Alta, UT 1/1/2013

[Tony Crocker]

We're saying that in any given year, there is a high likelihood that a 2-3 week period around the middle of January is going to be dryer and more prone to inversion.

Really, given that it's almost certain there was no such pattern for 20 years prior to 2005??? Maybe it's climate change. :stir:

Again, a dry spell in January is undoubtedly more likely to produce the Salt Lake valley inversion that in months when the sun is stronger. But the probability of the dry spell itself is not materially different than in other winter months.

We asked specifically how the ~99" Tony cited was distributed over the entire month.

I did that and documented over the last 8 years an extreme dry period of 32% of expected snowfall from Jan. 12-17, bounded by two dates with over 150% expected snowfall, a pattern which strongly implies random noise from a too small data set.

You want the whole data set?
Sum of 2005-2012 Alta snowfall by date:
31-Jan 37.0
30-Jan 25.0
29-Jan 20.0
28-Jan 35.0
27-Jan 24.5
26-Jan 23.0
25-Jan 28.0
24-Jan 29.0
23-Jan 26.5
22-Jan 28.5
21-Jan 53.0
20-Jan 22.0
19-Jan 32.0
18-Jan 41.5
17-Jan 6.5
16-Jan 8.5
15-Jan 20.5
14-Jan 4.0
13-Jan 6.0
12-Jan 4.0
11-Jan 46.5
10-Jan 20.0
9-Jan 37.5
8-Jan 34.5
7-Jan 15.5
6-Jan 53.0
5-Jan 42.0
4-Jan 26.0
3-Jan 8.0
2-Jan 33.5
1-Jan 6.0

 

[Admin]

Really, given that it's almost certain there was no such pattern for 20 years prior to 2005???

Precisely where, sir, did you document evidence of that?

We asked specifically how the ~99" Tony cited was distributed over the entire month.

I did that and documented over the last 8 years an extreme dry period of 32% of expected snowfall from Jan. 12-17, bounded by two dates with over 150% expected snowfall, a pattern which strongly implies random noise from a too small data set.

Or it implies precisely what we're trying to tell you. You're seeing what you want to see in your data.

Sent from my Samsung Galaxy Note II using Tapatalk 2

 

[socal]

We asked specifically how the ~99" Tony cited was distributed over the entire month.

I did that and documented over the last 8 years an extreme dry period of 32% of expected snowfall from Jan. 12-17, bounded by two dates with over 150% expected snowfall, a pattern which strongly implies random noise from a too small data set.

Or it implies precisely what we're trying to tell you. You're seeing what you want to see in your data.

Sent from my Samsung Galaxy Note II using Tapatalk 2

Yeah, I'm no actuary but look at your chart, the 12th through the 17th are WAY lower in snowfall than the rest of the month. I think you said it best yourself but just couldn't help but add something to it to support your original thesis:

over the last 8 years an extreme dry period of 32% of expected snowfall from Jan. 12-17,

 

[Tony Crocker]

look at your chart, the 12th through the 17th are WAY lower in snowfall than the rest of the month.

Yes, 32% of normal. For the same 8 years on the same dates Mammoth is 34% of normal. Sorry that Alta's daily snowfall data is not available online for prior years but Mammoth's is. Both long term data and observation of individual storm tracks show that a lot if not the majority of storms that hit the Sierra continue on and dump in the Wasatch also. So the failure of this "mid-January dry spell" to exist for 20+ years prior to 2005 in the Sierra strongly implies that it did not exist in the Wasatch either.

We can probably dig up some Wasatch SNOTEL data before 2005 if anyone really wants to pursue this further.

 

[Admin]

Sorry that Alta's daily snowfall data is not available online for prior years but Mammoth's is. Both long term data and observation of individual storm tracks show that a lot if not the majority of storms that hit the Sierra continue on and dump in the Wasatch also. So the failure of this "mid-January dry spell" to exist for 20+ years prior to 2005 in the Sierra strongly implies that it did not exist in the Wasatch either.

Seriously? Are you for real? You're now trying to correlate Utah inversions to Mammoth snowfall and you honestly believe that we'll buy that crap over personal observation?

Hard to believe but you've outdone yourself this time.

We can probably dig up some Wasatch SNOTEL data before 2005 if anyone really wants to pursue this further.

Yes, please.


Sent from my Samsung Galaxy Note II using Tapatalk 2

 

[Tony Crocker]

Yes, please.

Here you go :
30 years 1971-2000 for Silverlake Brighton 8,700 feet, the weather station admin and I actually drove up to observe in May 2005 I think. http://www.wrcc.dri.edu/cgi-bin/cliNORM2000t.pl?utsilv
DOY M DY MAX #YRS TMIN PRECIP #YRS SD MAX SD MIN
1 1 1 29.9 30. 8.4 30. 0.162 30. 9.272 10.692
2 1 2 29.9 30. 8.5 30. 0.162 30. 9.267 10.754
3 1 3 29.9 30. 8.7 30. 0.163 30. 9.283 10.862
4 1 4 30.0 30. 8.9 30. 0.167 30. 9.307 10.910
5 1 5 30.1 30. 9.0 30. 0.169 30. 9.254 10.883
6 1 6 30.3 30. 9.1 30. 0.165 30. 9.192 10.860
7 1 7 30.4 30. 9.1 30. 0.163 30. 9.151 10.781
8 1 8 30.5 30. 9.2 30. 0.162 30. 9.150 10.700
9 1 9 30.7 30. 9.2 29. 0.163 30. 9.175 10.553
10 1 10 30.6 29. 9.2 30. 0.161 30. 9.169 10.515
11 1 11 30.5 30. 9.2 30. 0.159 29. 9.088 10.526
12 1 12 30.5 29. 9.2 30. 0.159 29. 9.060 10.510
13 1 13 30.6 30. 9.2 30. 0.164 30. 9.054 10.527
14 1 14 30.7 30. 9.2 30. 0.167 30. 9.091 10.482
15 1 15 30.8 30. 9.2 30. 0.163 30. 9.099 10.437
16 1 16 30.8 30. 9.3 30. 0.170 30. 9.019 10.416
17 1 17 30.9 30. 9.5 30. 0.170 29. 8.984 10.363
18 1 18 31.0 30. 9.6 30. 0.168 30. 9.072 10.416
19 1 19 31.0 30. 9.6 30. 0.166 30. 9.081 10.394
20 1 20 31.1 30. 9.5 30. 0.164 30. 9.069 10.364
21 1 21 31.3 30. 9.6 30. 0.157 30. 9.060 10.333
22 1 22 31.4 30. 9.6 30. 0.156 30. 9.030 10.356
23 1 23 31.4 30. 9.6 30. 0.152 30. 9.107 10.441
24 1 24 31.5 30. 9.6 30. 0.152 30. 9.085 10.460
25 1 25 31.6 30. 9.5 30. 0.149 29. 9.091 10.539
26 1 26 31.7 30. 9.4 30. 0.148 28. 9.115 10.636
27 1 27 31.7 30. 9.4 30. 0.148 29. 9.086 10.706
28 1 28 31.8 30. 9.4 30. 0.148 30. 9.043 10.677
29 1 29 31.8 30. 9.4 30. 0.148 29. 8.963 10.579
30 1 30 31.8 30. 9.4 30. 0.149 30. 8.882 10.522
31 1 31 31.9 30. 9.4 30. 0.154 30. 8.842 10.486

So much for the "mid-January dry spell." The 16th and 17th are the highest dates :lol: , which has about as much significance as the random deviation of the past 8 years.

 

[Marc_C]

This is so frustrating. Tony, STOP LOOKING AT AVERAGES!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! THEY DON"T MATTER IN THIS DISCUSSION!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

If we're going to have a major dry spell with inversion during any given winter, it's more likely to happen in January, and it's going to tend to happen mid-month-ish, as it typically does most years. One year it might be from the 5th to the 16th (and we get hit with a 48" storm on the 17th). The next year it may be from the 15th to the 25th, right after the 60" cycle that ended on the 14th. There may even be two 8 or however many day events in January, separated by a few days of storm. Averages don't show this.

Your 30 years of *average* snowfall data is WORTHLESS in this case.

 

[Tony Crocker]

This is so frustrating.

Yes it is. The inversions are a subset of the dry spells. Cold temperatures are a key factor in the inversions, so I accept that the inversions are more common in January. I do NOT accept that the dry spells are more common in January.
1) Since the January average is the same (day by day per the 1971-2000 data as well as for the month as a whole), for there to be more dry spells there must also be more huge dumps as an offset. Any meteorological explanation for that?
2) Alta standard deviations for January are higher than for February/March but lower than for December.

One year it might be from the 5th to the 16th (and we get hit with a 48" storm on the 17th). The next year it may be from the 15th to the 25th, right after the 60" cycle that ended on the 14th. There may even be two 8 or however many day events in January, separated by a few days of storm.

You'll find the same pseudo patterns of dry spells in other months. Like during my Iron Blosam week from 1996-2008 for instance. :lol:
One week or longer periods of no snow are a fact of life in western ski resorts, even in elite microclimates like LCC. Recall that 1/4 of Alta weeks between Dec. 1 and Mar. 31 have NO days with 6+ inches new snow. With that level of frequency it is not surprising that it can happen for several years during the same week, creating a spurious pattern in the eye of the beholder. It's like the Super Bowl and the stock market, an interesting observation with no predictive value whatsoever.

I suppose it might be possible to dig up daily SNOTEL data, if that's what's necessary to lay this myth to rest.

 

[Admin]

This is so frustrating.

Yes it is. The inversions are a subset of the dry spells. Cold temperatures are a key factor in the inversions, so I accept that the inversions are more common in January. I do NOT accept that the dry spells are more common in January.

:-?

You might wish to reexamine your logic on that one, Sparky.

Sent from my Samsung Galaxy Note II using Tapatalk 2

 

[Tony Crocker]

The SilverLake Brighton data by date did not have standard deviations for precipitation but it did have standard deviations for temperature max and min. I'm fairly sure that dry winter days in the mountains have a wider temperature range than snowy days. So on the dates where dry spells are more common shouldn't we see higher temperature standard deviations?

The 30 year SilverLake Brighton data is there for every date of the year: http://www.wrcc.dri.edu/cgi-bin/cliNORM2000t.pl?utsilv. Temperature deviations are higher when it's cold than when it's warm. For max temps the SD is highest from late September to late January. For min temps the SD is highest from late November to late February. Like most weather data by date, all of these change gradually in smooth curves, with no short period of a week or two sticking out as an anomaly.

Raw daily data from WRCC requires permission to download. I've left a message. WRCC has a big data set for Alta too: http://www.wrcc.dri.edu/cgi-bin/cliMAIN.pl?ut0072

 

[Admin]

The SilverLake Brighton data by date did not have standard deviations for precipitation but it did have standard deviations for temperature max and min. I'm fairly sure that dry winter days in the mountains have a wider temperature range than snowy days. So on the dates where dry spells are more common shouldn't we see higher temperature standard deviations? The SilverLake Brighton data is there for every date of the year: http://www.wrcc.dri.edu/cgi-bin/cliNORM2000t.pl?utsilv .

Just when I can't think that you couldn't stretch credulity any further you go ahead and exceed my expectations one again! Maybe you could find an analogy between Utah inversions and sunspot activity visible by telescope on clear days from Botswana?

Unbelievable. Just unbelievable.


Sent from my Samsung Galaxy Note II using Tapatalk 2

 

[Marc_C]

I think we should start counting rings on wooly mammoths or something and correlate that with the average number of seeds in a lemon.

Kinda reminds me of this chart:

 

[Tony Crocker]

STOP LOOKING AT AVERAGES

Perhaps this will help:

There's 50+ years of Alta data behind these graphs. Dry spells should show up in the probability of at least .01 inch of precipitation. Big dumps should show up in the probability of at least 1.00 inch of precipitation. I've thrown in thresholds of .25 and .50 inch days as well.

The raw daily data is still volatile as shown in the first graph. WRCC gives you an option to smooth the data by any number of days. The second graph uses one week of smoothing. The "mid-January dry spell" should show up in that graph, which it does not. Feel free to try other options. http://www.wrcc.dri.edu/cgi-bin/PCPNqty_form.pl?ut0072

Do you Utah folk trust personal memory over documented evidence in other areas? Just wondering. Sometimes these arguments remind me of the, to oversimplify it, "Moneyball" fights in baseball: the old-school/"see it with my eyes" types vs. the new-school/have video & spreadsheet proof.

A rather apt analogy. I thought the 30-year averages by date at SilverLake Brighton would be convincing. We'll see if this explicit count of days with no new snow does any better.

 

[Admin]

Marc_C, give it up. I know that I have. Tony apparently only lives in a world of averages.

Sent from my Samsung Galaxy Note II using Tapatalk 2

 

[markiel]

I have the Alta daily data set...

1905-2008...

 

[socal]

Marc_C, give it up. I know that I have. Tony apparently only lives in a world of averages.

Sent from my Samsung Galaxy Note II using Tapatalk 2

Well yeah, but that's kinda a different point.... But isn't it hard to argue with the fact that the average precip on any given day of January for the past 30 years (and especially the middle days) is basically just as likely as any other day during the winter to have snow?

It's always nice to prove him wrong but the 30 years of data kinda speaks for itself, no? Isn't it just a fact that Jan doesn't historically have the dry period that it "seems" like it does?

 

[Admin]

Crocker himself agrees that the propensity for a persistent inversion is greatest in January... which is my premise that somehow started all of this bullshit.

Sent from my Samsung Galaxy Note II using Tapatalk 2

 

[SoCal Rider]

Look at that: You made a Utahan swear. :shock:

 

[Admin]

Utahan by way of New York. I could make a longshoreman blush.

Sent from my Samsung Galaxy Note II using Tapatalk 2

 

[TRam]

In other news did Devils Castle open today? It shows open now on the website. Anyone up there?