Plant Community Analysis and Dating of the Asulkan Glacier: A Study of Plant Community Succession Over Time in a Glacial Forefield By: Ciara Sharpe, Laura Grant and Karl Domes
Instructor: Dr. Dan Smith Department of Geography University of Victoria December 2008
Table of Contents Abstract................................................................................................................................3 Introduction..........................................................................................................................4 Study Site.............................................................................................................................6 Methods................................................................................................................................7 Dendrochronology...................................................................................................7 Lichenometry............................................................................................................8 Plant Community Analysis.....................................................................................10 Results................................................................................................................................11 Dendrochronology.................................................................................................11 Lichenometry..........................................................................................................12 Plant Community Analysis.....................................................................................14 Discussion..........................................................................................................................20 Dendrochronology.................................................................................................20 Lichenometry..........................................................................................................21 Developing a timeline of retreat............................................................................23 Plant Community Analysis.....................................................................................24 Conclusion.........................................................................................................................30 Acknowledgements............................................................................................................31 References..........................................................................................................................31
2
Abstract Succession of plant communities after a disturbance is characterized by the progressive replacement of one community by another. The slow rate of retreat of a glacier allows for a unique opportunity to create a representative view of succession as a function of time. In this study, performed at the Asulkan Glacier in British Columbia, plant community succession was assessed using species richness and indicator species. A transect was constructed from the terminal moraine to the glacial tongue along which 5m x 5m plots were studied every 100 m. Dendrochronology and lichenometry were used to generate a timeline of the glacier’s retreat along the transect. Species richness increased from the glacial tongue to the terminal moraine. Presence or absence of nitrogen soil indicators further confirmed that the sites closer to the terminal moraine, exposed for over 100 years, were in early to secondary successional stage. The sites closer to the tongue of the glacier, exposed for less than 100 years, were in pioneer successional stage. The ability to predict plant community succession as a function of time will allow an understanding of ecosystem response to increasingly frequent natural disturbances.
3
Introduction Webster’s dictionary defines succession as ‘the act or process of following in order’ (Merriam and Webster, 2010). In an ecological sense, it is the gradual process of change in an ecosystem following a disturbance. Characterized by the progressive replacement of one community by another, succession continues until a stable end stage or climax is reached. Though succession can be viewed in many ways it is most commonly examined by assessing community composition and species diversity. Mechanisms in determining the sequence of species found after a perturbance fall under three alternative models. All models state that certain species will appear first because they have evolved colonizing characteristics (Connell and Slatyer, 1977). Where the models differ are in the mechanisms that determine how new species appear later in the sequence of succession. The most common model is the facilitation model, which predicts that early successional species modify the environment, in turn determining the later successional species (Whittaker, 1975). This model is applicable to situations where substrate has not been influenced by organisms beforehand, such as the barren ground exposed by receding glaciers. Glacial retreat provides a unique setting for the study of primary succession. The newly exposed substrate lacks plant life; therefore, a glacier’s gradual recession allows for progression of primary succession in the glacial forefield. The advance and retreat of a glacier is dependent on annual mass balance fluctuations; if more mass accumulates in the winter months than is lost through melt during the summer months, the glacier will advance over time. Conversely, if ablation exceeds accumulation, the glacier will recede. Since the Little Ice Age, which ended in the late 19th century, glaciers across the globe
4
have been gradually receding, yielding a marked increase in retreat rate in the past two decades (IPCC, 2007). As the retreat rates of British Columbia’s glaciers increase, the importance of understanding subsequent environmental responses becomes increasingly significant. Dendrochronology and lichenometry are methods in which the historic retreat of glaciers can be outlined. Dendrochronology is the use of tree growth rings to date environmental events and conditions. The rings, which grow radially outwards from the center of the tree trunk, are formed annually. These rings can be visually distinguished by the variation in shade from the lighter spring growth to the darker fall growth. The age of a tree can be determined by simply counting its number of growth rings. Lichenometry was developed in the 1950’s as a means to date the time of exposure of a substrate. Lichen growth rate and age can be extrapolated by measuring the width of the lichen thallus (Luckman, 1977). The most commonly used species to date substrate is Rizocarpon geographicum because it grows slowly, relatively radially, and has a long life span (Luckman, 1977). Using these methods, a timeline for succession can be developed and compared to plant communities found within the glacial forefield. Ecologists often describe habitats by species assemblages or indicator species (Dufrene and Legendre, 1997). An indicator species is any biological species that defines a specific trait or characteristic about the environment. Indicator species and species diversity indices assist us in analyzing succession; habitat change as a function of time. For example, the presence of red alder (Alnus rubra) indicates nitrogen-rich environments, whereas the presence of Alpine Fireweed (Epilobium latifolium) indicates nitrogen-poor soils (Klinka et al.,1989). This idea can be applied to various ecosystems
5
including those found in the glacial forefield. As a glacier recedes over time, ground previously barren to plant life is gradually exposed. Therefore, indicator species in theory should illustrate a time line for glacial recession. Species found closer to the tongue of glacier should be early successional, whereas those found closer to the terminal moraine should be later successional. In conclusion, one would predict that from the tongue of a glacier to the terminal moraine, succession is evident. Though succession of bacterial communities inhabiting the glacial forefields has been thoroughly investigated in the past, plant communities in the forefield are relatively unexplored (Crivii et al., 2002). In this study, plant community composition was assessed along the Asulkan glacial forefield in Glacier National Park, British Columbia, Canada. One would expect that from the tongue of the glacier to the terminal moraine species diversity would increase, as the time the community has had to establish is greater. Furthermore, soil and nutrient establishment should increase as a function of time and thus species found closer to the terminal moraine should indicate richer soils. These measures in conjunction with a timeline created through dendrochronology and lichenometry, should illustrate a clear example of succession. This approach in determining plant community composition as a function of time, along the Asulkan glacial forefield, may provide a tool for future analysis of succession after natural disturbance.
Study site The Asulkan glacier is located in Glacier National Park (117°29’00’’W, 51°41’00’’N). In the Selkirk range of southeastern BC’s Columbia Mountains, this glacier is found between Revelstoke and Rogers Pass. Regionally, the climate is
6
relatively mild compared to the rest of the province, characterized by high annual precipitation levels and heavy snow packs. Lower elevation ecosystems are dominated by cedar-hemlock temperate rainforests, and higher elevations are characterized by mountain hemlock and subalpine fir. The study site was located at the tree line in the Asulkan glacial forefield, and can be accessed from the Asulkan valley trail via the terminal moraine. A 1400 m transect was constructed from the terminal moraine to the tongue of the glacier with 5m x 5m study plots every 100 m. The transect followed a braided river system which ran from the terminal moraine to the 600 m mark through quartzite dominated glacial till. At the 600 m mark the substrate changed abruptly to an exposed quartzite bedrock canyon, which continued for 400 m until reaching an exposed bedrock ridge. This ridge continued for 300 m until the tongue of the glacier and the end of the transect was reached. At from 1000m from the terminal moraine, the proposed study plot was inaccessible and is omitted from our data collection.
Methods Dendrochronology Cores were taken from the largest subalpine fir (Abies lasiocarpa) at every site along the 1400 m transect, from the terminal moraine to the glacial tongue. The largest trees were assumed to be the oldest thus effort was made to core these trees at each site. However, after the first 200 m, tree growth became sparse, and selection was limited for tree cores. In addition, there were no trees present in the vicinity of plots 8 to 13 therefore lichenometry methods were employed in lieu.
7
Trees were cored using a 5 mm increment borer. The cores were taken perpendicular to the stem, and as close to the ground as possible. This was done to ensure maximum presence of growth rings, and to minimize risk of underestimation of tree age. After allowing the cores to dry, they were mounted into grooved boards and sanded until the growth rings were clearly visible. Under a microscope, the rings on each core were counted from the outer bark to the pith and the age of each tree was determined with an error margin of +/- 2 years. Once the ages from all the samples were determined, only data from the oldest tree at each site was kept. In order to date the time of exposure, a dendrochronology ecesis interval was developed for the study area. The ages of three sites were determined through the use of historical photographs: the terminal moraine, and two recessional moraines. Photographs showed the historical margin of the glacier at these sites, dating the terminal moraine to 1889 and the recessional moraines to 1905, and 1918. The differences between the date of exposure, as determined photographically, and the date of sampling (2010) were calculated for each site. The known tree ages were then subtracted from these differences to generate an ecesis interval for the Asulkan Glacier with an error margin of +/- 5 year.
Lichenometry Lichen measurements were taken from Rizocarpon geographicum growing on exposed quartzite boulders and bedrock at 10 of the 14 study plots along the transect. No lichen growth was present at sites 8, 12, and 14, and due to inaccessibility, site 10 was not sampled. Measurements were also taken from the foundation of the Glacier House resort, down-valley from the Asulkan Glacier. The foundation was exposed in 1929 after
8
the resort’s demolition (Rogers Pass National Historic Site). Digital calipers were used to measure lichen size to the nearest tenth of a millimeter. However, due to rainy weather, the calipers short-circuited before completion of the study. A ruler was used in its place, and subsequent measurements were made to the nearest millimeter. The lichen’s largest diameter was measured; a thin black perimeter defining the outermost edges of growth. Measurements were only taken for those lichen that had a circular shape and were not immediately bordered by other lichen. Lichen growth is maximized in favorable conditions, therefore, measurement of the largest lichen ensures similar sampling methods and represents the maximum time of exposure for each site. (Luckman, 1977; Bingham and Pitman, 2008). In order to determine the year of exposure at each site, the lichen diameters were plotted on a growth curve published by McCarthy 2003, as seen in Figure 1. This curve was developed for lichen growing on moraines of the neighboring Illecillewaet glacier. The curve took into account the timed rate of growth of R. geographicum after a determined ecesis interval at various locations on the Illecillewaet glacier. Ideally, a growth curve specifically for the Asulkan glacier would be used. However, the Illecillewaet curve was the only available option and it was assumed that due to its proximity to the Asulkan, it represented a similar environment. Ages of substrate exposure for the various sites were extrapolated from McCarthy’s curve with a suggested error margin of +/- 10 years (McCarthy, 2003) (Figure 1).
9
Figure 1. Growth curve for R. geographicum developed for the Illecillewaet glacier. After age of plots was extrapolated from lichenometry and dendrochronology data, anomalous results were assessed and subsequently omitted from the development of the retreat timeline. The retreat timeline was created using dendrochronology data and historical photographical evidence. Plot 1 (the terminal moraine), was dated using a historical photograph, and plots 3 to 7 were dated using tree ring counts. Dates for plots 2 as well as plots 8 to 14 were extrapolated using Microsoft Excel linear regression.
Plant Community Analysis At each 5m x 5m plot along the transect, plant community composition was assessed. Species present in each plot were identified using Plants of Coastal British Columbia and the Handbook of the Canadian Rockies (Gadd, 1996; Mackinnon and Pojar, 2005). The presence or absence of species was used to determine species richness (number of species) at each site. Species richness values at each plot along the transect were compared in order to assess how species richness changes as a function of time.
10
Percent cover of each species was evaluated and recorded, as well as overall bryophyte and debris cover for each plot. Using a Bray-Curtis Similarity Analysis, plant percent cover data was used to assess similarity of plant community composition between the sites (PRIMER 6). In order to demonstrate these relationships, percent similarities between adjacent plots were plotted on a multidimensional scaling (MDS) plot. In addition, the percent cover data was used to create a cross-correlation matrix (SPSS Statistics). This cross-correlation matrix was constructed in order to determine the relationship between specific plant species found within the plots. Cross-correlation involves assessing the influence of one species on the presence or absence of another species. Species indicating early succession were chosen using Indicator Species of British Columbia and the Plant Indicator Guide for Northern British Columbia (Klinka et al., 1989; Beaudry et al., 2002). Compositions of these indicators among the plots were compared to illustrate species-specific relationships, which were used to extrapolate the successional stage of the landscape. Plant community analysis was combined with dendrochoronology and lichenometry data in order to develop an idea of succession as a function of time. Factors recorded at each site such as debris cover and substrate were considered in this process.
Results Dendrochronology The general trend of the dendrochronology data moving from the terminal moraine towards the glacial nose, shows a decrease in age, as seen in Figure 2.
11
Figure 2. Tree ring counts along the transect, with anomalous data points circled in red This trend demonstrates the linear proportional relationship predicted between substrate age and tree growth. This is the case for all the study plots, except plots one and two, which fail to conform to the trend. The oldest tree cored at plot 1 was 89 (+/- 2) years old, while the oldest tree cored at plot two was 98 (+/- 2) years old. This suggests that tree age decreases with increased distance from the glacier. In addition, an ecesis interval of 50-60 years was developed from the known dates on the transect. Using this ecesis, plot 2 dates to 1857, which is older than the known terminal moraine date of 1889.
Lichenometry The general trend of the lichenometry data, moving from the terminal moraine towards the glacial nose, was a decrease in lichen diameter (Figure 3).
12
Figure 3. Lichen diameters along the transect, with anomalous data points circled in red This trend demonstrates the proportional relationship predicted between substrate age and lichen growth (indicated by size). This is the case for all the study plots except plots 1, 9, and 11. The lichen diameters from plots 9 and 11 are both larger than the lichen diameter from plot 7, suggesting that these plots are older than plot 7, which is unlikely. Secondly, the largest lichen diameter measured at plot 1 was smaller than the largest diameter found at plot 2. Therefore, data points from plots 1, 9, and 11 were deemed anomalous. The retreat dates given from the lichnometry data are older than those suggested by the dendrochronology data (Figure 4). As the distance from the glacial tongue increases, the difference between suggested dates from each method increases in value. A correlation between the data from both methods is apparent.
13
Figure 4. A comparison of dates of retreat determined from both dendrochronology and lichenometry data. The timeline that was developed for the recession of the Asulkan glacier ranges over 81 years (Figure 5). The terminal moraine was dated to 1889, and plot 14 at 1300 m from the terminal moraine was dated to1970. The average rate of retreat is 16 m per year.
Figure 5. Timeline for the retreat of the Asulkan glacier Plant Community Analysis The plant communities changed along the transect from the terminal moraine to the tongue of the glacier, exposed from glacial recession in 1889 and 1958 respectively. Table 1 indicates the presence of plant species found in plots every 100m across the glacial forefield. There is a wide range of plants present in all the plots, however, some
14
species are exclusively present in plots closer to the terminal moraine while others are only found closer to the glacial tongue. In addition, species such as Epilobium latifolium were present in almost all plots across the transect. The plots closer to the terminal moraine are characterized by higher species richness and an increase in the presence of unique species (Table 1, Figure 5). Along the transect, species richness decreases closer to the glacial tongue. At 600m there is a dramatic decrease, this corresponds to the sudden change in substrate; a higher proportion of bedrock and debris cover are present towards to glacial tongue (Figure 5 and 6).
Figure 5. Species richness of vegetation plots along the transect from the terminal moraine to the tongue of the Asulkan glacier.
15
Table 1. Presence of plants found in plots from the terminal moraine to the tongue of the Asulkan glacier. X= plant presence, = unique species, = 1st discovery, = final appearance. 0 Species Luetkea pectinata Unidentified sp. 2 Sorbus sp. Veratrum viride Vaccinium ovalifolium Thuja plicata Lycopodium complanatum Pedicularis bracteosa Unidentified sp. 3 Cryptogramma sterlleri Gaultheria procumbens Senecio Canus Asplenium viride Hieracium sp. Cryptogramma crispa Antennaria neglecta Aquilegia flavescens Erigeron sp. Lycopodium selago Lomatium sp. Adiantum pedatum Botrychium lunaria Unidentified sp. 1 Athyrium felix-femina Alnus crispa Vaccinium vitis-idaea Leptarrhena pyrolifolia Parnassia fimbriata Menziesia ferruginea Tsuga mertensiana Phyllodoce sp. Vaccinium membranaceum Arnica augustifolia Salix sp. Petasites sp. Cassiope blanche Abies lasiocarpa Senecia triangularis Epilobium latifolium Castilleja sp. Picea sp. Polystichum lonchitis Antennaria alpina Arnica amplexicalis Oxyria digyna Woodsia scopulina Saxifraga lyalli Dryopteris expansa Gymnocarpium dryopteris Veronica wormskjoldii Stellaria calycantha Unidentified sp. 4 Species Richness
X X X
Distance from Terminal Moraine 100 200 300 400 500 600 700 800 900 1100 1200 1300 X X X X X X X X
X
X X X X X X X X X X X X X X
X
X X X X X X X X
X X X X X X X X X X X X X
X
X X X X X X X X X X
X X X X X X X X
X X X X
X
X X X X X X
X X X X X X X
X X X
X X
X X X X
X X
X X X
X X
X
X X X X X X X X X X X X X X
X X X X X
X
X X X X X X X X
X X X X X
X
X
X X
X X
X
X
X X
X
X X
X X X X X X
X X X
X X X X X X X X X
X X X X X X X X X X X X X X
X X
X X X
X X
X X X
X
X X 13
24
17
26
23
16
14
9
15
7
15
16
2
16
Figure 6. Percent cover of bryophyte and debris in plots from the terminal moraine to the glacial tongue. Upon comparison, the plant communities exposed earlier (closer to the terminal moraine) were far more similar to each other than those exposed more recently (Figure 7 and 8). Community composition differs between each sample site, although there are greater differences between sites closer to the glacial tongue (Figure 7, R= 0.449, p= 0.023, ANOSIM). Figure 7 illustrates the similarity of community composition for sites located between 0-600m in the glacial forefield. For example, plots located at 0m and 100m were more similar (61%) that those located at 1100m and 1200m (29%); (Figure 8, SIMPER).
17
Figure 7. MDS plot of percent cover site similarity between sites along the transect of the Asulkan glacier from the terminal moraine (0m) to the glacial tongue (1300m).
Figure 8. Percent similarity of adjacent vegetation community composition from the terminal moraine to the tongue of the glacier. Next, Figure 9 and 10 illustrate the cross-correlation of indicator species. There is a relationship in percent cover across the transect between early successional species: Salix sp., Epilobium latifolium and Leptarrhena pyrolifolia (Figure 9). The abundance of one species is correlated to the abundance of another, and all species have a higher percent cover composition in plots closer to the terminal moraine. Furthermore, 18
Epilobium latifolium is present in every plot (with one exception) up to the glacial tongue. In addition, there is a similar correlation between the presence of trees (Tsuga mertensiana) and other plant species such as: Phyllodoce sp. and Vaccinium membranaceum; species requiring soils with higher nutrient concentrations (Figure 10). All species are present with correlated compositions towards the terminal moraine, and rarely appear in plots past 600m.
Figure 9. Percent cover composition of correlated species from the terminal moraine to the glacial tongue of the Asulkan glacier: Salix sp., Epilobium latifolium, Leptarrhena pyrolifolia
Figure 10. Percent cover composition of correlated species from the terminal moraine to the glacial tongue of the Asulkan glacier: Tsuga mertensiana, Phyllodoce sp. and Vaccinium membranaceum.
19
Discussion: Dendrochronology The trend in tree-ring data shows the two anomalous data points (Figure 2). Possible explanations for this discrepancy are either the oldest tree at plot 1 was not cored, or the tree that was cored was secondary growth. The presence of secondary growth trees at plot 1 could be due to two factors. Firstly, the first generation trees may have died from natural causes. Subalpine fir naturally completes its life history at around 100 years of age (Aplet et al., 1988). Alternatively, a disturbance event such as a landslide, rockfall, or fire could have wiped out the first growth of trees at the site. Plot 1 was at the base of a large scree slope with evidence of a rock glacier, which is suggestive that the latter cause was probable. Plot 2, the second anomalous data point, dates older than the terminal moraine and represents a discrepancy in the ecesis interval. One explanation is that the glacier did not scrape this area to bedrock, and an isolated pocket of soil was present after retreat allowing for more rapid plant colonization. Alternatively, microhabitats and environmental variation could have encouraged a faster accumulation of soil at that site, reducing the ecesis interval. Characteristics of retreat and advance of this glacier indicate that this site may have had a different environmental history than further up the transect (Gerlib, 1961). However, further data collection would be required in order to develop a more robust theory. For the purposes of this study, the plot 2 data was simply assumed to show an anomaly in the ecesis interval, and was omitted from the development of the retreat timeline. The ecesis interval that was developed for the Asulkan glacier was a period of 50
20
to 60 years. This interval is comparable to the ecesis developed by McCarthy for the neighbouring Illicellewaet glacier (McCarthy, 2003). Furthermore, previous Geography 477 dendrochronology studies illustrated similar eceses for the Asulkan (Bingham and Pitman, 2008; Anastasiades et al., 2007). Using the ecesis interval of 50 to 60 years, the plots for the first 600 m of the transect were dated from 1989 at the terminal moraine, to 1918 at 700 m. These dates are similar to dates of glacial retreat for the Asulkan determined by Anastasiades et al. (2007).
Lichenometry The trend in the lichenometry data shows three anomalous data points (Figure 3). Firstly, the abnormally large lichen diameters at plots 7 and 9 can be explained by diverse environmental factors that facilitate more rapid establishment and growth for the lichen in these locations. Environmental factors greatly affect the eceses developed through lichenometry. McCarthy found that lichen eceses on the Illecillewaet were site-specific, ranging from 35 to 135 years (2003). Due to proximity, one can assume similar variation in the growth environment at the Asulkan. Additional sources of error include data collection irregularities such as the measurement of two separate lichen thali that were growing together. This error cannot be attributed to miscalibration of the calipers, as a ruler was used for data collection at these plots. Secondly, the largest lichen diameter measured at plot 1 was smaller than the largest diameter found at plot 2. This contradicts the proportional relationship predicted between substrate age and lichen size. In this case, sources of error may have arisen from either data collection or variation of environmental factors. It is possible that the largest
21
lichen at the site was not measured or environmental conditions at plot 1 allowed for a shorter lichen ecesis interval. In addition, a disturbance event like a rockfall could have eliminated the first growth lichen. The latter is likely the case. The site in question was located at the base of a scree slope showing evidence of rockfalls and a moving rock glacier. A rockfall could have led to anomalous data in two ways: data may have been collected from lichen that began growing after the event, or data may have been collected from lichen that was already established on a rock that originated from somewhere else. The trend in lichenometry data for plots 1 and 2 correspond with the dendrochronology data for the same plots (Figure 4). The plot exposure dates that were extrapolated from the lichenometry data range from 1740 at 100 m to 1920 at 1200 m, and show a non-linear glacial retreat timeline. These dates are up to 88 years older than retreat dates given from Bingham and Pitman (2008), and up to 110 years older than the dates extrapolated from our dendrochronology data. Furthermore, the lichenometry dates do not correlate with the historical photographs used in the establishment of a dendrochronology ecesis. For these reasons, the lichenometry dates were omitted from the development of a glacial retreat timeline. The most probable reason for the uncorrelated lichenometry and dendrochronology data was the use of an inappropriate lichen growth curve. The previously mentioned growth curve was developed by McCarthy for the Illecillewaet glacier. It is possible that there were enough environmental differences between these two sites to result in differing lichen growth rates. Moreover, when plotted together, the dates extrapolated from both the lichenometry and dendrochronology data show correlations in trends over distance (Figure 4). This correlation further suggests that the
22
error lies not in the data, but rather the extrapolation of dates from the growth curve. The validity of the lichenometry data is reinforced by the accurate lichenometric dating of the exposure of the Glacier House foundation. The lichenometrically derived date of 1925 (+/- 10 years) correlates positively with the known date of 1927. The most probable environmental condition that would have resulted in a faster growth rate for the lichen at the Asulkan site is the high local moisture content. Moisture has been found to be a limiting factor for lichen growth, with high-moisture environments conducive to faster growth (Armstrong, 1983). The transect in this study followed a braided river environment for the first 600 m, and small bedrock canyons for the next 400 m. Further evidence of these high-moisture conditions is indicated by the presence of hydrophilic species (ferns, mosses) along the transect.
Developing a timeline of retreat Using the dendrochronology data, a timeline was developed for the retreat of the Asulkan glacier from the terminal moraine to the nose of the glacier (Figure 5). The timeline for the glacier’s recession along the transect ranges over 81 years, from 1889 to 1970. The average retreat rate is 16 m per year, which is comparative to rates of retreat of the Illecillewaet glacier compiled by Champoux and Ommanny (1986).
This study demonstrates the advantages and limitations of using dendrochronology and lichenometry as dating tools. The main advantage of dendrochronology is the ability to accurately date the age of a tree from raw data, which helps to minimize error in estimating the substrate age. The potential disadvantages of
23
dendrochronology include not coring the oldest tree at the site, and tree mortality due to environmental factors. These sources of error can be minimized by collecting a larger sample size at each site, and by ensuring that the dates found at a given site correlate with local trends. Alternatively, the main advantage of using lichenometry as a dating tool is the widespread presence of R. geographicum, which allows for the dating of sites in which tree growth is not favorable. However, there are two main disadvantages of lichenometry. Firstly, as with dendrochronology, there is the possibility that the oldest sample at the site was not measured, or that measurements were taken from second growth lichen. Secondly, the age of the lichen cannot be determined from raw data, and a pre-established growth curve must be used for the determination of the substrate age. As a result, there is an intrinsic increase in error in the resulting age approximations. Both dendrochronology and lichenometry are limited by the life spans of the trees and lichens being studied (Aplet et al., 1988; Calkin and Ellis, 1980). Although useful in the determination of absolute dates, they can be used with more confidence as tools for relative dating.
Plant Community Analysis The Asulkan glacier is an alpine environment characterized by plants adapted to harsh growing conditions such as cold weather, a short growing season and low nutrient availability. In such a severe environment, facilitation, plant life history traits and seed dispersal ability are the major mechanisms that cause successional change (Matthews, 1992; Chapin et al., 1994). Colonizing plants supplement the available resources to levels required for subsequent plants to survive and thus facilitate community change
24
(Baumeister and Callaway, 2006; Callaway and Walker, 1997; Chapin et al., 1994). This facilitative effect is derived primarily from the addition of organic matter and the associated nutrients from pioneer plant species (Chapin, 1994; Ohtonen, et al., 1999). In addition, long-distance seed dispersal ability, short generation time and ability to survive harsh environments, allow for plants to colonize disturbed environments (Fastie, 1995). The Asulkan glacier began its current recession from the terminal moraine in 1889 and has continued the retreat to its present-day position. As the glacier recedes, these successional mechanisms ultimately determine the plant community composition of the glacial forefield. Through plant community analysis, this dynamic composition of glacial landscape and alpine ecosystems can be observed over time. Similar studies have demonstrated that communities become more complex over time, characterized by higher biodiversity and an increase in species-specific relationships (Connell and Slatyer, 1977; Chapin et al., 1994; Fastie, 1995). Biodiversity can be described as the variation of life within a particular ecosystem and is often used to evaluate the health of a community (Harrington et al, 2010). Species richness and the number of unique species are commonly used as a biodiversity index, as they contribute to this variation (Harrington et al., 2010). Landscape closer to the terminal moraine of the Asulkan glacier demonstrated increased species richness and a higher number of unique species (Table 1, Figure 5). This increased biodiversity can be attributed to close proximity to the seed source, as well as the increased time elapsed (since glacial recession) for facilitation to occur. This facilitation creates favorable soils by increasing soil biomass and nutrients, thus enabling further species to colonize (Ohtonen, et al., 1999). Non-vegetated soils slowly accumulate biomass as the microbial community
25
changes to effectively manipulate the newly available resources exposed by glacial retreat (Ohtonen, et al., 1999). In addition, the microbial community shifts from bacterial-dominated to fungal-dominated communities, which further increases the uptake of substrate into soil, and helps facilitate plants through mutualistic associations such as with mycorrhizal fungus (Ohtonen, et al., 1999). This improvement of environmental conditions, allows for higher biodiversity a wider range of species are able to survive. The community composition was first analyzed by comparing the presence or absence of plant species across the transect (Table 1). Some species present are widely distributed across the plots, while others are only present closer to the terminal moraine or the tongue of the glacier. This distribution can be observed to see how plant communities change across the landscape. For each plot, the plant community composition is different in respect to the other plots (Figure 6 and 7). In addition, plant communities closer to the terminal moraine were more similar to each other than those after 600m; closer to the glacial tongue. Plant communities close to the terminal moraine had similar bryophyte and debris cover facilitating a similar plant community composition (Figure 5). The similarity of environments creates the opportunity for a more uniform pattern of vegetation across the landscape. In comparison, plant communities closer to the glacial tongue were characterized by increased topographic diversity. At 600m, a change in debris cover with a higher proportion of bedrock existed. This change continued to the glacial tongue. Varying factors such as slope, moisture availability, cliffs and boulders create a diverse array of microhabitats (Kruckeberg 2002). These different habitats ultimately contribute to the dissimilarity of plant communities discovered near
26
the glacial tongue. In the future, progressive replacement of plant communities near the glacial tongue by those characterized by increased similarity in topography and biodiversity may occur. These communities represent a more mature ecosystem and later successional stage. The change in topography at 600m could potentially impact plant succession and community development by decreasing the chances of plant survival, caused by poor soil and nutrient availability (Matthews, 1992). On the other hand, plant pioneer species are shown to increase soil biomass and nutrients and create a more favorable environment over time for successive plants to colonize (Ohtonen, et al., 1999). More research should be performed to determine whether bedrock is shaping plant communities in the Asulkan glacier or vice versa. Both initial site conditions and facilitative effects of plant colonization have been shown to influence the rate of succession (Chapin, 1994). In addition, this study did not evaluate the effect of slope on plant communities. The effect of substrate and slope could be important mechanisms in the succession of these alpine communities. Indicator plants can be used to extrapolate about environmental conditions and successional stages. Salix sp., Epilobium latifolium and Leptarrhena pyrolifolia are early successional indicator species, present in environments characterized by full sunlight and moderate soil nutrients (Klinka et al.,1989; Beaudry at al. 2003). The composition of these plants was positively correlated (Figure 8); the presence of one influencing the presence of the other. In addition, factors such as proximity to an available seed source (surrounding vegetation), length of life histories and dispersal mechanisms contribute to their presence in these disturbed sites. Epilobium latifolium increases in abundance in
27
early pioneer seral stages and thus should be present near the tongue of the glacier. In our study, Epilobium latifolium was found continually from the terminal moraine to the tongue of the glacier, which supports its presence as an indicator species for pioneer seral stages. Although the recently deglaciated landscape appears inhospitable, often characterized by a high pH and low nutrients, these small concentrations of nutrients and available substrate are enough to support such pioneer species (Matthews, 1992). Such pioneer seral stages favor plants such as Epilobium latifolium that have smaller, easily dispersed seeds and faster life histories (Whittaker, 1993; Chapin et al., 1994). In comparison, species such as Salix sp., with a slower lifecycle fail to demonstrate the same response as shown by pioneer species. Provided with a close seed supply and adequate moisture these species can be found in the early successional stages instead of the initial pioneer stages (Whittaker, 1993). Trees require increased soil biomass and nutrients to survive, therefore they are usually found in early to secondary successional stages (Klinka et al.,1989; Beaudry at al. 2003). Furthermore, they are usually found in conjunction with higher plant diversity, as the conditions able to support tree species are usually favorable for a whole host of shrubs and herbaceous plants. In the forefield of the Asulkan glacier, Tsuga mertensiana increased in abundance in plots closer to the terminal moraine (Figure 10). This corresponds to the presence of Phyllodoce sp. and Vaccinium membranaceumi, two high elevation herbaceous plants found in nutrient medium soils. Increased abundance of these herb and tree species increases habitat complexity and provides resources important for animals such as birds, bears and small mammals (Fastie, 1995; Beaudry at al. 2003). Specifically, Phyllodoce sp. does not seem to develop healthy, locally regenerating
28
populations until terrain is in excess of 50 years of age (Whittaker, 1993). The correlation of these plants indicate that closer to the terminal moraine, the glacial forefield is in early to secondary successional stages. The increased biodiversity found within these plots is indicative of a more advanced successional stage. When compared to the timeline of the glacier, it can be said that such changes arose approximately over 100 years after glacial recession. In comparison, the landscape exposed for less than 100 years, closer to the glacial tongue, remains in the pioneer seral stage. After 600m, biodiversity decreases, plants requiring more soil nutrients disappear and pioneer species such as Epilobium latifolium persist. In Glacier Bay, Alaska, Chapin et al. (1994), assessed succession in glacial forelands. This study found similar results in regards to succesional change over time. Study sites showed progress from pioneer seral stages to more advanced communities from 50 to 100 years after glacial recession. After 100 years, early successional species (Alnus crispa) began to be overtaken by mature successional species (Picea sp.) The plant communities present across the forefield of the Asulkan glacier are part of a dynamic succession and will eventually change. Pioneer plants such as Epilobium latifolium and Alnus crispa will eventually be shaded out by later successional plants such as Picea sp. The time line created with dendrochronology and lichenometry, can be used to approximate plant community composition and successional seral stage following disturbance. When combined with these dating techniques, the rates of plant succession could be measured and applied to changing ecosystems.
29
Conclusion Succession is the process in which a community of organisms changes over time after a disturbance such as the retreat of a glacier. Species richness and indicator species were used to assess the dynamic succession of plant communities along the forefield of the Asulkan glacier. Progression of succession is accompanied by increasing biodiversity. It was found that communities close to the terminal moraine had higher biodiversity than those close to the glacial tongue, indicating later successional stages. Furthermore, the presence of nitrogen-rich soil indicators, such as Tsuga mertensiana, in communities closer to the terminal moraine were indicative of later successional stages. As seen in communities closer to the glacial tongue, presence of nitrogen-poor soil indicators, such as Epilobium angustifolium, are indicative of earlier seral stages. It was found that those plots from the terminal moraine to 600 m had been exposed for over 100 years and were in early to mature successional stage. Those plots from 600 m to the glacial tongue were found to have been exposed for less than 100 years and were in pioneer successional stage. The retreat of the Asulkan glacier began in 1889 and progressed to its current position at an average rate of 16m per year. Severe environments, such as the forefield of the Asulkan glacier, employ many mechanisms that contribute to the succession of plant communities over time. Diverse topography, seed-source proximity, and soil facilitation all influences ecosystem change. In a world that is destined to continually undergo natural disturbances, such as glacial recession, understanding succession enables us to predict the future of dynamic ecosystems.
30
Acknowledgements We would like to extend a big thank you to Miss Kara Pitman for all her help and support throughout this course. Thank you to Jim Gardner for sharing your remarkable stories about the mountains with us while braving the wind and rain in the field, we couldn’t have kept such a positive attitude without you. Dr. Dan Smith, many thanks for such a wonderful and inspiring course, the mountains of Glacier National Park truly captured our imaginations. We also send our thanks to our wonderful friends who made Geog 477 such a fun experience, we hope to meet you in the mountains again someday soon!
References Anastasiades et al. (2007). Mapping the Retreat of the Asulkan Glacier in Glacier National Park, British Columbia, Canada. Geography 477 Field School, University of Victoria. Aplet, G. H., Laven, R.D., and Smith, F. W. (1988). Patterns of Community Dynamics in Colorado Englemann Spruce-Subalpine Fir Forests. Ecology, 69: 312-319. Armstrong, A. (1983). Growth Curve of the Lichen Rhizocarpon Geographicum. New Phytology, 94: 612- 622. Baumeister, D. and Callaway, R.M. 2006. Faciliatation by Pinus Flexilis during succession: A hierarchy of mechanisms benefits other plant species. Ecology. 87 (7): 1816-1830. Beaudry, L., Coupe, R., DeLong, C., and Pojar, J. 2002. Plant indicator guide for Northern British Columbia: the Northern portion of the MS and ICH biogeoclimatic zones. Res. Br., B.C. Min. For., Victoria, B.C. Tech. Rep. Bingham, C. and Pitman, K. (2008). Dating Morraines of the Asulkan Glacier Through Lichenometry and Dendrochronology. Geography 477 Field School, University of Victoria. Calkin, R. E., and Ellis, J. M. (1980). A Lichenometric Dating Curve and Its Application to Holocene Glacier Studies in the Central Brooks Range, Alaska. Arctic and Alpine Research, 12 (3): 245-264. Champoux, A., and Ommanney, C. S. I. (1986). Evolution of the Illecillewaet Glacier, Glacier National Park, B.C., Using Historical Data, Aerial Photography and Satellite Image Analysis. Annals of Glaciology, 8: 31-32.
31
Chapin, F.S., Walker, L.R., Fastie, C.L. and Lewis, C. 1994. Mechanisms of primary succession following deglaciation at Glacier Bay, Alaska. Eco. Monog. 64(2): 149-175. Connell, J.H., and Slatyer, R.O. 1977.Mechanisms of succession in natural communities and their role in community stability and organization. Am. Nat. 111(982): 11191144. Crivii, S., Zeyer, J., and Sigler, W.V. 2002. Bacterial succession in glacial forefield soils characterized by community structure, activity and opportunitistic growth dynamics. Microb. Ecol. 44(4): 306-316. Dufrene, M., and Legendre, P. 1997. Species assemblages and indicator species: the need for a flexible asymmetrical approach. Eco. Monog. 67(3): 345. Fastie, C.L. 1995. Causes and Ecosystem Consequences of Multiple Pathways of Primary Succession at Glacier Bay, Alaska. Ecology. 76 (6): 1899-1916 Gerlib, R. K. (1961). Recent Glacier Fluctuations in the Selkirk Range, B.C.. The Canadian Alpine Journal: 87-91 Harrington, R., Aton, C., Dawson, T.P., de Bello, F., Feld, K.C., Haslett, J.R., Kluvankova-Oravska, T., Kontogianni, A., Lavorel, S., Luck, G.W., Rounsevell, D.A., Samways, M.J., Settele, J., Skourtos, M., Spangenberg, J.H., Vandewalle, M., Zobel, M. and Harrison, P.A. 2010. Ecosystem services and biodiversity conservation: concepts and a glossary. Biodivers. Conserv. 19(19): 2773-2790. IPCC Fourth Assessment Report: Climate Change 2007 (2007). Intergovernmental Panel on Climate Change. < http://www.ipcc.ch/publications_and_data/ar4/wg1 /en/contents .html>. Accessed December 6, 2010. Klinka, V.J., Karjina, A., Ceska, A., and Scagel, A.M. 1989. Indicator plants of coastal British Columbia. Government of Canada, Province of British Columbia. Kruckeberg, A. R. 2002. Geology and plant life: the effects of landforms and rock on plants, 1st ed. University of Washington Press, Seattle. Luckman, B. H. (1993). Glacier Fluctuations and Tree-Ring Records for the Last Millenium in the Canadian Rockies. Quaternary Science Reviews, 12: 441-450. Luckman, B. H. (1977). Lichonomentric dating of Holocene moraines at Mount Edith Cavell, Jasper, Alberta. Canadian Journal of Earth Science, 14: 1809-1822. McCarthy, D.P and Luckman, B.H. (1993). Estimating Ecesis for Tree-Ring Dating of Moraines: A Comparative Study from the Canadian Cordillera. Arctic and Alpine Research, 25: 63-68
32
Matthews, J. A. 1992. The ecology of recently-deglaciated terrain, a geoecological approach to glacier forelands and primary succession. Cambridge University Press, Cam- bridge, England McCarthy, D.P. (2003). Estimating Lichenometric Ages by Direct and Indirect Measurement of Radial Growth: A Case Study of Rizocarpon agg. at the Illecillewaet Glacier, British Columbia. Arctic, Antarctic, and Alpine Research, 35(2): 203-213 Merriam, C., and Webster, N. 2010. Merriam-Webster Online Dictionary.http://www. merriam-webster.com/dictionary/succession (accessed Dec. 17th 2010) Ohtonen, R., Fritze, H., Pennanen, T., Jumpponen, A. and Trappe, J. 1999. Ecosystem Properties and Microbial Community Changes in Primary Succession on a Glacier Forefront. Oecologia. 119 (2) : 239-246 Ragan, M.C. and Walker, L.R. 1997. Competition and Facilitation: A Synthetic Approach to Interactions in Plant Communities. Ecology. 78 (7): 1958-1965. Rogers Pass National Historic Site Glacier House information sign. Rogers Pass National Historic Site, Parks Canada, Rogers Pass, BC. Whittaker, R.H. 1953. A consideration of climax theory: The climax as a population and pattern. Ecol. Monog. 23:41-78. Whittaker, R.J. 1993. Plant Population Patterns in a Glacier Foreland Succession: Pioneer Herbs and Later-Colonizing Shrubs. Ecography. 16 (2) : 117-136
33
