Consider the possibility that FN actually used the original flight plan that was incorrect. I am not 100% sure but I believe that the return trip that was laid out has a true course of 77.07 degrees (257.07 - 180 = 77.07). It seems like a simple enough mistake that could have been made and not caught unless you re-worked the flight plan to verify it. As proof I would ask how many people have looked at this flight plan and not found this simple error previously? I did not spot it after looking at it on a spread sheet for a while.

This would place you at about 106SM North of Howland at 9:12GMT if you were right on the bogus flight line.

First, keep in mind that there is no proof that Noonan used William's plans since Noonan was a more experienced flight navigator than Williams was and most likely did his own computations, not trusting the work of others. I know I would have in his position and I have always done my own computations in the past, I don't rely on anybody else for this type of work.

What mistake are you complaining about?

I guess you do not understand how a navigator plans a great circle course. To actually fly the great circle would require the plane to be constantly turning because the great circle course is never constant and this is impossible to do. So you lay out intermediate points along the great circle spaced at a convenient distance, Williams used 2° 30' of longitude, and you calculate the latitude at which the great circle cuts those longitudes. Then you calculate the

**RHUMB** line courses between adjoining pairs of points, not great circle courses. That way you do not have to be constantly changing your heading, you fly a constant course for the entire segment leg. Then, at the next point, you make a change in course. Since each segment leg is a rhumb line you can reverse the direction of the flight by adding 180° to each leg. This method approximates the great circle and, for the LAE to Howland flight, adds less than one-tenth of a mile (nautical or statute, take your pick) to the flight compared to flying a perfect great circle.

There are so many places where uncertainty enters into the computation of the compass heading to fly it is silly to do these computations to the high level that you are attempting. First you compute the true course and round off to the whole degree thereby introducing a + / - half degree uncertainty into the true course, a total of a full degree of uncertainty. Then you add in magnetic variation, again rounded to the whole degree, and unlikely to be accurate even to that level, which adds an additional full degree of uncertainty into the magnetic course. Then you allow for the wind and the wind correction angle is unlikely to be more accurate than one or two degrees and usually not that good if using forecast winds so we are up to plus and minus 4 degrees at this point. Then you apply the deviation from the compass correction card, again rounded to the whole degree, so we are up to 5 degrees of uncertainty. And the compass correction card itself was determined by swinging the plane and using another compass to determine the heading of the plane on the ground and the bearing read off the testing compass is only read to a whole degree and is probably not that accurate so we are up to 6 degrees of uncertainty in the compass heading. So you can see it is silly to do your computations to the level that you are attempting. To make this clear by an analogy, let's say you filled your backyard swimming pool with a bucket and then compute how much water is in the pool to the nearest teaspoon. You are doing your computations to the nearest teaspoon.

You are attempting to put too fine a point on these computations.

I

posted this before:

Another thing that people get hung up on is about the need to fly the great circle course instead of the rhumb line course. A rhumb line maintains the same true direction for the entire flight while to follow the great circle you must calculate and then make periodic changes in your heading. The great circle is shorter than the rhumb line so that is why people think you must follow the great circle. However this really only makes a difference at higher latitudes but makes virtually no difference when flying near the equator. The great circle distance between the exact coordinates used by Williams for this leg, 06° 47.000' south, 147° 00.000' east for Lae and 00° 49.000' north, 176° 43.000' west for Howland is 2556.1 SM and the rhumb line is 2556.2 SM, exactly one-tenth of a statute mile longer. I can see poor Mr. Williams computing each segment (14 in all) of the great circle between Lae and Howland by hand using logarithmic trig tables only to save 1/10th of a statute mile. Leaving Lae, the initial great circle course is 079.4° true and it changes in steps so that the GC course approaching Howland is 077.6° true. The rhumb line for the entire flight is 078.1° true, only 1.3° difference. And the two course lines lie close to one another, never more than 9 SM apart which is so close that Noonan would not have been able to tell the difference, he would not know if he was on the great circle course line or on the rhumb line course line. Here is

a link to Mr. Williams chart. and his data form is attached.

The reason that I specified those coordinates so exactly was so that I could compute the distances to the nearest one-tenth of a statute mile. Williams and all flight navigators would only use coordinates to the nearest one minute of latitude and longitude, one nautical mile of precision. When the input data is only accurate to one nautical mile it is improper to calculate a distance to a greater precision than that of the original data but many people do this and it is not valid. Using the coordinates as Williams did, only good to one minute, would make the distance for the GC course 2556 SM and for the rhumb line also 2556 SM, there is no difference based on the level of precision of the data used by flight navigators.

So when you do your calculations give some thought as to what the numbers actually signify.

gl