An Example of the use of Synthetic 3.9 µm GOES-12 Imagery for Two-Moment Microphysical Evaluation

In preparation for the launch of the next generation of Geostationary Satellites (GOES-R), considerable effort has been placed on developing new products and algorithms for operational purposes. In addition to satellite-based products and algorithms, satellite imagery can be used to evaluate numerical weather prediction models. Important first steps have already been undertaken to produce synthetic satellite imagery from numerical model output. By comparing synthetic imagery to observed imagery, model performance can be evaluated with a relatively new metric.

In this study, synthetic GOES-12 imagery was used to improve the two-moment prediction of pristine ice in the RAMS mesoscale model. This model was used as part of the GOES-R risk reduction activities. A thunderstorm event that occurred on 27 June 2005 over the central plains of the United States was chosen for study. Because GOES-R observations are yet to be available, synthetic GOES-12 3.9 µm imagery of RAMS output was compared to observed GOES-12 3.9 µm imagery. A discrepancy between brightness temperatures of two anvils of thunderstorms led to an improvement in the prediction of pristine ice number concentrations. After the model was re-run, subsequent synthetic GOES-12 3.9 µm imagery of one anvil exhibited a reduced discrepancy with observed imagery. The second anvil became too warm, an issue that may be related to model-specified CCN concentrations. This example highlights the potential importance of using synthetic imagery to evaluate numerical weather prediction models.