Growth and Erosion: A Reflection on Salt Marsh Evolution in the St. Lawrence Estuary Using HGIS

Matthew G. Hatvany

Introduction

Salt marshes are the focus of attention of scientists and government actors studying the impact of climate change on Quebec’s St. Lawrence Estuary.1 Viewed as some of the world’s richest ecosystems, under optimal conditions salt marshes are capable of producing several times more organic material per acre than the best agricultural lands. Historically, they were highly valued by Amerindians and colonial peoples as important hunting and fishing grounds and as fertile sources of natural pasture and fodder. Since the industrial period, however, they were frequently treated with disdain, being diked, drained, or filled for intensive agriculture, road construction, and urban expansion. In post-industrial times, however, appreciation of the marshes has come full circle, and they are now recognized for their “ecosystem services” as producers of organic material that forms the base of the estuarine food chain, filters pollutants from water, sequesters carbon, and buffers coasts against storm events.2

In the St. Lawrence Estuary, there are some 44 km2 (4,418 ha) of salt marsh. Since the beginning of the post-industrial period (1970–), there has been growing anxiety over the future of the marshes because of a perceived erosion of the coastline. Marsh erosion, coastal managers argue, menaces not only important ecosystem services but also the built environment (houses, roads, railroads, dikes, etc.).3 According to a survey of some eighty studies published between 1977 and 2004, the magnitude of that erosion is unsurpassed in modern history and driven by human activities.4 Anthropogenic causes, the survey argues, are responsible for intensifying storm events, accelerating sea-level rise and reducing protective winter ice cover, all of which result in erosion of coastal marshes. Additionally, activities like dam construction on major rivers traps sediments upstream, reducing their downstream availability in the estuary for marsh maintenance and building. As a result, one specialist notes, the sedimentary coastlines of the St. Lawrence Estuary and gulf have been eroding at a rate of 0.5 m to 2 m annually since 1931.5

As a citizen profoundly interested in society’s relationship with the environment, I am alarmed by the threat that erosion poses to the future of the St. Lawrence salt marshes. However, as a historical geographer of wetlands, I am also interested in their natural and cultural evolution. Regrettably, that concern with looking both backward and forward in time is not widely embraced by most natural scientists. The problem of “presentism” – or only looking forward – stems in large part from the underlying interest of funding agencies and governments in better understanding current and future problems. And yet, the study of the past has made me keenly aware that the current discourse of marsh erosion is quite novel – a product of the post-industrial era. In fact, it contradicts the historical understanding shared by medieval, colonial, and industrial period observers that salt marshes are constantly growing as a result of geomorophological processes like sedimentation and biological processes like plant succession. In the process of studying the medieval and early modern history of prominent salt marshes like those of the bays of Mont-Saint-Michel and the Aiguillon in France, it has been impossible for me to ignore the fact that the marshes of Western France have been growing at impressive rates. So rapid, in fact, that many marshes have been successively diked over the last millennium – new dikes being regularly constructed in rhythm with the seaward advance of the marshes.6

Many of the early colonists settling the banks of the St. Lawrence Estuary came from areas in France where successive marsh diking was known; thus, it is not surprising that in my study of the Kamouraska region of Quebec (Fig. 9.1) – where some of the largest salt marshes in the St. Lawrence Estuary are located – farmers employed many of the same marsh diking techniques as those employed in the Old World.7 From the 1850s until as late as 1983, lay and scientific observers all noted that the marshes of Kamouraska were growing at annual rates varying between 30 to 75 cm and more. Underscoring this point is the fact that the word “erosion” (érosion) is rarely mentioned in the plethora of archival documents on the marshes written before the 1970s.8

This conspicuous dichotomy between past and present discourses on salt marsh evolution is a conundrum that coastal researchers have failed to address.9 Could hundreds of years of historical observation have misrepresented marsh growth? Could modern science, with highly sophisticated tools for observing and measuring erosion, be distorted? Or, might there be other explanations? Have anthropogenic activities provoked a radical environmental shift in marsh evolution over the last forty years that might explain the apparent contradiction between the historical and current literature? Looking backwards and forwards along the environmental timeline can provide a more nuanced perspective on this question, on the impact of anthropogenic activities, and on the future of the marshes.

One way to address these questions is to turn to Historical Geographical Information Systems, or HGIS. With HGIS, it is possible to trace the physical evolution of one of the largest salt marshes in the St. Lawrence Estuary – the Bay of Kamouraska – using historical maps from the eighteenth and nineteenth centuries, aerial photographs from the twentieth century, and GPS (Global Positioning Systems) coordinates from the twenty-first century. HGIS is an evolving tool and methodology that over the last decade has shown substantial promise in answering pressing environmental questions. Outstanding examples from France and the United States of the use of HGIS in the study of salt marshes include Virginie Bouchard et al.’s work on the evolution of the marshes of the Bay of Mont-Saint-Michel; Fernand Verger’s study of the progression of the marshes of the Bay of the Aiguillon; Franchomme and Schmitt’s analysis of wetlands in the Nord Pas-de-Calais; and Keryn Bromberg Gedan and Mark Bertness’s quantitative study of human impacts on New England salt marshes.10

Clearly, the capacity to wed historical records with avant-garde computer-assisted mapping technologies makes HGIS an appealing research method for today’s environmentalists. And yet, as innovative as HGIS appears, it has a proven pedigree rooted in nearly a century of historical and geographical thinking about how best to visually represent natural and cultural change of landscapes over time. In the 1940s, historical geographer Andrew Hill Clark encouraged his students to combine historical and spatial data, arguing that the geographer’s “muddy boots” must go to the archives in order to master the temporal aspects of landscape evolution.11 Conversely, using HGIS obliges the temporal specialist to go into the field to observe and confirm spatial patterns revealed in historical maps, aerial photographs and textual descriptions. It is in this sense of a holistic vision of the natural and cultural landscape informed by field observations and archival work that I see the ultimate potential of HGIS as a tool for the study of environmental change.

Methodology

The salt marsh of the Bay of Kamouraska (Fig. 9.1) is an ideal study site to explore the utilization of HGIS as a tool and method of environmental analysis. Located on the South Shore of the St. Lawrence Estuary, some 150 km east of Quebec City, the marsh is considered “intermediate” in size for the St. Lawrence Estuary, about 6.5 km long and up to several hundred metres wide. Thanks to the presence of an active seigneur who mapped the marshes in the colonial era, the nearby establishment of an agricultural society and an agricultural school in the early industrial period, and significant interest in the marshes of the region by a succession of farmers, government ministries, historians, environmentalists, and geographers in the last century, the Bay of Kamouraska is the most historically documented salt marshes in the St. Lawrence Estuary in regard to both natural and cultural evolution.12

While most people intrinsically understand what a salt marsh is, there is ongoing debate over precise definitions and delineations. Marsh characterization is highly dependent upon disciplinary background.13 In this study, I use a simple description inspired by the interdisciplinary work of Desplanques and Mossman (2004) on the salt marshes of the Bay of Fundy. “Salt marsh” refers to a flat low-lying coastal area of vegetated marine soils that can be a few metres to several kilometres long and wide. It is periodically inundated by tidal salt water so that only plants adapted to saline conditions survive. This definition excludes unvegetated tidal flats below mean sea level.14 Salt marsh evolution is analyzed through a time-series study of the lateral movements of the lower limit of the marsh using methods inspired by Champagne et al., Bouchard et al., and Verger and based upon data obtained from historical maps, aerial photographs, GPS coordinates, and my own field observations of the Bay of Kamouraska over seventeen years (Fig. 9.2).15

Fig. 9.2. Forms of salt marsh limits on the Bay of Kamouraska. While many texts provide examples of growing (A) and eroding (B) marshes, field observations proved to be vital to identifying the existence of both process on the same marsh (C). These illustrations were drawn from field notes and photographs and then produced using Adobe Illustrator.

Tidal regimes play an important role in the geological and biological delineation of marshes. In the Kamouraska region, tides are macrotidal (several metres), rising and falling twice daily (semi-diurnal) with mean tides of 4.2 m. High water (HW) is 6.2 m, low water (LW) is 0.1 m, and mean sea level (MSL) is 3.16 m.16 In relation to these measures, the salt marsh is divided into low and high marsh according to the amount of time (daily, monthly, and annually) that the marsh is submerged by the tides. The inferior limit of the low marsh, submerged twice daily by the tides, begins around mean sea level and is associated with the growth of Spartina alterniflora, a salt-tolerant plant adapted to daily submersion. The carpet-like growth of Spartina alterniflora is easily identified on aerial photographs, making it possible to accurately determine the area covered by the low marsh.17 The high marsh, situated above the line of average high tides (known as mean high water or MHW), is submerged irregularly by only the highest tides and is dominated in Kamouraska by Spartina patens, Spartina pectinata and other plants capable of tolerating periodic submersion in salt water. Spartina patens and Spartina pectinata, while harder to individually identify than Spartina alterniflora, can nevertheless be distinguished on aerial photographs and used to accurately determine the area and limits of the high marsh.18

Over the last seventeen years of field observations, I have noted three forms of marsh limit, or seaward edge of the marsh, on the Bay of Kamouraska. These limits are here represented in schematic form using Adobe Illustrator (Fig. 9.2). The first form (Fig. 9.2A) is denoted by an abrupt scarp, varying between 10 and 30 cm in height, located at mean high water (MHW). This scarp neatly separates the high marsh from tidal flat and is indicative of a marsh undergoing active erosion. A second form (Fig. 9.2B) features a gradual transition between the low marsh and the tidal flat where isolated colonies of marsh grass are sometimes seen colonizing the tidal flat. This form is indicative of an actively growing marsh. The third form (Fig. 9.2C) combines both a scarp at high water (HW) and a gradual transition between the low marsh and the tidal flat. While this third type of marsh form is not documented in the St. Lawrence literature, since the 1960s Verger has noted in French studies that this form of marsh limit is indicative of simultaneous movements of both erosion and growth.19

Historic cadastral plans of the Bay of Kamouraska salt marsh exist in the provincial archives for 1781 and 1826, while a series of aerial photographs of the salt marsh at low tide exist for 1929, 1948, 1974, and 1985 at various provincial and national repositories. After scanning and importing into a GIS program the 1826 cadastral plan and all of the aerial photographs, my research team georeferenced them onto a standard projection to a scale of 1:20,000.20 Because of problems of distortion with the 1826 map and older photographs, overall precision is estimated to be ± 5 m.21 Once these spatial data were loaded into a GIS, it was possible to determine the overall area of salt marsh and to identify and trace the geographic evolution of the seaward edge of the salt marsh from 1929 to 1985.

The next step was to determine the seaward edge of the marsh in the present landscape. To do so, my research team made several field excursions to the Bay of Kamouraska in July 2008. We traced the seaward marsh edge on foot using a hand-held differential GPS (dGPS) that registered geographical coordinates every few metres. The final step in the research procedure was to combine the dGPS coordinates with the spatial data from the aerial photographs. GIS software allowed us to view the spatial data from the aerial photographs and GPS coordinates as a separate layer and then to merge all of the layers into a single image showing the evolution of the salt marsh limit over the period 1929–2008. Then, using the measuring tool in ArcGIS, we were able to note the differences in marsh limit at diverse dates and tabulate those changes in an Excel spreadsheet.

Anthropogenic Change (1781–2008)

Most study of the evolutionary trends of the salt marshes of the St. Lawrence Estuary is limited in time to the last eighty years, corresponding with the beginnings of aerial photography in the late 1920s. To go further back in time is possible but requires the use of more complicated and significantly less precise methods relying on proxy sources (paleoecologic and stratigraphic analysis of soil cores) combined with costly and imprecise radio-carbon dating techniques. However, in the case of the Bay of Kamouraska, I turned to another source of information – historical maps – that allowed me to extend the spatiotemporal range of study by an additional 140 years to the late eighteenth century. The detailed cadastral plan of part of the Seigneurie de Kamouraska by Jeremiah McCarthy in 1781 (Fig. 9.3A) was ideal, providing significant spatial information on the central and western sections of the Bay of Kamouraska. Clearly visible on the plan is the seaward edge of the salt marsh at mean sea level (MSL), in addition to information allowing me to identify the mean high water level (MHW) and the high water level (HW) on the salt marsh.22

A second cadastral plan of the entire Seigneurie de Kamouraska by Joseph Hamel in 1826 (Fig. 9.3B) affords the first complete view of the entire salt marsh, or prairies de grève, of the Bay of Kamouraska.23 This plan likewise provides information making it possible to identify the seaward edge of the marsh at mean sea level (MSL), and the upper limits of the salt marsh at mean high water (MHW) and high water (HW). These data are exceptional for reconstructing the environmental history of the marsh because they can be used to establish a baseline for delimiting the total area of the salt marsh before intensive anthropogenic activities began in the industrial period (after 1850). By using ArcGIS to draw a polygon around the entire marsh as represented in the plan by Hamel, I determined that the total area of the salt marsh in 1826 was 309.8 ha with a length of 6.7 km and mean width of 535 m.

The first aerial photographs of the salt marsh come from a 1929 mosaic archived in the Université Laval Centre GéoStat. Unfortunately, the coverage of this mosaic provides only a partial view of the central and western sections of the marsh.24 Fig. 9.4A shows the geographical coordinates of the seaward edge of the marsh in 1929, indicated in this HGIS layer by an overlain dashed line. The next series of aerial photographs covering the entire marsh, taken in 1948, was obtained from the National Air Photo Library in Ottawa (Fig. 9.4B). Here again the coordinates of the seaward edge of the marsh were noted in the HGIS layer and indicated by a dashed line.25

By placing the 1948 mosaic under magnification (using the zoom function in ArcGIS), I identified a dike (aboiteau) that was constructed on the interface between high and low marsh. Using archival documents, I ascertained that it was constructed in 1937–38 by the provincial government (Departments of Agriculture and Colonization) in collaboration with local farmers in order to augment agricultural production during the Great Depression. In addition, I noted a series of smaller secondary dikes several hundred metres south of the primary dike. Archival information revealed that these secondary dikes were built after 1941, following the breaching of the main dike by a severe storm event. These secondary dikes remained in place for over thirty years until the main dike was rebuilt in 1979–80.26 With this information, it was possible using GIS to draw a polygon around all of the remaining salt marsh in 1948. In doing so, I determined that diking had decreased the average width of the marsh from 535 m in 1826 to 428.6 m in 1948, reducing overall marsh area by about 18 per cent from 309.8 ha to 254 ha.

The next two series of aerial photographs analyzed come from the Université Laval Centre GéoStat (1974 series) and from the Quebec Ministry of Natural Resources (1985 series).27 As with the previous photos, I noted the coordinates of the seaward edge of the salt marsh in the GIS, creating separate layers for 1974 and 1985 (Figs. 9.4C and 9.4D). Placing the 1985 series of photos under magnification revealed the reconstruction of the primary dike in 1979–80 and the demolition of the secondary dikes (digues dormantes) after 1980. By drawing a polygon around the total area of marsh in the 1985 GIS layer, I determined the extent of marsh destruction due to reconstruction of the primary dike. By 1985 the average width of the salt marsh had been reduced to 214 m, while the area of the marsh decreased to 115 ha. This equals a loss of nearly 63 per cent of the marsh since 1826 as a result of diking and draining.

Natural Change (1929–2008)

Traditionally, when data on environmental change are presented textually, readers are forced to imagine landscape evolution. Mapping data provides the reader with an optical representation that enhances understanding by directly engaging the reader’s sense of sight.28 This is the objective of Fig. 9.5, where I combined the 1948, 1974, and 1985 GIS layers into a HGIS map and then added the dGPS coordinates obtained in 2008. By merging these separate layers into a single image, my goal was to produce a dynamic map illustrating the natural evolution of the salt marsh over the last seventy-seven years. To facilitate interpretation, the data were colour-coded by year and six observation points (A, B, C, D, E, and F) were overlaid on the map in order to measure and tabulate the movements of the seaward edge of the marsh at logically spaced intervals. The resulting map and embedded table (Fig. 9.5) provide a dynamic representation of the marsh, clearly illustrating both erosion and growth events.

Erosion is plainly visible on the map in the western half of the marsh at points A (–162 m), B (–70 m), C (–38 m) and D (–55 m). These findings are supported by the field observations made in 2008, where I noted a clearly visible scarp in the western section of the marsh separating the high marsh from the tidal flat (Fig. 9.6A). Equally visible on the map are areas of growth of the salt marsh in the eastern half of the bay at point E (107 m). Here, I noted in my field observations, both a scarp near the high water line and a gradual transition between the low marsh and the tidal flat at mean sea level (Fig. 9.6B). This combination of forms, as noted earlier, is indicative of both growth and erosion. Finally, the map and table illustrate significant lateral growth of the marsh at point F of more than 200 m. In my field observations for this sector, I noted the absence of any kind of scarp. On the contrary, there was abundant evidence of sedimentation and a gradual transition between the low marsh and tidal flat (Fig. 9.6C).

In regard to natural processes, the data in Fig. 9.5 illustrate that over the last seventy-seven years there has been both lateral growth and erosion of the salt marsh. In the western half of the salt marsh, overall change to the marsh edge is negative, with a loss of some –175,254.68 m2. In the eastern half of the salt marsh, on the contrary, change to the marsh edge has been positive, with an advance of some 153,650.68 m2. When these two data sets are combined, we note that changes in location of the marsh edge resulting from natural phenomena (erosion and sedimentation) have resulted in a net loss of about 2.1 ha (–21,604 m2).

By referring to the historical plans, it is possible to compare the visual form, location, and curvature of the salt marsh in 1781 and 1826 with that represented in Fig. 9.5. Through such a comparison, one immediately notes that the semi-circular form of the lower marsh edge has remained remarkably stable for nearly two centuries. This is especially poignant when I contrast the evolution of the Bay of Kamouraska salt marsh with that of other well-studied salt marshes in the United States and France. For instance, in the 1970s, Redfield demonstrated that the salt marshes of Cape Cod (Barnstable, Massachusetts) experienced radical changes in form over the last two centuries primarily due to sedimentation and growth resulting from natural processes.29 Similarly, Verger’s work on the Genêts marsh on the Bay of Mont-Saint-Michel (Normandy, France) demonstrates that it has eroded and regrown several times during the last century due to natural processes.30

Overall Change to the Bay of Kamouraska Salt Marsh since 1781

The data illustrate that over the last 230 years sedimentation and erosion on the Bay of Kamouraska salt marsh have been in relative equilibrium, the lower marsh limit retaining its basic form since the late eighteenth century. The same argument is not true when it comes to discussing direct human impacts on the marsh landscape. During Amerindian and colonial periods, human impacts on the marsh were negligible. However, with industrialization, anthropogenic activities played a significant role in reducing the area of high marsh as a result of diking and draining. According to the baseline data derived from the 1826 plan, the overall area of salt marsh before intensive agricultural practices began was approximately 309.8 ha. By 1948, the marsh had been reduced to 254 ha by diking and draining. Reconstruction of the primary dike in 1979–80 led to a further reduction of marsh area to 115.3 ha by 1985. Since then, an additional 28.3 ha were lost due to the complete drainage of all former tidal marsh behind the primary dike. Contrasting the data on anthropogenic and natural change of the salt marsh over the last two centuries, it is clear that humans – and specifically industrial society – have had the greatest impact on the evolution of the salt marsh. Over the last 182 years, 0.6 per cent (2 ha) of total marsh area changed as a result of natural processes (sedimentation and erosion). On the other hand, nearly 72 per cent (223 ha) of the marsh was transformed due to diking and draining.

Conclusions

What are we to make of the findings revealed by the HGIS, especially when juxtaposed with the historical discourse of marsh growth due to natural processes and the current discourse of marsh erosion resulting from climate change? Why does the evolution of the Bay of Kamouraska salt marsh correspond with neither interpretation? The answer is in part methodological. The combining of archival materials (historical texts and maps) with spatial data (aerial photographs and GPS coordinates) within a GIS provides us with the ability to look both forwards and backwards on the environmental timeline. What surely appeared to previous scholars as unidirectional marsh movements when viewed through the lens of a single lifetime or context of society becomes considerably more complex and multidirectional when viewed across several lifetimes and historical contexts.

The answer is also in part epistemological. Stephen J. Gould, in Time’s Arrow, Time’s Cycle, suggests that in dealing with complex questions human nature tends to reduce problems to more simple dichotomies – yes or no, black or white. Yet these dichotomies, he warns, have the unintended result of reducing the richness of reality. 31 Gould’s understanding of this problem perfectly fits the historical and current literature on the salt marshes of the St. Lawrence Estuary where marsh evolution has been reduced to simple dichotomies – unidirectional movements of either growth or erosion.

Building on Gould’s thesis, the nonconformity of the data generated by this HGIS suggests the vital importance of situating the interpretation of marsh evolution within the time and place out of which it emerges. In the industrial period, one cannot separate the prevailing scientific interpretation of growing marshes from the socioeconomic discourse of land expansion through diking, draining, and mastery of nature. The transformation of salt marshes to benefit human needs that occurred at this time influenced how those who came afterward – post-industrial scholars – interpret marsh evolution. In other words, the post-industrial discourse of generalized marsh erosion as a result of climate change and sea-level rise cannot be easily separated from the discourse of anthropogenic destruction of a fragile environment. In the case of the Kamouraska salt marshes, the negative response to the destruction of 72 per cent of the salt marsh in the industrial period created a discursive echo that today reverberates in the interpretation of the impact of climate change and sea-level rise on marsh evolution.32

Objective minds, history tells us, do not exist outside of particular contexts of time and space. Unidirectional understanding of marsh growth and erosion, therefore, must be viewed as mental constructions imposed on data rather than demanded by them. HGIS, by combining the spatial preoccupations of geographers with the temporal preoccupations of historians, makes it possible to study environmental change, not only from the present context in time, but equally from the vantage point of other contexts. In so doing, HGIS brings to the study of the environment, not just the ability to empirically observe change over time, but also variations in the interpretation of those changes according to different contexts of society. In the case of the growing and eroding salt marshes of the St. Lawrence Estuary, combining empirical observation of the physical environment with the epistemological preoccupations of the social sciences is key in explaining why salt marsh evolution in Kamouraska County corresponds with neither the historical nor the current understanding of marsh growth and erosion.

notes

1 J.-É. Joubert and É. Bachard, Un marais en changement, caractérisation du marais salé de la baie de Kamouraska (Rimouski, QC: Comité ZIP du Sud-de-l’Éstuaire, 2012).

2 M. Hatvany, “Wetlands,” Encyclopedia of American Environmental History 4 (2011): 1380–83; C. Desplanque and D. Mossman, “Tides and Their Seminal Impact on the Geology, Geography, History, and Socio-Economics of the Bay of Fundy, Eastern Canada,” Atlantic Geology 40, no. 1 (2004): 1–118.

3 P. Bernatchez and C. Fraser, “Evolution of Coastal Defence Structures and Consequences for Beach Width Trends, Québec, Canada,” Journal of Coastal Research 28, 6 (2012): 1550–1566.

4 P. Bernatchez and J.-M. M. Dubois, “Bilan des connaissances de la dynamique de l’érosion des côtes du Québec maritime laurentien,” Géographie physique et Quaternaire 58, no. 1 (2004): 45–71.

5 Comité ZIP de la Rive Nord de l’Estuaire, Forum citoyen 2007 sur l’érosion des berges et l’occupation du territoire en Côte-Nord (Baie Comeau, QC: Webcréation, 2007), 11.

6 M. Hatvany, “Wetlands and Reclamation,” in International Encyclopedia of Human Geography, ed. R. Kitchen and N. Thrift, vol. 12 (Oxford: Elsevier, 2009), 241–46; V. J. Chapman, Salt Marshes and Salt Deserts of the World (New York: Interscience); V. Bouchard, F. Digaire, J.-C. Lefeuvre, and L.-M. Guillon, “Progression des marais salés à l’ouest du Mont-Saint-Michel entre 1984 et 1994,” Mappemonde 4 (1995): 28–33; F. Verger, Zones humides du littoral français : estuaries, deltas, marais et lagunes (Paris: Belin).

7 H. Charbonneau and N. Robert, “Origines françaises de la population canadienne, 1608–1759,” in Atlas historique du Canada, I : Des origins à 1800, dir. R. C. Harris (Montréal : Les Presses de l’Université de Montréal, 1987), plate 45; M. Hatvany, “The Origins of the Acadian Aboiteau: An Environmental Historical Geography of the Northeast,” Historical Geography 30 (2002): 121–37; Y. Suire, Le Marais poitevin : une écohistoire du XVIe à l’aube du XXe siècle (La Roche-sur-Yon : Centre vendéen de recherches historiques).

8 J.-D. Schmouth, “Mise en culture des terrains envahis par les eaux salées,” École d’agriculture de Sainte-Anne, Sainte-Anne-de-la-Pocatière, unpublished text, 1874, reprinted in La Gazette des Campagnes, 15 septembre 1942, 152–54; A. Hamel, “La récupération et la mise en valeur des alluvions maritimes du St-Laurent,” Agriculture 20, no. 3 (1963): 77–83; A. C. Redfield, “Development of a New England Salt Marsh,” Ecological Monographs 42, no. 2 (1972): 201–37; G. Gourde, Les aboiteaux : comté de Kamouraska (Québec : Ministère de l’agriculture, des pêcheries et de l’alimentation, 1980); J.-B. Sérodes and M. Dubé, “Dynamique sédimentaire d’un estran à spartines (Kamouraska, Québec),” Le naturaliste canadien 110 (1983): 11–26; P. Champagne, R. Denis, and C. Lebel, Établissement de modèle caractérisant l’équilibre dynamique des estrans de la rive sud du moyen estuaire du Saint-Laurent, Rapport manuscript canadien des sciences halieutiques et aquatiques no. 1711 (Quebec : Ministère des pêches et des océans, 1983); M. Hatvany, Marshlands: Four Centuries of Environmental Change on the Shores of the St. Lawrence (Sainte-Foy : Les Presses de l’Université Laval, 2003).

9 See, for instance, Joubert and Bachard, Un marais en changement.

10 Bouchard et al., “Progression des marais salés”; Verger, Zones humides; M. Fanchomme and G. Schmitt, “Les zones humides dans le Nord vue à travers le cadastre napoleon: les Systèmes d’Informations Géographiques comme outil d’analyse,” Revue du Nord 36 (2012): 661–80; K. D. Bromberg and M. K. Bertness, “Reconstructing New England Salt Marsh Loss Using Historical Maps,” Estuaries 26, no. 6 (2005): 823–32.

11 A. H. Clark, “Field Work in Historical Geography,” Professional Geographer 4 (1946): 13–23.

12 Hatvany, Marshlands, 164.

13 Hatvany, “Wetlands”; J.-C. Dionne, “Âge et taux moyen d’accrétion verticale des schorres du Saint-Laurent esturaien, en particulier ceux de Montmagny et de Sainte-Anne-de-Beaupré, Québec,” Géographie physique et Quaternaire 58, no. 1 (2004): 74–75.

14 Desplanque and Mossman, “Tides and Their Seminal Impact.”

15 Champagne et al., Établissement de modèle caractérisant l’équilibre dynamique des estrans; Bouchard et al., “Progression des marais salés”; F. Verger, Marais et wadden du littoral français : etude de géomorphologie (Bordeaux: Biscaye Frères, 1968); Verger, Zones humides.

16 Fisheries and Oceans Canada, “Tides, Currents, and Water Levels (Pointe aux Orignaux),” http://www.waterlevels.gc.ca/; Champagne et al., Établissement de modèle caractérisant l’équilibre dynamique des estrans, 7.

17 K. L. McKee and W. H. Patrick, Jr., “The relationship of smooth cordgrass (Spartina alterniflora) to tidal datums: A review,” Estuaries 11, no. 3 (1988): 143–44.

18 M. Hatvany, Paysages de marais : Quatre siècles de relations entre l’humain et les marais du Kamouraska (La Pocatière: Société historique de la Côte-du-Sud et Ruralys, 2009), 21.

19 On marsh limits, see Champagne et al., Établissement de modèle caractérisant l’équilibre dynamique des estrans, 11; Verger, Marais et wadden, 277, 289–93; and Verger, Zones humides, 57–58.

20 The preliminary collection and analysis of aerial photographs and GIS mapping of the Bay of Kamouraska was done under my supervision and that of D. Cayer by C. Careau as part of his thesis entitled “Les marais intertidaux du Saint-Laurent : Complexités et dynamiques naturelles et culturelles,” MS thesis, Dépt. de géographie, Université Laval, 2010.

21 Québec, Base de données topographiques du Québec (BDTQ) à l’échelle de 1/20 000, Normes de production, version 1.0, Québec, Direction de la cartographie topographique, Ministère des Ressources naturelles et de la Faune, 1999.

22 J. McCarthy, “Plan and Survey of Cap-au-Diable in Kamouraska,” 1781, Bibliothèque et Archives nationales du Québec à Québec, CA301, S45, no. 3.

23 J. Hamel, “Plan de la seigneurie de Kamouraska,” 1826, Bibliothèque et Archives nationales du Québec à Québec, P407, famille Taché, no. 2.

24 Centre d’information géographique et statistique, Bibliothèque générale de l’Université Laval, Mosaïque aérienne du Québec, 1929, F82, p. 3.

25 National Air Photo Library (Ottawa, Canada), A11660-290, 1948.

26 Hatvany, Marshlands, 134–38.

27 Centre d’information géographique et statistique, Bibliothèque générale de l’Université Laval, Q74316-94-95-96 and Q74313-133-134-135 (1974); Ministère des Ressources naturelles, Gouvernement du Québec, Q85913-143 (1985).

28 D.C.D. Pocock, “Sight and Knowledge,” Transactions of the Institute of British Geographers 6 (1981): 385–93.

29 Redfield, “Development of a New England Salt Marsh,” 235.

30 Verger, Zones humides, 191.

31 S. J. Gould, Time’s Arrow, Time’s Cycle: Myth and Metaphor in the Discovery of Deep Geological Time (Cambridge, MA: Harvard University Press, 1987), 199–200.

32 This discursive echo is most clearly seen today in Joubert and Bachard, Un marais en changement and in Bernatchez and Dubois, “Bilan des connaissances de la dynamique de l’érosion des côtes du Québec maritime laurentien.”