Like the famous quip from Charles Dickens' A Tale of Two Cities, occlusion is the best of times and it's the worst of times. In less literary terms, occlusion is the stage in the life cycle of a mid-latitude cyclone during which the surface low moves back into the cold air in search of life-sustaining divergence at 500 mb. Why would I liken an occlusion to A Tale of Two Cities? I'll explain the basis for invoking Dickens throughout this section. Just so I don't leave you hanging until the end, the bottom line is that an occluding low's search for upper-air divergence is very fruitful in the early stages of occlusion. Indeed, the cyclone reaches its maximum strength just as it starts to move back into the cold air. In the latter stages of occlusion, however, the surface pressure rises at the center of the low, signaling that the low is filling and on its way to the graveyard for storms. So the stage of occlusion marks a low's best of times and it's worst of times.
A telltale sign that a classic low pressure system is about to occlude and temporarily reach its maximum intensity can be gleaned from the "tilt" of the 500-mb trough. Let me explain what I mean. In the early stages of cyclone development, the tilt of the 500-mb trough line is usually positive, slanting from the southwest to the northeast. But, as cold-air advection occurring west of the surface low shifts its orientation in response to changing wind direction, the tilt of the trough makes a transition from a positive tilt to a neutral tilt (south to north) to a negative tilt (southeast to northwest). In the process of the 500-mb short-wave trough changing tilt and intensifying, please note that the distance from the 500-mb trough line to the surface low shortens in time. That's a very poignant point with regard to your understanding of what happens during occlusion.
For a dose of hard reality, please note the negative tilt of the 500-mb short-wave trough in the relatively early stages of occlusion of the surface low responsible for producing the "Blizzard of 1993". Note that the central pressure of the surface low is less than 968 mb (sea-level pressures bottomed out near 960 mb along the Middle Atlantic Seaboard, which is extremely low for a mid-latitude low pressure system). Also note the relative short distance between the negatively tilted 500-mb trough and the newly occluded surface low.
As the tilt of the 500-mb trough becomes negative, the area of maximum divergence aloft starts to shift westward (animation). In concert with the reshuffling pattern of upper-air divergence, the surface low, in an attempt to stay under the maximum area of divergence, moves back into the cold air northwest of the low. As the low heads back into the cold air, an occluded front (colored purple with alternating triangular barbs and circles on the eastern side of the occluded front) now appears at the surface. The occluded front stretches from the low's center to the "triple point", which marks the intersection of the southern end of the occluded front with the system's cold front and warm front. In the cyclone model of a classic low pressure system that occludes east of the Rockies, the occluded front is the boundary between advancing cold air to the west and retreating cold air to the east. Warm air is no longer at the surface at and southeast of the low's center, like it was while the low developed. Instead, warm air gets shoved upward, cooling on ascent.
In the early stages of occlusion, the low reaches its maximum intensity (lowest barometric pressure) as it continues to feed off a large supply of upper-level divergence from a negatively tilted 500-mb trough. One reason why a negatively tilted trough signals a rich supply of upper-air divergence is that a 300-mb jet streak simultaneously arrives from the southwest, putting its divergence-rich left-exit region over the surface low. In the case of the Blizzard of 1993, a formidable jet streak, with winds in excess of 150 knots at its core (over 170 miles an hour), arrived from the southwest, placing the occluding but still deepening surface low in the favorable left-exit region of the jet streak.
The low's good fortune of having a wealth of upper-air divergence is short-lived, however. Indeed, as the low continues to move closer to the now closed 500-mb low (weather forecasters refer to the system becoming vertically stacked), the death knell begins to sound for the low (animation). That's because the upper-level divergence associated with the negatively tilted 500-mb trough continues to zip ahead to the northeast away from the northwestward-moving surface low. With most of its upper-level divergence moving away from the low (and weakening), the low's fate is sealed and it starts to "fill". To see an animation showing the entire life-cycle of the cyclone click here.
To get a better mental picture of what transpires in the latter stages of occlusion, consider my "cyclone machine". Before I relate the machine to occlusion, let me describe how the machine works. A pizza pan covers a hole in the bottom piece of wood. Before I started to film, I filled the pizza pan with water. A hot plate underneath the pizza pan (out of your view) then heats the water to its boiling point (and, thus causes water to evaporate). In the somewhat cooler air above the very hot pizza pan, there's net condensation of water vapor, resulting in a "steam cloud". Meanwhile, air pressure lowers near the center of the pizza pan as a result of warming the air and lowering its density. In response, air rushes in toward the center of the pizza pan through an open side in each of the four Plexiglas windows. The open sides are arranged in order to make inward rushing air circulate in counterclockwise fashion. In effect, the four open sides mimic the Coriolis force.
I know that this skinny "vortex" looks more like a tornado than a mid-latitude cyclone (remember that vertical stretching always accompanies mass convergence), but it will help me to drive home my point. Atop the machine is a piece of metal piping, which essentially serves as a "vent" by providing the means for air to diverge into the room after it converged and rose above the center of the pizza pan. Now watch what happens when I place a book on top of the metal pipe, essentially cutting divergence to zero. In quick fashion, the low fills and dissipates. Although the demise of this tiny cyclone occurred much faster than the death of much bigger mid-latitude low pressure systems, it mimics what happens to "real" lows during the latter stages of occlusion. Indeed, as the surface and 500-mb lows become vertically stacked, upper-air divergence dwindles to nothing, much like it did when I placed a book over the cyclone machine's metal vent.