Browsing by Author "Rasmussen, Kristen, committee member"

Now showing 1 - 15 of 15

Open Access
An analysis of total lightning characteristics in tornadic storms: preparing for the capabilities of the GLM
(Colorado State University. Libraries, 2017) Reimel, Karly Jackson, author; Rutledge, Steven, advisor; Miller, Steven, advisor; Rasmussen, Kristen, committee member; Eykholt, Richard, committee member
Numerous studies have found that severe weather is often preceded by a rapid increase in the total lightning flash rate. This rapid increase results from numerous intra-cloud flashes forming around the periphery of an intensifying updraft. The relationship between flash rates and updraft intensity is extremely useful to forecasters in severe weather warning decision making processes, but total lightning data has not always been widely available. The Geostationary Lightning Mapper (GLM) will be the first instrument to detect lightning from geostationary orbit, where it will provide a continuous view of lightning over the entire western hemisphere. To prepare for the capabilities of this new instrument, this thesis analyzes the relationship between total lightning trends and tornadogenesis. Four supercellular and two non-supercellular tornadic storms are analyzed and compared to determine how total lightning characteristics differ between dynamically different tornadic storms. Supercellular tornadoes require a downdraft to form while landspout tornadoes form within an intensifying updraft acting on pre-existing vertical vorticity. Results of this analysis suggest that the supercellular tornadoes we studied show a decrease in flash rate and a decrease in lightning mapping array (LMA) source density heights prior to the tornado. This decrease may indicate the formation of a downdraft. In contrast, lightning flash rates increase during landspout formation in conjunction with an intensifying updraft. The total lightning trends appear to follow the evolution of an updraft rather than directly responding to tornadogenesis. To further understand how storm microphysics and dynamics impact the relationship between lightning behavior and tornadogenesis, two of the tornadic supercells were analyzed over Colorado and two were analyzed over Alabama. Colorado storms typically exhibit higher flash rates and anomalous charge structures in comparison to the environmentally different Alabama storms that are typically normal polarity and produce fewer flashes. The difference in microphysical characteristics does not appear to affect the relationship between total lightning trends and tornadogenesis. The capabilities of GLM are yet to be determined because the instrument is still in its calibration/validation stages. However, as part of the GLM cal/val team, we were in a unique position to examine the first-light GLM data and contribute to the assessment of its performance for noteworthy thunderstorm events during the Spring/Summer seasons of 2017. The final chapter of this thesis displays a preliminary analysis of GLM data. A first look into GLM performance is established by comparing GLM data with data from other lightning detecting instruments. Overall, GLM appears to detect fewer flashes than other lightning detecting networks and instruments in Colorado storms, more so for intense storms compared to weaker storms.
Open Access
Characteristics of linear mesoscale convective systems during DYNAMO
(Colorado State University. Libraries, 2019) Messina, Joseph, author; Rutledge, Steven, advisor; Xu, Weixin, committee member; Rasmussen, Kristen, committee member; Eykholt, Richard, committee member
Mesoscale convective systems (MCSs) have long been known to play a large part in the vertical transport of horizontal momentum. They also contribute to the vertical redistribution of heat and radiative forcing. The Madden Julian Oscillation (MJO) is a tropical disturbance that propagates across the central Indian Ocean and western Pacific Ocean with an intraseasonal cycle of 30-60 days. Many studies have explored the kinematic characteristics and organization of MCSs in the tropics, while others investigated the characteristics of convective systems within the MJO. However, there remains a gap in current literature on the connection between MJO phase and kinematics of precipitating tropical convection. Those studies that did examine MCSs in the tropical environment did so with limited observations. This study used radar, sounding, and meteorological data from the Dynamics of the Madden Julian Oscillation (DYNAMO) field campaign in the central Indian Ocean to examine the influence of vertical shear on the orientation of linear MCSs, effects of cold pools on propagation of linear systems, and the mesoscale flow features of the MCSs over the phases of the MJO. DYNAMO took place from October-December 2011 and produced a vast dataset for the analysis of tropical convection during multiple MJO events. Our results show that convection during DYNAMO was consistent with studies from previous tropical field campaigns. That is, convective lines are frequently oriented perpendicular to strong low-level shear. In the absence of strong low-level shear, they are oriented parallel to strong mid-level shear. Linear systems were more prevalent during active MJO phases. Cold pools did not play a substantial role in tropical squall line propagation. Kinematic features are also consistent with previous works. The presence of a jump updraft and descending rear inflow were ubiquitous in our samples. The absence of a downdraft outflow was common. This result shows that the MCSs studied were transporting front to rear horizontal momentum from low- to mid-levels and rear to front horizontal momentum from low- to mid-levels.
Open Access
Data-driven improvements to GPROF-based satellite snowfall retrievals with a focus on mountain snowfall
(Colorado State University. Libraries, 2025) Gonzalez, Ryan L., author; Kummerow, Christian, advisor; Liston, Glen, committee member; Chiu, Christine, committee member; Rasmussen, Kristen, committee member; Notaros, Branislav, committee member
Snowfall is a critical component of Earth's hydrological and climate system despite only 5% of Earth's annual precipitation falling as snow. Satellite-based snowfall estimates, particularly those obtained from the Global Precipitation Measurement (GPM) Microwave Imager (GMI), struggle to accurately estimate the total annual snowfall accumulations, especially in mountainous regions of the world. Part of the challenge is due to the reference precipitation used in the GMI-based algorithms, while radiometers struggle to distinguish between the microwave signatures of surface snowpack and snowfall. The aim of this dissertation is to evaluate the impact machine learning-based GMI retrievals have on snowfall estimates, explore how temperature and climatological adjustments to the reference precipitation can provide additional information to the retrieval, and asses if these changes lead to improved snowfall accumulations required for modeling the lifecycle of snow. A key objective of this study is to improve snowfall accumulation estimates in mountainous areas, where snowpack is a critical component of water storage. First, snowfall rates estimated from the Goddard Profiling Algorithm (GPROF) for GMI are compared using three types of GPROF algorithms: one Bayesian (GPROF V7) and two neural network versions (GPROF-NN 1D and GPROF-NN 3D). The highest detection and quantitative statistics are observed using GPROF-NN 3D with both neural network retrieval algorithms outperforming the Bayesian version. It is shown that artificial biases in the retrieval statistics can result from the selected threshold for snow/no-snow classification. Coincident in-situ snowfall and radar data are also used to evaluate the temperature dependency of the reflectivity-snowfall (Z-S) relationship and how it impacts the GPROF retrievals. Second, an evaluation of the three GPROF algorithms is conducted in the mountains of the western United States. Using data from a snow reanalysis dataset, water year snowfall accumulations from the Multi-Radar Multi-Sensor (MRMS) are adjusted to produce more realistic snowfall magnitudes and spatial patterns. These adjustments were found to decrease errors in snowfall accumulation estimates for all three retrieval algorithms, resulting in significant improvements when compared to independent SNOTEL observations. These results provide a positive outlook for snowfall retrievals in mountainous regions by incorporating additional information to the retrieval algorithm. Finally, a framework for incorporating satellite precipitation estimates into a snow evolution model in the western United States is presented that offers a flexible design to account for different study domains. The objective of this framework is to present an approach for deriving snow water equivalent (SWE) from satellite precipitation estimates given the difficulties of directly measuring SWE from passive microwave sensors. A UNet-based retrieval model is used to estimate precipitation at 30 minute time resolution across the currently available passive microwave and infrared sensors. The initial precipitation estimates were found to have a systematic bias across the study period, which, after correction, produced realistic spatial patterns of snow depth and snow water equivalent, but underestimated the magnitudes compared to two reference snow model simulations.
Open Access
Ensemble-based analyses of liminal extreme rainfall events near Taiwan and northern Colorado
(Colorado State University. Libraries, 2022) Cole, Alexandra S., author; Bell, Michael M, advisor; Rasmussen, Kristen, committee member; Nelson, Peter, committee member
Heavy rainfall is a phenomenon that impacts a variety of climates around the world, from the moisture-rich, tropical northwestern Pacific to the drier Northern Colorado plains. Improvements over decades of numerical weather prediction have allowed for increased accuracy in simulations of heavy rainfall cases, but there are still improvements yet to be made. This thesis, in conjunction with the Prediction of Rainfall Extremes Campaign in the Pacific (PRECIP) field campaign, aims to study the mechanisms behind these heavy rainfall events to increase understanding of their underlying processes and improve the modeling of them. Two weather events are investigated in detail, one in which heavy rainfall was not forecast by global models but greater than 600 mm of rainfall accumulated, and a contrasting case in which heavy rainfall was forecast but little to no rainfall accumulated. On 09 June 2020 near Taiwan, heavy rainfall was produced by quasi-stationary back-building mesoscale convective systems (MCS) associated with a mei-yu front. Peak rainfall amounts totaled over 600 mm with widespread rainfall totals greater than 100 mm. Global model forecast skill was poor in both location and intensity of rainfall. The mesoscale ensemble showed liminal conditions between heavy rainfall or little to no rainfall. The two most accurate and two least accurate ensemble members are selected for analysis via validation against radar-estimated rainfall observations. All members feature moisture-rich environments and moist neutral soundings with low levels of free convection (LFC) and sufficient instability for deep convection, and the synoptic setups do not suggest such different outcomes. Through our analysis, we find that stronger gradients in 100 m virtual potential temperature in the two most accurate members associated with a near-surface frontal boundary provide the primary lifting mechanism for enhanced rainfall. In the two heaviest rain-producing members, air moves north/northeastward and ascends the virtual potential temperature isentropes and rises above the LFC, producing back-building deep, moist convection. The near-surface gradients are weaker and more confined along Taiwan's coast in the two least accurate members, which leads to less rainfall that is misplaced from reality. The analyses suggest that subtle details in the simulation of frontal boundaries and meso-scale flow structures can lead to bifurcations in producing extreme or almost no rainfall. A contrasting event occurred in Northern Colorado on 31 July 2021, where heavy rainfall was forecast and flood warnings were issued, but little to no rainfall and flooding took place in the forecast area. Synoptic and mesoscale conditions were ripe for heavy rainfall, with anomalously high precipitable water values and moderate values of CAPE. Similar to the 09 June 2020 case, the mesoscale ensemble showed a wide spread in rainfall totals, related in part to the variability of surface boundaries and forcing across the ensemble. Weak surface forcing led to very little rainfall in this case despite the high moisture, suggesting similar physical mechanisms and predictability challenges across both the analyzed cases. Implications for improved probabilistic forecasts, increased forecast accuracy, and thus increased public safety for heavy rainfall events are discussed.
Open Access
From rain gauges to retweets: using diverse datasets to explore overlapping hazards and human experiences in landfalling tropical cyclones
(Colorado State University. Libraries, 2021) Mazurek, Alexandra C., author; Schumacher, Russ, advisor; Henderson, Jen, committee member; Morrison, Ryan, committee member; Rasmussen, Kristen, committee member
Landfalling tropical cyclones (LTCs) are responsible for numerous hazards, including damaging winds, storm surge, inland flooding, and tornadoes. Furthermore, multiple hazards may threaten an area at the same time, which raises challenges from a prediction, warning operations, and human impacts standpoint. Previous research has approached overlapping tornado and flash flood events—which exemplify these challenges because the recommended protective actions can be in conflict—in continental systems from multidisciplinary perspectives, but less work has been done to explore these phenomena in LTC environments. Because LTCs also introduce other hazards, additional complexities may exacerbate already challenging circumstances. This work integrates meteorological and social sciences to broadly advance the understanding and implications of simultaneous flash flood and tornado events in LTCs. Part I of this thesis investigates the relationship between two predecessors to tornadoes and flash floods—meso- to storm-scale rotation and heavy rainfall rates, respectively—using observations. Motivated by previous work that has drawn linkages between these two processes in continental convective storms, this connection is explored in Tropical Storm Imelda, a system that was among the wettest LTCs on record to impact the contiguous United States (CONUS), producing rainfall accumulations in excess of 1000 mm when it made landfall on the western Gulf Coast in September 2019. First, a synoptic and mesoscale overview of the tropical cyclone (TC) is presented as motivation for its utility in examining overlapping embedded rotation and extreme rainfall rates. Then, rain gauges from a high-density observing network in southeast Texas are analyzed alongside polarimetric radar data to compare rainfall rates that occur in the presence of embedded rotation to those that occur when no rotation is evident on radar. According to these results, 5-minute rainfall rates that followed subjectively-identified meso- to storm-scale rotation on radar tended to be statistically significantly greater, and when accumulated over time, more than twice as much rainfall was recorded at gauge sites when rotation was present near the gauge compared to when there was no rotation located nearby. To further quantify the spatial and temporal relationships of embedded rotation and heavy rainfall rates, quantitative precipitation estimates (QPE) and rotation tracks from the Multi-Radar Multi-Sensor system are compared in time and space. A positive correlation was found to exist between the hourly-accumulated 0-2 km rotation tracks and hourly local gauge bias-corrected QPE, suggesting that more rain tends to fall in the presence of low-level rotation. In Part II of this thesis, social science methods are used to investigate another LTC: Hurricane Harvey (2017)—an unprecedented event that became the wettest LTC on record to impact CONUS and spawned over 50 tornadoes when it affected the western Gulf Coast. This work aims to explore the notion of experience as it evolves on Twitter in real-time during Harvey among a group of users who were located in areas that were impacted by the LTC and its overlapping hazards. Though a significant amount of research has investigated experience through surveying and interview techniques after LTCs occur, much less work has been done to study experience as it is shared live during an event or through the lens of social media. Using this motivation and drawing on the overarching theme of concurrent hazards, this research begins with a database of tweets composed during the period surrounding Hurricane Harvey that reference tornadoes and flash flooding. The sample is refined through a multi-step querying process, ultimately resulting in a group of 39 users who shared 158 tweets about "past events"—that is, events related to LTCs and/or the hazards that are associated with them. These tweets are thematically analyzed by individual users, by individual past events, and over time. The results of these analyses show that Twitter users referenced past events during Harvey for two main reasons: first, because the user has a personal connection to the event and second, because the past event is helping them to make sense of various aspects of the situation that is unfolding around them. Understanding what roles past events may play in a real-time crisis is useful to leaders and decision-makers, such as meteorologists, local politicians, and emergency managers, as it provides insight on the evolving needs and concerns of the public that they serve as they change and are modulated by various events that unfold throughout the overarching crisis.
Open Access
Improving predictions and generating actionable forecast insights for downslope windstorms with machine learning
(Colorado State University. Libraries, 2025) Zoellick, Casey L., author; Schumacher, Russ, advisor; Rasmussen, Kristen, committee member; Barnes, Elizabeth, committee member; Nelson, Peter, committee member
Downslope windstorms are an extreme weather phenomenon characterized by accelerating winds down the lee slope of a mountain with gusts often exceeding 45 m s-1. These impact society through damage directly related to the high winds, ground transportation concerns in the vicinity of the windstorm, aviation impacts through the accompanying mountain wave turbulence, and fueling the rapid intensification and spread of wildfires such as the 2018 Camp Fire, the 2021 Marshall Fire, and the 2023 Lahaina Fire. Despite improvements in numerical weather prediction and observational datasets, predictability of these windstorms still rarely exceeds 12 hours further exacerbating their impacts. Recent advances have made machine learning (ML) more accessible to researchers and have shown promise in improving forecasts of other extreme weather phenomena. We first present models driven by two different types of ML architectures that classify wind events as moderate or high at three locations along the Rocky Mountain Front Range: Cheyenne, Wyoming; Fort Collins, Colorado; and Boulder, Colorado. The first type of architecture is the random forest (RF), which is comprised of multiple decision trees, and the second type is the convolutional neural network (CNN), which is a deep learning method that excels at image recognition. These models make forecasts at the Day 1 and Day 2 lead times based on predictors derived from a 12-km version of the WRF operated at Colorado State University. The results show improvement over the direct weather model forecasts. CNNs show enhanced event detection capability compared to the RFs but with a higher false alarm rate limiting their utility in some cases. Next, explainable artificial intelligence (XAI) techniques are presented. Feature importances indicate that the ML models rely on predictors at geographic locations that align with known atmospheric variables important to downslope wind forecast along the terrain. Also, a framework for reducing the dimensions of the predictor data and clustering these data with a Gaussian mixture model yields insights to the forecast ML models' performance and the synoptic conditions in which downslope windstorms along the Front Range occur. The ML models perform better in regimes characterized by prominent synoptic features such as cold air advection or the presence of the jet stream aloft. Lastly, we investigate whether increasing the resolution of the traditional weather model creating the ML predictors results in performance improvements. We use NOAA's High Resolution Rapid Refresh (HRRR) model to derive input predictors for newly trained CNNs and observe a decrease in false alarms that results in an overall performance boost over the direct HRRR forecasts. A case study on the Marshall Fire is conducted and indicates that the HRRR-based CNN is able to correctly forecast the subsequent downslope wind event before the wind event is explicitly depicted in the HRRR output itself. This study is an example of how ML fused with current weather models closes the forecast gap in these impactful weather phenomena with incomplete physical understandings.
Open Access
Mineral dust lofting and interactions with cold pools
(Colorado State University. Libraries, 2021) Bukowski, Jennifer, author; van den Heever, Susan, advisor; Chiu, Christine, committee member; Rasmussen, Kristen, committee member; Jathar, Shantanu, committee member; Barth, Mary, committee member; Miller, Steven, committee member
Convective dust storms, or haboobs, form when strong surface winds loft loose soils in convective storm outflow boundaries. Haboobs are a public safety hazard and can cause a near instantaneous loss of visibility, inimical air quality, and contribute significantly to regional dust and radiation budgets. Nevertheless, reliable predictions of convective dust events are inhibited by a lack of understanding regarding the complex and non-linear interactions between cold pools, dust radiative effects, and land surface processes, and their associated uncertainties in numerical models. In this dissertation, model simulations of real and idealized haboobs are used to address limitations in regional dust modeling, the direct radiative effect of mineral dust on cold pool properties and dynamics, and feedbacks between haboobs and the land surface. In the first study, we assess the influence of horizontal resolution, specifically parameterized versus convection-allowing resolution, on dust lofting, vertical transport, and aerosol heating rates in the WRF-Chem regional model. On average, convection-permitting simulations exhibit higher surface wind speeds, enhanced convective activity, and drier soil, which leads to more dust emissions to the atmosphere. More frequent and stronger vertical velocities also transport dust further aloft and increase the atmospheric lifetime of these particles. We conclude that tuning dust emissions in coarse-resolution regional simulations can only improve the results to first-order and cannot fully rectify discrepancies in the representation of convective dust transport in terms of aerosol distributions or the net aerosol radiative effect. The second study, WRF-Chem is utilized to simulate the effect dust radiation interactions have on a long-lived haboob case study that spans three distinct radiative regimes: day (high shortwave), evening (low shortwave), and night (longwave only). A sophisticated algorithm, known as TOBAC, is used to track and identify the numerous cold pool boundaries and assemble statistics that represent the impact of including dust radiative effects. To first order, dust scattering of shortwave radiation in the day leads to a colder, dustier, and faster moving cold pool. In the transition period of early evening, the shortwave effects diminish while longwave absorption by dust leads to warmer and slower cold pools that loft less dust as they propagate onward. At night, the haboob is again warmer due to dust absorption, but gustier in the more stable nocturnal surface layer. Lastly, the third study focuses on feedbacks between parameters that affect both dust mobilization and cold pool dynamics. The Elementary Effects statistical method is applied to an ensemble of 120 idealized RAMS simulations of daytime and nighttime haboobs. This sensitivity analysis identifies and ranks the importance of different input factors in predicting haboob properties as: initial cold pool temperature, surface type, soil type, and finally soil moisture. Most of these parameters modify the cold pool via their impacts on surface fluxes, although the effect of surface type is dominated by the change in roughness length. A semi-linear connection between haboob dust and cold pool temperature is detected in the statistics, and a relationship between dust flux and cold pool temperature is proposed which relates haboob strength to the thermodynamic environment.
Open Access
Projecting end-of-century human exposure to eastern Colorado tornadoes and hailstorms: meteorological and societal perspectives
(Colorado State University. Libraries, 2020) Childs, Samuel J., author; Schumacher, Russ, advisor; Demuth, Julie, committee member; Ojima, Dennis, committee member; Rasmussen, Kristen, committee member; Rutledge, Steven, committee member
The eastern half of Colorado is one of the most active areas for hailstorms and tornadoes in the U.S. An average of 39 tornadoes and 387 severe hail reports are tallied each year over this domain, and a number of damaging events, particularly hailstorms, have occurred in recent years. In an era of climate change, it is of worth to project how the frequency, geography, and severity of tornadoes and hailstorms may change over time, and doing so on a localized scale can shed light on the small-scale complexities that broader analyses miss. It is important to consider both meteorological and non-meteorological effects when projecting the changing human risk and exposure to these hazards in the future, as human factors such as population growth means that more people may potentially be exposed to tornadoes and hailstorms regardless of how climate change may influence storm characteristics. As such, this doctoral study employs a multidisciplinary, multi-perspective approach to investigate how the tornado and severe hail footprint may change across eastern Colorado by the end of the 21st century, and in turn how the impacts on those who live and work in this area may be exacerbated. A baseline climatology of tornadoes and hailstorms across eastern Colorado is established using Storm Prediction Center data records. Both hazards show increasing frequency since the 1950s, but when the temporal range is limited to 1997–2017, tornado reports and days show decreasing trends while severe hail reports and days continue to show upward trends. Population bias is inherent in the data records of both hazards and manifests itself as a clustering of reports near urban centers and along major roadways where people live and travel. However, the increasing number of severe hail days and proportion of hail reported at larger sizes is less likely to be influenced by population growth and thus may have a meteorological origin. Convective parameters output from high-resolution dynamical downscaling simulations of control and future climate scenarios using the Weather and Forecasting model are used as proxies to create and compare synthetic tornado and hail reports between the two simulations. Up to three more severe hail days and one more tornado day per year on average by the period 2071–2100 is found, maximized in the north-central part of the domain. This result is combined with population projections from the Shared Socioeconomic Pathways in Tornado and Hail Monte Carlo models to simulate changes in the number of people living underneath tornado tracks and hail swaths by the year 2100. Human exposure evolution is sensitive to the overlap of population and hazard spatial footprints, but the model predicts worst-case scenarios of a 178% increase in exposure to severe hail and a 173% increase in exposure to tornadoes by the end of the 21st century. In addition, population effects outweigh meteorological effects when simulated independently. Some simulations yield a decreasing human exposure to severe hail due to the greatest projected increases in hailstorms over rural, agricultural land. This finding provides motivation for an interview study of eastern Colorado farmers and ranchers to measure perceptions of exposure and sensitivity to severe hail. Most interviewees view hailstorms as a common nuisance throughout eastern Colorado and are most concerned with small hail that falls in large volumes or is driven by a strong wind since these scenarios cause the most damage to crops. Respondents express anxiety and dejection toward hailstorms, as they can significantly affect crop yields and in turn impact their livelihoods and local economy. Understanding this agricultural perspective validates ongoing research into hail surface characteristics and can promote stronger partnerships between the forecasting and farming communities. The synthesis of results from this dissertation, with its unique localized look at the human and meteorological factors contributing to a changing exposure, can be of great worth to forecasters, urban planners, emergency managers, insurance agents, and other local decision-makers. Moreover, this work will help to educate the local public about the past, present, and future of tornadoes and severe hailstorms within eastern Colorado, with the aim of protecting lives and property from their negative impacts.
Open Access
Quantifying and understanding current and future links between tropical convection and the large-scale circulation
(Colorado State University. Libraries, 2020) Jenney, Andrea M., author; Randall, David A., advisor; Barnes, Elizabeth A., advisor; Maloney, Eric, committee member; Rasmussen, Kristen, committee member; Anderson, Georgiana Brooke, committee member
Tropical deep convection plays an important role in the variability of the global circulation. The Madden Julian Oscillation (MJO) is a large tropical organized convective system that propagates eastward along the equator. It is a key contributor to weather predictability at extended time scales (10-40 days). For example, variability in the MJO is linked with variability in meteorological phenomena such as landfalling atmospheric rivers, tornado and hail activity over parts of North America, and extreme temperature and rainfall patterns across the Northern Hemisphere. Links between the MJO and atmospheric variability in remote locations are heavily studied. This is in part because the current skill of weather forecasts at extended time scales is mediocre, and because of evidence suggesting that the potential predictability offered by the MJO may not be fully captured in numerical prediction models. In the first part of this dissertation, I develop a tool for these types of studies. The "Sensitivity to the Remote Influence of Periodic Events" (STRIPES) index is a novel index that condenses the information obtained through composite analysis of variables after a periodic event (such as the MJO) into a single number, which includes information about the life cycle of the event, and for a range of lags with respect to each stage of the event. I apply the STRIPES index to surface observations and show that the MJO signal is detectable and significant at the level of individual weather stations over many parts of North America, and that the maximum strength of this signal exhibits regionality and seasonality. Tropical convection affects the extratropics primarily through the excitation of Rossby waves at the places where the upper-tropospheric divergent outflow associated with deep convection interacts with the background wind. In a future warmer climate, the strength of the mean circulation and convective mass flux is expected to weaken. A potential consequence is a weakening of Rossby wave excitation by tropical convective systems such as the MJO. In the second part of this study, I analyze a set of idealized simulations with specified surface warming and superparameterized convection and develop a framework to better understand why the mean circulation weakens with warming. I show that the decrease in the strength of the mean circulation can be explained by the slow rate at which atmospheric radiative cooling intensifies relative to the comparatively fast rate that the tropical dry static stability increases. I also show that despite a decrease in the mean convective mass flux, the warming tendency of the convective mass flux over the most deeply- convecting regions is not constrained to follow that of the global mean. In the final part of this dissertation, I consider how changes in the MJO and of the mean atmospheric state due to warming from increases in greenhouse gas concentrations may lead to changes in the MJO's impact over the North Pacific and North America. Specifically, I show that changes to the atmosphere's mean state dry static energy and winds have a larger impact on the MJO teleconnection than changes to MJO intensity and propagation characteristics.
Open Access
Rain and RELAMPAGO: analysis of the deep convective storms of central Argentina
(Colorado State University. Libraries, 2023) Kelly, Nathan Robert, author; Schumacher, Russ, advisor; Rasmussen, Kristen, committee member; Bell, Michael, committee member; Nelson, Peter, committee member
When, where and how much precipitation falls are fundamental questions to research interests spanning the weather to climate spectrum, yet are difficult to solve. The various methods used to answer "how much" each have sources of error, making it important to obtain knowledge about the characteristics of an individual dataset. This is especially true for rare events such as extreme precipitation. IMERG, TRMM 3B42, MERRA2 and ERA5 precipitation datasets were regridded to the same resolution and compared for 3-hourly heavy rainfall (99th and 99.9th percentile) in subtropical South America, which has some of the strongest convective storms on Earth. Seasonal and dirunal distribution are compared, with similar seasonal distributions between the datasets but at the diurnal scale MERRA2 and ERA5 show more afternoon events than TRMM and IMERG. Thermodynamic environments were compared with MERRA2 events tending to occur in more marginal environments than TRMM 3B42 and ERA5 environments over much of the analyzed region. Overall the satellite datasets showed the highest amounts. Brief case studies are included to illustrate these differences, which reinforce that choice of dataset can be an important factor in precipitation research. How the precipitation falls is also addressed using a case study from the RELAMPAGO field program in Argentina. Many observations are available of this case, which occurred during the mobile operations period of the field program. Mobile surface stations, increased temporal resolution from fixed sounding sites, and six mobile sounding systems provide a high level of detail on the evolution of this storm system. Additionally, a trove of radar data and a GOES mesoscale sector are available. This case is demonstrative of a common occurrence in the region: a strong MCS (Mesoscale Convective System) over the Sierras de Córdoba mountain range. The extent of the backbuilding observed with this MCS was not predicted by the operation convective allowing models used for field program forecasting. To study this event two simulations are presented: one in which backbuilding of the MCS occurs and one where such backbuilding does not occur. The difference between these simulations is the number of vertical levels used in the model which impacts moisture availability upstream of the system via the effect of mountain wave downslope winds.
Open Access
Satellite-based investigation of convection and precipitation in tropical cyclone intensity change
(Colorado State University. Libraries, 2022) Razin, Muhammad Naufal Bin, author; Bell, Michael, advisor; Kummerow, Christian, committee member; Rasmussen, Kristen, committee member; Kirby, Michael, committee member
Multiple hypotheses on the role of convection and precipitation in tropical cyclone intensity change have been proposed, but a scientific consensus has not yet been obtained. Some recent studies emphasize the importance of asymmetric deep convection in intensification, while others emphasize the importance of more symmetric precipitation that is not necessarily deep. To help provide clarity on this issue, this dissertation analyzes a large dataset of satellite passive microwave observations in order to obtain a sufficient sample size, which enables us to minimize the impact of the tropical cyclone environmental variability on intensity. The first part of this dissertation involves compiling an extensive research-quality and open-access dataset of low-Earth orbit (LEO) satellite passive microwave observations centered on tropical cyclones, called the Tropical Cyclone Precipitation, Infrared, Microwave, and Environmental Dataset (TC PRIMED). TC PRIMED consists of tropical cyclone-centric 1) inter-calibrated, multi-channel, multi-imager microwave brightness temperatures, 2) retrieved rainfall from NASA's Goddard Profiling algorithm (GPROF), 3) nearly coincident geostationary satellite infrared imagery, and 4) auxiliary data such as tropical cyclone position and intensity, ERA5 fields and derived environmental diagnostics, and satellite precipitation radar variables. TC PRIMED includes observations from over 168,000 LEO satellite overpasses of 2,101 tropical cyclones from 1998 to 2019. A simple composite analysis demonstrates the huge potential of TC PRIMED. The second part of this dissertation uses Global Precipitation Measurement (GPM) satellite observations from TC PRIMED to train a random forest model to classify the precipitation type (i.e., convective and stratiform). The model uses raw and derived passive microwave brightness temperature variables as input predictors and observations from the GPM dual-frequency precipitation radar as reference. This approach leverages the wider swath of the GPM passive microwave observations to obtain a larger sample size of precipitation type observation for the subsequent study on the importance of convection and precipitation in tropical cyclone intensity change. The random forest model performs very well at delineating clear air from stratiform and convective precipitation, and captures key features like the tropical cyclone eye, eyewall, and primary rainbands. However, the model struggles to detect some finer features like randomly scattered convection. Analysis of the model's performance demonstrates the importance of passive microwave texture information for precipitation type classification. The final part of this dissertation uses the tools developed in parts one and two to investigate the distribution of convection and precipitation in tropical cyclone intensity change. This analysis minimizes the impacts of the environmental variability on intensity by only selecting observations of tropical cyclones in favorable environments. Results reveal that intensifying minor tropical cyclones have a more symmetric distribution of rainfall that is not necessarily convective, while intensifying major tropical cyclones have more numerous and intense deep convection in the upshear quadrants. The results lead to the following hypotheses: i) for minor tropical cyclones, a more symmetric distribution of rainfall is more efficient at intensifying the tropical cyclone, and ii) the occurrence of deep convection in the upshear quadrants of major tropical cyclones is optimal for intensification. This dissertation provides the scientific community with an extensive set of tools to investigate tropical cyclones using satellite passive microwave observations and ancillary data in the form of TC PRIMED. Subsequently, the second part of this dissertation demonstrates the utility of a machine learning model in classifying multiple precipitation types from satellite passive microwave observations, with discussions on model deficiencies providing a framework to further improve such approach. Finally, this dissertation leverages a large dataset of tropical cyclone observations to clarify the role of convection and precipitation in tropical cyclone intensification.
Open Access
Spatiotemporal variations in liquid water content in a seasonal snowpack: implications for radar remote sensing
(Colorado State University. Libraries, 2020) Bonnell, Randall Ray, author; McGrath, Daniel, advisor; Fassnacht, Steven, committee member; Rasmussen, Kristen, committee member
Mountain snowpacks act as seasonal reservoirs, providing a critical water resource to ~1.2 billion people globally. Regions with persistent snowpacks (e.g., mountain and polar environments) are responding quickly to climate change and are warming at faster rates than low-elevation temperate and equatorial regions. Since 1915, snow water equivalent (SWE) in the western U.S. snowpack has declined by 21% and snow covered area is contracting in the Rocky Mountains. Despite the clear importance of this resource and the identification of changes affecting it, no current remote sensing approach can accurately measure SWE at high spactiotemporal resolution. L-band (1-2 GHz) Interferometric Synthetic Aperture Radar (InSAR) is a promising approach for detecting changes in SWE at high spatiotemporal resolution in complex topography, but there are uncertainties regarding its performance, particularly when liquid water content (LWC) is present in the snowpack. LWC exhibits high spatial variability, causing spatially varying radar velocity that introduces significant uncertainty in SWE-retrievals. The objectives of this thesis include: (1) examine the importance of slope, aspect, canopy cover, and air temperature in the development of LWC in a continental seasonal snowpack using 1 GHz ground-penetrating radar (GPR), a proxy for L-band InSAR, and (2) quantify the uncertainty in L-band radar SWE-retrievals in wet-snow. This research was performed at Cameron Pass, a high elevation pass (3120 m) located in north-central Colorado, over the course of multiple survey dates during the melt season of 2019. Transects were chosen which represent a range in slope, aspect and canopy cover. Slope and aspect were simplified using the northness index (NI). Canopy cover was quantified using the leaf area index (LAI). Positive degree days (PDD) was used to represent available melt-energy from air temperature. The spatiotemporal development of LWC was studied along the transects using GPR, probed depths, and snowpit measured density. A subset of this project substituted Terrestrial LiDAR Scans (TLS) for probed depths. Surveys (17 in total, up to 3 surveys per date) were performed on seven dates which began on5 April 2019, where LWC values were ~0 vol. %, and ended on 19 June 2019 where LWC values exceeded 10 vol. %. Point measurements of LWC were observed to change (ΔLWC) by +9 vol. % or -8 vol. % over the course of a single day, but median ΔLWC were ~0 vol. % or slightly negative. LAI was negatively correlated with LWC for 13 out of the 17 surveys. NI was negatively correlated with LWC for 10 out of the 17 surveys. Multi-variable linear regressions to estimate ΔLWC identified several statistically significant variables (p-value < 0.10): LAI, NI, ΔPDD, and NI x ΔPDD. Snow-on Terrestrial LiDAR Scans (TLS) were conducted twice during the melt season, and a snow-off scan was conducted in late summer. Snow-on scans were differenced from the snow-off scan to produce distributed snow depth maps. TLS-derived snow depths compared poorly with probe-derived depths, which is attributed to poor LiDAR penetration through the thick vegetation present during the snow-off scan. Finally, radar measurements of SWE (SWE-retrievals), if coupled with velocities derived from dry-snow densities, overestimated the mean SWE along transects by as much as 40% during the melt season, highlighting a potential issue for water managers during the melt season. Future work to support the testing of L-band radar SWE-retrievals in wet-snow should test radar signal-power attenuation methods and the capabilities of snow models for estimating LWC.
Open Access
Surface heat fluxes and MJO propagation through the Maritime Continent
(Colorado State University. Libraries, 2022) Hudson, Justin, author; Maloney, Eric, advisor; Rasmussen, Kristen, committee member; Rugenstein, Jeremy, committee member
The 'barrier effect' of the Maritime Continent (MC) is a known hurdle in understanding the propagation of the Madden-Julian Oscillation (MJO). To understand the differing dynamics of MJO events that propagate versus stall over the MC, a new MJO tracking algorithm utilizing 30-96 day filtered NOAA Interpolated OLR anomalies is presented. Using this algorithm, MJO events can be identified, tracked, and described in terms of their propagation characteristics. Latent heat flux from CYGNSS and OAFLUX as well as CYGNSS surface winds are used to compare large-scale patterns for MJO events that do and do not propagate through the MC. Local area-averaged surface fluxes and OLR anomalies are 7-14% and 18-22% of the value of precipitation anomalies, respectively. While differences in these contributions do not change substantially for propagating versus terminating events, precipitation events that successfully propagate through the MC demonstrate surface flux anomalies that are stronger and more spatially-coherent. The spatial scale of precipitation events that propagate through the MC region is also larger than terminating events. It is also shown that large-scale enhancement of latent heat fluxes near and to the east of the Dateline accompanies MJO events that successfully propagate through the MC. This large-scale enhancement of latent heat fluxes to the east of the Dateline is equally driven by dynamic and thermodynamic effects. These findings are placed in the context of recent theoretical models of the MJO in which latent heat fluxes are important for propagation and destabilization. The tracking algorithm is also used to show for historical and greenhouse gas warming scenarios in CESM2 that MJO propagation speed increases and precipitation anomalies propagate further east with warming. However, the CESM2 inadequately represents the 'barrier effect' of the MC region on propagating MJO events.
Open Access
The effects of temperature-elevation gradients on snowmelt in a high-elevation watershed
(Colorado State University. Libraries, 2022) Sears, Megan G., author; Fassnacht, Steven, advisor; Kampf, Stephanie, committee member; Rasmussen, Kristen, committee member
The majority of snowmelt in the western U.S. occurs at high elevation where hydrometeorological measurements needed for monitoring snowpack processes are often in complex terrain. Data are often extrapolated based on point measurements at lower elevation stations and the elevation to be modeled. In this study, we compute near-surface air temperature-elevation gradients and dew point temperature-elevation gradients (TEG and DTEG, respectively) and compare values to widely accepted rates (e.g., environmental lapse rate). Further, the implications on snowmelt modeling of TEG and DTEG versus accepted temperature-elevation gradients are quantified using two index snowmelt models, 1) temperature and 2) temperature and radiation. TEG and DTEG were found to be highly variable and during nighttime often influenced by cold air drainage. Several modeling scenarios were applied that manipulated air temperature and dew point temperature, via incoming longwave radiation. When compared to the control scenario, these scenarios ranged in snow-all-gone date by -1 to +6 days. The model utilizing observed air temperature and an estimated DTEG performed most similarly to the control scenario. Thus, the estimated DTEG is adequate for index snowmelt models used in similar domains; however, further investigation should be done prior to applying the environmental lapse rate or other estimated TEG values.
Open Access
Turning night into day: the creation, validation, and application of synthetic lunar reflectance values from the day-night band and infrared sensors for use with JPSS VIIRS and GOES ABI
(Colorado State University. Libraries, 2023) Pasillas, Chandra M., author; Kummerow, Christian, advisor; Bell, Michael, advisor; Miller, Steven D, committee member; Rasmussen, Kristen, committee member; Reising, Steven, committee member
Investigation of the dynamics of tropical cyclone precipitation structure using radar observations and numerical modeling Satellite remote sensing revolutionized weather forecasting and observing in the 1960s providing a true bird's eye view of the weather beyond what could be achieved from balloon and aircraft reconnaissance. With advances in observing systems came the desire for more capabilities and a better understanding of the Earth system, leading to rapid increases in satellite imaging capabilities. The most popular imager products come from solar reflective radiation in the form of visible imagery as they are the most intuitive to users. Similar benefits were later made possible by equivalent nighttime imagery; first available through the operational lines can system (OLS) and then the Day/Night Band (DNB), but these sensors have limited revisit time due to their low Earth orbits. A day-night band sensor in geostationary orbit would greatly enhance the utility of this measurement for now casters, but it does not exist. Work towards a pseudo-nighttime visible imagery to fill this gap has been done with varying results (Chirokova et al., 2018; Kim et al., 2019; Kim and Hong, 2019; Mohandoss et al., 2020; Harder et al., 2020). This dissertation demonstrates the creation and implementation of a machine learning model to turn night into day by transforming satellite radiance observations into representative full moon lunar reflectance values that provide quantifiable metrics and visible-like imagery to its users. In Chapter 2, a method is described that utilizes a feed-forward neural network model to replicate DNB lunar reflectance using brightness temperatures and brightness temperature differences in the short and long-wave infrared (IR) spectrum as the primary input. The goal was to improve upon the performance of the DNB during new moon periods, and lay the foundation for transitioning the algorithm to the Geostationary Operational Environmental Satellite (GOES) Advanced Baseline Imager (ABI). Results from this method are the first to quantitatively validate low-light visible nighttime imagery with lunar reflectance calculated from DNB radiances. This work further demonstrated that there is a relationship between full moon lunar reflectance and IR that can be captured to create imagery that is visually consistent across the full lunar cycle regardless of moon phase and angle. In Chapter 3, the machine learning (ML) nighttime visible imagery (NVI) model is applied to the GOES ABI utilizing wavelength relationships and satellite inter-calibrations information. This demonstrates that a model trained and validated on VIIRS polar orbiting imagery can work on sensors aboard geostationary satellites. It also confirms why the 10.3μm channel is the preferred substitution for the 10.7μm centered band over the 11.2μm channel. Furthermore, it demonstrates that lunar reflectance derived from IR can be replicated across sensors with similar spectral response functions providing enhanced geographic and temporal resolution that is not possible on the JPSS platforms. The final section of the dissertation transitions into forecaster applications by examining case studies concerning tropical cyclones and fog in greater detail. Focused on low cloud detection, NVI provides additional information not possible from IR and current analysis products available. It can detect tropical cyclone low-level circulations through cirrus cloud and identify fog extent more readily. The findings in this doctoral study will advance remote sensing of clouds at night, further reducing weather now-casting errors and increasing weather-related safety. In Chapter 2, a method is described that utilizes a feed-forward neural network model to replicate DNB lunar reflectance using brightness temperatures and brightness temperature differences in the short and long-wave infrared (IR) spectrum as the primary input. The goal was to improve upon the performance of the DNB during new moon periods, and lay the foundation for transitioning the algorithm to the Geostationary Operational Environmental Satellite (GOES) Advanced Baseline Imager (ABI). Results from this method are the first to quantitatively validate low-light visible nighttime imagery with lunar reflectance calculated from DNB radiances. This work further demonstrated that there is a relationship between full moon lunar reflectance and IR that can be captured to create imagery that is visually consistent across the full lunar cycle regardless of moon phase and angle. In Chapter 3, the machine learning (ML) nighttime visible imagery (NVI) model is applied to the GOES ABI utilizing wavelength relationships and satellite inter-calibrations information. This demonstrates that a model trained and validated on VIIRS polar orbiting imagery can work on sensors aboard geostationary satellites. It also confirms why the 10.3 µm channel is the preferred substitution for the 10.7 µm centered band over the 11.2 µm channel. Furthermore, it demonstrates that lunar reflectance derived from IR can be replicated across sensors with similar spectral response functions providing enhanced geographic and temporal resolution that is not possible on the JPSS platforms. The final section of the dissertation transitions into forecaster applications by examining case studies concerning tropical cyclones and fog in greater detail. Focused on low cloud detection, NVI provides additional information not possible from IR and current analysis products available. It can detect tropical cyclone low-level circulations through cirrus cloud and identify fog extent more readily. The findings in this doctoral study will advance remote sensing of clouds at night, further reducing weather now-casting errors and increasing weather-related safety.