Scholarship & Research

Permanent URI for this communityhttps://scholarworks.montana.edu/handle/1/1

Browse

Search Results

Now showing 1 - 10 of 30

A framework for the quantitative assessment of new data streams in avalanche forecasting
(Montana State University - Bozeman, College of Letters & Science, 2023) Haddad, Alexander Sean; Co-chairs, Graduate Committee: Eric A. Sproles and Jordy Hendrikx
Data used by avalanche forecasters are typically collected using weather stations, manual field-based observations (e.g., avalanche events, snow profiles, stability tests, personal observations, public observations, etc.) and weather forecasts ("traditional observations"). Today, snow cover observations can be delivered via remote sensing (e.g., satellite data, UAV, TLS, time-lapse camera etc.). Forecasting operations can also use statistical forecasting, weather models, and physical modeling to support decisions. This paper presents a framework and methodology to quantify the impact these new, complex data streams have on the formulation of, and associated uncertainty of, avalanche forecasting. We use data from a case study in Norway. Avalanche forecasters in Norway assessed size (D), likelihood, avalanche problem, and hazard level for a highway corridor in Grasdalen, Stryn Norway. The control groups were given access to traditional observations. The experimental groups were given access to the same traditional data, but also near-real-time snow surface LiDAR data ("RS+"). In case study one the RS+ (n=10) consensus findings were a hazard level two steps lower than the control group (n=10). In case study two the traditional (n=10) and RS+ groups' (n=7) consensus findings assessed the northeastern avalanche path at the same hazard level. Assessing the southwestern slide path, the traditional group (n=10) and RS+ group (n=9) had the same consensus finding for hazard level. In 2 of 3 case studies, the RS+ groups had fewer selections for size, likelihood, and avalanche problem which indicates reduced uncertainty in their forecasts. Throughout the 2022-2023 winter season Norwegian Public Roads Administration avalanche forecasters performed a real-time experiment throughout the season - with and without additional RS+ data when forecasting. They agreed on hazard level in 6 of 10 forecasts. In the other 4 forecasts, RS+ forecasters assessed the hazard level higher than traditional data forecasts. When RS+ data reveals aspects of conditions that traditional observations did not detail, RS+ forecasters adjust their selections in the hazard matrix, resulting in greater clustering of their predictions, indicating reduced uncertainty. Due to uncertainty associated with avalanche forecasting, this framework for assessment should be used to track avalanche forecast efficacy and build a qualitative and quantitative historical record.
Meteorological controls on wind slab properties
(Montana State University - Bozeman, College of Letters & Science, 2023) de Leeuw, Nathalie Marika; Chairperson, Graduate Committee: Jordy Hendrikx
Snow avalanches are dangerous phenomena, which can be made increasingly consequential through wind transport of snow. Wind-deposited snow has a broad range of possible physical and mechanical properties which can vary greatly over short distances, creating inconsistent and thus difficult avalanche conditions. This variability causes particular challenges for avalanche workers in data-sparse regions where important snowpack information at desired scales may be unavailable. Instead, snowpack properties are commonly inferred from available meteorological data. Though wind slab properties vary in space and time as meteorological conditions change, previous work has not explicitly studied these relationships at the slope-scale. In this research I aim to better understand how changes in meteorological variables relate to changes in wind slab physical properties. During two winters I recorded temperature, humidity, and wind speed at study sites in Montana's Madison Range (45.237, -111.424) and collected snowpack data during or immediately following blowing snow events. I found that average wind speeds at 0.5m and 1.5m above the snow surface were significantly higher during hard wind slab formation than soft wind slab formation, while unobstructed wind speed, maximum gust, and the length of time of wind transport were not associated with wind slab hardness. Temperature was higher during hard than soft wind slab formation, while humidity was not different between the two hardness categories. Although wind speed at 1.5m had a significant positive linear relationship with both wind slab density and blade hardness gauge force, it was a poor predictor of actual values for both of these parameters. Our findings help improve the understanding of the impact of near surface winds on wind slabs, which will aid avalanche forecasting and mitigation planning particularly in windy climates.
Novel models and observations of energetic events in the solar transition region
(Montana State University - Bozeman, College of Letters & Science, 2021) Parker, Jacob Douglas; Chairperson, Graduate Committee: Charles C. Kankelborg; Dana Longcope was a co-author of the article, 'Modeling a propagating sawtooth flare ribbon as a tearing mode in the presence of velocity shear' in the journal 'Astrophysical journal' which is contained within this dissertation.; Charles Kankelborg was a co-author of the article, 'Determining the spectral content of MOSES images' submitted to the journal 'Astrophysical journal' which is contained within this dissertation.; Roy Smart, Charles Kankelborg, Amy Winebarger and Nelson Goldsworth were co-authors of the article, 'First flight of the EUV snapshot imaging spectrograph (ESIS)' submitted to the journal 'Astrophysical journal' which is contained within this dissertation.
The solar atmosphere is an energetic and violent place capable of producing eruptions that affect us on earth. In order to better understand these events, so that we might improve out ability to model and predict them, we observe the sun from space to diagnose the local plasma conditions and track its evolution. The transition region, a thin region of the solar atmosphere separating the chromosphere from the corona, is where the solar atmosphere transitions rapidly from ten thousand, to one million kelvin and is therefore thought to play an important roll in the transfer of mass and energy to the hot corona. The sun's magnetic field, and magnetic reconnection, are thought to contribute to the increased temperature of the corona, since the cooler lower solar atmosphere cannot heat it via thermal conduction or convection. Explosive events, small solar eruptions likely driven by magnetic reconnection, are frequent in the transition region, making it an attractive area of the atmosphere to study and gather information on the processes. Using Computed Tomography Imaging Spectrographs (CTIS), capable of measuring spectral line profiles over a wide fields of view at every exposure, we find many eruptive events in the transition region to be spatially complex, three dimensional, and to evolve on rapid timescales. This demonstrates the utility of, and need to continue developing, CTIS style instruments for solar study since they provide a more complete picture of solar events, allowing us to improve our understanding of our closest star.
Spatio-temporal analysis of large magnitude avalanches using dendrochronology
(Montana State University - Bozeman, College of Letters & Science, 2020) Peitzsch, Erich Hans; Chairperson, Graduate Committee: Jordy Hendrikx; Jordy Hendrikx, Daniel K. Stahle, Gregory T. Pederson, Karl W. Birkeland and Daniel B. Fagre were co-authors of the article, 'A regional spatio-temporal analysis of large magnitude snow avalanches using tree rings' submitted to the journal 'Natural hazards and Earth systems sciences' which is contained within this dissertation.; Gregory T. Pederson, Jordy Hendrikx, Karl W. Birkeland and Daniel B. Fagre were co-authors of the article, 'Trends in regional large magnitude snow avalanche occurrence and associated climate patterns in the U.S. northern Rocky Mountains' submitted to the journal 'Journal of climate' which is contained within this dissertation.; Chelsea Martin-Mikle, Jordy Hendrikx, Gregory T. Pederson, Karl W. Birkeland and Daniel B. Fagre were co-authors of the article, 'Vegetation characterization in avalanche paths using LIDAR and satellite imagery' submitted to the journal 'Arctic, antarctic, and alpine research' which is contained within this dissertation.
Snow avalanches are a natural hazard to humans and infrastructure as well as an important landscape disturbance affecting mountain ecosystems. In many mountainous regions, records of avalanche frequency and magnitude are sparse or non-existent. Inferring historic avalanche patterns to improve forecasting and understanding requires the use of dendrochronological methods. In this dissertation, we examine a regional tree-ring derived large magnitude avalanche dataset from northwest Montana in the northern Rocky Mountains, USA, to produce avalanche chronologies at three distinct scales (path, sub-region, and region), assess seasonal climate drivers of years with large magnitude avalanche occurrence on a regional scale, and characterize vegetation in select avalanche paths. By implementing a strategic spatial sampling design and collecting a large dataset of tree-ring samples, we: (1) assessed scaling in the context of a regional avalanche chronology, reconstructed avalanche chronologies for 12 avalanche paths in four subregions, and examined the effects of two methods of sampling indexing on the resultant avalanche chronology; (2) identified specific climate drivers of large magnitude avalanche years across a region and identified trends in avalanche year probability through time; and (3) tested the feasibility of using remote sensing products to quantify vegetation types in avalanche paths and characterized the vegetation composition based on return periods within specific avalanche paths. This dissertation is organized into 3 key chapters/manuscripts (Chapters 2, 3, and 4) and two supporting chapters (Chapters 1 and 5) that address the problem of assessing large magnitude avalanche frequency at various spatio-temporal scales using a tree-ring dataset. The results contribute toward a better understanding of reconstructing regional avalanche chronologies, a more accurate assessment of avalanche-climate relationships, and improved methods to characterize vegetation characteristics within avalanche path return periods. This work has applications for regions with sparse avalanche records.
Extracting abstract spatio-temporal features of weather phenomena for autoencoder transfer learning
(Montana State University - Bozeman, College of Engineering, 2020) McAllister, Richard Arthur; Chairperson, Graduate Committee: John Sheppard
In this dissertation we develop ways to discover encodings within autoencoders that can be used to exchange information among neural network models. We begin by verifying that autoencoders can be used to make predictions in the meteorological domain, specifically for wind vector determination. We use unsupervised pre-training of stacked autoencoders to construct multilayer perceptrons to accomplish this task. We then discuss the role of our approach as an important step in positioning Empirical Weather Prediction as a viable alternative to Numerical Weather Prediction. We continue by exploring the spatial extensibility of the previously developed models, observing that different areas in the atmosphere may be influenced unique forces. We use stacked autoencoders to generalize across an area of the atmosphere, expanding the application of networks trained in one area to the surrounding areas. As a prelude to exploring transfer learning, we demonstrate that a stacked autoencoder is capable of capturing knowledge universal to these dataspaces. Following this we observe that in extremely large dataspaces, a single neural network covering that space may not be effective, and generating large numbers of deep neural networks is not feasible. Using functional data analysis and spatial statistics we analyze deep networks trained from stacked autoencoders in a spatiotemporal application area to determine the extent to which knowledge can be transferred to similar regions. Our results indicate high likelihood that spatial correlation can be exploited if it can be identified prior to training. We then observe that artificial neural networks, being essentially black-box processes, would benefit by having effective methods for preserving knowledge for successive generations of training. We develop an approach to preserving knowledge encoded in the hidden layers of several ANN's and collect this knowledge in networks that more effectively make predictions over subdivisions of the entire dataspace. We show that this method has an accuracy advantage over the single-network approach. We extend the previously developed methodology, adding a non-parametric method for determining transferrable encoded knowledge. We also analyze new datasets, focusing on the ability for models trained in this fashion to be transferred to operating on other storms.
Data analytics and software to support avalanche forecasting decisions
(Montana State University - Bozeman, College of Engineering, 2021) Ottsen, Peter Kenneth; Chairperson, Graduate Committee: Sean Yaw
Avalanches are a very powerful force of nature and pose significant risk for ski areas and mountainous roads. Avalanche forecasting and mitigation are a very important part of keeping the public safe. Terrestrial laser scanning lidar systems have proven useful in more accurate forecasting and mitigation efforts, but utilizing them can be time consuming. The goal of this project is to operationalize a workflow and create algorithms and ultimately produce a software product that can rapidly analyze snow covered mountainous terrain, allowing avalanche forecasters to make informed decisions on where to focus their mitigation efforts. In this dissertation, I first present algorithms that were designed to align scans, identify trees and cliffs, grid scans, and calculate snow depth. I then introduce a software package that was implemented incorporating these algorithms with a point cloud visualization tool. This software package allows a user to control and visualize the analysis process to make more informed avalanche mitigation decisions. Algorithms were parameterized and validated with a field study consisting of data collection events at Bridger Bowl, Bear Canyon, and the Yellowstone Club in Montana. A Riegl VZ-6000 TLS lidar system was used for all data collection efforts. This dissertation documents the design of this analytics workflow by presenting the algorithms developed, discussing the software implemented, and presenting the data collection efforts that guided the design of the algorithms and served to validate their efficacy.
Snow avalanche identification using Sentinel-1: detection rates and controlling factors
(Montana State University - Bozeman, College of Letters & Science, 2021) Keskinen, Zachary Marshall; Chairperson, Graduate Committee: Jordy Hendrikx; Jordy Hendrikx, Karl Birkeland and Markus Eckerstorfer were co-authors of the article, 'Snow avalanche identification using Sentinel-1 backscatter imagery: detection rates and controlling factors' submitted to the journal 'Natural hazards and Earth system sciences' which is contained within this thesis.
Snow avalanches present a significant hazard that endangers lives and infrastructure. Consistent and accurate datasets of avalanche events is valuable for improving forecasting ability and furthering knowledge of avalanches' spatial and temporal patterns. Remote sensing-based techniques of identifying avalanche debris allow for continuous and spatially consistent datasets of avalanches to be acquired. This study utilizes expert manual interpretations of Sentinel-1 synthetic aperture radar (SAR) satellite backscatter images to identify avalanche debris and compares those detections against historical field records of avalanches in the transitional snow climates of Wyoming and Utah. This study explores the utility of Sentinel-1 (a SAR satellite) images to detect avalanche debris on primarily dry slab avalanches. The overall probability of detection (POD) rate for avalanches large enough to destroy trees or bury a car (i.e., D3 on the Destructive Size Scale) was 64.6%. There was a significant variance in the POD among the 13 individual SAR image pairs (15.4 - 87.0%). Additionally, this study investigated the connection between successful avalanche detections and SAR-specific, topographic, and avalanche type variables. The most correlated variables with higher detection rates were avalanche path lengths, destructive size of the avalanche, incidence angles for the incoming microwaves, slope angle, and elapsed time between the avalanche and a Sentinel-1 satellite passing over. This study provides an initial exploration of the controlling variables in the likelihood of detecting avalanches using Sentinel-1 backscatter change detection techniques. This study also supports the generalizability of SAR backscatter difference analysis by applying the methodology in different regions with distinct snow climates from previous studies.
Investigation of crack arrest fracture toughness of laboratory-manufactured polycrystalline ice
(Montana State University - Bozeman, College of Engineering, 2021) Alcorn, Derek West; Chairperson, Graduate Committee: Edward E. Adams
Approximately 50% of ice mass loss from ice sheets is due to icebergs breaking off in a process called calving. Icebergs are created through the incremental growth of crevasses, which are large fractures in the ice. Crevasse propagation and iceberg calving predictions within ice sheet models conflict with direct observations of crevasse processes. Current ice sheet models assume that a crevasse will propagate until it reaches a depth where the stress intensity factor at the crack tip is less than that of crack initiation, however, this is likely an oversimplification as current models over estimate crevasse depth. A more robust model would also account for the crack arrest fracture toughness, a measure of how well a material can stop an already propagating crack. Here, we calculate crack arrest fracture toughness for samples of laboratory-manufactured polycrystalline ice. These samples were created using a radial freezing technique with a reproducible grain size distribution of 0.95 mm + or - 0.28 mm analyzed by cross-polarized light. Specimens were notched and brought to failure via a short-rod fracture toughness test at controlled temperatures and a constant displacement rate in a commercial mechanical testing apparatus with an environmental chamber. The presented data agrees with short-rod fracture toughness data collected from ice cores at the Filchner- Ronne Ice Shelf in Antarctica, demonstrating quasi-stable crack growth behavior. Results show the crack arrest fracture toughness of laboratory-manufactured polycrystalline ice is approximately 25 - 50% of fracture toughness. Using the crack arrest fracture toughness determined in this study would further increase modeled crevasse depth, indicating more analysis is required. Future studies can incorporate these data to more accurately determine crevasse penetration depth and improve iceberg calving predictions within ice sheet models.
An investigation modeling risk of wildlife-vehicle collisions in Montana, USA
(Montana State University - Bozeman, College of Engineering, 2019) Bell, Matthew Andrew; Chairperson, Graduate Committee: Yiyi Wang and Damon Fick (co-chair)
Road ecologists and transportation engineers have been exploring new methods to adapt to the environmental and motorist safety concerns involving wildlife-vehicle collisions. There are over one-million crashes with large-bodied wildlife every year in the U.S. that result in substantial property damage and personal injuries. Recent studies modeling these collisions identify where they cluster, and the landscape, road, and driver characteristics that influence the likelihood of a collision along short road corridors and small geographic regions. This research expands on current knowledge and attempts to model the risk of wildlife-vehicle collisions on a large geographic scale. This research investigates different analysis methods and creates predictive models that will estimate the risk of a wildlife-vehicle collision as drivers travel across multiple ecosystems. Different analysis units were created to extract two similar datasets that are modeled against two different response variables -- reported collisions and roadkill locations. Regularization is used to help with feature selection. Negative binomial regression models are built to predict risk. Random forest machine learning helps better understand the percent of variance explained by the variables in each model. A range of statistical measurements were taken to compare the non-nested models. The best performing model is applied to the seasonal division of data. Yearlong and seasonal risk is mapped onto the road network and color-coded to show the differences in risk on Montana's road network. The maps capture the changes in risk throughout the year, they generally match where wildlife-vehicle collisions actually happen, and even coincides with published work on the locations of collision hotspots in Montana. This research is the basis for future complex real-time risk-mapping models that can be integrated into smart technology and developed into on-board driver alert systems. With the advancements of autonomous vehicle, it is possible to incorporate real-time driving data into models that will analyze wildlife-vehicle collision risk based on vehicle location, season, time of day and driving habits. This can increase driver safety by informing them when they are traveling in areas where wildlife-vehicle collisions are more likely to happen, and can be especially helpful while driving on unfamiliar roads.
Atmospheric processes related to deep persistent slab avalanches in the western United States
(Montana State University - Bozeman, College of Letters & Science, 2019) Schauer, Andrew Robert; Chairperson, Graduate Committee: Jordy Hendrikx
Deep persistent slab avalanches are a natural hazard that are particularly difficult to predict. These avalanches are capable of destroying infrastructure in mountain settings, and are generally unsurvivable by humans. Deep persistent slab avalanches are characterized by a thick (> 1 m) slab of cohesive snow overlaying a weak layer in the snowpack, which can fail due to overburden stress of the slab itself or to external triggers such as falling cornices, explosives, or a human. While formation of such snowpack structure is controlled by persistent weather patterns early in the winter, a snowpack exhibiting characteristics capable of producing a deep persistent slab avalanche may exist for weeks or months before a specific weather event such as a heavy precipitation or rapid warming pushes the weak layer to its breaking point. Mountain weather patterns are highly variable down to the local scale (1-10 m), but they are largely driven by atmospheric processes on the continental scale (1000 km). This work relates atmospheric circulation to deep persistent slab events at Mammoth, CA; Bridger Bowl, MT; and Jackson, WY. We classify 5,899 daily 500 millibar geopotential height maps into 20 synoptic types using Self-Organizing Maps. At each location, we examine the frequency of occurrence of each of the 20 types during November through January during major deep persistent slab seasons and compare those frequencies to seasons without deep persistent slab avalanches. We also consider the 72-hour time period preceding deep persistent slab avalanches at each location and identify synoptic types occurring frequently, as well as those rarely occurring prior to onset of activity. At each location, we find specific synoptic types that tend to occur at a higher rate during major deep persistent slab years, while minor years are characterized by different circulation patterns. We also find a small number of synoptic types dominating the 72-hour period prior to onset of deep slab activity. With this improved understanding of the atmospheric processes preceding deep persistent slab avalanches, we provide avalanche practitioners with an additional tool to better anticipate a difficult to predict natural hazard.

Scholarship & Research

Browse

Filters

Settings

Sort By

Results per page

Search Results