• Agriculture and Agri-Food Canada, cited 2013: Non-contributing portions of the incremental gross drainage areas of the AAFC watersheds project. [Available online at http://data.gc.ca/data/en/dataset/a5cbed50-4a18-4658-b312-af13192bbaaf.]

  • Akinremi, O. O., , McGinn S. M. , , and Cutforth H. W. , 1999: Precipitation trends on the Canadian Prairies. J. Climate, 12, 29963003, doi:10.1175/1520-0442(1999)012<2996:PTOTCP>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Bao, Z., , Ye H. , , and Taylor P. A. , 2005: Summer regional rainfall over southern Ontario and its associations with outgoing long wave radiation and moisture convergence. Meteor. Appl., 12, 161167, doi:10.1017/S135048270500160X.

    • Search Google Scholar
    • Export Citation
  • Beltaos, S., , Burrell B. , , and Ismail S. , 1996: 1991 ice jamming along the Saint John River: A case study. Can. J. Civ. Eng., 23, 381394, doi:10.1139/l96-043.

    • Search Google Scholar
    • Export Citation
  • Blöschl, G., , Nester T. , , Komma J. , , Parajka J. , , and Perdigão R. A. P. , 2013: The June 2013 flood in the upper Danube basin, and comparisons with the 2002, 1954 and 1899 floods. Hydrol. Earth Syst. Sci. Discuss., 10, 95339573, doi:10.5194/hessd-10-9533-2013.

    • Search Google Scholar
    • Export Citation
  • Boening, C., , Willis J. K. , , Landerer F. W. , , Nerem R. S. , , and Fasullo J. , 2012: The 2011 La Niña: So strong, the oceans fell. Geophys. Res. Lett., 39, L19602, doi:10.1029/2012GL053055.

    • Search Google Scholar
    • Export Citation
  • Brimelow, J. C., , Stewart R. S. , , Hanesiak J. M. , , Kochtubajda B. , , Szeto K. , , and Bonsal B. , 2014: Characterization and assessment of the devastating natural hazards across the Canadian Prairie provinces between 2009 and 2011. Nat. Hazards, 73, 761785, doi:10.1007/s11069-014-1107-6.

    • Search Google Scholar
    • Export Citation
  • Brown, R. D., , and Mote P. W. , 2009: The response of Northern Hemisphere snow cover to a changing climate. J. Climate, 22, 21242145, doi:10.1175/2008JCLI2665.1.

    • Search Google Scholar
    • Export Citation
  • Budikova, D., , Coleman J. S. M. , , Strope S. A. , , and Austin A. , 2010: Hydroclimatology of the 2008 Midwest floods. Water Resour. Res., 46, W12524, doi:10.1029/2010WR009206.

    • Search Google Scholar
    • Export Citation
  • Burrows, W. R., , and Kochtubajda B. , 2010: A decade of cloud-to-ground lightning in Canada: 1999–2008. Part 1: Flash density and occurrence. Atmos.–Ocean, 48, 177194, doi:10.3137/AO1118.2010.

    • Search Google Scholar
    • Export Citation
  • Chen, G., , Iwasaki T. , , Qin H. , , and Sha W. , 2014: Evaluation of the warm-season diurnal variability over East Asia in recent reanalyses JRA-55, ERA-Interim, NCEP CFSR, and NASA MERRA. J. Climate, 27, 55175537, doi:10.1175/JCLI-D-14-00005.1.

    • Search Google Scholar
    • Export Citation
  • Christidis, N., , Stott P. A. , , Karoly D. J. , , and Ciavarella A. , 2013: An attribution of the heavy rainfall over eastern Australia in March 2012 [in “Explaining Extreme Events of 2012 from a Climate Perspective”]. Bull. Amer. Meteor. Soc., 94 (9), S58S61.

    • Search Google Scholar
    • Export Citation
  • Dankers, R., , and Feyen L. , 2009: Flood hazard in Europe in an ensemble of regional climate scenarios. J. Geophys. Res., 114, D16108, doi:10.1029/2008JD011523.

    • Search Google Scholar
    • Export Citation
  • Deacu, D., , Fortin V. , , Klyszejko E. , , Spence C. , , and Blanken P. D. , 2012: Predicting the net basin supply to the Great Lakes with a hydrometeorological model. J. Hydrometeor., 13, 17391759, doi:10.1175/JHM-D-11-0151.1.

    • Search Google Scholar
    • Export Citation
  • Dee, D. P., and et al. , 2011: The ERA-Interim reanalysis: Configuration and performance of the data assimilation system. Quart. J. Roy. Meteor. Soc., 137, 553597, doi:10.1002/qj.828.

    • Search Google Scholar
    • Export Citation
  • Diffenbaugh, N. S., , Scherer M. , , and Ashfaq M. , 2013: Response of snow-dependent hydrologic extremes to continued global warming. Nat. Climate Change, 3, 379–384, doi:10.1038/nclimate1732.

    • Search Google Scholar
    • Export Citation
  • Dunne, T., , and Black R. D. , 1971: Runoff process during snowmelt. Water Resour. Res., 7, 11601172, doi:10.1029/WR007i005p01160.

  • Ebita, A., and et al. , 2011: The Japanese 55-year Reanalysis “JRA-55”: An interim report. SOLA, 7, 149152, doi:10.2151/sola.2011-038.

    • Search Google Scholar
    • Export Citation
  • Field, C. B., and et al. , Eds., 2012: Managing the Risks of Extreme Events and Disasters to Advance Climate Change Adaptation. Cambridge University Press, 582 pp.

  • Gan, T. Y., , Barry R. G. , , Gizaw M. , , Gobena A. , , and Balaji R. , 2013: Changes in North American snowpacks for 1979–2007 detected from the snow water equivalent data of SMMR and SSM/I passive microwave and related climatic factors. J. Geophys. Res. Atmos., 118, 76827697, doi:10.1002/jgrd.50507.

    • Search Google Scholar
    • Export Citation
  • Goulding, H. L., , Prowse T. D. , , and Bonsal B. , 2009: Hydroclimatic controls on the occurrence of break-up and ice-jam flooding in the Mackenzie Delta, NWT, Canada. J. Hydrol., 379, 251267, doi:10.1016/j.jhydrol.2009.10.006.

    • Search Google Scholar
    • Export Citation
  • Government of Manitoba, 2013a: Manitoba 2011 flood review task force report. Rep. to the Minister of Infrastructure and Transportation, 192 pp. [Available online at www.gov.mb.ca/asset_library/en/2011flood/flood_review_task_force_report.pdf.]

  • Government of Manitoba, 2013b: 2011 flood: Technical review of Lake Manitoba, Lake St. Martin and Assiniboine River water levels. Tech. Rep., 123 pp. [Available online at www.gov.mb.ca/mit/floodinfo/floodproofing/reports/pdf/assiniboine_lakemb_lsm_report_nov2013.pdf.]

  • Granger, R. J., , Gray D. M. , , and Dyck G. E. , 1984: Snowmelt infiltration to frozen prairie soils. Can. J. Earth Sci., 21, 669677, doi:10.1139/e84-073.

    • Search Google Scholar
    • Export Citation
  • Gray, D. M., , Landine P. G. , , and Granger R. J. , 1985: Simulating infiltration into frozen Prairie soils in streamflow models. Can. J. Earth Sci., 22, 464472.

    • Search Google Scholar
    • Export Citation
  • Gray, D. M., , Toth B. , , Zhao L. , , Pomeroy J. W. , , and Granger R. J. , 2001: Estimating areal snowmelt infiltration into frozen soils. Hydrol. Processes, 15, 30953111, doi:10.1002/hyp.320.

    • Search Google Scholar
    • Export Citation
  • Hall, J., and et al. , 2013: Understanding flood regime changes in Europe: A state of the art assessment. Hydrol. Earth Syst. Sci. Discuss., 10, 15 52515 624, doi:10.5194/hessd-10-15525-2013.

    • Search Google Scholar
    • Export Citation
  • Hancock, S., , Huntley B. , , Ellis R. , , and Baxter R. , 2014: Biases in re-analysis snowfall found by comparing the JULES land surface model to Globsnow. J. Climate, 27, 624632, doi:10.1175/JCLI-D-13-00382.1.

    • Search Google Scholar
    • Export Citation
  • Hayashi, M., , van der Kamp G. , , and Schmidt R. , 2003: Focused infiltration of snowmelt water in partially frozen soil under small depressions. J. Hydrol., 270, 214229, doi:10.1016/S0022-1694(02)00287-1.

    • Search Google Scholar
    • Export Citation
  • Higgins, R. W., , Silva V. B. S. , , Shi W. , , and Larson J. , 2007: Relationships between climate variability and fluctuations in daily precipitation over the United States. J. Climate, 20, 35613579, doi:10.1175/JCLI4196.1.

    • Search Google Scholar
    • Export Citation
  • Kane, D. L., 1980: Snowmelt infiltration into seasonally frozen soils. Cold Reg. Sci. Technol., 3, 153161, doi:10.1016/0165-232X(80)90020-8.

    • Search Google Scholar
    • Export Citation
  • Kharin, V. V., , Zwiers F. W. , , Zhang X. , , and Wehner M. , 2013: Changes in temperature and precipitation extremes in the CMIP5 ensemble. Climatic Change, 119, 345357, doi:10.1007/s10584-013-0705-8.

    • Search Google Scholar
    • Export Citation
  • King, A. D., , Lewis S. C. , , Perkins S. E. , , Alexander L. V. , , Donat M. G. , , Karoly D. J. , , and Black M. T. , 2013: Limited evidence of anthropogenic influence on the 2011–2012 extreme rainfall over southeast Australia [in “Explaining Extreme Events of 2012 from a Climate Perspective”]. Bull. Amer. Meteor. Soc., 94 (9), S55S58.

    • Search Google Scholar
    • Export Citation
  • Klehmet, K., , Geyer B. , , and Rockel B. , 2013: A regional climate model hindcast for Siberia: analysis of snow water equivalent. Cryosphere, 7, 10171034, doi:10.5194/tc-7-1017-2013.

    • Search Google Scholar
    • Export Citation
  • Komori, D., and et al. , 2012: Characteristics of the 2011 Chao Phraya River flood in central Thailand. Hydrol. Res. Lett., 6, 4146, doi:10.3178/hrl.6.41.

    • Search Google Scholar
    • Export Citation
  • Lavers, D. A., , and Villarini G. , 2013: Were global numerical weather prediction systems capable of forecasting the extreme Colorado rainfall of 9–16 September 2013? Geophys. Res. Lett., 40, 6405–6410, doi:10.1002/2013GL058282.

    • Search Google Scholar
    • Export Citation
  • Lawford, R. G., , Prowse T. D. , , Hogg W. D. , , Warkentin A. A. , , and Pilon P. J. , 1995: Hydro-meteorological aspects of flood hazards in Canada. Atmos.–Ocean, 33, 303328, doi:10.1080/07055900.1995.9649535.

    • Search Google Scholar
    • Export Citation
  • LeComte, D., 2011: U.S. weather highlights of 2010: A year of extremes. Weatherwise, 64, 1320, doi:10.1080/00431672.2011.566815.

  • Lee, S., , Klein A. G. , , and Over T. M. , 2005: A comparison of MODIS and NOHRSC snow-cover products for simulating streamflow using the Snowmelt Runoff Model. Hydrol. Processes, 19, 29512972, doi:10.1002/hyp.5810.

    • Search Google Scholar
    • Export Citation
  • Liebmann, B., , and Smith C. A. , 1996: Description of a complete (interpolated) outgoing longwave radiation dataset. Bull. Amer. Meteor. Soc., 77, 12751277.

    • Search Google Scholar
    • Export Citation
  • Lim, G. H., , and Wallace J. M. , 1991: Structure and evolution of baroclinic waves as inferred from regression analysis. J. Atmos. Sci., 48, 17181732, doi:10.1175/1520-0469(1991)048<1718:SAEOBW>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Loukas, A., , Vasiliades L. , , and Dalezois N. R. , 2000: Flood producing mechanisms identification in southern British Columbia, Canada. J. Hydrol., 227, 218235, doi:10.1016/S0022-1694(99)00182-1.

    • Search Google Scholar
    • Export Citation
  • Mahfouf, J., , Brasnett B. , , and Gagnon S. , 2007: A Canadian Precipitation Analysis (CaPA) Project: Description and preliminary results. Atmos.–Ocean, 45, 117, doi:10.3137/ao.450101.

    • Search Google Scholar
    • Export Citation
  • Marengo, J. S., , Tomasella J. , , Soares W. R. , , Alves L. M. , , and Nobre C. A. , 2012: Extreme climatic events in the Amazon basin. Theor. Appl. Climatol., 107, 7385, doi:10.1007/s00704-011-0465-1.

    • Search Google Scholar
    • Export Citation
  • Mekis, É., , and Hogg W. D. , 1999: Rehabilitation and analysis of Canadian daily precipitation time series. Atmos.–Ocean, 37, 5385, doi:10.1080/07055900.1999.9649621.

    • Search Google Scholar
    • Export Citation
  • Milrad, S. M., , Atallah H. E. , , and Gyakum J. R. , 2013: Precipitation modulation by the Saint Lawrence River valley in association with transitioning tropical cyclones. Wea. Forecasting, 28, 331352, doi:10.1175/WAF-D-12-00071.1.

    • Search Google Scholar
    • Export Citation
  • Peters, D. L., , Prowse T. D. , , Pietroniro A. , , and Leconte R. , 2006: Flood hydrology of the Peace-Athabasca Delta, northern Canada. Hydrol. Processes, 20, 40734096, doi:10.1002/hyp.6420.

    • Search Google Scholar
    • Export Citation
  • Phillips, D., 2012: Top 10 Canadian weather stories for 2011. CMOS Bulletin, Vol. 40, No. 1, Canadian Meteorological and Oceanographic Society, Ottawa, ON, Canada, 1222. [Available online at www.cmos.ca/uploaded/web/members/Bulletin/Vol_40/b4001.pdf.]

  • Rannie, W. F., 2002: The role of the Assiniboine River in the 1826 and 1852 Red River floods. Prairie Perspect.,5, 56–75. [Available online at http://pcag.uwinnipeg.ca/Prairie-Perspectives/PP-Vol05/Rannie.pdf.]

  • Roy, L., , Leconte R. , , Brissette F. P. , , and Marche C. , 2001: The impact of climate change on seasonal floods of a southern Quebec River basin. Hydrol. Processes, 15, 31673179, doi:10.1002/hyp.323.

    • Search Google Scholar
    • Export Citation
  • Saint-Laurent, D., , Mesfioui M. , , and Evin G. , 2009: Hydroclimatic variability and relation with flood events (southern Quebec, Canada). Water Resour., 36, 4356, doi:10.1134/S0097807809010047.

    • Search Google Scholar
    • Export Citation
  • Satyamurty, P., , da Costa C. P. W. , , Manzi A. O. , , and Candido L. A. , 2013: A quick look at the 2012 record flood in the Amazon basin. Geophys. Res. Lett., 40, 13961401, doi:10.1002/grl.50245.

    • Search Google Scholar
    • Export Citation
  • Shanley, J. B., , and Chalmers A. , 1999: The effect of frozen soil on snowmelt runoff at Sleepers River, Vermont. Hydrol. Processes, 13, 18431857, doi:10.1002/(SICI)1099-1085(199909)13:12/13<1843::AID-HYP879>3.0.CO;2-G.

    • Search Google Scholar
    • Export Citation
  • Shaw, D. A., , van der Kamp G. , , Conly F. M. , , Pietroniro A. , , and Martz L. , 2012: The fill–spill hydrology of Prairie wetland complexes during drought and deluge. Hydrol. Processes, 26, 3147–3156, doi:10.1002/hyp.8390.

    • Search Google Scholar
    • Export Citation
  • Sillmann, J., , Kharin V. V. , , Zwiers F. W. , , Zhang X. , , and Bronaugh D. , 2013: Climate extremes indices in the CMIP5 multimodel ensemble: Part 2. Future climate projections. J. Geophys. Res. Atmos., 118, 24732493, doi:10.1002/jgrd.50188.

    • Search Google Scholar
    • Export Citation
  • Simmons, A. J., , Poli P. , , Dee D. P. , , Berrisford P. , , Hersbach H. , , Kobayashib S. , , and Peubey C. , 2014: Estimating low-frequency variability and trends in atmospheric temperature using ERA-Interim. Quart. J. Roy. Meteor. Soc.,140, 329–353, doi:10.1002/qj.2317.

  • Spry, C. M., , Kohfeld K. E. , , Allen D. M. , , Dunkley D. , , and Lertzman K. , 2014: Characterizing pineapple express storms in the lower mainland of British Columbia, Canada. Can. Water Resour. J., 39, 302323, doi:10.1080/07011784.2014.942574.

    • Search Google Scholar
    • Export Citation
  • Szeto, K. K., 2007: Assessing water and energy budgets for the Saskatchewan River basin. J. Meteor. Soc. Japan, 85A, 167186, doi:10.2151/jmsj.85A.167.

    • Search Google Scholar
    • Export Citation
  • Takala, M., , Luojus K. , , Pulliainen J. , , Derksen C. , , Lemmetyinen J. , , and Kärnä J.-P. , 2011: Estimating Northern Hemisphere snow water equivalent for climate research through assimilation of space borne radiometer and ground-based measurements. Remote Sens. Environ., 115, 35173529, doi:10.1016/j.rse.2011.08.014.

    • Search Google Scholar
    • Export Citation
  • Trenberth, K. E., , and Fasullo J. T. , 2013: North American water and energy cycles. Geophys. Res. Lett., 40, 365369, doi:10.1002/grl.50107.

    • Search Google Scholar
    • Export Citation
  • van der Kamp, G., , and Hayashi M. , 2009: Groundwater–wetland ecosystem interaction in the semiarid glaciated plains of North America. Hydrogeol. J., 17, 203214, doi:10.1007/s10040-008-0367-1.

    • Search Google Scholar
    • Export Citation
  • van Dijk, A. I. J. M., , Renzull L. J. , , Wada Y. , , and Tregoning P. , 2013: A global water cycle reanalysis (2003–2012) reconciling satellite gravimetry and altimetry observations with a hydrological model ensemble. Hydrol. Earth Syst. Sci. Discuss., 10, 15 47515 523, doi:10.5194/hessd-10-15475-2013.

    • Search Google Scholar
    • Export Citation
  • Vincent, L. A., 1998: A technique for the identification of inhomogeneities in Canadian temperature series. J. Climate, 11, 10941104, doi:10.1175/1520-0442(1998)011<1094:ATFTIO>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Vincent, L. A., , and Gullett D. W. , 1999: Canadian historical and homogeneous temperature datasets for climate change analyses. Int. J. Climatol., 19, 13751388, doi:10.1002/(SICI)1097-0088(199910)19:12<1375::AID-JOC427>3.0.CO;2-0.

    • Search Google Scholar
    • Export Citation
  • Vincent, L. A., , and Mekis É. , 2006: Changes in daily and extreme temperature and precipitation indices for Canada over the twentieth century. Atmos.–Ocean, 44, 177193, doi:10.3137/ao.440205.

    • Search Google Scholar
    • Export Citation
  • Vincent, L. A., , van Wijngaarden W. A. , , and Hopkinson A. , 2007: Surface temperature and humidity trends in Canada for 1953–2005. J. Climate, 20, 51005113, doi:10.1175/JCLI4293.1.

    • Search Google Scholar
    • Export Citation
  • Zhang, X., , Vincent L. A. , , Hogg W. D. , , and Niitsoo A. , 2000: Temperature and precipitation trends in Canada during the 20th century. Atmos.–Ocean, 38, 395429, doi:10.1080/07055900.2000.9649654.

    • Search Google Scholar
    • Export Citation
  • Zhang, X., , Alexander L. , , Hegerl G. C. , , Jones P. , , Tank A. , , Peterson T. C. , , Trewin B. , , and Zwiers F. W. , 2011: Indices for monitoring changes in extremes based on daily temperature and precipitation data. Wiley Interdiscip. Rev.: Climate Change, 2, 851870, doi:10.1002/wcc.147.

    • Search Google Scholar
    • Export Citation
  • Zhou, T., , Song F. , , Lin R. , , Chen X. , , and Chen X. , 2013: The 2012 North China floods: explaining an extreme rainfall event in the context of a longer-term drying tendency [in “Explaining Extreme Events of 2012 from a Climate Perspective”]. Bull. Amer. Meteor. Soc., 94 (9), S49S51.

    • Search Google Scholar
    • Export Citation
  • View in gallery

    Location of the ARB (red boundary) study area. Locations of stream gauges and cities are shown by colored symbols. Numbers represent the Assiniboine River (1), Qu’Appelle River (2), and Souris River (3).

  • View in gallery

    Monthly discharge (dam3) at (a) Wawanesa, (b) Brandon, and (c) Headingley for 2010 and 2011. The normal values are for 1981–2010, whereas the monthly record values are the max values observed in the record before 2011.

  • View in gallery

    Accumulated annual discharge (dam3) at Headingley, located near the terminus of the ARB, for 1913–2011. The discharge at Holland, upstream of the Portage diversion, is shown from 1968 onward. After 1970, the annual discharge at Headingley was calculated by adding the discharge from the Portage diversion.

  • View in gallery

    Accumulated annual discharge (dam3) by month for 2011 at Headingley. Also shown are the accumulated annual discharges for the 1976 and 1955 floods, and the mean-adjusted (1981–2010) accumulated discharge at Headingley.

  • View in gallery

    Basin-averaged CANGRD monthly departures (from the 1971–2000 mean) for (a) precipitation and (b) temperature for the ARB between May 2010 and December 2011.

  • View in gallery

    Monthly GlobSnow basin-averaged SWE for 1980–2011.

  • View in gallery

    Weekly GlobSnow SWE distribution over the ARB for selected weeks ending on (a) 1 Feb, (b) 1 Mar, (c) 15 Mar, (d) 1 Apr, (e) 7 Apr, and (f) 15 Apr.

  • View in gallery

    Standardized 0–47-cm soil moisture anomalies of basin-averaged soil moisture from JRA-55 relative to the 1976–2011 climatology.

  • View in gallery

    Gridpoint counts for GE10 and GE25 events for the (a),(b) 2010 and (c),(d) 2011 warm seasons (from 1 May through 30 Sep) from the CaPA data.

  • View in gallery

    Relative contribution of GE10 and GE25 events to warm-season rainfall for (a),(b) 2010 and (c),(d) 2011 from the CaPA data.

  • View in gallery

    Gridpoint counts for GE10 events during the 2010/11 cold season (from 1 Oct through 30 Apr) from the CaPA data.

  • View in gallery

    Accumulated daily precipitation at Estevan for the water year starting on 1 Sep. Also shown are the accumulated precipitation amounts for the 1954–55, 1975–76, and climatological (1971–2000) mean water years.

  • View in gallery

    Gridcell counts for GE10 and GE25 events for (a),(b) May and (c),(d) June 2011from the CaPA data.

  • View in gallery

    Monthly maps for May 2011 for (a) CANGRD precipitation departure from normal (%), (b) mean 500-mb heights (gpm), (c) 500-mb height anomalies (relative to 1981–2010 climatology), (d) 500-mb vertical motion (Pa s−1), (e) mean sea level pressure (mb), and (f) 250-mb zonal wind anomalies (m s−1). Data in (b)–(f) are from ERA-I.

  • View in gallery

    As in Fig. 14, but for June 2011.

  • View in gallery

    Hovmöller diagram of daily remotely sensed OLR anomalies from NOAA for a domain extending from 45° to 55°N and from 120°W to 80°E between 20 Apr and 30 Jun 2011. Solid black vertical lines indicate the approximate position of the ARB. Red rectangles indicate timing of heavy precipitation events.

  • View in gallery

    Accumulated amounts (mm) for the surface water budget terms from JRA-55 from 1 Mar through 30 Jun for (a) 1980–2011 mean, (b) 1976, (c) 1995, and (d) 2011.

  • View in gallery

    Timeline of significant events relevant to the 2011 ARB flood. Rankings for precipitation are relative to records going back to 1900 and for soil moisture and SWE are relative to records going back to 1979. Discharge records go back to 1906, 1913, and 1974 at Wawanesa, Headingley, and Brandon, respectively.

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 78 78 7
PDF Downloads 68 68 3

Hydroclimatic Aspects of the 2011 Assiniboine River Basin Flood

View More View Less
  • 1 * University of Manitoba, Winnipeg, Manitoba, and Environment Canada, Edmonton, Alberta, Canada
  • | 2 Environment Canada, Downsview, Ontario, Canada
  • | 3 Environment Canada, Saskatoon, Saskatchewan, Canada
  • | 4 University of Manitoba, Winnipeg, Manitoba, Canada
  • | 5 Environment Canada, Edmonton, Alberta, Canada
  • | 6 McMaster University, Hamilton, Ontario, Canada
© Get Permissions
Full access

Abstract

In the spring and early summer of 2011, the Assiniboine River basin in Canada experienced an extreme flood that was unprecedented in terms of duration and severity. The flood had significant socioeconomic impacts and caused over $1 billion (Canadian dollars) in damage. Contrary to what one might expect for such an extreme flood, individual precipitation events before and during the 2011 flood were not extreme; instead, it was the cumulative impact and timing of precipitation events going back to the summer of 2010 that played a key role in the 2011 flood. The summer and fall of 2010 were exceptionally wet, resulting in above-normal soil moisture levels at the time of freeze-up. This was followed by record high snow water equivalent values in March and April 2011. Cold temperatures in March delayed the spring melt, resulting in the above-average spring freshet occurring close to the onset of heavy rains in May and June. The large-scale atmospheric flow during May and June 2011 favored increased cyclone activity in the region, which produced an anomalously large number of heavy rainfall events over the basin. All of these factors combined generated extreme flooding. Japanese 55-year Reanalysis Project (JRA-55) data are used to quantify the relative importance of snowmelt and spring precipitation in contributing to the unprecedented flood and to demonstrate how the 2011 flood was unique compared to previous floods. This study can be used to validate and improve flood forecasting techniques over this important basin; the findings also raise important questions regarding floods in a changing climate over basins that experience pluvial and nival flooding.

Denotes Open Access content.

Corresponding author address: Julian Brimelow, Environment Canada, Eastgate Offices, 9250 49th Street, Edmonton, AB T6B 1K5, Canada. E-mail: julian.brimelow@ec.gc.ca

Abstract

In the spring and early summer of 2011, the Assiniboine River basin in Canada experienced an extreme flood that was unprecedented in terms of duration and severity. The flood had significant socioeconomic impacts and caused over $1 billion (Canadian dollars) in damage. Contrary to what one might expect for such an extreme flood, individual precipitation events before and during the 2011 flood were not extreme; instead, it was the cumulative impact and timing of precipitation events going back to the summer of 2010 that played a key role in the 2011 flood. The summer and fall of 2010 were exceptionally wet, resulting in above-normal soil moisture levels at the time of freeze-up. This was followed by record high snow water equivalent values in March and April 2011. Cold temperatures in March delayed the spring melt, resulting in the above-average spring freshet occurring close to the onset of heavy rains in May and June. The large-scale atmospheric flow during May and June 2011 favored increased cyclone activity in the region, which produced an anomalously large number of heavy rainfall events over the basin. All of these factors combined generated extreme flooding. Japanese 55-year Reanalysis Project (JRA-55) data are used to quantify the relative importance of snowmelt and spring precipitation in contributing to the unprecedented flood and to demonstrate how the 2011 flood was unique compared to previous floods. This study can be used to validate and improve flood forecasting techniques over this important basin; the findings also raise important questions regarding floods in a changing climate over basins that experience pluvial and nival flooding.

Denotes Open Access content.

Corresponding author address: Julian Brimelow, Environment Canada, Eastgate Offices, 9250 49th Street, Edmonton, AB T6B 1K5, Canada. E-mail: julian.brimelow@ec.gc.ca

1. Introduction

During the spring and summer of 2011, the Assiniboine River basin (ARB; Fig. 1) in Canada experienced an unprecedented and devastating flood. The flood was exceptional in terms of both the volume of water and its longevity, and it was rated as Canada’s number one weather event for 2011 (Phillips 2012).

Fig. 1.
Fig. 1.

Location of the ARB (red boundary) study area. Locations of stream gauges and cities are shown by colored symbols. Numbers represent the Assiniboine River (1), Qu’Appelle River (2), and Souris River (3).

Citation: Journal of Hydrometeorology 16, 3; 10.1175/JHM-D-14-0033.1

The impacts of the 2011 flood over the ARB were widespread and costly. The Canadian Wheat Board estimated that almost 5.5 million ha of farmland did not produce crops in 2011 because of flooding. Additionally, critical infrastructure (such as roads and bridges) sustained significant damage, and over 7000 people were displaced (Government of Manitoba 2013a). Impacts of the flood are ongoing at the time of writing, with thousands of people still displaced, and the current accumulated cost is estimated at $1.2 billion (Canadian dollars; Government of Manitoba 2013a).

Floods are not uncommon in the ARB, and at least 12 major events have been documented since official records started in 1913 (Government of Manitoba 2013b; Rannie 2002). Records indicate that the largest flood over the ARB prior to 2011 occurred in 1976. The unique and unprecedented nature of the 2011 flood is underscored by the fact that the annual discharge measured at the city of Brandon, Manitoba (Fig. 1), for 2011 was higher than the discharge from the entire basin for the year of the 1976 flood; the flow at Brandon typically represents only 55% of the basin’s discharge. In addition, the Assiniboine River at Brandon was at flood stage for a record 110 days in 2011, and the Portage diversion was operated for a record 126 days and was required to function above its design capacity for a record 31 days between early May and late June 2011 (Government of Manitoba 2013b).

Extreme floods occur because of numerous factors that can act individually or in concert. On the Canadian Prairies, April and May are the most common months for flooding, due to either runoff from snowmelt (i.e., nival) or heavy precipitation (i.e., pluvial) from large-scale systems, or a combination thereof (Lawford et al. 1995). For river basins with seasonal freezing of the soil (such as the ARB), snow accumulation and snowmelt play a key role in modulating spring flood potential (e.g., Lee et al. 2005). Spring floods are also affected by net solar radiation, soil state [frozen or not; soil moisture content as discussed by, e.g., Gray et al. (1985)], intensity and amount of precipitation, and timing and rate of snowmelt (e.g., Shanley and Chalmers 1999). Flooding during the summer months tends to be localized and associated with convective rainfall, especially over saturated soils.

The preponderance of extreme floods in recent years in North America (e.g., Lavers and Villarini 2013), Europe (e.g., Blöschl et al. 2013), South America (e.g., Satyamurty et al. 2013; Marengo et al. 2012), Asia (e.g., Zhou et al. 2013; Komori et al. 2012), and Australia (e.g., Christidis et al. 2013; King et al. 2013) has led to concern that floods are increasing globally as the hydrological cycle accelerates in response to anthropogenic global warming (Hall et al. 2013). The hydroclimate over the temperate regions of North America is also changing in ways that may affect spring floods in the ARB (Brown and Mote 2009; Vincent et al. 2007; Vincent and Mekis 2006). Exactly how the spring streamflow and flooding potential over the ARB will respond to future changes in the hydroclimate is unclear, because there is uncertainty as to how the key variables relevant to severe flooding identified above will change (Field et al. 2012). For example, Diffenbaugh et al. (2013) found that some high-latitude regions, despite increasing cold-season temperatures, will remain cold enough to ensure high March snow water equivalent (SWE) values on account of increasing winter snowfall, thereby leading to increased spring runoff. In contrast, a decrease in winter snowpack in regions where the temperature response overwhelms the increase in winter precipitation would reduce spring runoff (e.g., Dankers and Feyen 2009). It may be that the response of individual variables is not as critical as how the sequence of events, or interaction between variables, may play a role in muting or exacerbating a flood. Consequently, it is important to undertake studies of significant events such as the 2011 flood to identify the key factors at play and their interactions.

Despite the aforementioned issues and the socioeconomic costs associated with flooding over the ARB, to the authors’ knowledge, no previous peer-reviewed article has undertaken a detailed hydroclimatic analysis of floods over the ARB or similar basins in Canada. Studies have focused on flooding associated with ice jams (e.g., Goulding et al. 2009; Peters et al. 2006; Beltaos et al. 1996), rainfall-driven floods (e.g., Spry et al. 2014; Roy et al. 2001), or spring snowmelt floods (e.g., Saint-Laurent et al. 2009; Loukas et al. 2000). The results of this study will address the aforementioned knowledge gap, will be of value to stakeholders in the basin, and may also be applicable to other basins globally that experience pluvial and/or nival floods (e.g., the Ural and Don River basins).

The 2011 flood is well suited for a detailed analysis because of the wealth of available data. We used these to fulfill the following two objectives: 1) to characterize and quantify the hydroclimatic conditions responsible for the 2011 flood and 2) to determine why the 2011 flood was so dramatically different from previous floods in the ARB that were preceded by similar conditions such as heavy winter snowpack or heavy spring rains. To this end, surface and atmospheric conditions observed prior to and during the 2011 flood will be characterized and, where possible, placed in historical context.

The paper is organized as follows. Section 2 outlines the study area and datasets. In section 3, the hydrologic aspects of the ARB and the 2011 flood are discussed and compared with previous major floods. Section 4 documents the hydroclimatic factors prior to and during the flood, including the spatiotemporal variability of the surface precipitation and temperature, SWE, and soil moisture. The occurrence and distribution of heavy and very heavy precipitation events are reviewed in section 5. Section 6 examines the exceptionally wet spring of 2011 and attendant large-scale atmospheric forcing, and section 7 investigates the surface water balance to quantify the relative contributions of snowmelt and precipitation to the flood and to determine what made the 2011 flood so extreme and unique. The summary and conclusions are provided in section 8.

2. Study area and data

The ARB encompasses an area of 162 000 km2 over southeastern Saskatchewan, southwestern Manitoba, and northern North Dakota (Fig. 1). The main basin consists of three subbasins: the Souris River (62 000 km2), the Qu’Appelle River (58 000 km2) and the Assiniboine River (42 000 km2). Land use is dominated by annual cropland, with patchy mixed evergreen–deciduous forest found over the extreme eastern portions of the Assiniboine subbasin. The elevation declines from ~640 m near the source to 230 m at Headingley. The ARB is located within the Prairie Pothole Region, which was formed by glaciation during the Pleistocene epoch. The potholes exert an important influence on the basin’s water balance (and peak runoff) on account of their large potential storage capacity (Hayashi et al. 2003). Unlike more “traditional” river basin drainage systems, these potholes cause the ARB’s contributing area to vary from year to year, thereby complicating the basin’s hydrology (e.g., Shaw et al. 2012). Furthermore, the potholes do not always contribute to the drainage network, but under extreme wet conditions can do so by connecting to streams through the so-called “fill and spill” mechanism (e.g., van der Kamp and Hayashi 2009).

The ARB receives 400–500 mm of precipitation annually (1971–2000 baseline), of which about 65%–70% falls in the warm season (May–September; http://climate.weather.gc.ca). June is typically the wettest month (basin-averaged precipitation of almost 80 mm), whereas February is the driest (only 15 mm of water-equivalent precipitation). Mean annual accumulated snowfall over the basin is near 100 cm.

To accomplish the objectives of this study, an analysis of several datasets was required. These included precipitation and SWE data, in addition to atmospheric variables. Identifying suitable datasets to quantify the spatial and temporal variability of key hydrometeorological variables at the spatial and temporal scales relevant for this study (i.e., from subdaily through decadal time scales) required drawing upon a diverse range of datasets, as no single one met all the requirements. With this in mind, the following products were chosen because of their suitability for investigating particular aspects of the ARB’s hydroclimate.

Precipitation and temperature data from the 50-km Canadian Gridded Temperature and Precipitation (CANGRD) data product (Zhang et al. 2000) were used at monthly and seasonal time scales, whereas precipitation data (on a 0.2° grid) from the Canadian Precipitation Analysis (CaPA; Mahfouf et al. 2007) project were used for daily and subdaily time scales. CANGRD temperature and precipitation values are derived from climate station data that have been adjusted for homogeneity issues caused by station relocation and changes to instrumentation and observing times (Vincent and Gullett 1999; Mekis and Hogg 1999; Vincent 1998). In addition, precipitation data have been adjusted for gauge undercatch, wetting loss, and trace events (Mekis and Hogg 1999). The CaPA product has proven valuable in recent hydrometeorological research in Canada (e.g., Milrad et al. 2013; Deacu et al. 2012). Organized precipitation systems over or near the ARB were tracked using NOAA’s interpolated outgoing longwave radiation (OLR) product (Liebmann and Smith 1996). Previously, Bao et al. (2005) used OLR data in Ontario, Canada, for a similar purpose.

SWE data from GlobSnow (Takala et al. 2011) were used to track the temporal and spatial evolution of the snowpack. GlobSnow is a hybrid product, in that it assimilates spaceborne passive radiometer and ground-based measurements. A comparison between GlobSnow SWE data and those from the Canadian Meteorological Centre showed excellent agreement between 1999 and 2011. GlobSnow data were selected because they are available for the entire satellite era (since 1979) and are superior to other pure Earth-observation-derived products for monitoring the seasonal SWE cycle (Hancock et al. 2014). Furthermore, GlobSnow data have been used widely for hydrometeorological research (e.g., van Dijk et al. 2013). Snow cover extent data from the Rutgers Global Snow Lab (http://climate.rutgers.edu/snowcover/) were also incorporated. Streamflow data were extracted from the archived hydrometric dataset maintained by the Water Survey of Canada (http://wateroffice.ec.gc.ca/index_e.html). Lightning data were from the Canadian Lightning Detection Network (Burrows and Kochtubajda 2010).

To explore the physical processes associated with the 2011 flood, atmospheric variables including mean sea level pressure, vertical motion, winds, and geopotential heights from the state-of-the-art ERA-Interim (hereafter ERA-I) product (Dee et al. 2011) were analyzed. ERA-I has been used extensively for hydroclimatic studies (e.g., Klehmet et al. 2013; Trenberth and Fasullo 2013) but has the disadvantage that it covers only the satellite era. To address this, the recently released Japanese 55-year Reanalysis Project (JRA-55; Ebita et al. 2011) product was selected to investigate the surface water budget over the ARB because it extends back to 1958, thereby allowing us to compare the 2011 flood with those occurring prior to the satellite era. Chen et al. (2014) found that, over East Asia, the JRA-55 was comparable (or superior to) other reanalysis products in reproducing the diurnal cycle. Simmons et al. (2014) noted that JRA-55 atmospheric temperatures compared well with those from ERA-I. More importantly, the suitability of using JRA-55 for this research was assessed by comparing the JRA-55 terrestrial water budget over the ARB (between 1979 and 2012) with budgets from other products used in hydroclimatic studies (e.g., ERA-I, GLDAS-VIC, and GLDAS-Noah). This analysis revealed that, while the annual residual for JRA-55 (~70 mm yr−1) was larger than those for the GLDAS products, it was smaller than the residual for ERA-I (not shown). The JRA-55 also successfully captured the seasonal cycle of the key water budget terms.

3. 2011 hydrometric data in context

The streamflow of rivers in the ARB follows a clear seasonal cycle—increasing rapidly after February, reaching a peak in April or May during the spring freshet, and decreasing rapidly thereafter (Fig. 2). The spring peak is primarily in response to runoff from snowmelt, although the peak daily flow for the year can occur after the spring peak, in response to heavy rains.

Fig. 2.
Fig. 2.

Monthly discharge (dam3) at (a) Wawanesa, (b) Brandon, and (c) Headingley for 2010 and 2011. The normal values are for 1981–2010, whereas the monthly record values are the max values observed in the record before 2011.

Citation: Journal of Hydrometeorology 16, 3; 10.1175/JHM-D-14-0033.1

To prevent frequent flooding along the Assiniboine River downstream of Portage la Prairie, the Portage diversion (completed in 1970) was constructed to divert water to Lake Manitoba. The amount of water diverted at Portage varies from year to year, with an average of ~325 000 cubic decameters (dam3; 1 dam3 = 1000 m3) per year between 1970 and 2010.

The degree and duration of flooding in 2011 was unprecedented in records extending back to the early twentieth century (see Fig. 3). This is also underscored by the fact that in 2011 the Portage diversion operated for a record 126 days and diverted 5.83 million dam3 of water. This is exceptional when compared to the previous maximum of 65 days and 1.72 million dam3 associated with the 1976 flood (Government of Manitoba 2013b).

Fig. 3.
Fig. 3.

Accumulated annual discharge (dam3) at Headingley, located near the terminus of the ARB, for 1913–2011. The discharge at Holland, upstream of the Portage diversion, is shown from 1968 onward. After 1970, the annual discharge at Headingley was calculated by adding the discharge from the Portage diversion.

Citation: Journal of Hydrometeorology 16, 3; 10.1175/JHM-D-14-0033.1

The availability of hydrometric data for river gauges discussed here varies considerably. To facilitate comparison with data among gauges in the basin, statistics in Table 1 are based only from 1974 to 2013.

Table 1.

Streamflow statistics for three key gauges (Wawanesa, Brandon, and Holland) in the ARB. Statistics were calculated for the 40-yr period from 1974 to 2013. Return period estimates for the peak daily flow in 2011 are from Government of Manitoba (2013b). Max values are those that were observed in 2011.

Table 1.

The Souris subbasin accounts for almost 40% of the ARB’s area and typically contributes ~30% to the ARB’s annual discharge. The mean peak daily flow measured at the hydrometric station at Wawanesa (on the Souris River) is ~140 m3 s−1. Discharge between April and July contributes ~80% of the annual discharge (Table 1). In 2010, the flow tapered off more slowly than usual because of late summer rain (Fig. 2a). Discharge records were set for every month between May and October 2011; the new records were between 10 and 30 times greater than their respective long-term mean monthly values. The record peak daily flow of ~790 m3 s−1, set in early July 2011, was almost over 5 times higher than the mean peak value, and it occurred much later in the year than did the previous record peak daily flow of ~740 m3 s−1, in April 1976. The expected return period for the peak daily flow recorded in 2011 is ~130 years (Government of Manitoba 2013b). The cumulative discharge for 2011 was ~5.7 million dam3, which is almost 10 times greater than the 1974–2013 mean. In 2011, the Souris basin contributed ~45% to the ARB’s annual discharge (at Holland), which is about 20% greater than the average.

The hydrometric station at Brandon measures the cumulative discharge from the middle and upper Assiniboine River subbasin and the Qu’Appelle subbasin (55%, or ~89 000 km2, of the ARB’s area). The mean peak daily flow for 1974–2013 is ~220 m3 s−1, with about two-thirds of the annual discharge typically occurring from April through July (Table 1). New discharge records were set at Brandon from September 2010 onward (Fig. 2b), and the abnormally high flow continued through the winter. Records were set in January and February 2011 and in May–August 2011. In May 2011, a new record discharge for any month was set at 2.1 million dam3 (previous record was 1.3 million dam3, in May 1995). The record peak daily flow of ~1280 m3 s−1 set in May 2011 was about 6 times higher than the mean value. The return period for the peak daily flow recorded in the 2011 flood is ~250 years (Government of Manitoba 2013b). In 2011, the flow at Brandon contributed ~6.6 million dam3 (or 50%) to the ARB’s annual discharge that year. This was about 5 times greater than the mean annual discharge for this site, and higher than that measured for the entire ARB during the year of the 1976 floods (~5.5 million dam3).

Hydrometric data from Holland, rather than from Headingley, are discussed here to avoid complications of using peak daily flow data from Headingley because it lies below the Portage diversion. Holland represents the flow from 160 000 km2, or almost 99%, of the ARB. The mean peak daily flow at Holland is near 355 m3 s−1, with a mean April–July total discharge of 1.62 million dam3 contributing ~70% to the annual discharge (Table 1). New discharge records were set in August and September 2010 (Fig. 2c), and all-time highs were set in October–December 2010. Monthly discharge amounts started off relatively high in 2011, and in May a record high of ~1.4 million dam3 was set (the previous record was 1.3 million dam3 in May 1923). New monthly records were also set between June and September 2011, with records in August and September superseding those set in 2010. The cumulative discharge at Holland in 2011 was ~13 million dam3, which is about double the previous record observed during the 1976 floods, and 3 times greater than observed during the year of the 1995 flood (Fig. 4). Furthermore, the accumulated discharge in 2011 was over 5 times greater than the mean annual discharge. The peak daily flow at Holland in 2011 had an estimated return period of 230 years.

Fig. 4.
Fig. 4.

Accumulated annual discharge (dam3) by month for 2011 at Headingley. Also shown are the accumulated annual discharges for the 1976 and 1955 floods, and the mean-adjusted (1981–2010) accumulated discharge at Headingley.

Citation: Journal of Hydrometeorology 16, 3; 10.1175/JHM-D-14-0033.1

4. Contributing hydrometeorological factors

a. Precipitation and temperature

Both 2010 and 2011 were exceptionally wet over the ARB. In particular, basin-averaged CANGRD precipitation data reveal that 2010 was the wettest year since 1900. The summer of 2010 [June–August (JJA)] was the third wettest, followed by the fifth-wettest fall [September–November (SON)] on record. The water year between September 2010 and August 2011 was the fifth wettest on record.

The period between May 2010 and August 2011 was characterized by persistent moist conditions over most of the ARB (Fig. 5a), with only one month (March 2011) recording below-average precipitation. Also, the accumulated precipitation from September 2010 to June 2011 was the highest since 1900. The 2010 growing season (April–September) was associated with exceptionally high precipitation (Fig. 5a), with basin-averaged accumulated precipitation of 508 mm, nearly 160% above normal. May 2010 ranked as the seventh-wettest May since 1900, with large portions of the ARB receiving >200% of average precipitation (52 mm). The wet conditions persisted into June when basin-averaged precipitation of 115 mm was near 150% above normal. Conditions dried out somewhat in July 2010, but precipitation was still near-to-above average over most of the basin. Exceptionally wet conditions returned in August and September 2010, with precipitation over 200% above average over parts of the ARB. Except for the far western ARB, October 2010 was also wet, with portions of the eastern half of the basin receiving 150%–200% of normal precipitation. A heavy rain event between 25 and 27 October (storm totals of 60–70 mm) contributed to the high rainfall. This event was associated with an extremely deep depression over Minnesota and Wisconsin that set a new record for the lowest central pressure (955 hPa) over the contiguous United States (LeComte 2011) on 26 October.

Fig. 5.
Fig. 5.

Basin-averaged CANGRD monthly departures (from the 1971–2000 mean) for (a) precipitation and (b) temperature for the ARB between May 2010 and December 2011.

Citation: Journal of Hydrometeorology 16, 3; 10.1175/JHM-D-14-0033.1

In November, apart from drier conditions over the narrow arm of the basin between Brandon and Winnipeg (hereafter referred to as the panhandle), the remainder of the basin received well-above-normal precipitation, with the basin average of 37.5 mm (~200% above average). Precipitation was also well above average in January 2011, with a basin average of 32 mm (~160% above normal), making it the seventh-wettest January since 1900. In February 2011, basin-averaged precipitation of 17 mm was ~115% above average, whereas in March 2011 drier conditions (80% of average) were observed. As was the case in 2010, the 2011 growing season started out wet (Fig. 5a). In April and May, area-averaged precipitation amounts were 130% and 170% above average, respectively; however, the northern third of the basin was persistently dry (<60% of average precipitation). Wet conditions were observed basinwide in June, with a mean of 114 mm (near 150% above average). Starting in July 2011, there was a transition to relatively drier conditions that continued into the autumn. The panhandle area was especially dry in July, and Winnipeg experienced its driest July (only 10 mm of rain) since records began in 1887.

Large departures in temperature were also observed (Fig. 5b). Between May 2010 and January 2011, basin-averaged temperatures were near normal (±1°C). Two exceptions were May 2010, when temperatures were 1.5°C below average, and October 2010, when temperature anomalies were near +3°C. February–May 2011 experienced well-below-average temperatures (at least 1°C below normal), especially in March, when basin-averaged anomalies were below −4°C. Near-average temperatures in June 2011 transitioned to above-average temperatures into the autumn and early winter. Thus, the prolonged wet period was not associated with persistent warm or cold temperature anomalies. The importance of the spring 2011 temperatures is discussed in sections 4b,c and 7.

b. Snow water equivalent

The high SWE values of the winter snowpack and rapid spring melt played an important role in the ARB flood. Basin-averaged SWE in January 2011 was near 65 mm, which was ~20 mm above the 1981–2010 average (Fig. 6). By February the SWE had increased slightly to 70 mm, which was ~25 mm above average and the third-highest value for February in the satellite record. Although the basin-averaged SWE decreased slightly to 65 mm in March, this was 40 mm above average and was the highest SWE for March since 1980 (Fig. 6). The mean SWE in April was ~20 mm—this too was the highest value in the satellite record. Weekly data (Fig. 7) indicate that, by the end of January 2011, SWE decreased from a maximum near 120 mm in the northeast to 50–60 mm in the far southwest. From 1 February through mid-March, the distribution and basin-averaged SWE changed very little, with basin averages near 70 mm. Before basinwide melt commenced in early April, peak SWE values were >130 mm over the northeastern ARB.

Fig. 6.
Fig. 6.

Monthly GlobSnow basin-averaged SWE for 1980–2011.

Citation: Journal of Hydrometeorology 16, 3; 10.1175/JHM-D-14-0033.1

Fig. 7.
Fig. 7.

Weekly GlobSnow SWE distribution over the ARB for selected weeks ending on (a) 1 Feb, (b) 1 Mar, (c) 15 Mar, (d) 1 Apr, (e) 7 Apr, and (f) 15 Apr.

Citation: Journal of Hydrometeorology 16, 3; 10.1175/JHM-D-14-0033.1

Analysis of SWE data from 1980 to 2011 indicates that the snowmelt onset displays marked interannual variability on the Prairies. SWE values typically peak in late February, with the snowpack usually undergoing rapid melting from late March onward. The well-below-normal temperatures (−4°C) in March 2011, however, delayed the thaw. Specifically, daily mean temperature data at Estevan and daily streamflow data suggest that the onset of the thaw in 2011 was delayed by about 2 weeks (i.e., until 5–10 April).

Snowmelt first occurred in the south and over the panhandle and then spread to the west and north. By the end of the first week of April, rapid melt had greatly reduced the snowpack, with maximum SWE values >70 mm present only over the northeastern ARB; elsewhere, SWE values were typically <40 mm (Fig. 7). By 15 April, bare ground was present over parts of the basin and SWE values were <20 mm over large areas. Daily snow extent data from the Rutgers Global Snow Lab indicate that by 22 April 2011 the basin was essentially snow-free. Despite its delayed onset, the spring thaw was rapid and the basin was snow-free only about 3 days later than the 1971–2000 mean.

c. Soil moisture

Soil moisture data from JRA-55 (1976–2011) were used to calculate basin-averaged soil moisture anomalies for the top 47 cm of the soil column. Focus is placed on standardized anomalies for three key periods: September–November (prior to freeze-up), February–March (the period before snowmelt commenced), and May–April (Fig. 8).

Fig. 8.
Fig. 8.

Standardized 0–47-cm soil moisture anomalies of basin-averaged soil moisture from JRA-55 relative to the 1976–2011 climatology.

Citation: Journal of Hydrometeorology 16, 3; 10.1175/JHM-D-14-0033.1

Following the heavy rain during the warm season of 2010, the standardized anomaly for the 0–47-cm-layer soil moisture was at two standard deviations above average (Fig. 8). Soil moisture in the fall of 2010 was the highest going back to 1976. For February–April 2011, soil moisture in the 0–47-cm layer was over one standard deviation above average and ranked second highest on record behind February–April (FMA) 1996 (Fig. 8). For the period between May 2010 and April 2011, the 0–47-cm soil moisture was the highest between 1976 and 2011, with standardized anomalies over two standard deviations above normal (Fig. 8). Collectively, these data underscore the importance of the heavy rain (section 4a) during the 2010 warm season in increasing the soil moisture levels over the ARB both prior to freeze-up and before the snowmelt commenced in early April 2011.

Soil temperature data for Estevan, Yorkton, and Winnipeg indicate that the soil typically starts to freeze in late November. By February, the frost line extends down to 50 cm over the southern ARB, but to at least 100 cm in the north. The soil has usually thawed by early to mid-April, following the snowmelt. The high soil moisture content during the winter of 2010–11 would have likely increased the amount of ice in the soil, thereby increasing runoff in the spring (e.g., Gray et al. 2001; Granger et al. 1984; Kane 1980). Snowmelt and soil thaw dates for 2011 are not known, but Hayashi et al. (2003) found that snowmelt at a Canadian Prairie site started when the daily mean temperature exceeded 0°C, and Dunne and Black (1971) observed that the soil had typically thawed within a day after the snow had melted. Thus, daily mean temperatures from Estevan and GlobSnow snow extent data suggest that the soil would have been frozen over most of the ARB until at least mid-April 2011. Consequently, there was less time for the upper soil layer to dry out before the heavy events started at the end of April (see section 4a), and this would have increased runoff.

5. Heavy precipitation events

The frequency, timing, and severity of heavy precipitation events over the ARB were critical because they increased soil moisture (2010 warm season), the winter snowpack (winter of 2010/11), and subsequent spring runoff in 2011. Here, CaPA data were used to identify the spatial distribution and frequency of heavy daily precipitation events.

Heavy and very heavy precipitation days were defined as days with at least 10 and 25 mm, respectively. Specifically, CaPA grid points with ≥10 mm of precipitation (GE10) in a 24-h period (between 0600 and 0600 UTC) and those with ≥25 mm in ≤24 h (GE25) were identified. Zhang et al. (2011) suggested a threshold of 10 mm day−1 to identify heavy precipitation days. The very heavy precipitation threshold of 25 mm day−1 has been used in several studies (e.g., Budikova et al. 2010; Higgins et al. 2007; Akinremi et al. 1999) over North America. We only considered very heavy events (GE25) during the warm season (from 1 May through 30 September). An analysis of daily rainfall data for warm seasons between 1981 and 2010 at three stations in the vicinity of the ARB (Estevan, Prince Albert, and Winnipeg) found that this region typically experiences 10 GE10 events in a warm season, with a maximum of 16 events (Table 2). On average, only two GE25 events are recorded, with a maximum of six events per season (Table 2), underscoring the fact that 25 mm day−1 is atypical for this region.

Table 2.

Seasonal mean and max counts for heavy (GE10) and very heavy (GE25) precipitation events over the ARB. Climatological mean and max values are based on station data from Estevan, Winnipeg, and Prince Albert from 1981 to 2010. For example, these data show that, on average, 10 GE10 events are observed in this region annually, with an annual max of 16. Values in parentheses represent the mean and max values for October–March. Data for 2010 and 2011 are calculated for CaPA grid points within the ARB, with contributions of GE10 and GE25 events to the seasonal accumulated CaPA precipitation also shown.

Table 2.

Between May and September 2010, a total of 5689 GE10 occurrences were recorded at the 505 grid points in the ARB (Fig. 9a). The basin average was 11.3 occurrences, with a gridpoint maximum of 19 (i.e., above average). GE10 events contributed 36%–68% to the warm-season rainfall totals (Fig. 10a), with the basin-averaged contribution being 203 mm (53%). In 2011, the frequency of GE10 occurrences was not as high, at 4158 (Fig. 9c) with a maximum of 17 (i.e., near normal) near Estevan. GE10 events contributed between 21% and 80% to the warm-season rainfall in 2011 (Fig. 10c); the mean contribution was 163 mm (59%).

Fig. 9.
Fig. 9.

Gridpoint counts for GE10 and GE25 events for the (a),(b) 2010 and (c),(d) 2011 warm seasons (from 1 May through 30 Sep) from the CaPA data.

Citation: Journal of Hydrometeorology 16, 3; 10.1175/JHM-D-14-0033.1

Fig. 10.
Fig. 10.

Relative contribution of GE10 and GE25 events to warm-season rainfall for (a),(b) 2010 and (c),(d) 2011 from the CaPA data.

Citation: Journal of Hydrometeorology 16, 3; 10.1175/JHM-D-14-0033.1

The GE25 fields were more spatially heterogeneous than the GE10 fields. Almost 860 GE25 occurrences were recorded over the ARB during the 2010 warm season (Fig. 9b), with a basin average of 1.7 events and a gridpoint maximum of five events. The contribution of GE25 events to 2010 warm-season rainfall was 58 mm (~15%), with a gridpoint maximum of 44% (Fig. 10b). In 2011, GE25 events were near normal, but higher than 2010, at 1182, with an average of 2.3 events and a maximum of 4 events (Fig. 9d). The contribution of GE25 events to 2011 warm-season rainfall was 63 mm (~23%), with a maximum contribution of 52% (Fig. 10d).

For the cold season (from 1 October to 30 April), on average, the ARB can be expected to receive from three to four GE10 events (Table 2), with the maximum near nine events. For GE25 events, the mean is less than one event, with a maximum of two. From October 2010 to April 2011, a total of 2876 GE10 events were identified, and the basin average was 5.7 events (Fig. 11), so above average. The highest incidence of GE10 events was over the far southeastern ARB, where 10 events were identified. GE10 events contributed a mean of 44% to the total cold-season precipitation over the basin. The maximum number of four GE25 events over the far northeastern parts of the basin was unusually high for this region during the cold season.

Fig. 11.
Fig. 11.

Gridpoint counts for GE10 events during the 2010/11 cold season (from 1 Oct through 30 Apr) from the CaPA data.

Citation: Journal of Hydrometeorology 16, 3; 10.1175/JHM-D-14-0033.1

Of 11 heavy cold-season precipitation events (>10 mm water equivalent) between October 2010 and April 2011, eight were associated with lows that developed over Wyoming, Montana, or Alberta and then moved either over or just south of the ARB. These systems typically had strong upper-level support (e.g., cutoff lows), with a strong poleward flow ahead of the systems advecting relatively warm and moist air over the basin. Reanalysis data indicated that the vertically integrated water content for these events was typically well above average and was observed in concert with strong ascent over the basin. Of the three remaining events, two were associated with slow-moving surface lows, while another was associated with a weak surface low and associated Arctic front.

In section 4c, it was noted that the exceptionally heavy warm-season rainfall in 2010 led to well-above-normal soil moisture levels over the ARB prior to freeze-up and that the soil moisture anomalies persisted into the early spring of 2011. Additionally, there was a higher-than-average occurrence of GE10 and GE25 events over the eastern and southeastern portions of the basin that contributed to the well-above-average SWE content of the snowpack.

6. The exceptionally wet spring of 2011

By April 2011, the stage was set for potential flooding because of the snowpack’s high SWE and the high soil moisture. However, exceptionally wet conditions in the spring exacerbated an already serious situation. Given that heavy rainfall events between late April and June worsened and extended the flooding beyond the typical snowmelt peak in late April, it is worthwhile to examine the nature of these events and their attendant large-scale atmospheric circulation in greater detail.

In 2011, the ARB experienced its fifth-wettest May–June in the satellite era and the ninth wettest since the beginning of the twentieth century. The southern half of the ARB (south of 50.5°N) was especially wet, experiencing its third-wettest May–June since the beginning of the twentieth century. Specifically, Estevan experienced its wettest May–June since 1935 (the date from which continuous records are available) with 335 mm (Fig. 12). This was 240% above the 1981–2010 average (~140 mm) and represented over 70% of the mean annual total precipitation (~435 mm) in only 61 days. An important aspect of precipitation in May–June was that heavy and very heavy rainfall events contributed markedly to the extremely high spring and warm-season precipitation totals.

Fig. 12.
Fig. 12.

Accumulated daily precipitation at Estevan for the water year starting on 1 Sep. Also shown are the accumulated precipitation amounts for the 1954–55, 1975–76, and climatological (1971–2000) mean water years.

Citation: Journal of Hydrometeorology 16, 3; 10.1175/JHM-D-14-0033.1

a. Heavy and very heavy events

Historically (1981–2010), Winnipeg and Estevan receive an average of four heavy (GE10) events in May and June combined, with a maximum of seven to nine events. On average, one very heavy (GE25) event can be expected for May–June, with a maximum from three to five. In May 2011, between six and eight GE10 events were identified at grid points in the far southwest ARB (Fig. 13a). In June 2011, the occurrence of GE10 events was slightly lower (Fig. 13c), although the focus of activity was still over the southwestern part of the basin with up to six events. For May and June 2011 combined (Fig. 13a), the maximum number of GE10 events observed at any grid point was 13 in the southwest. A total of 2829 GE10 events occurred in May–June, which accounts for almost 70% of all the GE10 events over the ARB during the warm season. The contribution of GE10 events in May–June to the warm-season total count exceeded 50% at over 85% of the grid points. Averaged over the basin, GE10 events contributed almost 65% to total May–June total precipitation and almost 40% to total warm-season precipitation (Table 2).

Fig. 13.
Fig. 13.

Gridcell counts for GE10 and GE25 events for (a),(b) May and (c),(d) June 2011from the CaPA data.

Citation: Journal of Hydrometeorology 16, 3; 10.1175/JHM-D-14-0033.1

Over the southern half of the basin, some grid points recorded up to two GE25 events in May (Fig. 13b). In June, the GE25 events shifted to the western half of the basin (Fig. 13d), with some locations in the southwest receiving up to three events. For May–June, a maximum of five GE25 events were observed at any grid point and a total of 553 events at all grid points. This represents almost 50% of the 1182 GE25 events identified over the basin for May–September. In May–June, the contribution of GE25 events to the warm-season total count exceeded 50% at over 60% of the grid points. Averaged over the basin, GE25 events contributed almost 30% to total May–June total precipitation and almost 20% to warm-season precipitation (Table 2).

b. Atmospheric circulation

1) May 2011

The monthly-mean 500-mb geopotential height map for May 2011 (Fig. 14b) shows a split flow, with a negatively tilted trough along the Canadian and U.S. west coasts and an omega block extending northwestward from the Canadian Prairies. Below the 49th parallel, the flow was relatively zonal. Consistent with this regime, 500-mb heights were above average over most of western Canada (500-mb height anomalies over the Canadian Arctic were >90 gpm, the highest for May for this region since 1979), with below-average heights along the U.S. West Coast (−60 gpm) extending eastward toward the Great Lakes (Fig. 14c).

Fig. 14.
Fig. 14.

Monthly maps for May 2011 for (a) CANGRD precipitation departure from normal (%), (b) mean 500-mb heights (gpm), (c) 500-mb height anomalies (relative to 1981–2010 climatology), (d) 500-mb vertical motion (Pa s−1), (e) mean sea level pressure (mb), and (f) 250-mb zonal wind anomalies (m s−1). Data in (b)–(f) are from ERA-I.

Citation: Journal of Hydrometeorology 16, 3; 10.1175/JHM-D-14-0033.1

Strong descent was observed over the northern portions of the ARB (Fig. 14d) in association with the blocking high, with enhanced ascent over the southern ARB and U.S. Northwest downstream of the west coast trough. At the surface (Fig. 14e), a persistent surface low over the northern Great Plains of the United States drew warm, moist air northward to the ARB. Thus, the large-scale flow favored increased cyclone activity over the northern Great Plains and attendant precipitation over southern Saskatchewan and southern Manitoba. Zonal wind anomalies at 250 mb indicate that the upper-level flow was >10 m s−1 weaker than the long-term mean over the Canadian Prairie provinces (Fig. 14f). The location of the jet stream suggests a preferred storm track extending from the Four Corners region northeastward to the Great Lakes.

2) June 2011

The omega block over western Canada broke down in late May. The trough over the west coast shifted southward and weakened slightly, while a ridge formed downwind over the midwestern United States and extended to a blocking high between Baffin Island and Greenland (Fig. 15b), where the 500-mb height anomalies were the highest for June since 1979. Height anomalies at 500 mb were near −30 gpm upwind of the ARB. Strong ascent was observed over the ARB (Fig. 15d). Surface pressures were below average over the northern Great Plains of the United States (Fig. 15e), and this acted to draw moist air northward over the ARB. The 250-mb zonal wind anomalies indicate that the jet stream was weaker than average over the Canadian Prairies and northern Great Plains (Fig. 15f). Over the same region, the meridional wind component at 500 mb was over 4 m s−1 stronger than normal. These data suggest that the large-scale flow favored increased cyclone activity and precipitation over and in the vicinity of the ARB, with systems typically tracking south of the ARB.

Fig. 15.
Fig. 15.

As in Fig. 14, but for June 2011.

Citation: Journal of Hydrometeorology 16, 3; 10.1175/JHM-D-14-0033.1

c. Outgoing longwave radiation

The monthly maps shown in Figs. 14 and 15 provide insight into the dominant weather patterns during a given month, but they smooth out the details concerning the evolution of the atmospheric flow. To assess the sequence of events over the ARB, a time–longitude Hovmöller diagram of remotely sensed OLR anomalies was examined (Fig. 16). Negative OLR anomalies (blue shades) typically represent high cloud tops and enhanced precipitation, whereas positive anomalies are indicative of little or no precipitation. Additionally, comparison between Hovmöller diagrams for OLR anomalies and 500-mb vertical motion anomalies (not shown) indicate that OLR is a good proxy for baroclinic activity. This is consistent with the high correlations found by Lim and Wallace (1991) between tropospheric vertical motion and OLR.

Fig. 16.
Fig. 16.

Hovmöller diagram of daily remotely sensed OLR anomalies from NOAA for a domain extending from 45° to 55°N and from 120°W to 80°E between 20 Apr and 30 Jun 2011. Solid black vertical lines indicate the approximate position of the ARB. Red rectangles indicate timing of heavy precipitation events.

Citation: Journal of Hydrometeorology 16, 3; 10.1175/JHM-D-14-0033.1

The eastward progression of precipitation areas is clearly evident by the slope of the negative OLR anomalies in Fig. 16. Further, multiday precipitation events over a given latitude band spanning days are also evident (e.g., between 19 and 22 May). Hovmöller diagrams of OLR anomalies indicate that six organized systems affected the ARB between late April and the end of June 2011 (Fig. 16). This is corroborated by surface and 500-mb upper-air analysis maps. Thus, rather than a single very heavy precipitation event causing the flooding, at least six systems yielding very heavy precipitation affected the ARB. Only three events occurred in June, and all but two were associated with a deep surface low and attendant upper-air cold low.

In late April 2011, a deep low-pressure system (992 mb) and associated upper cold low that developed near Colorado traversed the ARB (Fig. 16), producing up to 70 mm of precipitation over the eastern parts of the basin, some of it as snow. Between 10 and 11 May, another surface low-pressure system affected the ARB. This system produced 40–50 mm of rain over the extreme southern portion of the ARB. The positive OLR anomalies from 11 to 18 May were the result of subsidence associated with an upper-air ridge over the Prairies. A deep surface and upper cold low (that originated over Wyoming) moved south of the ARB on 21 May 2011 and produced 30–40 mm of rain (Fig. 16). On 31 May, a deep Colorado low (996 mb) was located over southern Manitoba and produced a wide swath of heavy rain (40–50 mm) over the eastern and southeastern ARB, with isolated locations receiving from 70 to 80 mm of rain.

Apart from intermittent positive OLR anomalies in early June 2011, Fig. 16 suggests that frequent precipitation episodes affected the ARB region. These events were associated with systems spawned by a persistent longwave trough along the west coast. Specifically, on 7 June a deep low (994 mb) that originated over Wyoming produced very heavy rain (50–60 mm) over the southern ARB. Between 20 and 21 June, a persistent trough of low pressure extending from western Saskatchewan to the Dakotas produced a wide swath of heavy rain (up to 80 mm) over the central and western ARB. Lightning data (not shown) suggest that, unlike the prior events, a significant portion of the rainfall for the 7 June and 20–21 June events were associated with thunderstorms. The relative contribution of thunderstorm rain to precipitation totals was investigated using the same technique as Brimelow et al. (2014). Thunderstorm rain is of interest because the high rainfall rates lead to high surface runoff that can exacerbate flooding. Thunderstorm rain was estimated using coincident 6-hourly cloud-to-ground lightning flash and CaPA data. For May and June 2011, 26 mm (~15%) of the basin-average precipitation of 170 mm was attributed to thunderstorms. By comparison, in May and June 2010, over 50 mm (~25%) of the basin average of 200-mm total was attributable to thunderstorms. Hence, if May–June 2011 had as much thunderstorm rain as in 2010, the flooding may have been even worse.

7. Surface water budget and runoff generation

Streamflow is integrated runoff, which is in turn related to water cycling components in a drainage basin by way of water mass conservation, as in the following equation:
e1
where S, SM, P, E, and R are basin-average SWE, soil wetness, precipitation, evapotranspiration, and runoff, respectively. In this section, we use JRA-55 data to examine the surface water budget to determine why the 2011 flood was so dramatically different from previous record floods in the ARB (e.g., 1976 and 1995) and why record floods were not observed in years when the snowpack was deeper than in 2011 or when the spring rains were heavier than in 2011.

The application of reanalysis data to estimate water budgets in the neighboring Saskatchewan River basin was discussed in Szeto (2007). It was found in that study that the estimated basin-averaged water budget could vary substantially between different reanalysis datasets. Here, we found that the JRA-55 budget compares quite favorably with available observations and other modern reanalysis data such as ERA-I. In particular, the interannual variability of the JRA-55 spring [March–May (MAM)] precipitation agrees well with those of ERA-I and CANGRD, with correlations of 0.91 and 0.83, respectively. However, JRA-55 exhibits a wet bias, with P roughly 27% higher than CANGRD for MAM, declining to ~5% for JJA.

The JRA-55 SWE compares well with GlobSnow, with contemporaneous correlations of 0.92 for their January–February SWE values. There are no observations of E or SM to validate the JRA-55 values, but its E for MAM compares well with those of ERA-I, with a correlation of 0.7, and the two products’ long-term means for E are within 0.1 mm day−1 of each other. Likewise, the JRA-55 soil wetness data correlate well with those of ERA-I, with the correlation coefficient of their monthly values >0.7. It is difficult to directly compare the model runoff values with streamflow data because of the extensive coverage (~62% of total area) of noncontributing areas (NCA) in the ARB (Agriculture and Agri-Food Canada 2013). Additionally, although the archived model runoff values exhibit high (>0.8) 1-month lag correlations with measured streamflow data at Headingley over the 1970–2011 period, the values were found to be not in balance with the rest of the budget terms. As such, R was instead estimated from the residual from balancing the budget terms in Eq. (1). The runoff values diagnosed using this approach also exhibit high (>0.8) 1-month lag correlations with measured streamflow data at Headingley. In addition, with a couple of exceptions, years with the highest March–June (MAMJ) runoff (in decreasing order of magnitude: 2011, 1976, 1974, 2010, 1997, 1995, and 1999) also correspond well with the years with highest observed MAMJ discharge volume at Headingley (in decreasing order of magnitude: 2011, 1976, 1974, 1995, 1975, and 1999). These results demonstrate that our approach for quantifying the surface ARB water budget reasonably captures the interannual variability and extremes of at least some processes, including spring runoff, in the basin. Although the effects of NCA are not modeled in the reanalysis, an analysis of the water budget should still provide valuable insight into runoff generation in the basin. Additionally, if we make the assumption that the relative contribution of P and dS to changes in R and dSM during the melting period are proportional to the their relative magnitudes, then we can also keep track of the relative contributions of P, dS, and initial dSM to runoff in subsequent periods.

The accumulated daily P, E, dS, and dSM from 1 March to 30 June for the 1980–2011 mean and for 1976, 1995, and 2011 are presented in Figs. 17a–d, respectively. The long-term average budgets (Fig. 17a) show that snowmelt typically occurs from March to early April in the ARB. Runoff and soil moisture recharge occur at roughly the same rate during the melting period. After the spring thaw, runoff decreases and the soil wetness is depleted through evapotranspiration and runoff. Both P and E increase steadily throughout the period in step with increasing temperatures, with P being slightly greater than E during the early spring, resulting in nearly constant accumulated PE of ~11 mm during the latter part of the period. Consequently, runoff after the melt is mainly sustained by the subsurface flow, and accumulated SM becomes negative by the end of the period. For the spring melt period, P and dS contribute about 52% and 48%, respectively, to the runoff. In contrast, between 1 March and 30 June, P and dS contribute about 76% and 21% to the runoff, respectively, with a marginal contribution of only 3% from dSM.

Fig. 17.
Fig. 17.

Accumulated amounts (mm) for the surface water budget terms from JRA-55 from 1 Mar through 30 Jun for (a) 1980–2011 mean, (b) 1976, (c) 1995, and (d) 2011.

Citation: Journal of Hydrometeorology 16, 3; 10.1175/JHM-D-14-0033.1

The 1976 budget (Fig. 17b) confirms the importance of rapid and significant snowmelt in causing the flood event that year. SWE was higher in 1976 than in 2011, with the result that large surface runoff and rapid soil moisture recharge occurred during the melt from late March to early April, after which it closely followed variations in P. It should be noted that short-duration decreases in the accumulated runoff are artifacts that arose in using the current approach of budget balance to estimate the total runoff. An intense rainfall event in mid-April contributed to enhancing the surface runoff shortly after the snowmelt. This was followed by an extended dry period that lasted until the end of May. Consequently, SM was depleted to below its original value by early May. Similar to the observed discharge (Fig. 4), runoff tapers off rapidly in May. Although the accumulated PE was somewhat close to the long-term mean values during March and April, the two accumulated terms closely balanced each other throughout May and June as temperatures rose above normal in late April and May. Soil moisture in the early spring was well above normal, so antecedent SM for the 1976 flood could have initially had a greater impact on the spring runoff than in 2011. The relative contribution of P and dS to the accumulated runoff between 1 March and 30 June in 1976 were ~30% and ~70%, respectively. This is the complete opposite of the climatological values and stems from the fact that, for the spring melt period, P and dS contributed 13% and 87%, respectively, to the accumulated runoff. These data underscore the importance of the snowmelt in the 1976 flood.

There was substantially lower SWE in JRA-55 in 1995 than in 1976 (Fig. 17c) and 2011 (Fig. 17d), which is also consistent with the 2011 GlobSnow data. There were two short melt periods in mid-March and again in mid-April. Precipitation and the March snowmelt contributed to rapid runoff during the second half of March, and runoff was maintained at a slower (but steady) rate until late May, when it tapered off. The main difference between the 1995 and 1976 surface budgets is found in the somewhat-higher accumulated PE throughout the period, which can be partially attributed to the below-average temperatures in April and May 1995 and to the well-above-normal P in March and April. Although SWE was higher in 1976, the higher PE in 1995 reduced the difference in the accumulated runoff at the end of the period, as did the fact that rainfall after 1 May was near unexceptional (i.e., near normal). The relative contributions of P (77%) and dS (22%) to the accumulated runoff through 30 June were consistent with the long-term climatology. The contributions of P (70%) and dS (30%) during the spring melt reflect the elevated role of P compared to dS, on account of the relatively modest snowpack.

Accumulated budgets for 2011 are presented in Fig. 17d. The budgets clearly illustrate the reasons why runoff was substantially higher than in the previous two cases. First, although the SWE was slightly lower in 2011 than in 1976, it was markedly higher than the values in 1995. More importantly, the accumulated PE was increasing throughout the period, to a value that was substantially higher than for either of the two earlier flood events. The enhanced PE can be attributed to the frequent and heavy P and the cold spring temperatures in 2011. As noted earlier, the cold spring delayed the melt until April. The delayed melt, combined with the frequent cloudy conditions that accompanied the rain events, reduced E, and the large difference in PE largely accounts for the difference in accumulated runoff by the end of the period. As in 1995, the relative contributions of P (72%) and dS (24%) to the accumulated runoff through 30 June were consistent with the long-term climatology. However, during the spring freshet, the contributions of P (30%) and dS (70%) to the accumulated runoff were essentially reversed, reflecting the snowmelt from the record SWE. It must also be kept in mind that these are relative contributions and that the actual (absolute) contributions from P and dS were well above normal. These data suggest that, had it not been for the delayed freshet and the heavy and persistent rains in May and June 2011, the 2011 flood would have likely instead been of a similar magnitude to the 1976 flood.

The importance of the sequence of events for the 2011 flood is highlighted by the fact that river flooding did not necessarily follow years with either well-above-average SWE, SM, or spring P. For example, April–June precipitation totals in 1991 and 2010 were ranked first and second highest on record, respectively, yet there was no flooding. For 1991, the February and March SWE were well below average, and the soil moisture prior to freeze-up the preceding fall was almost one standard deviation below average. As for 2010, the budgets (not shown) show that although SWE in the beginning of March was slightly higher than normal, melt also occurred early in March due to warm spring temperatures (~3°C above normal in March and April). In addition, spring P was low while E was higher than normal throughout the warm March and April. As a result, the accumulated PE remained low until late May, and runoff was close to the normal value at the end of May, despite the higher-than-normal SWE and the high rainfall that started in the latter part of May. A large amount of runoff was generated in June from the exceptional rainfall, and this runoff was reflected in the high July discharge (ranked third during the period 1970–2011) observed for the year. The year 1997 provides an example of a high-SWE case that produced no flooding. Specifically, 1997 had the second-highest March SWE in the satellite record, but there were no spring floods. The behavior of the accumulated water budget terms (not shown) is remarkably similar to that in 1976, and the accumulated runoff by the end of the period is similar in magnitude to that in 1995. The JRA-55 data are, therefore, not very useful in providing clues to explain the absence of a flood in 2010. Examining the budgets in greater detail reveals that the JRA-55 precipitation for MAM 1997 was 50% higher than that in the CANGRD data, and the observed April–June P was actually well below average (sixteenth lowest on record). The lower observed P, along with the potentially higher E, could have accounted for the spring flow volume being relatively modest despite the high SWE.

8. Summary and conclusions

In the spring and early summer of 2011, the ARB experienced an extreme flood that was unprecedented in terms of its duration and the volume of water involved. In this study, we drew on a diverse set of datasets to characterize the surface and atmospheric conditions prior to and during the flood, in addition to elucidating the key hydroclimatic processes that led to the flood.

This research has led to important insights regarding the 2011 flood. In particular, the flooding over the ARB in the spring of 2011 was the result of a sequence of events starting in the summer of 2010 (Fig. 18). Contrary to what one might expect for such an extreme flood, individual precipitation events before and during the 2011 flood were not extreme; instead, it was the cumulative impact and timing of precipitation events going back to the summer of 2010 that played a key role in the 2011 flood. This corroborates the concept that extreme natural hazards can arise from the compounding impacts from a series of nonextreme events (Field et al. 2012). The sequence of events between 2010 and 2011 may be summarized as follows.

  1. The summer of 2010 was the third wettest on record, followed by the fifth-wettest fall. Consequently, soil moisture anomalies for the 0–47-cm layer were almost two standard deviations greater than average going into the winter, and the Assiniboine River had relatively high streamflow through the winter.
  2. Precipitation for the winter of 2010/11 was above average, and there was a higher-than-average occurrence of heavy and very heavy precipitation events over the eastern parts of the ARB that contributed to high SWE values. In fact, the March and April basin-averaged SWE values were the highest in the satellite record.
  3. At the time of the snowmelt, soil moisture anomalies (0–47-cm layer) were over one standard deviation above average. Temperatures in March were well below average, and this delayed the spring melt by weeks. The impact of the delayed thaw was threefold: 1) higher incident solar radiation and temperatures resulted in a rapid thaw, 2) above-average streamflows from the snowmelt were peaking just as the first in a series of heavy rainfall events affected the ARB, and 3) there was little time available for the upper soil layer to dry out before the heavy rains started, thereby increasing runoff.
  4. Events in May and June 2011 exacerbated the already wet conditions. The southern half of the ARB was especially wet, experiencing its third-wettest May–June period since the beginning of the twentieth century, with area-averaged precipitation in only about 60 days representing ~50% of the annual mean precipitation. May–June 2011 was also a time when heavy and very heavy rainfall events were frequent when compared to the rest of the season and historical data, and the heavy precipitation would have enhanced surface runoff. During this time, the large-scale circulation favored increased cyclone activity and stronger-than-average ascent and precipitation over and in the vicinity of the ARB. At least six major systems produced very heavy precipitation between late April and the end of June.
Fig. 18.
Fig. 18.

Timeline of significant events relevant to the 2011 ARB flood. Rankings for precipitation are relative to records going back to 1900 and for soil moisture and SWE are relative to records going back to 1979. Discharge records go back to 1906, 1913, and 1974 at Wawanesa, Headingley, and Brandon, respectively.

Citation: Journal of Hydrometeorology 16, 3; 10.1175/JHM-D-14-0033.1

This research has also allowed us to determine why the 2011 flood was so markedly different from previous floods in the ARB (i.e., 1976 and 1995); ultimately, it was the particular sequence, timing and magnitude of events in 2010–11 and their cumulative impact that distinguished the 2011 flood from previous floods. We caution, however, that the possible role of land use change in the 2011 flood has not been considered here.

Winter and spring temperatures in Canada have increased markedly in recent decades (Vincent et al. 2007), with a concomitant reduction in spring snow depth and snow cover (Brown and Mote 2009). Model projections suggest that winter and spring temperatures will continue to increase, with an associated increase in the frequency and intensity of heavy precipitation events (Kharin et al. 2013; Sillmann et al. 2013). Although some precursors to the 2011 flood were consistent with these expected changes—for example, the preponderance of heavy precipitation events—it is not possible to use a single season to support or argue against these observed or projected changes. It is also important to consider the regional deviations from the larger-scale trends when examining how the changing hydroclimate might affect the likelihood of extreme spring floods in Canadian regions. For example, over the ARB, CANGRD data show no trend in spring temperature between 1980 and 2011, and contrary to expectations (e.g., Brown and Mote 2009) and SWE studies over North America (e.g., Gan et al. 2013), GlobSnow data indicate that February and March SWE over the ARB has been increasing on the order of a few millimeters per decade (p = 0.10) from 1980 to 2011. This is corroborated by a similar increase in December–February (DJF) precipitation over the ARB for the same period. Should this trend continue, it could have important implications for spring flooding because of the potential for increased runoff. The discrepancy in the satellite data warrants research to determine whether or not the increase in the March snowpack over the ARB is real and, if so, why other products have not identified the increase in SWE. Thus, it would be prudent for stakeholders to keep the possibility of such atypical SWE conditions in mind when considering future flood potential even though the future hydroclimate is expected to become less favorable for spring floods.

Another issue arising from this work is, what caused the atmospheric circulation regimes and particular sequence of events that led to the persistent heavy precipitation between the fall of 2010 and June 2011? Were these circulation regimes (and the sequence of events) related to stochastic internal variability, ENSO, or perhaps some mechanism related to warming at the high latitudes?

Boening et al. (2012) found that the strong La Niña in 2010/11 was associated with a significant increase in terrestrial moisture storage along a swath extending from the northwestern United States to Hudson Bay. This finding warrants a closer examination of the linkages between the 2010/11 La Niña and heavy precipitation over the Canadian Prairies, and it also raises the question as to whether or not the heavy rain events in May and June 2011 could have been predicted weeks (or longer) in advance. Our research sets the stage for using the 2011 flood as a test bed for answering such questions and testing the viability of improving forecasting of such events in the future.

Acknowledgments

This research was supported by the Canadian Foundation for Climate and Atmospheric Sciences and the Changing Cold Regions Network funded by the Natural Sciences and Engineering Research Council of Canada. The authors appreciate the assistance of Lucie Vincent, Éva Mekis, and Ewa Milewska for providing the CANGRD data and Vincent Fortin and Bruce Davison for providing the CaPA data. Daily outgoing longwave radiation data and monthly ERA-I plots were provided by the NOAA/OAR/ESRL PSD, Boulder, Colorado, United States, from their website at www.esrl.noaa.gov/psd/. ERA-I data were provided by the European Centre for Medium-Range Weather Forecasts. JRA-55 data were provided by the Japan Meteorological Agency.

REFERENCES

  • Agriculture and Agri-Food Canada, cited 2013: Non-contributing portions of the incremental gross drainage areas of the AAFC watersheds project. [Available online at http://data.gc.ca/data/en/dataset/a5cbed50-4a18-4658-b312-af13192bbaaf.]

  • Akinremi, O. O., , McGinn S. M. , , and Cutforth H. W. , 1999: Precipitation trends on the Canadian Prairies. J. Climate, 12, 29963003, doi:10.1175/1520-0442(1999)012<2996:PTOTCP>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Bao, Z., , Ye H. , , and Taylor P. A. , 2005: Summer regional rainfall over southern Ontario and its associations with outgoing long wave radiation and moisture convergence. Meteor. Appl., 12, 161167, doi:10.1017/S135048270500160X.

    • Search Google Scholar
    • Export Citation
  • Beltaos, S., , Burrell B. , , and Ismail S. , 1996: 1991 ice jamming along the Saint John River: A case study. Can. J. Civ. Eng., 23, 381394, doi:10.1139/l96-043.

    • Search Google Scholar
    • Export Citation
  • Blöschl, G., , Nester T. , , Komma J. , , Parajka J. , , and Perdigão R. A. P. , 2013: The June 2013 flood in the upper Danube basin, and comparisons with the 2002, 1954 and 1899 floods. Hydrol. Earth Syst. Sci. Discuss., 10, 95339573, doi:10.5194/hessd-10-9533-2013.

    • Search Google Scholar
    • Export Citation
  • Boening, C., , Willis J. K. , , Landerer F. W. , , Nerem R. S. , , and Fasullo J. , 2012: The 2011 La Niña: So strong, the oceans fell. Geophys. Res. Lett., 39, L19602, doi:10.1029/2012GL053055.

    • Search Google Scholar
    • Export Citation
  • Brimelow, J. C., , Stewart R. S. , , Hanesiak J. M. , , Kochtubajda B. , , Szeto K. , , and Bonsal B. , 2014: Characterization and assessment of the devastating natural hazards across the Canadian Prairie provinces between 2009 and 2011. Nat. Hazards, 73, 761785, doi:10.1007/s11069-014-1107-6.

    • Search Google Scholar
    • Export Citation
  • Brown, R. D., , and Mote P. W. , 2009: The response of Northern Hemisphere snow cover to a changing climate. J. Climate, 22, 21242145, doi:10.1175/2008JCLI2665.1.

    • Search Google Scholar
    • Export Citation
  • Budikova, D., , Coleman J. S. M. , , Strope S. A. , , and Austin A. , 2010: Hydroclimatology of the 2008 Midwest floods. Water Resour. Res., 46, W12524, doi:10.1029/2010WR009206.

    • Search Google Scholar
    • Export Citation
  • Burrows, W. R., , and Kochtubajda B. , 2010: A decade of cloud-to-ground lightning in Canada: 1999–2008. Part 1: Flash density and occurrence. Atmos.–Ocean, 48, 177194, doi:10.3137/AO1118.2010.

    • Search Google Scholar
    • Export Citation
  • Chen, G., , Iwasaki T. , , Qin H. , , and Sha W. , 2014: Evaluation of the warm-season diurnal variability over East Asia in recent reanalyses JRA-55, ERA-Interim, NCEP CFSR, and NASA MERRA. J. Climate, 27, 55175537, doi:10.1175/JCLI-D-14-00005.1.

    • Search Google Scholar
    • Export Citation
  • Christidis, N., , Stott P. A. , , Karoly D. J. , , and Ciavarella A. , 2013: An attribution of the heavy rainfall over eastern Australia in March 2012 [in “Explaining Extreme Events of 2012 from a Climate Perspective”]. Bull. Amer. Meteor. Soc., 94 (9), S58S61.

    • Search Google Scholar
    • Export Citation
  • Dankers, R., , and Feyen L. , 2009: Flood hazard in Europe in an ensemble of regional climate scenarios. J. Geophys. Res., 114, D16108, doi:10.1029/2008JD011523.

    • Search Google Scholar
    • Export Citation
  • Deacu, D., , Fortin V. , , Klyszejko E. , , Spence C. , , and Blanken P. D. , 2012: Predicting the net basin supply to the Great Lakes with a hydrometeorological model. J. Hydrometeor., 13, 17391759, doi:10.1175/JHM-D-11-0151.1.

    • Search Google Scholar
    • Export Citation
  • Dee, D. P., and et al. , 2011: The ERA-Interim reanalysis: Configuration and performance of the data assimilation system. Quart. J. Roy. Meteor. Soc., 137, 553597, doi:10.1002/qj.828.

    • Search Google Scholar
    • Export Citation
  • Diffenbaugh, N. S., , Scherer M. , , and Ashfaq M. , 2013: Response of snow-dependent hydrologic extremes to continued global warming. Nat. Climate Change, 3, 379–384, doi:10.1038/nclimate1732.

    • Search Google Scholar
    • Export Citation
  • Dunne, T., , and Black R. D. , 1971: Runoff process during snowmelt. Water Resour. Res., 7, 11601172, doi:10.1029/WR007i005p01160.

  • Ebita, A., and et al. , 2011: The Japanese 55-year Reanalysis “JRA-55”: An interim report. SOLA, 7, 149152, doi:10.2151/sola.2011-038.

    • Search Google Scholar
    • Export Citation
  • Field, C. B., and et al. , Eds., 2012: Managing the Risks of Extreme Events and Disasters to Advance Climate Change Adaptation. Cambridge University Press, 582 pp.

  • Gan, T. Y., , Barry R. G. , , Gizaw M. , , Gobena A. , , and Balaji R. , 2013: Changes in North American snowpacks for 1979–2007 detected from the snow water equivalent data of SMMR and SSM/I passive microwave and related climatic factors. J. Geophys. Res. Atmos., 118, 76827697, doi:10.1002/jgrd.50507.

    • Search Google Scholar
    • Export Citation
  • Goulding, H. L., , Prowse T. D. , , and Bonsal B. , 2009: Hydroclimatic controls on the occurrence of break-up and ice-jam flooding in the Mackenzie Delta, NWT, Canada. J. Hydrol., 379, 251267, doi:10.1016/j.jhydrol.2009.10.006.

    • Search Google Scholar
    • Export Citation
  • Government of Manitoba, 2013a: Manitoba 2011 flood review task force report. Rep. to the Minister of Infrastructure and Transportation, 192 pp. [Available online at www.gov.mb.ca/asset_library/en/2011flood/flood_review_task_force_report.pdf.]

  • Government of Manitoba, 2013b: 2011 flood: Technical review of Lake Manitoba, Lake St. Martin and Assiniboine River water levels. Tech. Rep., 123 pp. [Available online at www.gov.mb.ca/mit/floodinfo/floodproofing/reports/pdf/assiniboine_lakemb_lsm_report_nov2013.pdf.]

  • Granger, R. J., , Gray D. M. , , and Dyck G. E. , 1984: Snowmelt infiltration to frozen prairie soils. Can. J. Earth Sci., 21, 669677, doi:10.1139/e84-073.

    • Search Google Scholar
    • Export Citation
  • Gray, D. M., , Landine P. G. , , and Granger R. J. , 1985: Simulating infiltration into frozen Prairie soils in streamflow models. Can. J. Earth Sci., 22, 464472.

    • Search Google Scholar
    • Export Citation
  • Gray, D. M., , Toth B. , , Zhao L. , , Pomeroy J. W. , , and Granger R. J. , 2001: Estimating areal snowmelt infiltration into frozen soils. Hydrol. Processes, 15, 30953111, doi:10.1002/hyp.320.

    • Search Google Scholar
    • Export Citation
  • Hall, J., and et al. , 2013: Understanding flood regime changes in Europe: A state of the art assessment. Hydrol. Earth Syst. Sci. Discuss., 10, 15 52515 624, doi:10.5194/hessd-10-15525-2013.

    • Search Google Scholar
    • Export Citation
  • Hancock, S., , Huntley B. , , Ellis R. , , and Baxter R. , 2014: Biases in re-analysis snowfall found by comparing the JULES land surface model to Globsnow. J. Climate, 27, 624632, doi:10.1175/JCLI-D-13-00382.1.

    • Search Google Scholar
    • Export Citation
  • Hayashi, M., , van der Kamp G. , , and Schmidt R. , 2003: Focused infiltration of snowmelt water in partially frozen soil under small depressions. J. Hydrol., 270, 214229, doi:10.1016/S0022-1694(02)00287-1.

    • Search Google Scholar
    • Export Citation
  • Higgins, R. W., , Silva V. B. S. , , Shi W. , , and Larson J. , 2007: Relationships between climate variability and fluctuations in daily precipitation over the United States. J. Climate, 20, 35613579, doi:10.1175/JCLI4196.1.

    • Search Google Scholar
    • Export Citation
  • Kane, D. L., 1980: Snowmelt infiltration into seasonally frozen soils. Cold Reg. Sci. Technol., 3, 153161, doi:10.1016/0165-232X(80)90020-8.

    • Search Google Scholar
    • Export Citation
  • Kharin, V. V., , Zwiers F. W. , , Zhang X. , , and Wehner M. , 2013: Changes in temperature and precipitation extremes in the CMIP5 ensemble. Climatic Change, 119, 345357, doi:10.1007/s10584-013-0705-8.

    • Search Google Scholar
    • Export Citation
  • King, A. D., , Lewis S. C. , , Perkins S. E. , , Alexander L. V. , , Donat M. G. , , Karoly D. J. , , and Black M. T. , 2013: Limited evidence of anthropogenic influence on the 2011–2012 extreme rainfall over southeast Australia [in “Explaining Extreme Events of 2012 from a Climate Perspective”]. Bull. Amer. Meteor. Soc., 94 (9), S55S58.

    • Search Google Scholar
    • Export Citation
  • Klehmet, K., , Geyer B. , , and Rockel B. , 2013: A regional climate model hindcast for Siberia: analysis of snow water equivalent. Cryosphere, 7, 10171034, doi:10.5194/tc-7-1017-2013.

    • Search Google Scholar
    • Export Citation
  • Komori, D., and et al. , 2012: Characteristics of the 2011 Chao Phraya River flood in central Thailand. Hydrol. Res. Lett., 6, 4146, doi:10.3178/hrl.6.41.

    • Search Google Scholar
    • Export Citation
  • Lavers, D. A., , and Villarini G. , 2013: Were global numerical weather prediction systems capable of forecasting the extreme Colorado rainfall of 9–16 September 2013? Geophys. Res. Lett., 40, 6405–6410, doi:10.1002/2013GL058282.

    • Search Google Scholar
    • Export Citation
  • Lawford, R. G., , Prowse T. D. , , Hogg W. D. , , Warkentin A. A. , , and Pilon P. J. , 1995: Hydro-meteorological aspects of flood hazards in Canada. Atmos.–Ocean, 33, 303328, doi:10.1080/07055900.1995.9649535.

    • Search Google Scholar
    • Export Citation
  • LeComte, D., 2011: U.S. weather highlights of 2010: A year of extremes. Weatherwise, 64, 1320, doi:10.1080/00431672.2011.566815.

  • Lee, S., , Klein A. G. , , and Over T. M. , 2005: A comparison of MODIS and NOHRSC snow-cover products for simulating streamflow using the Snowmelt Runoff Model. Hydrol. Processes, 19, 29512972, doi:10.1002/hyp.5810.

    • Search Google Scholar
    • Export Citation
  • Liebmann, B., , and Smith C. A. , 1996: Description of a complete (interpolated) outgoing longwave radiation dataset. Bull. Amer. Meteor. Soc., 77, 12751277.

    • Search Google Scholar
    • Export Citation
  • Lim, G. H., , and Wallace J. M. , 1991: Structure and evolution of baroclinic waves as inferred from regression analysis. J. Atmos. Sci., 48, 17181732, doi:10.1175/1520-0469(1991)048<1718:SAEOBW>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Loukas, A., , Vasiliades L. , , and Dalezois N. R. , 2000: Flood producing mechanisms identification in southern British Columbia, Canada. J. Hydrol., 227, 218235, doi:10.1016/S0022-1694(99)00182-1.

    • Search Google Scholar
    • Export Citation
  • Mahfouf, J., , Brasnett B. , , and Gagnon S. , 2007: A Canadian Precipitation Analysis (CaPA) Project: Description and preliminary results. Atmos.–Ocean, 45, 117, doi:10.3137/ao.450101.

    • Search Google Scholar
    • Export Citation
  • Marengo, J. S., , Tomasella J. , , Soares W. R. , , Alves L. M. , , and Nobre C. A. , 2012: Extreme climatic events in the Amazon basin. Theor. Appl. Climatol., 107, 7385, doi:10.1007/s00704-011-0465-1.

    • Search Google Scholar
    • Export Citation
  • Mekis, É., , and Hogg W. D. , 1999: Rehabilitation and analysis of Canadian daily precipitation time series. Atmos.–Ocean, 37, 5385, doi:10.1080/07055900.1999.9649621.

    • Search Google Scholar
    • Export Citation
  • Milrad, S. M., , Atallah H. E. , , and Gyakum J. R. , 2013: Precipitation modulation by the Saint Lawrence River valley in association with transitioning tropical cyclones. Wea. Forecasting, 28, 331352, doi:10.1175/WAF-D-12-00071.1.

    • Search Google Scholar
    • Export Citation
  • Peters, D. L., , Prowse T. D. , , Pietroniro A. , , and Leconte R. , 2006: Flood hydrology of the Peace-Athabasca Delta, northern Canada. Hydrol. Processes, 20, 40734096, doi:10.1002/hyp.6420.

    • Search Google Scholar
    • Export Citation
  • Phillips, D., 2012: Top 10 Canadian weather stories for 2011. CMOS Bulletin, Vol. 40, No. 1, Canadian Meteorological and Oceanographic Society, Ottawa, ON, Canada, 1222. [Available online at www.cmos.ca/uploaded/web/members/Bulletin/Vol_40/b4001.pdf.]

  • Rannie, W. F., 2002: The role of the Assiniboine River in the 1826 and 1852 Red River floods. Prairie Perspect.,5, 56–75. [Available online at http://pcag.uwinnipeg.ca/Prairie-Perspectives/PP-Vol05/Rannie.pdf.]

  • Roy, L., , Leconte R. , , Brissette F. P. , , and Marche C. , 2001: The impact of climate change on seasonal floods of a southern Quebec River basin. Hydrol. Processes, 15, 31673179, doi:10.1002/hyp.323.

    • Search Google Scholar
    • Export Citation
  • Saint-Laurent, D., , Mesfioui M. , , and Evin G. , 2009: Hydroclimatic variability and relation with flood events (southern Quebec, Canada). Water Resour., 36, 4356, doi:10.1134/S0097807809010047.

    • Search Google Scholar
    • Export Citation
  • Satyamurty, P., , da Costa C. P. W. , , Manzi A. O. , , and Candido L. A. , 2013: A quick look at the 2012 record flood in the Amazon basin. Geophys. Res. Lett., 40, 13961401, doi:10.1002/grl.50245.

    • Search Google Scholar
    • Export Citation
  • Shanley, J. B., , and Chalmers A. , 1999: The effect of frozen soil on snowmelt runoff at Sleepers River, Vermont. Hydrol. Processes, 13, 18431857, doi:10.1002/(SICI)1099-1085(199909)13:12/13<1843::AID-HYP879>3.0.CO;2-G.

    • Search Google Scholar
    • Export Citation
  • Shaw, D. A., , van der Kamp G. , , Conly F. M. , , Pietroniro A. , , and Martz L. , 2012: The fill–spill hydrology of Prairie wetland complexes during drought and deluge. Hydrol. Processes, 26, 3147–3156, doi:10.1002/hyp.8390.

    • Search Google Scholar
    • Export Citation
  • Sillmann, J., , Kharin V. V. , , Zwiers F. W. , , Zhang X. , , and Bronaugh D. , 2013: Climate extremes indices in the CMIP5 multimodel ensemble: Part 2. Future climate projections. J. Geophys. Res. Atmos., 118, 24732493, doi:10.1002/jgrd.50188.

    • Search Google Scholar
    • Export Citation
  • Simmons, A. J., , Poli P. , , Dee D. P. , , Berrisford P. , , Hersbach H. , , Kobayashib S. , , and Peubey C. , 2014: Estimating low-frequency variability and trends in atmospheric temperature using ERA-Interim. Quart. J. Roy. Meteor. Soc.,140, 329–353, doi:10.1002/qj.2317.

  • Spry, C. M., , Kohfeld K. E. , , Allen D. M. , , Dunkley D. , , and Lertzman K. , 2014: Characterizing pineapple express storms in the lower mainland of British Columbia, Canada. Can. Water Resour. J., 39, 302323, doi:10.1080/07011784.2014.942574.

    • Search Google Scholar
    • Export Citation
  • Szeto, K. K., 2007: Assessing water and energy budgets for the Saskatchewan River basin. J. Meteor. Soc. Japan, 85A, 167186, doi:10.2151/jmsj.85A.167.

    • Search Google Scholar
    • Export Citation
  • Takala, M., , Luojus K. , , Pulliainen J. , , Derksen C. , , Lemmetyinen J. , , and Kärnä J.-P. , 2011: Estimating Northern Hemisphere snow water equivalent for climate research through assimilation of space borne radiometer and ground-based measurements. Remote Sens. Environ., 115, 35173529, doi:10.1016/j.rse.2011.08.014.

    • Search Google Scholar
    • Export Citation
  • Trenberth, K. E., , and Fasullo J. T. , 2013: North American water and energy cycles. Geophys. Res. Lett., 40, 365369, doi:10.1002/grl.50107.

    • Search Google Scholar
    • Export Citation
  • van der Kamp, G., , and Hayashi M. , 2009: Groundwater–wetland ecosystem interaction in the semiarid glaciated plains of North America. Hydrogeol. J., 17, 203214, doi:10.1007/s10040-008-0367-1.

    • Search Google Scholar
    • Export Citation
  • van Dijk, A. I. J. M., , Renzull L. J. , , Wada Y. , , and Tregoning P. , 2013: A global water cycle reanalysis (2003–2012) reconciling satellite gravimetry and altimetry observations with a hydrological model ensemble. Hydrol. Earth Syst. Sci. Discuss., 10, 15 47515 523, doi:10.5194/hessd-10-15475-2013.

    • Search Google Scholar
    • Export Citation
  • Vincent, L. A., 1998: A technique for the identification of inhomogeneities in Canadian temperature series. J. Climate, 11, 10941104, doi:10.1175/1520-0442(1998)011<1094:ATFTIO>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Vincent, L. A., , and Gullett D. W. , 1999: Canadian historical and homogeneous temperature datasets for climate change analyses. Int. J. Climatol., 19, 13751388, doi:10.1002/(SICI)1097-0088(199910)19:12<1375::AID-JOC427>3.0.CO;2-0.

    • Search Google Scholar
    • Export Citation
  • Vincent, L. A., , and Mekis É. , 2006: Changes in daily and extreme temperature and precipitation indices for Canada over the twentieth century. Atmos.–Ocean, 44, 177193, doi:10.3137/ao.440205.

    • Search Google Scholar
    • Export Citation
  • Vincent, L. A., , van Wijngaarden W. A. , , and Hopkinson A. , 2007: Surface temperature and humidity trends in Canada for 1953–2005. J. Climate, 20, 51005113, doi:10.1175/JCLI4293.1.

    • Search Google Scholar
    • Export Citation
  • Zhang, X., , Vincent L. A. , , Hogg W. D. , , and Niitsoo A. , 2000: Temperature and precipitation trends in Canada during the 20th century. Atmos.–Ocean, 38, 395429, doi:10.1080/07055900.2000.9649654.

    • Search Google Scholar
    • Export Citation
  • Zhang, X., , Alexander L. , , Hegerl G. C. , , Jones P. , , Tank A. , , Peterson T. C. , , Trewin B. , , and Zwiers F. W. , 2011: Indices for monitoring changes in extremes based on daily temperature and precipitation data. Wiley Interdiscip. Rev.: Climate Change, 2, 851870, doi:10.1002/wcc.147.

    • Search Google Scholar
    • Export Citation
  • Zhou, T., , Song F. , , Lin R. , , Chen X. , , and Chen X. , 2013: The 2012 North China floods: explaining an extreme rainfall event in the context of a longer-term drying tendency [in “Explaining Extreme Events of 2012 from a Climate Perspective”]. Bull. Amer. Meteor. Soc., 94 (9), S49S51.

    • Search Google Scholar
    • Export Citation
Save