The flood waves are still moving through Manitoba, but already it is clear that in many areas of eastern Saskatchewan and western Manitoba, 2014 has seen the largest flood since settlement. It is also the most exceptional flood -- caused by rainfall, rather than snowmelt and it occurred in June and July, rather than April and May.
The extraordinary nature of the deluge taxed provincial resources and pushed water managers to innovate with water control structures that were originally designed to deal with spring snowmelts that typically give longer lead times for response. The 2011 flood of record in the Assiniboine River was caused primarily by snowmelt as had all other recorded floods in the river.
This shift indicates a dramatic regime change in Manitoba hydrology from snowmelt to rainfall that may be partly due to climate change. This flood was not just a manifestation of weather extremes due to climate change. It was likely magnified by the impact of artificial drainage of prairie wetlands across fields in Saskatchewan and Manitoba. This drainage increases both the volume of flow and the peak flow rate -- with devastating consequences downstream.
Our study of Smith Creek, a watershed near Langenburg, Sask., showed 24 per cent wetland coverage in 1958 was reduced to 11 per cent by 2008. We reproduced the 1958 wetland coverage in a hydrological model, and then ran the same model to predict the 2011 flood stream-flows. We found the stream-flow volumes and the flood peak of 2011 rose by roughly a third, due to that drainage of the wetlands.
This drainage has occurred extensively across the Prairies and in the Assiniboine River basin, which drains into Manitoba. It almost certainly played a part in raising flood volumes and peak flows in Manitoba in this summer's flood.
This flood started in the fall of 2013, when high snowfall began to cover the eastern prairies. It was then magnified by an exceptionally hard winter with a high snowpack and a late spring melt that saturated soils, filled sloughs and replenished reservoirs. Then came the rains: 200 millimetres in some places to late June, almost the yearly total, capped at the end of the month by heavy, extensive downpours of another 200 mm in some places. With soil saturated, and sloughs filled, the water cascaded into channels that flowed rapidly into the Assiniboine River. In too many fields, sloughs were already artificially drained, spilling more water downstream into ditches.
These heavy rains are the culmination of long term trends that started in the early 20th century -- many more multiple-day rainfall events that not only occur over a long time, but over a large area, and large enough to cause flooding. Such extensive rainfall is associated with trends for 4 C of winter warming, earlier snowmelt, more rainfall in March and less spring snowfall, seen in the last 70 years in the eastern Saskatchewan headwaters of the Assiniboine River.
While slough or wetland drainage by itself does not cause flooding, it is combining with torrential rainfalls in a changing climate to deliver much more water downstream.
In few places in North America is this more evident than in Manitoba.
Manitoba is downstream of Alberta, Saskatchewan, Ontario and a large portion of the north-central U.S., so the impacts of climate change and wetland drainage are concentrated here and manifested as repeated flooding. These events would have been considered exceptional in the 20th century, but we are now seeing them every year in Canada, as is the rest of the world.
We can take steps now to protect against floods:
-- Better prediction: improved seasonal weather predictions three to six months in advance, better severe weather prediction days in advance and precise forecasting for rivers, streams, groundwater and lakes.
-- Avoidance: land-use zoning to regulated development, based on careful and continuously updated flood-plain mapping -- regardless of whether the floods are caused by rainfall, snowmelt, overland flood, wind storms or bank erosion.
-- Mitigation: dikes, flood-control structures such as dams and retention ponds, channel modifications, and agricultural water management. These structures, modifications or land-use practices must be designed for extremes of water flow.
For the Prairies, wetland restoration is the most effective mitigation option.
Advancing the science and technology of prediction is not enough. Canada is exceptional among developed nations in its fragmented responsibility and capability for water prediction between multiple levels of government.
Our federal government collects most weather and water data, and predicts weather and drought -- but it does not forecast floods, stream-flow, lake levels and water supply nationally. The provinces have authority for water-resource management and to predict floods, map flood zones and deliver frontline disaster management, reimbursement and repair.
So, the 2014 flood that started in Saskatchewan and moved into Manitoba along the Assiniboine River system was forecast as weather by the federal government, and then as stream flow by two different provinces, with very different methods, focal points and capabilities. A co-ordinated effort would be more efficient and effective.
The U.S. recently built the National Water Centre in Alabama to provide enhanced nation-wide flood and drought analyses and predictions. Environment Canada closed its federal, western-based National Hydrology Research Institute in the late 1990s.
Canada must move to reduce its exposure to flood and drought by developing a national strategy to co-ordinate river basin-wide delivery of prediction and information for flood forecasting, water management and implementation of flood prevention and mitigation strategies.
Manitoba has just lived through an excellent case for why this can't be solved by one province alone. It should promote the development of such a system initially for the Assiniboine River and ultimately for the whole Lake Winnipeg Basin.
John Pomeroy is the Canada Research Chair in Water Resources and Climate Change and Director of the Centre for Hydrology at the University of Saskatchewan. He led a multi-year study that examined the impact of wetland drainage and climate change on the hydrology of Smith Creek in the Upper Assiniboine River Basin. Parts of the study were published last month.