SEARCH

SEARCH BY CITATION

Keywords:

  • climate change;
  • experimental warming;
  • International Tundra Experiment;
  • open-top chamber;
  • passive warming;
  • resistance;
  • tundra;
  • vegetation change

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

1. Identifying plant communities that are resistant to climate change will be critical for developing accurate, wide-scale vegetation change predictions. Most northern plant communities, especially tundra, have shown strong responses to experimental and observed warming.

2. Experimental warming is a key tool for understanding vegetation responses to climate change. We used open-top chambers to passively warm an evergreen-shrub heath by 1.0–1.3 °C for 15 years at Alexandra Fiord, Nunavut, Canada (79 °N). In 1996, 2000 and 2007, we measured height, plant composition and abundance with a point-intercept method.

3. Experimental warming did not strongly affect vascular plant cover, canopy height or species diversity, but it did increase bryophyte cover by 6.3% and decrease lichen cover by 3.5%. Temporal changes in plant cover were more frequent and of greater magnitude than changes due to experimental warming.

4.Synthesis. This evergreen-shrub heath continues to exhibit community-level resistance to long-term experimental warming, in contrast to most Arctic plant communities. Our findings support the view that only substantial climatic changes will alter unproductive ecosystems.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

General circulation models predict that during the 21st century, anthropogenic climate forcing will lead to significant warming and induce changes to the global climate system that may cause rapid and sometimes drastic changes to vegetation (IPCC 2007). Although many ecosystems will be sensitive to these impacts (Callaghan et al. 2005), others may not be. For example, some plant communities have exhibited strong resistance to simulated climate change manipulations (e.g. Grime et al. 2008), where resistance is defined as the ability of a community to maintain its composition and structure in the face of environmental change. Identifying which vegetation types are most sensitive to climate change will be critical for predicting large-scale ecosystem responses in the coming decades.

Passive warming experiments are an important tool for understanding the effects of climate change on low-productivity ecosystems, especially tundra (Wookey 2008). Standardized, plot-level experiments using open-top chambers (OTCs) are conducted in the International Tundra Experiment (ITEX, Henry & Molau 1997), a network of more than 20 Arctic and alpine tundra sites. The ITEX network has completed several multi-site syntheses to examine the effects of experimental warming on plant traits (Henry & Molau 1997; Arft et al. 1999), community composition and abundance (Walker et al. 2006) and carbon flux (Oberbauer et al. 2007). Using a meta-analysis of 17 experiments, Walker et al. (2006) found that passive warming alters species composition and abundance, increases canopy height and the cover of deciduous shrubs and graminoids, and decreases cryptogam cover and species diversity. Site-specific cover responses to warming have also been detected in dry, moist and wet tundra ecosystems (e.g. Chapin et al. 1995; Graglia et al. 2001; Hollister, Webber & Tweedie 2005; Jónsdóttir et al. 2005; Wahren, Walker & Bret-Harte 2005). Despite the importance of local factors, we expect that the altered thermal, nutrient and competitive regimes created by warming will control the magnitude of large-scale vegetation change in the Arctic (Shaver et al. 2000; Hollister, Webber & Tweedie 2005; Walker et al. 2006).

Our understanding of how tundra plant communities will be affected by climate change currently remains limited because of several factors, including too few long-term studies. The durations of field studies are typically determined by standard research funding cycles, but meaningful responses often take much longer to appear because tundra ecosystems respond slowly to perturbation. Most studies are discontinued after <5 years (Walker et al. 2006), even though changes in community composition may take more than a decade to become detectable (Epstein et al. 2004; Rinnan et al. 2007). Moreover, short-term experiments are not generally good predictors for longer-term changes (Chapin et al. 1995; Hollister, Webber & Tweedie 2005).

Climate-change experiments allow us to simulate future environmental conditions, examine subsequent vegetation responses, and determine how these responses will affect ecosystem properties and other trophic levels (see Callaghan et al. 2005). To accurately forecast vegetation responses to climate change, numerous studies across a range of plant communities and regions will be required, due to the spatially heterogeneous nature of both tundra and climate change. Here, we describe results from the longest-running passive warming experiment in the Canadian Arctic. In this article, we test the hypothesis that temperature affects community-level variables (plant cover, species diversity and canopy height) as predicted by Walker et al. (2006), in a High Arctic evergreen-shrub heath subjected to 15 years of passive warming.

Materials and methods

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Study site

The study was conducted on the 8-km2 coastal plain (sandur) at Alexandra Fiord (78°53′ N, 75°55′ W), Ellesmere Island, Nunavut, Canada, that is described in detail in Svoboda & Freedman (1994). The coastal lowland is bound by bare upland plateaus, outlet glaciers and the waters of Alexandra Fiord. It receives an influx of meltwater from snow and thawing soils from the surrounding uplands. The site typically receives 0–60 mm of rain from June to August (G.H.R. Henry & J.M.G. Hudson, unpublished 2000–07 data).

The study community is an evergreen-shrub heath dominated by several shrub species [Cassiope tetragona L. D. Don., Dryas integrifolia Vahl. and Salix arctica Pall.; vascular plant flora = Porsild & Cody (1980)] and bryophytes [Aulacomnium turgidum Schwaegr., Dicranum spp. Hedw. and Tomenthypnum nitens Hedw.; bryophyte flora = Ireland et al. (1980)]. Vascular plants cover between 65% and 75% of the ground surface (Hudson & Henry 2009). Hummocks and hollows are well-vegetated and rocks of varying sizes are common. The organic layer is 0–6 cm deep and soils are poorly developed Static Cryosols (Muc, Freedman & Svoboda 1989). The area is underlain by permafrost and maximum seasonal thaw is 50–70 cm (Hudson & Henry 2009). Snowmelt occurs in early to mid-June and the growing season, defined as the consecutive number of days with mean air temperatures >0 °C, is 60–90 days long (Hudson & Henry 2009). Soils remain moist throughout the snow-free season due to their low rates of evapotranspiration, even in years with no precipitation.

Experimental design

In 1992, 28 permanent 1 m2 plots were randomly located based on the presence of several target species for long-term monitoring of phenology, growth and reproduction. In 1995, eight more plots were added. Half of these 36 plots were experimentally warmed, while the other half were left as control plots. OTCs passively warmed the treatment plots for 12–15 years. The OTCs are small, inexpensive, easy to maintain, do not require electrical power and are left in place year-round. They are inclined hexagonal cones, 1.5 m diameter at the top and 0.5 m tall. To minimize edge effects, the chambers are slightly larger than the plots. The OTCs are constructed of 1.0 mm thick fibreglass (Sun-Lite HP Solar Components Corp., Manchester, NH, USA; Marion et al. 1997). The material has high solar transmittance in the visible wavelengths (>85%) and low transmittance in the infrared wavelengths (<5%). The chambers allow transmittance of solar radiation, decrease wind speeds and trap convective heat (Marion et al. 1997). Experimental warming with OTCs typically increases mean daily temperatures during the snow-free season by 1–2 °C (Marion et al. 1997). In 1993, experimental warming increased mean daily temperatures by 1.3 °C in this plant community (Marion et al. 1997). This increase is conservative relative to forecasted warming scenarios for the Arctic (Anisimov et al. 2007), but we elected to keep a lower degree of warming in order to minimize experimental artefacts. Unwanted side effects of OTCs are discussed elsewhere in great detail, and include increased temperature extremes; altered light intensities and moisture and gas concentrations; and interference with animals (Kennedy 1995a,b; Marion et al. 1997; Shen & Harte 2000). The OTC approach was adopted and standardized by the ITEX network (Henry & Molau 1997; Marion et al. 1997), has been validated at the plot level (Hollister & Webber 2000) and remains the preferred method of passive warming for Arctic and alpine ecosystems (Wookey 2008).

Environmental monitoring

Temperature was measured 10 cm above the ground using HOBO thermistors (= 6, Onset Computer Corp., Pocasset, MA, USA) and copper–constantan thermocouples attached to a data logging system (= 18). Each spring, we visually inspected all plots daily for snow-free dates (>90% of ground surface visible). Soil water content was measured in 12–36 plots, depending on the year. In 1995, 2002 and 2007, soil water content was measured (within the treatment plots but outside of the area where vegetation is monitored) three times during the growing season either gravimetrically (in 1995) with soil plugs (−5 cm, 2 cm diameter) or with a Hydrosense probe (−12 cm, Campbell Scientific Canada Corp., Edmonton, Alberta, Canada). At the end of the 2007 field season (early August), we measured maximum depth of thaw in the centre of all plots. During summer 2008, we measured bioavailable nitrogen in the soils of all plots from 19 June to 16 August with ion exchange membranes (PRSTM-Probes, Western Ag Innovations Inc., Saskatoon, Saskatchewan, Canada).

Vegetation measurements

In 1996, 2000 and 2007, we measured community composition and abundance using the standard ITEX point-intercept method (Molau & Mølgaard 1996). At 100 grid points in the 1-m2 point frame, the identity (either species or genus name) and state (alive or dead) of all layers were recorded. Only point-intercept pin hits that touched living plants were included in the analyses. From the raw point-intercept data, we calculated four species diversity metrics: the Shannon–Wiener index, Simpson’s index, species richness and Pielou’s evenness. We divided the species into six plant functional types: bryophytes, lichens, deciduous shrubs, evergreen shrubs, forbs and graminoids to determine changes in cover. In 2000 and 2007, canopy height was also measured at a subset of 25 standardized grid points per plot. Because this is a long-term experiment, all plant sampling was non-destructive.

Statistical analyses

We used the plant cover data to test for changes in community composition and abundance. First, a permutational multivariate analysis of variance (permanova, Anderson 2001) was conducted using PRIMER (Clarke & Gorley 2006). Next, we calculated Bray–Curtis dissimilarity measures using the ‘vegdist’ procedure in the vegan package (1.13–2, Oksanen et al. 2008) of R version 2.5.1 (R Development Core Team 2007). A dissimilarity score for each plot in each year was calculated based on the plot’s Bray–Curtis distance from the average plant cover of the control plots in 1996. To test these dissimilarities, we used a generalized least-squares (GLS) modelling function in the nlme package (3.1–86, Pinheiro et al. 2007). The model extends the basic linear mixed-effects model to include correlated within-group errors, which are present in repeated-measures experiments (Pinheiro & Bates 2000). In this study, we ran estimated generalized least-squares (EGLS) models rather than GLS models because GLS assumes that the error-covariance matrix is known, whereas we estimated the correlations from the sample data. Model predictor variables were Warming, Year and Warming × Year. A Continuous AutoRegressive order of 1 [CAR(1)] correlation structure for years within plots was used, because the re-measurement period was irregular (p. 229, Pinheiro & Bates 2000). We also tested for changes in the cover of dominant species, plant functional types, species diversity, canopy height, soil water content and spring snow-free dates using the same EGLS model structure. Variables that did not meet the assumptions of normality and homogeneity of variance were transformed (graminoids and Oxyria digyna were square-root transformed and D. integrifolia was log transformed).

The effect of duration of the experiment was also tested (recall 28 plots were established in 1992 and 8 in 1995) and we found no difference in the results, except that the interaction between warming and year became significant for Papaver radicatum (EGLS, F1,82 = 13.35, < 0.01). Thus, for simplicity, we present the results of the model without accounting for the duration of the experiment.

Because we measured vegetation infrequently throughout the period, we wanted to verify that our sampling could detect long-term trends and not just inter-annual variability. To test for inter-annual variation in plant cover, a subset of nine 1-m2 plots were re-sampled using the point-intercept method at peak season in 1995 and 1996. We tested for differences in the cover of abundant vascular species and plant functional types with the same EGLS model structure that was described above. In 1995, non-vascular plants were not identified to species level, so we could not compare the cover of abundant cryptogams.

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Environmental monitoring

From 1992 to 2007, the mean annual air temperature in the plots was about −15 °C and the mean July temperature was about 7 °C. During the 2007 growing season, the OTCs increased daily mean temperature by 1.0 ± 0.1 °C (95% CI) and average daily maximum temperature by 2.4 ± 0.2 °C. The OTCs did not alter spring melt-out dates (EGLS, F1,392 = 1.4, = 0.23), soil water content (EGLS, F1,86 = 0.01, = 0.91), maximum thaw depth in 2007 (t-test, t1,34 = 1.48, = 0.15) or bioavailable soil nitrogen in 2008 (t-test, t1,34 = −0.39, = 0.70). The average date of melt-out was 17 June (day 168), average soil water content was 24.7 ± 1.8%, average maximum thaw depth in 2007 was 70.0 ± 2.1 cm, and bioavailable soil nitrogen for the 2008 season (58 days) was 13.6 ± 1.5 μg N 10 cm−2.

Plant community

On the whole, after 12–15 years, this plant community was not strongly affected by experimental warming (Table 1). Year affected Bray–Curtis scores, while both year (FY2,80 = 7.90, PY < 0.01) and warming (FW1,80 = 4.29, PW < 0.01) were significant in the permanova (R2 values for year and warming were <0.10). Species diversity and canopy height were not affected by warming.

Table 1.   Effects of experimental warming (W), year (Y) and their interaction (W × Y) on dissimilarity, plant cover, species diversity and canopy height for 36 permanent 1-m2 plots in an evergreen-shrub heath at Alexandra Fiord, Ellesmere Island, Nunavut, Canada. The data were analysed with estimated generalized least-squares models [variables = W, Y and W × Y, correlation structure = CAR(1) on Year|PlotID]. For all models, d.f. = 1,82 for W, Y and W × Y, except for canopy height which had d.f. = 1,68. NS is non-significant (α = 0.05). Means ± 95% confidence intervals are also shown for control (CON) and warmed (OTC) treatment levels
 Model resultsMeans ± 95% confidence intervals
199620002007
WYW × YCONOTCCONOTCCONOTC
Dissimilarity
 Bray–CurtisNS<0.01NS0.13 ± 0.070.17 ± 0.030.20 ± 0.040.18 ± 0.030.19 ± 0.030.22 ± 0.03
Plant cover
 Evergreen shrubsNSNSNS44.0 ± 10.757.1 ± 5.741.5 ± 3.644.6 ± 5.353.1 ± 9.754.4 ± 10.7
 Bryophytes0.01<0.01NS39.6 ± 10.041.9 ± 7.435.3 ± 5.842.9 ± 5.549.0 ± 6.761.2 ± 5.9
 Lichens0.01NSNS15.8 ± 7.47.7 ± 1.85.3 ± 1.23.5 ± 1.28.5 ± 2.56.0 ± 1.4
 Deciduous shrubsNS0.04NS10.2 ± 7.97.9 ± 2.99.2 ± 2.37.9 ± 2.511.4 ± 2.69.3 ± 2.5
 GraminoidsNSNSNS3.2 ± 0.75.7 ± 4.04.4 ± 1.44.1 ± 1.93.1 ± 1.33.4 ± 1.2
 ForbsNS<0.01NS3.2 ± 1.74.4 ± 1.93.4 ± 1.04.8 ± 1.61.8 ± 0.92.1 ± 0.7
Species diversity
 Shannon-WienerNSNSNS2.33 ± 0.172.22 ± 0.142.03 ± 0.081.98 ± 0.022.11 ± 0.082.13 ± 0.03
 Simpson’sNSNSNS0.84 ± 0.050.82 ± 0.020.81 ± 0.020.80 ± 0.020.82 ± 0.020.82 ± 0.03
 Species richnessNSNSNS18.6 ± 0.819.8 ± 1.513.6 ± 0.813.3 ± 1.015.7 ± 1.316.4 ± 1.4
 Pielou’s evennessNSNSNS0.12 ± 0.010.12 ± 0.010.15 ± 0.010.15 ± 0.010.13 ± 0.010.13 ± 0.01
Height
 Canopy heightNSNSNSNANA36.6 ± 7.829.8 ± 8.434.9 ± 8.332.7 ± 7.9

The abundance of several plant functional types was affected by warming and year. Warming decreased lichen cover by 3.5% and increased bryophyte cover by 6.3% but total cryptogam cover was unaffected. Over the study period, bryophytes increased by 16.2%, deciduous shrubs increased by 3.1% and forbs decreased by 2.3%.

Dominant species responses were relatively minor and therefore similar to plant functional type responses. Of the nine dominant species, which together represent c. 65% of the living vegetation in the community, only A. turgidum cover was affected by warming, increasing by 3% (Table 2). Papaver radicatum and O. digyna decreased by 1–2% over time, while D. integrifolia and S. arctica cover both increased by 3%. The analyses of rarer species were unreliable, given that these species were absent from most plots.

Table 2.   Effects of experimental warming (W), year (Y) and their interaction (W × Y) on the plant cover of the most abundant species from 1996 to 2007 for 36 permanent 1-m2 plots in an evergreen-shrub heath at Alexandra Fiord, Ellesmere Island, Nunavut, Canada. Species are ordered by their abundances in 2007. Together, these species constitute c. 65% of the living vegetation in the community. The data were analysed with estimated generalized least-squares models [variables = W, Y and W × Y, correlation structure = CAR(1) on Year|PlotID]. For all models, d.f. = 1, 82 for W, Y and W × Y. NS is non-significant (α = 0.05)
Species nameCommon namePlant functional typeWYW × Y
Cassiope tetragonaArctic white heatherEvergreen shrubNSNSNS
Dryas integrifoliaArctic avenEvergreen shrubNS0.02NS
Aulacomnium turgidumN/ABryophyte<0.01NSNS
Dicranum spp.N/ABryophyteNSNSNS
Salix arcticaArctic willowDeciduous shrubNS0.04NS
Luzula arcticaArctic woodrushGraminoidNSNSNS
Collema ceraniscumN/ALichenNSNSNS
Papaver radicatumArctic poppyForbNS<0.01NS
Oxyria digynaMountain sorrelForbNS0.02NS

In the plot re-sampling test between 1995 and 1996, we did not detect inter-annual variability in the cover of any of the functional types (all six > 0.05). The four most abundant vascular species also did not change (all four > 0.05), but P. radicatum increased (EGLS, F1,16 = 7.19, = 0.02), and O. digyna decreased in cover (EGLS, F1,16 = 6.16, = 0.02). Together, these two forb species represented 2% of the total plant cover in 2007.

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

This High Arctic evergreen-shrub heath largely resisted 15 years of experimental warming. Given that many plant (Walker et al. 2006) and microbial (Allison & Martiny 2008) communities have responded strongly and rapidly to warming manipulations, our findings are surprising; especially since three other plant communities (wet sedge, moist graminoid/evergreen shrub and dry deciduous shrub) at the same site were among the most responsive communities to warming in the recent ITEX meta-analysis (Walker et al. 2006). The moist evergreen-shrub heath, which was not included in the meta-analysis, is similar in temperature, growing season length, nutrient flux and functional type composition to these three communities; however, species abundances differ widely (J.M.G. Hudson & G.H.R. Henry, unpublished data). The evergreen-shrub heath is dominated by relatively long-lived, stress-tolerant, conservative species with inflexible growth strategies and slow decomposition rates (Grime 1979; Chapin, Autumn & Pugnaire 1993); the other communities are dominated by species that have been shown to be more responsive to warming such as deciduous shrubs and graminoids (Walker et al. 2006). The lack of substantial vegetation change in the heath is not unique. At other Arctic sites, lichen, bryophyte and evergreen-shrub dominated heaths were less responsive to experimental warming than other plant communities (Hollister, Webber & Tweedie 2005; Jónsdóttir et al. 2005; Wahren, Walker & Bret-Harte 2005).

Although this plant community exhibited overall resistance to passive warming, there was a decrease in lichen cover (crustose, foliose and fruticose lichens). Lichens also declined across the six High Arctic communities in the recent meta-analysis (Robinson et al. 1998; Hollister, Webber & Tweedie 2005; Walker et al. 2006). The decline was attributed to increases in vascular plant abundance (Walker et al. 2006). Other factors must have caused the decline in our plots because we observed no change in vascular plants. The consequences of tundra-wide lichen decreases may be important, because some lichens are nitrogen fixers, some are an important source of winter forage for herbivores such as Rangifer tarandus (caribou and reindeer), and some influence soil temperatures, water regimes and biogeochemical cycling (Longton 1997).

In tundra communities, bryophytes influence ecosystem functions such as carbon, nitrogen, and water cycling (Turetsky 2003). In this study, bryophyte cover increased due to warming and over time. Elsewhere, OTCs have typically decreased bryophyte cover (Hollister, Webber & Tweedie 2005; Walker et al. 2006), while cover has sometimes increased over time (Hollister, Webber & Tweedie 2005). Warmer temperatures combined with drier conditions inhibit bryophyte growth, while warmer, wetter conditions encourage growth (Potter et al. 1995). In our study, OTCs did not lower soil moisture, which may explain why bryophyte cover was able to increase. It appears that faster-growing bryophytes are outcompeting lichens in our study community. Overall, warming had no net effect on cryptogam cover.

We have identified and attempted to address three of the limitations of our study. First, because the earliest measurements were taken 1–4 years after the treatments were initiated, we do not have pre-manipulation data and we could not use Before-After-Control-Impact statistical analyses. As a result, we cannot rule out that warming responses were actually caused by initial plot characteristics (i.e. plots had more of one species and less of another). However, the experiment was established in a visually homogeneous area and our level of replication (n = 18) was relatively high. Second, we infrequently measured the vegetation; however, studies that used the point-intercept method to monitor low Arctic vegetation more frequently than we did show slow, gradual changes and little inter-annual variability (Graglia et al. 2001; Wahren, Walker & Bret-Harte 2005). As well, the repeated-year (1995–96) sampling in our study detected no substantial abundance changes. Third, all simulated climate-change experiments have artefacts and are not perfect substitutes for ambient climate change (Kennedy 1995a,b; Marion et al. 1997; Shen & Harte 2000; Wookey 2008). For example, changes to microbial communities may drive vegetation responses (Rinnan et al. 2007), but the composition of these communities and their diversity and evenness have not been affected by experimental warming at the site (Fujimura, Egger & Henry 2008). Nevertheless, the ITEX approach has been highly effective at causing and detecting community-level changes (e.g. Graglia et al. 2001; Hollister, Webber & Tweedie 2005; Jónsdóttir et al. 2005; Wahren, Walker & Bret-Harte 2005; Walker et al. 2006). For these reasons, we have high confidence in our findings of limited plant compositional changes.

Experimental warming altered leaf morphological and chemical traits of several dominant plant species in our study community (Tolvanen & Henry 2001; J.M.G. Hudson & G.H.R. Henry, unpublished data), but these changes did not translate into significant shifts in community composition, even after 15 years. There are at least three possible reasons for this lack of community-level vegetation change. First, tundra ecosystems are usually slow to react to perturbations. For example, a sub-Arctic evergreen-shrub heath needed more than 10 years of warming to change its below-ground community composition (Rinnan et al. 2007), and above-ground communities may take as many as two decades to change (Epstein et al. 2004). Second, environmental factors other than temperature may have limited the ability of the vegetation to respond to warming; however, we are unable to identify a limiting factor. For example, low-dose nutrient additions and watering had no effect on the standing crop of this community (Henry, Freedman & Svoboda 1986) and other factors, such as light, UV-B and atmospheric carbon dioxide concentrations affect Arctic vegetation, but are generally not limiting (Callaghan et al. 2005). It is possible that a combination of these factors restrained this system, but we could not test for this in our design. Third, the plant community may be accustomed to minor changes in environmental factors. As mentioned, it was not affected by single low-dose nutrient additions (Henry, Freedman & Svoboda 1986), short-term passive warming, growing season length manipulations (Johnstone 1995), or long-term passive warming; however, high rates of nutrient addition strongly decreased total vascular plant production (Henry, Freedman & Svoboda 1986). Hence, this plant community appears resistant to minor perturbations and sensitive to more substantial ones.

The evergreen-shrub heath community became more productive between 1981 and 2008, largely due to regional warming (Hudson & Henry 2009). Over this period, mean July air temperatures increased by c. 2.5 °C (Hill & Henry in press; Trefry et al. in press), which was much greater than the experimental warming reported here. Control plots from both this study and Hudson & Henry (2009) show a similar temporal trend: bryophytes increased in abundance. In this study, forbs and deciduous shrubs changed slightly in abundance, and while evergreen shrubs did not increase overall, they did increase in many individual plots. Therefore, the control plots from this study appear to be responding in the same way as the plots in the 1981–2008 observational study (Hudson & Henry 2009). By contrast, responses to ambient and experimental warming differed. While ambient warming affected three of the vascular functional groups, experimental warming only affected bryophyte and lichen cover. The lack of vegetation change in response to experimental warming may have been due to the treatment’s inability to affect environmental variables other than temperature. For example, warming did not affect active layer depth, date of spring melt-out, nutrient availability or soil water content in our study, even though all of these variables are predicted to be affected by climate change (Chapin, McFadden & Hobbie 1997). At other sites, temporal changes observed in unmanipulated vegetation have often been greater than changes due to experimental warming (Chapin et al. 1995; Hollister, Webber & Tweedie 2005; Wahren, Walker & Bret-Harte 2005).

Strong and rapid plant responses to experimental warming have been detected across the Arctic (e.g. Walker et al. 2006). However, our results of limited changes are supported by a number of studies that have documented resistance in other plant communities (Table 3). This leads us to ask the question: why are some plant communities resistant to climate change, while so many others are not? Grime et al. (2000) proposed four non-mutually exclusive site-characteristic hypotheses: (a) diverse communities may be more resistant and resilient (species richness hypothesis), (b) communities that frequently experience climate extremes may be more resistant to extremes of a similar nature in the future (previous exposure hypothesis), (c) mature plant communities may be more resistant to change than early successional ones (successional status hypothesis) and (d) plant traits, such as life history and growth rate, influence community response to climate change and other stresses (functional composition hypothesis). Species richness is unlikely to be important because several responsive communities at our site have lower species richness scores than the evergreen-shrub heath described here (Walker et al. 2006). Previous exposure to climate extremes and successional status do not promote resistance (Jónsdóttir et al. 2005). Only hypothesis (d) is likely. The eight resistant communities have similar functional compositions, because they are all relatively unproductive and dominated by stress-tolerant, long-lived species capable of withstanding strong inter-annual environmental variation (Grime 1979). They also have conservative nutrient-use strategies that are associated with relatively slow growth rates, long tissue life spans and high root : shoot ratios. We suggest that these vegetation types, made up of conservative, unproductive species, have inherently developed community-level resistance to minor shifts in environmental variables, such as temperature. It is clear that further study is needed to identify the exact mechanisms responsible because so few plant communities may be able to resist the collective effects of global change.

Table 3.   List of plant communities that exhibit community-level resistance to ≥4 years of experimental warming. ‘Community responses’ = effects of OTCs on plant functional type (PFT) cover (bryophytes, deciduous shrubs, evergreen shrubs, forbs, graminoids and lichens). ‘Plant community’ = dominant PFT and community type. ‘Temperature increase’ = reported warming effect of the open-top chambers. All other studies except ours describe (i) communities that were sensitive to non-warming manipulations or (ii) additional plant communities that were sensitive to warming
SiteCommunity responsesPlant communityTemperature increase (°C)Number of yearsSource
Abisko, SwedenNo changeEvergreen-shrub heath2.0–4.09Richardson et al. (2002)
Alexandra Fiord, CanadaBryophyte cover increased, lichen cover decreasedEvergreen-shrub heath1.0–1.315This study
Atqasuk, AlaskaNo changeLichen heath0.6–2.25Hollister, Webber & Tweedie (2005)
Barrow, AlaskaBryophyte and lichen cover decreasedLichen heath0.6–2.27Hollister, Webber & Tweedie (2005)
Finse, NorwayNo changeEvergreen-shrub heath1.54Klanderud (2008)
Ny Ålesund, SvalbardLichen cover decreasedEvergreen-shrub polar semi-desert3.55Robinson et al. (1998)
Thingvellir, IcelandNo changeBryophyte heath1.35Jónsdóttir et al. (2005)
Toolik, AlaskaNo changeEvergreen-shrub heath1.58Wahren, Walker & Bret-Harte (2005)

The High Arctic evergreen-shrub heath we monitored was largely resistant to 15 years of 1.0–1.3 °C warming. This is the longest-running experiment in the Arctic to observe community-level resistance to climate-change manipulations. Our findings suggest that not all plant communities will be sensitive to minor warming and that ecosystem-level responses may only be observed after a temperature threshold is crossed.

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Financial support for the study has been provided by grants to G.H.R.H. from the Natural Sciences and Engineering Resource Council (NSERC) of Canada, the Northern Scientific Training Program, ArcticNet and the Government of Canada International Polar Year program. Additional funding to J.M.G.H. has been provided by NSERC and the Garfield Weston Foundation. Logistical support was provided by the Polar Continental Shelf Project and the Royal Canadian Mounted Police. We thank Catherine La Farge and Trevor Goward for bryophyte and lichen identification. The UBC Tundra Ecology Laboratory and discussion group, Valerie LeMay, Roy Turkington and two anonymous referees provided helpful comments on earlier drafts. We thank all of the field crews who contributed to this project through the years.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  • Allison, S.D. & Martiny, J.B.H. (2008) Resistance, resilience, and redundancy in microbial communities. Proceedings of the National Academy of Sciences of the United States of America, 105, 1151211519.
  • Anderson, M.J. (2001) A new method for non-parametric multivariate analysis of variance. Austral Ecology, 26, 3246.
  • Anisimov, O.A., Vaughan, D.G., Callaghan, T.V., Furgal, C., Marchant, H., Prowse, T.D., Vihjálmsson, H. & Walsh, J.E. (2007) Polar regions (Arctic and Antarctic). Climate Change 2007: Impacts, Adaptation and Vulnerability. Contribution of Working Group II to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change (eds M.L.Parry, O.F.Canziani, J.P.Palutikof, P.J.Van Der Linden & C.E.Hanson), pp. 653685. Cambridge University Press, Cambridge, UK.
  • Arft, A.M., Walker, M.D., Gurevitch, J., Alatalo, J.M., Bret-Harte, M.S., Dale, M. et al. (1999) Responses of tundra plants to experimental warming: meta-analysis of the International Tundra Experiment. Ecological Monographs, 69, 491511.
  • Callaghan, T.V., Björn, L.O., Chapin III, F.S., Chernov, Y., Christensen, T.R., Huntley, B., Ims, R. & Johansson, M. (2005) Arctic tundra and polar desert ecosystems. Arctic Climate Impact Assessment: Scientific Report (eds C.Symon, L.Arris & B.Heal), pp. 243352. Cambridge University Press, Cambridge, UK.
  • Chapin III, F.S., Autumn, K. & Pugnaire, F. (1993) Evolution of suites of traits in response to environmental stress. American Naturalist, 142, S78S92.
  • Chapin III, F.S., McFadden, J.P. & Hobbie, S.E. (1997) The role of arctic vegetation in ecosystem and global processes. Ecology of Arctic Environments (eds S.J.Woodin & M.Marquiss), pp. 97112. Blackwell, Oxford, UK.
  • Chapin III, F.S., Shaver, G.R., Giblin, A.E., Nadelhoffer, K.J. & Laundre, J.A. (1995) Responses of Arctic tundra to experimental and observed changes in climate. Ecology, 76, 694711.
  • Clarke, K.R. & Gorley, R.N. (2006) PRIMER v6: User Manual/Tutorial. PRIMER-E, Plymouth.
  • Epstein, H.E., Calef, M.P., Walker, M.D., Chapin III, F.S. & Starfield, A.M. (2004) Detecting changes in arctic tundra plant communities in response to warming over decadal time scales. Global Change Biology, 10, 13251334.
  • Fujimura, K.E., Egger, K.N. & Henry, G.H.R. (2008) The effect of experimental warming on the root-associated fungal community of Salix arctica. The ISME Journal, 2, 105114.
  • Graglia, E., Jonasson, S., Michelsen, A., Schmidt, I.K., Havström, M. & Gustavsson, L. (2001) Effects of environmental perturbations on abundance of subarctic plants after three, seven and ten years of treatments. Ecography, 24, 512.
  • Grime, J.P. (1979) Plant Strategies and Vegetation Processes. John Wiley & Sons, Chichester, UK.
  • Grime, J.P., Brown, V.K., Thompson, K., Masters, G.J., Hillier, S.H., Clarke, I.P., Askew, A.P., Corker, D. & Kielty, J.P. (2000) The response of two contrasting limestone grasslands to simulated climate change. Science, 289, 762765.
  • Grime, J.P., Fridley, J.D., Askew, A.P., Thompson, K., Hodgson, J.G. & Bennett, C.R. (2008) Long-term resistance to simulated climate change in an infertile grassland. Proceedings of the National Academy of Sciences of the United States of America, 105, 1002810032.
  • Henry, G.H.R., Freedman, B. & Svoboda, J. (1986) Effects of fertilization on three tundra plant communities of a polar desert oasis. Canadian Journal of Botany-Revue Canadienne De Botanique, 64, 25022507.
  • Henry, G.H.R. & Molau, U. (1997) Tundra plants and climate change: the International Tundra Experiment (ITEX). Global Change Biology, 3, 19.
  • Hill, G.B. & Henry, G.H.R. (in press) Responses of a High Arctic wet sedge tundra to climate warming since 1980. Global Change Biology, doi:10.1111/j.1365-2486.2010.02244.x.
  • Hollister, R.D. & Webber, P.J. (2000) Biotic validation of small open-top chambers in a tundra ecosystem. Global Change Biology, 6, 835842.
  • Hollister, R.D., Webber, P.J. & Tweedie, C.E. (2005) The response of Alaskan arctic tundra to experimental warming: differences between short- and long-term responses. Global Change Biology, 11, 525536.
  • Hudson, J.M.G. & Henry, G.H.R. (2009) Increased plant biomass in a High Arctic heath from 1981-2008. Ecology, 90, 26572663.
  • IPCC (2007) Climate Change 2007: Synthesis Report. Contribution of Working Groups I, II and III to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. IPCC, Geneva, Switzerland.
  • Ireland, R.R., Bird, C.D., Brassard, G.R., Schofield, W.B. & Vitt, D.H. (1980) Checklist of the Mosses of Canada. National Museums of Canada, Ottawa.
  • Johnstone, J. (1995) Responses of Cassiope tetragona, a High Arctic Evergreen Dwarf Shrub, to Variations in Growing Season Temperature and Growing Season Length at Alexandra Fiord, Ellesmere Island. MSc Thesis, University of British Columbia, Vancouver, Canada.
  • Jónsdóttir, I.S., Magnússon, B., Gudmundsson, J., Elmarsdóttir, A. & Hjartarson, H. (2005) Variable sensitivity of plant communities in Iceland to experimental warming. Global Change Biology, 11, 553563.
  • Kennedy, A.D. (1995a) Simulated climate change: are passive greenhouses a valid microcosm for testing the biological effects of environmental perturbations. Global Change Biology, 1, 2942.
  • Kennedy, A.D. (1995b) Temperature effects of passive greenhouse apparatus in high-latitude climate change experiments. Functional Ecology, 9, 340350.
  • Klanderud, K. (2008) Species-specific responses of an alpine plant community under simulated environmental change. Journal of Vegetation Science, 19, 363372.
  • Longton, R.E. (1997) The role of bryophytes and lichens in polar ecosystems. Ecology of Arctic Environments (eds S.J.Woodin & M.Marquiss), pp. 6996. Blackwell, Oxford, UK.
  • Marion, G.M., Henry, G.H.R., Freckman, D.W., Johnstone, J., Jones, G., Jones, M.H., Lévesque, E., Molau, U., Mølgaard, P., Parsons, A.N., Svoboda, J. & Virginia, R.A. (1997) Open-top designs for manipulating field temperature in high-latitude ecosystems. Global Change Biology, 3, 2032.
  • Molau, U. & Mølgaard, P. (1996) International Tundra Experiment (ITEX) Manual. Danish Polar Center, Copenhagen, Denmark.
  • Muc, M., Freedman, B. & Svoboda, J. (1989) Vascular plant communities of a polar oasis at Alexandra Fiord (79°N), Ellesmere Island, Canada. Canadian Journal of Botany-Revue Canadienne De Botanique, 67, 11261136.
  • Oberbauer, S.F., Tweedie, C.E., Welker, J.M., Fahnestock, J.T., Henry, G.H.R., Webber, P.J., Hollister, R.D., Walker, M.D., Kuchy, A., Elmore, E. & Starr, G. (2007) Tundra CO2 fluxes in response to experimental warming across latitudinal and moisture gradients. Ecological Monographs, 77, 221238.
  • Oksanen, J., Kindt, R., Legendre, P., O’Hara, B., Simpson, G.L., Stevens, M.H.H. & Wagner, H. (2008) Vegan: Community Ecology Package. R package version 1.13-2.
  • Pinheiro, J. & Bates, D. (2000) Mixed-Effects Models in S and S-PLUS. Springer, New York, USA.
  • Pinheiro, J., Bates, D., DebRoy, S., Sarkar, D. & the R Core team (2007) nlme: linear and Nonlinear Mixed Effects Models. R package version 3.1-86.
  • Porsild, A.E. & Cody, W.J. (1980) Vascular Plants of Continental Northwest Territories, Canada. National Museums of Canada, Ottawa, Canada.
  • Potter, J.A., Press, M.C., Callaghan, T.V. & Lee, J.A. (1995) Growth responses of Polytrichum commune and Hylocomium splendens to simulated environmental change in the sub-arctic. New Phytologist, 131, 533541.
  • R Development Core Team (2007) R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria.
  • Richardson, S.J., Press, M.C., Parsons, A.N. & Hartley, S.E. (2002) How do nutrients and warming impact on plant communities and their insect herbivores? A 9-year study from a sub-Arctic heath. Journal of Ecology, 90, 544556.
  • Rinnan, R., Michelsen, A., Baath, E. & Jonasson, S. (2007) Fifteen years of climate change manipulations alter soil microbial communities in a subarctic heath ecosystem. Global Change Biology, 13, 2839.
  • Robinson, C.H., Wookey, P.A., Lee, J.A., Callaghan, T.V. & Press, M.C. (1998) Plant community responses to simulated environmental change at a high arctic polar semi-desert. Ecology, 79, 856866.
  • Shaver, G.R., Canadell, J., Chapin III, F.S., Gurevitch, J., Harte, J., Henry, G.H.R., Ineson, P., Jonasson, S., Melillo, J., Pitelka, L. & Rustad, L. (2000) Global warming and terrestrial ecosystems: a conceptual framework for analysis. BioScience, 50, 871882.
  • Shen, K.P. & Harte, J. (2000) Ecosystem climate manipulations. Methods in Ecosystem Science (eds O.E.Sala, R.B.Jackson, H.A.Mooney & R.B.Howarth), pp. 353369. Springer, New York, USA.
  • Svoboda, J. & Freedman, B. (1994) Ecology of a Polar Oasis: Alexandra Fiord, Ellesmere Island, Canada. Captus University Publications, Toronto, Canada.
  • Tolvanen, A. & Henry, G.H.R. (2001) Responses of carbon and nitrogen concentrations in high arctic plants to experimental warming. Canadian Journal of Botany-Revue Canadienne De Botanique, 79, 711718.
  • Trefry, S.A., Freedman, B., Hudson, J.M.G. & Henry, G.H.R. (in press) Breeding bird surveys at Alexandra Fiord, Ellesmere Island, Nunavut (1980–2008). Arctic, 63(3).
  • Turetsky, M.R. (2003) The role of bryophytes in carbon and nitrogen cycling. The Bryologist, 106, 395409.
  • Wahren, C.H.A., Walker, M.D. & Bret-Harte, M.S. (2005) Vegetation responses in Alaskan arctic tundra after 8 years of a summer warming and winter snow manipulation experiment. Global Change Biology, 11, 537552.
  • Walker, M.D., Wahren, C.H., Hollister, R.D., Henry, G.H.R., Ahlquist, L.E., Alatalo, J.M. et al. (2006) Plant community responses to experimental warming across the tundra biome. Proceedings of the National Academy of Sciences of the United States of America, 103, 13421346.
  • Wookey, P.A. (2008) Experimental approaches to predicting the future of tundra plant communities. Plant Ecology & Diversity, 1, 299307.