|BOTANICAL ELECTRONIC NEWS|
|No. 418 December 3, firstname.lastname@example.org||Victoria, B.C.|
This is a note on what appears to be the first observation of the invasive European lineage of Common Reed (Phragmites australis (Cav.) Trin. ex Steud subsp. australis) in the Canadian Prairie Provinces. To place this observation in context, I commence with a very brief summary of the literature on the known distribution of invasive Common Reed in Canada. I then provide the details of recent observations made in Winnipeg, Manitoba. In conclusion, I offer a few remarks on the potential impacts of invasive Common Reed in the Canadian Prairies.
Less than a decade ago, it came to be recognized that two subspecies of the arundineceous grass, Phragmites australis (Cav.) Trin. ex Steud., occur in Canada. One of these, the endemic North American Common Reed, has very recently been classified as Phragmites australis (Cav.) Trin. ex Steud. subsp. americanus Saltonstall, P.M. Peterson and Soreng (Saltonstall et al. 2004). The other, which has been shown to be of Eurasian origin, has been referred to Phragmites australis (Cav.) Trin. ex Steud. subsp. australis (Catling 2006, 2007a). It is often called European Common Reed or invasive Common Reed. Several field characters have been identified by means of which these two subspecies can be differentiated (Catling et al. 2007; cf. Catling et al. 2003, Catling 2007a, Robichaud and Catling 2003).
Knowledge of the respective distributions of the two subspecies in Canada has progressed at almost the same pace as the understanding of how to differentiate them. The native Common Reed has been found to occur in all Canadian provinces and territories except the Yukon, Nunavut and Newfoundland and Labrador (CBIF 2007). The invasive Common Reed was initially recognized in southern Ontario and Quebec. It was then confirmed to be in the British Columbia interior, a small number of stands having established within the Okanagan Valley (Martin 2003, Schueler et al. 2003). Finally, its distribution in the Maritime Provinces was described, where it occurs at Stephenville in Newfoundland, Beaver Dam and Sackville in New Brunswick, and several localities in Nova Scotia (Catling et al. 2004). The best representation of its distribution in these provinces is probably that given by the Google Earth map which can be generated from the CBIF (2007) Phragmites of Canada database. (See in this issue: Catling and Mitrow, Canadian Phragmites Database – Update Notes for Use.)
What is the situation in the Canadian Prairie Provinces? The endemic North American Common Reed is widespread in the eastern prairie and boreal zones of Manitoba, Saskatchewan and Alberta (CBIF 2007). Until recently, the invasive Common Reed has been unknown from these areas. (See in this issue: Catling and Mitrow, Where is the Invasive Alien Phragmites Going in Canada.) However, on 15 October 2009, I observed a stand of over a hundred plants growing in the city of Winnipeg, Manitoba.
As is often the case in other provinces, the stand was found in a roadside ditch within an urbanized commercial area. It was discovered in the Fort Garry section of the city, on the western side of Route 90, north of Route 155: 49.8263° N. 97.2066° W. UTM coordinates are 14U 0628984 5520862 (WGS 84).
Most of the plants had inflorescences and approximately twenty five had reached anthesis. Unlike native plants, observed the same day between Winnipeg and the Ontario border, these plants were not yet senescing.
I was able to document this occurrence with both photographs and specimens. The former have been accessioned in the National Vascular Plant Herbarium (DAO) and the latter will be in the near future.
A second stand of invasive Common Reed is suspected on the south side of Highway 17, approximately ten minutes drive east of Winnipeg where road construction was taking place on the date of observation. Traffic conditions did not permit a closer investigation of this occurrence.
The introduced Common Reed has been recognized as a top priority invasive alien plant of natural habitats in Canada (Catling 2005). It is known to displace native wetland vegetation in the Great Lakes and Saint Lawrence River regions of this country (Gilbert et al. 2009, Lavoie et al. 2003, Wilcox et al. 2003). Similar impacts may be expected in prairie wetlands, including potholes, deltas and fens. This in turn might have significant impacts on the waterfowl species that depend on these habitats.
To prevent these outcomes, reconnaissance should be undertaken along the roadside ditches of major highways and high traffic routes in urban areas of southern Manitoba to assess the extent of the invasion. It may still be possible to eliminate invasive Common Reed stands as they develop. Without a well-organized control program, the grass became a serious problem in eastern North America within a decade.
Dr. Paul Catling very kindly provided remarks on the first draft of this article and assisted in arrangements for its publication in BEN - Botanical Electronic News.
We frequently receive requests for information on Phragmites in Canada. Useful major published compilations have been produced by Haslam (1958) and Mal and Narine (2004), but this was in advance of the realization that the rapid invasion of Phragmites was a result of a foreign genotype rather than simply the spread of the native genotype. As Phragmites australis subsp. australis has spread through eastern North America, it has attracted much attention and the number of articles and reports that mention it has increased sharply. Consequently much recent work that compares the two genotypes or concerns the invasive genotype (subsp. australis) is not included in the synthesis of Mal and Narine (2004). A few other websites listed below are very informative, but none of these has all the content of a list that we recently developed to assist those with questions. Most of the questions we receive relate to identification, ecology, distribution and control and our list covers primarily those areas. Since the older literature is available elsewhere (e.g. Mal and Narine 2004), our list begins in 2000 with the exception of a few valuable older reviews. This list is based on: (1) a library search of Biosis (biological abstracts), CAB abstracts and Agricola (agricultural on-line access); (2) documents found through a Google search of websites; and (3) reports and articles that have been sent to us. It has been developed especially for Canada but includes research based in other parts of the world that can be applied to Canada. To access the list, available here as an appendix to this issue of BEN: http://bomi.ou.edu/ben/418/phragmites_references.pdf This list will continue to grow and we expect to release updates in the future. Therefore we will be pleased to be informed of additions. Since much useful information is available on the web, we have listed several valuable websites below along with the two major reviews.
List of recent references for Phragmites in Canada http://bomi.ou.edu/ben/418/phragmites_references.pdf
Phragmites australis has been extensively studied by European and Asian scientists since the time of the International Biological Programme (IBP) in the 1960s and 1970s. The management and use of Phragmites-dominated wetlands is summarized by Gopal and Mazing (1990) and Haslam et al. (1998). Main production-ecological features are given by Květ et al. (1998). Both earlier and IBP studies brought the first information on ecotypic variation of Phragmites australis (Björk 1967, van der Toorn 1972).
The recent history of Phragmites australis subsp. australis in Europe encompasses examples of both decline and spread. First indications of decline were described by Klötzli (1971, 1974). A thorough description of its symptoms was given by Ostendorp (1989), and the knowledge acquired until the late 1980s is presented in Den Hartog (1989). Both national and international projects were carried out in the 1990s with the aim to elucidate mechanisms of the decline (Van der Putten 1997, Brix 1999). Although the reed decline had always site-specific features, some general mechanisms seem to emerge. First, declining populations always occurred on sites with stabilized water levels and fairly deep water columns (at least 0.5 m), which prevented generative reproduction (Rea 1996). The vegetatively maintained reed stands coped with this situation usually fairly well until an additional stress event occurred (e.g., strong mechanical damage or a flood event associated with a further marked increase of the water column). Eutrophication of wetland habitats probably predisposed the reed plants to decline because of a strengthened oxygen deficiency in the sediment (antrůčková et al. 2001), followed by damage to root and rhizome tips by toxic products of microbial metabolism (Armstrong et al. 1996).
The genetic research of declining reed populations in German lakes pointed to their low genetic diversity (Koppitz et al. 1997). It was suggested that in originally diverse populations, only such clones would persist that would be able to survive under the particular site conditions; and this would be reflected by a reduced genetic variability within the given population. The resulting populations, consisting of only a few large monoclonal stands, were supposed to be more susceptible to rapid changes in environmental conditions such as eutrophication, because the low genetic diversity would not provide the stands with a sufficient phenotypic plasticity needed for successful adaptation. Indirect support to the above hypothesis was provided by a study of genetic diversity of populations of different age, which showed that the genetic diversity decreased with the population age (Čurn et al. 2007).
Reed expansion has been observed at some other European sites (Güsewell et al. 2000, Čížková, unpublished results, and personal communications of nature conservation managers of sedge meadows along the Labe (Elbe) River, Czech Republic). These sites provide waterlogged or shallowly flooded habitats that were originally oligotrophic, but have recently been exposed to a higher nutrient supply. Under such conditions, the competitive ability of the common reed seems to be supported at the expense of species of the original sedge-dominated vegetation.
The GBIF website (http://www.gbif.org/) including the "Phragmites of Canada" databasae allows some niche modelling using climate data with the niche modeling library "openModeler" (http://openmodeller.sourceforge.net/). To determine the potential distribution, i.e. to predict spread, of the alien invasive Phragmites australis subsp. australis in Canada, we produced a map (approximation based on models) of the presence of climate similar to the climate already occupied by Phragmites australis subsp. australis in eastern Canada. Such maps are easily produced using instructions provided elsewhere in this issue. The procedure used a lax bioclimatic envelope algorithm called "envelope score": http://openmodeller.sourceforge.net/index.php?option=com_content&task=view&id=61&Itemid=4
Basically we assume that the plant could occupy all parts of Canada with climate equivalent to places where it already occurs. This area is the "fundamental niche". For many native species, dispersal capability, barriers, competition, pathogens, and other factors (e.g. MacArthur 1972) spatially reduce the fundamental niche to a "realized niche". For alien invasives on the other hand, dispersal is usually effective, competition is less severe and natural controls are fewer (at least for a period of time) so that the fundamental niche may correspond to the potential realized niche. Thus the assumption of potential range of an invasive plant being determined by climate and soil is not unreasonable. In the present case of Phragmites australis subsp. australis, edaphic factors appear to be less important to spread than usual because it has a broad tolerance of soil moisture ranging from continuously inundated to periodically very dry. Although it may prefer alkaline substrates, it can spread in extensive regions of acid soil along roadside ditches that are alkaline due to the use of de-icing salt. In fact there is good evidence that the major expansion of Phragmites australis subsp. australis in eastern Canada has been by fragmentation along roads (Catling & Carbyn 2006, Lelong et al. 2005, 2007). It does well in calcareous (CaCO3) and saline (NaCl) subtrates ranging from pure organic material to pure clay, beach sand, or rock and various combinations. Consequently we assume that climate will be much more important than soils in determining the potential distribution and within climate, its broad moisture tolerance suggests that temperature will be the ultimate factor. The lack of importance of precipitation is reinforced by the abundance of water throughout most of southern Canada and the fact that much of the spread of the plant is by rhizome fragments with roots and buds rather than seeds (Catling & Carbyn 2006) which may be more prone to the effects of periodic drought.
OpenModeller reads the corresponding environmental values for each occurrence point. The data points then become samples representing the environmental conditions at each location. An algorithm is then used to find a representation of the species niche in the environmental space. In the present case, for each environmental variable the envelope score algorithm finds the minimum and maximum at all occurrence sites. During model projection, the probability of occurrences is determined as: p = layers within min-max threshold / number of layers. We used 12 temperature layers (because Phragmites australis subsp. australis is capable of dealing with much variation in water availability in eastern Canada as noted above). The 20 climate layers included annual mean temperature, mean diurnal range, isothermality, temperature seasonality, maximum temperature of warmest month, minimum temperature of coldest month, temperature annual range, mean temperature of wettest quarter, mean temperature of driest quarter, mean temperature of warmest quarter, mean temperature of coldest quarter, annual precipitation, precipitation of wettest month, precipitation of driest month, precipitation seasonality, precipitation of wettest quarter, precipitation of driest quarter, precipitation of warmest quarter, precipitation of coldest quarter. This square km climate grid data (Hijmans et al. 2005) is provided by WorldClim (http://www.worldclim.org/).
Currently Phragmites australis subsp. australis is frequent in the southern parts of Ontario and Quebec with isolated occurrences in BC (Martin 2003, Schueler et al. 2003), the maritime provinces (Catling et al. 2004), and northwestern Ontario (http://data.gbif.org/species/16140669). The potential distribution map using only temperature suggested a much more extensive distribution in Canada including a large area of the prairie ecozone in the southern prairie provinces [Note: no sooner is a prediction made than it comes true – see article by E. Snyder in this issue, Ed.], and much of lower elevation of the montane cordillera ecozone in central BC as well as western parts of the Boreal Plains ecozone (for Canadian ecozones see http://ontario.on.ec.gc.ca?wildlife/wildspace/wsimages/ws-map-caneco.gif ). This map also suggests extensive colonization of parts of the Canadian Shield ecozone and the Atlantic Maritime Ecozone, although impact on natural habitat and agriculture in this region may be less due to acidic substrates. These results provide a picture of potential distribution that is supported by other models (in preparation) including other environmental measures such as growing degree days.
The potential distribution of Phragmites australis subsp. australis in Canada is very alarming because the potential environmental damage is very substantial. Prairie wetlands for example, are host to a large native biodiversity that includes waterfowl of great economic importance. Wild rice may be extensively displaced. Substantial costs to agriculture may occur as a result of invasion of irrigation systems in the western prairie region. Sport fishing may be impacted by general decline in wetland biodiversity in parts of southern and northwestern Ontario. Invasion of maritime salt marsh in eastern Canada will also likely be accompanied by substantial biodiversity loss. Prediction allows some hope for containment since it allows preparation. Control can protect biodiversity until either natural enemies catch up and/or until biocontrol agents become effective.
Send submissions to email@example.com
BEN is archived at http://www.ou.edu/cas/botany-micro/ben/