Anomaly detection from time-changing environmental sensor data streams

This thesis stems from the project with real-time environmental monitoring company EMSAT Corporation. They were looking for methods to automatically ag spikes and other anomalies in their environmental sensor data streams. The problem presents several challenges: near real-time anomaly detection, absence of labeled data and time-changing data streams. Here, we address this problem using both a statistical parametric approach as well as a non-parametric approach like Kernel Density Estimation (KDE). The main contribution of this thesis is extending the KDE to work more effectively for evolving data streams, particularly in presence of concept drift. To address that, we have developed a framework for integrating Adaptive Windowing (ADWIN) change detection algorithm with KDE. We have tested this approach on several real world data sets and received positive feedback from our industry collaborator. Some results appearing in this thesis have been presented at ECML PKDD 2015 Doctoral Consortium.

Independent Projects Environmental Studies 2010: From Climate Change Laggard to Leader: Policy recommedations for Newfoundland and Labrador

Dyer, along with many others reflecting international consensus in the scientific community, argues that countries must be carbon neutral by 2050 to avoid the worst impacts of climate change (Dyer, 2008, p. xii). to accomplish this by 2050, we need to act now: governments need to cooperate with the scientific community to ensure our society makes the changes to combat climate change. How is Canada reacting to this situation? As describes by the Sierra Club of Canada, "federal government continues to drag its feet, and delay any action to reduce emissions" (2008, p. 18). Given this federal reluctance, individual provinces must act to reduce emissions. While some provinces take a required action, primarily Quebec and British Columbia, others, like Newfoundland and Labrador, contribute little in terms of emissions reductions. Newfoundland and Labrador is ranked as "poor"-among the worst performers regarding climate change policy in the country-by the David Suzuki Foundation in a cross Canada evaluation of various provincial climate change policies (David Suzuki Foundation, 2008, p. 9). The Province of Newfoundland is not doing enough to address climate change. In this paper I argue that to improve this situation, the province could follow the example of the two leading jurisdictions, Quebec and British Columbia, to refine and introduce its own hybrid policy that directly affects decision making processes. But, this can be complicated when convincing the government that it is important to accept stronger policy. The government must consider what climate change impacts Newfoundland and Labrador experiences and willconinue to experience and what Newfoundland and Labrador is contributing to the problem of climate change. To evaluate these issues of climate change, I first survey the positive policies Newfoundland and Labrador is currently implementing/ discussing and then outline action taken by the provincial environment leaders, Quebec and British Columbia.Then I describe the strong pieces in the Quebec and British Columbia climate change action plan that Newfoundland and Labrador can emulate. Finally, I consider if thes polices are politically feasible for the government of Newfoundland and labrador. Hence, this paper aims to give a blueprint of what Newfoundland and Labrador has to act on to make itself an environmental leader.

Toxic Legacies, Slow Violence, and Environmental Injustice at Giant Mine, Northwest Territories

For fifty years (1949–99) the now-abandoned Giant Mine in Yellowknife emitted arsenic air and water pollution into the surrounding environment. Arsenic pollution from Giant Mine had particularly acute health impacts on the nearby Yellowknives Dene First Nation (YKDFN), who were reliant on local lakes, rivers, and streams for their drinking water, in addition to frequent use of local berries, garden produce, and medicine plants. Currently, the Canadian government is undertaking a remediation project at Giant Mine to clean up contaminated soils and tailings on the surface and contain 237,000 tonnes of arsenic dust that are stored underground at the Giant Mine. Using documentary sources and statements of Yellowknives Dene members before various public hearings on the arsenic issue, this paper examines the history of arsenic pollution at Giant Mine as a form of “slow violence,” a concept that reconfigures the arsenic issue not simply as a technical problem, but as a historical agent of colonial dispossession that alienated an Indigenous group from their traditional territory. The long-term storage of arsenic at the former mine site means the effects of this slow violence are not merely historical, but extend to the potentially far distant future.