Session U22: Geology, Atmospheric Sciences, Other

The Influence of Continental Geometry and Land Surface Properties on Temperature Variability

The highest temperature variability on Earth is over central North America, with twice as much variance in the daily-mean temperatures as Eurasia. Researching temperature variability is important because of its implications on society now and because of its role in understanding climate change and future climates. We research the effects of continental geometry and land surface properties on this temperature variability. In this study, we use the idealized ISCA model to run simulations with varied continental shapes, latitudes, and topographies. We probe the simulations to see how these factors affect the variance of near-surface temperature. Applying these findings to the shapes, locations, and topographies of North America and Eurasia could help further explain the large difference of temperature variability among the two continents.   

Obtaining Accurate Aerosol Extinction and Backscatter Coefficients Using Ceilometers

How different types of aerosols in our atmosphere explicitly affect quality and visibility on any given day is not quite clear. The aerosol extinction coefficient - $\alpha $ - and backscatter coefficient - $\beta $ -- can indicate these characteristics. Ceilometers can measure the backscatter power due to aerosols, but when diffuse sunlight reaches the ceilometer's sensors, it can alter the retrieval signal. Using this data leads to less accurate calculations. To obtain accurate calculations, a smoothing function was created to alleviate background noise. The results of the function were compared to Savitzky-Golay filtering of the backscatter power; the respective method's gradients showed the generated function produced more distinct atmospheric layers. With this new smoothing method, it is hypothesized that more accurate values for $\beta $ and $\alpha $ can be evaluated with the LIDAR Equation. The proposed process uses ceilometer data to calculate $\beta $ and the LIDAR ratio, which together can yield $\alpha $. The accuracy of the calculated $\alpha $, and thus the effectiveness of the smoothing function, can be tested by comparing the Aerosol Optical Depth (AOD), derived from $\alpha $, to sun photometer data AOD measurements.   

Gravity Wave Study in the MLT using ICON-MIGHTI Temperature and Wind Observations

Atmospheric gravity waves have important roles in driving atmospheric coupling processes from the lower atmosphere to the mesosphere, thermosphere, and ionosphere. NASA’s Ionospheric Connection Explorer (ICON) satellite was launched on 10 October 2019 and has been observing atmospheric temperatures and winds in the latitude range of 10S-40N. Michelson Interferometer for Global High-resolution Thermospheric Imaging (MIGHTI) is one of the instruments on ICON measuring temperatures and winds. Using both temperature and wind measurements from ICON-MIGHTI observations, small-scale perturbations (< wavenumber 6) are extracted and analyzed in the altitude range of 90-105 km and are considered to be gravity waves in this work. Obtained gravity waves will be compared to other satellite observations including TIMED/SABER.   

Rhodes College CubeSat Program: RHOK-SAT

The Rhode College CubeSat program (RHOK-SAT) is a partnership with the University of Oklahoma which is currently developing a 1U CubeSat. The satellite has an intended launch date of June 2023 and will fly in a Low Earth Orbit (LEO). RHOK-SAT will carry and space test novel photovoltaic cells being developed by the University of Oklahoma's Photovoltaic Materials and Devices group. This lightning talk will provide a general overview of the project including a look at the work that has occurred up to this point as well as some of the next steps. It will focus especially on the development of a ground station on the Rhodes campus, which will connect to the satellite to receive data from an onboard computer. The ground station includes a radio and transceiver which is prepared to downlink data from the satellite as it passes over. This talk will also briefly cover some of the on board hardware, like a computer or AMU chip which will help the satellite communicate with the ground station.   

Novel Photovoltaic Cell Characterization on a 1U CubeSat, RHOK-SAT

Rhodes College is leaving its footprint in space with a nanosatellite (a CubeSat), projected to be ready for launch in June of 2023. The payload's mission of the satellite is to characterize novel photovoltaic cells (PV) in low earth orbit. These cells, developed by the Photovoltaics Materials and Devices Group at the University of Oklahoma, show promise for providing remote power generation for future crewed and uncrewed space missions. The three types of cells are; (1) state-of-the-art flexible copper indium gallium selenide cells with a silver absorption layer (ACIGS), (2) gallium arsenide antimonide (GaAsSb) cells and (3), state of the art perovskite cells. These specific types of cells have not been tested in long duration space environments, where thermal cycling and radiation effects can effect cell performance. While in orbit, the experiment will perform current-voltage sweeps simultaneously across all test cells. This data will be stored on the on board computer (OBC), until it is downlinked to our groundstation in Memphis, TN. Graphical analysis of these current-voltage curves will provide insight into their respective space-hardiness for future space use, as in deep space missions and prolonged human presence on the lunar surface.   

Timeseries Analysis of Seasonal Variations of Pacific Arctic Sea Ice-Cloud Cover Feedback

It is predicted that by the year 2050, Arctic summers will be sea ice free. The complexity of the Earth's climate system and its feedback entails that the effects of melting of sea ice has coupled effects in the hydroclimate, especially for cloud cover. In my research, I have created a time-lapse of the sea ice and cloud cover in the same region to understand the extent of their relationship in the Arctic. An international consortium of scientists has implemented the Distributed Biological Observatory (DBO) which is a change detection array for the identification and consistent monitoring of biophysical responses to environmental change in the Pacific Arctic Region. The data for my project has been collected from the DBO sites. Sea Ice concentrations and Sea surface temperatures have been accumulated from their archives. The cloud fraction has been compiled from the NASA's MODIS AQUA satellite. This has been analyzed by using the Earth Trend's modeler using TerrSet software. By mapping annual trends, we understand whether climate change is causing a positive or negative feedback impact on the sea ice-cloud cover relationship. My findings show a negative feedback, that due to the decreasing sea ice there is increasing cloud cover. Since cloud cover is one regulator in minimizing temperatures, these feedbacks are crucial to understand for accurately modeling climate change and would help guide analysis in the future. My research is aimed at bridging the gap that exists in current research in identifying the extent of cloud cover in the Arctic Climate and the degree of its impact that inevitable sea ice disappearance is going to have.   

Molecular Modeling of a Bijel

Bicontinuous Interfacially Jammed Emulsion Gels, bijels, are emulsions of two immiscible liquids ``jammed'' into a network created by colloids. Bijels could have the capability to improve energy conversion, catalytic reactions, and electrical conductivity and given its potential we aim to understand the thermodynamic conditions for obtaining a Bijel. This involved initially tuning the binary liquid interactions. The competition between vapor-liquid phase separation and liquid-liquid phase separation (LLPS) among the three components complicates the system. To entirely suppress vapor-liquid equilibrium, the well-depth between the liquids is set to 1. Colloids were then incorporated into the system by introducing three parameters: $\varphi _{\mathrm{C}}$ which represents the colloidal volume, V that sets the strength of the colloidal attractions, and $\kappa $ which sets the range of attraction. Finally, the interaction between the three components was chosen to be neutral. $\kappa $ was set at 30 for the entire study while V and $\varphi_{\mathrm{C}}$ varied. The mean-squared displacement and the radial distribution function were plotted and compared to visual snapshots to classify the behavior of the material. The results of these simulations helped conclude that $\varepsilon =$1 is better at suppressing void formation and the degree of separation between the liquids decreases with increasing colloidal volume fraction. There's also evidence that bijels can form or LLPS can only occur when the colloids form a gel that creates large voids for phase separation. In relation to the overall goal, the thermodynamic parameters that provided the ``best'' route for a Bijel was $\varepsilon =$1, $\kappa =$30, V$=$5, and $\varphi _{\mathrm{C}}=$0.15.   

A Statistical Physics Description of Glacier Calving Behavior in Ice-Shelf Evolution

Ice-shelves provide buttressing forces for massive ice sheets, preventing land-based glacier ice from entering the ocean. The collapse of an ice-shelf due to calving could cause additional ice to enter the oceans and contribute to global sea-level rise, motivating a need for estimates of the likelihood of these types of events. However, significant challenges persist in incorporating the dynamics of calving glaciers into large-scale ice sheet simulations. In this study we introduce passive tracers with fiber bundles of random strengths into a one-dimensional shallow-shelf model to simulate the strains and stresses which lead to discrete fracturing and calving behaviors. These tracers advect along with the ice until the fiber bundles reach a critical elongation based on randomly distributed strengths, then break, initiating a calving event in the shelf. We will discuss preliminary results from this model that reproduce the style of calving at Erebus ice-tongue in Antarctica, including reproducing and quantifying the average number of calving events for Erebus. We propose that this model will be successful in estimating average long-term ice-shelf calving events, as well as determining whether there is a condition under which normal calving becomes a catastrophic collapse.   

Modeling discrimination in societies of neural networks

This ongoing project deals with the quantitative study of polarization and formation of groups holding opposite opinions on a set of issues in the context of agent based models where the agents are neural network classifiers. The agents exchange binary opinions on a set of multidimensional issues. In this case, polarization is driven by adaptive affective distrust and the inclusion of irrelevant features to the problem being discussed. We consider the case where some agents extend the correctly parsed assertion with a set of numbers that are irrelevant to the classification problem, but depend only on the emitter agent. This irrelevant addition acts like disrupting noise, driving agents to effectively learn from a group of similar agents and to unlearn from other agents. We use the Entropic Dynamics learning algorithm for neural networks (EDNNA) which has been extended to model societies where the agents are perceptrons. At a given time of a discrete dynamics a pair of agents is chosen at random, one acts as the emitter and the other as the receiver of a pair (input vector - label) information. The dynamics is performed by the update of the weights and the distrust of the receiver towards the emitter. We present some preliminary results from simulations.