M.P. Trudeau, Corresponding author A303 Loeb Building, Carleton University, 1125 Colonel By Drive, Ottawa, Ontario K1S 5B6 Canada Tel.: +00 1 613 231 3537 E-mail address: email@example.com
Murray Richardson A329 Loeb Building, Carleton University, 1125 Colonel By Drive, Ottawa, Ontario K1S 5B6 Canada Tel.: +00 1 613 520 2600, ext. 2574 E-mail address: Murray_Richardson@carleton.ca
Urbanization is generally known to change natural stream flow regimes (e.g. Schueler et al., 2009) but a lack of high-resolution, long-term records for flow, rainfall and land use have hampered a full characterization of these changes. This unique empirical study examines 42 years of hydrologic data, between May and November, at a temporal resolution of 15-minutes. This resolution allows estimations of total flow as well as peak flow responses to rain events, rising limb event flows and rising limb event flow accelerations with the objective of quantifying changes in flow regime attributable to increased urban area.
The study includes temporal and spatial analyses of flows within 33 sub-catchments of 11 watersheds confluent with Lake Ontario and Lake Erie in the Greater Toronto Region of the Canadian Great Lakes Basin. This region experienced heavy urbanization during the study period, 1969 to 2010, making the City of Toronto now the fourth largest city by population in North America (City of Toronto, 2014). The watersheds and sub-watersheds in the study vary in percent urban land cover and range in size from 806 km2 for a Humber Watershed catchment to 37.5 km2 for a sub-catchment of the Don Watershed. Flow regime perturbations associated with stress to aquatic communities are of particular interest for the study (e.g. Clausen and Biggs, 1997; Gibbins et al., 2001).
Examination of trends over four decades on a watershed basis at a very fine temporal resolution allows quantification of the cumulative effects of changing land cover on flow regime. The spatial analysis of catchments of various sizes, and with varying degrees of urban land use, provides a robust empirical estimation of flow characteristics associated with urban land use.
Data have been accessed or derived from several sources. The instantaneous (15-minute) hydrometric dataset and one-hour rainfall records were obtained from Environment Canada. Historic aerial photographs were purchased from Natural Resources Canada and urban areas were digitized by our team for 5 watersheds and their sub-watersheds. (Note that digitization work is on-going until December 2014). Province of Ontario government land and watershed data sets were used to calculate catchment areas and slopes. The Credit Valley Conservation Authority and the Toronto and Region Conservation Authority provided data and support for the land use database, as did P. Thompson (2013).
The study approach is to develop statistical empirical models for total flow, peak rain event flow and rising limb flow acceleration for both temporal and spatial variables. An extensive analysis of available rainfall data was also undertaken to determine whether or not there were concurrent trends in rainfall patterns. Because the study region experiences winter conditions, all flow regime analyses were undertaken for the seasonal period May 26th to November 15th in order to avoid introduced flow variation from spring frechette and winter freeze-up.
Results and Discussion
A temporal analysis of total seasonal flow (May to November) for a Don Watershed catchment (311 km2) and a Humber Watershed catchment (806 km2), from 1969 to 2010 inclusive, indicates an increase of about 45% over four decades, with no concurrent trend in rainfall. During the same timeframe, peak rain events flow rates increased by about 1m3s-1yr-1. In the Don, the ratio of rising limb event flows to median flow increased from 13 in the 1970's to 21 in the 2000's and rising limb event flow acceleration increased 2.4-fold over 4 decades. Similar analyses in the Humber showed higher variation in flow response but also indicated statistically significant increases in event flows and event flow acceleration. The spatial analysis to associate urban land cover with flow regime attributes is on-going. For 28 records in 8 sub-watersheds ranging in size from 43 km2 to 770 km2, a statistical model with one term, the interaction of percent urban cover with total rain, explains over 88% of the variation in total flow standardized by watershed area (in millimeters depth). By contrast, total rain depth alone explains 52% of the variation. Results for peak event flow and rising limb event flow acceleration are in preparation.
Overall, the study demonstrates marked temporal alterations in total and event flow regimes. Spatial analysis of total flow indicates that changing urban land cover is a statistically significant variable in predicting total seasonal flow. The study demonstrates that long-term, high temporal resolution hydrological records can quantify cumulative changes on a watershed basis. The dramatic changes identified signal the importance of managing water resources on a basin scale. Quantifying cumulative effects of urban land cover has potential implications for land development decisions, flood risk assessment, urban infrastructure design and ecological habitat protection. Alterations to hydrologic baselines resulting from urban land use are also relevant for assessments of urban infrastructure resilience in the face of changing climate trends (Milly et al., 2008).
1. City of Toronto, 2014, Toronto Facts, URL (Accessed October 2014): http://www1.toronto.ca/wps/portal/contentonly?vgnextoid=57a12cc817453410VgnVCM10000071d60f89RCRD
2. Clausen, B. and B. Biggs, 1997. Relationships between benthic biota and hydrological indices in New Zealand streams, Freshwater Biology, 38(2):327-342.
3. Gibbins, C.N., Dilks, C.F., Malcolm, R., Soulsby, C., Juggins, S.S. 2001. Invertebrate communities and hydrological variation in Cairngorm mountain streams. Hydrobiologia 462:205-219.
4. Milly, P. C. D., Betancourt, J., Falkenmark, M., Hirsch, R. M., Kundzewicz, Z. W., Lettenmaier, D. P., & Stouffer, R. J. (2008). Stationarity Is Dead: Whither Water Management? Science, 319(5863), 573-574.
5. Schueler, Fraley-McNeal, and Cappiella. 2009. Is impervious cover still important? Review of recent research. Journal of Hydrologic Engineering 14(4):309-315.
6. Thompson, P. 2013. Event Based Characterization of Hydrologic Change in Urbanizing Southern Ontario Watersheds via High Resolution Stream Gauge Data, Master of Applied Science Thesis, Waterloo, Ontario.