Hristo G. Chipilski - June 21

School of Meteorology Colloquium

ASSIMILATION OF GROUND-BASED REMOTE SENSORS FOR IMPROVING

THE PREDICTION OF BORE-GENERATING NOCTURNAL CONVECTION

Hristo G. Chipilski

Monday, June 21^st

9:00am

Google Meet link: https://meet.google.com/ssj-ytfx-zeo

Or dial: ‪(US) +1 903-471-8006 PIN: ‪194 533 171#

        Advances in Numerical Weather Prediction require synchronous enhancements to the underlying Global Observing System. However, our recent progress in convective-scale modeling has not been accompanied by a commensurate increase in the number of high-resolution measurements. Over the last two decades, for example, we have identified a striking gap in the observation coverage within the Planetary Boundary Layer (PBL). This gap is particularly detrimental for convective-scale models since their accuracy depends strongly on the analysed PBL structure. In an effort to address this problem, scientists and engineers have begun to develop novel ground-based remote sensing technology. The ability of these new instruments to accurately depict the fine-scale processes taking place in the PBL is expected to bring considerable benefits for the next generation of convective-scale models.

            In this colloquium, the above claim will be supported by demonstrating how the assimilation of ground-based remote sensors can systematically improve the forecasts of bore-generating convective systems observed during the Plains Elevated Convection at Night (PECAN) field campaign. Aggregate verification statistics from 10 diverse PECAN cases show that a combination of in situ and remote sensing instruments produces the largest increase in forecast skill, both in terms of the parent mesoscale convective systems as well as the explicitly resolved atmospheric bores. Our statistics also indicate that it is advantageous to collocate thermodynamic and kinematic profilers. Nevertheless, the impact of different ground-based networks tends to be highly flow-dependent, with thermodynamic (kinematic) instruments playing a dominant role in cases with relatively low (high) convective predictability. Deficiencies in the data assimilation method as well as inherent complexities in the governing moisture dynamics can further limit the forecast value extracted from single remote sensing profilers.