<p> Eversource Energy, formerly Public Service of New Hampshire (PSNH), has worked closely with Plymouth State University (PSU) in the past, and present, to better predict weather-related power outage events and maximize the efficiency with which Eversource responds to them. This research paired weather data from thirteen stations throughout New Hampshire, Vermont, and Massachusetts with Eversource Trouble Report and Unsatisfactory Performance of Equipment Report (TRUPER) data in an effort to quantify weather situations that lead to power outages. The ultimate goal involved developing a predictive model that uses weather data to forecast the magnitude of power outages. The study focused on the Eversource Western/Central service territory and utilized data from 2006-2010. The first four years, 2006-2009, were analyzed using Classification and Regression Tree (CART) statistical analysis. The results of this CART analysis trained a predictive model, while the fifth year, 2010, served as the testing set for the predictive model. </p><p> To conduct the statistical analysis, a database was created pairing TRUPER reports with the closest available hourly weather observations. The database included nine weather variables matched with three variables from the TRUPER data: 1) customers, 2) customer minutes, and 3) outage duration. While the entire Eversource service territory saw 91,286 TRUPERs from 2006-2010, the Western/Central service territory, the focus of this study, accounted for 29,430. Before conducting the CART analysis, correlations between single weather variables and TRUPER data were calculated and, in general, proved xi weak. In addition to analyzing the complete four-year training data set, many portions/variations of the data set were analyzed. The analyses included a yearly analysis, time lag analysis, cold/warm-season analysis, and a single-station analysis. Although individual years and smaller data sets showed moderately higher correlations between weather and outage data, consistent relationships throughout the data set were fairly weak. CARTs were then created to examine the joint effect of the entire set of weather variables, such as interactions and nonlinear relationships, to improve the overall predictability of power outages. After creating the trees from the four-year training data set, their predictive ability was tested using the final year of data. </p><p> The CART predictive models showed that among Eversource TRUPER variables, the hardest to predict was customers per TRUPER. The best performing model predicted customers per TRUPER to an average error of 96 customers, or a percent mean average error (PMAER) of 131% of 2010 customers per TRUPER. This result could deal with the high variability seen in customer outages per TRUPER, across a single weather event, driven by widelyvarying population and customer density. The most accurately predicted TRUPER variable, outage duration, saw average PMAER values of 60% of the mean (e.g., if mean duration per TRUPER for the year was 100 minutes, the model would miss on average by 60 minutes). Overall, the model results show surface weather data has a weak correlation to the TRUPER variables analyzed. The model can predict situations when one would expect longer duration outages but is unable to accurately predict the magnitude of these variables. When xii adapting the predictive models to smaller portions of the data set, warm-season data showed the greatest predictability, considerably outperforming the other data sets (cold-season, 2006-2009, and single station). Cold-season showed the greatest volatility and, not surprisingly, proved the most difficult to predict.</p>

<p> A myriad of processes acting singly or in concert may contribute to tropical cyclogenesis, including convectively coupled waves, breakdown of the inter-tropical convergence zone (ITCZ), or upper-level troughs. This thesis investigates the role that topographic effects from the island of Sumatra may play in initiating tropical cyclogenesis (TC genesis) in the eastern Indian Ocean. If easterly flow is split by the mountains of Sumatra, counter-rotating lee vortices may form downstream. Because Sumatra straddles the equator, though the wake vortices rotate in opposite directions, they will both be cyclonic when winds are easterly upon Sumatra, and may intensify further into tropical cyclones. The phenomenon of crossequatorial cyclone pairs, or &ldquo;twin&rdquo; tropical cyclones, in the Indian Ocean originating from Sumatra was first noted by Kuettner (1989). TC genesis appears to be particularly favored during the pre-onset phase of the Madden Julian Oscillation (MJO), when easterly flow encroaches upon Sumatra and the resulting cyclonic wake vortices encounter convectively coupled waves and enhanced moisture associated with the MJO in the Indian Ocean.</p><p> Operational analysis data from the Year of Tropical Convection (YOTC) and Dynamics of the Madden Julian Oscillation (DYNAMO) campaigns were used to evaluate the impacts of Sumatra's topography upon the flow. The YOTC data encompass two years, from May 2008 to April 2010, while the special observing period of DYNAMO was conducted from October to December 2011. This research also presents three case studies of twin tropical cyclones west of Sumatra in the Indian Ocean, which were all determined to originate from Sumatran wake vortices and occurred between October and December of 2008, 2009, and 2011.</p><p> Multiple cyclonic wake vortices and vorticity streamers were observed downstream of Sumatra during periods of easterly flow, most frequently between October and December. Froude numbers calculated for the region upstream of Sumatra with regard to easterly flow between October and December favored flow blocking and splitting, more so for Sumatra's northern tip due to the higher terrain there. Correlations between zonal wind and relative vorticity are more significant near Sumatra's northern tip than near and downstream of the island's southern tip. Cyclonic vorticity was maximized at the level of Sumatra's topography for most easterly wind days west of both the north and south ends of the island, suggesting that topography was contributing to vorticity generation.</p><p> Thirteen tropical cyclones in the Indian Ocean during the YOTC and DYNAMO campaigns were determined to develop from cyclonic wake vortices downstream of Sumatra, including three tropical cyclone pairs. Over 75% of these tropical cyclones formed between October and December. In four cases, wake vortices were generated by anomalously easterly low-level flow that preceded the active phase of the Madden Julian Oscillation. These vortices proceeded to encounter the MJO convective envelope, which is frequently accompanied by convectively coupled waves and may have altered the environment to be more moist and favorable for tropical cyclogenesis. In many cases, equatorial westerly winds, which may have been related to westerly wind bursts from the MJO or to convectively coupled equatorial Rossby waves, intensified low-level cyclonic circulations. It is suggested that diabatic heating in the vicinity of twin tropical cyclones may disturb the atmosphere enough to invigorate extant convectively coupled Kelvin waves, or contribute to the formation of a Kelvin wave. </p><p> The research presented herein describes the interaction of the flow with steep topography on Sumatra and its role in tropical cyclogenesis over the Indian Ocean, a mechanism for TC genesis that has heretofore received little attention. iv</p>

<p> An analysis was conducted to assess the relationship between the Arctic Oscillation (AO) and wintertime climatology of the Midwest and Tennessee Valley regions of the USA. In particular, this study focuses on variation in wintertime temperatures and snowfall totals during the top ten most positive and negative AO winters for the aforementioned regions. In addition, NCEP/NCAR reanalysis composite maps were created and examined to evaluate the relationship between certain atmospheric parameters and the opposing phases of the AO. In the Midwest and Tennessee Valley regions, variation in wintertime mean temperature and snowfall totals are associated with strong phases of the AO. The top ten most negative AO winters resulted in below average temperatures and above average snowfall totals. In contrast, the top ten most positive AO winters resulted in above average temperatures and below average snowfall totals. In addition, variation in mean wintertime temperature and snowfall totals is less significant during positive AO winters in comparison to negative AO winters. The top 10 most positive and negative AO winters appear to provide a significant link between anomalous middle and upper atmospheric circulation and atypical surface weather patterns across the Midwest and Tennessee Valley regions.</p>

<p>ABSTRACT Hail poses a substantial threat to life and property in the state of Florida. These losses could be minimized through better understanding of the relationships between atmospheric variables that impact hail formation in Florida. Improving hail forecasting in Florida requires analyzing a number of meteorological parameters and synoptic data related to hail formation. NOAA archive data was retrieved to create a database that was used to categorize text files of hail days. The text files were entered into the National Oceanic and Atmospheric Administration Earth System Research Laboratory website to create National Centers for Environmental Prediction/National Center for Atmospheric Research Reanalysis maps of atmospheric variables for Florida hail days as well as days leading to the hail event. These data were then analyzed to determine the relationship between variables that affect hail formation, in general, across different regions and seasons in Florida using Statistical Product and Service Solutions. The reasoning for the differing factors affecting hail formation between regions, seasons and hail sizes were discussed, as well as forecasting suggestions relating to region and month in Florida. The study found that the majority of all hail that occurs in Florida is during the wet season. A low Lifted Index, high Precipitable Water and lower than average Sea Level Pressure, in most cases, is present during hail days in Florida. Furthermore, results show that Vector Wind magnitude increases as hail size increases. Additionally, several atmospheric variables useful to studying hail events, such as Lifted Index, Precipitable Water, Sea Level Pressure, Vector Wind and Temperature have significant correlations with each other depending on the region and season being observed. Strong correlations between low Lifted Index, high Precipitable Water values and the occurrence of hail events are discussed, as well as the relationship between temperature anomalies at various pressure levels and the occurrence of hail events.

<p> Deep convection is an important component of atmospheric circulations that affects many aspects of weather and climate. Therefore, improved understanding and realistic simulations of deep convection are critical to both operational and climate forecasts. Large-eddy simulations (LESs) often are used with observations to enhance understanding of convective processes. This study develops and evaluates a nested-LES method using the Weather Research and Forecasting (WRF) model. Our goal is to evaluate the extent to which the WRF nested-LES approach is useful for studying deep convection during a real-world case. The method was applied on 2 September 2013, a day of continental convection having a robust set of ground and airborne data available for evaluation. A three domain mesoscale WRF simulation is run first. Then, the finest mesoscale output (1.35 km grid length) is used to separately drive nested-LES domains with grid lengths of 450 and 150 m. Results reveal that the nested-LES approach reasonably simulates a broad spectrum of observations, from reflectivity distributions to vertical velocity profiles, during the study period. However, reducing the grid spacing does not necessarily improve results for our case, with the 450 m simulation outperforming the 150 m version. We find that simulated updrafts in the 150 m simulation are too narrow to overcome the negative effects of entrainment, thereby generating convection that is weaker than observed. Increasing the sub-grid mixing length in the 150 m simulation leads to deeper, more realistic convection, but comes at the expense of delaying the onset of the convection. Overall, results show that both the 450 m and 150 m simulations are influenced considerably by the choice of sub-grid mixing length used in the LES turbulence closure. Finally, the simulations and observations are used to study the processes forcing strong midlevel cloud-edge downdrafts that were observed on 2 September. Results suggest that these downdrafts are forced by evaporative cooling due to mixing near cloud edge and by vertical perturbation pressure gradient forces acting to restore mass continuity around neighboring updrafts. We conclude that the WRF nested-LES approach provides an effective method for studying deep convection for our real-world case. The method can be used to provide insight into physical processes that are important to understanding observations. The WRF nested-LES approach could be adapted for other case studies in which high-resolution observations are available for validation.</p>

<p> Upper-tropospheric thermodynamic processes can play an important role in tropical cyclone (TC) structure and evolution. Despite its importance, until recently few <i>in-situ</i> observations were available in the upper levels of TCs. Two recent field campaigns&mdash;the National Aeronautics and Space Administration (NASA) Hurricane and Severe Storm Sentinel (HS3) and the Office of Naval Research Tropical Cyclone Intensity (TCI) experiment&mdash;provided a wealth of high-altitude observations within TCs. These observations revealed that the upper-level static stability and tropopause structure of TCs can change dramatically with both space and time. </p><p> The TCI dropsonde dataset collected during the rapid intensification (RI) of Hurricane Patricia (2015) revealed dramatic changes in tropopause height and temperature within the storm's inner core. These changes in tropopause structure were accompanied by a systematic decrease in tropopause-layer static stability over the eye. Outside of the eye, however, an initial decrease in static stability just above the tropopause was followed by an increase in static stability during the latter stages of RI. </p><p> Idealized simulations were conducted to examine the processes that might have been responsible for the tropopause variability observed in Hurricane Patricia. A static stability budget analysis revealed that three processes&mdash;differential advection, vertical gradients of radiative heating, and vertical gradients of turbulent mixing&mdash;can produce the observed variability. These results support the theoretical assumption that turbulent mixing plays a fundamental role in setting the upper-level potential temperature stratification in TCs. The existence of turbulence in the upper troposphere of TCs is corroborated by the presence of low-Richardson number layers in a large number of rawinsonde observations. These layers were more common in hurricanes than in weaker TCs, as hurricanes were characterized by both smaller static stability and larger vertical wind shear in the upper troposphere. </p><p> HS3 dropsondes deployed within and around TC Nadine (2012) observed two distinct upper-level stability maxima within the storm's cirrus canopy. Outside of the cirrus canopy, however, only one stability maximum was present in the upper levels. This maximum, just above the tropopause, was stronger over the cirrus canopy than outside of the cirrus canopy. Observations from a large rawinsonde dataset also show this structure, with a stronger temperature inversion located above the tropopause within regions of cold cirrus than outside of cold cirrus. It is hypothesized that vertical gradients of radiative heating, differential advection within the upper-tropospheric outflow layer, and vertical gradients of turbulence all could contribute to producing multiple stability maxima and the stronger temperature inversions in the lower stratosphere. </p><p>

<p> A mechanistic understanding of the North American Monsoon (NAM) is suggested by incorporating local- and synoptic-scale processes. The local-scale mechanism describes the effect sea surface temperature (SST) in Gulf of California (GC) and how it contributes to the low-level moisture during the 2004 NAM. Before NAM onset, the strong low-level temperature inversion exists over the GC, but this inversion weakens with increasing GC SST and generally disappears once SSTs exceed 29.5&deg;C, allowing the moist air, trapped in the MBL, to mix with free tropospheric air. This leads to a deep, moist layer that can be transported toward the NAM regions to produce thunderstorms. The synoptic scale mechanism is based on climatologies from 1983 to 2010 and explains that the warmest SSTs moving up the coast contributes to NAM convection and atmospheric heating, and consequently advancing the position of the anticyclone and the region of descent northward. </p><p> In order to improve microphysical properties of ice clouds, this study develops self-consistent second order polynomial mass- and projected area-dimension (m-D and A-D) expressions that are valid over a much larger size range, compared to traditional power laws. Such expressions can easily be reduced to power laws for the size range of interest, in order to use in cloud and climate models. This was done by combining field measurements of individual ice particle m and D with airborne optical probe measurements of D, A and estimates of m. The resulting m-D and A-D expressions are functions of temperature and cloud type (synoptic vs. anvil), and are in good agreement with m-D power laws developed from recent field studies. These expressions also appear representative for heavily rimed dendrites occurring over the Sierra Nevada Mountains. By using the m-D field measurements of rimed and unrimed particles, and by developing theoretical methods, an approach was suggested for calculating rimed m and A, which has the benefit of accounting for the degree of riming, and therefore it produces a gradual and continuous growth from unrimed ice particles to graupel. The treatment for riming includes a parameterization for collision efficiency as a function of droplet size and ice particle size using the available numerical studies. </p><p> A rimed snow growth model (RSGM) was developed based on the growth processes of vapor diffusion, aggregation, and riming. The RSGM uses a measured radar reflectivity at cloud top for initialization, and then predicts the vertical evolution of size spectra. The RSGM is based on the zeroth- and second- moment conservation equations with respect to mass, and thus conserves the number concentration and radar reflectivity, respectively. The size spectra predicted by the RSGM are in good agreement with observed spectra during Lagrangian spiral descents through frontal clouds. The snowfall rate with the inclusion of riming is significantly greater than that produced by the vapor deposition and aggregation alone. Snowfall rates are found to be sensitive to the cloud drop size distribution.</p>

<p> This works presents the results for the first study to ever attempt to analyze the full potential and limitations of incorporating aerosols within a truly physical SST retrieval for operational weather forecasting purposes. This is accomplished through the application of a satellite sea surface temperature (SST) physical retrieval for satellite split-window and hyperspectral infrared (IR) sensors that allows a better representation of the atmospheric state under aerosol-laden conditions. The new algorithm includes 1) accurate specification of the surface emissivity that characterizes the surface leaving radiance and 2) transmittance and physical characterization of the atmosphere by using the Community Radiative transfer model (CRTM). This project includes application of the NEMS-Global Forecasting System Aerosol Component (NGAC) fields, which corresponds to the first global interactive atmosphere-aerosol forecast system ever implemented at NOAA&rsquo;s National Center for Environmental Prediction (NCEP). </p><p> A number of limiting factors were identified by analysing brightness temperatures and SST outputs biases as a function of latitude, zenith angle, wind and moisture for cases in January and November 2013. SST ouputs are validated against a bulk SST (Reynolds SST) and a parameterized SST derived from operational products and partly against observed measurements from the eastern Atlantic Ocean, which is dominated by Saharan dust throughout most of the year and that is also a genesis region for Atlantic tropical cyclones. These observations are obtained from the NOAA Aerosols and Ocean Science Expeditions (AEROSE). The improved physical SST methodology has the potential to allow for improved representation of the geophysical state under dust-laden conditions. </p>

<p> The Intertropical Convergence Zone (ITCZ) is a feature that results from the ocean-atmosphere interactions in the tropics around the world. The ITCZ is characterized by surface wind convergence, tall storm clouds, and it forms a belt of high time-averaged precipitation around the globe. The ITCZ undergoes seasonal migrations between 5&deg;S and 15&deg;N roughly following the subsolar point on Earth with the seasons, with a mean annual position located slightly above the Equator, between 2&deg; and 5&deg;N. </p><p> This study tested the hypothesis that there was a northward shift in the median position of the ITCZ in the first decade of the 2000&rsquo;s relative to the 1900&rsquo;s. This hypothesis has been posed in the literature given a weakening in the intensity of the Trade Winds observed in the southern Caribbean Sea during the first decade of the 2000&rsquo;s, with concomitant ecological impacts due to weakening in coastal wind-driven upwelling. The hypothesis was tested by analyzing variations in the monthly latitudinal position of the ITCZ over the Atlantic Ocean relative to the median position computed for the period 1987&ndash;2011. The position of the ITCZ was derived from satellite-derived ocean surface wind measurements collected from 1987 to 2011. A Mann-Kendall analysis and a Monte Carlo simulation were used to test for trends in the median cross-basin latitudinal position of the ITCZ. The study included an analysis of regional changes across the tropical central Atlantic (50&deg;W to 15&deg;W), the Western Atlantic (50&deg;W to 30&deg;W), and the Eastern Atlantic (30&deg;W to 15&deg;W) within the tropics. The results show a slight southward trend in the median position of the ITCZ over the central Atlantic and also in the Eastern Atlantic in the first decade of the 2000&rsquo;s relative to the 1990&rsquo;s. While this trend is barely significant, it is likely simply due to interannual variation in the average annual position of the ITCZ. </p><p> The data were also examined for the timing and persistence of a double ITCZ in the Atlantic. The double ITCZ over the Atlantic appeared every year in February or March, with the largest separation between the northern and southern branches of the ITCZ observed in June and July. </p><p> The possible effects of changes in the average latitudinal position of the ITCZ on the upwelling in the Cariaco Basin (southeastern Caribbean Sea off Venezuela) were also examined. Anomalies of the median of the latitudinal position of the ITCZ in the Atlantic were compared with anomalies of in-situ temperature collected during the 1990&rsquo;s and the first decade of the 2000&rsquo;s by the CARIACO Ocean Time-Series program and with anomalies of satellite SST (from the Advanced Very High Resolution Radiometer satellite; AVHRR) from 1995 to 2016. Correlation analysis were performed between anomalies of water temperatures at various depths and anomalies of satellite SST with anomalies of the monthly mean ITCZ position with lags up to 3 months for the time series, and also just for the Cariaco basin upwelling months (December-April). </p><p> For the whole Cariaco time series there were no significant correlations between the anomalies of the ITCZ position and anomalies in subsurface temperatures in the Cariaco Basin. However, during the upwelling period, the central Atlantic and Western Atlantic ITCZ position anomalies were directly correlated with Cariaco Basin temperature anomalies with no-lag (r = 0.20), and the central and Eastern Atlantic ITCZ position anomalies were inversely correlated with Cariaco Basin temperatures (r ~ -0.22 to -0.28) with ITCZ leading Cariaco temperatures by 3 months. However, these correlations were low, indicating that other factors than the position of ITCZ latitudinal position play bigger role on the Cariaco basin upwelling variability. </p><p> Interannual variability in oceanographic and meteorological characteristics of the Atlantic Ocean are expected as a result of large-scale changes in other regions of the world, including due to changes such as the El Ni&ntilde;o Southern Oscillation (ENSO) and the Atlantic Multidecadal Oscillation (AMO). Six oceanic-atmospheric variables are used to monitor ENSO over the tropical Pacific, while the AMO is determined by monitoring SST over the Atlantic. Correlations with lags of up to &plusmn; 6 months were conducted with those climate indices and the anomalies of the median monthly latitudinal position of the ITCZ. Significant direct correlations with ENSO (Multivariate ENSO Index) were seen in the Atlantic and Western Atlantic (r = 0.15), with ENSO leading the position of the ITCZ anomalies by 3 months. This implies that within three months after an El Ni&ntilde;o event (warm ENSO anomaly in the Pacific) the ITCZ over the mid-Atlantic and Western Atlantic Ocean tends to shift to a more northerly position. The AMO also had a direct influence on the anomalies of the ITCZ position (r = 0.13) in the Central and the Western Atlantic, with the AMO leading ITCZ anomalies by 1 month (i.e. a warming of the North Atlantic led to a northward shift in the ITCZ one month later). Correlations between AMO and the ITCZ anomalies in the Eastern Atlantic were also direct but with no lag. Although significant, these correlations were low. </p><p> An inverse correlation (~ -0.35) was found between ENSO and anomalies of water temperature of the Cariaco Basin. ENSO lagged ocean temperature anomalies by 3 to 4 months for both the whole Cariaco time series and for the upwelling months of CARIACO data. Correlations with AMO were direct (~ 0.4); for the whole time series AMO led Cariaco temperature anomalies by 3 months, but for the upwelling months AMO lagged Cariaco temperature anomalies by one month.</p>

<p> Low coastal stratiform clouds (stratus, stratocumulus, and fog), referred to here as coastal low cloudiness (CLC), are a persistent seasonal feature of continental west coasts, including California. The importance of CLC ranges across fields, with applications ranging from solar resource forecasting, growth of endemic species, and heat wave expression and related health impacts. This dissertation improves our understanding of California&rsquo;s summertime CLC by describing its variability and influences on a range of scales from multidecadal to daily and continental to local. A novel achievement is the development of a new 19-year satellite-derived low cloud record. Trained on airport observations, this high resolution record plays a critical role in the description of CLC at finer spatial and shorter timescales. </p><p> Observations at coastal airports from Alaska to southern California reveal coherent interannual to interdecadal variation of CLC. The leading mode of CLC variability, accounting for nearly 40% of the total variance, and the majority of individual airports, exhibit decreasing low cloudiness from 1950 to 2012. The coherent patterns of CLC variability are organized by North Pacific Sea Surface Temperature (SST) anomalies, linked to the Pacific Decadal Oscillation (PDO). </p><p> The new satellite-derived low cloud retrieval reveals, in rich spatial texture, considerable variability in CLC within May-September. The average maximum cloudiness moves northward along the coast, from northern Baja, Mexico to northern California, from May to early August. Both component parts of lower tropospheric stability (LTS), SST and free-troposphere temperature, control this seasonal movement. The peak timing of cloudiness and daytime maximum temperatures are most closely aligned in northern California. </p><p> On weather timescales, daily CLC anomalies are most strongly related to stability anomalies to the north (climatologically upwind) of the CLC region. CLC is strongly linked to stability in northern (southern) California throughout (only in early) summer. Atmospheric rather than oceanic processes are responsible for the cloud dependence on stability at daily timescales. The spatial offset of the LTS-CLC relationship reveals the roles of advective processes, subsidence, and boundary layer characteristics. Free-tropospheric moisture additionally impacts CLC, implicating the North American monsoon as a factor affecting southern California&rsquo;s coastal climate in late summer.</p>