UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The impacts of High Arctic permafrost disturbances on vegetation and carbon flux dynamics Cassidy, Alison Elizabeth 2016

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

Item Metadata

Download

Media
24-ubc_2016_november_cassidy_alison.pdf [ 9.45MB ]
Metadata
JSON: 24-1.0315351.json
JSON-LD: 24-1.0315351-ld.json
RDF/XML (Pretty): 24-1.0315351-rdf.xml
RDF/JSON: 24-1.0315351-rdf.json
Turtle: 24-1.0315351-turtle.txt
N-Triples: 24-1.0315351-rdf-ntriples.txt
Original Record: 24-1.0315351-source.json
Full Text
24-1.0315351-fulltext.txt
Citation
24-1.0315351.ris

Full Text

THE IMPACTS OF HIGH ARCTIC PERMAFROST DISTURBANCES ON  VEGETATION AND CARBON FLUX DYNAMICS by  Alison Elizabeth Cassidy  B.A.&Sc., McGill University, 2009 M.Sc., Queen’s University, 2011  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES (Geography)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)  November 2016  © Alison Elizabeth Cassidy, 2016 ii  Abstract Changing climate and disturbance regimes can have widespread ecosystem impacts, especially in the Arctic. Vegetation recovery and carbon flux dynamics were examined to determine the impacts of thermokarst disturbance on patterns and processes in High Arctic tundra ecosystems. Ecosystem responses to two forms of permafrost disturbance, active layer detachment slides and retrogressive thaw slumps, were studied on the Fosheim Peninsula, Ellesmere Island, Canada during the 2012, 2013, and 2014 growing seasons. The impacts of disturbance on vegetation and recovery were determined by sampling active retrogressive thaw slumps and recovered active layer detachment slides that were investigated nearly 20 years ago. Comparison of historic and modern data indicates distinct vegetation communities exist in differently aged disturbances with unique vascular plant species defining various zones and ages of disturbance. Differences were also found in site characteristics (including soil moisture, temperature, active layer depth, and soil nutrient concentrations) indicating the impacts of permafrost disturbance on the landscape. In addition, four active layer detachment slides measured in 1994 had transitioned to active retrogressive thaw slumps, which may be a response to the progressively warming climate. Carbon dioxide fluxes between the surface and the atmosphere were measured using a static chamber system and the eddy covariance technique at three sites on the Fosheim Peninsula. Over the studied growing seasons, disturbed landscapes sequestered significantly less carbon than their surrounding undisturbed tundra. In some sites, this resulted in the shift of the system from a net sink of CO2 to a net source.  A dual eddy covariance sampling approach was found to be preferable over a single tower setup with separation of fluxes based on wind partitioning as disturbed and undisturbed fluxes were simultaneously measured throughout the growing season using this method. Overall, active layer iii  detachments and retrogressive thaw slumps alter vegetation and carbon flux dynamics, and these changes may persist over many years. With predicted increases in the frequency and magnitude of these permafrost disturbances, impacts on tundra ecosystems will be evident at the landscape scale.   iv  Preface  Outside contributions to this thesis are as follows: Chapter 2 uses data collected by Manon Desforges as part of a MSc. degree, which was conceived and supervised by Greg Henry. Her thesis entitled, “Plant succession after active layer detachment slides, in a high Arctic tundra, Fosheim Peninsula, Ellesmere Island, Canada” was completed in 2000. She conducted field research in 1994 and identified active layer detachment slides and established the study design, which was repeated in 2012 to determine the long-term responses to permafrost disturbance.   I conceived the idea for Chapter 3, with input from Andreas Christen and Greg Henry. I executed the experiment, analyzed the data, and wrote the manuscript, with input from Andreas Christen and Greg Henry.   Andreas Christen, Greg Henry, and I conceived the idea for Chapter 4. I executed the experiment, analyzed the data, and wrote the manuscript, with input from Andreas Christen and Greg Henry.   A version of Chapter 4 has been published: Cassidy, AE, A Christen, and GHR Henry. 2016. The effect of a permafrost disturbance on growing-season carbon dioxide fluxes in a High Arctic tundra ecosystem. Biogeosciences 13: 2291-2303. I conducted all the testing and wrote most of the manuscript. The section on “Turbulent source area model” was originally drafted by Christen, A. v  Table of Contents  Abstract .......................................................................................................................................... ii	  Preface ........................................................................................................................................... iv	  Table of Contents ...........................................................................................................................v	  List of Tables .................................................................................................................................. x	  List of Figures .............................................................................................................................. xii	  List of Abbreviations ................................................................................................................... xv	  Acknowledgements .................................................................................................................... xvi	  Chapter 1: Introduction ................................................................................................................1	  1.1	   Overview ............................................................................................................................ 1	  1.2	   Literature review ................................................................................................................ 2	  1.2.1	   The Arctic ecosystem .................................................................................................. 2	  1.2.2	   Permafrost & Arctic ecosystem fluxes ....................................................................... 3	  1.2.2.1	   Permafrost thaw implications .............................................................................. 3	  1.2.2.2	   Net ecosystem exchange ...................................................................................... 4	  1.2.3	   Disturbance ecology .................................................................................................... 5	  1.2.4	   Thermokarst disturbances ........................................................................................... 7	  1.2.4.1	   Active layer detachment slides (ALDS) .............................................................. 8	  1.2.4.2	   Retrogressive thaw slumps (RTS) ..................................................................... 10	  1.2.5	   Climate impacts of ALDS and RTS .......................................................................... 13	  1.3	   Research objectives .......................................................................................................... 15	  1.4	   Structure of the thesis ....................................................................................................... 17	  vi  1.5	   Introduction to the study area .......................................................................................... 18	  1.5.1	   Previous research at the study location ..................................................................... 19	  Chapter 2: Vegetation succession after permafrost disturbance on the Fosheim Peninsula, Ellesmere Island ...........................................................................................................................21	  2.1	   Overview .......................................................................................................................... 21	  2.2	   Introduction ...................................................................................................................... 22	  2.3	   Methods............................................................................................................................ 24	  2.3.1	   Study site ................................................................................................................... 24	  2.3.2	   Vegetation sampling (1994 and 2012) ...................................................................... 28	  2.3.2.1	   Active layer detachment slides .......................................................................... 28	  2.3.2.2	   Retrogressive thaw slumps ................................................................................ 29	  2.3.3	   Comprehensive site characterization (2012) ............................................................. 30	  2.3.3.1	   Soil moisture and active layer depth .................................................................. 30	  2.3.3.2	   Soil nutrient availability ..................................................................................... 30	  2.3.3.3	   Soil temperature ................................................................................................. 30	  2.3.4	   Statistical analysis ..................................................................................................... 31	  2.3.4.1	   Time comparison ............................................................................................... 31	  2.3.4.2	   Vegetation composition and cover ..................................................................... 31	  2.3.4.3	   Soil moisture and active layer depth .................................................................. 32	  2.3.4.4	   Soil temperature ................................................................................................. 33	  2.4	   Results .............................................................................................................................. 33	  2.4.1	   Comparison of historical and modern data ............................................................... 33	  2.4.1.1	   Vegetation community composition and cover ................................................. 33	  vii  2.4.2	   Comparison of plant communities in disturbances sampled in 2012 ....................... 36	  2.4.3	   Retrogressive thaw slump vegetation analysis ......................................................... 40	  2.4.4	   Site characteristics of disturbances sampled in 2012 ................................................ 42	  2.4.4.1	   Soil moisture ...................................................................................................... 42	  2.4.4.2	   Active layer depth .............................................................................................. 43	  2.4.4.3	   Soil temperature ................................................................................................. 45	  2.4.4.4	   Soil nutrients ...................................................................................................... 47	  2.5	   Discussion ........................................................................................................................ 50	  2.5.1	   Successional response ............................................................................................... 50	  2.5.2	   Soil characteristics .................................................................................................... 54	  2.5.3	   Permafrost thaw implications ................................................................................... 55	  2.6	   Conclusion ....................................................................................................................... 56	  Chapter 3: Variability of the impacts of active retrogressive thaw slumps on net ecosystem exchange in the Canadian High Arctic ......................................................................................58	  3.1	   Overview .......................................................................................................................... 58	  3.2	   Introduction ...................................................................................................................... 59	  3.3	   Materials and methods ..................................................................................................... 62	  3.3.1	   Study area .................................................................................................................. 62	  3.3.2	   Site selection ............................................................................................................. 63	  3.3.3	   Vegetation sampling ................................................................................................. 66	  3.3.4	   Soil characteristics .................................................................................................... 67	  3.3.5	   Portable CO2 efflux chamber system ........................................................................ 68	  3.3.6	   Eddy covariance measurement of NEE .................................................................... 70	  viii  3.3.7	   Flux data processing ................................................................................................. 71	  3.3.8	   Statistical analysis ..................................................................................................... 72	  3.4	   Results .............................................................................................................................. 73	  3.4.1	   Micrometeorological conditions ............................................................................... 73	  3.4.2	   Vegetation ................................................................................................................. 75	  3.4.3	   Thermal regime ......................................................................................................... 77	  3.4.4	   Moisture regime ........................................................................................................ 79	  3.4.5	   Nutrient availability .................................................................................................. 81	  3.4.6	   Ecosystem respiration ............................................................................................... 82	  3.4.7	   Net ecosystem exchange (NEE) ................................................................................ 83	  3.5	   Discussion ........................................................................................................................ 85	  3.6	   Conclusions ...................................................................................................................... 89	  Chapter 4: The effect of a permafrost disturbance on growing-season carbon dioxide fluxes in a High Arctic tundra ecosystem .............................................................................................91	  4.1	   Overview .......................................................................................................................... 91	  4.2	   Introduction ...................................................................................................................... 92	  4.2.1	   Study area .................................................................................................................. 95	  4.3	   Materials and methods ..................................................................................................... 96	  4.3.1	   Eddy covariance measurements of carbon dioxide fluxes ........................................ 96	  4.3.2	   Turbulent source area model ..................................................................................... 99	  4.3.3	   Portable chamber system ........................................................................................ 102	  4.3.4	   Environmental variable sampling ........................................................................... 104	  4.4	   Results ............................................................................................................................ 105	  ix  4.4.1	   Environmental conditions during the study period ................................................. 105	  4.4.2	   NEE of disturbed and undisturbed tundra ............................................................... 108	  4.4.3	   Partitioning of NEE ................................................................................................. 112	  4.5	   Discussion ...................................................................................................................... 115	  4.6	   Conclusion ..................................................................................................................... 119	  Chapter 5: General conclusions ................................................................................................120	  5.1	   Summary ........................................................................................................................ 120	  5.2	   Future directions ............................................................................................................ 123	  5.3	   General implications ...................................................................................................... 124	  References ...................................................................................................................................126	  Appendix A ................................................................................................................................. 139	  Appendix B ................................................................................................................................. 141	  B.1	   Sub-Appendix ............................................................................................................ 142	   x  List of Tables  Table 2.1 Age characteristics of ALDS at Hot Weather Creek .................................................... 28	  Table 2.2 Results of ANOSIM comparison between plots sampled in 1994 and 2012 ................ 35	  Table 2.3 ANOSIM results comparing disturbed vegetation of several ALDS of varying ages with undisturbed tundra ................................................................................................................ 38	  Table 2.4 Indicator species analysis results from 2012 vegetation data ....................................... 39	  Table 2.5 Community characteristics of ALDS sampled in 2012 ................................................ 40	  Table 2.6 Indicator species for active RTS ................................................................................... 41	  Table 2.7 Results of ANOSIM comparison of species composition between undisturbed and disturbed tundra at RTS ................................................................................................................ 41	  Table 2.8 ANOVA table for comparisons of nutrient availability ................................................ 48	  Table 3.1 Site characteristics of sampled RTS ............................................................................. 63	  Table 3.2 ANOSIM pairwise comparisons between disturbed and undisturbed tundra ............... 75	  Table 3.3 Indicator species by site ................................................................................................ 76	  Table 3.4 Summary of soil characteristics by site and treatment ................................................. 79	  Table 3.5 Summary of soil nutrient availability at three study locations ..................................... 81	  Table 3.6 Results of soil nutrient availability analysis ................................................................. 82	  Table 4.1 Summary of net ecosystem exchange, soil temperatures, and soil moisture from disturbed and undisturbed tundra ................................................................................................ 108	  Table 4.2 Summary of measurements from the portable chamber system ................................. 115	  Table A.1 Community characteristics of ALDS sampled in 2012 ............................................. 139	  Table A.2 SIMPER analysis comparing vegetation composition of young and old ALDS ....... 140	  xi  Table A.3 SIMPER analysis of 2012 ALDS .............................................................................. 140	   xii  List of Figures  Figure 1.1 Photograph of ALDS morphology including three zones ............................................. 9	  Figure 1.2 Study location on the Fosheim Peninsula, Ellesmere Island ....................................... 20	  Figure 2.1 Average annual temperature and total annual precipitation measured at the Eureka Weather Station ............................................................................................................................. 26	  Figure 2.2 Study area at Hot Weather Creek with two types of permafrost disturbances ............ 27	  Figure 2.3 Schematic diagram showing distinct zones of an active layer detachment slide and corresponding transects ................................................................................................................. 29	  Figure 2.4 NMDS ordination of sites comparing 1994 and 2012 vegetation data ....................... 34	  Figure 2.5 Total vegetation cover based on zone, age, disturbance, and year .............................. 36	  Figure 2.6 NMDS ordination of data collected in 2012 ................................................................ 37	  Figure 2.7 NMDS ordination of vegetation measured in active RTS and corresponding undisturbed areas .......................................................................................................................... 41	  Figure 2.8 Mean soil moisture measured from scar, track, and toe zones of young, mature, and old disturbances during the 2012 growing season ........................................................................ 42	  Figure 2.9 Variability in soil moisture measurements taken throughout the 2012 growing season....................................................................................................................................................... 43	  Figure 2.10 Early season active layer depths in zones and adjacent undisturbed areas of ALDS of three general age classes in 2012 .................................................................................................. 44	  Figure 2.11 Maximum active layer depth in disturbance zones and undisturbed areas of ALDS of three different general age classes in 2012 ................................................................................... 45	  Figure 2.12 Daily mean soil temperature over 1 year in zones of select disturbances ................. 46	  xiii  Figure 2.13 Soil temperature deviation  from select disturbances ................................................ 47	  Figure 2.14 Box plots of soil nutrient availabilities across 4 zones of 3 ALDS ........................... 49	  Figure 3.1 Study location, Fosheim Peninsula, Ellesmere Island in 2013 .................................... 64	  Figure 3.2 Site sampling at locations RTS-1 and RTS-2 .............................................................. 66	  Figure 3.3 Portable closed chamber system and corresponding collar  and EC tower ................. 69	  Figure 3.4 Conditions throughout the 2013 growing season ........................................................ 74	  Figure 3.5 2013 NMDS results with site scores ........................................................................... 75	  Figure 3.6 Total vegetation cover at each site .............................................................................. 77	  Figure 3.7 Active layer depth among sites and zones ................................................................... 78	  Figure 3.8 Mean soil temperature throughout July 2013 at each site ........................................... 79	  Figure 3.9 Mean soil moisture in July at three sites ..................................................................... 80	  Figure 3.10 Box plots of ecosystem respiration at each site ......................................................... 83	  Figure 3.11 Mean NEE during the July 2013 growing season ..................................................... 84	  Figure 3.12 Seasonal mean NEE flux measured by eddy covariance ........................................... 85	  Figure 4.1 Aerial image of the dual eddy covariance system setup .............................................. 96	  Figure 4.2 Turbulent source areas for two time-steps. .................................................................. 99	  Figure 4.3 Meteorological conditions during the 2014 growing season ..................................... 107	  Figure 4.4 Ensemble average diurnal course of soil temperatures ............................................. 109	  Figure 4.5 Ensemble diurnal course of CO2 fluxes ..................................................................... 110	  Figure 4.6 Average daily net CO2 flux for the three sampling periods ...................................... 112	  Figure 4.7 Comparison of NEE measurements from static chamberand calculated from the two EC systems .................................................................................................................................. 113	  xiv  Figure 4.8 Partitioning of NEE data from static chamber measurements into component fluxes, GPP and Re for the undisturbed and disturbed sites .................................................................... 114	  Figure B.1 Tilted open-path and inlet for closed-path system .................................................... 142 Figure B.2 Difference between CO2 fluxes determined by open- and closed-path systems.…..142  xv  List of Abbreviations  Acronym  Description    Units ALD   Active layer thaw depth  cm ALDS    Active layer detachment slide EC   Eddy covariance  Re   Ecosystem respiration   µmol m-2 s-1   GPP   Gross primary productivity  µmol m-2 s-1 IRGA   Infrared gas analyzer NEE   Net ecosystem exchange  µmol m-2 s-1 PAR   Photosynthetically active radiation µmol m-2 s-1 RH   Relative humidity   % RTS    Retrogressive thaw slump SOM   Soil organic matter SM   Soil moisture    %  ST   Soil temperature   °C WD   Wind direction   ° xvi  Acknowledgements  Firstly, I’d like to thank both of my supervisors, Andreas Christen and Greg Henry, for their support and encouragement. Thanks to Greg for taking me to Ellesmere Island and letting me explore the landscape for three summers. Thanks to Andreas for entrusting me with high tech equipment in the Arctic tundra and for the helpful discussions on data processing and interpretation. I also thank my supervisory committee, Trevor Lantz and Toni Lewkowicz, who provided valuable input throughout the initial study design and final synthesis. Andy Black, Paul Jassal, Vincent St. Louis, and Craig Emmerton also provided essential equipment and advice. Many people assisted with fieldwork, from initial preparations to execution. I would like to thank Rick Ketler and Tanja Christen for their assistance with fieldwork preparations and equipment testing. Field assistance was provided by Andrew Baylis (2012), Jason Paul (2013), and Derek van der Kamp (2014). This research would not have been possible without their hard work and endless energy whilst carrying car batteries. I am grateful for the support of colleagues in the Henry Lab: Ali Beamish, Anne Bjorkman, Noémie Boulanger Lapointe, Sarah Desrosiers, Sarah Elmendorf, Esther Frei, Chris Greyson-Gaito, Isla Myers-Smith, and Sam Robinson. Thank you for the valuable feedback and suggestions.  Logistical support was provided by Polar Continental Shelf Program, Natural Resources Canada. Travelling though Resolute and catching up with the crew at PCSP was a great bookend to each field season. Thank you to the many pilots who helped locate sites and carefully transported gear and equipment. A special thank you to John Innis, whose mail and cookie drops were much appreciated.  xvii  Financial support was provided by the Natural Sciences and Engineering Research Council (NSERC) Discovery Frontiers Grant (ADAPT) to Greg Henry, NSERC Discovery Grants to both Greg Henry and Andreas Christen, the Canadian Foundation for Innovation, the Northern Scientific Training Program, and the Association of Canadian Universities for Northern Studies and Environment Canada.  Lastly, thank you to the wonderful family and friends that provided me with motivation and support to finish this thesis.  A special thank you goes to my parents and Derek for pep talks and never-ending encouragement.   1  Chapter 1: Introduction  1.1 Overview Projected climate change including increased air and ground temperatures and precipitation are expected to increase the extent of thermokarst disturbances (disturbances caused by thawing of ice-rich permafrost) in the Arctic (ACIA, 2005; Vincent et al., 2011). In the High Arctic, one of the most common forms of thermokarst are active layer detachment slides (ALDS), which occur when the thawed (or active) layer of ground breaks away from the underlying ice-rich permafrost. This results in a rapid mass movement of soil and vegetation downslope. ALDS often occur after prolonged periods of increased temperatures and precipitation and may evolve into retrogressive thaw slumps (RTS), a different form of disturbance (Lewkowicz, 1990). RTS are characterized by the exposure of ground ice within a headwall. As ground ice thaws, further material slumps downslope and the headwall regresses and will remain active until falling blocks of soil and vegetation insulate exposed ice and prevent further thaw (Burn and Friele, 1989). The intensification of thermokarst processes is important in the context of carbon dynamics and landscape heterogeneity.   The initiation requirements and geomorphological characteristics of these ALDS in the Arctic have been analyzed, but there are few studies examining the recovery of tundra ecosystems over decadal time scales. Additionally, climate warming may cause differential responses in disturbed and undisturbed tundra. The response of plants to increases in temperature may be intensified when disturbance occurs (Lantz et al., 2009). In fact, disturbance may play a more significant role in vegetation modification than temperature changes, particularly at the fine scale (Lantz et al., 2009). In addition, large quantities of carbon are currently stored in permafrost soils, with 50% of global soil carbon estimated to be stored in tundra and taiga systems (Tarnocai et al., 2009). This carbon has the 2  potential to be released upon thaw through oxidation (microbial respiration) and through the flow of water into aquatic systems (Abbott et al., 2016).  My research examines the impacts of ALDS and RTS on High Arctic tundra ecosystems. Research objectives and hypotheses are presented in Section 1.3 in addition to an overview of the dissertation. This research is significant as understanding the long-term effects of ALDS and RTS on the landscape, vegetation, and the carbon balance of tundra ecosystems is essential for predicting the effects on future carbon balance and tundra biodiversity and mitigating the impacts of disturbance.  1.2  Literature review  1.2.1 The Arctic ecosystem The Arctic tundra biome is characterized by vegetation with low species diversity and low annual productivity (Bliss et al., 1973; Babb and Bliss, 1974; Forbes et al., 2001). The Arctic is especially sensitive to disturbance, due to its short growing season, cold temperatures, and resource poor permafrost soils (Stow et al., 2004). The Arctic has a range of climatic and hydrologic conditions and is often classified into two sub-regions: High Arctic and Low Arctic. The High Arctic is characterized by more extreme weather conditions, including lower temperatures and a shorter growing season than the Low Arctic, and it contains more prostrate plants and a lower vascular plant cover and diversity that results in lower productivity (Forbes et al., 2001). Areas in the High Arctic are generally more sensitive to disturbance than Low Arctic areas, as they are slow to return to pre-disturbance conditions, and areas remain bare for long periods (Babb and Bliss, 1974; Billings, 1973; Bliss and Wein, 1972; Forbes et al., 2001; Svoboda and Henry, 1987), whereas recovery may occur more quickly in the Low Arctic. However, when disturbance results in enhanced nutrient availability and decomposition rates, plant growth can be stimulated and enhanced in both Low and High Arctic areas (Forbes et al., 2001).  High Arctic tundra vegetation is governed by few factors, with summer temperature acting as 3  one of the main controls on growth, as it determines species that can establish during the growing season, and the rates of growth and reproduction, as well as soil nutrient processes (Young, 1971). Soil moisture is usually limited to snow melt water, given the low precipitation rates.  However, wetland areas can develop in lowlands or along water courses where drainage is impeded by permafrost. The most exposed areas with little snow cover in the winter and very low soil moisture, and hence very low plant cover, are known as polar deserts (Babb and Bliss, 1974; Bliss and Matveyeva 1992). Differences in substrate are reflected in vegetation communities, and generally can be assigned to either acidic or basic areas.  Most of the central high Arctic is underlain by sedimentary rock with alkaline soils, while the eastern regions have extensive areas of acidic metamorphic rocks (Walker et al. 2005). 1.2.2 Permafrost & Arctic ecosystem fluxes The northern circumpolar permafrost region encompasses an area of 18.783 x 106 km2, with 65% of this area located in Eurasia and the remainder in North America and Greenland (Tarnocai et al., 2009). Regions can be characterized along a latitudinal gradient by varying degrees of permafrost cover, ranging from discontinuous (30-80 % frozen ground) to continuous (> 80 % frozen ground). 1.2.2.1 Permafrost thaw implications The quantity of soil organic carbon (SOC) stored in the top 3 m of frozen and unfrozen soils in permafrost regions is estimated to be 1400-1850 Pg, encompassing up to 50% of worldwide below ground organic carbon (Grosse et al., 2011; Hugelius et al., 2014; Tarnocai et al., 2009; Schuur et al., 2015). Current estimates are likely an underestimation (by as much as a factor of two) due to difficulties measuring and uncertainty regarding carbon storage in cryoturbated soils (Hugelius et al., 2013).   As permafrost thaws, this organic carbon has the potential to be released to the atmosphere (Schuur et al., 2008). It has been estimated that thawing permafrost will contribute 120±85 Gt carbon 4  to the atmosphere by 2100, increasing global temperatures by 0.29±0.21°C (Schaefer et al., 2014). As temperatures rise, this formerly frozen organic carbon becomes available for microbial decomposition and may be released in the form of carbon dioxide (CO2) or methane (CH4) depending on moisture conditions. This could create a positive feedback, as more carbon released leads to warmer temperatures, thus exacerbating thaw and leading to a further release of carbon. Organic carbon contents are highest in peat and cryoturbated mineral soils; therefore thaw in these areas could have significant impacts (Tarnocai et al., 2009). Conversely, a negative feedback may result if soil carbon inputs offset decomposition as the balance between litter accumulation and decomposition determines the effects on climate (Davidson and Janssens, 2006; Cornelissen et al., 2007). Shrub expansion, especially shrub species with recalcitrant leaf litter, and tree line advances may alter feedbacks to warming and offset or enhance warming effects on litter decomposition (Cornelissen et al., 2007; McGuire et al., 2006). Vegetation succession may also create a negative feedback and act to protect permafrost from large temperature increases (Grosse et al., 2011). 1.2.2.2 Net ecosystem exchange One method of quantifying carbon fluxes between the surface and the atmosphere is the measurement of net ecosystem exchange (NEE), a measure of carbon flux (usually CO2) within a system calculated as the difference between ecosystem respiration (Re) and gross primary production (GPP): negative values of NEE represent an uptake by the ecosystem (sink), and positive values represent release to the atmosphere (source). (Another term used is NEP, net ecosystem productivity, which is GPP minus Re). NEE in Arctic terrestrial systems is influenced by temperature, light and moisture levels (Baldocchi, 2008). Growing season NEE can vary between source and sink depending on vegetation type and summer climate, while wintertime NEE values show a very slow steady loss of carbon to the atmosphere (Baldocchi, 2008; Welker et al. 2004). The timing of snowmelt and snow accumulation varies each year, and influences the start and end of the growing 5  season. Shortly after snowmelt, plant growth and leaf area expansion is rapid and in most tundra types, net carbon uptake occurs through the growing season. The loss of leaves and leaf function in the late summer and early autumn generally precedes snowfall, and the autumn is dominated by respiration and the loss of carbon (Lüers et al., 2014). Although tundra communities are typically sinks for carbon during the growing season, they can shift to sources due to changes in cloud cover and the water table, or drought conditions (Baldocchi, 2008).   Experimental warming was found to increase sink potential through a longer growing season (Welker et al., 2004). At Alexandra Fiord, Ellesmere Island, experimental warming impacted NEE differently based on moisture, due to increased respiration at wet sites (Welker et al., 2004). Across a latitudinal gradient, warming tended to increase respiration, with the greatest increases found in dry ecosystems (Oberbauer et al., 2007). At Lake Hazen, Ellesmere Island, a polar desert site was found to be a weak sink, while wet sedge tundra sequestered significantly more CO2 (Emmerton et al., 2015). At the polar desert site, increases in moisture corresponded with increased respiration and CO2 emissions (Emmerton et al., 2015). Previous studies examining Arctic NEE have found large interannual variability among sites and this variability has been substantial enough to shift the growing season from a carbon sink to carbon source (Griffis and Rouse, 2001; Kwon et al., 2006; Merbold et al., 2009). In addition, there are few measurements of NEE from High Arctic tundra sites, which combined with the variability found in Arctic tundra ecosystems makes it difficult to determine NEE fluxes across the Arctic (Lafleur et al., 2012). 1.2.3 Disturbance ecology Disturbances alter hydrologic, nutrient, and thermal regimes, which are reflected through soil and vegetation characteristics within disturbed areas (Walker, 1996). These changes alter vegetation community development and create unique environments. Areas may be differentially affected, resulting in the formation of patches of differing severity across the landscape.  As these patches 6  recover they create spatial and temporal mosaics of unique plant communities (Forbes et al., 2001; Geertsema and Pojar, 2007).  Disturbance plays an important role in ecosystem function and vegetation development. In the Arctic, disturbance can be triggered by anthropogenic activities or can be naturally occurring. Disturbances associated with anthropogenic activity tend to be relatively small in scale, occupying areas ranging from 10−1 to 106 m2 (Forbes et al., 2001; Walker and Walker, 1991). Anthropogenic disturbances are mainly associated with industrial activity and heavy equipment usage, e.g. seismic lines, vehicle tracks, mining activities and oil drilling. These activities may result in thermokarst, or surface subsidence due to the melting of ground ice (Mackay, 1970; Raynolds et al., 2014). Disturbance also occurs naturally, and can include overgrazing, tundra fires and a range of disturbances related to thermokarst processes. Thermokarst refers to ground subsidence associated with the melting of ground ice. Two forms of thermokarst include active layer detachment slides and retrogressive thaw slumps.  Climate warming is predicted to increase the frequency and extent of disturbances, including tundra fires and permafrost disturbances (Anisimov et al., 2002; ACIA, 2005). Climate warming may cause differential responses in disturbed and undisturbed tundra and the response of plants to increases in temperature may be intensified when disturbance occurs (Lantz et al., 2009). Disturbance may play a more significant role in vegetation modification than temperature changes, particularly at the local scale (Lantz et al., 2008).   The removal of vegetation during disturbance can reduce albedo by a factor of two (Babb and Bliss, 1974). Removal of the soil organic layer reduces albedo and can increase soil temperatures and deepen the active layer, however the active layer may also increase due to the disturbance itself (Bliss and Wein, 1972; Auerbach et al., 1997; Lantz et al., 2009). An increase in active layer depth provides plant roots with more space in the soil layer (Bliss and Wein, 1972). Plants that have deep 7  roots, including species such as Calamagrostis canadensis and Eriophorum angustifolium are common invaders in some disturbed tundra sites due to their ability to take advantage of extra soil space (Chapin and Shaver, 1981). Changes to the physical environment following a disturbance event can also influence vegetation recovery through the creation of a unique microenvironment, including increased nutrient availability and ground temperatures, in the disturbance (Lantz et al., 2009). 1.2.4 Thermokarst disturbances Thermokarst disturbances vary temporally and spatially, and include both long-term and short-term events. They are influenced by changes in climate or hydrologic regimes (Walker and Walker, 1991; Kokelj and Jorgenson, 2013). Permafrost disturbances are occurring due to climatic changes that directly impact permafrost and active layer depths, and warming temperatures in the Arctic are predicted to increase the number and severity of permafrost disturbances (ACIA, 2005; Vincent et al., 2011).  There are several forms of mass wasting in the Arctic that are linked to thawing ice-rich permafrost on slopes, and they affect soil temperature, water quality and soil nutrients (Mackay, 1970; Lamoureux and Lafrenière, 2009; Lantz et al., 2009; Kokelj and Lewkowicz, 1998, 1999). These landslides include active layer detachment slides and retrogressive thaw slumps. Active layer detachment slides (ALDS), initiated by increased temperatures and precipitation, occur when the active layer detaches from the underlying ice-rich permafrost and results in a mass movement of soil and vegetation downslope (Lewkowicz, 1990; Lacelle et al., 2010). Retrogressive thaw slumps (RTS) are initiated by the exposure of ground ice (sometimes linked to coastal erosion and to ALDS) and result in the removal of soil and vegetation as the slump retreats further upslope (Lantuit and Pollard, 2008). Both forms of landslides alter the physical environment and vegetation communities. Within the overall landscape, these distinct landforms differ from the surrounding terrain and often contain unique microenvironments resulting in increased landscape heterogeneity (Ukraintseva, 8  2008; Lantz et al., 2009; Bosquet, 2011). 1.2.4.1 Active layer detachment slides (ALDS) Active layer detachment slides (ALDS), a shallow form of landslide, are defined by three morphological zones (Figure 1.1): a bare scar area, a track zone, and depositional toe (where soil compression can be great or minimal, depending on the run out zone). Shapes of these disturbances vary and include compact, elongated, and complex (Lewkowicz and Harris, 2005a). The formation of ALDS requires a combination of meteorological conditions and preconditioning of the ground surface. Meteorological triggers include warm summer temperatures, extreme summer precipitation events, and humid conditions during the previous year (Leibman, 1995). The initiation of ALDS require periods of extended surface heating and rapid thaw and do not necessarily form when the active layer is at its maximum depth, but when thaw rates are at their maximum (McRoberts and Morgenstern, 1974). Rapid thaw results in high pore water pressures increasing the potential for failure. Antecedent ground conditions necessary for failure include reduced soil shear strength over time through weathering of soils and increases in ice content at the base of the active layer. These conditions can take years to develop, and thus limit the reactivation potential of ALDS. Harris and Lewkowicz (2000) estimated ALDS require a minimum of 10-15 years for sufficient conditions to be established for reactivation. Preconditioning of the ground surface occurs through solifluction, as annual solifluction processes deform the soil and reduce shear strength over time (Harris et al., 2001).  ALDS can be triggered by permafrost thaw or other disturbances, such as forest fires (that result in permafrost thaw) (Lipovsky et al., 2005; Lewkowicz and Harris, 2005a,b). Lewkowicz and Harris (2005b) studied ALDS in discontinuous permafrost associated with fire initiation and found that they occurred weeks to months after forest fire activity, with no site preconditioning of the soil possible. Massive ice, which has an ice content greater than 250% (ice to dry soil weight) (Harry et 9  al., 1988), is not necessary for the occurrence of ALDS (which differs from RTS); however, they can occur in areas with massive ice and later may transition into RTS (Lewkowicz, 1990).  Figure 1.1 Photograph of ALDS morphology including three zones (scar, track, and toe). ALDS located at Hot Weather Creek, Ellesmere Island. Note the person standing above the scar zone for scale.    ALDS can occur on both shallow and steep slopes (Lewkowicz, 1990). Generally, steep slopes have rapid drainage and therefore reduced pore water pressures. Consequently, ALDS typically form on relatively shallow slopes if conditions such as an ice rich basal layer is present and rapid thaw occurs. Failure can involve one movement or multiple movements progressively upslope (Harris and Lewkowicz, 1993). Grain size is also an important determinant of slide morphology, as seen through differences in disturbance slope locations based on grain size variations across the Fosheim Peninsula, Ellesmere Island, Nunavut; slides form on low to medium plasticity clays, where fine sand and silt bands may be present (Harris and Lewkowicz, 1993). ALDS occurring on relatively sandy substrates are shorter, narrower, and steeper than those located on silty landscapes (Lewkowicz, 1990). On the Fosheim Peninsula, ALDS and RTS are generally located below the glacial marine limit where substrates include colluvium composed of till and marine sediments, sands, silts, clays, alluvium, and till and the permafrost is saline and ice rich (Bell, 1996).  Once ALDS occur, revegetation begins with pioneer species including sedges and grasses 10  colonizing bare soil sites, often in the absence of bryophytes (Ukraintseva, 2008). In the High Arctic, as ALDS continue to recover, shrub cover increases and bryophytes begin to appear (Desforges, 2001). On the Yamal Peninsula, Russia, sedge dominated communities transition to moss-lichen- shrub dominated areas (Leibman et al., 2003). In both areas, the greatest shrub cover and the tallest shrub species are found in the oldest ALDS, with shrubs, cushion plants, moss and lichen dominant in the final stages of succession (Desforges, 2001; Leibman et al., 2003). Recovery of these landslides can take 50 to 100 years, as the disturbed soil system is slow to recover (Lewkowicz, 1990; Desforges 2001). Bosquet (2011) analyzed the short-term (< 5 years) effects of High Arctic ALDS on vegetation and the physical environment and found increased snow cover resulted in cooler soils and shallower active layer depths, due to longer snow retention within the disturbed areas than undisturbed terrain. Longer snow retention also resulted in later phenological development for vascular plants, with delayed leaf bud burst found for all vascular species measured, including Ranunculus nivalis, Potentilla vahliana, and Salix arctica.  Loss of substrate occurs following ALDS activity, and this can lead to increased nutrient concentrations in disturbed areas resulting in enhanced plant productivity (Ukraintseva, 2008; Bosquet, 2011). Increased concentrations of K+, Ca2+, Mg2+, Cl-, SO42-, and PO33-, resulted in greater growth of Salix spp. on the Yamal Peninsula, Siberia (Ukraintseva and Leibman, 2000). Increases in soil salinity were also found to correlate with increased growth of Salix species located on disturbed slopes (Ukraintseva and Leibman, 2000). On the Fosheim Peninsula, salt efflorescence accumulations were frequently found at disturbed sites located below the marine limit, as dissolved solids that were previously trapped in frozen sediments were released and redistributed downslope (Kokelj and Lewkowicz, 1999). 1.2.4.2 Retrogressive thaw slumps (RTS) Retrogressive thaw slumps (RTS) are associated with the thawing of ice-rich permafrost in sloping 11  terrain (Mackay, 1966; McRoberts and Morgenstern, 1974). RTS are characterized by a steep headwall, gentle scar zone, and concave morphology and may be accompanied by long debris flows (Harry et al., 1988; Kokelj et al., 2013b). When ice-rich permafrost is exposed at the headwall, rapid melt occurs and saturated sediments flow downhill away from the headwall. Ongoing ground-ice ablation drives upslope headwall retreat (Lewkowicz, 1987). RTS require the presence of ground ice, either high ice contents and/or massive ice, as rapid ablation only occurs with adequate ice content in soils (Mackay, 1966). Slumps can be triggered by ground ice exposure caused by coastal erosion or other forms of disturbance including fire and ALDS, and increases in precipitation causing erosion.  In some areas, the erosion gullies found within scars of ALDS can cause ALDS to transition into RTS through further degradation of the ice-rich permafrost at the headwall (Lantuit and Pollard, 2008; Lewkowicz, 1990). If ice wedges are present and exposed in ALDS, troughs may develop leading to the formation of RTS (Lewkowicz and Harris, 2005a). French and Egginton (1973) note RTS require distinct meteorological conditions to form, including warm air and ground temperatures.  RTS can remain active for many years, even decades, or can stabilize in a few years, with an average life span of 3-25 years (Lewkowicz, 1987). They will remain active as long as fresh ice-rich substrate is exposed and continues to ablate. Headwall retreat occurs when maximum air temps are greater than 0°C and retreat can increase with rain events that remove material quickly (Burn and Friele, 1989). Kokelj et al. (2015) found recent intensification in rainfall significantly increased the number and size of RTS. As ground ice within the headwall of RTS melts, soil and vegetation located above the headwall fall into the slump. Slumps will remain active until either ice is depleted or blocks of soil and vegetation from above insulate the remaining ground ice and prevent further melt (Mackay, 1966; Burn and Friele, 1989). Due to this gradual retreat, a chronosequence of vegetation development and recovery can usually be identified within RTS, whereby the youngest vegetation and shortest duration of recovery is present directly beneath the most recent headwall 12  location (Bartleman et al., 2001). Although slumps can remain active for long periods, some years may exist where the slump is inactive, especially in cool years (Mackay, 1966).  Vegetation plays an important role in the recovery of the ground thermal regime in slumps as it insulates ground ice and decreases melt (Burn and Friele, 1989). However, even vegetated slumps show increased thaw depths when compared to undisturbed tundra because of increased snowpack (Kokelj et al., 2009). Slumps expose nutrient-rich mineral substrate, which can assist in vegetation recovery by increasing plant productivity and seed viability in disturbed zones (Lantz et al., 2009). Plant nutrient regimes may be elevated, as soluble materials in the frozen ground may be released with ground ice melt associated with disturbance, exemplified by increased nutrient (calcium and sulfate) concentrations in stable RTS (Lantz et al., 2009). RTS also displayed higher sulfate and calcium concentrations regardless of slump stability (Lantz et al., 2009). Elevated nutrient levels allow colonization to occur rapidly, beginning near the edge of RTS, with the expansion of surviving vegetation (Lantz et al., 2009; Ukraintseva, 2008). Snow may accumulate in slump depressions and the presence of shrubs in stable zones trap greater amounts of snow (Lantz et al., 2009). The accumulation of snow can result in freezeback of the ground occurring at a later date, and an increase in permafrost temperatures (Lantz et al., 2009; Kokelj et al., 2009). Warmer and more nutrient rich microenvironments are thus created in the retrogressive thaw slumps, relative to the surrounding areas.   Revegetation in thaw slumps is aided by blocks of vegetation, which fall into the slump or remain intact as blocks within the slump and stimulate recovery (Burn and Friele, 1989). In addition to multiple headwall retreat events, these islands of surviving vegetation create a mosaic of different vegetation communities throughout the slump. Bartleman et al. (2001) examined RTS near Mayo, Yukon and found when slumps are still active and meltwater from the headwall is abundant, areas on the slump floor are colonized by Funaria spp., Senecio spp., Calamagrostis canadensis, 13  Eriophorum scheuchzeri and Arctagrostis latifolia. In other studies of Low Arctic RTS, Equisetum spp. and Salix spp. appear in drier zones, with Salix spp. and Betula spp. appearing furthest downslope (Bartleman et al., 2001; Lambert, 1972; Lantz et al., 2009). Some species have been found exclusively in disturbed zones, including Senecio spp., Artemisia spp., and Epilobium spp. (Lantz et al., 2009). However, slump vegetation is typically less diverse and has a lower bryophyte cover (Burn and Friele, 1989). Changes in soil moisture in slumps can also modify vegetation development; for example, when slumps stabilize, the headwall no longer supplies meltwater to vegetation in the slump, and hydrophilic species will no longer dominate in these areas (Bartleman et al., 2001). The vegetation community in the oldest disturbances is most similar to the current undisturbed vegetation communities, however usually with lower cover and richness (Burn and Friele, 1989; Bartleman et al., 2001; Cannone et al., 2010). Differences in composition may persist for centuries (Cray and Pollard, 2015).  1.2.5 Climate impacts on ALDS and RTS There is evidence that the frequency of ALDS in the Canadian High Arctic has increased. On the Fosheim Peninsula the number of ALDS has doubled since 1975, which has been linked with changing climatic conditions (Lewkowicz and Harris, 2005b). Erosional processes following disturbance may result in ALDS becoming indistinguishable from the surrounding undisturbed terrain in a relatively short timespan, and make it difficult to identify relict disturbance on historical imagery resulting in a bias of underestimating the historical frequency of these disturbances (Lewkowicz, 1990; Lewkowicz and Harris, 2005b).  Increased summer temperatures could potentially increase the frequency of disturbance, however as initiation usually requires intense solar radiation, if greater temperatures are accompanied by an increase in cloud cover this may decrease the initiation of disturbances and could decrease the frequency of ALDS and RTS (Lewkowicz, 2007). Increased temperatures and periods of rapid thaw 14  could alter the frequency of permafrost disturbance. However, ground preconditioning is necessary for reducing shear strength (through weathering and sufficient ice content at the base of the active layer) and initiating failure (Lewkowicz and Harris, 2005b; Lewkowicz, 1990).  Certain areas may be more susceptible to changing disturbance regimes. In Siberia, Leibman et al. (2003) found longer return intervals (350-500 years) for ALDS based on radiocarbon dating, which can be contrasted with the Fosheim Peninsula, where return intervals are much shorter (25 years) as four different episodes have been identified over the past 100 years, according to Lewkowicz and Harris (2005b). In the Low Arctic, changing fire regimes (including frequency and burn severity) alter the surface vegetation mat and may increase susceptibility to both ALDS and RTS (Lewkowicz and Harris, 2005a). As permafrost disturbance frequency is dependent on fire return intervals, changes in fire frequency will control potential changes in the frequency of ALDS (Lewkowicz and Harris, 2005b). Increased erosion, especially along coastlines could expose more ground ice, and thus make the landscape more prone to retrogressive thaw slumping (McGuire et al., 2006). However, as RTS are limited to areas with high ice content, this will act as a control on where slumping will occur and thus RTS frequency.  Climate warming and changes in precipitation may affect disturbance regimes directly as both ALDS and RTS are triggered by meteorological conditions, in conjunction with other factors (ACIA, 2005; Kokelj et al., 2015; Segal et al., 2016). Recent increases in rainfall intensity have resulted in an increase in the total number of slumps and their overall extent (Kokelj et al., 2015). Warmer summer temperatures may increase ground temperatures and lead to rapid thaw of the active layer, making the ground prone to failure through sliding and slumping (Lewkowicz and Harris, 2005b). Indirectly, alteration of other disturbance regimes (including fire) and increased erosion may initiate ALDS and RTS. Overall, climate change will impact disturbance regimes in the future and is predicted to increase both magnitude and frequency of permafrost disturbances.  15   Similar to other types of disturbance, RTS and ALDS create patches of heterogeneous vegetation of varying sizes and ages across the landscape (Burn and Friele, 1989; Bartleman et al., 2001; Forbes et al., 2001; Bosquet, 2011; Cassidy, 2011). These forms of thermokarst have potentially significant climate implications, as large quantities of carbon are stored within permafrost soils (Grosse et al., 2011). As permafrost thaws, CO2 and CH4 emissions may increase, which could amplify climate change (Schuur and Abbott, 2011). Limited research has examined the impacts of these disturbances on carbon emissions and NEE (Beamish et al., 2014; Jensen et al., 2014; Abbott and Jones, 2015). The impact of spatial heterogeneity, including patches of differing vegetation communities within the overall landscape needs to be addressed to determine the impacts of RTS and ALDS on NEE. Scaling up of carbon fluxes to the landscape and regional scales is essential for determining overall sink vs. source characteristics. Increases in the frequency and magnitude of permafrost disturbances will create more heterogeneous landscapes throughout the Arctic (ACIA, 2005), which has important implications for vegetation, soil, animal species and the carbon balance at the landscape and regional scale.  1.3 Research objectives The purpose of this research was to examine the effects of active layer detachment slides and retrogressive thaw slumps on terrestrial ecosystem structure and function in the High Arctic. Previous studies have examined the impacts of thermokarst disturbances on species composition and abundance, with a focus on the Low Arctic. In this research project, we examine the effects of RTS and ALDS on community composition and evaluate the impacts of RTS on net ecosystem exchange of carbon with the atmosphere, using eddy covariance and static chambers. To date, no study has measured the impacts of thermokarst disturbance on NEE using eddy covariance.  By analyzing the effects of ALDS and RTS on vegetation and CO2 fluxes on the 16  Fosheim Peninsula, Ellesmere Island, this research extends current understanding of thermokarst disturbance. In addition, I also re-established and sampled study plots that were initially sampled during the summer of 1994 to allow for comparison of short- and long-term vegetation recovery following ALDS.  In quantifying the ecosystem impacts of permafrost disturbances, three specific research objectives and accompanying hypotheses are addressed. These studies are presented as stand-alone manuscripts in Ch. 2-4. The ecological impacts of ALDS and RTS were explored in three specific projects. The research objectives and accompanying hypotheses in each of these projects are detailed below.  Objective 1) Determine if the patterns of community composition and vegetation structure ecosystem recovery within ALDS have remained stable over the past 20 years, by synthesizing historical data collected from ALDS of varying ages in 1994 with data collected after nearly two decades of recovery. Evaluate the successional trajectories predicted by Desforges (2001). Hypothesis 1) Apart from reactivation of recovered disturbances, recovery is consistent with successional trajectories posited by Desforges (2001). Species composition differs based on location within disturbance and age of disturbance, with the oldest disturbances containing vegetation communities most similar to undisturbed adjacent tundra. Select species can be used as indicators of time since ALDS activity. The presence of mosses and shrub species are limited to the oldest disturbances as sufficient time is required before successful recolonization. Objective 2) Quantify the ecosystem effects of permafrost disturbance. Across time, determine the effects of RTS, by comparing fine scale measurements of vegetation and environmental variables from multiple disturbances that have been stable for different periods with undisturbed terrain. Assess the effects of active RTS and impacts of disturbance on productivity by 17  measuring net ecosystem exchange (NEE of CO2). Hypothesis 2a) Recovered ALDS have unique microclimatic conditions, distinguished by significant differences in soil temperatures, soil nutrients, and active layer thicknesses which influence local vegetation composition. Hypothesis 2b) NEE differs between active RTS and undisturbed control tundra, due to the lack of vegetation found within active RTS, resulting in less carbon uptake in disturbances during the growing season.  Objective 3) Evaluate the feasibility of utilizing eddy covariance to measure NEE from permafrost disturbances.  Hypothesis 3) The impact of permafrost disturbance can be measured using eddy covariance. Two techniques were tested: flux partitioning based on wind direction and a dual eddy covariance sampling approach.  1.4 Structure of the thesis This research examines the impacts of disturbance on both ecosystem structure and ecosystem function. I return to sites that were visited 20 years ago and determine changes in composition at ALDS across time (Ch. 2). Sampling over time complements the space for time substitution utilized originally (Desforges, 2000). In Chapter 3 I also examine the spatial variability of RTS in terms of site characteristics and ecosystem fluxes. To determine the impacts of RTS on ecosystem function, I used a portable chamber system and eddy covariance to measure the exchange of CO2 between the surface and the atmosphere. To date, eddy covariance has not been used to quantify the impacts of thermokarst disturbances. The unique methodology presented in this thesis (Ch. 4) allows a comparison of fluxes from disturbed and undisturbed tundra simultaneously to determine these impacts.  A synthesis of the research project and its significance is presented in the final chapter (Ch. 5).  18  1.5 Introduction to the study area  Research was conducted at locations on the Fosheim Peninsula, Ellesmere Island, Nunavut (79° 58’ N 84° 17’ W; Figure 1.1). The flora of the Fosheim Peninsula consists of at least 140 vascular plant species on uniform, weakly alkaline to neutral cryosols derived from marine silts and glacial outwash materials (Edlund et al., 1989). The dominant plant community is Salix-Dryas hummocky tundra, occurring across moderately drained sites with nearly continuous cover and across drier polar desert areas as isolated patches, while wet sedge communities dominate lower slopes and valley bottoms (Edlund et al., 1989). The surficial geology of the region is comprised of sandstones of the Eureka Sound group (Bell, 1996).  The marine limit (upper limit of marine inundation at the end of the last glaciation) is located at approximately 140 m above sea level; areas above the marine limit are dominated by bedrock and till and contain minimal vegetation.    Initial field sampling in 2012 occurred at Hot Weather Creek (HWC). Active layer detachment activity has been widespread in the past at this location and the ages of historical disturbances have been determined from air photo analysis and past field sampling. Select ALDS studied at Hot Weather Creek in 1994 (Desforges, 2001) were revisited and resampled during the 2012 field season. Several of these original ALDS had transitioned into active RTS. Two additional field locations were visited during 2013 and 2014, both located east of HWC, after being surveyed in 2012. The 2013 field site was characterized by chains of active RTS and widespread slumping. The 2014 field site was selected as it was characterized by one isolated RTS, allowing us to test a novel method using two simultaneously operated towers on either side and using footprint modelling for source attribution.  19  1.5.1 Previous research at the study location Comprehensive research has been undertaken on the Fosheim Peninsula, specifically HWC, as it was the location of the High Arctic Global Change Observatory, which was established in 1989 by the Geological Survey of Canada and was active until 1994 (Edlund et al., 2000). Active layer detachment slide activity at Hot Weather Creek was widespread; particularly warm temperatures during the summer of 1988 resulted in the initiation of numerous ALDS (Edlund et al., 1989). Research at this site has focused on the nature of the ground ice, initiation, stabilization, and the hydrological and morphological impacts of disturbance (Lewkowicz, 1990; Kokelj and Lewkowicz, 1998; Kokelj and Lewkowicz, 1999; Robinson, 2000; Lewkowicz and Harris, 2005). Wetland features in the area of HWC may be indicative of subsurface ground ice that supplies these systems with moisture (Edlund et al., 1989). Overall, there is a lack of research on the ecological impacts of high Arctic permafrost disturbances. The only previous research on vegetation change after ALDS on the Fosheim Peninsula focused on a single season of data collection and used space for time substitution to determine vegetation recovery trajectories (Desforges, 2000; Cannone et al., 2010).  20   Figure 1.2 Study location on the Fosheim Peninsula, Ellesmere Island (Worldview-2 image). The study region sampled in 2012 and three study areas with EC tower locations in 2013 and 2014 are noted.          21  Chapter 2: Vegetation succession after permafrost disturbance on the Fosheim Peninsula, Ellesmere Island  2.1  Overview Active layer detachment slides located at Hot Weather Creek, Ellesmere Island, were studied during the growing season of 1994 and revisited during 2012 to determine the short- and long-term impacts on vegetation and ecosystem processes. Distinct vegetation communities exist in differently aged disturbances with unique species defining various zones and ages of disturbance. Species found in disturbed areas include Puccinellia spp. (in the scar zone), Alopecurus magellanicus (in the track zone), and Potentilla hyparctica and Carex rupestris (in the toe zone). Zonal differences illustrate the varying responses of the ecosystem to disturbance and differing modes of recovery. The successional trajectory found corresponds with succession in marginal environments. Disturbances affect site soil characteristics over the long-term, exemplified through difference in soil nutrients (e.g. nitrate), soil moisture, and active layer depths measured between ALDS and undisturbed tundra during the 2012 sampling period. Additionally, multiple sites that were originally sampled as active layer detachment slides at Hot Weather Creek had transitioned into retrogressive thaw slumps. Retrogressive thaw slumps also displayed significantly different community composition, and indicator species included Arctagrostis latifolia, Poa arctica, and Polygonum viviparum. 22  2.2 Introduction Thermokarst disturbances in the Arctic are predicted to increase in frequency and extent with global climate change and increased ground temperatures (ACIA, 2005; Vincent et al., 2011; Kokelj and Jorgenson, 2013; Segal et al., 2016). In the High Arctic, these disturbances commonly take the form of active layer detachment slides (ALDS). ALDS occur when the thawed (or active) layer breaks away from the underlying ice-rich permafrost resulting in a mass movement of soil and vegetation downslope.  With predicted changes in the rates of occurrence of ALDS (Lewkowicz and Harris, 2005), it becomes essential to determine both the short- and long-term impacts of disturbance on the landscape and underlying processes as they may drastically alter these processes. ALDS may transition into another form of permafrost disturbance, retrogressive thaw slumps (RTS), through the exposure of ground ice (Lantuit and Pollard, 2008; Lewkowicz, 1990). RTS remain active until ground ice is depleted or its surface is covered and insulated by soil and vegetation preventing further thaw (Burn and Friele, 1989). On the Fosheim Peninsula, Ellesmere Island, Canada, ALDS and RTS are widespread across the landscape (Kokelj and Lewkowicz, 1999).   Succession following disturbance has been studied across the Arctic (Chapin et al., 1992). Most traditional models of succession are based on temperate or subarctic environments; High Arctic ecosystems are characterized by low productivity and limited vegetation cover and may not be representative of models of succession developed elsewhere (Chapin and Shaver, 1985; Svoboda and Henry, 1987). One model of succession in marginal environments proposed by Svoboda and Henry (1987) incorporates varying levels of environmental severity. In severe environmental conditions, they posit that species depend on individual strategies for survival and slight ameliorations of the environmental conditions in order to become established and grow. 23  Factors of environmental resistance include low nutrient availability, limited moisture, short growing seasons and low temperatures, which discourage the dispersal and establishment of individuals. This is balanced by biological driving forces, intrinsic biological agents including life history traits that allow species to establish and survive. The balance between these two components, environmental resistance (ER) and biological driving forces (BDF), determines the rate and direction of succession. If biological driving forces outweigh environmental resistance, successional change occurs. Three pathways of succession were proposed: 1) directional replacement occurs in low resistance environments, where BDF is greater than ER, such as most temperate environments and in subarctic tundra and polar oases; 2) directional non-replacement occurs when biological driving forces are just greater than or equal to environmental resistance, and expansion of established species occurs without displacing earlier established species. This is common in polar semi-deserts; 3) non-directional non-replacement is found when environmental resistance outweighs biological driving forces, such as polar desert regions, and is characterized by the establishment of few species during periods of lower ER, but they may be eliminated in poor years. When conditions are extremely adverse, a community may undergo retrogression, whereby no establishment is successful and established plants do not survive (Svoboda and Henry, 1987). Returning to disturbances in the High Arctic where successional trajectories were assessed nearly two decades ago allows us to evaluate ecosystem recovery and whether the successional pathway determined by Desforges (2000) has continued.  The objective of this study was to determine the impact of permafrost disturbances, specifically ALDS and RTS, on tundra vegetation, over time, using data from 1994 and 2012 and to analyze environmental characteristics influencing vegetation recovery and to evaluate successional trajectories.  24  2.3 Methods 2.3.1 Study site Research was conducted in 1994 and 2012 at Hot Weather Creek (79° 58’ N, 84° 27’ W), located on the Fosheim Peninsula, Ellesmere Island. The nearest weather station is located at Eureka, approximately 30 km west. The Fosheim Peninsula has a known flora of 140 vascular plants, occurring on uniform, weakly alkaline to neutral cryosols (Edlund et al., 1989). Sandstones of the Eureka Sound group characterize the regional surficial geology (Bell, 1996).  The marine limit is located at approximately 140 m above current sea level; areas above the marine limit are dominated by bedrock and till and contain minimal vegetation (Bell, 1996). Hot Weather Creek (HWC) is located below the marine limit, and vegetation here is characterized as prostrate dwarf shrub tundra (Cannone et al., 2010). The dominant plant community at HWC is Salix-Dryas hummocky tundra, occurring across moderately drained sites with nearly continuous cover and across drier polar desert areas as isolated patches. Lower slope and valley bottoms are comprised of Cassiope heath and wet sedge communities (Edlund et al., 1989).  Due to the presence of ice rich permafrost throughout this region and increased summer temperatures and precipitation (Lesins et al., 2010) over the past twenty years (Fig. 2.1), active layer detachment activity has become widespread at Hot Weather Creek. When the ice rich permafrost degrades these slides can transition into retrogressive thaw slumps (RTS), thus lengthening the duration of active disturbance activity and delaying landscape recovery and re-vegetation (Lewkowicz, 1990).  Ecosystem recovery was analyzed at 10 ALDS at Hot Weather Creek in 1994 (Desforges, 2000). Several (4) of these originally studied ALDS had transitioned into active retrogressive thaw slumps when revisited in 2012. The ages of historical disturbances were determined using 25  air photo analysis and past field sampling (Desforges, 2000), and were classified into young, mature, and old age categories (Table 2.1). These original sites were revisited and resampled during the 2012 summer season to determine the changes that had occurred in the intervening 18 years (Figure 2.2). As a space for time substitution method is usually used to determine the impacts of disturbance over time, this method allowed a determination of changes over time directly in each of the age classes.  Desforges (2000) identified multiple vegetation classes associated with the recovery of ALDS. Young disturbances were colonized by ruderal grasses and forbs (e.g. Puccinellia spp., Braya purpurescens.). This was followed by the establishment of a mid to late sere of grass and forbs (mature disturbances were colonized by Poa spp., Taraxacum hyparctica, Oxyria digyna, Melandrium spp., Potentilla spp.). The undisturbed and oldest disturbances were characterized by shrubs and cushion plants and dominated by Dryas integrifolia, Salix arctica, and Cassiope tetragona.  26   Figure 2.1 Average annual temperature and total annual precipitation measured at the Eureka Weather Station over the past four decades (Environment Canada, 2015).  Research on these disturbances (ALDS and RTS) has predominantly focused on geomorphological characteristics (Lewkowicz & Harris, 2005) and impacts on soil temperature, water quality, soil nutrients, water chemistry, and DOC (Lamoureux & Lafrenière, 2009, Lantz et al., 2009, Kokelj & Lewkowicz, 1998, Kokelj & Lewkowicz, 1999). In the Low Arctic, the impacts of ALDS on vegetation and recovery have been examined (Burn and Friele, 1989; Bartleman et al., 2001; Lantz et al., 2009; Cray and Pollard, 2015). Vegetation studies of disturbed sites in the High Arctic have been limited to one year of sampling and no analysis over time has been completed to date (Desforges, 2000; Cannone et al., 2011), thus returning to the ALDS studied by Desforges allows us to re-evaluate the vegetation recovery of these sites and provide verification of the space-for-time method used in previous studies. −22−20−18−16501001970 1980 1990 2000 2010Mean Avg. Temp  (oC)Precipitation (mm/year)27   Figure 2.2 Study area at Hot Weather Creek with two types of permafrost disturbances (ALDS and RTS) and corresponding sites and location in the Canadian High Arctic (inset). Adapted from Desforges (2000).          N!!!!Hot Weather Creek!ALDS!RTS!Lake!Creek!0! 50! 100! 200 m!RTS5 ALD4 RTS6 RTS7 RTS8 ALD3 ALD1 ALD2 ALD9 ALD10 28  Table 2.1 Age characteristics of ALDS at Hot Weather Creek ALDS           Number Name    Age (1994)      Age (2012)          (Age in 1994) Young   2  ALD1, ALD3   6   24 Mature   1  ALD2    6-19   24-38 Old   3  ALD4, ALD9, ALD10 20 +   38 +  2.3.2 Vegetation sampling (1994 and 2012) 2.3.2.1 Active layer detachment slides ALDS (N=6) were sampled during the 1994 and 2012 growing seasons. ALDS that transitioned into RTS (N=4) were examined separately (see Section 2.3.2.2) and only sites sampled in both years are included in the following analysis. Within each disturbance, three transects running from undisturbed through disturbed to undisturbed terrain were situated along contours perpendicular to the slide direction (Figure 2.3). At each site, separate transects were established to include the three zones found within ALDS; 1) the upper scar zone, from which material is removed, 2) the track zone, where material is transported and 3) the toe zone, located furthest downslope, where material is deposited and compressed. Material is removed from the scar area. In the track and toe zones material, including clumps of vegetation and soil, is deposited. Control plots were located in undisturbed tundra on both sides of the ALDS and were placed at least 1 m away from the edge to prevent the influence of edge effects. A total of 450 plots were located using haphazard stratified sampling, with a trowel blindly thrown every 2-3 m along each transect and plots established where the trowel fell (N (per zone) = 25). Percent cover of vascular plants, mosses, lichens, and litter was visually estimated by the same observer using a 50 cm x 29  50 cm quadrat, subdivided into 5 cm x 5 cm sections. Vascular plants were identified to the species level, and total cover was estimated for mosses and lichens. 2.3.2.2 Retrogressive thaw slumps Four RTS that were previously classified as ALDS in 1994 (Desforges, 2000) were also sampled in this study. Due to the active nature of these disturbances, we were unable to sample from within the center of any RTS. As such, only areas that were accessible and stable (located on the periphery of the disturbance) were sampled. Similar to ALDS sampling, undisturbed vegetation located beyond the areas impacted by disturbance was sampled for comparison using identical sampling procedures detailed above. As RTS were active, undisturbed plots were located at least 5 m beyond the edge of the disturbance to prevent edge effects and plots becoming disturbed.    Figure 2.3 Schematic diagram showing distinct zones of an active layer detachment slide (scar, track, toe, and undisturbed) and corresponding transects used for vegetation and site sampling. Adapted from Desforges (2000).   1 2 3 4 1) Scar 2) Track 3) Toe 4) Undisturbed 30  2.3.3 Comprehensive site characterization (2012) 2.3.3.1 Soil moisture and thaw depth Soil moisture was measured weekly throughout July 2012 at all plots where vegetation was measured using a calibrated HydroSense II Soil Water Time Domain Reflectometry (TDR) sensor (Campbell Scientific Canada, Edmonton, AB) with 12 cm rods. Following precipitation events, measurements were delayed for 24 hours. Thaw depth was measured by inserting a thin metal probe into the ground until the depth of refusal, and was measured in conjunction with soil moisture.  2.3.3.2 Soil nutrient availability  Ion exchange membranes (PRS Probes, Western Ag, Saskatoon, SK) were installed at three sites on 28 June 2012 and retrieved 6 August 2012. Pairs of cation and anion exchange membranes (N=48) were installed vertically to a depth of 10 cm in three recovered active layer detachments, two young ALDS (ALD1 and ALD3), and one old ALDS (ALD4).  Four pairs of IEMs were installed across each of the scar, track, and toe zones, and in control areas beyond the disturbance adjacent to zone. During 2013, nutrient probes were installed at one site, ALD1, in the same areas as the previous year to determine inter-annual differences. Probes were installed on 28 June 2013 and retrieved on 26 July 2013. Nutrient availabilities (including Total N, NO3-, NH4+, Mg2+, Ca2+, and K+) were calculated for each burial period in micrograms/10 cm2/burial length (days).   2.3.3.3 Soil temperature Soil temperature loggers (HOBO Pendant (Model UA-002-64), Onset Co., Bourne, MA, USA) were installed within the upper 5 cm of soil on June 28, 2012 and ran throughout the growing season at 1 min sampling frequency. The data from each logger were downloaded and loggers 31  were re-launched to continue over the winter at a reduced frequency (5 min).  Loggers were installed in the scar, track, and toe of three ALDS (2 young and 1 old) and in the control area (N=12). One logger failed through the winter in the track zone of one ALDS and two loggers were lost within another ALDS (scar and undisturbed).  2.3.4 Statistical analysis 2.3.4.1 Time comparison All plots were pooled based on age classification (identified in 1994) and year collected. To compare data from 1994 and 2012, the two extremes were analyzed, specifically age of disturbance in 1994 (young and old) and zone (scar vs. toe), as well as the control areas. Four ALDS in 1994 had transitioned into RTS by 2012 and were omitted from the analysis, as these disturbances were active. Analysis was constrained to focus on active layer detachments slides, to allow for a comparison between the historical and current dataset. 2.3.4.2 Vegetation composition and cover All data analysis and graphing was completed using R statistical language, Version 3.1.2 (R Core Team, 2014). Dissimilarity matrices were computed based on the Bray-Curtis distances method. Ordinations were computed using the metaMDS function in the vegan package (Oksanen et al., 2012: Version 2.2). A three dimensional ordination displayed the least stress and was repeated 100 times to reach the best solution for each nonmetric multidimensional scaling (NMDS) (Legendre and Legendre, 1998).  Analysis of similarities (ANOSIM) was used to test for compositional differences among groups (using the vegan package (Oksanen et al., 2012)). Pairwise ANOSIM tests were used to determine compositional differences between 1994 and 2012 datasets and disturbed and undisturbed tundra. The RANOSIM value ranges from 0 to 1 and indicates similarity between groups, with values of <0.25 representing groups that cannot be 32  separated (Clarke and Gorley, 2001). Similarity percentages (SIMPER) analysis (in the vegan package) was used to determine species contributing to largest group dissimilarities. Indicator species were identified using the indval function within the indicspecies package (Caceres et al., 2011: V1.7.5). Indicator species analysis calculates an indicator value (!"!!) for species i in group j based on relative abundance (specificity: !!") and relative frequency (fidelity: !!"): !!" =    !!"!!!         Equation 1 !!" =    !!"!!            Equation 2 !"!" =   !!"  ×  !!"  ×  !""       Equation 3 where !!" is the mean cover of species i within group j, !!!  is the sum of mean cover of species i in all groups, !!" is the number of samples in group j occupied by species i, and !! is the total number of samples in group j (McCune and Grace, 2002). !"!" ranges between 0 and 100 and strong indicators are those with !"!" > 25.  Total live cover was computed based on individual values of species cover for the 1994 and 2012 datasets. Fixed effect 2-way analyses (Age x Zone) of variance (ANOVAs) were used to test for differences and interactions across all sites, zones, and ages of disturbance for total cover and nutrient concentrations.  2.3.4.3 Soil moisture and active layer depth Measurements of soil moisture and active layer thickness were compared using 1-way ANOVAs using a grouping cluster as the variable incorporating disturbance classification (disturbed, undisturbed), site age (young, mature, old), and zone (scar, track, toe). Soil moisture measurements were aggregated to calculate an average soil moisture and variability throughout 33  the season based on zone and location. Early season (late June) active layer depth and maximum active layer depth (early August) were compared across sites. Any datasets that did not meet statistical assumptions or normality and equal variance were log-transformed. Post hoc Tukey tests with Bonferroni correction were used to determine differences among means.  2.3.4.4 Soil temperature Soil temperatures during the growing season were compared within sites by calculating soil temperature deviation, Vi (1) to determine deviations from the mean soil temperatures (A) measured at each site, i at each time step (t). Deviation was calculated by comparing values from each sensor with the average of all sensors from all sites (Eq.4): V! t = A! t -­‐A(t)(t)       Equation 4    2.4 Results 2.4.1 Comparison of historical and modern data 2.4.1.1 Vegetation community composition and cover Non-metric multidimensional scaling (NMDS) ordinations showed compositional differences between disturbed plots sampled in 1994 and 2012 (Figure 2.4). Variation in disturbed and undisturbed vegetation was evident, however there was clustering based on disturbance at each site. Significant differences in vegetation communities exist in some zones at some sites of disturbance (scar and toe) between 1994 and 2012 sampled data year (Table 2.2). Pairwise ANOSIM comparisons indicate compositional differences between disturbed sites sampled in 1994 and in 2012 in some zones. Using SIMPER analysis, the youngest disturbances (as classified in 1994) were compared between 1994 and 2012, and the scar differed as 2012 disturbances were characterized with greater cover of Puccinellia spp. while the toe differed with 34  increased cover of Salix arctica and Puccinellia spp., and a decrease in Elymus alaskanus over the 20 year period. When the oldest disturbances were compared, in the scar region Dryas integrifolia and Salix arctica increased in cover in 2012 while Cassiope tetragona, moss, and lichen decreased. The toe zones differed with a decrease in Dryas integrifolia and Salix arctica, and an increase in Cassiope tetragona and Stellaria longipes in 2012 (Appendix A.2).  Figure 2.4 NMDS ordination of sites comparing 1994 and 2012 vegetation data of total cover per species per quadrat from all ALDS (stress=0.12, k=3, nonmetric R2=0.99, linear R2=0.92). Each point represents a quadrat: colours represent different sites, open symbols are 1994 samples while closed symbols are 2012 samples.             ï2 ï1 0 1 2ï1.5ï0.50.5NMDS1NMDS2ALD1ALD3ALD4ALD9ALD10ALD1 ALD3 LD4 LD9 LD10  ALD1(young) ALD3(young) ALD4 (old) ALD9 (old) ALD10 (old)      -2      -1     0    1  2      -1.5          -0.5     0.5   NMDS2 NMDS1 35  Table 2.2 Results of ANOSIM comparison between plots sampled in 1994 plots and resampled in 2012. Significant compositional differences are noted in bold.  Comparison Site Zone RANOSIM P-value 1994 vs. 2012 ALD1 (young) Scar -0.073 0.709 Undisturbed 0.066 0.193 Toe 0.366 0.031 Undisturbed 0.374 0.015 ALD3 (young) Scar 0.829 0.002 Undisturbed 0.011 0.376 Toe 0.485 0.001 Undisturbed n/a ALD4 (old) Scar 0.062 0.214 Undisturbed 0.141 0.128 Toe 0.536 0.005 Undisturbed 0.081 0.290 ALD9 (old) Scar 0.604 0.002 Undisturbed 0.083 0.203 Toe 0.175 0.076 Undisturbed 0.198 0.152 ALD10 (old) Scar 0.468 0.011 Undisturbed 0.087 0.123 Toe 0.125 0.179 Undisturbed 0.050 0.328  Results from the 2-way ANOVA indicate significant differences in total plant cover based on zone (F(7, 268) = 33.06, p<0.001) and an interaction between zone and year (F(7, 268) = 4.51, p<0.001) (Figure 2.5). Within the youngest age category, vegetation cover was lower in the scar and toe compared to undisturbed tundra (Figure 2.5). This differed from the oldest age category, as total plant cover was greater in disturbed zones (in both the scar and the toe) in both 1994 and 2012. Post hoc Tukey tests revealed vegetation cover was lower in the scar of young disturbances while cover in the scar of old disturbances was greater than controls in 1994. In 2012, resampled sites still showed decreased cover in young disturbances and increased cover in old disturbances, however, these cover differences were not significant.  While there was 36  variability among the control plots in both years, there were no significant differences between years or among sites.  Figure 2.5 Total vegetation cover based on zone (scar and toe), age (young and old), disturbance (undisturbed and disturbed) and year (1994 and 2012). Data are means with SE bars. Different letters indicate significant differences among all categories.     2.4.2 Comparison of plant communities in disturbances sampled in 2012  Disturbances sampled in 2012 of all age categories (which were originally classified as young, mature, and old in 1994) and all zones (scar, track, toe, and corresponding controls) were compared using ordination analysis. NMDS revealed compositional differences between disturbed and undisturbed plots (Fig. 2.6). ANOSIM also indicated differences in composition within the 2012 data when compared to undisturbed plots in adjacent tundra (Table 2.3). These differences were evident in the young disturbances, where the scar (ALD1), track (ALD1, ALD3) and toe (ALD3) displayed significant compositional differences from the undisturbed tundra. ANOSIM found no significant differences in the one mature disturbance (ALD2). Among the oldest disturbances, ALD4 contained significantly different vegetation in all zones, ALD10, contained similar vegetation in all zones when compared with undisturbed terrain and Old Young0255075025507519942012scar toe scar toeZoneTotal Cover (%)ControlDisturbedOld Young0255075025507519942012scar toeZoneTotal Cover (%)ControlDisturbedOld Young0255075025507519942012scar toeZoneTotal Cover (%)ControlDisturbedU disturbed Disturbed e!d e! c d e! c  ! c d ! c d ! b c  ! b c  ! b c  !  b c  ! a b c  ! a b c  ! a b c  ! a b c  ! a b   ! a    !37  ALD9 was characterized by a different track zone. SIMPER analysis indicated that species contributing to dissimilarity included Salix arctica and Dryas integrifolia; the youngest disturbances had the lowest cover of these two species when compared to undisturbed tundra and older disturbances (Appendix A.3). The youngest disturbances differed compositionally from undisturbed tundra due to increased Puccinellia spp. cover in disturbed areas, while Carex rupestris contributed to the greatest dissimilarity in the mature disturbances. The oldest disturbances contained Cassiope tetragona and Stellaria longipes, which are most dissimilar from undisturbed tundra.   Figure 2.6 NMDS ordination of data collected in 2012 (stress=0.13, k=3, nonmetric R2=0.98, linear R2=0.91). Different colours represent different sites, open circles are undisturbed plots and closed circles are disturbed plots. An additional age category (mature) and zone (track) were included in this analysis.     −2 −1 0 1 2−1.5−1.0−0.50.00.51.0NMDS1NMDS2ALD1ALD2ALD3ALD4ALD9ALD10ALD1(young) ALD2(mature) ALD3(young) ALD4(old) ALD9(old) ALD10(old) 38   Table 2.3 ANOSIM results comparing disturbed vegetation of several ALDS of varying ages with undisturbed tundra based on data collected in 2012. Significant differences are noted in bold.  Comparison Site (age) Zone RANOSIM P-value Disturbed vs. Undisturbed ALD1 (young) Scar 0.941 0.002 Track 0.739 0.004 Toe 0.068 0.286 ALD3 (young) Scar 0.152 0.123 Track 0.528 0.004 Toe 0.475 0.009 ALD2 (mature) Scar 0.193 0.105 Track 0.041 0.301 Toe 0.318 0.065 ALD4 (old) Scar 0.424 0.011 Track 0.716 0.011 Toe 1.000 0.003 ALD9 (old) Scar 0.110 0.178 Track 0.754 0.002 Toe 0.273 0.053 ALD10 (old) Scar 0.095 0.742 Track 0.1025 0.229 Toe 0.014 0.316  Indicator species were found for disturbed and undisturbed areas (Table 2.4). In the youngest age class of disturbance Puccinellia spp. was the indicator species for the scar zone while Potentilla hyparctica was the indicator for the toe zone. In the mature age category, the disturbed track and toe were found to have unique indicator species: Pedicularis capitata and Carex rupestris, respectively. In the old age category the toe in the disturbance was characterized by Stellaria longipes, Cassiope tetragona, and Alopecurus alpinus. Only two undisturbed areas contained indicator species, Erigeron compositus adjacent to the mature scar and Minuartia rubella and Lesquerella arctica adjacent to the mature toe.       39  Table 2.4 Indicator species analysis results from 2012 vegetation data (Σ probabilities=8.701, ΣIV=4.23, # Significant IV = 11) Species                  Cluster*       Indicator Value     p value Puccinellia  spp.          2  14.83        0.046   Potentilla hyparctica    6  15.66        0.033 Erigeron compositus    7  23.78      0.015  Pedicularis capitata    10  17.89      0.032 Potentila vahliana       11  21.50      0.006 Minuartia rubella        11  13.46      0.041 Lesquerella arctica     11  13.11      0.047 Carex rupestris           12  13.85      0.034 Stellaria longipes        18  23.65      0.005 Cassiope tetragona    18  17.65      0.010 Alopecurus alpinus      18  15.17      0.032 *Cluster codes as follows: 1) young scar undisturbed, 2) young scar disturbed, 3) young track undisturbed, 4) young track disturbed, 5) young toe undisturbed, 6) young toe disturbed, 7) mature scar undisturbed, 8) mature scar disturbed, 9) mature track undisturbed, 10) mature track disturbed, 11) mature toe undisturbed, 12) mature toe disturbed, 13) old scar undisturbed, 14) old scar disturbed, 15) old track undisturbed, 16) old track disturbed, 17) old toe undisturbed, 18) old toe disturbed.      Species richness and diversity (Shannon Index) were also determined for all 2012 ALDS (Table 2.5). Species richness was significantly different based on clusters (identical to those in Table 2.3), F(17,220) = 6.98, p<0.001. Post hoc Tukey tests (Appendix A; Table A1) revealed differences among the mean values of richness with the fewest species found in the scars of the youngest disturbances. Young track zones also had few species, while there were no richness differences found in the toe zone of the youngest disturbances. No other significant differences were apparent between undisturbed and disturbed zones in each age category.  Species diversity (as measured using the Shannon H Index) was also significantly different, F(17,220) = 7.16, p<0.001. The youngest disturbed track and scar areas were significantly less diverse than their respective controls. However, diversity did not differ between disturbed and undisturbed areas in any other age or zone class.  40  Table 2.5 Community characteristics of ALDS sampled in 2012, with mean (± SE) vascular plant species richness and Shannon’s H Index values.  Age Zone Treatment Richness Shannon's H Young (in 1994) scar undisturbed 5 (±0.5) 1.03 (±0.11) disturbed 1 (±0.4) 0.10 (±0.10) track undisturbed 6 (±0.4) 1.37 (±0.10) disturbed 3 (±0.4) 0.52 (±0.12) toe undisturbed 5 (±0.4) 1.19 (±0.08) disturbed 6 (±0.7) 1.19 (±0.16) Mature (in 1994) scar undisturbed 4 (±0.6) 0.79 (±0.17) disturbed 5 (±0.8) 1.09 (±0.18) track undisturbed 3 (±0.8) 0.46 (±0.20) disturbed 5 (±0.4) 1.10 (±0.22) toe undisturbed 5 (±0.9) 0.98 (±0.21) disturbed 4 (±0.5) 1.05 (±0.07) Old (in 1994) scar undisturbed 4 (±0.3) 0.99 (±0.08) disturbed 5 (±0.4) 1.01 (±0.08) track undisturbed 3 (±0.3) 0.51 (±0.08) disturbed 4 (±0.4) 0.88 (±0.11) toe undisturbed 4 (±0.5) 0.78 (±0.12) disturbed 5 (±0.5) 1.13 (±0.10)  2.4.3 Retrogressive thaw slump vegetation analysis Analysis of total vegetation cover in the four RTS that had formed in previous ALDS (2 young ALDS and 2 mature ALDS in 1994) showed no differences between disturbed and undisturbed plots (F(1,32) = 0.48, p = 0.49). Additionally, there were no differences among the sites (F(3,32) = 0.42, p = 0.75). However, vegetation composition was impacted by disturbance. Indicator species analysis (Table 2.6) revealed Dryas integrifolia was an indicator species in undisturbed tundra while disturbed tundra was characterized by three indicator species (Polygonum viviparum, Arctagrostis latifolia, and Poa arctica). Ordination analyses supported composition differences as disturbed and undisturbed vegetation communities clustered separately. ANOSIM results indicate disturbed and undisturbed vegetation at each site were well separated (Table 2.7).  41  Table 2.6 Indicator species for active RTS (Σ probabilities=5.33, ΣIV=2.36, # Significant IV = 4) Species      Zone     Indicator Value p value            Dryas integrifolia  Undisturbed  74.5  0.005 Polygonum viviparum  Disturbed  54.8  0.019 Arctagrostis latifolia  Disturbed  53.4  0.049 Poa arctica   Disturbed  53.3  0.031  Table 2.7 Results of ANOSIM comparison of species composition between undisturbed and disturbed tundra at RTS previously sampled as ALDS. Significant differences noted in bold.  Site RANOSIM P-value RTS5 0.936 0.005 RTS6 0.384 0.03 RTS7 0.1469 0.099 RTS8 0.308 0.007   Figure 2.7 NMDS ordination of vegetation measured in active RTS (blue) and corresponding undisturbed (red) areas (stress = 0.13, k=3, non metric fit R2 = 0.98, linear fit R2 = 0.86).  −2 −1 0 1 2−1.5−1.0−0.50.00.51.0NMDS1NMDS242  2.4.4 Site characteristics of disturbances sampled in 2012 2.4.4.1 Soil moisture One way ANOVA revealed differences in mean soil moisture based on combinations of zone and age (see clusters listed below Table 2.1; F(17,217)=15.95, p<0.001). Soil moisture was greatest in the scar area of young disturbances and differed from the undisturbed area (Fig. 2.8). All zones in disturbances of the mature and old age categories had lower mean soil moisture values than the scar and track zones of young disturbances; however, there were no differences between the disturbed and undisturbed areas. In the undisturbed sites, soil moisture tended to be greater in plots adjacent to the toe of the disturbances. Variability in soil moisture was analyzed through ANOVA of the standard deviation of the mean soil moisture throughout the season, and greater variability was found in the young disturbances (F(17,217)=9.90, p<0.001), specifically within the scar and toe areas (Fig. 2.9).    Figure 2.8 Mean soil moisture (±SE) measured from scar, track, and toe zones of young, mature, and old disturbances during the 2012 growing season. Top panel displays soil moisture measured from adjacent undisturbed areas while bottom panel shows disturbed plots. Different letters above bars indicate significant differences in soil moisture across all categories (p<0.05). Young Moderate Old01020300102030UndisturbedDisturbedscar track toe scar track toe scar track toeZoneSoil Moisture (%)a! a b!a b c! a b c d!b c d! c d! c d e!  c d e! c d e! d e! d e!  d e! d e! d e!  d e!  e! e!  e! Mature l  43   Figure 2.9 Variability (standard deviation) in soil moisture measurements taken throughout the 2012 growing season from the scar, track, and toe areas of young, mature, and old ALDS. Top panel displays undisturbed values while bottom panel represents measurements from disturbed areas. The data are deviation from mean soil moisture values of seasonal measurements. Different letters above bars represent significantly different soil moisture variability across all categories (p<0.05).   2.4.4.2 Thaw depth Early season thaw depth thickness (µ) showed differences among disturbance, age, and zone (F(17,210)=5.35, p<0.001). Post hoc Tukey tests indicated that the shallowest depth occurred in the disturbed scar of the youngest age category (µ=45.6±4.5 cm), however this mean was not significantly different from the immediately adjacent undisturbed tundra. Younger disturbed areas were shallower than older more recovered disturbances and undisturbed areas had thicker active layer depths than disturbed plots (Fig. 2.10). Very few differences among classes and zones were found in active layer thickness (µ) at the end of the season (F(17,210)=2.80, p<0.001). Post hoc tests showed that mean thickness in the track of young ALDS was thicker (µ=88.2 cm ±10.3) than in undisturbed areas in the same zone (µ=73.1 cm ±2.1; Fig. 2.11). The scar of the oldest disturbances had the shallowest mean Young Moderate Old051015051015UndisturbedDisturbedscar track toe scar track toe scar track toeZoneSoil Moisture Deviationaa b!a b!a b c!a b c d!a b c d!b c d! c d!  c d! c d!  d!  d!   d!  d!  d!  d!  d!  d! Mature l  44  active layer thickness (µ=71.5 cm ±8.5). No significant differences were found between young and mature disturbances.  The amount of thaw (calculated as the difference in thaw depth between early season and late season thaw) was also compared to determine if disturbance impacts seasonal thaw. No differences in the magnitude of thaw were found (F(17,210)= 1.65, p=0.054).   Figure 2.10 Early season active layer thicknesses in zones (bottom panel) and adjacent undisturbed (top panel) areas of ALDS of three general age classes in 2012. Data are means ± SE and different letters indicate significant differences between means across all categories (p<0.05).  Young Moderate Old02040600204060UndisturbedDisturbedscar track toe scar track toe scar track toeZoneEarly Season Active Layer (cm)a!a b!a b!a b c!a b c d!a b c d!a b c d!a b c d! b c d! b c d! b c d! c d!a b c! b c d! b c d! b c d! b c d! d! Mature ld 45    Figure 2.11 Maximum active layer thickness in disturbance zones and undisturbed areas of ALDS of three different general age classes in 2012. Data are means ± SE and different letters indicate significant differences between means across all categories (p<0.05).  2.4.4.3 Soil temperature During the winter (Fig. 2.12), the coldest 5 cm soil temperatures were reached within the toe of ALD1 reaching -38 °C and ALD3 reaching -42 °C. However soil temperatures within the track of the old disturbance (ALD4) were the coldest (reaching at minimum of -30 °C) at that site.  Over the non-growing season, the maximum difference in mean daily soil temperatures between zones within ALDS was present in ALD3, however the scar and undisturbed locations had similar soil temperatures at this site.  Deviation in growing season soil temperature differed by disturbance class and zone within the disturbance (Fig. 2.13). Within all ALDS, during the peak growing season (DOY~205) deviation was similar regardless of location (scar, track, toe, and undisturbed tundra). However, the early season and late season at this ALD1 (young disturbance) were characterized by warmer soils (and greater deviation) at the measurement location in the scar. In ALD3 (also a young disturbance) similar soil temperatures were found throughout the peak Young Moderate Old02550750255075UndisturbedDisturbedscar track toe scar track toe scar track toeZoneEarly Season Active Layer (cm)a! a! a!a!a b!a b! a b!a b!a b!a b!a b!a b!a b!a b!a b!a b!a b!b!Maximum Active Layer Depth (cm)! Mature ld 46  growing season, however, an opposite pattern was found in that the scar was characterized by lower more variable soil temperatures, which were especially striking during the early season. Within ALD4 (an old disturbance), the scar and track had similar soil temperatures, while the undisturbed tundra had warmer soils and the toe was characterized by colder soils.   Figure 2.12 Daily mean soil temperature over 1 year in zones of select disturbances (ALD1 (young), ALD3 (young), and ALD4 (old)) measured over July 2012 – July 2013 from one sensor per zone.  Ald1Ald3Ald4ï40ï20020ï40ï20020ï40ï20020Jul 2012 Oct 2012 Jan 2013 Apr 2013 Jul 2013DateSoil Temperature (degC)scar track toe undisturbedSoil Temperature (°C) 47   Figure 2.13 Soil temperature deviation (e.g. difference from mean of all sensors) from select disturbances (ALD1 (young), ALD3 (young), and ALD4 (old)) during the growing season of 2012 (DOY=day of the year).   2.4.4.4 Soil nutrients A two-way ANOVA was used to determine differences in plant available nutrients (NH4+, NO3-, Ca2+, K+, and Mg2+) based on zone (scar, track, toe, control) and treatment (disturbed and undisturbed tundra) (Table 2.8; Fig. 2.14). Controls from scar, track, and toe were pooled as no significant differences were found in nutrient concentrations from each set of controls. Nutrient availability was then compared based on zone within disturbance (including control) (Table 2.8). Total N (calculated as the sum of NO3- and NH4+) differed significantly between disturbed and undisturbed tundra (treatment) and among locations within the ALDS (zone). Post hoc Tukey tests revealed total N within the scar was greater than those in the toe, track, and control.  The differences in Total N were due to differences in NO3-.as there were no significant differences found in NH4+. No differences among means were found in the availability of Ca2+ or K+, but Ald1Ald3Ald4ï202ï202ï202180 190 200 210 220DOYSoil Temperature Variability (degC)scartracktoeundisturbedSoil Temperature Deviation (°C) 48  Mg2+ concentrations were greater in the disturbance (Fig. 2.14). Variability was also greater for K+ and Mg2+ from disturbed areas. Table 2.8 ANOVA table for comparisons of nutrient availability of five ions measured using ion exchange membranes in three ALDS. Zone includes scar, track, and toe areas of the disturbance while treatment includes disturbed or undisturbed tundra. Values in bold represent significant differences as measured by the F statistic.  Nutrient F p value Total N treatment zone NO3-  F(1,42) = 15.229 F(4,42) = 15.307  <0.001 <0.001  treatment F(1,42) = 31.939 <0.001 zone F(4,42) = 12.303 <0.001 NH4+   treatment F(1,42) = 1.2136 0.277 zone F(4,42) = 1.4222 0.243 Ca2+   treatment F(1,42) = 0.5649 0.456 zone F(4,42) = 2.5361 ≤0.05 Mg2+   treatment F(1,42) = 8.4394 0.0058 zone F(4,42) = 0.6657 0.619 K+   treatment F(1,42) = 3.9948 ≤ 0.05 zone F(4,42) = 0.5264 0.717    Ion exchange membranes that were installed the following season (2013) at one young ALDS (ALD1) for approximately the same burial period were compared with 2012 values from this ALDS. Values were consistent with those measured in 2012 and no significant differences in any nutrient concentrations were found. Significant differences between the disturbed and undisturbed zones in the same ions were also found in 2013, supporting the 2012 results.   49   Figure 2.14 Box plots of soil nutrient availabilities across 4 zones of n=3 ALDS during the 2012 growing season as measured by ion exchange membranes. Availabilities of all nutrients are in µg 10 cm-2 40 days-1; boxes show the 25th and 75th percentiles, dots are the outliers, and horizontal lines are medians. Different letters above the bars indicate significant differences between means (p<0.05)  510152025control scar toe trackLocationNH4undisturbed        scar              toe              track ZONE K+  Ca2+ NH4+ Mg2+  NO3- 0 200 400 600 200 100 2000 1000 0 50 100 5 10 15 20 25 NUTRIENT AVAILABILITY (µg 10cm -2 40 days -1) a a a a a b b b a a a a a ab ab b a a a a 50  2.5 Discussion 2.5.1 Successional response ALDS sampled in 1994 were revisited in 2012 to evaluate long-term recovery patterns and succession in High Arctic tundra. When vegetation sampled in 2012 was compared with the  1994 dataset, we found similar recovery patterns in both years. For example, the differences in total live cover were essentially the same in 2012 as in 1994. In disturbances that were classified as young in 1994, decreased total cover was found in the scar zone in both years as compared to undisturbed tundra. In contrast, an increase in total cover was found in the scar region within old disturbances in both sampling years, likely as a result of the slight depression in these areas that allowed deeper snow and greater water availability. Other important consequences of deeper snow include reduced thaw depth and winter bud protection. However, while the patterns found between disturbances and undisturbed terrain in total cover present in 2012 were similar to those measured in 1994, the differences were no longer significant between the disturbances and surrounding controls, indicating further development of vegetation in the disturbances over the past 20 years.  Differences in plant cover among the disturbance zones were still found indicating the long lasting impacts of ALDS.  Desforges (2000) concluded that a combination of directional replacement and directional nonreplacement models of succession explain recovery at this site (Svoboda and Henry, 1987). In 2012, we found that each age was characterized by different indicator species. This indicated replacement over time is occurring and supports the directional replacement model of succession. This is also consistent with significant compositional differences between disturbances sampled in 1994 and 2012, indicative of species replacement over time. 51  Other evidence of directional replacement succession has been found in glacier forelands at Alexandra Fiord, Ellesmere Island (Jones and Henry, 2003). Forelands were characterized by the progression of vegetation from directional non-replacement and non-directional non-replacement succession was also found at select glaciers. Bliss and Gold (1994) found evidence of directional species replacement on a succession of beach ridges in high Arctic coastline on Devon Island. At these sites, vegetation progressed in wet regions from Puccinellia spp., Dupontia spp., Carex spp., Salix spp. to cushion plant lichen communities and in drier environments from Puccinellia spp. to rosette herb to cushion plant lichen communities (Bliss and Gold, 1994).  As space is not limited in recovering young ALDS, competition is unlikely a driving force or dominant successional control, as in found in other environments (Svoboda and Henry, 1987). No differences were found in between disturbance class and zones in species richness apart from the scar of the youngest age category. Few species were present in all plots, with significantly fewer found in the scar area characteristic of primary succession. Estimates of total cover indicate that it rarely reaches 100% and bare ground is present throughout. Matthews (1992) proposed that early succession is controlled by environmental conditions whereas biological factors control the later stages. Within young scar regions, early invaders able to tolerate the harsh environmental conditions, including Puccinellia spp., are found. Other species appear over time as conditions become more favourable for establishment, therefore biological processes such as growth and litter input, which is able to retain moisture and trap seeds, may be important in improving the sites for plant recruitment (Desforges, 2000; Robbins and Matthews, 2010).  52  Despite the progression in site recovery, exemplified by reduced differences in total cover over time between the ALDS and undisturbed tundra, unique species were still present within different zones and ages of disturbance. Early successional species (Poa arctica, Arctagrostis latifolia, and Polygonum viviparum) were found to be prominent within the active RTS. Poa arctica has been found to be an early invader in disturbed areas where compacted soils limit root penetration (Forbes, 2001). Arctagrostis latifolia is a nitrophilous grass able to thrive on enhanced nutrients, establish from seeds and ramets, and produce root systems that can grow deep into the soil (Bliss and Wein, 1972). These species are also found in thaw slumps in the Low Arctic (Cray and Pollard, 2015).  Polygonum viviparum, which are able to produce bulbils rapidly, are found within disturbed terrain impacted by thaw slumps, and are also associated with ALDS on the Fosheim Peninsula appearing as a rapid invader 2-3 years following disturbance activation (Cannone et al., 2011).  After stabilization of ALDS, the youngest scars are characterized by Puccinellia spp., which have been noted as early successional species in both the High Arctic and Low Arctic (Bliss and Gold, 1994; Cray and Pollard, 2015). Puccinellia spp. can be found within a wide variety of habitats, with varying levels of soil moisture and enhanced nutrients and are often found as colonisers on landslides (Aiken et al., 2007). Within ALDS, compacted floors of the disturbance may limit root penetration (Forbes et al., 2001), therefore rhizomatous grasses are found here, including Puccinellia spp. and Alopecurus magellanicus, another common invader in disturbed areas (Babb and Bliss, 1974).  Moving from the base of the toe toward the headwall, vegetation proceeds from regions characterized by Potentilla hyparctica, which is tolerant of dry conditions and varying levels of organic matter, to Carex rupestris, a species common in dry barrens (Aiken et al., 2007). Over 53  time, the species composition changes and the toes are characterized by species such as Alopecurus magellanicus, Stellaria longipes, and dwarf shrubs such as Cassiope tetragona. Cassiope tetragona has been found to be a late successional species on glacial forelands (Jones and Henry, 2003).  The time required for recovery in the High Arctic can take up to 300 to 1000 years, while succession may proceed more rapidly in the Low Arctic (Tishkov, 1986; Svoboda and Henry, 1987; Chapin et al., 1992; Jones and Henry, 2003). However, RTS in the Low Arctic have been found to require multiple centuries to return to pre-disturbance vegetation composition (Cray and Pollard, 2015). Our results indicate that, while vegetation development has occurred over 20 years, considerable more time is still required for successional recovery in ALDS at Hot Weather Creek as undisturbed vegetation differs from zones located even in the oldest ALDS, which have been estimated to have occurred more than 40 - 70 years ago.  We found that the local distribution of species across the disturbance gradient is likely a response to microenvironmental conditions that arise following ALDS. ALDS and RTS alter the ground thermal and moisture regime, which influences the nutrient supply and affects species composition and abundance for decades (Bartleman et al., 2001; Burn and Friele, 1989). Becker et al. (2015) found that disturbed terrain associated with thermokarst induced by melting ice wedges resulted in altered species composition and species turnover and may have resulted in a shift from polar desert to wetland tundra in areas of ice rich permafrost.  In the active RTS, we were only able to identify one disturbed zone for analysis. This differs from the work of Bartleman et al. (2001) as areas within their studied RTS stabilized at different periods of time, creating a chronosequence of disturbance recovery that was sampled and classified based on year of stabilization. Areas that are currently active and that may 54  potentially stabilize could be visited in the future to determine the long-term progression of vegetation recovery within these RTS.  2.5.2 Soil characteristics Nutrients, including NO3-, Mg2+, and K+ were elevated in scar zones of the ALDS. Increased nutrient availability has been found in other permafrost disturbances (Kokelj and Lewkowicz, 1999; Lantz et al., 2009; Ukraintseva, 2000; Ukraintseva and Leibman, 2008). Across the Fosheim Peninsula, Kokelj and Lewkowicz (1999) noted the presence of salt efflorescence (surface salt accumulations) in areas impacted by ALDS and other forms of disturbance, as soluble materials previously stored in permafrost of these marine sediments are released upon thaw. At one ALDS, they found concentrations of Na+ in surface runoff reached nearly 5 g l-1. Salt concentrations this high can result in decreased plant growth and increased plant mortality (Srivastava and Jefferies, 1995). Despite the presence of graminoids able to tolerate saline soils (including Puccinellia spp.) enhanced salinity in some areas affected by disturbances may be too elevated even for salt tolerant species. In the Low Arctic, retrogressive thaw slumps were associated with elevated sulfate, calcium, and nitrate availability (Lantz et al., 2009). In Siberia, Ukraintseva (2000) found increased concentrations of K+, Ca2+, Mg2+, Cl-, SO22-, and PO43- in soil water that interact with plant root systems in ALDS. Nutrient enrichment of soil, water and vegetation resulted in enhanced productivity enabling the spread of Salix species at ALDS in Siberia (Ukraintseva and Leibman, 2008). Given the variability in nutrient concentrations among the zones and successional sequences, potential long-term impacts of these permafrost disturbances on soil characteristics are difficult to decipher.  There was variability in soil temperatures among zones, especially in the winter and shoulder seasons. Over the growing season, soil temperatures varied with position within 55  disturbed and the undisturbed terrain however, these differences were minimal. We lacked full soil temperature records due to missing instrumentation. Other studies of active layer detachment slides have found soil temperatures elevated in disturbed regions (Bosquet, 2011) due to increased snow accumulation. Variation in snow cover imposed by microtopography in the ALDS likely plays an important role in soil temperatures during the non growing season.  Elevated soil moisture was found in the scars of young disturbances, influencing vegetation cover. We also found greater plant cover in the depressions of scars, which is likely partly due to greater snow depth. Bartleman et al. (2001) found vegetation recovery trajectories to be influenced by moisture supply, as vegetation found in areas that were supplied with flow from the headwall was different from vegetation found when this flow stopped. Taken together, this suggests that variation in soil moisture has an important influence on vegetation recolonization in these permafrost disturbances.  2.5.3 Permafrost thaw implications Over 18 years between 1994 and 2012, 4 of the 10 ALDS that were characterized and sampled in 1994 transitioned to RTS. As these sites are currently active, ecosystem recovery is delayed and recovery trajectories may deviate from those of ALDS due to the active transport of sediment from the headwall, which prevents plant establishment. Material that may have been deposited in the toe area of ALDS is subsequently removed when they develop into RTS due to the flow of material associated with melting ground ice. As these active RTS border Hot Weather Creek, this material flows into the creek.  Due to the quantity of material removed (McRoberts and Morgenstern, 1974; Dallimore et al., 1996; Lewkowicz and Kokelj, 2002; Lewkowicz, 2007), vegetation recovery would be predicted to follow the trajectory of the scar of ALDS as it is also characteristic of primary succession. Transitioning of disturbances from active layer detachment 56  slides into retrogressive thaw slumps delays recovery and may alter the recovery trajectory through modified site characteristics, including soil moisture. Since slumps usually remain active for longer than ALDS, ecosystem fluxes of carbon dioxide will also be impacted for longer periods of time (see Chapters 3 and 4).  Lewkowicz and Harris (2005a) predict an increase in ALDS across the Fosheim Peninsula if increased temperatures occur in conjunction with low cloud cover. As ALDS transition into RTS, the implications of increased disturbance potential and positive climate feedbacks are significant. Successional trajectories may be altered. Organic carbon currently stored within the permafrost could be released with greater permafrost thaw and become available for microbial decomposition (Schuur et al., 2008). As depth of permafrost thaw increases, so does ecosystem respiration (Hicks Pries et al., 2013). These disturbances have direct effects on energy, snow trapping, and herbivory. ALDS and RTS are likely to modify the carbon balance in many Arctic ecosystems thus understanding their short- and long-term impacts on ecosystems is essential for predicting how these systems will change.  2.6 Conclusion Vegetation has continued to develop in stable ALDS over the past 20 years, especially in scar and toe zones, and there are no longer major differences in total vegetation cover between disturbed and undisturbed zones. Unique species are still present within different scar and toe zones reflecting differing modes of succession and different ages of disturbance. For example, following 20 years of recovery, toe areas in the oldest age category (> 38 years) still have different vegetation from the surrounding undisturbed environment indicating compositional differences. Total cover and species diversity was lower in young disturbances in both the scar and toe zones, but both had increased in older disturbances, indicating a directional replacement 57  pathway that had been predicted by Desforges (2000). Soil nutrients, including Total N, NO3-, and Mg2+ were different among the disturbed zones, with concentrations elevated in the scar zone compared to other zones. Differences also existed in soil moisture and active layer depths, indicating long-lasting impacts of permafrost disturbance on these high Arctic tundra ecosystems. Greater recovery in the scars and toes is enhanced by soil moisture and nutrient availability in these zones relative to surrounding tundra. If predicted increases in frequency and magnitude of disturbances occur across the Fosheim Peninsula, this will have widespread ecosystem impacts. With the transition of multiple ALDS into RTS, we predict that in the future, RTS will hinder vegetation recovery through the continual disturbance of the landscape.   58  Chapter 3: Variability of the impacts of active retrogressive thaw slumps on net ecosystem exchange of CO2 in the Canadian High Arctic  3.1 Overview Permafrost disturbances are widespread across the Fosheim Peninsula, Ellesmere Island, Canada, where they take the form of active layer detachment slides (ALDS) and retrogressive thaw slumps (RTS). These disturbances likely affect the net ecosystem exchange of CO2 between the tundra and atmosphere, but the impacts have not been studied in the High Arctic. During the 2013 growing season impacts were assessed for three different RTS located in areas with different vegetation types. In each of the three RTS, vegetation composition, site environmental characteristics, and growing season carbon dioxide (CO2) exchange were measured and compared to the adjacent undisturbed tundra. Eddy covariance and static chamber measurements were used to determine net ecosystem exchange (NEE) and ecosystem respiration (Re), respectively. Two eddy covariance towers were established at the boundary between disturbed and undisturbed terrain and wind direction was used to separate NEE from RTS and the surrounding undisturbed terrain.  Eddy covariance measurements indicated an overall decrease in NEE in disturbed areas. At one site, this decrease shifted the system from a weak net sink (-0.05±0.02 g C m-2 day-1) to a weak net source (+0.07±0.04 g C m-2 day-1) over the entire growing season. Vegetation cover was greatly decreased in the active disturbances, relative to the surrounding tundra and cover affected the overall impact of disturbance on CO2 fluxes. Disturbances increased Re compared to surrounding undisturbed tundra. The three RTS showed differential impacts on soil temperature 59  (increases in two RTS and a decrease in one RTS compared to undisturbed tundra) and exhibited an increase in soil moisture (+14 %) and in the contribution of trace elements (specifically nitrogen). In summary, RTS have a profound impact on soil climates, vegetation, and carbon exchange, yet responses differ.  3.2 Introduction Permafrost disturbances are expected to increase in frequency and magnitude with predicted climate change (ACIA, 2005; Vincent, 2011; Kokelj and Jorgenson, 2013). In the High Arctic, permafrost disturbances frequently take the form of active layer detachment slides (ALDS) and retrogressive thaw slumps (RTS). RTS occur when ground ice is exposed, perhaps after an ALDS, and ongoing ground ice ablation causes slump growth and thawed material moves downslope; soil and vegetation are removed further and further upslope as the ground ice melts and the slump retreats (Lantuit and Pollard, 2008). Slumps remain active until ground ice is depleted or further thaw is prevented by falling blocks of soil and vegetation (Burn and Friele, 1989).  Ecosystem responses to these permafrost disturbances have focused on hydrological impacts (Kokelj and Lewkowicz, 1998; Kokelj and Lewkowicz, 1999; Lamoureux and Lafrenière, 2009). In aquatic ecosystems, physical and chemical changes of sediment and water following slumping impact invertebrates, macrophytes, and diatoms, altering composition and abundance (Chin et al., 2016; Mesquita et al., 2010; Moquin et al., 2015; Thienpont et al., 2013). The ecological impacts of active layer detachment slides have been analyzed at a few sites in the High Arctic and changes to the physical environment following a disturbance event have been shown to influence vegetation recovery (Bosquet, 2011; Cannone et al., 2010, Cassidy, 2011 Desforges, 2001). The impacts of retrogressive thaw slumps on vegetation development have 60  focused on Low Arctic ecosystems (Bartleman et al., 2001; Burn and Friele, 1989; Cray and Pollard, 2015; Lantz et al., 2009). Thaw slumps modify vegetation communities with changes persisting for centuries (Cray and Pollard, 2015).  Due to the harsh conditions (low temperatures, short growing season) in the High Arctic, recovery is predicted to occur more slowly than in the Low Arctic (Svoboda and Henry, 1987). The removal of vegetation in the disturbance can reduce albedo by up to 50% (Babb and Bliss, 1974). A reduction in albedo and removal of the soil organic layer can increase soil temperatures and deepen the active layer, however the active layer may also increase due to the disturbance itself (Bliss and Wein, 1972; Auerbach et al., 1997; Lantz et al., 2009). An increase in the active layer provides plant roots with more space in the soil layer (Bliss and Wein, 1972). Plants that have deep roots, including species such as Calamagrostis canadensis and Eriophorum angustifolium are common invaders in some disturbed tundra sites due to their ability to take advantage of extra soil volume and ability to disperse (Chapin and Shaver, 1981).  Carbon fluxes between the surface and the atmosphere can be quantified by measuring net ecosystem exchange (NEE) of CO2, the difference between CO2 released to the atmosphere due to ecosystem respiration (Re) and CO2 removed from the atmosphere by gross primary production (GPP), and can be measured by eddy covariance (EC) (e.g. Emmerton et al., 2015; Humphreys and Lafleur, 2011; Lafleur et al., 2012). Negative values of NEE represent uptake by the ecosystem, and positive values represent emissions to the atmosphere. NEE is influenced by temperature, moisture and light levels among other factors and differs among plant community types (Baldocchi, 2008). Measurements of NEE in tundra systems during the growing season have found they are typically carbon sinks, but can easily shift to carbon sources due to changes in temperature, moisture, and the water table (Oberbauer et al., 2007; Vourlitis et al., 2000). 61  Chamber studies have found small but steady losses of CO2 during the winter, with experimental warming increasing sink potential through a longer growing season (Welker et al., 2004). At Alexandra Fiord, Ellesmere Island, experimental warming impacted NEE differently based on soil moisture, due to respiration differences between wet and dry sites (Welker et al., 2004). Across a latitudinal gradient, warming tended to increase respiration, with the greatest increases found in dry ecosystems (Oberbauer et al., 2007). Previous studies examining Arctic NEE have found large inter-annual variability within and among sites. This variability has been substantial enough to shift the growing season from a CO2 sink to CO2 source (Griffis and Rouse, 2001; Kwon et al., 2006; Merbold et al., 2009). However, there are few measurements of NEE from Arctic tundra sites, which combined with the inherent variability in Arctic tundra ecosystems makes it difficult to determine relative NEE across the Arctic and their contribution to regional and global fluxes (Lafleur et al., 2012). Vast amounts of carbon are stored in permafrost soils, with Hugelius et al. (2014) estimating these soils contain 50% of worldwide below ground organic carbon. These estimates are likely an underestimation (by up to a factor of two) due to measurement difficulties and uncertainty regarding carbon storage in cryoturbated soils (Hugelius et al., 2013). Schuur et al. (2008) found organic carbon that is unfrozen can be released as permafrost thaws. If permafrost thaws, this formerly frozen carbon has the potential to be released to the atmosphere through microbial respiration.  In this study, I assess how the changes that result from permafrost disturbances in the High Arctic influence spatial heterogeneity in carbon exchange. I focus on changes in vegetation, the unique soil characteristics, and the fluxes of carbon that are associated with thaw slumps at different stages of recovery.  The research objective was to determine the variability of 62  vegetation composition and cover, active layer depth, soil thermal and moisture characteristics, soil nutrients, and CO2 fluxes between the three RTS and adjacent undisturbed tundra.  3.3 Materials and methods 3.3.1 Study area Research was conducted on the Fosheim Peninsula, western Ellesmere Island (79° 58’ 21” N; 84° 17’ 41” W, 100 m asl). This area was selected as the study site due to the widespread nature of thaw slumps in this area. Multiple sites with RTS were identified during summer 2012 and monitored during summer 2013. Vegetation in the area is characterized as dwarf-shrub-graminoid tundra; 140 species of vascular plants are found on the Fosheim Peninsula on uniform, weakly alkaline to neutral cryosols (Edlund et al., 1989).  Geology is sandstone of the Eureka Sound group (Bell, 1996) with marine silts and sands varying in thickness above the bedrock (Robinson and Pollard, 1998). The upper limit of marine inundation at the end of the last glaciation is approximately 140 m above sea level (Bell, 1996), and limited vegetation is found above this level. This region is an area of ice rich permafrost. Due to the nature of the permafrost and increased summer temperatures and precipitation over the last 20 years, there has been an increase in the occurrence of permafrost disturbances (Lewkowicz and Harris, 2005).  Previous research on the Fosheim Peninsula has focused on the initiation, stabilization, and hydrological impacts of ALDS (Kokelj and Lewkowicz, 1998; Kokelj and Lewkowicz, 1999; Lewkowicz, 1990; Lewkowicz and Harris, 2005; Lewkowicz, 2007; Robinson, 2000), however limited work has focused on the ecological impacts, especially on vegetation and CO2 fluxes (Cannone, 2010; Desforges, 2001).  63  3.3.2 Site selection Three RTS were selected for detailed classification and analysis, based on vegetation characteristics and the nature of the disturbances (Table 3.1). Air photo analysis using photos taken in 1949 show these sites were not visible on the landscape and indicate that these disturbances formed after 1949. Therefore, these disturbances could be 50+ years old, however, as numerous disturbances formed during 1988 and 1998, these disturbances may have occurred during these seasons (Lewkowicz and Harris, 2005).  Two of these disturbances were situated on east facing slopes, and are referred to as sites RTS-1 and RTS-2 (Figure 3.1) and were selected as they are both part of a larger string of disturbances. Eddy covariance (EC) flux towers were established at RTS-1 and RTS-2 on the periphery of the disturbance (Figure 3.2; Figure 3.3). Both EC flux towers were located on the northern edge of the RTS; the area north of both towers was undisturbed tundra, while that south was influenced by the RTS and in part by the other disturbances in the vicinity.  Table 3.1 Site characteristics of sampled RTS Site Type of disturbance Extent (m2) Description* Slope outside disturbance (%) Slope inside disturbance (%) RTS-1 Active RTS 31645 Multi-disturbance  9 12 RTS-2 Active RTS 32304 Multi-disturbance  6 8 RTS-3 Stable RTS (recently reactivated) 17562 Isolated 1 5 *Description: indicates whether disturbances are located independently and are not connected to other RTS (isolated) or are part of a larger chain of disturbances (multi-disturbance).   Two sampling transects were established at RTS-1 and RTS-2. The two transects at each RTS were each 60 m in length and were established perpendicular to one another with the EC flux tower located at their intersection on the undisturbed periphery of the RTS (Figure 3.2). In 64  this way, each transect encompassed both undisturbed and disturbed tundra. Plots (5 m x 5 m) were established 15 m from the edge of the disturbance to minimize edge effects at 15 m intervals. The active center of these disturbances was not sampled due to accessibility issues. Increased activity and flows throughout the latter part of the season inundated a segment of one disturbed plot at RTS-2.  Figure 3.1 Study location, Fosheim Peninsula, Ellesmere Island in 2013 with three sites labeled and outlined (in white), looking west (a). RTS-1 and RTS-2 are located on the periphery of two severely disturbed landscapes, while RTS-3 is an isolated disturbance, with a recovered revegetated region. Ground view of RTS-1 towards the west (b).   65  A third disturbance, RTS-3, was located on a west-facing slope. A portion of the headwall within this disturbance had been insulated preventing further thaw and so was inactive and a revegetated (recovered) zone was selected and compared with the undisturbed surrounding terrain. This zone spanned the northern and southern edges of the current active headwall. The southern recovered region covered an area of 1250 m2 (with a maximum length of 65 m and a maximum width of 35 m), the northern recovered region occupied 835 m2 (maximum length of 45m and maximum width of 30 m), and both regions accounted for approximately 12% of the total area of disturbance. At RTS-3, undisturbed and disturbed (and recovered) terrain was sampled using four transects (length = 50 m). Two transects were established across the disturbance (across the southern and northern recovered areas, at a minimum distance of 10 m from the recovered headwall) and the other two were positioned across the surrounding undisturbed terrain. Six plots (5 m x 5 m) were located in each zone (recovered and undisturbed), at a distance of at least 3 m from the current active region and 3 m from the recovered headwall, respectively. Plots were located at distances of 5 m along the transects.   66   Figure 3.2 Site sampling at locations RTS-1 and RTS-2 indicating transect locations relative to EC tower, sensors, and sampling locations.  3.3.3 Vegetation sampling Vegetation was sampled at each plot along transects using a 50 cm x 50 cm quadrat separated into 5 cm squares. Vegetation composition and abundance (cover) were estimated inside five quadrats in each plot: one at the centre of the plot, and four times surrounding the plot centre, located along the transect and orthogonal to it, located by throwing a trowel and placing the quadrat where the trowel fell with the corner of the quadrat placed next to the end of the tip of the trowel. Vascular plant cover was visually estimated for each species, and total cover of moss and lichen was determined.  Disturbed Undisturbed 0 10 m Plot EC Tower Collar Soil T Logger Transect Upslope Downslope 67  3.3.4 Soil characteristics Active layer depth was measured using a thin metal probe, which was inserted into the ground until the depth of refusal. Active layer depths were measured at the end of the field season on 27 July 2013 at each site. At each slump, active layer depth was measured 18 times while outside the slump active layer depth was measured 24 times in the undisturbed tundra.  Soil temperature was measured every minute using Hobo Pendant Temperature/Light Data Loggers (Model UA-002-64, Onset Computer Corp., Bourne, MA, USA) installed at a depth of 5 cm below ground. These sensors were deployed on 2 July 2013 and retrieved on 29 July 2013. A total of 18 loggers were deployed at disturbed (d) and undisturbed (c) plots at RTS-1 and RTS-2 (Nd=8, Nc=10). Loggers were positioned in the centre of each plot shown in Figure 3.2 and an additional logger was placed at the EC tower.  Four loggers were deployed at RTS-3 (Nd=2, Nc=2), with one placed along each transect, in the centre plot. All loggers were placed in the centre of the plots along transects.  Soil moisture was measured manually using a calibrated Time Domain Reflectometry (TDR) probe (HydroSenseII Soil Water TDR, Campbell Scientific Inc.) with 12 cm rods. Soil moisture was recorded at three locations within each plot along each transect and three locations between plots at 10 day intervals throughout the season (Nd=18, Nc=18). Additional measurements were taken at both flux towers and 10 m upslope and downslope of the tower (N=9).  Reflectivity was estimated using a handheld pyranometer (LI200X, Campbell Scientific Inc., Logan, UT, USA), mounted at a height of 1 m to allow for instrumentation to be leveled for all measurements. Measurements were made upwards and downwards during cloud free conditions on 10 July 2013 (between 10:00 – 14:00) at each plot along the transects.  68  Soil nutrient availability was measured using ion exchange membranes (PRS Probes, Western AG, Saskatoon, SK, Canada). Four cation and four anion membranes (each 17.5 cm2 membrane was mounted in a plastic holder 15 cm x 3 cm x 0.5 cm) were inserted 10 cm into the soil in the centre of each plot.  These were deployed on 2 July 2013 and retrieved 26 July 2013 for a 24 day sampling period. At RTS-1 and RTS-2, probes were placed along the transects for a total of Nd = 6 and Nc=9 sites. At RTS-3, probes were placed in the centre of each plot (Nd = 6 and Nc=6).  3.3.5 Portable CO2 efflux chamber system  On 30 June 2013, 28 opaque PVC collars (10 cm diameter, Area = 78.5 cm2, Depth = 6 cm) were installed at sites RTS-1 (N=8), RTS-2 (N=8), and RTS-3 (N=12), for a total of 14 collars in both disturbed and undisturbed tundra. The collars were inserted 4 cm into the soil to minimally disturb soil and vegetation and they extended 2 cm above the ground surface. Collars were spaced across transects at each plot (Figure 3.3). Due to the variability of vegetation cover, collars were placed on both vegetated and unvegetated tundra within the disturbed and undisturbed tundra.  A non-steady state portable chamber system similar to Jassal et al. (2005) was used to measure soil respiration from each collar using an opaque chamber (Figure 3.3). The measurement head was a PVC chamber with a volume of 1.4 x 10-3 m3 (height: 15.6 cm, diameter: 10.7 cm). The chamber head was placed on each collar for 2-minute intervals. A foam gasket was used to seal the connection between collar and chamber head. A pump (flow rate: 600 cm3 min-1) circulated air from the chamber head into a portable battery operated infrared gas analyzer (IRGA) (LI-840, LI-COR Inc., Lincoln, USA) and back into the chamber through a 69  closed circuit. The IRGA determined CO2 mixing ratios ([CO2] in ppm) and water vapour concentrations at a temporal resolution of 1 Hz during each run.   Figure 3.3 Portable closed chamber system (a) and corresponding collar (b) and eddy covariance (EC) tower (c) used to measure Re and NEE, respectively.  Respiration (Re) was calculated (as Fc) from Δ[CO2]/Δt (linear regression over 2 min, discarding the first 10 seconds), using Eq. 5: !" =    !  !!!   ![!"!]!!         Equation 5      where ρ is the molar density of air (mol m-3) calculated from measured air temperature, D is dilution considering [H20], Δ[CO2dry]/Δt is the rate of change of CO2 mixing ratio over time (µmol mol-1 s-1), and V and A are chamber volume and area, respectively. Measurements were made on two days during the season and were taken between 10:00 and 16:00 on each day to minimize diurnal changes in light and temperature. 70  3.3.6 Eddy covariance measurement of NEE We used an eddy covariance (EC) tower with conditional sampling using wind direction to measure NEE from disturbed and undisturbed tundra. Two towers were located at RTS-1 and RTS-2, and each tower was established 2 m from the edge of the disturbance (and 50 m and 60 m away from the headwall, respectively). Both EC systems were mounted on tripods and ran continuously from 26 and 27 June 2013, respectively, until 29 July 2013 (Figure 3.3). Each system included: an ultrasonic anemometer (CSAT-3, Campbell Scientific Inc., Logan, UT, USA) and a co-located open-path infrared gas analyzer (IRGA) (LI-7500, LI-COR Inc., Lincoln, NE, USA, installed tilted by 30°) both established at 1.75 m; a temperature and humidity sensor (HMP, Campbell Scientific Inc.) at 1.3 m; a net radiometer (NR Lite, Kipp & Zonen B.V., Delft, The Netherlands; height: 1.2 m; a quantum sensor (SQ-110, Apogee Instruments Inc., Logan, UT, USA; height: 1.2 m; and all data were stored on a data logger (CR1000, Campbell Scientific Inc.) at a sampling rate of 10 Hz.  Both towers were placed with the IRGA and sonic anemometer parallel to the slump edge to avoid any flow distortion effects from preferred wind sectors (data were removed in sectors with flow distortion due to terrain). Conditional sampling using wind direction allowed measurement of fluxes from disturbed and undisturbed tundra. Friction velocity (u*) thresholds of 0.10 m s-1 were applied to remove data under low-turbulence conditions. All IRGAs (EC tower and portable chamber) were calibrated prior to the field season using a two-point calibration in the lab against standards from the Greenhouse Gas Measurement Laboratory (GGML), Meteorological Service of Canada (using a zero gas and span gas of known mixing ratio). Partitioning of fluxes was appropriate as disturbed and undisturbed fluxes were characterized by similar conditions (i.e. average temperatures and wind speeds) throughout the measurement period.  71  3.3.7 Flux data processing  Molar fluxes of CO2 (Fc in µ mol m-2 s-1) were computed in EddyPro (V5.1.1, LI-COR Inc.) with a missing sampling allowance of 30%. Fc was calculated over a 30 minute averaging interval using double rotation for tilt correction, block average detrending, contact time lag detection, and WPL corrections using mixing ratios (Burba et al., 2012). Data quality controls based on the flagging system proposed by Mauder and Foken (2004) were used and data categorized as level 2 was discarded. At low temperatures measurements from open-path LI-7500 CO2 sensors have been shown to overestimate uptake rates of CO2 (Burba et al., 2012). Despite the study location, measurements were made during the warmest month of the year, with mean temperatures of 8°C across our study location. As such, surface heating issues have been shown to be minimal (especially with tilted instruments) and thus we did not perform corrections on flux data (Cassidy et al., 2016; see Appendix B for a detailed discussion of this correction).   When winds were parallel to the edge of disturbance and terrain could not be separated as disturbed or undisturbed, calculated NEE (Fc) values were unstable, and fluxes from unstable sectors were removed. Fluxes with a difference from the daily average greater than 5 standard deviations of the 30 min values (of the same day) were removed. We averaged half hour NEE data into hourly fluxes. If one of the half hour values was not available and from the same segment (disturbed/undisturbed), the hourly value was then based on the single 30 min flux measurement. We filled remaining hourly gaps using the following methods: a) gaps of less than 2 hours from the same segment were filled using linear interpolation; and b) gaps greater than 2 hours were filled using aggregate averaging over a rolling five day window selecting the same time of day and same segment. The dataset was comprised of 90% original data and 10% gap filled data, as 173 of the 1723 data points were modeled.   72  3.3.8 Statistical analysis  All statistical analysis was completed using the R programming language (Version 3.1.2 (R Core Team, 2013) to analyze differences between sites (RTS-1, RTS-2 and RTS-3) and the impact of disturbance (two zones, disturbed and undisturbed). To determine differences in community composition of vegetation among sites (RTS-1, RTS-2, and RTS-3) and between disturbed and undisturbed tundra (zone), non-metric multidimensional scaling, a multivariate ordination technique based on Bray Curtis distance matrices derived from percent cover data, was used. A two dimensional ordination displayed the least stress and was repeated 100 times to reach the best solution (Legendre and Legendre, 1998).  ANOSIM was used to test for differences among groups (using the vegan package (Oksanen et al., 2012)).   Indicator species analysis was used to determine species characteristic of each zone and site.  Indicator species analysis calculates an indicator value (!"!") for species i in group j based on relative abundance (specificity: !!") and relative frequency (fidelity: !!") !!" =    !!"!!!         Equation 6 !!" =    !!"!!            Equation 7 !"!" =   !!"  ×  !!"  ×  !""       Equation 8 where !!" is the mean cover of species i within group j, !!!  is the sum of mean cover of species i in all groups, !!" is the number of samples in group j occupied by species i, and !! is the total number of samples in group j (McCune and Grace, 2002). !"!" ranges between 0 and 100 and strong indicators are those with !"!" > 25.  Total cover was calculated as the sum of percent cover of live green material in each plot.  73  Differences in environmental variables and CO2 fluxes (total vegetation cover, soil moisture, active layer depth, soil temperatures) were tested using two-way ANOVA (site x zone) with Bonferroni correction. Data were transformed to meet normality assumptions, when necessary. Post hoc Tukey tests were used to conduct pairwise comparisons. Soil nutrient availability data were tested with a non-parametric Kruskal-Wallis test to determine the effect of site and zone.  3.4 Results 3.4.1 Micrometeorological conditions There were minimal differences in micrometeorological conditions at both flux tower locations (Figure 3.4). Air temperatures rose steadily throughout the season and reached a peak on 16 July 2013 (with average temperatures 7.9°C at RTS-1 and 8.6°C at RTS-2) before cooling down after 24 July 2013. Vapour pressure deficit also rose, albeit at a more modest rate, with a maximum deficit on 16 -17 July 2013 before decreasing at the end of the sampling season consistent with air temperature decreases. PAR and net radiation displayed consistent diurnal fluctuations throughout the season (Figure 3.4).  74   Figure 3.4 Conditions throughout the 2013 growing season, averaged between RTS-1 and RTS-2. Air temperature is plotted over the study period in the top panel, photosynthetically active radiation is plotted in the panel second from the top, net radiation in the panel second from the bottom, and vapour pressure deficit is plotted in the bottom panel.  051015180 190 200 210DOYAir Temp (deg C)02505007501000180 190 200 210DOYPAR (micromol/m^2s)0100200300180 190 200 210DOYNet Radiation (Wm2)0.00.51.0180 190 200 210DOYVapour Pressure Deficit (kPa)Air Temperature (°C) PAR (µmol m-2 s-1) Net Radiation (W m-2) Vapour Pressure Deficit (kPa) 75  3.4.2 Vegetation Significant differences were found in plant community composition among the sites (Figure 3.5). ANOSIM indicated differences in composition based on site and disturbance status (ANOSIM R=0.4312; p=0.001, based on 999 permutations). Pairwise comparisons at each site also indicate compositional differences (Table 3.2).  Figure 3.5 NMDS results with site scores represented by circles coloured based on disturbance status (stress=0.09, k=2, nonmetric fit R2=0.992, linear fit R2=0.989).  Table 3.2 ANOSIM pairwise comparisons between disturbed and undisturbed tundra at each site. Significant compositional differences are indicated in bold. Site RANOSIM   P-value RTS-1 0.452 0.001 RTS-2 0.392 0.001 RTS-3 0.739 0.001  Indicator species analysis (Table 3.3) supported the differences in composition among the sites and between zones (disturbed and undisturbed tundra) as shown using NMDS and ANOSIM. Undisturbed tundra was characterized as dwarf shrub graminoid tundra, with shrubs UndisturbedDisturbed–101–2 –1 0 1 2NMDS1NMDS276  Salix arctica and Dryas integrifolia found at these sites. Disturbed areas were dominated by rhizomatous grasses and sedges. At RTS-1, undisturbed tundra was characterized by Dryas integrifolia and Carex nardina, however, in adjacent disturbed tundra vegetation cover was characterized by grasses, Poa glauca and Alopecurus magellanicus. At RTS-2, undisturbed tundra was characterized by Salix arctica and moss, however no unique vegetation was characteristic within the disturbance at RTS-2, which was largely unvegetated. At RTS-3, undisturbed tundra was characterized by Salix arctica, Dryas integrifolia, Puccinellia spp., and lichen. The stabilized slump at RTS-3 contained notable amounts of Carex aquatilis and Alopecurus magellanicus. Common species were found at multiple sites, as Dryas integrifolia was found at RTS-1 and RTS-3 and Salix arctica was present at RTS-2 and RTS-3, however differences were found in the overall community composition. Disturbed tundra was characterized by species that are able to recover and recolonize quickly and tolerate site conditions present in retrogressive thaw slumps.  Table 3.3 Indicator species by site Site Disturbance Classification Species IV P-value RTS-1 undisturbed Dryas integrifolia 79.2 0.001 Carex nardina 61.7 0.005 disturbed Poa glauca 31.2 0.007 Alopecurus magellanicus 25.0 0.026 RTS-2 undisturbed Salix arctica 74.1 0.001 moss 48.1 0.004 RTS-3 undisturbed Salix arctica 77.8 0.003 lichen 66.7 0.009 Dryas integrifolia 55.6 0.033 Puccinellia phyganoides 55.6 0.029 stabilized Carex aquatilis 100 0.001 Alopecurus magellanicus 66.7 0.007  77  Total cover (Figure 3.6) was not significantly different between disturbed and undisturbed tundra (F(1,878)= 2.52, p=0.12) however differences in cover were found among sites (F(2,10534)=15.08, p< 0.001).  The interaction between zone (disturbed and undisturbed tundra) and site (RTS-1, RTS-2, RTS-3) was significant F(2,9574)=13.71, p<0.001, as at RTS-1 and RTS-2, total cover decreased within the disturbance, however, this decrease was only significant at RTS-2 while at RTS-3, the reverse pattern was found, as a significant increase in cover characterized disturbed terrain. The lack of overall significance of disturbance is based on stabilization of RTS-3 and the active nature of the tower sites.   Figure 3.6 Total vegetation cover at each site. Unshaded green bars represent undisturbed tundra, while shaded brown bars represent disturbed tundra. Data are means with ± SE of the mean, and different letters above bars indicate significant differences.  3.4.3 Thermal regime Maximum active layer depth ranged from 50– 81 cm. Overall, when maximum active layer was compared, no differences were found between disturbed and undisturbed tundra (Figure 3.7), (F(1,36)=2.10, p=0.16). However, there were significant differences in active layer depths among sites (Table 3.3; F(2,36)=7.59, p<0.01), with the greatest depths found at RTS-2 (depth: 020406080b c a b  c d a c d SiteTotal cover (%)UndisturbedDisturbedRTS 1 RTS 2 RTS 378  71±2 cm). There was also a significant interaction between disturbed and undisturbed tundra and site (F(2,36)=6.00, p<0.01). Spatial variability in active layer depths was evident across sites, as undisturbed tundra at RTS-2 had a significantly deeper active layer (depth: 72±1 cm) than the undisturbed terrain at RTS-1 (depth: 58±2 cm) and RTS-3 (depth: 60±1 cm). The active layer depths in the disturbance were slightly greater at RTS-1 and RTS-3, and slightly shallower at RTS-2, but these differences were not significant.  Figure 3.7 Active layer depth among sites and zones. Bars show the mean ± SE of the mean. Different letters above bars indicate statistically significant differences.  Soil temperatures over the growing season were impacted by disturbance (Table 3.4; F(1,14286) = 251.9, p<0.001) as increased temperatures were found within disturbed soils at RTS-1 and RTS-2 (9.3°C ± 0.1 and 8.6°C ± 0.1, respectively) when compared to adjacent undisturbed soils (7.2°C ±0.1 and 8.3°C ±0.1, respectively). At RTS-3, disturbed soils displayed lower temperatures (6.5°C ±0.1) than undisturbed soils (8.2°C ±0.1). This divergence in the direction of soil temperatures in active and stabilized slumps is evident in Figure 3.8. Soil 0255075100a c ca a b cb c b c SiteActive layer depth (cm)UndisturbedDisturbedRTS 1 RTS 2 RTS 379  temperatures were significantly different among sites F(2,14286)=140.7, p<0.001), with the coolest soil temperatures found at RTS-3 and the overall warmest temperatures at RTS-2.  Table 3.4 Summary of soil characteristics by site and treatment (mean ± SE) Site Disturbance Classification Active Layer Depth (cm) Soil Moisture (%) Soil Temperature (°C) RTS-1 Undisturbed 58.1 (±2.5) 26.9 (±1.0) 7.2 (± 0.1)  Disturbed 71.2 (± 2.6) 38.5 (± 0.8) 9.3 (± 0.1) RTS-2 Undisturbed 72.4 (± 1.4) 25.9 (± 0.9) 8.3 (± 0.1)  Disturbed 67.9 (± 4.8) 35.8 (± 2.4) 8.6 (± 0.1) RTS-3 Undisturbed 60.3 (± 0.9) 16.9 (± 0.6) 8.2 (± 0.1)  Disturbed 60.8 (± 2.9) 41.3 (± 0.7) 6.5 (± 0.1)   Figure 3.8 Mean soil temperature (-5 cm) throughout July 2013 at each site (n=4). Plot shows hourly means based on 5 minute samples.   3.4.4 Moisture regime Soil moisture was greater (38.5 ± 0.9 %) in the disturbed tundra than the undisturbed terrain (24.0 ± 0.6 %) at all three sites (Figure 3.9; F(1,370)=218.5, P<0.01). Differences in soil 51015510155Jul 05 Jul 15 Jul 22 Jul 291015DateSoil temperature (degC)UndisturbedDisturbedRTS 1RTS 2RTS 3Soil temperature (°C) 80  moisture were also found among sites, as greater soil moisture was measured at RTS-3 than RTS-1 and RTS-2 (Table 3.4; F(2,370)=1.96, p=0.14). There was a significant interaction between site and disturbance (F(2,370)=19.8, p<0.01) as soil moisture measured within disturbed soils at RTS-3 was greater than in RTS-2. Soil moisture at RTS-1 was not significantly different thatn RTS-2 or RTS-3. Undisturbed soils in RTS-1 and RTS-2 had similar soil moisture levels, however these undisturbed soil moisture values were greater than those measured within the undisturbed tundra at RTS-3.  Figure 3.9 Mean soil moisture (±SE) in July at three sites (n=115 per disturbance). Different letters above bars represent statistically significant differences.  The average (±SE) reflectivity measured on 10 July 2013 at RTS-1 was 0.18 ± 0.00 in undisturbed tundra and 0.15 ± 0.01 in the disturbance. At RTS-2, undisturbed tundra had a slightly greater reflectivity (0.19 ± 0.01) than at RTS-1, while reflectivity in the disturbed portion of this site (RTS-2) (0.21 ± 0.01) was notably greater than at RTS-1.  With all measurements 01020304050a b c c bdaSiteSoil moisture (%)UndisturbedDisturbedRTS 1 RTS 2 RTS 381  combined, average reflectivity was very similar between the undisturbed tundra (0.18 ± 0.004) and the disturbed tundra (0.18 ± 0.01). Reflectivity was not measured at RTS-3.  3.4.5 Nutrient availability Summary values of nutrient availabilities as measured using IEMs are presented in Table 3.5. When macronutrients were compared between disturbed and undisturbed soils, total nitrogen was greater in disturbed soils. Differences in total N were driven by NO3-, which was also higher in the disturbances. NH4+ availability did not differ (Table 3.6).  Site variability was apparent in differences in the concentrations of K+ and NO3-, with greater availability of K+ and NO3- found at RTS-1 and RTS-2, compared to RTS-3. Concentrations of micronutrients including SO42-, B, Cu+, Fe2+, and Mg2+ were impacted by disturbance as increased concentrations of these nutrients were found in disturbed soils. In addition, increased concentrations of Cu+ were found in disturbed soils at RTS-2 and RTS-3 but not RTS-1.  Table 3.5 Summary of soil nutrient availability at three study locations (mean ± SE), RTS-1, RTS-2, and RTS-3 in disturbed (d) and undisturbed (c) tundra. All concentrations are µg cm-2 x 24 days.  RTS-1 RTS-2 RTS-3 Nutrient Disturbed Undisturbed Disturbed Undisturbed Disturbed Undisturbed N 44.2±21.6 4.1±0.5 16.3±5.5 10.0±1.0 8.7±1.9 6.8±1.8 NO3- 41.8±21.9 1.4±0.3 14.3±5.6 7.3±1.0 6.2±2.2 5.0±1.6 NH4+ 2.5±0.3 2.4±0.6 2.2±0.3 2.7±0.4 2.7±0.3 2.0±0.3 Ca2+ 895.8±271.9 1425.0±195.1 963.8±156.9 1106.2±55.3 1442.7±108.1 1312.7±130.6 Mg2+ 473.7±31.4 393.9±24.5 593.5±56.2 464.6±27.0 670.5±61.0 371.2±33.3 K+ 104.0±20.9 59.7±11.4 105.8±14.8 115.6±12.5 47.2±9.9 48.8±7.1 PO33- 1.4±0.5 1.1±0.2 0.9±0.2 0.9±0.2 1.3±0.2 0.7±0.2 Fe2+ 13.6±8.3 6.6±1.2 12.3±1.8 7.2±1.4 53.5±18.6 4.9±0.6 Mn2+ 0.0±0.0 0.1±0.1 1.4±0.9 0.5±0.3 1.0±0.4 0.0±0.0 Cu+ 0.3±0.1 0.4±0.1 1.0±0.3 0.5±0.1 0.8±0.2 0.2±0.0 Zn2+ 0.1±0.1 0.4±0.2 0.1±0.0 0.2±0.0 0.5±0.2 0.0±0.0 B 1.4±0.1 1.2±0.1 1.4±0.1 0.9±0.1 1.6±0.2 1.3±0.3 SO42- 161.0±75.7 141.9±21.8 277.0±59.9 189.4±33.4 463.5±94.2 132.2±25.2 Al3+ 28.8±2.0 27.4±1.6 28.7±1.7 25.6±1.1 34.0±2.7 30.8±3.0  82  Table 3.6 Results of soil nutrient availability analysis with significant differences between zones (disturbed/undisturbed) and among sites shown in bold Nutrient Zone  Site             χ2       p value            χ2       p value Al3+ 3.15 0.08 4.43 0.11 SO42- 4.19 0.04 5.3 0.07 B 10.08 < 0.001 1.93 0.38 Zn2+ 0.03 0.86 1.11 0.57 Cu+ 4.01 0.04 3.9 0.14 Mn2+ 2.33 0.12 3.9 0.14 Fe2+ 12.42 < 0.001 1.83 0.40 PO33- 0.78 0.38 1.82 0.40 K+ 0.21 0.65 16 < 0.001 Mg2+ 13.21 < 0.001 3.38 0.18 Ca2+ 1.42 0.23 4.82 0.09 NH4+ 0.72 0.39 0.24 0.89 NO3- 5.21 0.02 7.22 0.03 Total N 5.33 0.02 5.68 0.06  3.4.6 Ecosystem respiration  Ecosystem respiration (Re) measured over the growing season (Figure 3.10) was greater in disturbed zones (0.47±0.09 µmol m-2s-1) than undisturbed zones (0.27±0.06 µmol m-2s-1), and these fluxes were marginally significantly different (F(1,44)=4.17, p=0.05). When individual sites were examined, differential responses in respiration were found. At RTS-2 and RTS-3, increased respiration occurred in disturbed zones (0.84±0.43 µmol m-2s-1 and 0.52±0.08 µmol m-2s-1, respectively) compared to undisturbed zones (0.43±0.15 µmol m-2s-1 and 0.25±0.09 µmol m-2s-1, respectively). This differed from RTS-1, as respiration from undisturbed tundra (0.21±0.07 µmol m-2s-1) was similar to respiration from disturbed tundra (0.20±0.05 µmol m-2s-1).  Spatial variability in respiration was greater within the disturbed zone of RTS-2. However at RTS-1 and RTS-3, variability was similar in both zones. Maximum respiration values were twice as large in disturbed areas (1.89 µmol m-2s-1) compared to undisturbed areas (0.96 µmol m-2s-1). When 83  respiration was compared between all sites and zones, differences in respiration were found between sites (F(2,44)=4.49, p=0.02).   Figure 3.10 Box plots of ecosystem respiration at each site, pooled by disturbance status. Boxes show the 25th and 75th percentiles, dots are outliers, and horizontal lines are medians.  3.4.7 Net ecosystem exchange (NEE)   When EC measured NEE was separated by wind direction, undisturbed areas (wind from North) at both sites were a net sink of CO2 (Figure 3.11). At RTS-1, the undisturbed tundra sequestered an average of 0.19 ±0.03 µmol m-2s-1 CO2 while the disturbed tundra sequestered four times less CO2 (0.05 ±0.04 µmol m-2s-1; Figure 3.12). At RTS-2, the undisturbed tundra sequestered 0.05 ±0.02 µmol m-2s-1 while the disturbed tundra released 0.07 ±0.04 µmol m-2s-1 CO2 over the measurement period. Variability in fluxes in undisturbed tundra were evident between these two sites, as the CO2 sequestered in undisturbed tundra at RTS-2 was similar to that sequestered by the disturbed tundra at RTS-1. Comparison of fluxes showed the influence of disturbance, as NEE from disturbed areas (NEEd = 0.00±0.03 µmol m-2s-1) significantly differed from NEE in 00.5121.5SiteR e (µmol m-2 s-1) UndisturbedDisturbedRTS 1 RTS 2 RTS 384  undisturbed areas (NEEc= - 0.13±0.02 µmol m-2s-1), (F(1,697)=11.22, p<0.0001). Site was also a significant factor as fluxes differed at RTS-1 and RTS-2 (F(1,697)=11.85, p<0.0001). When averaged, Tukey tests showed significant differences in NEE from undisturbed tundra at RTS-1 and all other locations and zones (Figure 3.12).   Figure 3.11 Mean NEE (±SE) during the July 2013 growing season in disturbed and undisturbed portions of RTS-1 and RTS-2 measured by eddy covariance.  ïïïï   DOYAverage daily NEE (ѥmol m-2s-1)576RTS 2UndisturbedDisturbed85   Figure 3.12 Seasonal mean NEE flux measured by eddy covariance during 2013 sampling period (June 27 - July 29) at RTS-1 and RTS-2. Data are mean and error bars are SE.  3.5 Discussion Spatial mosaics across the Arctic landscape are evident on the Fosheim Peninsula as a result of disturbance. Site characteristics that result from morphology and modifications that result following retrogressive thaw slumping are dependent on site characteristics present before disturbance. These permafrost disturbances increase spatial heterogeneity at the fine scale. At our study location, each site was characterized by different combinations of plant species, possibly due to the initial site conditions. We found vegetation differences across the three disturbances: the newly disturbed tundra in RTS-1 was characterized by the grass Poa glauca while the recovered tundra at RTS-3 was dominated by Carex aquatilis, which were likely the result of different soil moisture levels, with higher moisture at RTS-3, due to the exposure of ground ice and continued flow in nearby areas. These conditions allow species that prefer high moisture –0.2–0.100.10.2SiteNEE (ѥmol m-2s-1)UndisturbedDisturbedRTS 1 RTS 286  content, such as Carex aquatilis, to colonize (Chapin et al., 1992).  Greater soil moisture relative to undisturbed tundra was also present at RTS-1 and RTS-2, and field observations suggest that differences were influenced by the quantity of meltwater provided by ground ice thaw at the headwall in addition to flow patterns present in the disturbance. The dynamic nature of these flow patterns can greatly modify vegetation composition during recovery. Bartleman et al. (2001) found recovery trajectories were altered by the depletion of ground ice and lack of resulting flow. This resulted in moisture rich areas that were dominated by Equisetum spp. and drier areas of the slump that were characterized by Salix spp.  Additionally, active layer depths are impacted by disturbance. Differences in active layer depth at our study sites corresponded with reduced reflectivity. Lower reflectivity at RTS-1 resulted in greater absorption of solar radiation and therefore a deeper active layer, while greater reflectivity corresponded with a shallower active layer at RTS-2. Thaw is also impacted by snow accumulation, soil texture, soil moisture, and organic layer depth. At sites impacted by vehicle disturbance, Emers et al. (1995) found both increases and decreases in active layer depth associated with disturbed tundra, with shallower depth associated with insulating vegetation and complete removal of vegetation resulting in deeper depths. We found increased soil temperatures within disturbed soils at RTS-1 and RTS-2. However, at RTS-3 greater soil temperatures were found in the undisturbed terrain, which were driven by an increase in total vegetation cover in the disturbance at this site, resulting in increased shading. Less vegetation was found on the undisturbed surrounding terrain, thus albedo is likely greater in the disturbance.  As soil temperatures and the overall ground thermal regime may be impacted for up to 100 years following disturbance (Burn and Friele, 1989), both active and stabilized disturbances at our 87  study site were characterized by altered soil temperatures as compared to undisturbed surrounding terrain.  Plant nutrient availability may be elevated by permafrost thaw as soluble materials in the frozen ground may be released with ground ice melt associated with disturbance, exemplified by increased plant available nitrate, sulfate and calcium concentrations in RTS in the Low Arctic (Lantz et al., 2009). Increased concentrations of various soil nutrients (including K+, Ca2+, Mg2+, Cl-, SO42-, and PO33-) have been found in soils impacted by cryogenic landslides on the Yamal Peninsula, Russia; these elevated concentrations resulted in increased productivity of Salix spp. (Ukraintseva and Leibman, 2000; Ukraintseva, 2008). Ukraintseva (2008) has also noted increases in soil fertility associated with increases in nitrogen and potassium in disturbed soils. At our study site, enhanced nutrient concentrations (including Total N, NO3-, Mg2+, Fe2+, Cu+, B, and SO42-) were found within disturbed terrain, however we also saw evidence of high variation within sites. Although elevated nutrient concentrations are found within disturbed areas, runoff from disturbances removes material released from thawing permafrost. The fluvial impacts following permafrost disturbance (ADS and RTS) include pulses in sediment transport and increases in discharge and turbidity (Kokelj et al., 2015; Lamoureux and Lafrenière, 2009).   On the Fosheim Peninsula, salt efflorescence accumulations have been found at sites located below the glacial marine limit, and are largely related to disturbance as dissolved solids that were previously trapped in frozen sediments were released and redistributed downslope (Kokelj and Lewkowicz, 1999). Efflorescences were found within scar floors of the ALDS and RTS at our sites and downstream from disturbances. Large concentrations of Na+ and salts in the active layer and runoff could negatively affect plant growth and revegetation of disturbed terrain, 88  increasing the duration of modified drainage and enhancing erosion and may remain in the terrestrial system for 30 years or more (Kokelj and Lewkowicz, 1999).  Measurements of ecosystem respiration (Re) were consistent with those measured from other permafrost disturbances (Beamish et al., 2014; Cassidy et al., 2016) as fluxes from disturbed sites differed from those in undisturbed sites. Re measured from disturbed tundra was nearly double compared to Re from undisturbed tundra at one location. However, we found differential responses of Re based on site, as disturbed tundra was characterized by greater Re at sites RTS-2 and RTS-3, while undisturbed tundra at RTS-1 had similar Re rates compared to the disturbance. Disturbed tundra that has undergone some revegetation, such as the recovered site at RTS-3, may also have larger GPP, which would offset increases in Re at these sites. However, greater GPP could also enhance autotrophic respiration contributions.  Loss of old carbon from permafrost has been associated with disturbance. Schuur et al. (2008) found a positive relationship between ecosystem respiration and old carbon release associated with permafrost thawing. On the Fosheim Peninsula, old carbon may be released due to permafrost disturbances, which may be responsible for increases in respiration in disturbed areas relative to undisturbed tundra at some sites; however these increases were not significant.  Our NEE measurements are comparable to those measured in other high Arctic locations (specifically Lake Hazen, Ellesmere Island and Cape Bounty, Melville Island), which ranged between uptake of 0.2 – 2.2 g C m-2 day -1 (Lafleur et al., 2012). Despite the small magnitude of these fluxes, the impact of disturbance is still evident.  Although both disturbed and undisturbed tundra at RTS-1 sequestered a minimal quantity of CO2, the undisturbed tundra sequestered nearly four times the CO2 of the disturbed tundra. In addition, at RTS-2 the undisturbed terrain sequestered CO2 throughout the season whereas the disturbed tundra released CO2 to the 89  atmosphere. The NEE of the undisturbed tundra at RTS-1 was substantially greater than at RTS-2, which was likely due to greater vegetation cover at this site. This indicates that the initial conditions of undisturbed tundra influence the resulting carbon exchange of disturbed tundra through vegetation colonization and recovery within the disturbance.  Rates of NEE measured at High Arctic sites are generally smaller than those measured at Low Arctic sites (Lafleur et al., 2012). If disturbances occur in areas that sequester the most C in a landscape the effect on the landscape C balance will be greater than in areas with lower initial sequestration. In the Low Arctic, very large slumps could have large impacts on the carbon balance of these landscapes as this region is carbon rich and has higher fluxes (Lafleur et al., 2012). The lower fluxes measured and the smaller amounts of carbon in the soils of the High Arctic indicate that the impacts of disturbances here will not have the same effects on regional carbon balance or the atmosphere as in the Low Arctic. Understanding the spatial variation in thermokarst will be critical to understanding the magnitude of disturbance effects on the climate system.  3.6 Conclusions Based on the data presented here we draw the following conclusions: 1) Differences among the three undisturbed sites for many parameters were greater than the impacts of disturbance within one site; and 2) RTS impact NEE. Heterogeneity was amplified as a response to disturbance, resulting in unique combinations of vegetation communities, site characteristics, and carbon fluxes. The establishment of flux towers at two locations allowed us to determine the impacts of permafrost disturbances on CO2 fluxes simultaneously throughout the season and examine the role of heterogeneity on NEE. However, as fluxes were dependent on wind direction, only 90  disturbed or undisturbed tundra could be measured at each site at any one time. It would have been beneficial to have fluxes measuring both conditions within each disturbance simultaneously for comparison. This setup is further elaborated in Ch. 4 where NEE over a disturbed and undisturbed tundra were measured simultaneously over the growing season with two towers over the same RTS.   As the magnitude and frequency of disturbance is predicted to increase with increasing temperatures and precipitation, it becomes important to determine the impacts of disturbance both at the plot- scale but also at the landscape scale. Variations among studied disturbances indicated that upscaling to the landscape level would provide a more comprehensive view of the effects of the disturbances. Understanding both scales will help us predict the potential ecosystem changes that may result from modified disturbance regimes in the Arctic.           91  Chapter 4: The effect of a permafrost disturbance on growing-season carbon dioxide fluxes in a High Arctic tundra ecosystem  4.1 Overview Soil carbon stored in high-latitude permafrost landscapes is threatened by warming, and could contribute significant amounts of carbon to the atmosphere and hydrosphere as permafrost thaws. Thermokarst and permafrost disturbances, especially active layer detachments and retrogressive thaw slumps, are present across the Fosheim Peninsula, Ellesmere Island, Canada. To examine the effects of retrogressive thaw slumps on net ecosystem exchange (NEE) of CO2 in high Arctic tundra, we used two eddy covariance (EC) tower systems to simultaneously and continuously measure CO2 fluxes from a disturbed site and the surrounding undisturbed tundra. During the 32- day measurement period in the 2014 growing season the undisturbed tundra was a small net sink (NEE = -0.1 g C m-2 d-1); however, the terrain disturbed by the retrogressive thaw slump was a net source (NEE = +0.4 g C m-2 d-1). Over the measurement period, the undisturbed tundra sequestered 3.8 g C m-2, while the disturbed tundra released 12.5 g C m-2. Before full leaf out in early July, the undisturbed tundra was a small source of CO2, but shifted to a sink for the remainder of the sampling season (July), whereas the disturbed tundra remained a source of CO2 throughout the season. A static chamber system was also used to measure CO2 fluxes in the footprints of the two towers, in both disturbed and undisturbed tundra, and fluxes were partitioned into ecosystem respiration (Re) and gross primary production (GPP).  Average GPP and Re found in disturbed tundra were smaller (+0.40 µmol m-2 s-1 and +0.55 µmol m-2 s-1, respectively) than those found in undisturbed tundra (+1.19 µmol m-2 s-1 and +1.04 µmol m-2 s-1, 92  respectively). Our measurements indicated clearly that the permafrost disturbance changed the high Arctic tundra system from a sink to a source for CO2 during the growing season.   4.2 Introduction Permafrost soils in the Arctic store vast amounts of carbon. The northern permafrost zone carbon inventory estimates the quantity of soil organic carbon stored in the top 3 m of frozen and unfrozen soils in northern circumpolar permafrost regions to be up to 1035 ±150 Pg, or approximately 50% of worldwide soil organic carbon (Tarnocai et al., 2009, Grosse et al., 2011, Hugelius et al., 2013, Schuur et al., 2015). Measurement difficulties and uncertainty regarding carbon storage in cryoturbated soils suggest that this is an underestimate of these carbon stocks (Hugelius et al., 2013).  As ground temperatures increase due to global climate change and permafrost thaws, this organic carbon becomes available for microbial decomposition (Schuur et al., 2008). McGuire et al. (2006) noted the implications for feedbacks to Arctic climate resulting from disturbance and enhanced decomposition including positive feedbacks as more CO2 released leads to warmer temperatures, thus exacerbating thaw and leading to further release of CO2. Conversely, a negative feedback may result if soil carbon inputs offset decomposition, as the balance between litter accumulation and decomposition determines the net effect on climate (Davidson et al., 2006, Cornelissen et al., 2007). Climate change is expected to increase the frequency and extent of land surface disturbances in the Arctic (ACIA, 2005, Vincent et al., 2011; Kokelj and Jorgenson, 2013). These disturbances are usually linked to thermokarst and affect soil temperature, water quality and soil nutrients (Mackay, 1970, Lamoureux and Lafrenière, 2009, Lantz et al., 2009, Kokelj and Lewkowicz, 1998, Kokelj and Lewkowicz, 1999). In the High Arctic, these disturbances commonly take the form of retrogressive thaw slumps (RTS). RTS are initiated by the exposure 93  of ground ice (sometimes linked to coastal erosion) and result in the removal of soil and vegetation as the slump headwall retreats further upslope (Lantuit and Pollard, 2008). As ground ice thaws, the headwall regresses and will remain active until falling blocks of soil and vegetation insulate exposed ice and prevent further thaw (Burn and Friele, 1989). Within the overall landscape, these distinct landforms often create unique microclimates resulting in increased landscape heterogeneity (Ukraintseva, 2008, Lantz et al., 2009, Bosquet, 2011). Climate warming may cause differential responses in disturbed and undisturbed tundra. For example, the response of plants to increases in temperature may be intensified when disturbance occurs (Lantz et al., 2009). Lantz et al. (2009) suggested disturbances play a more significant role in vegetation modification than temperature changes, particularly at the fine scale. We hypothesize that those changes in the landscape (slumping and vegetation loss) will have a significant effect on the carbon balance of tundra systems.  However, few direct measurements of net ecosystem exchange (NEE) and its component fluxes, ecosystem respiration (Re) and gross primary productivity (GPP), have been completed to determine the effect of these permafrost disturbances, and no measurements have been made in the High Arctic.  Eddy covariance (EC) has been used to quantify NEE in the Arctic and measurements vary greatly, depending on location and ecosystem type.  The magnitude of CO2 fluxes are generally greater at low latitudes than in the high Arctic (Lafleur et al., 2012) and in wet sedge areas than dry heath tundra (Kwon et al., 2006; Groendahl et al., 2007). Variability may be explained by nutrient availability, substrate quantity and soil organic matter (Mbufong et al., 2014). Typical mean daily values measured during the peak growing season (July only) showed net CO2 uptake ranged between 0.2 and 2.2 g C m-2 d-1 at a wide range of Arctic sites (Lafleur et al., 2012). Previous studies have found large inter-annual variability within and among sites, 94  which can shift the site from a carbon sink to carbon source (Griffis and Rouse, 2001; Kwon et al., 2006; Merbold et al., 2009). Large variability in tundra vegetation communities over short distances increases the difficulty in assessing NEE fluxes across the Arctic, and determining their responses to disturbance and environmental change (Lafleur et al., 2012).  Static chamber systems, which partition NEE into component fluxes GPP and Re, are an alternative method of measuring ecosystem fluxes. Chamber studies in the Arctic have found a loss of carbon during the winter, and increasing sink potential with a longer growing season (Welker et al., 2000; Welker et al., 2004). At Alexandra Fiord, Ellesmere Island, experimental warming impacted NEE differently based on soil moisture, with a greater increase in respiration at dry compared to wet sites (Welker et al., 2004). Across a latitudinal gradient, warming tended to increase respiration, with the greatest increases found in dry ecosystems (Oberbauer et al., 2007).  While NEE values are generally similar between chamber and EC methods, differences are attributed to the scale of the measurements (Stoy et al., 2013). Fox et al. (2008) showed there was large bias in upscaling chamber measurements, relative to EC values in a tundra ecosystem, due to microscale surface heterogeneity of the landscape. Further, with 24 hours of daylight during which the sun remains relatively high above the horizon, the usual partitioning methods for EC measurements into component fluxes (Reichstein et al., 2012) are not applicable, as they rely on nighttime measurements, or measurement during low light conditions. Consequently, to measure the impact of a RTS on the CO2 exchange of the high Arctic tundra we used both EC and chamber measurements.  In this study, we analyze the impacts of RTS on CO2 exchange in a high Arctic tundra ecosystem. Our main research objective was to examine how growing season NEE and its 95  component fluxes vary between a RTS and undisturbed tundra. We hypothesize that RTS will significantly impact NEE. 4.2.1 Study area Our research was conducted on the Fosheim Peninsula, located on western Ellesmere Island, Canada (79° 58’ 56” N 84° 23’ 55” W (WGS-84), elevation 100 m asl).  This field site had an isolated retrogressive thaw slump (RTS) (6300 m2) within a relatively flat area and wind patterns were constrained (NNE-SSW) by its location near a shallow valley bottom (Fig. 4.1; Fig. 4.2). Ice-rich permafrost is found throughout the study region and increased summer temperatures and precipitation over the past 20 years have resulted in greater occurrence of active layer detachment slides and RTS (Lewkowicz and Harris, 2005a). The bedrock is mainly comprised of sandstones of the Eureka Sound group (Bell, 1996) with marine deposits of silts and fluvial sandy soils varying in thickness above bedrock (Robinson and Pollard, 1998). The limit of ocean inundation at the end of the last glaciation in the area lies at approximately 140 m above sea level (Bell, 1996), with limited vegetation above this level. Our study location was located below the marine limit, where vegetation was a relatively uniform dwarf-shrub-graminoid community on moderately drained, slightly alkaline soils. Vegetation located in the undisturbed tundra was dominated by Salix arctica, Dryas integrifolia, Carex nardina, moss, and lichen. Within the disturbance, the dominant plant species was Puccinellia angustata, which is able to colonize the disturbed area and proliferate. Vegetation cover within the RTS varied based on moisture and proximity to undisturbed vegetation, and was much lower than the surrounding undisturbed areas (with estimates of cover averaging (±SE) 3(±0.5)% and 27(±1.5)% total cover, respectively). The nearest weather station, Eureka, is located 40 km to the west and has a mean temperature of 96  6.1°C and mean precipitation of 14.5 mm in July over the 1981-2010 period (Environment Canada, 2015).   Figure 4.1 Aerial image of the dual eddy covariance system setup with the location of both flux towers indicated. The valley trends NNE-SSW. View is to the south.  4.3 Materials and methods 4.3.1 Eddy covariance measurements of carbon dioxide fluxes An appropriate sampling design was necessary to quantify the CO2 fluxes between land surface and atmosphere simultaneously from disturbed and undisturbed sites in close proximity (Hollinger and Richardson, 2005). We used a dual eddy covariance approach, which was advantageous over a single eddy covariance tower as we were able to measure fluxes 97  simultaneously from disturbed tundra and the surrounding undisturbed tundra (Fig. 4.1; Fig. 4.2). However, direct placement of an EC system within the disturbance was not possible due to the active mass movements in the RTS creating risk for researchers and equipment. Two towers were established on opposite sides of the RTS, at the boundary between disturbed and undisturbed terrain (Fig. 4.1). Tower 1 was established on the southern boundary of the RTS and Tower 2 was established on the northern boundary at a distance of 90 m from Tower 1. Disturbed tundra were areas impacted by RTS, while undisturbed tundra were areas located outside the boundary of the RTS. This set-up allowed the measurement of fluxes containing signals from both areas simultaneously. By using turbulent source area modeling (see below) we then estimated the contribution of disturbed and undisturbed tundra to each of the signals. Both EC systems were established on tripods located on the periphery of the active RTS on 26 June 2014 and were operated continuously until 28 July 2014. On each system the instrumentation included: an ultrasonic anemometer (CSAT-3, Campbell Scientific Inc., Logan, UT, USA) and a co-located infrared gas analyzer (IRGA) (LI-7500, LI-COR Inc., Lincoln, NE, USA). The IRGA was tilted 30° from the vertical to minimize issues associated with sensor heating and reduce pooling of moisture on the windows. Both IRGA and ultrasonic anemometer were established at a height of 1.3 m on both towers; a temperature and humidity sensor (HMP, Campbell Scientific Inc.) at 1 m; a quantum sensor (SQ-110, Apogee Instruments Inc., Logan, UT, USA; height: 1m); a net radiometer (NR Lite, Kipp & Zonen B.V., Delft, The Netherlands; height: 1m); and all sensors were attached to a data logger (CR1000, Campbell Scientific Inc.) and sampled at a rate of 10 Hz. This dual EC sampling technique allowed for simultaneous sampling of fluxes from the disturbed tundra and the surrounding undisturbed (control) terrain for most time steps. Previous knowledge of wind direction (!) based on the location of the 98  disturbance within a valley constrained winds along the valley axis into up-valley wind (0º < ! < 40º) and down-valley wind (160º <  ! < 200º) directions, which resulted in aligning the sector facing towards 290°, having a sector free of flow distortion from 140º to 80º (distorted sector 80º to 140º). The towers were established at a distance of 3 m from the slump edge to ensure stability, and were moved periodically throughout the season due to recession of the slump edge. Additionally, the potential impacts of step changes due to the placement of the flux tower at the boundary of disturbed and undisturbed tundra was minimized through the use of friction velocity thresholds and removing data with wind along the discontinuity with obvious flow distortion due to sensors. Both IRGAs were calibrated prior to the field season using a two-point calibration in the lab against standards from the Greenhouse Gas Measurement Laboratory (GGML), Meteorological Service of Canada using a zero gas and span gas of known mixing ratio.  Fluxes of CO2 (FC) were computed in EddyPro® (V5.1.1, LI-COR Inc.) with a missing sample allowance of 30%. FC was calculated over a 30 minute averaging interval using double rotation for tilt correction, block average detrending, contact time lag detection, and density corrections using mixing ratios (Burba et al., 2012). Data were quality checked using the flagging system proposed by Mauder and Foken (2004).  99   Figure 4.2 Turbulent source areas for two time-steps on DOY 186:  a) 09:00 and b) 18:00, with ellipses displaying areas contributing to the given percentage of the signal from each instrument tower (T1 and T2). Three ellipses from each tower represent the 50%, 80%, and 90% cumulative source area.  The shaded area represents a signal from the disturbed part of the surface. The white polygon represents the furthest extent of headwall retreat, as the initial image was taken in July 2013 (Worldview-2) and significant retreat occurred between image acquisition and the summer 2014 sampling period. A wind rose of the study location is inset in the first time-step.    4.3.2 Turbulent source area model To estimate the instantaneous turbulent source area that influences sampled NEE, a 2-dimensional gradient diffusion and crosswind dispersion model (Kormann & Meixner, 2001) was run for all 30 min periods between 26 June 2014 and 28 July 2014 at a 1 m grid resolution over a domain of 300 x 300 m with the tower situated in the centre (see Fig. 4.2). Model inputs included wind direction ! (º), standard deviation of the lateral wind component !! (m), roughness length !! (m) and Obukhov length L (m) separately for each tower and for each time 100  step. !, σv and L were calculated directly by EddyPro® based on measurements by the two ultrasonic anemometers. Roughness length varied depending on whether the upwind surface in a particular time period was in the RTS or representing undisturbed tundra. !! was determined separately for 10º wind direction bins based on the ensemble of measurements from the entire dataset following Paul-Limoges et al. (2013).  For each wind sector !, !! was calculated for cases with near-neutral stability (-0.05 < z/L < +0.05) using Eq. 9:     !! ! = !    exp   − !!!∗  Equation 9  where z is the measurement height (1.3 m), k is the von Kármán constant (0.4), ! is the measured mean horizontal wind (m s-1) from this wind direction, and !∗ is the simultaneously measured friction velocity (m s-1) calculated as !∗ = (!!!!! + !′!′!)!.!" where !!!! and !′!! are covariances of longitudinal (u), lateral (v) and vertical (w) wind components. Mean wind ! and covariances were calculated by EddyPro® based on measurements from the ultrasonic anemometer. The disturbed sectors of both towers had an average !! = 0.032 m whereas the undisturbed sectors had an average !! = 0.017 m. Gridded flux footprints (or vertical per unit point source) ϕ (x,y) were calculated with 1 m resolution for each 30 min step following Christen et al. (2011). A fraction of the flux footprint was predicted to be outside the 300 m study area, which was assumed to represent an undisturbed (control) surface (as no additional permafrost disturbances were located within proximity of the towers).  The 300 m x 300 m model domain included the entire disturbance and a spatial mask I (x,y) of the domain was created with a value of 1 inside the disturbance boundary and 0 for 101  undisturbed tundra. For each grid-cell, I (x,y) was multiplied by ϕ (x,y), and then summed to determine the fraction of the footprint that originates from inside the RTS (Eq. 10):  Φ! = ! !,! !(!,!)!""!!!!""!!!  Equation 10  Φ! is the fraction of the tower signal (from 0 to 1) influenced by the disturbed surface of the RTS. The fraction of the signal influenced by the undisturbed tundra  Φ! is then calculated as   Φ! = 1−      Φ!. By solving a set of linear equations (Eq. 11 and Eq. 12), we are able to partition the component fluxes of CO2  (Fig. 4.2) from the disturbed tundra (NEE!) and from the undisturbed tundra (NEE!) from both towers (T1 and T2):   NEE T1 =Φ! T1 NEE! +Φ! T1 NEE! Equation 11   NEE T2 =Φ! T2 NEE! +Φ! T2 NEE! Equation 12  Turbulent source areas calculated for each time step over the sampling period are shown in Fig. 4.2. These two example time steps from Fig. 4.2 can be solved as follows. In the first time-step (09:00),   Φ! for T1 is 1, therefore the NEE T1 = !""! = -0.17 µmol m-2 s-1. For T2, 88% (  Φ!) was disturbed while the remaining 12% was allocated as undisturbed (  Φ!), so NEE!  and NEE!   were solved with NEE T2  = 1.20 µmol m-2s-1 and resulted in NEE! =1.39  µmol m-2 s-1. Corresponding to the second time step from Fig. 4.2 (18:00), T1 is influenced by both undisturbed and disturbed NEE as Φ!    !1 =   0.73 and Φ!    !1 = 0.27  and NEE T1  is 0.38 µmol m-2 s-1. At T2, Φ! is 0, while Φ! is 1, so NEE T2 = NEE! = -0.03 µmol m-2 s-1. Consequently, NEE! =   −0.03  µmol m-2s-1 and NEE! = 0.54  µmol m-2s-1. 102  Calculations of NEEd and NEEc were numerically unstable under multiple combinations of surface fractions, including when winds were parallel to the edge of the disturbance and when Φ! and Φ!   were roughly equal to one another. As a result, values where the absolute difference between Φ! and Φ!   was less than 0.05 were removed and fluxes during these periods were gap-filled as detailed below during these periods.   The resulting NEEc and NEEd were compared and fluxes that had a difference from the daily average that was greater than 5 standard deviations of the 30 min values of the same day (applied iteratively) were removed. For further analysis, half hour fluxes were averaged to calculate hourly fluxes. If one of the two 30 min values was invalidated, then the hourly value was calculated based on the remaining other 30 min period. Hourly gaps that still existed were then filled using the following methods: a) gaps in NEEc and NEEd of less than 2 hours were filled using a linear interpolation; and b) gaps greater than 2 hours were filled using aggregate averaging over a rolling three-day window selecting the same time of the day. The cleaned and filled dataset is composed of 86% original data and 14% gap filled (of a total of 750 data points, 106 of these were modeled).  Data were also removed during times of maintenance, when the towers were moved and when manual chamber or vegetation measurements were made within the tower footprint.  4.3.3 Portable chamber system On 27 June 2014, 63 opaque PVC collars (10 cm diameter, A = 78.5 cm2, 6 cm depth) were installed across the source areas of T1 and T2, in both the disturbed and undisturbed zones (disturbed tundra N=21, undisturbed tundra N=42). Collars were inserted 4 cm into the ground so as to minimally disturb soil and vegetation and left to protrude 2 cm above the soil surface. As moss cover was minimal and discontinuous, the location of the ground surface could be easily 103  identified as the upper surface of the soil. Collar locations were randomly determined based on the generation of random distances and angles from the flux tower within disturbed and undisturbed flux source areas, with a minimum distance of 2 m and a maximum distance of 30 m from the towers. The disturbed areas of the RTS were not entirely devoid of vegetation, as clumps of soil and plants existed sporadically throughout the disturbance; 9 of 21 collars contained at least one individual plant. Measurements of CO2 fluxes began on 29 June, to allow the immediate disturbance effects of installation to dissipate.  A non-steady state vented portable chamber system similar to Jassal et al.(2005) was used to measure fluxes from each collar using transparent and opaque chambers. The measurement head was a PVC chamber with a volume of 1.4 x 10-3 m3 (height: 15.6 cm, diameter: 10.7 cm). Fluxes from all collars were measured six times throughout the study period, at 5-day intervals.  The chamber head was placed on each collar and a foam gasket sealed the connection between the collar and the chamber head. Measurements were made for two minutes while a pump (flow rate 600 cm3 min-1) circulated air from the chamber head into a portable, battery operated infrared gas analyzer (IRGA) (LI-840, LI-COR Inc., Lincoln, USA) and back into the chamber head through a closed circuit. The IRGA determined CO2 mixing ratios ([CO2] in ppm) and water vapour concentrations at a temporal resolution of 1 Hz during the run. The IRGA was calibrated in the laboratory prior to sampling using a two-point calibration, against standards from the Greenhouse Gas Measurement Laboratory (GGML), Meteorological Service of Canada, using a zero gas and span gas of known mixing ratio. The IRGA has been calibrated in the laboratory for effective volume, which exceeds geometric volume by 10% due to absorption of CO2 on the walls of the chamber and contribution of near-surface soil porosity (Jassal et al., 2012). This calibration was carried out in the laboratory by determining the 104  difference between two flux measurements, one immediately following the other, where the second measurement included a known rate of injection of CO2 into the chamber. Fluxes were calculated from  Δ[CO2]/Δt (linear regression over 2 min, discarding the first 10 seconds), using Eq. 13:   !" =   ρ  !V!   Δ[CO!]Δ!  Equation 13  where ρ is molar air density (mol m-3) calculated from measured air temperature, D is  dilution considering [H2O], Δ[CO2]/Δt  is the rate of change in CO2 mixing ratio over time (µmol mol-1 s-1), and V and A are chamber volume and area, respectively. To obtain measurements of NEE, the transparent chamber head was used on each collar. For ecosystem respiration (Re) measurements, the chamber was removed and allowed to equilibrate to ambient [CO2] before being replaced on the collar, and a shroud was placed over the transparent chamber head to block out all PAR. GPP was calculated based on GPP = Re – NEE, where NEE is negative if GPP > Re and both Re  and GPP are positive values. NEE and Re measurements were taken within minutes at each collar allowing for comparison. Measurements were completed over a 7-hour sampling period and were always completed between 10:00 and 18:00 CDT to reduce diurnal changes in light and temperature. 4.3.4 Environmental variable sampling Soil temperature loggers (HOBO Pendant Temperature/Light Data Loggers, Model UA-002-64, Onset Computer Corporation, Bourne, MA, USA) were installed at randomly selected collars throughout the study area within 0.5 m of the collar. A total of 21 HOBO sensors (14 sensors located in undisturbed tundra and 7 sensors in disturbed tundra) measured soil temperatures at 5 105  cm every minute throughout the sampling season. The soil temperatures were aggregated into hourly averages to allow for comparison with hourly EC data. Soil moisture was measured adjacent to collars every five days as volumetric water content (%) using a calibrated Time Domain Reflectometry (TDR) sensor (HydroSenseII Soil Water TDR, Campbell Scientific Inc., Logan, UT, USA) with 12 cm rods. After rain events, measurements were delayed for 24 hours.   4.4 Results 4.4.1 Environmental conditions during the study period The measured variations over the study period in air temperature (Ta), net radiation (Q*, over undisturbed tundra), incoming photosynthetically active radiation (PAR), and vapour pressure deficit (VPD) measured at Tower 2, and soil temperature from the disturbance and undisturbed tundra area near Tower 2 are shown in Fig. 4.3.  Three distinct periods (early, peak and late season) were identified throughout the study period based on plant phenological development and environmental conditions (Fig. 4.3; Fig. 4.4). These periods varied in their duration (see Table 4.1). The early season was characterized by clear skies, however the middle of July was dominated by a period of cloudy, cooler conditions (exemplified by decreased Q*, Fig. 4.3). Air and soil temperatures showed distinct diurnal and seasonal patterns (Fig. 4.3; Fig. 4.4), characterized by an increase in both temperatures early in the season, which were sustained through the peak season, followed by decreases in both during the end of the season. During the measurement period, Ta increased from 10.5ºC in the early season (DOY=175-185) to 12.2°C during the peak of the growing season (DOY=186-202) and then decreased to 7.2°C by the end of July. On a diurnal basis, disturbed soils reached greater temperatures than undisturbed soils earlier in the season (12.6°C and 11.6°C, respectively), but cooled off quickly later in the season (7.8°C and 8.1°C, respectively), likely due to the lack of insulating vegetation, lack of litter, and 106  more moisture resulting from greater thermal admittance, conductivity and diffusivity (Mann Whitney U-Test (V = 181992, p < 0.01); Fig. 4.4). In undisturbed terrain, soil moisture decreased during peak season, while soil moisture increased steadily in disturbed tundra (Table 4.1). Overall, soil moisture values were significantly greater (Mann Whitney U-Test (V = 7023, p < 0.01)) in disturbed soils (24.1% ± 0.9) than in undisturbed soils (13.9% ± 0.4).   107   Figure 4.3 Meteorological conditions during the 2014 growing season at T2 with early season and end season shaded grey while peak growing season is unshaded. Height of all instrumentation on the tower was 1 m above the canopy. Soil temperatures were measured at a depth of  -5 cm, and mean temperature is shown for the disturbance (dashed line; n=7) and undisturbed tundra (solid line; n=14). DOY = day of the year.  51015200.51.51000100–100200300050010004081216Air temperature(°C)PARѥPROP–2 s–1)1HWUDGLDWLRQ(W m–2)9DSRXUSUHVVXUHdeficit (kPa)6RLOWHPSHUDWXUH(°C)180 190 200 210DOYPeakEarly End108  Table 4.1 Summary of net ecosystem exchange (NEE), soil temperatures (Soil T) and soil moisture (Soil M) from disturbed (d) and undisturbed (c) tundra, and air temperature (measured at T2) throughout the growing season. Min and max refer to hourly values.  Variable Early Season Peak Season End Season DOY 175-185 186-202 203-210  NEEc  (µmol m-2 s-1)*  0.080 ± 0.03  -0.28 ± 0.03  -0.015 ± 0.05  NEEd (µmol m-2 s-1)*  0.55 ± 0.03  0.25 ± 0.03  0.58 ±0.13  Air Temperature (°C)  mean (±SE) min/max   10.5 ± 0.17 5.3 / 15.0    12.2 ± 0.10 6.9 /16.1   7.2 ± 0.22 2.0 / 12.0  Soil Tc (°C)* at 5 cm   mean (±SE) min/max   11.6 ± 0.05 5.4/ 19.8   11.9 ± 0.29 6.9/19.8   8.1 ± 0.05 2.6 / 16.2  Soil Td (°C)* at 5 cm mean (±SE) min/max    12.6 ± 0.07 6.6 / 19.1   11.9 ± 0.04 6.8 / 19.5   7.8 ± 0.07 2.1/ 15.2  Soil Mc (%)* mean (±SE) min/max   14.4 ± 0.5 3.4 / 28.4    12.9 ± 0.4 1.1 / 31.6   16.9 ± 1.0 0.6 / 34.2  Soil Md (%)* mean (±SE) min/max   20.5 ± 0.8 9.7 / 41.2   24.8 ± 1.0 4.1 / 45.4    30.5 ± 1.6 6.9 / 44.8  * NEEc = average net CO2 flux from undisturbed (control) tundra. * NEEd = average net CO2 flux from disturbed tundra. * Soil Tc = average soil temperature from undisturbed tundra * Soil Td = average soil temperature from disturbed tundra *Soil Mc = average soil moisture from undisturbed tundra *Soil Md = average soil moisture from disturbed tundra  4.4.2 NEE of disturbed and undisturbed tundra The early season was characterized by leaf emergence, cool temperatures, and elevated soil moisture (Table 4.1) due to recent snowmelt. The peak season was characterized by maximum leaf area and flowering of vegetation (e.g. Salix arctica, Dryas integrifolia, and grass species) 109  and a decrease in surface soil moisture as warm air temperatures and large VPD persisted (Fig. 4.3).  The late season was characterized by the beginning of leaf senescence, dry soils, and the greatest thaw depth. Precipitation was minimal throughout the season (1.2 mm at Eureka), with isolated rain events occurring on July 17, 21, and 26. There was a windstorm beginning on 22 July that lasted 24 hours with wind speeds (as determined from the 20 Hz measurements) of up to 21 m s-1.  Figure 4.4 Ensemble average diurnal course of soil temperatures in the disturbed and undisturbed sites throughout the season: Early season =  24 June – 4 July; Peak season = 5 July – 21 July; End of season = 22 July – 29 July. Boxes show the 25th and 75th percentiles, dots are the outliers, horizontal lines are medians.  NEEc and NEEd were analyzed separately for three periods (early, peak and late season). In the undisturbed tundra, NEEc was initially a small source in the early period and transitioned to a small sink as photosynthesis increased during the peak season. In the late season, NEEc EarlyPeakEnd5101520510152050000 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23101520UndisturbedDisturbedHour of daySoil temperature (°C)110  became a small source consistent with decreased air and soil temperatures and the beginning of leaf senescence. This was in contrast with fluxes measured in the disturbed area (NEEd), which remained a CO2 source throughout the sampling period and displayed only slightly dampened values during peak season. Overall, NEEc and NEEd were significantly different throughout the sampling period (Mann-Whitney U-Test (V = 45839, p < 0.01)).    Figure 4.5 Ensemble diurnal course of CO2 fluxes from the retrogressive thaw slump (disturbed) and undisturbed tundra separated into the three sampling periods: top (early season), middle (peak season) and bottom (end season). Boxes show the 25th and 75th percentiles, black circles are outliers, horizontal lines are medians.  Aggregate fluxes calculated over the study period showed an overall loss of CO2 from disturbed tundra, and a modest CO2 sink in the undisturbed tundra (Fig. 4.6). Daily averages of EarlyPeakEndïï03ïï03ïï030 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23Hour of day1((ѥPROP-2 s-1)UndisturbedDisturbed111  NEEc ranged from -0.89 to +0.54 g C m-2 day-1. NEEd ranged from -0.29 to +1.63 g C m-2 day-1. During the early season, the average daily NEEc was a small source of CO2 to the atmosphere (+0.07 g C m-2 day-1) while disturbed tundra was a greater source of CO2 (NEEd = +0.55 g C m-2 day-1). During peak growth, this shifted as the undisturbed tundra sequestered -0.29 g C m-2 day-1 and disturbed tundra continued to emit CO2 at an average of +0.26 g C m-2 day-1. During the end of the sampling season, the undisturbed was a very small sink of CO2 with mean NEE of  -0.02 g C m-2 day-1 and the NEE of the disturbed tundra was +0.47 g C m-2 day-1. Over the duration of the entire sampling period, the disturbed tundra was a source of CO2 with an average of +0.39 g C m-2 day-1 while the undisturbed tundra was a sink for CO2 with an average uptake of -0.12 g C m-2 day-1 (Fig. 4.5). In total, the undisturbed tundra sequestered 3.8 g C m-2, while the disturbed tundra released 12.5 g C m-2 over the 32-day measurement period.  Diurnal NEE from the tower systems corresponded with soil temperatures. In disturbed areas, as soil temperatures warmed, CO2 emissions increased, consistent with increased respiration. However, fluxes in undisturbed areas showed increased sequestration during midday, due to greater photosynthetic activity dominating over respiration increases.  Temporal patterns of fluxes and climatic and environmental variables were analyzed for disturbed and undisturbed areas. In the disturbed area, regression analysis revealed strong positive relationships between NEE and soil temperature, PAR, Ta and VPD for the early and peak season periods (p<0.001), while PAR was the most important control during the late season (r=0.71, p < 0.001). In the undisturbed tundra, correlations between NEE and environmental variables varied throughout the sampling period. During the early season, PAR was most strongly correlated (r=0.40, p<0.001) with NEE, however, during the peak season temperature (r=0.28, p<0.001) and vapour pressure deficit (VPD) (r=0.28, p<0.001) became important 112  controls on NEE. At the end of the sampling season PAR was once again most strongly correlated with NEE (r=0.50, p<0.001) in the undisturbed tundra.  Figure 4.6 Average daily net CO2 flux (±SE) for the three sampling periods as measured by the two EC systems and the net effect for the entire season.  4.4.3 Partitioning of NEE Measurements from the static chamber system were allocated to one of the three seasonal periods, allowing comparison with EC data (Fig. 4.7). The NEE values measured using the chamber technique supported the EC measurements, but allowed fluxes to be partitioned into their component parts. The chamber measurements showed that the magnitude of GPP and Re were roughly similar, resulting in minimal NEE in both disturbed and undisturbed areas (Table 4.2; Fig. 4.8). Variability in GPP was greater in the undisturbed tundra with values up to 8.03 µmol m-2 s-1 while the maximum GPP in the disturbed tundra reached 2.47 µmol m-2 s-1. Re Early Peak End Total00.2–0.2–0.40.40.6Average NEE (g C m-2 day-1)Undisturbed Disturbed0.390.070.550.260.47-0.12-0.29-0.02113  ranged up to +5.92 µmol m-2 s-1 in the undisturbed tundra and to +2.23 µmol m-2 s-1 in the disturbed tundra. Over the sampling season in the disturbed areas, chamber-measured GPP averaged 0.40 µmol m-2 s-1 increasing during peak season to 0.45 µmol m-2 s-1, before falling to 0.24 µmol m-2 s-1 in the late season. Respiration was greatest during the early season with +0.70 µmol m-2 s-1, decreasing to +0.53 µmol m-2 s-1 during peak season, and finally to +0.35 µmol m-2 s-1 during the late season. These opposing fluxes resulted in the disturbed tundra being a small source for CO2 throughout the entire sampling season. NEE measured by the chamber system varied between +0.05 and +0.41 µmol m-2 s-1 in the disturbance with the largest NEE occurring early in the season due to high respiration.  Re was always greater in magnitude than GPP over disturbed tundra resulting in positive NEE values.  Figure 4.7 Comparison of NEE measurements from static chamber (square) and calculated from the two EC (circle) systems. Open symbols represent measurements from undisturbed tundra while closed symbols are measurements in the disturbed areas. Measurements were made in 21 collars in each of the disturbed and both undisturbed footprint areas of the EC towers.  ïï0   DOY$YHUDJHGDLO\1((ѥPROP-2  s-1 )UndisturbedDisturbedChamber EddyPeakEarly End114   Figure 4.8 Partitioning of NEE data from static chamber measurements into component fluxes, GPP and Re for the undisturbed and disturbed sites. Measurements were made in 21 collars in each of the disturbed and both undisturbed footprint areas of the EC towers.  The undisturbed areas were small sources of CO2 early in the season as Re outpaced productivity. In the undisturbed tundra, during the early season GPP averaged 0.85 µmol m-2 s-1, nearly doubling during peak season to 1.47 µmol m-2 s-1, before falling to 1.00 µmol m-2 s-1 late in the season. Re in the undisturbed tundra ranged from +0.62 µmol m-2 s-1 to +1.14 µmol m-2 s-1, with the greatest respiration occurring during peak growth. Both GPP and Re peaked during the middle of the sampling period (mid July), before decreasing at the end of the season, but GPP was always greater in magnitude than Re.     Undisturbed Disturbedïï0       DOY1((ѥPROP-2 s-1)NEEReGPP (-)PeakEarly End PeakEarly End115  Table 4.2 Summary of measurements (mean ± SE) from the portable chamber system (in µmol m-2 s-1) Variable Location Early Peak End Total DOY  175-185 186-202 203-210   NEE  undisturbed  0.25 ± 0.10  -0.33 ± 0.15  -0.37 ± 0.15  -0.14 ± 0.13  disturbed 0.31 ± 0.12 0.07 ± 0.13 0.11 ± 0.07 0.15 ± 0.06 GPP undisturbed 0.85 ± 0.16 1.47 ± 0.26 1.00 ± 0.19 1.19 ± 0.19  disturbed 0.39 ± 0.14 0.45 ± 0.16 0.24 ± 0.08 0.40 ± 0.03 Re undisturbed 1.10 ± 0.13 1.14 ± 0.15 0.62 ± 0.06 1.04 ± 0.12  disturbed 0.70 ± 0.08 0.53 ± 0.10 0.35 ± 0.05 0.55 ± 0.06  4.5 Discussion During the 2014 growing season, the RTS at our high Arctic site was a CO2 source while undisturbed tundra was a small sink. All fluxes were quite low, but similar to those measured in other high Arctic sites (Lafleur et al. 2012; Emmerton et al., 2015). Multi-year measurements of NEE in high Arctic tundra indicate that initial uptake of carbon coincides with snowmelt and increases in CO2 emission rates correspond with deep and long-lasting snowpack (Lund et al., 2012). Arctic sites show significant inter-annual variability, which is controlled by temperature among other variables; increased temperatures may result in enhanced emissions (Griffis and Rouse, 2001; Kwon et al., 2006; Merbold et al., 2009; Lund et al., 2012). In the High Arctic, soil moisture differences result in variations in ecosystem respiration (measured using chamber systems) and may enhance the impacts of warming (Welker et al., 2004). Warming has been 116  found to increase respiration along a latitudinal gradient with the greatest increases found in dry ecosystems (Oberbauer et al., 2007).  Based on chamber measurements, we found permafrost disturbance alters carbon dynamics by decreasing GPP and Re (Fig. 4.7). However, reductions to GPP are greater than reductions to Re, resulting in the disturbance becoming a net carbon source. Decreases in GPP are due to lower vegetation cover within disturbed terrain. Decreases in respiration have been found within slumps and slides and are linked with carbon export from the disturbed area and fewer plants thus lower autotrophic respiration (Abbott and Jones, 2015; Beamish et al., 2014). Respiration measured in other high Arctic polar desert sites was positively correlated with soil moisture (Emmerton et al., 2015).  This balance between reduced Re as a result of disturbance and potential increases as a result of increased soil moisture may result in the greater magnitude of Re relative to GPP and thus the overall shift to carbon source within the disturbance.  Despite the small magnitude of these high Arctic fluxes, there was a considerable effect of the permafrost disturbance as the net CO2 emissions from the disturbance were approximately three times larger than the net sequestration in the undisturbed tundra. Overall, the double EC system approach coupled with a source area model proved to be an effective method of accurately partitioning measured fluxes into undisturbed and disturbed contributions, and values were consistent with the static chamber measurements. By separating the growing season into three periods related to plant phenology, we were able to identify differences in NEE between undisturbed and disturbed tundra throughout the growing season. Initial sampling corresponded with leaf emergence, and as the season progressed, plant growth and leaf area increased, resulting in increased photosynthetic activity. The changes in NEE also corresponded to differences in PAR during the three periods of the 117  growing season. Phenological changes, especially in leaf emergence, growth and senescence, can be compared to the shift in CO2 fluxes as initially the undisturbed tundra was a source of CO2, but during peak growth there was a distinct shift to it being a CO2 sink. By the end of the sampling season, vegetation had begun to senesce, and this was reflected in reduced sink strength of NEEc in the undisturbed tundra. The disturbed areas contained low vegetation cover, resulting in very low GPP. Throughout the season, the environmental controls on CO2 fluxes in the disturbed tundra were PAR, Ta and VPD during the early and peak season, while PAR was a control in the late season. Estimates of landscape level impacts of permafrost disturbances in an 81 km2 ice-free land area on the Fosheim Peninsula, which included the area used for our study, were determined from satellite imagery and ground truthing in 2013. The analyses revealed that permafrost disturbances currently accounted for 0.34 km2 or only 0.4% of the landscape (A.C.A. Rudy, personal communication, 2015). Although the landscape area directly impacted by disturbance at this time is minimal, indirect impacts such as the lateral export of dissolved and particulate organic matter (hence, carbon) through streams and the hydrologic network are also important (Lamoureux and Lafrenière, 2009, Kokelj and Lewkowicz, 1998, Kokelj and Lewkowicz, 1999). The frequency and magnitude of these land surface disturbances appear to be increasing across the Fosheim Peninsula (and elsewhere in the Arctic) as a result of the warming climate, thus exacerbating these impacts (Lewkowicz, 1990; Lewkowicz and Harris, 2005b; Lantz and Kokelj, 2008; Segal et al., 2016). The increasing frequency and magnitude of these disturbances will affect the carbon balance at the landscape scale and could result in increased net CO2 emissions from these areas in the future. Organic carbon stored within permafrost has the potential to be released to the atmosphere as permafrost thaws (Schuur et al., 2008; Hicks Pries et al., 2011; 118  Hicks Pries et al., 2013). We quantified this release to the atmosphere and demonstrated that these permafrost disturbances are sources of CO2 over the measurement period during the growing season.  Potentially, some of the carbon in the soils could also be released in the form of methane (Anisimov, 2007; IPCC, 2007; Walter Anthony et al., 2012). Soil oxygen availability has been found to influence permafrost carbon that is released as both carbon dioxide and methane, and under aerobic conditions significantly more carbon is released as CO2 than CH4 (Lee et al., 2012). Consequently, we expect that methane release was relatively minimal from both the undisturbed and disturbed sites because of the aerobic conditions present in the moderately well-drained soils found in our study location. However, we also expect increased release of carbon with the deepening of the active layer and the increase in frequency and magnitude of permafrost disturbances. In addition, inorganic carbon released with the dissolution of carbonates and weathering may result in ventilation of CO2 and thus increased emissions (Lovett et al., 2006; Perez-Priego et al., 2013; Serrano-Ortiz et al., 2010). With increasing soil moisture, soil ventilation associated with carbonates may increase overall Re (Emmerton et al., 2015). However, slow carbon evolution in tundra soils (as a result of the release of inorganic carbon from carbonates) would limit this influence (Billings et al., 1977).  Shoulder season and winter respiration have been shown to be significant in various studies for year-round estimates of the effects on the carbon cycle (Nordstroem et al., 2001; Welker et al., 2004; Johansson et al., 2006; Humphreys and Lafleur, 2011; Wang et al., 2011; Lund et al., 2012), however only growing season fluxes were considered in our study. Due to logistical constraints, our sampling season was limited to approximately 30 days after snowmelt had occurred. As these disturbances were dynamic in nature, the site could not be left alone as 119  personnel were needed to monitor the slide edge location and adjust the equipment as needed. Leaving the site unmanned would have put the equipment at risk. Starr and Oberbauer (2003) have found photosynthetic activity in vascular plants under snow further indicating the importance of fluxes outside the snow free period. These fluxes were not considered in our study and could alter the annual carbon balance.  However, year round measurements of carbon exchange in areas impacted by permafrost thaw in Alaska indicate these areas act as sources of carbon over multiple years (Vogel et al., 2009). 4.6 Conclusion Using a dual EC sampling approach, in combination with the turbulent source area model and complemented by static chamber measurements, we were able to determine fluxes from one representative retrogressive thaw slump nearly continuously over a majority of the 2014 growing season. We found that these disturbances modify the NEE of the tundra, changing it from a net sink to a source of CO2. With a reduction in plant cover (~ 3 % vs. 27 % in undisturbed tundra), the disturbance reduced the magnitude of both Re and GPP, although reductions to GPP were greater. The dual EC approach in combination with the source area model allowed accurate assessments of the contributions of disturbed and undisturbed areas to CO2 fluxes so we could quantify the effect of permafrost disturbance on NEE. This approach may be preferable to measurements taken using manual portable chamber systems due to the continuous sampling frequency and spatial integration of the signal.      120  Chapter 5: General conclusions 5.1 Summary To examine the ecosystem effects of permafrost disturbance in the Canadian High Arctic, the impacts of active layer detachment slides (ALDS) and retrogressive thaw slumps (RTS) on vegetation and CO2 flux were analyzed over the 2012 – 2014 growing seasons. Active and recovered disturbances of various ages were compared to determine the nature of vegetation recovery. Multiple study locations were sampled on the Fosheim Peninsula, Ellesmere Island, to examine variability. Vegetation and site characteristics were measured to determine the impacts on ecosystem structure (Ch. 2 & Ch. 3). Static chamber systems and the eddy covariance sampling technique were used to determine the impacts of permafrost disturbance on carbon flux dynamics (Ch.3 & Ch. 4).   To analyze the patterns of tundra ecosystem recovery within ALDS that have remained stable over the past 20 years (Objective 1), historical data collected from ALDS of varying ages in 1994 was synthesized with data collected in the same sites after two decades of recovery (Ch. 2). In the chronosequence of ALDS, species composition and cover changed over the 20 year period. We showed that unique species are indicative of the stages of succession present in the toe and scar zone and unique vegetation communities are still present in the disturbances. It is likely that differences in community composition correspond with differences in soil characteristics, including soil moisture and nutrient availability. Recovered ALDS and RTS contain unique microclimate conditions (Objective 2), distinguished from surrounding undisturbed terrain by significant differences in soil moisture, soil nutrients, and active layer thickness which influence local vegetation composition. Soil moisture values were greatest in the youngest ALDS, especially in scar zones. Increased soil moisture was also found in RTS. 121  Increased nutrient availability was found in disturbed soils of both ALDS and RTS, in particular nitrate and magnesium. Active layer thickness differed based on age of disturbance, with increase in thickness found in old ALDS, while decreased thickness was found in younger ALDS. Both increases and decreases in active layer thickness were found in RTS. In some cases, differences between different disturbances were greater than the differences between disturbed and undisturbed plots. Fine scale heterogeneity created as a result of disturbances increases landscape heterogeneity in the Arctic. The impacts of disturbance were analyzed by measuring NEE and Re in several active and one stable thaw slump (Objective 2). NEE differed between active RTS and undisturbed control tundra, due to decreased vegetation found within active RTS, resulting in less CO2 uptake in disturbances during the growing season. This finding was confirmed at multiple disturbances over the 2013 and 2014 growing seasons. Disturbed areas in active RTS sampled in 2013 and 2014 were characterized by a decrease in carbon uptake when compared to surrounding undisturbed areas. For example, over the measurement period in 2014, the undisturbed tundra was a small net sink (NEE = -0.1 g C m-2 day-1) while the terrain disturbed by the RTS was a net source (NEE = +0.4 g C m-2 day-1). However, I also found spatial variability in carbon uptake from undisturbed tundra, as two sites sampled in 2013 differed significantly in the magnitude of carbon uptake (-0.05 g C m-2 day-1 and -0.20 g C m-2 day-1). Differences in NEE and Re were evident at all sites; as such, responses following disturbance are contingent on landscape characteristics prior to disturbance.   I tested the feasibility of utilizing eddy covariance to measure NEE from permafrost disturbances in the High Arctic (Objective 3; Ch. 3 & Ch. 4). It was found that the impact of permafrost disturbance can be measured using eddy covariance even in these areas with small CO2 fluxes. Flux partitioning based on wind direction and a dual eddy covariance sampling 122  approach in combination with source area modelling were developed and tested. The latter approach is preferable due to the simultaneous measurement of both disturbed and undisturbed fluxes, however isolated disturbances are required and they may be difficult to identify in heavily disturbed landscapes. If this is the case, partitioning fluxes based on wind direction allows disturbed and undisturbed fluxes to be separated. Measurements from flux towers were consistent with values measured using a portable chamber system indicating agreement between chamber and EC approaches.   Over the 2013 and 2014 growing season, decreases were found in carbon uptake over the growing season associated with active RTS. These changes corresponded with decreases in GPP greater relative to Re that resulted in an overall decrease in NEE.  In both years, at two sites, the disturbed sites were a carbon source, while the undisturbed tundra was a small sink. These shifts corresponded with vegetation and environmental characteristics of disturbed environments, including reduced vegetation cover, and increased soil moisture and temperature.  Reduced vegetation cover was also noted in ALDS during the 2012 growing season. These sites likely also differ in their NEE and component fluxes as compared to surrounding undisturbed tundra. These compositional differences that persist through recovery may be therefore reflected in corresponding ecosystem fluxes from these disturbances.    Rates of NEE measured at High Arctic sites are generally smaller than those measured from Low Arctic sites (Lafleur et al., 2012). Our measurements are comparable to those measured in other High Arctic locations, including Lake Hazen, Ellesmere Island and Cape Bounty, Melville Island (Lafleur et al., 2012; Emmerton et al., 2015). Despite the small magnitude of measured fluxes, the impact of disturbance is still evident, as seen through measured values of NEE at three RTS in 2013 and 2014. Our results contrast increases in 123  growing season NEE and GPP associated with permafrost thaw at sites in the Low Arctic, likely due to productivity differences between these sites (Trucco et al., 2012). As fluxes are smaller in the High Arctic, the impacts of disturbances here will not have the same effects on the regional carbon balance or the atmosphere as in the Low Arctic.  5.2 Future directions We visited ALDS nearly two decades after they were initially sampled to determine the long-term impacts of disturbance. As vegetation recovery is slow in High Arctic locations, future sampling of these sites is necessary to further understand the long-term impacts of disturbance. In the sub-Arctic, the ground thermal regime can exhibit effects of disturbance for a century following disturbance (Bartleman et al., 2001). In fact, Cray (2015) found residual impacts after multiple centuries. As such, visiting these sites in the future is essential to determine the long-term patterns of recovery.  Shoulder season fluxes have been shown to be an important component of annual carbon budgets (Nordstroem et al., 2001; Welker et al., 2004; Johansson et al., 2006; Humphreys and Lafleur, 2011; Wang et al., 2011; Lund et al., 2012). Extending growing season measurements of CO2 and incorporating shoulder season measurements would allow a comprehensive record of annual fluxes. Photosynthetic activity can occur in vascular plants underneath snow (Starr and Oberbauer, 2003), therefore measurements during the non-growing season may be important to the annual carbon budget. Measurements of CO2 over the non-growing season indicate a loss of CO2 over the wintertime, with increases in the magnitude of this loss associated with warming temperatures, and resulting in a shift in some tundra communities from C sink to C source (Oechel et al., 2014; Webb et al., 2015). However, due to logistical constraints at our remote location, we could not measure shoulder season or wintertime fluxes.  124  Expanding flux measurements to include methane (CH4) fluxes would provide a more comprehensive understanding of carbon flux dynamics, especially given the potential CH4 released from permafrost soils which can occur following permafrost thaw (Anisimov, 2007; IPCC, 2007; Walter Anthony et al., 2012). Utilizing a field deployable fast methane analyzer in conjunction with the IRGA on EC towers would provide estimates of methane concentrations from disturbed and undisturbed terrain. This could be complemented with measurements made using a portable chamber system to determine CH4 efflux and uptake at a finer scale. However, the fine scale spatial variability of CH4 will present a challenge to representatively measure CH4 fluxes.  As noted in Ch.2, multiple ALDS transitioned to active RTS between 1994 and 2012. An inventory of disturbances in the area including type and extent would allow us to examine the rate of transition. Revisiting sites frequently, and using high resolution remote sensing would allow a comparison of the active duration of RTS vs. ALDS and monitoring the transition from ALDS to RTS. Future analysis could be extended to incorporate these characteristics. Determining the conditions and timing during which ALDS transition to RTS would allow us to better understand these disturbances and potentially predict future occurrences and land-climate feedbacks. Detecting disturbances using remote sensing would allow us to better understand the overall magnitude of disturbance at a landscape scale, the types of disturbance present (ALDS vs. RTS), and associated changes in carbon fluxes.   5.3 General implications Ecosystem structure and function are impacted by permafrost disturbances. With increasing frequency and magnitude of these disturbances, based on the current study there is evidence that the carbon budget of tundra ecosystems will be significantly altered, in some cases shifting the 125  system from a carbon sink to a carbon source over the growing season. Disturbance may result in the release of previously stored carbon, alter hydrological networks, influence water quality, and decrease wildlife habitat (Walker, 1996; Schuur et al., 2008; Lamoureux and Lafrenière, 2009; Kokelj and Lewkowicz, 1998, 1999). The initial formation of ALDS and transition to RTS increases the duration and magnitude of landscape disturbance; with climate warming, this could have widespread ecosystem impacts. Understanding the dynamics of these disturbances and their overall ecosystem impacts is essential for predicting future landscape changes associated with climate change. As flux measurements are rare in the High Arctic, there is a need for more evidence to determine whether findings in this study are universal for Ellesmere Island or even the entire Arctic. The Arctic is warming faster than elsewhere on Earth (Weller et al., 2005). Using recent climate model projections (CMIP5), the business-as-usual Representative Concentration Pathway (RCP) scenario 8.5 predicts an Arctic wide temperature increase of 13 °C in fall and 5°C in spring by 2100 (Overland et al., 2013). However, feedback processes are not included in these climate models. When the permafrost carbon feedback is included in global climate models, an additional increase in temperatures is estimated to range between 0.13 – 1.69°C by 2300 (MacDougall et al., 2012). As this region is so heavily impacted by climate change, understanding Arctic ecosystems and how they respond to changes is essential for predicting future changes.   126  References Abbott, BW, and JB Jones. 2015. Permafrost collapse alters soil carbon stocks, respiration, CH4, and N2O in upland tundra. Global Change Biology. doi:10.1111/gcb.13069. Abbott, BW, JB Jones, EA Schuur, FS Chapin III, WB Bowden, MS Bret-Harte, et al.  2016. Biomass offsets little or none of permafrost carbon release from soils, streams, and      wildfire: an expert assessment. Environmental Research Letters, 11(3), 034014. ACIA. 2005. Arctic Climate Impact Assessment and Arctic Monitoring and Assessment Programme and Program for the Conservation of Arctic Flora and Fauna and International Arctic Science Committee. Cambridge University Press. Aiken, SG, MJ Dallwitz, LL Consaul, CL McJannet, RL Boles, GW Argus, JM Gillett, et al. 2007. Flora of the Canadian Arctic Archipelago: Descriptions, illustrations, identification, and information retrieval. Ottawa: NRC Research Press, National Research Council of Canada. Anisimov, OA, AA Velichko, PF Demchenko, AV Eliseev, II Mokhov, and VP Nechaev. 2002. Effect of climate change on permafrost in the past, present, and future. Izvestiya Atmospheric and Oceanic Physics, 38(1), S25. Anisimov, OA. 2007. Potential feedback of thawing permafrost to the global climate system through methane emission. Environmental Research Letters. 2 (4): 045016, doi:10.1088/1748-9326/2/4/045016. Auerbach, NA, MD Walker, and DA Walker. 1997. Effects of roadside disturbance on substrate and vegetation properties in Arctic tundra. Ecological Applications 7 (1): 218–35.  Babb, TA, and LC Bliss. 1974. Effects of physical disturbance on Arctic vegetation in the Queen Elizabeth Islands. Journal of Applied Ecology 11 (2): 549-562. Baldocchi, D. 2008. Turner Review No. 15: ‘Breathing’ of the terrestrial biosphere: Lessons learned from a global network of carbon dioxide flux measurement systems. Australian Journal of Botany 56 (1). CSIRO: 1–26. Bartleman, AP, K Miyanishi, CR Burn, and MM Cote. 2001. Development of vegetation communities in a retrogressive thaw slump near Mayo, Yukon Territory: A 10-Year Assessment. Arctic 54 (2): 149–56.  Beamish, A, A Neil, I Wagner, and NA Scott. 2014. Short-term impacts of active layer detachments on carbon exchange in a High Arctic ecosystem, Cape Bounty, Nunavut, Canada. Polar Biology 37: 1459–68. 127  Becker, MS, TJ Davies, and WH Pollard. 2015. Ground ice melt in the High Arctic leads to greater ecological heterogeneity. Journal of Ecology. doi:10.1111/1365-2745.12491. Bell, T. 1996. The last glaciation and sea level history of Fosheim Peninsula, Ellesmere Island, Canadian High Arctic. Canadian Journal of Earth Sciences 33 (7): 1075–86.  Billings, WD, KM Peterson, GR Shaver, and AW Trent. 1977. Root growth, respiration, and carbon dioxide evolution in an Arctic tundra soil. Arctic and Alpine Research 9 (2): 129–37. Billings, WD. 1973. Arctic and alpine vegetations: Similarities, differences, and susceptibility to disturbance. BioScience 23 (12): 697–704.  Bliss, LC, GM Courtin, DL Pattie, RR Riewe, DWA Whitfield, and P Widden. 1973. Arctic tundra ecosystems. Annual Review of Ecology and Systematics, 359–99. Bliss, LC, and WG Gold. 1994. The patterning of plant communities and edaphic factors along a High Arctic coastline: Implications for succession. Canadian Journal of Botany 72 (8): 1095–1107.  Bliss, LC, and NV Matveyeva. 1992. Circumpolar arctic vegetation. In Arctic Ecosystems in a changing climate: An ecophysiological perspective (ed: FS Chapin III, RL Jefferies, JF Reynolds, GR Shaver, J Svoboda and EW Chu). San Diego, California. Academic Press, pp. 59-89. Bliss, LC, and RW Wein. 1972. Plant community responses to disturbances in the Western Canadian Arctic. Canadian Journal of Botany 50: 1097-1109.  Bosquet, LM. 2011. The effects of observed and experimental climate change and permafrost disturbance on tundra vegetation in the Western Canadian High Arctic. Queen’s University, Kingston, Ontario: Canadian theses. Burba, G, A Schmidt, RL Scott, T Nakai, J Kathilankal, G Fratini, C Hanson, et al. 2012. Calculating CO2 and H2O eddy covariance fluxes from an enclosed gas analyzer using an instantaneous mixing ratio. Global Change Biology 18: 385–99.  Burn, CR, and PA Friele. 1989. Geomorphology, vegetation succession, soil characteristics and permafrost in retrogressive thaw slumps near Mayo, Yukon Territory. Arctic 42 (1): 31–40. Caceres, MD, F Jansen, and MM De Caceres. 2011. Package ‘Indicspecies.’ Probability.ca, 1–16, available at: http://www.probability.ca/cran/web/packages/indicspecies/indicspecies.pdf, (last access: 30 May 2016). Cannone, N, AG Lewkowicz, and M Guglielmin. 2010. Vegetation colonization of permafrost-related landslides, Ellesmere Island, Canadian High Arctic. Journal of Geophysical Research 115 (G4): G04020.  128  Cassidy, AE. 2011. The effects of recent and relict permafrost disturbances on tundra vegetation, Cape Bounty, Melville Island, Nunavut. Queen’s University, Kingston, Ontario: Canadian theses. Cassidy, AE, A Christen, GHR Henry. 2016. The effect of a permafrost disturbance on growing-season carbon dioxide fluxes in a high Arctic tundra ecosystem. Biogeosciences 13: 2291-2303. Chapin III, FS, and GR Shaver. 1985. Individualistic growth response of tundra plant species to environmental manipulations in the field. Ecology 66 (2): 564.  Chapin III, FS, W Eugster, JP McFadden, AH Lynch, and DA Walker. 2000. Summer differences among Arctic ecosystems in regional climate forcing. Journal of Climate 13 (12): 2002–10.  Chapin III, FS, JF Reynolds, GR Shaver, and J Svoboda. 1992. Arctic Ecosystems in a Changing Climate: An Ecophysiological Perspective. Academic Press. San Diego, California. Chapin III, FS, and GR Shaver. 1981. Changes in soil properties and vegetation following disturbance of Alaskan Arctic tundra. Journal of Applied Ecology 18 (2): 605–17. Chin, KS, J Lento,  JM Culp,  D Lacelle, and SV Kokelj. 2016. Permafrost thaw and intense thermokarst activity decreases abundance of stream benthic macroinvertebrates. Global Change Biology. Christen, A, NC Coops, BR Crawford, R Kellett, KN Liss, I Olchovski et al. 2011. Validation of modeled carbon dioxide emissions from an urban neighborhood with direct eddy-covariance measurements. Atmospheric Environment, 45: 6057–6069. Clarke, KR and RN Gorley. 2001. Primer v5: Users Manual/Tutorial. Primer-E Ltd, Plymouth, Maine.  Clements, FE. 1916. Plant succession, an analysis of the development of vegetation. Carnegie Institution of Washington 242: 1–512.  Cornelissen, JHC, PM Van Bodegom, R Aerts, TV Callaghan, RSP Van Logtestijn, J Alatalo, FS Chapin, et al. 2007. Global negative vegetation feedback to climate warming responses of leaf litter decomposition rates in cold biomes. Ecology Letters 10: 619–27. Cray, HA, and WH Pollard. 2015. Vegetation recovery patterns following permafrost disturbance in a Low Arctic setting: Case study of Herschel Island, Yukon, Canada. Arctic, Antarctic, and Alpine Research 1: 99–113. 129  Dallimore, SR, SA Wolfe, and SM Solomon. 1996. Influence of ground ice and permafrost on coastal evolution, Richards Island, Beaufort Sea Coast, N.W.T. Canadian Journal of Earth Sciences 33 (5): 664–75.  Davidson, EA, and IA Janssens. 2006. Temperature sensitivity of soil carbon decomposition and feedbacks to climate change. Nature 440: 165–73.  Desforges, M. 2000. Plant succession after active layer detachment slides in High Arctic tundra, Fosheim Peninsula, Ellesmere Island, Canada. The University of British Columbia, Vancouver, BC: Canadian theses. Edlund, SA, BT Alt, and M Garneau. 2000. Vegetation patterns on Fosheim Peninsula, Ellesmere Island, Nunavut. Geological Survey of Canada Bulletin 529: 129–43. Edlund, SA, BT Alt, and KL Young. 1989. Interaction of climate, vegetation, and soil hydrology at Hot Weather Creek, Fosheim Peninsula, Ellesmere Island, Northwest Territories. Geological Survey of Canada Current Research 1989-D.  Emers, M, JC Jorgenson, and MK Raynolds. 1995. Response of Arctic tundra plant communities to winter vehicle disturbance. Canadian Journal of Botany 73 (6): 905–17.  Emmerton, CA, VL St. Louis, ER Humphreys, J Gamon, JD Barker, and GZ Pastorello. 2015. Net ecosystem exchange of CO2 with rapidly changing High Arctic landscapes. Global Change Biology doi:10.1111/gcb.13064. Environment Canada, Canadian Climate Normals 1981-2010 Station Data, Environment  Canada, available at: http://www.climate.weather.gc.ca (last access: 15 April 2015), 2015. Forbes, BC, JJ Ebersole, and B Strandberg. 2001. Anthropogenic disturbance and patch dynamics in circumpolar Arctic ecosystems. Conservation Biology 15 (4): 954-969. Fox, AM, B Huntley, CR Lloyd, M Williams, and R Baxter. 2008. Net ecosystem exchange over heterogeneous Arctic tundra: Scaling between chamber and eddy covariance measurements. Global Biogeochemical Cycles 22: GB2027. French, HM, and P Egginton. 1973. Thermokarst development, Banks Island, Western Canadian Arctic. In 2nd International Conference on Permafrost, 1:203–12. Geertsema, M, and JJ Pojar. 2007. Influence of landslides on biophysical diversity—A perspective from British Columbia. Geomorphology 89 (1-2): 55–69.  Gleason, HA. 1926. The individualistic concept of the plant association. Bulletin of the Torrey Botanical Club 53 (1): 7–26.  130  Griffis, TJ, and WR Rouse. 2001. Modelling the interannual variability of net ecosystem CO2 exchange at a subarctic sedge fen. Global Change Biology 7 (5): 511–30. Groendahl, L, T Friborg, and H Soegaard. 2007. Temperature and snow-melt controls on interannual variability in carbon exchange in the High Arctic. Theoretical and Applied Climatology 88: 111–25. Grosse, G, J Harden, M Turetsky, AD McGuire, P Camill, C Tarnocai, S Frolking, et al. 2011. Vulnerability of high-latitude soil organic carbon in North America to disturbance. Journal of Geophysical Research 116 (G4): G00K06. Harris, C, and AG Lewkowicz. 1993. Form and internal structure of active-layer detachment slides, Fosheim Peninsula, Ellesmere Island, Northwest Territories, Canada. Canadian Journal of Earth Sciences 30: 1708–14. Harris, C, B Rea, and M Davies. 2001. Scaled physical modelling of mass movement processes on thawing slopes. Permafrost and Periglacial Processes 12 (1): 125–35. Harry, DG, HM French, and WH Pollard. 1988. Massive ground ice and ice-cored terrain near Sabine Point, Yukon Coastal Plain. Canadian Journal of Earth Sciences 25: 1846-1856.   Hicks Pries, CE, EAG Schuur, and KG Crummer. 2013. Thawing permafrost increases old soil and autotrophic respiration in tundra: Partitioning ecosystem respiration using δ(13) C and ∆(14) C. Global Change Biology 19 (2): 649–61. Hicks Pries, CE, EAG Schuur, and KG Crummer. 2011. Holocene carbon stocks and carbon accumulation rates altered in soils undergoing permafrost thaw. Ecosystems 15 (1): 162–73. Hollinger, DY, and AD Richardson. 2005. Uncertainty in eddy covariance measurements and its application to physiological models. Tree Physiology 25: 873–85.  Hugelius, G, C Tarnocai, G Broll, JG Canadell, P Kuhry, and DK Swanson. 2013. The Northern Circumpolar Soil Carbon Database: Spatially distributed datasets of soil coverage and soil carbon storage in the northern permafrost regions. Earth System Science Data 5: 3–13.  Hugelius, G, J Strauss, S Zubrzycki, JW Harden, EAG Schuur, CL Ping, et al. 2014. Estimated stocks of circumpolar permafrost carbon with quantified uncertainty ranges and identified data gaps. Biogeosciences, 11(23): 6573–6593. Humphreys, ER, and PM Lafleur. 2011. Does earlier snowmelt lead to greater CO2 sequestration in two Low Arctic tundra ecosystems? Geophysical Research Letters 38: L09703.  IPCC. 2007. Climate Change 2007: The Physical Science Basis. Intergovernmental Panel on Climate Change 446: 727–28. 131  Jassal, R, A Black, M Novak, K Morgenstern, Z Nesic, and D Gaumont-Guay. 2005. Relationship between soil CO2 concentrations and forest-floor CO2 effluxes. Agricultural and Forest Meteorology 130: 176–92.  Jensen, AE, KA Lohse, BT Crosby, and CI Mora. 2014. Variations in soil carbon dioxide efflux across a thaw slump chronosequence in northwestern Alaska. Environmental Research Letters 9 (2): 025001.  Johansson, T, N Malmer, PM Crill, T Friborg, JH Åkerman, M Mastepanov, and TR Christensen. 2006. Decadal vegetation changes in a northern peatland, greenhouse gas fluxes and net radiative forcing. Global Change Biology 12: 2352–69.  Jones, GA, and GHR Henry. 2003. Primary plant succession on recently deglaciated rerrain in the Canadian High Arctic. Journal of Biogeography 30: 277–96.  Kokelj, SV, TC Lantz, SL Smith, JCN Kanigan, and R Coutts. 2009. Origin and polycyclic behaviour of tundra thaw slumps, Mackenzie Delta region, Northwest Territories, Canada. Permafrost and Periglacial Processes 20: 173–84.  Kokelj, SV, and AG Lewkowicz. 1999. Salinization of permafrost terrain due to natural geomorphic disturbance, Fosheim Peninsula, Ellesmere Island. Arctic 52 (4): 372–85.  Kokelj, SV, and AG Lewkowicz. 1998. Long-term influence of active-layer detachment sliding on permafrost slope hydrology, Hot Weather Creek, Ellesmere Island, Canada. Proceeding of the Seventh International Conference on Permafrost, Yellowknife, Canada, Nordicana 55: 583–89.  Kokelj, SV, and MT Jorgenson. 2013. Advances in Thermokarst Research. Permafrost and Periglacial Processes 24 (2): 108–19. Kokelj, SV, D Lacelle, TC Lantz, J Tunnicliffe, L Malone, ID Clark, and KS Chin. 2013. Thawing of Massive Ground Ice in Mega Slumps Drives Increases in Stream Sediment and Solute Flux across a Range of Watershed Scales. Journal of Geophysical Research: Earth Surface 118 (2): 681–92.  Kokelj, SV, J Tunnicliffe, D Lacelle, TC Lantz, KS Chin, and R Fraser. 2015. Increased Precipitation Drives Mega Slump Development and Destabilization of Ice-Rich Permafrost Terrain, Northwestern Canada. Global and Planetary Change 129: 56–68. Kormann, R, and FX Meixner. 2001. An analytical footprint model for non-neutral stratification. Boundary-Layer Meteorology 99: 207–24.  Kwon, HJ, WC Oechel, RC Zulueta, and SJ Hastings. 2006. Effects of climate variability on carbon sequestration among adjacent wet sedge tundra and moist tussock tundra ecosystems. Journal of Geophysical Research: Biogeosciences 111: G03014.  132  Lacelle, D, J Bjornson, and B Lauriol. 2010. Climatic and geomorphic factors affecting contemporary (1950-2004) activity of retrogressive thaw slumps on the Aklavik Plateau, Richardson Mountains, NWT, Canada. Permafrost and Periglacial Processes 21: 1–15.  Lafleur, PM, ER Humphreys, VL St. Louis, MC Myklebust, T Papakyriakou, L Poissant, JD Barker, M Pilote, and KA Swystun. 2012. Variation in peak growing season net ecosystem production across the Canadian Arctic. Environmental Science & Technology 46: 7971-7977.  Lambert, JDH. 1972. Plant succession on tundra mudflows: Preliminary observations.” Arctic 1: 99-106.  Lamoureux, SF, and MJ Lafrenière. 2009. Fluvial impact of extensive active layer detachments, Cape Bounty, Melville Island, Canada. Arctic, Antarctic, and Alpine Research 41 (1): 59–68.  Lantuit, H, and WH Pollard. 2008. Fifty years of coastal erosion and retrogressive thaw slump activity on Herschel Island, Southern Beaufort Sea, Yukon Territory, Canada. Geomorphology 95 (1): 84–102. Lantz, TC, and SV Kokelj. 2008. Increasing rates of retrogressive thaw slump activity in the Mackenzie Delta Region, NWT, Canada. Geophysical Research Letters 35 (6): L06502. Lantz, TC, SV Kokelj, SE Gergel, and GHR Henry. 2009. Relative impacts of disturbance and temperature: Persistent changes in microenvironment and vegetation in retrogressive thaw slumps. Global Change Biology 15 (7): 1664–75.  Lee, H, EAG Schuur, KS Inglett, M Lavoie, and JP Chanton. 2012. The rate of permafrost carbon release under aerobic and anaerobic conditions and its potential effects on climate. Global Change Biology 18: 515–27. Legendre, P, and L Legendre. 1998. Numerical Ecology. Numerical Ecology Second English Edition 20 (20): 870. Leibman, MO. 1995. Cryogenic landslides on the Yamal Peninsula, Russia: Preliminary observations. Permafrost and Periglacial Processes 6: 259–64.  Leibman, MO, and AI Kizakov. 2003. Dynamics of landslide slopes and their development on Yamal Peninsula. Permafrost/ Proceedings of the 8th international conference on permafrost. Swets and Zeitlinger. Lisse 651-656. Lesins, GT, J Duck, and JR Drummond. 2010. Climate Trends at Eureka in the Canadian High Arctic. Atmosphere-Ocean 48 (2): 59–80. 133  Lewkowicz, AG. 1987. Headwall retreat of ground-ice slumps, Banks Island, Northwest Territories. Canadian Journal of Earth Sciences 24 (6): 1077–85. Lewkowicz, AG. 1990. Morphology, frequency and magnitude of active layer detachment slides, Fosheim Peninsula, Ellesmere Island, NWT. In Permafrost-Canada, Proceedings of the Fifth Canadian Permafrost Conference: Université Laval, Nordicana, 56:111–18. Lewkowicz, AG, and SV Kokelj. 2002. Slope sediment yield in arid lowland continuous permafrost environments, Canadian Arctic Archipelago. Catena 46 (4): 261–83.  Lewkowicz, AG, and C Harris. 2005. Frequency and magnitude of active-layer detachment failures in discontinuous and continuous permafrost, Northern Canada. Permafrost and Periglacial Processes 16 (1): 115–30. Lewkowicz, AG, and C Harris. 2005. Morphology and geotechnique of active-layer detachment failures in discontinuous and continuous permafrost, Northern Canada. Geomorphology 69 (1-4): 275–97.  Lewkowicz, AG. 2007. Dynamics of active-layer detachment failures, Fosheim Peninsula, Ellesmere Island, Nunavut, Canada. Permafrost and Periglacial Processes 18 (1): 89–103. Lipovsky, PS, J Coates, AG Lewkowicz, and E Trochim. 2005. Active-layer detachments following the summer 2004 forest fires near Dawson City, Yukon. Yukon Exploration and Geology, Yukon Geological Survey, 175-194. Lovett, GM, JJ Cole, and ML Pace. 2006. Is net ecosystem production equal to ecosystem carbon accumulation? Ecosystems 9 (1): 152–55.  Lüers, J, S Westermann, K Piel, and J Boike. 2014. Annual CO2 budget and seasonal CO2 exchange signals at a High Arctic permafrost site on Spitsbergen, Svalbard Archipelago. Biogeosciences 11: 6307–22. Lund, M, JM Falk, T Friborg, HN Mbufong, C Sigsgaard, H Soegaard, and MP Tamstorf. 2012. Trends in CO2 exchange in a High Arctic tundra heath, 2000-2010. Journal of Geophysical Research: Biogeosciences 117: G02001. MacDougall, AH, CA Avis, and AJ Weaver. 2012. Significant contribution to climate warming from the permafrost carbon feedback. Nature Geoscience, 5: 719 – 721. Mackay, JR. 1966. Segregated epigenetic ice and slumps in permafrost, Mackenzie Delta area, NWT. Geographical Bulletin 8 (1): 59–80. Mackay, JR. 1970. Disturbances to the tundra and forest tundra environment of the Western Arctic. Canadian Geotechnical Journal 7: 420-432.  134  Matthews, JA. 1992. The Ecology of Recently Deglaciated Terrain: A Geoecological Approach to Glacier Forelands. Cambridge University Press.   Mauder, M, and T Foken. 2004. Documentation and Instruction Manual of the Eddy Covariance Software Package TK2. Mbufong, HN, M Lund, M Aurela, TR Christensen, W Eugster, T Friborg, BU Hansen, et al. 2014. Assessing the spatial variability in peak season CO2 exchange characteristics across the Arctic tundra using a light response curve parameterization. Biogeosciences 11 (17): 4897–4912.  McCune, B, and JB Grace. 2002. Analysis of Ecological Communities. Gleneden Beach: MjM Software Design. McGuire, AD, FS Chapin, JE Walsh, and C Wirth. 2006. Integrated regional changes in Arctic climate feedbacks: Implications for the global climate system. Annual Review of Environment and Resources 31 (1): 61–91. McRoberts, EC, and NR Morgenstern. 1974. The stability of thawing slopes. Canadian Geotechnical Journal 11 (4): 447-469. Merbold, L, WL Kutsch, C Corradi, O Kolle, C Rebmann, PC Stoy, SA Zimov, and ED Schulze. 2009. Artificial drainage and associated carbon fluxes (CO2/CH4) in a tundra ecosystem. Global Change Biology 15 (11): 2599–2614. Mesquita, PS, FJ Wrona, and TD Prowse. 2010. Effects of Retrogressive Permafrost Thaw Slumping on Sediment Chemistry and Submerged Macrophytes in Arctic Tundra Lakes. Freshwater Biology 55 (11): 2347–58.  Moquin, PA and FJ Wrona. 2015. Effects of permafrost degradation on water and sediment quality and heterotrophic bacterial production of Arctic tundra lakes: An experimental approach. Limnology and Oceanography, 60 (5), 1484-1497. Nordstroem, C, H Soegaard, TR Christensen, T Friborg, and BU Hansen. 2001. Seasonal carbon dioxide balance and respiration of a High-Arctic fen ecosystem in NE-Greenland. Theoretical and Applied Climatology 70: 149–66. Oberbauer, SF, CE Tweedie, JM Welker, JT Fahnestock, GHR Henry, PJ Webber, RD Hollister, et al. 2007. Tundra CO2 fluxes in response to experimental warming across latitudinal and moisture gradients. Ecological Monographs 77 (2). Eco Soc America: 221–38. Oechel, WC, CA Laskowski, G Burba, B Gioli, and AAM Kalhori. 2014. Annual patterns and budget of CO2 flux in an Arctic tussock tundra ecosystem.  Journal of Geophysical Research: Biogeosciences 119: 323–339. 135  Oksanen, J, FG Blanchet, R Kindt, P Legendre, RB O’Hara, GL Simpson, P Solymos, M Henry, H Stevens, and H Wagner. 2012. Vegan: Community Ecology Package. R Package Version 2.2-1.  Overland, JE, M Wang, JE Walsh, and JC Stroeve. 2013. Future Arctic climate changes: Adaptation and mitigation time scales. Earth’s Future, 2, doi: 10.1002/2013ER000162.  Paul-Limoges, E, A Christen, NC Coops, TA Black, and JA Trofymow. 2013. Estimation of aerodynamic roughness of a harvested Douglas-Fir forest using airborne LiDAR. Remote Sensing of Environment 136: 225–33.  Pérez-Priego, O, P Serrano-Ortiz, EP Sánchez-Cañete, F Domingo, and AS Kowalski. 2013. Isolating the effect of subterranean ventilation on CO2 emissions from drylands to the atmosphere. Agricultural and Forest Meteorology 180: 194–202.  Prach, K 1993. On the rate of succession. Oikos 2 (66): 343–46. R Core Team. 2014. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. UR: http://www.R-project.org/. Raynolds, MK, DA Walker, KJ Ambrosius, J Brown, KR Everett, M Kanevskiy, GP Kofinas, VE Romanovsky, Y Shur, and PJ Webber. 2014. Cumulative Geoecological Effects of 62 Years of Infrastructure and Climate Change in Ice-Rich Permafrost Landscapes, Prudhoe Bay Oilfield, Alaska. Global Change Biology 20 (4): 1211–24. Reichstein, M, PC Stoy, AR Desai, G Lasslop, and AD Richardson. 2012. Partitioning of net fluxes. In Eddy Covariance, 263–89. Robbins, JA and JA Matthews. 2010. Regional variation in successional trajectories and rates of vegetation change on glacier forelands in south-central Norway. Arctic, Antarctic, and Alpine Research, 42(3): 351-361. Robinson, SD. 2000. Thaw-slump-derived thermokarst near Hot Weather Creek, Ellesmere Island, Nunavut. Geological Survey of Canada Bulletin 529: 335–45.  Robinson, SD, and WH Pollard. 1998. Massive ground ice within Eureka Sound Bedrock. In Permafrost, Proceedings of the Seventh International Conference on Permafrost, Yellowknife (Canada), Collection Nordicana, 949–54. Schaefer, K, H Lantuit, VE Romanovsky, EAG Schuur, and R Witt. 2014. The impact of the permafrost carbon feedback on global climate. Environmental Research Letters 9 (8): 085003. Schuur, EAG, AD McGuire, G Grosse, JW Harden, DJ Hayes, G Hugelius, CD Koven, and P Kuhry. 2015. Climate change and the permafrost carbon feedback. Nature 520: 171–79. 136  Schuur, EAG, and B Abbott. 2011. High risk of permafrost thaw. Nature 480: 32–33. Schuur, EAG, J Bockheim, JG Canadell, E Euskirchen, CB Field, SV Goryachkin, S Hagemann, et al. 2008. Vulnerability of permafrost carbon to climate change: Implications for the global carbon cycle. BioScience 58 (8): 701–14. Segal, RA, TC Lantz, and SV Kokelj. 2016. Acceleration of thaw slump activity in glaciated landscapes of the Western Canadian Arctic.Environmental Research Letters, 11(3), 034025. Serrano-Ortiz, P, M Roland, S Sanchez-Moral, IA Janssens, F Domingo, Y Goddéris, and AS Kowalski. 2010. Hidden, abiotic CO2 flows and gaseous reservoirs in the terrestrial carbon cycle: Review and perspectives. Agricultural and Forest Meteorology 150 (3): 321–29.  Srivastava, DS, and RL Jefferies. 1995. The effect of salinity on the leaf and shoot demography of 2 Arctic forage species. Journal of Ecology 83 (3): 421–30. Starr, G, and SF. Oberbauer. 2003. Photosynthesis of Arctic evergreens under snow: Implications for tundra ecosystem carbon balance. Ecology 84: 1415–20. Stow, DA, A Hope, and D McGuire. 2004. Remote sensing of vegetation and land-cover change in Arctic tundra ecosystems. Remote Sensing of Environment 89 (3): 281–308.  Stoy, P, M Williams, J Evans, A Prieto-Blanco, M Disney, T Hill, H Ward, T Wade, and L Street. 2013. Upscaling tundra CO2 exchange from chamber to eddy covariance tower. Arctic, Antarctic, and Alpine Research 45: 275–84.  Svoboda, J, and GHR Henry. 1987. Succession in marginal Arctic environments. Arctic and Alpine Research 19 (4): 373-84. Tarnocai, C, JG Canadell, EAG Schuur, P Kuhry, G Mazhitova, and S Zimov. 2009. Soil organic carbon pools in the northern circumpolar permafrost region. Global Biogeochemical Cycles 23, GB2023. Thienpont, JR, KM Rühland, MFJ Pisaric, SV Kokelj, LE Kimpe, JM Blais, and JP Smol. 2013. Biological Responses to Permafrost Thaw Slumping in Canadian Arctic Lakes. Freshwater Biology 58 (2): 337–53.  Tishkov, AA. 1986. Primary succession in Arctic tundra on the west coast of Spitsbergen (Svalbard). Polar Geography and Geology 10 (2): 148–56. Trucco, C, EAG Schuur, SM Natali, EF Belshe, R Bracho, and J Vogel. 2012. Seven-year trends of CO2 exchange in a tundra ecosystem affected by long-term permafrost thaw, Journal of Geophysical Research 117: G02031. 137  Ukraintseva, NG. 2008. Vegetation response to landslide spreading and climate change in west Siberian tundra. In Proceedings of the 9th International Conference on Permafrost, Fairbanks, Alaska, 1793–98. Ukraintseva, NG, and MO Leibman. 2000. Productivity of willow-shrub tundra in connection with landslide activity. In Proceedings of the 30th Arctic Workshop Programme and Abstracts, INSTAAR, University of Colorado, Boulder, 16–18. Vincent, WF, TV Callaghan, D Dahl-Jensen, M Johansson, KM Kovacs, C Michel, T Prowse, JD Reist, and M Sharp. 2011. Ecological implications of changes in the Arctic cryosphere. Ambio 40: 87-99. Vogel, J, EAG Schuur, C Trucco, and H Lee. 2009. Response of CO2 exchange in a tussock tundra ecosystem to permafrost thaw and thermokarst development. Journal of Geophysical Research: Biogeosciences 114: G04018. Vourlitis, GL, Y Harazono, WC Oechel, M Yoshimoto, and M Mano. 2000. Spatial and temporal variations in hectare‐scale net CO2 flux, respiration and gross primary production of Arctic tundra ecosystems. Functional ecology 14(2): 203-214. Walker, DA. 1996. Disturbance and Recovery of Arctic Alaskan Vegetation. In Landscape Function and Disturbance in Arctic Tundra, 35–71. Walker, DA, and MD Walker. 1991. History and pattern of disturbance in Alaskan Arctic terrestrial ecosystems: A hierarchial approach to analyzing landscape change. Journal of Applied Ecology 28: 244–76. Walker, DA, MK Raynolds, FJ Daniëls, E. Einarsson, A Elvebakk, WA Gould, AE Katenin, SS Kholod, CJ Markon, ES Melnikov, NG Moskalenko, SS Talbot, BA Yurtsev, and the CAVM team. 2005. The Circumpolar Arctic vegetation map. Journal of Vegetation Science 16: 267-282. Walter Anthony, KM., P Anthony, G Grosse, and J Chanton. 2012. Geologic methane seeps along boundaries of Arctic permafrost thaw and melting glaciers. Nature Geoscience 5 (6): 419-426. Wang, T, P Ciais, SL Piao, C Ottlé, P Brender, F Maignan, A Arain, et al. 2011. Controls on winter ecosystem respiration in temperate and boreal ecosystems. Biogeosciences 8: 2009–25.  Webb, EE, EAG Schuur, SM Natali, KL Oken, R Bracho, JP Krapek, D Risk, and NR Nickerson. 2016. Increased wintertime CO2 loss as a result of sustained tundra warming. Journal of Geophysical Research: Biogeosciences 121: G002795. 138  Weller, G, E Bush, T Callaghan, R Corell, S Fox, C Furgal, HA Hoel, H Huntington, E Kallen, VM Kattsov, DR Klein, H Loeng, ML Martello, M MacCracken, M Nutall,  TD Prowse, LO Reiersen, JD  Reist,  A Tanskanen, JE Walsh, B Weatherhead, and FJ Wrona. 2005. Arctic Climate Impact Assessment: Summary and Synthesis of the ACIA. Pages 990-1020. (C. Symon, L Arris, and B. Heal, Eds.). Cambridge University Press, New York. Welker, JM, JT Fahnestock, and MH Jones. 2000. Annual CO2 flux in dry and moist Arctic tundra: Field responses to increases in summer temperatures and winter snow depth. Climatic Change 44: 139–50.  Welker, JM, JT Fahnestock, GHR Henry, KW O’Dea, and RA Chimner. 2004. CO2 exchange in three Canadian High Arctic ecosystems: Response to long-term experimental warming. Global Change Biology 10 (12): 1981–95. Young, SB. 1971. The vascular flora of Saint Lawrence Island, with special reference to floristic zonation in the Arctic regions. Contributions from the Gray Herbarium of Harvard University 201: 11–115.               139  Appendix A   Table A.1 Community characteristics of ALDS sampled in 2012 with Tukey HSD in vascular plant species richness and diversity (Shannon’s H). Different letters represent significant differences in species richness and diversity.  Age Zone Treatment Sp. Richness Shannon's H Young  scar undisturbed abc abc disturbed d d track undisturbed a a disturbed cd cd toe undisturbed ab ab disturbed ab ab Mature  scar undisturbed abcd abc disturbed abc abc track undisturbed cd cd disturbed abc abc toe undisturbed abc abc disturbed abcd abc Old  scar undisturbed abcd abc disturbed abc abc track undisturbed cd cd disturbed bcd abc toe undisturbed bcd bc disturbed abc ab          140  Table A.2 SIMPER analysis comparing vegetation composition of young and old ALDS (scars and toes) in 1994 and 2012  Age Zone Species Cumulative Dissimilarity (%) Old scar Dryas integrifolia 20.35 Salix arctica 38.61 Cassiope tetragona 52.59 moss 61.89 lichen 72.93 toe Dryas integrifolia 49.78 Salix arctica 26.41 Cassiope tetragona 66.58 Stellaria longipes 72.59 Young scar Puccinellia spp.  49.72 toe Salix arctica 31.75 Puccinellia spp. 54.30 Elymus alaskanus  79.40  Table A.3 SIMPER analysis of 2012 ALDS (all ages and zones) with adjacent undisturbed tundra Age Zone Species Cumulative Dissimilarity (%) Young scar Salix arctica, Dryas integrifolia, Puccinellia spp. 36.69/57.67/75.52 track Salix arctica, Dryas integrifolia, Puccinellia spp., Carex rupestris 33.35/49.37/60.22/70.31 toe Salix arctica, Dryas integrifolia, Puccinellia spp., Carex rupestris 57.27/30.55/72.05/64.76 Mature scar Salix arctica, Dryas integrifolia, Puccinellia spp., Carex rupestris 63.64/36.63/73.08 track Salix arctica, Dryas integrifolia, Puccinellia spp., Carex rupestris 43.65/64.41/75.59 toe Salix arctica, Dryas integrifolia, Puccinellia spp., Carex rupestris 66.80/39.99/82.14 Old scar Salix arctica, Dryas integrifolia, Cassiope tetragona, Carex nardina 53.21/27.60/63.39/71.63 track Salix arctica, Dryas integrifolia, Cassiope tetragona, Puccinellia spp.  32.93/57.47/78.41/69.71 toe Salix arctica, Dryas integrifolia, Cassiope tetragona, Stellaria longipes 27.03/51.06/68.50/74.83  141  Appendix B   Under cold conditions instrument surface heat may critically affect the ability to measure CO2 fluxes with an open-path sensor, in particular with an upright mounted sensor (Burba et al., 2008). The EddyPro® manual summarizes “When CO2 and H2O molar densities are measured with the LI-7500 in cold environments (low temperatures below -10 °C), a correction should be applied to account for the additional instrument-related sensible heat flux, due to instrument surface heating/cooling.” (https://www.licor.com/env/help/eddypro6/Content/Calculating_Off-season_Uptake_Correction.html).  This correction was not applied because it would lead to unrealistic values (‘overcorrection’) for the following three reasons: • Measurements were collected during July, with 24 hours sunlight. The average air temperature was +10ºC. Temperatures never dropped below 2°C, and reached as high as 16°C. This is not falling within the critical range mentioned in Burba et al. (2010) of < -10ºC where fluxes are affected.  • In addition, sensors (LI7500) were mounted tilted at an angle of 30º to minimize issues associated with heating and reduce pooling of moisture on the windows. The correction cannot be employed with a tilted sensor. Burba et al. (2008) write on the correction approach: "… assumes that the instrument is mounted in a near-vertical orientation.". Previous work by our colleague Prof. A. Black (UBC, pers. comm.) has shown that a tilted sensor does not cause differences between open-path and closed-path systems. The comparison between fluxes measured with a tilted Li-7500 sensor and a closed path system as function of temperature can be found in the appendix to this response. • In our current approach (i.e. without correction) closed chamber measurements and the EC approach match well (RMSE = 0.6 µmol m-2 s-1).  If we would correct fluxes according to Burba et al. (2008) assuming the sensor is upright mounted, we found that the correlation between chamber measurements and EC measurements becomes worse (RMSE = 1.4  µmol m-2 s-1). To us, this is a strong argument that the incorrectly applied correction (assuming a vertically mounted sensor) would not improve our dataset.    142  B.1 Sub-Appendix Demonstration that a tilted sensor mounting of the Li-7500 sensor causes minimal sensor disagreement with a closed-path system between 2ºC and 16ºC.  To investigate whether there are any systematic differences between measuring CO2 fluxes with a tilted open-path (OP, Li-7500) vs. a closed-path (CP) analyzer in the range of the currently observed temperatures, we used a dataset sampled by the UBC Biometeorology group (Prof. A. Black, pers. comm.) over a forest clear-cut in Saskatchewan, Canada (Fluxnet site HJP02, Zha et al. 2009, Figure 1). We used data from June – December 2004 when a LI-7500 open path analyzer tilted by about 30º (same tilt as in the current submission) and a closed path analyzer (LI-7000, Fluxnet Canada standard) were operated. Temperatures analyzed covered the range from -15 to +20ºC.   Figure B.1 Tilted open-path and inlet for closed-path system at HJP02 (Photo: Zoran Nesic, UBC)  The half-hourly differences between all FCO2 measured by the tilted OP and FCO2 measured by the CP dataset were binned by air temperature (2 K bins). Figure 2 shows that for the range between 0 and 20ºC, the systems are not systematically different, while below 0ºC the OP 143  instrument starts to systematically underestimate fluxes, which becomes a major issue < -10ºC - presumably due to the sensor heating effect. As we measure in this thesis always between 2ºC and 16ºC and on average 10ºC we do not see any evidence that our fluxes are compromised with a tilted sensor mounting of the Li-7500.   Figure B.2 Difference between CO2 fluxes determined by open-path and closed-path systems from 1 July to  31 December 2004 at the Canada Fluxnet Site HJP02     10 19 17 29  49 79 190 340  276 219  104 115  70  95  35  28+-36DVNDWFKHZDQ-XOï'HFOpen path is Li-7500 tilted by ~30ºï ï 0 5 10 15Air temperature (ºC)ïï2SHQïSDWKïFORVHGïSDWKLQ—PROP-2 s-1  Temperature range of current studyPossible instrument heat !ux e"ect?Di!erence in measured CO2 "uxes

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}]}"
                            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:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0315351/manifest

Comment

Related Items