Presentation Summary

Overland flooding has a profound socio-economic, food security, public safety, and environmental impact on agricultural land, rural properties, and public infrastructures.

The current flood risk analysis work is predominantly focused on small-scale scenarios within densely populated urban zones, resulting in a notable information gap on flood susceptibility across a vast agricultural land in the Canadian prairies.

To bridge this critical information gap, this study is designed to use small-scale flood simulation analysis results and various publicly available data, such as hydrometric, digital elevation models (DEM), physical terrain and vegetation groundcover characteristics and meteorological data, to train artificial intelligence (AI) systems to determine if AI systems can accurately predict overland flood risk in South Saskatchewan in a near real-time.

This research will entail the utilization of geospatial data processing techniques, facilitated by Geographic Information Systems (GIS), coupled with various flood risk models and applications.  

The study will employ the latest artificial intelligence methodologies for data processing and predictive risk modeling approach.  To future proof this initiative, various climate change scenarios will be incorporated into the development of the AI models. The following two abstracts are subsections under the current research project. 

Abstract from Xin Zhou

Machine learning techniques, such as Random Forests (RF), Support Vector Machines (SVM), and Artificial Neural Networks (ANN), can quickly and accurately classify landcover from large satellite imagery  To gauge the effectiveness of each machine learning approach in classifying landcovers, Ms. Zhou used publicly accessible Sentinel-2 satellite imagery sourced from the European Space Agency. The imagery data were used to train the machine learning systems.  Each machine learning system were evaluated for its accuracy, speed, and computer resource demands.  This presentation will provide the advantage and disadvantage of each machine learning system for land cover classifications.

Abstract from Amin Hassanjabbar

Mr. Hassanjabbar applied  Artificial Neural Network (ANN) method to investigate the impacts of climate change on the water quantity and quality of the Qu’Appelle River in Saskatchewan, Canada. First, the second-generation Canadian earth system model (CanESM2) was adopted to predict future climate condition. The Statistical DownScaling Model (SDSM) was then applied to downscale the generated data. To analyze water quality of the river, concentrations of the Dissolved Oxygen (DO) and Total Dissolved Solids (TDS) from the river were collected. Using the collected climate and hydrometric data, the ANN networks were trained to simulate (i) the ratio of snowfall to total precipitation based on temperature, (ii) river flow rate based on temperature and precipitation; and (iii) DO and TDS concentrations based on river flow and temperature. Finally, the generated climate change data were used as inputs to the ANN model to investigate the climate change impacts on the river flow as well as DO and TDS concentrations within the selected region. Hydrologic alteration of the river was evaluated via the Range of Variability Approach (RVA) under historical and climate change scenarios. This presentation will summarize the results of this analysis.

Researcher Bios

Dr. Peng Wu presently holds the position of Associate Professor at the University of Regina within the Environmental Systems Engineering department. He became a part of the university's faculty in 2015 subsequent to the successful completion of his PhD in hydraulic engineering, with a distinct emphasis on river flooding and cold region research. His ongoing research endeavors are currently backed by notable institutions such as NSERC, CFI, and the Ministry of Agriculture of Saskatchewan.

Miss Xin Zhou is actively pursuing her PhD in Environmental Systems Engineering. She earned her bachelor's degree in Geomatics Engineering from China and proceeded to obtain her master's degree from the University of Regina. Her research pursuits primarily revolve around Geographic Information Systems (GIS) and Artificial Intelligence, showcasing her expertise in these domains.

Mr. Amin Hassanjabbar, a PhD candidate in Environmental Systems Engineering, boasts a background encompassing both a bachelor's degree in civil engineering and a master's degree in water engineering. His research focal points encompass Artificial Neural Networks (ANN) and comprehensive data analysis, which demonstrate his adeptness in these specialized areas. 

Industry Technical Advisors

Dr. Kevin McCullum, P.Eng., is the principal owner of KRM & Associates (Environmental Engineering) with specialization in real-time large data environmental analysis and interpretation using statistical and machine learning analysis.

Todd Han, P.Biol., PAg, is a principal Environmental Scientist with Pathway Environmental Solutions Inc., specializing in vegetation health impact analysis using machine learning system related to oil and gas development and operation activities. 

Event Information

Date: September 27, 2023

Time: Virtual 12:00 PM - 1:00 PM MST

Registration Fee(s): 

*Corporate members, please login to your corporate member account to manage employee accounts and provide access to your team members to register at the member rate. To inquire, please contact us.

Registration Deadline: September 25, 2023

The webinar access information will be available to you upon successful registration for the event.

This Webinar Proudly Sponsored by: