The blending of observation-driven heuristic storm (0 - 2 h) nowcasts with numerical weather prediction model output can lead to improved skill in very short-term (1 - 6 h) forecasting of convective storms. However, it remains to be shown how the blending may best be accomplished. In this study, several blending techniques are discussed and evaluated for initial use in the FAA-supported Consolidated Storm Prediction for Aviation (CoSPA) development using one month's worth of data from the summer of 2006. A flexible (“plug-and-play”) evaluation tool has been developed that enables easy selection among several input data streams and blending procedures, and computes various statistical assessment metrics. The input data streams currently include extrapolations (TITAN-based and variational echo tracker) and model data (RTFDDA MM5 and cold start WRF). The current set of algorithms under evaluation includes: (1) a simple weighted averaging technique; (2) an image matching approach; (3) a rules-based approach; and (4) a variational method. The evaluation tool is easily expandable to include additional data streams and blending techniques. Statistical results are presented from an initial intercomparison of these blending techniques, model forecasts and extrapolations using standard skill scores. The discussion includes illustrative examples of success and failure for the various algorithms. In addition, new evaluation techniques that provide more descriptive information on the performance of the blending techniques relevant to aviation planning (e.g., storm area, organization, orientation) are highlighted as well.