A review of erosion and hydrology research at eriss, December 1997

Department of the Environment, 1997

Internal Report 268
Evans KG & Johnston A
Supervising Scientist Division

About the report

A major issue that has been the subject of research at eriss for a number of years has been the fate and subsequent environmental impact of material that will be eroded from the landform at ERA Ranger Mine (ERARM) following rehabilitation. There are three readily identifiable stages in the dispersal of eroded material from the rehabilitated site at ERARM: erosion of the rehabilitated landform itself; dispersion and deposition in the tributaries that link the mine site to Magela Creek; and dispersal and deposition in the Magela Creek and floodplain.

Past research has focused on sediment distribution in downstream areas of Magela Creek and the Magela Plain. The proportion of tailings from the mine reaching the Plain with a constant sediment delivery ratio is a function of time of total erosion of tailings. At the natural denudation rate (25±19 t/km2/y) tailings would take 1.4x105 y to erode with tailings being 1 % of the annual load of 5000t reaching the Plain from Magela Creek. If the time is ≤1800 y tailings would comprise 50% or more of the load reaching the Plain. The proportion of tailings at any northing can be calculated for a wide range of erosion times. These calculations are reliable only if assumptions underlying the sediment routing model used are valid. It is likely that turbidity would increase as a result of sediment transport from the mine. The environmental impact of this should be investigated. If erosion of the tailings impoundment occurs at the natural rate it is probable that there will be minimal effect on biota of the Magela Plain. If erosion occurs at ≥4 times the natural rate increases in the bioavailable concentration of metals will occur. The biological and radiological impact of this should be investigated.

Present research is addressing the refinement and validation of a landform evolution model SIBERIA. This model can be used to predict the long term geomorphic development of a proposed rehabilitated landform. A calibrated empirical erosion loss model has predicted a sediment loss rate of approximately 3100 t/km2/y from an unvegetated, unripped cap site on the waste rock dump and approximately 140 t/km2/y from a ripped and vegetated site. These values were calculated for slopes of approximately 3%. The erosion rates are orders of magnitude greater than the natural rate. SIBERIA simulations showed that for the unvegetated and unripped case, the landform at 1000 y would be dissected by localised erosion valleys (maximum depth = 7.6 m) with deposited fans (maximum depth = 14.8 m) at the outlet of the valleys. For the vegetated and ripped case reduced valley development (maximum 1000 y depth = 2.4 m) and deposition (maximum 1000 y depth = 4.8 m) occurred. For the vegetated and ripped condition simulated maximum valley depth in the capping over the tailing containment structure was about 2.2 m. Incision was approximately 5 m for unripped and unvegetated conditions. By modelling valley incision decisions can be made on the minimum depth of tailings cover required to prevent tailings from being exposed to the environment within a certain time frame. Validation of SIBERIA aims to test the long tenn erosion predictions of SIBERIA against natural processes in the field to strengthen confidence in the modelling process.

SIBERIA modelling to date has used present day parameters which assumes that the initial surface conditions of the waste rock dump will remain constant throughout the simulation period and there is no temporal change in parameters due to soil formation or ecosystem development. Simulations also assume there is no spatial variation in parameter values over the simulated area. Spatial variation results from surface treatments on the waste rock dump or the undisturbed land surface being included in the simulations. Temporal changes in parameter values are being addressed in current research.

Future research will continue the refinement of SIBERIA by incorporating spatial variation in input parameter values for the model. Sediment storage and transport in the small catchments between the mine site and Magela Creek will be investigated and this information will provide the link between the mine site and downstream areas of Magela Creek. This should strengthen the predicted sediment budget previously based on limited data. The impact of sediment and solute loads on the biota of Magela Creek should be investigated so that this can be related to the predicted sediment budget. Little is currently known of extreme flood events on Magela Creek and its tributaries. Floods at least 10 times greater than the mean annual flood have totally destroyed many channels and floodplains in Australia. It is essential that the probability of extreme storm events and floods is more accurately defined because of their significance for soil erosion, sediment transport, channel changes and landform evolution modelling by SIBERIA The palaeoflood record should be analysed so that the effect of catastrophic events can be incorporated in modelling.

The research program is total catchment management orientated. Completion of the program will provide modelling techniques which can be used to assess sediment movement from a proposed rehabilitated landform through the mine site catchment to Magela Creek and assess the impact of this sediment on the ecosystem.

The need for, and the extent of, future research on erosion of rehabilitated sites in the ARR is considered in the light of possible decisions that may be made in the near future on the final repository for tailings arising from the processing of ore from the ERARM and Jabiluka mine. Recommendations are given for the priorities that should be attributed to future research proposals.