Fundamental to an understanding of El Niño/Southern Oscillation climate fluctuations is an understanding of the interannual equatorial Pacific surface flows, which advect the surface waters and change the sea surface temperature. While some knowledge of the observed interannual flows has already been obtained, some key features are still not fully understood. Using the long records of satellite altimeter data, together with long in situ records of current, salinity and temperature from the TAO/TRITON array in the equatorial Pacific, the observed interannual surface flows, their dynamics and link to the El Niño Prediction can be understood better. In the first half of the thesis, I used theoretical arguments and a wind-forced ocean model to understand why the equatorial eastern Pacific flow leads sea level, eastern equatorial thermocline displacement and El Niño indices. This half of the thesis is based on the result that for large zonal scales and low frequencies, wind-forced sea level, even near the equator, can be described by wind-forced long Rossby waves. In the eastern equatorial Pacific where the interannual wind forcing is small, these waves are essentially locally unforced and propagate westward from the boundary. At the boundary the wave’s sea level is in phase because of geostrophy and no normal flow to the boundary. However, because the waves propagate more slowly with increasing latitude, west of the boundary lag increases as latitude increases. Consequently a northward sea level gradient is like a time derivative, and the zonal geostrophic flow is like a time derivative of the sea level. This implies that the equatorial flow should lead the equatorial sea level by about 9 months on El Niño time scales. Analysis shows that when dissipation of the large-scale flow is taken into account, this lead is reduced to about 3 months. This lead time is approximately the dissipation time scale of the second vertical mode, which dominates the zonal surface flow. Since the eastern equatorial Pacific sea level is proportional to eastern equatorial thermocline displacement and El Niño, the zonal equatorial flow leads El Niño indices. Analysis further shows that the zonally-averaged equatorial Pacific sea level leads El Niño, and that this lead is associated with the geostrophic zonal velocity and the long Rossby wave physics in the eastern equatorial Pacific. The second part of this work addresses the influence of the heavy precipitation on the Western equatorial Pacific Ocean. Surface and subsurface salinity and temperature measurements at 137oE, 147oE, and 156oE since the late 1990s from the western equatorial Pacific TRITON moored array indicate that the large interannual sea surface salinity (SSS) fluctuations there change little with depth over the top 50 m of the water column. Beneath this surface layer the SSS signal decreases, and is usually much smaller at about 100 m depth. The isothermal layer depth (ILD) ranges from about 50–70 m and estimates of dynamic height relative to the ILD indicate a near-surface salinity-driven contribution to the monthly sea level anomaly that is uncorrelated with, and smaller than, interannual sea surface height (SSH) estimated from altimeter data. Despite the smaller size of , its meridional gradient dominates the total sea level meridional gradient and thus the corresponding shallow equatorially-trapped interannual fresh water jet dominates the near-surface zonal interannual flow. This jet-like flow has a meridional scale of only about 2–3o of latitude, an amplitude of 23cm/s, and is associated with the zonal back and forth displacement of the western equatorial warm/fresh pool that is fundamental to El Niño. The jet is not forced by the interannual fresh water surface flux but rather by wind stress anomalies that are mostly east of the warm/fresh pool edge during La Niña and mostly west of it during El Niño. / A Dissertation submitted to the Department of Earth, Ocean and Atmospheric Science in partial fulfillment of the requirements for the degree of Doctor of Philosophy. / Spring Semester 2017. / March 9, 2017. / El Nino Prediction, Equatorial Pacific Currents, Oceanography Dynamics, Physical Meteorology and Climatology, Rossby waves, Salinity / Includes bibliographical references. / Allan J. Clarke, Professor Directing Dissertation; Christopher Tam, University Representative; Mark A. Bourassa, Committee Member; William Dewar, Committee Member; William M. Landing, Committee Member.

Florida has a high frequency of tornadoes that occur throughout the United States. Together, Florida's large population and expensive property, provides a great risk for injuries, fatalities, and damage to structures for when a tornado occurs. This risk of death or damage continues to increase as the population expands. The goal of this research is to better understand the tornado hazard in Florida by creating a climatology of Florida tornadoes through examining occurrence rates, casualties, and property loss. The tornado reports are obtained from the Storm Prediction Center's Severe Weather database. Descriptive statistics are used to analyze temporal distributions, characteristics, and geographical distributions of tornadoes. Tropical cyclone tornado data from 1995 through 2013 is used for examining temporal distributions throughout the state. In addition, a new property value dataset put together by Georgianna Strode at the Florida Resources and Environmental Analysis Center is used to evaluate property loss from tornadoes throughout the state. Inferential statistics are used for testing hypotheses and modeling future tornado paths using a Monte Carlo simulation. Over the period from 1987 though 2016, there were 1,765 tornado reports in the state. The peak frequency occurs during the month of June with the overall tornado distribution mimicking the tropical cyclone distribution of the North Atlantic hurricane season. Majority of tornadoes occur in the peninsular region of the state, with tornadoes in the panhandle likely being stronger. There is a strong positive correlation between the amount of property exposed and the number of casualties produced by tornadoes. Although the majority of tornadoes that occur throughout Florida are very weak, the path length and width are shown to be increasing in recent years. Additionally, the annual average property loss estimate from tornadoes in Florida is $53 Million. Results of the Monte Carlo simulation indicate a 5% chance that the annual loss will exceed $203 million, a 1% chance that it will exceed $430 million, and a 0.1% chance that it will exceed $1 billion. / A Thesis submitted to the Department of Geography in partial fulfillment of the requirements for the degree of Master of Science. / Spring Semester 2018. / April 6, 2018. / Casualties, Climatology, Florida, Occurrence rates, Property Losses, Tornado / Includes bibliographical references. / James B. Elsner, Professor Directing Thesis; David C. Folch, Committee Member; Mark W. Horner, Committee Member.

Hurricane Harvey (2017) spawned from a westward propagating tropical wave in the Atlantic and then tracked across the southern Caribbean Sea, the Yucatán Peninsula, and lastly over the Gulf of Mexico, where it quickly intensified into a category 4 (on the Saffir-Simpson Scale) tropical cyclone. As a mature hurricane, Harvey underwent an eyewall replacement cycle which led to structural and intensity changes hours before making landfall over the Texas central coast. This study investigates the structure and frequency of secondary eyewalls in 20 forecast simulations of Tropical Cyclone Harvey (2017) as produced by the 2017 operational Hurricane Weather Research and Forecast (HWRF) System. To understand the predictability of secondary eyewalls, the secondary eyewall-producing simulations must be distinguished from the non-secondary eyewall-producing simulations. Thus, a diagnostic method of subjectively detecting secondary eyewalls in forecast data is developed. The diagnostic method identifies specific secondary eyewall traits that have been studied and documented in literature. The results show that most of the simulations (~80%) produce a secondary eyewall. While the all secondary eyewall-producing simulations are initialized over the ocean, the unsuccessful simulations, on the other hand, are initialized over or just west of the Yucatán Peninsula. To study the relationship between land-storm interaction and secondary eyewall simulation, a comparison is made between the successful simulations initialized over the Caribbean Sea (which tracked over the Yucatán Peninsula) and the unsuccessful runs. For both sets of simulations, the effect of land-storm interaction led to temporary storm weakening while over the Yucatán Peninsula. However, this interaction has respectively a greater negative effect on vortex spin-up and organization on those simulations initialized over land. A comparison between the over land evolution of a non-SE producing and aSE-producing simulation is made. The results show that both storms maintain a similar dynamic structure as they move west over the Yucatán Peninsula. However, the SE-producing simulation is in a more favorable thermodynamic environment with higher RH values above the storms and more convective activity near its center when compared to the non-SE producing simulation. Based on these results, it is speculated that deep moist convective feedback processes enhanced by a thermodynamically favorable conditions within and near the Caribbean Sea initialized storms act as an additional intensification mechanism which lacks in the over land initialized storms. The relatively drier air mass and less convective activity associated with the land simulations produces a less favorable environment and limits the intensification rate of these storms over once over water. It is speculated that slower intensification rates inhibit these storms from reaching an adequate TC intensity and structure conducive for SEF before making landfall over Texas/Mexico and weakening. / A Thesis submitted to the Department of Earth, Ocean, and Atmospheric Science in partial fulfillment of the requirements for the degree of Master of Science. / Summer Semester 2018. / July 5, 2018. / 2017, Harvey, Hurricane, HWRF, Secondary, Structure / Includes bibliographical references. / Jeffrey Chagnon, Professor Directing Thesis; Robert E. Hart, Committee Member; Philip Sura, Committee Member.

"This study examines atmospheric boundary layer over and around the Gulf of Mexico during return flow events. The locations investigated are on both the Texas coast and the Florida coast. Moreover, several inland stations such as Jackson (JAN), Mississippi and Shelby Co Airport (BMX), Alabama are added to the analyses to make some comparisons. In addition to examining the boundary layer structure of these coastal and inland stations, an attempt is made to investigate thermadynamic structure on the continental shelf of the Gulf of Mexico in terms of both synoptic analysis and model applications. Return flow events of the Gulf of Mexico are mainly examined by using a three dimensional Air Mass Transformation (AMT) model and a one-dimensional Planetary Boundary Layer (PBL) model. The models are intende for short-range weather forecasts of the temperature profiles in the lower atmosphere and the structure of the boundary layer"--Abstract. / Typescript. / "1996." / "Submitted to the Department of Meteorology in partial fulfillment of the requirements for the degree of Master of Science." / Advisor: Paul H. Ruscher, Professor Directing Thesis. / Includes bibliographical references.

Equatorial waves have been studied extensively due to their importance to the tropical climate and weather systems. Historically, their activity is diagnosed mainly in the wavenumber-frequency domain. Recently, many studies have projected observational data onto parabolic cylinder functions (PCFs), which represent the meridional structure of individual wave modes, to attain time-dependent spatial wave structures. The non-orthogonality of wave modes has yet posed a problem when attempting to separate data into wave fields where the waves project onto the same structure functions. We propose the development and application of a new methodology for equatorial wave expansion of instantaneous flows using the full equatorial wave spectrum. By creating a mapping from the meridional structure function amplitudes to the equatorial wave class amplitudes, we are able to diagnose instantaneous wave fields and determine their evolution. Because all meridional modes are shared by some subset of the wave classes, we require constraints on the wave class amplitudes to yield a closed system with a unique solution for all waves' spatial structures, including IG waves. A synthetic field is analyzed using this method to determine its accuracy for data of a single vertical mode. The wave class spectra diagnosed using this method successfully match the correct dispersion curves even if the incorrect depth is chosen for the spatial decomposition. In the case of more than one depth scale, waves with varying equivalent depth may be similarly identified using the dispersion curves. The primary vertical mode is the 200 m equivalent depth mode, which is that of the peak projection response. A distinct spectral power peak along the Kelvin wave dispersion curve for this value validates our choice of equivalent depth, although the possibility of depth varying with time and height is explored. The wave class spectra diagnosed assuming this depth scale mostly match their expected dispersion curves, showing that this method successfully partitions the wave spectra by calculating wave amplitudes in physical space. This is particularly striking because the time evolution, and therefore the frequency characteristics, is determined simply by a timeseries of independently-diagnosed instantaneous horizontal fields. We use the wave fields diagnosed by this method to study wave evolution in the context of the stratospheric QBO of zonal wind, confirming the continuous evolution of the selection mechanism for equatorial waves in the middle atmosphere. The amplitude cycle synchronized with the background zonal wind as predicted by QBO theory is present in the wave class fields even though the dynamics are not forced by the method itself. We have additionally identified a time-evolution of the zonal wavenumber spectrum responsible for the amplitude variability in physical space. Similar to the temporal characteristics, the vertical structures are also the result of a simple height cross-section through multiple independently-diagnosed levels. / A Dissertation submitted to the Geophysical Fluid Dynamics Institute in partial fulfillment of the requirements for the degree of Doctor of Philosophy. / Spring Semester 2016. / March 23, 2016. / QBO, Tropics, Waves / Includes bibliographical references. / Ming Cai, Professor Directing Dissertation; Xufeng Niu, University Representative; Allan Clarke, Committee Member; Kevin Speer, Committee Member; Philip Sura, Committee Member.

Ensemble forecasts are the primary tool used operationally to assess forecast uncertainty. Studies of ensemble forecasts, however, have shown that forecast verifications too frequently lie outside of the ensemble's range of possibilities, meaning that uncorrected ensemble forecasts suggest more confidence than is justified. To make ensemble forecasts more representative of the actual range of possibilities, we present a technique to post-process ensemble forecasts by replacing member forecasts with verifications of what actually occurred when past forecasts were similar. To maximize the information that can be extracted from an archive of past forecasts and verifications, we allow analogs to come from different locations in space. We evaluated our procedure to post-process NCEP ensemble precipitation forecasts for the United States for 15-day periods in July 2005 and January 2006. Our analog correction technique significantly improved the ensemble's ability to forecast the probability of precipitation, in particular correcting the NCEP Global Ensemble's ``wet' bias at low precipitation amounts. Brier Skill Scores for 6-hour accumulated precipitation during the winter indicated that uncorrected ensemble forecasts were less skillful at predicting the probability of precipitation than forecasting zero precipitation as indicated by negative Brier Skill Scores (roughly -2.5). Post processed forecasts had Brier Skill Scores as high as 0.34. The tendency of the ensemble to underforecast heavy precipitation events, however, was not well corrected by our post-processing technique. Examinations of analog locations during heavy precipitation events indicated that analogs were taken from regions where precipitation patterns differed from those at the forecast point. This indicates that analogs must be chosen using more information than merely the similarity of ensemble precipitation forecasts to past forecasts. / A Thesis submitted to the Department of Meteorology in partial fulfillment of the requirements for the degree of Master of Science. / Degree Awarded: Spring Semester, 2007. / Date of Defense: March 29, 2007. / Forecasting, Post-Processing, Ensemble Forecasts, Ensemble, Analogs / Includes bibliographical references. / Jon E. Ahlquist, Professor Directing Thesis; T. N. Krishnamurti, Committee Member; Xiaolei Zou, Committee Member.

Accurate hurricane intensity prediction is at the forefront of atmospheric science today, and improvements to mesoscale modeling of these storms continue to be major components of refining the accuracy of intensity forecasting. The primary goal of this study is to improve mesoscale modeling of hurricane intensity via the comparison of field campaign observations of Hurricane Erin 2001 from the Fourth Convection And Moisture Experiment (CAMEX-4) and Hurricane Dennis 2005 from the Tropical Cloud Systems and Processes (TCSP) mission with simulated results of improved microphysical parameterization in a mesoscale model that utilizes the Krishnamurti, et al (1991) technique of rain rate initialization (RRI). Comparison of the simulated results with field observations collocated with satellite observations provides a way to validate many different aspects of the simulated hurricane's structure and intensity. The mesoscale model used in this research is the Weather Research & Forecasting (WRF) model version 2.1 (ARW). Much of the existing microphysical parameterization of this model is built from results of mid-latitude observations. Substantial improvement to the model's intensity forecasting in the tropics can be made via proper parameterization of the model microphysics for hurricanes. With a foundation of results from other hurricane mesoscale modeling initial/boundary conditions, dynamics and physics studies, basic options for modeling hurricanes Erin (2001) and Dennis (2005) are chosen and held constant during a series of microphysical sensitivity experiments for each storm. These are specifically designed to isolate the individual effects of altering one microphysical parameter at a time on the hurricane's intensity forecast and are carried out in a doubly or triply nested way. The initial and boundary conditions used in the innermost grid with finer resolution are obtained from the respective outermost grids where rain rate initialization is invoked. All of the results are illustrated for the highest-resolution innermost domain, which is integrated using an explicit microphysics scheme. Each of these experiments are integrated for a forty-eight hour forecast period, adequately capturing the mature and intensification stages of the two hurricanes. Skill scores are obtained from the results of the two sets of experiments. Root Mean Square Errors (RMSE) and Anomaly Correlations (AC) are computed by comparing the model output of each experiment to NCEP's final analysis (fnl) available at one-degree horizontal resolution and six-hour temporal resolution interpolated to the respective model grid. Taking into account the way that each experiment performs in terms of simulated storm intensity as well as optimized RMSE and AC, the optimal combination of microphysical processes (i.e. melting, evaporation, fall speed of hydrometeors) for each storm is determined. Then a final forty-eight hour forecast of each hurricane is made utilizing this optimal microphysical parameterization combination. The results from each final run are compared to observations, skill scores are computed, and the final intensity improvements for both hurricanes Erin and Dennis are shown. The results of this study strengthen the evidence that RRI and proper microphysical parameterization in mesoscale hurricane modeling are both useful and effective techniques, and combine to improve hurricane intensity forecasting in a mesoscale model. / A Thesis submitted to the Department of Meteorology in partial fulfillment of the requirements for the degree of Master of Science. / Degree Awarded: Fall Semester, 2007. / Date of Defense: October 26, 2007. / Modeling, Mesoscale, Forecast, Microphysics, Parameter / Includes bibliographical references. / Tiruvalam N. Krishnamurti, Professor Directing Thesis; Guosheng Liu, Committee Member; Paul Ruscher, Committee Member; Robbie Hood, Committee Member.

"Nineteen years of wind data over the North Atlantic are used to calculate a field of wind stress curl. An EOF analysis is performed on this field resulting in variance-qualified spatial patterns of wind stress curl and associated time series. A Monte Carlo technique is used to establish the statistical significance of each spatial pattern. The first four statistically significant EOF modes represent more than 50% of the curl variance. The spatial patterns of curl associated with these modes exhibit many elements of North Atlantic climatology. The associated time series are spectrally analyzed. Most of the variance is contained in annual and semiannual frequencies. Features observed include the individual annual variation of the subtropical high and the subpolar low, the annual oscillation of intensity between the above pressure centers, the influence of localized strong SST gradients and associated cyclogenesis regions, and the constant nature of the trades. The EOF curl patterns are in the form of simple standing waves with wavelength on the order of basin size"--Abstract. / Typescript. / "April, 1986." / "Submitted to the Department of Meteorology in partial fulfillment of the requirements for the degree of Master of Science." / Advisor: James O'Brien, Professor Directing Thesis. / Includes bibliographical references (leaves 35-36).

No description available.

Lau, Chi Shing = 一個應用於航空的高分辨率、快速更新的氣象數據分析系統 / 柳巳丞. / Thesis (M.Phil.)--Chinese University of Hong Kong, 2008. / Includes bibliographical references (leaves 76-78). / Abstracts in English and Chinese. / Lau, Chi Shing = Yi ge ying yong yu hang kong de gao fen bian lu, kuai su geng xin de qi xiang shu ju fen xi xi tong / Liu Sicheng. / Abstract --- p.i / Acknowledgement --- p.iii / Chapter 1 --- Introduction --- p.1 / Chapter 1.1 --- Overview --- p.1 / Chapter 1.2 --- Review on Windshear --- p.2 / Chapter 2 --- Review of the Weather Radar System --- p.5 / Chapter 2.1 --- Introduction --- p.5 / Chapter 2.2 --- Reflectivity Measurement --- p.8 / Chapter 2.3 --- Velocity Measurement --- p.11 / Chapter 2.4 --- The Doppler Dilemma --- p.14 / Chapter 2.5 --- TDWR and LIDAR used in Hong Kong --- p.16 / Chapter 3 --- Design of the System --- p.19 / Chapter 3.1 --- The Wind Analysis --- p.19 / Chapter 3.2 --- The Cloud Analysis --- p.25 / Chapter 3.3 --- Settings of the Domain --- p.26 / Chapter 4 --- Data Preparation --- p.31 / Chapter 4.1 --- Background Field --- p.31 / Chapter 4.2 --- Non-radar Observation Data --- p.33 / Chapter 4.3 --- The Radar Data --- p.33 / Chapter 5 --- A Study on Sea Breeze --- p.37 / Chapter 5.1 --- The Physical origin of Sea Breeze --- p.37 / Chapter 5.2 --- Case Study on 10 March 2006 --- p.41 / Chapter 6 --- A Study on Tropical Cyclone --- p.46 / Chapter 6.1 --- The Physics of Tropical Cyclone --- p.46 / Chapter 6.2 --- Case Study on 3 Aug 2006 --- p.51 / Chapter 7 --- A Study on Microburst --- p.57 / Chapter 7.1 --- The Physical origin of Microburst --- p.57 / Chapter 7.2 --- Case Study on 8 June 2007 --- p.60 / Chapter 8 --- Discussions and Conclusions --- p.67 / Chapter 8.1 --- Discussions --- p.67 / Chapter 8.2 --- Conclusions --- p.69 / Chapter A --- Derivation of Radar Equation --- p.70 / Chapter A.1 --- Radar Equation for Point Target --- p.70 / Chapter A.2 --- Radar Equation for Distributed Targets --- p.71 / Chapter B --- Technical Details --- p.73 / Chapter B.1 --- Hardware and Timing --- p.73 / Chapter B.2 --- Programming issues --- p.75 / Bibliography --- p.76