Urban land development causes significant changes to river channel form and function. Outcomes are strongly conditioned by local circumstances although responses such as channel enlargement and incision are common but are highly variable in extent and detectability. Many case studies focus on small watersheds (or short reaches) and before-after conditions, and direct causation and temporal trajectories of change are not well known but often presumed to be a single step change and response. Well-documented examples in fully-urban, larger watersheds are rare, and rarer still are cases covering the full trajectory from pre-urban to fully urban land cover with accompanying data on the direct drivers of change such as discharge and stream power. Highland Creek, Toronto, has a drainage area of 102 km2 and between the early 1950s and 2000s underwent a land cover change from agricultural to completely urban in very low relief terrain with channels in narrow valleys incised locally into glacial sediments. Streamflow monitoring, land cover data and aerial photography over this seven-decade period, along with a calibrated runoff model and digital elevation data, give a complete picture over multiple epochs of the causes and changes in fluvial hydrology and geomorphology. The complete transformation of the flow regime, along with channel straightening and steepening, resulted in increases in average stream power (total and specific) of the order of ten and six times, respectively. The channel is now exceptionally energetic with maximum total and specific stream power for 2-year flood approaching 5000 Wm−1 and 300 Wm−2, respectively. The semi-alluvial and supply-limited character of the channels is also influential. Channel response is variable along the 20–25 km of channel measured and is notably different between the sub-watersheds, related to timing of development and its intersection with engineering intervention, and changing conceptions of the channel over time. In places the channel widened by a factor of four or five and changed from high sinuosity meanders to wide, bar-chute morphology, consistent with regime expectations for gravel-bed channels. In other places the process of urbanization has involved progressive engineering of the channel to convey stormwater (channelized), protect infrastructure and mitigate erosion, so that responses in these reaches are much more restricted and fluvial functions such as bend and bar development have been eliminated. Substantial channel transformation can be seen to have occurred episodically resulting from two major flood events as well as the increases in magnitude and frequency of peak flows, which limit post-flood recovery. Urbanization effects and consequences have developed gradually over a long period, rather than in a single urbanization step change, and they do not match simple models of channel evolution. In many ways urbanization of the river is ongoing, particularly in the form of geomorphically-referenced design of the channels as a response to damage from large floods and overall institutional watershed planning. The ongoing urbanization story of Highland Creek illustrates some extreme effects of urbanization as well as the complexity and contingency of response to the physical and institutional processes influencing the trajectory of urban river evolution.
Read full abstract