Thursday, November 13, 2014

Winter 2014/15 Outlook

One of our most asked questions is "What is the winter going to be like?"  In this blog, we'll try to answer that question as best we can.  As most people know, looking this far into the future is more of an "Outlook" than a "Forecast".  

Most everyone has heard of El Nino and La Nina, especially when we're talking about the winter outlook.  We often refer to this as ENSO (El Nino Southern Oscillation)  There are other "oscillations" in the atmosphere that affect our weather.  However, ENSO is the only one that is linked to the ocean temperatures.  This is important for two reason.  First, we have some skill at forecasting ocean temperatures. Second, the ocean temperatures don't suddenly change.  These two facts allow us to predict the atmosphere several months into the future.  Most of the other oscillations are purely atmospheric.  As such, we can predict them only 1-2 weeks into the future with any skill.  What does all of this mean?  ENSO is just one of the influences on our winter weather, and the only one that we can predict with any skill.  The other oscillations are less predictable, and could alter or even override the effects of ENSO.  So let's first take a look at what is the state of ENSO.

The area of the Pacific Ocean that is monitored for El Nino is shown in the figure below.
ENSO Monitored Regions

The Nino 3.4 is the most critical area, although the other regions also play a role.  What is monitored is the Sea Surface Temperatures (SSTs) in this area.  And the main way to look at these is to observe how the temperature compares to what we "normally" observe in this area.  This comparison is referred to as an "anomaly".  For El Nino, the anomaly must be 0.5C warmer than normal for 5 months.  La Nina is defined as an anomaly of 0.5C cooler than normal.

Currently, the SST anomaly in the 3.4 region is very close to 0.5C.  It's actually been close to that value since early summer, but decline a bit in August before warming this autumn.

Sea Surface Temperature Anomalies in the 4 NINO regions

So the Nino3.4 region is already close to the 0.5C criteria for El Nino.  So what is the forecast for the SSTs?  Here's a figure that shows a number of numerical predictions from several different countries.

This graph shows that the majority of forecasts lie between the 0.5C and 1.0C lines through the winter and next spring.  This would equate to a "weak" El Nino.  There are a few that expect a somewhat stronger El Nino, as well as a few that are predicting no El Nino.  

Here's the prediction from a number of runs from the U.S. CFS model:

Again, the majority (but not all) of the forecasts are between 0.5 and 1.0C, with additional warming during the summer.  The European climate model has a similar forecast:

So the take-away from all of this is that the majority of computer forecast are expecting a weak El Nino to persist through the winter and into the Spring of 2015.  The latest NOAA Climate Prediction Center discussion states that there is a 58% chance of an El Nino.  

The affects of El Nino on North America weather are fairly well understood.  And these are reflected in the official winter outlook from NOAA:

For the Pacific Northwest, the expectation is for warmer and drier conditions than normal.  And this fits what we commonly see in our area during an El Nino winter.

Here's a bar chart showing the observed temperatures at Spokane for the winters since 1949/50.  The bars are colored in red (El Nino), blue (La Nina), and white (neutral).  The black horizontal line represents a "normal" winter.

As you can see, just about every red bar (El Nino winters) is above the black line, meaning that the El Nino winters are warmer than normal.  

A similar graph for precipitation is found below:

In this instance, you can see more variability with the red bars.  Some El Nino winters are wetter than normal, others are drier than normal.

The third graph shows the occurrence of snow.

This is a fairly clear signal.  We haven't seen an above-normal snowfall El Nino winter since 1977/78.  Most El Ninos bring below normal snowfall to the Inland Northwest.

So the outlook for this winter in the Inland Northwest is for above-normal temperatures, with below-normal snowfall.  But as was earlier stated, this is by no means a done deal.  The other atmospheric oscillations could still alter the weather beyond the affects of ENSO.  Additionally, this is an outlook for the entire winter.  So there could still be a frigid or snowy week or two.  But when the winter is all said and done, we'll probably look back at it as a milder winter.

Friday, November 7, 2014

Winter Arrives Tuesday

We've had unprecedented warmth this Autumn season.  But all that is about to end.  First, let's take a look at how warm it's been.  

October was a record warm month is some locations.  Here's a quick table showing the warmth:

  •  Wenatchee
    • Oct 2014: 56.6F - 2nd warmest October
    • Record: 57.3F in 1988
  • Spokane
    • Oct 2014: 53.3F - 5th warmest October
    • Record: 54.5F in 1952
  • Lewiston
    • Oct 2014: 57.9F - 3rd warmest October
    • Record: 58.7F in 1988
The warmth actually extends back into mid-September.  Here's a graph of the temperatures at Spokane Airport for the last 2 months:

The blue bars are the observed high and low temperature this year.  The brown shading is what we typically see for temperatures on these days.  And the blue and red lines are the coldest/warmest temperatures ever observed.  So you can see that we didn't set any daily records at Spokane.  But notice how the observed temperatures were consistently above normal, with just a few cool spells.  Add to that, Spokane didn't reach the freezing mark until Nov 2nd, which ties with 2005 as the latest ever for the Airport first freeze.  

But as we said, all of that is about to change, and in a big way.  Here's the weather pattern that is going to do it.  Below is a depiction of the temperatures at about 5000' above sea level for this afternoon (Friday 7 November):

850mb Temperatures Friday afternoon 7 Nov 2014

This image shows that the coldest air is over north-central Canada with mild air over all of the western US.  Below is a forecast for Monday morning:

850mb Temperatures for Monday morning 10 Nov 2014

Now the cold Canadian air has penetrated into the northern tier of the US.  You can see that the Inland Northwest is right on the edge of the cold air.  By Wednesday morning, even colder air has made its way into the lower 48:

850mb Temperatures for Wednesday morning 13 Nov 2014

By this time, cold air has completely moved into all of Washington and Oregon east of the Cascades.  Even Portland will probably see some cold air seep through the I-84 Gorge.  Also note how far south the cold air will move in the central US, all the way down to the Texas Panhandle.  The coldest air will be over Montana into the Dakota, but the Idaho Panhandle will still be plenty cold.

What's causing this pattern shift?  In part, it's Super-Typhoon Nuri in the western Pacific.  Nuri developed east of the Philippines on Halloween, rapidly strengthened, and "recurved" into the westerlies.  The CIMMS blog has some great satellite images of it.   

As Nuri moved northward, it converted to an extra-tropical cyclone, and has currently become a massive storm in the Bering Sea.  So why does this affect our weather?  First, he's a graphic showing the jet stream today.

Jet Stream analysis Friday 7 Nov 2014

The yellow line represents the jet stream.  As you can see, the jet stream is coming at us from a very southern latitude, just north of Hawaii.  But as Nuri's remnants develop and move into the Bering Sea, the weather pattern will shift.  

Jet Stream forecast Monday 10 Nov 2014

By Monday, the jet stream will be directed from the central Pacific into Alaska, where it will pick up cold Canadian air, and bring it southward into the lower 48 states.

So how cold will it get?  Here's the forecast high temperatures for Tuesday.

That's right.  High temperatures on Veteran's Day will be below freezing for much of the Inland Northwest.  And here's the forecast low temperatures for Wednesday morning:

Yes, you're reading that right.  Low temperatures in the teens in many places.  This is way colder than anything we've seen this Autumn.  If you've been putting off those Fall Chores (e.g. blowing out sprinklers, putting away hoses, etc), this weekend will be your last chance to do that.

And with these temperatures, people will naturally be wondering if there's any snow that will come with it.  About the only chance for snow will be in the northern Panhandle and northeast Washington on Sunday Night and Monday Morning.  Here's the current forecast:

Friday, October 31, 2014

How's tonight's trick-or-treat weather looking?

Yesterday we discussed what the odds of a wet Halloween would be and whether or not the holiday revelers would need to accessorise with umbrellas or additional rain gear. Well, now that the event is nearing our confidence in where rain will fall during the evening is growing. First lets look at the latest satellite picture. The feature of concern is the north-south band of clouds and moisture (while and green band) associated with a cold front (blue dotted line). This cold front is slowly working its way to the east and will likely it's trek through the evening.

3pm Water Vapor imagery with approximate cold front position in blue

So while the satellite was showing the cold front moving slowly to the east, what did that translate to in regards to precipitation? As you notice on the radar loop below, much of the preciptation has shifted east of the Cascades and Yakima (aside from some isolated showers over western Washington and Oregon) and was making a slow northeast path toward Spokane and northern Idaho. Based on this, it seems certain that rain will fall this evening over these locations, however the front is still expected to weaken during the evening. So where do we expect to see the best chances of rain during the evening between 6 pm and 9 pm?

Radar Mosaic from 130pm-300pm

To answer this question we will refer to the same ensemble forecast as yesterday (SREF model) as well as some hourly model guidance (HRRR model).

According to the SREF model, the best chances will occur over the extreme northeast portions of Washington and adjacent portions of north Idaho (areas shaded in purples and reds). This would impact trick-or-treaters in Sandpoint, Bonners Ferry, and possibly Colville. However notice the SREF isn't expecting much chance of measurable precipitation over Spokane or Coeur d'Alene(greens are 20% chance of less). This is because the front is still expected to stretch and weaken as it moves to the east this evening. The other area of concern for rain is over the extreme southeast corner of Washington, near the Blue Mountains and Pomeroy. Based on current radar trends, this seems overdone, but bears watching. Let's now check on the hourly HRRR guidance to see if it agrees with the SREF.
SREF measurable rain chances from 5pm-11pm

So below you will see the hourly HRRR model data for the period between 6pm-9pm. The images represent what the radar might be showing during those times. Greens and blues show where the model thinks there will be rain. Generally speaking it agrees quite well with the SREF, by keeping most of the rain to the north and east of Spokane. In fact, most of the rain generally remains fixed over northeast Washington and doesn't really hit north Idaho in force until after 8pm, and by then the front looks quite weak. Also notice that over southeast Washington the forecast looks dry until 9pm and then later into the night (not shown here).

6pm simulated radar from the SREF

7pm simulated radar from the SREF

8pm simulated radar from the SREF

9pm simulated radar from the SREF
Does this mean we can count on a dry evening over Spokane and Coeur d'Alene? We don't think so (the HRRR is showing a few tiny specks of green and blue around the area), The front has enough moisture and lift to produce some light rain in this area, but it won't likely amount to more than a few sprinkles and thus rain gear will be optional.

Rain or not, one thing is certain, this will be another mild Halloween. Much more so than what the region endured back in 2002. That evening, the holiday revelers had to endure temperatures in the teens. 2003 was not much warmer. However, since 2007, every Halloween evening has been fairly mild. Here's a look at trick-or-treat temperatures since 2000.

Halloween temperatures at the Spokane Airport since 2000

Thursday, October 30, 2014

Will the Halloween revelers need umbrellas this year?

Aside from planning where to gather candy and what costume to wear on the evening of Halloween, many folks are likely wondering if they need to pack an umbrella as well. The answer to that question will be fairly straightforward for a good portion of the Inland Northwest, for other sections, the outlook is a little less clear. Before we get to that answer though, let's look at what a typical Halloween day brings from a climatological perspective.

The last day of October can actually be quite wet depending on your location. In Spokane and Lewiston, measurable precipitation is reported around 1 out of 3 years (the exact number is 36% in Spokane and 37% in Lewiston) while in locations such as Wenatchee, Moses Lake, and Omak rain is expected about 1 out of every 4 years.  Here's a table of the Halloween rainfall frequency from sites across the Inland Northwest.

Rainfall frequency on 10/31

So as you can see, rain on Halloween isn't that rare, especially over the Idaho Panhandle and extreme eastern Washington. Now that we have the climatological background, lets delve into the specifics for this year's festivities. First we will take a look at the regional satellite imagery below. The water vapor image below shows the current weather will  be dominated by a cluster of low pressure areas off the coast combined with an elongated cold front (blue dotted line) stretching from British Columbia south to off the California coast. There also was a weak warm front extending into Washington (red dotted line).
Water Vapor Satellite-- 12 pm 10/30/14

These fronts were producing a widespread blanket of rain covering much of Washington as seen on the radar image below. All indications are that this rain will persist through most of today and into tonight. But what happens to this rain shield tomorrow, and most importantly, what will become of the rain by Halloween night? To best answer this question, we will want to track the cold front, and try to place it by tomorrow evening. 

1200 pm Radar Image
So looking at most of the model guidance below, the front (represented by green shading extending from BC to northern California) is placed near the Cascades by early Halloween morning. This is where we'd expect most of the precipitation to occur. Rain chances look very good in these areas.

5am Halloween 
By midday, the front continues it slow march to the east, nearly reaching the Idaho/Washington border. However, during the morning, the front begins to weaken and stretch apart. Notice how much narrower the green shaded band is compared to the image from 6 hours earlier. This is a sign that the chances of rain associated with the front will begin to decrease. 

11am Halloween

By 5pm, the front continues it trend of stretching and weakening as it crosses the Washington/Idaho border. Although the front is still weakening, the rainfall chances will likely be bolstered somewhat by the upslope wind flow pushing into the Panhandle mountains. Thus, we expect rain to be found over most of north Idaho and extreme eastern Washington at this time. Amounts may not be heavy where most people live though. But with most trick-or-treating occurring after dark we really are concerned with what's expected in the window from 6pm-9pm (sunset is around 530 pm). 

5pm Halloween

During this time, the front really begins to lose its eastward momentum. Most of this is due to the deepening of the offshore trough (orange lines). When troughs deepen, they typically slow the forward (eastward) progression of any fronts found ahead of them. This is certainly the case here which then poses a problem with the weather forecast. How fast will the front really move?

8pm Halloween
By 11pm, the front has barely moved as the offshore trough continues to deepen (you can better visulise the movement of the front and deepening of the trough in the second movie image below). Notice locations from southeast Washington and across most of the Idaho Panhandle remain covered by the green shading. Does this translate to wet conditions over these areas?

11 pm Halloween

Loop of  frontal progression between 8am 10/30 through 11pm 10/31

To answer that question, let's take a look at the amount of precipitation expected between 5pm-11pm. Here's a look at 4 different models for that period and the amount of rain expected (green shading represents light rain while blues equate to moderate rain). For trick-or-treaters in Wenatchee and Omak the news is good with little if any threat of evening rain. Meanwhile holiday revelers in Sandpoint, Kellogg and Lewiston may want to think about incorporating an umbrella into the costumes. So what's our advice for locations on the western edge of the precipitation shield such as Spokane, Colville, Ritzville  and Newport? This is where the forecast gets tougher. A little slowing of the front will equate to persistent rain in these locations, but what are the odds of such slowing?

That's not an easy question to answer at this time. It will become clearer tomorrow, but until then we can rely on such tools as ensemble forecasting. For some background on ensemble forecasting, you can refer to this earlier blog entry. Anyway when utilizing ensemble model input we can see that most of the model perturbations or variations place the bulk of the precipitation threat (between 5pm-11pm) from Northport through the Idaho Panhandle, and south-soutwest toward Pullman, and Walla Walla (purple shading). This is where we have high confidence in a rainy forecast and this largely reflects the 4-panel of precipitation amounts seen above. Meanwhile, there is some agreement the front could still deliver some precipitation to locations such as Spokane, Pomeroy, and Colville (represented by blue shading). The blue shading simply states that 30-40% of the various model variations are producing measurable rain in these locations. 
Ensemble chances of measurable rainfall between 5pm-11pm Halloween

So for now our advice is pack an umbrella if you are trick-or-treating in Idaho or southeast Washington. If your plans are in east-central or northeast Washington, stay tuned for later forecasts and be prepared for the possibility of a slowing front and some rain drops. Even if rain does occur, it should be quite light meaning Halloween celebrations will not be a complete rain out.

Friday, October 24, 2014

An Update on Windy Weekend and Hurricane Ana

The Inland Northwest will see some interesting weather over the next 4 days.  First, we have a strong Pacific storm moving through the area this weekend, then the potential for remnants of Hurricane Ana.  First, let's update the forecast for the weekend wind.

The satellite shows a beautiful storm off the West Coast.  The warm front of this storm is the cloud band stretching from Washington into Montana.  The cold front is the cloud band that extends from western Washington southward off the California coast.  The warm front is bringing us rain this evening.  It will lift northward into BC.  Then we'll have a dry start to Saturday, before the cold front brings more rain Saturday afternoon/evening. 

Infrared Satellite at 730pm PDT 24 Oct 2014
Here's the UW WRF forecast of wind gusts as the low approaches the coast.

The dark blue colors along the southern Oregon coast are gusts around 60 kts (70 mph).  There's already a high wind warning out for that area.

As the low moves onshore and eastward across the US/Canadian border, winds will increase Saturday night over the Inland Northwest.  The cold front should move over the Cascades around midnight, and reach the Panhandle before sunrise.  That's when the winds will be strongest.  Below is the forecast for 5 am Sunday:

The green shading in southeast Washington is 35 kts (40 mph).  There's a bulls-eye of 60 kts over the Blue Mountains.

But this computer forecast looks too conservative.  The pattern of a deep low passing just to our north is a common pattern for high winds here.  Actually, the perfect pattern for really strong winds is where the low tracks just a little farther to the north.

So for the area from Spokane/Cd'A down to Lewiston and over to Tri Cities, expect wind gusts to 50 mph Saturday night into Sunday morning.

Then our attention turns to Hurricane Ana, churning northwest of Hawaii.

IR Satellite image 830pm PDT 25 Oct 2014

Here's the GFS forecast Friday evening.  The little white circle with the yellow shading is Ana.

GFS forecast of SLP (white lines) and precipitation (shading) for 11pm 25 Oct 2014

By Saturday afternoon, the Pacific low has picked up Ana and is moving her remnants to the north.

GFS forecast of SLP (white lines) and precipitation (shading) for 5pm 26 Oct 2014
24 hours later, the low is moving into the Gulf of Alaska towards the Canadian coast.

GFS forecast of SLP (white lines) and precipitation (shading) for 5pm 27 Oct 2014
And finally by Tuesday morning, the rain has reached the Inland Northwest.

GFS forecast of SLP (white lines) and precipitation (shading) for 5am 28 Oct 2014
Here's the track forecast from the Central Pacific Hurricane Center in Hawaii:

We can't remember ever seeing a situation quite like this.

So on Tuesday morning, as the rain falls on your windshield as you drive to work, you can thank Hurricane Ana for providing some of the moisture to the Inland Northwest.  

Once again, by the time this reaches the Pacific Northwest, it will not be a tropical storm, just the moisture remnants of it.  Still, it's kinda neat to think that our rain came in part from a Hurricane.

Thursday, October 23, 2014

A windy weekend expected

In yesterday's blog post, we talked about the potential for the remnants of Tropical Storm Ana to head to the Inland Northwest early next week.  In that blog, we briefly mentioned a possible wind storm on the weekend, before Ana arrives.  Let's take a look at some details of this event.

First, we'll look at the overall set up.  Here's the GFS model forecast for rain (shading) and sea-level pressure (red lines).

The forecast shows a deep low pressure center off the West coast Friday afternoon.  The precipitation extending across northern Oregon into central Idaho is associated with the warm front that will move into our area.  By this time, strong winds will be developing along the southern Oregon coast.

Overnight, the low deepens and moves north.  Here's the forecast for Saturday morning.

GFS Forecast SLP (red contours) and precipitation (shading) for Saturday morning 25 Oct
Now the low pressure is down to 990mb, and impressive low.  The warm front has lifted up to the WA/BC border.  The tightly packed pressure lines (red contours) indicate wind strength.  The tighter the packing, the stronger the winds.  Over the Inland Northwest, winds will be from the east on Saturday, blowing into the low.  Along the Oregon Coast it will be extremely windy.

By Saturday afternoon (below), the low moves onto the Olympic Peninsula and fills to 995mb.  

GFS Forecast SLP (red contours) and precipitation (shading) for Saturday afternoon 25 Oct

And by Sunday morning, the low will be in southeast British Columbia.

GFS Forecast SLP (red contours) and precipitation (shading) for Sunday 26 Oct

It's this Saturday night and Sunday morning period where the winds could be rather strong across the Inland Northwest.  On the one hand, the pattern for this storm is one that is common for strong winds, with the low moving by just to the north of the US/Canadian border.  On the down side, it's moving through at night, which tends to inhibit winds somewhat.  But in this case, that probably won't matter.

Now let's look a bit closer at the details.  First, here's the University of Washington WRF model forecast for Saturday.

This shows forecast gusts of 70+ mph along the central Oregon coast.  Not good beach umbrella weather.

And here's the WRF forecast for wind gusts Saturday evening:

The area of red and orange in southeast Washington is a forecast of up to 57 kts, or 65 mph.  Now, before you say "glad I don't live in southeast Washington", this is just one model forecast.  The next model run (tonight) could move that area of strong winds to a slightly different area, such as Tri Cities, or Pullman, or Spokane.  At this point, it's too early to tell exactly where the strongest winds will occur, and just how strong they'll be.

As for Tropical Storm Ana, the GFS and ECMWF models are still showing solutions similar to yesterday's blog.  Here's the GFS forecast for Tuesday afternoon:

GFS Forecast SLP (red contours) and precipitation (shading) for Tuesday 29 Oct
We'll write more about this Friday evening.