into the code. Though someone else (like Andy) may be able to beat me to an answer.

That said, I was a little surprised because in my testing I got some behavior I didn't expect. Namely, I got a grid of Chc R-S- when PoP was 53 followed by a grid with Chc R-S- and Sch ZR- when PoP was 54. I am not sure why I didn't get ZR- produced in the former given Top-down produced a PotFreezingRain grid there with a value of 26% (so it should have had a Sch mention of ZR- too).

--Ed T

Dennis Vancleve:

Hi Ed! It's been a while...hope you're doing well!

I was able to easily recreate this situation today and have a few more images below. I do understand that there are background calculations going on when dewpoints are low enough to consider the surface Tw. That surface Tw calculation is clearly being made for temps of 33 degrees, but then ZR is only being added for the 33 degree areas.

I do wonder if ZR makes sense for TwAloft of just 0.1. Seems like a sleet situation if anything, with only partial melting.

Also, the pop >53 is not a coincidence...I can consistently recreate this. It doesn't affect the PotZR, but rather the final wx grid...only applying the PotZR for pops greater than 53.

I'd be happy to be wrong about all this (certainly not the first time!), but it really seems like something goofy is going on.

Thanks for the reply!

Denny

Area where PoP is greater than 53. Slight chance ZR cuts off to the east of this edit area:

Area where T = 33:

PotFreezingRain:

Temps:

Dewpoints:

Andy Just, modified 7 Years ago.

RE: ZR Concerns

Youngling Posts: 89 Join Date: 6/2/15 Recent Posts

Denny and Ed,

Looks like this is a 2 part question and everything is working as expected. Hopefully my explanation here will work but please feel free to respond if it doesn't. There's a lot going on in this scenario.

1. Seeing the values the PotFreezingRain grid immediately stands out the situation described in FRAM training. If you have MaxTwAloft > 0, a temperature greater than 32 but less than 36 F, and wet-bulb less than 0 C, which were all true in your case, will trigger a situation of low freezing rain probabilities. The assumption made here is that from at least 2000 ft AGL where the MaxTwAloft resides, down to the surface, your temperature profile stays above 0C. In this scenario, enough warm air exists to allow for full melting, but then the dewpoint plays a role in allowing the potential for a little icing to occur at the surface, thus the 10 and 26% probabilities. The FRAM research comes up with the probabilities. Now, back to the assumption. If the temperature profile went back below freezing below the MaxTwAloft, then yes a sleet scenario would be more likely. This is where we would like to adjust and or replace the ProbRefreezeSleet grid to give us some idea if there is a subfreezing zone below the MaxTwAloft. That process is still in research.

So after #1, if your MaxTwAloft was <= 0C, it assumed a rain/snow scenario. Raising the T to 34 may have been enough to increase the Tw to 0C

2. How the PoP is interacting to yield ZR or not in the Wx grid. From the PoWT knobology training, In mixed precipitation type scenarios, the probability of type(s) with the highest value is given the value of the PoP. In your case Rain and Snow I'm guessing were equivalent. For all other types, we utilize PoP * Type / 100. So at a 53 PoP and a 26 ZR, that equates to 13.78, which is less than 14, the value set as statistically high enough to mention in the Wx grid. Increase the PoP to 54, and you get 14.04 and thus the SChc of ZR gets mentioned. One change I will make here is to raise 14 to 14.4, as the code just checks for greater than, and would like to get it the smidge closer to 15 that we are normally used to. Nevertheless this shows why 53 and 54 interacted differently for this case.

FYI ProbRefreezeSleet would not play any role in this case. ProbRefreezeSleet is only incorporated when MaxTwAloft >= 3C.

Again please let me know if you have any questions on this. In summary, your dewpoint and the "strength" of your PoP value played a role in having the ZR mentioned.

Dennis Vancleve, modified 7 Years ago.

RE: ZR Concerns

Youngling Posts: 4 Join Date: 11/24/12 Recent Posts

Thanks for the reply, Andy! I'm glad to hear that the code is functioning as expected. That being said, I think it's worth discussing whether this output is desirable.

For the T=33 setup, I believe you are saying that the T of 33 and MaxTwAloft greater than 0 produces the assumption of temps above freezing from the surface to the point of MaxTwAloft. Makes sense, though is it proven that temps of only a fraction of a degree above freezing through a 2 kft layer would be enough for full melting? I should take a peek at some of the training again. :-) Also, if the T of 33 at the surface plays a role in the assumption of full melting, how then can we use the Tw at the surface to suggest a possibility of freezing rain? Seems it should be one or the other. Lastly, I think it's questionable that you can have a slight chance of freezing rain only at 33 degrees given this specific setup. I've seen this multiple times now. I know it follows the coding, but it just doesn't seem to make meteorological sense.

As for the PoP issue, I get how the coding works and I remember it from the training. I do think that increasing pops allowing for increasing # of P types is a bit concerning though. Seems # of precip types should rely strictly on thermal profile. It was just odd and not very meteorological to see a slight chance for freezing rain advect in from the northwest with the higher pops. I just don't think this is desired behavior.

Thanks for considering my concerns!

Denny