Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Long-term experimental warming effects on tundra plant sexual reproduction in the high Arctic Klady, Rebecca A. 2006

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata

Download

Media
831-ubc_2007-0142a.pdf [ 8.31MB ]
Metadata
JSON: 831-1.0100758.json
JSON-LD: 831-1.0100758-ld.json
RDF/XML (Pretty): 831-1.0100758-rdf.xml
RDF/JSON: 831-1.0100758-rdf.json
Turtle: 831-1.0100758-turtle.txt
N-Triples: 831-1.0100758-rdf-ntriples.txt
Original Record: 831-1.0100758-source.json
Full Text
831-1.0100758-fulltext.txt
Citation
831-1.0100758.ris

Full Text

LONG-TERM EXPERIMENTAL WARMING EFFECTS ON TUNDRA PLANT SEXUAL REPRODUCTION IN THE HIGH ARCTIC by REBECCA A. KLADY B.Sc, Nipissing University, 2001 Diploma, Canadore College, 2001 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in THE FACULTY OF GRADUATE STUDIES (Geography) THE UNIVERSITY OF BRITISH COLUMBIA November 2006 ' Rebecca A. Klady, 2006 A B S T R A C T Predictions that climate warming wi l l enhance plant sexual reproduction in the High Arctic were examined using a field experiment at a polar oasis and a polar semi-desert site in the eastern Canadian High Arctic. Small open top chambers (OTCs), which simulated climate warming, were established in plant communities along a soil moisture gradient in 1992. Over two growing seasons, fresh and over-wintered seeds across a range of species were collected from aerial seed banks exposed to experimental warming and ambient conditions. Seeds were weighed and germinated to measure changes in reproductive effort and success in response to experimental warming. OTCs increased within-plot growing season air temperatures by 1 - 2°C, which is within range of general circulation model (GCM) predictions for climate warming in the Arctic. Reproductive effort and success of fresh seeds were enhanced by warming in most species, depending on initial site conditions. Enhanced reproductive effort and success may be attributed to warming conditions, which advanced dates of snowmelt and extended the growing season. Similar effects on over-wintered seeds were likely, but seed dispersal prior to over-wintered seed harvests confounded these results. Inter-annual variability in reproductive success appeared to be diminished by experimental warming. Further testing wi l l verify i f this result is an indicator of long-term (> 10 y) warming effects. Results of this study confirm predictions that long-term warming wi l l enhance sexual reproduction in high arctic plants. These changes wi l l have implications for plant demographics at the community-level and the rate and extent of bare-ground colonization, particularly i f rates of seedling establishment also increase. T A B L E O F C O N T E N T S Abstract ii Table of Contents iii List of Tables v List of figures . .. vi Acknowledgments •• ix 1. Introduction 1 2. Methods 8 2.1 Study Location • 8 2.1.1 Site A rea Description 8 2.1.2 Site Descriptions 9 2.1.3 Experimental Design 10 2.1.4 Target Species H 2.2 Measures of Reproductive Effort and Success (Biomass and Germination) 12 2.2.1 Aboveground Biomass 12 2.2.2 Seed Harvests 13 2.2.3 Germination Trials ; 14 2.2.4 Regional Variability in Germination 14 2.3 Statistical Analysis 15 2.3.1 General Criteria 15 2.3.2 Parametric Tests 15 2.3.2.1 G L M and Tests for Normality and Heteroscedasticity 16 2.3.2.2 G L M and Transformations 16 2.3.3 Non-Parametric Tests 16 2.3.3.1 Logistic Regression 16 2.3.3.2 Genmod • 17 2.3.4 Analytical Procedure 17 2.3.5 Reproductive Effort (Biomass) 18 2.3.6 Reproductive Success (Germination) : 18 3.0 Results : 20 3.1 Environmental Data 20 3.2 Reproductive Effort: Vegetative and Reproductive Biomass: 21 3.2.1 Sample Size 21 3.2.2 Combined Species Response 22 3.2.3 Shrubs - Dry as integrifolia, Salix arctica 24 3.2.4 Forbs - Papaver radicatum and Oxyria digyna 29 3.2.5 Graminoids - Festuca brachyphylla, Eriophorum angustifolium subsp. triste, Luzula spp. and Carex misandra 30 3.3 Reproductive Success: Cumulative, Peak and Rate of Germination 32 3.3.1 Sample Size 32 3.3.2 Combined Species Response 32 iii 3.3.3 Shrubs - Dryas integrifolia, Salix arctica 34 3.3.4 Forbs - Papaver radicatum, Oxyria digyna, Draba spp 42 3.3.5 Graminoids - Festuca brachyphylla, Eriophorum angustifolium subsp. triste, Luzula spp. and Carex misandra : 43 3.4 Spatial Variability in Germination of Dryas spp.: International Comparisons 45 3.5 Reproductive Effort and Success: Correlations Between Seed Biomass and Germination 45 3.6 Summary of Results 46 4.0 Discussion 48 4.1 Warming Effects on Reproductive Effort and Success in a Polar Oasis '. 48 4.2 Warming Effects on Reproductive Effort and Success in a Polar Semi-Desert 53 4.3 Warming Effects on Over-wintering Seeds (F2004 vs. OW2004) 54 4.4 Inter-annual Variability in Reproductive Success 55 4.5 Local and Regional Variability in Reproductive Effort and Success ; 57 4.6 Correlations Between Reproductive Effort and Success ..59 4.7 Long-term Effects of Warming 60 5.0 Conclusions and Recommendations 61 References Cited 63 Appendix A 70 Appendix B ; '. 71 Appendix C 72 Appendix D . 73 Addendum 74 i v LIST OF TABLES Table 1. List of target species and their growth forms and associated sites from which each species was sampled 22 Table 2. Mean fresh (F2004) cumulative seed germination (%) ± SD from Alexandra Fjord. Germination was averaged over plots by species, treatment (C = control, W = warming) and site. ** p <0.05 for warming effects, site effects not shown. Sample sizes are reported in Appendix B 35 Table 3. Mean over-winter cumulative seed germination (%) ± SD for species sampled in June 2004 (OW2003) and June 2005 (OW2004) from Alexandra Fjord. Germination was averaged over plots by species, treatment (C = control, W = warming) and site. ** p <0.05 for warming effects, site effects not shown. Sample sizes are reported in Appendix B .'. 36 Table 4. Mean peak germination (%) ±SD, and mean time elapsed to day of peak germination (day) across a range of species for fresh (F2004) and over-winter (OW2003, OW2004) seeds. Percent germination values were averaged by species over treatment plots (C = control, W = warming) and within sites. ** p <0.05 for warming effects, site effects not shown. Sample sizes provided in Appendix B 39 Table 5. Change (A) in Dry as integrifolia mean cumulative seed germination (%) and germination rate at different sites within the polar oasis at Alexandra Fjord from control and warmed treatments. A germination rate was calculated using Timson's Index values (Section 2.3.6). Negative values (-) indicated reduced cumulative or rate of germination. Samples were averaged over plots within sites and treatments. ** p <0.05. Sample sizes are reported in Appendix B 40 Table 6. Summary of Kendall's Tau-b (x) correlations between seed biomass and cumulative germination for fresh (F2004), over-winter (OW2003) and international seed collections, and associated probabilities (* p < 0.1, ** p < 0.05) and sample sizes (n-plots). Data were pooled by species (All), or analysed separately by species, over treatment (control = C, warming = W) and site. Site names are described in Section 2.1.2 46 Table 7. Summary of sample population increases (+) or decreases (-) in reproductive effort (RE) and reproductive success (RS) in fresh (F2004) and over-wintered (OW2003, OW2004) seeds in response to warming. 'SB' seed biomass, 'FB' flower biomass, 'PB' photosynthetic bract biomass, 'n/a' not available. *p <0.1,"p <0.05 47 LIST OF FIGURES Figure 1. Alexandra Fjord is located on east-central Ellesmere Island, Nunavut, Canada (78° 53' N, 75° 55' W) (inset map, star symbol). An aerial view of Alexandra Fjord is shown in the larger photo with the location and orientation of study sites labelled with white Xs and abbreviated site names. The lowland polar oasis sites include the central cluster of S-CP-DS, CP-DS(C), CP-DS(D) and DDS-G sites; upland polar semi-desert sites are located in the right-hand edge of the aerial photo (PSD-G and PSD-D). For scale, the distance from the coast to the tip of the glacier is approximately 3 km. (Photo courtesy of G. Henry) 5 Figure 2. Open top chamber (OTC) used to simulate climate warming. Six translucent fibreglass sheets with high solar transmittance in the visible spectrum are 0.5 m high and are angled at 60° from horizontal. Sample plots (1 m2) were nestled within the OTC. (Photo courtesy of G. Henry) 11 Figure 3. Mean date of snowmelt (+SD) in 2004 averaged over treatment plots (n = 6-10 plots) from the lowland sites at Alexandra Fjord, including the Sedge Meadow [S-CP-DS], Cassiope Heath [CP-DS(C)], Dryas Heath [CP-DS(D)] and Deciduous Dwarf Shrub-Graminoid [DDS-G] sites 20 Figure 4. (A) Mean air temperature (°C) at the Sedge Meadow [S-CP-DS] and (B) the Dryas Heath [CP-DS(D)] lowland sites at Alexandra Fjord 21 Figure 5. Mean combined species biomass +SD of A) total, annual and flower biomass and B) fresh (F2004) seed biomass averaged over treatment plots from Alexandra Fjord. ** p <D.05 for warming effects, site effects not shown. Sample sizes are reported in Appendix A 23 Figure 6. Mean combined over-winter seed biomass +SD of species sampled in June 2004 (OW2003), and averaged over treatment plots from the Sedge Meadow (S-CP-DS) and Deciduous Dwarf Shrub-Graminoid (DDS-G) sites at Alexandra Fjord. * p <0.1, ** p <0.05 for warming effects, site effects not shown. Sample sizes are reported in Appendix A 24 Figure 7. Dryas integrifolia mean +SD aboveground A) wood biomass [WB], flower biomass [FB] and annual vegetative biomass [AVB], and B) fresh (F2004) seed biomass averaged over treatment plots from the lowland sites at Alexandra Fjord. ** p <0.05 for warming effects, site effects not shown. Sample sizes are reported in Appendix A 25 Figure 8. Mean over-winter seed biomass +SD for species sampled in June 2004 (OW2003) averaged over treatment plots from the Sedge Meadow (S-CP-DS) and Deciduous Dwarf Slmib-Graminoid (DDS-G) lowland sites at Alexandra Fjord. * p <0.1, ** p <0.05 for warming effects, site effects not shown. Sample sizes are reported in Appendix A '. 26 Figure 9. A) Female Salix arctica mean +SD aboveground wood biomass [WB], current-year-fascicle [CYF], photosynthetic bract [PB] and flower biomass [FB], and B) fresh (F2004) seed biomass averaged over treatment plots from the lowland and polar-semi desert sites at Alexandra Fjord. * p < 0.1, ** p <0.05 for warming effects, site effects not shown. Sample sizes are reported in Appendix .A : 27 v i Figure 10. Male Salix arctica mean +SD aboveground wood [WB], current year fascicle [CYF], and flower [FB] biomass averaged over treatment plots from the lowland and polar-semi desert sites at Alexandra Fjord. Sample sizes are reported in Appendix A 29 Figure 11. Mean fresh (F2004) seed biomass +SD for Papaver radicatum and Oxyria cligyna averaged over treatment plots from the Deciduous Dwarf Shrub-Graminoid site at Alexandra Fjord. * p <0.1, ** p <0.05 for warming effects, site effects not shown. Sample sizes are reported in Appendix A . . 30 Figure 12. Festuca brachyphylla, Eriophorum angustifolium subsp. triste, Carex misandra, and Luzula spp. A) mean flower biomass +SD, and B) fresh (F2004) seed biomass for treatment plots from lowland sites at Alexandra Fjord. * p <0.1, ** p <0.05 for warming effects, site effects not shown. Sample sizes are reported in Appendix A 31 Figure 13. Mean combined fresh (F2004) cumulative germination throughout the trial period (35 d). Data were averaged by plots over 3-day intervals within sites and treatments from all the lowland sites, including A) S-CP-DS, B) CP-DS(C), C) CP-DS(D), D) DDS-G and E) PSD-G sites at Alexandra Fjord. Sample sizes are reported in Appendix B. 33 Figure 14. Combined mean over-wintered cumulative germination throughout the germination trial of seeds sampled in June 2004 (OW2003) at the A) S-CP-DS site and B) the DDS-G site. Data were averaged by plots over 3-day intervals within sites and treatments. Sample sizes are reported in Appendix B '. 34 Figure 15. Dryas integrifolia mean fresh (F2004) cumulative germination throughout the trial period. Data were averaged by plots over 3-day intervals within sites and treatments from the A) S-CP-DS, B) CP-DS(C), C) CP-DS(D) and D) DDS-G lowland sites at Alexandra Fjord. ** p <0.05 for warming effects, site effects not shown. Sample sizes are reported in Appendix B 37 Figure 16. Dryas integrifolia mean over-wintered cumulative germination throughout the trial period (35 d) of seeds sampled in June 2004 (OW2003) and June 2005 (OW2004) at the (A & C) S-CP-DS site, (B & D) DDS-G site, and E) the CP-DS(C) and F) CP-DS(D) sites. Data were averaged by plots over 3-day intervals within sites and treatments at Alexandra Fjord. ** p <0.05 for warming effects, site effects not shown. Sample sizes are reported in Appendix B 38 Figure 17. Salix arctica mean seed cumulative germination throughout the germination trial period (35 d) of A) over-wintered seeds sampled in June 2004 (OW2003) and (B - F) fresh (F2004). Data were averaged by plots over three-day intervals within sites and treatments from the lowland sites at Alexandra Fjord. **p <0.05 for warming effects, site effects not shown. Sample sizes are reported in Appendix B 41 Figure 18. Mean fresh (F2004) cumulative gerrnination throughout the trial period (35 d) for (A) Papaver radicatum, (B), Oxyria digyna, (C - D) and Draba spp. Data were averaged by plots over 3-day intervals within sites and treatments from two lowland sites at Alexandra Fjord. * p <0.1 for warming effects, site effects not shown. Sample sizes are reported in Appendix B 43 vu Figure 19. Mean cumulative germination throughout the trial period (35 d) from fresh (F2004) seeds of (A) Festuca brachyphylla, (B - C) Eriophorum triste, and Luzula spp. from (D - F) F2004 and (G) OW 2003. Data were averaged by plots over 3-day intervals within sites and treatments at Alexandra Fjord. ** p <0.05 for warming effects, site effects not shown. Sample sizes are reported in Appendix B 44 Figure 20. Regional variability of fresh (F2004) mean cumulative seed germination (+SD) of Dryas integrifolia and D. octopetala from ambient conditions. Values were averaged over plots within sites. "Alex Fjord Total" includes germination within all lowland polar oasis sites at Alexandra Fjord. Sample sizes are reported in Appendix B. All sites differed (p < 0.0001) •. 45 Vll l ACKNOWLEDGMENTS I would like to thank the following: Professor Greg Henry, who by introducing me to Alexandra Fjord helped me to realize a dream, and was most gracious about letting me "take the reins"; Professors Valerie Lemay and Roy Turkington, to whom I am grateful for their support and their critical eyes; my field assistants, Kerri-Ann Down and Sarah Bogart, who shared freely laughter, learning and love of good food; my lab assistant, Paige Brown, whose heroic work counting and weighing ridiculously small and unwieldy seeds will not be forgotten; other ITEX researchers, including Borgthor Magnusson, Paddy Sullivan and Ulf Molau for kind donations of sample material; additional field/lab support was given by Kyle Terry, the Cape Herschel Crew from U of Toronto and the filming crew for the Adolphus Greely kayaking expedition; for editorial comments, my supervisor and committee members, as well as Kai Hilpert, Geoff Hill and Josh Caulkins; for logistical support & funding, PCSP, ArcticNet, NSERC, NSTP and the RCMP provided generously, while all those wonderful folks who piloted planes or helicopters and cooked delicious meals in the Arctic did so in ways that acknowledged the Arctic's beauty and the hunger of her admirers. I would like to extend a final expression of gratitude to my partner, Kai Hilpert, as well as friends and family, who so generously gave love, listening and laughter throughout the creation of this work. And of course, my thanks to Alexandra Fjord, for sharing her secrets and source (see Addendum). Shalom, Namaste, Liebe. IX 1. I N T R O D U C T I O N The Arctic tundra lies north of the latitudinal tree line and is characterized by long, cold winters, and a short (50 - 70 d) growing seasons (Bliss, 1977; Chapin III & Shaver, 1985). Both winter and growing season conditions affect Arctic plant growth and reproduction (Billings & Mooney, 1968), but growing season conditions, including air and soil temperatures and soil-nutrient and soil-moisture status, are particularly critical (Billings, 1987; Billings & Mooney, 1968; Chapin III & Shaver, 1985). The growing season commences in June, and continues through to late July or mid-late August (Chapin III & Shaver, 1985), depending on local climate, topography and latitude. At high latitudes, incoming solar radiation is 24-h throughout most of this short growing-season, and mean growing season temperatures can range from 3.0 to 5.0°C (Chapin III & Shaver, 1985; Freedman et al, 1994). The annual surface energy balance is negative, which leads to low air and soil temperatures and widespread permafrost (Bliss et al, 1973). Low temperatures and permafrost indirectly contribute to low soil nutrient status and poor drainage in the rooting zone (Chapin III & Shaver, 1985; Webber et al, 1980), which affect plant, growth and reproduction. According to Polunin (1951), Arctic tundra may be divided into two major sub-regions, the High and Low Arctic (Bliss, 1997; Polunin, 1951), in which the distinguishing feature is the extent of vascular plant cover. The Low Arctic is characterised by continuous plant cover, but in the High Arctic, plant cover is discontinuous and patchy. Using a photo-interpretive approach, cover in the High Arctic has been estimated to range from 50% to <5%, with diminishing cover as latitude increases (Walker et al, 2005). Generally speaking, vascular plant reproduction in the Arctic occurs in two ways: (1) asexually, by vegetative expansion through fragmentation and the production of vegetative parts, such as bulbils, tillers or rhizomes, and (2) sexually, by the production of seed (Harper, 1977). In the Low Arctic, plant community dynamics are largely constrained by competition for resources, and establishment by seed is rare (Arft et al, 1999; Bliss & Peterson, 1992; Callaghan & Carlsson, 1997; Hobbie et al, 1999). In contrast, in the High Arctic bare ground dominates and 1 colonization by seed represents an important contribution to plant demographics (Arft et al, 1999; Jones & Henry, 2003; Larsson, 2002; Walker & Chapin, 1987; Welker et al, 1997), despite low flowering frequency and seed set, and high inter-annual variability of these events (ACIA, 2004). In the High Arctic, constraints to plant growth and reproduction, such as low soil nutrient status and short, cold growing seasons, are interdependent (Billings, 1987; Billings & Mooney, 1968; Chapin III & Shaver, 1985). For example, increases in soil temperature have been shown to enhance soil decomposition processes (Grogan & Chapin, 2000; Nadelhoffer et al., 1992), and increase nutrient availability for plant uptake (Nadelhoffer et al., 1997; Rolph, 2003). Constraints on growth and reproduction, which lead to low total plant production within any given growing season, result in a trade-off between allocation to vegetative versus sexual structures in tundra plants (Chambers, 1995). ' Tundra plants also exhibit a trade-off within the process of sexual reproduction itself: "reproductive effort", defined as an organism's investment in reproductive tissues, is more decoupled, or separate from "reproductive success", the final outcome of that investment (Molau & Shaver, 1997), than in plants of any other biome (Molau, 1993; Molau & Shaver, 1997). For example, reproductive effort on a whole-plant basis is often quite high, while reproductive success, measured as seed viability or seed set, is low (ACIA, 2004). This is a function of the severe constraints on plant growth in the Arctic, and possibly a mechanism that reduces the energetic burden of producing sexual structures in years when growing season conditions are unfavourable. Seasonal fluctuations in environmental quality and challenges to growth and reproduction are also accommodated by the development of flower primordia one to several years in advance of flowering (Mooney & Billings, 1961; Sorenson, 1941). This helps ensure that flower maturation and subsequent seed development occurs during particularly favourable growing seasons. Despite such characteristics, reproductive success is persistently low due to high ovule and/or embryo abortion, associated with adverse microclimatic conditions and possibly pollen limitation (Molau & Shaver, 1997). 2 Recent rates of global climate change are unprecedented and are considered largely anthropogenic (ACIA, 2004). Since the 1950s, mean annual temperatures in the Arctic and Antarctic Peninsula have increased by 2 - 3°C (ACIA, 2004). The predicted trend for the circumpolar Arctic, which is partly reliant on current levels of atmospheric CO2, is an increase of 4 - 5°C by 2080 (ACIA, 2004). Since Arctic tundra ecosystems are strongly constrained by temperature (Billings & Mooney, 1968), the effects of climate warming are predicted to be most pronounced at high latitudes (ACIA, 2004; IPCC, 2001; Maxwell , 1992). Already, warming effects have been observed in the reduced extent of sea ice (ACIA, 2004; Hansen et al, 2006), melting polar ice caps (ACIA, 2004; IPCC, 2001), and a decrease in the extent of regions underlain by permafrost (ACIA, 2004; Anisimov & Nelson, 1997; Demchenko et al, 2001; IPCC, 2001). Predicted effects of continued Arctic warming include increases in air and soil temperatures (ACIA, 2004) and enhanced rates of nutrient cycling (Eviner & Chapin, 2003; Nadelhoffer et al, 1992). These changes are expected to affect plant community structure and composition, and colonisation of bare ground (ACIA, 2004; Arft et al, 1999; Dormann & Woodin, 2002; Svoboda & Henry, 1987). Observed and expected changes wi l l ultimately influence global air and ocean circulation, and the global energy and carbon budgets (ACIA, 2004). ; Short-term (<7 y) experimental warming studies that have applied continuous, but low-temperature increases (1 - 5°C) have already demonstrated effects on plant communities in the High Arctic. For example, vegetative biomass increased (Arft et al, 1999; Callaghan et al, 1999; Savile, 1972), and earlier dates of spring snowmelt, which extend the growing season, have been shown to indirectly advance plant phenology and enhance reproductive success (Arft et al, 1999; Welker et al, 1997). Shifts in nutrient allocation over time are expected to favour the production of heavier and more germinable seed (Arft et al, 1999; Henry & Molau, 1997; Sandvik & Totland, 2000; Shaver & Kummerow, 1992; Wookey et al., 1993), as rates of nutrient mineralization are enhanced (Nadelhoffer et al, 1997; Rolph, 2003), resulting in increases in plant-available nutrients that can be allocated toward sexual reproduction. Observations of enhanced reproductive success in tundra plants (Arft et al, 1999), such as increased frequency or degree of seed set (Dormann & Woodin, 2002; Welker et al, 1997; Welker et al, 2005; Wookey et al, 1993), indicate that temperature-driven restrictions on seed 3 viability and seedling survival in the Arctic may diminish with long-term warming. These changes wi l l affect plant demographics at the community-level (Walsh et al, 2002), and indicate the potential for an increase in the rate and extent of bare-ground colonisation (ACIA, 2004; Arft et al, 1999; Bliss & Gold, 1999; Dormann & Woodin, 2002; Larsson, 2002). At high latitudes, changes to these barren-ground landscapes wi l l have implications for primary consumers (Hinzman et al, 2005) and the global carbon budget (ACIA, 2004) The emergence of clear patterns of plant response to environmental perturbations can be slow (Epstein et al, 2004; Tilman, 1982). Despite the rapid rates of change forecasted for the Arctic (Epstein et al, 2004; Tilman, 1982), normal constraints to plant growth limit the rate of ecosystem response (Arft et al, 1999; Chapin et al, 1995), and observed ecosystem responses to warming, such as increases in active layer depth and enhanced nutrient mineralization rates, have been slow (ACIA, 2004; Grogan & Chapin, 2000; Nadelhoffer et al, 1992; Nadelhoffer et al, 1997; Rolph, 2003). These conditions and observations that short-term responses differ from long-term responses (Callaghan et al, 1999; Chapin et al, 1995; Epstein et al, 2000; Hartley et al, 1999) indicate the need for long-term (>10 y), rather than more short-term (<7 y) temperature manipulation studies on plant sexual reproduction in the Arctic (ACIA, 2004; Arft et al, 1999; Chapin et al, 1995; Dormann & Woodin, 2002; Henry & Molau, 1997; Hollister et al, 2005a; Walker et al, 2006). In the Canadian High Arctic, polar deserts and semi-deserts represent a large area of the ice-free land: polar deserts have 1-2% vascular plant cover, which accounts for about 44% of the land area, and 49% of the land is classified as polar semi-desert, with ~ 20% vascular plant cover (Robinson et al, 1998). In contrast, polar oases represent only about 6% of the Canadian Arctic Archipelago (Fig. 1, inset), and 1% of the Queen Elizabeth Islands, the northern cluster of islands within the Archipelago (Bliss, 1977; Freedman et al, 1994), and have comparatively high plant cover, which may approach 100% (Bliss, 1977; Freedman et al, 1994). As the climate warms, bare-ground colonization via changes in plant demographics wi l l be most substantial in polar desert and semi-desert communities, but polar oases wi l l play an important role in future warming as a seed source for colonisation in polar desert and semi-desert landscapes (Arft et al, 1999). • 4 Figure 1. Alexandra Fjord is located on east-central Ellesmere Island, Nunavut, Canada (78° 53' N , 75° 55' W) (inset map, star symbol). A n aerial view of Alexandra Fjord is shown in the larger photo with the location and orientation of study sites labelled with white Xs and abbreviated site names. The lowland polar oasis sites include the central cluster of S-CP-DS, CP-DS(C), CP-DS(D) and D D S - G sites; upland polar semi-desert sites are located in the right-hand edge of the aerial photo (PSD-G and PSD-D). For scale, the distance from the coast to the tip of the glacier is approximately 3 km. (Photo courtesy of G. Henry). In the past, tundra plant response to experimental warming has been highly species- and site-specific (Chapin etal, 1995; Dormann & Woodin, 2002; Jonsdottir etal, 2005), which suggests that multi-species and multi-site studies of experimental warming will enable more accurate forecasting of community changes at high latitudes (ACIA, 2004; Dormann & Woodin, 2002). A multi-species approach to studying plant community dynamics provides two additional benefits. The first is an opportunity to assess the validity of grouping plant responses 5 into broader categories, such as by growth form or functional group, in order to establish "indicator species" and focus future climate warming research (ACIA, 2004; Dormann & Woodin, 2002). The second benefit is that late-flowering species, which may reflect long-term responses to warming better than species that flower early in the season, are more likely to be included in the range of species tested. The responses of a range of growth forms exposed to long-term (>10 yr) simulated warming are expected to provide essential information about both small- and large-scale changes to tundra ecosystems in a warming climate (Callaghan et al., 1999; Chapin et al, 1995; Hartley et al, 1999). The International Tundra Experiment (ITEX) is a collaborative network of Arctic and alpine researchers who study the impacts of climate change on tundra ecosystems. At the 6 t h I T E X Workshop, several studies were proposed to assist with forecasting change in a warming Arctic, including demographic studies, which have been identified as a necessary link between plant growth/reproduction and ecosystem function (Dormann & Woodin, 2002; McGraw & Fetcher, 1992). In keeping with the I T E X directives, I used measures of plant demography to test the hypothesis that long-term experimental warming has enhanced sexual reproduction in tundra plants at a high arctic polar oasis and polar-semi desert. I measured sexual reproduction as reproductive effort and success, using fresh and over-wintered serotinous seed (attached to mother plant) from the aerial seed bank. Measures of reproductive effort and success from the aerial, rather than soil seed bank, provided a direct measure of plant response, without the added complexity of multi-year seed associated with soil seed bank studies. The use of over-wintered seed provided a measure of the effects of warming on seeds exposed to natural vernalisation, the period of cold required for germination in some species such as Carex (Baskin & Baskin, 1998; Chapin & Shaver, 1985). Warming was achieved using small open top chambers (OTCs) that enhanced within-plot temperatures during the growing season from 1-2°C, within range of general circulation model predictions for the Arctic by 2050 (Marion et al, 1997; Maxwell , 1992). OTCs were installed in a range of plant communities, along a soil moisture gradient, since site conditions have been shown to influence both the direction and degree of plant response warming (Arft et al, 1999; Hollister et al, 2005a; van Wijk et al, 2004). Sampling from plant communities with different soil-moisture conditions also provided the opportunity to scale up the results of this study, in an effort to forecast change at larger scales. A range of 6 species and growth forms were also sampled based on previously observed species-specific differences in nutrient allocation (Maessen et al, 1994) and response to warming (Chapin et al, 1996). This study was intended to form a connection between plant demographics and climate warming in the High Arctic, which wi l l contribute to the understanding of plant community dynamics under a climate-warming scenario, and improve the ability to forecast change in the High Arctic. Secondary study locations in Greenland, Sweden and Iceland were used to assess baseline regional variability in seed germination. 7 s 2. METHODS 2.1 Study Location 2.1.1 Site Area Description Alexandra Fjord is located on the coast of east-central Ellesmere Island (78° 53' N , 75° 55' W) (Fig. 1). The research sites were located in a lowland polar oasis and an upland polar semi-desert. The lowland polar oasis (8 km 2 , 0 - 100 m.a.s.l.) has gently sloping topography (1-3% to the north) bounded by steep upland terrain (500 - 700 m.a.s.l.). Melt water from glaciers in the southern part of the lowland is drained by one large and four smaller streams. The combination of slope aspect and other features of the local topography contribute to favourable climatic conditions in which slightly warmer temperatures result in an extended growing season, from early June until mid/late August (~ 75 d). These conditions permit the establishment of the mostly closed-canopy plant communities of the polar oasis, although the vegetation is less than 0.5 m high, and slightly higher plant species diversity and productivity compared to the surrounding, more typical polar semi-desert and polar desert (Freedman et al, 1994). Sites were also established in an upland polar-semi desert just west of the lowland (~ 500 m.a.s.l.; Fig. 1). Polar semi-desert represents the transition between polar oasis and polar desert, where growing conditions are only slightly more favourable than polar desert, owing to seepage and soils that tend to retain moisture throughout the growing season (~ 45 d). Soil moisture conditions at the polar semi-desert site have been described as xeric, but are sufficient for development of cryptogamic crusts and associated sparse plant communities. The bedrock material is initially dolomite, but grades upslope into sandstones and granite. During the growing-season, daily mean air temperatures within the lowland fluctuate around 5°C. Occasionally maximum values are in excess of 15°C (Svoboda and Freedman 1994). Precipitation events are rare at the beginning of the growing season, but increase in frequency and duration, especially once the sea ice has melted. A ten year precipitation record from Eureka (approximately 235 km west of Alexandra Fjord, 79° 58' N 85° 55' W) averaged 4 mm in June, 13 mm in July and 9 mm in August, but unpublished values at Alexandra Fjord have been steadily increasing over the last decade (G.H.R. Henry, pers. com.). For additional 8 background information on the environment and ecology of Alexandra Fjord, see Svoboda and Freedman (1994). 2.1.2 Site Descriptions A total of six sites at Alexandra Fjord were chosen for this study and represent a range of plant community types and soil moisture conditions within a lowland polar oasis and an upland polar semi-desert (Fig. 1). The lowland polar oasis sites were named according to the dominant plant type, while the upland polar semi-desert sites were named according to their parent material. Physical and biotic characteristics of each community type have been described in detail by Muc et al. (1989) and Muc et al. (1994). In brief, the four polar oasis sites located within the lowland (Fig. 1) had relatively high species diversity and almost complete canopy cover of low shrubs, forbs and/or grasses. On the hyrdic end of the soil moisture spectrum, the Sedge Meadow site was labelled [S-CP-DS], which stands for sedge (S), cushion plant (CP), dwarf shrub (DS). The remaining three lowland polar osis sites generally ranged from moist-mesic to mesic-mesic to xeric-mesic. The Cassiope Heath site was labelled [CP-DS(C)], which stands for cushion plant (CP) and dwarf shrub (DS), in which the dominant dwarf shrub was Cassiope tetragona (C). The Dryas Heath site [CP-DS(D)], was labelled similarly to the Cassiope Heath site, with the exception that the dominant dwarf shrub was Dryas integrifolia (D). Last was the Deciduous Dwarf Shrub (DDS)-Graminoid [DDS-G] site, where "G" stands for graminoid. In the remaining upland polar semi-desert sites, bare-ground dominates, despite marginally higher plant cover than the adjacent polar desert! These sites included the polar semi-desert (PSD) Granite site [PSD-G] and the polar semi-desert Dolomite site [PSD-D]. The Sedge Meadow site [S-CP-DS] is characterised by hydric soil with standing or slowly flowing water throughout the growing season in a matrix of hummocks and hollows dominated by sedges and grasses, but moss cover is also high (Henry et al., 1990; Muc et al., 1989). On hummocks, Dryas integrifolia and Salix arctica are also relatively abundant. The Cassiope Heath [CP-DS(C)] and Dryas Heath [CP-DS(D)] sites are dominated by the evergreen shrub Cassiope tetragona and the semi-deciduous shrub Dryas integrifolia, respectively. The soils at each site are mesic and moderately-drained. The Deciduous Dwarf Shrub-Graminoid [DDS-G] site is dominated by Salix arctica, Luzula confusa, Poa arctica, and Festuca brachyphylla, and soils are well-drained, xeric-mesic (Muc et al, 1989). The soils of all three mesic sites, including CP-DS(C), CP-DS(D) and ' ' 9 DDS-G, are described as poorly developed Static Cryosols (Muc et ai, 1994). In the adjacent upland polar semi-desert, the PSD-G site has granitic parent material, and immediately east the PSD-D site has dolomitic parent material (Fig. 1). At this polar semi-desert location, soil moisture is generally xeric, but the PSD-G site has slightly higher soil moisture conditions and pH ranges from 4.9 - 5.5 (Stenstrom et al, 1997). The PSD-D site has higher pH than the PSD-G site (pH 7.9) (Muc et ai, 1989), which is attributable to the composition of the parent material. Vascular plant cover in the polar semi-desert is sparse, with occasional patches of vascular plants surrounded by largely bare-ground. Species most common to the PSD-G site include Salix arctica, with only occasional patches of Dryas integrifolia. At the PSD-D site Saxifraga oppositifolia, Dryas integrifolia and Salix arctica occur. Draba spp., Papaver radicatum, Pedicularis spp., Luzula spp., Festuca brachyphylla and Poa arctica are scattered sparsely throughout both communities, but are more common at the PSD-G site. 2.1.3 Experimental Design This study is part of a more comprehensive network of Arctic and Alpine researchers working collaboratively as the International Tundra Experiment (ITEX), to study the impacts of climate change on tundra ecosystems. As part of ITEX, open top chambers (OTCs) were permanently installed in six plant communities along a soil-moisture gradient at Alexandra Fjord at the beginning of the growing season in June of 1992. OTCs were left out year round since 1992 to passively warm air and soil temperatures 1 - 3°C during the growing season (Marion et al., 1997), which is within G C M predictions for the Arctic (Maxwell, 1992). Warming and control plots were established using a randomized block design (n = 10-18 plots per site), where each site (block) represented a different plant community. In each site, six (n = 6 plots) of the randomly located control and warmed plots were sampled, except at the Cassiope Heath site, where n = 4 for control and warmed plots. Warming chambers were made of six Sun-Lite® HP translucent fibreglass sheets (1.0 mm thickness) with high light transmittance (85 - 90%) in the visible spectrum, and were manufactured by Solar Components Corp. Each sheet was 0.5 m high, and was angled at 60° from horizontal to form a hexagon measuring 1.5 m across the top, covering 1.8 m (Fig. 2). The plot area used for monitoring was 1 m 2 nestled within the centre of the OTC. Control plots were delimited to prevent trampling, and their dimensions were approximately that of the 10 standard OTCs (1.8 m ). For detailed analysis and discussion of OTC design and application considerations refer to Marion et al. (1997) and Hollister and Webber (2000). Hobo® loggers recorded air temperature every 8 minutes at 10 cm above the soil surface in designated warming and control plots at the Sedge Meadow (n = 3 plots) and Dryas-Heath (n = 3 plots) sites. Thermistors connected to Pocket Loggers measured soil temperatures at these same two lowland sites. Copper-constantan thermocouples connected to a CR10 data logger (Campbell Scientific) measured and recorded the air, surface and soil temperatures in a set of experimental plots at the Cassiope Heath and Deciduous Dwarf Shrub-Graminoid sites. A similar instrument design was used to measure soil (-10 cm) temperature at the polar semi-desert sites. Some of the standard OTCs at the PSD-D site were replaced with smaller models to reduce wind damage in 2000 (Henry, pers. comm.). The smaller OTCs were 0.3 m high and 0.5 m across the top, covering 0.8m2, but otherwise were exact replicas of the larger OTCs. Snowmelt across all sites was recorded as the day of the year when a plot was 95% snow-free. Mean dates of snowmelt were calculated as an average of the day of the year. Figure 2. Open top chamber (OTC) used to simulate climate warming. Six translucent fibreglass sheets with high solar transmittance in the visible spectrum are 0.5 m high and are angled at 60° from horizontal. Sample plots (1 m2) were nestled within the OTC. (Photo courtesy of G. Henry). 2.1.4 Target Species A number of criteria were considered important when selecting target species, but most important was the ability to reproduce by seed. All species selected for this study are known to 11 reproduce sexually; however, some species inherently allocate more resources to the production of seed than others (Billings, 1987; Grime, 1977), and this may vary according to micro-site conditions (Bliss, 1956; Grime, 1977). Secondary criteria included the need to use plants that were relatively abundant, widely distributed, and occurred in as many of the sampled plant communities as possible, although few species occur at all sites. Finally, target species from a range of different growth forms were selected, given that plant response to warming differs among growth forms (Arft et al., 1999). Growth forms can be defined using a range of key characteristics, and in this study groupings were based first on woody tissue presence (shrubs) or absence (forbs/graminoids), and then on reproductive morphology for herbaceous (non-woody) plants, which included dicots (forbs) or monocots (graminoids). Detailed descriptions of target species are included in Aiken et al. (1999) and Porslid and Cody (1980), but the overall habit of the target species is low, with average heights around 15 cm for most species and maximum values around 25 cm (Bliss, 1956). 2.2 Measures of Reproductive Effort and Success (Biomass and Germination) 2.2.1 A boveground Biomass ' Details of aboveground plant biomass harvests are described in greater detail below. Generally aboveground biomass of target species was harvested at peak production during the growing season, between late July and early August 2004 from warmed and control plots in each of the six sites at Alexandra Fjord. Two biomass samples per species were collected haphazardly (non-random, but without intentional bias) from what appeared to be different individuals or genets in each control and warmed plot. An "individual" was identified as all aboveground biomass that was obviously associated with a single stem that issued from the ground (Maessen et al., 1994). Samples were separated by plant component into labelled paper envelopes and then air-dried prior to oven drying at 65°C for 48 hours in a drying oven. Plant components measured for biomass included the following categories and descriptions: wood biomass (WB), which included all woody material from the terminal bud of a flowering branch back to the first major node or branch, excluding any flower material; annual vegetative biomass (AVB), including all above-ground tissues that were not sexual structures or woody; flower biomass (FB), including flowers and associated flower structures; current year fascicle 12 (CYF) biomass, the current year's stem, and photosynthetic bract (PB) biomass, including green bracts on the flowering stem. Current year fascicle and photosynthetic bract biomass are produced only by Salix arctica. Annual vegetative biomass of grasses only included single stems from separate clumps whenever possible, and where multiple stems were present from a basal roseatte, such as with Papaver radicatum, all stems and associated flower biomass was harvested. Seed biomass (SB) was obtained by haphazardly removing sub-samples of approximately 50 seeds of each target species from sample packets with fresh and over-winter harvest collections (Section 2.2.2, Seed Harvests). Each sub-sample contained seeds from only one species and plot for a given treatment and site (total seeds = -50 seeds x 1 target species x plot x treatment x site). Seeds lacking endosperm were excluded whenever possible. These seed sub-samples were then dried and weighed using the same protocols as with the other biomass components. Seed biomass included seeds and any attached protective material or dispersal mechanisms such as awns or perigyna, with the exception of Luzula spp. in which perigyna were excluded. Biomass was weighed using an O H A U S Adventurer™ analytical balance to an accuracy of ±1 ug)-2.2.2 Seed Harvests Seeds attached to parent plants and produced during the growing season (June - August, 2004) were harvested from control and warmed plots in each of the six plant communities in mid-August, 2004, as close to the time of full seed maturity as possible. Two seed heads of each target species from what appeared to be different genets were collected and placed into labelled paper envelopes. These were dried for one week in the field lab (~ 25°C), placed in a fridge at approximately 1°C for two weeks, and then exposed to a one-month stratification period at -20°C. Seeds from this harvest were labelled F2004, with " F " indicating that harvested seeds were fresh, rather than over-wintered seed. Serotiny is defined by Lamont (1980) as the seed crop that is retained on the mother plant beyond seed maturity in any given year. Seed heads with serotinous over-wintered seed were harvested just after snowmelt (mid-June) in 2004 and 2005 to provide information about the 13 I effects of warming on seeds exposed to natural vernalisation, as well as other influences associated with over-wintering. In most cases, because of the reduced number of over-wintered seeds that remained in the aerial seed bank, multiple seed heads were collected and placed in labelled paper collection packets. Over-wintered seeds produced in the 2003 growing season were harvested in 2004. These seeds were stored for approximately 75 days at -1°C in the field, and then for one month in a freezer at -20°C, to provide the same treatment as was applied to fresh seeds, as well as minimize seed respiration. Seeds from this harvest were labelled OW2003, with " O W " indicating that harvested seeds had over-wintered, and therefore had been exposed to natural stratification conditions. Over-winter seeds produced during 2004 were harvested in 2005. These seeds were stored for one month in the field at approximately -1°C, and then an additional month in a freezer at -20°C, for the same reasons as with the OW2003 collection. This seed harvest was labelled OW2004. Differences in storage treatment among the two harvest years arose out of logistical constraints, but all seed harvests were exposed to a one-month stratification period at -20°C, as discussed previously. 2.2.3 Germination Trials In the germination trials, sub-samples of approximately 50 seeds of each target species were selected haphazardly (non-random, but without intentional bias) from sample packets of the F2004, OW2003, OW2004 and international (Section 2.2.4, Regional Variability in Germination) harvest collections. Sub-samples were then placed onto moist filter paper in separate 90 mm diameter Petri dishes. Each Petri dish contained seeds from only one species and plot for a given treatment and site (total seeds = -50 seeds x 1 target species x plot x treatment x site). Seeds clearly lacking endosperm were not included in the sub-samples. Seeds were germinated under optimal germination conditions in the greenhouse at the University of British Columbia, with temperatures from 20 - 27°C and a 24 h photoperiod using full spectrum lights (600 watt, 90,000 lumens). The germination trials ran for 35 days (Baskin & Baskin, 1998), during which time the filter paper was kept moist and Petri plates were rotated to minimize systematic bias. Petri plates were checked every three days and germinants were discarded after being counted. 14 2.2.4 Regional Variability in Germination Dryas integrifolia and D. octopetala are two circumpolar con-specifics, which hybridize easily in the Eastern Arctic where the two species were once sympatric (Hulten, 1968). Seed heads • with fresh seed were collected at the end of the 2004 growing season by collaborative I T E X partners from un-treated (control) locations at Alexandra Fjord (Ellesmere Island), Thule (Greenland), Latnjuajare (northern Swedish Lapland), and Audkuluheidi (Iceland). This work was intended to help establish the influence of regional differences on the germination of Dryas integrifolia and D. octopetala, in the absence of warming. The difference in germination between these two species was also evaluated to determine the appropriateness of assumptions about the germination of these two species as conspecifics. Dryas octopetala was collected at locations within Iceland and Sweden, while D. integrifolia was harvested from Greenland and the lowland polar oasis sites at Alexandra Fjord (Ellesmere Island). In the northern Swedish Lapland, samples were collected from a Dryas meadow community (68° 21 'N, 18° 30'E), which is described as sub-arctic alpine (1000 m.a.s.l.) with dry soils around pH 5.0. At the Iceland site (65° 14'N, 19° 43'W), another low arctic site, samples were collected from 490 m.a.s.l. from gently sloping dry gravel flats, with poorly developed soil overlying glacial til l (pH ~ 7). The site in Greenland (76° 29 'N, 68° 25'W) was a polar semi-desert at 232 m.a.s.l., with growing season mean volumetric water content (VWC) of approximately 20% and soil pH 6.5. Seed heads were harvested, stored, processed and germinated using the same methods as were applied to the F2004 collections, with only minor variations in the time between harvest and stratification treatment due to shipping delays. 2.3 Statistical Analysis 2.3.1 General Criteria Data were analysed using SAS (SAS Institute Inc. 1999, version 8.2) and Excel was used for graphical analyses. Graphical analyses showed sample means with standard deviation (SD) whenever appropriate. In all cases a = 0.05 and p-values were considered significant when p < 0.05. Data were removed from statistical analysis where n < 3 plots for a given species at any site, and where no germination occurred or all values for a species at any site were all zero; however, these data were included in graphical analyses. 15 2.3.2 Parametric Tests The analysis of ecological data typically requires the application of either parametric or non- 0 parametric statistical tests. While parametric tests are preferred for their power to detect important differences between or among data sets, non-parametric tests must be applied where the data do not meet the necessary assumptions for parametric statistics (normal distribution and heteroscedastic). The use of either parametric (GLM) or non-parametric (logistic regression, Genmod) statistics were identified in the text of the results section. 2.3.2.1 G L M and Tests for Normality and Heteroscedasticity Shapiro-Wilk, Kolmogorov-Smirnov, Cramer-von Mises and Anderson-Darling were used to test for normal distribution, and homogeneity of variance was determined using Bartlett and Levene tests. The Bartlett test was chosen for its documented success with normally distributed data (Conover.e? al, 1981) and the Levene test is reported to be quite robust to non-normality, and has high power (Conover et al, 1981). When assumptions of normality and heteroscedasticity were met, a parametric Generalised Linear Model (GLM) was applied to the data to test for differences in reproductive effort and success. 2.3.2.2 G L M and Transformations If the assumptions of G L M were not initially met, a range of transformations were applied to the data in an effort to achieve normality and heteroscedasticity. These included log, exponential, squared, square root, reciprocal and rank transformations. The results of rank transformed data were only accepted if all others transformations failed to produce acceptable p-values for normality and heteroscedasticity. The transformation that produced the best p-values in subsequent tests for normality and heteroscedasticity was then applied and the data were re-tested for differences in reproductive effort and success using GLM. 2.3.3 Non-Parametric Tests Data that failed to meet the assumptions of a G L M and could not be successfully transformed to meet these assumptions were tested using non-parametric statistical tests. These tests are maximum likelihood estimation techniques that are described in the following sub-sections. 16 2.3.3.1 Logistic Regression Logistic regression is a maximum likelihood estimation method that was applied to binomial data, such as cumulative germination, where percentage data were arranged as a fraction: the total number of germinants divided by the total number of seeds in a sub-sample/Petri dish (~ 50 seeds). 2.3.3.2 Genmod Genmod is a maximum likelihood estimation that was applied in two cases: either (1) model problems were identified in logistic regression, or (2) data were not binomial, in addition to not meeting the assumptions for a GLM, even after applying transformations. In the first case, the link function was "logit", to specify that the data were binomial, and "dist=binomial" specified a binomial distribution. Genmod uses a different search algorithm than GLM, and in the second case, when analysis of count data using GLM was unsuccessful, Genmod was used with either a link function "log" or "pow(-l)" (inverse power) and a poisson distribution with the link=log or a gamma distribution with link=pow(-l) to analyse the data. Gamma distribution has been shown in some cases to describe certain biological features, such as plant heights, better than single- or double-normal distributions, and does not produce unrealistic modeled values (negative values) (Barthram et al, 2005). 2.3.4 Analytical Procedure In addition to testing for treatment (warming) effects, owing to differences among plant communities along the soil moisture gradient, the influence of site was also tested, as well as the interactions between site and treatment. This was achieved by blocking data by site (habitat) to test for treatment (control vs. warming) effects on seed germination and biomass. ,In addition, interactions between these two variables (site*treatment) were also tested. Interactions were significant when p < 0.05. The differences of least square means (LS Means), or multiple comparisons of the interaction, were reported in the text when site*treatment interactions were identified. Separate site or treatment effects were ignored only when interactions were present. In logistic regression, site*treatment interactions were subsequently analysed on a site-by-site basis, with correction for post-comparison tests (Bonferroni). In Genmod, the differences of LS Means were assessed using pairs of means t-tests, with correction for post-comparison tests (Bonferroni). In the absence of interactions, site and/or treatment effects (p < 0.05) were 17 reported in the text. Correction for post-comparison tests (Bonferroni) was applied in GLM and Genmod, and resultant p-values were adjusted by the model. 2.3.5 Reproductive Effort (Biomass) Treatment (warming) effects on the biomass of all species combined were analysed and reported under the following category labels (Section 2.2.1): combined biomass (CB), including all aboveground woody, non-woody and sexually reproductive biomass; annual vegetative biomass (AVB); flower biomass (FB) and seed biomass (SB). Biomass data were also assessed at the species-level. At this level, additional biomass categories included the following: total biomass (TB), the sum of all biomass harvested, including all aboveground woody (where applicable), non-woody and sexually reproductive biomass, but excluding seeds, (similar to "CB" described previously); wood biomass (WB); current year fascicle (CYF) and photosynthetic bract (PB) biomass. Kendall's tau-b (Conover, 1980; Kendall, 1970) correlations were used to test for relationships between seed biomass and cumulative-germination, and GLM or Genmod (where appropriate) were used to test for differences in reproductive effort (biomass) between control and warming treatments. 2.3.6 Reproductive Success (Germination) Measures of reproductive success included cumulative percent germination, peak percent germination, and germination rate, which were subsequently tested using GLM, Logistic Regression or Genmod (Sections 2.3.3.1 & 2.3.3.2), are described here. Cumulative germination represents the germination potential within a growing season, while peak germination describes when the maximum reproductive potential is reached, and also the magnitude of that potential and the timing of seed germination in the field (Walker et al, 2006). Cumulative percent germination (C-r) was calculated as the total number of germinants (G-r) divided by the total number of seeds (ST) within all Petri dishes, and multiplied by 100: [ C T = (GT/ST)*100]. 18 Germination rate was calculated using a modified Timson's Index (Timson, 1965) ___G/t, where the sum of the germination values (EG) measured at 3-d intervals for 35 d was divided by the number of 3-d measurement periods (t) (Ungar, 1996); the maximum possible value of the TI in this study was 34. Peak germination may also represent the portion of the germinant cohort most likely to survive. Peak percent germination was the maximum percent germination value within any three-day interval over the 35-day germination trial. The change (A) in reproductive success over time was also evaluated qualitatively to determine the influence of over-wintering (F2004 vs. OW2004) and inter-annual variability (OW2003 vs. OW2004) on seed germination. These calculations were intended to clarify the magnitude and direction of changes in reproductive success in response to over-wintering, annual fluctuations in climate and environmental quality, all in addition to the effects of site and treatment. Calculations were achieved by subtracting the value of the most recent collection from the older collection. For example, to calculate inter-annual changes in cumulative germination, i f the mean value of fresh germination for a given species at a given site is 12%, and over-wintered (OW2004) seed germination of the same species from the same site is 4% then, A = 4%-12% = -8%. The resultant delta value indicates that over-wintering reduced seed viability by 8%. 19 3.0 RESULTS 3.1 Environmental Data In 2004, the mean date of snowmelt was generally earlier in OTCs relative to control plots (Fig. 3), indicating that the growing season was extended under warming conditions. Air temperature data were collected from the Sedge Meadow and Dryas Heath sites by hobo loggers in three plots (n = 3 plots) at each site from control and warming treatments. These data were averaged by day and are presented in Figure 4. At the Sedge Meadow site the growing season mean temperature was 1.5°C (± 0.82 SD) higher in the warming treatment. At the Dryas Heath site the growing season mean was also higher in the warming treatment, but only by 0.5°C (± 0.21) relative to the control. 167 166 165 164 163 162 161 160 159 158 157 ra a S-CP-DS Control Warmed Control Warmed CP-DS(C) CP-DS(D) Control | -Warmed DDS-G Site and Treatment Figure 3. Mean date of snowmelt (+SD) in 2004 averaged over treatment plots (n = 6-10 plots) from the lowland sites at Alexandra Fjord, including the Sedge Meadow [S-CP-DS], Cassiope Heath [CP-DS(C)], Dryas Heath [CP-DS(D)] and Deciduous Dwarf Shrub-Graminoid [DDS-G] sites. 20 0.0 -I , , , . 1 12-Jun 22-Jun 2-Jul 12-Jul 22-Jul 1-Aug 11-Aug 21-Aug Date Figure 4. ( A ) M e a n air temperature (°C) at the Sedge M e a d o w [ S - C P - D S ] and (B) the Dryas Heath [CP-DS(D) ] lowland sites at Alexandra Fjord. 3.2 Reproductive Effort: Vegetative and Reproductive Biomass 3.2.1 Sample Size The fo l lowing results are based on biomass calculations using 11 target species grouped by growth form (Table 1). A summary o f target species and associated biomass sample sizes for each species, by site and treatment (n = number o f plots per site) are provided for aboveground biomass in Append ix A . Seed biomass sample sizes are also reported in Appendix A . Overal l , the number o f control or warmed plots sampled was n < 6 plots. See Section 2.1.2 for site abbreviations. 21 Table 1. List of target species and their growth forms and associated sites from which each species was sampled. Species Growth form Site Dryas integrifolia Vahl. Shrub S-CP-DS, CP-DS(C), CP-DS(D), DDS-G Dryas octopetala L. Shrub Iceland, Sweden Salix arctica Pall. Shrub S-CP-DS, CP-DS(C), CP-DS(D), DDS-G, PSD-G, PSD-D Papaver radicatum A.L. de Jussieu Forb CP-DS(C), DDS-G Oxyria digyna (L.) Hill Forb DDS-G Drab a L. Forb CP-DS(C), DDS-G Festuca brachyphylla Schult. and Schult. f. Graminoid DDS-G Eriophorum angustifolium subsp. triste (Th. Fr.) Hulten Graminoid S-CP-DS, CP-DS(D) Luzula confusa Lindeb. Graminoid DDS-G Luzida arctica Blytt in Blytt and A. Blytt, Norges FI. {Luzula nivalis Spreng.) Graminoid CP-DS(C), CP-DS(D), DDS-G Carex fuliginosa Schkuhr subsp. Misandra (R. Br.) Nyman Graminoid S-CP-DS, CP-DS(D) 3.2.2 Combined Species Response Within the lowland, each of the four biomass measures from the F2004 collection, including total biomass, annual vegetative biomass and flower biomass differed among sites (p < 0.001, Genmod), but not between treatments. At all sites except the Deciduous Dwarf Shrub-Graminoid (DDS-G) site, mean total biomass, annual, vegetative biomass and flower biomass were heavier or the same weight under the control treatment; at the Deciduous Dwarf Shrub-Graminoid site, mean annual vegetative biomass was marginally, but not significantly heavier in the control relative to the warming treatment (Fig. 5A). Fresh (F2004) seed biomass showed a site*treatment interaction (p = 0.0396, GLM), and differences between treatments were detected for the Sedge Meadow site (p = 0.0012). F2004 mean seed biomass was heavier in the warming treatment at all sites, but the largest relative difference between treatments was at the Sedge Meadow (S-CP-DS) site (Fig. 5B). The analysis of the polar semi-desert sites included combined female and male Salix arctica biomass. In the polar semi-desert, combined and 22 annual vegetative biomass differed (p = 0.0364 and p = 0.0225, respectively, G L M ) between the PSD-G and PSD-D sites. Mean values were higher at the PSD-G site (Fig. 5A). 1.40 1.20 1.00 ~Z 0-80 n | 0.60 ho 0.40 0.20 0.00 r^-i I mm (A) I ft • Total • Annual • Flower I.. I mm o O ro S-CP-DS o o CP-DS(C) ro ro o O > CP-DS(D) DDS-G Site and Treatment c o O PSD-G o O ro 5 PSD-D 0.01200 0.01000 0.00800 | 0.00600 o I 0.00400 0.00200 0.00000 (B) L • Control • Warmed S-CP-DS S-CP-DS CP-DS(C) CP-DS(C) CP-DS(D) CP-DS(D) Site and Treatment Figure 5. Mean combined species biomass +SD of A ) total, annual and flower biomass and B) fresh (F2004) seed biomass averaged over treatment plots from Alexandra Fjord. ** p < 0.05 for warming effects, site effects not shown. Sample sizes are reported in Appendix A . 23 Over-winter (OW2003) seed biomass differed between the Sedge Meadow and Deciduous Dwarf Shrub-Graminoid sites, and between treatments (p = 0.0329, p = 0.0053 respectively, G L M ) . Mean seed biomass was slightly heavier under ambient conditions at both sites (Fig. 6). 0.0014 0.0012 0.001 ~ 0.0008 1 ra o 0.0006 m 0.0004 0.0002 0 Control Warmed Control Warmed S-CP-DS DDS-G Site and Treatment Figure 6. Mean combined over-winter seed biomass +SD of species sampled in June 2004 (OW2003), and averaged over treatment plots from the Sedge Meadow (S-CP-DS) and Deciduous Dwarf Shrub-Graminoid (DDS-G) sites at Alexandra Fjord. * p < 0.1, ** p < 0.05 for warming effects, site effects not shown. Sample sizes are reported in Appendix A . 3.2.3 Shrubs - Dryas integrifolia, Salix arctica Dryas integrifolia wood biomass differed among sites (p = 0.0010, G L M ) ; however, annual vegetative biomass (square-root-transformed) did not (p = 0.0709, G L M ) . Flower biomass showed a site*treatment interaction (p = 0.0002, G L M ) , where treatment effects were only different (p = 0.0012) at the Sedge Meadow site. Mean flower biomass was heavier in the control treatment at this site (Fig. 7A). Only F2004 seed biomass showed treatment effects (p = 0.0100, G L M ) . Mean seed biomass (F2004) was consistently heavier under the warming treatment at all sites (Fig. 7B). OW2003 D. integrifolia seed biomass was not affected by warming (p = 0.0699, G L M ) , but mean values were consistently higher under ambient conditions (Fig. 8). 24 0.110 0.070 (A) • FB 0.090 • AVB <§ 0.050 - * * • • T '-MUUU _ . . . Control Warmed Control Warmed Control Warmed Control Warmed -0.010 1 1 1 1 1 S-CP-DS CP-DS(C) CP-DS(D) DDS-G 0.030 0.010 -0.010 CP-DS(C) CP-DS(D) Site and Treatment 0.00045 0.0004 0.00035 0.0003 s </> 0.00025 in n | 0.0002 m 0.00015 0.0001 j 0.00005 0 (B) • Control • Warmed • S-CP-DS CP-DS(C) CP-DS(D) Site DDS-G Figure 7. Dryas integrifolia mean +SD aboveground A ) wood biomass [WB], flower biomass [FB] and annual vegetative biomass [ A V B ] , and B) fresh (F2004) seed biomass averaged over treatment plots from the lowland sites at Alexandra Fjord. ** p < 0.05 for warming effects, site effects not shown. Sample sizes are reported in Appendix A . 25 0.0006 0.0005 0.0004 in | 0.0003 o n 0.0002 0.0001 Control • Dryas • Salix • Papaver • Eriophorum • Luzula spp. Warmed Control Warmed S-CP-DS DDS-G Site and Treatment Figure 8. Mean over-winter seed biomass +SD for species sampled in June 2004 (OW2003) averaged over treatment plots from the Sedge Meadow (S-CP-DS) and Deciduous Dwarf Shrub-Graminoid (DDS-G) lowland sites at Alexandra Fjord. * p < 0.1, ** p < 0.05 for warming effects, site effects not shown. Sample sizes are reported in Appendix A . Measures of female Salix arctica biomass differed among sites (log-transformed), including current year fascicle (p = 0.0017, G L M ) , photosynthetic bract (p = 0.0447, G L M ) and flower (p = 0.0002, G L M ) biomass; however, no treatment effects were detected (Fig. 9A). Female S. arctica wood biomass (rank-transformed) and F2004 seed biomass differed among sites (p = 0.0243 and p < 0.0001, respectively, G L M ) , but only seed biomass showed treatment effects (p = 0.0012, G L M ) . Mean F2004 seed biomass was substantially heavier under the warmed treatment at the Sedge Meadow site (Fig. 9 B). Over-wintered (OW2003) S. arctica seed biomass showed significant (p = 0.0531) site, but not treatment effects, and mean seed biomass was heavier at the Sedge Meadow site (Fig. 8). 26 0.400 0.000 CP-DS(C) CP-DS(D) | DDS-G Site and Treatment 0.0012 0.001 0.0008 H <2 0.0006 DO 0.0004 0.0002 (B) • Control BWarmed S-CP-DS CP-DS(C) CP-DS(D) DDS-G Site PSD-G PSD-D Figure 9. A) Female Salix arctica mean +SD aboveground wood biomass [WB], current-year-fascicle [CYF] , photosynthetic bract [PB] and flower biomass [FB], and B) fresh (F2004) seed biomass averaged over treatment plots from the lowland and polar-semi desert sites at Alexandra Fjord. * p < 0.1, ** p < 0.05 for warming effects, site effects not shown. Sample sizes are reported in Appendix A . 27 In addition to testing for site and treatment effects, data were also organized into two groups: the first group encompassed the lowland sites at Alexandra Fjord, including Sedge Meadow (S-CP-DS), Cassiope Heath (CP-DS(C)), Dryas Heath (CP-DS(D)) and Deciduous Dwarf Shrub-Graminoid (DDS-G) plant communities, and the second group represented the two upland, polar semi-desert sites, PSD-G and PSD-D. This division was made on the assumption that environmental conditions affect plant biomass, and that environmental conditions of sites within the polar semi-desert and lowland habitats had a higher degree of intra- rather than inter-habitat similarity in environmental conditions affecting plant biomass. In this case, female S. arctica biomass from the lowland showed similar patterns to the prior analysis: current year fascicle (square root transformed), flower biomass (log transformed), wood biomass (rank transformed) and F2004 seed biomass (rank transformed) differed among sites (p = 0.0115, p < 0.0001, p = 0.0250 and p = 0.0038, respectively, GLM), but not treatment (p > 0.05). Only fresh seed biomass showed treatment effects (p = 0.0080, GLM). In the polar semi-desert, female S. arctica photosynthetic bract (log-transformed) showed a site*treatment interaction (p = 0.0377). At the PSD-G site mean photosynthetic bract biomass from ambient conditions was heavier than from the warming treatment (0.018 g ±0.024SD vs. 0.009 g ±0.011SD), but differences of least square means showed no effects (p = 0.0986, GLM). Male Salix arctica biomass data were also grouped into lowland and polar semi-desert categories for analysis. Male S. arctica wood biomass (rank-transformed), current year fascicle (log-transformed) and flower biomass from the lowland were variable by site (p = 0.0015, p 0.0004, p = 0.0054 respectively). Wood biomass showed a site*treatment effect (p = 0.0312), but differences in LS Means showed no treatment effects at any of the sites, or site effects for either the control or warming treatment. In the polar semi-desert, current year fascicle biomass differed between sites (p = 0.0409), and mean current year fascicle biomass was higher at the PSD-G site (Fig. 10). 28. 0.250 0.000 Control Warmed S-CP-DS Control |Warmed Control Warmed Control [Warmed Control (Warmed CP-DS(D) DDS-G PSD-G PSD-D Site and Treatment Figure 10. Male Salix arctica mean +SD aboveground wood [WB], current year fascicle [CYF] , and flower [FB] biomass averaged over treatment plots from the lowland and polar-semi desert sites at Alexandra Fjord. Sample sizes are reported in Appendix A . 3.2.4 Forbs - Papaver radicatum and Oxyria digyna Papaver radicatum F2004 seed biomass at the Deciduous Dwarf Shrub-Graminoid site differed between treatments (p = 0.0102, G L M ) , and mean seed biomass from the warming treatment was heavier (Fig. 11). Over-winter (OW2003) P. radicatum seed biomass did not differ between treatments (p = 0.8895). Experimental warming effects did not affect F2004 Oxyria digyna seed biomass (p = 0.0928) (Fig. 11). 29 0.00120 • Control • Warmed 0.00100 s en E o 3 CI 0.00000 0.00040 0.00020 0.00060 0.00080 Papaver radicatum Oxyria digyna Species Figure 11. Mean fresh (F2004) seed biomass +SD for Papaver radicatum and Oxyria digyna averaged over treatment plots from the Deciduous Dwarf Shrub-Graminoid site at Alexandra Fjord. * p < 0.1, ** p < 0.05 for warming effects, site effects not shown. Sample sizes are reported in Appendix A . 3.2.5 Graminoids - Festuca brachyphylla, Eriophorum angustifolium subsp. triste, Luzula spp. and Carex misandra Festuca brachyphylla flower biomass from the Deciduous Dwarf Shrub-Graminoid site was greater (p = 0.0267, G L M ) under the warming versus control treatment (0.023 g ±0.0062SD vs. 0.015 g ±0.0045SD, respectively) (Fig. 12A). Eriophorum angustifolium subsp. triste flower biomass differed (p = 0.0048) between the Sedge Meadow and Dryas Heath sites, but not between treatments. Fresh (F2004) E. triste seed biomass did not differ between treatments (p = 0.0871), although mean values appeared heavier under the warming treatment (Fig. 12B). 30 0.3 0.25 0.2 in I 0.15 E o 0.1 0.05 • Control • Warmed DDS-G S-CP-DS CP-DS(D) S-CP-DS \ CP-DS(D) Festuca Eriophorum triste Carex misandra brachyphylla Site and Species A CP-DS(C) | CP-DS(D) DDS-G Luzula spp. 0.00040 0.00035 -| 0.00030 w 0.00025 in | 0.00020 i§ 0.00015 | 0.00010 0.00005 0.00000 DDS-G CP-DS(D) S-CP-DS CP-DS(D) Festuca Eriophorum Carex misandra brachyphylla triste • Control • Warmed ** L i DDS-G Luzula spp. Site and Species Figure 12. Festuca brachyphylla, Eriophorum angustifolium subsp. triste, Carex misandra, and Luzula spp. A) mean flower biomass +SD, and B) fresh (F2004) seed biomass for treatment plots from lowland sites at Alexandra Fjord. * p < 0.1, ** p < 0.05 for warming effects, site effects not shown. Sample sizes are reported in Appendix A . 31 3.3 Reproductive Success: Cumulative, Peak and Rate of Germination The data used to examine germination were collected from the primary study location at Alexandra Fjord, described in Section 2.1. Species-specific, spatial and temporal variability in germination were examined. Also, the inherent spatial variability in Dryas germination under ambient (control) conditions, in the absence of warming effects, was studied using data collected from four circumpolar locations. Results are based on germination trials using approximately 11 target species, and results are presented in groups by growth form (Table 1). 3.3.1 Sample Size A summary of target species and associated germination trial sample sizes from each treatment and site are provided in Appendices B and C. Both fresh and over-wintered seeds were collected from the aerial seed bank to provide a direct measure of changes in reproductive effort and success in response to warming. Over-wintered serotinous seed provided an added treatment effect of natural vernalisation, in addition to the controlled vernalisation period to which fresh seeds were exposed. See Section 2.1.2 for site abbreviations. 3.3.2 Combined Species Response Combining species response to experimental warming provides an overview of warming effects on reproductive effort and success, measured as biomass allocation and seed germination. Cumulative germination of all F2004 seeds differed by site (p < 0.0001) and treatment (p < 0.0001). Cumulative germination was higher under warmed conditions (Fig. 13A-E). At the PSD-D site all seeds failed to germinate; therefore, these data were removed from statistical analysis. Cumulative germination (log-transformed) of OW2003 seeds was not different between treatments or among sites (p > 0.05), although mean values appeared to vary by site (Fig. 14). 32 0 3 6 9 12 15 18 21 24 27 30 33 35 Time (days) Figure 13. Mean combined fresh (F2004) cumulative germination throughout the trial period (35 d). Data were averaged by plots over 3-day intervals within sites and treatments from all the lowland sites, including A) S-CP-DS, B) CP-DS(C), C) CP-DS(D), D) DDS-G and E) PSD-G sites at Alexandra Fjord. Sample sizes are reported in Appendix B. The rate of fresh (F2004) seed germination of all species combined showed a site*treatment interaction (p = 0.0008). This interaction was significant between treatments at the Cassiope Heath site (p = 0.0289). The interaction was also identified by treatment among sites: for the 3 3 control treatment, the Sedge Meadow differed from the Cassiope Heath (p < 0.0001), Dryas -Heath (p < 0.0001) and Deciduous Dwarf Shrub-Graminoid (p < 0.0001) sites, and for the warming treatment, the Sedge Meadow and Deciduous D w a r f Shrub-Graminoid sites differed (p - 0.0234). M e a n germination rate was highest at the Deciduous Dwarf Shrub-Graminoid site relative to the other sites, but under the control treatment (Appendix D)., A t this site the germination rate was lower under the warming treatment (Appendix D) ; however, germination rate was higher in the warming treatment for the other four lowland sites (Fig. 13, Appendix D) . Over-winter (OW2003) germination rate showed a site*treatment interaction (p = 0.0288), but differences in least square means were identified only between control treatments at the Sedge Meadow and Deciduous D w a r f Shrub-Graminoid sites. O W 2 0 0 4 germination rate o f Dryas integrifolia did not differ among sites or between treatments (p < 0.05). A) S-CP-DS Control —1—Warmed / " _ _ —— , A ~ " 12 15 18 21. 24 27 30 33 ' 35 Time (days) 30 _ 25 | 20 to c "E ai 15 B) DDS-G -Control 1 -Warmed • 12 15 18 21 24 27 30 33 35 Time (days) Figure 14. Combined mean over-wintered cumulative germination throughout the germination trial o f seeds sampled in June 2004 (OW2003) at the A ) S - C P - D S site and B ) the D D S - G site. Data were averaged by plots over 3-day intervals within sites and treatments. Sample sizes are reported in Appendix B . 3.3.3 Shrubs - Dryas integrifolia, Salix arctica Fresh (F2004) Dryas integrifolia cumulative germination showed site and treatment effects (p = 0.0198 and p < 0.0001, respectively, Genmod). Fresh seed germination was enhanced by warming (Fig. 15). For this analysis, only the Cassiope Heath and Sedge Meadow sites were tested since all seeds from the control treatment failed to germinate. Cumulative germination o f over-winter seeds harvested in 2004 (OW2003) showed a site*treatment interaction (p < 0.0001, Genmod), which was identified between treatments at the Sedge Meadow site (p = 0.0090). 34 Cumulative germination was also enhanced by warming at this site (Fig. 16). Over-winter seeds collected in 2005 (OW2004) showed no site or treatment effects (p > 0.05, log-transformed, GLM). Inter-annual variability and over-wintering effects on the cumulative germination of Dryas integrifolia from all collections (OW2003, OW2004 and F2004) was analysed by site. At the Sedge Meadow site, warming enhanced cumulative germination (p = 0.0462, square root transformed, GLM) (Tables 2 & 3, Figs. 15A & 16A). At the Deciduous Dwarf Shrub-Graminoid site, inter-annual variability was indicated in a year*treatment interaction (p < 0.0001, Genmod). Differences in least square means were between over-wintered seed collections (OW2003 vs. OW2004) from ambient conditions (p < 0.0001). An over-wintering effect was indicated in the differences in cumulative germination under the warming treatment between fresh and 2003 over-wintered seed (F2004 vs. OW2003, p = 0.0109) and between fresh and 2004 over-wintered seed (F2004 vs. OW2004) (p = 0.0003). At the Cassiope Heath site, there was a year*treatment interaction (p = 0.0111, rank transformed, GLM), but no over-wintering effects were detected in the differences in LS Means (p > 0.05). At the Dryas Heath site, a year*treatment interaction (p < 0.0001, Genmod) also indicated an over-wintering effect. At this site, differences in LS Means (p < 0.0001) were identified between warmed fresh (F2004) and over-wintered seeds produced in 2004 (OW2004) (Table 2 vs. 3, Fig. 15C vs. 16E). Table 2. Mean fresh (F2004) cumulative seed germination (%) ± SD from Alexandra Fjord. Germination was averaged over plots by species, treatment (C = control, W = warming) and site. ** p < 0.05 for warming effects, site effects not shown. Sample sizes are reported in Appendix B. Dryas Salix Papaver Oxyria Draba Festuca Eriophorum Luzula Treatment integrifolia arctica radicatum digyna spp. brachyphylla angustifolium spp. S-CP-DS C 1±2 63±30 4±6 \v •; 24"4 2i. • ;, 76" : 14 : ,- . r 24"±19 CP- c 2±5 42±17 28±6 0±0 32±22 DS(C) \v . 34"±23 92" : 5 " 11::: 12 3±2 , • 54±47 ' CP- c 0±0 42±22 13±7 39±25 DS(D) 31**±24 -. 79**: 23 24"±14 24±26 DDS-G c 0±0 58±17 27±20 77±20 55±35 1±1 0±0 W'':'}:••' -~' 33": 27 78"; 20. .. 55.-25 i 62.:.27 68±41 19"±14 18±7. PSD-G c 3±5 w - .» 5" 19 . : PSD-D c 0±0 ^ O l O l M ; - ~J . . : ' • - : Y . , -A 35 The germination rate of fresh (F2004) D. integrifolia seeds was different between treatments (p = 0.0229, GLM, log-transformed). Mean germination rate of warmed seeds was higher for the duration of the trial, and persisted longer relative to the control (Appendix D, Fig. 15). Similar trends were observed in the germination rate for over-wintered (OW2003 & OW2004) D. integrifolia (Appendix D, Fig. 16), but no differences were detected (p > 0.05). Table 3. Mean over-winter cumulative seed germination (%) ± SD for species sampled in June 2004 (OW2003) and June 2005 (OW2004) from Alexandra Fjord. Germination was averaged over plots by species, treatment (C = control, W = warming) and site. ** p < 0.05 for warming effects, site effects not shown. Sample sizes are reported in Appendix B. Treatment OW2003 OW2004 'Dryas integrifolia Salix arctica Eriophorum angustifolium Luzula spp. Dryas integrifolia S-CP-DS C 17±13 0±0 12±16 w" --^ • 28*":2l 0:0 •'. CP-DS(C) c 9±11 CP-DS(D) c 8"±18 . w DDS-G c 42±18 11±10 5±10 0±0 ' ; 21 : 19: -.7: 10 • . Peak germination of fresh (F2004) Dryas integrifolia seed (Table 4) did not differ among sites or between treatments (p > 0.05, GLM); however, only the Sedge Meadow and Cassiope Heath sites could be tested because of the failure of seeds from the control plots to germinate. Mean peak germination was consistently higher in seeds from the warming treatment (Table 4), although the differences were not significant. The peak germination of 2003 over-wintered D. integrifolia seeds (OW2003) showed no site*treatment interactions (p = 0.0637). OW2004 peak germination also did not differ by site or treatment (p = 0.5313). The complexity of the germination response of D. integrifolia was simplified by examining the change in cumulative germination between over-winter and fresh collections (Section 2.3.6 for calculations), which helped to illustrate over-wintering effects on seed germination, in addition to warming effects. In order to determine the change in cumulative germination after seeds over-wintered, OW2004 D. integrifolia seed germination was subtracted from fresh (F2004) values. These calculations showed that cumulative germination was reduced under warming 36 • conditions, between F2004 and OW2004. In contrast, cumulative germination increased under ambient conditions (Table 5). In the warming treatment, the greatest magnitude of change was at the Dryas Heath site; the lowest was at the Sedge Meadow site. In the control treatment, changes in germination were greatest at the Sedge Meadow site, and lowest at the Deciduous Dwarf Shrub-Graminoid site, almost the reverse trend observed under warming conditions (Table 5). 50 45 40 35 30 25 20 15 10 5 -I 0 A) S-CP-DS, ' Control +-— Warmed I I I I I I I •> Ml »> <b d <1, N«> # & <? 50 -j 45 -40 -o 35 -| 30 a 25 -at a> .i 20 -ulat 15 E 3 10 -o 5 0 -- Control - Warred C) CP-DS(D) .4 | | | | | | + I I I I I I I MMM °> b °> $ # ^ ^ £ # # Time (days) 50 -45 -40 -| 35 • I 30 -1 25 -I 20 -1 15 -J 10 . 5 • 0 -O 50 45 & 25 | 20 i i s 5 0 B) CP-DS(C), ** Control —1 Warmed M 1 1 1 1 1 1 1 1 1 1 r \ J NV ty f\r <y Time (days) D) DDS-G — Control —-H— Warmed O -b <o Q, £ N« ^ ^ ^ # Time (days) Figure 15. Dryas integrifolia mean fresh (F2004) cumulative germination throughout the trial period. Data were averaged by plots over 3-day intervals within sites and treatments from the A) S-CP-DS, B) CP-DS(C), C) CP-DS(D) and D) DDS-G lowland sites at Alexandra Fjord. ** p < 0.05 for warming effects, site effects not shown. Sample sizes are reported in Appendix B. 37 C) OW2004, S-CP-DS, -Control - Warmed ui s ro r n m CM CM CM CO CO CO Time (days) D) OW2004, DDS-G - Control -Warmed 4-+ I M I +-H-f-rn N N o) n Time (days) E) OW2004, CP-DS(C) i- co in T- co m Control —*+•— Warmed 11 i 11 i ' Time (days) 50 45 35 4 0 .1 3 5 ™ .= 30 -E 25 To 1 1 5 O 10 5 0 F) OW2004, CP-DS(D) Control — + — Warmed +T+T+ CM (N O Time (days) Figure 16. Dryas integrifolia mean over-wintered cumulative germination throughout the trial period (35 d) of seeds sampled in June 2004 (OW2003) and June 2005 (OW2004) at the (A & C) S-CP-DS site, (B & D) DDS-G site, and E) the CP-DS(C) and F) CP-DS(D) sites. Data were averaged by plots over 3-day intervals within sites and treatments at Alexandra Fjord. ** p < 0.05 for warming effects, site effects not shown. Sample sizes are reported in Appendix B. 38 Table 4. Mean peak germination (%) ±SD, and mean time elapsed to day of peak germination (day) across a range of species for fresh (F2004) and over-winter (OW2003, OW2004) seeds. Percent germination values were averaged by species over treatment plots (C = control, W = warming) and within sites. ** p < 0.05 for warming effects, site effects not shown. Sample sizes provided in Appendix B. Dryas integrifolia Salix arctica Papaver radicatum Oxyria digyna Draba spp. Eriophorum triste Festuca brachyphylla Luzula spp. mean ±SD >> Q mean ±SD >. ca a mean ±SD cd a mean ±SD >^  ca • Q mean ±SD ca a mean ±SD >> ca a mean ±SD ca a mean ±SD >. ca Q Fresh 2004 S-CP-DS C 1±2 9 33±24 3 1±2 18 :'.W 14: i : 6 63: 17 3 9±10 . 18 CP-DS(C) c 1±3 6 27±19 3 n/a 6±8 12 .,W; 24±17 ' 6 68±22 * * •3- • 2±3 . 21 17±19 15 CP-DS(D) c n/a 30±17 3 6±6 15 9±10 15 w 25:IS : 6 62:23 **- -• 3 7±5 ** 15 9±14 18 DDS-G c n/a 40±16 3 10±11 15 61 ±20 6 59±34 6 Oil 15 n/a 30±26 6 60:17' , - : * * : 3 16±12 12 •43*23, 6 49±41' : 6 7±8 15 , 10±7 30 PSD-G C 2±3 6 \v 4=7 - * * 6 — j 1: PSD-D c \\ Over-winter 2003 S-CP-DS c 10±7 6 w 21 !17: 6 ; ~f:ij-: DDS-G c 10±8 9 4±6 6 3±5 27 w 14+14 6 •3- .f >..''_." 3±6 . 15 Over-winter 2004 S-CP-DS c 6±7 9 w 7X10 " 12 CP-DS(C) c 6±8 6 w 6113 - 6; ' "«?-:" ' ' ; . •' CP-DS(D) c Sill 9 Mf \ ™ - •' DDS-G c ;w : 6+10 * 2- I it:. There was a year*treatment interaction (p < 0.0001, Genmod) in cumulative germination between over-wintered seed populations of D. integrifolia (OW2003 vs. OW2004), indicating inter-annual differences in cumulative germination. These differences were identified at the Deciduous Dwarf Shrub-Graminoid site (p < 0.0001), under ambient conditions. At this site, 39 under the control treatment, cumulative germination decreased by 42% between over-wintered seed populations (Table 5). Cumulative germination in OW2004 was lower than OW2003 (Tables 3 & 5). Germination rates of D. integrifolia differed (p < 0.0001, GLM) after over-wintering under warmed conditions (p = 0.0044). Generally, conditions germination rate appeared to decrease after over-wintering under ambient, but seeds harvested from warming conditions showed an increased germination rate (Table 5). In contrast to trends for cumulative germination, germination rate appeared to be greater in over-wintered seed collections from 2003 (OW2003) than from 2004 (OW2004); however, the greatest magnitude of change was also at the Deciduous Dwarf Shrub-Graminoid site. Table 5. Change (A) in Dryas integrifolia mean cumulative seed germination (%) and germination rate at different sites within the polar oasis at Alexandra Fjord from control and warmed treatments. A germination rate was calculated using Timson's Index values (Section 2.3.6). Negative values (-) indicated reduced cumulative or rate of germination. Samples were averaged over plots within sites and treatments. Sample sizes are reported in Appendix B.' F2004 - OW2004 OW2003 - OW2004 CONTROL WARMED CONTROL WARMED Site A Cumulative germination (%) S-CP-DS 11 -5 -CP-DS(C) 7 .. •'•.':':-?3i:;.,/. CP-DS(D) 8 -31 , DDS-G 0 -18 -42 A Germination rate (Timson's Index) S-CP-DS -3.32 :: 6.86 1.23 • 1.50 ' CP-DS(C) -2.14 9.48 ; CP-DS(D) -2.14 0.09 i . DDS-G 0 7.40 . 12.56 ••' 2.62 Cumulative germination of F2004 Salix arctica seed differed among sites (p<0.0001) and between treatments (p<0.0001, GLM). Cumulative germination was higher under the warming treatment at all sites (Table 2). The change in cumulative germination between treatments was lowest at Sedge Meadow and highest at Cassiope Heath (Table 2). At the PSD-D site, all germination was 0% (Table 2) and these data were excluded from statistical analysis of site and treatment effects. Fresh S. arctica cumulative germination (F2004) at the PSD-G site was quite low relative to the lowland sites, and germination began approximately three days later; 40 x however, at the PSD-G site germination levels early in the trial were twice as high under the warmed relative to the control treatment (Fig. 17F). Over-wintered S. arctica seeds from 2003 (OW2003) did not differ in cumulative germination at the DDS-G site between treatments (p > 0.05, square-root transformed, GLM) (Fig. 17A, Table 3). 100 90 80 70 60 50 40 30 20 10 0 A) OW2003, DDS-G Control Warmed i¥!^r?tt-Hr++^^+-*-^ -H-fc+4 Time (days) 100 90 80 70 60 50 40 30 20 10 0 B) F2004, S-CP-DS I I I I I I I I 4 Control —t— Warmed ,-H-l I I I I I I I I I I I ' N I I I I I O 1> <b <* Nl< N*> n> H?" ^ # Time (days) ,++++-| | | | I I I I I I I I I +1 M I N I M U M C) F2004, CP-DS(C) * Control —I— Warmed Time (days) D) F2004 , CP-DS(D) ** + + + + I I I I I I I I I I I I I I I I I I I -Control -Warmed Time (days) E) F2004, DDS-G ** 4-I I I I I I I I I I I I I I I I I I Control Warmed t-H"+ Time (days) 100 90 c 0 inat 70 E 60 to > 50 re 40 3 E 3 30 u c 20 re 0} S 10 0 F) F2004, PSD-G " Control —+—Warmed , , , „ I I I I I I I I I I I I i t - n - w Time (days) Figure 17. Salix arctica mean seed cumulative germination throughout the germination trial period (35 d) of A) over-wintered seeds sampled in June 2004 (OW2003) and (B - F) fresh (F2004). Data were averaged by plots over three-day intervals within sites and treatments from the lowland sites at Alexandra Fjord. ** p < 0.05 for warming effects, site effects not shown. Sample sizes are reported in Appendix B. 41 F2004 Salix arctica seed germination rate was differed between treatments (p = 0.0002, G L M ) . Germination rate was greater under warming conditions at all sites except the PSD-D site, where all seeds failed to germinate (Appendix D). Cumulative germination over time is illustrated in Figure 17. There was no difference in mean germination rate of over-wintered seeds (OW2003) between treatments (p > 0.05) (Fig. 17; Table 3; Appendix D). Peak germination of F2004 seeds was consistently higher in seeds from the warming treatment (Table 4), particularly at the lowland sites, and the actual percentage of seeds germinating at the peak differed between treatments (p < 0.0001, G L M ) . 3.3.4 Forbs - Papaver radicatum, Oxyria digyna,Draba spp. Warming treatment effects on F2004 Papaver radicatum cumulative germination did not differ by treatment (p = 0.0686, G L M ) at the Deciduous Dwarf Shrub-Graminoid site (Fig. 18A). The germination rate and peak germination of fresh (F2004) and over-wintered (OW2003) seed did not differ among sites or between treatments (p > 0.05, G L M ) . At the Deciduous Dwarf Shrub-Graminoid site, germination began approximately six days earlier, and continued over a longer period of time (35 vs. 33 days) under warmed conditions. Peak germination was both higher and earlier (d. 12 vs. d. 15) under warming versus control conditions (Appendix D, Fig. 18A), but differences were not meaningful (p < 0.05). Warming appeared to decrease cumulative and peak germination of F2004 Oxyria digyna seeds relative to the control (Table 2, Table 4), but no differences were detected (p > 0.05, G L M ) . F2004 seeds of O. digyna from the warming treatment continued to germinate over a longer period of time, but germination rate was also not different between treatments (p > 0.05) (Fig. 18B, Appendix D). Cumulative germination of Draba spp. from the F2004 collection also showed no treatment effects (p > 0.05, G L M ) (Table 2, Fig. 18C& D). Peak Draba spp. seed germination (F2004) appeared higher under the warming treatment at both the Deciduous Dwarf Shrub-Graminoid and Cassiope Heath sites (Table 4), but no meaningful differences between treatments were identified (p > 0.05, G L M ) . 42 80 70 | 60 o I 50 E 0 40 cn <u £ 30 ra 1 20 3 o 10 0 A) Papaver radicatum, DDS-G * -Control -Warmed 0 3 6 9 12 15 18 21 24 27 30 33 Time (days) 70 ? 60 c o I 50 'I 40 CD I 30 n 3 § 70 C) Draba spp., CP-DS(C) Conlrol —H— Warmed -H-r-f-pH 12 15 18 21 24 27 30 33 Time (days) B) Oxyria digyna, DDS-G - Control —Warmed 9 12 15 18 21 24 27 30 33 Time (days) D) Draba spp., DDS-G 0 3 6 -Control - Warmed 12 15 18 21 24 27 30 33 Time (days) Figure 18. Mean fresh (F2004) cumulative germination throughout the trial period (35 d) for (A) Papaver radicatum, (B), Oxyria digyna, (C - D) and Draba spp. Data were averaged by plots over 3-day intervals within sites and treatments from two lowland sites at Alexandra Fjord. * p < 0.1 for warming effects, site effects not shown. Sample sizes are reported in Appendix B. 3.3.5 Graminoids - Festuca brachyphylla, Eriophorum angustifolium subsp. triste, Luzula spp. and Carex misandra Festuca brachyphylla fresh (F2004) cumulative seed germination showed treatment effects (p = 0.0143, rank-transformed, GLM); at the Deciduous Dwarf Shrub site. Cumulative germination was higher under the warming treatment (Fig! 19A, Table 2), and only two seeds of F. brachyphylla from ambient conditions germinated. Festuca brachyphylla germination rate was affected by treatment (p = 0.0140, GLM), where germination rate was enhanced by warming, but peak germination was not (p > 0.05, GLM).(Table 4). Throughout the germination trial, F2004 F. brachyphylla seeds from the warming treatment showed higher cumulative germination relative to the control, and germination persisted until the end of the germination trial period, whereas control germination did not (Fig. 19A). 43 A) F 2 0 0 4 , Festuca brachyphylla, D D S - G - Conlrol - Warmed 4.4.+.+.+•+•+•+'• •x 40 c E B) F 2 0 0 4 , Eriophorum triste ,** S - C P - D S ———Control —+—Warmed 15 18 Time (days) 21 24 27 30 33 0 4-h+r+,+.+i+rh' 0 3 6 9 12 15 18 21 24 27 30 33 T i m e (days) C ) F 2 0 0 4 , Eriophorum triste, " Control C P - D S ( D ) —H— Control ,-+-•+ . . . . . ^ 9 12 15 18 21 24 27 30 33 T i m e {days) vj, ' ...... i_u*u... -)—|—|-C P - D S ( C ) / 7" H—1—1—' — Control J 1 1—— Warmed K 12 15 18 21 24 27 30 33 T ime (days) 8 s 10 % '20 i 3 10 E ) F 2 0 0 4 , Luzula spp. C P - D S ( D ) F ) F 2 0 0 4 . Luzula spp. Control D D S - G —<—— Warmed r ( ( , , , pi-j.-1.4.4..t..*.o-.t~4--fH-HH-0 3 6 9 12 15 18 21 24 27 30 33 T i m e (days ) I 1 20 G ) O W 2 0 0 3 , Luzula spp. Control D D S - G — I — Warmed . . - ± ± * J 0 3 6 9 12 15 18 21 24 27 30 33 T ime (days) Figure 19. Mean cumulative germination throughout the-trial period (35 d) from fresh (F2004) seeds of (A) Festuca brachyphylla, (B - C) Eriophorum triste, and Luzula spp. from (D - F) F2004 and (G) OW 2003. Data were averaged by plots over 3-day intervals within sites and treatments at Alexandra Fjord. ** p < 0.05 for warming effects, site effects not shown. Sample sizes are reported in Appendix B. 1 44 ( 3.4 Spatial Variability in Germination of Dryas spp.: International Comparisons When testing for spatial variability in germination of Dryas spp., these data failed to meet the assumptions for G L M , logistic regression or Genmod, even when transformations were applied. Results of G L M are reported here, since this test is considered the most robust to deviations from normality and heteroscedasticity, but should be interpreted with caution. Dryas integrifolia and D. octopetala cumulative seed germination from control plots in Greenland, Iceland, Sweden and Alexandra Fjord (F2004) differed among sites (p < 0.0001, GLM)(F ig . 20). Harvest dates for D. integrifolia seeds were on August 18, 2004 from Greenland, and around August 08, 2004 from Alexandra Fjord. Dryas octopetala seeds were harvested in Iceland on September 22, 2004 and on August 16, 2004in Sweden. 35 _ 3 0 c 2 25 CO c I 20 a 1 15 5 E 3 10 c n 3 S 5 • Dryas integrifolia • Dryas octopetala Iceland Sweden Greenland S-CP-DS CP-DS(C) CP-DS(D) DDS-G Alex Fjord (Alex (Alex (Alex (Alex Total Fjord) Fjord) Fjord) Fjord) Site Figure 20. Regional variability of fresh (F2004) mean cumulative seed germination (+SD) of Dryas integrifolia and D. octopetala from ambient conditions. Values were averaged over plots within sites. "Alex Fjord Total" includes germination within all lowland polar oasis sites at Alexandra Fjord. Sample sizes are reported in Appendix B. A l l sites differed (p < 0.0001). 3.5 Reproductive Effort and Success: Correlations Between Seed Biomass and Germination When seed germination and biomass data were pooled over all species and sites, Kendall's Tau-b (T) correlations between seed biomass and cumulative percent germination were positive and different from zero in both control (p = 0.0014) and warming (p = 0.0028) treatments (Table 13). At the site-scale, the strength of pooled species correlations diminished. There was no relationship between biomass of over-winter seeds (OW2003) and cumulative germination 45 (Table 6). Correlations between fresh (F2004) Salix arctica seed biomass and cumulative germination were only different from zero (p = 0.010) under ambient conditions, when the data were pooled over all sites (Table 6). Correlations for other collections and species, including over-winter (OW2003) and international samples, appeared to be strong, but not significantly different from zero. This was likely due to low sample sizes (Table 6). Table 6. Summary of Kendall's Tau-b (x) correlations between seed biomass and cumulative germination for fresh (F2004), over-winter (OW2003) and international seed collections, and associated probabilities (* p < 0.1, ** p < 0.05) and sample sizes (n-plots). Data were pooled by species (All), or analysed separately by species, over treatment (control = C, warming = W) and site. Site names are described in Section 2.1.2. Species Site CAV T P n - plots F2004 : ' '/s'fr All All C 0.48 0.0014** 24 w 0.44 0.0028** 24 CP-DS(C) c 0.50 .0.173 6 DDS-G w 0.66 0.174 4 PSD-G c -0.82 0.221 3 w -0.50 0.48 3 Dryas integrifolia CP-DS(C) w 0.60 0.09* 6 Salix arctica All c 0.44 0.010** 20 CP-DS(C) c 0.69 0.06* 6 CP-DS(D) c -0.71 0.18 4 S-CP-DS c -0.55 0.28 4 Luzula spp. All w -0.82 0.221 3 '%<•,-. ••' OVV2003 • ::f:-W:-y- : \-All S-CP-DS c -0.66 0.174 4 DDS-G c 0.66 0.174 4 w 0.67 0.174 4 Dryas integrifolia All w 0.55 0.062* 8 DDS-G w 0.67 0.174 4 Luzula spp. A l l : c 0.80 0.126 .4 ' . '.'..... ', '. ... 7 '; International Dryas octopetala Iceland c -0.95 0.051* 4 46 3.6 Summary of Results Under warmed conditions, enhanced reproductive effort in response to warming was exhibited as greater seed or flower biomass of all genera but two, enhanced cumulative germination in five genera, higher peak germination and faster germination in three genera (Table 7). Observations of enhanced reproductive success included enhanced cumulative germination in four genera, while two genera showed higher peak germination and five germinated faster (Table 1). Table 7. Summary of sample population increases (+) or decreases (-) in reproductive effort (RE) and reproductive success (RS) in fresh (F2004) and over-wintered (OW2003, OW2004) seeds in response to warming. 'SB' seed biomass, 'FB' flower biomass, 'PB' photosynthetic bract biomass, 'n/a' not available. * p < 0.1, p < 0.05. RE RS Reproductive biomass Cumulative germination Peak germination Germination rate : • V i : • F2004 " ' " Combined response (all species) SB +" ** + n/a ** + Shrubs Dryas integrifolia SB+" FB -** + ** + Salix arctica SB+**PB-** + ** + + Forbs Papaver radicatum SB +** Oxyria digyna SB +* Draba spp. n/a Graminoids Festuca brachyphylla FB +** + + Eriophorum angustifolium subsp. triste SB + ** + + * + Luzula spp. SB +" FB +** + -Carex fuliginosa subsp. misandra ^QM:M}U' . (>v\2003 ir:::xU£M:'^fl--.: ;..7..j Combined response (all species) SB-" Dryas integrifolia SB-* ••ry.- OW2004 . .:. . Dryas integrifolia 47 4.0 DISCUSSION Primary constraints on plant reproduction in the High Arctic include low soil nutrient status, a short growing season, and low air and soil temperatures (Billings, 1987; Billings & Mooney, 1968; Bliss, 1977; Moore et al, 1998). While short-term amelioration of these constraints by experimental warming in Arctic tundra (ACIA, 2004; Anisimov et al, 2001; Suzuki & Kudo, , 1997) has shown relatively immediate changes in vegetative expansion, the expected long-term response is enhanced sexual reproduction, which wi l l affect the barren ground landscape of the High Arctic (ACIA, 2004; Arft et al, 1999; Dormann & Woodin, 2002). Changes to reproductive effort and success wil l affect the demographics of both established plant communities and bare-ground at high latitudes. These expectations are supported by experimental and observational studies showing that reproductive effort and success are strongly influenced by temperature (Bliss & Gold, 1999; Henry & Molau, 1997; Molau & Shaver, 1997; Shaver & Kummerow, 1992). For example, Molau and Shaver (1997) showed that both seed weight and germinability of Eriophorum vaginatum were enhanced by experimental warming. Bliss and Gold (1999) found that seeds produced in the warmest years of sampling germinated faster and had a higher germinability. Results of the current study showed that the reproductive effort and success of high arctic aerial seed banks were enhanced by 12 years of experimental warming, while the direction and magnitude of response varied by species, plant community and year. 4.1 Warming Effects on Reproductive Effort and Success in a Polar Oasis Despite generally low allocation to reproductive tissues in Arctic plants (Maessen et al, 1994), in this study, fresh seeds (F2004) harvested from most species exposed to experimental warming showed enhanced reproductive effort as seed biomass, and reproductive success as cumulative germination and germination rate. Reproductive effort and success were both enhanced in shrubs and graminoids in response to warming, while only reproductive effort was enhanced in forbs. Observed differences in the frequency and magnitude of response in this study and Others (Henry & Molau, 1997; Levesque & Svoboda, 1992; Molau & Shaver, 1997; Woodley & 48 Svoboda, 1994) may indicate that, under conditions of climate warming, shrubs and graminoids wi l l be the first to colonise and establish areas of bare ground via enhanced sexual reproduction, while forbs may be slower to respond or colonise. As a result of differences in species response, plant demographics would be altered under a climate-warming scenario. These responses to wanning may be explained in part by certain qualities associated with each growth form. Dwarf shrubs, such as Salix arctica and Dryas integrifolia, produce persistent woody tissue, which reduces the need for annual renewal of structural support relative to other growth forms. This feature may also act as a repository for nutrients, effectively providing a store of nutrition that can support the production of viable seed. If woody material does store nutrients that can be later utilized for sexual reproduction, it may be reasonable to expect that increases in nutrient availability associated with experimental or climate warming wi l l either result in the production of proportionately more viable seed per plant, or more consistent production of viable seed over time (diminished inter-annual variability). In this study, the graminoid Eriophorum angustifolium subsp. triste showed a positive response to warming in almost all measures of reproductive effort and success, in contrast to other graminoids tested. While structural support in graminoids is renewed annually, they typically invest quite heavily in a below-ground system of roots and rhizomes (Billings, 1987). Eriophorum is noted in particular for its ability to translocate carbohydrates from an annual root system to the over-wintering stem, as well as root growth that persists even at near freezing temperatures (Billings, 1987). This may indicate that the nutrient storage system for graminoids is similar to shrubs, with facilitated sexual reproduction, but using the root, rather than the shoot system. Since roots have both nutrient storage and acquisition capacity, graminoid response to warming may be similar to shrub response. In contrast, forb growth and reproduction relies on annual renewal. Chapin et al. (1996) found that dwarf-shrubs and graminoids with non-aerenchymatous roots were more closely linked with each other than with forbs, using cluster analysis based on a range of traits expected to influence ecosystem processes under conditions of rapid climate change in the Arctic. Species with aerenchymatous roots, such as Eriophorum or Carex, were grouped quite distinctly from all other groupings. 49 Another possibility proposed for the lack of response to warming in forbs is increased competition for resources with shrubs and graminoids (Chapin et al, 1996). Using selective plant removal experiments, Hobbie et al. (1.999) concluded that competition did not influence species response to wanning in the short term; however, this does not necessarily preclude competitive effects under long-term experimental warming. Finally, if warmer air temperatures associated with experimental warming result in increased rates of evapotranspiration and/or high soil moisture deficits (Nosko & Courtin, 1995), species that are poorly adapted to these conditions may experience additional stresses, such as low soil moisture, which may result in reduced production of viable seed. The importance of soil moisture for viable seed production by forbs is uncertain. I According to Bazzaz et al. (2000), the question of how plants will respond to different environmental pressures depends on their degree of plasticity. Significant differences in both integrated and genera-specific effects on reproductive success in this study were largely restricted to cumulative germination, with frequent enhancements in germination rate; enhancements in the timing of peak germination occurred only infrequently. Although temperature is considered an important factor for seed development (Gutterman, 2000), these results may indicate that inherent controls in the timing of seed germination are not affected by warming; instead, enhanced seed production and viability in tundra plants may be more likely to be observed as a higher proportion or abundance of germinable seed, rather than seeds that tend to germinate earlier in the growing season. The relative importance of the different measures of reproductive success used in this study has several implications for future warming research. For example, rather than considering measures related to peak germination, studies may focus on measures of cumulative germination and germination rate. As the nature of plant response to experimental warming in the Arctic . continues to shift over time (Chapin et al, 1995), the results of this current study do not necessarily preclude changes in peak germination, but indicate that these may be unlikely over a 12 y period. Also, under conditions of climate warming, increasing germination may indicate that seeds will germinate earlier, as dates of snowmelt are advanced and the growing season extended. This may improve the chances for plant colonization and establishment (Willson & 50 Traveset, 2000), as the extended growing season permits germinants to become established in the absence of shading, competitive effects and/or grazing. Temporal changes in the nature of plant response and the timing of seed germination are important for plant demographics. Future long-term warming studies will benefit from incorporating measures of germination rate as a potential reflection of changes in reproductive success. Observed differences in the date of snowmelt between treatments at any given site in this study were relatively low, with a maximum difference of 4 days at the Sedge Meadow site and a minimum difference of zero at the Cassiope Heath site. Although these differences may not appear meaningful, tundra plants are well adapted to maximize favourable growing season conditions, which means that in some species plant growth begins even before the snow cover is completely gone (Billings & Mooney, 1968). Pre-formed flower buds also reduce the time required for flowering and seed development in any given season (Sorenson, 1941), and flower and seed maturation can begin as soon as critical temperatures are achieved (Chapin & Shaver, 1985). Once temperatures are suitable for growth, tundra plants are able to fully utilize the 24-h photoperiod, which renders hourly changes in the microenvironment as more important than daily changes (Chapin & Shaver, 1985). Advanced snowmelt dates associated with warming in the Arctic are predicted to enhance soil microbial activity and the production of plant-available nutrients, as the growing season is extended (Grogan & Chapin, 2000; Nadelhoffer et al., 1992). The combined effect are expected to increase nutrient allocation to sexual reproduction, which will likely enhance reproductive success in the form of increased production of viable seed, and a higher incidence of germination and seedling establishment where competition is not limiting (Arft et al., 1999; Bliss & Gold, 1999; Henry & Molau, 1997; Sandvik & Totland, 2000; Shaver & Kummerow, 1992; Wookey et al., 1993). Observed site-specific advances in spring snowmelt associated with experimental warming, which extended the growing season and likely contributed to the production of heavier seeds with higher cumulative germination across a range of species, may confirm these predictions, but expected changes are difficult to test directly. 51 Changes in plant phenology provide critical insights into the constraints on an organism's growth (Murray & Miller, 1982), and flower phenology is reported to have the greatest effect on plant reproductive success (Molau et al, 2005; Thorhallsdottir, 1998). Experimental warming has been shown to advance early-season phenophases, such as flowering, over the short-term (Arft et al, 1999; Hollister et al, 2005; Hulten, 1968; Johnstone, 1995). In this study, enhanced reproductive effort and success in fresh seeds harvested from warming treatments may indicate that long-term experimental warming has indirectly extended late-season phenophases. This may provide further evidence that bare-ground colonization in the High Arctic wi l l proceed via changes in sexual reproduction (ACIA, 2004; Arft et al, 1999; Bliss & Gold, 1999; Dormann & Woodin, 2002; Larsson, 2002). Species that are otherwise limited by a relatively slow germination rate, observed in this study as Papaver radicatum, Draba spp., Luzula spp., Festuca brachyphylla, and Eriophorum angustifolium subsp. triste, may experience higher rates of establishment where warming results in an extended growing season and conditions favourable for the growth and establishment of germinants are prolonged. However, under warming conditions, i f germination persists late into the growing season, species slow to germinate may also be at an increased risk of mortality due to moisture stress, in sites where soil moisture supplies are restricted to early-season snowmelt, or climatic conditions associated with the onset of winter. None the less, this risk wi l l l ikely 1 remain low, since germination of most tundra species is light-limited, and light levels diminish toward the end of the growing season (Baskin & Baskin, 1998; Densmore, 1997; Heide, 2005). Although warming enhanced the reproductive effort and success of some species, the absence of a response for other species does not necessarily preclude warming effects; in particular, plants that are already light-limited due to competition. For example, diminishing light quality associated with increasing cloud cover tends to result in the cessation of plant growth and affects the production of germinable seed (Olson & Richards, 1979), and is predicted to increase with future warming in the Arctic (Chapin & Shaver, 1985). Also, freshly matured seeds of most shrub, herb and many graminoid species are non-dormant (Baskin & Baskin, 1998; Chapin & Shaver, 1985); however, some species of Carex and Festuca show some specialized germination requirements, indicating that dormancy may have affected germination success (Baskin & 52 Baskin, 1998; Chapin & Shaver, 1985). These possible counteractive effects of warming need to be tested directly. 4.2 Warming Effects on Reproductive Effort and Success in a Polar Semi-Desert Climate-mediated changes in the aerial seed banks of well-vegetated polar oases provide important information about potential seed sources for the surrounding barren landscape (Bliss, 1958; Svoboda & Henry, 1987). In the vast landscapes of polar semi-deserts, where vascular plant cover ranges from 25% to < 5% (Walker et al, 2005), experimental warming research will yield information on both a potential seed source and future dynamics associated with bare-ground colonization (Svoboda & Henry, 1987). Growing conditions in polar semi-deserts are not strongly ameliorated by the presence of established plant communities or even cryptogamic crusts (Gold & Bliss, 1995), and abiotic factors, including soil temperature (Chapin, 1983), soil texture (Sheard & Geale, 1983), and nitrogen and phosphorus availability (Miller, 1982), as well as experimental influences such as the date of seed harvests (Bliss & Gold, 1999), can determine the absolute germinability of seeds produced in these environments. Two important conclusions were derived from observed values of reproductive success at the polar semi-desert site in this study, where Salix arctica seeds collected only from the site with Granitic parent material (PSD-G site) germinated. First, site-specific differences in environmental conditions between the two polar semi-desert sites may have determined not only the extent of reproductive success, but the absolute presence or absence of it (Chapin, 1983; Miller, 1982; Sheard & Geale, 1983). The parent material at the PSD-D site was dolomitic and soils were slightly drier and more basic than the PSD-G site, where granitic parent material permitted slightly better drainage and potentially more suitable pH for plant growth. Second, even with only small increases in air and soil temperature associated with experimental warming, and possibly in combination with more favourable site-specific qualities, such as soil moisture (Gold & Bliss, 1995), soil composition and pH (Gold & Bliss, 1995), resulted in the production of more viable seed at the PSD-G site, relative to the PSD-D site. The apparently anomalous relationship between high reproductive effort (seed weights), but low reproductive success at the PSD-D site relative to the PSD-G site may be explained by the inverse relationship between soil moisture and diaspore mass (Baskin & Baskin, 1998; Dormann et al, 2002), where lighter seeds from the more moist PSD-G site with lower weights germinated, while heavier seeds from the PSD-D site failed to germinate. 53 4.3 Warming Effects on Over-wintering Seeds (F2004 vs. OW2004) While experimental warming clearly enhanced reproductive success of fresh (F2004) seeds in some species, but dependent on site, over-wintered seeds (OW2003, OW2004) responded quite differently. Over-wintered Dryas integrifolia seeds showed decreased reproductive success, but only for those seeds from the warmed treatment. Incomplete development or maturation of seed is not uncommon in the High Arctic. Bliss and Gold (1999) noted that seeds were seldom ripe by the end of the growing season in which they were produced. Bell and Bliss (1980) found that the germination of several different high arctic species ranged from 0 - 40%; low germination was attributed to the combined effects of low temperature and slow imbibition (water absorption). In addition, even if freshly matured seed of a tundra plant does happen to be viable, given the harsh conditions of the Arctic winter, the chances of successful germination in seeds that over-winter will likely be further diminished. In contrast, seeds formed in unusually warm growing seasons have been found to germinate faster and have higher germinability relative to seeds produced in other years (Bliss & Gold, 1999). In some species, over-wintering has even enhanced germination success (Humlum, 1980). Under conditions of climate warming, advanced snowmelt and phenological development are also expected to reduce the risk of late-season seed abortion, extend the growing season (Hinzman et al, 2005; Hollister et ai, 2005), and increase the proportion of viable seed for certain species (Arft et al, 1999; Hollister et al, 2005; Krannitz, 1996). In combination with indirect effects of warming, such as increased nutrient availability attributable to enhanced soil microbe activity (Nadelhoffer et al, 1992), experimental warming could theoretically be expected to result in a higher proportion of seeds achieving full maturity. This hypothesis is supported in part by evidence that reproductive success of fresh seeds from warming conditions was enhanced across a range of species. However, the question remains: Why did warmed over-wintered seeds, which were from the same population as the fresh harvest, fail to show similar responses, such as enhanced reproductive success? There may be two possible explanations for the observed decrease in germinability of over-wintered seeds (OW2003, OW2004) collected from warmed plants. First is that somehow over-54 wintering diminished seed viability, as discussed previously. The second explanation involves seed dispersal. Assuming that fully developed and matured fresh seeds subsequently dispersed as seed rain, thereby exiting the aerial seed bank, the important consequence of this may have been that the remaining aerial seed bank population, post dispersal, may have been largely comprised of seeds with proportionately lower viability relative the control population. In the control population, seeds did not mature to the point of dispersal by the end of the growing season, which meant that the remnant populations could easily have higher viability than the post-dispersal seed population from the warming treatment. In other words, observed decreases in reproductive success of over-wintered seeds exposed to warming conditions may indicate that over-wintering diminished seed viability, while seeds exposed to ambient conditions benefited from additional maturation time at the end of the 2004 growing season, and generally resulted in enhanced cumulative germination. In conclusion, the result was that proportionately higher seed dispersal in the warming treatment appeared to indicate that warming, in particular, reduced the viability of over-wintered seed, although this effect was observed in both warmed and control conditions. Selective seed predation by birds and sub-nivean fauna, or sampling bias associated with rare or marginal species that may exhibit higher or lower reproductive success as an artefact of micro-site conditions, are two additional factors that could have contributed to apparent decreases in warmed seed viability, or variability in germination responses among site. Warming-induced seed respiration may also have influenced over-wintered seed viability in warming treatments, but this seems unlikely, since seeds stored at +22°C for one year have not been found to exhibit reduced seed viability due to seed respiration (Laine et ai, 1995). Despite challenges in timing associated with obtaining fresh seed from the aerial seed bank at the end of the growing season, the challenges associated with analyzing germination data from the over-wintered aerial seed bank are much greater. On the other hand, this question would benefit from directly measuring dispersal frequency of fresh seed. 4.4 Inter-annual Variability in Reproductive Success Challenges to growth and reproduction of tundra plants include a short growing season, poor soil nutrient status, and low precipitation/plant-available soil moisture (Billings, 1987; Billings 55 & Mooney, 1968; Bliss, 1977; Moore et al, 1998). Fluctuations in climate intensify or diminish these challenges to varying degrees, and contribute to inter-annual variability in plant performance. Interactions between inter-annual variability in climate and limitations to growth and reproduction inherent in the Arctic environment can be substantial (Walker et al, 1995). For example, among-year variation in phytomass in alpine plant communities ranged from 1 5 -40% (Walker et al, 1994). In this study, inter-annual differences were only identified in seeds that developed under ambient conditions, and likely indicate that experimental warming probably diminished inter-annual variability in reproductive success. Through trade-offs, a short growing season and inter-annual variations in climate and environmental quality determine the allocation potential for sexual reproduction in any given plant (Chambers, 1995). Under conditions of climate warming, sexual reproduction may show more consistent enhancements, or reduce inter-annual variability in performance over the long-term as the growing season is extended through advanced spring snowmelt (ACIA, 2004; Anisimov et al, 2001; Suzuki & Kudo, 1997), and soil nutrient status is improved via enhanced soil microbial activity (Nadelhoffer et al, 1992; Rolph, 2003). Reduced inter-annual variability in reproductive success, in combination with decreased seedling mortality, may enhance bare-ground colonization and, possibly increase competition intensity in established plant communities such as polar oases. Over consecutive years, measures of inter-annual variability in reproductive success in response to warming wi l l help determine the accuracy of these predictions. The magnitude of inter-annual variability was also influenced by site. For example, at the Deciduous Dwarf Shrub-Graminoid site, where soil moisture conditions are mesic, over-wintered Dryas integrifolia seeds from ambient conditions germinated to nearly 50% one year, but the following year, all seeds failed to germinate. In contrast, at the Sedge Meadow site, where soil moisture conditions are hydric, the difference in germination between years was only ~ 5%. These results may indicate that plant response to warming in hydric habitats is low, but with less annual variation in performance from year to year relative to mesic habitats, which experience relatively high inter-annual fluctuations in plant response. Over the long-term, the sum total germination response in mesic habitats, and potential contributions to bare-ground colonization, may be quite high under conditions of climate warming, particularly in mast years (Callaghan & Carlsson, 1997; Willson & Traveset, 2000), when there is an especially high 56 production of seed, and when seedling survival is high (Svoboda & Henry, 1987). However, as years with low seed production cancel out the effects of high production years, the identification of clear warming trends in seed germination may be substantially diminished. In contrast, in hydric sites where germination may be low, but perhaps less affected by inter-annual changes in climate, it is possible that the long-term effects of low, but consistent germination, may produce an equal or greater contribution to plant community dynamics and colonization of bare ground. 4.5 Local and Regional Variability in Reproductive Effort and Success Site effects, and/or interactions between site and treatment, indicate that the magnitude of warming response in this study was affected by the unique biotic and abiotic characteristics of each site. Abiotic characteristics, such as soil moisture conditions, have been shown to affect plant growth and reproduction throughout the Arctic (Chapin & Shaver, 1985), as well as specifically at Alexandra Fjord (Jones, 1995). Experimental warming studies at other Arctic locations have shown that soil moisture influenced the magnitude of plant response (Chapin et al, 1996; Illeris et al., 2004). In this study, germination and biomass of fresh Dryas integrifolia seeds were enhanced by warming at the two heath sites, which have mesic soil-moisture conditions. In contrast, the magnitude of the warming response was low at the Sedge Meadow site, where soil moisture was hydric. Over-wintered seed germination shared similar patterns of response, particularly when certain effects were interpreted as seed dispersal. These results indicate that plant sexual reproduction may exhibit a greater magnitude of warming response where soil moisture conditions are mesic, while the magnitude will be lower under hydric soil-moisture conditions. Findings from this study were consistent with other experimental warming research in the Arctic, which indicated that warming effects will be less intense in areas where soil moisture conditions are saturated for much of the growing season (Nadelhoffer et al, 1992; Nadelhoffer et al, 1997). In Toolik Lake, Alaska Shaver and Jonasson (1999) attributed slow or reduced plant response to warming to the effect of underlying permafrost, which lowered soil temperatures and reduced nutrient mineralization rates. Jonsdottir et al. (2005) reported an insulating effect of moss in sedge communities that prevented deepening of the active layer in association with experimental warming. Hydric soil moisture conditions may buffer rapid 57 environmental fluctuations, resulting in a stable community that is more resistant to external changes such as climate warming (Margalef, 1968). The combination of high soil moisture conditions that limit soil heat absorption, and maintain low soil temperatures, in addition to a thin active layer (Muc, 1977), likely help to reduce the response potential of these communities, and may indicate that plant sexual reproduction will be more responsive to warming where soil moisture conditions are mesic. However, as discussed previously, high annual variability in germination in this study observed in mesic plant communities relative to communities with more hydric soil moisture conditions, may reduce these effects. In this study, certain site conditions appeared to affect plant sexual reproduction more than the date of snowmelt. For example, at the hydric Sedge Meadow site, despite a snowmelt date that was four days earlier in the warmed treatment, the magnitude of plant response in control versus warming treatments was less than at the mesic Cassiope Heath site, where the date of snowmelt did not differ between treatments. The suitability of a given species to a particular habitat also appeared to be an important influence on changes in reproductive effort and success in response to warming. For example, in hydric habitats, Dryas integrifolia is somewhat rare, and is typically found on hummocks, where soil moisture conditions are slightly moderated (Courtin & Labine, 1977; Muc, 1977). The incidence of floral abnormalities in this species, at sites with high soil moisture, is high relative to drier sites (Hart & Svoboda, 1994). This indicates that sub-optimal growth conditions for any given plant may reduce the potential response to warming. In this study, where Dryas integrifolia and other species such as Papaver radicatum were relatively abundant, cumulative germination was enhanced under warming conditions; at sites where these species were rare, the trend was reversed, and germination was higher under ambient conditions. Cumulative germination of Eriophorum angustifolium subsp. triste at both the Sedge Meadow and Dryas Heath sites was improved under warming conditions, but peak and rate of germination were higher only at the Sedge Meadow site, where species like E. triste are common (Aiken et al, 1999). Under conditions of climate warming, these results suggest that plant response may be limited where growth and reproduction are already restricted by available "safe sites" (Harper et al, 1961) or where site-specific abiotic stresses, such as high soil 58 moisture, are present. Increases in reproductive success may also be limited to dominant species, unless environmental parameters and constraints to plant growth and reproduction changes dramatically in a given plant community. An example of a situation where a dominant species may be negatively, rather than positively impacted through changes in environmental parameters associated with warming is with Carex stans. Carex starts is a species of sedge that is well adapted to plant communities with high soil moisture (Aiken et al, 1999). While results of this study suggest that hydric communities may be more resistant to environmental changes, such as increasing temperature, it is not only the possibility of increased drainage associated with melting permafrost that could impact hydric communities under conditions of climate warming (ACIA, 2004). In years with unusually warm growing season temperatures, high rates of evapotranspiration have been shown to reduce the productivity of species adapted to hydric conditions, such as Carex stans, through water stress (Nosko & Courtin, 1995). ' Differences among international locations in control D. integrifolia and D. octopetala germination suggest that the range of germination potential for these genera at a larger spatial scale is highly differentiated. Implications of this finding are that genera such as Drays spp., which exhibit regional differences in germination, may not be viable options for extending the spatial range of comparisons of plant response to experimental warming. 4.6 Correlations Between Reproductive Effort and Success Although seed biomass and cumulative germination each appeared to show increases in response to warming when tested independently, correlations were not significantly different from zero. Most likely, correlations were not detected because of low statistical power and the conservative nature'of the statistical test (Conover et al, 1981). Others have shown that seed weight is positively correlated in other species, especially above a lower threshold mass (Molau & Shaver, 1997). In addition, the process of pooling data across species, sites and treatments contributes additional variability to the data set, which tends to mask important relationships. 59 4.7 Long-term Effects of Warming Differences in the germination of Dryas integrifolia between comparable control and warmed populations provide an indication of inter-annual variability. Differences in measures of reproductive success between years were more common in seeds grown under ambient conditions, while annual differences between sample populations from the warming treatment were infrequent. These differences, which have not been observed in short-term studies, may be indicative of long-term effects of experimental warming. However, repeated measures over consecutive years will provide a more accurate indication of long-term trends. 60 5.0 CONCLUSIONS AND RECOMMENDATIONS Studies of experimental warming effects on high arctic aerial seed banks are rare, despite expectations that colonization of bare ground will proceed via changes in sexual reproduction under conditions of climate warming. Increasing plant cover will affect plant community dynamics, primary consumers and the global carbon budget. Results of this study showed that both reproductive effort and success were enhanced by experimental warming for most of the high arctic species studied, with responses dependent on initial site conditions. These results confirm observations of short-term warming studies, and indicate that changes in bare-ground colonisation will likely proceed via enhanced plant sexual reproduction, with polar oases functioning as seed source for the surrounding bare-ground landscape. In addition, plant response to warming will be enhanced or modified by local, regional and inter-annual influences, while dominant species and mesic-habitats will likely be most responsive to changing temperature. Earlier dates of snowmelt, associated with experimental warming, appear to have indirectly extended the growing season and shifted plant phenophases. These shifts appear to have enhanced reproductive effort and success by allowing more seeds to achieve full maturity and disperse. Changes in plant phenophases can be expected to indirectly affect plant community structure and the rate and extent of colonization in the barren landscapes of the High Arctic, but will be restricted where site conditions, such as soil moisture, are limiting. Phenophase shifts resulting in enhanced reproductive effort and success may increase rates of establishment for species that are normally slow to germinate, which will affect species dominance and possibly competition under a climate-warming scenario. In marginal habitats, such as polar semi-deserts and polar deserts, changes in reproductive effort and success will be slow or may not occur, but the potential for colonization from outside sources (polar oases), in combination with temperature amelioration, may enhance the potential for and rate of colonization of these bare-ground landscapes under conditions of climate warming. Reproductive effort and success will also tend to have a lower magnitude of response 61 to warming in hydric sedge meadow communities. Although, over the long-term, diminished inter-annual variability in these habitats may support steady increases in reproductive effort and success relative to habitats with more mesic soil moisture conditions, unless there are dramatic changes in biotic or abiotic factors, such as competition or drainage. Mesic habitats will exhibit variable responses according to annual changes in climate, due to comparatively higher inter-annual variability in plant response. Observations that warming reduced inter-annual variability in cumulative germination, may be further evidence that reproductive effort and success will increase under conditions of climate warming. However, plant response will also be constrained by environmental factors that limit growth and reproduction, especially in polar semi-desert environments or in habitats where plant growth and establishment is restricted on a species-specific basis. Future warming research in the High Arctic will benefit from recruitment studies that focus on in-situ germination success and seedling survival in polar semi-desert and polar desert communities, where plant competition is low but colonization potential is high. With climate warming, these investigations will be important as reproductive success is enhanced progressively over time, and conditions that contribute to high mortality rate simultaneously diminish (Callaghan & Carlsson, 1997). In addition, forecasting efforts will benefit from long-term warming studies that incorporate multi-year, multi-species studies that include a range of measurements of reproductive effort and success, in concert with micro-site measures of soil nutrients, moisture and other variables. Finally, these studies should also focus on the collection of fresh, rather than over-winter seed, collected as close to the end of the growing season as possible. Collectively, findings of the current study indicate that climate warming in the High Arctic may enhance reproductive effort and success of tundra plants; however, responses will likely be species-specific, and dependent on initial site conditions. These results may also confirm predictions that bare-ground colonisation in the High Arctic will proceed via enhanced sexual reproduction and reproductive success. 62 REFERENCES CITED ACIA (2004) Impacts of a warming arctic: arctic climate impact assessment. Cambridge University Press, Cambridge, UK. Aiken, S., Dallwitz, M., Consaul, L., McJannet, C , Gillespie, L., Boles, R., Argus, G., Gillett, J., Scott, P., Elven, R., LeBlanc, M., Brysting, A., & Solstad, H. (1999) Flora of the Canadian Arctic Archipelago: Descriptions, Illustrations, Identification, and Information Retrieval. Anisimov, O., Fitzharris, B., Hagen, J.O., Jefferies, R., Marchant, H., Nelson, F., Prowse, T., & Vaughan, D. (2001). Polar regions (Arctic and Antarctic). Cambridge University Press. Anisimov, O. & Nelson, F. (1997) Influence of climate change on permafrost in the Northern Hemisphere. Russian Meteorology and Hydrology, 5, 47-53. Arft, A.M., Walker, M.D., Gurevitch, J., Alatalo, J.M., Bret-Harte, M.S., Dale, M., Diemer, M., Gugerli, F., Henry, G.H.R., Jones, M.H., Hollister, R.D., Jonsdottir, I.S., Laine, K., Levesque, E., Marion, G.M., Molau, U., Molgaard, P., Nordenhall, U., Raszhivin, V., Robinson, C.H., Starr, G., Stenstrom, A., Stenstrom, M., Totland, O., Turner, P.L., Walker, L.J., Webber, P.J, Welker, J.M, & Wookey, P.A. (1999) Responses of Tundra Plants to Experimental Warming: Meta-Analysis of the International Tundra Experiment. Ecological Monographs, 69, 491 -511. Barthram, G , Duff, E , Elston, D , Griffiths, J , Common, T, & Marriott, C. (2005) Frequency distributions of sward height under sheep grazing. Grass and Forage Science, 60, 4-16. Baskin, C. & Baskin, J. (1998) Seeds: Ecology, Biogeography, and Evolution of Dormancy and Germination Academic Press. Bazzaz, F.A, Ackerly, D.D, & Reekie, E.G. (2000). Reproductive allocation in Plants. In Seeds. The ecology of regeneration in plant communities, (ed M. Fenner), pp. 1-29. CABI Publishing. Bell, K.L. & Bliss, L.C. (1980) Plant reproduction in a high arctic environment. Arctic and Alpine Research, 12, 1-10. Billings, W.D. (1987) Constraints to plant growth, reproduction, and establishment in arctic environments. Arctic and Alpine Research, 19, 357-365. Billings, W.D. & Mooney, H.A. (1968) The ecology of arctic and alpine plants. Biological Reviews, 43,.481-529. Bliss, L. & Peterson, K. (1992). Plant succession, competition, and the physiological constraints of species in the Arctic. In Arctic Ecosystems in a Changing Climate: An Ecophysiological Perspective (eds F. Chapin III, R. Jefferies, J. Reynolds, G. Shaver & J. Svoboda), pp. 111-133. Academic Press, Orlando, Florida. Bliss, L.C. (1956) A comparison of plant development in microenvironments of arctic and alpine tundras. Ecological Monographs, 26, 303-337. Bliss, L.C. (1958) Seed germination in arctic and alpine species. Arctic, 11, 180-188. Bliss, L.C. (1977). General summary, Truelove Lowland ecosystem. In Truelove Lowland, Devon Island, Canada; A High Arctic Ecosystem, (ed L.C. Bliss). University of Alberta Press, Edmonton, Alberta. Bliss, L.C. (1997). Arctic ecosystems of North America. In Polar and alpine tundra (ed F. Wielgolaski), pp. 551-683. Elsevier, Amsterdam, NL. Bliss, L .C, Courtin, G , Pattie, D , Riewe, R, DWA, W, & Widden, P. (1973) Arctic Tundra Ecosystems. Annual Review of Ecology and Systematics, 4, 359-399. 63 Bliss, L.C. & Gold, W.G. (1999) Vascular plant reproduction, establishment and growth and the effects of cryptogamic crusts within a polar desert ecosystem, Devon Island, NWT, Canada. Canadian Journal Of Botany, 11, 623-636. Callaghan, T. & Carlsson, B.A. (1997). Impacts of climate change on demographic processes and population dynamics in arctic plants. In Global change and arctic terrestrial ecosystems (eds M. Caldwell, G. Helmaier, O. Lange, H. Mooney, E. Schulze & U. Sommer), Vol. 124, pp. 129-152. Spring-Verlag, New York. Callaghan, T.V., Press, M.C., Lee, J.A., Robinson, D.L., & Anderson, CW. (1999) Spatial and temporal variability in the responses of Arctic terrestrial ecosystems to environmental change. Polar Research, 18, 191-197. Chambers, J.C. (1995) Disturbance, life history strategies, and seed fates in alpine herbfield communities. American Journal of Botany, 82, 421-433. Chapin, F.S. (1983) Direct and indirect effects of temperature on arctic plants. Polar Biology, 2, 47-52. Chapin, F.S., Bret-Harte, M., Hobbie, S., & Hailin, Z. (1996) Plant functional types as predictors of transient responses of arctic vegetation to global change. Journal of Vegetation Science. Chapin, F.S. & Shaver, G. (1985). Arctic. In Physiological ecology of North American plant communities, (eds B. Chabot & HA. Mooney), pp. 16-40. Chapman and Hall. Chapin, 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, 16, 694-711. Chapin III, F.S. & Shaver, G. (1985). Arctic. In Physiological ecology of North American plant communities, (eds B. Chabot & H.A. Mooney), pp. 16-40. Chapman and Hall. Conover, W. (\980).Practical nonparametric statistics, 2nd edn. John Wiley & Sons. Conover, W., Johnson, M., & Johnson, M. (1981) A comparative study of tests for homogeneity of variances, with applications to the Outer Continental Shelf Bidding Data. Technometrics, 23, 351-361. Courtin, G. & Labine, C. (1977). Microclimatological studies on Truelove Lowland. In Truelove Lowland, Devon Island, Canada: a High Arctic Ecosystem (ed L. Bliss), pp. 73-106. University of Alberta Press, Edmonton. Demchenko, P., Eliseev, A., Molkhov, I., Obukhov, A., Nechaev, V., & Velichko, A. (2001) Sensitivity of permafrost cover in the Northern Hemisphere to climate change. CLIVAR Exchanges, 6, 9-11. Densmore, R. (1997) Effect of day length on germination of seeds collected in Alaska. American Journal of Botany, 84, 274-278. Dormann, C , Albon, S., & Woodin, S. (2002) No evidence for adaptation of two Polygonum viviparum morphotypes of different bulbil characteristics to length of growing season: abundance, biomass and germination. Polar Biology, 25, 884-890. Dormann, C. & Woodin, S. (2002) Climate Change in the Arctic: Using Plant Functional Types in a Meta-Analysis of Field Experiments. Functional Ecology, 16, 4-17. Epstein, H.E., Calef, M.P., Walker, M.D., Stuart Chapin, F., & Starfield, A.M. (2004) Detecting . changes in arctic tundra plant communities in response to warming over decadal time scales. Global Change Biology, 10, 1325-1334. Epstein, H.E., Walker, M.D., Chapin, F.S., & Starfield, A.M. (2000) A transient nutrient-based model of Arctic plant community response to climatic warming. Ecological Applications, 10,824-841. 64 Eviner, V.T. & Chapin, F. (2003) A Conceptual Framework for Predicting Multiple Plant Effects on Ecosystem Processes. Annual Review of Ecology and Evolution and Systematics, 34, 455-485. Freedman, B, Svoboda, J , & Henry, G. (1994). Alexandra Fiord - An ecological oasis in the polar desert. In Ecology of a polar oasis: Alexandra Fiord, Ellesmere Island, Canada. (eds J. Svoboda & B. Freedman), pp. 1-9. Captus University Publications, Toronto. Gold, W.G. & Bliss, L.C. (1995) Water limitations and plant community development in a polar desert. Ecology, 76, 1558-1568. Grime, J. (1977) Evidence for the existence of three primary strategies in plants and its relevance to ecological and evolutionary theory. The American Naturalist, 111, 1169-1194. Grogan, P. & Chapin, F.S. (2000) Initial effects of experimental warming on above- and belowground components of net ecosystem C02 exchange in arctic tundra. Oecologia, 125,512-520. Gutterman, Y. (2000). Maternal effects on seeds during development. In Seeds. The ecology of regeneration in plant communities, (ed M. Fenner), pp. 59-84. CABI Publishing. Hansen, A.H, Jonasson, S, Michelsen, A , & Julkunen-Tiitto, R. (2006) Long-term experimental warming, shading and nutrient addition affect the concentration of phenolic compounds in arctic-alpine deciduous and evergreen dwarf shrubs. Oecologia, 147, 1-11. Harper, J. (1977) Population biology of plants Academic Press, London; New York. Harper, J , Clatworthy, J , McNaughton, I, & Sagar, G. (1961) The evolution and ecology of closely related species living in the same area. Evolution, 15, 209-227. Hart, G. & Svoboda, J. (1994). Autecology of Dryas integrifolia in Alexandra Fiord Lowland habitats. In Ecology of a Polar Oasis, Alexandra Fiord, Ellesmere Island, Canada (eds J. Svoboda & B. Freedman), pp. 145-156. Captus Press, Inc., North York. Hartley, A.E, Neill, C , Melillo, J , Crabtree, R, & Bowles, F. (1999) Plant performance and soil nitrogen mineralization in response to simulated climate change in subarctic dwarf shrub heath. Oikos, 86,331-343. Heide, O. (2005) Ecotypic variation among European Arctic and Alpine populations of Oxyria digyna. Arctic, Antarctic, and Alpine Research, 37, 233-238. Henry, G.H.R. & Molau, U. (1997) Tundra plants and climate change: the International Tundra Experiment (ITEX). Global Change Biology, 3, 1-9. Henry, G.H.R, Svoboda, J , & Freedman, B. (1990) Standing crop and net production of sedge meadows of an ungrazed polar desert oasis. Canadian Journal Of Botany, 68, 2660-2667. Hinzman, L , Bettez, N , Bolton, W, Chapin, F , Dyurgerov, M , Fastie, C , Griffith, B , Hollister, R, Hope, A , Huntington, H , Jensen, A , Jia, G , Jorgenson, T, Kane, D , Klein, D , Kofinas, G , Lynch, A , Lloyd, A , McGuire, A , Nelson, F , Oechel, W, Osterkamp, T , Racine, C , Romanovsky, V , Stone, R, Stow, D , Sturm, M , Tweedie, C , Vourlitis, G , Walker, M , Walker, D , Webber, P, Welker, J , Winker, K , & Yoshikawa, K. (2005) Evidence and Implications of Recent Climate Change in Northern Alaska and Other Arctic Regions. Climatic Change, 72, 251. Hobbie, S, Shevtsova, A , & Chapin III, F.S. (1999) Plant responses to species removal and experimental warming in Alaskan tussock tundra. Oikos, 84, 417-434. Hollister, R.D, Webber, P, & Bay, C. (2005) Plant response to temperature in northern Alaska: implications for predicting vegetation change. Ecology, 86, 1562-1570. Hollister, R.D. & Webber, P.J. (2000) Biotic validation of small open-top chambers in a tundra ecosystem. Global Change Biology, 6, 835-842. 65 Hollister, R.D., Webber, P.J., & Tweedie, C.E. (2005a) The response of Alaskan arctic tundra to experimental warming: differences between short- and long-term responses. Global Change Biology, 11, 525-536. Hulten, E. (1968) Flora of Alaska and Neighboring Territories Stanford University Press, Stanford. Humlum, C. (1980) Germination ecology in the arctic-alpine species Oxyria digyna and the alpine species Oxyria elatior. Botanisk Tidsskrift, 75, 173-180. Illeris, L., Christensen, T.R., & Mastepanov, M. (2004) Moisture effects on temperature sensitivity of C02 exchange in a subarctic heath ecosystem. Biogeochemistry, 70, 315-330. IPCC (2001). Climate Change 2001: Impacts, Adaptation, and Vulnerability, Contribution of Working Group II to the Third Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge Univ. Press, Cambridge, U.K. Johnstone, J.F. (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, University of British Columbia, Vancouver. Jones, G. & Henry, G.H.R. (2003) Primary plant succession on recently deglaciated terrain in the Canadian High Arctic. Biogeography, 30, 277-296. Jones, M.H. (1995) Experimental investigations in to effects of climate change on high arctic plants. Ph.D., University of British Columbia. Jonsdottir, I.S., Magnusson, B., Gudmundsson, J., Elmarsdottir, A., & Hjartarson, H. (2005) Variable sensitivity of plant communities in Iceland to experimental warming. Global Change Biology, 11, 553-563. Kendall, M.G. (1970) Rank Correlation Methods, 4th edn. Griffin, London. Krannitz, P.G. (1996) Reproductive ecology of Dryas integrifolia in the high Arctic semi-desert. Canadian Journal of Botany-Revue Canadienne De Botanique, 74, 1451-1460. Laine, L., Malila, E., & Siuruainen, M. (1995) How is annual climatic variation reflected in the production of germinable seeds of arctic and alpine plants in the northern Scandes? In Global change and arctic terrestrial ecosystems (eds T.V. Callaghan, W. Oechel, T. Gilmanov, J. Holten, B. Maxwell, U. Molau, B. Sveinbjomsson & M. Tyson), pp. 89-95. Commission of the European Communities Ecosystems Research Report, Oppdal, Norway. Larsson, E.-L. (2002) Seed banks and seed dispersal in subarctic and arctic environments, Goteborg University, Goteborg, Sweden. Levesque, E. & Svoboda, J. (1992) Growth of arctic poppy in contrasting habitats of a polar oasis and a polar desert. Musk-Ox, 39, 148-156. Maessen, O., Freedman, B., Nams, M., & Svoboda, J. (1994). Resource allocation in high-'arctic vascular plants of differing growth form. In Ecology of a Polar Oasis: Alexandra Fiord, Ellesmere Island, Canada (eds J. Svoboda & B. Freedman), pp. 113-121. Captus Press Inc., North York. Margalef, R. (1968) Perspectives in Ecological Theory The University of Chicago Press, Chicago. Marion, G., Henry, G., Freschman, D., Johnstone, J., Jones, G., Jones, M., Levesque, E., Molau, U., Molgaard, P., Parsons, A., Svoboda, J., & Virginia, R. (1997) Open-top designs for manipulating field temperature in high-latitude ecosystems. Global Change Biology, 3, 20-32. 66 Maxwell, B. (1992). Arctic Climate: Potential for Change under Global Warming. In Arctic Ecosystems in a Changing Climate: An Ecophysiological Perspective (eds F. Chapin III, R. Jefferies, J. Reynolds, G. Shaver & J. Svoboda). Academic Press, Inc. McGraw, J. & Fetcher, N. (1992). Response of Tundra Plant Populations to Climatic Change. In Arctic Ecosystems in a Changing Climate - An Ecophysiological Perspective (eds F.S. Chapin, R. Jefferies, J. Reynolds, G.R. Shaver & J. Svoboda), pp. 359-376. Academic Press. Miller, P. (1982) Environmental and vegetational variation across a snow accumulation area in montane tundra in central Alaska. Holarctic Ecology, 5, 85-98. Molau, U. (1993.) Relationship between flowering phenology and life history strategies in tundra plants. Arctic and Alpine Research, 25, 391-402. Molau, U , Nordenhall, U , &'Eriksen, B. (2005) Onset of flowering and climate variability in an alpine landscape: a 10-year study from Swedish Lapland. American Journal of Botany, 92, 422-431. Molau, U. & Shaver, G.R. (1997) Controls on seed production and seed germinability in Eriophorum vaginatum. Global Change Biology, 3, 80-88. Mooney, H.A. & Billings, W.D. (1961) Comparative physiological ecology of arctic and alpine populations of Oxyria digyna. Ecological Monographs, 31, 1 - 29. Moore, R, Clark, W, & Vodopich, D. (1998) Botany, Second edn. WCB McGraw-Hill. Muc, M. (1977). Ecology and primary production of Sedge-moss Meadow communities, Truelove Lowland. In Truelove Lowland, Devon Island, Canada: A High Arctic Ecosystem (ed L.C. Bliss), pp. 157-184. University of Alberta Press, Edmonton. Muc, M , Freedman, B , & Svoboda, J. (1989) Vascular plant communities of a polar oasis at Alexandra Fiord, Ellesmere Island, Canada. Canadian Journal Of Botany, 67, 1126-1136. Muc, M , Svoboda, J , & Freedman, B. (1994). Soils of an extensively vegetated polar desert oasis, Alexandra Fiord, Ellesmere Island. In Ecology of a polar oasis, Alexandra Fiord, Ellesmere Island, Canada (eds J. Svoboda & B. Freedman), pp. 41-52. Captus Press Inc., North York. Murray, C. & Miller, P. (1982) Phenological observations of major plant growth forms and species in montane and Eriophorum vaginatum tussock tundra in central Alaska. Holarctic Ecology, 5, 109-116. Nadelhoffer, K , Giblin, A , Shaver, G.R, & Linkins, A. (1992). Microbial processes and plant nutrient availability in Arctic soils. In Arctic ecosystems in a changing climate: An ecophysiological perspective (eds F. Chapin, R. Jefferies, J. Reynolds, G. Shaver & J. Svoboda), pp. 281-300. Academic Press. Nadelhoffer, K , Shaver, G , Giblin, A , & Rastetter, E. (1997). Potential Impacts of Climate Change on Nutrient Cycling, Decomposition, and Productivity in Arctic Ecosystems. In Global Change and Arctic Terrestrial Ecosystems (eds W. Oechel, T. Callaghan, T. Gilmanov, J. Holten, B. Maxwell, U. Molau & B. Sveinbjornsson), Vol. 124. Springer. Nosko, P. & Courtin, G. (1995) The water relations of Carex stans in Wet Sedge-Moss Tundra at a High Arctic Oasis, Devon Island, NWT, Canada. Arctic and Alpine Research, 27,137-145. Olson, R. & Richards, J. (1979) Temperature responses of germination in arctic poppy (Papaver radicatum Rottb.) seeds. Arctic and Alpine Research, 11, 343-348. Polunin, N. (1951) The Real Arctic: Suggestions for its Delimitation, Subdivision and Characterization. The Journal of Ecology, 39, 308. 67 Porslid, A. & Cody, W. (1980) Vascular plants of continental Northwest Territories, Canada National Museum of Canada, National Museum of Natural Sciences, Ottawa. 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, 856-866. Rolph, S. (2003) Effects of a Ten-Year Climate Warming Experiment on Nitrogen Cycling in High Arctic Tundra. Master of Science, University of British Columbia. Sandvik, S.M. & Totland, O. (2000) Short-term effects of simulated environmental changes on phenology, reproduction, and growth in the late-flowering snowbed herb Saxifraga stellar is L. Ecoscience, 7,201-213. Savile, D. (1972) Arctic adaptations in plants (ed C.D.o. Agriculture), Vol. 6, pp. 1-81. Canada Department of Agriculture. Shaver, G. & Kummerow, J. (1992). Phenology, resource allocation, and growth of arctic vascular plants. In Arctic ecosystems in a changing climate: An ecophysiologicalperspective (eds F. Chapin, R. Jefferies, J. Reynolds, G.R. Shaver & J. Svoboda), pp. 193-211. Academic Press. Shaver, G.R. & Jonasson, S. (1999) Response of Arctic ecosystems to climate change: results of long-term field experiments in Sweden and Alaska. Polar Research, 18, 245-252. Sheard, J. & Geale, D. (1983) Vegetation studies at Polar Bear Pass, Bathurst Island, N.W.T. II. Vegetation-environment relationships. Canadian Journal Of Botany, 61, 1637-1646. Sorenson, T. (1941) Temperature relations and phenology of the northeast Greenland flowering plants. Meddr Gronland, 125, 1-305. Stenstrom, M., Gugerli, F., & Henry, G. (1997) Response of Saxifraga oppositifolia L. to simulated climate change at three contrasting latitudes. Global Change Biology, 3, 44-54. Suzuki, S. & Kudo, G. (1997) Short-term effects of simulated environmental change on phenology, leaf traits, and shoot growth of alpine plants on a temperate mountain, northern Japan. Global Change Biology, 3, 108-115. Svoboda, J. & Henry, G. (1987) Succession in marginal arctic environments. Arctic and Alpine Research, 19, 373-384. Thorhallsdottir, T. (1998) Flowering phenology in the central highland of Iceland and implications for climatic warming in the Arctic. Oecologia, 114, 43-49. Tilman, D. (1982) Resource competition and community structure Princeton University Press, New Jersey. Timson, J. (1965) New method of recording germination data. Nature, 207, 2016-2017. Ungar, I., A. (1996) Effect of Salinity on Seed Germination, Growth, and Ion Accumulation of Atriplex Patula (Chenopodiaceae). American Journal of Botany, 83, 604-607. ' van Wijk, M.T., Clemmensen, K.E., Shaver, G.R., Williams, M., Callaghan, T.V., Chapin, F.S., Cornelissen, J.H.C., Gough, L., Hobbie, S.E., Jonasson, S., Lee, J.A., Michelsen, A., Press, M.C, Richardson, S.J., & Rueth, H. (2004) Long-term ecosystem level experiments at Toolik Lake, Alaska, and at Abisko, Northern Sweden: generalizations and differences in ecosystem and plant type responses to global change. Global Change Biology, 10, 105-123. Walker, D.A., Raynolds, M.K., Daniels, F.J.A., Einarsson, E., Elvebaak, A., Gould, W.A., Katenin, A.E., Kholod, S.S., Markon, C.J., Melnikov, E.S., Moskalenko, N.G., Talbot, S.S., & Yurtsev, BA. (2005) The Circumpolar Arctic vegetation map. Journal of Vegetation Science, 16, 267-282. 68 Walker, L. & Chapin, IF. (1987) Interactions among processes controlling successional change. Oikos, 50, 131-155. Walker, M., Wahren, C , Hollister, R.D., Henry, G.H.R., Ahlquist, L., Alatalo, J.M., Bret-Harte, M.S., Calef, M., Callaghan, T.V., Carroll, A., Epstein, H.E., Jonsdottir, I.S., Klein, J., Magnusson, B., Molau, U., Oberauer, S., Rewa, S., Robinson, C.H., Shaver, G.R., Suding, K., Thompson, C , Tolvanen, A., Totland, Q., Turner, P.L., Tweedie, C , Webber, P.J., & Wookey, PA. (2006) Plant community responses to experimental warming across the tundra biome. PNAS, 103, 1342-1346. Walker, M.D., Ingersoll, R.C., & Webber, P.J. (1995) Effects of Interannual Climate Variation on Phenology and Growth of Two Alpine Forbs. Ecology, 76, 1067. Walker, M.D., Webber, P.J., Arnold, E.H., & Ebert-May, D. (1994) Effects of Interannual Climate Variation on Aboveground Phytomass in Alpine Vegetation. Ecology, 75, 393. Walsh, J., Kattsov, V., Chapman, W., Govorkova, V., & Pavlova, T. (2002) Comparison of arctic climate simulations by uncoupled and coupled global models. American Meteorological Society, 1429-1446. Webber, P.J., Miller, P., Chapin III, F.S., & McCown, B. (1980). The Vegetation: Pattern and Succession. In An Arctic Ecosystem: The Coastal Tundra at Barrow, Alaska (ed J. Brown), Vol. 12, pp. 571. Dowden, Hutchinson & Ross, Inc., Stroudsburg Pennsylvania. Welker, J., Molau, U., Parsons, A., Robinson, C , & Wookey, P. (1997) Responses of Dryas octopetala to ITEX environmental manipulations: a synthesis with circumpolar comparisons. Global Change Biology, 3, 61-73. Welker, J., Rayback, S., & Henry, G.H.R. (2005) Arctic and North Atlantic Oscillation phase changes are recorded in the isotopes (8I80 and 813C) of Cassiope tetragona plants. Global Change Biology, 11, 997-1002. Willson, M. & Traveset, A. (2000). The Ecology of Seed Dispersal. In Seeds The Ecology of Regeneration in Plant Communities (ed M. Fenner), pp. 85-110. Woodley, E. & Svoboda, J. (1994). Effects of habitat on variations of phenology and nutrient concentration among four common plant species of the Alexandra Fiord Lowland. In Ecology of a polar oasis: Alexandra Fiord, Ellesmere Island, Canada (eds J. Svoboda & B. Freedman), pp. 157-175. Captus Press. Wookey, P.A., Parsons, A.N., Welker, J.M., Potter, J.A., Callaghan, T.V., Lee, J.A., & Press, M.C. (1993) Comparative Responses Of Phenology And Reproductive Development To Simulated Environmental-Change In Sub-Arctic And High Arctic Plants. Oikos, 67, 490-502. 69 APPENDIX A Aboveground vegetative and reproductive biomass, fresh (F2004) seed biomass, and over-winter (OW2003 & OW2004) seed biomass sample sizes showing the number of plots per site and treatment by species Sample sizes were consistent across biomass categories (e.g. current-year fascicle, flower biomass, etc., Section 2.2.1). Material was harvested from Alexandra Fjord in 2004 (F2004 & OW2003) and 2005 (OW2004) from control (C) and warmed (W) plots in each of the six plant communities (Section 2.1.2). Site C/W Dryas Salix Salix Papaver Oxyria Festuca Eriophorum Luzula Carex integrifolia arctica 5 arctica 3 radicatum digyna brachyphylla triste spp. misandra S-CP- c 4 4 1 6 6 DS w 2 . 2 1 6 5 CP- c 6 6 6 DS(C) w 6 6 6 CP- • c 6 6 6 6 6 DS(D) w 6 6 6 4 4 DDS- c 6 6 6 6 6 6 6 6 -CP w 6 6 6 . 6 6 6 6 6 PSD- c 6 - 6 G w 6 6 PSD- c 6 6 D w 5 4 Aboveground Vegetative and Reproductive Biomass : F2004 Seed Biomass Site C/W Dryas integrifolia Salix arctica Papaver radicatum Draba SPP-Oxyria digyna Festuca brachyphylla Eriophorum triste S-CP-DS W CP-DS(C) W CP-DS(D) W DDS-G W PSD-G W PSD-D W Site C/W Dryas Salix Papaver Festuca brachyphylla Eriophorum Luzula Carer integrifolia arctica radicatum digyna triste spp. misandra 'r~.~ ' -' : OW2003 Seed Biomass ; S-CP-DS C 8 3 3 w 7 2 1 DDS-G c 8 5 5 w 5 2 6 OW2004 Seed Biomass . • , ' S-CP-DS c 4 4 4 w 4 2 2 3 CP- c 6 6 DS(C) w 6 4 CP- c 4 4 4 4 DS(D) w 4 4 4 4 DDS-G c 4 4 4 4 4 4 w 3 4 4 4 4 4 PSD-G c 5 w 5 PSD-D c 3 w 4 70 APPENDIX B Sample sizes of fresh (F2004) and over-wintered (OW2003 & OW2004) seeds harvested from control (C) and warmed (W) plots in each of the six plant communities (Section 2.1.2) at Alexandra Fjord in 2004 (F2004, OW2003) and 2005 (OW2004). Sample sizes describe the number of plots*- 50 seeds/plot. F2004 Site C/W Dryas integrifolia Salix arctica Papaver radicatum Draba spp. ' Oxyria digyna Festuca brachyphylla Eriophorum triste Luzula spp. S-CP-DS c 6 6 6 w 5 4 6 CP-DS(C) c 6 6 2 4 6 w 6 5 2 ' 4 6 CP-DS(D) c 6 6 4 6 w 6 6 4 6 DDS-G c 6 5 6 6 6 6 3 w 4 5 5 5 6 6 3 PSD-G c 3 w 3 PSD-D c 3 w 3 Site c/w Eriophorum triste Luzula spp. S-CP-DS c 4 3 w 4 3 DDS-G c 4 4 4 w 4 . 4 4 : OW2004 - -'-.y.\ - • . Site c/w S-CP-DS c 4 w 2 CP-DS(C) c 6 w 6 CP-DS(D) c 6 w 6 DDS-G c 6 w 5. 71 A P P E N D I X C Sample sizes of fresh seeds harvested from several international locations (Section 2.2.4) under ambient conditions, as well as from the four lowland sites at Alexandra Fjord. These samples were collected from control (C) plots. Site Dryas integrifolia Dryas octopetala Greenland C 7 Sweden C 1 Iceland C 6 Alexandra Fjord (Total) C 24 72 APPENDIX D Mean germination rate ± SD, including fresh (F2004) and over-winter (OW2003, OW2004) seed collections for control (C) and warming (W) treatments from lowland and polar semi-desert sites at Alexandra Fjord. Germination rates were calculated using Timson's index of germination velocity (Section 2.3.6). Site abbreviations are described in Section 2.1.2. ** p < 0.05. S-CP-DS CP-DS(C) CP-DS(D) DDS-G PSD-G c | w c | w c | w . c | vv c | w F2004 All 17±9 31± I4 19±10 44±13" 21±9 42±9 . 75±14 59±19 1±1 2±3 Dryas integrifolia 0±0 7 ± 6 " 1±1 10±7" 0±0 10±7" 0±0 11:8" Salix arctica 17±9 24±5" 13±5 2 8 ± l " 13±7 25:7" 14±9 18±12" 1±1 2±3-Papaver radicatum 6±4 1"7 Oxyria digyna 23±6 • 17:7 ' Draba spp. 0±0 0±0 0±0 0±0 ' 16±11 20±T3,--Eriophorum triste 0±1 5 ± 3 " 3±2 4:2 • Festuca brachyphylla y l 0±0 •3±2";'\ ' : Luzula spp. 5±5 11±10" 5±4 • 4 :4" 0±0 . 2 :1" , OW2003 All 5±3 8±6 16±4 9:3 ' • Dryas integrifolia 5±3 - ; 8±6 13±5 . 6:5 Salix arctica 3±3 . 2±3' ; -Eriophorum triste 0±0 0±0 : • ' : • . X Luzula spp. 1±4 ' 1 !2 -s. OW2004 Dryas integrifolia 4±5 7.:. 10 3±3 y 3:5 y 2±5 0 -0 : 0±0 3±6 73 A D D E N D U M The following pieces of writing and the conte drawings have been included in this document as an addendum in testament to the "work" that was undertaken at Alexandra Fjord, Ellesmere Island over the summers of 2004 and 2005. Seedling Quadrats d. 223 (21:30hrs, 2004) My sense is this: Once we are gone from here, After we have left, bringing our work-stained clothing and our consumptive ways Back to respective "homes", After we have finished with our measuring and monitoring the bounty of your labours, the fruit and fallow of your lands, There will be a space once again reclaimed, Where creatures shy or wary of the noise, human sounds, and of course, human intent, Will venture. A tentative approach, Into that once busy and odiferous place. They will sniff, start, flee and return. Someone will urinate, by necessity or instinct, Or sheer defiance; The first act of restoration. These thoughts occur to me, are moulded, shaped and mulled while I perch on hummocks of Cassiope, moss and willow, or while striding unevenly through puddle and tussock, to or from "work"; Always in the golden light of the sun, Which dips to meet the mountain for the first time tonight, to be extinguished and done. Wafts of Cassiope d.184 (July 02, 2004) Wafts of Cassiope, the warm spice of mountain heather crushed or sun-baked. Dryas petals spread wide, cream-coloured and curved to welcome the sun, yellow anthers and stigmas, an offering to each other and the efforts of Bombus. Straw-coloured Arctagrostis of last years' crop, amongst green leaf and stem and purple-green inflorescence of the new. Salix heads, tufted halos backlit. Males with pollen-dusted parts; females with lip-sticked stigmas, awaiting .the kiss of the wind, while I serve as a host to whining mosquitoes, hungry flies, and discover within myself the listener and lover of plant, rock, water and flesh of this fjord. Ode to ITEX fd. 184, 2004) Sitting against a bank of Cassiope, Listening to a Bombus bee. Trickling water over rocks. Sun-warmed wind. Mosquito whispers, Happy. 74 Post "Dome" hike d. 216 (August, 2004) Sun dappled air, burnished skin and goldened hairs. A mossy glade where Silene and Salix reside; I am given the privilege of their beauty, and the peace of this miniature oasis, on the side of a mountain, Where one must bow down to consume the sweetly mineraled elixir, Which flows as birth waters, from between the legs of the lady herself. Alexandra, I will drink the water that has washed your skin, and flowed through your veins, to become pure and clean and life giving. Aware d. 184 (July 02, 2004 Who are we to enter this world, which is untouched and alive? Data loggers, thermistors, rain gauges, Fibreglass OTCs. Footprints fall into the warm, wet folds of her earth and remain. We come here to measure and monitor The wonders of growth, warmth and bounty; Blind to the struggles, the sleeping, the power of wind and ice and snow, Which in our absence, our warm-blooded absence, Continues without pause, wreaks havoc To create the landscape we love, Over months, and through time infinite and finite; She opens her world for use to see. Alexandra Fiord - Sphynx View 75 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            data-media="{[{embed.selectedMedia}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0100758/manifest

Comment

Related Items