Real-Time Flood Forecasting System for Cooksville Creek, Ontario.

Neelam Gupta, Rob James, Karen Finney and Tiehong Xiao

ABSTRACT

Cooksville Creek, located in Mississauga, Ontario, discharges to Lake Ontario between the Credit River and Etobicoke Creek. At approximately 3600 ha and heavily urbanized, the Cooksville Creek watershed is considered to be 95% developed and comprises over 50% directly connected impervious area. With little online storage, Cooksville Creek watershed has a short rainfall-response time, and has a history of flooding in the middle and lower regions of the watershed. Complicating the flooding issues are the approximately 300 buildings currently located within the flood plain.

Credit Valley Conservation (CVC) is the conservation authority charged with flood management for the larger Credit Valley Watershed, which includes Cooksville Creek. As part of this responsibility, CVC has a need to provide adequate and timely flood warning to property owners and local emergency response agencies. In an effort to improve the accuracy and lead time of flood forecasting, a remote-sensing flood forecasting and decision support system was developed. Utilizing PCSWMM Real-Time, NOAA’s NWS NEXRAD WSR-88D radar, US EPA SWMM5, HTML5 and Google Maps, this system forecasts near-future water surface elevations throughout Cooksville Creek and produces products to help flood duty staff make appropriate and informed decisions.

The PCSWMM Real-Time system’s four main components include: real-time radar rainfall acquisition, processing and forecasting, real-time continuous SWMM modeling, real-time flood inundation and flood vulnerable asset analysis, and real-time web-based decision support. Advantages of the approach include the remote sensing nature of radar, high resolution, physically-based, dynamic modeling, continuous model calibration, the utilization of all available data, and comprehensive decision support with anywhere access.

This paper highlights the key features of this system, evaluates its performance and presents lessons learned from developing and integrating the system components.


Permanent link: