gl sez: "...what they thought was a forest fire."
If they thought the smoke would rise up like smoke from a forest fire, then they would have been looking for the wrong shape, especially among scattered clouds (a quick search didn't find any evidence that they even knew the smoke would be produced, so the point may be moot). I agree that they would have followed any clue they spotted, but I'm suggesting that a person searching for "smoke" might be looking for a rising column of smoke, like forest fires produce, rather than a trail of smoke on the water.
To Heath's point, would a trail of smoke be expected to remain visible after an hour and 20 miles? I have no experience with smoke from boilers, only smoke from common sources, such as forest fires.
I was discussing this same issue with a friend on August 14, 2007 when I sent him the following:
--------------------------------------------------------
There is a brush fire burning north of Santa Barbara California and I
can see the smoke from my office in Thousand Oaks California. I
measured the distance on a Sectional chart and the distance is a
little more than 60 NM. Of course the brush fire is making a greater
volume of smoke than Itasca but it does show that you can see smoke
from this distance.
AE many not have been able to see smoke from 50 NM due to their low
altitude as I explained in a previous message but is should have been
visible at least 20 NM since that was the visibility reported by
Itasca. (Actually they reported visibility of "9" on a "1" to "9"
scale which means the visibility was at least 20 NM and may have been
a thousand NM, though unlikely, but we don't know so using 20 NM is being conservative.)
Itasca's estimate that the smoke would be visible for 40 NM might have
been optimistic but it is the only estimate that we have so we have to
accept it as being at least approximately correct., surely it exceeded
20 NM.
The main point of my post is that the smoke screen eliminated any
navigation problem associated with Noonan aiming for the wrong
coordinates for Howland. It also eliminates Long's theory that they
flew past Howland southeast bound passing to the west of Howland as
they would have seen the smoke. It leaves the most probable place for
the aircraft as too far to the northwest, more than 40 NM, under a low
cloud deck.
------------------------------------
A short explanation for why this is the case. The distance that an object can be seen at sea is limited by the horizon, the edge of the earth, getting in the way of the line-of-sight. The distance to the horizon in nautical miles is determined by the formula 1.144 times the square root of the height, in feet. (For statute miles the multiplication factor is 1.32 or one-third more than the square root of the height.) These distances are listed in table 8 of Bowditch.
To do the calculation you calculate the distance to the horizon from the height of the object and then you add to this the distance to the horizon from the height of eye of the observer. If the smoke went up only 50 feet the distance to the horizon from that height is 8.1 NM; if it was up 100 feet then 11.4 NM; and 200 feet then 16.2 NM. The distance to the horizon from a submarine's periscope, three feet high, is only 2.0 NM so the distance the smoke could be seen from the periscope at the various heights are 10.1 NM, 13.4 NM and 18.2 NM. The smoke at 200 feet is visible almost twice as far away as at 50 feet from the periscope.
The distance to the horizon for a plane flying at 1000 feet is 36.2 NM so the distance that the smoke could be seen from the plane for the various heights of the smoke are 44.3 NM, 47.6 NM, and 52.4 NM respectively.
This is the reason for "crow's nests" on ships. A lookout in a crow's nest 100 feet above the waterline will be able to see more that 3 times farther compared to an observer on deck only 10 feet above the waterline, 11.4 NM versus 3.6 NM. This is also the reason that lookouts were stationed up in the periscope shears, as high as possible on a surfaced submarine, when searching for enemy ships in WW 2.
The Commanding officer of the Lexington said that the "Itasca was laying a heavy smoke screen which hung for hours." He also said that "the Itasca's smoke plume could have been seen 40 miles or more." Now the captain of the Lexington was not there at the time, so he was relying on reports from others, but the captain of the Lexington would know the capabilities and characteristics of smoke made by ships in 1937. See the
Dowell report.
gl