Comparisons of monthly daytime cloudiness time series of total cloud cover (TCL) and cloud types' occurrences between the ground-based and International Satellite Cloud Climatology Project (ISCCP) D2 datasets were conducted over the contiguous United States. Despite its large bias (+6.7% to +20.8%), TCL in ISCCP D2 is highly correlated with the surface observations in all seasons. Close correspondence of seasonal and interannual variability between these two datasets is also seen in low stratiform and cirrus clouds. Reconstruction of these cloud characteristics using ISCCP data was performed for the 1990s when the Automated Surface Observation System (ASOS) implementation irreversibly broke the homogeneity of ground-based cloudiness time series over the United States. Verification of reconstructed cloudiness characteristics was conducted using the independent in-situ climatic variables such as diurnal temperature range and precipitation.