Satellite mapping of the optical depth in midlatitude and tropical regions has illustrated that cirrus clouds are frequently finite in nature and display substantial horizontal variability. Vertical inhomogeneity of the ice crystal size distribution and ice water content (IWC) has also been demonstrated in balloon-borne replicator sounding observations, as well as the time series of backscattering coefficients derived from lidar returns. However, methodology for the mapping of three-dimensional (3D) inhomogeneous clouds in space and time based on observational data has not been well developed at this point. We have innovated a remote sensing technique involving the construction of 3D field of IWC and ice crystal mean size of cirrus clouds based on a combination of satellite (AVHRR and MODIS) and ground-based cloud remote sensing observations. From the time series of vertical distribution of cloud parameters, which are derived from ground-based mmCR radar and Raman lidar data, normalized vertical profiles of IWC and mean effective size can be obtained. The 3D cloud parameter fields are determined by combining the normalized profiles with collocated satellite retrieved horizontal distribution of optical depth and mean effective size. We have applied this algorithm to collocated and coincident AVHRR/NOAA and DOE/ARM mmCR and lidar data gathered at the SGP-CART site in the northern Oklahoma area involving a single-layer cirrus on April 18, 1997 and March 9, 2000 during ARM Cloud-IOPs. For satellite retrievals, we use the 0.63/3.7 µm channel radiance correlation technique. The 3D cloud parameter fields thus constructed are validated with ice crystal size distributions independently derived from measurements by optical probes on board the University of North Dakota Citation. The retrieved cirrus cloud optical depths and mean effective sizes are comparable with those determined from the collocated and coincident in situ analyses.