Rough Guide To Snow Level in the PNW

[IbrChris]

Keyword is "rough guide"...there isn't a definitive set of criteria for determining whether it will snow in the lowlands, although we can arrive at guidelines for a rough 50/50 odds of snow vs rain. While 0c H8 temp (850 mb temp) and 540 thickness are generally the rough values for areas east of the Rockies due to the maritime influence dominant in the PNW those same values (0c and 540) generally equate to a snow level around 3,000 feet in the Cascades and temps closer to 45-50 along the I-5 corridor.

 

Let's examine some predictors:

• Surface temp: If <= 0c either freezing rain, sleet or snow is likely, though if the sub-freezing layer at the surface is very shallow and/or WAA (warm air advection) is occurring either freezing rain or even plain rain are likely. The surface temp alone isn't a great predictor of snow vs rain, but is a good predictor of wintry precip vs rain.
• 925 mb temp: A better measure of snow vs rain or freezing rain because the warm layer in an overrunning/warm front event is often noticeable at 925 mb (2500'). If this layer is above 0c snow can generally be ruled out and freezing rain as well unless the surface temp is at or below freezing. In a moist adiabatic or isothermal profile a 925 mb temp of 0 to -1c can be enough for snow down to the hills (500-1000 feet) and even flakes or a mix below that.
• 850 mb temp: Similar to 925 mb temp, generally if both 925 mb temp and 850 mb temp are above freezing any surface cold air is very shallow and the only potential is for freezing rain. In an isothermal or inverted layer, the 850 mb temp can be 0c and snow can still occur assuming lower levels are at or below freezing. This isn't a common scenario in western WA/OR except during periods of easterly/offshore flow near the Cascade gaps and the Gorge. In a moist adiabatic profile (usually present with some component of onshore flow) generally a 850 mb temp of at least -8c is necessary for snow down to the surface (below 500 feet).
• 850 mb thickness: Less commonly utilized (but arguably superior) metric in determining precip type at low elevations. 1290-1292 meters indicates the entire 1000-850 mb layer is at or below freezing and suggests a 90% chance of frozen precip type (snow generally). 1300 is a typical rain/snow guess east of the Rockies.
• 500 mb thickness: Probably the worst metric for determining precip type, but the "first guess" for meteorologists east of the Rockies. It's fairly useless over and west of the Rockies. Studies on precip type vs 500 mb thickness in the UK have established a rough 50/50 rain/snow value at around 522 at coastal locations near sea level in onshore flow scenarios. A rough 50/50 value of 526 based on observations at Portland, OR.
• Other considerations: Modeled wet bulb zero elevation (either in terms of pressure level or in meters/feet). Not always available or easily determined from model output. If available a very useful value along with 850 mb thickness.

Surface temp should always be considered, but when model values are utilized in my opinion the order of importance is:

 

1. 850 mb thickness: look for values at or below 1300 meters south and west of your location, or at a lower elevation than your location.

2. 925 mb temp: look for values at or below 0c, preferably -1 or -2c for better odds of accumulation. Beware of 925 mb temps even a degree above freezing.

3. 850 mb temp: look for values at or below -8c unless offshore flow is present and surface dewpoint depressions will allow for evaporative cooling.

4. 500 mb thickness: if this is all you have look for values at or below 525 or 526 (usually the maps are in increments of 3 dam or 30 meters so ideally you would look for 525 and 522 lines). This isn't a guarantee of snow but values above 526 are fairly unlikely to yield snow in the lowlands without offshore flow.

 

Rough guide to snow level based on 500 mb thickness and 850 mb temps, assuming moist adiabatic thermal profile between the surface and 700 mb

 

Sea level (coastal): 524 and -9c

Sea level (inland): 526 and -8c

500': 528 and -7c

1000': 530 and -6c

2000': 534 and -4c

3000': 538 and -2c

4000': 542 and 0c

5000': 546 and +2c

6000': 550 and +4c

 

Note: NWS Salt Lake gives a critical 500 mb thickness of 543 at Salt Lake City (elev 4,200') which suggests the above list is fairly accurate at mid-slope level in the Cascades. Marine influence is most pronounced in the lower levels.

[IbrChris]

An example of a 850 mb thickness map (1000-850 mb thickness) from today's 12z GFS run. Note the black 130 line which is traditional rain/snow line. We would prefer to see the 129 line over the I-5 corridor to be guaranteed snow. This map indicates Bellingham and probably Seattle would be good candidates for snow if moisture is present.

 

[IbrChris]

500 mb thickness for the same hour on the 12z GFS...thickness along the I-5 corridor is only in the neighborhood of 537-538. So we can see the bulk of the cold air (in terms of anomalies) is in the lower levels. Since this is way out there and in the GFS lower resolution period the 850 mb thickness is probably too low since the model doesn't incorporate terrain as well at lower resolutions. Based on 500 mb thickness and 850 mb thickness I'd say it's a very borderline situation, but the potential would be there for snow or ZR especially near the Gorge.



 

[IbrChris]

Looking at additional parameters at hour 300 we find a modeled 925 mb temp of -2c and 850 mb temp of -4c...again low level cold is exaggerated in the low-resolution extended range but these parameters would suggest snow potential at face value when considered from most important to least important:
 

• 850 mb thickness: ~1295 (129-130 dam) inconclusive
• 925 mb temp: -2c snow
• 850 mb temp: -4c rain normally but given the offshore flow snow is possible
• 500 mb thickness: ~5380 (538 dam) rain/mix

[IbrChris]

Here's another interesting and pertinent map...maximum temperature in the surface to 500 mb layer which is about 1 or 2c at PDX. A look at surface temps confirms this maximum occurs at the surface (2 meters). If the value is below 0c it means the entire lower atmosphere column is below freezing and snow is the only possible precip type.

[SilverFallsAndrew]

I have found the 500-1000mb thickness value to be the most accurate indicator of snow at my location. I have seen it rain with -6 or 7C 850mb temps if the precip is light and thickness is in the 530s. On the other hand I have seen it snow and accumulate with 850s around -3C but 528 or lower thickness. 

 

I do not have enough knowledge to know why that would be, but I actually do not really pay attention to the 850mb temps anymore when looking to see if a pattern could produce snow. 

[IbrChris]

I have found the 500-1000mb thickness value to be the most accurate indicator of snow at my location. I have seen it rain with -6 or 7C 850mb temps if the precip is light and thickness is in the 530s. On the other hand I have seen it snow and accumulate with 850s around -3C but 528 or lower thickness. 

 

I do not have enough knowledge to know why that would be, but I actually do not really pay attention to the 850mb temps anymore when looking to see if a pattern could produce snow. 

 

That is odd...my only explanation would be that because you're at 1,600' the low levels (below 925 mb or so) are not particularly relevant. I'm surprised 850 mb temp doesn't seem to be relevant either, although a 50/50 probability leaves room for odd "tail" events.

 

I am surprised that -7c 850 mb temp doesn't pretty much guarantee snow there given the moist adiabatic lapse rate would yield about -2c at your location, though the dry adiabatic rate would yield +3c, dry meaning any situation where T =/= Td for a parcel.

 

Near guarantees for snow at your location would still seem to be situations where both 925 and 850 mb temps are -1c or lower and where 850 mb thickness is 130 or lower. The higher up you go the more accurate 500 mb thickness gets however, because the mean temp in the layer tends to more closely approximate the temperature within the precipitating layer. I'd still argue 850 mb thickness is a more accurate value for snowfall prediction even at your elevation.

[SilverFallsAndrew]

That is odd...my only explanation would be that because you're at 1,600' the low levels (below 925 mb or so) are not particularly relevant. I'm surprised 850 mb temp doesn't seem to be relevant either, although a 50/50 probability leaves room for odd "tail" events.

 

I am surprised that -7c 850 mb temp doesn't pretty much guarantee snow there given the moist adiabatic lapse rate would yield about -2c at your location, though the dry adiabatic rate would yield +3c, dry meaning any situation where T =/= Td for a parcel.

 

Near guarantees for snow at your location would still seem to be situations where both 925 and 850 mb temps are -1c or lower and where 850 mb thickness is 130 or lower. The higher up you go the more accurate 500 mb thickness gets however, because the mean temp in the layer tends to more closely approximate the temperature within the precipitating layer. I'd still argue 850 mb thickness is a more accurate value for snowfall prediction even at your elevation.

 

Yes, I have never seen it rain with 925mb temps -1C or lower. I have had it snow with 925mb temps as high as 2C, I assume due to precip rates. Also in overrunning situations my location will scour out even before Salem as a warm tongue of air usually noses up the foothills. This was very pronounced in February 2014 when Salem stayed snow about 24 hours after my location had turned to rain. 

[Heavy Snow]

Keyword is "rough guide"...there isn't a definitive set of criteria for determining whether it will snow in the lowlands, although we can arrive at guidelines for a rough 50/50 odds of snow vs rain. While 0c H8 temp (850 mb temp) and 540 thickness are generally the rough values for areas east of the Rockies due to the maritime influence dominant in the PNW those same values (0c and 540) generally equate to a snow level around 3,000 feet in the Cascades and temps closer to 45-50 along the I-5 corridor.

 

Let's examine some predictors:

• Surface temp: If
• 925 mb temp: A better measure of snow vs rain or freezing rain because the warm layer in an overrunning/warm front event is often noticeable at 925 mb (2500'). If this layer is above 0c snow can generally be ruled out and freezing rain as well unless the surface temp is at or below freezing. In a moist adiabatic or isothermal profile a 925 mb temp of 0 to -1c can be enough for snow down to the hills (500-1000 feet) and even flakes or a mix below that.
• 850 mb temp: Similar to 925 mb temp, generally if both 925 mb temp and 850 mb temp are above freezing any surface cold air is very shallow and the only potential is for freezing rain. In an isothermal or inverted layer, the 850 mb temp can be 0c and snow can still occur assuming lower levels are at or below freezing. This isn't a common scenario in western WA/OR except during periods of easterly/offshore flow near the Cascade gaps and the Gorge. In a moist adiabatic profile (usually present with some component of onshore flow) generally a 850 mb temp of at least -8c is necessary for snow down to the surface (below 500 feet).
• 850 mb thickness: Less commonly utilized (but arguably superior) metric in determining precip type at low elevations. 1290-1292 meters indicates the entire 1000-850 mb layer is at or below freezing and suggests a 90% chance of frozen precip type (snow generally). 1300 is a typical rain/snow guess east of the Rockies.
• 500 mb thickness: Probably the worst metric for determining precip type, but the "first guess" for meteorologists east of the Rockies. It's fairly useless over and west of the Rockies. Studies on precip type vs 500 mb thickness in the UK have established a rough 50/50 rain/snow value at around 522 at coastal locations near sea level in onshore flow scenarios. A rough 50/50 value of 526 based on observations at Portland, OR.
• Other considerations: Modeled wet bulb zero elevation (either in terms of pressure level or in meters/feet). Not always available or easily determined from model output. If available a very useful value along with 850 mb thickness.

Surface temp should always be considered, but when model values are utilized in my opinion the order of importance is:

 

1. 850 mb thickness: look for values at or below 1300 meters south and west of your location, or at a lower elevation than your location.

2. 925 mb temp: look for values at or below 0c, preferably -1 or -2c for better odds of accumulation. Beware of 925 mb temps even a degree above freezing.

3. 850 mb temp: look for values at or below -8c unless offshore flow is present and surface dewpoint depressions will allow for evaporative cooling.

4. 500 mb thickness: if this is all you have look for values at or below 525 or 526 (usually the maps are in increments of 3 dam or 30 meters). This isn't a guarantee of snow but values above 526 are fairly unlikely to yield snow in the lowlands.

 

Rough guide to snow level based on 500 mb thickness and 850 mb temps, assuming moist adiabatic thermal profile between the surface and 700 mb

 

Sea level (coastal): 524 and -9c

Sea level (inland): 526 and -8c

500': 528 and -7c

1000': 530 and -6c

2000': 534 and -4c

3000': 538 and -2c

4000': 542 and 0c

5000': 546 and +2c

6000': 550 and +4c

 

 

Thanks for sharing.

[Sometimesdylan]

Lots of good information. 

 

I use mostly meteostar GFS for getting the information, wish they had a 850mb thickness. I have been noticing a lot lately that the 500mb thickness has been letting me down a lot lately. 

 

Will be watching the 925mb temps more closely now. 

Thank you guys.

[Deweydog]

I have found the 500-1000mb thickness value to be the most accurate indicator of snow at my location. I have seen it rain with -6 or 7C 850mb temps if the precip is light and thickness is in the 530s. On the other hand I have seen it snow and accumulate with 850s around -3C but 528 or lower thickness.

 

I do not have enough knowledge to know why that would be, but I actually do not really pay attention to the 850mb temps anymore when looking to see if a pattern could produce snow.

I too much prefer to use 500-1000mb thicknesses as a guide during cold advection. 520dm is typically a fool-proof cut off for sticking snow here with any kind of decent precip rates. 850mb temps typically vary at least a little with each set up. I suppose the advantage to using thicknesses is it's a representation of about an 18,000 foot column of air as opposed to a specific cross section.

[bainbridgekid]

The atmosphere is obviously far too complex to use any one magic bullet, but I have found 925mb temps to be the most consistent indicator. Especially on Bainbridge during cold air damming events.

 

I've always lived about 300-400 feet, but my experience has been:

 

Above 0 = No frozen precip

0 to -1 = Cold rain or a mix with onshore flow and wet snow with offshore flow or in the CZ.

-1 to -2 = nonsticking snow with onshore flow and sticking snow with offshore flow or in the CZ.

Anything below -2 is usually sticking snow regardless of the wind.

[BLI snowman]

Ana-fronts are my favorite wildcard snow producer in the PNW. The stalled cold front boundary on March 21, 2012 was fascinating to watch. Eugene saw 7" of snow with 1000-500mb thickness around 538m and 850mb temps around -2c (925mb temps around 0c).

 

http://mp1.met.psu.edu/~fxg1/NARR/2012/us0321.php

 

The same exact setup (displaced westward) had occurred shortly before that on March 12-13, 2012, which produced the historic snowstorm at the OR coast.

 

http://mp1.met.psu.edu/~fxg1/NARR/2012/us0312.php

 

November 19, 2003 and January 1-2, 2009 are a couple of other recent examples of this pattern, but the March 2012 storms were the most dramatic. 

[IbrChris]

I too much prefer to use 500-1000mb thicknesses as a guide during cold advection. 520dm is typically a fool-proof cut off for sticking snow here with any kind of decent precip rates. 850mb temps typically vary at least a little with each set up. I suppose the advantage to using thicknesses is it's a representation of about an 18,000 foot column of air as opposed to a specific cross section.

 

The problem is 1000-500 mb thickness won't resolve EWLs (elevated warm layers) which are common around here in overrunning events, generally around 900 mb. Since 700-500 mb layer is almost invariably sub-freezing and rarely is there an above-freezing layer at 850-700 mb in marginal snow situations west of the Cascades the primary consideration is the surface to 850 mb layer, 850 mb thickness below 1290 implies the entire layer is at or below 0c. 925 mb temps sample a critical portion of that layer where a warm tongue often exists.

 

Case in point a 522 dam 500 mb thickness can have a EWL at 900 mb perhaps shown by a 925 mb temp of 0c or even +1c. Result will be rain or ZR.

 

A 537 dam thickness can have no EWL with 925 mb temp of -2c and 850 mb temp of -5c. Result will be snow. The sfc-500 mb layer in the latter case has a much lower lapse rate (closer to isothermal) so the mean virtual temperature of the layer is warmer (and hence the thickness is higher).

 

http://www.wpc.ncep.noaa.gov/research/snowfcst/snow.pdf

[IbrChris]

The ideal method is analysis of forecast soundings but the Euro for instance doesn't provide that product (at least not on WxBell or other vendors I've utilized). GFS has forecast soundings available.

[IbrChris]

The atmosphere is obviously far too complex to use any one magic bullet, but I have found 925mb temps to be the most consistent indicator. Especially on Bainbridge during cold air damming events.

 

I've always lived about 300-400 feet, but my experience has been:

 

Above 0 = No frozen precip

0 to -1 = Cold rain or a mix with onshore flow and wet snow with offshore flow or in the CZ.

-1 to -2 = nonsticking snow with onshore flow and sticking snow with offshore flow or in the CZ.

Anything below -2 is usually sticking snow regardless of the wind.

Yeah this seems pretty spot-on for my location as well. -1c with showery precip can bring the snow level down to near sea level, however with light/moderate stratiform it won't, -2c usually delivers for everyone.

[IbrChris]

On Dec 31, 1968 Portland had 19/11 with 1.11" of precip in what is probably the only case of a day with a high below 20 and over an inch of precip in its period of record. Much of it was snow, though it appears areas further south toward Salem saw some freezing rain or sleet.

The 12z sounding on Dec 31st is perhaps one of the strangest I've seen here (temp and pressure level):

 

700 mb: -2.1c

750 mb: 0.2c

800 mb: 0.0c

850 mb: -0.3c

900 mb: 2.4c (EWL)

950 mb: -3.5c

1000 mb: -13.4c
1015 mb: -12.8c

 

Pretty extreme inversion...it was mid 30s at 3000' and 9 at the surface.

This is a case where 500 mb thickness was likely well over 530 dam due to deep-layer WAA between 700 and 900 mb.


 

[IbrChris]

Two interesting studies in the UK utilizing 500 mb thickness for rain vs snow near sea-level:

Probability of Snow vs Thickness from Boyden

90%: 518
70%: 524
50%: 526

30%: 529

10%: 533

 

Murray (1959)

Rain is rare: 519
50/50 rain/snow: 523
Snow is rare: 530

Boyden also utilized 850 mb thickness:

Probability of snow
90%: 1279

70%: 1287
50%: 1293
30%: 1297
10%: 1302

 

[IbrChris]

A comparison of critical thickness by elevation

 

[IbrChris]

[BLI snowman]

On Dec 31, 1968 Portland had 19/11 with 1.11" of precip in what is probably the only case of a day with a high below 20 and over an inch of precip in its period of record. Much of it was snow, though it appears areas further south toward Salem saw some freezing rain or sleet.

 

The 12z sounding on Dec 31st is perhaps one of the strangest I've seen here (temp and pressure level):

 

700 mb: -2.1c

750 mb: 0.2c

800 mb: 0.0c

850 mb: -0.3c

900 mb: 2.4c (EWL)

950 mb: -3.5c

1000 mb: -13.4c

1015 mb: -12.8c

 

Pretty extreme inversion...it was mid 30s at 3000' and 9 at the surface.

 

This is a case where 500 mb thickness was likely well over 530 dam due to deep-layer WAA between 700 and 900 mb.

 

 

 

 

That was probably the most amazing low level cold event through the gorge on record.

 

And actually most of Portland's precip on 12/31/1968 was sleet. Only 3.4" of snow was measured that day at PDX and news reports that I've read indicate that the snow transitioned to sleet (then zr) pretty quickly that day, after the heavy snowfalls on the previous few days. 

[SilverFallsAndrew]

Ana-fronts are my favorite wildcard snow producer in the PNW. The stalled cold front boundary on March 21, 2012 was fascinating to watch. Eugene saw 7" of snow with 1000-500mb thickness around 538m and 850mb temps around -2c (925mb temps around 0c).

 

http://mp1.met.psu.edu/~fxg1/NARR/2012/us0321.php

 

The same exact setup (displaced westward) had occurred shortly before that on March 12-13, 2012, which produced the historic snowstorm at the OR coast.

 

http://mp1.met.psu.edu/~fxg1/NARR/2012/us0312.php

 

November 19, 2003 and January 1-2, 2009 are a couple of other recent examples of this pattern, but the March 2012 storms were the most dramatic. 

 

Both of those were huge snow producers at my location. I got 10" with the March 12-13th storm and then just under 20" with the one on the 21st. What was interesting though was that during the middle of the snow event on the 21st the band of heavy precip pivoted during mid-afternoon and was aligned over the Willamette Valley and my foothills location was just getting light precip. The precip changed to rain and the temp shot up to 37 degrees, meanwhile down in Silverton on the valley floor it was snowing heavily and 33. Then around 6pm the heavy precip moved back in and everything changed back to snow...It rained for about 4 hours during the middle of that snow event, if it had been snow the whole time I could have gotten up around 25" or so probably. 

[SilverFallsAndrew]

On Dec 31, 1968 Portland had 19/11 with 1.11" of precip in what is probably the only case of a day with a high below 20 and over an inch of precip in its period of record. Much of it was snow, though it appears areas further south toward Salem saw some freezing rain or sleet.

 

The 12z sounding on Dec 31st is perhaps one of the strangest I've seen here (temp and pressure level):

 

700 mb: -2.1c

750 mb: 0.2c

800 mb: 0.0c

850 mb: -0.3c

900 mb: 2.4c (EWL)

950 mb: -3.5c

1000 mb: -13.4c

1015 mb: -12.8c

 

Pretty extreme inversion...it was mid 30s at 3000' and 9 at the surface.

 

This is a case where 500 mb thickness was likely well over 530 dam due to deep-layer WAA between 700 and 900 mb.

 

 

 

 

Such a powerful low level cold airmass...But such a low level airmass. 

 

December 31st

 

Silverton: 15/7

Silver Falls: 33/12

[BLI snowman]

Both of those were huge snow producers at my location. I got 10" with the March 12-13th storm and then just under 20" with the one on the 21st. What was interesting though was that during the middle of the snow event on the 21st the band of heavy precip pivoted during mid-afternoon and was aligned over the Willamette Valley and my foothills location was just getting light precip. The precip changed to rain and the temp shot up to 37 degrees, meanwhile down in Silverton on the valley floor it was snowing heavily and 33. Then around 6pm the heavy precip moved back in and everything changed back to snow...It rained for about 4 hours during the middle of that snow event, if it had been snow the whole time I could have gotten up around 25" or so probably. 

 

Yeah, the fact that the front initially pushed SE, then pivoted back to the NW (giving the Portland area a widespread snowfall) on the evening of the 21st, before being pushed east again is part of what made that event so fascinating to watch. I can't recall similar movement, or lack thereof, with such a well defined Pacific front before. The fact that it was an ana-front (the precipitation placed behind the boundary) made it all the more interesting.

 

Worth noting that that event was during the peak of the record-shattering March 2012 heat event across North America. It took a pretty crazy upper level pattern to cause that type of action in our region.

[SilverFallsAndrew]

Yeah, the fact that the front initially pushed SE, then pivoted back to the NW (giving the Portland area a widespread snowfall) on the evening of the 21st, before being pushed east again is part of what made that event so fascinating to watch. I can't recall similar movement, or lack thereof, with such a well defined Pacific front before. The fact that it was an ana-front (the precipitation placed behind the boundary) made it all the more interesting.

 

Worth noting that that event was during the peak of the record-shattering March 2012 heat event across North America. It took a pretty crazy upper level pattern to cause that type of action in our region.

 

I have lived 5 full winters now at my location and March 2012 was the snowiest month by far I have had with 40" of snow. I also recorded 20.55" of precip which is 2nd only to December 2015. Between February 29th and March 22nd I had 3 storms which delivered storm totals over over 10". All 3 were similar in terms of being ana-fronts. The one on February 29-March 1st did deliver T-2" amounts across much of the Willamette Valley as well, with 16" total at my location. 

[Deweydog]

The problem is 1000-500 mb thickness won't resolve EWLs (elevated warm layers) which are common around here in overrunning events, generally around 900 mb. Since 700-500 mb layer is almost invariably sub-freezing and rarely is there an above-freezing layer at 850-700 mb in marginal snow situations west of the Cascades the primary consideration is the surface to 850 mb layer, 850 mb thickness below 1290 implies the entire layer is at or below 0c. 925 mb temps sample a critical portion of that layer where a warm tongue often exists.

 

Case in point a 522 dam 500 mb thickness can have a EWL at 900 mb perhaps shown by a 925 mb temp of 0c or even +1c. Result will be rain or ZR.

 

A 537 dam thickness can have no EWL with 925 mb temp of -2c and 850 mb temp of -5c. Result will be snow. The sfc-500 mb layer in the latter case has a much lower lapse rate (closer to isothermal) so the mean virtual temperature of the layer is warmer (and hence the thickness is higher).

 

http://www.wpc.ncep.noaa.gov/research/snowfcst/snow.pdf

 

That's why I specified their use during only free-air, cold advection situations.  Obviously 500-1000mb thicknesses become pretty irrelevant in overrunning situations.  

[IbrChris]

That's why I specified their use during only free-air, cold advection situations. Obviously 500-1000mb thicknesses become pretty irrelevant in overrunning situations.

Yes definitely more representative in situations where the lower levels are stable (neither inverted nor conditionally unstable). A steep lapse rate will produce a fairly low 500 mb thickness due to cold mid levels, even though the sounding will be too warm for snow at the surface.

[IbrChris]

I've started going through SLE soundings where snow was reported within an hour of the balloon launch (so at 11z and 23z) to determine the mean thicknesses and 925/850 temps for snow events at Salem. So far with 25 occurrences of snow during a balloon launch I have a mean 500 mb thickness of 5258 or 526 dam. A mean 850 mb thickness of 1284, obviously skewed downward a bit by the presence of some cold events like 2014, 2008, 1998, 1990.

[bainbridgekid]

I've started going through SLE soundings where snow was reported within an hour of the balloon launch (so at 11z and 23z) to determine the mean thicknesses and 925/850 temps for snow events at Salem. So far with 25 occurrences of snow during a balloon launch I have a mean 500 mb thickness of 5258 or 526 dam. A mean 850 mb thickness of 1284, obviously skewed downward a bit by the presence of some cold events like 2014, 2008, 1998, 1990.

Interesting.

 

Though like you say, the mean isn't a good indicator of the situation around the snow level since there are plenty of time they got snow and the snow level would have been below Salem and none where the snow level was above it.

[IbrChris]

Interesting.

 

Though like you say, the mean isn't a good indicator of the situation around the snow level since there are plenty of time they got snow and the snow level would have been below Salem and none where the snow level was above it.

Probably something between +1 SD and maximum is more reflective of snow level...maximum is an extreme case so not representative and +1 SD would basically mean 68% of the events occurred with lower thicknesses and 32% with higher...a much better (but still not fully ideal) depiction of what constitutes critical thickness at Salem.