3.2.1. CO2 Storage Potential in the Regional Deep Saline Aquifer Dugi otok (DSA Dugi otok)

Initially formed as a foreland basin ahead of SW-propagating Dinaric thrust system, sedimentation of siliciclastic deposits in the Dugi otok basin took place during the Late Eocene, Oligocene and Miocene [25,47,48]. Miocene series is made of chalk limestones with marly and sandstone interbeds. Chalk limestones and sandstones have good reservoir properties; their porosity is in 15–25% range. Based on regional seismic interpretation (seismic facies), Miocene sediments are mostly comprised of a stacked sequence of sandstone and marl layers, with some subordinate lateral lithology variations. Lower and Middle Miocene sandstone layers are the ones where CO2 might be injected. More precisely, these are the layers and lenses of silty sandstones at depth range of 700–2100 m, regionally SW-dipping in the form of monocline unconformably covered by Pliocene marls, thus are considered as prospective for geological CO2 storage.

In Table 2, the main characteristics of DSA Dugi otok are presented together with the parameters used to calculate its theoretical CO2 storage capacity. An aquifer is treated as if it makes a consistent single large unit with average depth and porosity values and an estimated small proportion of pores that will eventually be filled with carbon dioxide once its plume spreads throughout the unit. This is an oversimplification of the effect of many processes that will eventually contribute to geological storage, in line with the so-called "conservative approach" taken in EU GeoCapacity CO2 storage atlas [7]. Outline of the aquifer is shown in Figure 4. Until a detailed exploration of Miocene units in this area is made, their presence was estimated in the region where the total thickness of Eocene to Holocene sediments exceeds 3000 m. Both the storage efficiency coefficient estimate (taken as 0.02 after [49] as P50 value for clastic regional deep saline aquifers) and the way in which the unit is mapped are major contributors to the large uncertainty in the calculation of the storage capacity. This means that the number of 327.075 Mt presented in Table 2 is just a first numerical estimate of the potential and should by no means be directly compared with the numbers given in Table 1, where the potential in depleted gas fields was estimated.


**Table 2.** Characteristics of the regional deep saline aquifer Dugi otok.

The CO2 storage capacity was calculated using the compressibility method (after [50]) and the volumetric method as described in [49]. The obtained capacities were than summed up, following the approach suggested by [51] that assumes that additional pore volume will be available for CO2 storage due to compressibility of pores and initially present pore water. If pressure increase is considered, pore compressibility should be included to storage assessment:

$$c\_p = 1/\delta V \times (\delta V / \delta p) \tag{3}$$

where: *cp*: pore compressibility (bar−1); δ*V*: change in pore volume resulting from the change in pressure (m3); δ*p*: change in pressure due to injection (bar).

The maximum pore pressure was estimated to be 10% above the initial pore pressure, which is significantly less than what is estimated for structurally defined aquifers in carbonates. The reasoning behind this is that the initial pore pressure could not be expected to be intensively increased within the volume of the entire regional deep saline aquifer; the intensive pressure increase would be limited to volumes of the regional deep saline aquifer in the surroundings of the injection well, but overall average pore pressure should not increase as significantly, as it does for structurally defined deep saline aquifers; i.e., it should not be comparable to fracture pressure. The average porosity value was estimated from regional data on Miocene sandstones in the study area. Pore compressibility was estimated using the correlation of pore compressibility with the net confining pressure developed for Bandera sandstones [52], which have a similar initial porosity to the Miocene sandstones of DSA Dugi otok, amounting to 16.5%. The pore water compressibility was calculated after following equation [53]:

$$\mathcal{L}\_{\overline{w}} = -\frac{1}{B\_{\overline{w}}} \left( \frac{\partial B\_{\overline{w}}}{\partial p} \right)\_{T} = \frac{1}{\left[ 7.033 \, p + 541.5 \, \text{C}\_{\text{NaCl}} - 537.0 \, T + 403.300 \right]} \tag{4}$$

where *cw* is water compressibility (1/psi), *Bw* is water formation volume factor, *p* is pressure (psi), *T* is temperature (◦F), and *CNaCl* is salinity of pore water (g NaCl/l).

For the salinity of pore water, the value of 35 g NaCl/l, corresponding to salinity of seawater was taken. Initial pore pressure and overburden pressure were calculated using brine density of 1025 kg/m<sup>3</sup> and bulk wet density of overlying sediments of 2400 kg/m3. Density of CO2 was calculated based on the estimated values of pressure and temperature, using the equation of state after [46]. Average temperature was estimated using the geothermal gradient of 1.57 ◦C/100 m, which was calculated from data on temperature of the sea bottom [54] and the regional isothermal map of formation temperatures at the depth of 3000 m [31]. Pressure was calculated assuming the hydrostatic pressure gradient according to [10].
