Reconstructing Pore System Evolution with Digital Image Analysis (DIA): A case study from Miocene Lacustrine Carbonates (Southern Germany)

Carbonates often experience a heterogeneous distribution of pore system properties caused by overprinting of the primary pore fabric by multiple phases of diagenetic modifications, leading to high uncertainties in predicting porosity and permeability. This has been also observed in the Middle Miocene marginal lacustrine carbonates of the Nördlinger Ries crater lake in Southern Germany, where primary pore types are overprinted by subsequent phases of diagenetic processes and their products, such as the formation of secondary pores by dissolution and the occlusion of pores by calcite cement. These modifications in the pore system occur patchily, vary at sub-centimeter scale and show no facies-dependence. The goal of this study was to precisely capture diagenetic features on thin section images by using Digital Image Analysis (DIA) and then quantify the impact of secondary pore formation and cementation on porosity and permeability generation. This further allows for reconstructing the pore system throughout each stage of its evolution, both numerically and visually. At first, thin section images have been acquired from rock cylinders which have been used to determine total porosity and permeability with routine gas-injection method, revealing a non-relationship between both petrophysical parameters in the studied lacustrine carbonate succession. The images were then processed with DIA in order to segment the pore space, classify primary and secondary pore types, and detect the calcite cement. Various pore geometry parameters have then been measured to infer porosity and permeability. By processing each thin section image with a graphic software, pore system representations for each diagenetic stage have been produced and then repeatedly analysed with DIA. The resulting quantitative data have then been compared with the previously obtained values of the original pore system, resulting in distinct values of porosity-permeability change for diagenetic stage. Due to the formation of secondary pores, porosity has increased to 14.42 % in average and permeability has increased towards 227.07 mD. Cementation of pore space instead caused a decrease in porosity towards 8.44 %, whereas permeability has decreased to 154.53 mD. The percentage of change in porosity (- 41.47 %) and permeability (- 31.94 %) by cementation can then be used as quantitative measures characterizing the impact of cementation on porosity and permeability. Since cementation occurs patchily, varies at a sub-centimeter scale and can therefore not be precisely located in the studied outcrop, these distinct values serve as lower threshold values for porosity and permeability estimations. This allows for assessing the origin of the previously observed heterogeneous porosity-permeability distribution which cannot be realized by using routine petrophysical measurements, solely. With the hereby presented approach it is demonstrated how quantitative information obtained from DIA applied on thin section images highly improve the prediction of porosity and permeability as well as the distribution of diagenetic features both in the outcrop and the subsurface.