EXERCISE CATALINA QUEST 2006 - CATALINA AJ160 WRECK SEARCH AND SURVEY

On 2 Feb 44 RAF Catalina AJ160 crashed while taking off from the water strip adjacent to Kelai Island in the northern Maldives. The aircraft was a total loss and sank with all hands on board. One of the aims of Exercise CATALINA QUEST 2006 was to search and locate the final resting place of Catalina AJ160

Kelia Island was home to No 30 Advanced Flying Boat Base which was part of 222 Gp, RAF Koggala, Ceylon. It was formed on 1 Oct 43 and closed on 15 Jun 45. The principle function of the base was as a forward refuelling, rearming and alighting facilities for three flying boats. Generally the facility was only used outside of the monsoon season and was established for 2 Officers and 100 other ranks. In reality the base was rarely up to full strength and contemporary reports suggest that establishment for other ranks was way below strength at only 33 other ranks in Sep 44. During the period of operation 2 Catalina Sqns operated from Kelia; 205 Sqn and 240 Sqn, and during the period May 44 to May 45, 230 Sqn operated with Sunderland flying boats.

RAF Catalina

Aerial photograph of Kelia and lagoon.

The island of Kelai is situated on Tiladummati Atoll in the northern area of the Maldives in the Indian Ocean, Lat N06º 58” Long E073º 12”. The island is approximately 1.7 NM long (North/South) and 0.7 NM wide (East/West). The island is inhabited and is mostly used for agricultural purposes although there is some fishing for local consumption. To the west of the island is a large lagoon which stretches to the west of the island and is approximately 4.1 NM East/West and 2.4 NM North/South. The lagoon is enclosed from the Indian Ocean by an external reef wall which runs the full circumference. The East side the reef wall encloses the island of Kelai. The lagoon is between 15 and 20m throughout apart from around the edges where there was a noticeable shelf of sand against the reef wall, the sand shelf was approximately 6m deep and the depth dropped quickly into the lagoon. There were several navigable entrances into the lagoon, all to the South and probably man made.

The Catalina Squadrons utilized a pre marked strip in the Kelia lagoon to the west of the island. Fixed flare markers were used to mark out the strip, the 3^rd flare marker being central and the only permanent marker, the other 4 were lifted and moved as necessary to facilitate the change in wind directions. For a typical take-off the RAF tender would position to the left of the markers once they had been set. The flying boat would position to the right of the markers adjacent to the 1^st. The take-off would be conducted to the right of the markers with the aircraft becoming airborne after 1,190 yards and between the 3^rd and the 4^th markers.

Catalina AJ160 was tasked with a search and rescue mission on 2 Feb 44. Reports state that the wind was light and North and that AJ160 carried out a conventional take-off on the marked water strip adjacent to Kelai Island. The aircraft became airborne at some point after the 3^rd flare marker and climbed for a short period before crashing back into the sea beyond the 4^th flare marker. The aircraft was destroyed as a result of the crash and a fire was seen to burn on the surface. Although an RAF tender went immediately to the scene no survivors were found. No attempt was made to investigate the crash or retrieve the wreckage. A replacement Catalina was launched to conduct the original search and rescue mission.

Map

Metal Detectors

. Initial contact from the UK with local sources suggested that the location of the Catalina wreck was known within the lagoon. The island of Kelai, which is adjacent to the water strip used by the Catalina detachment, has been inhabited by an indigenous population continuously since the period of the crash to the present day. Consequently, the probability of the local fisherman, or village elders, knowing the exact whereabouts of the wreck was considered high. As a consequence bulky wreck locating equipment such as a magnetometer and echo sounder were left behind to avoid excessive baggage costs.

The expedition arrived mid afternoon at Kelai and a small team went ashore almost immediately to make contact with the locals in order to try and establish the wreck location. We met with the Chief Islander and it was at this point that we learnt a significant proportion of the aircraft wreckage had been salvaged by the locals from another island approximately 30 years earlier in the ‘70’s. Although the Chief Islander did not know the wreck location he was able to put us in contact with a local man who had been involved with the salvage operation.

That afternoon we were taken out to the site of a known piece of wreckage that had been dragged onto the 6m sandy shelf from the main wreck site for salvage purposes. The salvager was in his teens when he was involved in the task and although he knew the location of this piece of wreckage he could not identify the location of where the wreckage had been dragged from or the main wreck site. The wreckage had all the hallmarks of aircraft structure, solid and complex construction, too elaborate to be part of a ship. We were able to snorkel down to the wreckage, which was in approximately 6m of water, and identified a long flat double skinned structure approximately 0.9m in depth by 3.1m wide by 5.3m long. There was little to identify which part of the aircraft it was, roughly symmetrical in construction the wreckage had fixing points on the upper surface of the structure. Study of the aircraft diagram identified the wreckage as a wing section; the symmetrical fixings were hard points for the weapon racks and would have been on the underside of the aircraft wing. We were not able to identify that it was a section of the port wing until much later. Coordinates and details for the port wing are plotted on Chart 1.

Port Wing

Barge on the Shoreline

. In order to better understand the possible wreck location it was important to establish the old RAF pier/jetty location, which was described in detail within the contemporary reports. In particular it was hoped to then track the location of the only permanent fixed flare marker (flare marker 3) that made up the take-off run. Accurate bearing and distance had been provided from the jetty to the marker and finding the 3^rd flare marker provided a known point, which AJ160 would have transited past on its take-off run, and ultimately would have been useful in locating the main wreck site.

On surveying the shore line around the location that the locals reported the jetty to have been, it was not possible to see any conclusive evidence of the jetty. One area was located as probable, due to a clearing in the trees where a track provided access to the old RAF camp (now derelict), a wooden drainage channel still existed leading down across the beach. There was no evidence of any piers or jetty mounds on the shore line either on the beach or in the shallows. A couple of concrete mooring bollards were located in the shallows, but it was not possible to establish whether this was their original location, or whether they had been relocated. Coordinates for the areas identified as the jetty and the barge location are on Chart 1.

The port wing wreck in the shallows, although lying on a clear bearing from the deeper water, did not offer a terrific amount of information regarding the location of the main wreck site. Additionally, it was considered that the wing would have detached during the crash from the main wreckage and it may well have been positively buoyant for a period of time after the crash due to the fuel contents and could quite possible have sunk separately from the main fuselage. It was therefore felt that the location of the fixed middle marker would be able to provide valuable information into understanding the possible main wreck location. However, it was understood that although there were a significant number of variable errors in our estimated position for the 3^rd flare marker, ultimately it provided an arbitrary position in the lagoon in which we had reasonable confidence that AJ160 had transited past on take-off prior to crashing. Details of how the position of the 3^rd Marker Flare was estimated and the 3 possible take-off runs from AJ160 are plotted onto Chart 2.

The first dive in the Kelai Lagoon put four pairs in the water at cardinal points 20m from the plotted location of the third marker using the approximate location of the jetty and the contemporary reports of the third marker location from the jetty. The four pairs were briefed to carry out individual circular searches which would overlap at the extremes of search and cover an area of seabed approximately 80m². The search would be for the permanent concrete fixing point for the 3^rd flare marker. The water was approximately 9m deep in this area; visibility was good at approximately 10m; however, circular searching was hampered by the large heads of coral. All four pairs returned reporting nil finds.

Chart 1

Chart 2

On returning to the hard boat, which had repositioned mooring from the morning and was further away from Kelia Island (and the mosquitoes), the expedition received a considerable piece of luck in that another unknown section of wreckage had been sighted from the top deck of the hard boat. The new piece of wreckage turned out to be the whole of the starboard wing, minus engine but complete with wing float. Coordinates for the starboard wing are at Annex B and are plotted onto Chart 1.

As a concurrent activity to the first phase of diving, a shore party had been on Kelai to locate the main RAF detachment camp and try and more accurately plot the location of the jetty. Although the camp location was found it was not possible to obtain any further information on the jetty. There were so many errors with the estimated position of the 3rd marker buoy, the jetty location could have been up to 30m north or south from the location plotted. The contemporary report that the marker was on a bearing of 175°M and 1750 yards from the jetty could also have had inaccuracies. The consequence of this meant that our potential search area for the fixed marker could be vast for what was a relatively small target; furthermore, it could have been buried under sand, encrusted in coral and undistinguishable as an anchor, or simply have been lifted and removed. If we found the anchor, it would still not provide a great deal of information as to the actual wreck site location, it was an arbitrary point in the sea which the aircraft would have passed adjacent to on its take-off run; however, what bearing the aircraft was on when it took off, and where it subsequently crashed were all subject to fierce debate. What was known was the location of the starboard wing, the section of the port wing and the fact that finding the 3^rd flare marker could take hundreds of man-dives to achieve. The next phase of diving would focus on this information.

Chart 3

Chart 4

With no clear local leads as to the whereabouts of the main wreck location we were left with large tracts of the lagoon that could have been potential locations for the wreck site. Time, as always, was limited and it was important to try and eliminate as many areas as quickly and as effectively as possible. In order to narrow down the search area several assumptions had to be drawn up as follows:

Wreckage is within the lagoon and is deeper than 6m. The wreckage lies inside the lagoon and is in the deeper water. The furthest extremity north being the shallow water sand shelf that marks the end of the lagoon before the coral wall and open ocean. (It was believed that if the wreck was any shallower than 6m then the site would be well known to the locals).

Wreckage is no further south than the 4^th Flare Marker. Although some of the wreckage may well have floated after the crash, there were no noticeable currents during the period that we were in the lagoon. Additionally, the contemporary reports that the wind was light from the north. It was therefore considered that although the wreckage may have floated for a period it was unlikely to have moved far from the original crash site, and the crash site would have been beyond the 4^th Flare Marker.

With these assumptions in force it still left a large tract of lagoon to search. Although we now knew the main wreck had been lifted for salvage it was hoped that debris from the crash site would still be evident on the lagoon bottom. The likely area of the wreckage is plotted on Chart 3. The strategy for the second phase diving was as follows:

Stbd Wing to 4^th Flare Marker. Eliminate the area immediately around the starboard wing location and in particular the area in the immediate direction of the take-off runs.

NW, N and NE Take-Off to 4^th Flare Marker. Eliminate the area to the north of the 4^th flare marker on the take-off runs NW, N and NE.

To carry out both these tasks divers were put in along a fixed line at 10m intervals to swim a set bearing while scouring the bottom for wreckage. The method was very similar to a jackstay search but without the jackstay. Although fairly crude in practice, with divers veering of course by several degrees over a distance of 1.0 to 1.5 km it did allow for large areas of the sea bed to be covered with relatively few man dives.

Starboard Wing to 4^th Flare Marker. The first search was set from the starboard wing with 2 waves of 4 divers swimming in a NE direction towards the approximate location of the 4^th flare markers. The seabed in this area shelved steadily up from 16m at the wing to 10 to 12m when the divers reached the approximate location of the flare marker. Apart from coral heads no sign of wreckage of any description was found.

NW Take-Off to 4^th Flare Marker. The second search started at the edge of the lagoon at the most NW extremity of the NW take-off run. One wave of divers entered the water and swam on a bearing towards the 4^th flare marker. The water shelved down from 6m to approximately 10 to 12m. Other than occasional coral heads no sign of wreckage was found.

N and NE Take-Off to 4^th Flare Marker. A similar search to that conducted on the NW take-off path was planned for the N and NE runs. However, the depth of water in both these areas was found to be little deeper than 6m. As a result both areas were discounted as likely wreck locations.

Serial	Item	Coordinates	Notes
(a)	(b)	(c)	(d)
1	RAF Barge	N 06° 56 698 E 073° 12 875	Wrecked steel barge on beach, reported to be an abandoned RAF barge that was used to move ammunition and stores out to the aircraft.
2	Wooden drain	N 06° 56 539 E 073° 12 869	Wooden drainage ditch running down the beach from a clearing for a track in the jungle. Most likely area for the old RAF jetty.
3	Concrete mooring blocks	N 06° 56 681 E 073° 12 858	Abandoned concrete mooring blocks, possibly used to the barges.
4	3^rd Flare Marker	N 06° 56 536 E 073° 12 041	Estimated position of the fixed 3^rd flare marker. Position derived from a bearing of 259°T at 1700 yards.
5	Port Wing	N 06° 56 882 E 073° 11 273	Heavily salvaged section of port wing situated on the shallow water shelf at 6m. Wreckage is 5.3m x 3.1m x 0.9m lying on an orientation of 150°M.
6	Starboard Wing	N 06° 56 491 E 073° 12 484	Complete starboard portion of wing minus engine. Starboard wing float stowed within wing tip, orientation of wing is 025°M.
7	AJ160 Main Wreck Site	N 06° 56 638 E 073° 11 444	Centre of wreck site of Catalina AJ160. Area heavily salvaged with wreckage spread over an area of approximately 100²m. Area of seabed around wreck site is heavily contaminated with oil.

Co-ordinates

Search Briefing

Compass Search

Success - Engine Cowl Panel in Debris Field

From analysis of the second phase of diving a more detailed picture of the shape and depths of the lagoon was developed. We were confident that the main wreck did not lie to the NE of the starboard wing. Additionally, we were able to discount the N and NE take-off run, in both these areas the water was only 5 to 8m deep and it was considered unlikely that the wreck would be in this area and remain unknown to the locals. The NW take-off run was slightly deeper but did not produce any conclusive results. A more detailed inspection of the starboard wing had revealed that it was relatively intact and certainly showed no signs of salvage action; as a result we concluded that the wing had remained untouched from when the aircraft crashed. The port wing had obviously been dragged and heavily salvaged; however, the longitudinal bearing of the wreckage (169°M) lined up with the area of the starboard wing.

We had now eliminated a great deal of the search area. The main area which remained was the area of lagoon bottom between the starboard wing and the port wing and the area to the East of this line towards the 4^th Flare Markers. It had been noted earlier that the bearing that the port wing lay on was directly from the starboard wing and this suggested that when the wing had been dragged from the deep water, it had been brought up by the shortest route from the same area as the starboard. There was therefore a good possibility that the main wreck lay in this position.

The search area was divided down into search boxes of 185m x 185m, one corner of each box was shot. Two pairs of divers were put into each box, one pair per wave, and swam a bearing timed pattern to search that box. Again crude, this search method gave us the capability to saturate an area of the lagoon bottom and quickly eliminate it from our search while maintaining a reasonable degree of confidence that any significant area of debris would be discovered. A detailed plot of the boxes to be searched is included on Chart 4.

During the subsequent search period one pair of dives swam on to the edge of a debris field. The search pattern was relocated to this box and a debris field of approximately 100m² was found. There were no pieces of wreckage larger than 1m² though some of it could have been buried under the sand. Noticeably in the centre of the wreck site was a large area of sand/silt that was contaminated with oil. There were a number of items of wreckage that looked familiar with the Catalina diagram; in particular items such as radio racks from the main fuselage, and engine exhaust ports from the engine cowlings. One pair of divers in an adjacent box reported finding a small anchor similar to the type used on Catalina aircraft; however, the site was not marked and despite extensive diving the anchor was not relocated.

The quantity of wreckage in this area, the types of wreckage and its proximity to the starboard wing strongly suggested that this was the location of the main crash and that this was the site that the aircraft had been salvaged from. No sign of the aircraft engines were found either at the main wreckage area or at the wing location. Additionally, there was no sign of depth charges that the aircraft was reportedly carrying. Local salvers did make comment of items with the description of depth charges and reported that they returned them to the lagoon, though their exact location is unknown.

All of the fuselage of Catalina AJ160 had been salvaged along with a substantial amount of the port wing. Although the starboard wing was largely intact it was not possible to locate identifying marks, serial numbers or identification plaques. Both engines were missing from the wreckage and it was not clear whether these had been salvaged with the fuselage, or were lying elsewhere within the lagoon. Although no positive identification could be found there are no other reported Catalina losses within the Kelai lagoon area, furthermore, the starboard wing was unmistakably that of a Catalina in size and shape; the starboard wing float was present and stowed in the wing tip and provided further evidence. Although conclusive positive identification could not be achieved there is 99% confidence that the main wreck site and remaining wings are from the Catalina aircraft AJ160.

Contemporary reports state that Catalina AJ160 used the N take-off path on the night of 2 Feb 44 before crashing into the lagoon shortly after take off. The evidence found during the search neither verifies, or disagrees with this account though the wreckage is someway from where the anticipated crash sight would be for a northerly take-off. The lack of main fuselage due to salvaging severely limits any further possibility of establishing the true cause of the crash.

Analysis of the location of wreckage within the lagoon and reports from the salvers suggests it is likely that the wings broke free from the fuselage as a result of the impact with the sea and the two parts separated from the main fuselage. At the same time the force of the crash is likely to have catapulted the engines free from their mounts to sink immediately. The salvers reported that when they found the main fuselage it was split in two behind the wing box and it is likely that this damaged occurred as a result of the crash. The main fuselage would have sunk relatively quickly and this would explain the relatively small and compact debris field. The wings, now free from the fuselage, without the weight of the engines and full of aviation fuel could have remained positively buoyant for a longer period. Low in the water they would not have been significantly influenced by the northerly wind and with no substantial currents within the lagoon they settled relatively close to the main fuselage, though thankfully the starboard wing was sufficiently far from the main site not to be found be the salvers.