The reduction of carbon dioxide emissions is a prerequisite for stabilization of its concentration in the atmosphere. After PACALA & SOCOLOW (2004), it can be obtained only by implementation of various measures, including carbon capture and geological storage. Concept of carbon capture and storage includes the separation of CO2 from flue gas and its capture at large stationary sources (thermal power plants, cement mills, refineries, natural gas processing plants), safe transport by pipelines or ships to the storage sites and its injection deep into underground. Geological storage can be performed under different geological conditions in sedimentary basins: in depleted oil/gas reservoirs, in oil/gas reservoirs in production (Enhanced Oil/Gas Recovery), in unmineable coal seams and in deep saline aquifers (BACHU, 2000; 2003). Additionally, CO2 can be injected in caverns in salt formations, in fissures (fractures) within basalts and in shales rich in organic matter. The Republic of Croatia has signed the Kyoto protocol in 1999 and ratified it in 2007, thus taking obligation for reducing greenhouse gasses for 5% in the period from 2008 to 2012, with respect to the initial year 1990. Since almost 25% of CO2 emissions in Croatia come from large stationary sources, CO2 capture and storage technology offers a possibility for significant emission reduction. Preliminary screening revealed that favourable natural conditions for CO2 geological storage in the Republic of Croatia exist in oil and gas reservoirs as well as in deep saline aquifers (EU GEOCAPACITY, 2009). It is reasonable to assume that under the circumstances of constant increase in oil and gas prices and development of new methods for enhanced oil/gas recovery, partially depleted hydrocarbon reservoirs in Croatia will stay in production in the near future and will not be available for CO2 storage. Hence, deep saline aquifers (porous and permeable layers situated at a depth greater than 800 m, saturated with highly mineralized water) remain the only possible storage objects. Reliable capacity estimate represents the most important task in the first phase of assessing the certain area for CO2 geological storage. This requires standardized methodology, similar to the methodology developed for estimation of mineral resources. The most frequently cited methodology for estimation of carbon storage capacity in deep saline aquifers is given in studies of Task Force for Review and Identification of Standards for CO2 Storage Capacity Estimation of Carbon Sequestration Leadership Forum (CSLF) from 2005 and 2007. For theoretical storage capacity estimations on a regional scale the authors of CSLF studies suggest calculation of capacity by static trapping, representing the mass of CO2 that can be stored within the pore space of each structural and/or stratigraphic trap present within the regional aquifer, assuming that geometries of all traps are known, which is often not the case. Somewhat different approach has been proposed by National Energy Technology Laboratory of Fossil Energy Bureau of US Department of Energy (US DOE, 2007). It takes into account pore space of the entire aquifer and thus does not require data on number, nor size of the traps. Capacity is calculated as a product of deep saline aquifer surface, its average effective thickness, average porosity, average density of CO2 under conditions of pressure and temperature anticipated for a regional aquifer and storage efficiency factor. A new method for regional capacity estimate is proposed, taking into consideration spatial variations of thickness, porosity and depth of permeable layers. The proposed method is tested on the example of Upper Miocene sandstones corresponding to the Poljana Sandstones of the Kloštar Ivanić Formation in the western part of Sava depression. These sandstones were deposited in turbiditic and deltaic facies that caused the specific morphology of the sandstone bodies. Namely, they are characterised by considerable variations of thickness and porosity. On the other hand, the structure of sedimentation basin (Sava depression) causes pronounced depth variability. Spatial variations of parameters were investigated through the construction of maps based on the data obtained from exploration wells. These data primarily include spontaneous potential and specific resistivity log measurements which were used to define depth and thickness of sandstone layers and logs that were used to define porosity values – density, neutron and sonic logs. Apart from that, in order to determine CO2 density in p, T conditions of a deep saline aquifer, values of pressure and temperature in the aquifer are required. The boundaries of the regional aquifer in Poljana Sandstones are defined with respect to aquifer depth, with the southern boundary being set simply a boundary of estimation area, mostly due to data availability. Based on the interpretation of electric logs, marl directly overlying Poljana Sandstones was chosen as a regional isolator, since it can be identified on all the available electric logs and it is situated at suitable depth. Further investigation is needed to prove its ability to act as a barrier to CO2 migration, including laboratory capillary pressure measurements as well as measurements of relative permeability for brine and CO2. On the basis of spatial variations of effective thickness, porosity and CO2 density, the specific storage capacity is calculated andmapped. For that purpose, the area of deep saline aquifer was divided in 178 blocks – quadric prisms and the specific capacity calculated for each of the prisms using the values read from the centre of each block. The largest specific storage capacity was estimated in the central and south-eastern part of the investigated area, where the deep saline aquifer is of largest thickness and is situated at the greatest depth. However, the thickness of the regional isolator is reduced in this area, so the special attention should be dedicated to its detailed characterization. The most important parameter in estimates of specific storage capacity for units characterized by significant variability of thickness is the effective thickness that can be relatively reliably defined, while the spatial variability of porosity, which is far more difficult to define, has lesser impact on estimated values of specific storage capacity. The proposed method can be used in the early phase of exploration with aim of defining deep saline aquifers in rocks with intergranular porosity and estimating their potential for CO2 geological storage. Mapping of specific storage capacity should enable the authorities to determine the regions for which exploration allowances might be issued and the results of this second phase would eventually lead to planning and construction of the subsurface CO2 storage facilities.