Meteorology Commons^™

Spatiotemporal Prediction Of Atmospheric Events Through Recurrent Deep Learning Model, Brian W. F. Popick

Theses and Dissertations

The main contributions of this research is to add to the growing library of literature on the use of deep learning algorithms for the spatiotemporal prediction of dangerous atmospheric and hydrologic phenomena. Specifically, we develop novel attention-based and non-attention-based recurrent neural network frameworks to produce short-range sequential forecasts for lightning and tornado occurrences. Additionally, we introduce methods that account for and include error in the model tuning process to generate more reliable models. Furthermore, we have created a lightweight spatiotemporal tornadic prediction dataset that we plan to make publicly available. The first component of this research develops three novel spatiotemporal …

Evaluating Weather Effects On Sortie Generation Using Discrete Event Simulation, Markus Case

Theses and Dissertations

United States Air Force (USAF) operations rely on sortie generation, a complex system involving aircraft maintenance, operational planning, munitions, security forces, and aircrew. Failures in any of these areas can jeopardize a mission, and extreme weather events such as lightning, high winds, and snow further complicate operations. This thesis examines the impact of extreme weather on sortie generation, focusing on developing a data-driven discrete-event simulation (DES) to predict generation timelines and identify high-risk areas. The model allows users to adjust key inputs, including the month, number of aircraft, processing times, and personnel/equipment availability. By simulating real-world conditions, the model helps …

Predictability Limit Of The 2021 Pacific Northwest Heatwave From Deep‐Learning Sensitivity Analysis, P. Trent Vonich, Gregory J. Hakim

Faculty Publications

The traditional method for estimating weather forecast sensitivity to initial conditions uses adjoint models, which are limited to short lead times due to linearization around a control forecast. The advent of deep‐learning frameworks enables a new approach using backpropagation and gradient descent to iteratively optimize initial conditions, minimizing forecast errors. We apply this approach to the June 2021 Pacific Northwest heatwave using the GraphCast model, yielding over 90% reduction in 10‐day forecast errors over the Pacific Northwest. Similar improvements are found for Pangu‐Weather model forecasts initialized with the GraphCast‐derived optimal, suggesting that model error is an unimportant part of the …

Center Fixing Tropical Depressions And Tropical Storms Using Machine Learning-Nighttime Visible Imagery, Nathan K. Stanford

Theses and Dissertations

The first step in most TC-retrieval algorithms is determining the storm’s central position. In mature TCs, the center is highlighted by a distinct eye and curved band pattern; however, in intensifying and decaying storms, the center is often obscured by thick clouds or overlying cirrus. This study assesses the benefits of incorporating machine learning-derived nighttime visual imagery to improve analysis of center fix in intensifying and decaying TD- and TS-strength TCs during periods of darkness and when polar orbiting satellites are unavailable. The study is divided into two parts: the first, an objective analysis using the ARCHER-2 algorithm, and the …

Seasonal Variability And Predictability Of Monsoon Precipitation In Southern Africa, Matthew F. Horan, Fred Kucharski, Moetasim Ashfaq

Faculty Publications

Rainfed agriculture is the mainstay of economies across Southern Africa (SA), where most precipitation is received during the austral summer monsoon. This study aims to further our understanding of monsoon precipitation predictability over SA. We use three natural climate forcings, El Niño–Southern Oscillation, Indian Ocean Dipole (IOD), and the Indian Ocean Precipitation Dipole (IOPD)—the dominant precipitation variability mode—to construct an empirical model that exhibits significant skill over SA during monsoon in explaining precipitation variability and in forecasting it with a five-month lead. While most explained precipitation variance (50%–75%) comes from contemporaneous IOD and IOPD, preconditioning all three forcings is key …

Lightning Forecast From Chaotic And Incomplete Time Series Using Wavelet De-Noising And Spatiotemporal Kriging, Jared K. Nystrom, Raymond Hill, Andrew J. Geyer, Joseph J. Pignatiello Jr., Eric Chicken

Faculty Publications

Purpose: Present a method to impute missing data from a chaotic time series, in this case lightning prediction data, and then use that completed dataset to create lightning prediction forecasts.

Design/Methodology/Approach: Using the technique of spatiotemporal kriging to estimate data that is autocorrelated but in space and time. Using the estimated data in an imputation methodology completes a dataset used in lighting prediction.

Findings: The techniques provided prove robust to the chaotic nature of the data, and the resulting time series displays evidence of smoothing while also preserving the signal of interest for lightning prediction.

Abstract © Emerald Publishing …

Fragility Of The Florida Panhandle's Electrical Transmission Grid To Hurricanes, Zachary D. Schumann

Theses and Dissertations

The increased frequency and intensity of extreme weather events from climate change necessitates understanding impacts on critical infrastructure, particularly electrical transmission grids. One of the foundational concepts of a grid’s resilience is its robustness to extreme weather events, such as hurricanes. Resilience of the electric grid to high wind speeds is predicated upon the location and physical characteristics of the system components. Previous modeling assessments of electric grid failure were done at the systems level with assumptions on location and type of specific components. To facilitate more explicit adaptation metrics, accurate component-level information is needed. In this study, we build …

Atmospheric Meteorological Effects On Forecasting Daily Lightning Occurrence At Cape Canaveral Space Force Station, Jon Saul, Torrey J. Wagner, Eric G. Mbonimpa, Brent T. Langhals

Faculty Publications

As the Cape Canaveral Space Force Station and Kennedy Space Center increase their launch rate, any process that could assist in the automation of the currently-manual lightning forecast would be valuable. This work examines the possibility of machine-learning assistance with the daily lighting forecast which is produced by the 45th Weather Squadron. A dataset consisting of 34 lightning, pressure, temperature and windspeed measurements taken from 334 daily weather balloon (rawinsonde) launches in the timeframe 2012-2021 was examined. Models were created using recursive feature elimination on logistic regression and XGClassifier algorithms, as well as Bayesian and bandit optimization of neural network …

Feasibility Of Obtaining Surface Layer Moisture Flux Using An Ir Thermometer, Steven T. Fiorino, Lance Todorowski, Jaclyn Schmidt, Yogendra Raut, Jacob Margraf

Faculty Publications

This paper evaluates the feasibility of a method using a single hand-held infrared (IR) thermometer and a mini tower of wet and dry paper towels to psychometrically obtain surface layer temperature and moisture gradients and fluxes. Sling Psychrometers have long been standard measuring devices for quantifying the thermodynamics of near-surface atmospheric gas–vapor mixtures, specifically moisture parameters. However, these devices are generally only used to measure temperature and humidity at one near-surface level. Multiple self-aspirating psychrometers can be used in a vertical configuration to measure temperature and moisture gradients and fluxes in the first 1–2 m of the surface layer. This …

Effect Of Trigonometric Transformations On The Machine Learning Prediction And Quality Control Of Air Temperature, Andrea Fenoglio, Torrey J. Wagner, Paul Auclair, Brent T. Langhals

Faculty Publications

Conducting effective quality control of weather observations in real time is vital to the 14th Weather Squadron’s mission of providing authoritative climate data. This study explored automated quality control of weather observations by applying multiple machine learning techniques to 43,487 surface weather observations from 5 years of data at a single location. Temperature predictors were evaluated using recursive feature elimination on linear regression and XGBoost algorithms, as well as using a neural network hyperparameter sweep. Modeling was repeated after calculating trigonometric transforms of temporal variables to give the models insight into the diurnal heating cycle of the Earth. All models …

Deep Learning For Weather Clustering And Forecasting, Nathaniel R. Beveridge

Theses and Dissertations

Clustering weather data is a valuable endeavor in multiple respects. The results can be used in various ways within a larger weather prediction framework or could simply serve as an analytical tool for characterizing climatic differences of a particular region of interest. This research proposes a methodology for clustering geographic locations based on the similarity in shape of their temperature time series over a long time horizon of approximately 11 months. To this end an emerging and powerful class of clustering techniques that leverages deep learning, called deep representation clustering (DRC), are utilized. Moreover, a time series specific DRC algorithm …

Correction Of Back Trajectories Utilizing Machine Learning, Britta F. Gjermo Morrison

Theses and Dissertations

The goal of this work was to analyze 24-hour back trajectory performance from a global, low-resolution weather model compared to a high-resolution limited area weather model in particular meteorological regimes, or flow patterns using K-means clustering, an unsupervised machine learning technique. The duration of this study was from 2015-2019 for the contiguous United States (CONUS). Three different machine learning algorithms were tested to study the utility of these methods improving the performance of the CFS relative to the performance of the RAP. The aforementioned machine learning techniques are linear regression, Bayesian ridge regression, and random forest regression. These results mean …

Ccsfs/Ksc Total Lightning Warning Radii Optimization For Merlin Using Preexisting Lightning Areas, Kimberly G. Holland

Theses and Dissertations

The purpose of this research is to optimize lightning warning radii specifications for the 45th Space Wing (45 SW), thus reducing the number of unnecessary warnings that delay ground processing needed for space launch execution at Kennedy Space Center and Cape Canaveral Space Force Station. This thesis sought to answer two key research questions addressing: 1) What radius reduction effectively balances both safety and operations and do reduction recommendations from previous research align with results from the new detection system? 2) What insights can be gained from comparing measurement results for seasonal lightning events as well as lightning types? This …

Behavior Of Lightning In Developing Storms, Erick A. Tello

Theses and Dissertations

Air Force weather squadrons issue a warning when lightning activity is observed within 5 nautical miles (NM) of protected areas. Upon receiving this warning, personnel outdoors are expected to pause work and move inside. Studies sponsored by the 45th Weather Squadron (45 WS) have concluded that the 5 NM warning radius can be safely reduced for well-developed storms. This thesis investigates whether radii for storms in early development can also be reduced. Our research develops algorithms to partition lightning sensor data into storms. Next, storms are filtered to their earliest lightning events, and the study calculates distances between successive early …

Machine Learning Modeling Of Horizontal Photovoltaics Using Weather And Location Data, Christil Pasion, Torrey J. Wagner, Clay Koschnick, Steven J. Schuldt, Jada B. Williams, Kevin Hallinan

Faculty Publications

Solar energy is a key renewable energy source; however, its intermittent nature and potential for use in distributed systems make power prediction an important aspect of grid integration. This research analyzed a variety of machine learning techniques to predict power output for horizontal solar panels using 14 months of data collected from 12 northern-hemisphere locations. We performed our data collection and analysis in the absence of irradiation data—an approach not commonly found in prior literature. Using latitude, month, hour, ambient temperature, pressure, humidity, wind speed, and cloud ceiling as independent variables, a distributed random forest regression algorithm modeled the combined …

An Analysis Of A Lighting Prediction Threshold For 45th Weather Squadron Electric Field Mill Data, Charles A. Skrovan

Theses and Dissertations

The mission of the 45th Weather Squadron (45 WS) is to “exploit the weather to assure safe access to air and space” for Patrick Air Force Base, Cape Canaveral Air Force Station (CCAFS), and Kennedy Space Center (KSC) in support of various operations (United States Air Force, n.d.). To support that mission the 45 WS hosts a suite of weather detection instruments that include a lightning warning system that consists of an array of 31 electric field mills (EFM) and a lightning detection and ranging system (Department of the Air Force, 1976). Electric field mills at Cape Canaveral continuously record …

E700xd Portable Doppler Radar Energy Systems Analysis, Brandon M. Bailey, Torrey J. Wagner, Jada Williams

Faculty Publications

Occurring in industrialized nations, inexpensive and abundantly available power is routinely taken for granted. However, energy resilience and to a lesser extent price are key concerns when considering potential solutions for disaster response, disaster relief, or military operations. The Department of Defense (DoD) currently uses a 5 kW generator to power the E700XD portable Doppler radar system when grid power is unavailable [1]. While the radar has an approximate power consumption of 2.5 kW, there is a potential for higher demand due to weather conditions [2]. This paper examines the cost of operating a currently installed generator, compared to the …

Comparing Dual-Polarization Radar Lightning Forecast Methods Across Southwest Utah, Daniel O. Katuzienski

Theses and Dissertations

Lightning poses a significant hazard to space launch operations to include ground processing, launch window forecasts, and rocket-triggered lightning events. Two lightning initiation forecast methods using weather radar developed in Gremillion and Orville (1999) and Travis (2015) for Cape Canaveral Air Force Station (CCAFS) and Kennedy Space Center (KSC), Florida are tested in a new geographical region. This is accomplished by applying the highest-performing radar parameters from Gremillion and Orville (1999): reectivity (Z) ≥ 40 dBZ for two consecutive volume scans at the -10°C thermal height and Travis (2015): Z ≥ 36.5 dBZ with differential reflectivity (ZDR) ≥ 0.31 dB …

Coupled Atmospheric Surface Observations With Surface Aerosol Particle Counts For Daytime Sky Radiance Quantification, Scott S. Wolfmeyer

Theses and Dissertations

This study investigates the radiative transfer code, Laser Environmental Effects Definition and Reference (LEEDR), developed by the Center for Directed Energy at the Air Force Institute of Technology. Many multi- and hyperspectral applications are limited to the nighttime due in large part to daytime solar background noise and it is advantageous to be able to quantify this background noise using LEEDR. Real-time meteorological surface observations, numerical weather prediction, and aerosol particle concentrations were used to investigate the accuracy of LEEDR radiances simulations. Comparisons between simulations and measured values show that aerosol concentrations, weather predictions, and meteorological observations provide enough information …

Effect Of Using Probabilistic Contingency Tables To Modify Forecast Predictions, Sarah A. Gold

Theses and Dissertations

The 45th Weather Squadron (45 WS) records daily rain and lightning probabilistic forecasts and the associated binary event outcomes. Subsequently, they evaluate forecast performance and determine necessary adjustments with an implemented verification process. For deterministic outcomes, weather forecast analysis typically utilizes a Tradition Contingency Table (TCT) for verification, however the 45 WS uses an alternative tool, the Probabilistic Contingency Table (PCT). Using the TCT for verification requires a threshold, typically at 50%, to dichotomize probabilistic forecasts. The PCT maintains the valuable information in probabilities and verifies the true forecasts being reported. Simulated forecasts and outcomes as well as 2015-2018 45 …

Lightning Prediction Using Recurrent Neural Networks, Dominick V. Speranza

Theses and Dissertations

Cape Canaveral Air Force Station (CCAFS), Kennedy Space Center (KSC), and Patrick Air Force Base (PAFB) all reside in the thunderstorm capital of the United States. According to the Florida Climate Center, these installations experience more thunderstorms per year than any other place in the United States. It is the mission of the 45th Weather Squadron to provide timely and accurate warnings of weather conditions such as lightning that pose a risk to assets and personnel CCAFS, KSC and PAFB. To aid 45th Weather Squadron forecasters, a network of 30 Electric Field Mills (EFM) was installed in the area in …

The Geographic Distribution Of Downburst Frequency Across Spaceport Florida, Tania M. Garza

Theses and Dissertations

Strong winds from downbursts pose a significant hazard to personnel and launch operations at Cape Canaveral Air Force Station (CCAFS) and NASA Kennedy Space Center (KSC). The CCAFS/KSC complex has a robust mesonet from which an 18 year (1995-2012) warm-season convective wind climatology (WSCWC) was developed (Koermer 2017). While the frequency of downbursts in the area has been determined, the frequency at the individual tower locations has not. The 5-minute peak wind data from the WSCWC was analyzed to determine the geographic distribution of downburst frequency across Spaceport Florida. For this project a downburst was determined by the threshold of …

Methodology To Analyze Tropical Cyclone Intensity From Microwave Imagery, Matthew W. Perkins

Theses and Dissertations

Satellites with microwave remote sensing capabilities can be utilized to study atmospheric phenomena through high-level cloud cover (particularly cirrus), an advantage over visible and infrared bands, which only sense cloud tops. This unique capability makes microwave imagery ideal for studying the cloud structures of tropical cyclones (TCs) in detail, and relating these features to TC intensity. Techniques to estimate the intensity of TCs using infrared imagery, such as the Dvorak technique, have been used in TC forecasting for 40 years. However, due to the inherent temporal limitations of microwave imagery, no such similar technique exists for the microwave spectrum. This …

A New Analysis Of The Gálvez-Davison Index For Convective Forecasts In Northern Africa, Gabriel D. Donndelinger

Theses and Dissertations

Severe wind gusts and thunderstorms have been difficult to forecast in Africa. Traditional convective forecast tools (e.g. Total Totals Index, Lifted Index, K Index (KI) and Convective Available Potential Energy) do not accurately portray potential for thunderstorms in Africa. This research effort used the Gálvez-Davison Index (GDI), a convective index created for the tropics, and assess its applicability to northern Africa. GDI was produced for the Caribbean and Central America, and utilized temperature, moisture, mid-level stability, dry air entrainment and an elevation factor to calculate convective potential. In this research, GDI and KI were calculated using Global Forecast System (GFS) …

Forecasting Lightning Cessation Using Dual-Polarization Radar And Lightning Mapping Array Near Washington, D.C., Nancy M. Holden

Theses and Dissertations

Accurate forecasts of thunderstorms are important to space launch, aviation, and public safety. While prior studies have primarily focused on atmospheric conditions leading to lightning onset, less research has been dedicated to the challenging problem of predicting lightning cessation. This study verifies the probabilistic lightning cessation model developed by Joseph Patton (2017) at Florida State University for use by the U.S. Air Force's 45th Weather Squadron at Cape Canaveral Air Force Station (CCAFS) and the National Aeronautics and Space Administration (NASA) Kennedy Space Center (KSC). The Washington, D.C. greater metropolitan area, which presents a climate different to that of central …

Estimation Of Temporal Variations In Path-Averaged Atmospheric Refractive Index Gradient From Time-Lapse Imagery, Santasri Basu, Jack E. Mccrae, Steven T. Fiorino, Jared Przelomski

Faculty Publications

The sea level vertical refractive index gradient in the U.S. Standard Atmosphere model is −2.7×10⁻⁸  m⁻¹ at 500 nm. At any particular location, the actual refractive index gradient varies due to turbulence and local weather conditions. An imaging experiment was conducted to measure the temporal variability of this gradient. A tripod mounted digital camera captured images of a distant building every minute. Atmospheric turbulence caused the images to wander quickly, randomly, and statistically isotropically and changes in the average refractive index gradient along the path caused the images to move vertically and more slowly. The temporal variations of …

Wartime Construction Project Outcomes As A Function Of Contract Type, Ryan M. Hoff, Gregory D. Hammond, Peter P. Feng, Edward D. White

Faculty Publications

The United States has spent more than $23 billion on construction in Afghanistan since 2001. The dynamic security situation created substantial project uncertainty, and many construction projects used cost-plus-fixed-fee contracts (CPFF) instead of the firm-fixed-price (FFP) norm. Using a dataset of 25 wartime construction projects managed by the Air Force Civil Engineer Center, the authors sought to confirm that both contract types yield project outcomes consistent with the established literature. As expected, they found CPFF contracts had greater cost and schedule growth than FFP. However, they did not find differences regarding as-built quality. Additionally, the authors sought to determine whether …

Meeting The Dod’S Tactical Weather Needs Using Cubesats, Shayna K. Mckenney

Theses and Dissertations

This thesis investigates a CubeSat design that uses Commercial-Off-The-Shelf (COTS) components to capture, store, process, and downlink collected terrestrial weather data at resolutions near stat-of-the-art. The weather phenomena to be detected and transmitted in a timely manner are cloud formations, wind profiles, ocean currents, sea state, lightning, temperature profiles, and precipitation. It is hypothesized and shown that the proposed design will provide an improvement on the current U.S. tactical weather collection satellites because of the anticipated increased reliability and lowered cost to build and maintain the proposed CubeSat constellation. The methodology employed a multi-phase approach through the collective research of …

Forecasting Sea Breeze Enhanced Thunderstorms At Eglin Air Force Base: A Comparison Between Empirical Methods And The High Resolution Rapid Refresh Model, Matthew P. King

Theses and Dissertations

Summer thunderstorms pose the greatest forecast challenge for the United States Air Force's 96th Weather Flight (96WF) which provides weather support for Eglin Air Force Base (AFB). Located on Florida's panhandle, Eglin AFB encompasses the Department of Defense's largest range complex which covers 134,000 sq mi (347,000 sq km) of water and 724 sq mi (1875 sq km) of land. Due to the base's coastal location and unique coastline geometry, sea breeze is a dominant forecast consideration for thunderstorms. The 96 WF currently utilizes an empirical method, called WINNDEX, developed by former staff meteorologist Roger Winn to aid in thunderstorm …

The Reliability And Skill Of Air Force Weather’S Ensemble Prediction Suites, Derek A. Burns

Theses and Dissertations

The Ensemble Prediction Suites (EPS) used at the 557th Weather Wing (557 WW) provide probability based forecasts for thousands of worldwide locations. Point Ensemble Probability (PEP) bulletins are tailored specifically to the United States military and its criteria for operationally significant weather. This study assesses the performance of PEPs from each EPS at 17 Continental United States (CONUS) locations. Results from April to October of 2015 show that each EPS is underforecasting ceilings and visibility for most forecast hours at several locations. The underforecasting of ceilings is most severe at Vandenberg AFB, an area prone to frequent marine layer fog …