Mesoscale precipitation bands within extratropical cyclones may lead to extreme precipitation events, causing significant societal disruption, economic losses, injuries, and even casualties. To extend the work of such precipitation bands over the United States and understand this weather system more thoroughly, this thesis first discussed a five-year climatology and composite study of precipitation bands associated with extratropical cyclones over the British Isles from April 2017 to March 2022. The database includes 249 single bands associated with 167 parent extratropical cyclones. The geographical distribution of these bands varied seasonally. Bands were frequently observed west of Scotland (30% and 27%, respectively) in winter and spring, whereas they were frequently observed west of Ireland (30%) in summer and autumn. A greater number of bands developed over coastal waters compared to inland regions, with most exhibiting a north-south orientation. Bands were classified into six categories based on their locations relative to the front. Specifically, occluded-frontal bands (19 year-1), warm-frontal bands (11 year-1), and cold-frontal bands (10 year-1) were the three most common categories of bands over the British Isles. Occluded-frontal and warm-frontal bands typically formed west of Scotland, stretching northwest to southeast, whereas cold-frontal bands were more frequently formed southwest of Great Britain, stretching southwest to northeast. The low-level jet provides moisture to the bands, matching the location and scale of composite bands, similar to an atmospheric river. These findings are compared to earlier research on band formations in the United States. Based on the database of precipitation band cases over the British Isles, an extensive series of simulations using the Weather Research and Forecasting (WRF) model were conducted. However, these simulations consistently struggled to accurately replicate banded precipitation. The uncertainty of cloud microphysics schemes is one of the important sources of simulation biases. To further understand the performance of different cloud microphysics schemes on the simulations of precipitation bands, a narrow cold-frontal rainband (NCFR) and a wide cold-frontal rainband (WCFR) over the UK on January 24, 2018, were chosen. This research used three schemes: the WRF double-moment 6-class microphysics (WDM6), Thompson, and Morrison schemes. All three simulations overpredicted the intensity and area of NCFRs, whereas they underpredicted those of WCFRs compared to the Met Office radar data. The WDM6 simulation showed the most accurate results. Specifically, the scheme predicted more intense NCFR in the development stage but weaker NCFR in the dissipating stage than the Thompson and Morrison mainly due to rain mixing ratio at 1 km altitude. The broader and stronger WCFR in the WDM6 simulation was linked to higher amounts of low-level snow, graupel, and rain, attributed to faster aggregation, accretion and melting rates, respectively, in the WDM6 simulation. Therefore, differences in low-level ice-phase processes across microphysics schemes significantly impacted the WCFR intensity, whereas differences in warm-rain processes were less pronounced. The smaller raindrop size in the WDM6 simulation also contributed to a broader and stronger WCFR by transporting raindrops rearward from the cold front. The evaluation of three microphysics parameterization schemes in a precipitation band case led to a question of whether different microphysics parameterization schemes have systematic biases in simulating precipitation. To address this problem, a large language model-based review of 2,699 publications on precipitation simulations using the WRF model was conducted to investigate systematic biases in different microphysics parameterizations. The publications, identified from Web of Science and Scopus prior to June 22, 2023, were analyzed using GPT-4 Turbo to extract details on model configurations and performance. The research highlighted nine common parameterizations, with one-moment parameterizations dominating before 2020 and double-moment parameterizations becoming more popular afterwards. The Lin, WRF Single-Moment (WSM versions 3,5,6), and WRF Double-Moment (WDM versions 5,6,7) were preferred for use in South Asia and China. In contrast, the Ferrier, WSM6, Thompson, and Morrison were preferred for use in the United States. Seven of the nine parameterizations generally overestimated precipitation. The Ferrier showed the lowest median RMSE (2.19 mm/day), while GCE had the highest median correlation coefficient (0.83), though both had smaller sample sizes (39 and 68, respectively).
| Date of Award | 24 Jan 2025 |
|---|
| Original language | English |
|---|
| Awarding Institution | - The University of Manchester
|
|---|
| Supervisor | David Schultz (Main Supervisor) & Zhonghua Zheng (Co Supervisor) |
|---|