For one essential ingredient of permeable fracture networks (degree of fracture pore-space preservation in large fractures), I show how the characterization challenge presented by sparse fracture sampling can be overcome by measuring a surrogate, the abundance of rock-mass cement that precipitated after fractures ceased opening. Sampling limitations are overcome because the surrogate is readily measured in small rock samples, including sidewall cores and cuttings, permitting site-specific diagnosis of the capacity of fractures to transmit fluid over a wider range of sample depths than conventional methods allow. A diverse core database shows that this surrogate correctly predicts where large fractures are sealed. Information on timing of fracture opening relative to cement sequence can be obtained in two ways. First, evidence of fracture-movement history and cement sequences in sparse large fractures can be extrapolated to areas having only cement data. Alternately, evidence of fracture timing can be acquired from sealed, micrometer-scale fractures. Distribution of porosity-reducing cement is commonly heterogeneous (from bed to bed and location to location) in siliciclastic and carbonate rocks. However, because patterns of sealed or open fractures cannot be delineated using fracture observations alone, surrogates have practical value for production fairway mapping and other applications in which identifying open fractures is essential. This study highlights the vital interplay among structural and diagenetic processes for fracture-porosity preservation or destruction.