Uncertainty analysis of CSG- induced GAB flux and water balance changes in the Narrabri Gas Project area

[摘要] The proposal for coal seam gas (CSG) development in the Pilliga forest in northern NSW has raised several environmental concerns. The Pilliga Sandstone aquifer in this region is an important fresh water source that is used for irrigation, stock and domestic uses. The Pilliga forest is also the main recharge area for the Pilliga Sandstone aquifer, which is part of the Great Artesian Basin (GAB) aquifers. There is concern that depressurization of coal seams for producing gas may potentially impact groundwater pressure in the Pilliga Sandstone aquifer and affect the quantity of water recharged into the GAB. This report provides a probabilistic quantification and uncertainty of potential groundwater flux and water balance changes in the GAB aquifer caused by coal seam gas development by the Narrabri Gas Project in the Gunnedah Basin.A probabilistic groundwater modelling method was applied for the preliminary assessment of potential flux and water balance changes and associated uncertainties in the GAB aquifer – the Pilliga Sandstone caused by coal seam gas development through the Narrabri Gas Project. The groundwater model built for the Namoi subregion in the Bioregional Assessments Programme (http://www.bioregionalassessments.gov.au/) was used for this purpose. The changes in flux and water balance induced by extraction of water from the coal seams was quantified as the difference between the CSG development and the baseline scenarios of groundwater flow. Uncertainty in the CSG water production rates, hydraulic characteristics of the geologic formations and groundwater flow components including recharge were accounted for by varying their respective parameters in the model in a wide range. Three thousand five hundred sets of model parameters sampled from a uniform distribution were initially evaluated to characterise the potential model states and to compare the model predictions to the limited amount of available observations. Five hundred posterior parameter sets were selected from these 3500 parameter sets by using an objective function that characterised the difference between the model predictons and available observations. These 500 parameter sets were then used to undertake the predictive analysis of CSG induced GAB flux changes. The results of the analyses indicated that CSG development could potentially induce flux changes in the GAB aquifer – the Pilliga Sandstone. One of the most important variables of interest in the prediction analyses was the increase in flux from the Pilliga Sandstone to the deeper formations due to the lowering of groundwater pressure in the coal seams due to gas and water extraction. This increase in the rate of flow from the Pilliga Sandstone to the deeper formation could be considered as temporary flux losses from the Pilliga Sandstone. The median value of simulated maximum flux losses from the Pilliga Sandstone to deeper formations is 84.52 ML/year. The 5th and 95th percentile of the distribution are respectively 0.28 to 2299.21 ML/year. The median value corresponds to approximately 0.29% of the Long Term Annual Average Extraction Limit of 29.68 GL/y from the GAB groundwater source in this area. The median value of 84.52 ML/year is comparable to the corresponding values of 60 ML/year simulated for the base case water production scenario reported in Santos’ Groundwater Impact Assessment report (CDM Smith, 2016). In their groundwater impact assessment Santos considered 3 cases of water production viz, the base case, the low case and the high case. Unlike that, in our study uncertainty in the water production rates were explicitly accounted in the modelling as uncertain variables. The 5th and 95th percentiles of the total CSG water extraction simulated by this approach are respectively 4.41 GL and 107.11 GL. This range encompasses the total water production of the Base (37.5 GL), Low (35.5 GL) and High cases (87.1 GL) of water production that Santos reported in their Groundwater Impact Assessm

[发布日期] 2017-10-20 [发布机构] CSIRO

[学科分类] 地球科学(综合)