Assessment of the ability of the standardized precipitation evapotranspiration index (SPEI) to model historical streamflow in watersheds of Western Canada. Issue 1 (3rd April 2021)

• Record Type: Journal Article
• Authors: Gurrapu, Sunil; Hodder, Kyle R.; Sauchyn, David J.; St. Jacques, Jeannine Marie
• Is Part Of: Canadian water resources journal =. Volume 46:Issue 1/2(2021)
• Page Start: 52
• Page End: 72
• Publication Date: 2021-04-03
• Subjects: Water resources development -- Canada -- Periodicals; Water conservation -- Canada -- Periodicals; 333.9100971
• DOI: 10.1080/07011784.2021.1896390
• Languages: English
• ISSNs: 0701-1784

Abstract

Abstract: Knowledge of present-day spatial and temporal distribution of water resources is vital for successful water management and policies for planned adaptation to climate change. Measured quantities of hydroclimatic variables, including precipitation, evapotranspiration, streamflow, etc., are the primary indicators of water availability, and indices derived using several such primary variables provide a means to express water availability across a range of spatio-temporal scales. In this study, the ability of one such multi-scalar index, the Standardized Precipitation Evapotranspiration Index (SPEI), computed at a range of time scales, was examined to see how well it could model historically observed warm season monthly and annual streamflow in 24 natural-flowing watersheds of western Canada. The empirical relationships between the SPEI, computed at 1-, 3-, 6-, 9-, 12- and 24-month time scales, and monthly and annual streamflow were analyzed, showing significant correlations for all watersheds. The time scale of the SPEI with the strongest correlations varied seasonally. Based on these results, SPEI-based principal component regression (PCR) equations were calculated to model warm season monthly and annual historical streamflow. These PCR equations are able to adequately capture historical streamflow in these watersheds. Annual streamflow variability was better captured (mean R adj 2 = 0.46) than monthly variability (mean R adj 2 = 0.30 over March–October). Summer and fall streamflow variability was better captured (mean R adj 2 = 0.42 over June–September) than spring variability (mean R adj 2 = 0.15 over March–April).