Warning This Web page has been archived on the Web.

Archived Content

Information identified as archived on the Web is for reference, research or recordkeeping purposes. It has not been altered or updated after the date of archiving. Web pages that are archived on the Web are not subject to the Government of Canada Web Standards, as per the Policy on Communications and Federal Identity.

Skip booklet index and go to page content

COSEWIC Assessment and Status Report on the Yellow Rail in Canada

Habitat

Habitat Definition

The Yellow Rail prefers marsh habitat with dense, fairly low herbaceous vegetation with little or no standing water (generally 0-12 cm) where the substrate remains saturated throughout the summer. It can be found in damp fields and meadows, on the floodplains of rivers and streams, in the herbaceous vegetation of bogs, and at the upper levels (drier margins) of estuarine and salt marshes (Bookhout 1995, Robert 1996, Robert and Laporte, in prep.).

Previous studies have found that sites harbouring Yellow Rails during the breeding season are usually in marshes large enough to support several pairs (>10 ha) (Walkinshaw 1939, Terrill 1943, Brewer et al., 1991, Gibbs et al., 1991, Robert and Laporte, in prep.). However, Bookout (pers. comm.) has found them during the breeding season in sedge meadows as small as 0.5 ha in Michigan. They also have been found during the breeding season in numerous 1 ha sites in the Cold Lake area of Alberta (W. Smith, pers. comm.), several 2-3 ha sites at Churchill and in southern Manitoba (R. Koes, pers. comm.), and in numerous 3-4 ha sites in the Rainy River region of northwestern Ontario (D. Elder, pers. comm.). In Alberta, T. Sadler (pers. comm.) has never encountered Yellow Rails in sedge meadows larger than 2 ha; rather, he has found them in numerous sedge meadows less than 1 ha.

Nesting Yellow Rails are typically associated with marshes dominated by sedges, true grasses, and rushes, particularly by fine-stemmed emergents of the genera Carex, Spartina, Juncus, Calamagrostis, Scirpus, Eleocharis, and Hierochloe. Nesting habitats usually have a dry mat of dead vegetation from previous growing seasons (Dawson 1921, Peabody 1922, Roberts 1932, Fuller 1938, Devitt 1939, Walkinshaw 1939, Huber 1960, Stalheim 1974, Savaloja 1981, Stenzel 1982, Bookhout et Stenzel 1987, Gibbs et al., 1991, Grimm 1991, Stern et al., 1993, Bookhout 1995, Robert and Laporte 1996, Robert and Laporte, in prep.). Although one of twelve nests discovered by Maltby (1915) did not have one, this concealing canopy nevertheless is quite characteristic of Yellow Rail nesting habitat, and females pull it back over the nest if disturbed (Stalheim 1974, Stenzel 1982, Robert and Laporte, in prep.). Stenzel (1982) suggested that a senescent vegetation canopy allows the birds to move freely without visual detection. It might be particularly important during the first weeks after spring arrival of the rails, when the green vegetation has not yet grown enough to offer them much concealment (Robert and Laporte, in prep.).

Although Carex sedges often characterize habitats used by Yellow Rails during summer (Bookhout and Stenzel 1987, Gibbs et al., 1991, Stern et al., 1993, Robert and Laporte, in prep.), many other fine-stemmed plants may also predominate. The importance of Carex in Yellow Rail habitats may simply be due to the hydrophytic nature of the genus, its tendency to occupy shallow or damp areas and, more importantly, its extensive variation and distribution in temperate latitudes of North America (Marie-Victorin 1995). Because of this, Robert and Laporte (in prep.) believe that Carex sedges should not be considered as the only indicator plant species (Stenzel 1982, Gibbs et al., 1991) of Yellow Rail habitats. Rather, the Yellow Rail may tolerate considerable variation in certain subtle features of its summer habitats, such as plant species composition, stem density, and height of vegetation, as do other rail species (Rundle and Fredrickson 1981, Flores and Eddleman 1995, del Hoyo et al., 1996). Habitat selection might be influenced primarily by plant physiognomy and maximum water levels.

The Yellow Rail uses a greater diversity of habitat types during migration and winter than during the breeding season. In fall, as well as using wet meadows and coastal marshes, it may occur in rice fields, dry hay fields, and cereal fields (Bent 1926, Ripley 1977, Savaloja 1981, Godfrey 1986, Cardiff and Smalley 1989). On Texas wintering grounds, it is often found in coastal marshes dominated by Spartina patens (Stalheim 1974, Anderson 1977, Lane and Tveten 1984), while in Louisiana, it is often found in rice fields (Lowery 1974, Cardiff and Smalley 1989). It is also observed in coastal Spartina marshes during spring migration (Shoch 1990). In South Carolina it is found in short-grass low meadows and power-line right of ways running through certain wetlands that are mowed on a regular basis (J. Cely, pers. comm.).

Habitat Distribution

In Canada, a large proportion of habitat suitable for Yellow Rails occurs along the west coast of Hudson Bay (Manitoba and Ontario) and along the James Bay coast (Ontario and Québec). In our view, this poorly known region is potentially a major stronghold for the species. In Ontario, the shores of James Bay and Hudson Bay are extremely flat, forming an extensive coastal plain that is poorly drained and dominated by open fens and bogs interspersed with treed fens and bogs of Black Spruce (Picea mariana) and Tamarack (Larix laricina) (Cadman et al., 1987). The Hudson Bay Lowland of Ontario comprises about 260 000 km2, or one quarter of the province (Wilson and McRae 1993), and it would be useful to attempt to assess the proportion of the whole region that could be suitable for Yellow Rails.

Even though no Yellow Rail surveys have been conducted on the Québec side of James Bay, we already have a good idea of the amount of potential habitat there (Robert et al., 1995). There may be about 20 000 ha of marsh south of the Castor River (Consortium Gauthier & Guillemette - GREBE 1992b), including both low and high-marshes. High-marshes, littoral meadows and back marshes respectively cover 1635, 1075 and 211 ha of Cabbage-Willows Bay and 1073, 37 and 1222 ha of Boatswain Bay. Elsewhere in the Rupert Bay estuary, these cover 1186, 239 and 700 ha, respectively (Consortium Gauthier & Guillemette - GREBE 1992c). This represents, in theory, a little more than 7000 ha of potential Yellow Rail habitat in Boatswain Bay and Rupert Bay, which cover the majority of potential habitat for this species on the Québec side of James Bay. Northward of the Castor River, the coastal marshes are less numerous and less vast than in south-eastern James Bay. Of the 264 marshes mapped by (Dignard et al., 1991), less than 10 are larger than a few square kilometres. They are mostly located in the upper portions of large bays such Dead Duck Bay, Aquatuc Bay, Baie des Oies, Paul Bay and Bay of Many Islands.

Eastern Alberta seems to contain considerable amounts of potential habitat, as do the Parklands and southern Boreal Forest of Saskatchewan, and the Parklands and southern Boreal Forest of Manitoba. The Peace Parklands of northwestern Alberta, which extend about 300 km north, almost to the border with the Northwest Territories, is an isolated extension of aspen parkland. It contains sedge meadows that, once dry, are used for nesting by several waterfowl species (B. Caverley, pers. comm.). Yellow Rails have not been found in this region, but likely little if any surveying has been done. Southern Ontario has lost a large proportion of its wetlands (Austen et al. 1994), and the population, which likely used to have a widespread breeding distribution in the province, has been reduced to one or two sites (Cadman et al., 1987).

In southern Québec, sedge meadows cover 1480 ha between Cornwall (Ontario) and Trois-Rivières and they are restricted to the Lake Saint-François region (Aménatech 1992a). From Trois-Rivières to Montmagny, most suitable habitats are located at Île aux Grues and Cap Tourmente and cover more or less 900 ha (Robert et al., 1995, Robert and Laporte 1996). Overall, approximately 2400 ha of marshes between Cornwall and Montmagny are probably suitable for breeding Yellow Rails. Elsewhere along the St. Lawrence River, no precise estimates exist of the area covered by high-marshes like the ones inhabited by Yellow Rails. A number of marshes are located on the south shore of the St. Lawrence estuary, between Saint-Jean-Port-Joli and Matane, a result of the large areas of fine deposits found there. The largest marshes on the south shore are those of Isle-Verte (1369 ha), Rivière-du-Loup (325 ha), Trois-Pistoles (252 ha), Saint-André (200 ha) and Kamouraska (150 ha). Along the north shore of the St. Lawrence estuary, the steep shoreline largely precludes the presence of marshes. The largest marshes are those of Outardes Bay (593 ha), Baie-Saint-Paul (304 ha), Milles-Vaches Bay (249 ha) and the Îlets Jérémie (121 hectares) (Desponts et al., 1995, Robert et al., 1995). Along the shoreline of the gulf of the St. Lawrence River, large marshes are rare and usually confined to river-mouths where fine sediments accumulate, particuliarly in sites protected by long littoral tongues and in deep bays. On the Gaspé Peninsula, the largest marshes (200-400 ha) are those of Gaspé, Malbaie, Paspébiac and Restigouche (Desponts et al., 1995). Of course only a fraction of all these areas consists of high-marsh that could be suitable for Yellow Rails. Finally, it is possible that some large fens in the Abitibi and Lac-Saint-Jean regions harbour some Yellow Rails.

Potential habitat in New Brunswick includes the Grand Lake Meadows portion of the St. John River estuary, plus about 30 sedge meadow sites in the province’s interior (P. Kehoe, pers. comm.).


Trend in Quality and Quantity of Critical Habitat

General

Loss of wetlands to human activity is probably the most serious factor affecting Yellow Rail populations (Bookhout 1995). Negative effects on Yellow Rail habitat might be particularly important because of this species’ occupation of the drier parts of marshes (Eddleman et al., 1988).

Another very important factor affecting the amount of sedge meadow habitat at a given place and time is climatic conditions: the number of sites in an area varies from year to year according to water levels. If wet conditions persist for several years, sedge meadows will become cattail/rush marshes, whereas if dry conditions persist, willows and grasses will encroach (W. Smith, pers. comm.).


United States

About half of the coastal wetlands in many eastern states have been lost to dredging and filling (Eddleman et al., 1994), and since the mid-1950s, estuarine wetland loss in the U.S. coastal zone has accelerated to about 0.5% annually (Schneider and Pence 1992). By the mid-1970s, only 46% of the original wetlands in the conterminous U.S. remained, and the rate of loss continued at more than 160 000 ha annually a decade ago. The most threatened habitats include palustrine and riverine wetlands, which are important for several rail species. Agricultural development is responsible for 87% of recent losses in the U.S., with urban, industrial and reservoir development accounting for the rest (Eddleman et al., 1988).

An important factor in the loss of both breeding and wintering habitat in the U.S. is that laws concerning the draining of wetlands may or may not apply to drier wetlands of the type that Yellow Rails use. It is very easy to farm the edges of wetlands that Yellow Rails use. Thus, habitat loss is a large concern regarding Black Rails (Laterallus jamaicensis) and Yellow Rails in the U.S. In the midwest, for example, very little or no habitat remains for Yellow Rails, as this habitat disappeared a long time ago. Unfortunately, there seems to be little information on the status of the Yellow Rail in the midwest beforehand (B. Eddleman, pers. comm.).

The apparent increase in the Oregon Yellow Rail population documented in the last few years may be due to a delayed effect of the end of the drought in the early 1990's. There is probably more acceptable nesting habitat due to increased water depths, as well as better nesting habitat due to more live and senescent vegetation (K. Popper, pers. comm.). However, the Oregon population is under similar threats faced by other populations, mainly the ditching and draining of wet meadows for agricultural use and overgrazing of nesting habitat. One of the problems in the Klamath Basin is the cleaning of existing ditches. Not only do people drain a field in which the water table has risen in response to the ditches slowly filling in with sediment and vegetation, but the ditches may be deepened further than their original depth. This may have negative effects not only on the land surrounded by the ditches and canals, but also on adjacent land where Yellow Rails breed. On the other hand, Yellow Rail nesting areas on public (federal) lands may become flooded late in the season (July) due to flood-irrigation occurring on adjacent private lands. Water levels strongly affect the presence and movement of the rails, so water management is of great importance to the continued health of the populations in Oregon and elsewhere in the U.S. The current habitat trend in south-central Oregon is relatively stable, with fluctuation from year to year depending on precipitation/water levels. Historically, there has been a drastic reduction, with only about 15% of the historic wetlands remaining in the Upper Klamath Basin (K. Popper, pers. comm.).

Habitat loss for wintering Yellow Rails has been so extensive in the U.S. that the wintering range may no longer be contiguous. It would be very useful for research to be done on this possibility (T. Bookout, pers. comm.). In Texas and elsewhere in the U.S. wintering grounds, the species seems to be largely restricted to a narrow band (e.g. 48 km) of coastline (K. Mizell, K. Arnold, pers. comm.). Coastal marshes throughout the Gulf states are threatened. In Texas, for example, all of the coastal marshes used by this species, which are located along the northern portion of the coast, are threatened by development or agriculture. A major factor is the reduction of federal government subsidies for rice farming, which is inducing farmers to sell their land for development or to use it for cattle grazing (K. Mizell, pers. comm.). In addition, rice fields along the Gulf coast are being converted to other uses. The rate of conversion is not significant, but it is steady. In Texas, where pumping ground water is necessary for rice culture, rice fields are being converted to pasture and residential areas. In Louisiana, where both ground water and stored water (in canals) are used for growing rice, the trend is to replace rice with sugar cane. The federal rice subsidy has been steadily reduced each year, and in a couple more years it will be unavailable to farmers. For this reason, many farmers are either selling out to other interest groups or, in the case of Louisiana, are converting to sugar cane which still benefits from a federal subsidy. The general trend in native wetland acreage in Texas and Louisiana is also down (C. Cordes, pers. comm.).


Canada

Hudson/James Bays

In the Hudson Bay Lowlands of northern Manitoba, a major influence on coastal habitat, possibly including habitat for Yellow Rails, is overgrazing by Snow Geese (D. Hussell, pers. comm.). This species has increased tremendously in numbers over the past three decades. It has learned to exploit farmland on its U.S. wintering grounds and on its U.S. and Canadian migratory stop-over sites, rather than relying on saltwater and freshwater marshes, which are not always as productive. As a result, birds tend to be in peak physical condition when they arrive on their breeding grounds, and breeding success has been much higher than in the past. However, the population has become so large that overgrazing is becoming a considerable problem, not only for the geese, a number of which starve before they fledge, but also for other species (Abraham and Jeffries 1997, K. Ross pers. comm.).

The high density area of the Snow Goose colony at La Pérouse Bay, east of Churchill, for example, has moved a number of kilometres away from its original location, leaving behind severely altered habitat (exposed peat, extensive salt pannes, dead willow thicket) (K. Ross, pers. comm.). The geese have caused substantial changes to all inter-tidal habitats at La Pérouse Bay. There is much less vegetation, and large areas of dead willows are present in the coastal zone. Nearly all shoots of Carex aquatilis are grazed up to 10 km from the coast. In the vicinity of the coast, extensive moss carpets are present (Abraham and Jeffries 1997). Regarding La Pérouse Bay, R. Rockwell (pers. comm.) writes: “In the 1960s and early 1970s we used to hear them [Yellow Rails] regularly. I have not heard any since 1982. For the last 3 years [1996-1998] I have made a concerted effort with dragging, listening, etc. and have been unsuccessful. My best guess is they are no longer in the region they once occupied. [This is] not a big surprise since much of that is a degraded wasteland.”

This habitat degradation is at various stages along the James Bay and southern Hudson Bay coasts, which coincide with the Yellow Rail’s breeding range. There could be an impact on the Manitoba and Ontario populations, although at present the effect would be quite local (K. Ross, pers. comm.). All areas along the Hudson/James Bay coast that have Snow Geese nesting colonies have been impacted, and many staging areas as well (K. Abraham, pers. comm.). 

Northwest Territories

Habitat changes in the Slave River Lowlands of the Northwest Territories, which fall within the Yellow Rails’s very small known range in the territory, are causing concern for Moose (Alces alces). Logging has been ongoing in the area for over 50 years. In addition, prescribed burning of willow-choked meadows is being done by the territorial government to improve the habitat for American Bison (Bison bison) (Bradley et al., 1996). Without knowing specifics of exactly where Yellow Rails may breed and where these habitat changes are occurring, it is difficult to comment on potential effects.

Prairies

In prairie Canada, the number of wetlands available to waterfowl (and presumably also to Yellow Rails) is subject to broad annual fluctuations, and this reflects the highly variable nature of temperatures and precipitation throughout the midcontinent (Turner et al., 1987). From a largely pristine environment at the turn of the century, the prairie-parkland region now comprises the largest expanse of agricultural land in Canada. The demand for increased agricultural production was formerly met by breaking large tracts of new lands. As the supply of suitable new lands decreased, farmers strived for increased production through more-intensive operations on their existing land holdings. These activities include draining, filling, haying and cultivating of wetlands, and the clearing and cultivating of marginal lands, resulting in the progressive and incremental degradation of waterfowl (and presumably Yellow Rail) habitat (Turner et al., 1987). As of 1986, about 40% of the original wetlands in Prairie Canada had been lost (Canada/United States Steering Committee 1986).

A study conducted on more than 10 000 potential wetlands across the prairie-parkland region from 1981 to 1985 showed that wetland margins (the area contiguous with and extending 10 m beyond the outer edge of the wet meadow zone of the wetland) were consistently altered at a faster rate than were the wetland basins (defined as the centre of the wetland to the outer edge of the wet meadow zone). Degradation of wetlands decreased with wetland permanency, and this likely reflected the higher vulnerability of less permanent wetlands. Degradation rates of both wetland basins and margins were found to be increasing (Turner et al., 1987).

Large dyking and damming projects were traditionally used to create or improve waterfowl habitat; such management was usually applied to areas 10 ha or greater. Although sedge meadows were flooded and therefore rendered useless for Yellow Rails, new sedge meadows would apparently develop around the edge of the impoundment and these new sedge meadows were secured from drainage. Furthermore, while the natural sedge meadows had been subject to random water level fluctuations, sedge meadows that formed after damming would be managed to keep them as sedge meadow over the long term (T. Sadler pers. comm., B. Calverley pers. comm.).

Recent research has shown that one of the greatest limiting factors for successful waterfowl breeding is nesting habitat for upland nesting species. As a result, the new Prairie CARE program under the North American Waterfowl Management Plan, which became fully operational in 1991, focusses mainly on creating and improving upland nesting cover. In the Parkland region, extensive upland areas are seeded to create nesting cover, while grazing and haying have been eliminated except for wildlife management purposes. This kind of management, which is done in cooperation with landowners through a number of different types of agreements, is targeted in areas with high wetland densities (e.g. 100-150 potholes/square mile). It helps not only waterfowl, but apparently also species such as Sedge Wrens (Cistothorus platensis), Le Conte’s Sparrows (Ammodramus leconteii) and Yellow Rails, which use areas of dense nesting cover surrounding wetlands (T. Sadler pers. comm., Brett Calverley pers. comm.).

Alberta

In Alberta, industrial development is occurring in many areas of the province at a very rapid rate. Conventional oil and gas development is expanding throughout much of the province, whereas heavy oil extraction is increasing in the north. All oil and gas development creates infrastructure that can affect habitat, including roads, above-ground and below-ground pipelines, powerlines, and seismic lines (G. Hamilton pers. comm., D. Hervieux pers. comm.). This development is expanding on the landscape scale and is thus affecting many habitat types, including wetlands. Two bird species that often nest in the same areas as Yellow Rails, but for which enough is known to assess their status in Alberta, are the Sedge Wren and the Willet (Catoptrophorus semipalmatus). Both are on the Yellow B list, meaning that they they warrant management attention. Both have suffered breeding habitat losses from drought conditions combined with wetland drainage (Alberta Environmental Protection 1996).

Peatland extraction is a new industry that is extending into non-forested portions of some northern Alberta areas (D. Moyles, pers. comm.). Even though Yellow Rails are not directly associated with peatlands, this activity could potentially affect water levels of the mined area and its surroundings, which could harbour sedge meadows. Recent documentation of the occurrence of Yellow Rails in the extensive Red Lake Peatland of Minnesota suggests that, with the continued destruction of wetlands in other parts of the species’ range, peatlands may constitute an important refuge for the species (Coffin and Pfannmuller 1988).

Land clearing and drainage of wetlands for crop and pasture are problems in both known Alberta Yellow Rail strongholds: the Cold Lake and Buffalo Lake areas. Logging is another factor in the Cold Lake area. When land clearing and logging do not destroy wetlands directly, the hydrological regime is usually altered such that the shallow wetlands remain dry for longer periods of time and disappear after a few years (W. Smith pers. comm.).

Saskatchewan

The recent elimination by the federal government of the Crow Rate, a subsidy for grain production that stimulated land clearing, has made it much less feasible economically to clear land for agriculture. This should have a positive effect on wildlife habitat in general, including Yellow Rail habitat. During the past four years, wet conditions have reflooded many areas in southern Saskatchewan. Nevertheless, in the long term, wetland habitat continues to be lost to agriculture. An estimated 40 000 ha of wildlife habitat are lost each year in Saskatchewan, mostly to forest clearing, cultivation of grassland, and pothole drainage. Most of the drainage is occurring in the south-east quarter of the province. Fortunately, about 10 000 ha of habitat are being protected each year by Ducks Unlimited and other partners through the North American Waterfowl Management Plan, of which some may be suitable for Yellow Rails (B. Macfarlane, pers. comm.).

Manitoba

In southern Manitoba, there is still a considerable amount of wetland being drained for agriculture. Even though drainage has become less feasible economically (reduced public subsidies) and politically (increased cooperation between agencies) as in Saskatchewan, and in fact is illegal, it is still done on a small scale by individual farmers, particularly with small seasonal potholes/basins. These changes are difficult to quantify, particularly at the natural community level (e.g. sedge meadow). However, it is clear that Yellow Rail habitat continues to be lost to agriculture, albeit at a reduced level than formerly (D. Hudd, pers. comm.).

Ontario

The Canadian Shield north of road or rail access has been little affected by direct human activity (Cadman et al., 1987). However, the forest industry is pushing north and logging will likely continue north to the tree line (Cumming 1997). Yellow Rails in the Rainy River region have lost a considerable number of small (3-4 ha) sites in which they could be heard 25 years ago to wetland drainage (D. Elder, pers. comm.). However, one large site remains and is protected.

In southern Ontario, habitat has declined to the point where the species has almost been extirpated as a breeder. Of the 2.3 million ha of wetlands originally found in southern Ontario south of the Canadian Shield, fewer than 12% remain today. This loss has been largely a result of human settlement and agriculture. The area in which the greatest proportion of wetlands has been drained is the extreme southwest. Probably the best Ontario example of a known Yellow Rail breeding site being greatly degraded is Holland Marsh, on the south shore of Lake Simcoe, which has been surrounded by a drainage ditch since the 1920s largely for agriculture (Day 1927), and where only small habitat patches remain (R. Harris, pers. comm.). In 1927, about 280 ha of the marsh were reclaimed and brought under cultivation for market gardens (Devitt 1967). Already in the 1960s Devitt (1967) wrote that: “this section of the marsh has been drained and given over to market gardens causing the Yellow Rails to abandon the area”. The Le Conte’s Sparrow, which also used to breed at the Holland Marsh, was also much affected. Even though the Yellow Rail may still breed there (G. Bennett, pers. comm.), this area is only a ghost of its former self.

Numerous small wet meadows in the Rainy River district have been drained with ditches over the past 25 years. D. Elder (pers. comm.) has witnessed the loss of Yellow Rails from many of these sites during this period.

Quebec

In Québec, loss of habitat by the draining and filling in of many wetlands found along the St. Lawrence River is the most important limiting factor for the Yellow Rail (Robert et al., 1995). There is no doubt that Yellow Rail sites along the St. Lawrence River are rarer today than in the past, because the extent of wetlands there has declined considerably over the past few decades. About 40% of the original coastal marshes along the St. Lawrence River have been converted (lost) from 1950 to 1978 (Bouchard et Millet 1993). Furthermore, it seems that many of the converted marshes were high-marshes because, being drier than the lower-marshes, they are easier to drain and dyke (Robert et al., 1995). For example, the construction of dykes in the St. Lawrence estuary caused the highest portion of salt marsh to retreat several metres (Couillard and Grondin 1986) and also caused the disappearance of many sites that might have been used by Yellow Rails. Also, the intertidal marshes of the Kamouraska area, which stretch over 26 km between Pointe Saint-Denis and Rivière des Caps on the south shore of the St. Lawrence River, and which used to cover 9.33 km2, covered only 3.91 km2 in 1986. Thus, 542 ha disappeared.

The changes to these intertidal marshes are of anthropogenic origin. They were largely a result of dyking to convert lands for agriculture, and they occurred between 1930 and 1986 (Pelletier et al., 1990). Many lagoons and marshes along the St. Lawrence River were also converted by filling and by various types of constructions. For example, the installation of harbour infrastructure at Cacouna beginning in 1965 resulted in the direct loss of 1.08 km2 of marsh, as well as changes to the hydrodynamic and sedimentation regimes of the site that changed the plant community structure (Pelletier et al., 1990). In addition, infrastructure put in place along the St. Lawrence River, for example Highway 20 near Sainte-Anne-de-la-Pocatière and Rivière-du-Loup , and Highway 138 between Québec City and Sainte-Anne-de-Beaupré, must also have contributed to the loss of numerous Yellow Rail marshes in southern Québec (see Campagna 1931). Overall, at least 50% of the potential Yellow Rail habitat along the St. Lawrence River and the Saguenay River disappeared over the past few decades (Robert et al., 1995).

Potential effects on Yellow Rails of dyking areas to increase water levels for the benefit of waterfowl have not been examined in Québec. Nevertheless, the Marais Antoine in the Abitibi region was flooded in 1996 and the species was not found in the area in 1997 (M. Robert, pers. obs.). No Yellow Rails were heard during the 1998 nesting season; however, one individual was heard during the first week of August (R. Deschênes, pers. comm.). This site is known to have had Yellow Rails and considerable habitat for the species before flooding. While flooding may have initially reduced Yellow Rail habitat, J.-P. Laniel (pers. comm.) suggests that the amount of such habitat is likely to increase over the years following flooding because expansion of the wetland should have resulted in a larger perimeter in which sedge meadow would colonize (as often occurs after such management in the prairies). It remains to be seen to what extent sedge meadow re-colonizes and how many years it takes for Yellow Rails to return.

Maritimes

In the Maritimes, wetlands suffered more from human actions since 1600 than most other habitats, although only dumping and sewage were widespread influences on them before 1900. Dyking of saltmarshes for hayland and pasture began around the upper Bay of Fundy in the late 1600s (Erskine 1992), and this habitat would have been lost to any Yellow Rails present, as it was to Nelson’s Sharp-tailed Sparrows (Ammodramus nelsoni) and Willets. The drainage of fresh marshes, ponds, and lakes for agriculture was done locally and on a small scale before World War II, but became widespead afterward and continued until about 1960, with a slowing or even a slight reversal of that trend since then. The reversal of the trend would have been caused by the development of impounded marshes, especially for waterfowl (Erskine 1992). An estimated 10% of New Brunswick’s freshwater wetlands have been lost, whereas about 65% of its coastal wetlands have been lost (K. Connor, pers. comm.).

Along the Lower St. John River, there were likely few habitat changes to the 8000 ha floodplain wetlands before the Mactaquac hydroelectric dam was built upstream of Fredericton in the 1970s. The effects of this development on the hydrology and nutrient cycling of downstream wetlands are unclear (P. Kehoe, pers. comm.). More recently, conversion has been occurring due to dyking of semi-permanent meadows for waterfowl (G. Forbes, pers. comm.). In fact, almost all sedge meadows in the province could be managed in some form, e.g. impoundment or open water management through either level ditching or blasting potholes. However, this is unlikely to occur in New Brunswick because of the financial constraint and the environmental awareness that is now present through the Department of Natural Resources and the Coastal Habitat program. Until the mid-1980s the New Brunswick government had little interest in wetland conservation, and virtually no regulatory process. In the 1970s and 1980s many wetlands were managed (i.e. altered) through Ducks Unlimited Canada’s program of water stabilization. This program was and still is beneficial to ducks as well as a variety of species (e.g. Black Terns (Chlidonias niger) and Pied-billed Grebes (Podilymbus podiceps)). Until the late 1980s no one saw a need to regulate or change the strategy. Impoundments were attractive to many species, while undeveloped sedge meadow seemed endless and devoid of the wildlife diversity found on managed sites (P. Kehoe, pers. comm.). However, Ducks Unlimited Canada’s projects have converted an estimated 2300 ha of seasonally flooded wetlands to permanently flooded wetlands (K. Connor, pers. comm.).

All this is changing. Ducks Unlimited Canada, along with the provincial and federal governments through the Eastern Habitat Joint Venture partnership, is developing a management strategy for the St. John River floodplain that will conserve habitat for its own value as well as delineate the areas where some management, impoundment and restoration may be desirable for multi-species goals. The first step is the profiling of floodplain communities and their habitat associations. This phase is in its second year (P. Kehoe, pers. comm.).


Rate of Habitat Change

The wintering habitat may be more limiting than breeding habitat for Yellow Rails because of the great pressures along the Gulf Coast and because the wintering range is much smaller than the breeding range. It is difficult to estimate the rate of loss of wintering habitat, because little is known about which habitats are used by Yellow Rails and which ones are not used. The loss of rice fields that is occurring and that will continue because of economic factors is a concern. K. Mizell’s work and other work planned by the U.S. Fish and Wildlife Service on wintering habitats of Yellow Rails and Black Rails (C. Cordes, pers. comm.) should shed some light on the Yellow Rail’s status on the wintering grounds.

Yellow Rail habitat loss was likely fairly rapid earlier this century throughout southern Canada, but it seems to have slowed somewhat. The same situation has occurred in the U.S. breeding range. In Canada, partners of the North American Waterfowl Management Plan play a great role in protecting and re-establishing wildlife habitat, particularly in Prairie Canada, which is probably a stronghold for the Yellow Rail. These agencies are becoming increasingly sensitive to the plight of species other than waterfowl. The recent focus on upland nesting habitat for waterfowl and other birds is a positive step for a number of bird species, including the Yellow Rail. However, habitat loss is occurring particularly rapidly in Alberta at a landscape scale as a result of intense industrial activity.

The Snow Goose problem is also a concern because it is affecting the only pristine stronghold of the Yellow Rail. Bag limits have been increased recently to attempt to curb the problem (S. Wendt, pers. comm.). However, it is not clear that this will solve the problem (C. Gratto-Trevor, pers. comm.).


Protection of Habitats

Available information suggests that a sizable proportion of the Yellow Rail’s global population may breed in the Hudson Bay/James Bay Lowlands, which does not appear to have any major threats despite the Snow Goose problem and the fact that there is little official habitat protection in the region. Special management to impede vegetational succession (e.g. burning, flooding, drying) is not an issue in this region, where the Yellow Rail habitat is kept as such by the natural tides. We do not have detailed information on the proportion of Yellow Rail sites in Prairie Canada that are protected, but some clearly are. In Ontario, the Big Marsh in the Rainy River region is protected by the provincial government (D. Elder, pers. comm.). Richmond Fen is not protected, but is under no immediate threat. In Québec, about half of the sites harbouring Yellow Rails are in protected areas (Robert et al., 1995) (e.g. the Lake Saint-François National Wildlife Area, the Cap Tourmente National Wildlife Area, and Boatswain Bay).

Nevertheless, in Québec as elsewhere, simple habitat protection does not guarantee that Yellow Rail habitat will be maintained because plant succession may gradually transform the area to a site no longer suitable for rails (Jean and Bouchard 1991, Robert et al., 1995, Robert and Laporte 1996). In addition, protected marshes may be subject to wildlife management implemented to favour certain species, notably waterfowl; this management may be to the detriment of Yellow Rail habitat because wetlands of greatest importance to rails are shallower and have greater coverage by emergent vegetation than those typically managed for waterfowl (Eddleman et al., 1988, M. Robert, pers. obs.). However, waterfowl management on wetlands used for breeding by inland rail species (e.g. Yellow Rails) can be compatible with maintenance of rail habitat. For example, appropriate timing of dewatering and shallow flooding, using gradual dewatering, maximizing perennial vegetation that serves as nesting habitat, flooding different impoundments in different years, and avoiding land-levelling, are all methods that can be used to benefit Yellow Rails without compromising waterfowl. Techniques and timing to be implemented vary according to whether the site is used by Yellow Rails in spring migration, fall migration, breeding or wintering (Eddleman et al., 1988).


Degree of Specialization

Yellow Rail habitat selection might be influenced primarily by plant physiognomy and maximum water levels (Robert and Laporte, in prep.) According to Robert and Laporte (in prep.), the species may tolerate considerable variation in certain subtle features of its summer habitats, such as plant species composition, stem density, and height of vegetation, as do other rail species (Rundle and Fredrickson 1981, Flores and Eddleman 1995, del Hoyo et al., 1996). For example, habitats recently described in Québec were located in hydrographic environments that differ from one another: Cacouna was characterized by infrequent high salt-water tidal flooding, Île aux Grues by infrequent very high brackish-water tidal flooding, and Lake Saint-François National Wildlife Area by very poor drainage in a tide-free freshwater environment. Previous studies conducted in Maine (Gibbs et al., 1991) and Michigan (Bookhout and Stenzel 1987) have also suggested that habitat variables often differ from one location to another. Furthermore, although Carex sedges often characterize habitats used by Yellow Rails during summer (Bookhout and Stenzel 1987, Gibbs et al., 1991, Stern et al., 1993, Robert and Laporte, in prep.), many other fine-stemmed plants may also dominate their habitat. Therefore, Carex sedges should not be considered as the only indicator plant species (Stenzel 1982, Gibbs et al., 1991) of Yellow Rail habitats (Robert and Laporte, in prep.).

Although the presence of a senescent vegetation mat is a significant feature of nesting areas, Yellow Rails also inhabit marshes not covered by such a mat at other times and for purposes other than nesting (Robert and Laporte, in prep.). Furthermore, fire and/or mowing may also have an effect on Yellow Rails, by temporarily eliminating habitat and reducing the area of potential nesting habitat by the lack of the senescent vegetation mat. It should be noted, however, that mowing, like burning, can sometimes be beneficial if it is timed correctly, because it prevents the usual vegetative succession and thus may help perpetuate this habitat type (Stenzel 1982). For example, periodic burning is necessary for the maintenance of breeding areas at Seney National Wildlife Refuge in Michigan (R. Urbanek, pers. comm.), where the meadows dominated by the sedge Carex lasiocarpa are a transition stage preceding the establishment of wet conifer swamps (Stenzel 1982). Fire may also have played an important role in maintaining Yellow Rail habitats in the Lake Saint-François National Wildlife Area; alder (Alnus) scrubs have rapidly invaded the wet meadows there since property acquisition by the Canadian government in 1971, when burning was usually suppressed (Jean and Bouchard 1991). It seems that Yellow Rails do not use freshly burned marshes, at least not until the green vegetation has grown sufficiently to offer concealment to the birds. Furthermore, because fire destroys the senescent vegetation mat, burned areas will not be useable for nesting by Yellow Rails until the first or second spring after a fire, depending upon whether the fire occured in the spring or in the fall (Robert and Laporte, in prep.).

Habitat needs of the Yellow Rail may be less restrictive during migration than during the breeding season. During migration, the species is occasionally found in dry meadows and even on croplands (Salt and Salt 1976). As is the case with all rail species, Yellow Rails may be found in atypical habitats during migration, such as in cities, for example. This is likely due to the fact that they migrate at night (see General biology) and land at day break; given that habitat loss for wintering Yellow Rails has been so extensive in the U.S., it is not difficult to imagine that some rails must be forced to land even if it cannot see any suitable habitat.