Three-dimensional fracture networks, combined from seismic scale to wellbore scale, can greatly enhance the knowledge of fracture contribution to hydrocarbon storage and flow inside the reservoir. This paper presents some techniques used at ENI-Agip for fracture network simulation at wellbore scale. First, new imaging techniques adopted for enhancing and facilitating the data acquisition from orientated cores will be shown. We will discuss how the fracture data, acquired from the combined analysis of orientated core and wellbore image logs, have to be processed in order to calculate the geometrical parameters of each feature (dip direction, dip, size, terminations) and to classify them according to their filling (oil, water, shale, calcite, etc.). Second, we will illustrate how these kind of data are processed in order to extract the fracture representative parameters needed for stochastic simulation of fracture network at the wellbore scale: spatial distribution along the cored/logged interval, number of fracture sets, representative orientation and statistical distribution of each set, distribution laws of the fracture length and relevant minimum radii and fracture aperture estimate. Fracture porosity evaluation, matrix block size, fracture network connectivity at wellbore scale constitute the outputs of such simulations: they are used to better characterize a fractured reservoir and to describe its behaviour. The synthetic results of the application of such a methodology on a real case (tight packstone/wackestone of carbonate platform from a southern Apennine italian oil reservoir) complete the paper.The main target of the reservoir geologist is the petrophysical characterization of the reservoir in order to estimate: 9 the original hydrocarbons in place; 9 the fluid flow path; and thus 9 the reserves.The building of a geological model, which reasonably describes the lateral and vertical variations of the petrophysical characteristics, i.e. porosity, permeability and hydrocarbon saturation, stands at the root of these estimates. The hydrocarbons in place are, in fact, estimated on the basis of this geological model, and very often a three-dimensional (3D) numerical model, which schematically represents the geological reality, is built in order to estimate the reserves.The petrophysical characterization of fractured reservoirs is more complicated than conventional non-fractured reservoirs because of the presence of two different pore systems: the matrix pore system and the fault-fracture network pore systemThe recognition of the presence of a fracture network is, therefore, crucial because it has a significant effect on reservoir fluid flow. Considering values of fracture porosity that become progressively more significant when compared to the matrix porosity, there are four ways in which fractures can affect a reservoir. Schematically, four types of reservoir can be recognized according to the different contribution of fractures: type 1: type 2: type 3: type 4: fractures impart no positive reservoir quality bu...