2.6.2.5.2 Coal seam gas wells

For numerical modelling purposes the CRDP was finalised in October 2015, and while AGL withdrew from its proposed Gloucester Gas Project in December 2015, as described in the companion submethodology M04 (as listed in Table 1) for developing a coal resource development pathway (Lewis, 2014), the CRDP was not revisited.

While the CRDP includes both Stage 1 and Stage 2 of the AGL gas field development area, only Stage 1 is modelled due to the availability of information (see companion product 1.2 for the Gloucester subregion (Hodgkinson et al., 2014)). In Stage 1, 110 wells are planned. For the purposes of the modelling, it is assumed that the wells are drilled in a four-year period between 2013 and 2017. Each well will be actively exploited in the model for a 20-year period. As the exact location of these wells is not known at this stage, 110 wells are randomly distributed within the AGL Stage 1 gas field development area (Figure 14) subject to the following rules, in line with the locational principles outlined in Parsons Brinckerhoff (2015):

The locational principles in Parsons Brinckerhoff (2015) are more detailed and include rules with regards to distance to existing residences (minimum 200 m) and watercourses (minimum 40 m for major watercourses, 20 m for minor watercourses) as well as rules to avoid significant vegetation and riparian areas and heritage sites. As these detailed rules are not incorporated in the random placement of coal seam gas wells in the GW AEM due to operational constraints, it is unlikely that the CSG wells will eventuate in the locations indicated in Figure 14. Note that in this figure a number of CSG wells are located within the Avon alluvium MODFLOW model domain. This merely indicates that some wells will depressurise coal seams directly underneath the alluvium, it does not indicate water being extracted from the alluvium.

In the analytic element modelling code Ttim (Bakker, 2015), it is not possible to switch off a head-dependent well during a transient run. In order to simulate recovery, a two-stage process is adopted. In the first stage, water production for each individual well screen is computed with a transient run that covers the exploitation period, with the CSG wells implemented as head-dependent wells with a well diameter and entry resistance nominally set to 0.25 m and 100 days for all wells. In the subsequent run, that covers the entire simulation period, the CSG wells are implemented as flux-specified wells, with a flux during production from the previous simulation and zero flux after exploitation has stopped. For each parameter combination, the pumping rate will vary and is therefore recorded as a yearly time series as part of the model output. The volume of water required to depressurise the targeted model coal seams to the specified groundwater head will be simulated by the numerical model.