RESEARCH INTERESTS: Predictive modeling of ecological and environmental processes via big data analytics; Wildland fire occurrence prediction; modeling of spatio-temporal processes.
Seeking academic or industry partnerships in the area(s) of: Forest Fire Analytics, Agriculture and Environmetrics, Quantitative Ecology
Available positions for grads/undergrads/postdoctoral:
- Research projects for both MSc and PhD programs are currently available in my lab. Prospective graduate students with an academic background focusing on statistics, biostatistics or a closely related field and having strong programming skills (e.g RStudio) are welcome to apply.
- International doctoral students with a minimum 80% (A-) admission average are strongly encouraged to apply and avail an opportunity to apply for the International Doctoral Student Scholarship (see details here).
- I also co-supervise graduate students in the Bioinformatics program at the University of Guelph.
My research focuses on leveraging innovative techniques and methods in computing and statistical machine learning to extract meaningful and actionable insights from massive volumes of high-dimensional data. His work focuses on predictive modeling of ecological and environmental processes via big data analytics. Key research themes include:
- Wildland fire occurrence prediction. My research harnesses the availability of large scale historical environmental, wildland fire and demographic data in Canada to spatially predict severe and large forest fires for two weeks ahead into the future. Apart from contributing to improvements to the Canadian Interagency Forest Fire Centre’s (CIFFC) management, coordination and information services, this research has implications for sustainable forest management in the face of changing climate and increasing human anthropogenic activity. The outcomes of this ongoing work will have significant impact on Canada’s ability to efficiently respond to the danger of severe wildland fires and to understand fire-weather dynamics in the wake of changing Earth climate.
- Agro-environmental science. Within agro-environmental science, data across multiple sources and scales are increasingly becoming available, encompassing the full spectrum of four V’s of big data: volume, variety, velocity, and veracity. Sources of these datasets range from agronomic data through precision agriculture such as monitoring site-specific soil characteristics and harvest yield on geo-spatial scales; to functional crop genomics data, to agroeconomics data. Data availability in such volume and variety presents an opportunity to integrate and analyze these data streams for evidence-based decision making in implementing sustainable agricultural practices. I am interested in: i) leveraging the availability of varied data sources to answer important large-scale comprehensive research questions related to agricultural productivity and its interaction with the environment, ii) developing a digital framework to identify, collect and integrate relevant data from various sources to develop databases and decision-support tools, and iii) employing advanced statistical and computational techniques, including machine learning and artificial intelligence methods, to develop predictive models for answering the key research questions, and to deploy them in a decision-support system. Key agricultural research problems to explore are: i) how climate change will impact food productivity and security in Canada and to what extent northern areas will become suitable for agriculture, and ii) how biodiversity components and ecosystem services interact across a wide range of agricultural landscapes and farming practices.
- CEPS News: Fired Up
Food from Thought
- Food from Thought, University of Guelph: Appointment of Assistant Professor in Statistics to support Food from Thought through computational statistics and big data analytics