2.6.1.5 Uncertainty

Summary

In the quantitative uncertainty analysis a large number of parameter combinations are evaluated and combined with the results of the groundwater modelling to obtain the maximum raw change, the year of maximum change and the relative change between baseline and coal resource development pathway (CRDP) conditions for the nine hydrological response variables.

The fit between observed and simulated historical hydrological response variable values is used to constrain the Australian Water Resources Assessment landscape model (AWRA-L) parameters while the simulated coal seam gas extraction rate and the simulated surface water – groundwater flux at Casino is used to constrain the groundwater model results. Only parameter combinations that are considered to meet all of these three criteria are accepted as behavioural parameter combinations that are used to make predictions. This set of behavioural parameter combinations is different for each hydrological response variable.

The quantitative uncertainty analysis is followed by a qualitative assessment of the effect of model assumptions on the prediction. The largest sources of uncertainty not captured in the quantitative uncertainty analysis are considered to be the availability of calibration catchments, the selection of objective functions for the uncertainty analysis and the interaction with the groundwater model.