Just before Christmas 2003 Southland developed a heating problem which became a fire on Christmas day. The mine has been closed since then with efforts focused on assessing the possibility of reentering the mine.

Southland Colliery is located near Cessnock in the Lower Hunter coalfield of NSW. The mine development was originally commenced as a replacement for the now closed Ellalong colliery. The tenement is adjoined to the west and north by old workings. The mine exploits the Greta Seam which is some 4.8 – 7.0 m thick in the mining reserve area, producing a low ash, high fluidity semi-hard coking coal product.

There is a substantial history of development of heatings in the relatively high vitrinite content (higher oxidation susceptibility) Greta seam coals. Published data regarding R70 test results (Highton, 1995) indicate a value of 0.80, which falls into the medium risk category in regard to propensity to spontaneous combustion. Cross over test data from the same source indicate IRH values of 1.89 and 1.85 and TTR values of 6.05 and 4.93 for Pelton (Ellalong) and Abedare North samples from the Greta seam respectively. The risk index under this test method falls into the high risk propensity category.

Development roadways at Southland are typically driven in the lower section of the seam at around 3.2 – 3.5 m height, with the longwall extracting a similar profile. The upper section of the seam contains high sulphur (pyritic) coal that is not extracted, and which caves into the void behind the longwall. The roof immediately above the seam is highly laminated. Current longwall mining is in SL4 (longwall block 4), with development of SL5 in process.

The SL2 longwall panel (extracted) prior to SL3 was mined by the advancing longwall retreat method (i.e. longwall extracting in the same direction as ongoing roadway development. To facilitate requirements of this method and allow driveage of the additional length of the SL2 maingate for use in SL3 tailgate, the majority of SL2 Maingate was driven with 3 headings.

The mining of SL3 was delayed with a series of roof falls in the first third of the panel, followed by relatively trouble free mining for the remainder of the block. The mining horizon initially allowed for leaving a coal floor.

After limited retreat (around 130m from the installation roadway), poor ground conditions were encountered and shortly after a series of face falls occurred, greatly slowing the rate of retreat over a period between August / September 2002 and February 2003.

The remainder of the panel was mined at a relatively rapid rate of production of some 500,000 tonnes per quarter. The media has reported nominal panel reserves of between 1.50 and 1.70 Mt per block, however based on the quarterly production figures, it would appear that some 1.75 Mt was mined (suggesting a total of 50,000 - 250,000 of fallen top coal and stone).

SL4 commenced late October 2003. Based on media reports, SL4 contains some 1.70 Mt of reserves. IMC estimate that approximately 372,000 tonnes or just over 20% of the block had been extracted prior to the development of the mine fire, placing the longwall face adjacent to the area of roof falls from SL3 based on press releases, Coal Services quarterly longwall production statistics and Gympie Gold’s Q4 2003 production figures.

The media stated that the Southland heating initiated somewhere in Longwall Panel SL3, and subsequently the Gympie Gold press release of January 8, 2004 confirmed that the site of the fire was adjacent to the current SL4 longwall face position.

Coal, oxygen and the right airflow conditions are required to support development of a heating. In light of these factors, some observations regarding the Greta seam and the Southland mine layout include ;

· high sulphur top coal remains in Soutland longwall goaves with partial seam extraction. The oxidation of pyrite to iron sulphate and sulphuric acid is highly exothermic, producing two-thirds as much heat as the combustion of carbon in coal

· the rate of retreat in mining the inbye end of SL3 was very slow, allowing extended exposure of goaf coal to ventilation (oxygen)

· a heating problem became apparent when mining in SL4 came within proximity of the zone where the roof falls in SL3 occurred

The role of goaf seals, the mine ventilation system, and the nature of the goaf zone in SL3 where the falls occurred is unclear due to little or no detail in public source information.

The incubation period for a given coal, mining method and ventilation system may be defined as the time required for oxidation of goafed coal to cause a rise in temperature to its ignition point. Vutukuri and Lama (1986) cite low rank coal incubation times typically falling in the range from 3 to 6 months, and high rank coal between 9 and 18 months.

These periods relate to time from commencement of panel extraction (i.e. after completion of first workings). Ventilation conditions may extend (e.g. through very tight ventilation management) or substantially decrease (e.g. with air ingress, poor ventilation management) these times.

Simulation of conditions to ascertain incubation time is very difficult, so in practice incubation time is usually determined through experience. While cases of very rapid incubation are known, it would normally be expected that a heating would develop over a period of time, with transition of the heating into an open fire only at a very advanced stage.

Part 2 of this special report will appear Monday