Saturday, January 14, 2017

A wintry mess before the very wet and milder arrives next week?

So as we mentioned in our previous blog post, we are expecting to finally break out of this latest cold and dry weather pattern by early next week. So what is going to be responsible for this change? Here is a look at the current weather maps which are supported by what we see on satellite (a good sign if we want to follow a particular weather model):

500 mb map (left) with Integrated Vapor Transport (shows atmospheric rivers denoted by bright colors). Water vapor satellite imagery  on right. 
The most noticeable things on the left chart are the large high pressure ridge centered over the Inland Northwest (Large H) as well as the atmospheric river developing north and west of Hawaii. Over the next several days we expect the presence of the high to persist over our area. This means more fog, low clouds, and very cold temperatures in the valleys with light winds. This is because high pressure systems in the winter typically lead to strong inversions (temperature increasing with height). This is especially true when there is widespread snow cover over an area and clearing skies. This is certainly the case right now as we have seen many mornings recently with sub-zero temperatures and mainly clear skies. Here is a look at just how strong the inversion was this morning at our office.

1/14/17 4am weather sounding at NWS Spokane. The red line shows how the temperature varied with elevation. The farther to the right the red line is the warmer the atmosphere is. Notice how the temperature increases all the way from the ground up to 4700'. That is what we would call a strong inversion. 
So why are we telling you about this inversion? Because what happens between now and Monday night will be critical into producing an accurate weather forecast. But more on that later. So first, what do we expect to happen to that atmospheric river forming north of Hawaii? By Monday night here is what most of the weather models are showing.
Atmospheric river position 10 pm Monday 1/16/17
According to this we should see the ridge being rapidly replaced by the atmospheric river moving in from the Pacific. Not only is this air stream moist, but it will be warm as well. So that should mean we would see a rapid transition from the current chilly air mass to a much milder one, right?. In other words we should see any threat of snow changing to rain and probably in a short amount of time. But if only it were that easy! This upcoming system will likely provide us with all sorts of weather conundrums simply based on the inversions we showed you earlier.   Here is an example from Moses Lake.

Weather Sounding from Moses Lake 4am Tuesday with the freezing line denoted in pink.
So as the precipitation arrives it could begin as all snow as the entire atmosphere will have temperatures below freezing. But as the atmospheric river comes into the area (being ushered in by southwest winds of 30-40 mph 3000-10,000 ft above the ground) the temperatures above the ground will warm faster than those near the ground and this will create what we call a melting layer or warm nose of air aloft. What this means is any snow which forms in upper portions of the atmosphere will melt as it traverses the melting layer. If the melting layer is deep enough it will melt anything which falls into it. If it's not deep enough it will only partially melt the snow falling into it. In this case we are looking at a melting layer around 4000-5000' thick which should be plenty deep to melt most of the snow falling through it. That's not the whole equation though. What happens after that depends on how deep the freezing layer near the ground is. For the sounding above the freezing air near the ground is around 1500' thick so the question is "is this deep enough to completely freeze what melted before hitting the ground"? If the answer is yes the melted snow would completely refreeze producing sleet or ice pellets. If its not deep enough the melted snow will not refreeze and instead fall as a very cold rain which can easily refreeze once it contact a sub-freezing object on the ground. Here is a nice weather schematic the NWS office in Omaha created that shows the complexity of forecasting winter weather precipitation types.

So what does the profile in the Moses Lake sounding favor? That is always a tough question to answer as any perturbation in the strength/depth of the above ground melting layer and near ground freezing layer has a direct impact on the precipitation type.  If nothing were to change in the models we think this would lead to freezing rain. However the no change argument isn't one that often appears as much as we would desire. Since this event is still a few days away there are numerous answers as to what might occur. While we are confident we will see precipitation, we are far less confident in what the prevailing type of precipitation will consist of. Each model is advertising a different scenario. When we see such great differences we sometimes like to defer to what we call an ensemble model solution. This is were we take the same initial model and introduce very small changes at the very beginning of the run. These small initial changes can lead to vastly different solutions as you get farther into the forecast. So here is a snapshot of the various precipitation chances associated with the incoming weather make during the Tuesday morning commute.
4-panel ensemble solution for the precipitation type on Tuesday morning. Upper left corner=Sleet chance, upper right=Snow chance, lower left=freezing rain chances, lower right=measurable precipitation chances

The image above is about as messy as a weather forecast can get around here. As for sleet or ice pellets (upper left), we'd expect to see the best chances to occur over the western Columbia Basin and into the northern valleys east of Omak. The snow chances (upper right panel) would be highest near the north Cascades, northeast Washington, and over the northern Panhandle. Meanwhile the threat of freezing rain looks most prevalent over the remainder of the the Inland Northwest. But that is what the latest ensemble model runs suggest and this could change significantly as the event nears. Previously we were expecting to see a lot more snow with a brief changeover to freezing rain and then all rain. If we see snow first the freezing rain impacts could be somewhat mitigated as freezing rain falling onto fresh snow has much less impact on driving conditions then freezing rain falling onto bare pavement. Stay tuned to the latest changes in this evolving portion of the forecast.

Eventually (how long is uncertain) the atmospheric river will bring the warm moist over the entire region and mix it down to the ground with the most of the snow or freezing rain chances remaining near the north Cascades or over the far northern valleys. Here is what the temperature profile looks like for Moses Lake at 10 am Tuesday. Notice it's entirely above freezing which means we would see all rain.

10am Tuesday temperature profile for Moses Lake
Once the lower atmosphere warms above freezing we expect it to remain above freezing through the end of the event.The end of the event will likely consist of a good amount of time as the atmospheric river remains fixed over the area. Here is the expected position on Wednesday afternoon and then on Thursday.
Atmospheric River position on Wednesday 1/18/19 4pm.
Atmospheric river position by Thursday morning 1/19/17

That means the atmospheric river will be over the region beginning Monday night and persisting into early Thursday. Needless to say that can result in a lot of precipitation. Just how much are we expecting? Here are four distinct model runs.

72 hour precipitation ending 4 pm Thursday
Surprisingly there is very good agreement for a forecast this far out. We are quite confident that the precipitation over the western Columbia Basin and in the lee of the Cascades is a little bit overdone, and near the Cascade Crest and over north Idaho it might be underdone. Nonetheless these are some very impressive precipitation amounts. Looks like the eastern third of Washington and the Idaho Panhandle should generally see anywhere from 1.50"-3.00 inches. If there were no snow on the ground this alone could produce some issues with low-land flooding and perhaps some ice jam flooding as the ice breaks up on area rivers. However that is not the case as we have an appreciable amount of snow on the ground. Here is a look at the approximate amount of water tied up in the snow pack.

That is quite a bit of water or potential runoff sitting on the ground and will only add to the runoff totals. Based on the forecast temperatures, we expect to see temperatures remaining above freezing from Tuesday through much of Thursday (including nighttime low temperatures). The above freezing temperatures will combine with breezy winds resulting in a great setup for appreciable melting. The best melting potential will occur the SE third of our forecast area with the least amount near the Cascades and far northern valleys. We still expect to see most if not all the snow to melt over the Palouse, LC Valley, and eastern Columbia Basin. So combining the expected rainfall with the water from the snow melt could equate to up to 2.50 to nearly 5.00 inches of water which could potentially runoff. This increases the odds for flooding appreciably. Here is a map of where we anticipate the greatest flooding problems to occur.

We expect most of the flooding issues will involve low-land flooding, ice jam flooding, and urban flooding (think snow clogged storm drains...clear them now if you can!). Small rivers and streams could also flood, especially over the Palouse and nearby areas. As of now we are not expecting any mainstem flooding, however that could change especially if we melt more snow than expected.

If we look into the longer range weather forecasts, we don't expect to totally dry out, but we will cool considerably. Not down to current levels, but certainly enough to refreeze what doesn't melt. The pristine snow cover we have over us will look completely different by next Friday and could feature hard snowpack and frozen slush.

Please stay tuned to the latest forecasts for this upcoming will likely be a messy situation and is certainly subject to change. Oh yes...the countdown to spring is now 64 days and counting.

Tuesday, January 10, 2017

Much warmer weather is expected next week after prolonged cold!

If you think this winter has a rather drastic change from the past several winters your senses are not deceiving you. When we began the month of November is seemed we would see yet another mild November, however as the calendars changed to December that notion came to an abrupt halt. The start of 2017 has continued that trend. Look at these statistics for Spokane and this will put things into perspective.

It certainly seems the mild winters of the past several years are a distant memory. This has been the snowiest winter since 2008-09 and the coldest since the early 90s. But the cold wasn't just in Spokane. Check out this list of sites that have seen their record coldest December-early January period on record (keep in mind some of these station records don't go back too far, see last column for details).

So we can see it's been cold and somewhat snowy. To see how snowy its been look at this comparison graphic of satellite pictures from early December vs. early January. 

Satellite comparison of snow cover via visible satellite. The January image suggests just about all of the Inland NW is covered with snow. 
This is quite a change. But its winter and it's not unusual to see widespread snow cover at the beginning of January across the entire Inland Northwest is it? Here is a look at the amount of water in the snow (Snow Water Equivalent- SWE) on each January 10th since 2008. 

Based on this, we actually saw the entire region covered by snow in 2016 as well. But this year and that were the only ones since 2008.  So if there is snow covering the entire region should this lead to concern? Well it depends. If we could melt all the snow at once it could cause a great concern due to runoff and potential flooding. This time of year it is improbable that we can melt all the snow over the mountains (where you can see the dark blue, purple, and red shading) as it doesn't get warm enough but we can melt some. But it is quite a bit easier to melt a significant (if not all the snow) in the valleys where we see the abundance of white and pale blue shading. Any area shaded in white has at least an inch of water tied up in the snow pack while the light blue shading represents values ranging from 2-4" of water. So what does it take to melt this? Obviously we need to see temperatures above freezing for a prolonged period. But that's not all. It also helps immensely to add wind into the equation. Warm winds and prolonged temperatures above freezing are an enemy of the snow and can cause it to quickly disappear. Imagine taking that 1 to 4" of water trapped in the snow and sending it into the rivers and streams. How about taking that same 1 to 4" of water over urban areas and sending it toward snow-clogged storm drains? That can cause quite a mess and obviously would lead to flooding issues. This scenario can be further complicated by rivers and streams choked with thick ice. 

So why are we telling you this? Well it just so happens we are looking at a dramatic change in the weather pattern early next week. Since December, we have been stuck in a persistent weather pattern featuring cold weather systems coming from our northwest. These are notoriously cold systems, and generally pose little threat from a hydrology standpoint. 
Mean 500 mb (18,000' above ground) since December. Note how the colors (or heights) buckle north into Alaska and then drop SE into the remainder of the county. This is indicative of cold NW flow. 

When we compare this weather pattern to what's normal for this time of year we see a clear signal. The blue shaded areas represent below normal heights which generally result in cold weather whereas the orange, yellow, and green shading represent warm conditions. 
500 mb height anomaly values. 
This is quite unusual. Typically we see some warmer systems invading from the southwest tied to a nice tap of relatively warm sub-tropical moisture. When these systems arrive they can bring a rapid warming in temperatures as well as significant precipitation and snow melt. Well after a long respite from such systems, we see hints of one on the horizon. Here is what many of the forecast models are hinting at for next week. 

Weather pattern for next Monday evening (500 mb heights in white, precipitable water is shaded)
What this reveals is our northwest flow of late will be replaced by southwest flow early next week. Of even more concern is the swath of green and blue shading headed into the Pacific Northwest. This indicates we will be subject to a very moist and warm air mass, which could do a very effective job of melting the low-level snow pack. Now if this feature were only going to be over us for a day it may not do much to the snow pack. However look at this map below for Wednesday. Not much difference is there?

Weather pattern for next Wednesday (500 mb heights in white, precipitable water is shaded)
So this at least indicates the potential for warm weather. Here are our forecast highs for next Wednesday. 
Forecast temperatures for next Wednesday
Notice most of the valley temperatures are above freezing. This will be good for potentially melting the snow, especially over the SE portions of Washington and NC Idaho where you see the greens (temperatures above 45°F) during the day. But what we aren't showing here is many locations across the same area will also see nighttime temperatures remaining above freezing. So its conceivable that some locations across SE Washington and NC Idaho could see temperatures above freezing anywhere from 48-72 hours. We will also see robust south to southwest winds during portions of this period. Below is the wind forecast for Tuesday. These are sustained winds. The gusts will likely range from 20-35 mph. 

Tuesday wind forecast
So remember earlier when we said about above freezing temperatures and good winds can result in some rapidly melting snow. It's conceivable that a good portion (if not most) of the snow over the Palouse, southern Columbia Basin, LC Valley, and even parts of the the Spokane/CdA could melt during this episode. So that's a significant chunk of the 1 to 4 inches of water tied up in the snow pack which could runoff into the area rivers. What we haven't talked about yet is how much precipitation can we expect in this upcoming weather pattern. The map below shows the GFS model forecast for precipitation amounts possible in a 48-hr period ending Wednesday afternoon. 

Tuesday-Wednesday precipitation forecast
This is a very impressive amount and is one of the wetter model runs. Locations shaded in yellow are amounts ranging from 1.50" (around Spokane) to nearly 3.00" over parts of the Idaho Panhandle. We suspect the valley precipitation will likely be overdone, but perhaps the mountains aren't showing enough precipitation. The combination of snow melt combined with ample precipitation could lead to some bad things across the region including:
  • Flooding of small rivers and streams (especially over SE Washington and the southern ID Panhandle).
  • Lowland flooding where the ground is frozen prohibiting infiltration into the soil. 
  • Ice jams could give way and enhance the flooding potential.
  • Urban flooding is a distinct possibility given snow-clogged storm drains.

But since we like to be purveyors of positivity let's not dwell on the negative and instead focus on some of the potential good:
  • We will finally be able to walk on snow-free sidewalks, 
  • We can drive on snow-free roads, 
  • Just imagine how warm and refreshing 40s will feel compared to more single digit and teens. 
  • The ice dams on snow-covered roofs may finally melt thus alleviating our worries of potential roof damage.
  • Much smaller worry of frozen pipes.  
  • We won't have to bundle up in countless layer of clothes just to shovel the latest dump of snow. 
  • Spring is only 68 days away!!

Wednesday, January 4, 2017

How much does our snow weigh?

Many people are beginning to wonder how December of 2016 compared to the snowiest winter that the Inland Northwest had back in 2008. Along with that, we have received calls from people concerned about the weight of the snow on their roof.

We will first look at how the snow totals for a couple locations across the Inland Northwest from December 2008 to December 2016.

Snow Totals for the Month of December in 2008 and 2016 in inches. Rankings is based on how much snow the entire month of December has received from records dating back to 1892.

We can see that the places such as Spokane and Bonners Ferry did not receive snow totals close to that of 2008. On the other hand, locations such as Kellogg and Stehekin received almost double the amount of snow that they received in 2008.

When  looking for the possibility of a roof collapsing, we want to take into consideration that different weights that snow may hold due to water content. Light, dry snows weight ranges from 1 to 3 lb/ftwhile wet, heavy snow can reach up to 21 lb/ft2.Once the snow has fallen, the snow can accumulate more liquid through humidity, the compression of snow, and any additional rain that may have fallen. Snow tends to act like a sponge and even though it may look like the snow depth is decreasing, the snow is actually absorbing the excess water. In other words, the depth of the snow can be misleading compared to the weight of the snow. 

When looking at the maps below, we can see the weight of the snow (in lbs/square foot) at worst case scenario being placed on the roofs. This means that if there was no day time melting that occurred during that period, there could potentially be up to the stated weight on the roofs. But chances are that there was daytime melting that was occurring bringing the total weight down.

Snow load weight in  lb/ft for December 2008

Snow load weight in  lb/ft for December 2016

Above we can see how the weight of the snow differs from the different years. To get a clearer picture on which locations have increased and decrease, a table has been included for a side to side comparison. Note that the table does not include all locations as many of the locations did not have data for both years.

Red: increase in the weight load from 2008 to 2016
Blue: decrease in the weight load from 2008 to 2016

The storms of 2008 caused 95 structured to fail in some way across the region. 91% of those buildings were at least 20 years old. Newer buildings have to be built to code so that the they can sustain at least the normal amount of snow that the region receives.

Here are some suggested signs of snow load problems.
  • Cracks in the walls that were not previously there or have extended
  • Frequent popping, creaking, or cracking sounds 
  • The roof visibly sagging
  • Doors, windows that are noticeably harder to open or close, especially interior doors

For more information about structural snow loads, feel free to visit FEMA Snow Load Safety Guide