This study uses the high-resolution infrared radiation AVHRR (Advanced Very High Resolution Radiometer)-only and microwave radiation AMSR (Advanced Microwave Scanning Radiometer)+AVHRR sea surface temperature (SST) datasets to analyze and compare non-Gaussian statistics and extreme events for SSTs. Since the primary difference between the two datasets is the lack of AVHRR data in regions of cloud cover, higher correlations between the datasets are expected in areas of low percent daily-averaged total cloud cover. These are regions where both sensors usually are capable of detecting SSTs and do not rely on the process of optimum interpolation to fill missing data. Probability density functions, skewness, kurtosis, autocorrelation time scale, and standard errors are used to reveal non-Gaussianity (i.e., statistically extreme events) in the datasets, while the correlation coefficient between the datasets is used to explore extreme events beyond a certain threshold. Non-Gaussianity is present in both SST datasets, and the highest correlations of extreme events between the datasets were within positive anomalies above a certain threshold for regions of low percent daily-averaged total cloud cover. / 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. / Fall Semester, 2010. / August 3, 2010. / extreme events, non-Gaussian, skewness, kurtosis, SST, AVHRR, AMSR / Includes bibliographical references. / Philip Sura, Professor Directing Thesis; Carol Anne Clayson, Committee Member; Sharon Nicholson, Committee Member.

We will describe the development of a high-resolution, gridded forecast guidance product for warm season cloud-to-ground (CG) lightning in Florida. Four warm seasons of analysis data from the 20-km Rapid Update Cycle (RUC) and lightning data from the National Lightning Detection Network are used to examine relationships between observed atmospheric parameters and the spatial and temporal patterns of CG lightning over Florida. The most important RUC-derived parameters then are used in a perfect prognosis (PP) technique to develop equations producing 3-hourly spatial probability forecasts for one or more CG flashes, as well as the probability of exceeding various flash count percentile thresholds. Binary logistic regression is used to develop the equations for one or more flashes, while a negative binomial (NB) model is used to predict the amount of lightning, conditional on one or more flashes occurring. When applied to the dependent sample of RUC analyses, the equations show forecast skill over a model containing only persistence and climatology (L-CLIPER). We also evaluate the lightning forecast scheme when applied to output from three mesoscale models during an independent test period (the 2006 warm season). The evaluation is performed using output from NCEP's 13-km RUC, the NCEP 12-km NAM-WRF, and local runs of WRF for a domain over South Florida that were initialized with NCEP 1/12th degree sea-surface temperatures (SST) and data from the Local Analysis and Prediction System (LAPS) (WRF-LAPS). During the most active lightning period (1800-2059 UTC), the three models forecast between 80-90% of the lightning events having one or more flashes, and between 30-60% of the events with flash counts meeting or exceeding the 95th percentile. Of the three mesoscale models, WRF-LAPS generally produces the best verification scores during 1800-2059 UTC. Forecasts from all three mesoscale models generally show positive skill with respect to L-CLIPER and persistence through the 2100-2359 UTC period, demonstrating that the PP scheme is model independent. Although the exact timing and placement of forecast lightning is not perfect, there generally is good agreement between the forecasts and their verification, with most of the observed lightning occurring within the higher forecast probability contours. / A Dissertation Submitted to the Department of Meteorology in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy. / Spring Semester, 2007. / March 22, 2007. / Weather Research, Rapid Update Cycle, Negative Binomial, Logistic Regression, Perfect Prognosis, Lightning Forecasting, Forecasting Model / Includes bibliographical references. / Henry E. Fuelberg, Professor Directing Dissertation; James B. Elsner, Outside Committee Member; Paul H. Ruscher, Committee Member; Jon E. Ahlquist, Committee Member; Robert Hart, Committee Member.

Anthropogenic and biomass burning emissions impact atmospheric chemistry and many other natural processes that affect air quality and human health. Biomass burning emissions released in the boundary layer can be quickly lofted to the free troposphere by deep convection. Accurately simulating this process in chemical transport models (CTMs) will improve our understanding of the link between local pollution sources and global scale transport. This study investigated the convective transport of biomass burning emissions during the summer phase of NASA's Arctic Research of the Composition of the Troposphere from Aircraft and Satellites (ARCTAS) campaign using the Weather Research and Forecasting (WRF) model with chemistry (WRF-Chem). Three cumulus parameterization schemes were tested to identify which performs best: Kain-Fritsch (KF), Betts-Miller-Janjic (BMJ), and Grell-Devenyi (GD). To test the cumulus parameterizations, simulated meteorological parameters were quantitatively compared against point observations, and daily precipitation fields were compared against the Global Precipitation Climatology Project (GPCP) dataset using the Method for Object-Based Diagnostic Evaluation (MODE), an object-based verification tool. CO vertical mass fluxes were evaluated at various altitudes and times during the simulation period. Daily averaged total column CO and mixing ratios at three altitudes were quantitatively compared against daily averaged values from the Atmospheric InfraRed Sounder (AIRS) using MODE. Results show that the choice of cumulus parameterization is critical when simulating the convective transport of biomass burning emissions using WRF-Chem. Although spatial differences are not great at most individual times, they accumulate over time leading to large magnitude differences in precipitation, upward CO mass flux, and long-range CO plume transport. The KF cumulus parameterization scheme vertically transports more CO than the BMJ and GD schemes, and outperforms the other schemes when compared to GPCP and AIRS dataset. In situations similar to this study, not using KF cumulus parameterization may underestimate the convective transport of CO and subsequently its long-range transport. The current results demonstrate that the choice of a cumulus parameterization scheme in a CTM can affect many aspects of its output. / 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. / Spring Semester, 2011. / March 10, 2011. / Numerical weather prediction, Cumulus parameterization, Pollution transport, Forecast verification / Includes bibliographical references. / Henry Fuelberg, Professor Directing Thesis; Guosheng Liu, Committee Member; Vasubandhu Misra, Committee Member.

Fourteen years of cloud-to-ground lightning data from the National Lightning Detection Network, and radiosonde releases from Miami and West Palm Beach, are used to develop statistical guidance equations that forecast the amount of warm season afternoon and evening lightning that is expected over two areas of South Florida that are serviced by Florida Power and Light Corporation (FP&L)-- the eastern halves of Miami-Dade and Broward Counties. A total of 54 parameters are calculated from the soundings to serve as candidate predictors for the equations. These include parameters that describe wind direction and speed in various layers, moisture, temperature, and stability. Day number, persistence, and same day morning lightning also are included as potential predictors of afternoon lightning. A variety of statistical modeling techniques is attempted initially, but many are found to be inappropriate. The best results are obtained by creating four quartile groups of flash count based on climatology, and then using binary logistic regression to develop three prediction equations for each domain, one giving the conditional probability of a quartile one (Q1) lightning event, another for the probability of a quartile three (Q3) or greater event, and a third equation giving the probability of a quartile four (Q4) lightning event. Principal component analysis is used to select a subset of non-redundant predictors that have the greatest physical relevance to convection and lightning in South Florida. The final candidate sounding predictors are the vector mean 1000-700 hPa cross-shore wind component and speed, the K-index, modified Lifted Index, and the temperature at 900 hPa. Non-linear effects are considered by including second, third, and fourth order terms as additional candidate predictors. A combination of stepwise screening and cross-validation is used to select the variables that best generalize to independent data. To determine the most likely quartile of lightning activity, a decision tree scheme is constructed using probability thresholds for the three equations. Finally, the resulting prediction schemes are tested independently using k-fold cross-validation. The dominant effect in each of the equations is the component of the wind perpendicular to the coastline which is found to have a significant non-linear relationship with lightning activity. Other important variables are the K-index and modified Lifted Index. Day number, persistence, and same day morning activity also are selected as important indicators of afternoon lightning in the two domains. When each year is treated independently, the Miami-Dade scheme correctly forecasts the quartile ~ 37% of the time and is correct to within one quartile of the observed ~ 79% of the time. The scheme for eastern Broward County forecasts the correct quartile ~ 36% of the time and is correct to within one quartile ~ 77% of the time. The prediction schemes generally are superior to persistence and climatology for both the dependent data and during k-fold cross-validation. Thus, they possess real forecast skill. For example, when forecasting the correct quartile, these results are a ~ 4-6 percentage point improvement over persistence, and ~ 11-12 percentage point improvement over climatology. In terms of correctly predicting to within one quartile of the observed, the two schemes are an improvement over persistence by ~ 6-8 percentage points and over climatology by ~ 14-17 percentage points. Further analysis shows that the two schemes rarely forecast the upper two quartiles when no activity is observed. Additionally, correct predictions of Q4 events are shown to increase with flash count within the Q4 category. Overall, the cross-validation results show only a 1-2% reduction in skill from what is obtained for the fourteen years of dependent data, demonstrating that the two schemes are statistically robust, and can be expected to achieve similar results when implemented operationally. / A Thesis Submitted to the Department of Meteorology in Partial Fulfillment of the Requirements for the Degree of Master of Science. / Summer Semester, 2004. / June 22, 2004. / South Florida, lightning climatology, statistical forecasting, logistic regression / Includes bibliographical references. / Henry E. Fuelberg, Professor Directing Thesis; Jon Ahlquist, Committee Member; Paul Ruscher, Committee Member; Andrew I. Watson, Committee Member.

The Weather Research and Forecasting Model (WRF) was developed by the National Center for Atmospheric Research as the next generation mesoscale meteorology model. The inclusion of a chemistry module (WRF-Chem) allows transport simulations of chemical and aerosol species such as those observed during NASA's Arctic Research of the Composition of the Troposphere from Aircraft and Satellites (ARCTAS) during 2008. The ARCTAS summer deployment phase during June and July coincided with large boreal wildfires in Saskatchewan and Eastern Russia. One of the most important aspects of simulating wildfire plume transport is the height at which emissions are injected. WRF-Chem contains an integrated one-dimensional plume rise model to determine the appropriate injection layer. The plume rise model accounts for thermal buoyancy associated with fires and the local meteorological stability. This study compares results from the plume model against those of more traditional injection methods such as filling the planetary boundary layer or a layer 3-5 km above ground level (AGL). Fire locations are satellite-derived from the GOES Wildfire Automated Biomass Burning Algorithm (WF_ABBA) and the MODIS thermal hotspot detection. Two preprocessing methods for these fires are compared: the prep_chem_sources method included with WRF-Chem, and the Naval Research Laboratory's Fire Locating and Monitoring of Burning Emissions (FLAMBE). Satellite products from the AIRS, MISR and CALIOP sensors provide data for verifying the simulations. Observed near-source plume heights from MISR's stereo-height product are compared with the plume rise model's simulated injection heights. Long range plume transport is evaluated qualitatively in the horizontal using AIRS's total column carbon monoxide product. Qualitative vertical evaluation uses CALIOP's high vertical resolution and aerosol identification algorithm. Horizontal plume structures are further tested quantitatively using an object-based methodology. The plume rise model produces the best agreement with satellite-observed injection heights. Filling the planetary boundary layer or the 3-5 km AGL layer with emissions exhibit less agreement with the observational plume heights. Results indicate that WRF-Chem can accurately transport chemical plumes throughout the ten-day simulation. However, differences in injection heights produce different transport pathways. Small differences in injection height are ameliorated when synoptic scale features such as warm conveyor belts quickly loft the emissions to higher altitudes. In scenarios where large scale lofting is delayed, the plume rise simulations creates the most accurate simulated plumes. / A Thesis Submitted to the Department of Meteorology in Partial Fulfillment of the Requirements for the Degree of Master of Science. / Summer Semester, 2010. / April 15, 2010. / Numerical Weather Prediction, ARCTAS / Includes bibliographical references. / Henry Fuelberg, Professor Directing Thesis; Guosheng Liu, Committee Member; Robert Hart, Committee Member.

The role of mesovortices in the eyewalls of sheared unstable, high-Rossby number vortices is investigated. A high-resolution numerical model is used to simulate dry vortices in an attempt to unite ideas from previous works. The simulations are used to investigate the dynamical, adiabatic interactions between potential vorticity (PV) mixing dynamics and shear forcings of barotropically unstable, high-Rossby number barotropic vortices. Previous work has investigated barotropic vortices in shear, while other previous work has studied barotropically unstable ring vortices. This work will combine those two avenues of research by shearing barotropically unstable barotropic ring vortices because ring vortices are more representative of tropical cyclones. Quantitative and qualitative analysis of the tilt and of the internal dynamics are presented. Using such as metrics as PV power spectra, PV palinstrophy, and a linear energy equation that incorporates the effects of the shear forcing, it is found that impact of the shear forcing on the initial breakdown of the ring is merely slight; however, the breakdown of the ring of high PV into smaller mesovortices – and the subsequent rearrangement of PV into a monopolar structure – is quite significant when considering the tilt evolution. As the vortex mixes, the storm weakens. This acts as a detriment to the ability of the vortex to keep itself upright and resistant to the shear forcing, as the penetration depth of each layer of the vortex decreases to below the scale height after mixing. In terms of the energetics, it is found that the barotropic energy conversion term is consistently the largest, which is expected. When sheared, the shear forcing acts to generally counteract the effects of mixing and reduce eddy kinetic energy. Additionally, it is found that the shear forcing induces a trochoidal oscillation at levels of highest background flow. The sensitivity of the results is investigated by comparing and contrasting two different centroid metrics - a pressure-ring centroid and a PV-cubed centroid. PV centroid metrics historically have been used to investigate inner-core tilt while geopotential centroid metrics have been to used to investigate larger-scale tilt. For the first time, these two approaches are being compared and contrasted. It is found that during a dynamical mixing event, the PV centroid is not very resistant to the rapidly changing inner-core PV field, and that this has nontrivial effects on the calculations of center-sensitive fields such as Fourier decompositions in the azimuth and determining radial and tangential wind structures. When using a pressure-ring centroid centeredon a pressure contour that resides far enough outside the core yet radially inward enough not to be impacted by the environment, it is found that this method is much more resistant to inner core processes. / A Dissertation Submitted to the Department of Earth, Oceanic, and Atmospheric Science in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy. / Spring Semester, 2011. / March 22, 2011. / dynamics, hurricane, adiabatic, vortex in shear / Includes bibliographical references. / Robert Hart, Professor Directing Dissertation; T. N. Krishnamurti, Professor Co-Directing Dissertation; William Dewar, University Representative; Vasu Misra, Committee Member; Robert Ellingson, Committee Member.

Warm seclusions are large midlatitude storms that have the potential to substantially influence the turbulent heat fluxes and global energy budget. These storms have not been previously investigated from an energy and flux perspective. They have large areas of strong surface winds and rapidly moving cold fronts, which are associated with large air-sea differences of temperature and humidity. These regions contain large air-sea fluxes of latent and sensible heat. Therefore, errors in model representation of warm seclusions may introduce significant bias and uncertainty to the energy budget. The turbulent heat fluxes associated with three specific warm seclusions in different ocean basins are examined through an intercomparison of satellite- derived flux data and numerically derived flux data. The satellite data includes the SeaFlux version 0.75 data derived from SSM/I (Special Sensor Microwave/Imager), and model-derived reanalysis data includes CFSR, ERA-Interim, MERRA, and NCEP-R2 reanalysis data sets. Latent and sensible heat fluxes are computed in a physically consistent manner though the use of a bulk flux parameterization A single warm seclusion, which typically lasts between three and seven days, is responsible for approximately one quarter of the total time-integrated monthly fluxes for the ocean basin containing the warm seclusion, depending on the storm and data set under consideration. The large area of extremely large fluxes is associated with the mature phase of the cyclone. Proper representation of these fluxes is critical to determining accurate monthly- averaged, basin-wide fluxes. / 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, 2011. / March 29, 2011. / Satellite, Reanalysis, Air Sea Interaction, Turbulent Heat Fluxes, Intercomparison, Warm Seclusion / Includes bibliographical references. / Mark A. Bourassa, Professor Directing Thesis; Carol Anne Clayson, Professor Co-Directing Thesis; Philip Sura, Committee Member.

Lightning data provide a valuable tool for examining interactions between multi-scale weather phenomena. Weather events are determined by complex atmospheric interactions at various spatial and temporal scales. Long-term climatologies facilitate discussion of average meteorological conditions and can help isolate the relative influence of multi-scale systems (e.g., synoptic scale, mesoscale, etc.) on local weather patterns. Lightning datasets allow the development of large-scale, long-term climatologies. These lightning climatologies then are compared with additional atmospheric observations (e.g., numerical models and radar) to examine the regional, seasonal, and storm-scale variability of thunderstorm characteristics. The National Lightning Detection Network (NLDN) underwent a major upgrade during 2002–2003 that increased its sensitivity and improved its performance. Therefore, this study applies the same methodology to pre- and post-upgrade NLDN datasets to allow direct quantitative comparisons between them and thereby examine the influence of the recent upgrade on regional distributions of cloud-to-ground (CG) lightning characteristics. Although seasonal variability must be understood to better define apparent relationships between storm properties and lightning production, seasonal differences are best described on the regional scale. Therefore, this study also examines Florida's seasonal, regional, and storm-scale CG variability during 2004–09. Since lightning data are recorded instantaneously and typically reported every minute, they also provide valuable information on storm-scale development and evolution. Automated procedures are developed to create grids of lightning and radar parameters, cluster individual storm features, and data mine the lightning and radar attributes of many individual storms. These procedures facilitate detailed analysis of relationships between lightning and radar-derived parameters in many individual storms in the Mid-Atlantic Region during 2007–09. A major goal of this research is to combine information about the near-storm environment, radar-defined storm structure, and both intra-cloud (IC) and CG lightning characteristics to better quantify relationships between storm structure, lightning production, and storm severity. / A Dissertation Submitted to the College of Arts and Sciences in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy. / Spring Semester, 2011. / November 09, 2010. / Lightning, Severe Storms, Radar, NLDN, WDSS-II, GIS / Includes bibliographical references. / Henry E. Fuelberg, Professor Directing Dissertation; J. Anthony Stallins, University Representative; Carol Ann Clayson, Committee Member; Robert E. Hart, Committee Member; Guosheng Liu, Committee Member.

The need for a high-resolution precipitation database over Florida is evident by the ever-increasing impact of water resources on society. The Multisensor Precipitation Estimator (MPE) software was developed by the National Weather Service to combine hourly rain gauge- and radar-derived precipitation estimates optimally. The MPE procedure creates hourly rainfall estimates on a 4 x 4 km&178; grid. This research uses the final product from the MPE procedure (MMOSAIC) to create a high-resolution rainfall climatology over the Florida peninsula from 1996 through 2001. An objective rain gauge quality control (QC) procedure is developed to minimize the amount of erroneous gauge data that is input to the MPE procedure. This objective scheme compares the gauge data with corresponding raw radar data. A rigorous examination of rain gauge data and the encompassing radar-derived rainfall estimates revealed four main scenarios that are included in the QC procedure. The scenario responsible for removing the most data occurs when a radar-derived estimate reports heavy rainfall (> 1 in./h), while the gauge within that radar grid cell reports a value near zero. The QC procedure removes most of these suspect gauge data, thereby limiting their corrupting effect on the bias calculations within the MPE procedure. The other three QC scenarios removed a smaller number of gauges that would have adversely affected the calculations. 1 in./h), while the gauge within that radar grid cell reports a value near zero. The QC procedure removes most of these suspect gauge data, thereby limiting their corrupting effect on the bias calculations within the MPE procedure. The other three QC scenarios removed a smaller number of gauges that would have adversely affected the calculations. An analysis of the different rainfall products produced by MPE demonstrates the positive impact of the extensive quality control efforts. An independent set of rain gauges is used to evaluate the MPE products statistically for selected periods during 1999 and 2001. Results show that the final MPE product (MMOSAIC) outperforms both the radar and gauge data alone. Gauges alone generally cannot accurately represent the spatial details of warm season convective type precipitation events. Conversely, radar data depict convective precipitation events quite well; however, radars do a poor job of detecting cold season stratiform precipitation events. Analyses reveal that the MMOSAIC product utilizes the strengths of the gauges and the radars in an optimum way. Seasonal case studies comparing the independent set of gauge observations to the final MPE product show that there is good agreement between the two hourly sources (i.e., biases &61; -0.004 in., r &61; 0.78, and RMSD &61; 0.12 in.). Agreements are found to improve over daily and monthly accumulation periods. Results of this study describe the problems and uncertainties associated with quantitatively measuring Florida rainfall through a multisensor analysis / A Thesis submitted to the Department of Meteorology in partial fulfillment of the requirements for the degree of Master of Science. / Summer Semester, 2004. / June 21, 2004. / Rainfall Variability, MPE Verification / Includes bibliographical references. / Henry E. Fuelberg, Professor Directing Thesis; Paul H. Ruscher, Committee Member; Jon E. Ahlquist, Committee Member.

Ocean vector winds measured by the SeaWinds scatterometer onboard the QuikSCAT satellite can be validated with in situ data. Ideally the comparison in situ data would be collocated in both time and space to the satellite overpass; however, this is rarely the case because of the time sampling interval of the in situ data and the sparseness of data. To compensate for the lack of ideal collocations, in situ data that are within a certain time and space range of the satellite overpass are used for comparisons. To determine the total amount of random observational error, additional uncertainty from the temporal and spatial difference must be considered along with the uncertainty associated with the data sets. The purpose of this study is to quantify the amount of error associated with the two data sets, as well as the amount of error associated with the temporal and/or spatial difference between two observations. The variance associated with a temporal difference between two observations is initially examined in an idealized case that includes only Shipboard Automated Meteorological and Oceanographic System (SAMOS) one-minute data. Temporal differences can be translated into spatial differences by using Taylor's hypothesis. The results show that as the time difference increases, the amount of variance increases. Higher wind speeds are also associated with a larger amount of variance. Collocated SeaWinds and SAMOS observations are used to determine the total variance associated with a temporal (equivalent) difference from 0 to 60 minutes. If the combined temporal and spatial difference is less than 25 minutes (equivalent), the variance associated with the temporal and spatial difference is offset by the observational errors, which are approximately 1.0 m2s-2 for wind speeds between 4 and 7 ms-1 and approximately 1.5 m2s-2 for wind speeds between 7 and 12 ms-1. If the combined temporal and spatial difference is greater than 25 minutes (equivalent), then the variance associated with the temporal and spatial difference is no longer offset by the variance associated with observational error in the data sets; therefore, the total variance gradually increases as the time difference increases. / 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. / Fall Semester, 2010. / October 25, 2010. / QuikSCAT, Winds, SAMOS, Error variance, Collocation / Includes bibliographical references. / Mark A. Bourassa, Professor Directing Thesis; Vasubandhu Misra, Committee Member; Zhaohua Wu, Committee Member; Shawn R. Smith, Committee Member.