Supervising Scientist Division

2 Environmental assessments of uranium mines (Ranger)

Supervising Scientist Annual Report 2004–2005

Supervising Scientist, Darwin, 2005
ISBN 0 642 24395 6
ISSN 0 158-4030

2.2 Ranger

2.2.1 Developments

Mining and milling of uranium ore at Ranger continued throughout 2004–05, with further development of the orebody in Pit 3 (Orebody 3). This development has resulted in the extension of Pit 3 to the east totally subsuming Djalkmara Billabong (see section 2.2.2). Ranger mine site is shown in Figure 2.1.

Exploration drilling to the east of Pit 3, outside of the bunded area, to further delineate the extent of Orebody 3, commenced during the 2005 dry season (see next section).

The Ranger mill produced 5544 t of uranium oxide (U3O8) during 2004–05 from 2 231 000  t of treated ore (Table 2.1). Production statistics for the milling of ore and the production of U3O8 at Ranger for the years 1999–2000 to 2004–05 are shown in Table 2.2.

 

Table 2.1 Ranger Production Activity For 2004-05 By Quarter
  1/07/2004 to
30/09/2004
1/10/2004 to
31/12/2004
1/01/2005 to
31/03/2005
1/04/2005 to
30/06/2005
Total
Production (drummed tonnes of U3O8 ) 1 422 1 408 1 464 1 250 5 544
Ore Treated ('000 tonnes) 529 572 564 566 2 231
Table 2.2 Ranger Production Activity for 2000-2001 to 2004-05
  2000-2001 2001-2002 2002-2003 2003-2004 2004-2005
Production (drummed tonnes of U3O8 ) 4 612 3 815 5 312 4 666 5 544
Ore Treated ('000 tonnes) 1 840 1 429 2 153 1 880 2 231

Ranger contributed approximately 50% of Australia’s total annual production of U3O8 in 2004–2005.

On-site activities

Exploration

Exploration drilling in the area to the east of Pit 3, but within the access road bund, commenced during the 2004 dry season and was completed by December 2004. This exploration identified that the ore body continued to the east. Following this, ERA advised that further exploration to the east will occur during the 2005 dry season. This will involves drilling from an area at the base of the access road bund and adjacent to the Magela Land Application Area. In order to minimise potential impacts on the Land Application Area or sensitive riparian vegetation along Magela Creek, drilling multiple holes in a fan formation from one each of a number of sites is being undertaken.

Map of the Ranger mine site

Figure 2.1 Ranger minesite

Water Treatment Plant

Construction of a Water Treatment Plant (WTP) began in April 2005 (Figure 2.2) and is due for completion in November 2005. The WTP was identified as the preferred treatment option during ERA investigations into reducing the water inventory, which has increased over the last few years. It is designed to treat both process and pond water prior to its release from site.

Figure 2.2 Water Treatment Plant under construction June 2005

Figure 2.2 Water Treatment Plant under construction June 2005

Pond water inventory has increased due to the expansion of Pit 3 and the requirement for the lower levels of the pit to be accessible for mining all year round. The removal of Djalkmara Billabong by the expansion of Pit 3 coupled with the extensions of the Western Stockpile for waste and very low-grade ore represent an increase of 30 hectares in the catchment of pond water (equating to 220 ML/annum with average rainfall). Further works around Pit 1 and the high grade ore stockpiles will see between 300 and 500 ML/annum of stockpile runoff diverted from entry into the process water circuit to the pond water system.

Apart from passive evaporation, there is currently no mechanism available on site to reduce the volume of contained process water. Surplus water in Pit 1 reduces the available storage capacity and may compromise the consolidation rate of deposited tailings.

The WTP will utilise lime softening, micro-filtration, reverse osmosis and wetland filtration to treat process water during the dry season. The salt brine and the solid waste stream from the WTP will be sent to the tailings dam for disposal. The preferred treatment option for pond water will utilise ferric flocculation, micro-filtration and reverse osmosis with the solid waste being used in the process to recover uranium, the brine being sent to the tailings circuit, and the treated water discharged beyond the wetland filters during the wet season.

Seepage barrier in Pit 1

A seepage containment barrier is currently being constructed in Pit 1. Construction of the first stage was completed prior to the 2004–05 wet season with the second lift currently under construction during the 2005 dry season. The barrier is part of a strategy by ERA to store additional tailings in Pit 1 as an interim measure until Pit 3 is ready for tailings deposition. The deposition of neutralised tailings in Pit 1 above RL0 to RL12 had not been approved by the Supervising Authorities, however, due to time constraints it was considered judicious by ERA to construct the barrier as part of a separate application.

The barrier is being constructed of locally derived clay-rich regolith material supported and protected by a layer of waste rock armour. Geotechnical and geochemical testing of the materials used for construction undertaken by the University of New South Wales and ANSTO indicates that the integrity of the barrier will be maintained over the long term.

The proposal for the deposition of tailings above the currently authorised level at RL0 will permit the company to meet regulatory requirements that all tailings are placed in the mined out pits at the end of mine operations, maximise the volume of tailings stored in Pit 1, and achieve the planned life-of-mine for the Ranger project with minimal disruptions to, and maximum flexibility in, its mining plan.

2.2.2 On-site environmental management

Water management

Throughout the reporting period the water management system at Ranger operated satisfactorily. The rainfall for 2004–05 was 1303 mm, 204 mm below the long-term average of 1507 mm (Figure 2.3).

Figure 2.3 Annual rainfall Ranger mine 1971-72 to 2004-05

Figure 2.3 Annual rainfall Ranger mine 1971–72 to 2004–05

Djalkmara Billabong was consumed by mining in Pit 3 in June/July 2004. The remaining catchment of Djalkmara Billabong has been contained by a diversion bund, which delineates the final extent of Pit 3 under the current mine plan. Water that once reported to Djalkmara Billabong is now diverted to the remnant Djalkmara sump adjacent to the access road and is then pumped to RP2 or discharged depending on water quality.

Process water system

Under the Ranger Environmental Requirements, water that is in direct contact with uranium ore during processing (process water) must be maintained within a closed system. It may only be released by evaporation or after treatment in a manner and to a quality approved by the Supervising Scientist. There were no releases of process water from the circuit during the reporting period.

As of 31 March 2005, the process water system contained a total of 7835 ML. This includes all free water from the Tailings Dam and Pit 1.

Pond water system

The pond water system contains water that has been in contact with stockpiled mineralised material and operational areas of the site other than those contained within the process water system. This also includes water from Pit 3. The water is managed in accordance with the Water Management Systems Operation Manual. The manual describes a system whereby water is managed according to source and quality. The pond water system consists of Retention Pond 1, Retention Pond 2, Retention Pond 3 and Pit 3. Retention Pond 1 (RP1) overflows via a constructed weir into Coonjimba Creek every wet season and was observed to discharge between 15 January 2004 and 19 April 2005. Water from Retention Pond 2 (RP2) or Pit 3 may not be released without prior treatment through wetland filtration and/or irrigation. In recent years management of the pond water system has changed from a proscribed regime based on catchment type to one in which water is managed according to water quality. Concurrently there has been an optimisation of the wet season stockpile design to reduce the production of pond water. Despite a lower than average 2004–05 wet season, the total volume in the pond water system at the end of March (generally considered the peak storage time) increased from 1427 ML in 2003 to 2512 ML in 2005. This is a reflection of the increased catchment size of Pit 3 and the removal of Djalkmara Billabong through Pit 3 expansion. Prior to this, water from Djalkmara Billabong was periodically discharged to Magela Creek at a rate that depended on water quality, thereby reducing the total inventory on site.

Stockpile sheeting

In October 2004 a presentation was given to the MTC detailing a new stockpiling strategy that would take ERA through to the current end of mine life. The proposed strategy involves stockpile development to the north of the current grade 1 waste-rock stockpile and to the west of Pit 3. Stockpile development will include construction of a base (made up of compacted clayey waste material) that will direct seepage towards RP2. Water will be sheeted from the surface of the stockpile, depending on quality, towards the unused RP1(2) wetland filter for sediment removal, prior to discharge into RP1. The stockpile will eventually cover RP1(2) and extend to the RP1 wetland filter and RP1. The stockpile will also cover the old access road and utilities such as power and water supply to Jabiru East, which will need to be moved to the current access road. Works on the grade 1 waste-rock stockpile to the west of Pit 3 commenced early in the 2004–05 wet season.

Currently ERA has a requirement to divert the first 200 mm of runoff from sheeting works on the southern stockpile and southern grade 2 waste-rock stockpile into the RP2 catchment. As in the previous two seasons, ERA applied to remove this requirement and discharge directly into the Corridor Creek Wetland Filter. The MTC was not able to gain consensus on the issue prior to the milestone of 200 mm of precipitation passing and subsequently this runoff was directed to the RP2 catchment. The application is still current and the MTC expects to achieve a resolution on the issue prior to the 2005–06 wet season.

Wetland filters/Land Application

Two wetland filter systems operated during 2004–05. The Corridor Creek system and the RP1 constructed wetland filter in the RP1 catchment.

The RP1 wetland filter operated successfully throughout the dry season providing polished water for land application on the RP1 and Djalkmara Land Application Areas.

The Corridor Creek Wetland Filter was not used during the 2004 dry season. Ground water pumped from a local bore into cells 1 and 6 maintained vegetation throughout the dry season.

Land application commenced in June/July 2004 and continued until the week of 22 November for the Magela Land Application Area and 6 December for the Djalkmara Irrigation and Djalkmara Extension Areas and the RP1 irrigation area. Both Djalkmara irrigation areas and the RP1 irrigation area operated in rotating shifts of 8 hours over a 24-hour period. Supply to these areas is regulated by pumping from cell 9 of the RP1 wetland filter. This year an electric pump replaced the previous diesel pump. The new system with superior controls has dramatically improved the efficiency of the system and reduced the requirements for maintenance due to clogging of the irrigation lines. Retention Pond 2 (RP2) water continued to be irrigated directly on the Magela Land Application Area.

Tailings and waste management
Tailings

Since August 1996, no process residue from the milling of ore has been deposited into the tailings dam with Pit 1 now the sole receptor. Over this time (up to 12 May 2005) a total of 17 969 000 tonnes of tailings have been deposited in Pit 1, which apart from 1 809 000 tonnes of tailings dredged from the tailings dam, is derived directly from ore processing. Transfer of tailings into Pit 1 from the milling and processing of ore is currently by central sub-aqueous deposition.

The average density of process residue in Pit 1 at 12 May 2005 was 1.38 t/m3, which meets the minimum target density of 1.2 t/m3.

For a majority of the reporting period deposition occurred in the northeast corner of the pit raising the level of the tailings in this local area to the approved limit of Relative Level 0 m (RL0). Due to the constraint of RL0, and the difficulty in managing tailings below this level in the northeast corner, more recent deposition has been directed towards the south and southeast where tailings depths are up to 20 m lower.

Process water rose above RL0 in Pit 1 during the 2004–05 wet season and if mining and tailings management proceeds as planned, it is estimated that tailings will reach RL0 sometime in 2006. Under the current authorisation ERA is only permitted to deposit tailings up to RL0 within Pit 1, as it is known that a zone of higher permeability rock at and above RL0 exists in the southeast corner. Anticipating that process water would rise above RL0 in the 2004–05 wet season, ERA began construction of a barrier in the southeast corner of Pit 1 in October 2004. The first raise of the barrier to RL8 (ie 8 m above RL0) was completed prior to the onset of the 2004–05 wet season. Water during the 2004–05 wet season rose within Pit 1 to approximately RL5.

The current proposed application for tailings above RL0 is an interim operational strategy and ERA will have to undertake further research and investigative work to provide a final tailings containment solution to the Supervising Authorities for approval. If the interim strategy is not proven to meet the requirements of the MTC for final containment, the Supervising Scientist has advised that tailings should be removed from Pit 1 to a scientifically justifiable level approved by the Supervising Authorities.

Waste stockpiles

Waste material continued to be placed on the northern grade 1 stockpile, which reached its final height of RL68 along the southern corridor haul road. Instability and slumping of the northern edge caused by an excess of lateritic material prevented the construction of the stockpile to its design height limit of RL78. In order to accommodate the remaining material it was necessary to extend the footprint of the stockpile by a further 9 ha. The design of the extension incorporates seepage collection to RP2, a rock windrow on the batter perimeter to protect the slope against erosion, and sediment control and wetland polishing of surface runoff via the RP1 wetland filter.

Landfill/waste management

During the reporting period ERA continued to improve the operation of the on-site landfill and the management of waste across site. A waste management plan, completed in the previous reporting period, has now been implemented. A number of strategies including recycling, drum identification, waste awareness, signage etc, are in place. This combined with a housekeeping initiative instituted in the previous reporting period has seen a significant improvement in the general tidiness of the site. The Mine Department has been given responsibility for all access to the landfill, which is located adjacent to Pit 1. Appropriate signage and security are now in place to limit incorrect disposal, and regular cover is placed on the waste pit to reduce scavenging. A number of bioremediation stations have also been constructed adjacent to the landfill in order to treat hydrocarbon-contaminated soils.

Radiological exposure to employees

The radiation dose limit for workers recommended by the International Commission on Radiological Protection (ICRP 60) in 1990 and adopted in Australia by the National Health and Medical Research Council (NHMRC) is 100 milliSieverts (mSv) in five years, with no single year exceeding the dose limit of 50 mSv. Furthermore, workers are categorised at Ranger as designated or non-designated workers. Designated workers are those workers who may receive a dose greater than 5 mSv in a year.

Table 2.3 shows the doses received by designated and non-designated workers in 2004, and a comparison with the average doses from the year before. The average and maximum radiation doses received in 2004 were approximately 5% and 23% respectively of the recommended ICRP 60 annual dose limits.

Table 2.3 Annual Radiation Doses Received by Workers at Ranger Uranium Mine
  Annual dose in 2004 Annual dose in 2003
  Average mSv Maximum mSv Average mSv Maximum mSv
Non-designated worker Not calculated 1 0.7 Not calculated 1.0
Designated worker 1.0 4.6 1.6 6.5

1 A hypothetical maximum radiation dose to non-designated employees is calculated using the gamma exposure results of employees of the Emergency Services Group, and dust and radon results measured at the Acid Plant. Consequently, the dose is conservative and would exceed actual doses received by non-designated employees, and are hence considered maximum doses.

Average annual doses to designated employees have been steadily declining since 2000. The decline in maximum doses received by designated employees is partly due to introducing a protection factor of 50 for the respiratory protection worn in the product packing area when calculating the doses received from the inhalation of dust.

The three primary radiation exposure pathways to workers at the mine are:

In 2004 the main contributor to radiation dose received by designated employees in the mine and mill production was external gamma radiation (58–67%) followed by the inhalation of radon (17–25%) and dust (17%), whereas mill maintenance workers and electricians received three-quarters of their annual dose via the dust inhalation pathway. Mine maintenance workers are no longer deemed designated employees since the September 2004 quarter. This decision has been made by ERA based on the fact that radiation doses for this work group have not exceeded 5 mSv in the past three years. Also, mine maintenance workers spend the majority of their time in the heavy engineering workshop, which is a supervised area.

A minor contribution to employee dose can be the ingestion pathway. A visual inspection of all work areas at Ranger mine is conducted monthly for the mine and mill areas. Priority areas for monitoring are areas such as lunch and crib rooms. If any surface contamination is found it is cleaned up to the satisfaction of the Radiation Safety Officer, to minimise intake of radionuclides with food and drink.

Surface contamination is also monitored when equipment is leaving site. Following incidents in 2004, when contaminated equipment left the mine, action has been taken and the vehicle fleet at Ranger has been separated into controlled and non controlled vehicles. Controlled vehicles are now the only vehicles on site that can enter controlled areas such as the pit area or stockpiles, and they cannot leave the site without a Radiation Release Certificate. Furthermore, short-term lease equipment is no longer to be used in controlled areas on site.

Audit outcomes
2004 Audit Review

The follow-up Environmental Audit Review to the 2004 annual audit of the Ranger operation (reported in the 2003–2004 Annual Report of the Supervising Scientist) was undertaken on 10–11 November 2004. The review team consisted of representatives of oss, NLC and DBIRD.

The two Category 1 non-conformances identified during the May 2004 audit (discussed in the 2003–2004 Annual Report) remained as Category 1 non-conformances. Both of these related to the process/potable water incident on 23 March 2004. Of the three Category 2 non-conformances, two remain Category 2. In addition, another three Category 2 non-conformances arose, two from issues that were considered Conditional and one from an issue that was Not Verified during the May 2004 audit.

There were 22 Conditional issues in the May 2004 audit. Of these, 14 were revised to Acceptable and two became Category 2 non-conformances (as mentioned above).

2005 Environmental Audit

The objective of the 2005 audit was to review the Water Management Systems Operations Manual (WMSOM) and determine the extent of the implementation, and the effectiveness, of the commitments made therein.

The scope of the audit included:

The overall outcome of the audit was deemed satisfactory, with the following findings:

Minesite Technical Committee

The Ranger Minesite Technical Committee (MTC) met eight times during 2004–05. Dates of meetings and significant issues discussed are shown in Table 2.4.

Table 2.4 Ranger Minesite Technical Committee meetings
Date Significant additional agenda items
11 August 2004
(full list included)
Process water treatment programme update, RL0 application update, status of Mine Management Plan and approvals, threshold levels for reporting, review of radiological monitoring, first flush, radiation clearance procedures, forum for providing information on previous years research by EWLS, potable water incident, Ranger Rehab Plan, Ranger Project Area exploration.
16 September 2004 No new items
18 October 2004 SSD revised trigger values, review of radiological monitoring, potable water incident, stockpile runoff proposal, waste rock dump extension
1 December 2004 No new items
27 January 2005 Reporting requirements, content and appropriate presentation of data and discussion of trends
30 March 2005 Gulungul Creek monitoring site location
29 April 2005 Ranger bund audit and maintenance program, product packing exhaust stack incident update, vegetation monitoring
8 June 2005 Product packing exhaust stack incident update, Commonwealth and NT working arrangements, Northern stockpile extension

Table 2.5 Ranger Authorisation Changes/Approvals
Date Issue
12 January 2005 Approval for the construction of a Water Treatment Plant
25 January 2005 Approval for discharge of treated water from the Water Treatment Plant
31 January 2005 Approval of Ranger Rehabilitation Plan 30
16 February 2005 Approval to extend the Northern Grade One Stockpile
Incidents

There were several reportable incidents at Ranger during the year.

On 29 October 2004 ammonium diuranate (ADU) entered the plant compressed air system within the product packing and precipitation buildings. This led to an employee being exposed to ADU when it spilt out of the system and on to the ground. The incident was the subject of an on-site investigation by the Supervising Scientist and is discussed in detail in section 2.2.5 of this Annual Report.

On 4 November 2004 several contractors at the Ranger mine site were sprayed with process water whilst undertaking maintenance in close proximity to an existing process water pipe. One of the contractors claimed to have swallowed some process water and felt unwell as a result. The contractors were offered medical assistance, which was not accepted, and no further effects have been reported.

On 2 April 2005 the Supervising Scientist was informed that a substance was observed emitting from the product packing scrubber and subsequently falling on the roof of the calciner and precipitation buildings. The scrubber was immediately shut down and the investigation discovered that the substance contained product. An investigation was immediately carried out by ERA. Approximately 0.5 kg of product is believed to have emitted from the scrubber falling on the product packing roof and down the side of the calciner building. The root cause of the incident was identified as the scrubber working outside of its operational limits. The scrubber was subsequently repaired and recommissioned. Exposure to employees was calculated to be below limits. The employee who discovered the incident was wearing appropriate Personal Protective Equipment (PPE).

On 19 May 2005 an ERA employee tripped on a bund in which ADU was being temporarily stored and fell partly into the water covering the ADU but was not submerged. He was wearing full PPE at the time, which he removed prior to immediate use of the safety shower, followed by further showering and changing his clothes. There was an indication that he may have ingested a small amount of ADU as a result of the incident or during the response. From urine monitoring results and other analyses it was concluded that some ingestion had occurred. Based on these results and the biokinetic model for uranium published by the ICRP, the committed effective dose to the operator as a result of ingestion of uranium was determined to have been approximately 60 microSieverts ( � Sv). This dose is roughly equal to that received in a modern X-Ray and small compared to the 2000  � Sv that the average Australian receives each year from natural background.

2.2.3 Off-site environmental protection

Surface water quality

Under the Ranger General Authorisation, ERA is required to monitor and report on water quality in Magela and Gulungul Creeks adjacent to the mine. Specific water quality objectives must be achieved in Magela Creek. These objectives were recently reviewed and updated by the Office of the Supervising Scientist.

The Authorisation specifies the sites, the frequency of sampling and the analytes to be reported. Each week during the wet season, ERA reports the water quality at key sites at Ranger, including Magela and Gulungul Creeks, to the major stakeholders (SSD, NTDBIRD and NLC). A detailed interpretation of water quality across the site is provided at the end of each wet season in the ERA Ranger Annual Wet-season Report.

In addition to ERA’s monitoring program, the Supervising Scientist conducts an independent surface water monitoring program that includes chemical and physical monitoring in Magela and Gulungul Creeks and biological monitoring of numerous water bodies in the region. Key results (including time-series charts of key variables of water quality) are reported on the Internet at www.deh.gov.au/ssd/monitoring/index.html and will be reported in a Supervising Scientist Report (in preparation) on the Surface Water Monitoring programme . The highlights of the monitoring season are summarised below.

Magela Creek water quality objectives

The Commonwealth Environmental Requirements (ERs) for Ranger require that key variables that indicate the health of the Magela Creek system be defined so that any unusual change can be identified. Water quality criteria, in the form of a hierarchy of values that trigger increasingly stringent management responses (focus, action and guideline/limit trigger values), were set for key variables in Magela Creek to provide a framework for detecting, interpreting and acting on changes in water quality. The criteria were described in the Supervising Scientist’s Annual Report for 2001–2002.

An extensive review1 of those water quality criteria was undertaken prior to the 2004–05 wet season to align the process and criteria with key elements of the Australian and New Zealand Guidelines for Fresh and Marine Water Quality.2 Major stakeholders – representatives of Commonwealth and Territory Governments, ERA and the Northern Land Council – were involved in the review.

Narrative statements have now been coupled with revised trigger values to form water quality objectives. The trigger values support the primary management aim of aquatic ecosystem protection and allow scientific interpretation of the data. The narrative statements support the secondary management aim of minimising water quality changes downstream of the mine, in line with the wishes of the Traditional Owners.

The trigger values are based on either local biological effects data from ecotoxicological testing (as for uranium), dietary modelling (for radium-226) or natural variations in water quality, established through analysis of reference site data collected over many years. Exceedances of triggers based on natural variations are expected occasionally and the management actions to be initiated have been modified to take account of this. The main change is that where trigger values are based on natural variations in water quality the upper triggers (and lower for pH) are now ‘guidelines’. Where the value is based on local biological effects data or dietary modelling (as for uranium and radium respectively) the upper trigger remains a ‘limit’.

The water quality objectives for Magela Creek and the responses triggered by an exceedance are fully described in Iles 2004 (see footnote 1) and reproduced in Appendix 2 as Explanatory Material on the Ranger Environmental Requirements.

Chemical and physical monitoring of Magela Creek

The first water chemistry samples for the Supervising Scientist’s 2004–05 wet season surface water monitoring program were collected from the Magela Creek downstream statutory compliance point on 21 December 2004, o ne day after the commencement of flow past the site. Weekly spot-sampling continued throughout the wet season with the last of the routine monitoring samples collected on 25 May 2005, the week before flow past the downstream compliance point ceased.

All indicators remained within limits/guidelines3 throughout the 2004–05 wet season. The measured values are indicative of the pattern of improved quality seen in the past three wet seasons, exemplified in the uranium results of Figure 2.4.

Statistical analysis of the data is still being conducted and a detailed interpretation will be published, along with the results of other monitoring programs, as a Supervising Scientist Report later in the year.

The upstream and downstream key water quality data from both the SSD and ERA programmes are summarised in Table 2.6 with uranium concentrations shown in Figure 2.5. There appears to be good agreement between the datasets.4 Uranium, magnesium and sulfate wet season median concentrations from both datasets were higher downstream of the mine but the concentrations were very low and not of environmental concern. Uranium was less than 3% of the limit throughout the season, and mostly less than 1% (Figure 2.5). Electrical conductivity (EC), whose guideline value provides a management control for the magnesium and sulfate concentrations, was also slightly higher downstream but compared to the guideline value the difference was small. The manganese, pH, and turbidity medians are similar at both sites for each dataset.

The water quality objectives set to protect the aquatic ecosystems downstream of the mine (see section on Magela Creek Water Quality Objectives) were achieved during the 2004–05 wet season. Available biological monitoring data (described later in this section) also indicate that the environment remained protected throughout the season.

Figure 2.4 Uranium concentrations in Magela Creek since the 2000–01 wet season (SSD data)

Figure 2.4 Uranium concentrations in Magela Creek since the 2000–01 wet season (SSD data)

Figure 2.5 Uranium concentrations measured in Magela Creek by SSD and ERA during the 2004–05 wet season

Figure 2.5 Uranium concentrations measured in Magela Creek by SSD and ERA during the 2004–05 wet season

Table 2.6 Summary of Magela Creek 2004–05 Wet Season Water Quality Upstream and Downstream of Ranger
  Median Range
Parameter Guideline
or Limit*
Organisation Upstream Downstream Upstream Downstream
pH 5.0 – 6.9 SSD 6.3 6.3 5.4 – 6.7 5.8 – 6.7
ERA 6.1 6.2 5.7 – 6.5 5.8 – 6.4
EC
( �S/cm)
43 SSD 13 17 5.9 – 17 7.3 – 23
ERA 12 14 7.9 – 18 8.4 – 19
Turbidity
(NTU)
26 SSD 2.2 2.6 0.9 – 15 0.9 – 12
ERA 2. 3. <1 – 25 <1 – 5
Sulfate ‡
(mg/L)
Limited by
EC
SSD 0.2 0.7 0.1 – 0.3 0.2 – 1.8
ERA 0.2 1.0 <0.1 – 0.8 0.4 – 3.2
Manganese ‡
( �g/L)
26 SSD 5.0 5.3 3.3 – 8 3.0 – 22
ERA 6.0 5.7 2.7 – 13 1.5 – 17
Uranium ‡
( �g/L)
6 SSD 0.015 0.031 0.004 – 0.065 0.014 – 0.145
ERA 0.015 0.035 <0.005 – 0.048 0.018 – 0.174

ERA data taken from the ERA Weekly Water Quality Report 5 July 2005; dissolved (<0.45 �m);

* A compliance limit applies to uranium, management guidelines apply all other parameters shown

Radium-226 in Magela Creek

Radium-226 ( 226Ra) values measured in Magela Creek by the Supervising Scientist since the 2001–02 wet season (when routine monitoring by the Supervising Scientist began) are shown in Figure 2.6. The concentrations are generally very low, including downstream of the Ranger mine. The 226 Ra concentration of 8.8 mBq/L in the sample collected in February 2005 at the upstream site is probably due to a higher contribution of 226 Ra-rich soil or finer sediments that are present naturally in Magela Creek.

The limit for total 226 Ra activity concentrations has been defined for human radiological protection purposes. The median of all data collected over the whole of the season at the upstream site is calculated and then subtracted from the median of the downstream data for the whole of the season . This difference between the median values for the wet season, called the wet season median difference , denotes any increase at the downstream site and should not be more than 10 mBq/L.

Based on the available data, the wet season median difference (shown by the solid line in Figure 2.6) for each of the wet seasons is approximately zero indicating that 226Ra levels in Magela Creek are due to the natural occurrence of radium in the environment and that radium from the Ranger mine has not caused any impact on human health.

Figure 2.6 Radium-226 activity concentrations in Magela Creek upstream and downstream of Ranger mine and the wet season median differences compared to the limit

Figure 2.6 Uranium concentrations measured in Magela Creek by SSD and ERA during the 2004–05 wet season

Chemical and physical monitoring of Gulungul Creek

The first water chemistry samples for the Supervising Scientist’s 2004–05 wet season surface water monitoring program were collected from Gulungul Creek on 23 December 2004 , the first day of flow in the creek for the wet season. Weekly spot-sampling continued throughout the season with the last of the routine monitoring samples collected on 8 June 2005, during the week in which flow at the downstream site ceased.

The overall water quality and seasonal trends were comparable to those seen in previous years. This is demonstrated by the uranium concentrations shown in Figure 2.7. Statistical analysis of the data is still being conducted and a detailed interpretation will be published, along with the results of other monitoring programs, in a forthcoming Supervising Scientist Report.

The upstream and downstream key water quality data from both the SSD and ERA programs are summarised in Table 2.7 with uranium concentrations shown in Figure 2.8. There appears to be good agreement between the two datasets. Small differences could be attributable to different sampling times.

Uranium concentrations in Gulungul Creek are naturally higher than those in Magela Creek (due to different geochemistry and hydrology influences in the respective catchments). Like Magela Creek, the concentrations at the Gulungul Creek downstream site are slightly higher than those at the upstream site. However, the uranium concentrations in Gulungul Creek were well below the limit throughout the season, with the concentration at both the upstream and downstream sites ranging between about 1% and 4% of the limit (Figure 2.8).

Figure 2.7 Uranium concentrations in Gulungul Creek since the 2000–01 wet season (SSD data)

Figure 2.7 Uranium concentrations in Gulungul Creek since the 2000–01 wet season (SSD data)

Figure 2.8 Uranium concentrations measured in Gulungul Creek by SSD and ERA during the 2004–05 wet season

Figure 2.8 Uranium concentrations measured in Gulungul Creek by SSD and ERA during the 2004–05 wet season

Table 2.7 Summary of Gulungul Creek 2004–05 wet season water quality upstream and downstream of Ranger
Median Range
Parameter Company Upstream Downstream Upstream Downstream
pH SSD 6.6 6.6 5.7 – 6.9 5.8 – 6.9
ERA 6.3 6.3 8.9 – 6.8 6.0 – 6.8
EC
( �S/cm)
SSD 17 21 9.0 – 26 9.4 – 28
ERA 14 18 10 – 21 13 – 23
Turbidity
(NTU)
SSD 1.4 1.8 <0.5 – 4.0 0.6 – 6.7
ERA 1. 2. <1 – 20 <1 – 7
Sulfate ‡
(mg/L)
SSD 0.2 0.5 <0.1 – 0.9 0.1 – 1.4
ERA 0.2 0.6 0.1 – 1.0 0.1 – 2.3
Magnesium ‡
( mg/L)
SSD <0.9 <1.1 0.4 – 1.6 0.4 – 1.5
ERA 1.0 1.2 0.7 – 1.6 0.8 – 1.4
Manganese ‡
( �g/L)
SSD 2.7 3.0 1.2 – 8.0 0.7 – 15
ERA 2.1 3.0 1.1 – 9.5 0.8 – 7.8
Uranium ‡
( �g/L)
SSD 0.058 0.093 0.031 – 0.237 0.058 – 0.212
ERA 0.069 0.110 0.034 – 0.253 0.063 – 0.249

‡ dissolved (<0.45 �m), * limit = 6  �g/L

Sulfate concentrations in Gulungul Creek were similar to those seen in previous years, with the downstream concentrations generally higher than those upstream, but still well below levels of environmental concern.5 The small difference in electrical conductivity (EC) between the upstream and downstream sites shows that the increase in sulfate has little effect on the overall solute levels downstream of the mine. The EC trend at both upstream and downstream sites closely follows that of magnesium. Magnesium, manganese, pH and turbidity were similar at the upstream and downstream sites indicating that these variables are not, or only slightly, influenced by the mine.

Biological monitoring in Magela Creek

Based on eriss research since 1987, biological monitoring techniques have been developed that can be used to assess the environmental impact of uranium mining on aquatic ecosystems downstream of the Ranger mine. Two broad approaches are used: early detection studies and assessment of overall ecosystem-level responses.

Creekside monitoring is used for early detection of effects in Magela Creek arising from any dispersion of mine waters during the wet season. For ecosystem-level responses, benthic macroinvertebrate and fish communities from Magela and Gulungul Creek sites are compared with historical data and data from control streams. Results of creekside monitoring, macroinvertebrate (Magela and Gulungul Creek sites only) and fish community studies conducted during the 2004–05 wet and early d ry seasons and macroinvertebrate studies carried out in 2004 are summarised here. The data and results will be fully reported in the forthcoming Supervising Scientist Report on the Surface Water Monitoring Programme.

Creekside monitoring

In this form of monitoring, effects of Ranger mine wastewater dispersion are evaluated using responses of aquatic animals held in tanks on the creek side. The responses of two test species are measured over a four-day period:

Animals are exposed to a continuous flow of water pumped from upstream of the mine site (control site) and from the creek just below gauging station GS8210009, some 5 km downstream of the mine. Tests usually commence in December and cease in early April, the period of significant creek flow in Magela Creek.

Eight creekside tests were conducted in the 2004–05 wet season. Details and results of those tests have been reported at www.deh.gov.au/ssd/monitoring/explanatory-bio.html#04-05. The results are summarised here.

The results of the creekside tests are plotted as part of a continuous time series of actual and ‘difference’ (between upstream and downstream) data in Figure 2.9A for snail egg production, and in Figure 2.9B for larval fish survival. Snail egg production at upstream and downstream sites was similar across all tests conducted for the wet season (Figure 2.9A) and ‘difference’ values for 2004–05 were comparable with those from previous years (no significant difference was found (P>0.05)).

Across all fish tests, larval fish survival at upstream and downstream sites was similar (Figure 2.9B ), apart from reduced survival at the upstream site relative to the downstream site during the third creekside test in particular (an observation commonly noted in previous years for this test species – possible causes are discussed in the Supervising Scientist annual report for 2002–03).

Larval fish survival at the downstream relative to upstream site in Magela Creek during 2004–05 was consistent with the same relative survival rates observed in the previous five wet seasons.

From the collective creekside results, it was concluded that there were no adverse effects of dispersed Ranger mine wastewaters to Magela Creek on either of the creekside test species over the 2004–05 wet season.

Monitoring using macroinvertebrate community structure

Macroinvertebrate communities have been sampled from a number of sites in Magela Creek at the end of significant wet season flows, each year from 1988 to the present. The design and methodology have been gradually refined over this period (changes are described in the 2003–04 Supervising Scientist Annual Report). The design is now a balanced one comprising upstream and downstream sites at two ‘exposed’ streams (Gulungul and Magela Creeks) and two control streams (Burdulba and Nourlangie Creeks).

Figure 2.9A and 2.9B Creekside monitoring results for freshwater snail egg production for wet seasons between 1992 and 2004

Figure 2.9A and 2.9B Creekside monitoring results for freshwater snail egg production and larval black-banded rainbowfish survival for wet seasons between 1992 and 2004

Samples were collected from each site at the end of each wet season (between April and May). For each sampling occasion and for each pair of sites for a particular stream, a dissimilarity index is calculated. This index is a measure of the extent to which macroinvertebrate communities of the two sites differ from one another. A value of ‘zero’ indicates identical macroinvertebrate communities while a value of ‘one’ indicates totally dissimilar communities, sharing no common taxa.

Disturbed sites will have significantly ‘higher’ dissimilarity values compared with undisturbed sites. Analysis of the full macroinvertebrate data set from 1988 to 2004, and data from the paired sites in the two ‘exposed’ streams, Magela and Gulungul Creeks for 2005, has been completed and results are shown in Figure 2.10. This Figure plots the paired-site dissimilarity values using family-level (log-transformed) data, for the two ‘exposed’ streams and the two ‘control’ streams.

Figure 2.10 Paired upstream-downstream dissimilarity values (using the Bray-Curtis measure) calculated for community structure of macroinvertebrate families in several streams in the vicinity of the Ranger uranium mine for the period 1988 to 2005

Figure 2.10 Paired upstream-downstream dissimilarity values (using the Bray-Curtis measure) calculated for community structure of macroinvertebrate families in several streams in the vicinity of the Ranger uranium mine for the period 1988 to 2005. The dashed vertical lines delineate periods for which a different sampling and/or sample processing method was used. Dashed horizontal lines indicate mean dissimilarity across years. Processing of samples collected from the two control streams in 2005, Burdulba and Nourlangie, had not been completed by the time this annual report was prepared.

Inferences that may be drawn from the data shown in Figure 2.10 are weakened because there are no pre-mining (pre-1980) data upon which to assess whether or not significant changes have occurred as a consequence of mining. Notwithstanding, the plots show that the mean dissimilarity value for each stream across all years is approximately the same (~0.3) and that the values are reasonably constant over time. Confirming this, single-factor ANOVA (a statistical comparison) shows no significant difference in the mean dissimilarities between the ‘control’ and ‘exposed’ streams.

Another statistical technique (multivariate ordination) used to compare community structure also indicated that macroinvertebrate communities were similar at the ‘control’ and ‘exposed’ sites. The data will be fully reported in the forthcoming Supervising Scientist Report on the Surface Water Monitoring Programme.

Collectively, t hese results provide good evidence that changes to water quality downstream of Ranger as a consequence of mining in the period 1994 to 2005, at least, are not sufficient to have adversely affected macroinvertebrate communities.

Monitoring using fish community structure

Sampling of fish communities in billabongs is conducted in late April to the end of June of each year. Data are gathered, using non-destructive sampling methods, from ‘exposed’ and ‘control’ sites in deep channel billabongs and shallow weedy lowland billabongs. Details of the sampling methods and sites are provided in the 2003–04 Supervising Scientist Annual Report. For both deep channel and shallow lowland billabongs comparisons can be made between: (i) directly exposed billabong(s) versus control billabong(s) from independent catchments (Nourlangie Creek, East Alligator River, Wirnmuyirr Creek); and/or (ii) directly exposed versus indirectly exposed billabongs in Magela Creek, recognising that this second approach is confounded by possible movement of fish between the two lowland billabong types in the same stream system.

Channel billabongs

The similarity of fish communities at Mudginberri Billabong (directly exposed site downstream of Ranger) and Sandy Billabong (control site in the Nourlangie catchment) was determined using multivariate dissimilarity indices. These indices and rationale for their use are explained above in the section ‘Monitoring using macroinvertebrate community structure’. A plot of the dissimilarity values from 1994 to the present is shown in Figure 2.11.

In the Supervising Scientist Annual Report for 2003–2004, the significant decline noted in the paired-site dissimilarity measures over time (Figure 2.10) was attributed to the particularly high abundances of chequered rainbowfish ( Melanotaenia splendida inornata) and to a lesser extent glassfish ( Ambassis spp) in Mudginberri Billabong in the early years of the study, relative to Sandy Billabong. Chequered rainbowfish have declined in Mudginberri Billabong since sampling commenced in 1989. This issue is examined in more detail in section 3.6 of this Annual Report where a number of environmental correlates of this decline are reported. The decline in rainbowfish numbers and, by association, the paired billabong dissimilarity value, does not appear to be related to any change in water quality over time associated with water management practices at Ranger (section 3.6).

Figure 2.11 Paired control-exposed dissimilarity values (using the Bray-Curtis measure) calculated for community structure of fish in Mudginberri (‘exposed’) and Sandy (‘control’) billabongs in the vicinity of the Ranger uranium mine over time

Figure 2.11 Paired control-exposed dissimilarity values (using the Bray-Curtis measure) calculated for community structure of fish in Mudginberri (‘exposed’) and Sandy (‘control’) billabongs in the vicinity of the Ranger uranium mine over time. Values are means ( � standard error) of the 5 possible (randomly-selected) pairwise comparisons of transect data between the two billabongs, while line about the means illustrates the fitted regression ( R2 = 0.2, P = 0.0003)

Another statistical technique (multivariate ordination) was also used to compare fish community structures in the channel billabongs. The results of this analysis indicated that the fish communities at the ‘control’ and ‘exposed’ sites were not too dissimilar. The data will be fully reported in the forthcoming Supervising Scientist Report on the Surface Water Monitoring Programme.

Shallow lowland billabongs

The abundance of fish in shallow billabongs for the period 1993 to 2005 is shown in Figure 2.12. Abundances in 2005 were similar to, or slightly greater than in, 2004. The changes in fish abundances in 2005 were, with one exception, all within the range of natural variation observed for each location since sampling commenced in 1993. Fish abundance in Coonjimba Billabong was the highest so far recorded for this site.

The number of fish species recorded in each billabong has varied only slightly over the 1993 to 2005 period by comparison with fish abundance (data not shown here).

The temporal pattern of multivariate dissimilarity values for control and direct exposure sites is shown in Figure 2.13. These values and rationale for their use are explained above in the section ‘Monitoring using macroinvertebrate community structure’. There were no significant differences among years and the slight trend of increase over time is not statistically significant, nor indicative of changes or trends observed amongst the numerically dominant fish species present in the billabongs (see previous section ‘Channel billabongs’).

Another statistical technique (multivariate ordination) was also used to compare fish community structures in the shallow lowland billabongs. The results of that analysis, which will be fully described in the forthcoming Supervising Scientist Report on the Surface Water Monitoring Programme , also indicate no discernible effect of mining activity on fish communities.

The small variation in dissimilarity values for fish communities in channel billabongs and the similarity of 2005 fish communities in shallow and channel billabongs to the respective communities found in previous years indicate there is no evidence of any adverse effects of mine waste waters arising from the Ranger mine site on fish communities of Magela Creek.

Figure 2.12 Relative abundance of fish in billabongs with different degrees of exposure to contaminants from Ranger uranium mine, 1993–2005. Not all sites have been sampled each year

Figure 2.12 Relative abundance of fish in billabongs with different degrees of exposure to contaminants from Ranger uranium mine, 1993–2005. Not all sites have been sampled each year

Figure 2.13 Paired dissimilarity values (using the Bray-Curtis measure) calculated for community structure of fish in ‘directly exposed’ Magela and ‘control’ Nourlangie and Magela billabongs in the vicinity of the Ranger uranium mine over time

Figure 2.13 Paired dissimilarity values (using the Bray-Curtis measure) calculated for community structure of fish in ‘directly exposed’ Magela and ‘control’ Nourlangie and Magela billabongs in the vicinity of the Ranger uranium mine over time. Values are means ( � standard error) of the (up to) 3 possible pairwise billabong comparisons: Coonjimba vs Buba, Gulungul vs Winmurra, and Georgetown vs Sandy billabongs. Line about the means illustrates the fitted, but non-significant, regression.

Radiological exposure of the public

The International Commission on Radiological Protection recommends that the annual dose received from a practice such as uranium mining and milling should not exceed 1 milliSievert (mSv) per year. This dose is on top of the radiation dose received naturally, which averages to approximately 2 mSv per year in Australia, but typically varies between 1–10 mSv per year.

Ranger is the main potential source of radiation exposure to the community in the Alligator Rivers Region. There are two main pathways of potential exposure to the public during the operational phase of a uranium mine: the inhalation pathway, which is a result of dispersion of radionuclides from the mine site into the air; and the ingestion pathway, which is caused by the uptake of radionuclides into bush-foods from the Magela Creek system downstream of Ranger.

The Supervising Scientist monitors the two main airborne pathways:

The main areas of habitation are Jabiru, Mudginberri and Jabiru East, consequently the monitoring focuses on those three population centres in the region. Airborne RDP and LLAA concentrations are measured monthly and the results compared with ERA’s quarterly atmospheric monitoring results.

Figure 2.14 shows Jabiru and Jabiru East RDP data and a comparison with ERA data from January 2002 up to April 2005. Differences in sampling time and location may be the cause of the slight differences in RDP concentrations observed at Jabiru, with ERA values being slightly higher than values measured by eriss. The annual exposure due to the inhalation of radioactivity trapped in or on dust has been shown to be trivial and is less than 1% of the public dose limit.

Figure 2.14 Radon decay product concentration measured by eriss and ERA in Jabiru and Jabiru East from 2002 to early 2005

Figure 2.14 Radon decay product concentration measured by eriss and ERA in Jabiru and Jabiru East from 2002 to early 2005

Table 2.8 shows the average annual doses received from the inhalation of radon decay products in the air, as calculated from the RDP concentration data from ERA and eriss (in brackets) at Jabiru. This is assuming an occupancy of 8760 hrs (1 year) and a dose conversion factor for the public of 0.0011mSv per �J � h/m3. Mine derived annual doses from the inhalation of radon progeny are shown, as calculated by ERA using a wind correlation model developed by eriss, which correlates wind direction with airborne radon decay product concentration.

Table 2.8 Average Annual RDP Concentration at Jabiru East and Jabiru and Doses Received from the Inhalation of RDP at Jabiru
2002 2003 2004
RDP concentration [ �J/m 3 ] Jabiru East 0.095 (0.085) 0.075 (0.101) 0.103 (0.095)
Jabiru 0.077 (0.047) 0.065 (0.043) 0.079 (0.063)
Total annual dose [mSv] at Jabiru 0.74 (0.45) 0.63 (0.41) 0.76 (0.61)
Mine derived dose [mSv] at Jabiru   0.03 0.011 0.014

The effective dose to members of the public, especially the Aboriginal population living downstream of Ranger, would be dominated by the ingestion of radium-226 in freshwater mussels. Consequently, the ingestion pathway is monitored by measuring the radionuclide concentration in mussels from Mudginberri Billabong.

Radionuclide concentrations in mussels, collected between 2000 and 2003 by eriss, Traditional Owners and Aboriginal people from the Mudginberri community, are low and show no trend of increasing radionuclide concentrations with time, for mussels from the same location and of the same age.

Assuming a 10-year old child consumes 2 kg of freshwater mussels per year from Mudginberri Billabong, the total committed effective dose from 226Ra and 210Pb would amount to 0.27 mSv. Based upon the water quality data presented earlier in this section, this dose is primarily due to radionuclides present naturally in the billabong. The contribution of mine origin radionuclides to this total is difficult to quantify, but can be estimated by a radionuclide transport and fate model that has been developed by eriss. For the case of water releases from Retention Ponds 1 and 4 during the wet season (the most conservative historical case), the effective mine derived contribution would be less than 0.2% of the total dose derived from the consumption of the mussels.

2.2.4 Outcome of investigations into incidents at Ranger in 2003–04

The Supervising Scientist’s reports on the incidents, Investigation of the potable water contamination incident at Ranger mine March 2004 (Supervising Scientist Report 184) and Investigation of radiation clearance procedures for vehicles leaving the Ranger mine (Supervising Scientist Report 185), were tabled in Parliament on 30 August 2004 and subsequently made available to all major stakeholders in hard copy format and through the Supervising Scientist’s website.

Following consideration of issues raised in the reports, the Minister for Industry, Tourism and Resources, the Hon Ian Macfarlane MP, wrote to ERA requiring the company to comply with a series of conditions under the Atomic Energy Act 1953. These conditions were based on the recommendations in the Supervising Scientist’s two reports. Mr Macfarlane required that the conditions be met in accordance with a timeframe involving deadlines of 10 September 2004, 31 October 2004 and 31 December 2005.

The Department of Industry, Tourism and Resources then commissioned the Australian Nuclear Science and Technology Organisation (ANSTO) and the Australian Radiation Protection and Nuclear Safety Agency (ARPANSA) to conduct a series of audits to assess ERA’s compliance with the deadlines and conditions. These audits were held on 13 September 2004, 3–4 November 2004 and 19 January 2005. The Supervising Scientist attended each of the on-site audits as an observer.

On 10 March 2005, Mr Macfarlane wrote to ERA to advise it that, on the basis of the audit reports from ANSTO and ARPANSA, he had concluded that ERA had, with the exception of the implementation of workplace safety standard AS4801 by 30 September 2005, complied with all of the Minister’s conditions. He also advised ERA that the Supervising Scientist would provide him with regular updates on the adequacy of measures taken by ERA in response to all of his recommendations and would also keep him informed about ERA’s progress in achieving accreditation against AS4801.

Minister Macfarlane also decided that, to ensure continued openness and transparency in the reporting on these incidents, it would be appropriate to make the ANSTO and ARPANSA audit reports publicly available. To implement this decision, the audit reports have been posted on the Supervising Scientist’s web site.

The Ranger Minesite Technical Committee has, at each of its recent meetings, reviewed the status of compliance by ERA against the Supervising Scientist’s full set of recommendations. In addition, the Supervising Scientist has engaged ARPANSA to conduct an additional detailed audit of the Radiation Safety Practices at the Ranger mine. The purpose of this audit will be to examine the steps that ERA has taken to upgrade its radiation management system and, as a result, to address concerns about the radiation protection culture at Ranger. This audit is scheduled to occur in July 2005.

By 30 June 2005, ERA had made substantial progress towards compliance with AS4801. It is expected that the certification audit will take place in September 2005.

Following completion of the AS4801 certification audit and the new ARPANSA audit, the Supervising Scientist will provide further advice to the Minister for Industry, Tourism and Resources on the status of ERA’s implementation of all of the Minister’s conditions and of the recommendations of the Supervising Scientist related to the potable water contamination incident and the radiation clearance incident.

2.2.5 Compressed air contamination incident

On 29 October 2004 the Supervising Scientist was notified of an incident involving a substance assumed to be ammonium diuranate (ADU) emitting from a ‘rattle gun’ being used to seal lids on a product container in the Precipitation and Product Packing area. The operator had tried to use the ‘rattle gun’ but noticed that it was clogged and when he detached the ‘rattle gun’ from the compressed air line, a quantity of liquid containing a yellow precipitate flowed out from the airline and onto the floor, splashing a small amount onto the operator’s boot. The supervisor was notified and product packing halted until the incident was investigated. Initial testing of the substance indicated it was radioactive and possibly ADU. This was later confirmed. The compressed air line was isolated at the entrance to the Precipitation and Product Packing area, and the buildings cordoned off. Staff from the Supervising Scientist and DBIRD went to the mine site to undertake investigations. On arrival a meeting was held to update stakeholders on the situation. ERA advised that it was likely ADU had entered the compressed air system within the precipitation building through a couple of locations. Staff then commenced the investigation.

Investigation

Supervising Scientist and DBIRD staff thoroughly examined the Product Packing room and the Precipitation building. The two main theories on the origin of contamination were the potential connection of the compressed air line to the gravity pump ADU supply line to the calciner, and a potential backflow into the compressed air line from the barren strip sand filter.

Connection of compressed air line to ‘Bredel Pump’ on ADU supply line to calciner

A fitting was identified on the pressure side of the Bredel pump that pumps ADU into the calciner. If a compressed air line was connected to that point whilst the pump was operating, and the valve was open, the pressure at the point of the connection would be, under certain conditions, greater than the pressure in the compressed air line. This would force ADU into the compressed air system. This location was considered a potential contamination point after an employee indicated that compressed air had been connected to the fitting on the pressure side of the Bredel pump in the past to free up blockages in that line, however, operators working in the area during and prior to the incident indicated that no such connection had been made in recent memory. No connection was evident at the time of investigation, and examination of the fittings provided no evidence that such a connection had recently been made. Examination of the supervisors’ log books also indicated that there had been no blockage recorded in the ADU line in the past week.

Backflow into the compressed air line from the barren strip sand filter

The barren strip sand filter is used to filter liquor that has been stripped of ADU to remove any remaining residual ADU in the liquor. This filter is periodically backwashed to free filtered ADU and clean the sand. To do this a compressed air line feeds an air bed within the sand filter that delivers air into the sand to agitate it during backwashing. The air line incorporates a control valve that is designed to open during backwashing operations and a non-return valve that is designed to allow air to pass into the sand filter and to prevent barren strip solution from exiting the sand filter via the compressed air line. Both the control valve and the non-return valve were removed, inspected and determined to be faulty. Evidence of barren strip solution and ADU was observed in both valves and on the upstream side of both valves in the compressed air line. It was established that barren strip solution and ADU passed through both of these valves into the compressed air line, and may have been doing so for some time. It was established that under normal operational conditions, the pressure of the barren strip solution within the sand filter would be greater than the pressure in the compressed air line at that point. However, given that both valves in the compressed air line were faulty, any positive pressure differential between the barren strip in the sand filter and the compressed air line would result in barren strip solution entering the compressed air line. The compressed air system pressure log indicated that at the source of the compressed air (a building approximately 100m away), the pressure is nominally lower than that generated in the backwash of the sand filter. However, pressure losses through the length of the system prior to utilisation at the sand filter could result in the pressure of the operating sand filter being greater than that in the compressed air line and hence barren liquor solution containing a small percentage of ADU could have entered the compressed air line.

Potential issues

The compressed air system is used throughout the site for pneumatic purposes ie ‘rattle guns’ for fastening nuts, and historically at times for breathing air supplies. At the time of the incident, the only use of air had been for pneumatic purposes apart from a breathing apparatus used by a sand blasting team at the other end of the plant. The breathing apparatus included a comprehensive carbon filter system for cleaning of the compressed air. This apparatus was quarantined for testing by the investigation team.

The operator using the ‘rattle gun’ had been splashed with liquid containing ADU. The operator at the time was wearing appropriate PPE and it was determined that radiation exposure to this employee would have been very low and well within limits. As the substance was in liquid form, the potential for an air pathway for exposure was negligible. However, depending on the extent of contamination of the compressed air line, there could have been the potential for aerosols containing ADU to be expressed at other locations throughout the site.

Monitoring

The Supervising Scientist undertook initial investigative sampling that included:

All samples were tested for activity levels to detect possible contamination.

The samples confirmed that ADU was the source of contamination within the compressed air line. They also confirmed that both the fittings surrounding the Bredel pump, and the compressed air lines entering the sand filter had detectable levels of ADU present.

Monitoring of the air stream, and swabs taken from the compressed air line confirmed that contamination had not spread beyond the isolation point outside the precipitation area. The air stream samples also confirmed that there was no detectable activity within the airstream above background.

Activity analysis were also undertaken by eriss on the carbon filters within the breathing apparatus used for sand blasting. It was assumed that if this apparatus was connected to the compressed air line at the time the contamination was occurring, contamination would be contained within the carbon filter system. As this system would have been in operation for lengthy periods of time, it would be a good long-term sample source to determine whether the contamination had spread beyond the precipitation building to the extremities of the site. Samples taken indicated only background levels of activity.

Conclusions

The most likely cause of ADU entering the compressed air line was through a faulty non-return and control valve connected to the barren strip sand filter. ADU had entered this line over time due to the fluctuating pressure differences between the operational sand filter and the compressed air system. ADU had progressed over time through the compressed air system within the precipitation and product packing building and settled at low points such as the outlet point within the product packing area where an operator first noticed the contamination.

The possibility that ADU could have entered the compressed air line through a deliberate connection to the ADU supply to the calciner was disproved during the investigation.

Contamination was restricted to the precipitation and product packing buildings. All lines were flushed clean with water and air to the satisfaction of the Supervising Scientist prior to recommencement of operations.

The faulty non-return and control valves were replaced to the satisfaction of the Supervising Authorities.

Prior to recommencement of the use of the compressed air system throughout the site, the Supervising Authorities sought the following assurances from ERA:

Based on the investigation and sampling undertaken, there was no evidence of radiation exposure to employees from the incident.

Footnotes

1 Iles M 2003. Review of water quality triggers, November 2003 progress report. Internal Report 463, November, Supervising Scientist, Darwin. Unpublished paper. & Iles M 2004. Water quality objectives for Magela Creek – revised November 2004. Internal Report 489, December, Supervising Scientist, Darwin. Unpublished paper.

2 ANZECC & ARMCANZ 2000. Australian and New Zealand guidelines for fresh and marine water quality. National Water Quality Management Strategy Paper No 4, Australian and New Zealand Environment and Conservation Council & Agriculture and Resource Management Council of Australia and New Zealand, Canberra.

3 The terms ‘limit’, ‘guideline’ and ‘objective’ are described above in ‘Magela Creek Water Quality Objectives’.

4 Results of statistical analyses of these and Gulungul Creek datasets will be reported in a Supervising Scientist Report (in preparation) on the Surface Water Monitoring Programme.

5 Toxicity tests on local Hydra and Lemna species demonstrated that SO4 (as Na2SO4) exhibits very little, if any, toxicity to these species below 200 mg L-1.