Multi-Century Spring Flood Reconstruction in Eastern Boreal Canada from Novel Application of Wood-Cell Anatomy
- 1University of Quebec AT, Forest Research Institute, Rouyn-Noranda, QC, Canada (alexandreflorent.nolin@uqat.ca)
- 2The University of Winnipeg, Center for Forest Interdisciplinary Research, Winnipeg, MB, Canada (j.tardif@uwinnipeg.ca)
- 3University of Arizona, Laboratory of Tree-Ring Research, Tucson, AZ, United States (dmeko@ltrr.arizona.edu)
- 4Industrial Chair CRSNG-UQAT-UQAM in sustainable forest management, Rouyn- Noranda, QC, Canada (yves.bergeron@uqat.ca)
The streamflow regimes of eastern boreal Canada are snow-melt and ice-melt driven with the highest flows occurring in spring. Over the last few decades, a positive streamflow trend has been observed, with increasing severity and frequency of spring flooding. Further changes in flood dynamics are projected as a consequence of global climate change. The validity of projections is restricted by the lack of long and spatially well-replicated observations. High-resolution proxy records are needed to better understand the natural range of variability in spring runoff and associated atmospheric controls.
Recent research has shown that riparian black ash trees (Fraxinus nigra Marsh.) exposed to periodic submersion produce “flood rings” whose earlywood cross-sectional vessel area is linearly associated with the severity of flooding. Twelve continuous chronologies of ring width and earlywood vessel anatomy were developed for Lake Duparquet to extend the record of Harricana River mean spring flow. A visually determined index of flood rings was also developed to determine i) the spatial coherency of the spring flood signal and ii) the coherency of the flood signal among natural, regulated and unflooded rivers.
The reconstruction spans the period 1770-2016 and captures more than 65% of the variance of Harricana river spring flow. Trend analysis indicates an increase in both magnitude and frequency of the major floods starting at the end of the Little Ice Age (LIA, 1850-1890), with highest peaks after 1950. Time-frequency analysis shows non-stationarity: a stable 30-year periodicity during the LIA is replaced by a decadal pattern starting around 1850, and evolves into a more high-frequency pattern after 1930. The signal is strongly coherent between watersheds for natural rivers and weaker for regulated basins. Field correlations with gridded climate data indicate the broad spatially coherent pattern of spring high flows across much of central/eastern north Canada is positively associated with April-May precipitation and snow cover, and negatively associated with March-April maximum temperature.
These large-scale associations support atmospheric forcing of inter-annual hydroclimatic variability. While the Artic and North Atlantic Oscillations have previously been found to influence winter and spring climate conditions in eastern Quebec, our results contrast with a significant negative association with El-Niño Southern Oscillation from January to May, and the Pacific Decadal Oscillation from December to February. In Lake Duparquet, warm and wet air from Pacific-South Ocean (El-Niño) are associated with early spring and small floods, while cold and dry air masses (La-Niña) correlate to late thaw and high floods in spring. The association with sea surface temperature and 200mb geopotential field heights reveal a clear atmospheric connection between eastern north boreal Canada and the tropical Pacific Ocean.
The novel application of wood-cell anatomy to hydroclimatology underscores an increase in flood frequency and severity since the end of the 18th century in northeastern Canada. More broadly, the application highlights how analysis of tree rings from riparian trees can be used to extend the flood history of boreal rivers.
How to cite: Nolin, A. F., Tardif, J. C., Conciatori, F., Meko, D. M., and Bergeron, Y.: Multi-Century Spring Flood Reconstruction in Eastern Boreal Canada from Novel Application of Wood-Cell Anatomy, EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-5572, https://doi.org/10.5194/egusphere-egu2020-5572, 2020