Date of Award

1-1-2023

Language

English

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

College/School/Department

Department of Atmospheric and Environmental Sciences

Content Description

1 online resource (xiv, 101 pages) : illustrations (some color)

Dissertation/Thesis Chair

Everette Joseph

Committee Members

Cheng-Hsuan Lu, Ryan Torn, Scott Miller

Keywords

Weather forecasting

Subject Categories

Atmospheric Sciences

Abstract

In New York State (NYS), the prediction of high-impact severe weather event has presented a challenging forecasting problem. The NYS Mesonet (NYSM) has the potential to improve the severe weather forecasting through its continuous in situ and remote sensing measurements on the lower troposphere. The dense observing network can capture the evolution of mesoscale motions with high temporal and spatial resolution. The objectives of this dissertation are (1) to assess whether the assimilation of NYSM observations into numerical weather prediction models could improve model analysis and short-term weather forecasting, and (2) to figure out the important key factor and physical process, which could be complemented by observations, in the prediction of targeted weather system. A severe weather system that is specifically aimed for this study is a convective event on 21 June 2021, especially its phase of the convection re-intensification triggered by the southerly Mohawk-Hudson Convergence (MHC).In the first part of the dissertation, several data assimilation (DA) experiments are conducted to investigate the impact of NYSM data using operational DA system Gridpoint Statistical Interpolation with rapid update cycles. The assimilated datasets include National Centers for Environmental Prediction Automated Data Processing global upper-air and surface observations, NYSM surface observations, Doppler lidar wind profiles, and microwave radiometer (MWR) temperature and specific humidity profiles at NYSM profiler sites. In comparison with the control experiment that assimilates only conventional data, the timing and location of the convection re-intensification was significantly improved by assimilating NYSM data, especially the Doppler lidar wind profiles. Our analysis indicate that the improvement could be attributed to improved simulation of the southerly MHC. However, the MWR DA resulted in degraded forecasts, likely due to large errors in the MWR temperature retrievals. Overall, the results suggest the positive impact of assimilating NYSM surface and profiler data on forecasting summertime severe weather. In the second part of the dissertation, three experiments are conducted by using Mellor-Yamada-Janjic (MYJ), Yonsei University (YSU), and Mellor-Yamada Nakanishi Niino (MYNN) level 3 PBL schemes in the Advanced Weather Research and Forecasting (WRF) Model. The results show that the onset of MHC-triggered convection is primarily affected by the development of its antecedent convective system which is sensitive to the choice of PBL schemes. Among the experiments, MYJ (MYNN) predicted earlier (later but accurate) onset of MHC-triggered convection due to shallow (weak) vertical mixing and higher (lower) convective available potential energy (CAPE). YSU predicted a deepest and strongest vertical mixing with the onset time between MYJ and MYNN. The reason for weak vertical mixing in MYNN is that higher low-level cloud cover inhibited the incoming solar radiation. Overall, the boundary layer moisture distribution, turbulent mixing, and cloud cover in the pre-convective environment play a key role in the prediction of MHC-triggered convection, which could be beneficial from high-resolution moisture profiling measurements.

Download

Share

COinS