WC130 MayJune2023 - Magazine - Page 8
INNOVATION
Data Potential
Advancing AI and machine learning in Canada’s water sector
BY CHRIS GERRITS
T
Chris Gerrits
Chris Gerrits, M.Sc., P.Eng., is a water
resource engineer/hydrogeologist and senior
project manager with Crozier Consulting
Engineers. Chris is also the mayor of the
Township of Amaranth in Southern Ontario.
Special contributor Harsh Panchal, M.Eng.,
AI Engineering Analyst, at Crozier
Consulting Engineers.
8
WATER C AN ADA • M AY/JUNE 2023
In a similar fashion, many conservation authorities
operate a number of water monitoring stations that
collect stream flow data, water quality data, and operate weather stations, often collecting information at
regular intervals (hourly, weekly, monthly) resulting
in a large amount of data. Environment Canada has
approximately 1,631 weather stations in the Province of Ontario alone that have been collecting data
since 1840. Yet, widespread use of historical data is
not common outside of highly scientific studies or
academic settings.
It is time for a shift in how the water industry in
Canada views and values data as an asset. All available
data can be leveraged and inform the decision-making process. The emergence of artificial intelligence
(AI) and machine learning (ML) technology in recent
years could result in a revolutionary advancement in
how we utilize data to aid in decision-making. While
actual measured and verified data is still irreplaceable, there is a role for new technology to help with
modelling and predicting future outcomes based on
past results.
AI in action
There are currently examples of AI/ML models being
utilized around the world. For example, Malta is utilizing AI predictive models to optimize the operation
of their water system, and the University of Kentucky
and the Kentucky Geological Survey are leveraging
ML models to identify karst sinkholes from remote
sending data.
Crozier Consulting Engineers
HE WATER SECTOR IN CANADA is very
data intensive. Whether you are a water system operator, or are with a local, provincial,
or federal agency, a conservation authority, or
some other party, you are most likely collecting a large amount of data that is then stored, going
largely unused. For example, following the recommendations laid out in Part 2: Report of The Walkerton Inquiry, most municipalities in Ontario began
collecting supervisory control and data acquisition
(SCADA) data for their water supply systems. SCADA data is dense and has multiple fields including
water level, pump speed, and discharge. A number
of water systems are also equipped with continuous
water quality monitoring systems that collect data for
water quality parameters such as chloride, hardness,
turbidity, among others.
