Tank Cave was first dived in the early 1960's. The entrance to the Cave is a 2.5m high vertical entrance tube which was originally covered by a very heavy cement-lined water tank. It was necessary to move this water tank to access the cave, hence the name Tank Cave. From the base of the entrance tube you had to crawl through a 300mm gap which ran between a mud heap and the roof of the cave. This led into a chamber containing the very muddy entrance lake leading to the as yet unknown extensive tunnel system.

Because of the difficulties in accessing the tunnel beneath the water very little exploration was done at this time. The first 10 metres from the entrance lake down a low, very silty passage with line traps proved difficult. It was not until 1983/4 that Tank Cave was dived again. Around 150 metres of various passages were found at that time. Again the difficulties in negotiating the entrance tunnel discouraged further exploration.

In May 1989 a further attempt was made. This time the cave was dived to a distance of 240m from the entrance (to the present B3 intersection) with a number of potential leads observed.

After this dive it was decided that a considerable amount of effort needed to be put into further exploration. Approval was obtained from the owner to dig an access path through the mud heap in the dry section of the cave to allow easier access to the water. Work on opening up the first 8m of the water filled passage was also begun. This involved moving a number of large rocks from the tunnel and also removing some large slabs from the ceiling to allow back mounted twin 100's to be used. A heavy line from the surface to A2 needed to be laid and securely pegged in position to avoid the line traps. This section was extremely silty.

Further exploration over the next two months extended the length of the system to over 3.5km. Several major tunnels with interconnecting tunnels running between them were discovered. The more the cave was explored, the more complex the maze of interconnecting tunnels became.

Chris Brown and Phil Prust in Tank Cave | © Peter Rogers

It became obvious, that due to the complexity of the tunnels, a guidance system was necessary. Divers that did not have a good knowledge of the cave found difficulty in navigating to reach a given point in the tunnel system. To overcome this problem it was decided to number all intersections, both underwater and on the map. Initially tags were put on the intersections already discovered, and as new tunnels were found, these followed on. The numbering started at A1-A2-A3 and so on. These were later changed to reflect what tunnel/section you were in (ie all the "S" section labels start with "S" ; all the "L" section start with "L"; etc).

Tank Cave did not give its secrets up easily. Entrances to new sections were difficult to find. You could look in ninety nine dead ends and the one hundreth could be the next big find. Major finds were the "S" section in June 1993, the Easter Extension in April 1994 and the Gateway Rockpile tunnel in May 1995.

Tony Richardson & Greg Bulling after discovering Easter Extension | © Greg Bulling

One of the most difficult sections to break into was through the Pole Hole by F6. Looking through a small hole 200mm high, 1.0 m wide and 2.0 m long, there appeared to be a large chamber beyond. It was decided to dig this area out to access the chamber. This would involve shifting a large number of boulders and a lot of silt to allow access. It involved working in zero visibility and took several dives to achieve.

Mapping of Tank Cave started after the first few dives. Initially a stick map was produced showing all the numbered intersections. Because of the complexity of the Cave it was important to survey and produce the map ASAP as it was difficult for divers to find their way around the cave to given points.

Entrance area on previous Stick Map

All line installed was knotted every 5 metres, with an arrow every 20 metres indicating the direction to the entrance. The distance from the entrance was also marked on each arrow to assist with the surveying. There were also two arrows installed at every intersection indicating the direction out. This enabled divers to draw their planned route on their slate showing the numbered intersections. They could then refer to the slate on the dive if there was any confusion on the route to be taken.

Once all of the known cave was mapped, work then began on adding the wall details to the map. At that time all mapping was done using a good quality compass with two degree or less increments on it. The compass was attached to a slate at least 250mm long and when taking a bearing the slate would be held against the line and the bearing read from the compass; every time the line changed direction this would be repeated. The distances the bearings were taken at would also be recorded. At the base and top of rockpiles depths had to be recorded with distances, making allowance for the rockpile.

Entrance area on updated Wall Map

One method used to check if the survey was accurate was to survey a circuit. If it did not meet at the end when put on the map, there was an error. To recheck what had been surveyed, you would then resurvey coming from the opposite direction. As there are many areas in Tank Cave where circuits could be surveyed, this method was effective in checking for accuracy. This method was also used to check specific points: Starting from Point A and doing a circuit, then starting again from Point A and doing a different circuit: you could prove the accuracies of other points on both circuits.

When a tunnel was being surveyed which could not be checked by closing a circuit, you would survey on the way in and again on the way out and compare the two. This was particularly important when surveying the "S" and "G" sections as a small error at the beginning could put you out by 20 or 30 metres at the end of these sections.

When surveying there needs to be care taken to minimise interference to the compass, particularly when using side mount steel tanks as they are much closer to the compass than back mounted tanks. There are also considerable differences with interference between different brands of steel tanks. Another common problem is holding powerful lights too close to the compass. Most divers already experience errors when surveying underwater because of the extra task loading, misreading the bearing or perhaps a wrong measurement etc and interference to the compass can add to this.

I can remember some sections in Tank Cave having to be remapped several times as the errors were often 10 to 15 degrees out, or circuits which should have closed within 1 or 2 metres being 15 metres out. This is frustrating to both the diver having to recheck their survey and the person inputting the data for the map.

The recent radio direction finding and surveying developments are providing new views of the site and have enabled a high level of precision to be applied to the underwater map. Whilst the new information has resulted in numerous adjustments of the original data, it is noteworthy that the changes to the original map have not been significant. Given the size of the site and the limitations of using compass and line techniques this is a credit to the original mapping team of about 20 divers who took part in the mapping project up until June 1998.