Permeable pavement functions as a porous infrastructure interface allowing the infiltration and evaporation of rainfall-runoff while functioning as a relatively smooth load-bearing surface for vehicular transport. Hydraulic conductivity (k) of permeable pavement is an important hydraulic property and is a function of the pore structure. This study examines k for a cementitious permeable pavement (CPP) through a series of pore-structure models. Measurements utilized include hydraulic head as well as total porosity, (phi(t)), effective porosity (phi(e)), tortuosity (L-e/L) and pore size distribution (PSD) indices generated through X-ray tomography (XRT). XRT results indicate that the permeable pavement pore matrix is hetero-disperse, with high tortuosity and phi(t) not equal phi(e). Power law models of k-phi(t), and k-phi(e) relationships are developed for a CPP mix design. Results indicate that the Kruger, Fair-Hatch, Hazen, Slichter, Beyer and Terzaghi models based on simple pore-structure indices do not reproduce measured k values. The conventional Kozeny-Carman model (KCM), a more parameterized pore-structure model, did not reproduce measured k values. This study proposes a modified KCM utilizing phi(e), specific surface area (SSA)(pe) and weighted tortuosity (L-e/L)(w). Results demonstrate that such permeable pavement pore-structure parameters with the modified KCM can predict k. The k results are combined with continuous simulation modeling using historical rainfall to provide nomographs examining permeable pavement as a low impact development (LID) infrastructure component. (c) 2010 Elsevier B.V. All rights reserved.