Physical Sciences and Mathematics Commons^™

Fine Scale Mapping Of Laurentian Mixed Forest Natural Habitat Communities Using Multispectral Naip And Uav Datasets Combined With Machine Learning Methods, Parth P. Bhatt

Dissertations, Master's Theses and Master's Reports

Natural habitat communities are an important element of any forest ecosystem. Mapping and monitoring Laurentian Mixed Forest natural communities using high spatial resolution imagery is vital for management and conservation purposes. This study developed integrated spatial, spectral and Machine Learning (ML) approaches for mapping complex vegetation communities. The study utilized ultra-high and high spatial resolution National Agriculture Imagery Program (NAIP) and Unmanned Aerial Vehicle (UAV) datasets, and Digital Elevation Model (DEM). Complex natural vegetation community habitats in the Laurentian Mixed Forest of the Upper Midwest. A detailed workflow is presented to effectively process UAV imageries in a dense forest environment …

Regional Impacts Of Invasive Species And Climate Change On Black Ash Wetlands, Joseph Shannon

Dissertations, Master's Theses and Master's Reports

For more than a decade intensive research on the ecohydrology of black ash wetland ecosystems has been performed to understand these systems before they are drastically altered by the invasive species, emerald ash borer (EAB). In that time there has been little research aimed at the scale and persistence of the alterations. Three distinct but related research articles will be presented to demonstrate a method for moderate resolution mapping of black ash across its entire range, understand the relative impacts of EAB and climate change on probable future wetland conditions, and develop an experimental and modeling approach to quantify and …

Reconstruction Of The 2018 Anak Krakatau Collapse Using Planetscope Imaging And Numerical Modeling, Davide Saviano

Dissertations, Master's Theses and Master's Reports

The study was focused on the Anak Krakatau sector collapse that occurred on 22 December 2018 in the Sunda Strait (Indonesia). The goal of the study was to monitor and analyze changes of the volcanic edifice and to try to understand causes that may have predisposed and triggered the sector collapse.

The use of different remote sensing techniques allowed the acquisition of thermal data, SO₂ emission data, structural data and the identification and analysis of the eruptive events that occurred on Anak Krakatau in the period from 1° January 2016 to 28 February 2019. The acquisition of 1221 thermal …

Rain Generated Lahars Prior To The 2018 Catastrophic Eruption Of Fuego Volcano, Guatemala, Claudia Buondonno

Dissertations, Master's Theses and Master's Reports

Fuego volcano is one of the most active and hazardous volcanoes in the world. It is located in the northern part of the Central American Volcanic Arc in Guatemala and its activity can be characterized by long term, low-level background activity, and sporadic larger explosive eruptions. Its historical observations of eruptions date back to 1531, but it has been erupting vigorously since 2002 with major activity throughout 2018, producing three main eruptions in February, June and November. Its almost persistent activity generates major ashfalls, pyroclastic flows, lava flows; when heavy rains mobilize its deposits, they can form damaging lahars. Phenomena, …

Assessing The Accuracy Of A Uav Snow Depth Survey: Utqiagvik (Barrow), Alaska Calm Grid, Ian Nichols

Dissertations, Master's Theses and Master's Reports

Active layer depth and snow depth are annually collected across the Circumpolar Active Layer Monitoring (CALM) Network to observe the response of the active layer and near-surface permafrost to climate change over decadal-time scales. Snow depth is typically measured using a graduated steel probe at each grid node but, in this paper, we explore the viability of using Unmanned Aerial Vehicle (UAV) (drone) technology to collect snow depth measurements at the 1 km² Utqiagvik (Barrow), Alaska CALM grid. This is achieved by comparing estimated UAV snow depths to measured snow depths collected using a MagnaProbe (MP) at each of …

Application Of A Distributed Hydrologic Model For The Analysis Of Land Use Change In Kedougou, Senegal, Francisco Perez

Dissertations, Master's Theses and Master's Reports

West Africa has experienced extensive land use/land cover (LULC) change since the Sahel drought of the 1970s and 80s. Demographic changes, rapid urbanization, and other anthropogenic impacts on land cover in West Africa have potentially altered hydrological regimes. This work was undertaken while I served as a Peace Corps Volunteer in the Kedougou region of Senegal, an area that has experienced significant and rapid infrastructural improvements and urbanization since the early 2000s. The Gambia River flows near Kedougou, where flooding is an extensive problem, making access and livelihood practices difficult during the rainy season. Changes in climate and land use …

Testing The Use Of Vegetation Indices As Indicators Of Pre-Eruptive Volcanic Unrest At Kilauea Volcano, Hawaii, Quelyn Bekkering

Dissertations, Master's Theses and Master's Reports

Satellite remote sensing has become an integral part of pre-eruptive and long-term volcano monitoring due to its low cost, spatial and temporal coverage, and time-series analysis capabilities. Pre-eruptive unrest at volcanoes can include increased fluxes of heat and volcanic gases, which can be manifested either directly (e.g., the appearance of thermal anomalies or fumaroles) or indirectly (e.g., through impacts on vegetation health and extent). We are evaluating the use of novel, high spatial and temporal resolution data from the Planet Labs cubesat constellation to monitor changes in vegetation related to volcanic unrest. We present results from a study of key …

Application Of Remote Sensing And Machine Learning Modeling To Post-Wildfire Debris Flow Risks, Priscilla Addison

Dissertations, Master's Theses and Master's Reports

Historically, post-fire debris flows (DFs) have been mostly more deadly than the fires that preceded them. Fires can transform a location that had no history of DFs to one that is primed for it. Studies have found that the higher the severity of the fire, the higher the probability of DF occurrence. Due to high fatalities associated with these events, several statistical models have been developed for use as emergency decision support tools. These previous models used linear modeling approaches that produced subpar results. Our study therefore investigated the application of nonlinear machine learning modeling as an alternative. Existing models …