Regional trends in evaporation loss and water yield based on stable isotope mass balance of lakes: The Ontario Precambrian Shield surveys

Abstract Stable isotopes of water, oxygen-18 and deuterium, were measured in water samples collected from a network of 300 lakes sampled in six ∼100 km 2 blocks (centred at 49.72°N, 91.46°W; 48.49°N, 91.58°W; 50.25°N, 86.62°W; 49.78°N, 83.98°W; 48.24°N, 85.49°W; 47.73, 84.52°W) within Precambrian shield drainages in the vicinity of Lake Superior, northern Ontario, Canada. Additional sampling was also conducted within the Turkey Lakes watershed (47.03°N, 84.38°W), a research basin situated in the Algoma region located 50 km north of Sault Saint Marie, Ontario. The studies were undertaken to gain a better understanding of hydrology and geochemistry of watersheds in the region in order to better predict acid sensitivity of lakes. The main objective of this paper is to describe the hydrologic variations observed based on stable isotope results. Evaporative isotopic enrichment of lake water was found to be systematic across the region, and its deviation from the isotopic composition of precipitation was used to estimate the evaporation/inflow to the lakes as well as runoff (or water yield) based on a simple isotope mass balance model. The analysis illustrates significant variability in the water yield to lakes and reveals a pattern of positively skewed distributions in all six widely spaced blocks, suggesting that a high proportion of lakes have relatively limited runoff whereas relatively few have greater runoff. Such basic information on the drainage structure of an area can be valuable for site-specific hydrologic assessments but also has significant implications for critical loads assessment, as low runoff systems tend to be less buffered and therefore are more sensitive to acidification. Importantly, the Turkey Lakes sampling program also suggests that isotope-based water yield is comparable in magnitude to hydrometric gauging estimates, and also establishes that uncertainty related to stratification can be as high as ±20% or more for individual lakes, although it likely has only a minor influence on regional survey results. While further analysis in gauged lake watersheds would be beneficial to constraining the accuracy of the method or calibrating it for operational use, it is nevertheless a powerful tool in its present form for lake-to-lake and regional runoff inter-comparisons.