Squall Strategy

Re-printed with permission from Latitude 38, July 1994.
Copyright Max Ebb Publications and Latitude 38, 1994

     "What's the event tonight?" I asked the bartender. "I had to
drive all the way back to the second overflow parking lot to find
a space!"
     But the bar was almost empty. Just a few people I didn't
recognize buying drinks, and a member serving.
     "It's the weather briefing for the race to Hawaii. Standing
room only in the dining room."
     "Ah, that explains it. Big entry list this year. Who's
     The speaker was a well-known name in Transpac navigation,
all the more reason for the full house. So I ordered a gin and
tonic for myself and wandered towards the dining room, to see if
I could squeeze in for the rest of the presentation.
     "If you're on a sled," advised the speaker, "you can
actually sail as fast as the squall is moving. VMG downwind will
be about half the windspeed, with no upper limit."
     He was referring to a table of numbers projected on the
screen, showing the tacking angle, apparent wind angle, boat
speed, and VMG for a large ultralight.
     "The squall typically moves at 15 knots, but the wind in the
squall is around 30. That's enough to get you down the wind fast
enough to stay in the area of strongest wind, so you can actually
jibe back and forth across the face of the squall many times.
Remember, the object is to stay *in* the squall"
     This generated a chuckle from the audience. Having seen the
entry list, I knew that most of them were cruisers before racers.
The next chart flashed up on the screen, the numbers for a much
older and heavier 40-footer.
     "But in a slower or smaller boat, one pass is usually all
you'll get. How you exit the squall is critical."
     This was interesting. I walked along the back perimeter of
the darkened room, navigating between the packed bodies and re-
arranged tables and chairs. No seats, but it seemed like it might
be worth standing and listening for a while.
     "On a slow boat - and that means anything but a sled - you
should generally `exit stage left' when the squall ends. That is,
leave to the left, on port pole. Never get caught right on the
ceterline of the squall's track behind it, or to the right. And
especially not just before dawn! On the average, the squall wind
will be a starboard-tack lift, for reasons that I explained
earlier. So you'll probably have jibed onto port pole anyway.
Once you leave the squall, you want to get away from the light
air behind the squall as quickly as you can."
     It sounded like I had missed something important.
     "Rule of thumb:" summarized the speaker - "Once you're out
of the squall itself, if you want the wind conditions to change
get on port tack. If you want them to stay the same, sail on
starboard. This is also the rule of thumb for increasing your
chances of staying under the 'wind stripes' that you'll see as
lines of clouds during the day."
     "Why?" I thought to myself. I had definitely missed
something important, walking in in the middle of the talk.
     "Another rule of thumb." he continued. "If you're in a sled
and are trying to jibe back into the squall, jibe as soon as you
get headed. It sounds crazy to jibe on a header - but I've
repeatedly observed the wind direction aimed inward towards the
track of the squall. The header is often the first indicator that
you're getting to one side of the strongest wind, and it's time
to go back."
      Just about everyone in the room was taking notes furiously,
copying down a diagram showing wind speed and direction around a
squall cloud.
     "I don't know why this happens," he confessed. "You'd think
that the strongest wind would just radiate outward from the
center of the squall. But for some reason, the wind is directed
inward like this."
     He emphasized the wind arrows on his diagram, drawn in a
"toed-in" orientation on the left and right front corners of the
     As my eyes became adjusted to the low light level, I could
pick out many of my sailing friends in the crowd. There were a
few of my competitors from the YRA fleet, and the handful of
yacht club members that had berths on the Hawaii race were
scattered in the crowd. There was my sailmaker, over by the wall
on the right. And even naval architecture student Lee Helm, who
sometimes can be persuaded to crew for me, was over on the left
side of the room taking notes along with all the others.
     "Maybe you can explain this phenomenon," the speaker asked
the sailmaker.
     All eyes were on the sailmaker, himself a veteran of many
successful races to Hawaii. Lee noticed me standing along the
back of the room, and she waved acknowledgment when we made eye
     "Oh no," answered the sailmaker. "I know enough to not guess
at questions like this, especially in this crowd!"
     "Anyone else?"
     Now the room was silent, and I saw the sailmaker looking at
Lee. I looked at Lee to see if she would respond. She looked back
at the speaker. The speaker looked back at Lee. I looked at the
sailmaker again. He shrugged. Lee shrugged. The speaker shrugged.
And for good measure, I shrugged.
     "Let's take a 10-minute intermission," announced the
speaker. "After that we'll cover wind stripes, effects of
tropical storms, and best approaches to the finish."
The house lights went on, and a large number of people made for
the bar. I made my way over Lee Helm's table, borrowing a
temporarily vacant chair.
     "I'm surprised you didn't offer an explanation for that wind
shift question," I said as I sat down.
     "Like, I'm just auditing this class," she joked. "Collecting
those `rules of thumb.'"
     "Do you have a spot on the race this year?"
     "I wish. But like, I really have to finish my thesis this
summer. So I'm on the beach again. I mean, I'll be up for it next
year, though. Especially if you know someone doing the race to
     "I'll be on the lookout for a berth for you, Lee. But I
still can't believe you don't have an explanation for that wind
     "For sure, there are ways to explain it. The main thing is
to think of systems of convection cells, instead of just a single
source of wind from an isolated downrush column. When one
convective cell is collapsing, it's almost certainly triggering
new ones ahead of it. So a squall system often resembles a kind
of dipole, a source-sink pair with the strongest wind right
between he two."
     "Ah ha! Of course!" exclaimed a racer who was sitting at the
same table, until now absorbed in studying the notes from the
previous part of the lecture. "If you simperimpose a dipole flow
field on the surface wind - taking into account the veered upper
flow - you get exactly the wind shown on that last diagram!"
     "Wait, wait, back up," I said. "What on earth are you
talking about?"
     "Okay, Max," said Lee patiently. "I'll try to explain this
for the differently clued."
     She turned over the page on her yellow note pad, and drew a
graph showing temperature versus altitude.
     "The new word you need to know is *lapse rate*. This is the
observed vertical temperature gradien in a column of air, and
it's typically about 3.5 degrees Fahrenheit per thousand feet of
altitude. It's like if you sent a thermometer up in a balloon,
and measured temperature with respect to altitude as the balloon
went up, you like get this line."
     "Okay, I'm with you."
     "Cool. Now, there are two more lapse rates to deal with, the
*dry adiabatic lapse rate* and the *wet adiabatic lapse rate.*
These refer to the rate at which an imaginary piece of air would
change temperature if it's moved up or down without any heat
being allowed to flow in or out - hence `adiabatic.' A typical
value for dry adiabatic lapse rate is like 5.4 degrees per
thousand feet. So you start with a container of air, move it up a
thousand, let the container expand to match the reduced pressure,
the temperature will be drop by 5.4 degrees. That's for dry air -
if the humidity in the air is 100% when you start, then some of
the water vapor will be forced to condense, because at the lower
pressure the air has less capability to hold water vapor. This
involves a change of state for the water vapor - from gas to
liquid - and the heat of vaporization is released into the air
when the water condenses. This keeps the air warmer, so the wet
adiabatic lapse rate is less than he dry rate - typically 3.2
degrees per thousand feet."
     She drew two more lines on her graph, representing the two
additional rates of cooling.
     "Now, the good stuff. Suppose it's a typical day, and the
actual measured lapse rate is the average 3.5. If the surface
temperature is 75 degrees, what would happen if you took some air
from near the surface and raised it a thousand feet?"
     "Dry air or wet air?" I asked.
     "Good question! Let's say dry, for now."
     "Okay, you said the dry rate was 5.4 degrees per thousand
feet, so it cools 5.4 degrees, and end up at just under 70, at -
let's see - 69.6 degrees."
     What's the temperature of the surronding air at that
     "I get it," I said. "The surrounding air is 3.5 degrees
cooler, according to the lapse rate curve, so the surrounding air
is at 71.5 degrees. The air that we lifted up is now cooler than
the surrounding air, so it would sink back down."
     "Very good! We have stable air. No convection cells here.
You remember from the last time we went through this, huh?"
     "Okay, but what does this have to do with squalls?"
     "Hang on. What if you have air that's fully saturated with
water, and raise it a thousand feet?"
     "Use the wet lapse rate," prompted the other racer at the
table, when I hesitated."
     "Thanks," I said as I did some more arithmetic in may head.
"Now the air only drops in temperature by 3.2 degrees, to 71.8."
     "And?" asked Lee.
     "Compare with ambient," suggested the racer.
     "Oh, I see. It's a little warmer than the air around it this
time, so it would rise up."
     "So, I guess it would keep rising."
     "Which means the air is unstable. That's why cumulus clouds
billow up - the moist air keeps rising, water vapor adding heat
to the rising column of air. Whenever the lapse rate - what you
measured with the balloon - is steeper than the adiabatic rate -
wet or dry, as appropriate for the altitude and the amount of
moisture - the air will be unstable. Push some air up, and like,
it keeps going up. Push it down, and it keeps sinking."
     "Last time we discussed this you mentioned the lava lamp." I
     "Right. If heating near the ground causes the measured lapse
rate to be steeper than the dry adiabatic lapse rate," added the
other racer at the table, "even dry air will be unstable, and
boil up in a column of rising air, or a thermal. I know all about
this stuff because I fly gliders. When the air reaches `cloud
base,' pressure drops to where the water vapor saturates the air,
then the wet lapse rate takes over, and it becomes even more
unstable. The thermal in the cloud is generally stronger, and
more turbulent."
     "Okay, back to squalls," said lee. "The ocean surface is
heated slightly by the sun during the day, but it doesn't change
temperature nearly as fast or as much as the air. At night the
upper air cools, but the air near the surface stays warm. The
actual lapse rate becomes very steep - relatively warm at the
surface, much cooler a little way up - so the air is unstable,
even the unsaturated `dry' air right at the surface is unstable.
Rising columns of air form. But the air doesn't have to rise very
far before it becomes saturated, causing it to cool at the
slower, wet adiabatic rate, which makes it even more buoyant
relative to the surrounding air, which makes it rise even faster,
which makes it even more buoyant still. You have a humongous
towering cumulus cloud, transfering hot air up from the surface,
trying hard to bring the temperature gradient in the atmosphere
back to normal."
     Lee was gesturing with her hands as she spoke, trying to
depict clouds rising to the stratosphere.
     "And it works the other way too." she continued. "When the
moist air at the top of these clouds cools down, it's ready to
start falling. Extra liquid water in the form of cloud droplets
and rain are re-evaporated back into the descending air,
effectively refrigerating it as the pressure increases. The
`downrush column' picks up speed, and as long as the lapse rate
of the surrounding air is steeper than the wet adiabatic lapse
rate, the downdraft air just keeps sinking faster as it falls
through the cloud. If there's rain falling out of the cloud, then
the wet rate applies right down to the surface, because there's
still water evaporating into the air."
     "That checks with my experience," I noted. "Cold rain in
     "That's the standard description of how an isolated squall
works," said Lee.
     "Right," added the racer. "They build all evening, and start
collapsing later at night or in the early morning hours. The
biggest squalls of all are the ones that hit just before dawn."
     "So you'd think," Lee continued, "that the wind field around
a squall would be a strong outward flow of cold air, from the
downrush column hitting the surface and spreading out in all
directions. I mean, you have to add to that wind pattern the
existing trade wind field, so the two winds reinforce each other
in front of the squall, and cancel out behind it leaving you
becalmed. You also have to add the wind component from the motion
of the squall cell itself, which will be deflected to the right
relative to the surface wind. That's because the upper air
follows the isobars, but down low the wind is slowed by surface
friction and tends to be distorted along the pressure gradient,
away from the center of the high."
      "That explains the usual starboard-tack lift in a squall,"
said the racer. "At least it's a starboard tack lift slightly
more than 50% of the time."
     "I'm not sure I got that last part," I allowed, "but
everything else agrees with what the books say. The squall
behaves like a strong downrush of air, fanning out from a point
right under the clouds, adding to the average wind in front and
subtracting from the wind in back."
     "Except when it doesn't," said the racer/pilot. "The shift
is to the right most of the time, but how do you explain the
times when it shifts the other way? And how do you explain those
toed-in wind arrows on that diagram?"
     "There's a few more things going on," said Lee. "First off,
you really don't know where the convection cell is in its cycle
of developing and collapsing. That affects wind speed and
rainfall. But more important, squalls hardly ever exist as
isolated cells. The night air is unstable, and when the downrush
air turns horizontal and flows out in front of the squall, it
forms a cold wedge, like a miniature cold front, that lifts a lot
more warm air up from the surface. This sets off a new convection
cell of rising air immediately in front of the squall."
     "So the air from the downrush gets sucked right back up into
the new thermal?" I said, as the idea finally sunk home. "Now I
see why the wind direction turns inward!"
     "Not exactly. I don't think the downrush air is going to
warm up quickly enough to power a strong upward convection. So
it's not a true dipole flow field, in that sense. But I think
there's converging flow into the new cell right above the
downrush air, and this tends to deepen the layer of cold air,
making it converge into a narrower band of wind like the diagram
     "Interesting theory," said the other racer. "It suggests
that the strongest squalls would be double cells like that, to
get the effects of downflow and upflow combined."
     "Does it explain the old rhyme about wind and rain?" I
     "What's that, Max?"
     I recited a rhyme I remembered from an obscure book about
nautical folklore:
     "First the rain and then the wind,
      Topsail sheets and halyards mind.
      First the wind and then the rain,
      Hoist your topsails up again.
     "Owe! That's a rhyme?" Lee scoffed. "It uses `wind' and
`mind,' then `rain' and `again.' I mean, call the rhyme police!"
     "It must be very old," noted the other racer. "Those words
might have sounded okay as rhymes a thousand or more years ago."
     "But the amazing thing is," I said, "it seems to be true!"
The strongest squall winds come right after the rain. According
to the usual description of a squall, it always seemed to me that
the wind should hit first."
     "It checks with my dipole model," said Lee, "despite the
dorky rhymes. Like, you'd expect to find rain under the new,
strongly driven convection cell in front of the main squall
cloud. Then the strongest wind would follow the rain. But if the
rain comes after the wind, then the squall is already over and
you're in for a period of calm in the squall's wake."
     "So how can we use all this to advantage during the race?"
we asked.
     "By understanding that some clouds are going up, and some
are going down. Some people like to think of cumulus clouds as
pistons - when one collapses in a squall, it pushes another one
up. Then there's the gravity wave theory. Researchers have
actually identified big atmospheric waves, like water waves,
causing new convections cells to appear near existing cells. So
it gets pretty gnarly."
     Meanwhile the room was filling up again, but fortunately
whoever had a prior claim to my chair had found a better view
from another part of the room. Lee flipped her pad to another new
sheet of paper as the lights dimmed.
     "Is any of this actually going to help us get to Hawaii?"
the racer asked again.
     "For sure," said Lee. "Just use all the rules of thumb!"
                                                   max ebb
1) "Incoming lane" for squall: squall should be just abaft the
beam on starboard tack.
2) Jibe on headers to stay in the squall (sled only).
3) Always exit stage left.
4) Don't be anywhere near the last squall just before dawn. Wind
dissipates at first light.
5) Best lure: blue & white feathers with silver head.