Natural and unnatural complexities: flood control along Manitoba's Assiniboine River

https://doi.org/10.1016/j.jhg.2009.04.017Get rights and content

Abstract

Flooding and flood control long have been topics of concern for geographers and historians of North America, from Gilbert White's foundational work to attempts to understand the recent disaster in New Orleans. This paper considers the problem of flood control along the Assiniboine River in the province of Manitoba, Canada. The first section addresses the local landscape and the larger context, explaining changes in the relevant human and physical geography. The second section examines efforts to mitigate flooding, focusing on how the risk of inundation was shifted in location (not eliminated altogether) and compounded through time (not progressively resolved). The third focuses on a further source of complexity: the contested character of government jurisdiction. This final section also addresses major mid twentieth century changes in mitigation efforts: in the light of flooding on an unprecedented scale, governments became more open to compromise and more willing to undertake substantial projects. Gilbert White asserted that while floods are natural occurrences, flood damages are the responsibility of humanity. The question of responsibility might seem relatively clear in relation to flood mitigation, given that dykes are human constructions. The history of flood mitigation along the Assiniboine would suggest, however, that assigning responsibility for the failures and successes of flood mitigation is not so straightforward.

Introduction

On 14 May 1923, the Winnipeg Free Press printed a long commentary about flooding along the Assiniboine River on the Canadian Prairies. The author, identified only as J. M. L. (engineer), explained that the water situation was particularly troublesome in this region, arguing that it is ‘doubtful if anywhere on the continent a more complicated series of factors in flood conditions are met with than those whose effects are observed in Manitoba’. In the southern portion of the province, vast river basins encompassed areas that varied greatly in topography and climate, and human alterations for purposes of agriculture, settlement and transportation had only compounded matters. The basic question of which government should be responsible for managing the river was intensely debated. As the writer put it: ‘The natural complexities have been made more intricate by unnatural ones…’. A copy of J. M. L.'s published piece is contained in the files of the Canadian government's Department of Public Works, surrounded by documents from the mid 1940s.1 Adjoining materials make clear that it is included not as a window into how an earlier generation perceived the flood situation, but as a useful analysis that spoke as much to the contemporary situation as to the historical context. Whether natural or unnatural, the complexities certainly had not disappeared.

In the mid twentieth century, geographer Gilbert White set the tone for much academic scholarship on natural hazards with the observation that while flooding is beyond human control (an Act of God, in his phrasing), flood damages are largely the work of humanity.2 By arguing that all flood disasters had both natural and cultural components, White broadened J. M. L.'s earlier assessment of the Assiniboine River situation. Following from White's work, early hazards geographers focused on analyzing the social, political and economic components of various disasters. Typically, these scholars concentrated on a single hazardous episode or on comparisons between hazardous episodes in the same or different locations, rather than on extensive study of the long-term development of particular hazard-prone areas.3 More recently, historical geographers and environmental historians have filled in the histories of many flood-vulnerable locations. Scholars such as Ari Kelman, Craig Colten, and Georgina H. Endfield have provided rich historical studies of cities and regions that have endured high water.4 Other scholars have concentrated more particularly on the question of flood control.5 This article follows from much of this research, explaining the human as well as the environmental factors that contributed to flooding while focusing on the history of flood prevention along the Assiniboine River in the Province of Manitoba, Canada.

Recently, historian Stéphane Castonguay argued that flooding along the St. Francis River in the Province of Québec, a regular occurrence to which residents had long adjusted, was redefined as a catastrophe because of the needs of early to mid twentieth century industry.6 The situation was different along the stretch of the Assiniboine River examined in this article, as in this rural area high water posed a serious threat to the extensive agricultural settlement desired by the Federal, Provincial, and Municipal governments as well as by individual settlers. Despite some recognition that floods helped enrich soils leading to better crops, the prevailing view from the late nineteenth century onward was that Assiniboine River floods were a problem to be solved. But the idea of flood control on the Assiniboine – what this meant and who should be responsible for it – was no more straightforward than the idea of flooding along the St. Francis. Rather than the emergence of a new idea, as Castonguay discusses in relation to flooding as a catastrophe, this article addresses a long-running debate over what is appropriate flood control and who is responsible for it.

The first section of the paper addresses the local landscape and the larger context, explaining the relevant human and physical geography. The second examines efforts to mitigate flooding, focusing on how the risk of inundation was shifted in location (not eliminated altogether) and compounded through time (not progressively resolved). Governments and settlers alike scrambled to protect their interests, often without regard for the effects on other interested parties. Predictably, upstream flood protection worsened the downstream situation. And all regions along the river were affected by change over time as farmers altered how they used their land and so altered their preferred method of managing high water. What emerged was a patchwork flood control effort, satisfactory to neither residents who felt their interests were not served nor governments who fielded continuing complaints.

The third section of the paper focuses on a further source of complexity: the contested character of government jurisdiction. Karen O'Neill has argued that, in the American case, local interests successfully influenced the federal government's flood control efforts.7 The Assiniboine River situation suggests the articulation between local problems and federal action was more contested in Canada. Governments (provincial and federal, with some activity at municipal levels) and departments (the federal Departments of Public Works and Agriculture, largely) agreed that flood protection was necessary but disagreed over which entity should take this on. The final section also addresses major mid twentieth century changes in Canadian flood mitigation efforts: mainly in light of flooding on an unprecedented scale, governments became more open to compromise and more willing to undertake substantial projects. What emerged was a new impetus to deal with flooding, catalyzed in part through a particularly catastrophic episode of high water. This article explores the question of responsibility in relation to flood mitigation. If floods were, as Gilbert White asserted, beyond human control and flood damages were a human responsibility, who or what was responsible for the failures and successes of flood mitigation efforts along the Assiniboine River?

Section snippets

The Assiniboine River

The Province of Manitoba, Canada, is widely known to be afflicted by flooding. In 1950, much of the provincial south as well as a significant portion of the City of Winnipeg was under water. Images of the submerged province were circulated nationally and internationally, and significant financial aid was received from Britain and the United States. In the mid-1960s, the construction of the Winnipeg Floodway, an enormous channel meant to divert excess flows around the city, was a subject of

Flood protection along the Assiniboine

Beginning with an 1886–87 federal investigation that cost over $4000, governments invested significant resources in investigating the Assiniboine. From relatively early on, the focus was not the river but the dykes and cutoffs along its length. There was less attention to how settlement might adapt to the river and more concern with how the river might be altered to accommodate the farms in its floodplain. The intention of government and the hope of many farmers was that flood danger would be

Government jurisdiction over flood control

The events of the early 1920s drove home the severity of flooding along the Assiniboine. 1922 saw extensive inundation due primarily to climatic conditions.42 30,351 hectares (75,000 acres) of wheat land were flooded, including 7.28 hectares (18,000 acres) that had already been seeded. About $10,000 in damage was done to the highway between Portage La Prairie and Winnipeg. At Brandon, about 200

Conclusion

On 9 August 1949, Member of Parliament W. G. Weir wrote to Minister of Agriculture J. G. Gardiner on the subject of Assiniboine River flood control. Like so many other Manitobans who had written to the federal government regarding the Assiniboine River situation, Weir had a complaint about flooding. But Weir also had a broader observation about how flood protection had been executed. Continued flood problems were hardly surprising, given that only ‘a more or less patchwork job’ of flood

Acknowledgement

I am grateful to the editors and anonymous reviewers for the Journal of Historical Geography. I also would like to thank commentator Matthew Evenden and co-panelists Jennifer Bonnell and Jim Clifford as well as audience members who participated in the session of the Canadian Historical Association Annual Meetings at which I presented a version of this paper. Thanks to Sean Gouglas for his work on my maps. For financial support, I am grateful to the Social Sciences and Humanities Research

References (0)

Cited by (21)

  • Fidelity of reanalysis datasets in floodplain mapping: Investigating performance at inundation level over large regions

    2021, Journal of Hydrology
    Citation Excerpt :

    The discrepancy in the flooding extent is possibly due to the complex and diverse topography of the River Basin, with significant area contained within a large glacial meltwater channel (Blais et al., 2016). Moreover, the slope of the Assiniboine River channel has several undulations due to its confluence with the Red River (Ahmari et al., 2016) and meandering due to presence of large deposits of fine clay sediment (Bower, 2010), which may not be perfectly captured by the MERIT DEM topographic data. ERA-Interim and CFSR perform reasonably well with H-scores lying between 0.58 and 0.7, and C scores lying between 0.55 and 0.63 (Fig. 7b & f; Fig. 7c & g).

  • Environmental justice and flood prevention: The moral cost of floodwater redistribution

    2019, Landscape and Urban Planning
    Citation Excerpt :

    She reports that in the early 20th century, flood engineers, governments, and local residents were already well aware that “[w]hile dyking might prevent flooding some locations, it did so by in effect passing along the high water” (p. 62). Although this side-effect from floodwater redistribution by FCI is known (see Bower, 2010; Douglas et al., 2009), it continues to be neglected (Lebel & Sinh, 2009). Society tends to focus on the success of FCI in preventing most floods and the overall benefits of FCI, which are often measured solely in economic terms (Penning-Rowsell & Pardoe, 2012).

  • Trends in hydrological extremes in the Senegal and Niger Rivers

    2018, Journal of Hydrology
    Citation Excerpt :

    The stationarity assumption reigned during this “old hydrological world” (Milly et al., 2008). However, both increases and decreases of extreme discharges have been reported via the evaluation of historical series around the world (e.g. Kundzewicz et al., 2005; Bower, 2010; Condon et al., 2015). A main challenge of hydrological extremes thus concerns the validity of the stationarity assumption and the implications of its rejection.

  • Reflections on the American environment

    2014, Journal of Historical Geography
View all citing articles on Scopus
View full text