In this work, we address the challenge of modelling a complex, carbonate reservoir, where the fractures network, connected throughout a complex fault framework, represents large part of both the storage and the flow capacity of the system. The asset is a giant, onshore field, developed since the 90's by primary depletion through several horizontal wells, targeting anomalous fluid columns. Different culminations are characterized by specific production drive mechanisms. The objective is to integrate an impressive amount of data into a digital model, suitable to understand fluid flow behavior and support decision. The field is challenging in every geological and dynamic feature. The reservoir complexity ranges from the intricate structural framework (several hundreds of reverse faults), to the puzzling fractures network at different scales, to the unclear role of the low-porosity rock matrix, to the heterogeneous distribution - both laterally and vertically - of fluid properties, related to different combinations of hydrocarbon and acid components. The workflow is based on the adoption of Volume Based Modelling (VBM) to account for seismic faults. Then, large-scale fractures are modelled using a blend of stochastic and deterministic Discrete Fracture Networks (DFNs), while background fractures (BGF) are characterized using a Continuous Fracture Modeling (CFM) formulation. A Dual Porosity - Dual Permeability (DPDK) approach is then implemented for reservoir simulation. The model is finally reconciled with the production data by iterating between geology and simulated dynamic response. The whole modeling and simulation workflow, from static to dynamic model definition, is developed relying on company's top-class computational resources. The DPDK formulation, where DFN is the second medium while the first medium consists of BGF and rock matrix, allows us to simulate the main production mechanism: large-scale discontinuities – DFN – are withdrawal first, and then fluid is recharged by smaller scale features. Besides, the history matching phase, together with accurate production and Pressure-Volume-Temperature (PVT) data analysis, sheds light on the extreme heterogeneity of the field. Petrophysical properties, storage and effective apertures of discontinuities are calibrated according to the production history, and integrated into a comprehensive understanding of the reservoir. Eventually, we reveal how a robust history matched model may be used as a powerful tool to understand the impact of all the involved criticalities on the subsurface fluid behavior and movement in a complex fractured carbonate setting. The challenges addressed in this work provide relevant best practices for carbonate reservoir modelling, in particular highlighting the role of the integration between geology and reservoir engineering to minimize subsurface uncertainties. Furthermore, the PVT model developed in this study proposes new migration scenarios to explain the sour gas distribution. Finally, optimized procedures to tackle numerical criticalities using advanced reservoir simulators are disclosed.