2.6.2.4.4 Surface water – groundwater interactions

Surface water – groundwater interactions are represented as a linear feature in the GW AEM to which a constant groundwater level of zero metres is assigned. The linear feature consists of line segments approximating the main channel of the Belyando River (Figure 10). The line segments correspond to the stream reaches used in the surface water modelling (see companion product 2.6.1 for the Galilee subregion (Karim et al., 2018)).

The Belyando River receives some baseflow from shallow aquifers, as indicated in the potentiometric surface for the Cenozoic sediments in Section 2.1.3.3.2 in companion product 2.1-2.2 for the Galilee subregion (Evans et al., 2018). This groundwater discharge occurs either as a baseflow contribution to the Belyando River or through direct evapotranspiration by vegetation. This means the groundwater levels underneath the Belyando River system are mostly controlled by the riverbed elevation and the extinction depth of evapotranspiration, provided the Belyando River remains a regional discharge location.

The other rivers and creeks in the model domain are not considered to be discharge locations for regional groundwater flow and are not represented in the analytic element model. These ephemeral streams, when flowing, are likely to be losing streams – that is, they could locally recharge the alluvial aquifer. For the majority, they will be losing disconnected – that is, the loss rate is independent of the groundwater level in the aquifer (Brunner et al., 2009). In that case, any change in the groundwater level due to coal resource development will not affect the surface water – groundwater interaction flux. Not representing these streams will result in a conservative estimate of the drawdown as the simulated drawdown cannot be compensated by an increase in the water flux entering the aquifer through the stream. However, wherever drawdown is simulated under creek beds, local information needs to be sought to establish the connection status of the creek at that location to evaluate the potential change in surface water – groundwater flux.

For the Belyando River, any drawdown simulated in the Cenozoic/alluvium aquifer will not result in a drawdown underneath the Belyando River, but it will result in a reduction in the surface water – groundwater flux. The change in surface water – groundwater flux thus simulated is subtracted from the total streamflow for that stream segment calculated by the Australian Water Resources Assessment landscape model (AWRA-L) (see companion product 2.6.1 (Karim et al., 2018)). As pointed out in the methods section (Section 2.6.2.1) in Equation (1), the minimum resulting streamflow is zero.

This implementation of the river boundary results in an overestimate of the reduction in surface water – groundwater flux, as it comprises both the change in baseflow and groundwater evapotranspiration. It does this only locally, however, as in the immediate vicinity of the river channel drawdowns are under estimated.

The Carmichael River receives baseflow through discharge from the Clematis Group aquifer as well as through outflow from the Doongmabulla Springs complex, which is also sourced from the Clematis Group aquifer. Any change in the flow rate of these springs therefore has the potential to change the baseflow contribution to the Carmichael River. Previous modelling, such as Turvey et al. (2015) indicates that these changes are very small. The change in baseflow in the Carmichael River due to changes in spring flow rates is not simulated explicitly or incorporated in the AWRA-L model. This assumption is only valid if the changes in spring flow rate are very small. The Doongmabulla Springs complex is therefore included in the model to quantify the change and verify this assumption. Further discussion about the conceptual understanding of the Doongmabulla Springs complex is provided in companion product 3-4 (Lewis et al., 2018) for the Galilee subregion.