Return to search

A Climatological Analysis of Upper-Tropospheric Velocity Potential Fields using Global Weather Reanalysis, 1958-2020

Upper-tropospheric (200 hPa) velocity potential is useful in identifying areas of rising or sinking atmospheric motions on varying temporal scales (e.g., weekly, seasonal, interannual) especially in the global tropics. These areas are associated with enhancement (rising motion) or suppression (sinking motion) of tropical convection and subsequent weather phenomena dependent on these processes (e.g., tropical cyclones). This study employed three commonly used global weather reanalysis datasets (NCEP/NCAR Reanalysis 1, JMA JRA-55, ECMWF ERA5) to calculate and compare upper-tropospheric velocity potential fields on varying temporal scales and quantify any differences that existed between them from 1958 to 2020 over four key regions of variability (Equatorial Africa, Amazon Basin, Equatorial Central Pacific, and Equatorial Indonesia). To supplement this analysis, the highly correlated variables to velocity potential of outgoing longwave radiation (OLR) and daily precipitation rate were used and directly compared with independent OLR and precipitation datasets to determine the reanalysis' level of agreement with the independent data. The ECMWF ERA5 held the highest agreement to these data over all regions examined and was reasoned to have the highest confidence in capturing the variability of upper-tropospheric velocity potential fields for the study period. Confidence was decreased in the usefulness of the NCEP/NCAR Reanalysis 1 as it consistently performed poorly over much of the study domain. The results of this study also emphasized the usefulness in ensemble-based approaches to assessing climate variability and understanding potential biases and uncertainties that are inherent in the data sources. / Master of Science / Historical weather data across the globe is analyzed using global weather reanalysis datasets which provide the most complete picture of how the atmosphere has evolved over the course of the last several decades. This data is a vital component in today's research investigating climate change and variability over time. This study examined how the history of upper-tropospheric velocity potential was captured in three commonly used global weather reanalysis datasets (NCEP/NCAR Reanalysis 1, JMA JRA-55, ECMWF ERA5) from 1958 to 2020 over four key regions of variability (Equatorial Africa, Amazon Basin, Equatorial Central Pacific, and Equatorial Indonesia). The variable of velocity potential is useful in identifying areas of rising or sinking atmospheric motions on varying time scales (e.g., weekly, seasonal, interannual) especially in the global tropics. These areas are associated with enhancement (rising motion) or suppression (sinking motion) of tropical convection (i.e., thunderstorms) and subsequent weather phenomena dependent on these processes (e.g., tropical cyclones). The analysis conducted found that the newest of the reanalysis datasets, the ECMWF ERA5, held the highest agreement to independent weather observations over all regions examined was reasoned to have the highest confidence in capturing the variability of upper-tropospheric velocity potential fields for the study period. Confidence was decreased in the usefulness of the NCEP/NCAR Reanalysis 1, the oldest of the reanalysis datasets, as it consistently performed poorly over much of the study domain. The results of this study also emphasized the usefulness in ensemble-based approaches to assessing climate variability and understanding potential biases and uncertainties that can be found in the data sources.

Identiferoai:union.ndltd.org:VTETD/oai:vtechworks.lib.vt.edu:10919/110353
Date26 May 2022
CreatorsStanfield, Tyler Jarrett
ContributorsGeography, Ramseyer, Craig A., Zick, Stephanie E., Ellis, Andrew
PublisherVirginia Tech
Source SetsVirginia Tech Theses and Dissertation
LanguageEnglish
Detected LanguageEnglish
TypeThesis
FormatETD, application/pdf
RightsIn Copyright, http://rightsstatements.org/vocab/InC/1.0/

Page generated in 0.0025 seconds