Methods for cloud characterization for solar forecasting using sky imagery are presented. Sky images taken every 30 seconds were processed to determine cloud cover using a clear sky library (CSL) method. The CSL method utilizes spectral color information from single-frame sky image and detects clouds reliably outside the circumsolar region. A sky image sequence segmentation (SISS) method combing multi-frame sky images, contour detection, and random walk image segmentation is developed and shown to outperform the CSL method. Cloud motion vectors were generated by cross-correlating two consecutive sky images and variational optical flow (VOF). Cloud locations were forecasted using a frozen cloud advection method at 30 second intervals up to a forecast horizon of 15 min. A month of image data was analyzed to compare the accuracy of VOF forecast with cross-correlation method (CCM) and image persistence method. The VOF forecast with a fixed smoothness parameter was found to be superior to image persistence forecast for all forecast horizons for almost all days and outperform CCM forecast with an average error reduction of 39%, 21%, 19%, and 19% for 0, 5, 10, and 15 min forecasts respectively. Optimum forecasts may be achieved with forecast-horizon-dependent smoothness parameters. Cloud stability and forecast confidence was evaluated by correlating point trajectories with forecast error. Point trajectories were obtained by tracking sub- sampled pixels using optical flow field. Point trajectory length in minutes was shown to increase with decreasing forecast error and provide valuable information for cloud forecast confidence at forecast issue time. A technique for characterizing and predicting hours-ahead solar forecast error using cloud patterns as predictors is also presented. The dependence of forecast root-mean-square error (RMSE) on different cloud patterns is analyzed using multivariate linear regression and analog model. Image entropy or cloud randomness is shown to contribute the most to the forecast RMSE