Grad Research Day

April 09, 2019

Grad Research Day in this year will be Tuesday, April 9th, 9:00am - 12:00pm in RCB 6152.  All 2nd year Masters students and some PhD students will be presenting their research. This is a great opportunity to find out what kind of exciting research our students are up to, and to provide helpful feedback. For more details on the presentations please visit our YouTube page.

  • Date: Tuesday, April 9, 2019
  • Time: 9:00 am - 12:00 pm
  • Room: RCB 6152

Here's the brief description for each presentation.

9:15am  Liam Fox
Program: MA
Supervisor: Geoff Mann

“Regulating in the “public interest”: The National Energy Board, pipelines, and the problem of sovereign legitimacy”

Perhaps the most visible and pressing pipeline conflict in Canadian history, the proposed Trans Mountain Pipeline expansion (TMX) has yet to see shovels break ground as the project is bound up in a web of legal challenges and political controversy. At the centre of the debate is the National Energy Board (NEB) – Canada’s federal energy regulator, responsible for regulating international and interprovincial pipelines. Through its mandated environmental and socio-economic assessments, the NEB has the de facto “go/no-go” say on proposed project. This determination hinges on the NEB’s decision if the project is in the “public interest” of Canadians. Despite that billion-dollar projects hinge on this “public interest” mandate, the nebulous term is never explicitly defined in the NEB’s operating statutory framework, or anywhere, for that matter. Instead, these large-scale infrastructure projects are justified by an appeal to the abstract “common good,” bolstered by expert knowledge, law, and emotionally-charged framings of pipelines as nation-building projects. In this paper, I argue that the NEB, through its public interest mandate, operates as an institutional fix to produce political legitimacy for state sovereignty. In its pursuit of resource-based economic expansion, the Canadian state faces what Jürgen Habermas calls a legitimation crisis: if the state cannot secure legitimacy from civil society for its management of the national political economy, then its claim to sovereignty comes into question.

9:30am Taylor Anderson
Program: PhD
Supervisor: Suzana Dragicevic

“Geographic network automata for simulating complex spatial systems”

Geographical phenomena including ecological systems can be represented as spatial networks where nodes represent the location of system components and links represent spatial interactions between them. Traditional spatial network analysis seeks to characterize a network’s structure at one or several points in time to better understand network dynamics. However, this analysis is unable to capture the spatio-temporal complexity in these systems whereby the network structure evolves as a function of network dynamics. Therefore, this study proposes the development of a novel modelling framework, Geographic Network Automata (GNA), for representing and analyzing complex spatial systems as evolving networks. The GNA modelling approach is used to simulate the dispersal of the emerald ash borer (EAB) as it spreads across a network of forest stands in Michigan, US. Results indicate that based on the connectivity of the landscape, EAB dispersal is highly robust to localized eradication measures, providing valuable insight for decision-makers.

9:45am Shuoge Shen
Program: MSc
Supervisor: Suzana Dragicevic

“Suitability analysis of high-rise building development in Metro Vancouver Region: The LSP method”

High density urban development can alleviate the impact of the urban sprawl by following urban densification as a more sustainable practice. Therefore, it is necessary to find effective methods of suitability analysis for identifying the best location for high density urban developments. The main objective of this research study is to propose and implement the GIS-based Logic Scoring of Preference (LSP) method to evaluate suitable areas for urban densification. Geospatial data for Metro Vancouver Region has been used within ArcGIS software environment to implement the LSP method. The perspectives of two different stakeholders – urban developer and urban planner have been considered. The obtained results indicate that the GIS-based LSP method can represent the human reasoning and decision-making process, making it a suitable methodological tool for identification of locations for high density urban developments in the study area.

10:00am Nesha Wright
Program: MSc
Supervisor: Kirsten Zickfeld

“Reversibility of Arctic and Sub-Arctic Permafrost Thaw in Temperature Overshoot Scenarios”

Achieving the Paris Agreement target of remaining well below a 2°C atmospheric warming and pursuing efforts to limit warming to 1.5°C is increasingly challenging given current global fossil fuel consumption and insufficient pledges to reduce greenhouse gas emissions in the future.

Overshoot trajectories, which assume that the temperature target is restored after temporarily exceeding it, are increasingly prominent in policy discussions. This study explores the long-term response of Arctic and sub-Arctic permafrost spatial extent in temperature overshoot scenarios used for the 6th Phase of the Coupled Model Intercomparison Project and employed the University of Victoria Earth System Climate Model. Preliminary results suggest that the permafrost spatial extent across northern high latitudes in overshoot scenarios partially recovers under declining atmospheric concentrations of greenhouse gases and the subsequent atmospheric cooling associated with overshoot trajectories. However, the recovery of permafrost spatial extent lags the recovery in surface air temperature by several decades.

10:15am Claude-Michel Nzotungicimpaye
Program: PhD
Supervisor: Kirsten Zickfeld

“Modelling wetland methane emissions and the permafrost carbon feedback on the global climate”

Concerns have been raised that permafrost thawing across the northern high-latitude regions will expose substantial quantities of organic carbon to decomposition, releasing carbon dioxide (CO2) and methane (CH4) to the atmosphere, and accelerating the rate of future global warming. While a few Earth system models now incorporate the permafrost-CO2 feedback, no fully coupled models represent the contribution from CH4 to this feedback. We present a wetland CH4 emission model developed for and implemented into an Earth system model of intermediate complexity. Despite being a relatively simple model, results show that the wetland CH4 emission model is able to reproduce methanogenesis rates in northern wetlands as well as global annual CH4 emissions consistent with present-day estimates. The ultimate goal of this research is to quantify CH4 emissions associated with the projected thawing of permafrost in northern wetlands and the subsequent feedback on the global mean surface temperature.

10:30am  Michael Curran
Program: MSc
Supervisor: Jeremy Venditti

“Structural Control of Bedrock Canyon Morphology”

Bedrock canyons are the loci of incision in unglaciated terrain and link the processes of climate, tectonics, and landscape evolution. There are a variety of models that relate channel geometry to bedrock erosion rates, but little is known about the influence of geologic structure on these processes. This study aims to broaden understanding of bedrock canyon morphology in two ways. 1) By investigating the control of regional and reach-scale structure on the spatial distribution of canyons through a watershed. 2) Employing kinematic analyses of canyon walls to explore the links between joint orientations and densities and erosion patterns within a single canyon. Greater understanding of the structural control of differential erosion in bedrock canyons will help inform numerical models of landscape evolution and stream erosion.

10:45am Melora Koepke
Program: PhD
Supervisor: Eugene McCann

“Shelter, triage, border: Urban tactics for “exclusion within inclusion” in Paris, 2016-2018”

Drawn from my ongoing dissertation research, this paper investigates a key urban site of the European “migration crisis”, the Centre de Premier Accueil Paris-Nord (CPA) a.k.a. La Bulle (the Bubble). Opened in November 2016 by the Ville de Paris in response to the proliferation of informal migrant camps in the French capital, the CPA, which operated for seventeen months in Porte de La Chapelle, an intersection in northeastern Paris, was France’s first and Europe’s largest reception centre for primo-arrivants (newly-arrived migrants). Drawing on ethnographic fieldwork in Porte de la Chapelle for the entire duration of its existence, this research considers this experimental, temporary municipal dispositif de premier accueil  (apparatus of primary reception) as both a socio-spatial site of governance that controlled the bodily presences of primo-arrivants in the city and also shaped an emergent politics of exclusion across multiple scales and territories.

11:00am Anika Vassell
Program: MA
Supervisor: Valorie Crooks

"Telling the story of medical crowdfunding: Qualitative explorations of campaign narratives and news coverage"

Medical crowdfunding campaigns connect disparate people and places through the practice of individuals using (or exploiting) social networks to seek funding from others for health-related expenses. I have conducted two analyses to assist with addressing important knowledge gaps about this practice. First, I explored how Canadians managing, or seeking diagnosis for Lyme disease write campaign narratives. Lyme disease is a contested illness, and this reality frames important aspects of the illness experience. Through thematically exploring medical crowdfunding campaign narratives, I found that they typically touched upon four common issues central to the illness experience. The consistency in these narratives across campaigns suggests that there is some common awareness of how to portray the illness experience on crowdfunding platforms. Second, I conducted interviews with Canadian journalists to explore their experiences of writing news stories that are based on medical crowdfunding campaigns. These interviews revealed that journalists hold responsibilities to the news story being written, to the campaign and campaigner, and to their profession when participating in such news coverage.

11:15am  Cora Cao
Program: MSc
Supervisor: Suzana Dragicevic

“Convolutional neural networks method for land use change classification”

Worldwide population growth leads to the increasing demand for natural and land resources, as a result, changes in land use and land cover have caused the concerns from urban planners and land managers. Convolutional neural network (CNN) method has been used increasingly in land-use classification tasks, and this research study applied transferred CNN models to identify land use change on orthophotos from 2004 to 2017 of community of Cloverdale, in the northeastern part of the City of Surrey, Canada. Eight transferred CNN models were evaluated, CNN combined with Support Vector Machine (SVM) has shown efficient performance in this case study. Two sources of datasets were used for model training to solve the issue of limited data, improving the adaptation of classification models. Through analysis of classified land use maps from multiple years, the classification results accurately capture the trends of LU changes.

11:30am   Babak Kasraei
Program: MSc
Supervisor: Margaret Schmidt

“Using Airborne LiDAR-acquired Digital Elevation Data for Production of High-resolution Digital Soil Maps”

Digital soil mapping is a subdiscipline of soil science that can be used to produce high resolution maps of soil properties. It is expected that high resolution digital elevation models (DEM) generated from LiDAR technology may help make better maps of soil properties. The objective of this project is to evaluate the use of airborne LiDAR-acquired digital elevation data to produce high-resolution digital soil maps (DSMs). In summer 2018, 213 plots were sampled in the study area, the Eagle Hill Forest (150 ) located northwest of Kamloops lake, BC. The soil properties measured are depth to carbonate, soil depth, coarse fragment content, forest floor depth, soil total carbon content, organic carbon content, inorganic carbon content, total nitrogen, pH and texture (sand, silt and clay contents). Modeling methods used in this project are Random Forest, Multi Linear Regression, and Cubist decision tree. The validation method used is random holdback. Thus far several maps have been produced with different levels of validation accuracy. Data collection will continue in the spring of 2019 and improved methods of modeling soil properties using LiDAR will be investigated.

 Lunch 11:45-12:20