Pilots are flying with more weather information today than ever before, and it has made flying both easier and safer. If you don’t believe that, read this fascinating article from 70 years ago. Legendary author Wolfgang Langewiesche explains why the weather information pilots had in 1949 was so limited, and what could be done to improve the situation. Many of his wish list items have become a reality. It will make you thankful for datalink weather, ForeFlight, and so much more.
The Pilot and the Weather Bureau
Some suggestions that might well revolutionize the weather bureau
Does the Weather Bureau give the pilot what he needs? On a recent long tour, I asked myself that many times; and, like so many pilots, I often found: Not quite. I shall tell my troubles here, and suggest what might be done about them. I shall find it easy to do this in a friendly spirit, and without heat. I like the Weather Bureau; I have many friends among its people. I think they are doing a good job, doing it faithfully, cheaply, without bureaucratic empire-building. They don’t tell anybody what he can’t do. They send out no press releases. They simply work away.
How accurate are their reports? How good are their forecasts? I disagree with pilots who say that the Weather Bureau doesn’t know the weather. If the Voice says that the one-nine-three-one observation at Xville showed ceiling 1200, visibility 2, I believe that at that time the weather was just that. (If meanwhile it has changed, and they are not telling me—that’s another story.)
As for the forecasts — I agree, they are not good enough: only perfection would be good enough. But they are based on a science of which some keystones are still missing. (Just for example: nobody knows for sure what causes rain!) And they deal with a tough continent. Air Facts author Bob Buck, of TWA, has flown all over the world in deliberate search for bad weather; he says the world’s worst weather is right here, between Kansas City and New York. So, perfect forecasting is too much to expect. Nor have I ever seen proof that some other agency could do better than the Weather Bureau does.
One thing, however, I miss in these forecasts. “How sure are they of all this?” It can’t be that forecasts are made with equal confidence; they must range from sure bets to tentative guesses. A pilot wants to know which it is. I think the Weather Bureau would do pilots a favor, and would also shut off much undeserved ridicule from the general public, if they used more often such words as: “almost certainly,” “probably,” “possibly,” “some risk of,” “difficult to predict exactly.”
Here now I have already given the first example of my general proposition: I think the Weather Bureau knows the weather all right. My trouble is, I often can’t find out all they know. With all the teletypes rattling away; with all the weather maps forever a-drawing; with all the broadcasts, I still too often barge off into weather I don’t fully “see;” uncomfortably aware that I have somehow failed to get information which must have been lying right there. All my troubles come under that heading. If my experience is typical, therefore, what we need are effective ways of putting the weather dope across to the pilot.
Trouble: I can’t read a sequence report. They can jerk my license if they like; but I bet lots of commercial pilots can’t read those teletype symbols correctly; and private pilots are no longer even expected to. What makes it so hard are those space-saving tricks. Under some conditions, an item such as the ceiling is reported simply by being left out. This then shifts the relative position of all the other items in the report. I used to know these tricks; but they seem to change every few years. The changes are not advertised in any place I see. So, I don’t trust my reading of a sequence; I am afraid I might confuse an 800-foot ceiling with an 8-mile visibility. I ask for help every time; hence I get no practice, and so I have lost all the reading-facility I once had.
Remedy: It would help me if they would print the first few most important items of each report in a strict, unvarying sequence. For example, the ceiling would always appear in the same position. If it is unlimited, call it U. If it is missing, call it M. I think it would be even better to identify some of the most critical values with a prefix letter: the sky-symbols are unmistakable, but how about C8 for a ceiling of 800 feet, V3 for a visibility of three miles? Then any private pilot could learn in 5 minutes how to read these reports. It would save a lot of weather men a lot of time.
Trouble: I can’t find the sheet with the reports for my route; and on the sheet I can’t identify the stations I am interested in. This is of course especially true in sections with which I am not familiar (and where I therefore ought to check weather with extra care). At a small weather bureau or a communications station, the fellows have a little more time, and usually help you find and read the reports. But at a busy airport, it’s a problem. There are too many batches of teletype matter displayed along the shelves: too many people are too busy doing too much; and I seem to notice that sometimes the weather man himself, trying to find you what you want, gropes quite a bit.
Remedy: Perhaps the time has come to display the sequence reports so that the visiting pilot can normally help himself. He could do so if each sheet of reports were posted on a suitable frame, plainly labelled, “Xcity-Yville-Ztown” — the main points of the particular teletype circuit. Perhaps a wall map might appear next to the same frame, showing the stations represented. A string might lead from each station on the map to the corresponding line on the teletype sheet, fixed there by a thumb tack. A profile map might help you pick the hill stations from the valley stations and the pass stations. Coupled with a less cryptic way of writing the report itself, this would make it unnecessary for me to take up anyone’s time.
Trouble: I find it hard, at the busier weather bureaus, to get all the items of information I need. Information is going to waste unless I get 1) a full sequence on current weather; 2) the airway forecast; 3) terminal forecasts, probably several of them; 4) winds aloft; 5) where are the tops?; and 6) where, within the range of my airplane, is the good weather?
I fail to get this information because of the reasons already stated. I can’t find it for myself. If it is handed to me, I can’t read it reliably. And the weather man himself is so busy with so many duties, he cannot get together for me in one batch the stuff I need. I have to ask for each item separately, and each time have to catch him between other duties — he has to make an observation, or answer a phone, or tear sheets off the teletype and post them, or get his observation ready for teletyping, and so on. To extract from him all I need becomes a slow procedure: sometimes I seem to annoy him a little by wanting still another item. To be blunt about it: at several large-airport weather bureaus I have recently noticed a tendency to turn you away by some brief statement, such as, “That’ll be all right,” or that it was “all pretty good out that way.” Then a sort of curtain goes down, and the weather man volunteers no details. If you bring up such questions as might worry you, say, possible night ground fog, he discusses them more-or-less by memory. Well, I don’t know him. For that matter, he may be a clerk pinch-hitting for the real weather man. (You find that too, especially on the phone.) But I know Uncle Sam. I want a clear look at his official wording, clearly written out. For this somewhat off handed treatment, I can’t blame the weather man. He is busy. Most of his work has to be done on the dot of the scheduled minute, to keep the whole system working. There does not seem enough allowance made, in this schedule, for dealing with the pilot. Yet, telling the pilot is the pay-off of the whole deal. Whenever a pilot goes away hungry, all the manifold work of the Bureau has to that small extent gone to waste.
Remedy: The time is past when you could have a chummy conference with the weather man; the time has come to develop a streamlined, effective way of getting the dope across without such a conference. Several solutions suggest themselves. For example, would it not be possible to type out the airway and terminal forecasts of the region in plain speech, and post them on large, conspicuous frames, suitably labelled?
Or we might borrow from the Navy. Suppose that you wrote on a printed form your destination, IFR, or CFR, time of departure, fuel range, time en route. (Suppose, even, you bought this form from the Weather Bureau for a quarter?) Suppose a girl clerk then filled in your route weather, your terminal forecasts and all the rest? (The difference from military procedure would be that you still need nobody’s “clearance” on VFR.) At any rate, someone high up in the Weather Bureau should ask himself this question: A pilot walks into a busy weather bureau and says: “I want to fly to X.” What-all does the Weather Bureau think he ought to know? And what is to be the standard procedure of letting him know?
Trouble: In flight, the information you get by radio is too old at times when you need it worst — in marginal weather, or on sudden weather changes. Approaching X, the other night, I was proceeding on information, about an hour old, that X was clear; when an almost solid overcast formed below me. I called Y, the last range station this side of X, to check again on X weather: “Clear” was the answer. (Modified, to be sure, by the time of the observation — some time ago.) I told them it didn’t seem to check, and asked them to get me a more recent report, if possible. The operator down there was most helpful. He presently came back with a message from Flight Advisory that X now had a broken overcast and a 1000-foot ceiling, and that in about ten more minutes the overcast would be solid. This was fine work; it was an example of what our weather system will do for you — if you know what to ask for, and have the transmitter and/or the brass to ask all the questions you want. But hardly had my friend of the Y range station told me all this, when he turned around and started his regular weather broadcast. And sure enough: “X, clear.”
For this I don’t blame him. He was doing his duties according to the book. The whole broadcast was clearly labelled (well, it was labelled, if you were cagy enough to listen for the time-item at the opening of the broadcast) as giving the something-something observations, naming a clock time of a little while ago. But the fact remains: the information was out of date and dangerous. It was night; the thousand-foot ceiling at the airport meant weather definitely below VFR in the surrounding terrain, what with hills and tall buildings standing on hills. I happened to have an ADF; I happened to arrive over X during a traffic lull, and was allowed to duck down sort of informally. A fellow in a small, slow, less well-equipped airplane would have had to file an instrument flight plan and to make a standard instrument approach for which perhaps he would not have been prepared. He would have had one of those experiences that make people quit flying.
Remedy: We must realize that the Weather Bureau has to operate heavily “by the book.” It is a vast set-up, intricate, working to close schedules and high accuracies. To let too many people in that system make too many improvised shortcuts would cause confusion and distrust. To broadcast the weather more quickly after the observations are taken is probably not possible. To broadcast more frequently, say each quarter-hour, would clutter up the radio facilities and the teletype circuits. (Remember — each observation has to go by teletype to all the other stations in the region, and then has to be read off over the air. Remember, the range stations’ voice feature is busy also with air traffic control.) In average weather, such frequent broadcasts would also be quite unnecessary. But still, in marginal or suddenly-changing weather, a pilot may need a report often, say every ten minutes. Fortunately, he needs such ten-minute reports only concerning places which he can reach in ten or twenty minutes. Pilots who are still an hour out from a troubled place can be more patient. Perhaps the Weather Bureau could set up a secondary, less formal, more purely local system of weather warnings; something that would be analogous, in the weather field, to Approach Control in the traffic field. These broadcasts might go on only in marginal or rapidly-changing weather. When Xcity has weather trouble, the local weather bureau or the Tower might request the nearest range station, and perhaps the stations right around it, to put out its latest weather every few minutes, without loading up the whole reporting and broadcasting system of the entire region. Perhaps such a broadcast should also suggest an alternate airport.
In thinking about this, think particularly of the private pilot who is suddenly caught in bad weather and — very important — hasn’t the transmitter power to ask for information. When scud suddenly forms around him or the weather otherwise does something unexpected, he may have to make a quick judgment — go on, turn back, land right where he is, run for Z? In the instance cited — had I been a private pilot in a 100-mph ship with a receiver only, I would have had to know: was this a condition local or wide-spread? Had the same thing developed behind me also? Where was it still broken? If it was a rapid, wide-spread deterioration (it was) there would have been no time to lose in getting down; but where?
Meanwhile: in broadcasting weather, if the man said, “. . . observation of 20 minutes ago,” instead of, “. . . one-nine-three-one observation” — would that be amateurish?
Trouble: The forecasts are not sufficiently available in flight. You can get them if you ask for them, but perhaps lots of small ships can’t ask. Could not the airway and terminal forecasts be broadcast once an hour? In my particular case, the sudden worsening had been accurately forecast. But at my last stop the weather man had given me the casual treatment. And, since the weather then was unlimited everywhere, I had not gone through with the tedious job of asking to see all the pertinent forecasts.
Perhaps at times when a station’s current weather is good, while the forecast is bad, this fact could be included in each weather station’s sequence report, perhaps by some brief phrase such as: “Forecast negative.” This would serve as a warning: it would be up to the pilot to get the details. “Forecast steady,” might mean, “It will hold.”
We must recognize that many private flights start without benefit of weather information. To get weather while still on the ground would involve a tedious and expensive long-distance call — and during that phone call the problem would again come up of how to pull out of the weather man all he knows. (I called the Kansas City airport weather bureau recently to check weather, only to be told by a girl that they were busy and were “not taking any calls” till 10 o’clock, which was half an hour away.) Some small fields don’t even have a phone available; or it is on a party-line and some lady is talking. Once in flight, many pilots have no transmitter with which to ask for dope. Therefore, Mr. Weatherman: whatever you know — current weather, forecasts, positions of fronts, or what-not — put it on the air.
Imagine always that your best girl is flying around out there right now, headed for your station. She has only 60 hours total time; and she is one of those real dumb blondes. She doesn’t follow well when you talk fast and throw in big words. She is 45 minutes out, and she has only 1½ hours’ gas left. She is listening to your nearest range station. Is there anything you would like to tell her?
Trouble: A weather map is often hard to find for a private pilot. Too often it is at a field where only airline captains can see it; and the airlines have their own weather bureaus with their own weather maps. Check the situation in almost any metropolitan area: where are the private airplanes flying from, and where is the weather map?
Remedy: Obviously this is mostly a question of money. But it may be that a different type of weather map would be easier to spread more widely, and at the same time more useful. About this, more later.
Trouble: The Flight Advisory man is too hidden. Many private pilots don’t even know he exists, and what he could do for them. This Flight Advisory can be found at every place that has an Air Traffic Control Center. (As a matter of fact, Flight Advisory is handled by Air Traffic Control, and on the table of organization may not even be part of the Weather Bureau. But we mean here by “Weather Bureau” the whole weather setup — the Bureau itself, the Airways Communications Stations and this Flight Advisory Service.) At any rate, there is just about the kind of weather service a pilot needs. The Flight Advisory man is, I believe, a pilot. If he isn’t, he has somehow learned to think like one. He knows his region; he has the whole weather situation in his part of the country thoroughly digested, as only an expert weather man could who is also an experienced pilot. Instead of giving you “the weather,” he can give more nearly the finished product you really want: what the weather will be like in, over, through or under. He can also render you a service which I find hard to get from an ordinary weather bureau even when they have time and patience to try and render it: he can tell you — well, don’t go by this route, you’ll get weathered-in. Go by that route; or — no point pushing like hell right now, have another cup of coffee, waste an hour and then you’ll have clear sailing: the strategy, in other words, of how to get there the easiest way.
Unfortunately, though, this man is hard to find. In Kansas City, for example, he operates in a downtown office building! In St. Louis, he is hidden in the Air Traffic Control Office, to which you go through the weather bureau, with no signs pointing the way. In Cleveland, he is hidden in the “bomber plant;” his phone is not listed, and I’m not sure that you are welcome when you walk in on him. I suspect he is hidden on purpose. If enough pilots knew about him and his special map, he would be swamped with callers, by radio, by phone, and in person.
Remedy: You remember the one about the country store-keeper who refused to stock a certain kind of cake. Every time he ordered some, he said, people always bought it all up right away. So he had decided to hell with it. Sometimes the official mind (harassed by budgets and civil-service classifications and inter-bureau warfare) runs along similar lines, and restricts a service that threatens to be too popular. I hope in this case it won’t. I hope they will put that man out front where we can get at him; and if that makes him too busy, hire another.
Trouble: I find the weather map not as useful to me as it is supposed to be; and I hanker after another type of map which is actually in existence — the map the Flight Advisory man works by.
The weather map as we have it now may be a better tool for the forecaster than for the pilot. Sometimes I wonder if all the work of drawing all those maps at all those stations all the time really pays off: how many pilots can take a competent look at that map? And of those, how many are really helped by that look?
A Front may mean anything or nothing. Those station models (I mean the weather symbols grouped around a station) are hard to read. Even if you decipher them, their real meaning, for your purpose, is hard to figure. As Leighton Collins has pointed out, discussing the same subject here two years ago, “weather” to a pilot means clouds. He doesn’t really need a forecasters weather map. You might also say he is not interested in the “weather.” All he wants to know is — where is the air transparent and where is it opaque. He needs a cloud-map.
And that’s exactly what the Flight Advisory man has before him — a cloud-map, drawn casual-like with chalk on a blackboard. This map explains — along with his being a pilot — his amazing usefulness. A photograph of such a map at St. Louis was published in the April, 1947, issue of Air Facts and is re-published herewith. When I brazenly walked in on St. Louis Flight Advisory the other day, I got a good demonstration of what such a map could do for us.
The weather that day was a puzzling mess of zero-zero, unlimited, low overcasts with lower scattered, high overcasts with rain — all of it alternating all over the Middle West. Toward the East, where I was bound, instrument weather prevailed and the trends were bad. It was the unpleasant case of flying not only into bad weather, but into weather of which you cannot see the structure, rhyme or reason.
Actually now, the structure was simple and clear. The Polar Front lay right through St. Louis, Indianapolis, Cleveland, about as indicated on the “weather map” herewith. (This “weather map” makes no claim to any accuracy, other than to illustrate my point.) The weather was good-to-excellent right at the Front, and for 100 miles or so to the north and south of it. But the further you got away from the Front, the worse it got. This was, of course, contrary to all the writings, including mine; and therefore hard to discover: who would think of looking for the good weather right along the Front? A thorough study of the standard weather map, with expert help, would no doubt have disclosed that. But for such study there was no opportunity in the busy St. Louis bureau.
Cudgelling my brain for a cagy way to get to New York without sticking my neck out, I went into Flight Advisory. Well, I hardly needed to ask my question: one look at his cloud-map told you all. Here was this channel of cloudless weather, running from St. Louis to near Indianapolis and then up to Cleveland and beyond. All I had to do was follow that slight detour, and I would have to use no brains: no flight plans, no approaches, no clearances and traffic instructions, no fuss. The Flight Advisory man said to do it fast, because the channel would later be cut by over-running warm air from the South in the Pittsburgh-Akron area. And that was all.
A sketch, from memory, of that cloud map is attached. Here again there is no pretense at accuracy; it is merely to illustrate my point; not how different the actual flying weather was from what you would have guessed, giving an equally casual look at the standard weather map.
Now I don’t want to blow up a small occurrence into a big affair. But then again — put it in general terms: that day, I bet, 5 or 10 pilots set out along the Indianapolis-Pittsburgh airway for the East; and I bet most of them got dismally weathered-in, probably, as it turned out, for 2 days; or else they had to do some sharp instrument work. The Weather Bureau actually had the data about an easy, clear flight-channel; but they did not have it at a place where it was easy to get (many small ships are shy of Lambert Field traffic); nor did they put it (as far as standard weather services are concerned) in a form where an average pilot like myself could “get” it. But that cloud-map — any pilot, if he had a chance to see it, would have grasped the situation in three seconds, without any explanations. Your only questions were: Was it still true? And would it last?
I think a cloud-map is what we need. I don’t know what fine points and shadings the Flight Advisory man uses in making his map. He has, of course, lots of other weather information in addition, and his map may be too simple for use without such supporting data. To make it useful for the private itinerant pilot, it would have to be self-explanatory, so that it could be interpreted without help. It might have to have three shades: black for low clouds (say ceilings of less than 1000 feet); grey for ceilings of less than, say, a comfortable 2000; light grey for ceilings of 5000 feet or better; while blank spaces would mean blue sky or extremely high ceilings.
You can go on from there. Suppose we called those three shades “Able,” “Baker” and “Charlie.” Suppose that for a nickel you bought at any communications station an outline map of, say, the Middle West. This map might be divided into areas of, say, 50 by 50 miles, each area named by its main town or its most center range station. Suppose they broadcast the latest cloud map every hour, by reading off: Muncie, Charlie; Ft. Wayne, Baker; Toledo, Able. Could you make your own weather map in flight? Could a small airport make a weather map for itself when wanted? Would not such a map be comparatively simple to tele-print, by wire or radio?
This brings up another idea, cribbed from current practice of the Navy, and from past practice of the CAA. At a Navy clearance office you face a wall-map on which red, green and amber lights show which places are closed, open, or “instrument.” The CAA for a while used to classify weather the same way, and the first time in each weather report was the designation “closed,” “instrument” or “contact.” I liked that idea; I didn’t like the regulations that went with it, because they choked off too much flying. But we do need a one-word description of the flyability of weather.
To make it more useful, I think there should be five classes, rather than only three: say “closed,” “instrument,” “caution” (flyable with close attention, and don’t let it get any worse on you), “standard” (say at least a 2000-foot ceiling and at least four miles visibility), and “clear” — blue sky showing, visibility at least five miles. I admit that the names suggested here are open to obvious objections. I admit it would take much study to find the exact formula which would combine ceiling, visibility and perhaps precipitation into a true measure of “flyability.” The formula might vary slightly from region to region depending on terrain and on the reliability of the weather. It might allow for the fact that you will accept low visibility if due to smoke which would scare you if it were due to light fog. Main idea: give us a one-work statement of flyability of the weather; and perhaps a flyability-map.
Well, that’s off my chest. I am not proud of my troubles in getting weather dope—a good pilot should learn how to get the correct weather just as he learns not to pull the mixture control when he means to put on carburetor heat. But if we want flying to grow, we must try to clear away those secondary difficulties of flying: if they were gone, flying itself would seem much easier to many people.
I am certainly not proud of my “remedies.” It is easy to dream up something that would be nice to have. But to put such things into effect is almost as difficult as to put some dreamed-up improvement on an airplane. The thing as-it-is is about as good as the most expert minds, after years of effort and study, know how to make it. Our weather set-up, coupled with our airways system, is even now surely the world’s best. The forecasting elsewhere may be as good, but nowhere else is there as closely-spaced a reporting system (with an airport at almost every reporting station, and lighted airways connecting them). Nowhere else is the weather-dope quite as accessible to the little fellow who has perhaps only a dry-cell receiver. Nowhere else is the weather-information as happily coupled with navigational guidance as it is in our radio ranges with simultaneous voice-transmission—and the whole thing “gettable” with the cheapest, lightest radio equipment. This system, developed in 20 years of effort, is much more complex than appears on the surface; it functions like clockwork, and it is a sort of intricate clockwork of inter-meshing functions. It shares its communications system, in part, with Air Traffic Control. Then again, the information gathered serves not only us pilots, but also the forecasters, and thus the farmers and the nation as a whole. The weather observers are working (if I understand it right) part of each hour for Airways, part of each hour for the Weather Bureau. The schedules by which observations go on the teletype are closely timed, and the whole communications system is heavily loaded if not overloaded. In short, the technical angles, the legal angles, the personnel and financial angles are most complex; and to an expert eye some of the remedies suggested here may be obviously impractical.
Still, if we don’t let them know what we want, we sure won’t get it. That’s why this gripe list has been written. I feel like a heel, thinking of my pals for example in the Harrisburg, Pa. weather bureau and the one at Phoenix, Arizona, and all the trouble they just recently took to help me; also of NEwtown 9-7570, the La Guardia Flight Forecasters who are worth more to me in my Kollsman work than my left engine. There is so much knowledge and so much talent in the weather service; quite a few of the weather people have started flying privately, and that makes them even more valuable. Our problem is how to get at this knowledge more fully, and make more use of those talents. To that end, these suggestions are respectfully submitted.
- From the archives: What it takes to fly the President - March 15, 2023
- From the archives: The Airphibian - August 19, 2022
- Go or no go: how much ice is too much? - April 1, 2022
As good as it is today compared to 70 years ago, there’s still room for improvement.
My wish for my now VFR centric flying: more reporting stations, every public use airport should be sponsored to have an AWOS (seems like a better safety use of FAA capital funds than ramp repaving, alternatively, AOPA, NBAA, EAA adopt/sponsor gaps). In areas where that would still be sparse, work with public safety entities to host AWOS independent of an airport. Those reports should be updated into the system at least every 5 minutes. If there is concern about maintenance/accuracy, local maintenance remotely observed/approved (FaceTime, etc) and/or tier the AWOS certification levels and note in the reporting. Develop a validation routine running in the background of the data polling system that would ID stations for manual review that are wildly varying or routinely anomalous for the area.
I agree with the author 70 years on…”we always did it that way” still isn’t a good enough answer.
There’s no reason why the states couldn’t put in their own AWOS/ASOS. In fact some do.
Thank you for reprinting this. I love reading Wolfgang’s articles because he has such a keen mind, and such far reaching insight. Look at how much of what he requested is available to us now? There are so many weather tools available that it is hard to keep track of them all.
If you have read my article on 7-day VFR flight planning then you know what would be on my wish list: a 2-week graphical forecast that could be peeled back like the layers of an onion. It should be possible with today’s technology. TWC give local day & night forecasts out 2 weeks, and NOAA has the 7-day graphic forecast I wrote about in my article. There are a lot of weather tools available if you have the time and desire to look.
I see also you have reprinted more of Wolfgang’s articles. I have been away from this site for awhile, so I intend to read all of them. Thank you and keep up the good work.
Have FUN! Fly SAFE! Petehdgs