by Samuel Halpern
Copyright © 2007 Samuel Halpern, all rights reserved.


We all know that the Titanic sank about 2 hours and 40 minutes after colliding with an iceberg on her maiden voyage on the night of April 14, 1912. Thanks to the discovery of the wreck in 1985, we now know exactly where she foundered. But what does that tell us about the location of where the collision actually took place?

This question was one of several questions put before the accident investigators of the Marine Accident Investigation Branch (MAIB) of the Department of Transport in Southampton in 1992 as part of their reappraisal of evidence relating to the Californian affair. The conclusion of both MAIB investigators was that the Titanic was in the approximate position of 41° 47’ N, 49° 55’ W when she struck an iceberg at 23:45 (11:45 p.m.) ship’s time on April 14, and in position 41° 43.6’ N, 49° 56.9’ W when she foundered.

How was this collision point derived? As explained in the MAIB report written by the Deputy Chief Inspector James De Coverly:

“This is the one almost fully substantiated piece of new evidence since the 1912 Inquiry. Dr. Robert Ballard, leader of the expedition which found the wreckage, gives the position of the boilers and stern section, and the Inspector supports his view that these very heavy items will have sunk almost straight to the seabed: their position must therefore be very close indeed to the position of sinking. I agree. The position is 41° 43.6’ N, 49° 56.9’ W. This will not of course be the position of the collision, as the ship must have drifted some distance before she foundered; how far and in what direction will have depended entirely upon the current, for the night was calm with virtually no wind. The current is discussed below, for it is an important feature in this reappraisal: there is strong evidence that it was setting a little west of south at rather more than 1 knot. Allowing such a current, and working back from the position of sinking, the position yielded for collision with the berg is approximately 41° 47’ N, 49° 55’ W. This position is substantially different from that given by Titanic in her wireless distress messages and accepted by the Court of Formal Investigation, namely 41° 46’ N, 50° 14’ W.”

And in the discussion about the local current, De Coverly wrote:

“The position of the flotsam as given by the Californian, when compared with the position of sinking as now established, is further evidence of a southerly set and - assuming the position to be correct - allows it to be quantified: the direction of set was about 196° true and the rate about 1.3 knots. There is still further suggestive evidence in support of these figures in that, when run back to give the likely position of collision, the position arrived at, though different from that sent by Titanic in her distress calls does lie practically on the line of her course through that position. Perhaps the error in the position as transmitted was caused by the wrong distance being allowed along the course line from the last known position – a simple mistake to make under stress…I think there can be no reasonable doubt that a current setting about south by west at something like 1¼ knots existed in the area of the accident.”

What the report did not do is provide the specific details of their work.

In this article I will try to explain how the MAIB derived their estimate for the collision point that they gave. What I will also explain is why this may not be quite right even though it would not be too far off. I will also show how we can derive a reasonably confident estimate of where the Titanic came to a final stop in the water several minutes after the collision. I will then explain how we can get a better feel for where the collision with the iceberg took place when considering her post collision maneuvering.

The MAIB’s Collision Point

The key component in the MAIB’s derivation of the collision point was an estimate of the local current. Based on information provided in the log of the Californian, and presented in testimony before the British Wreck Commission, the Californian left the scene of the floating wreckage at 11:20 a.m. on the morning of April 15, 1912. The position of that wreckage was given as 41° 33’ N, 50° 01’ W. The position of the Titanic wreck site used by the MAIB was 41° 43.6’ N, 49° 56.9’ W, which is the position of the stern section on the bottom of the Atlantic as given by Dr. Robert Ballard. The Titanic was reported to have foundered at 2:20 a.m. What the MAIB inspectors did was to take the distance between those two locations, 11.04 miles, and divide that by 9 hours of time to get an average current drift of 1.23 knots, which was referred to as “something like 1¼ knots” in De Coverly’s report.[1] The course line between the two locations gave them a current set of 196.2° true. To get to their collision point they then applied this current in reverse for 2 hours and 35 minutes, the time difference between 11:45 p.m. on April 14 and 2:20 a.m. on April 15. That would take them 3.17 miles at 016.2° true (the reciprocal of 196.2°) to a point at 41° 46.6’ N, 49° 55.7’ W. But it also appears that they allowed the ship to drift about ½ mile westward after striking the iceberg before it became dead in the water. Thus they backed the stopping point by that amount to get an assumed collision point at 41° 47’ N, 49° 55’ W when they rounded the coordinates to the nearest minute of arc.

The chart below shows the key locations of the wreck site, the observed wreckage on the morning of April 15, the SOS position given out in Titanic’s wireless distress messages, the 266° true course line used by Titanic’s fourth officer Joseph Boxhall in getting to that position, and the estimated position of the collision point derived by the MAIB inspectors.

A Navigational Reappraisal

When one looks closely at the methodology used by the MAIB inspectors there are a few things that come to light. First was the assumption that the position of the flotsam reported in the log of the Californian was the correct position of the wreckage, and that the current can be estimated by dividing the distance between that position and that of the wreck site by 9 hours. A second observation is their apparent assumption that the collision point was slightly east and a little north from the ship’s final stopping point as can be seen in the chart above.

So let us examine both of these and see what we may need to change in order to get a slightly more realistic solution.

Captain Lord of the Californian had estimated the location of Titanic’s wreckage by working backward from his noontime observation of the sun that was taken after he departed the wreckage area Monday morning. This procedure, where you apply your course and speed for a known interval of time to derive another position is called dead reckoning, and the position you obtain is called a “dead reckoning position,” or simply a DR. It does not take into account the effect of any current or wind on the movement of the ship which is generally unknown to begin with. Captain Lord’s noontime position was fairly accurate as it involved a noon observation of the sun which put him at 41° 33’ N, 50° 09’ W. This was a noontime fix.[2] About the time this observation was taken, the Californian was coming out of a large ice field heading westward. The ice field stretch from far to the north to far to the south and was reported as being 5 to 6 miles in width. As the Californian was coming out of this pack ice the Frankfurt was seen steaming down parallel to the western side of the field coming from the northwest and heading SSE. Both ship’s spotted each other within minutes of local apparent noon. From the Frankfurt, the Californian was seen off their port bow. Captain Hattorff of the Frankfurt reported reaching 41° 35’ N, 50° 15’ W at a time that corresponded to 15:20 GMT,[3] which for that date and location, was 1 minute before local apparent noon. From these two position reports, the two ships would have been less than 5 miles apart.

So with this observational evidence, it appears that Californian’s position at local apparent noon on April 15 was correct. But what about the location of the wreckage?

According to Captain Lord’s testimony, the time they left the floating wreckage and abandoned lifeboats was 11:20 a.m. ship’s time.[4] Reading from his logbook: “11.20 proceeded on course N. 59 W. by compass.” When asked by Sir Robert Finlay what would that be in degrees true (as opposed to degrees magnetic), Lord replied: “I think I was intending to make N. 89 W [271° true]. I think that was my intention. The variation is 23½, and I think the deviation was 5.” Captain Lord was also asked about the speed they were going after leaving the wreckage. His reply was simply: “I went slow. I came through the ice full speed to the ship [Carpathia], but I went back slow.”

According to Californian’s Chief Officer George Stewart, between 11:20 a.m. and noon they traveled some four or five miles going through the ice. The actual distance between the reported locations is almost exactly 6 miles. But we also know from Captain Moore of the Mount Temple that the ice field was about 5 to 6 miles in width down in the vicinity of the wreckage. This tends to support the distance traveled that Stewart estimated and agrees very well with the 6 miles between the coordinates of the wreckage and the coordinates of their noon sun sight.[5]

How fast was the Californian going as it cut through the ice? GMT of local apparent noon for the Californian on April 15 at longitude 50° 09’ W was 24 hours and 11 minutes later than GMT of local apparent noon for April 14 when she was at 47° 25’ W on her way to Boston. This means that the time of local apparent noon on an unadjusted clock set at noon the previous day would be 11 minutes past 12 o’clock. Therefore, the difference in time between leaving the wreckage at 11:20 a.m. and local apparent noon when they took that sun sight was 51 minutes, not 40 minutes as one might assume it to be.[6] A distance of 6 miles in 51 minutes gives a speed over ground of just over 7 knots. As we have seen from Captain Lord’s statement, the ship’s heading was 271° true.

Finding Current Set and Drift

If the local average current is known, a navigator can apply that as a correction to a DR position to get what is called an “estimated position,” or more simply an EP. However, the local current on the morning of April 15 was not known, so the position of the wreckage was a DR position based off of the Californian’s noontime fix. What Captain Lord did was take his noontime location, 41° 33’ N, 50° 09’ W, and run it back at a speed of 7 knots till 11:20 a.m. on the reciprocal of his course heading, or 091° true. That gave him the position of 41° 33’ N, 50° 01’ W for the wreckage when rounded to the nearest minute of arc. It was a DR position. It was not a fix, nor was it an EP.

To determine the average current a navigator needs to determine the distance and azimuth angle between a second fix and its relation to the DR position for the same point in time. Course and speed changes made between the two fixes only affects how the DR is obtained, not the determination of the current.

As an example, take the situation shown on the chart below.

Here we have a tramp steamer heading due east at a speed of 7.5 knots. Their last fix was at 08:00 at 41° 40’ N, 51° 00’ W. Four hours later, at 12:00, they would have traveled 30 nautical miles due east to a DR location at 41° 40’ N, 50° 20’ W as shown. But at 12:00 a another fix was taken and they found that their ship was really at 41° 36’ N, 50° 16’ W. This location is 3 miles east and 4 miles south of their DR position, or 5 miles at 143.1° true from the DR. They got there by a current that took them 5 miles off their course over a period of 4 hours. The speed of the current was therefore 1.25 knots.

Now suppose that at 11:00 they had to stop because of engine problems which took over an hour to repair. At 12:00 they took another fix and found that instead of being 22.5 miles (3 hours at 7.5 knots) due east of their 08:00 fix, they were in fact 3 miles further east and 4 miles south of their 11:00 DR, which again is 5 miles at 143.1° true from that location. But since they were stopped for an hour, their DR position for 12:00, the time they took their second fix, is the same as their DR for 11:00 (which was at 41° 40’ N, 50° 30’ W). So once again they find that a current had set them 5 miles off their course over a period of 4 hours, and again the speed of that current works out to be 1.25 knots.

This second scenario is shown in the chart below.

In both scenarios we get the same current, running 143.1° true at 1.25 knots. The current was obtained both times by observing the difference between the location of their second fix and their DR position for that same exact time. The speed of the current was derived in both cases by taking the distance between the fix and the DR, which was 5 miles, and dividing that distance by the time between the two fixes, which was 4 hours apart in both cases. What was different between the two scenarios is the track of the ship over the ground as can be seen by comparing the two diagrams. But the derived current, both speed and direction, came out the same.

In the MAIB solution, the starting time was 2:20 a.m. at the position where the Titanic sank, a position that can certainly be regarded as a precise navigational fix. Their ending time was 11:20 a.m. when the Californian departed the wreckage area. Their result for the strength of the current was overestimated somewhat because they used the wrong time interval. They should have used the time that the noon sun sight was taken upon which that DR location for the wreckage was based. They also assumed, as in the British Wreck Commission’s report, that ship’s time on the Titanic was the same as ship’s time on the Californian. Although the time on their clocks were not too far apart, they were not the same.

We will correct these apparent oversights in the next section. We will also take into account the evidence that strongly suggests that the Titanic turned northward immediately following the collision, and at one point, her engines were put ahead again for a short period of time after first coming to an initial stop. This will then be used to get a somewhat more realistic estimate for where the collision with the iceberg took place.

Re-Working the Problem

In looking to solve the problem, the approach I took was to run time in reverse. My starting point was the April 15 noontime fix of the Californian, 41° 33’ N, 50° 09’ W. The first leg of the journey is to run at a speed of 7 knots on a heading of 091° true (the reciprocal of 271° true) for 51 minutes to get to the DR where the wreckage was seen last. Then, as in scenario 2 above, we will remain dead in the water floating with the wreckage in reverse until we take another fix. That fix is the location where the Titanic sank. The time for this event is taken at 3 minutes before 2:20 a.m., when all of her lights went out.

Lookout George Symons in lifeboat No. 1:

“I saw her lights go out, all her lights. The next thing I saw was her poop...It righted itself without the bow; in my estimation she must have broken in half...I should think myself it was abaft the after expansion plate.”

Symons said the stern disappeared from view about 2 to 3 minutes after he saw the ship break in two.

Over on the upper bridge of the Californian the lights of the steamer that had been firing white rockets had suddenly disappeared from view. As observed by Apprentice James Gibson:

“Just after two o'clock, she was then about two points on the port bow, she disappeared from sight and nothing was seen of her again.”

Gibson was then sent down to inform Captain Lord that the lights of the steamer they were observing had disappeared. When asked for the time, Gibson replied, “Five minutes past two by the wheelhouse clock.”

Titanic’s clock’s had been set last to read 12:00 at local apparent noon on April 14.[7] Her longitude at noon on April 14 was approximately 44° 31’ W. Californian’s clocks were set at local apparent noon on April 14 to read 12:00 when she was at a longitude of 47° 25’ W. The difference in their noontime longitudes would have meant about a 12 minute difference in their respective clocks. A time of 2:05 on the Californian corresponded to 2:17 on the Titanic, or 3 minutes before the stern section disappeared beneath the surface. And as observed by George Symons, this was the time that Titanic’s lights went out as the ship apparently broke apart.

Using Californian time throughout, the difference in time from 2:05 a.m. to when Californian’s officers took their noon sighting of the sun is 10 hours and 6 minutes,[8] the time interval between fixes that we are looking for.

For the position of our second fix, the Titanic wreck site location, I took as the center of the boiler field given by Dr. Robert Ballard. This is located at 41° 43.5’ N, 49° 56.8’ W.[9] The difference in distance between this fix and the DR position for the wreckage (at 41° 33’ N, 50° 01’ W) is 11.0 nautical miles. Dividing this distance by 10 hours and 6 minutes gives a current drift of 1.09 knots. The set of the current would be the course line from the center of the boiler field to the DR location which works out to 196.7° true.

Now that we have a good estimate for the average current, we can use this to get an estimated position for the wreckage at 11:20 a.m. instead of using a DR position. This is obtained by applying the current in reverse for 51 minutes of time starting from the DR position and traveling on a heading of 016.7° true. The result is an EP for the wreckage at 11:20 a.m. at 41° 33.9’ N, 50° 0.6’ W, almost a mile northward from the DR.

All of these key locations are shown in the area chart below.

Titanic’s Final Stopping Point

With the derivation of the local average current we can now find the position of the Titanic when she came to a final stop after colliding with the iceberg. This not the same as the collision point. To find the collision point requires some knowledge about the after collision movement of the ship. But the final stopping point can be estimated with a higher degree of certainty because it does not depend on any assumption other than the ship was dead in the water at a certain point in time.

We know that the Titanic did not stop immediately after striking the iceberg. We know this from several sources including greasers Frederick Scott and Thomas Ranger, trimmer Thomas Dillon, Second Officer Charles Lightoller, fireman Alfred Shiers, and passengers Henry Stengel, George Rheims, and Lawrence Beesley. The engines were reported to have ran on for about a minute or two before coming to a stop and were then backed slowly to take the way off the ship before they stopped again. We also have some evidence that the engines were restarted ahead again for a very short time from Scott, Dillon, and Beesley, and from quartermaster Alfred Olliver and passenger William Carter. There is some speculation that this was done to move the ship away from some nearby ice shortly before they gave the order to uncover the lifeboats. Beesley had observed the ship moving ahead slowly just a few minutes before going off the boat deck. On his way down the second class staircase he noticed a crew member starting to uncover lifeboat No. 16 on the port side. The order to uncover the boats came about 15 to 20 minutes after the collision.

Over on the Californian Third Officer Charles Groves observed the steamer that had approached them from abaft their starboard beam appeared to have stopped with most of her deck lights shut out from his field of view. The time on the Californian was 11:40 p.m. He knew the time because the quartermaster struck one bell to inform the oncoming watch to get ready to come up in 20 minutes to take over the watch on deck. Captain Lord joined Groves on the upper bridge about five minutes later, 11:45 p.m., to look at the lights of the stopped steamer. On the Titanic it now would have been 11:57 p.m., and close to the time when the order was given to uncover the boats.

What I have done was assume that by 11:45 p.m. Californian time (11:57 p.m. on the Titanic), the Titanic was dead in the water for all practical purposes. As noted above, the time when her lights disappeared from view was 2:05 a.m. Californian time. The difference in time is 2 hours and 20 minutes during which the ship would have drifted with the current. Working back for that amount of time from the wreck site location places the final stopping point of the Titanic at 41° 45.9’ N, 49° 55.8’ W, which when rounded to nearest minute of arc is 41° 46’ N, 49° 56’ W. This final stopping point is shown on the chart below along with the location of the wreck site and the EP of the wreckage at 11:20 a.m.

Collision Point

To come up with an estimate for the collision point we need to know something about the movement of the Titanic before, during and after the collision. According to lookouts Frederick Fleet and Reginald Lee, the Titanic had veered to port shortly before striking the iceberg. This is supported by Quartermaster Robert Hichens who said he received an order to hard-astarboard the helm shortly before the impact came. We also know from standby Quartermaster Alfred Olliver that an order for the helm to be put hard-aport was given as the berg was seen passing down the starboard side of the ship. We also know from Quartermaster George Rowe stationed on the poop that as the berg passed his location the ship was not under starboard helm. And we also know from seaman Joseph Scarrott and fireman Alfred Shiers that the stern of the ship was veering away from the berg which slowly disappeared off the starboard quarter. It is reasonably certain that first officer William Murdoch had ordered the helm shifted to port in order to clear the ship’s stern as the iceberg glided by along the starboard side. If he had not done so, the allision with iceberg would probably have opened up the entire starboard side of the ship resulting in the ship capsizing soon thereafter.

The collision sequence, derived by taking into account the known turning characteristics of the ship and its speed of approach, is shown in the animation sequence below. The animation frames are in 3.75 second increments but speeded up by a factor of three. The approach course was 266° true, and the scale shown is 250 feet per grid line. As depicted in a sketch drawn by Frederick Fleet, the iceberg was placed ahead of the ship fine off the starboard bow as the ship approached it at a speed of 22½ knots.

As seen in the animation above, the Titanic was turning to starboard soon after striking the berg under port helm. The ship also started to slow as the engines came to a stop and then were backed for a short time to take the way off the ship. Somewhere during the starboard turn the helm was steadied up. If a specific direction was not ordered, the helmsman, Robert Hichens, would have probably steadied the ship’s head onto one of the points on the compass which was directly in front of him. It is likely, but not a certainty, that he would have chosen a cardinal point. In this case it would be magnetic north. Indirect evidence for this comes from Quartermaster George Row who testified that “her head was facing north” when he was on the bridge assisting with the firing of distress socket signals and working one of the Morse signaling lamps out on the bridge wing. As a quartermaster, it would be quite natural for him to have looked at the compass as he moved about the bridge while doing that work. And his reference to direction would be magnet which differed by about 2 points from the ship’s geographic heading at that location. North magnetic would have corresponded to NNW true. As the ship came to a final stop, its head may have fallen off a bit as the rudder lost effectiveness as the ship slowed to a dead stop.[10]

The diagram below shows the probable track of the ship over ground from just before the collision to a few minutes after the collision based on the movements described above.

After the engines came to a stop after being backed for a minute or two to kill the ship’s headway, the ship was making about 4 to 6 knots through the water, something observed by Second Officer Charles Lightoller after he got out of his cabin a few minutes after the collision to look around. The ship would then have continued to drift forward for the next few minutes before the engines were restarted again before stopping for a final time. The net result is that the ship likely traveled as much as ¾ of a mile west and ½ mile north from the collision point before becoming dead in the water close to midnight on the Titanic. Working back from our EP for the Titanic’s final stopping point, the collision point would have been close to 41° 45.5’ N, 49° 55’ W, or about a mile and a half south of where the MAIB estimated it took place.

The diagram below shows this collision point in relation to the estimated position of the final stopping point and the wreck site location. The track of the ship over ground is also shown, and includes the effect of the local current on the ship’s course line prior to the collision.


Based on a confirmed noontime position of the Californian on April 15, 1912, her reported DR position when she departed the area of the wreckage at 11:20 a.m., and the location of the Titanic wreck site, we were able to derive the average local area current for the region. Using this estimate we were then able to derive an estimated position (EP) for the wreckage at 11:20 a.m., and an EP for where the Titanic came to a final stopping point close to midnight. Based on the likely movement of the Titanic that followed the collision, we were also able to derive an approximate position of where the collision took place. This point was located approximately at 41° 45.5’ N, 49° 55’ W, or about a mile and half south of the collision point derived by the MAIB.

[1] The initial reference in the report of “about 1.3 knots” appears to be either a typographical error, or possibly a calculation made by De Coverly while reviewing the work of the appointed inspector Thomas Barnett. If the latter, it seems to have come about by measuring the distance from the wreck site taken along an extended line from the wreckage position through the wreck site to where that line intersected Boxhall’s 266° course line. That distance, wreck site to intersection, is about 3.5 miles. Dividing that by 2 hours 40 minutes gives a speed of about 1.3 knots for the drift of current.

[2] The noontime fix is really accurate in latitude. The longitude is not as accurate and has to be worked up to that time to get both coordinates. It is really what is called a running fix. A little after 7 a.m. that morning, the Californian exchanged positional information with the Mount Temple by wireless. The Mount Temple obtained a very accurate longitude line just a few minutes before 7 a.m. when she took what is called a Prime Vertical sight of the sun.

[3] George Behe, “The Frankfurt Incident,” Part 1, THS Commutator, 1990 Vol. 14, No. 3. The time of this sighting in this reference was given as 12:10 p.m. which was Californian’s time on a clock still set for April 14. But that time was 1 hour 50 minutes ahead of NY time, or 3 hours 10 minutes behind GMT. If we add 3 hours and 10 minutes to 12:10, we get 15:20 GMT, or one minute before local apparent noon for that particular longitude on that date.

[4] British Inquiry, 7022, 7267-7270.

[5] The heading was 271° true, which taken over a run of 6 miles west, results in going 0.1 mile north, which is irrelevant when rounding off the numbers to the nearest minute of arc.

[6] Californian’s clocks were last adjusted at local apparent noon on April 14.

[7] It was the practice on White Star Line vessels to set the ship’s clocks at midnight so that they would read 12:00 at local apparent noon about 12 hours later. If necessary, the clocks were corrected in the forenoon so they would be accurate when the noontime observation of the sun was taken.

[8] From 02:05 to 12:00 is 9 hours and 55 minutes. To this we add the 11 minutes to get to the time for local apparent noon on a clock that was set at noon the previous day.

[9] In terms of degrees, minutes and seconds, this location was given by Ballard as 41° 43’ 32’’ N, 49° 56’ 49’’ W. The heavy boilers, which came out of boiler room No.1 at the break location, would have gone almost straight down to the bottom at a relatively high rate of speed.

[10] There is some evidence to suggest that when the ship came to a final stop her head was pointing close to North-by-West ½ West magnetic, and then very slowly started to swing back toward magnet north before she foundered. (British Inquiry, 17667-17674.)

| Home | Back to Top |