Over-prediction of subsidence levels of mined land may be an issue for getting mining approval and tighter conditions in some places can be imposed to better inform mining operations and planning. Under-prediction of subsidence, present safety, social and environmental consequences during or after longwall extractions have occurred. In this study we collected subsidence data from reports related to underground coal mining environmental impact statements (EIS) and end of panel reports, and compared predicted subsidence data with the observed data to compare the accuracy of prediction. From the limited data, while it was difficult to substantiate a strong relationship between the predicted and observed subsidence data and mining characteristics, we found that where mining is less than 220 m below the surface subsidence prediction generally matches well with the observed subsidence and the ratio is closer to 1 than those mined at greater depths. Subsidence prediction analysed for mines in the Gunnedah Basin are found to be generally better than those in the Sydney Basin.The results presented here are based on limited data available in the public domain and while more data is needed to prove or disprove the discussed hypotheses, we consider there to be value in extending EIS requirements. Importantly, publication of both prediction and observed subsidence measurements would provide suitable transparency for improved, future predictions, and to provide social and stakeholder assurances that would meet growing legislative requirements.
Read full abstract