Gap Winds and QG Theory

Being in Seattle now, I have developed a healthy interest in our own local weather phenomena.  Today I woke up to a new weather alert that there was a wind advisory out for the passes and western slopes of the Cascades.  Sure enough, observations this morning are confirming 20-30 knot winds through all of the mountain passes in the central Cascades and in some foothills towns like Enumclaw.

But what's causing such strong winds through the passes and on the western slopes?  This is a phenomenon known as gap winds.  Think about water or air being forced through a funnel.  A lot of water moving through a narrower space is going to make the water accellerate.  That's what happens with gap winds.  Air wants to cross the Cascades, but can only find a way through in the low, narrow passes.  Since large volumes of air are trying to move through such narrow corridors, the flow is accelerated and we see really strong winds as the air goes through and comes out on the other side.  This is a crude description of what happens, but it gives you a good general idea.  Justin Sharp, a recent student of Cliff Mass (one of my advisors at Washington), did a study on some of these gap winds that can be found here.

But what is making the air "want" to cross the Cascades today, particularly from the east to the west? The answer is differences in air pressure.  Air tends to flow from regions of higher pressure to regions of lower pressure.  Thing about pumping up a flat basketball.  Initially after you attach the pump nothing happens.  But as you press down on the pump, you increase the pressure of the air in the pump.  There is then a pressure imbalance, as there is higher pressure in the pump but the same, original air pressure in the ball.  Therefore, the air starts flowing from the pump into the ball to try and decrease this pressure difference.  It flows from high to low pressure.

So how do we see this today?  Take a look at the sea-level pressures from several stations on both sides of the Cascades for the past 24 hours:

Fig 1--Sea-level pressures in inches of mercury for the past 24 hours at several Washington stations.  Figure courtesy of University of Washington Dept. of Atmos. Sci. website.

In the graph above, the left side represents almost 9AM PDT yeseterday and the right hand side represents this morning at around 8AM.  The red, green and blue lines represent Seattle, Bellingham and Tacoma respectively, all on the western side of the Cascades.  The pink, teal and yellow lines represent Wenatchee, Ellensburg and Yakima, all on the eastern side of the Cascades.  Yesterday morning, all the pressures were about the same.  But look at what has happened in the last 24 hours!  All of the stations west of the Cascades have seen their pressure drop while those on the eastern side have seen a slight pressure rise.  No wonder air is moving from the east to the west with a pressure imbalance like that!  If you're curious about some more details about this phenomenon, here is a study by Richard Reed (1981) that highlights the kinds of pressure differences required to produce winds of a certain strength in the Cascades.

Let's go a step further (and into a far more technical area) and ask one question more--why are we seeing such a pressure difference?  The answer partially lies in the fact that there is a strengthening ridge to the east of the Cascades (also due to a trough approaching from offshore, but that's a story for another time).  Those without a good meteorology background should probably stop here, but you can read on if you would like.  We can diagnose this strengthening by remembering the quasi-geostrophic height tendency equation.  Recall that a change in the geopotential height of a pressure surface (and correspondingly changes in surface pressure) is related to vorticity advection on that surface and the gradient of temperature advection through through surface.  A quick look at a vorticity advection map (not shown) didn't reveal much coherent going on in the area, so particularly, we will recall that:

Warm air advection decreasing with height contributes to a rise in geopotential heights.

Let's look at an objective analysis of warm air advection at 850 mb from this morning:

Fig 2--850mb temperature advection and geopotential height contours from 12Z, Nov 3rd, 2010. From the HOOT website.

We can see this soaring ridge across much of the inand northwest, as well as the strong height gradient offshore.  If we translate this to the surface, we can already see that pressure gradient we saw from the surface observations manifesting itself in the upper air pattern.  But note the very strong values for warm air advection along the British Columbia coast as indicated by the blue arrow.  That's a maximum of over 5E-4 degrees Celsius meters per second (crazy units, I know...).  And this area seems to extend near to or just west of the ridge axis.  Let's look a bit higher and see temperature advection at 500 mb:

Fig 2--500mb temperature advection and geopotential height contours from 12Z, Nov 3rd, 2010. From the HOOT website.

The same location near the British Columbia and Washington coasts is denoted by the blue arrow.  The warm air advection up at 500mb is considerably weaker! Only a maximum of around 3E-4 degrees Celcius meters per second, and over a somewhat smaller area.  So as we go up in height from 850mb to 500mb, our warm air advection is decreasing.  What does our "adage" from above say that this means about heights?

Warm air advection decreasing with height contributes to a rise in geopotential heights.

So we can expect that in the vicinity of that ridge axis area and to its immediate west, heights will rise.  This translates to an even stronger ridge and with that height gradient offshore--winds could really kick up.

A really astute observer might also notice that the cold air advection over ther northern plains is increasing with height.  This is the same thing as warm air advection decreasing with height--so we once again would look for height rises over the next day or so across the northern plains.  Tomorrow we'll see if the ridge did indeed evolve in this way...