Development of a Coupled GOES-R Legacy Sounding NearCast with Convective Initiation Products to Improve Convective Weather Nowcasts

To improve and reduce over-forecasting within the current version of the 0-1 h GOES-R convective initiation (CI) algorithm, 1-9 h NearCasts of GOES-R Legacy Sounding Moisture and Temperature products are incorporated within GOES-R CI as a means of better differentiating areas in which storms are most likely to grow from those where growth is less likely. This procedure maximizes use of all capabilities of the forthcoming GOES-R Advanced Baseline Imager (ABI) visible and infrared (IR) high time-resolution (1-5 min) imagery, as well as the 15-30 min interval clear-air profiles (especially moisture). The CI and NearCast datasets are physically consistent and complimentary, often providing a better depiction of evolving stability patterns in advance of storm development than are available from numerical weather prediction (NWP). It is known that NWP models suffer from significant forecast errors, especially with respect to convective-scale quantitative precipitation forecasts in the first 1-9 h of a forecast, and in summer when operational Threat Scores can be as low as 12%. GOES-R CI and NearCast methods are designed to mitigate these NWP forecast deficiencies, with NearCast providing a consistent, frequently updated depiction of the vertical and horizontal distribution of moisture in the pre- and near-storm environment and CI providing improved situational awareness of which radar returns are most likely to intensify once cloud growth has commenced. For this presentation, data from the current GOES-E satellite will be used.

Through fusion of GOES-R CI and NearCast, forecasters will gain several advantages over using both products individually, including low-level moisture and boundary detection, which becomes a method of enhancing GOES-R CI performance. The ability of NearCasts to improve the depiction of water vapor features and gradients at the full resolution provided by the satellite observations will help isolate boundaries of moisture in the near-storm environment that lead to CI and upscale convective storm development. The Lagrangian-transported NearCast temperature and moisture fields (used to deduce potential instability) identify general regions that are favorable for CI over the coming few hours. These fields will provide additional input to GOES-R CI (in a logistic regression framework), information that is presently lacking within the algorithm. High-impact severe weather event days over the U.S. will be evaluated with results presented depicting the benefits of NearCast/CI fusion.