Thursday, October 28, 2010

How to Kill People in Clouds

Luke Bucklin disappears in a Mooney, foolishly
See also
http://www.startribune.com/local/west/105871163.html?elr=KArks:DCiU1PciUiD3aPc:_Yyc:aUoaEYY_1Pc_bDaEP7U

(I am a commercially and instrument rated glider and airplane pilot, and an FAA-designated Sr. Aviation Medical Examiner and a general internist interested in aviation human factors.)

This tragedy illustrates the fact that 90% of aircraft accidents are due to the misperceptions, miscalculations, or mistakes of the most complex part of the aircraft: the flesh-and-bones thing operating the controls. Another pilot and I were chatting about the circumstances of this disappearance, based on media reports; he summarized by saying, “That was manslaughter! He should never have taken off in that plane in that weather!”

Why does he feel this way? Because the situation strongly suggests that this pilot experienced loss of control due to tailplane icing and plummeted to earth. And the pilot knew, or should have known, the dangers.

Factors:
1: the particular model of aircraft that he flew is not certified for flight into known icing conditions; in fact, federal regulations prohibit this and similar aircraft from being flown into known icing conditions. Debate does exist on just what “known” means, legally, but … if we know there’s gonna be more than trivial ice, we really shouldn’t go there.
2: Snowfall always means there is icing in the clouds! And the icing continues to accumulate until the airplane breaks out of the cloud tops (if indeed the airplane is capable of climbing to such an altitude) or descends to above-freezing temperatures.
3: The model aircraft he flew is not turbocharged, and has significantly decreased power above about 12,000 ft, leaving small margin for adverse conditions.
4: Icing conditions affect the efficiency of the propeller, decreasing power and speed; decreasing the efficiency of the wing, which both decreases lift and increases drag; and, most importantly, the tail (invisible to the pilot), decreasing the pilot’s ability to hold the nose up with the elevator control.
5: During the crash, ice breaks off, and quickly melts as soon as there’s warmth, removing the evidence from post-crash investigators.

So, knowing that this man took off into clouds from which was falling snow, at high altitude, with minimum enroute altitude of 15,800ft well above the best performance of his aircraft, over unlandable terrain, we can conclude that he has deliberately flown into a dangerous combination of risks. This is spelled R-E-C-K-L-E-S-S.

What evidence is available regarding this?

I looked up the flight trace and airplane registration on flightaware.com
http://flightaware.com/resources/registration/N201HF
This shows that the pilot was flying a 33 year old Mooney, a wonderful airplane, but one that is equipped neither with anti-icing equipment that could reduce ice accumulation, nor with engine turbocharging.
It is possible to purchase aftermarket anti-icing for this aircraft, but it reduces top speed slightly and load-carrying capacity significantly, and in any case is NOT approved for flight into known icing conditions (it’s useful for inadvertent icing encounters; I have such a system on my own Mooney).
It is possible to purchase an turbocharged engine for this aircraft, but it’s much more cost-effective to purchase the turbocharged model; I would be amazed if this aircraft had had such a conversion. A turbocharged Mooney can climb to about 22,000 ft if equipped with oxygen. But even this altitude may not outclimb icy clouds. When there’s snow, cloud tops are often in the range of 20s thousand feet altitude.
In addition, while the Mooney does have 4 seats, its load-carrying capacity with full fuel tanks seldom permits 4 people with luggage. I would be very surprised if this aircraft had not been somewhat over its maximum gross weight. This decreases the maximum altitude it can achieve and raises the stall speed; other adverse consequences are not related to this flight (less margin in bad turbulence, potential damage to landing gear on landing).

Now, let’s look at the flight trace:
http://flightaware.com/live/flight/N201HF

If you zoom in on the map, you will see that the airplane went straight and true after turning on course.
If you click on track log and graph http://flightaware.com/live/flight/N201HF/history/20101025/1730Z/KJAC/KPIR/tracklog you will see a graph of altitude (with a spurious ravine, a gap in data.

If you scroll down on that page, you will see a spreadsheet of data, that, to the knowledgeable person, has some very suggestive data (isn’t that just the way data works?).
First, we have to realize that this is “data” and subject to error. We need to look at consistencies.

What does this data tell us? There’s a story here.
1: Look at climb rates. The FAA standard expected climb for piston-engine aircraft in instrument conditions is 500 ft/min. This airplane at sea level is probably capable of almost 1500 ft/min. Climb rate decreases with altitude, as the engine’s maximum power decreases as the air gets less dense. As a rule of thumb, the instrument pilot is embarrassed or worried when an airplane is not able achieve 500 ft/min. If you look at this data, you see that the pilot was able to exceed this rate up to 11,600 ft., at 3:14 pm. He maintained about 12,000 ft from 3:15 to about 3:20, then climbed to about 14,000 until 3:30 (2000 ft in 10 min is 200 ft/min, a slow climb) After 3:14, we see that he never again had a rate of climb more than 240 ft/min, and he appears to have been maintaining 14,000 ft. (I read that he was assigned 15,800, but the minimum enroute altitude in this sector is 14,000 and there’s no clear sign in the trace that a climb was attempted. I would be surprised if a controller gave this particular aircraft a higher altitude than necessary.)
To see the minimum altitudes, you can go to skyvector.com, enter KJAC and click GO, then click on IFR charts, and use your mouse to look at the routes, e.g. http://skyvector.com/?ll=43.607333333,-110.73775&chart=411&zoom=3 the miminum enroute altitude for the airways (white band with black line) is printed above the small reverse-type airway designation (V298 and V330 in this case)
2: look at the ground speeds. This is airspeed +/- wind, but the wind is a constant factor. This airplane has a normal cruise speed around 150 kt. We don’t know what the winds aloft were that day; but winds above about 10,000 feet are almost always westerly. This plus airspeeds under 130 kt until 3:31 suggest that he was slow for some reason. I normally climb my Mooney at about 130kt and 500 fpm; he would be able to do this to about 10-12,000 ft and then performance would trail off. His climb speeds were about 100 kt; some Mooney pilots do climb at this speed.
However, ground speeds reached about 160 kt by 3:32 pm
3: Notice that after 3:34 pm, the ground speed rapidly decays to 118 kt at 3:49 (in 15 minutes) while maintaining 14,000 ft (to repeat, we have to look at consistencies; my experience is that this FAA-based data, reported from calculations of transponder returns, tends to fluctuate).

What could cause this slowing?
1: A decision by the pilot to save fuel. Unlikely: pilots prefer to go fast; otherwise they could drive.
2: A loss of engine power. Unlikely: pilots have engine-monitoring instruments, and generally get very anxious, especially over water or unlandable terrain with any engine malfunction.
3: Accumulation of ice. Likely: it’s insidious; there’s a certain bravado among some pilots (“Hey, I had nearly a inch of ice, and my Mooney handled it just fine! I kept my airspeed up, and just descended into warm air!”); and, most important, its accumulation on the tailplane is invisible from the cockpit.
The Mooney has a special characteristic: the elevator, in level flight, can be “locked” by ice, especially if the autopilot is used (the autopilot flies much more smoothly than a human, who continually moves the controls to make small corrections)
A very nice illustration of the Mooney’s elevator, the same model as the missing aircraft, is at http://boxybutgood.com/M20F/1024/00053.jpg
Note that the end of the elevator has an arm-like extension. This is a balance weight. In level flight, the elevator is angled downward to the rear; the front end of this balance weight is *above* the plane of the horizontal stabilizer to which it’s attached. Ice can fill this gap and trap the elevator from being deflected up if the nose needs to be raised. We cannot know whether Mr. Bucklin was using his autopilot, but I am sure that his aircraft had one and would be surprised if he had not had it engaged.

Why does an airplane crash in icing conditions? Usually, because of tailplane icing. Most pilots, in my experience, are ignorant of the risks of tailplane icing, discussed thoroughly and illustrated in this NASA video: http://video.google.com/videoplay?docid=2238323060735779946# It’s a 20-minute video, so I’ll summarize by saying that the tail holds the airplane up. That is, the heavy engine is well in front of the center of gravity, airplanes are deliberately nose-heavy, so that if they stall, they’ll descend nose-first and pick up speed going forwards. The horizontal tail is a small wing; it generates downward-directed lift to balance the engine’s weight. Changing the angle of the elevator permits climb and descent (technically, it adjusts the speed by changing the wing’s angle of attack, and the engine power is adjusted to manage the actual rate of climb or descent).
Ice reduces the ability of the horizontal tail to generate this balancing downward-directed lift; beyond a certain point, the tail cannot lift the nose of the aircraft, which then becomes a lawn dart.

Final datum: Look at 3:50 to 3:52. While not speeding up, Mr. Bucklin’s aircraft descends 1000 ft in 2 minutes, 1300 ft in one minute. Deviation from one’s assigned altitude by more than 300 ft is a violation of federal air regulations. No attentive pilot would have deliberately made such a descent. This strongly suggests that the earlier decay in airspeed was due to insidious loss of propeller and wing efficiency due to icing, and the rapid loss of altitude due to continued ice accumulation, probably rendering the tailplane relatively ineffective in holding the nose up in a level attitude.

Summary:
Mr. Bucklin took off into snowy clouds, assuredly containing some ice.
Mr. Bucklin appears slower than usual for his model aircraft.
Climb rates after 12000 ft are slower than usual
Aircraft speed mysteriously decreases in the last 6 minutes of flight
The aircraft begins a rapid, inappropriate descent after going slowly for 3 minutes.
If I were traveling with 3 sons, the autopilot would have been on, and there would have been much happily distracting conversation. There is no reason to think that Mr. Bucklin would have noticed insidious loss of speed in just 5 or 6 minutes.

Saturday, October 16, 2010

Grampa Leads Them Astray: The Acorn Collaboration

Mike and Kim, the Competent, Careful, and Responsible parents, went cheerfully off to a seminar on Science and Faith, leaving their innocent children under the care and supervision of the Competent, Careful, and Responsible gramma -- and Grampa.

While Gramma was distracted by the shower, Grampa was lounging around the kitchen in the general vicinity of Eleanor, who's 8 and pretty much self-starting, and Amanda, who's 6 and quietly self-starting, and Luke, who's 3 and pretty much in his own universe.

Luke constructed a very sensible and well-proportioned walking stick from Tinker Toy pieces, walked about using it reasonably; Amanda made airplanes from Tinker Toys and zoomed around the living room with them; and Eleanor slipped outside and returned with a handful of freshly-fallen acorns. Meanwhile, Grampa quietly sipped coffee and observed.

Eleanor found a stout toothpick and began patiently de-husking an acorn. This was rather tedious, and gave Grampa time to finish his coffee, which gave him time to realize that finding a better way to de-husk the acorns might be a real service to his Next-but-Eldest granddaughter.

Grampa fetched a freshly sharpened paring knife from the drawer and begin tediously de-husking an acorn. Eleanor said, relevantly, "I'm not allowed to use knives since Dad sharpened them." Grampa did not say, but thought, 'I'm not sure this is safe for my own fingers; it's pretty awkward.'

Then he had another thought. "Eleanor:" he said. "Maybe we should make some acorn bread."

She beamed. Eleanor generally finds insane ideas refreshing. There have been no switched babies in her lineage, that's sure. "OK," she said, "what do we do?"

Grampa, who had only a ghost of an idea, said (to buy time), "First we have to figure out a way to get these acorns peeled faster."

He set up an assembly line. The pile of fresh acorns at the left, the sun shining in the windows center and right; he put the cutting board on the table and picked the big, freshly-sharp chef's knife from the drawer. He put two cereal bowls in front of Eleanor. He said, "Here. I'll cut these things in quarters, and then you can just peel the shells off. Put the shells in one bowl and the meats in this other.

They hadn't worked long before Amanda and Luke joined the assembly line as helpers. Luke was occasionally helpful; he preferred to select particular acorns for Grampa to slice personally for him to peel, and he didn't always put the shells in the right bowl, and he didn't always peel them first; which mildly distressed Eleanor.

We soon ran through the first acorns. Amanda saw the need coming and harvested more; Luke slipped out and soon brought in a zip-lock bag containing about 2 dozen acorns immersed in fresh water. He set the bag on the table. "Luke, please be careful not to spill that," Grampa said optimistically. Luke carefully set on the table, paused to check that it was nicely balanced upright, and turned away. The bag fell over. Happily, there just happened to be 2 terry towels within reach...

Meanwhile, Grampa and Eleanor trudged along with the acorn-quartering and acorn-peeling task. Grandma emerged from her ablutions all fresh and shiny and helped. It is somewhat slow, this acorn-peeling process. This gave Grampa time to think about how to make acorn flour with the tools at hand.

In case you haven't done this lately, the first problem is that acorns are moist. Moist nuts don't grind well. And what do we use for a grinder? I must have asked this out loud, for Eleanor said, "Mom has an old coffee grinder." "Perfect," Grampa said.

We decided to dry the meats in the oven, and to chop them down from quarter-acorn size to coffee-bean size (it seemed logical). So the cereal-bowl full of meats was upended on the cutting board and the big knife chopped and chopped.

Grampa remembered how bitter the red-oak acorns were that he nibbled as a child. He remembered that bitterness can be water-extracted from some foods. So Eleanor found frying pan, filled it half-full of water, and dumped in the meats. Then she put them on the stove and simmered them for nearly a half hour, stirring gently. The water turned slowly brown. There's a lot of tannin in oak. It had quite a nice, nutty fragrance.

Then a cookie sheet with sides was found by Eleanor, and the chopped meats were spread upon it. The oven was rather arbitrarily set to 250 dF (hotter than boiling, maybe not hot enough to burn), and the grandchildren and Gramma drifted off while the oven did its work and Grampa operated the sink and dishtowel.

After half an hour, it was now 11 am; lunch was threatening to derail our continuity, and the project was nearly beyond our collective attention span. Grampa took a nutmeat from the oven. It was hot, and the outside was dry, but when he broke it, it seemed just a bit soft. What to do?

A nice burr grinder had been retired to a top shelf in the pantry, put out of service by the new coffeemaker having an integral grinder. Grampa brought it down and investigated. He upended it over the garbage disposer and shook out three stale coffee beans. He carefully dusted out the stale grounds and cleaned the receptacle.

He and Eleanor carefully considered how to approach this challenge. While the grinder seemed disused, we decided Dad would be annoyed if the burrs became gummed up with acorn meat. We decided that a little white rice mixed with the acorn meats would cleanse the burrs and not distort the fine acorn aroma.

And in fact this worked well. We set the grind on 'coarse' just to make sure we got the least gumminess, and got nearly 2 cups of acorn meal. Then Grampa put a couple tablespoons of pure rice in the grinder and ran it through to cleanse the system. It would have been a lot better if he'd attached the grounds-cup first, though: ground rice flooded the countertop and spilled on the floor.

Grampa got the broom and dustpan. Luke ran off, and came back with a tiny toy broom, perfect for the counter. Together we saved the day.

Now we had to figure out what ingredients we could combine to make a sort of bread or cracker. And we wanted to make it Amanda-safe, so we couldn't use egg or dairy or tree nuts.

(Grampa is an internist who has tried hard to stay up on allergy, and doesn't know of any cross-allergenicity between dietary tree nuts and acorns; and the Internet agreed: "I have found no evidence that acorns or chestnuts are dangerous for nut-allergic children or adults. Edible chestnuts do cause anaphylaxis in some people, but these are not particularly the people who react to peanuts or the other common tree nuts. Rather, such reactions are linked to latex allergy, quite a separate problem, although both problems happen in people who are 'atopic' (predisposed to common allergies." -- in http://www.users.globalnet.co.uk/~aair/nut_corr.htm)

We decided that that the acorn meal was sort of coarse. How to make a smooth batter that might cook up to a cracker-like endpoint? Gramma's expertise was mined. She found a small bag of whole-wheat flour, and Eleanor and Amanda stirred while Grampa added water a little at a time. With about 1/4th cup of whole-wheat flour, it still seemed a bit coarse. We sprinkled in about 1/8th cup of egg replacer (Amanda-safe), then another 1/8th cup. This smoothed it out quite nicely. There was a nearly-empty jar of wildflower honey nearby on the counter begging to be added. So Grampa upended the jar over the batter, adding about 1/3 cup of honey.

Now we were nearly done. A half-teaspoon of baking powder (to aerate it quickly) and a half-teaspoon of baking soda (to aerate it while baking) went in. Gramma and Eleanor spread a little vegetable oil on a cookie sheet. The batter was carefully smoothed onto the sheet.

The oven was rather arbitrarily set at 350 dF and cooking commenced. It began a light tan. Half an hour later, when we took it out and went to Red Robin for lunch, it looked brown. When we came home at the end of the afternoon, it looked chocolate.

Luke wanted to have the first taste. He took one bite, said quite firmly, "I don't like it." Grampa said, "You can throw it away." He seemed a little surprised to be permitted not to like it, and was quite willing to give it up. Allergic Amanda had a tiny taste, with no bad side effects -- but she didn't plead for more. Rumor has it that she ate one piece. Eleanor had one piece. Grampa had three pieces during the next hour. It isn't bitter; it does taste of honey; it leaves a slight, persistent astringency low behind the tongue. It actually has moderately complex, interesting flavor.

I am thinking that a little cinnamon, and maybe some raisins, might be good next time. And I do like these North Carolina oak trees: their acorns are definitely less bitter than Minnesota.

Sunday, October 10, 2010

Is Healthcare a Right?

This question is a political one: it is used rhetorically, with an assumed 'yes' or 'no' answer, depending on the values of the asker. But it's an interesting question.

The simple response is that the question, 'Is healthcare a right?', is nonsense. It's nonsense because the questioner always (in my experience) has not bothered to ask whether the responder understands what is meant by either 'healthcare' or 'right'.

An analogous question is 'Is food a right?'

I choose this question deliberately, because 'healthcare' is an even broader idea that 'food' -- yet it's obvious to everyone that a right not to starve is not the same as a right to gourmet dining. Like food, healthcare comes in many styles and prices. Like food, not all healthcare is actually healthy for the consumer (I say this as a primary care physician who's seen three decades of unintended medical adverse consequences). Just because it seems good doesn't mean necessarily that it is good.

It would be a months-long process to catalog all the things that are called 'healthcare' from wannabes like aromatherapy and therapeutic dance to truly life-saving things like emergency appendectomy and Caesarian section. Obviously, not every treatment tossed into the healthcare basket is essential to anyone.

But even if we can identify a set of health treatments that are actually lifesaving, we still have a problem with the other half of our question: 'Is healthcare a right?

What is a 'right'? Why do we take time and effort to define and debate rights? Are there different types of rights, as there are different kinds of precipitation? If people thoughtfully disagree on any right, can it be a right? That is, can anything actually be a right, if the fact of its being a right is seriously debated?

Still, there must be some things that are, in fact, rights. If there were none, there would be no point in having the word at all. Discussing and debating conflatabulation makes no sense if no one actually conflatabulates, or fabricates a conflatabulator.

What is a right?

First, a right is social. That is, it is relevant to interactions and relationships among people, or with respect to others' values. Someone walking in the woods, or blogging to no audience, may say anything at all. But this is not a 'right' because the words have no consequences to anyone else. Similarly, 'right' is irrelevant to one's thoughts. 'Right' applies to the manifestation of those thoughts in a way that affects someone else.

Second, there are two types of rights: to act, or to receive.
- The right to act is the ability to refrain from saying or doing a thing without coercion to do or prohibition from doing.
- The right to receive is the ability (or implicit permission) to obtain, from others or from one's surroundings, any thing or service.

Third, there are many social mechanisms for establishing and enforcing rights. Some are informal, between individuals and within groups. But a 'right', in any case, is inherently associated with the existence of some (social) mechanism to sustain the right. It is not possible to have a right without there also being an enforcement mechanism.

Enforcement of rights

Now, enforcement can be either re-enforcing, through explicit permission, encouragement, praise, payment, or thanks; or it can be punitive, through disappointment, anger, weeping, blows, ostracism, fines, imprisonment, or the like.

As a corollary, in order to have rights, there must also be things that are understood not to be rights, with consequences between individuals when one persons assumes a right that does not exist. This is not obvious, I think. A trivial example: if it is not my right to speak against a management decision at work, but I do so, then there must be some consequence, even if it's only the scowl of a co-worker, in order to maintain the boundaries establishing who does have that right.

I often see evidence that some people think 'consequences' are only punitive financial costs or liberty restrictions (fines, loss of licensure, jail time, etc.) As we've all experienced, 'consequences' that are strongest in establishing and enforcing behavior are the thanks, praise, approval, good humor, and acceptance of others. In fact the absence of these, or their withdrawal actually stimulate contrariness: the boy who has no nurturance, for example, becomes aggressive and at least socially damaging. And the strongest punishments are not fines and the threat of jail time, but the irritation, anger, withdrawal, spite, or superciliousness of friends.

Ask me again...

So, let's restate our question in a way that makes sense:
- Are there some healthcares that every person has a right to receive?
This acknowledges that there is a vast range of healthcares.

Now, the fact that a right to healthcare is a right to receive, and the fact that every right is social, means that for every such right, someone has an obligation to provide it.

This means that some individual, whether acting alone or as part of a huge healthcare organization, has a obligation to the individual holding the right to treatment.

Money (vs. health)

Now we have to bring in another social mechanism: money. Money is, at its essence, a social phenomenon that gives relative value to the time, effort, ownership, and skill that an individual brings, in creating a material object or providing a service to another.

It is immediately obvious that we have an intersection of two sets of social rights: any rights to any particular health treatments interact with whatever sets of rights exist with regard to the social phenomenon of money.

This is interesting and important, because some healthcares preserve life (are life-saving), and because in any complex society, the exchange of money is also essential to preserve life (as it permits purchase of essentials that each individual can't produce).

This also means that conflicts will occur; not the least is when a penniless individual is urgently in need of life-saving medical treatment that is costly for the provider to offer. The provision of treatment may threaten the existence of the provider while to withhold it more obviously threatens the existence of the sick individual.

How to Guarantee Healthcare is Available

There is only one way to resolve this conflict: to spread the cost of lifesaving healthcares across society through equitable sharing of all such cost, so that providers and their necessary physical resources exist when they are emergently needed; and so that (for example) Bart need not be denied neurosurgery for his epidural hematoma because providing it, at that moment, would bankrupt the hospital.

We see by this the essential difficulties in defining and establishing a right to healthcare. It involves establishing which treatments should be rights-based, and under what circumstances, and it also involves important parallel decisions about how to arrange finances -- how a healthcare system can and should be funded (what quality, what availability, how much reserve capacity; what payment systems), and what is 'fair' -- incomes, payments, and the like.

It's my view that there seems to be no real debate in our society that healthcares needed for preservation of life and amelioration of suffering are rights -- after all, clean water and sewage treatment, the most important healthcares, are without any debate funded by every community, and safe water and sewage treatment are required of individuals living outside of communities. This would imply that the universal provision of catastrophic health insurance would be universally accepted -- there would be debate, though, about what's a catastrophe!

But there is intense and continuing debate, with opinions swirling like the tides in a Fundy estuary, regarding funding mechanisms and levels. Ironically, the very people who are opposed to mandatory private health insurance are arguing as strongly for the socialized medicine of Medicare by saing, "Don't mess with my Medicare." This is exactly the British narcissism, "I've got mine, Jack!" This is not a debate about healthcare; it's purely selfishness.

It is impossible, in this little essay, to resolve which healthcares should be rights, or what the fairest payment mechanisms should be.

I will only observe that the current US system of healthcare payment is the least socially fair of any in the world, and the least efficient as well. Pick a thousand experts and ask each to redesign the system, and you'd get a thousand reforms that is each superior to the current chaos -- and each would be different from the others: inability to reach consensus -- the consequence of 'diversity' in our country -- is the central obstacle to agreement on reform.