AN EXAMINATION OF THE LIMNOLOGY, CHIRONOMID BIOGEOGRAPHY AND PALEOECOLOGY OF EASTERN CANADIAN ARCTIC AQUATIC ECOSYSTEMS
ANDREW S. MEDEIROS
A DISSERTATION SUBMITTED TO THE FACULTY OF GRADUATE STUDIES IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY
GRADUATE PROGRAM IN BIOLOGY YORK UNIVERSITY TORONTO, ONTARIO
AUGUST 2011 Library and Archives Bibliotheque et 1*1 Canada Archives Canada Published Heritage Direction du Branch Patrimoine de I'edition
395 Wellington Street 395, rue Wellington OttawaONK1A0N4 OttawaONK1A0N4 Canada Canada
Your file Votre reference ISBN: 978-0-494-80524-4 Our file Notre reference ISBN: 978-0-494-80524-4
NOTICE: AVIS:
The author has granted a non L'auteur a accorde une licence non exclusive exclusive license allowing Library and permettant a la Bibliotheque et Archives Archives Canada to reproduce, Canada de reproduire, publier, archiver, publish, archive, preserve, conserve, sauvegarder, conserver, transmettre au public communicate to the public by par telecommunication ou par I'lnternet, preter, telecommunication or on the Internet, distribuer et vendre des theses partout dans le loan, distribute and sell theses monde, a des fins commerciales ou autres, sur worldwide, for commercial or non support microforme, papier, electronique et/ou commercial purposes, in microform, autres formats. paper, electronic and/or any other formats.
The author retains copyright L'auteur conserve la propriete du droit d'auteur ownership and moral rights in this et des droits moraux qui protege cette these. Ni thesis. Neither the thesis nor la these ni des extraits substantiels de celle-ci substantial extracts from it may be ne doivent etre imprimes ou autrement printed or otherwise reproduced reproduits sans son autorisation. without the author's permission.
In compliance with the Canadian Conformement a la loi canadienne sur la Privacy Act some supporting forms protection de la vie privee, quelques may have been removed from this formulaires secondaires ont ete enleves de thesis. cette these.
While these forms may be included Bien que ces formulaires aient inclus dans in the document page count, their la pagination, il n'y aura aucun contenu removal does not represent any loss manquant. of content from the thesis.
1+1 Canada Acknowledgements
I am very grateful for the opportunity given to me by my supervisor, Roberto Quinlan, to reach my highest potential. He allowed me to write my own proposal, research questions, and funding applications to be able to conduct this research. I was fortunate to have a supervisor who believed in me and basically allowed me to do anything I wanted. I am also thankful for all of the assistance from Dr. Sarah Finkelstein (University of Toronto), whose input into manuscripts allowed for the successful completion of this dissertation.
Likewise, Charlotte Friel and Jane Devlin whom provided additional samples, feedback, and contributions.
Dr. Derek Muir, Xiaowa Wang, and colleagues at the NLET water quality laboratory provided all of the analysis of water chemistry, to which was instrumental in the completion of this project. I am also grateful for the assistance and support provided by Mary Ellen Thomas, Jamal Shirley, Dorothy Tootoo and the staff at the Nunavut
Research Institute. Likewise, field assistance by Andy Aliyak, Raymond Biastoch,
Andrew Dunford, Milissa Elliott, and Christopher Luszczek was appreciated.
This project was funded by a NSERC Discovery Grant and NSERC Northern
Research Supplement held by RQ, a NSERC Northern Research Internship, NSTP, and additional York University funding for graduate student research. This project had the following approvals: NRI #0501008N, 0300707N, 0301206N, NIRB 08YN023,
07YN035, DFO S-09/10-1011-NU, S-08/09-1002-NU, S-07/08-1015-NU, S-06/07-1023-
NU, and a Nunavut Department of Environment Wildlife Research Licence WL000839.
iv Contributions
Chapter 2: Patterns in the limnology of lakes and ponds across multiple local and regional environmental gradients in the eastern Canadian Arctic.
• Manuscript submitted to Inland Waters with authorship: Medeiros, A.S., Biastoch, R.G., Luszczek, C.E., Wang, X.A., Muir, D.G.C. & Quinlan, R. • Study design by ASM and Roberto Quinlan (RQ), limnological data collection by ASM, Raymond G. Biastoch (RGB) and Chris E. Luszczek (CEL), data analysis primarily by ASM, with contributions from RGB (geological data), and CEL (inter regional analysis), manuscript writing by ASM with editorial contributions by RQ. Water chemistry analysis was conducted by Xiaowa A. Wang, and Derek G.C. Muir.
Chapter 3: The distribution of the Chironomidae (Insecta: Diptera) along multiple environmental gradients in lakes and ponds of the eastern Canadian Arctic.
• Manuscript accepted for publication in the Canadian Journal for Fisheries and Aquatic Sciences with authorship: Medeiros, A.S., & Quinlan, R. • Study design by ASM and RQ, data analysis by ASM, manuscript writing by ASM with editorial contributions from RQ.
Chapter 4: A high resolution multi-proxy record of pronounced 20th century environmental change at Baker Lake, Nunavut
• Manuscript submitted to the Journal of Paleolimnology with authorship: Medeiros, A.S., Friel, C.E., Finkelstein, S. & Quinlan, R. • Study design and chironomid data collection and analysis by ASM, diatom data collection and analysis by Charlotte E. Friel (CEF) and Sarah Finkelstein (SF) (University of Toronto), manuscript writing primarily by ASM with diatom-related contributions by CEF and SF, and editorial contributions from RQ.
Appendix 1: Benthic biomonitoring in Arctic tundra streams; a community based approach in Iqaluit, Nunavut, Canada
• Manuscript published in Arctic as follows: Medeiros, A.S., Luszczek, C.E., Shirley, J. & Quinlan, R. 2011. Benthic biomonitoring in Arctic tundra streams; a community based approach in Iqaluit, Nunavut, Canada. Arctic, 64(1): 59-72. • Study design by ASM, data collection by ASM, Jamal Shirley (JS) and CEL, data analysis primarily by ASM with contributions by CEL, manuscript writing by ASM with editorial contributions from JS, CEL and RQ.
v Abstract
An examination of the Chironomidae was conducted to determine the environmental gradients that may influence their distribution in the eastern Canadian
Arctic. Subfossil chironomid head capsules, comprising 86 taxa, were sampled from surficial sediments of lakes and ponds that spanned from the tree-line in northern
Manitoba across multiple regions of the territory of Nunavut, Arctic Canada. In addition, relationships between the limnology of lakes and ponds across these regions were examined. Pond systems ( < 2 m depth) were found to have a higher variability in temperature, nutrients, and major ions when compared to lake systems. There was also a significant difference between regions based on their limnology, however, overall relationships between environmental variables were similar in lakes and ponds across all regions.
The water chemistry and environmental data were then compared to assemblages of taxa using multivariate analysis. The focus on both local and regional environmental gradients by sampling several systems within and between several regions allowed for the elucidation of a primary gradient that was represented by both temperature and total nitrogen. While temperature and trophic status were found to strongly influence the distribution of some taxa (e.g., Cladotanytarsus mancus-gr), partially constrained gradient analysis indicated that specific chironomid taxa could be used to indicate a primary response to climate regardless of trophic status. This allowed for a robust surface water paleo-temperature transfer function to be generated for application to sediment cores.
A high-resolution examination of the subfossil remains of the Chironomidae was then conducted on a sediment core from Baker Lake, a large, deep Arctic lake in Canada.
The core was sectioned at 0.5 cm resolution and Pb dating was used to establish a chronology. A downcore analysis of over 60 taxa indicated a pronounced gradual decline of several cold-water indicator taxa beginning at approximately 1940 AD and reaching
0% relative abundance at approximately 1990 AD. Several taxa indicative of warmer conditions first appear in sediments beginning in the 1940s and increase in abundance in more recent sediments. In addition, the arrival of Cladotanytarsus mancus gr., a warm- water adapted taxa indicative of higher nitrogen concentrations, in recent sediments (circa
1985 AD), increased to 12% of the total chironomid community by the surface sediment interval. The biostratigraphic results from Baker Lake were applied to the mid-summer surface water temperature inference model, and inference results indicated a 3°C increase in water temperature over the last 60 years. This also corresponded strongly to the instrumental record available since 1950. Thus, the gradual decline of cold-water adapted taxa, and subsequent increase of several taxa indicative of warmer regions, is a strong signal of recent environmental change within the area of Baker Lake, Nunavut.
vn Table of Contents COPYRIGHT PAGE II CERTIFICATE PAGE Ill ACKNOWLEDGEMENTS IV CONTRIBUTIONS V ABSTRACT VI TABLE OF CONTENTS VIII LIST OF FIGURES XI LIST OF TABLES XIII LIST OF ABBREVIATIONS XIV CHAPTER 1: INTRODUCTION AND LITERATURE REVIEW 1
THESIS OBJECTIVES 8 CHAPTER 2: PATTERNS IN THE LIMNOLOGY OF LAKES AND PONDS ACROSS MULTIPLE LOCAL AND REGIONAL ENVIRONMENTAL GRADIENTS IN THE EASTERN CANADIAN ARCTIC 18
ABSTRACT 19 INTRODUCTION 20 METHODS 22 Field sampling 24 Data Screening 25 Numerical Analyses 26 RESULTS 27 pH and conductivity 28 Nutrients and Productivity 29 Metals 30 Multivariate Analyses 31 DISCUSSION 34 Physical characteristics 36 Local geological influences 37 Temperature, nutrients, and productivity 38 Metals 43 CONCLUSION 44 ACKNOWLEDGEMENTS 46 REFERENCES 46 TABLES 51 FIGURES 57 CHAPTER 3: THE DISTRIBUTION OF THE CHIRONOMH)AE (INSECTA: DIPTERA) ALONG MULTIPLE ENVIRONMENTAL GRADIENTS IN LAKES AND PONDS OF THE EASTERN CANADIAN ARCTIC 63
ABSTRACT 64 INTRODUCTION 65 MATERIALS AND METHODS 69 Study Areas 69
viii Laboratory Analysis 72 Data Screening 74 Numerical Analyses 75 RESULTS 77 Ordination Analyses 78 Environmental Distributions 79 DISCUSSION 81 Influences of temperature and depth on Chironomidae 81 Nutrients and productivity 85 Regionality 88 Future community change 91 ACKNOWLEDGEMENTS 93 REFERENCES 94 TABLES 99 FIGURES 102 CHAPTER 4: A HIGH RESOLUTION MULTI-PROXY RECORD OF PRONOUNCED 20™ CENTURY ENVIRONMENTAL CHANGE AT BAKER LAKE, NUNAVUT 114
ABSTRACT 115 INTRODUCTION 117 Study Area 120 METHODS 121 Field Methods 121 Laboratory Methods 122 RESULTS 125 Chironomid model development 125 Core stratigraphy and chronology 727 Chironomid assemblages 727 Fossil diatom assemblages 729 Chironomid-based quantitative water temperature reconstruction 131 DISCUSSION 133 CONCLUSIONS 137 ACKNOWLEDGEMENTS 139 REFERENCES 139 TABLES 145 FIGURES 146 CHAPTER 5: CONCLUSIONS 155
REFERENCES 160 APPENDIX 1: BENTHIC BIOMONITORING IN ARCTIC TUNDRA STREAMS; A COMMUNITY BASED APPROACH IN IQALUIT, NUNAVUT, CANADA 161
ABSTRACT 162 INTRODUCTION 163 STUDY AREA 167 Airport Creek 168 Apex River. 769 METHODS 170 Benthic Invertebrate Sampling 770 Laboratory methods 777 Statistical Analysis 772 Multivariate Analysis 773 RESULTS 174
IX Taxonomic Resolution. 175 Assessment Metrics 777 Multivariate Analysis 778 DISCUSSION 181 Biological Response 787 Taxonomic Sufficiency and Simple Summary Indices 183 Multivariate Analysis 184 CONCLUSIONS AND RECOMMENDATIONS 185 ACKNOWLEDGEMENTS 187 REFERENCES 188 TABLES 191 FIGURES 196 APPENDIX 2: RAW WATER CHEMISTRY AND ENVIRONMENTAL DATA 207 APPENDIX 3: RAW COUNTS OF CHIRONOMH) HEAD CAPSULES IN SURFACE SEDIMENTS 214 APPENDIX 4: RAW COUNTS OF CHIRONOMID HEAD CAPSULES IN A SEDIMENT CORE FROM BAKER LAKE, NUNVUT 230
x List of Figures
Chapter 2:
Fig. 1: Location of sampling areas identified by region.
Fig. 2: Trends in selected environmental variables across identified regions.
Fig. 3: Principal Components Analysis of significant environmental indicators (excluding metals).
Fig. 4: Principal Components Analysis of significant environmental indicators (including metals).
Fig. 5: Canonical variate analysis of significant environmental indicators (including metals).
Chapter 3:
Fig. 1: Locations of all recent (>1980 AD) published examinations of subfossil Chironomidae assemblages within the territory of Nunavut, Canada.
Fig. 2: Diagnostic images for Zalutschia types found.
Fig. 3: Relative Abundances of selected taxa recovered from surface samples of Nunavut lakes and ponds.
Fig. 4: The classification of the 63 sampling locations across the eastern Canadian Arctic with the use of two-way indicator species analysis.
Fig. 5: Box-plots of selected environmental variables by TWINSPAN group.
Fig. 6: RDA of sampled locations and significant environmental indicators.
Fig. 7: RDA of species and significant environmental indicators.
Fig. 8: Species arrows in a singly constrained partial RDA of (a) MSSWT with TN as a covariable, and (b) TN with MSSWT as a covariable.
Fig. 9: Abundances of environmental indicator taxa plotted on the RDA with significant environmental indicators indicated.
xi Chapter 4:
Fig. 1: Location of Baker Lake coring site, and surface calibration-set sampling locations for sub-fossil Chironomidae across the eastern Canadian Arctic.
Fig. 2: The second component WA-PLS inference model.
Fig. 3: Baker Lake biostratigraphy of chironomid assemblages plotted as percent abundances.
Fig. 4: Diatom biostratrigraphy for the Baker Lake sediment core.
Fig. 5: Summary diagram of the Baker Lake stratigraphy.
Fig. 6: Principle correspondence analysis of the Chironomidae surface calibration set with the Baker Lake core intervals run passively.
Fig. 7: A comparison of the Baker Lake summary biostratigraphic results to the recorded mean annual temperature anomaly.
xn List of Tables
Chapter 2:
Table 1: Summary of environmental and limnological variables.
Table 2: Pearson correlation matrix of selected environmental variables.
Table 3: Regress sion matrix of selected significant environmental variables
Table 4: PCA scores for environmental and limnological variables.
Table 5: CVA regression coefficients, associated f-values, and Interset correlations for significantly different limnological variables for the 97 sampling locations defined by region.
Chapter 3:
Table 1: Descriptive statistics of environmental variables for the 63 sampled lakes and ponds.
Table 2: Eigenvalues of first four redundancy analysis (RDA) axes, species environment correlation for each canonical axis, and cumulative % variance of species data and species.
Table 3: Regression coefficients of the first two RDA axes (AXl and AX2), their associated Mest values, and interset correlations for backwards-selected environmental variables.
Chapter 4:
Table 1: Performance of each model type relating to MSWWT and chironomid variance.
Xlll List of Abbreviations
CA Correspondance analysis CCA Canonical correspondence analysis CONISS Constrained cluster analysis with incremental sum of squares partitioning CRS Constant rate of supply DCA Detrended correspondence analysis DCCA Detrended canonical correspondence analysis DO Dissolved oxygen DOC Dissolved organic carbon DIC Dissolved inorganic carbon LOI Loss on ignition MAT Modern anologue technique MSSWT Mid-summer surface water temperature ORP Oxygen reduction potential PCA Principle components analysis PLS Partial least squares regression RDA Redundancy analysis RMA Reduced major axis regression RMAT Recorded mean annual temperature RMSE(P) Root mean square error (of prediction) SA Surface area TP Total phosphorus TN Total nitrogen TWINSPAN Two way indicator species analysis WA Weighted averaging regression WA-PLS Weighted averaging partial least squares regression Zmax Maximum depth
XIV CHAPTER 1: Introduction and Literature Review
Arctic ecosystems are defined by large seasonal transitions that control many thermodynamic relationships defined by snow, ice cover, and solar radiation (Vincent et al. 2008). Most Arctic lakes typical of the eastern Canadian Arctic are nutrient-limited, alkaline, and clear (Douglas et al. 2000), and the majority of organic and nutrient inputs occur due to surface runoff and overland flow during the spring snow-melt (Young and
Woo 2006). This seasonal difference in inputs can result in a difference in the nutrient availability of aquatic systems during the drier summer months (Young and Woo, 2006).
The transition from the terrestrial to aquatic environment is a key component of allochthonous inputs of organic carbon and nutrients that flow into isolated lakes and ponds, which is strongly controlled by the physical characteristics and configuration of the drainage basin and varying amounts of catchment area vegetation (Wetzel 1983).
Physical factors, such as the underlying geomorphology, permafrost, topography, amount of snow-melt, and drainage basins can highly influence the limnology of aquatic systems in the Arctic (Hamilton et al. 2001).
While the limnology and biological communities of aquatic systems may respond to differences at both a local (physical factors) and regional (environmental factors) scale, several studies have focused on latitudinal based transect sampling to maximize gradients in temperature (e.g., Barley et al. 2006, Porinchu et al. 2009a). Indeed, the mechanisms behind the influence of climate change on Arctic systems are not yet fully understood, and are only beginning to be observed at the highest scales of organization (e.g. Inuit
1 accounts of delayed freezing and early breakup of ice, catastrophic lake disappearance, reduction of megafauna populations). While the rate and extent of climate change at high-levels of organization are difficult to predict (e.g. migration of temperate species, loss of native biodiversity), and may not be observable on small temporal scales, changes in the hydrological, biogeochemical, and biological regimes of Arctic systems may be realized and quantifiable over shorter periods. For example, the increased weathering of newly unfrozen glacial till has been observed in limnological trends from Lake Toolik
(Alaska), where summer alkalinity has doubled since 1975 (Hinzman et al. 2005). Other
Arctic lakes and streams have also shown increases in alkalinity, balanced primarily by increases in calcium and magnesium (Hinzman et al. 2005).
The shifts predicted in the geochemical processes due to increased temperatures
(Rouse et al. 1997), and increases in the hydrological influence of the active layer due to permafrost melt (Edwardson et al. 2003) have wide-scale implications for the overall productivity of Arctic systems (Michelutti et al. 2005). Phytoplankton are limited by low photosynthetic rates in Arctic habitats versus temperate systems due to temperature and light availability (Flanagan et al. 2003). If the length of the ice-free period increases, an increase in the available light and subsequent growing season for both terrestrial and aquatic systems is also expected (Douglas and Smol 1999). An increase in available light would therefore allow for an increase in total annual primary production and more complex periphytic communities to develop in freshwater systems (Rouse et al. 1997,
Michelutti et al. 2005). Arctic lakes and ponds are particularly sensitive to long term changes in climate, and of particular interest is identifying how changes in physical and
2 chemical properties of freshwater bodies, due to warmer climate, promote changes in ecological interactions across trophic levels.
Of the aquatic insects that occur in the Canadian Arctic, species of Chironomidae within the Order Diptera dominate in number of species and sheer abundance (Oliver
1971). In some areas just north of the treeline, members of other orders may be common, such as the mayflies (Ephemeroptera), caddisflies (Trichoptera), stoneflies (Plecoptera), beetles (Coleoptera), and dagger flies (Empididae), however, the abundance and diversity of these groups is reduced northward as the Diptera become even more dominant (Oliver
1971, Danks 1992). Chironomids are distributed globally, and make up between one- fifth and one-half of the total number of insect species in the Arctic (Oliver 1971). As environmental conditions become more severe, the Chironomidae increase in proportion to other families, and in the most severe conditions species of the sub-family
Orthocladiinae dominate (Oliver 1968, Danks 1992).
Brundin (1966) hypothesized that the Chironomidae originated in fast montane streams, were likely cold-adapted to survive periodic freezing, and spread to slower moving and warmer systems over time. Since individual species are able to tolerate large gradients of environmental conditions, such as temperature (Walker et al. 1991), oxygen
(Quinlan et al. 2001), productivity (Broderson and Lindegaard 1999), and salinity
(Henrichs et al. 2001), the biotopes they occupy are exceptionally wide-ranging
(Porinchu and MacDonald 2003). Indeed, many species are specially adapted for specific environments (Pinder 1995) with narrow ecological thresholds (Smol et al. 2005), thus,
3 the composition of the chironomid community in a particular habitat highly depends on both the terrestrial and aquatic environment.
Arctic lakes are traditionally dominated by various cold-water adapted species
(Walker 1990), of which the widest distributed is the subfamily Orthocladiinae (Oliver
1971). Unlike the Tanypodinae and Chironominae, their numbers decrease in warmer habitats (Oliver 1971). Conversely, the northern range of warm-water adapted species, which are more reflective of temperate regions (e.g., those of the Tribe Chironomini), may be strictly limited by temperature (Walker et al. 1991) and are rare or absent in many
Arctic habitats (Quinlan et al. 2005). Gajewski et al. (2005) found that lakes in northern and southern parts of the Canadian Arctic Archipelago had similar assemblages compared to central islands. The lowest diversity occurred in the Devon and Cornwallis
Islands, where temperatures were low and nutrients were limited (Gajewski et al. 2005).
In addition, Gajewski et al. (2005) found an absence of species from the Tribe
Chironomini across a broad swath of the Arctic Archipelago, while Walker and
MacDonald (1995) identified major shifts in Chironomidae assemblages north of the treeline, including the presence of several Chironomini species in the warmer areas of the western Canadian Arctic. This indicates that the survival of these species is not influenced by a loss of watershed vegetation, but likely temperature dependence (Walker etal. 1991).
While there are many species that are highly correlated to specific environmental conditions, some genera have adapted to multiple stressors across a variety of environmental gradients. For example, the larval development of Chironomus athracinus
4 in Alaskan ponds can be delayed by both temperature (in the winter) and low oxygen conditions in the summer (Butler 1982). Halpern et al. (2002) also found in laboratory experiments that the larval tube of some Chironomus species significantly protected them against chemical toxins than those without. Indeed, Butler (1982) found that the life- cycle of two tube-building Chironomus species was 7 years long in Alaskan ponds, even though sediments were frozen nine months of the year. This ability to adapt and survive under severe environmental stressors has allowed the Chironomidae to be indicators of a wide range of environmental conditions, including differences in their tolerance to anthropogenic pollution (Clements et al. 2000, Mousavi et al. 2003).
Chironomids in Arctic aquatic systems are especially sensitive to perturbations in environmental and landscape conditions (Walker 2001) and are often used as indicators of environmental conditions by looking at their diversity and abundance within a particular system or area. Qualitative relationships between biological communities and their environment have been used as a basis for comparing current communities to historic conditions. Walker (1990) collected surface sediments across multiple biogeographic regions to compare assemblages of the Chironomidae to the varying environmental conditions. The presence, absence, and overall abundance of a particular species were found to be indicative of specific environmental conditions. For example, genera of Heterotrissocladius, Protanypus, and Sergentia are common in the cold profundal regions of stratified lakes versus most genera of Chironomini,
Pseudochironomus, and Pentaneurini, which are common in temperate systems (Walker
1990). In addition, certain species have been shown to be reflective of multiple
5 environmental gradients (Broderson and Anderson 2002). Surveys across multiple spatial scales provide a basis for interpretation of the qualitative ecology of biological indicators and are used to define ecological reference conditions (Battarbee 1999).
The use of aquatic organisms that respond to direct (e.g. temperature and light availability) and indirect (nutrient and ionic concentrations) climate dependent variables allows for an inference of the environmental conditions that may have accounted for their abundances in the sediment profiles of lakes and ponds. Biostratigraphic inferences of environmental conditions have been successfully constructed using stable isotopes (Sauer et al. 2001), chemical indicators (e.g. pollutants, Fe/Mn profiles), amoebae (Ellison
1995), cladoceran fossils (Sweetman et al. 2008), chironomid head capsules (Walker
1991), diatoms (Smol et al. 2005), algal pigments (Michelutti et al. 2005), varve thickness (Lamoureaux 1999), pollen (Kerwin et al. 2004), ostracod shells (Porter et al.
1999), and various microfossils (Wolfe and Perren, 2001). While many indicators can be used as proxies of past environmental conditions and change, sensitivity to recent climate warming varies. For example, reconstructions by Kerwin et al. (2004) based on lacustrine pollen records indicate that regional warming (Baffin Island) and cooling
(Northern Quebec and Labrador) events have occurred since the Mid-Holocene, however, the vegetational response to climatic conditions is inherently delayed and therefore not sensitive to smaller temporal scales.
As chironomids are abundant in the sediments of freshwater systems, their fossilized remains can be used to infer the community structure. The examination of the chitinized head-capsules of chironomids has been used to infer several environmental
6 conditions, including climate (Walker et al. 1991), dissolved oxygen (Quinlan et al.
2001), trophic status (Brooks et al. 2001), and productivity (Broderson and Lindegaard
1999). The identification of chironomids in the sediments of lakes relies primarily on the morphology of the chitinous head capsule produced by each instar of the aquatic chironomid larvae. Spatial surveys of surface sediments across multiple bioregions in
Canada (Walker 1990, Gajewski et al. 2005) allow for the identification of the environmental conditions that currently account for the distribution of chironomids.
Walker et al. (1991) developed an inference model temperature reconstructions estimation from chironomid subfossil assemblages along a transect in Northern Labrador.
Their model predicted summer surface-water temperatures that were strongly correlated with recorded temperature (Walker et al. 1991), and subsequent refinements of this inference model (Walker et al. 1997) and studies where there was high agreement between chironomid-inferred climate changes and inferences from other climate proxies
(e.g. Levesque et al. 1994) strengthened the validity of using subfossil chironomids to quantitatively infer past changes in climate. In addition, Walker and MacDonald (1995) found that certain chironomid taxa were significantly influenced by both depth and distance from treeline in Western Canadian Arctic lakes. These relationships provided the basis for the application of transfer functions to reconstruct past environmental conditions from sediment core analysis of chironomid populations. Several additional inference models have since been developed using sub-fossil chironomids and applied to sediment cores across multiple regions in the eastern Canadian Arctic (e.g. Francis et al.
2006; Briner et al. 2006; Thomas et al. 2008; Porinchu et al. 2009b).
7 Thesis objectives
The lack of available instrumental data and long term monitoring of Arctic aquatic systems makes inferences about the influence of environmental change difficult.
Therefore, generating predictions of ecological change is difficult in areas of the Arctic that have had little, if any, previous sampling and research, such as the Kivalliq region of
Nunavut. Region-specific characteristics of climate, geology, etc., may result in limnological characteristics, and relationships amongst limnological variables, that differ compared to other Arctic regions. Additionally, the sampling regime of current limnological studies tends to maximize specific climate related gradients in order to reconstruct a desired gradient rather than explore local differences that may arise from physical characteristics of systems. Therefore, fundamental questions arise whether transect sampling approaches miss limnological and physical gradients that may only be apparent at a local scale. Chapter 2 of my thesis attempts to address these questions based on a sampling regime that attempts to capture larger local gradients in physical and environmental attributes (e.g., depth, surface area, local catchment inputs). An exploratory review of the limnology of lakes and ponds from under-sampled areas was conducted with additional analysis to address a hypothesis that physical and environmental factors would be apparent at both local and regional scales and influence lakes and ponds in a similar fashion. This was carried out by clustering samples within local areas, and then comparing relationships found within a particular area across a regional scale. In addition, the examination of the limnology of both lakes and ponds
8 independently attempts to elucidate relationships that may be dependant on the water depth of a system.
Chapter 3 of my thesis examines the distribution and abundance of chironomids in the Kivalliq region of Nunavut, which represents a large area of the eastern Canadian
Arctic that has few samples. Surveys across multiple spatial scales provide a basis for interpretation of the qualitative ecology of biological indicators and are used to define ecological reference conditions (Battarbee 1999). The examination of the Chironomidae at both local and regional scales is used to test the hypothesis that chironomids may respond to localized environmental gradients as well as regional gradients in climate.
The examination of local gradients as well as regional climatic patterns is hypothesized to show changes in species richness and diversity in invertebrate community structure as sampling is conducted across gradients in temperature and trophic condition.
The use of the subfossil remains of chironomids in a paleo-perspective provides researchers with methods to interpret historic changes in the environment, and put them into perspective with modern instrumental records. The long-term history of climatically sensitive lakes and ponds allows for an understanding of the mechanisms behind climate- induced changes in biological communities. Baker Lake, Nunavut, is one of the largest and most prominent lakes in the eastern Canadian Arctic. While the spatiotemporal influence of climate warming on Arctic systems has been quantified by paleo- reconstructions of algal and invertebrate communities for large temporal scales (e.g.,
Holocene- or Quaternary-scale), Chapter 4 of my thesis examines a high-resolution paleo-climate reconstruction of a large, deep lake [Baker Lake], which may have a
9 stronger 'climate memory' that is less sensitive to inter-annual variation in temperature compared to shallower systems (Abrosetti and Barbatini 1999). It is expected that a biostratigraphic examination of chironomids from Baker Lake would indicate a recent
shift in community structure in a warming climate from cold-water adapted species to
warm-water adapted species that are more characteristic of southern temperate systems.
Finally, Chapter 5 outlines the key findings of the thesis as a whole, as well as potential avenues for future investigations.
References
Abrosetti, W. and Barbatini, L. 1999. Deep water warming in lakes: an indicator of
climate change. Journal of Limnology, 58:1-9.
Battarbee, R.W. 1999. The importance of paleolimnology to lake restoration.
Hydrobiologia, 395:149-159.
Barley, E.M., Walker, I.R., Kurek, J., Cwynar, L.C., Mathewes, R.W., Gajewski, K., and
Finney, B.P. 2006. A northwest North American training set: distribution of
freshwater midges in relation to air temperature and lake depth. Journal of
Paleolimnology, 36:295-314.
Briner, J.P., Michelutti, N., Francis, D.R., Miller, G.H., Axford, Y., Wooller, M.J.,
Wolfe, A.P. 2006. A multi-proxy lacustrine record of Holocene climate change on
northeastern Baffin Island, Arctic Canada. Quaternary Research, 65:431-44.
Broderson, K.P. and Lindegaard, C. 1999. Classification, assessment and trophic
reconstruction of Danish lakes using chironomids. Freshwater Biology, 42:143-
157.
10 Broderson, K.P. and Anderson, J.N. 2002. Distribution of chironomids (Diptera) in low
arctic West Greenland lakes: trophic conditions, temperature and environmental
reconstruction. Freshwater Biology, 47:1137-1157.
Brooks, S.J., Bennion, H., and Birks, J.P. 2001. Tracing lake trophic history with a
chironomid total phosphorus inference model. Freshwater Biol. 46:511-532.
Brundin, L., 1966. Trans antarctic relationships and their significance, as evidenced by
chironomid midges with a monograph of the subfamilies Podonominae and
Aphroteniinae and the austral Heptagyiae. Kunglica Svenska
Vetenskapsakademiens Handlungar, W:\-\12 + plates.
Butler, M. G. 1982. A 7-year life-cycle for 2 Chironomus species in Arctic Alaskan
tundra ponds (Diptera, Chironomidae). Canadian Journal of Zoology, 60:58-70.
Clements, W.H., Carlisle, D.M., Lazorchak, J.M., and Johnson, P.C. 2000. Heavy metals
structure benthic communities in Colorado mountain streams. Ecological
Applications, 10:626-638.
Danks, H.V. 1992. Arctic insects as indicators of environmental change. Arctic, 45:159-
166.
Douglas, M.S.V. and Smol, J.P. 1999. Freshwater diatoms as indicators of environmental
change in the High Arctic, in The Diatoms: Applications for the environment and
earth sciences, edited by E.F. Stoermer and J.P. Smol, pp. 227-244. Cambridge
Univ. Press, New York.
11 Douglas, M.S.V., Smol, J.P., and Blake, W. Jr. 2000. Summary of paleolimnological
investigations of high Arctic ponds at Cape Herschel, east-central Ellesmere
Island, Nunavut. Bull. Geol. Surv. Can. No. 529. pp.257-269.
Edwardson, K.J., Bowden, W.B., Dahm, C, and Morrice, J. 2003. The hydraulic
characteristics and geochemistry of hyporheic and parafluvial zones in Arctic
tundra streams, north slope, Alaska. Advances in Water Research, 26:907-923.
Ellison, R.L. 1993. Paleolimnological analysis of Ullswater using testate amoebae.
Journal of Paleolimnology, 13:51-63.
Flanagan, K.M., McCauley, E., Wrona, F., and Prowse, T. 2003. Climate change: the
potential for latitudinal effects on algal biomass in aquatic ecosystems. Canadian
Journal of Aquatic Science, 60:635-639.
Francis, D., Wolfe, A., Walker, I.R., and Miller, G.H. 2006. Interglacial and Holocene
temperature reconstructions based on midge remains in sediments of two lakes
from Baffin Island, Nunavut, Arctic Canada. Palaeogeography,
Palaeoclimatology, Palaeoecology, 236:107-124.
Gajewski, K., G. Bouchard, S.E. Wilson, J. Kurek and L.C. Cwynar. 2005. Distribution
of chironomidae (Insecta: Diptera) head capsules in recent sediments of Canadian
Arctic lakes. Hydrobiologia, 549:131-143.
Halpern, M., Gasith, A., and Broza, M. 2002. Does the tube of a benthic chironomid larva
play a role in protecting its dweller against chemical toxicants? Hydrobiologia,
470:49-55.
12 Hamilton, P.B., Gajewski, K., Atkinson, D.E. & D.R.S. Lean, 2001. Physical and
chemical limnology of 204 lakes from the Canadian arctic archipelago.
Hydrobiologia; 457:133-148.
Henrichs, M.L., Walker, LA., Mathews, R.W. 2001. Chironomid-based paleosahnity
records in southern British Columbia, Canada: a comparison of transfer functions.
Journal of Paleolimnology, 26:147-159.
Hinzman, L.D., Bettez, N., Bolton, W., and 32 others. Evidence and implications of
recent climate change in northern Alaska and other Arctic regions. Climatic
Change, 72:251-298.
Kerwin, M.W., Overpeck, J.T., Webb, R.S. and Anderson, K.H. 2004. Pollen-based
summer temperature reconstructions for the eastern Canadian boreal forest,
subarctic, and Arctic. Quaternary Science Reviews, 23:1901-1924.
Lamoureaux, S. 1999. Spatial and interannual variations in sedimentation patterns
recorded in nonglacial varved sediments from the Canadian High Arctic. Journal
of Paleolimnology, 21:73-84.
Michelutti, N., Wolfe, A.P., Vinebrooke, R.D., Rivard, B., and Briner, J.P. 2005. Recent
primary production in Arctic lakes. Geophysical Research Letters, 32:L19715.
Mousavi, S.K. Primicerio, R., and Amundsen, P-A. 2003. Diversity and structure of
Chironomidae (Diptera) communities along a gradient of heavy metal
contamination in a subarctic watercourse. The Science of the Total Environment,
307:93-110.
13 Oliver, D. R. 1968. Adaptations of Arctic chironomidae. Paper 33 in the program
"Studies on Arctic Insects", Entomology Research Institute, Canada Department
of Agriculture, in collaboration with the Defence Research Board of Canada.
pp.111-118.
Oliver, D.R. 1971. Life History of the Chironomidae. Annual Review of Entomology;
16:211-230.
Pinder, L.C.V. 1995. The habitats of Chironomidae larvae. In: Armitage, P.D., Cranston,
P.S., and Pinder, L.C.V (eds). The Chironomidae: Biology and ecology of non-
biting midges. Chapman and Hall, London, pp.107-133.
Porinchu D. F. and G. M. MacDonald. 2003. The use and application of freshwater
midges (Chironomidae: Insecta: Diptera) in geographical research. Progress in
Physical Geography, 27:378-422.
Porinchu, D., RoUand, N., and Moser, K. 2009a. Development of a chironomid-based air
temperature inference model for the central Canadian Arctic. Journal of
Paleolimnology, 41: 349-368.
Porinchu D.F., MacDonald G.M., RoUand N. 2009b. A 2000 year midge-based
paleotemperature reconstruction from the Canadian Arctic archipelago. Journal of
Paleolimnology, 41:177-188.
Porter, S.C., Sauchyn, D.J., Delorme, D.L. 1999. The ostracode record from Harris Lake,
southwestern Saskatchewan: 9200 years of local environmental change. Journal of
Paleolimnology, 21:35-44.
14 Quinlan, R., and Smol, J.P. 2001. Chironomid-based inference models for estimating
end-of-summer hypolimnetic oxygen from south-central Ontario shield lakes.
Freshwater Biology, 46:1529-1551.
Quinlan, R., Douglas, M.S.V., and Smol, J.P. 2005. Food web changes in Arctic
ecosystems related to climate warming. Global Change Biology, 11:1381-1386.
Rouse, W., Douglas, M., Hecky, R., Kling, G., Lesack, L., Marsh, P., McDonald, M.,
Nicholson, B., Roulet, N. & J.P. Smol, 1997. Effects of climate change on the
freshwaters of Arctic and sub Arctic North America. Hydrological Processes,
11:873-902.
Sauer, P.E., Miller, G.H., Overpeck, J.T. 2001. Oxygen isotope ratios of organic matter in
arctic lakes as a paleoclimate proxy: field and laboratory investigations Journal of
Paleolimnology, 25:43-64.
Smol. J.P, Wolfe, A.P., Birks, J.B., Douglas, M.S., Jones, V.J., KorholaA., Pienitz R.,
Ruhland, K., Sorvari, S., Antoniades, D., Brooks, S.J., Fallu, M., Hughes,M.,
Keatley, B.E., Laing, T.E. Michelutti, N., Nazarova, L., Nyman, M., Paterson,
A.M., Perren, B., Quinlan, R., Rautio, M., Saulnier-Talbot, E., Siitonen, S.,
Solovieva, N., Weckstromi, J. 2005. Climate-driven regime shifts in the biological
communities of arctic lakes. PNAS, 102:4397-4402.
Sweetman, J.N., LaFace, E., Ruhland, K.M., and Smol, J.P. 2008. Evaluating the
Response of Cladocera to Recent Environmental Changes in Lakes from the
Central Canadian Arctic Treeline Region. Arctic, Antarctic, and Alpine Research,
40:584-591.
15 Thomas, E.K., Axford, Y., and Briner, J.P. 2008. Rapid 20th century environmental
change on northeastern Baffin Island, Arctic Canada inferred from a multi-proxy
lacustrine record. Journal of Paleolimnology, 40:507-517.
Vincent, W.F. & J. Laybourn-Parry. 2008. Polar Lakes and Rivers: Limnology of Arctic
and Antarctic Aquatic Ecosystems. Oxford University Press, NY, USA.
Walker, I.R. 1990. Modern assemblages of arctic and alpine Chironomidae as analogues
for late-glacial communities. Hydrobiologia, 214:223-227.
Walker, I.R., Smol, J.P., and Engstrom, D.R. 1991. An assessment of Chironomidae as
quantitative indicators of past climatic change. Canadian Journal of Fisheries and
Aquatic Science, 48:975-987.
Walker, I.R., and MacDonald, G.M. 1995. Distributions of Chironomidae
(Insecta:Diptera) and Other Freshwater Midges with Respect to Treeline,
Northwest Territories, Canada. Arctic and Alpine Research, 27:258-263.
Walker I.R. 2001. Midges: Chironomidae and related Diptera. In: Smol J.P., Birks H.J.B.,
and Last W.M. (eds). Tracking environmental change using lake sediments, vol.
4. Zoological Indicators, Kluwer Academic Publishers, Dordrecht, the
Netherlands, pp 43-66.
Wetzel, R. G. 1983. Limnology. 2nd Edition. Saunders College Publ., Philadelphia.
860pp.
Wolfe, A.P. and Perren, B.B. 2001. Chrysophyte microfossils record marked responses to
recent environmental changes in high- and mid-arctic lakes. Canadian Journal of
Botany, 79:747-752.
16 Wolfe, A.P., Frechette, B., Richard, P.J.H., Miller, G.H., and Forman, S.L. 2000.
Paleoecology of a >90000-year lacustrine sequence from Fog Lake, Baffin Island,
Arctic Canada. Quaternary Science Reviews, 19:1677-1699.
Woo, M.K. and Young, K.L. 2006. High Arctic wetlands: their occurrence, hydrological
characteristics, and sustainability. Journal of Hydrology, 320:432-450.
17 CHAPTER 2: Patterns in the limnology of lakes and ponds across multiple local and regional environmental gradients in the eastern Canadian Arctic.
Medeiros, A. S.1, Biastoch, R. G.1, Luszczek, C.E. \ Wang, X.A.2, Muir, D.C.G.2,
& R. Quinlan1
1. York University, Department of Biology, 4700 Keele St, Toronto, Ontario, M3J1P3.
2. Aquatic Ecosystem Protection Research Division, Canada Centre for Inland Waters,
Environment Canada, 867 Lakeshore Rd., Burlington, Ontario L7R 4R6
Corresponding Author: [email protected]
Manuscript style for submission to Inland Waters
SHORT TITLE: Limnology of lakes and ponds in the eastern Canadian Arctic.
Keywords: Limnology, Arctic, lakes, ponds, water chemistry, biogeochemistry, Nunavut
18 Abstract
This study examined regional patterns in water chemistry from 57 lakes and 56 ponds (< 2 m deep) across the eastern Canadian Arctic territory of Nunavut. Water samples were collected from sites that spanned several ecoclimatic regions, and gradients in environmental and geochemical variables were compared with the use of multivariate analysis. In a principal components analysis (PCA) lakes and ponds cluster along a primary gradient of temperature, nutrients, and conductivity. In addition, there were significant differences in the regional variation between lake and pond samples for nutrients [total nitrogen (TN), total phosphorus (TP), particulate organic nitrogen (PON)], chlorophyll-a (CHLA), and dissolved major ions determined via Canonical Variates
Analysis (CVA). Across all regions TN:TP ratios were high, indicating phosphorus limitation, and mid-summer surface water temperature was strongly correlated to dissolved nitrogen concentrations. Differences were also tested between pond and lake systems; ponds had higher concentrations of ions, and subsequently had higher conductivity, especially within the western Hudson Bay region, where local geology likely plays a large role in the limnology of these systems. Likewise, the concentration of nutrients and ions in ponds were strongly correlated to concentrations of dissolved organic carbon (DOC), likely indicating the influence of watershed inputs and resuspended materials on the limnology of ponds. While there was higher regional variation in the limnology of pond systems than lakes, the general patterns within each region were similar.
19 Introduction
The eastern Canadian Arctic is defined by the marshes, streams, lakes, and ponds that dominate its landscape. An estimated 6% of the Arctic and sub-Arctic in North America is covered by freshwaters (Rouse et al. 1997). The responses of invertebrates to changes in limnology and environmental conditions are often used as indicators of environmental change. For example, Smol et al. (2005) suggested that shallow Arctic ponds could be the most sensitive aquatic system to climate warming depending on the amount, type, and timing of precipitation these systems receive. These types of predictions are difficult to assess in areas of the Arctic that have few, if any, direct observations.
The geochemistry of Arctic aquatic systems is also likely strongly influenced by connections between terrestrial catchments, surface sediments, and bedrock composition
(Hutchinson 1957, Wetzel 1983). Physical factors, such as topographic relief, depth, surface area, landscape position, and vegetation cover of catchments are important for hyporheic exchange, chemical weathering, and nutrient dynamics (Hamilton et al. 2001).
Westover et al. (2009) noted differences between bedrock and sediment composition as a component of the limnological transition identified between regions along a sampling transect of latitude. The physical configution of and composition of drainage basins and catchment areas are also known to strongly control allochthonous inputs (Rasmussen et al. 1989). Thus, the spatial scale of watersheds also highly influences geochemical condition due to varying catchment:water volume ratios (Rasmussen et al. 1989). For example, dissolved organic carbon (DOC) concentrations in Arctic systems are known to
20 be correlated to the amount of catchment vegetation present (Pienitz et al. 1997). In addition, Lim et al. (2001) found that shallow ponds on Bathurst Island had higher concentrations of DOC than deeper, and subsequently dilute, lakes with less vegetated catchments.
While there are several studies on the limnology of Arctic lakes and ponds in the literature (Pienitz et al. 1997, Gregory-Eaves et al. 2000, Hamilton et al. 2001, Lim et al.
2001, Michelutti et al. 2002, Ruhland et al. 2003, Lim et al. 2005, Westover et al. 2009), there remains several large areas of the Canadian Arctic that have little to no baseline data, and very few long-term monitoring sites exist for several vast regions. This is especially true for the Kivalliq region of Nunavut, where the lack of baseline limnological data makes it difficult to identify which environmental variables are most important in structuring the biotic composition when there are potentially large regional differences in physical, landscape, and geochemical conditions. In addition, the focus on maximizing temperature gradients in latitudinal transect-based sampling (e.g., Pienitz et al. 1997, Westover et al. 2009) may miss significant local relationships between watershed specific processes and the local environment.
This study addresses the lack of limnological sampling across several under- sampled areas of the eastern Canadian Arctic through an exploratory analysis of limnological patterns. Included in our analysis is an investigation of potential differences between ponds (< 2 m depth) and lakes (> 2 m depth). In order to investigate the influence of local environmental gradients from regional gradients, samples were gathered from a limnological survey of 57 lakes and 56 ponds across several areas in the
21 eastern Canadian Arctic. Our study sites cross several ecoclimatic regions from the tree- line (north-west of Churchill, Manitoba) in the south, northward along the western coast of Hudson Bay, central Nunavut, and across several areas in the eastern Baffin region.
We specifically sampled multiple systems within each area in an attempt to capture large local gradients that could have direct influence over the limnology of lakes and ponds and compared relationships found across multiple regions. Thus, this investigation both improves the spatial coverage of Arctic datasets as well examines limnological patterns across multiple scales.
Methods
Water samples were collected from 57 lakes and 56 ponds within multiple regions of varying landscape and environmental characteristics across the territory of Nunavut.
Sampling locations were conducted within similar geographic areas to maximize local gradients, and between areas that spanned from the tree-line area north-west of Churchill,
MB extending northward across the Kivalliq region to the north-eastern islands of the
Baffin region (Fig. 1). Each region was characterized by variations in landscape characteristics, from the relatively flat (mean elevation of 12 m a.s.l.) western Hudson
Bay region, to the relatively high relief areas of the Baffin region (mean elevation 100 m a.s.l.).
The soils of the eastern Canadian Arctic are poorly developed due to recent deglaciation and the presence of permafrost region-wide (Ruhland et al. 2003). The majority of surficial materials in most regions was characterized by a till blanket (thick
22 and continuous glacial deposits), or till veneer, which is characterized as silty, sandy, and clay glacial deposits, typically discontinuous, thin, and often with widespread rock outcrops. The surficial material surrounding lakes in the southern coastal western Hudson
Bay region varied between till blanket and coarse-grained glaciomarine deposits. The underlying geology also differs between regions, with the central region dominated by
Archean metamorphic granite or diorite underlying % of the study lakes, and Archean sedimentary clastic or carbonate rock and volcanic igneous rock underlying % of the study lakes (Stockwell et al. 1976, Hanmer et al. 2004). Most of the western Hudson Bay study lakes are underlain by Archean volcanic igneous rock in the Rankin Inlet area, and
Archean intrusive granitoid rock in the Arviat area. Archean metamorphic gneiss underlies over half of the southern treeline region study lakes (Wheeler et al. 1997), and till blanket surficial material surrounds the majority of the lakes (Fulton 1995). The
Baffin region is characterized by Proterozoic intrusive granitoid and Archean metamorphic gneiss bedrock (Wheeler et al. 1997), and the sediment composition varied between till blanket, fine-grained, and till veneer surficial materials (Fulton 1995).
Vegetation was primarily represented by numerous grass species (e.g. Poa arcticus, Festuca rubra, and Elymus spp.), matted compact cushions of prickly saxifraga
(Saxifraga tricuspidata), and moss campion (Silene acaulis). Catchments consisted of moss cushions of several grasses and tundra plants intermixed with clumps of dwarf fireweed (Chamerion latifolium). Large stands of Salix spp. shrubs, approximately 1.0 -
1.5 m in height, were also frequently present in catchment areas in southern regions.
Dominant shrubs were found throughout the vicinity of the southern portion of the
23 Hudson Bay region, and tree-line region, which contrasted with other areas sampled in the central Kivalliq region and eastern Baffin region where the number and height of shrubs observed were less prominent.
Field sampling
Physicochemical variables were measured using a YSI-600QS multi-parameter probe, including water temperature, specific conductance at 25 ^C (COND), pH, and oxidation reduction potential (ORP). Depth was measured with a depth sounder and the GPS location of each mid-basin coring location was recorded. Epilimnetic water samples were collected at 0.5 m below the water surface in pre-cleaned, HCl acid-washed, polyethylene bottles and immediately treated in the field following the protocols outlined in the
Analytic Methods Manual of Environment Canada (Environment Canada 1994a). Water samples for each lake and pond sample were analyzed using standard operating procedures (SOP 02-2002) for major ions and nutrients (Environment Canada 1994a), as well as total and dissolved trace metals in water (Environment Canada 1994b) by in- bottle digestion and Inductively Coupled Plasma (ICP) Mass Spectrometry and ICP-
Optical Emission Spectrometry by the National Laboratory for Environmental Testing
(NLET) at the Canadian Centre for Inland Waters (CCIW), Burlington, Canada. In total,
41 variables were measured including: chlorophyll-a (CHLA) uncorrected for phaenophytin, dissolved organic carbon (DOC), dissolved inorganic carbon (DIC), particulate organic carbon (POC), particulate organic nitrogen (PON), calcium (Ca), magnesium (Mg), potassium (K), sodium (Na), chloride (CI), total nitrogen unfiltered
24 (TN), nitrate (N03), nitrite (N02), ammonia (NH3), sulphate (SO4), total phosphorus unfiltered (TP), silver (Ag), aluminum (Al), arsenic (As), boron (B), bismuth (Bi), cadmium (Cd), cesium (Cs), cobalt (Co), chromium (Cr), copper (Cu), iron (Fe), gallium
(Ga), lanthanum (La), lithium (Li), manganese (Mn), molybdenum (Mo), niobium (Nb), nickel (Ni), lead (Pb), platinum (Pt), rubidium (Rb), antimony (Sb), selenium (Se), tin
(Sn), strontium (Sr), titanium (Ti), uranium (U), vanadium (V), and zinc (Zn). Water samples were filtered through a GF/F 0.7 urn glass fibre filters for CHLA, and
PON/POC, wrapped in aluminum foil, frozen and stored until analysis.
Data Screening
Variables that had concentrations below the detection limit in a majority of sites sampled
(e.g. NO3, N02, SRP, Ag, Be, Bi, Cs, Li, Nb, Pt, Se, Sn, Tl, W) were deleted from the dataset. If any sample contained variables that were below the detection limit in less than half the study lakes, half the detection limit was substituted for use in numerical analysis.
Trace metals with extremely low concentrations (less than 0.05 ug L"1), and which were highly collinear (determined by unconstrained ordinations) with other environmental variables, were also removed from analysis (Cd, Cs, Co, Cr, Ga, La, Mo, Sb, Ti, and V).
In addition, trace metal variables, as well as POC/PON, were missing from 12 lakes and 4 ponds sampled in 2006 due to loss of samples during shipping. Environmental variables were transformed and checked for normality with a Kolmogorov-Smirnov test. Surface area, depth, COND, CI, Ca, S04, DOC, Mg, K, Na, Al, As, B, Ba, Cu, Fe, Mn, Ni, Pb,
Rb, Sr, U, and Zn were log(x+l) transformed, while elevation, ORP, NH3, CHLA, Si02,
25 DIC, TN, and TP were square-root transformed. Temperature and pH data were found to have normal distributions, therefore no transformation of these data was conducted.
Numerical Analyses
Environmental and limnological gradients were examined with a Principal Components
Analysis (PCA) performed using CANOCO v4.53 (ter Braak & Smilauer 1998), with variables centered and standardized. Geographical and non-limnological explanatory variables for limnological variation (e.g. latitude, elevation, MSSWT) were passively run in the PCA, as well as CHLA (biological response variable). Na was also run passively due to the high correlation between Na and CI (r = 0.99). The sample scores from the
PCA analysis were used to identify potential outlier lakes, defined as lakes that had PCA
sample scores that exceed the 95% confidence limits of the first two axes of limnological data. One site within our dataset was considered an outlier (AV01) as the sample score exceeded the 95% confidence limit of the PCA site scores. The removal of this site location from the PCA had no discernable impact on site and variable scores observed in the PCA biplot, therefore this location was included in the final analysis.
Relationships between environmental and limnological variables were examined through a Pearson correlation matrix with sequential Bonferroni correction (Rice 1989) for: (i) all sampling locations, (ii) only lake samples, and (iii) only pond samples. Ratios of TN:TP were calculated to indicate potential nutrient limitation, and Na:K ratios were calculated to relate to the influence of catchment vegetation (McNeely et al. 1979).
Individual environmental variables were evaluated with a series of reduced major axis
26 (RMA) regressions performed using PAST 1.61 (Hamilton et al. 2001) to evaluate the relationships for lakes, ponds, and the combined dataset. A Canonical Variates Analysis
(CVA) identified linear combinations of statistically significant limnological variables that best discriminated among regions. Variables that were collinear with other variables
(highly correlated) were sequentially removed in a backwards selection procedure until variance inflation factors (VIFs) were < 10 (ter Braak and Smilauer 1998). Physical and geographical / landscape descriptors (e.g. depth, latitude, surface area, elevation) were excluded from the CVA analysis. As several samples lacked the full water chemistry data
(trace metals), the CVA was conducted on both the reduced dataset excluding lakes without trace metal analysis, as well as on the full dataset excluding trace metals. As there was little difference between the CVA outputs we have chosen to present the results that included the metal analysis.
Results
A total of 57 lakes and 56 ponds were sampled that encompassed a large gradient in several environmental, limnological, and landscape variables across a large area of the eastern Canadian Arctic (Fig 1). Sites ranged in elevation from sea-level to 263 m a.s.l.
Most lakes sampled at low elevations were in the Arviat (south-west Hudson Bay) vicinity, which is characterized by large wetlands, low relief valleys, and sporadic eskers.
Mid-summer surface water temperature (MSSWT) ranged from 5.5 to 22.0 °C, with a mean temperature of 13.9 °C (Table 1). There was a 1.5 °C difference in MSSWT between lakes and ponds (mean 13.1 °C for all lakes, 14.7 °C for all ponds). Depth ranged
27 from 0.2 to 17.2 m, with a median maximum depth of 4.0 m for lakes and 1.0 m for ponds (Table 1). Surface area ranged from 0.5 to 4620 ha (not including Baker Lake,
188700 ha) for lakes (median 56 ha), and 0.5 to 448 ha for ponds (median 9 ha). Only three lakes sampled were both deep ( > 10 m) as well as large ( > 100 ha); Baker Lake, 15
(unnamed lake, Iqaluit area), and B16 (Whitehills Lake, Baker Lake area). pH and conductivity
The pH of lakes and ponds were similar, and both ranged from 6.3 to 8.9 (median 8.0).
Larger, deeper lakes generally had a lower pH, and smaller shallow ponds were more alkaline (Table 1). Ponds with a higher alkalinity corresponded to higher concentrations of DIC and DOC, and a higher specific conductivity (Table 2). pH varied widely in each region, but was generally highest in lakes and ponds in the central Nunavut region, and lowest in the tree-line area (Fig 2). Specific conductivity was also lowest in lakes closest to tree-line and highest in lakes in the western Hudson Bay region (Table 1). As expected in systems with higher conductivity, concentrations of Mg, Sr, SO4, Ca, B, and DIC were also high, with corresponding correlations of greater than 0.67 (Table 2). The most dilute samples were from deeper, larger lakes of high elevation, reflected by the significant negative correlations between elevation and conductivity (Table 2). The relative anion concentrations were CI > DIC > SO4, with means of 12.4, 4.9, and 3.0 mg L"1 respectively
(Table 1). Concentrations of K ranged from 0.07 to 4.83 mg L"1 but varied greatly in lakes and ponds in the Hudson Bay region.
28 Nutrients and Productivity
The ratio of TN:TP was calculated to determine whether lakes and ponds were phosphorus or nitrogen limited, with ratios by weight of TN:TP > 17 reflective of P limitation (Sakamoto 1966, Downing and McCauley 1992). Across all regions, ratios of
TN:TP ranged from 22:1 to 108:1 (median 50:1) by weight, indicating phosphorus
limitation (Table 1). While there was little difference in the mean TN:TP ratio for lakes
versus ponds (Table 1), total nitrogen (TN) concentrations were found to be higher in
warmer, shallow ponds, in the tree-line and southern Hudson Bay regions (Fig 2).
Likewise, the regional variability in TN concentrations was higher in ponds than lake
systems (Fig 2). Significant positive correlations were found between TN and
temperature, conductivity (and associated ions; CI, Ca, Mg, K), DOC, several metals
(e.g., Rb, Ni, Fe), and a negative correlation with depth (Table 2). Several of these relationships were found to be stronger when restricting comparisons to samples from ponds, with higher correlations between TN and MSSWT, DOC, PON, and POC (Table
3). Similar to nitrogen, the concentration of TP was highest in the Hudson Bay region and
lowest in the Baffin region (Fig 2). In addition, while the concentration of total phosphorus was generally lower in lakes (median 5.3 ug L"1) than ponds (median 7.2 ug
L"1), the regional variability of total phosphorus in pond systems was larger than in lake
samples (Fig 2). Several relationships between TP and environmental variables were also
stronger in ponds reflected by regressions between TP and temperature, Al, Fe, DOC,
PON, and POC stronger in pond systems than lakes (Table 3).
29 In contrast to nutrients, chlorophyll-a was similar for both lakes and ponds, ranging from 0.05 to 3.7 ug L"1 (median 1.0 ug.L"1). The largest concentrations of CHLA were found in lakes and ponds in the tree-line (median 1.6 ug L"1) and western Hudson
Bay regions (median 1.2 ug L"1), with the highest variability observed in pond systems
(Fig 2). The corresponding relationship between CHLA and nutrients was stronger than temperature, reflected by the significant correlations between CHLA and nitrogen (TN,
0.61), phosphorus (TP, 0.64), and particulate organic nitrogen (0.60). The importance of organically-bound nutrients, and subsequent CHLA concentrations, is reflected by significant correlations with TP and TN with DOC (Table 2). Concentrations of DOC overall were larger in pond systems, with the highest variability found in ponds in the southern Hudson Bay region (Fig 2). Correlations between DOC and TP, surface area,
Al, Fe, and Rb were higher when comparing pond samples (Table 2). The relationship between MSSWT and DOC, in particular, was significantly stronger only comparing pond samples (Table 3). In contrast, the concentrations of DIC were similar for both ponds and lakes, but highly variable in the western Hudson Bay region (Fig 2).
Metals
A total of 45 lakes and 52 ponds were sampled for trace metals during the summers of
2007 - 2009. The dominant metals found were Fe, Al, Sr, Mn, Ba, B, Cu, Zn, and Rb with median concentrations of 133.0, 20.0, 18.2, 6.7, 5.5, 3.1, 1.2, 1.0, and 0.8 ug L"1, respectively). The median concentration for metals was higher in ponds compared to samples from lakes (Table 1). For example, the largest concentration of Fe was found in
30 ponds sampled in the tree-line region (median 354 ug L" ) compared to western Hudson
Bay (median 160 ug L"1), the central Nunavut region (median 112 ug L"1), and Baffin regions (median 52 ug L"1).
There were several noteworthy relationships between metals and environmental
variables, with correlations between temperature and conductivity with B, Rb and Sr
(Table 2). There were also significant relationships between Fe and CHLA, TN and TP.
Likewise, significant relationships were found between Rb and POC, PON, TN, and TP
(Table 3). Other relationships were only found when comparing either samples from lakes or samples from ponds exclusively. For example, significant relationships were also found between Sr, Mg, and conductivity to DIC when comparing lake samples (Table 3).
In contrast, the relationships found between Al and Fe were only significant when comparing pond samples exclusively (Table 2). Likewise, significant relationships between Fe and POC and PON were only evident in ponds. In addition, ponds with larger
surface areas had positive correlations with temperature, DOC, and several elements (e.g.
B, Rb, Sr) in contrast to deeper lake systems (Table 2).
Multivariate Analyses
The PCA of limnological data identified that the limnological variation amongst the 113 lake and pond samples was along a primary gradient of conductivity, nutrients, and temperature. The first and second principal components (PCI and PC2) accounted for
44.7 and 22.0 % of the variance, respectively (Fig 5). The third and fourth principal components accounted for 10 and 7.2 % of the dataset variance, and are not discussed
31 further. The first two PC axes primarily represented positive associations between nutrients and variables associated with ionic concentration (e.g. Ca, CI, Mg, K, SO4, Na,
DIC) with PCI (Table 4). Negative relationships between PCI and elevation, ORP, and depth were also observed (Table 4, Fig 3). Gradients found along PC2 primarily consisted of a positive relationship with MSSWT, nutrients, DOC, and CHLA, and negative relationships with Si02, DIC, Ca, and pH (Table 4, Fig 3).
The PCA biplot of sample locations indicated a regional ordination, with Baffin region lakes and ponds oriented on the left of the PCA, and those along the western
Hudson Bay region found along the far right (Fig 3). Lakes and ponds in the southern region, within close proximity to tree-line, were oriented in the upper-left quadrant of the
PCA. This reflected the larger depth and surface area, and lower concentration of ions, for lakes in this inland region (Fig 3). Lakes and ponds in the central Nunavut region were primarily found in the center of the PCA, with the exception of larger lakes (e.g.
Baker Lake) that were found grouped with the larger lakes of the southern and Baffin regions (Fig 3). Ponds were generally found oriented within the same general clusters as lakes for most regions, with the exception of ponds in the coastal area of western Hudson
Bay, which were found further to the right in the PCA diagram than the lakes in this region (Fig 3).
A second principal components analysis explored the variation of metal concentrations in a reduced subset of 97 lakes and ponds. The first and second principal components (PCI and PC2) accounted for 34.5 and 18.3 %, respectively, of the variance
(Fig 4). The third and fourth principal components accounted for 9.6 and 6.2 % of the
32 dataset variance respectively. The first two PC axes primarily represented positive
associations between nutrients, and ions with PCI (Table 4). Similar to the PCA of all
sites, gradients found along PC2 primarily consisted of a positive relationship with DOC
and CHLA, and a negative relationship with pH and conductivity (Fig 4). The orientation
of site locations in the PCA of the reduced set of sample locations that included metal
concentrations (Fig 4) was similar to that of the full sample dataset (Fig 3). As such, lakes
and ponds within the Baffin region were oriented on the left of the PCA ordination, and
those along the western Hudson Bay region found along the far right (Fig 4).
Since several relationships between physical and limnological variables were
identified by RMA regressions as being stronger for lake samples than pond samples
(Table 3), additional sets of PCAs were conducted to compare lakes and ponds
independently (Table 4). The variance explained in each PC axis for both the lake dataset
and the pond dataset was similar (Table 4). The PCA scores for most variables were also
similar for both axes in each dataset, reflecting similar gradients in limnology for both
lakes and ponds. However, there was a larger gradient in DIC reflected by PCI for lakes,
versus a larger gradient in DOC and NH3 for ponds (Table 4). Likewise, a larger
temperature gradient was reflected by PC2 for lakes, while there was a larger
conductivity gradient for ponds (Table 4).
A CVA was used to analyze classified lakes and ponds into the four regions identified based on their limnology; Tree-line, Central Kivalliq, western Hudson Bay, and
Baffin (Fig 5). The first CVA axis accounted for 30.3 % of the difference between regions, while 26 % of the variance was explained by CVA axis 2 (Fig 5). There was a
33 significant correlation between CVA axis 1 and temperature, B, Ba, Ni, Rb, SO4, and
NH3, and a significant correlation between CVA axis 2 and pH, conductivity, Al, Ba, Fe,
DOC, and CHLA (Table 5). The CVA axis 1 gradient primarily represented the difference between lakes and ponds in the warmer Hudson Bay and tree-line regions that had higher concentrations of B, Rb, and Ni oriented to the right of the ordination, from the colder central and Baffin regions that had higher concentrations of SO4, Ba and NH3 that were oriented to the left (Fig 5, Table 4). The second CVA axis was represented by lakes and ponds in the tree-line region that had higher concentrations of CHLA, DOC and
Fe found in the upper quadrant of the ordination, compared to lakes in the Baffin region that had significantly higher pH and conductivity found in the lower-left quadrant (Fig 5).
While we sampled two distinct areas of western Hudson Bay, the vicinity of Arviat (Fig
lb) and the vicinity of Rankin Inlet (Fig Id), there was little difference found in the limnology of these two areas in the CVA analysis (Fig 5). Thus, samples from Arviat and
Rankin Inlet were grouped into the western Hudson Bay region for discussion. Intra- regional PCAs indicated similar limnological trends between lakes and ponds as were found between all regions.
Discussion
While the limnology of vast areas of the Canadian Arctic is still unknown, our analysis indicates common trends that may span multiple physiographic regions. Our sampling regime attempted to target gradients in the local physical and environmental characteristics as well as examine potential differences between pond systems and lakes.
34 This allowed for both inter- and intra-regional comparisons of the limnology of both lakes and ponds independently. Pond systems generally had higher concentrations and greater regional variability than lake systems (Fig 2), and some differences in the gradients of DOC, DIC, and pH (Table 4), however, the overall patterns were similar for both lakes and ponds for most variables (Table 4, Fig 4). Large gradients were found in several major ions (B, CI, K, Mg), metals (Rb, Sr), and nutrients for both lakes and ponds in our dataset (Table 4, Fig 4). To a lesser extent, a gradient in pH, CHLA, and metals
(Al, Fe, Mn) was also found along PCA axis 2 (Table 4). The CVA analysis, however, indicated four distinct regions with respect to limnological characteristics. Gradients of temperature, CHLA, nutrients, conductivity, and metals were observed between these regions (Fig 5). For example, lakes and ponds sampled within the western Hudson Bay region characteristically had higher conductivity and ion concentrations than other regions (Fig 4).
Intra-regional PCAs displayed many of the similar trends to those observed across physiogeopgrahic regions and trends were also generally conserved between regions. In each region, the main differences observed were the separation of lake and pond samples into opposite quadrats. Ponds were generally associated with increased ions and metals
(Cu, Al, Fe, Mn, Ni and Sr). There was a gradient of decreasing nutrient concentrations, and CHLA, along a latitudinal gradient, especially in pond systems. This is likely due to influence of higher nutrients on productivity among more southern sites as well as the smaller volume of pond systems resulting in higher turnover, remobilization of particles from the sediments, and the reduced water residency time in these bodies. Pond systems
35 are generally smaller, and hydrologically disconnected systems that can also be influenced by the concentration of elements during drought and dilution during large precipitation events (Macrae et al. 2006).
Physical characteristics
Across all sampling locations within our dataset, there were large differences in depth, elevation, and surface area within each region. The influence of these physical characteristics is most pronounced in the limnology of ponds (< 2 m depth) versus lakes
(Table 2), especially in the western Hudson Bay region where [Fe] and the conductivity of ponds was significantly higher than lakes (Fig 4). Higher inputs to lakes and ponds were found in low relief regions, but the relationship between the surface area of a lake or pond and its limnological characteristics is not clear (Table 2). Hamilton et al. (2001) found that larger lakes had lower conductivity, and lower concentrations of ions, nutrients, and carbon particulates, but our dataset did not indicate similar relationships with surface area (Table 2). This is likely due to the low topographic relief of the majority of our sampling locations across mainland Nunavut. The Kivalliq region, in particular, was the central maximum of the Laurentide ice sheet, which depressed the land surface, carved out lake basins, and eroded much of the region's soil horizon (Pielou
1992). Thus, the area is characterized by extensive wetlands and networks of chained lakes and ponds connected by dendritic rivers and streams (Vincent et al. 2008). The lakes in the south-western Hudson Bay region (Fig lb) are characterized by large surface areas with a relatively shallow maximum depth (median depth 4 m). These shallow,
36 large, and wind-swept systems have higher concentrations of ions, higher productivity, and higher nutrients (Fig 3), which could be due to increased interaction with organic- rich benthic sediments or because of wind-induced sediment resuspension (Larsen and
MacDonald 1993).
Local geological influences
The limnology of lakes and ponds in Arctic systems is highly influenced by the composition of the underlying lake sediments and input catchment areas. For example,
Arctic ponds are often alkaline due to the presence of calcareous glacial tills overlying
Precambrian bedrock (Douglas et al. 2000; Hamilton et al. 2001). Westover et al. (2009) found that the main difference in limnology between lakes and ponds from the mainland
Kitikmeot region and Victoria Island lakes was based on the solubility of elements in sediments, indicated by the dominance of Ca and Mg cations from dolomitic sediments characteristic of Victoria Island lakes, and Na for lakes throughout the marine sediment deposits characteristic of the mainland Kitikmeot region. In contrast, the tree-line, central, and Baffin regions all contained a relative cationic abundance of Ca > Na > Mg >
K, which reflects the relative solubility of silicate minerals from the igneous sources of the sediments in these areas (Wetzel 2001). Lakes and ponds in the coastally-influenced western Hudson Bay region contained higher concentrations of sodium (Na > Ca > Mg >
K), and subsequently high conductivity, than other regions. Pienitz et al. (1997) found that coastal lakes had higher concentrations of Na, and CI, and marine aerosols can strongly influence the chemistry of coastal lakes. While 8 lakes and ponds in our dataset
37 from the western Hudson Bay region were within 2 km of the coast, the majority were
10-100 km from the current coastline. However, this region was inundated by seawater after the retreat of the last glaciation, and many lakes and ponds in this region were isolated from marine environments by the slow isostatic uplift that has continued over the last 7500 years (Pielou 1992). This is reflected by the large glaciomarine sediment composition in this area. Likewise, some lakes within this region are underlain by
Achaean metamorphic rock, which can provide a significant source of sodium from feldspar weathering (Banks et al. 1998). As a result, it is likely that the limnology of these systems is influenced by the legacy marine sediments in the surficial geology of the
Hudson Bay lowlands.
Temperature, nutrients, and productivity
A large portion of nutrient inputs to Arctic soils occurs from precipitation, while limited nitrogen fixation by microbial and algal communities is temperature linked (Chapin et al.
1983). Since high TN:TP ratios indicating phosphorus limitation were found for both lake and pond systems across all regions (Table 1), the majority of N inputs likely occur from allochthonous sources. The prevalence of phosphorus limitation in Canadian Arctic aquatic systems is consistent with other limnological surveys (Gregory-Eaves et al. 2000,
Hamilton et al. 2001, Lim et al. 2001, Michelutti et al. 2002, Ruhland et al. 2003,
Westover et al. 2009). The highest concentrations of both phosphorus (Fig 2i) and nitrogen (Fig 2j) were in lakes and ponds in the southern Hudson Bay region, but similarly high concentrations were also found in lakes and ponds, in several regions, in
38 systems that had higher MSSWT (Fig 4). The significant relationship found in our results between nitrogen and temperature (Table 2) strongly suggests that nitrogen concentrations may be dependant on temperature; areas with higher mean temperatures may have increased microbial decomposition, larger amounts of vegetation, and high allochthonous inputs that correspond to higher nitrogen concentrations found. This is consistent with Hobara et al. (2006) who found that nitrogen fixation was several times greater than decomposition rates, and exponentially increased under higher temperature, moisture, and light availability. The short growing season, low temperatures, and limitations in available phosphorus and nitrogen are known to significantly reduce primary production in Arctic lakes and ponds (Schindler et al. 1974, Alexander et al.
1989). In addition, Chetelat et al. (2010) found that the primary energy source for Arctic food-webs came from the benthic algae community rather than pelagic algae typical of temperate systems.
Reduced decomposition rates in Arctic soils result in larger pools of soil organic matter, and subsequently larger amounts of bound phosphorus, nitrogen, and carbon, compared to temperate systems (Dowding et al. 1981, Chapin et al. 1983). While organic inputs from terrestrial sources and internal biogeochemical processing may be significant, rates of organic matter decomposition are lower than similar temperate systems due to lower temperatures and subsequently lower rates of autochthonous inputs (Hobbie 1973).
The overall concentrations of DOC in our dataset were low compared to temperate systems, but higher concentrations of DOC were found in lakes and ponds with higher corresponding MSSWT (Table 2), likely reflecting greater terrestrial inputs of available
39 dissolved organic matter during a warmer and more prolonged ice-free season.
Concentrations of nutrients were also found to be significantly higher in samples from ponds in the Hudson Bay region, and significantly correlated to systems that contained higher concentrations of DOC (Table 2, Fig 2), likely reflecting higher temperature- mediated, microbial activity within these systems. This is consistent with Kling (1995), who identified that bacterial respiration and production is strongly linked to inputs of labile dissolved organic matter in Lake Toolik, Alaska. Pools of organic materials in
Arctic soils also contain high concentrations of free amino acids, and are an important source of nitrogen to the Arctic vegetation community (Kielland 1994).
The higher variability in dissolved and particulate organic carbon in pond systems may also reflect temporary hydrologic connections to ephemeral streams and ponds in large watersheds (Macrae et al. 2004, 2006). Overall precipitation in Arctic regions is low, and the majority of water enters systems during the spring snowmelt. Water derivied from melting of the upper soil horizon during late spring is strongly enriched with organic compounds due to percolation through humic-rich soils before entering streams, and subsequently lakes, within large watersheds. Organic carbon concentrations therefore increase in parallel to ground water discharge, increasing water colour and decreasing water clarity among shallow lakes, ponds, and streams from the late spring to the summer
(Rouse et al. 1997; Douglas et al. 2000; Vincent et al. 2008).
Hedin et al. (1998) argued that terrestrial-aquatic hyporheic interfaces adjacent and beneath lakes, ponds, and streams acted as focal points for nutrient transformation, and are the most important control points along paths of nutrient flux from terrestrial to
40 aquatic ecosystems. These shallow pond systems have also been shown to be important for atmospheric carbon and nitrogen cycling (Bello & Smith 1990, Kling et al. 1991), and may undergo further limnological variability under climate warming if hyporheic areas expand (Rouse et al. 1997, Edwardson et al. 2003).
Biological productivity within freshwater systems is also highly influenced by temperature, nutrients, and organic carbon inputs (Rouse et al. 1997). Median concentrations of CHLA for the central and Baffin regions (0.9 and 0.7 \ig L"1 respectively) were comparable to values in the eastern N.W.T. (0 9 ug.L"1; Pienitz et al.
1997), the central-east tree-line areas of the N.W.T. and central-west region of Nunavut
(0.85 ug L"1; Ruhland et al. 2003), and in the north-western Kitikmeot region of Nunavut
(0.60 ug.L"1; Westover et al. 2009). CHLA was also significantly correlated to both phosphorus and nitrogen, indicating a primary nutrient limitation within lakes and ponds in our dataset (Table 2). In addition, CHLA was found to be significantly correlated (r =
0.62, p < 0.05) with Fe in lakes. Dillon & Molot (1991) noted that nutrient inputs to temperate lakes and ponds are organically bound. This is often reflected in significant
DOC-TP and DOC-TN correlations (Table 2; Dillon et al. 1991, 1997). Iron enhances phosphorus complexation with DOC, and can increase catchment fluxes of TP (Jones et al. 1988; Dillon et al. 1997). Internal recycling of phosphorus and iron also depends on the depth of the system, where shallow large lakes will have a higher potential for release of nutrients and other compounds from sediment due to wind-induced resuspension. For example, Bostrom et al. (1988) found that bioturbation and sediment resuspension can result in the release of iron and phosphorus from sediments.
41 The input of organic carbon to Arctic lakes and ponds greatly influences the limnology and productivity of these systems. For the 99 samples with particulate organic carbon (POC) data available, only 6 had POC:CHLA values of less than 200:1, reflecting the relatively large input of allochthonous carbon sources. The median ratio of 548:1 is similar to other Arctic limnology surveys (e.g. Lim et al. 2001; Hamilton et al. 2001;
Michelutti et al. 2002), which is generally representative of low-productivity systems north of the tree-line. In addition, Secchi depth was almost always equal to the maximum observed depth at all of our sampling locations, indicating the overall clarity of aquatic tundra systems and low concentrations of POC. Those systems with higher available concentrations of nutrients, and subsequent higher productivity, had higher concentrations of DOC. This is also likely due to inputs from the large abundance of wetlands in the southern regions of Hudson Bay (Fig 2), which are a key contributor of allochthonous sources of DOC (Wetzel, 1992).
The concentrations of dissolved and particulate inputs in Arctic lakes and ponds also have significant consequences on the biological communities within these systems.
The concentration of DOC in 55 % of lakes and ponds was less than 4 mg L"1, which has been shown to be a threshold below which there is greatly increased UV-B penetration in the water column (Pienitz and Vincent 2000). Since dissolved organic matter strongly absorbs radiation in the UVR region, systems with low concentrations of DOC have aquatic communities exposed to high levels of UV-B radiation that can damage invertebrates (Molot et al. 2004).
42 Metals
The concentration of the more abundant metals (Fe, Al, Mn, Rb) were positively associated with particulates (POC, PON), indicating likely allochthonous inputs
(Hamilton et al. 2001). In addition, Arctic lakes and ponds are primarily highly oxidizing environments. Low temperatures, low autochthonous organic matter input, and low sediment accretion leads to conditions where relatively high retention of diffusing Mn and Fe in sediments occurs (Cornwell and Kipphut, 1992). The median concentrations of
Al (20.25 ug L"1), Cu (1.39 ug L1), Mn (8.15 ug L"1), and Zn (1.17 ug L1) in our dataset were similar to those found by Westover et al. (2009), for their Takijuq Lake Upland zone of the Kitikmeot region (22.4 jag L"1, 1.22 ug L"1, 2.56 ug L"1, and 1.6 ug L"1 respectively). In contrast, the median concentration of Fe was higher in lakes sampled in the tree-line region (354 ug L"1) than western Hudson Bay (160 ug L"1), the central
Nunavut region (112 ug L"1), and Baffin regions (52 ug L"1). In comparison, Westover et al. (2009) reported that median Fe concentrations (61.8 ug L"1), were higher than other studies in the same area (Pienitz et al. 1997, Ruhland et al. 2003).
The pH of freshwater systems is an important component of the sediment/water interface, as it controls metal cation hydrolysis and specific sorption rates. Under alkaline conditions, sediments sorb Al and Zn, while Cu, Fe, and Mn precipitate (Jackson 1998), however, no correlations were found between pH and the concentrations of Fe or Mn regardless of alkaline conditions (Table 2). Hobbie (1973) indicated that high Fe inputs are primarily controlled by chelation with available organic acids that keep iron in the water column. Indeed, Molot and Dillon (2005) showed that the loss of Fe was negatively
43 correlated with DOC. High concentrations of Fe found in lakes sampled in the tree-line region suggest higher inputs from terrestrial organic matter sources. In addition, pond systems had a stronger relationship between DOC and Fe than lake systems (r = 0.62 vs
0.43), possibly due to the resuspension of Fe from wind-induced mixing that is common in ponds (Schlesinger 1997). The concentration of Fe was also correlated to Mn, CHLA, depth, and nitrogen (Table 2).
Concentrations of strontium (Sr) and rubidium (Rb) were correlated to several ions (Ca, Mg, K, CI, and SO4), nutrients, and temperature. Rb in particular was strongly correlated to temperature, particulates, K, nutrients, and CHLA (Table 2). The correlation with K may be due to competitive uptake with K by terrestrial vegetation (Murphy et al.
1955, Drobner and Tyler 1998). In addition, both Rb and K are found in higher concentrations in clay particulates, as clay soils have a high affinity for both Rb and K
(MacDougall and Harris 1969, Drobner and Tyler 1998). [Sr] was found to be elevated in lakes with high concentrations of Ca and Ba, and all three of these elements had significant positive correlations to DIC (Table 2). This is consistent with Puznicki (1996), who reported high concentrations of Ba and Sr in lakes with calcium carbonate-rich deposits in the central N.W.T.
Conclusion
While our dataset reflects lakes sampled in clusters that spanned a large spatial extent of the eastern Canadian Arctic, several significant gradients were found across all regions.
The relationships found between nutrients (TP, TN) and productivity (CHLA) was
44 similar to other Arctic limnology reviews, but the concentration of nutrients and CHLA were higher in our dataset, especially in the southern regions of Nunavut. In addition, mid-summer surface water temperature was found to be strongly correlated to nitrogen concentrations across all regions. High TN:TP ratios indicated that phosphorus limitation was prevalent in both lakes and ponds across all regions.
The CVA analysis indicated that several lake and pond samples could be grouped together into specific ecoregions based on their limnology. For example, lakes and ponds in the western Hudson Bay region shared common patterns in conductivity, DIC/DOC, temperature, and nutrients. These regional characteristics are likely due to the glaciomarine origin of sediments. In contrast, the tree-line region was characterized by lower conductivity, lower concentrations of major ions, and higher productivity than other regions. Shallow pond systems had a much higher variability in the concentration of major ions and metal concentrations than that of deeper lake systems, especially in the the western Hudson Bay region. The overall general patterns in limnological characteristics within each region, however, were similar for both lakes and ponds. Thus, anticipated future changes in climate may influence both lakes and ponds in a similar fashion within a particular region.
45 Acknowledgements
This project was funded by a NSERC Discovery Grant and NSERC Northern Research
Supplement held by RQ, NSERC Northern Research Internships (NRINT) held by ASM,
RGB, and CEL, the Northern Scientific Training Program (NSTP), and additional funding for graduate student research. Fieldwork assistance and support was provided by
Mary Ellen Thomas, Jamal Shirley, Dorothy Tootoo and the staff at the Nunavut
Research Institute and Nunavut Arctic College. We are also grateful for field sampling assistance from Andy Aliyak, Andrew Dunford, and Milissa Elliott.
References
Alexander V, Whalen SC, Klingensmith KM. 1989. Nitrogen cycling in Arctic lakes and ponds. Hydrobiologia 172:165-172. Banks D, Frengstad B, Midtgard AK, Krog JR, Strand T. 1998. The chemistry of Norwegian groundwaters: I. The distribution of radon, major and minor elements in 1604 crystalline bedrock groundwaters. Sci Total Environ. 222:71-91. Bello R, Smith JD. 1990. The effect of weather variability on the energy balance of a lake in the Hudson Bay Lowlands, Canada. Arctic and Alpine Res. 22:98-107. Bostrom B, Anderson JM, Fleischer S, Jansson M. 1988. Exchange of phosphorus across the sediment-water interface. Hydrobiologia 170:229-244. Chapin FS III. 1983. Direct and Indirect Effects of Temperature on Arctic Plants. Polar Biol. 2:47-52. Chetelat J, Cloutier L, Amyot M. 2010. Carbon sources for lake food webs in the Canadian High Arctic and other regions of Arctic North America. Polar Biol. 33:1111-1123. Cote G, Pienitz R, Velle G, Wang XA. 2010. Impact of geese on the limnology of lakes and ponds from Bylot Island (Nunavut, Canada). Int Rev Hydrobiol. 95:105-129. Cornwell JC, Kipphut GW. 1992. Biogeochemistry of manganese- and iron-rich sediments in Toolik Lake, Alaska. Hydrobiologia 240:45-59. Danks HV. 2007. How aquatic insects live in cold climates. Can Entomol. 139:443-471. Dillon PJ, Molot LA, Scheider WA. 1991. Phosphorus and nitrogen export from forested stream catchments in central Ontario. J Environ Qual. 20:857-864.
46 Dillon PJ, Molot LA. 1997. Effect of landscape form on export of dissolved organic carbon, iron, and phosphorus from forested stream catchments. Water Resour Res. 33:2591-2600. Douglas MSV, Smol JP, Blake W Jr. 2000. Summary of paleolimnological investigations of high Arctic ponds at Cape Herschel, east-central Ellesmere Island, Nunavut. In: Garneau M, Alt B. (eds), Environmental Response to Climate Change in the Canadian High Arctic. Bulletin of the Geological Survey of Canada 529:257-269. Dowding P, Chapin FS III, Wielgolaski FE, Kilfeather P. 1981. Nutrients in tundra ecosystems. In: Bliss LC, Heal OW, Moore JJ (eds). Tundra ecosystems: a comparative analysis. Cambridge University Press, Cambridge, pp 647-683. Downing JA, McCauley E. 1992. The nitrogen:phosphorus relationship in lakes. Limnol Oceanogr. 37:936-945. Drobner U, Tyler G. 1998. Conditions controlling relative uptake of potassium and rubidium by plants from soils. Plant Soil 201:285-293. Edwardson KJ, Bowden WB, Dahm C, Morrice J. 2003. The hydraulic characteristics and geochemistry of hyporheic and parafluvial zones in Arctic tundra streams, north slope, Alaska. Adv Water Res. 26:907-923. Ellis S. 1980. An investigation of weathering in some arctic-alpine soils on the northeast flank of Okaskolten, North Norway. J Soil Sci. 31:371-385. Environment Canada. 1994a.Manual of Analytical Methods, Vol 1. Major Ions and Nutrients. Environmental Conservation Service - ECD. Canadian Communications Group. Toronto, Ontario. Environment Canada. 1994b.Manual of Analytical Methods, Vol 2. Trace Metals. Environmental Conservation Service - ECD. Canadian Communications Group. Toronto, Ontario. Findlay S. 1995. Importance of surface-subsurface exchange in stream ecosystems: The hyporheic zone. Limnol Oceanogr. 40:159-164. Flanagan KM, McCauley E, Wrona F, Prowse T. 2003. Climate change: the potential for latitudinal effects on algal biomass in aquatic ecosystems. Can J Fish Aquat Sci. 60:635-639. Fulton RJ. 1995. Surficial materials of Canada, Geological Survey of Canada, Map 1880A. Gregory-Eaves I, Smol JP, Finney BP, Lean DRS, Edwards E. 2000. Characteristics and variation in lakes along a north-south transect in Alaska. Arch Hydrobiol 147:193-223. Hamilton PB, Gajewski K, Atkinson DE, Lean DRS. 2001. Physical and chemical limnology of 204 lakes from the Canadian arctic archipelago. Hydrobiologia 457:133-148. Hanmer S, Sandeman HA, Davis WJ, Aspler LB, Rainbird RH, Ryan JJ, Relf C, Peterson TD. 2004. Geology and Neoarchean tectonic setting of the Central Hearne supracrustal belt, Western Churchill Province, Nunavut, Canada. Precambrian Res. 134:63-83.
47 Hedin LO, von Fischer JC, Ostrom NE, Kennedy BP, Brown MG, Robertson PG. 1998. Thermodynamic constrains on nitrogen transformations and other biogeochemical processes at soil-stream interfaces. Ecology 79:684-703. Hobara S, McCalley C, Koba K, Giblin AE, Weiss MS, Gettel GM, Shaver GR. 2006. Nitrogen Fixation in Surface Soils and Vegetation in an Arctic Tundra Watershed: A Key Source of Atmospheric Nitrogen. Arc Antarc Alp Res. 38:363-372. Hobbie JE. 1973. Arctic limnology: a review. In: Britton ME (ed). Alaskan Arctic tundra. Technical paper No. 25. Arctic Institute of North America, Calgary, Alberta. pp. 127-168. Hobbie SE Chapin SF III. 1992. Winter regulation of tundra litter carbon and nitrogen dynamics. Biogeochemistry 35:327-338. Hutchins0n GE. 1957. A treatise on limnology. Vol. 1. Geography, physics and chemistry. John Wiley and Sons, NY, USA. 1015pp. Jackson TA, Hecky RE. 1980. Depression of primary productivity by humic matter in lake and reservoir waters of the boreal zone. Can J Fish Aquat Sci. 37:2300-2317. Jackson TA. 1998. The biogeochemical and ecological significance of interactions between colloidal minerals and trace elements. In: Parker A, Rae J (eds). Environmental Interactions of Clays. Clays in the Environment, Springer, pp. 1-5. Jones RI, Salonen K, de Haan H. 1988. Phosphorus transformations in the epilimnion of humic lakes: Abiotic interactions between dissolved humic materials and phosphate. Freshw Biol. 19:357-369. Kielland K. 1994. Amino acid absorption by Arctic plants: Implications for plant nutrition and nitrogen cycling. Ecology 75:2373-2383. Kling GW, Kipphut GW, Miller MC. 1991. Arctic lakes and streams as gas conduits to the atmosphere: implications for tundra carbon budgets. Science 251:298-301. Kling GW, O'Brien WJ, Miller MC, Hershey AE. 1992. The biogeochemistry and zoogeography of lakes and rivers in Arctic Alaska. Hydrobiologia 240:1-14. Kling GW. 1995. Land-water interactions: the influence of terrestrial diversity on aquatic ecosystems. In: Chapin FS III, Komer C (eds). Arctic and Alpine Biodiversity: Patterns, Causes and Ecosystem Consequences, Springer-Verlag, Berlin, pp 297- 310. Larson CPS, MacDonald GM. 1993. Lake morphometry, sediment mixing and the selection of sites for fine resolution palaeoecological studies. Quat Sci Rev. 12:781-792. Lim DSS, Douglas MSV, Smol JP, Lean DRS. 2001. Physical and chemical limnological characteristics of 38 lakes and ponds on Bathurst Island, Nunavut, Canadian High Arctic. Int Rev Hydrobiol. 86:1-22. Lim DSS, Douglas MSV, Smol JP. 2005. Limnology of 46 lakes and ponds on Banks Island, N. W.T., Canadian Arctic archipelago. Hydrobiologia 545:11-32. Marcae ML, Bello RL, Molot LA. 2004. Long-term carbon storage and hydrological control of CO2 exchange in tundra ponds in the Hudson Bay Lowland. Hydrol Process. 18:2051-2069. Macrae ML, Devito KJ, Creed IF, Macdonald SE. 2006. Relation of soil-, surface-, and ground-water distributions of inorganic nitrogen with topographic position in
48 harvested and unharvested portions of an aspen-dominated catchment in the Boreal Plain. Can J Forest Res. 36:2090-2103. McNeely RN, Neimanis VP, Dwyer L. 1979. Water Quality Sourcebook: A Guide to Water Quality Parameters. Inland Waters Directorate, Water Quality Branch: Ottawa, 89 pp. Molot LA, Keller W, Leavitt PR, Robarts RD, Waiser MJ, Arts MT, Clair TA, Pienitz R, Yan ND, McNicol DK, Prairie YT, Dillon PJ, Macrae ML, Bello R, Nordin RN, Curtis PJ, Smol JP, Douglas MSV. 2004. Risk analysis of dissolved organic matter-mediated ultraviolet B exposure in Canadian inland waters. Can J Fish Aquat Sci. 61:2511-2521. Molot LA, Dillon, PJ. 2005. Long-term trends in catchment export and lake retention of dissolved organic carbon, dissolved organic nitrogen, total iron, and total phosphorus: The Dorset, Ontario, study, 1978-1998. J Geophys Res. 110:G01002. MacDougall JD, Harris RC. 1969. The geochemistry of an Arctic Watershed. Can J Earth Sci. 6:305-315. Michelutti N, Douglas MSV, Lean DRS, Smol JP. 2002. Physical and chemical limnology of 34 ultra-oligotrophic lakes and ponds near Wynniatt Bay, Victoria Island, Arctic Canada. Hydrobiologia 482:1-13. Murphy WS, Hunter AH, Pratt PF. 1955. Absorption of rubidium by plants from solution and soils. Soil Sci Soc Am J. 19:433^135. Osterkamp TE, Romanovsky VE. 1999. Evidence for warming and thawing of discontinuous permafrost in Alaska. Permafrost Periglac. 10:17-37. Pielou EC. 1992. After the ice age: the return of life to glaciated North America. University of Chicago Press, Chicago, IL. 376pp. Pienitz R, Smol JP, Lean DRS. 1997. Physical and chemical limnology of 59 lakes located between the southern Yukon and the Tuktoyaktuk peninsula, Northwest Territories (Canada). Can J Fish Aquat Sci. 54:330-346. Pienitz R, Vincent WF. 2000. Effect of climate change relative to ozone depletion on UV exposure in subarctic lakes. Nature 404:484-487. Puznicki W. 1996. An overview of lake water quality in the Slave structural province area Northwest Territories. Water Resources Division, Natural Resources and Environmental Directorate. Prepared for the Department of Indian and Northern Affairs, Yellowknife, N.W.T., Canada: 153 pp. Prowse TD, Wrona FJ, Reist JD, Gibson JJ, Hobbie JE, Levesque LMJ, Vincent WF. 2006. Climate Change Effects on Hydroecology of Arctic Freshwater Ecosystems. Ambio 35:347-358. Rasmussen JB, Godbout L, Schallenberg M. 1989. The Humic Content of Lake Water and its Relationship to Watershed and Lake Morphometry. Limnol Oceanogr. 34:1336-1343. Rice WR. 1989. Analyzing tables of statistical tests. Evolution 43:223-225. Rouse W, Douglas M, Hecky R, Kling GW, Lesack L, Marsh P, McDonald GM, Nicholson B, Roulet N, Smol JP. 1997. Effects of climate change on the freshwaters of Arctic and sub Arctic North America. Hydrol Process. 11:873-902.
49 Ruhland KM, Smol JP, Wang XA, Muir DCG. 2003. Limnological characteristics of 56 lakes in the central Canadian Arctic tree-line region. J Limnol. 62:9-27. Sakamoto M. 1966. Primary production by phytoplankton community in some Japanese lakes and its dependence on lake depth. Arch Hydrobiol 62:1-28. Schindler DW, Welch HE, Kalff J, BrunskiU GJ, Kritsch N. 1974. Physical and chemical limnology of Char Lake, Cornwallis Island (75°N lat.). J Fish Res Board Can. 31:585-607. Serreze MC, Walsh JE, Chapin FS III, Osterkamp T, Dyurgerov M, Romanovsky V, Oechel WC, Morison J, Zhang T, Barry RG. 2000. Observational evidence of recent change in the northern high-latitude environment. Clim Change 46:159- 207. Schlesinger WH. 1997. Biogeochemistry: an analysis of global change. - Academic Press, San Diego, CA, U.S.A. 588pp. Smith LC, Sheng Y, MacDonald GM, Hinzman LD. 2005. Disappearing Arctic Lakes. Science 308:1429. Stockwell CH, McGlynn JC, Emslie RF, Sanford BV, Norris AW, Donaldson JA, Fahrig WF, Currie KL. 1976. Geology of the Canadian Shield. In: R.J.W. Douglas (Editor) Geology and Economic Minerals of Canada Part A. Economic Geology Report No. 1. Minister of Supply and Services Canada. Supply and Services Canada, Ottawa, Canada P 43-150. ter Braak CJF, Smilauer P. 1998. CANOCO reference manual and user's guide to CANOCO for windows: Software for canonical community ordination (version 4). Microcomputer Power, New York, USA. 352pp. Vincent WF, Laybourn-Parry J (eds) 2008. Introduction to the limnology of high-latitude lake and river ecosystems. Polar Lakes and Rivers: Limnology of Arctic and Antarctic Aquatic Ecosystems. Oxford University Press, NY, USA. 327pp. Westover KS, Moser KA, Porinchu DF, MacDonald GM, Wang XA. 2009. Physical and chemical limnology of a 61-lake transect across mainland Nunavut and southeastern Victoria Island, Central Canadian Arctic. Fund Appl Limnol. 175:93-112. Wetzel RG. 1983. Limnology (2nd ed.). Saunders Publishing, Philadelphia, PA. 767pp. Wetzel RG. 1992. Gradient-dominated ecosystems: sources and regulatory functions of dissolved organic matter in freshwater ecosystems. Hydrobiologia 229:181-198. Wetzel RG. 2001. Limnology: lake and river ecosystems (3rd ed). Academic Press, San Diego, CA, U.S.A. 1006pp. Wheeler JO, Hoffman PF, Card KD, Davidson A, Sanford BV, Okulitch AV, Roest WR. 1997. Geological map of Canada. Geological Survey of Canada, Map D1860A.
50 Tables
Table 1 Summary of environmental and limnological variables. * Excludes BI (Baker Lake). Full dataset is available in supplementary materials.
All Lakes Ponds
Variable Unit s -a (3 3 3 2 Area* Ha 0.1 4620 22.5 201 0.8 4620 56 364 0.0 448.3 9.0 38.3 Elev masl 0.0 263.0 59.0 76.5 0.0 260.0 58.0 75.0 4.0 263.0 63.0 78.1 Depth m 0.2 17.2 2.0 3.1 2.0 17.2 4.0 5.1 0.2 1.9 1.0 1.1 MSSWT °C 5.6 22.0 13.0 13.9 5.6 19.4 12.6 13.1 7.2 22.0 14.0 14.7 pH -log[H+] 6.3 8.9 8.0 7.8 6.5 8.8 7.9 7.8 6.3 8.9 8.1 7.9 COND uS cm"1 1 1006 48 74 1 255 45 57 8 1006 57 92 ORP mV 16 277 130 131 16 277 133 140 19 261 131 128 CI mgL" 1 0.2 235.0 4.8 12.4 0.2 51.3 4.3 8.1 0.2 235.0 5.8 16.9 Ca mgL" 1 0.2 27.5 5.0 6.6 0.2 27.5 4.5 5.7 0.7 26.3 5.2 7.4 Mg mgL" 1 0.2 18.3 1.3 1.8 0.2 4.6 1.1 1.3 0.4 18.3 1.5 2.3 K mgL" 1 0.1 4.8 0.6 0.8 0.1 3.0 0.6 0.8 0.1 4.8 0.6 0.9 Na mgL" 1 0.4 121.0 2.5 6.8 0.4 27.4 2.4 4.6 0.6 121.0 3.2 9.0 Si02 mgL" 1 0.0 4.1 0.3 0.7 0.0 3.5 0.2 0.6 0.0 4.1 0.4 0.7 Al rlgL"' 2 215 20 37 2 206 18 34 4 215 20 39 As MgL"' 0.0 0.7 0.2 0.2 0.0 0.7 0.1 0.2 0.0 0.4 0.2 0.2 B MgL" 0.8 66.1 3.1 7.0 0.9 35.8 2.9 5.5 0.8 66.1 3.3 8.3 Ba MgL" 0.9 25.5 5.5 7.1 0.9 25.5 3.8 6.1 0.9 20.3 6.8 8.0 Cu MgL" 0.3 29.2 1.2 1.8 0.3 5.3 0.9 1.4 0.5 29.2 1.4 2.1 Fe MgL" 7 1500 133 194 7 497 70 125 28 1500 168 253 Mn MgL" 0.9 30.4 6.7 8.6 0.9 20.0 6.3 6.7 1.2 30.4 8.1 10.3 Ni UgL" 0.1 4.2 0.4 0.4 0.1 4.2 0.3 0.4 0.1 1.1 0.4 0.5 Pb MgL"' 0.0 0.3 0.1 0.1 0.0 0.2 0.1 0.1 0.0 0.3 0.1 0.1 Rb MgL"' 0.2 3.3 0.8 1.1 0.2 2.6 0.7 0.9 0.2 3.3 0.8 1.3 Sr MgL"' 1.8 167.8 18.2 25.6 1.8 72.5 13.9 21.0 4.7 167.8 19.6 29.6 U MgL" 0.0 3.0 0.0 0.1 0.0 1.8 0.0 0.1 0.0 3.0 0.1 0.2 Zn MgL"1 0.2 80.3 1.0 2.2 0.2 2.3 0.8 0.9 0.4 80.3 1.2 3.2 S04 mgL" 1 0.0 21.2 2.1 3.0 0.0 13.9 2.1 2.6 0.1 21.2 2.3 3.4 DOC mgL" 1 0.6 64.0 4.0 6.0 0.6 21.2 3.7 4.5 1.6 64.0 4.7 7.6 DIC mgL" 1 0.6 19.4 4.3 4.9 0.6 15.9 3.7 4.4 1.1 19.4 4.5 5.4 CHLA MgL"1 0.1 3.7 1.0 1.2 0.1 2.9 1.0 1.1 0.1 3.7 1.0 1.2 NH3 MgL" 8 497 40 51 8 189 33 37 10 497 48 65 PON mgL" 1 0.02 0.30 0.05 0.07 0.02 0.20 0.05 0.06 0.02 0.30 0.06 0.08 POC mgL" 1 0.13 3.56 0.48 0.67 0.13 2.51 0.40 0.51 0.17 3.56 0.58 0.82 TP MgL"1 1.2 16.4 5.8 6.8 1.2 13.1 5.3 5.7 1.7 16.4 7.2 8.0 TN MgL"' 75 960 294 351 75 544 289 276 109 960 336 427 TN:TP 22:1 108:1 50:1 54:1 22:1 108:1 49:1 52:1 28:1 103:1 53:1 55:1
51 Table 2 Pearson correlation matrix (*100) with sequential Bonferroni-adjusted probabilities of selected environmental variables for all sampling locations1 (n=l 13), lakes only2 (n=57), and ponds only3 (n=56). Bolded values indicate P < 0.01, italicizedValues indicate P < 0.05.
SA ]Ele v Depth MSSWT pH COND ORP CI Ca Mg K Si02 B Fe Mn Rb Sr DOC DIC CHLA PON POC TP Elevl -36 2 -35 3 -41 Depth 1 46 -4 2 18 -8 3 41 17 MSSWT 1 25 -31 -25 2 34 -30 -41 3 45 -35 23 pHl -35 9 -13 -52 2 -46 21 -8 -29 3 -30 -2 -38 -69 COND 1 -11 -59 -21 30 11 2 -19 -48 -3 32 -2 3 22 -77 -33 25 21 ORP1 24 39 9 -23 -15 -55 2 25 41 -13 -23 -3 -63 3 19 39 30 -18 -27 -44 Cll 13 -73 -16 41 -2 85 -48 2 9 -68 -14 51 -15 82 -53 3 35 -80 -16 33 6 88 -42 Cal -36 -29 -22 -2 30 80 -44 50 2 -41 -17 3 4 13 80 -50 50 3 -14 -47 -51 -14 46 78 -34 49 Mgl -4 -51 -28 31 11 86 -38 82 71 2 -9 -45 -9 35 4 84 -45 79 74 3 26 -64 -23 24 14 91 -32 85 69 Kl 7 -64 -18 33 -1 84 -49 84 64 81 2 2 -64 -7 37 -22 81 -58 86 63 77 3 29 -75 -30 27 13 87 -38 82 64 85 Si02 1 -38 40 -3 -27 20 -16 11 -40 16 -15 -48 2 -4/ 46 14 -35 35 -18 15 -47 15 -12 -56 3 -35 32 -11 -25 6 -18 9 -38 14 -23 -48 B 1 20 -72 -18 51 -4 78 -42 82 49 77 81 -22 2 15 -60 -18 61 -5 76 -47 87 48 83 85 -24 3 49 -85 -1 42 -3 81 -34 79 49 75 78 -22 Fel 7 1 -44 48 -29 -11 3 5 -30 -4 0 -20 5 2 31 -7 -50 38 -1 -18 35 14 -39 2 11 -44 8 3 12 8 8 52 -54 -22 -14 -11 -43 -25 -21 4 -5 Mnl -1 8 -30 31 -19 -7 -12 11 -20 1 4 -28 0 73 2 21 -3 -31 44 -6 -1 8 21 -12 20 26 -40 15 72 3 -7 20 3 15 -28 -25 -22 -2 -43 -22 -21 -18 -19 70 Rbl 47 -67 -22 67 -24 60 -25 70 27 64 78 -45 79 31 13 2 52 -60 -26 72 -31 47 -19 77 8 55 76 -60 76 44 39 3 62 -76 -2 62 -19 69 -22 65 39 67 77 -35 81 15 -13 Sri -2 -55 -22 40 5 85 -43 78 72 78 80 -14 79 -2 6 56 2 -11 -39 -7 44 8 83 -44 71 80 80 80 -10 76 -9 15 47 3 47 -79 -7 32 5 86 -35 85 59 78 82 -20 82 -11 -14 63 D0C1 23 -15 -20 59 -35 21 -18 29 4 26 31 -21 30 55 45 54 26 2 31 -3 -11 50 -27 13 -12 21 5 18 23 -25 21 43 55 49 18 3 40 -33 2 65 -46 26 -19 34 -5 27 35 -19 35 63 29 56 28 DIC 1 -39 -21 -24 -2 29 67 -37 36 92 61 57 14 39 -21 -10 24 63 9 2 -39 -13 1 2 16 68 -39 41 93 70 58 12 44 -26 5 12 77 10 3 -25 -35 -49 -15 44 62 -31 29 91 54 55 13 32 -37 -42 32 41 -3 HLA1 30 -23 -13 47 -35 6 -3 21 -10 8 28 -34 22 56 -49 44 16 50 -9 2 34 -20 -31 48 -18 5 7 24 -5 18 36 -48 36 62 53 60 20 51 3 3 37 -26 -1 48 -48 8 -10 18 -17 1 21 -22 13 57 49 35 13 51 -23 PON1 17 -22 -29 55 -26 7 -13 17 -18 9 10 -17 35 59 38 54 17 49 -13 51 2 39 -42 -31 58 2 31 -21 45 -2 34 32 -35 62 60 44 76 37 49 5 60 3 33 -29 2 57 -40 4 -22 16 -25 5 10 -22 39 62 37 58 17 60 -23 54 POC1 19 -23 -31 61 -28 8 -13 19 -18 12 12 -19 40 64 32 59 20 52 -11 53 98 2 35 -37 ' -35 57 5 22 -16 38 -9 26 23 -32 56 60 43 71 32 45 -3 51 97 3 32 -20 7 60 -47 -3 -11 7 -30 -3 2 -12 31 63 33 51 8 56 -28 53 96 TP1 13 -46 -31 61 -35 25 -28 44 12 38 51 -41 54 55 45 73 42 59 18 64 62 66 2 23 -48 -23 48 -17 36 -24 53 20 42 66 -61 55 44 47 76 44 51 30 65 64 55 3 38 -52 -4 67 -51 27 -26 37 -4 28 40 -31 52 57 36 70 34 64 -4 66 74 68 TNI 5 -39 -39 65 -35 44 -30 48 29 45 53 -28 56 56 40 76 49 69 31 61 63 65 85 2 21 -37 -29 58 -24 50 -26 62 39 55 71 -56 64 46 45 80 59 63 42 67 55 48 81 3 30 -50 -13 66 -48 35 -29 40 14 33 43 -16 51 56 29 74 37 75 14 64 72 69 86 Table 3 Regression matrix of selected significant environmental variables for all sampling locations (n=l 13), lakes only (n=57), and ponds only (n=56). All values indicate P< 0.001.
all lakes ponds r r2 r r2 r r2 MSSWT-TN 0.65 0.42 0.55 0.30 0.67 0.45 TP-TN 0.85 0.72 0.78 0.61 0.86 0.74 DOC-TN 0.69 0.48 0.55 0.30 0.73 0.53 PON-TN 0.68 0.46 0.46 0.21 0.71 0.50 POC-TN 0.65 0.42 0.39 0.15 0.67 0.45 CHLA-TN 0.61 0.38 0.63 0.40 0.65 0.43 Rb-TN 0.76 0.57 0.68 0.47 0.72 0.52
MSSWT-TP 0.61 0.38 0.44 0.19 0.68 0.46 Fe-TP 0.54 0.29 0.36 0.13 0.57 0.32 Rb-TP 0.70 0.49 0.67 0.45 0.69 0.48 DOC-TP 0.64 0.40 0.41 0.17 0.65 0.42 CHLA-TP 0.64 0.40 0.62 0.38 0.67 0.45 PON-TP 0.68 0.47 0.56 0.31 0.71 0.50 POC-TN 0.63 0.39 0.47 0.22 0.65 0.42
MSSWT- CHLA 0.47 0.22 0.44 0.19 0.50 0.25 Fe-CHLA 0.52 0.27 0.54 0.29 0.55 0.30 Mn-CHLA 0.46 0.21 0.49 0.24 0.49 0.24 DOC-CHLA 0.50 0.25 0.44 0.19 0.55 0.30 PON-CHLA 0.54 0.29 0.46 0.21 0.56 0.32 POC-CHLA 0.50 0.25 0.38 0.14 0.55 0.30
DOC-PON 0.56 0.31 0.33 0.11 0.62 0.38 POC-PON 0.97 0.93 0.97 0.94 0.96 0.92
MSSWT-DOC 0.59 0.34 0.46 0.21 0.62 0.39
COND-DIC 0.67 0.45 0.68 0.46 0.62 0.39 Ca-DIC 0.92 0.85 0.93 0.87 0.91 0.82 Mg-DIC 0.61 0.37 0.69 0.48 0.54 0.29 Sr-DIC 0.58 0.33 0.70 0.48 0.37 0.14
DOC-Rb 0.51 0.26 0.34 0.12 0.57 0.32 PON-Rb 0.59 0.34 0.60 0.36 0.56 0.31 MSSWT-Rb 0.66 0.43 0.69 0.48 0.62 0.38 B-Rb 0.79 0.62 0.74 0.55 0.81 0.66
54 Table 4 PCA scores for environmental and limnological variables (adjusted for variance) for the 113 sampling locations (n=113), lakes only (n=57), and ponds only (n=56).
All sites Metals included Lakes Ponds Axis 1 Axis 2 Axis 1 Axis 2 Axis 1 Axis 2 Axis 1 Axis 2 EIG. 0.45 0.22 0.36 0.18 0.35 0.17 0.32 0.23 Lat -0.20 -0.59 -0.33 -0.52 -0.28 -0.38 -0.39 -0.69 SA -0.02 0.41 0.12 0.25 0.16 0.62 0.54 0.15 Elev -0.61 -0.08 -0.61 0.13 -0.50 -0.10 -0.83 0.19 Zmax -0.31 -0.03 -0.32 -0.10 -0.10 -0.18 -0.02 0.26 MSSWT 0.45 0.55 0.59 0.39 0.53 0.27 0.58 0.48 pH -0.04 -0.68 -0.16 -0.60 -0.09 -0.44 -0.18 -0.67 COND 0.88 -0.31 0.77 -0.49 0.73 -0.50 0.80 -0.47 ORP -0.50 0.12 -0.42 0.18 -0.33 0.40 -0.41 0.02 CI 0.84 -0.03 0.81 -0.23 0.82 -0.07 0.84 -0.22 Ca 0.72 -0.60 0.52 -0.71 0.56 -0.74 0.40 -0.77 Mg 0.88 -0.26 0.80 -0.43 0.82 -0.37 0.78 -0.46 K 0.90 -0.05 0.88 -0.26 0.95 -0.02 0.85 -0.36
S04 0.77 -0.40 0.63 -0.54 0.54 -0.67 0.71 -0.43 DOC 0.47 0.58 0.53 0.49 0.31 0.44 0.58 0.51 DIC 0.66 -0.52 0.49 -0.59 0.58 -0.61 0.30 -0.70
Si02 -0.31 -0.49 -0.36 -0.30 -0.42 -0.69 -0.38 -0.11 CHLA 0.34 0.58 0.41 0.59 0.46 0.58 0.37 0.64
NH3 0.49 0.34 0.51 0.16 0.31 0.14 0.54 0.17 TP 0.65 0.60 0.77 0.47 0.75 0.46 0.73 0.53 TN 0.74 0.52 0.81 0.40 0.83 0.33 0.74 0.47 Al 0.22 0.63 0.09 0.42 0.27 0.73 As 0.80 0.30 0.81 0.21 0.77 0.46 B 0.84 -0.24 0.86 -0.16 0.87 -0.24 Ba 0.63 0.05 0.69 -0.01 0.49 0.14 Cu 0.11 -0.01 0.24 -0.10 -0.13 -0.07 Fe 0.29 0.75 0.16 0.72 0.19 0.84 Mn 0.28 0.59 0.30 0.52 0.10 0.70 Ni 0.46 0.22 0.62 0.00 0.20 0.49 Pb 0.24 0.27 0.43 0.25 -0.03 0.21 Rb 0.83 0.08 0.77 0.39 0.88 -0.03 Sr 0.83 -0.37 0.85 -0.40 0.83 -0.32 U 0.19 -0.27 0.23 0.03 0.10 -0.43 Zn 0.12 0.49 0.02 -0.09 -0.04 0.59 Table 5 Canonical Variates Analysis (CVA) regression coefficients, associated ^-values, and Interset correlations for significantly different limnological variables for the 97 sampling locations defined by region (sites without metals removed). Significant values to an associated axis are bolded (P< 0.05).
Regression coefficient t-values Interset correlation AXl AX2 AXl AX2 AXl AX2
dSSWT 0.52 0.39 2.28 1.57 0.66 0.24 pH -0.11 -0.68 -0.63 -3.44 -0.26 -0.51 COND 0.16 -0.84 0.45 -2.21 0.57 -0.54 Mg -0.21 0.45 -0.63 1.25 0.63 -0.36 Al -0.24 0.37 -1.22 1.76 0.18 0.37 As -0.04 -0.22 -0.15 -0.73 0.53 -0.17 B 1.51 -0.43 4.66 -1.23 0.77 -0.35 Ba -0.50 -0.78 -2.49 -3.59 0.27 -0.44 Fe -0.37 0.63 -1.45 2.26 0.21 0.40 Ni 0.45 -0.06 2.27 -0.27 0.18 -0.18 Pb -0.02 0.07 -0.15 0.37 0.21 0.19 Rb 2.08 0.30 6.05 0.81 0.92 -0.02 S04 -0.78 -0.24 -2.82 -0.78 0.43 -0.45 DOC 0.01 0.57 0.06 2.77 0.48 0.32 DIC -0.10 0.20 -0.45 0.80 0.26 -0.49 CHLA 0.06 0.47 0.27 2.17 0.36 0.33 NH3 -0.32 0.07 -1.84 0.40 0.33 -0.12 TP 0.33 -0.37 0.97 -1.00 0.67 0.04 TN 0.40 -0.44 1.00 -1.01 0.70 0.07
56 Figures
Fig. 1: Location of sampling areas identified by region, a) Central Kivalliq, b) Western
Hudson Bay - Arviat area, c) Western Hudson Bay - Rankin Inlet area d) Baffin Region
- Iqaluit area.
Fig. 2: Trends in selected environmental variables across identified regions. Sampling locations are organized by significantly different region, with lakes on the left and pond samples on the right, a) MSSWT, b) CHLA c) pH, d) DOC, e) DIC, f) Rb, g) POC, h)
PON,i)TP, andj)TN.
Fig. 3: Principal Components Analysis of significant environmental indicators (excluding metals) for the 113 sampling locations. Limnological variables run actively are indicated by solid lines, while physical and geographical variables run passively are indicated by dotted lines. Region classification; • Tree-line, • Hudson Bay, A Central, and • Baffin.
Ponds are indicated by gray fill.
Fig. 4: Principal Components Analysis of significant environmental indicators (including metals) for the 97 sampling locations. Sites and variables are classified as in Fig 3.
Fig. 5: Canonical Variate Analysis of significant environmental indicators (including metals) for the 97 sampling locations. Sites classified as in Fig 3.
57 era Qikiqtaaluk ' 70N \ (Baffin) Region
dl
\ Quebec 2 4Km J
00 a. b. * 3 • X . •j- il r • i TT L T ^ ?5$ TTB •¥? 0 •
• d. • • * 1 • * Hr^ X s 9 ^#rl * * B T ^ 1
e. f. • J Jt r * a 6 T 1 e O ' 9a ^ Q
• L j- MO 15 • 1—1 X i MO *•. • & • MO X T • A C X T X 9 • MO • ^Q 1 i • 5 ? * ?£ 1 Hwktn GMM BMft* &** tMsm Cwtari Mfa Lakes Ponds Lakes Ponds
Fig. 2. ^ i
1 ®@ ! CHLA DOC TP A H A y* , /MS$WT* TN • ® Area,
% • ^A/s/xm
ORP •>• m _ Zl pa a a • JP 1 m c/ B H Elev Depth A * >2 A^ x^^^^^ • A A/ ^ ^W^\""^—-^_^ • /•* j;1 V • /^ / f! j o Z NV. • ^k COND ** / / 9 \. i^ u N^v • S04 S*02 ^ ; N^/c • 1: ^< •! Ca < •a .
1 -1.0 PCA axis 1=44.7% 1.0
Fig. 3
60 ©
* */ /A /i! n^^'^S^P^"'"* ^j? N. '•••-ooa^ • ^^^^^f re£B/ _ *4 * / ^ K &02 * \4 / • / k Sr • • ," / Mg Q \\ 1^ La/* r/ V* SCW COND pi/ i a VDIC Ca
1 —i 1 -1.0 PCA axis 1=34.5% 1.0
Fig. 4
61 • @ ® ! • © ® • • _• • ® 4 <§>• • ! ^ AL CHLA • Fel DOC 4f/ * MSSWT . s •^ • 1 / /tfyz^M • If// // TN • Rb •
v A __——"—-»• _ op • • ^NH3 • TP • 1 _ A % A j ! VV SKSvI • ••• A A A ^•wN s B " A ^ A AA A V* Mg A A A A / A A * A \\S04\ A AA \ _..-. • A A pH \ DIC riCOND Ba
'
1 1 1 1 -6 CVAaxis 1=30.3%
Fig. 5
62 CHAPTER 3: The distribution of the Chironomidae (Insecta: Diptera) along multiple environmental gradients in lakes and ponds of the eastern Canadian Arctic.
Medeiros, Andrew, S., and Quinlan, Roberto
York University, Department of Biology, 4700 Keele St, Toronto, Ontario, M3J1P3.
Corresponding Author: [email protected] Phone: 416 736 2100 x 40076
Fax: 416 736 5698
Manuscript Accepted for Publication in
the Journal of Canadian Fisheries and Aquatic Sciences, August 2011 issue.
63 Abstract
An examination of the Chironomidae, the dominant aquatic invertebrate taxa found in Arctic lakes and ponds, was conducted to determine the environmental gradients that may limit their geographical distribution in the eastern Canadian Arctic. Subfossil chironomid head capsules, comprising 86 taxa, were sampled from surficial sediments of
63 lakes that spanned from tree line (northern Manitoba) across multiple regions within the eastern Canadian Arctic. Water chemistry and environmental data were then compared with chironomid assemblages using multivariate analysis. The distribution of chironomids was found to primarily follow a temperature gradient, but additional significant relationships were also found along a nutrient/productivity gradient. Several species of the Tribe Chironomini, which generally represent warm-water adapted taxa, were also found far beyond tree line in the southern Kivalliq region of Nunavut, indicating a more northerly range than previously known. While temperature and trophic status were found to strongly influence the distribution of some taxa, partially constrained gradient analysis indicates that specific chironomid taxa could be used to indicate a primary response to climate regardless of trophic status. This may allow for more holistic inferences of how aquatic communities may respond to climate change as the range of temperature dependant species expand into Arctic systems.
64 Introduction
Changes in climate are expected to have cascading effects on northern aquatic invertebrate communities as aquatic food-web structure is altered, owing to a reduction of ice-cover and subsequent expansion of the localized growing season for algal communities (Rouse et al. 1997). Bio-indicators that respond to major limnological changes, such as temperature, dissolved oxygen, and nutrients, can be used to generate relationships between localized environmental conditions and subsequent biotic community structure. However, the use of these indicators to make interpretations about the ecological response of aquatic systems to long-term changes in climate requires baseline knowledge of the relationship between aquatic communities and their limnological environment.
The family Chironomidae (Insecta: Diptera) is one of the most dominant and most diverse groups in Arctic aquatic systems (Oliver and Dillon 1997). There are four basic stages in the life cycle of chironomids: egg, larva, pupa, and adult. While the adult terrestrial stage of the chironomid life cycle is short, the four-stage aquatic larval phase can persist for years until conditions are favourable for pupation and emergence (Butler
1982). The emergence period depends on the conditions of the environment, including particular heat-sun thresholds, oxygen concentrations of the water, water levels, food resources, and temperature (Danks 1971; Pinder 1995). Many species of chironomids are specially adapted for specific habitats and environmental conditions (Pinder 1995), and have been used as indicators of climatic conditions (Walker et al. 1991), trophic status
65 (Brooks et al. 2001), dissolved oxygen (Quinlan and Smol, 2001a), biomonitoring
(Medeiros et al. 2011), and productivity (Broderson and Lindegaard 1999a). The chitinized remains of the four larval stages are well preserved in lake sediments (Walker
2003), and the examination of the diversity and abundance of their sub-fossil remains in surficial (0-1.0 cm) lake sediments allows for an interpretation of the environmental conditions conducive to their survival and persistence in a given system.
Substantial portions of the eastern Canadian Arctic are under sampled for
Chironomidae distributions. For example, the Kivalliq region of Nunavut, formerly the district of Keewatin, encompasses 445 109 km2 (Statistics Canada 2006), roughly the size of Sweden (Figure 1). Several lakes were sampled by Porinchu et al. (2009) along a transect from the mainland Kitikmeot region, and far north-western area of the Kivalliq
(Figure 1). In addition, there have been a few faunal surveys of aquatic invertebrates
(e.g. Giberson et al. 2007). However, Walker's (1990) examination of four lakes for sub fossil chironomid assemblages may represent the single instance of this type of lake study within this large mainland area. The lack of recent research on the Chironomidae within the eastern Canadian Arctic represents a gap in knowledge of community composition in an area where climate warming may result in range extensions of several southern warm- water adapted taxa. A warming climate would allow for the establishment and proliferation of these taxa, and the ecological range of several species defined by warm- water temperature thresholds for survival, to extend into tundra environments. Thus, the recent appearance of warm-water adapted species should be reflected in lakes just north of their previous historical northern limits. However, the limited baseline knowledge of
66 Arctic chironomid communities challenges our ability to understand the ecological gradients that may define the range of many species, especially when the taxonomy of morphotypes is limited. Recent work with DNA bar-coding has made progress in identifying specific species and improving taxonomic keys. For example, Stur and Ekrem
(2011) noted several species of Tanytarsini that have long been grouped together, or misidentified owing to similar larval traits.
As the topography, geologic history, landscape characteristics, and climate can differ greatly across regions within the Arctic, the limnology and invertebrate diversity of aquatic systems likely also differ. The relationships between ecological interactions (e.g. competition and predation) and chironomid distributions in Arctic systems are mostly unknown. For example, the influence of bird populations on the limnology of lakes and ponds in nesting areas may significantly alter local nutrient regimes (Cote et al. 2010), and cause shifts in the abundance and diversity of local chironomid assemblages (Frisch et al. 2007; Michelutti et al. 2011). As Brodersen and Anderson (2002) indicated that the temperature optima for chironomid taxa and trophic variables were strongly correlated, it may be difficult to separate the influence of productivity from the influence of temperature on chironomids. Kernan et al. (2009) also found that regional differences in a pan-European study of several indicators, including midge distributions, could not solely be explained by environmental variables, suggesting that biogeographic influences
(e.g. dispersion and colonization) may also be important. In addition, while a gradient of richness and diversity in invertebrate community structure is well documented as one moves along a transect from lakes and ponds within areas defined by lower mean-July
67 temperature to those characterized by a higher mean-July temperature (Walker et al.
1997; Barley et al. 2006; Porinchu et al. 2009), there have been fewer studies examining possible gradients in available nutrients and localized environmental conditions that may influence the community structure within the same geographic area (Brodersen and
Anderson 2002).
In order to address these issues we sought to answer the following questions:
What is the community composition of the Chironomidae in lakes and ponds of the eastern Canadian Arctic? What environmental factors govern the distribution of these species in a particular environment? Could localized differences in temperature, nutrient availability, depth, or other environmental characteristics influence the chironomid community structure in lakes and ponds within the same geographical area? Can common environmental influences on chironomid distributions be found across multiple regions?
Therefore, in order to achieve a better understanding of the community structure of Chironomidae populations within lakes and ponds across several communities in
Nunavut, an examination of lakes and ponds was conducted within several under- sampled areas in the eastern Canadian Arctic. This analysis focused on the environmental factors that may influence the diversity and abundance of chironomid assemblages across several regions within Nunavut, as well as the localized environmental conditions that govern the range of chironomid taxa in the eastern Canadian Arctic.
68 Materials and Methods
Study Areas
Lakes and ponds near settlements were sampled across several regions, along a transect beginning at tree-line (Northern Manitoba), extending through the Kivalliq region towards the north-eastern islands within the Baffin region of Nunavut (Figure 1;
Table 1). These lakes were also sampled in groups of varying landscape and environmental characteristics, concentrated in the vicinity of several towns within the
Kivalliq region (Arviat, Baker Lake, Rankin Inlet, and Repulse bay).
The position of the southern-most town of Arviat (central-west Hudson's Bay,
61° 7' 0" N, 94° 3' 0" W), -260 km north of Churchill, Manitoba (58° 46' 0" N, 94° 10' 0"
W), makes it an ideal location to determine if the ranges of several species of the Tribe
Chironomini extend into the lakes and ponds north of treeline within southern Nunavut.
Several Chironomini species have been documented in the vicinity of the Town of
Churchill, Manitoba (R. Quinlan, unpublished data), but there are no records further north. The landscape of Arivat is relatively flat (mean elevation of 12 meters above sea level) and dominated by matted compact cushions of prickly saxifraga (Saxifraga tricuspidata) and moss campion (Silene acaulis). These cushions were intermixed with clumps of dwarf fireweed (Chamerion latifolium). Large stands of Salix spp. shrubs, approximately 1.0 - 1.5 m in height, were also frequently present in catchment areas.
Dominant shrubs found throughout the vicinity of Arviat contrasted with other areas sampled in the Kivalliq region, where the number and height of shrubs observed were less prominent.
69 The town of Rankin Inlet (62° 48' 41" N, 92° 6' 57" W), 200 km northeast of
Arviat, has extensive all-terrain vehicle trails allowing for the examination of several lakes across a large gradient of landscape characteristics (e.g. varying depth, elevation, catchment vegetation, and underlying hydrology). While lakes and ponds sampled within the vicinity of Arviat and Rankin Inlet were in valleys of low elevation (<50 m), lakes in other regions were generally found isolated at higher elevations (100-200 m). These included Iqaluit (63° 45' 8" N, 68° 33' 50" W), Clyde River (70° 27' 0" N, 68° 34' 0" W), and Repulse Bay (66° 31' 17" N, 86° 13' 29" W). Catchment vegetation for these areas were primarily represented by dwarf shrubs of alpine bearberry (Arctostaphylos alpina), mountain cranberry (Vaccinium vitis-idaea ssp. minus), and Arctic white heather
(Cassiope tetragona). Poaceae commonly found throughout catchment areas included species of wildrye (Elymus spp.), bluegrass (Poa spp.), and fescue (Festuca spp.). Dwarf fireweed (Chamerion latifolium) was also common in disturbed areas along all-terrain vehicle trails and roads.
Field sampling
Sediment and water samples for 57 lakes were collected during the ice-free seasons between June and August in each of the years from 2006-2009. Samples consisted of 16 lakes from the vicinity of Rankin Inlet, 10 from Arviat, 9 from the
Churchill vicinity, 7 from Baker Lake, 6 from Iqaluit, 6 from Repulse Bay, and 3 from
Clyde River. Of these 57 lakes, replicate cores were recovered from the approximate center and/or deepest portion of lakes using a Glew (1991) mini-corer (3.8 cm inner diameter) and/or Uwitec (http://www.uwitec.at) gravity corer (8.4 cm inner diameter)
70 deployed from a collapsible PakCanoe, inflatable Zodiac ® boat, or float-plane.
Sediment cores were extruded in the field at 0.5 cm resolution, stored in Whirl-Pak ® bags, and kept cool (4 °C) and dark until processed in the laboratory. Sediment and water samples for 11 lakes were also collected by S. Finkelstein (Dept. Geography, University of Toronto) between July 27, 2008 and July 31, 2008 within Sirmilik National Park (4
Northern Sirmilik Baffin Lakes from the Qorbignaluk headlands, 7 South/South-western
Bylot Island Lakes) and included in our dataset.
During sediment core collection, physicochemical variables were measured using an YSI600QS multi-parameter probe, including mid-summer surface water temperature
(MSSWT, measured at 0.5-1.0 m below surface of water), bottom water temperature
(measured 1.0 m above sediments at a mid-basin location), specific conductance, pH, and dissolved oxygen. Depth was measured with a depth sounder and the GPS location of each mid-basin coring location was taken with a Garmin 1200XL GPS device.
Epilimnetic water samples were collected at 0.5 m below the water surface in pre-cleaned polyethylene bottles and immediately treated in the field following the protocols outlined in the Analytical Methods Manual of Environment Canada (Environment Canada 1994).
Water samples for each lake were analysed by the National Laboratory for Environmental
Testing at the Canadian Centre for Inland Waters, Burlington, Canada (Table 1).
While air temperature has been found to be a significant environmental variable for directing Chironomidae community distributions (e.g. Walker et al. 1997; Barley et al. 2006; Porinchu et al. 2009), if lakes are within a relatively close geographical proximity then the use of air temperature (or a reconstructed air temperature estimate)
71 derived from interpolations amongst meteorological stations in the region is problematic.
This problem is particularly acute if there is a lack of a substantive altitudinal gradient
amongst sites from the same regional locale. Likewise, using air temperature, as opposed
to water temperature, does not incorporate catchment-specific characteristics that may
influence the water temperature of lakes and ponds, such as water clarity and light
penetration (e.g. catchment vegetation influencing allochthonous DOC inputs), water
source (e.g. glacially derived), basin morphometry (e.g. surface area (SA) and depth), and
local topography. These localized parameters can determine the total volume of a water
body and exposure to wind and solar insolation, which may influence water-body
response to climate change. For example, neighbouring water bodies with different local
topography may have substantively different responses to climate change (Keatley et al.
2008). In order to provide a control for spot measurements of temperature and
environmental variables, several lakes were sampled multiple times over the 4-year study
(June - August, 2006 to 2009). These duplicate samples, in both water chemistry and
chironomids from surficial sediments, were plotted passively into the ordination analysis
as a control, but removed from ordination diagrams owing to the repetitive nature of
these samples obscuring data-points.
Laboratory Analysis
Surface sediments (1-4 grams wet mass) were sub-sampled and washed through
nested 212 and 106 urn mesh sieves with DH2O. Retained residues were then washed with
95% ethanol in preparation for sifting. Residues were sifted using a stereomicroscope at
30-40x magnification, and subfossil chironomid headcapsules extracted with the use of
72 fine forceps. Specimens were permanently mounted on glass microscope slides in
Entellan® mounting medium for identification.
For each lake a minimum of 50 head capsules were enumerated. If the total number of head capsules found was fewer than 50, more sediment was sub-sampled until the total identified remains were greater than 50 (Quinlan and Smol 2001b). Surficial sediments from each lake represented the surface-water interval (0 - 0.5 cm sediment depth), with the exception of A3, where 0-1 cm of sediment was extracted because of low numbers within the first 0-0.5 cm interval.
Specimen identifications were made at 400-1000x magnification to the lowest taxonomic resolution possible based on Saether (1976), Oliver and Roussel (1983),
Wiederholm (1983), Walker (1988), Uutala (1990), Epler (2001), Rieradevall and Brooks
(2001), Brooks et al. (2007), and Walker (2007). The genus Heterotrissocladius and
Psectrocladius (Psectrocladius) were identified to species (e.g. H. maeaeri-type 1, H. maeaeri-type 2, H. grimshawi, H. marcidus, H. subpilosus; Psectrocladius
(Psectrocladius) sordidellus, P. (P.) barbimanus, P. (P.) psilopterus, and P. (P.) limbatellus gr.) but conservatively grouped for analysis because of a lack of clear demarcations in different character states of diagnostic taxonomic features.
There were three distinct types of Zalutschia found in our dataset (Figure 2). We feel that, from visual inspection, Zalutschia sp. B as described by Brooks et al. (2007) is a different species than the Zalutschia type B identified by Barley et al. (2006). Therefore we have referred to Zalutschia type B in Barley et al. (2006) as Zalutschia sp. C in our study. The diagnosis of this specimen is based on the wider asymmetrical median teeth,
73 which have two outwardly pointing peaks. The first lateral tooth is distinctly separate
from the median tooth, and half the size of the second lateral tooth. The second to sixth
lateral teeth are of comparable size and progressively subequal. The ventromental plates
curve upwards at the apex, but do not obstruct any of the lateral teeth. We did not find
any specimens of Zalutschia sp. B as described by Brooks et al. (2007). The diagnosis
for Zalutschia lingulata pauca is sometimes designated Zalutschia sp. A, however, we feel that this taxon has marked differences in diagnostic features from Brooks et al.
(2007) and Barley et al. (2006) that are available in Saether (1976) and pictured in
Wiederholm (1983). As the description of Zalutschia lingulata pauca is a European type,
we have designated our specimen as Zalutschia nr. lingulata pauca. The first lateral tooth
of Zalutschia lingulata pauca could be described as an accessory tooth as it is attached to
the median teeth at the base, is much thinner and less pronounced than the other laterals,
and can vary from equal in height to the second lateral to slightly subequal. The second to
sixth lateral teeth gradually diminish in size. The ventromental plates are elongated
compared to Zalutschia sp. C, can partially obscure the fourth to sixth laterals, and
markedly curve upwards at the apex.
Data Screening
Five lakes from the northern Baffin region missing primary environmental
variables (due to lack of processed water samples), were removed from analysis. Further
variables that had concentrations below the detection limit in a majority of sites sampled
(e.g. NO3, NO2, and SRP) were deleted from the dataset. If variables were below the
74 detection limit in fewer than half the study lakes, half the detection limit was substituted for use in numerical analysis.
Environmental and water chemistry variables were normalized (Table 1) and checked with a Kolmogorov-Smirnov test of normality. Testing for outliers was also conducted for both species and environmental data, where the 95% confidence limits were calculated for the sample scores on the first two axes of a principal components analysis (PCA) of environmental data and a detrended correspondence analysis (DCA) of species data (Gauch 1982). No sites within our data set were considered outliers as no sites had sample scores that exceeded the 95% confidence limit for both the PCA and the
DCA. Taxa are presented as square root transformed percent abundances, calculated as the percentage of total identifiable midges.
Numerical Analyses
Relationships between environmental variables and chironomid community composition were examined with the use of ordinations performed using CANOCO v4.53
(ter Braak and Smilauer 1998). A DCA, a variation of reciprocal averaging
(correspondence analysis), was conducted (with detrending by segments, square-root transformation of species abundance, and down-weighting of rare taxa) in order to observe the relationship between species and environmental data in ordinal space. The
DCA was also used to determine the gradient lengths of species composition along the first two axes in order to select between linear- or unimodal-based ordinations.
Environmental variables that explained a significant (P < 0.05) direction of variation in species data were determined through direct gradient analyses. The
75 significance of each environmental variable was then tested in an ordination singly constrained to each variable, using both redundancy analysis (RDA; linear model) and canonical correspondence analysis (CCA; unimodal model), performed with CANOCO using 999 unrestricted Monte Carlo permutations (reduced model). The selection of environmental variables was conducted with a backwards selection process for both RDA and CCA techniques and included only environmental variables that were found to be significant (P < 0.05) in singly constrained ordinations. The collinearity of environmental variables was minimized during this process, where variables with the highest variance inflation factor (VIF) were removed sequentially. The backwards selection procedure was repeated until all remaining environmental variables had VIFs <
10.
In order to examine whether there were significant differences between site locations based on the relative abundances of chironomid species, a two-way indicator species analysis (TWINSPAN; Hill 1979) was conducted. This analysis ordinates the site locations using reciprocal averaging, and creates a dichotomy using a calculated centroid line to make binomial divisions and create a hierarchical classification of sites (Hill
1979). Pseudo-species cut levels were defined as 0%, 2%, 5%, 10%, and 20% on the relative abundance data of chironomid species.
Each TWINSPAN division delineated sample locations into groups based on the chironomid community composition at each site. Tests for significant (P < 0.05) differences between groups was carried out with the use of parametric ANOVAs that compared the means of DCA axis scores for samples within each TWINSPAN group.
76 Significant differences in environmental variables amongst the TWINSPAN groups were then tested with subsequent parametric ANOVAs, or Kruskal-Wallis one-way ANOVA for those variables that violated statistical assumptions of normality. This elucidated the environmental preferences of taxa based on the placement of a location within a particular TWINSPAN division.
The relationship between selected taxa and site location was visualized using C2 software (Juggins 2003). The relative abundance of species was arranged in rank order, where RDA axis 1 species scores were singly constrained to MSSWT, with lakes ordered in terms of latitude from south to north. Attribute plots were created using CanoDraw
4.12 (ter Braak and Smilauer 1998), where relationships between specific taxa and environmental variables were based on significant (P < 0.05) correlations when examined by a Spearman's-rank correlation matrix with Bonferonni-corrected P-values. In addition, a series of partially constrained ordinations were run to examine relationships between primary gradients independently of each other. The RDA ordination was run in
CANOCO with MSSWT as the primary explanatory variable for chironomid distributions, with TN set as a covariable (and vice versa). Species that were significant to either constrained gradient were retained.
Results
Over 6500 head capsules were extracted and identified from the surface sediment intervals (0 - 0.5 cm), with an average of 102 head capsules per lake (median = 79.5; range 50-373). Eighty-six taxa were represented within the 63-lake dataset (Figure 3).
77 The range in surface temperature for all lakes sampled was from 4.9 - 18.3 °C (a gradient of 13.4 °C) and the range in mid-basin depth was 0.30 to 20.6 m (a gradient of 20.3 m).
All lakes were observed to be polymictic at the time of sampling (difference in surface temperature and bottom temperature < 1.0 °C).
Ordination Analyses
The gradient length of species data, determined from the DCA, were 2.6 and 2.1 standard deviation units for axis 1 and 2, respectively. These are representative of axes of intermediate lengths, therefore both linear- and unimodal-based ordination techniques were used. Ordinations constrained to each environmental variable identified the following significant (P < 0.05) environmental variables: mid-summer surface water temperature (MSSWT), bottom water temperature, depth, pH, total phosphorus (TP), total nitrogen (TN), dissolved organic carbon (DOC), dissolved inorganic carbon (DIC), silicon dioxide (Si02), potassium (K), sodium (NA), chloride (CI), magnesium (Mg), calcium (Ca), sulphate (S04), ammonia (NH3), and conductivity (COND) at 25 °C. The backwards selection process eliminated Ca, Mg, K, Na, COND, and bottom water temperature.
The species-environment correlation between species axis 1 and canonical axis 1 was similar in the CCA (0.919) as in the RDA (0.929); however, total species variance explained was higher in the RDA (48.3% in RDA, 36.6% in CCA). Thus, further analysis was based on linear RDA ordinations rather than unimodal-based ordinations
(CCA). The first two RDA axes accounted for 30.0% of the species variance (Table 2),
61.4% of the species-environment relation, and each axis explained a significant amount
78 of species variation (P < 0.001). Relationships between these environmental variables and the species data set were examined through canonical correlation coefficients of variables to the first two axes, their associated t-test value to each axis, and the interset correlations of the environmental variables (Table 3). Depth, mid-summer surface water temperature (MSSWT), CI, DIC, DOC, surface area (SA), and Si02 were significantly correlated (P < 0.01) with the first RDA axis. Depth, TN, CI, and Si02 were significantly correlated (P < 0.01) with the second RDA axis.
Environmental Distributions
A TWINSPAN analysis was conducted to examine relationships that may describe the environmental preferences of several taxa. The TWINSPAN analysis of the chironomid communities in the lake dataset indicated a significant division between lakes in colder regions from those in warmer regions (group C, Figure 4). This was characterized by a significant (P < 0.05) separation based on cold-water
(Heterotrissocladius gr., Zalutschia sp. C) and warm-water adapted taxa (Dicrotendipes,
Cladotanytarsus mancus gr.), lower TP, TN, CHLA, temperature, and higher depth
(Figure 5). The second TWINSPAN division was also significant, and separated sites based on SA, TN, CHLA, DIC, COND, and depth (group B, Figure 5). This was characterized by shallow sites in warmer regions represented by Chironomus anthracinus-type, and productive, deeper, larger lakes in warmer regions represented by
Corynocera ambigua, Arctopelopia gr., and Zavrelia /Stempellinella gr. (group A, Figure
4). Sites grouped according to these TWINSPAN divisions had significant differences (t- tests, P < 0.05) in either DCA axis 1 or axis 2 sample scores.
79 A RDA biplot of sites (Figure 6) shows that RDA axis 1 separates colder, deeper sites, with lower TN (e.g. CR04 and IQ05), from warmer, shallow sites with higher TN
(e.g. AV02 and RB01). Cold-water adapted species (e.g. Pseudodiamesa,
Hydrobaenus/Oliveridia gr., Abiskomyia, and Micropsectra type R) are found opposite of high MSSWT at the bottom right quadrant of the ordination, whereas warm-water adapted species (e.g. Polypedilum, Cryptochironomus, Endochironomus, Cladopelma, and Glyptotendipes) are found clustered towards the top left quadrant (Figure 7). Thus, for numerous subfossil chironomid taxa, distributions along a climate gradient agree with previous studies from the Arctic (e.g. Walker et al. 1997; Gajewski et al. 2005; Porinchu et al. 2009).
Redundancy analysis axis 2 primarily separated areas with warm, shallow (2-5 m), large lakes with higher available nutrients (e.g. Churchill locations), higher productivity, and higher DIC from those that are colder, deeper, with lower available nutrients, lower productivity, lower DIC, and a smaller SA (e.g. RB02 and RB17; Figure
6). The separation of lakes and ponds by RDA axis 2 were represented by the second
TWINSPAN division that separated warmer, shallow lakes and ponds with higher available nutrients (Figure 4, group A) from larger lakes with higher productivity (Figure
4, group B). Species that followed this pattern oriented along RDA axis 2 were
Microtendipes, Zavrelia/Stempellinella gr., Corynocera ambigua, Psectrocladius
(Monopsectrocladius) gr. (warmer, deeper, and larger lakes with higher productivity), and Limnophyes, Dicrotendipes, Procladius, Corynoneura arctica (colder, shallow, smaller lakes with lower productivity).
80 While specific taxa were found to have distinct relationships with both MSSWT and TN, partially constrained ordinations revealed that several taxa may have relationships with one variable that were independent of the other. In a partially constrained ordination with MSSWT as the sole explanatory gradient and TN as a covariable, strong correlations were found with MSSWT independent of TN (Figure 8a).
In time series data sets, sd either monitoring data or using paleoecological data in a biostratigraphic analysis, changes in these taxa may be indicative of changes that are tightly linked to climate (e.g. water temperature, ice-cover phenology, etc.) regardless of a particular trophic status. These included cold-water indicator taxa such as
Hydrobaenus, Micropsectra type R, and Zalutschia nr. lingulata pauca, and warm-water indicator taxa such as Parachironomus and Polypedilum. In a partially constrained ordination with TN as the sole explanatory gradient and MSSWT as a covariable, there were several taxa that still had a strong correlation with a TN gradient (Figure 8b). These included Cricotopus bicinctus, Dicrotendipes, Eukiefferiella/Tventia gr., Micropsectra contracta, Mesocricotopus, Micropsectra insignilobus, Micropsectra padulla, and
Parakiefferiella sp. B.
Discussion
Influences of temperature and depth on Chironomidae
Chironomids are able to survive and persist across long gradients of environmental conditions, such as temperature, oxygen, productivity, and salinity, which has allowed for a global distribution (Porinchu and MacDonald 2003). While the composition of the chironomid assemblages in a particular habitat depends on both the
81 terrestrial and aquatic environment (Pinder 1995), the relatively short duration of the adult stage compared with the larval stage, along with a general lack of feeding in the adult stage, suggests that environmental controls on the larval stage may be more important in determining range and distributions. Initial qualitative descriptions of chironomid species and the presence/absence and/or abundance of larvae in different lake types (Thienemann 1915; Brundin 1951; Saether 1979; Wiederholm 1980) were used to develop lake typology classification schemes, and have been used to relate assemblages to trophic status and to detect environmental stress (Johnson 1995; Wiederholm 1984;
Clements et al. 2000).
Within Arctic aquatic systems, chironomid assemblages have been identified to be especially sensitive to perturbations in environmental and landscape conditions
(Walker 2001). Walker et al. (1991) found that several warm-adapted taxa are primarily found in southern, temperate habitats; whereas cold-adapted taxa are found in northern, cold environments. Thus, the distribution and range of many Arctic chironomid taxa were thought to be primarily temperature dependent (Walker et al. 1991). This study's collection of surface sediment samples from across the eastern Canadian Arctic allowed for the representation of chironomid community structure across several regions, and across several environmental gradients within those regions. As expected, deep lakes in colder regions of higher latitude were found to be primarily represented by chironomid communities of the cold-water adapted Diamesinae and Orthocladiinae sub-families (e.g
Abiskoymia, Pseudodiamesa, Heterotrissocladius gr., and Hydrobaenus/Oliveridia gr.).
Lakes within these colder regions had taxa that primarily clustered in TWINSPAN group
82 C, and represented colder and deeper lakes. For example, lakes sampled within the
Iqaluit area (south-central Baffin Island), were predominantly represented by the cold- water adapted taxon Heterotrissocladius gr. This is consistent with Walker and
MacDonald (1995), where Heterotrissocladius, Parakiefferiella cf. triquetra, and
Protanypus were associated with the largest and deepest lakes in the western Canadian
Arctic. The association of these cold-water adapted taxa (e.g. Abiskoymia) and temperature (Figure 9A) are often used as a proxy indicator of periods of cold climate in paleolimnological studies (Francis et al. 2006).
In contrast, lakes and ponds within the warmer and more productive areas of
Arviat, Rankin Inlet (with the exception of RA08), Repulse Bay, and Bylot Island contained species that are considered warm-water adapted taxa of the Tribe Chironomini
(Oliver and Roussel 1983, Brooks et al. 2007) such as: Microtendipes, Dictotendipes,
Cladopelma, Polypedilum, Cryptochironomus, Parachironomus, Endochironomus, and
Glyptotendipes. These sites also had an absence of Heterotrissocladius remains.
Occurrences of warm-water adapted taxa in Arctic lakes have been documented in a few cases (e.g. Iceland, Langdon et al. 2008; Kiktitmeot tree line region, Porinchu et al.
2009), but have not previously been found in lakes within colder regions of the eastern
Canadian Arctic north of tree line (e.g. Kivalliq, Baffin regions). For example,
Cladopelma were found in nine lakes in our dataset, and ranged in abundance from 0-5% in lakes that were from 0.5 to 7 m deep. This is noteworthy as Cladopelma, a warm- stenotherm, has a Holarctic distribution in warm, low-latitude lakes (Walker et al. 1991;
Brooks et al. 2007), and in temperate systems is found primarily in shallow polymictic
83 lakes, with very rare occurrences in deeper, stratified lakes (R. Quinlan, unpublished data). Cladopelma was also found in shallow ponds within the vicinity of Iqaluit (IQ02), along with Dicrotendipes, Chironomus anthracinus-type, and a high abundance of
Tanytarsus lugens compared to other lakes sampled within the Baffin region. In addition, six lakes sampled within the vicinity of Iqaluit also contained Chaoborus mandibles, which were not found in sediments from any other region sampled. The presence of adult
Chaoborus in the southern Baffin region was noted in Lamontagne et al. (1994), but
Chaoborus are traditionally strongly associated with temperate systems, and were thought to be limited by tree line (Walker et al. 1997).
Although several lakes that were represented by cold-water adapted taxa were observed to separate primarily based on MSSWT (e.g. TWINSPAN group C), several lakes did not follow a simple latitude/temperature/depth relationship. For example, lakes in the treeline region were relatively shallow (2 - 5 m), but colder than similar lakes -100 km north in the Arviat area. Lakes clustered around this tree line area within the vicinity of Churchill had higher abundances of cold-water adapted taxa, such as
Heterotrissocladius gr. and Micropsectra insignilobus-type. Likewise, shallow ponds in the colder northern and central areas of Repulse Bay and Baker Lake contained several taxa of the Tribe Chironomini that are traditionally thought of as warm-water adapted
(e.g. Dicrotendipes and Cladopelma). The variation in the distribution of chironomid taxa dependant on localized conditions within a particular bioregion was also noted by
Kernan et al. (2009) who found that the inter-regional differences among lakes in Europe accounted for a large portion of the biotic response to large-scale variations in
84 environmental conditions. Thus, our analysis suggests that, while temperature is a primary factor governing the distribution of midges in Arctic lakes when conducting large spatial gradient analyses of the entire data set, there is within-region variability with
several lakes and ponds containing taxa that fall outside of a 'traditional' paradigm that invokes the presence of warm-water adapted Chironomidae taxa in lakes at southern latitudes, with shallow depth and high temperatures.
Nutrients and productivity
The role of nutrients and productivity in governing the distribution of chironomids has been alluded to in several recent studies (Brodersen and Anderson 2002,
Langdon et al. 2010). The difficulty of interpreting the influence of nutrients on the abundances of several taxa is complicated by the latitudinal transect-sampling approach of several Arctic surveys (Walker et al. 1991, Porinchu et al. 2009) that rely on the environmental conditions of relatively few lakes sampled along transects that cross large geographical areas (Telford and Birks 2009). While temperature gradients are maximized through latitudinal transect-sampling, other environmental factors may not be realized owing to short gradients of parameters responsive to local factors. This may produce results that may be scale-dependent, with only long environmental gradients generating statistically significant responses from assemblages (Walker et al. 1992; Velle et al. 2010). Our approach of sampling lakes and ponds from multiple regions of varying latitude and mean-July temperature, as well as multiple lakes from within the same geographical area, was to examine the 'pure' and interactive influences of nutrients/productivity and temperature on Chironomidae communities.
85 The significant influence of DIC and DOC on the first RDA axis is representative of larger, deeper lakes in colder regions that have lower nutrient inputs from catchment areas. Lakes and ponds ordinated in the upper-left quadrant of the RDA, reflective of higher MSSWT, higher inputs of DIC and DOC, and higher TN, had higher abundances of warm-water adapted Chironomini. The role of DIC and DOC in particular may be important in the distribution of several species (e.g. Cricotopus intersectus-type,
Endochironomus, Cladopelma, and Tanytarsus lugens/Corynocera oliveri gr.), based on their placement in RDA ordinations. In particular, Tanytarsus lugens/Corynocera oliveri were found to be positively associated with lakes with higher DOC concentrations, and were absent from lakes with low DOC (<1.0 mg'L"1) in the Clyde River vicinity.
Gajewski et al. (2005) also found that Tanytarsus lugens/Corynocera oliveri were found at higher DOC concentrations, but absent from the central Arctic Archipelago where watershed DOC inputs were limited. Likewise, Langdon et al. (2008) suggest a positive relationship between Tanytarsus lugens and total organic carbon. The restriction of zooplankton and benthic invertebrates due to high UVB penetration in Arctic ultra- oligotrophic systems with low DOC/DIC (Molot et al. 2004) could also be limiting Arctic aquatic food webs, subsequently influencing these taxa.
Several specific taxa were found to have higher abundances in lakes and ponds with both high MSSWT and TN (e.g. Cricotopus intersectus-type, Chironomus anthracinus, and Cladotanytarsus mancus gr.). In contrast, Micropsectra insignolobus- type was primarily found in colder lakes and ponds with lower TN, and was generally absent in lakes that were classified by TWINSPAN group C (productive, larger lakes).
86 Cricotopus intersectus-type in particular was found in lakes and ponds with higher available nutrients and higher MSSWT, which contrasts with most other species of
Cricotopus that are usually associated with cold-water conditions (Oliver and Roussel
1983). This relationship was also identified by Langdon et al. (2008), who found
Cricotopus intersectus-type in lakes with higher mean July temperature in northwest
Iceland. Herri and Lotter (2010) found that the coarse taxonomic resolution of several commonly grouped taxa reduces the effectiveness of chironomids as environmental indicators. Thus, the relationships of indicator taxa such as Cricotopus intersectus-type and Micropsectra insignolobus-type are of particular interest, as these chironomids are often grouped to a lower taxonomic resolution in other chironomid examinations in the
Arctic (Walker and MacDonald 1995; Gajewski et al. 2005; Francis et al. 2006).
Several relationships were also observed with the association with environmental variables significantly correlated to RDA axis 2, which is reflective of lakes and ponds in areas of higher depth and higher productivity. Dicrotendipes nervosus-type and
Microtendipes pedellus-type represented the highest relative abundance of warm-water adapted (Tribe Chironomini) species in several lakes. Walker and MacDonald (1995) related Microtendipes to the proximity of tree line. While lakes sampled along our tree- line gradient did indicate higher relative abundances of Microtendipes pedellus-type, they were also present several hundred kilometres north of tree line in warmer, larger, moderately deep (2-8 m), lakes with higher productivity (Figure 9b).
In contrast, Dicrotendipes nervosus-type was primarily found in shallow ponds in colder regions with higher total nitrogen (Figure 9d). Dicrotendipes is traditionally
87 considered a warm-water adapted taxon associated with tree line (Oliver and Roussel
1983; Walker and MacDonald 1995). However, Dicrotendipes were found at particularly high relative abundances farther north in shallow, buffered lakes around the town of
Repulse Bay (15-23%). In addition, Dicrotendipes were abundant (30 - 40%) in well- buffered lakes with higher nutrients on Victoria Island (Porinchu et al. 2009), and in warmer Icelandic lakes (Langdon et al. 2008). Three shallow lakes sampled on the south/south-western side of Bylot Island also contained small numbers of Dicrotendipes.
It is therefore likely that the combination of depth, nutrients, and temperature are the primary determinants of this species in Arctic systems, as they are primarily found in smaller, shallow lakes and ponds with higher available nutrients (represented by RDA axis 2, TWINSPAN group A). The pattern of either lake temperature and/or trophic status in strongly influencing the distribution of some taxa is similar to Brodersen and
Anderson (2002) and Langdon et al. (2010), who found that the temperature optima and trophic optima of chironomid taxa were strongly correlated.
Regionality
Unique regional patterns in the distribution of specific chironomid taxa were also observed. Lakes within the southern Kivalliq region were often dominated by several species of Tanytarsini (e.g. Cladotanytarsus mancus gr., Tanytarsus lugens, and
Corynocera ambigua). For example, Corynocera ambigua represented the most dominant species found in the Kivalliq region, and was found to comprise 20 - 80% of the chironomid community in warmer, deeper sites with higher productivity (Figure 9c).
This aptly named taxon has a perplexing environmental preference as it has been found to
88 be highly abundant in a variety of environments. Brodersen and Lindegaard (1999b) found Corynocera ambigua at high abundances in shallow, warm, eutrophic lakes in
Denmark. In contrast, Larocque-Tobler et al. (2010) found that Corynocera ambigua dominated a time period of inferred cold climate (8200 yr BP) in a Switzerland lake, and was presently absent from any lake sampled within the same area. Our results indicate that Corynocera ambigua is less abundant, or absent altogether, in shallow (< 3 m) lakes/ponds compared with deeper sites within the same geographical area. The relative abundances of other Tanytarsini were also higher in these shallow lakes.
Corynocera ambigua were not found in any of our northern surface samples from
Baffin Island, or extensive Baffin surveys by Francis et al. (2006). While this taxon's distribution suggests that its brachypterous adult stage is limiting its geographic distribution and dispersal from mainland regions of central Nunavut to Baffin Island, recent surveys by Porinchu et al. (2009) found high abundances of C. ambigua in several lakes on Victoria Island. Likewise, C. ambigua also occurs in several of the lakes sampled on Bylot Island (2.8 - 4.8 m depth), and comprised more than 18% of the chironomid community in Lake BY15. Walker and Cwynar (2006) and Barley et al.
(2006) have suggested that the distribution of this flightless chironomid species may be reflective of a Beringian refugium, as it has been previously found in very limited areas across the Arctic Archipelago. In contrast, Velle et al. (2005) found that high abundances of C. ambigua occurred at species-rich periods following deglaciation in Scandinavian lakes, suggesting that environmental factors, and not mobility, were the primary factor for early colonization. Our results reveal a pattern of C. ambigua dominance in larger
89 productive lakes, and absence in shallow un-productive lakes and ponds within the same geographical area. This suggests its distribution is influenced by environmental variables
(e.g. food availability and competition). C. ambigua's absence from island sites in eastern Nunavut also suggests its distribution is influenced by biogeographical patterns
(dispersion and/or colonization).
The chironomid communities of the Bylot Island lakes sampled were significantly different in each of the lakes sampled. Each lake was reflective of a different
TWINSPAN division, but were represented by taxa not commonly found at high abundances in any other region examined (Orthocladius type S, Tanytarsus lugens/Corynocera oliveri, and Zalutschia nr. lingulata pauca). These taxa contrasted with those found in lakes directly south in the Qorbignaluk headlands (72° 14' 37"N, 78°
42' 49"W) of Baffin Island, which contained mostly cold-water adapted Orthocladiinae taxa. For example, all three Bylot Island lakes contained small numbers of Dicrotendipes and C. ambigua, but very low numbers of Heterotrissocladius gr. that dominated the northern Baffin lakes. Likewise, the chironomid assemblages differ from other archipelago locales that were previously surveyed (Gajewski et al. 2005; Francis et al.
2006). The difference in the chironomid community found in each of the three Bylot
Island lakes was pronounced, and resulted in the placement of each lake in a different
TWINSPAN division. While each Bylot Island lake contained similar taxa, the composition of the most dominant taxa differed greatly between each lake. This may be explained by the alteration of nutrient regimes in Bylot Island lakes affected by high geese populations within the area (C6te et al. 2010). These highly seasonal localized
90 inputs may be influencing the direction of benthic invertebrate communities in specific lakes towards those more commonly found at higher temperatures owing to nutrient enrichment (Brodersen and Anderson 2002; Michelutti et al. 2011). The significant differences found in the chironomid assemblage composition of these Bylot Island lakes,
which are in a similar geographic and climatic area as the northern Baffin samples, highlight the influence of nutrients and productivity on the chironomid community composition of Arctic lakes. Thus, differences in nutrients and productivity at a local
scale can be reflected in significant differences in chironomid assemblages.
Future community change
Increases in temperature, reduced ice-cover, and increased thermal stratification
as a result of climate warming have shifted the structure of biological communities in
Arctic lakes (Walker et al. 1991; Douglas et al. 1994; Ruhland et al. 2003). The distribution of chironomid taxa within the eastern Canadian Arctic primarily follow a temperature gradient, however, our results indicate additional relationships along a nutrient/productivity gradient. In addition, while several species common in temperate systems are also cold-tolerant (Danks, 2004) and have been found to have successful multi-cohort extended aquatic life-cycles in Arctic environments (Butler, 1982), factors such as competition, predation, ice dynamics, and resource availability may be limiting the success of several warm-water taxa within Arctic systems beyond simple cold- tolerance. This is evident in the strong north-south gradient observed in the assemblage composition of several Chironomidae taxa.
91 From a perspective of using chironomid assemblages to examine climate warming within Arctic lakes and ponds, the presence of several genera of traditionally warm-water adapted taxa of the Tribe Chironomini (e.g. Cladopelma, Dicrotendipes, Microtendipes, and Polypedilum) could be an indicator of increased temperatures within these systems.
However, the response of this invertebrate community to future climate warming will also be dependent on localized climate-mediated changes to lake and catchment properties including food-web dynamics and catchment vegetation influence on allochthonous inputs of nutrients.
Partially constrained gradient analysis showed that some chironomid taxa could be responding to temperature and nitrogen independently. Those taxa that are strongly correlated to MSSWT, irrespective of the variation in the nitrogen gradient, could be used to differentiate between regional climatic gradients and local controls on nutrient levels.
Likewise, those specific taxa found to primarily follow a nitrogen gradient irrespective of the variation in MSSWT could be strong indicators of responses to changes in localized gradients. Thus, changes in specific indicator taxa may be able to separate the two gradients. These relationships could then be applied to biostratigraphic analyses to reconstruct past changes in lake environments, where paleolimnologists may be able to differentiate between climate effects and trophic status effects in influencing long-term changes in subfossil assemblage composition. Likewise, the biostratigraphic examination of taxa that are likely influenced by coincident changes in both temperature and nutrients/productivity (e.g. Cladotanytarsus mancus), may offer new insights into the ecological response of aquatic communities to changes in both these variables as a result
92 of possible future climate change. This study also suggests that, for more holistic inferences of how aquatic communities may respond to climate change (e.g. via quantitative inference models), sample selection should also be based on a nutrient/productivity gradient in addition to a large temperature gradient. This can be generated by sampling a number of lakes from within similar areas of varying landscape position and trophic status.
Acknowledgements
This project was funded by a Natural Sciences and Engineering Research Council of Canada NSERC Discovery Grant and NSERC Northern Research Supplement held by
R.Q., a NSERC Northern Research Internship held by ASM, the Northern Scientific
Training Program, and additional York University funding for graduate student research.
We are grateful to Stephen Brooks and Joshua Kurek for assistance with the diagnosis of our Zalutschia specimens, Sarah Finkelstein and Jane Devlin (University of Toronto) for sediment samples and water chemistry data for lakes and ponds within Sirmilik National
Park, and Derek Muir, Xiaowa Wang, and colleagues at the National Laboratory for
Environmental Testing for water chemistry analysis. Fieldwork assistance and support was provided by Mary Ellen Thomas, Jamal Shirley, Dorothy Tootoo, and the staff at the
Nunavut Research Institute and Nunavut Arctic College. We also thank Andy Aliyak,
Raymond Biastoch, Andrew Dunford, Milissa Elliott, and Christopher Luszczek for field sampling assistance.
93 References
Barley, E.M., Walker, I.R., Kurek, J., Cwynar, L.C, Mathewes, R.W., Gajewski, K., and Finney, B.P. 2006. A northwest North American training set: distribution of freshwater midges in relation to air temperature and lake depth. J. Paleolimnol. 36(3): 295-314. doi:10.1007/sl0933-006-0014-6. Brodersen, K.P. and Anderson, J.N. 2002. Distribution of chironomids (Diptera) in low arctic West Greenland lakes: trophic conditions, temperature and environmental reconstruction. Freshwater Biol. 47(6): 1137-1157. doi:10.1046/j.l365- 2427.2002.00831.x. Brodersen, K.P. and Lindegaard, C. 1999a. Classification, assessment and trophic reconstruction of Danish lakes using chironomids. Freshwater Biol. 42(1): 143- 157. doi:10.1046/j.l365-2427.1999.00457.x. Brodersen, K.P. and Lindegaard, C. 1999b. Mass occurance and sporadic distribution of Corynocera ambigua Zetterstedt (Diptera, Chironomidae) in Danish lakes. Neo- and palaeolimnological records. J. Paleolimnol. 22(l):41-52. doi: 10.1023/A: 1008032619776. Brooks, S.J., Bennion, H., and Birks, J.P. 2001. Tracing lake trophic history with a chironomid total phosphorus inference model. Freshwater Biol. 46(4):513-532. doi:10.1046/j.l365-2427.2001.00684.x. Brooks, S.J., Langdon, P.G. and Heiri, O. 2007. The identification and use of Palaearctic Chironomidae larvae in palaeoecology. QRA Technical Guide No. 10, Quaternary Research Association, London, UK. 276pp. Brundin, L. 1951. The relation of 02-microstratification at the mud surface to the ecology of the profundal bottom fauna. Report of the Institute for Freshwater Research, Drottningholm. 32:32-42. Butler, M. G. 1982. A 7-year life-cycle for 2 Chironomus species in Arctic Alaskan tundra ponds (Diptera, Chironomidae). Can. J. Zoolog. 60(l):58-70. doi:10.1139/z82-008. Clements, W.H., Carlisle, D.M., Lazorchak, J.M., and Johnson, P.C. 2000. Heavy metals structure benthic communities in Colorado mountain streams. Ecol. Appl. 10(2):626-638.doi:10.1890/1051-0761(2000)010[0626:HMSBCI]2.0.CO;2. Cote, G., Pienitz, R., Velle, G., and Wang, X.A. 2010. Impact of geese on the limnology of lakes and ponds from Bylot Island (Nunavut, Canada). Int. Rev. Hydrobiol. 95(2):105-129.doi:10.1002/iroh.200911151. Danks, H.V. 1971. Overwintering of some north temperate and arctic chironomidae. II. Chironomid Biology. Can. Entomol. 103:1875-1910. Danks, H.V. 2004. Seasonal Adaptations in Arctic Insects. Integr. Comp. Biol. 44(2):85- 94. doi: 10.1093/icb/44.2.85. Douglas, M.S.V., Smol, J.P., and Blake Jr, W. 1994. Marked post-18111 century environmental change in high Arctic ecosystems.Science : 416^419. Environment Canada. 1994. Manual of analytical methods. National Laboratory for Environmental Testing, Canadian Centre for Inland Waters, Burlington, Canada
94 Epler, J.H. 2001. Identification Manual for the larval Chironomidae (Diptera) of North and South Carolina. A guide to the taxonomy of the midges of the southeastern United States, including Florida. Special Publication SJ2001-SP13. North Carolina Department of Environment and Natural Resources, Raleigh, NC, and St. Johns River Water Management District, Palatka, FL. 526pp. Francis, D.R, Wolfe, A.P, Walker, I.R., and Miller, G.H. 2006. Interglacial and Holocene temperature reconstructions based on midge remains in sediments of two lakes from Baffin Island, Nunavut, Arctic Canada. Palaeogeogr. Palaeoclimatol. Palaeocol. 236(1-2): 107-124. doi:10.1016/j.palaeo.2006.01.005. Frisch, D., Green, A.J., and FiguerolaJ. 2007. High dispersal capacity of a broad spectrum of aquatic invertebrates via waterbirds. Aquat. Sci. 69(4):568-574. doi:10.1007/s00027-007-0915-0. Gajewski, K., Bouchard, G., Wilson, S.E., Kurek, J., and Cwynar, L.C. 2005. Distribution of Chironomidae (Insecta: Diptera) head capsules in recent sediments of Canadian Arctic lakes. Hydrobiologia. 549(1):131-143. doi:10.1007/sl0750- 005-5444-z. Gauch, H.G.Jr. 1982. Noise reduction by eigenvector ordinations. Ecology. 63(6): 1643- 1649. Giberson, D. J., Burian, S.K., and Shouldice, M. 2007. Life history of the northern mayfly Baetis bundyae in Rankin Inlet, Nunavut, Canada, with updates to the list of mayflies of Nunavut. Can. Entomol. 139(5):628-642. doi:10.4039/n06-089. Glew, J. 1991. Miniature gravity corer for recovering short sediment cores. J. Paleolimnol., 5(3):285-287. doi:10.1007/BF00200351. Heiri, O. and Lotter, A. F. 2010. How does taxonomic resolution affect chironomid-based temperature reconstruction? J. Paleolimnol. 44(2):589-601. doi:10.1007/sl0933- 010-9439-z. Hill, M.O. 1979. TWINSPAN. A Fortran program for ananging multivariate data in an ordered two-way table by classification of the individuals and attributes. Cornell University, Ithaca, NY. Johnson, R. K. 1995. The indicator concept in freshwater biomonitoring. Thienemann Lecture. In P. S. Cranston (ed). Chironomids - from genes to ecosystems. CSIRO, Melbourne. Pp 11-27. Juggins, S. 2003. C2 User Guide. Software for Ecological and Palaeoecological Data Analysis and Visualisation, University of Newcastle, Newcastle upon Tyne, UK. 69pp. Keatley B.E., Douglas M.S.V., and Smol J.P. 2008. Prolonged ice cover dampens diatom community responses to recent climatic change in High Arctic lakes. Arct. Antarct. Alp. Res. 40(2): 364-372. doi: 10.1657/1523-0430(06- 068)[KEATLEY]2.0.CO;2. Kernan, M., Ventura, M., Bitusik, P., Brancelj, A., Clarke, G., Velle, G., Raddum, G.G., Stuchlfk, E., Catalan, J. 2009. Regionalisation of remote European mountain lake ecosystems according to their biota: environmental versus geographical patterns. Freshwater Biol. 54(12):2470-2493. doi:10.1111/j.l365-2427.2009.02284.x.
95 Lamontagne, S., Donald, D.B., & Schindler, D.W. 1994. The distribution of four Chaoborus species (Diptera:Chaoboridae) along an elevation gradient in Canadian Rocky Mountain lakes. Can. J. Zool. 72(9): 1531-1537. doi: 10.1139/z94-203. Langdon, P.G., Ruiz, Z., Wynne, S., Sayer, CD., and Davidson, T.A. 2010. Ecological influences on larval chironomid communities in shallow lakes: implications for palaeolimnological interpretations. Freshwater Biol. 55(3):531-545. doi: 10.1 lll/j.l365-2427.2009.02345.x. Langdon, P.G., Holmes, N., and Caseldine, C.J. 2008. Environmental controls on modern chironomid faunas from NW Iceland and implications for reconstructing climate change. J. Paleolimol. 40(l):273-293. doi:10.1007/sl0933-007-9157-3. Larocque-Tobler, I., Heiri, O., and Wehrli, M. 2010. Late Glacial and Holocene temperature changes at Egelsee, Switzerland, reconstructed using subfossil chironomids. J. Paleolimnol. 43(4):649-666. doi:10.1007/sl0933-009-9358-z. Medeiros, A.S., Luszczek, C.E., Shirley, J., and Quinlan, R. 2011. Benthic Biomonitoring in Arctic Tundra Streams: A Community-Based Approach in Iqaluit, Nunavut, Canada. Arctic. 64(l):59-72. Michelutti, N. Mallory, M.L., Blais, J.M., Douglas, M.S.V., and Smol, J.P. 2010. Chironomid assemblages from seabird-affected High Arctic ponds. Polar Biol. 34(6):799-812.doi:10.1007/s00300-010-0934-5. Molot, L. A., Keller W., Leavitt P.R., Robarts R.D., Waiser M.J., Arts M.T., Clair T.A., Pienitz R., Yan N.D., McNicol D.K., Prairie Y.T., Dillon P.J., Macrae M., Bello R., Nordin R.N., Curtis P.J., Smol J.P., and Douglas M.S.V. 2004. Risk analysis of dissolved organic matter-mediated ultraviolet B exposure in Canadian inland waters. Can. J. Fish. Aquat. Sci. 61(12): 2511-2521. doi:10.1139/f04-165. Oliver, D.R. and Roussel, M.E. 1983. The insects and arachnids of Canada, Part 11: The genera of larval midges of Canada - Diptera:Chironomidae. Publication 1746, Agriculture Canada, Ottawa, ON. 263pp. Oliver, D.R., Dillon, M.E. 1997. Chironomids (Diptera: Chironomidae) of the Yukon Arctic North Slope and Hershel Island. In Insects of the Yukon, Danks, H.V., Downes J.A. (eds). Biological Survey of Canada (Terrestrial Arthropods): Ottawa; 615-635. Pinder, L.C.V. 1995. The habitats of Chironomidae larvae. In: Armitage, P.D., Cranston, P.S., and Pinder, L.C.V (eds). The Chironomidae: Biology and ecology of non- biting midges. Chapman and Hall, London, pp. 107-133. Porinchu, D., Rolland, N., and Moser, K. 2009. Development of a chironomid-based air temperature inference model for the central Canadian Arctic, J. Paleolimnol. 41(2): 349-368. doi:10.1007/sl0933-008-9233-3. Porinchu D. F. and G. M. MacDonald. 2003. The use and application of freshwater midges (Chironomidae: Insecta: Diptera) in geographical research. Prog. Phys. Geog. 27(3): 378-422. doi:10.1191/0309133303pp388ra Quinlan, R., and Smol, J.P. 2001a. Chironomid-based inference models for estimating end-of-summer hypolimnetic oxygen from south-central Ontario shield lakes. Freshwater Biol. 46(11): 1529-1551. doi:10.1111/j.l365-2427.2001.00763.x.
96 Quinlan, R., and Smol, J.P. 2001b. Setting minimum head capsule abundance and taxa deletion criteria in chironomid-based inference models. J. Paleolimnol. 26(3): 327-342. doi:10.1023/A:1017546821591. Rieradevall, M., and Brooks, S.J. 2001. An identification guide to subfossil Tanypodinae larvae (Insecta:Diptera:Chironomidae) based on cephalic setation. J. Paleolimnol. 25(1): 81-99. doi:10.1023/A:1008185517959. Rouse, W., Douglas, M., Hecky, R., Hersey, A., Kling, G., Lesack, L., Marsh, P., McDonald, M., Nicholson, B., Roulet, N. & Smol, J. P., 1997: Effects of climate change on the freshwaters of Arctic and sub Arctic North America. Hydrol. Process. 11(8): 873-902. doi:10.1002/(SICI)1099- 1085(19970630)ll:8<873::AID-HYP510>3.0.CO;2-6. Ruhland K., Priesnitz A., and Smol J.P. 2003. Paleolimnological evidence from diatoms for recent environmental changes in 50 lakes across Canadian Arctic treeline. Arctic Alpine Res, 35(1): 110-123 Saether, O.A. 1976. Revision of Hydrobaenus, Trissocladius, Zalutschia, Paratrissocladius, and some related genera (Diptera:Chironomidae). Bull. Fish. Res. Bd Can. 195:1-287. Saether, O.A. 1979. Chironomid communities as water quality indicators. Holarctic Ecol. 2(2):65-74. Available from: http://www.jstor.org/stable/3682659 [accessed 22 Dec 2010]. Statistics Canada, 2006. 2006 Census of Canada. Government of Canada, Ottawa. Stur, E. Ekrem, T. 2011. Exploring unknown life stages of Arctic Tanytarsini (Diptera: Chironomidae) with DNA barcoding. Zootaxa. 2743(18):27-39. Telford, R.J. and Birks, H.J.B. 2009. Evaluation of transfer functions in spatially structured environments. 28(13-14): 1309-1316. doi:10.1016/j.quascirev.2008.12.020 ter Braak, C.J.F. and Smilauer, P. 1998. CANOCO reference manual and user's guide to CANOCO for windows: Software for canonical community ordination (version 4). Microcomputer Power, NY. 352pp. Thienemann, A., 1915. Die Chironomidenfauna der Eifelmaare. Verhandlungen des Naturhistischer Verein der Rheinlande und Westfalens. 72:1-58. Uutala, A.F. 1990. Chaoborus (Diptera:Chaoboridae) mandibles - paleolimnological indicators of the historic status of fish populations in acid-sensitive lakes. J. Paleolimnol. 4(2): 139-151. doi:10.1007/BF00226321. Velle, G., Brodersen, K.P., Birks, H.J.B., and WiUassen, E. 2010. Midges as quantitative temperature indicator species: Lessons for palaeoecology. Holocene. 20(6):989- 1002. doi: 10.1177/0959683610365933. Velle, G., Brooks, S.J., Birks, H.J.B., and WiUassen, E. 2005. Chironomids as a tool for inferring Holocene climate: an assessment based on six sites in southern Scandinavia. Quat. Sci. Rev. 24(12-13): 1429-1462. doi:10.1016/j.quascirev.2004.10.010 Walker, I. R. 2007. The WWW Field Guide to Fossil Midges [online]. Available from http://www.paleolab.ca/wwwguide [accessed 5 May 2010].
97 Walker, I.R. and Cwynar, L.C, 2006. Midges and paleotemperature reconstruction - the North American experience. Quaternary Sci. Rev. 25(15-16): 1911-1925. doi:10.1016/j.quascirev.2006.01.014. Walker, I.R. 2003. Chironomid overview. In S.A. Elias (ed.) Encyclopaedia of Quaternary Science, Vol. 1. Elsevier, Amsterdam, pp 360-366. Walker I.R. 2001. Midges: Chironomidae and related Diptera. In: Smol J.P., Birks H.J.B., and Last W.M. (eds). Tracking environmental change using lake sediments, vol. 4. Zoological Indicators, Kluwer Academic Publishers, Dordrecht, the Netherlands, pp 43-66. Walker, I.R., Levesque, A. J., Cwynar, L.C, and Lotter, A.F. 1997. An expanded surface-water paleotemperature inference model for use with fossil-midges from eastern Canada. J. Paleolimnol. 18(2): 165-178. doi: 10.1023/A: 1007997602935. Walker, I.R., and MacDonald, G.M., 1995: Distributions of Chironomidae (Insecta:Diptera) and other freshwater midges with respect to treeline, Northwest Territories, Canada. Arct. Alp. Res. 27(3): 258-263. DOI: 10.2307/1551956. Walker, I.R., Smol, J.P., Engstrom, D.R., Birks, H.J.B., 1992. Aquatic invertebrates, climate, scale, and statistical hypothesis testing: A response to Hann,Warner, and Warwick. Can. J. Fish. Aquat. Sci. 49(6): 1276-1280. doi:10.1139/f92-143. Walker, I.R., Smol, J.P., Engstrom, D.R., and Birks, H. J. B. 1991. An Assessment of Chironomidae as quantitative indicators of past climate change. Can. J. Fish. Aquat. Sci. 48(6): 975-987. doi:10.1139/f91-l 14. Walker, I.R. 1990. Modern assemblages of Arctic and alpine Chironomidae as analogues for late-glacial communities. Hydrobiologia. 214(1): 223-227. doi: 10.1007/BF00050954. Walker, I.R. 1988. Late-Quaternary palaeoecology of Chironomidae (Diptera:Insecta) from lake sediments in British Columbia. PhD Thesis, Simon Fraser University, Burnaby, B.C. 204pp. Wiederholm, T. 1984. Incidence of deformed chironomid larvae (Diptera: Chironomidae) in Swedish lakes. Hydrobiologia. 109(3):243-249. Wiederholm, T. (eds). 1983. Chironomidae of the Holarctic region. Keys and diagnosis. Part 1. Larvae. Entomologica Scandinavica Suppl. No. 19. 457pp. Wiederholm T. 1980. Use of benthos in lake monitoring. J. Water. Pollut. Con. F. 52(3):537-547.
98 Tables
Table 1 Descriptive statistics of environmental variables for the 63 sampled lakes and ponds. Abbreviations: Mid-summer surface water temperature (MSSWT), surface area (SA), chlorophyll-a (CHLA), chloride (CL), sulphate (SO4), dissolved organic carbon (DOC), dissolved inorganic carbon (DIC), ammonia (NH3), silicon dioxide (Si02), oxygen reduction potential (ORP), total nitrogen (TN), total phosphorus (TP). S.D. = standard deviation of the mean, Min = minimum value, Max = maximum value. * Excludes BL01 (Baker Lake).
Variable Mean S.D. Min Max Transformation
SA (ha) 204.8* 413.9* 0.8 2214* log
Depth (m) 4.5 4.3 0.5 20.6 log
MSSWT (°C) 12.7 3.2 4.9 18.3 none
pH (-log[H+]) 7.7 0.6 6.3 8.9 none
CHLA (ug-L1) 1.1 0.8 0.1 3.7 V
CL (mg-L1) 12.1 31.0 0.2 235.0 log(x+l)
1 S04(mg.L ) 2.8 3.2 0.0 21.2 log(x+l)
DOC (mg'L1) 5.2 5.5 0.6 42.1 log
DIC (mg'L1) 4.9 3.8 0.6 19.4 ^
1 Si02 (mg-L- ) 0.6 0.8 0.0 3.5 V
ORP (mg-L1) 169.7 123.1 38.0 512.3 log
1 NH3 (mg-L- ) 43.0 38.5 2.5 189.0 tf
TP (mg-L1) 7.1 5.6 1.7 41.7 log
TN (mg-L-1) 349.2 161.2 89.0 840.0 V Table 2 Eigenvalues of first four redundancy analysis (RDA) axes, species environment correlation for each canonical axis, and cumulative % variance of species data and species.
Axis 1 Axis 2
Eigenvalues 0.194 0.106
Species-environment correlations 0.929 0.780
Cumulative percentage variance (%)
of species data 19.4 30.0
of species-environment relation 40.2 62.1
Sum of all eigenvalues 1.000
Sum of all canonical eigenvalues 0.483 Table 3 Regression coefficients of the first two redundancy analysis (RDA) axes (AXl and AX2), their associated r-test values, and interset conelations for backwards-selected environmental variables. Bolded ^-values are statistically significant (P < 0.05).
Regression Interset Environmental Coefficients Conelations r-values Variables AXl AX2 AXl AX2 AXl AX2
Depth 0.388 0.475 0.724 0.366 4.18 2.54
TN -0.142 -0.579 -0.752 0.035 -0.90 -1.83
MSSWT -0.322 0.229 -0.717 0.160 -3.03 1.07
TP -0.129 0.225 -0.545 0.149 -1.17 1.02
CL 0.313 -0.787 -0.522 0.080 2.15 -2.68
DIC -0.288 0.196 -0.537 -0.112 -2.83 0.95
Si02 0.438 -0.917 0.398 -0.468 4.03 -4.17
S04 -0.139 0.332 -0.487 -0.018 -1.01 1.19
NH3 0.014 0.138 -0.420 0.098 0.16 0.78
SA 0.208 0.049 0.157 0.456 2.38 0.28
CHLA 0.021 0.257 -0.283 0.311 0.25 1.52
pH 0.109 -0.171 -0.278 -0.237 1.41 -1.10
DOC -0.179 0.269 -0.398 0.106 -2.42 1.80
ORP -0.123 -0.198 0.448 -0.167 -1.17 -0.94 Figures
Figure 1: Locations of all recent (>1980 AD) published examinations of subfossil
Chironomidae assemblages within the territory of Nunavut, Canada.
Figure 2: Diagnostic images for Zalutschia types found: (a) Zalutschia sp. C. (Barley et al. 2006 as type B), (b) Zalutschia nr. lingulata pauca (Wiederholm, 1983), (c)
Zalutschia zalutschicola (Saether, 1976).
Figure 3: Relative Abundances of selected taxa recovered from surface samples of
Nunavut lakes and ponds from Jun 2006 - Aug 2009. Taxa are ananged according to the sample scores of the RDA constrained to mid summer surface water temperature, with cold-water taxa on the left of the diagram. Sites are ananged by latitude, south (top) to north (bottom).
Figure 4: The classification of the 63 sampling locations across the eastern Canadian
Arctic with the use of two-way indicator species analysis (TWINSPAN). Significant differences (P < 0.05) between TWINSPAN divisions via ANOVA of DCA Axis 1 sample scores (solid-line boxes) and DCA Axis 2 sample scores (dashed-line boxes) are indicated. Taxon names are reflective of the TWINSPAN indicator taxa.
Figure 5: Box-plots of selected environmental variables by TWINSPAN group: (a) temperature, (b) SA, (c) depth, (d) conductivity, (e) DIC, (f) CHLA, (g) TN, (h) TP.
Figure 6: RDA of sampled locations and significant environmental indicators.
TWINSPAN divisions significant to DCA axis 1 are indicated; • division A, • division
B, • division C.
102 Figure 7: RDA of species and significant environmental indicators. Environmental abbreviations are as in Table 1. Selected species are indicated, and abbreviated as follows; Abisk = Abiskomyia, Hydrob = Hydrobaenus/Oliveridia gr., Psecp =
Psectrocladius psectrocladius gr., ParakN = Parakiefferiella nigra, Pseudo =
Pseudodiamesa, Psecp = Psectrocladius psectrocladius gr. ZalutB = Zalutschia type B,
ZalutL = Zalutchia nr. lingulata pauca, ZalutZ = Zalutchia zalutschicola, Htotal =
Heterotrissocladius gr., Mesoc = Mesocricotopus, Cinter = Cricotopus intersectus-type,
CoryO = Corynocera Oliveri-type, Paracl = Paracladius, Cory Arc = Coryononuera arctica, Eukif = Eukiejferiella / Tvetenia gr., Limno = Limnophyes, OrthoS =
Orthocladius type S, Metro = Metrocnemus gr., Cladman = Cladotanytarsus mancus gr.,
Minsigs = Micropsectra insignilobus-type, Cambig = Corynocera ambigua, ZavStemp
= Zavrelia /Stempellinella gr., Protany = Protanypus, ParaT = Paratanytarsus gr., Dicrot
= Dicrotendipes nervosus-type, Microten = Microtendipes pedellus-type, Crypto =
Cryptochironomus, Cpelma = Cladopelma, Chiron = Chironomus anthracinus-type,
Polyp = Polypedilum, Harnesh = Harnischia /Paracladopelma gr., Proclad = Procladius,
Sticto = Strictochironomus gr.
Figure 8: Species anows in a singly constrained partial RDA of (a) MSSWT with TN as a covariable, and (b) TN with MSSWT as a covariable. Only species anows with a strong conelation (Irl > 0.5) are shown.
Figure 9: Abundances of environmental indicator taxa plotted on the RDA with significant environmental indicators indicated for (a) Abiskomyia, (b) Microtendipes pedellus-type., (c) Corynocera ambigua gr., and (d) Dicrotendipes nervosus-type.
103 TWINSPAN divisions significant to DCA axis 1 are shown; • division A, • division B,
• division C. Larger symbols represent higher relative abundances as in the Figure 7A inset legend.
104 //?%/ 7 1 Researcher •¥• Medeiros Q i k i q t a a I u k \ i B a f f i n )\ R e g i o n\ A Franois \ \ \ \ • Gajewski • Porinchu \ ,-- + Rolland -*''\ • Vtelker m m \ ••!••• Treeline
Fig.l
105 l%
*> *
V' £ Fig. 2
106 £'%]£ Sample location 5??8S8SSSagliilPFPggFgBBBi5BB?lSSigSSS5SSSSS?S|SSS?S|19
J nn -[. TTT—
:r—rr ^T 1 V -my ^T :: n
c % V TO*, \. -r-TTTn ~P~ uX>
IT IT T~^ ^T xXN ^'l-'l- | ii|'-- "T ^J^TT-T— n | | • . • • rr 1 X 1 1 - [ I ::JT - 'ir' "- - • 'IT • 1 i \ v tr rm i ••--•!--- FT 'il"'! II 'i • I I I 11 I • I • • I I I I\N I \ c % • • '• I I '• I • • 11 T-T r •• i' -mrr •— • • • \ x x. v * l-M-l ('I' -HIM -r^TTTn T -1 XX ir TV I '" '•'^X\" > Tnnpn nTTTTi \X\ IT TTI ^T*- ^r '• ,,,"l • I'M" ' ^
1 • "X\X ir •|" r • ii" I "• T ••• T ! ••'r--' I i[c\\\ IT N=48 N=15 \ Dicrotendipes nervosus-type Zalutschia type C Cladotanytarsus mancus-gr Heterotrissocladius gr. N=21 N=27 r 1 Chironomus Corynocera ambigua, Arctopelopia, anthracinus-type Zavrelia / Stempellinella gr. BL01.BL02, BL16, 1 CR04, CR01.QB07, IQ05, CR02, QB01, QB06, BY14, IQ04. IQ01.IQ03 AV07, BL06. BY16.AV02, N=14 N=13 AV05, RB01.RB02, RB04, RB14, RB17, BL35, BL17, AV01,AV04,AV06,AV11, Zalutschia type C Psectrocladius (mono) gr. RA10, RA11.RA07, RA02, Microtendipes IQ02 B1 B2 i RA17, RA19.AV09, CH09, CH08, CH07 RA20, RA15, RA08, CH06, CH04, CH05 RA13, RA16, RA05, AV10, RA12, BL11, RA06, RA04, RB07, CH01,CH02,AV03 BY15 Fig. 4 TN(Mg'L') DIC (mg'L-1) log Depth (m) MSSWT (°C) «a % i 1 § B. S I B fe£ tut )at fJ & * an » g s
> • 4- -!•• 4— 4- 1*
00 i f —h 4- I • —\*
00 - —!• < —!• to 4 I —!•
o - • 1 * 1 -+ •H ~i I i
CHLA(mg»L-1) tog COND (ug'L1) log Surface area (ha)
O »~* *»* Ki W I* O --* «J w s s i s § i S
• f- H> * •Hh
4- • 1 -H»
4 h •I— H •
o v© 1.0 !
AV09 A RA15 RA04 M1J l * * CH06 A A 1\ Area BL01 CHLA RA17 o£ ™» ^ Depth RA05AcH09S: A/ RA02«\ 4* A i A/ACH04 CD CH08 DOC \ fvfos m BL02/^ CR04 O MS R&i/iA\AV10 i 7 " s' m »SWTTP NH3 w^X™ ; / A QB06 i TN BL16 • CM '•-LIlj^J AV07 \ / CH05 103 •
AV : S0^4 RA20^-*" 3^-^AO7/-^^^>k A\A^V' V""^ ^A„ , X .. QB07 IQ05 03 03 DIC *B04 /SY«|A V B A^^ • QSOf Q CHOI ORP AV01 pH j«« By\ • • IQ04 • IQ01 CROf • CR02 AV02 • RB02 I AV04 IQ06 RB01* 9 *BL17 ; • RBI4 BYUm Si02 1.0 1 i —i -1.0 RDAaxis 1-19.4% 1.0
.6 RDAaxis 1 -19.4%
Fig. 7
in (b)
Hlolal
Synth Pva,^/\, Protany TN, ParaT ^O^rm^^y^Harnenh CricBI _ T!^^^^0 Orthocon Cpe!mq^geo„^:^~^'Mpadu TanyMend"* >>,4WsW Dicrot ZatutL ZalulC Mesoc
1.0 -1.0 1.0 pRDA axis 1 Fig. 8
112 08 08 A (a) (b) I Area Area Depth CHLA Depth */ /A MSSWT >^ * MSSWT NH3 \ , / • ^TP DOC \ /^ • ^TP^DOC \ CL CL ^r*a=^ \ \ '/ /^^ •"""A TN « ~—E5^ • • TN -^—-__^3~^£ - ' . - S04 SO.T ^^Z^-^/ DIC " DIC * / i' \ V" A ORP :. ORP • 15 20+% pH • A 20+% PH * * II 10 15% \ • A 1° 20* • 6 10% i CO i 5 10% • 3 6% • 1 3% • 1 5% . * SI02 -08 0% ™ Sl02 -0 8 0% -1 0 1 0 -10 1 0 08 08 Area Area • A + " A! A* A. Depth CHLA ' * Depth CHLA A A { M / •\A>J% •V.5V */* MSSWT M A y MSSVVT™*^ V ^TP NH3DOC^ / V / * / CL5=JW- A . TN - £5r£5 ^^ * m TN -. —^faJSSsa+V t^"^ SO* '~~^~- S04 •^•T-i^^^/ • DIC , A\ ^"^ DIC -"W / .» I« \ \ ^^^• A- ORP ~* ORP 80 *• \ A ** pH *' ^f 20*% A 50-60% j \ . P" A 10-20% J^ 40 50% , A 20 40% . • 3 6% i. 1 20% , 3 -0 8 0% * Si02 -0 8 V SI02 * J, * -10 10 -10 1 0 RDAaxis 1-194%
Fig. 9
113 CHAPTER 4: A high resolution multi-proxy record of pronounced 20 century environmental change at Baker Lake, Nunavut.
Medeiros AS1*, Friel CE2, Finkelstein SA2, and Quinlan R1
1 York University, Department of Biology, 4700 Keele St, 211 Lumbers Building,
Toronto, Ontario, M3J 1P3, Canada.
2 University of Toronto, Department of Geography, 100 St George Street, Rm 5047,
Toronto, Ontario, M5S 3G3, Canada
* Corresponding Author: [email protected]
Manuscript style for submission to Journal of Paleolimnology
114 Abstract
A multi-proxy approach was used to quantify recent environmental changes at Baker
Lake in Nunavut, Arctic Canada. Analyses of fossilized remains of Chironomidae, the dominant aquatic invertebrate group in Arctic lakes and ponds, and diatoms, the major primary producers in the system, were conducted on a gravity sediment core of 20 cm in length. Each interval of the core was sampled at 0.5 cm resolution and 210Pb dating established a sub-decadal chronology. A surface sediment training set of subfossil chironomid assemblages from 65 lakes across the eastern Canadian Arctic generated a robust surface water paleotemperature transfer function. The transfer function was applied to stratigraphic intervals from the Baker Lake sediment core to generate a paleotemperature reconstruction of sub-decadal resolution. The surface water temperature reconstruction inferred a 3° C increase in mid-summer surface water temperature for
Baker Lake over the last 60 years, which was corroborated by the available local instrumental record spanning the period of 1950-2007 AD. The chironomid subfossil record shows a pronounced decline, and subsequent extirpation, of several cold-water indicator taxa and the appearance of warm-water indicators in recent sediments. This shift in community structure began circa 1940 AD, and intensified after 1985 AD. The corresponding fossil diatom record also showed an increase in small planktonic
Cyclotella taxa over the past 60 years, intensifying in the last 5 years. These diatom changes further corroborate the inferences of warmer climate and longer ice-free seasons from the chironomid-derived water temperature reconstructions and the instrumental
115 record. The shifts in the diatom assemblages began later than the shifts in the chironomid assemblages, and were of lower magnitude, reflecting differences in sensitivity to climate changes due to differences in the life histories of these two groups. The results also confirm that large, deep lakes are responsive to recent climate warming despite a large water volume.
116 Introduction
It is now well established that climate changes significantly affect aquatic ecosystems, particularly in the Arctic (Wrona et al. 2006). Current research questions include identifying specific mechanisms and pathways of climate-mediated changes in freshwater systems. Of particular importance is identifying how changes in the physical and chemical properties of freshwater bodies, due a warmer climate, promote changes in community structure and ecological interactions across trophic levels. In the absence of long-term monitoring data in the Arctic, paleolimnological records supply valuable records of climate history and associated ecological change. Changes in the abundance and assemblage compositions of microscopic biological remains preserved in lake sediments can be used to reconstruct past environments, and to produce numerical estimates of paleo-temperatures or other variables of interest (e.g., Antoniades et al.,
2005). Organisms such as Chironomidae (Insecta: Diptera) and diatoms (Chromista:
Bacillariophyta) respond sensitively to climatic and climate-induced limnological changes including temperature (Walker et al. 1991), dissolved oxygen (Quinlan et al.
2001a), pH (Sorvari et al. 2002), and trophic status (Broderson and Lindegaard 1999;
Hadley et al. 2010), making them useful for reconstructing long-term responses of freshwater food webs to climatic changes (Brooks et al. 2001; Antoniades et al. 2005;
Quinlan et al. 2005).
Several surveys of surface sediments across varied spatial scales and multiple bioregions in the Canadian Arctic have related gradients in environmental conditions to the distribution of chironomid taxa (Walker 1990; Gajewski et al. 2005; Porinchu et al.
117 2009a; Medeiros and Quinlan 2011). Chironomids have an aquatic (larval) and terrestrial
(adult) stage, which makes them ecologically sensitive to the environmental and landscape conditions of both their freshwater habitats and the terrestrial environment
(Oliver 1971). In the Arctic, chironomid taxa are particularly sensitive to temperature, since many taxa are living at their ecological limits (Walker et al. 1991). Arctic lakes are often dominated by cold-water adapted chironomids of the sub-families Orthocladiinae and Diamesinae (Oliver 1971; Walker 1990), which unlike the Tanypodinae and
Chironominae, are known to decrease in abundance in warmer habitats (Oliver 1971).
Conversely, several chironomid taxa (such as those of the Tribe Chironomini) are primarily represented in warmer regions, as they are limited by mean summer temperatures (Walker et al. 1991). These warm-water adapted taxa can also remain in their aquatic larval stage for several years until conditions are suitable for successful pupation into the terrestrial environment (Butler 1982; Pinder 1995), and their development and emergence are also rapidly induced by warmer conditions (Oliver 1971;
Danks 2007). Thus, short periods of increased temperatures may allow for the quick establishment of southern, temperate chironomid taxa. Owing to this strong sensitivity to temperature, chironomid assemblages in sediment cores are powerful means of reconstructing past climates (Walker et al. 1991). This close relationship provides the basis for developing chironomid-based inference models for quantitative reconstructions of past temperatures in the Arctic (Barley et al. 2006; Francis et al. 2006; Porinchu et al.
2009b).
118 The algal productivity of freshwater ecosystems is also expected to be altered by climate warming (Wrona et al. 2006). Diatoms are important primary producers and support higher trophic levels, including chironomid larvae and other benthic invertebrates. Cold surface water temperatures, short open water seasons, and low nutrient concentrations substantially limit algal production, which is often confined to bottom substrates of Arctic lakes and ponds (Chetelat et al. 2010). Climate therefore plays a critical role in algal dynamics and diatom community structure (Ruhland et al.
2008), with climatically and limnologically mediated changes in diatom communities in turn influencing chironomid communities (Quinlan et al. 2005).
Studies of recent shifts in the structure of diatom communities of Arctic lakes have also suggested that reduced ice-cover and enhanced thermal stratification are important mechanisms that alter diatom communities in the context of warmer climate
(Ruhland et al. 2003; Smol et al. 2005). For example, an increase in the abundance of planktonic diatoms (e.g., Cyclotella spp.) relative to periphytic or benthic diatoms, especially Fragilaria spp., may indicate a decrease in the duration and extent of ice cover under warmer climates (Ruhland et al. 2008). A decrease in ice cover duration can result in an expansion of available diatom habitats within the water column via increased water depth, deepened light penetration and enhanced thermal stratification (Douglas and Smol
2010; Paul et al. 2010). Such diatom assemblage shifts have been noted in recent studies in both European and North American lakes (Sorvari and Korhola 1998; Ruhland et al.
2008). Thus, diatom and chironomid assemblages are both sensitive to temperature, particularly in the Arctic. Comparing the responses to recent warming of these two
119 communities in the same system can provide new insights into the impacts of climate change across multiple trophic levels.
The first objective of this study was to investigate the application of a chironomid-based temperature inference model applied to the down-core sediment profile of Baker Lake to determine the late Holocene temperature history for this continental region of the eastern Canadian Arctic. The Baker Lake site also has a yearly instrumental record available since AD 1950, which is unusually long for paleolimnological research sites in the Arctic and provides an opportunity to assess the robustness of the chironomid- based temperature estimates. The second objective of this study was to compare the timing and the magnitude of changes in both benthic invertebrate grazers and primary producers in response to climate warming. Quinlan et al. (2005) show synchronous changes in diatom and chironomid assemblages in response to warming at three sites in the Arctic, but with few samples over the time period of recent warming. Our study aims to examine the timing of chironomid and diatom responses to climate warming at high resolution in an Arctic lake sediment record.
Study Area
Baker Lake is one of the largest (surface area = 188,700 ha) and deepest (20+m) lakes in the Kivalliq region of the territory of Nunavut, an area that represents a large portion of the continental eastern Canadian Arctic. The Baker Lake area is characterized by a continental climate that is offset by localized lake-effect warming and cooling throughout the open-water season. A town, also named Baker Lake, is located on the north-western shores of the lake, and is home to approximately 1700 people. The meteorological station
120 at the Baker Lake airport provided an instrumental temperature record since 1950, which was obtained from the Meteorological Service of Environment Canada for the time period 1950-2007. Instrumental records indicated a polar climate with a mean annual temperature of -11.8 °C and total annual precipitation averaging 156.7 mm from 1971-
2001. The geology of the region is of Proterozoic - neoArchean origin generally characterized by weathered Archean granitoid gneiss overlain by feldspathic sandstone
(Rainbird et al. 2003). Catchment vegetation was primarily represented by dwarf shrubs
(Salix spp.), several grasses (Poaceae), and moss campion (Silene acaulis). Dwarf fireweed (Chamerion latifolium), Cloud Berry (Rubus chamaemorus), and Saussurea were also common.
Methods
Field Methods
Replicate sediment cores of 20.0 cm in length were recovered from the eastern basin of
Baker Lake, at a depth of 13.7 m, using an Uwitec gravity corer (8.4 cm diameter) deployed from a boat on August 5, 2007. Each core was extruded in the field at 0.5-cm intervals, stored in Whirl-Pak ® bags, and kept cool (4 °C) and dark until processed in the lab. Physicochemical variables were measured using an YSI-600QS multi-parameter probe every meter from the lake bottom. Measurements indicated that the lake had a consistent conductivity of -20 uS cm"1, temperature of 11.0 - 11.5 °C, and pH of 7.0 -
7.15 from the surface to the lake bottom, which indicated that the lake was not thermally stratified at the time of sampling. Replicate epilimnetic water samples were collected at
121 0.5 m below the water surface in pre-cleaned polyethylene bottles and immediately
treated in the field following the protocols outlined in the Analytic Methods Manual of
Environment Canada (Environment Canada 1994). Samples were analysed by the
National Laboratory for Environmental Testing (NLET) at the Canadian Centre for
Inland Waters (CCIW), Burlington, Canada, following protocols from Environment
Canada (1994).
Laboratory Methods
Chironomid analysis followed Medeiros and Quinlan (2011): Sediments were sub-
sampled (1-2 grams wet weight) and washed through nested 212 um and 106 um mesh
sieves. For each interval a minimum of at least 50 subfossil chironomid head capsules
were enumerated following Quinlan and Smol (2001b). Specimens were mounted on
glass slides and identified at 400-lOOOx magnification to the best taxonomic resolution possible. Identification of specimens follows Medeiros and Quinlan (2011), based on
Oliver and Roussel (1983), Wiederholm (1983), Epler (2001), Rieradevall and Brooks
(2001), and Brooks et al. (2007).
A chironomid-based surface water paleotemperature inference model was
developed using a training set from sediment and water samples from lakes and ponds
that were sampled in clusters of areas that spanned several regions, along a transect beginning at tree-line (Northern Manitoba), and extended through the Kivalliq region
towards the north-eastern islands within the Baffin region of Nunavut (Fig 1). Lakes in the training set were sampled in areas of varying landscape and environmental
122 characteristics in order to maximize the variability of limnological conditions within a particular area. A midge-based inference model was then developed based on a calibration set containing 65 lakes and 86 midge taxa as outlined in Medeiros and
Quinlan (2011). The R statistical language v2.13.0 (R Development Core Team 2011) was used to generate and evaluate transfer functions with the rioja library v0.5e6 (Juggins
2009). Tests for assessing the statistical significance of reconstructions were evaluated using palaeoSig vl.O (Telford and Birks 2011) and potential autocorrelation in the dataset was evaluated using gstat v0.9e81 based on the methods of Telford and Birks (2009).
The relative abundance of each taxon was calculated as the percent of total identifiable midges, and square root transformed. Chironomid taxa were constrained to mid-summer surface water temperature and evaluated with weighted average (WA), partial least squares (PLS), weighted average partial least squares (WA-PLS) transfer functions, and the modern analogue technique (MAT), and plotted using C2 software
(Juggins, 2003). To determine whether the chironomids assemblages down-core were represented by the surface sample calibration set, the core intervals were plotted passively in a Principal Correspondence Analysis (PCA) biplot using CANOCO v4.53
(ter Braak and Smilauer 1998). Additional redundancy analysis (RDA) and PCA were run with the vegan library vl.l7el0 in R (Oksanen et al. 2010).
Diatoms were analyzed from 23 samples at 1-cm intervals throughout the core with 3 additional samples at 1, 8 and 18 cm core depth. Processing followed Ruhland et al. (1999): subsamples of 0.5 ml were treated with 10% HCl, followed by treatment with nitric and sulfuric acids (50:50 molar ratio), and heated to fully digest the organic matter.
123 Due to the high clay content of the samples, a few drops of a 5% Calgon solution (sodium hexametaphosphate) were added to encourage disaggregation of the fine clastic material.
The samples were then successively rinsed with distilled water until neutral. The resulting diatom slurries were mounted onto glass microscope slides using Naphrax . A minimum of 500 diatom valves were identified and enumerated per sample along horizontal transects using a Zeiss microscope with differential interference contrast (DIC) optics and a lOOx oil immersion objective. Diatom taxonomy was based primarily on
Cumming et al. (1995), Fallu et al. (2000), Krammer and Lange-Bertalot (1986-1991) and Antoniades et al. (2008). Raw diatom counts were converted to relative abundance data for each sample based on the total number of identified diatoms in each sample.
Diatom diversity was expressed for each interval using Hill's N2 diversity index (Hill
1973).
Diatom and chironomid stratigraphic diagrams were produced using C2 v 1.4.1 software (Juggins 2003). Relative abundances of chironomid taxa were plotted in rank order of the axis 1 species scores when a RDA ordination was singly constrained to mid summer surface water temperature (MSSWT). Biostratigraphical changes in diatom and chironomid assemblages were noted by creating zones using constrained cluster analysis with incremental sum of squares partitioning (CONISS, Grimm 1987) and a squared chord distance dissimilarity coefficient (ZONE vl.2, Juggins 1992). Only diatom taxa with a mean abundance of at least 1% were retained in the analyses. The fossil diatom assemblages were then analyzed by PCA using CANOCO v4.5.3 (ter Braak and Smilauer
1998) to detect the timing and magnitude of major shifts in community composition in
124 the record. For this purpose, the data were square-root transformed to stabilize variance, and rare taxa were downweighted.
A chronology was constructed using the constant rate of supply (CRS) model of
Pb accumulation. Measurements of Pb activity were performed on 20 samples spanning the length of the core, and were conducted by MyCore Scientific Inc. Percent organic matter and carbonate were determined at 0.5-cm intervals throughout the core by measuring weight loss following sequential burns of dried sediment at 550°C and 950°C
(Heiri etal. 2001).
Results
Chironomid model development
The chironomid training set included lake and pond samples that spanned multiple regions of varying latitude and mean July temperature across the eastern Canadian Arctic, as well as sites within the same geographical area (Fig 1). This large spatial array as well as the sampling of multiple lakes within the same area allowed for the targeting of localized gradients of variables that may be influenced by differences in landscape and environmental characteristics, while still maintaining a long primary gradient of water temperature, which represented a MSSWT range of 13.4 °C (from 4.92 to 18.30 °C).
Distributions of the most abundant taxa and analyses of secondary gradients are presented in Medeiros and Quinlan (2011).
A detrended correspondence analysis (DCA) found that the gradient lengths of species composition along the first two ordination axes for the training set were 2.7 and
125 2.0 standard deviation units respectively. A detrended canonical correspondence analysis
(DCCA) indicated that the gradient length of assemblages constrained to MSSWT was
2.4 standard deviation units, and temperature was the dominant gradient found in a constrained RDA ordination with a A,iA,2 of 0.93. A range of chironomid inferred temperature transfer functions were analyzed to reconstruct past MSSWT in the Baker
Lake record, including the modern analogue technique (MAT), based on the lowest error and bias (Table 1). The 2-component weighted averaging partial least squares (WA-PLS) model gave the most robust performance statistic, with predicted water temperatures closely approximating actual values (Fig 2). The transfer function generated from the
WA-PLS second component was suitable for predicting MSSWT as it had a strong
2 correlation coefficient (r jack= 0.79), low root mean square error of prediction (RMSEP =
1.41 °C) and a low maximum bias (2.38 °C). These performance statistics are consistent with other inference models based on MSSWT, where the maximum bias ranged from
2.33-4.01 °C (Walker et al. 1997; Barley et al. 2006; Francis et al. 2006; Porinchu et al.
2009a). Tests for autocorrelation following Telford and Birks (2009) and Telford and
Birks (2011) indicated that transfer functions generated were robust, and significant, against potential autocorrelation affects in the dataset. In addition, lakes with the largest residual error values in the paleo-climate model were those that had measured surface water temperatures that were higher than were expected, primarily in the southern regions
(Fig 2). This error is likely due to the fact lakes and ponds sampled in these areas had a large surface area but were relatively shallow, and hence warmer than other samples in this dataset. The removal of sites at the coldest and warmest sites from the paleo-climate
126 model only showed slight improvement to performance statistics, thus potential model edge effects are not considered to influence the paleo-reconstruction of Baker Lake.
Core stratigraphy and chronology
The uppermost section of the core (0-2 cm; 2004 AD - present day) consisted of a reddish-brown organic horizon, overlying coarse black laminated sequences (2-3 mm in thickness per unit) and grey clay. The organic content, estimated by LOI analyses, was considerably higher in this upper layer, and decreased sharply at the 2 cm mark from
>25% by dry mass, to <5% for the remainder of the record. Percent carbonate by dry
910 mass was <1% through the record. Pb dating indicated a relatively high sedimentation rate (average 864 g/m2/yr) compared to other Arctic systems (Muir et al. 2004; Evenset et al. 2007). A chronology for the core was established from ~ 1885AD (210Pb background) at a depth of 12.5 cm downcore. Chironomid assemblages
Over 7100 chironomid head capsules, representing 66 taxa, were extracted and identified at 0.5-cm intervals (average = 180; median = 172.5; range 50-370). Overall abundance and diversity of fossil assemblages was high compared to surface assemblages in most lakes in the reference collection (Medeiros and Quinlan 2011). The biostratigraphic analysis of chironomid assemblages indicated three statistically significant zones within the core (Fig 3).
The lowermost biostratigraphic zone was demarcated at 12 - 20 cm, which represents the period prior to 1906 AD. This period was characterized by high relative
127 abundances of cold stenothermic taxa such as: Heterotrissocladius gr., Abiskomyia,
Orthocladius type S, Eukiefferiella/Tventia gr., and Parakiejferiella nigra. In addition,
Pseudodiamesa, Hydrobaenus, and Protanypus were present at abundances that ranged from 1-10%, and small abundances of Micropsectra type R were found sporadically.
These taxa represent those found in the coldest lakes of the surface sediment training set
(Medeiros and Quinlan 2011). While the genus Heterotrissocladius was the most abundant group found throughout this period, the relative abundance of
Heterotrissocladius maeaeri-type 2 was noted to decline from a peak of 30% relative abundance at 18 cm downcore to 7% at 12.5 cm.
The overlying biostratigraphic zone (4-12 cm, Zone 2), marked the period from
1906-1985 AD and was characterized by a gradual decrease of cold-water taxa
(Abiskomyia, Hydrobaenus, Parakiefferiella nigra, Eukiefferiella/Tventia gr, and
Orthocladius type S). Subsequently, several taxa first appear in sediments during this zone, and increase in abundances in more recent sediments (Harnishia/Paracladopelma gr., Parakiefferiella c.f. triquetra, Chironomus anthracinus, Constempellina, Procladius).
In particular, the relative abundance of Chironomus anthracinus went from trace amounts to comprising over 10% of the chironomid community during this period.
The upper zone (4-0 cm, Zone 3) reflected the period from -1985 AD to the surface of the core (2007 AD). This period was marked by the complete disappearance of several cold-water taxa (Pseudodiamesa, Abiskomyia, Hydrobaenus, Heterotrissocladius maeaeri-type 2) from the uppermost intervals. Likewise, there was a further reduction in the relative abundance of other cold stenotherms such as: Parakiefferiella nigra,
128 Eukiefferiella/Tventia gr., and Orthocladius type S. In contrast there was a marked introduction of Cladotanytarsus mancus gr, which increased in abundance to 12% of the total chironomid community by the top sediment interval (Fig 3). Likewise, the relative abundances of Micropsectra insignilobus increased during this period.
Fossil diatom assemblages
The Baker Lake diatom flora was species-rich, with a total of 163 diatom taxa representing 28 genera, and was composed of predominantly benthic taxa with the exception of a high relative abundance of the planktonic Aulacoseira genus, particularly in the lower half of the core (Fig 4). Diatom concentration remained high and exhibited little variation throughout the core with values ranging from 5.30 x 107 valves/g to a maximum of 9.9 xlO valves/g. Therefore > 500 diatom valves were counted in each sample. It should be noted that the reliability of diatom concentration values is based on the accuracy of the age-depth model, and therefore may reflect varying rates of sediment accumulation as opposed to any climatic factors (Smol 1981). Similar to the chironomids, overall diatom abundance and diversity was high compared to many Arctic lakes (Smith
2002).
Stratigraphically constrained cluster analysis of the diatom assemblages indicated three statistically significant zones, although changes in the diatom assemblage are largely subtle throughout the core (Fig 4). Zone 1 (20 cm - 8.5 cm) spans the period prior to - 1885 AD (background 210Pb) until -1951 AD. This zone is characterized by high relative abundances of taxa indicative of cold and turbulent conditions, including the
129 heavily silicified tychoplanktonic Aulacoseira lirata (>12%), and small benthic
Fragilaria taxa (cumulatively > 15%, Pseudostaurosira brevistriata, Staurosira construens var. venter and Pseudostaurosirella pseudoconstruens).
Zone 2 (1.5-8.5 cm, ~ 1951 - 2006 AD) is characterized by a relatively abrupt and sustained decrease in Aulacoseira taxa as well as a small and gradual increase in the smaller, planktonic warmer water taxa, Cyclotella stelligera and C. rossi from 1960 AD onwards (Fig 4). The relative abundance of the cold water and nutrient poor taxa in the
Fragilaria group begins high in zone 2 (> 5 %), but declines after 1965 AD. The abundances of some of the small benthic diatoms, including Karayevia clevei,
Rossithidium pusillum and Amphora pediculus, decrease in Zone 2 relative to Zone 1, while there are increases in Zone 2 in some of the larger benthic taxa considered to be more competitive in shallower water (Ruhland 2001), including: Nitzschia perminuta, N. palea, Diatoma tenue, Encyonema silesiacum and Navicula cryptotenella.
The uppermost zone (1.5-0 cm) 2006 - 2007 AD incorporates only the uppermost two samples. This zone is characterized by a sudden 10% increase in the relative abundance of small planktonic species Cyclotella stelligera and C. rossi accompanied by a continued decrease in the heavily silicified centric Aulacoseira taxa.
This abrupt increase is accompanied by increases in species indicative of higher nutrient concentrations, such as Tabellaria flocculosa and Achnanthidium minutissimum (Hadley et al. 2010). The relative abundances of all Fragilaria taxa reach their lowest values in zone 3 with the almost complete disappearance of Staurosirella pinnata and
Pseudostaurosira pseudoconstruens (< 1% abundance) and significant decrease of P.
130 brevistriata (to < 2% abundance). The relative abundances of Encyonema minuta and
Rossithidium pusillum reach a maximum of 5% and 10%, respectively, and the relative abundances of planktonic species Cyclotella stelligera and C. rossi remain higher than had been recorded throughout the core (> 5%) with the exception of the previous sample.
While the taxonomic diversity of the diatom assemblages does not increase in the uppermost sample, as documented for many other Arctic sites (Smol et al. 2005), there are large shifts in the abundances of different functional groups throughout this record.
Ordinations of fossil diatom assemblages using PCA indicated short gradient lengths for species and samples along the first two axes (1.2 and 0.9 standard deviation units). The
PCA sample scores support the zonation above, with more pronounced increases in axis 1 scores for samples above 9 cm, and variability in axis 2 scores through zones 2 and 3 (Fig
5).
This high resolution record permits close comparison of the timing of changes in the chironomid vs diatom assemblages. In the transitions from Zone 1 to 2, chironomid assemblages shift 40-50 years earlier than diatom assemblages, and about 20 years earlier in the transition from Zone 2 to Zone 3. Based on the ordination sample scores, the magnitude of the chironomid community changes is larger as well, with a greater turnover in species composition compared to the diatom assemblages.
Chironomid-based quantitative water temperature reconstruction
The zonation of the core intervals based on the constrained cluster analysis was visible when the samples were plotted passively on a PCA diagram along with the samples from
131 the surface calibration set (Fig 6). The upper sediment intervals in Zone 3 plotted closer to the centre of the ordination, indicating that these samples are most similar to lakes
sampled in warmer regions. This shift to warmer conditions in Zone 3 is reflected by a
shift in chironomid assemblages, with the loss of many cold-water indicator taxa (Fig 3).
Conversely, the Zone 1 and Zone 2 samples, containing higher abundances of cold-water indicator taxa, were plotted to the right of the ordination diagram, indicating the most
similarity with the coldest and deepest lakes in the surface calibration set (Fig 6).
The reconstruction of MSSWT based on the chironomid assemblages reflected this reduction in the relative abundances of cold-water indicator taxa (Fig 5). MSSWT inferences were found to be relatively constant (5-6 °C) throughout Zone 1, which represented the deeper intervals of the core (pre-1906 AD). The transition from the older, colder, intervals through Zone 2 was marked by a gradual increase in inferred MSSWT (6
- 9 °C) from 1906-1985 (Fig 5). Zone 3, the uppermost zone of the core (1985-2007), was represented by a further increase in inferred MSSWT (9-11 °C). This increase also corresponded to an increase in the deviation from the recorded mean annual air temperature (RMAT) normal (1971-2000 average) recorded by the Baker Lake weather
station (Fig 7). The increase in inferred MSSWT was primarily reflected in PCA axis 1
(Fig 5), which was significantly correlated to the RMAT record (r = 0.59, p < 0.01). The transition from Zone 2 to Zone 3 was also reflected by an increase in PCA axis 1 as well as PCA axis 2 scores for this period (Fig 7).
132 Discussion
The surface sediment training set of Chironomidae assemblages throughout the eastern
Canadian Arctic allowed for the estimation of the optima and tolerances for taxa across multiple ecological gradients (Medeiros et al. 2011). While the sampling regime incorporated both within and between bioregion variability, the overall primary gradient found to influence Chironomidae community structure was water temperature. This close relationship with temperature allowed for the development of a robust inference model of
MSSWT (Table 1; Fig 2), with comparable performance statistics to existing regional models (Walker et al. 1997; Francis et al. 2006; Porinchu et al. 2009a). It is noteworthy that the inferences produced by the paleo-climate model corresponded well to the 20th century climate record from the Baker Lake meteorological station (Fig 7).
The application of the midge-based paleo-climate model to the Baker Lake core inferred that summer water temperatures increased 3-4 °C over the last century (Fig 6).
This warming was inferred by the marked decline of several Orthocladiinae (e.g.,
Oliveridia /Hydrobaenus, Abiskomyia, Orthocladius type S, Eukiefferiellia / Tventia) and Diamesinae (Pseudodiamesa) cold-water indicator taxa through Zone 2 of the core
(Fig 3). The gradual reduction of these taxa intensified after -1940 AD, with the complete extirpation of Oliveridia /Hydrobaenus, Abiskomyia, and Pseudodiamesa around 1990 AD (Fig 3), suggesting that conditions have changed within Baker Lake to no longer favour these cold-water stenotherms (Walker et al. 1991; Danks 2007). This recent loss of cold-water indicator taxa is consistent with other paleolimnological investigations of midges in Arctic systems, but at a higher resolution for post-1800 AD
133 time period (Anthropocene era) compared to previous studies (Francis et al. 2006;
Thomas et al. 2008; Porinchu et al. 2009b).
The decline of these cold-water indicators corresponded to the introduction of
several other chironomid taxa generally reflective of warmer systems (Chironomus anthracinus-type, Procladius, Zavrelia/Stempellinella gr., Constempellina, and
Cladotanytarsus mancus gr). There was also a marked introduction of Cladotanytarsus mancus gr. around 1985 AD, which increases to 12% of the total chironomid community by the top sediment interval (Fig 3). Medeiros and Quinlan (2011) found that this taxon is highly indicative of warmer systems with higher concentrations of nitrogen. Likewise, the relative abundance of the diatoms Tabellaria flocculosa and Achnanthidium minutissimum, which are indicative of higher nutrient concentrations (Hadley et al. 2010), increased in the upper intervals of the core, suggesting that increased temperatures may have influenced nutrient availability and trophic structure. These shifts all occurred at a time when the mean annual temperature of Baker Lake increased 2 °C from the 1971-
2000 climate normals (Fig 7).
The fossil diatom assemblages of Baker Lake deposited prior to 1950 AD (Zone
1) further support the inferences made from the chironomids. The Zone 1 assemblages
are dominated by Aulacoseira and the Fragilaria group, among other small, benthic taxa.
In modern studies from the Canadian high Arctic, high abundances of small benthic
Fragilaria species are commonly associated with cold, oligotrophic, deep lakes with extensive ice cover (Ruhland et al. 2003; Lim et al. 2008). Similarly, the dominance of
Aulacoseira spp., in the bottom sediments also suggests conditions typical of cold, Arctic
134 tundra with strong winds, decreased thermal stratification and reduced nutrient cycling
(Ruhland and Smol 2005).
A marked shift in the diatom record with an increase in planktonic taxa after 1950
AD (Zone 2), mainly reflected by an increase in Cyclotella, suggests that there has been recent climate warming. Cyclotella species have increased across Arctic and sub-Arctic regions in the late 20th century (Ruhland et al. 2003; Ruhland et al. 2008), linked to increased air temperature in relation to a longer ice-free season and/or deeper sub-surface habitats where nutrient concentrations are slightly elevated, light properties are stabilized, and thermal stratification is enhanced (Fahnenstiel and Glime 1983; Ruhland et al. 2003).
In the Baker Lake record, the increases in Cyclotella spp. are accompanied by decreases in Aulacoseira spp and frequently in Fragilaria spp. This inverse correlation between the abundances of these two planktonic genera has been noted elsewhere in the Arctic
(Sorvari et al. 2002; Ruhland et al. 2003). For example, Ruhland and Smol (2005) found that the decline in the heavy and highly silicified Aulacoseira taxa at Slipper Lake, an inland sub-Arctic tundra lake in the Northwest Territories, was likely an indication of stronger thermal stratification, and reduced mixing. Similarly, in a 50-lake dataset from the Canadian Arctic treeline, strongest thermal stratification, which was recorded in the deepest sites (14-19 m), resulted in the greatest increases in Cyclotella species (Ruhland et al. 2003). Turbulent mixing is required to keep heavier planktonic diatom valves afloat in the euphotic zone (Reynolds 1993). In deep (< 27 m) subarctic lakes of Finnish
Lapland, thermal stratification was also found to be a major driver for expansion of planktonic diatoms (Sovari et al. 2002). Increased thermal stratification can be caused by
135 increases in air temperature, length of the ice-free season, and also by a decline in wind speeds. Changes in diatom assemblages infer that these types of limnological changes may have started in Baker Lake around 1950 AD, intensifying in the most recent few years (Zone 3).
The response of Baker Lake aquatic ecosystems to long-term temperature changes is detectable despite the large size and volume of Baker Lake, perhaps because lakes with larger water volumes are less susceptible to water level fluctuations and have lower variability in solute concentrations, including nutrient concentrations, resulting in a clearer response to direct temperature effects. The large thermal capacity allows for heat to be retained, and profundal layers are not as influenced by daily temperature changes as in small tundra ponds. This idea is consistent with the findings of Abrosetti and Barbatini
(1999) who noted that large, deep, lakes have a 'climatic memory', as deep lakes are less responsive to seasonal changes, and changes in water temperature at depth better integrate changes in climate at longer time scales compared to small, shallow lakes.
Kumke et al. (2004) and Chakraborty et al. (2010) also found that long-term temperature changes are readily detectable in large lakes.
The comparison of diatom and chironomid assemblages in the Baker Lake record shows that community changes took place at different times. While chironomid communities began to shift in favour of warmer water taxa from the early 20th century, the diatom assemblages indicate the onset of changes later, around 1950 AD. Similarly, the recent intensification of community shifts begins around 1985 AD for chironomids, but not until 2005 AD for the diatoms. This difference is likely due to the fact that the
136 mechanisms in which these two indicators respond to changes in the environment differ.
Chironomids are known to be limited directly by changes in both minimum and maximum air and water temperatures depending on competitive advantages during both their aquatic and terrestrial stages of their lifecycle (Butler 1982; Danks 2007). While diatoms are responsive to temperature changes, they may respond first to limnological changes such as pH, nutrient status or habitat availability (Anderson 2000). These limnological variables may be closely linked to climate, or they may be more related to catchment-scale hydrological and biogeochemical processes. As such, the increase in planktonic taxa that occurred at 8.5 cm downcore may not be directly related to air or water temperature, but perhaps related to temperature-mediated changes. These changes would include a lengthening of the ice-free season, change in degree of lake thermal stratification and associated decreases in water column mixing, and changes in the availability of photosynthetic radiation, possibly related to temperature-related changes in
DOC catchment export or in-lake DOC decomposition dynamics (Schindler et al. 1997;
Neff and Hooper 2002; Guo et al. 2007). While the diatom flora of deep lakes (> 20 m) is clearly responsive to climatic perturbation, the magnitude of community changes may be lower that what has been observed in smaller, shallower lakes (Michelutti et al. 2003;
Lim et al. 2008).
Conclusions
The application of a chironomid-based temperature inference model of MSSWT to the
Baker Lake biostratigraphy indicated a -3 °C increase in mid-summer surface water
137 temperature for Baker Lake over the last 60 years. The robustness of the model is further corroborated by the significant agreement with the instrumental temperature record available for the site. The high resolution comparison of both diatom and chironomid biostratigraphies showed differences in timing and magnitude of the response to warming at two distinct trophic levels. In addition, the sub-decadal resolution of both indicators produced a clearer biological and climate signal for recent decades relative to previous
Arctic paleo-reconstructions.
The high resolution record indicating the gradual shift of the biological indicators may be due to the large volume of Baker Lake, which masks variability from small-scale influences that may confound studies that have been applied to smaller lake systems. The large volume of Baker Lake may have also muted the diatom response. The appearance of diatom and chironomid taxa that are reflective of higher nutrient availability also occurs in the most recent sediment intervals, suggesting a corresponding shift in nutrients and trophic status with higher temperatures, as well as potentially increased thermal stratification and nutrient cycling in the lake.
138 Acknowledgements
This project was funded by NSERC Discovery Grants and Northern Research
Supplements held by RQ and SF, a NSERC Northern Research Internship (NRINT) held by AM, the Northern Scientific Training Program (NSTP), and additional York
University and University of Toronto funding for graduate student research. We are grateful to Dr. Derek Muir, Xiaowa Wang, and colleagues at the NLET water quality laboratory for the water chemistry analysis used in this study. Fieldwork assistance and support was provided by Bill Cooper, and the staff at the Nunavut Research Institute and
Nunavut Arctic College. We also thank Andy Aliyak, Raymond Biastoch, Andrew
Dunford, Milissa Elliott, Christopher Luszczek, and the Nukik family for field assistance.
References
Abrosetti W, Barbatini L (1999) Deep water warming in lakes: an indicator of climate change. J Limnol 58:1-9 Anderson NJ (2000) Diatoms, temperature and climatic change. Eur J Phycol 35:307- 314. Andrews JT, Nichols H (1981) Modern Pollen Deposition and Holocene Paleotemperature Reconstructions, Central Northern Canada. Arctic Alpine Res 14:387-408 Antoniades D, Douglas MSV, Smol, JP (2005) Quantitative estimates of recent environmental changes in the Canadian High Arctic inferred from diatoms in lake and pond sediments. J Paleolimnol, 33:349-360 Antoniades D, Hamilton PB, Douglas MSV, Smol JP (2008) Diatoms of North America: The Freshwater Floras of Prince Patrick, Ellef Ringnes and northern Ellesmere Islands from the Canadian Arctic Archipelago. Gantner Verlag, Ruggell, Liechtenstein Barley EM, Walker IR, Kurek J, Cwynar LC, Mathewes RW, Gajewski K, Finney, BP (2006) A northwest North American training set: distribution of freshwater midges in relation to air temperature and lake depth. J Paleolimnol 36:295-314
139 Birks HJB (1995) Quantitative palaeoenvironmental reconstructions. In Maddy D, Brew JS (eds), Statistical Modelling of Quaternary Science Data, Technical Guide 5, Quaternary Research Association, Cambridge, pp. 161-254 Broderson KP, Lindegaard C (1999) Classification, assessment and trophic reconstruction of Danish lakes using chironomids. Freshwater Biol 42:143-157 Brooks SJ, Bennion H, Birks JP (2001) Tracing lake trophic history with a chironomid total phosphorus inference model. Freshwater Biol. 46:511-532 Brooks SJ, Langdon PG, Heiri O (2007) The identification and use of Palaearctic Chironomidae larvae in palaeoecology. QRA Technical Guide No. 10, Quaternary Research Association, London. 276pp Butler MG (1982) A 7-year life-cycle for 2 Chironomus species in Arctic Alaskan tundra ponds (Diptera, Chironomidae). Can J Zoolog 60:58-70 Chakraborty K, Finkelstein SA, Desloges JR, Chow NA (2010) Holocene paleoenvironmental changes inferred from diatom assemblages in sediments of Kusawa Lake, Yukon Territory, Canada. Quaternary Res 74:15-22 Chetelat J, Cloutier L, Amyot M (2010) Carbon sources for lake food webs in the Canadian High Arctic and other regions of Arctic North America. Polar Biol 33:1111-1123. Cumming B, Wilson S, Hall R, Smol JP (1995). Diatoms from British Columbia (Canada) lakes and their relationship to salinity, nutrients and other limnological variables. Bibliotheca Diatomologica, Berlin Danks HV (2007) How aquatic insects live in cold climates. Can Entomol 139:443-471. Douglas MSV, Smol JP (2010) Freshwater diatoms as indicators of environmental change in the High Arctic, In: Smol JP, Stoermer E (eds) The Diatoms: Applications for the environmental and earth sciences, 2nd Edition. Cambridge University Press, Cambridge, pp 249-266 Environment Canada (1994) Manual of analytical methods. National Laboratory for Environmental Testing, Canadian Centre for Inland Waters, Burlington, ON Epler JH (2001) Identification Manual for the larval Chironomidae (Diptera) of North and South Carolina. A guide to the taxonomy of the midges of the southeastern United States, including Florida. Special Publication SJ2001-SP13. North Carolina Department of Environment and Natural Resources, Raleigh, NC, and St. Johns River Water Management District, Palatka, FL. 526pp Evenset A, Christensen GN, Carroll J, Zaborska A, Berger U, Herzke D, Gregor D (2007) Historical trends in persistent organic pollutants and metals recorded in sediment from Lake Ellasj0en, Bj0rn0ya, Norwegian Arctic. Environ Pollut 146:196-205 Fallu M, Allaire N, Pienitz R (2000) Freshwater diatoms from northern Quebec and Labrador (Canada) Bibliotheca Diatomologica 45, Berlin Fahnenstiel GL, Glime JM (1983) Subsurface chlorophyll maximum and associated Cyclotella pulse in Lake Superior, Internationale Revue der Gesamten Hydrobiology und Hydrographie 68:605-618 Francis DR, Wolfe AP, Walker IR, Miller GH (2006) Interglacial and Holocene temperature reconstructions based on midge remains in sediments of two lakes from Baffin Island, Nunavut, Arctic Canada. Palaeogeogr Palaeocl 236:107-124
140 Frey D (1988) Littoral and offshore communities of diatoms, cladocerans and dipterous larvae, and their interpretation in paleolimnology. J Paleolimnol 1:179-191 Gajewski K, Bouchard G, Wilson SE, Kurek J, Cwynar LC (2005) Distribution of Chironomidae (Insecta: Diptera) head capsules in recent sediments of Canadian Arctic lakes. Hydrobiologia 549:131-143 Grimm E (1987) CONISS: A FORTRAN 77 program for stratigraphically constrained cluster analysis by the method of incremental sum of squares. Comput Geol, 13:13-16 Guo L, Ping C-L, Macdonald RW (2007) Mobilization pathways of organic carbon from permafrost to arctic rivers in a changing climate. Geophys Res Lett 34:L13603 Hadley K, Douglas MSV, Blais J, Smol JP (2010) Nutrient enrichment in the High Arctic associated with Thule Inuit whalers: a paleolimnological investigation from Ellesmere Island (Nunavut, Canada). Hydrobiologia 649:129-138 Heiri O, Lotter AF, Lemcke G (2001) Loss on ignition as a method for estimating organic and carbonate content in sediments: reproducibility and comparability of results. J Paleolimnol 25:101-110 Hill MO (1973) Diversity and evenness: a unifying notation and its consequences. Ecology 36:427-431 Juggins S (1992) ZONE vl.2, University of Newcastle, Newcastle, UK Juggins S (2003) C2 (Version 1.4.1) user manual, University of Newcastle, Newcastle, UK Juggins S (2009) rioja: an R Package for the Analysis of Quaternary Science Data, v0.5-6 Kay PA (1979) Multivariate Statistical Estimates of Holocene Vegetation and Climate Change, Forest-Tundra Transition Zone, NWT, Canada. Quaternary Res 11:125- 140 Krammer K, Lange-Bertalot H (1986-1991). Bacillariophyceae. Susswasser-flora von Mitteleuropa 2(1-4). Gustav Fischer, Stuttgart Kumke T, Kienel U, Weckstrom J, Korhola A, Hubberten H-W (2004) Inferred Holocene Paleotemperatures from Diatoms at Lake Lama, Central Siberia. Arctic Alpine Res 36:624-634 Lim DSS, Smol JP, Douglas MSV, (2008) Recent environmental changes on Banks Island (N.W.T., Canadian Arctic) quantified using fossil diatom assemblages. J Paleolimnol 40: 385-398 Michelutti N, Douglas MSV, Smol JP (2003) Diatom response to recent climate change in a high arctic lake (Char Lake, Cornwallis Island, Nunavut). Global Planet Change 792: 1-15 Medeiros AS, Quinlan R. (2011 IN PRESS). The distribution of the Chironomidae (Insecta: Diptera) along multiple environmental gradients in lakes and ponds of the eastern Canadian Arctic. Can J Fish Aquat Sci Muir DCG, Omelchenko A, Grift NP, Savoie DA, Lockhart LW, Wilkinson P, BrunskiU GJ (2004) Spatial trends and historical deposition of polychlorinated biphenyls in Canadian mid-latitude and Arctic lake sediments. Environ Sci Technol 30:3609- 3617
141 Neff JC, Hooper DU (2002) Vegetation and climate controls on potential C02, DOC and DON production in northern latitude soils. Glob Change Biol 8:872-884 Oksanen J, Blanchet FG, Kindt R, Legendre P, O'Hara B, Simpson GL, Solymos P, Stevens MHH, Wagner H (2010) vegan: Community Ecology Package. R Package VI.17-10 Oliver DR (1971) Life history of the Chironomidae. Annu Rev Entomol 16:211-230 Oliver DR, Roussel ME (1983) The insects and arachnids of Canada, Part 11: The genera of larval midges of Canada - Diptera:Chironomidae. Agriculture Canada Publication 1746, 263pp Paul C, Ruhland K, Smol JP (2010) Diatom-inferred climatic and environmental changes over the last 9000 years from a low Arctic (Nunavut, Canada) tundra lake. Palaeogeogr Palaeocl 291:205-216 Pinder LCV (1995) The habitats of Chironomidae larvae. In: Armitage PD, Cranston PS, Pinder LCV (eds). The Chironomidae: Biology and ecology of non-biting midges. Chapman and Hall, London, pp. 107-133 Porinchu DF, Rolland N, Moser K (2009a) Development of a chironomid-based air temperature inference model for the central Canadian Arctic. J Paleolimnol 41:349-368 Porinchu DF, MacDonald GM, Rolland N (2009b) A 2000 year midge-based paleotemperature reconstruction from the Canadian Arctic archipelago. J Paleolimnol 41:177-188 Quinlan R, Smol JP (2001a) Chironomid-based inference models for estimating end-of- summer hypolimnetic oxygen from south-central Ontario shield lakes. Freshwater Biol, 46:1529-1551 Quinlan R, Smol JP (2001b) Setting minimum head capsule abundance and taxa deletion criteria in chironomid-based inference models. J Paleolimnol 26:327-342 Quinlan R, Douglas MSV, Smol JP (2005) Food web changes in arctic ecosystems related to climate warming. Glob Change Biol 11:1381-1386 R Development Core Team (2011) R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna Rainbird RH, Hadlari H, Aspler JA, Donaldson JA, LeCheminant AN, Peterson TD (2003) Sequence stratigraphy and evolution of the Paleoproterozoic intracontinental Baker Lake and Thelon basins, western Churchill Province, Nunavut, Canada. Precambrian Res 125:21-53 Reynolds CS (1993) The ecology of freshwater phytoplankton. Cambridge University Press, Cambridge. 384 pp Rieradevall M, Brooks SJ (2001) An identification guide to subfossil Tanypodinae larvae (Insecta:Diptera:Chironornidae) based on cephalic setation. J Paleolimnol 25:81- 99 Ruhland K, Karst T, Paterson A, Gregorgy-Eaves I, Smol JP, Cumming B (1999) Standard sediment sample preparation methods for siliceous microfossils (diatoms, chrysophyte scales and cycts). PEARL research laboratory, Queen's University, Kingston
142 Ruhland K (2001) Diatom assemblage shifts relative to changes in environmental and climatic conditions in the circumpolar treeline regions of the Canadian and Siberian Arctic, PhD Thesis, Queen's University, Kingston, Ontario, Canada 265 PP Ruhland K, Priesnitz A, Smol JP (2003) Paleolimnological evidence from diatoms for recent environmental changes in 50 lakes across Canadian Arctic treeline. Arctic Alpine Res, 35:110-123 Ruhland K, Smol JP (2005) Diatom shifts as evidence for recent Subarctic warming in a remote tundra lake, NWT, Canada. Palaeogeogr Palaeocl 226:1-16 Ruhland K, Patterson A, Smol JP (2008) Hemispheric-scale patterns of climate-related shifts in planktonic diatoms from North American and European lakes. Global Change Biol 14:1-15 Schindler DW, Curtis JP, Bayley SE, Parker BR, Beaty KG, Stainton MP (1997) Climate-induced changes in the dissolved organic carbon budgets of boreal lakes. Biogeochemistry 36:9-28 Smith IR (2002) Diatom-based Holocene paleoenvironmental records from continental sites on northeastern Ellesmere Island, high Arctic, Canada. J Paleolimnol 27:9- 28 Smol JP (1981) Problems associated with the use of "species diversity" in paleolimnological studies. QuatRes 15:209-212 Smol JP, Wolfe AP, Birks HJB, Douglas MSV, Jones VJ, Korhola A, Pienitz R, Ruhland K, Sorvari S, Antoniades D, Brooks SJ, Fallu MA, Hughes M, Keatley BE, Laing TE, Michelutti M, Nazarova L, Nyman M, Paterson AM, Perren B, Quinlan R, Rautio M, Saulnier-Talbot E, Siitoneni S, Solovieva N, Weckstrom J (2005) Climate-driven regime shifts in the biological communities of Arctic lakes. PNAS 102:4397-4402 Sovari S, Korhala A (1998) Recent diatom assemblage change in subarctic Lake Saanajarvi, NW Finnish Lapland. J Paleolimnol 20:205-215 Sovari S, Korhola A, Thompson R (2002) Lake diatom response to recent Arctic warming in Finnish lapland. Global Change Biol 8:171-181 Telford RJ, Birks HJB (2009) Evaluation of transfer functions in spatially structured environments. Quat Sci Rev 28:1309-1316 Telford RJ, Birks HJB (2011) A novel method for assessing the statistical significance of quantitative reconstructions inferred from biotic assemblages. Quat Sci Rev 30:1272-1278 ter Braak CJF, Smilauer P (1998) CANOCO reference manual and user's guide to CANOCO for windows: Software for canonical community ordination (version 4). Microcomputer Power, New York Thomas EK, Axford Y, and Briner JP (2008) Rapid 20th century environmental change on northeastern Baffin Island, Arctic Canada inferred from a multi-proxy lacustrine record. J Paleolimnol 40:507-517 Walker IR (1990) Modern assemblages of Arctic and alpine Chironomidae as analogues for late-glacial communities. Hydrobiologia 214:233-227 Walker IR, Smol JP, Engstrom DR, Birks HJB (1991) An Assessment of Chironomidae as Quantitative indicators of Past Climate Change. Can J Fish Aquat Sci 48:975- 987 Walker IR, Levesque AJ, Cwynar LC, Lotter AF (1997) An expanded surface-water paleotemperature inference model for use with fossil-midges from eastern Canada. J Paleolimnol 18:165-178 Wiederholm T (1983) Chironomidae of the Holarctic region. Keys and diagnosis. Part 1. Larvae. Entomologica Scandinavica Suppl. No. 19. 457 pp Wrona FJ, Prowse TD, Reist JD, Hobbie JE, Levesque LMJ, Vincent WF (2006) Climate Change Effects on Aquatic Biota, Ecosystem Structure and Function. Ambio 35:359-369
144 Tables
Table 1 Performance of each model type relating to MSWWT and chironomid variance. "% change" refers to the decline in predictive error (RMSEP) compared to a less complex model (e.g. WA models with inverse deshrinking, one-component PLS or WA-PLS models).
Inference Model Apparent Cross validation
RMSE (°C)i r RMSEP (°C)' r jack Max bias (°C) %change WA (inverse) 1.44 0.78 1.59 0.73 3.29 WA (classic) 1.64 0.78 1.72 0.74 3.36 -8.0 WAtoi (inverse) 1.40 0.79 1.69 0.70 3.01 WAtoi (classic) 1.57 0.79 1.75 0.70 3.53 -3.2 MAT 2.10 0.62 3.14 WMAT 1.99 0.65 2.96 PLS-1 1.71 0.69 1.87 0.63 3.54 PLS-2 1.30 0.82 1.58 0.74 2.97 15.6 WAPLS-1 1.44 0.78 1.59 0.73 3.27 WAPLS-2 1.03 0.89 1.41 0.79 2.38 11.8
145 Figures
Fig. 1: Location of Baker Lake coring site, and surface calibration-set sampling locations for sub-fossil Chironomidae across the eastern Canadian Arctic.
Fig. 2. The second component WA-PLS inference model for (a) the observed mid summer surface water temperature (MSSWT) versus the inferred MSSWT (°C) for the surface sediment dataset, and (b) the observed MSSWT versus residual values.
Fig. 3 Baker Lake biostratigraphy of chironomid assemblages plotted as percent abundances. Only selected taxa are shown. Taxa are arranged according to the sample scores of the RDA constrained to MSSWT, with cold-water taxa on the left of the diagram. The chronology was determined using 210Pb dating. Note that x-axis scaling varies.
Fig. 4 Diatom biostratrigraphy for the Baker Lake sediment core. Diatom abundances are plotted as percentages of the total number of valves counted. Only selected taxa are shown. Taxa are arranged by functional type, then alphabetically. Small benthics are defined as taxa generally smaller than 25 um in maximum length. Three zones were delineated using stratigraphically constrained cluster analysis (see Methods). The
910 chronology was determined using Pb dating. Note that x-axis scaling varies.
Fig. 5 Principle correspondence analysis of the Chironomidae surface calibration set with the Baker Lake core intervals run passively. Significant zones identified by the constrained cluster analysis of chironomid assemblages are indicated.
146 Fig. 6 Summary diagram of the Baker Lake stratigraphy with PCA ordination axis scores, chironomid-based inferred MSSWT, 210Pb activity, N2 is Hill's diversity index for the diatom community, and LOI (% lost at 550°C).
Fig. 7 A comparison of the Baker Lake summary biostratigraphic results to the recorded mean annual temperature (RMAT) anomaly (relative to the 1970-1990 AD average) from the instrumental record, for the period of 1950-2007 AD. The shaded area represents the upper-most significant zone in the chironomid assemblage.
147 s **^ ^C^s*^ftj)*VfoW\ ^^ Qikiqtaaluk *J>J'*J'—•' (Baffin) Res'
• Sample locations Treeline T0n
-{l| 0 S 10 20Km
• - -••• ^y>» . -i^. .,/.. n, ^ Figl
148 I I I I I I I I I I I I i i i i—i i i i i i i i i i—i i i i i 75 90 105 120135 150 165180 195 5 60 75 90 105120 135 150 165 180 195 Observed MSSWT (°C) Observed MSSWT (°C) Fig 2
149 2 2 fl & i: ] •••••II 1.1.
jJlilU.. I'll 1.1.'111. ] "«,. J ] x> • - 4 ] w i; xs ] Il ...
[ %• 3» "^ 1 •Uw^. •• « ] • I. .iV •11 •• !• • ]^- i.'.'w'iin lil..jii -MC «. 1 JJ_C: ] i.nii 1 Mi jiii^iiiM.1.11... .ii r: ]^j n*^.»yy>t^l ii hill ii.. Il ..F- vvv J JJ—•_ -••. .>_* ,•- L o X. JjL^-LijIrfllijpl'rt1*: # I •- • • 1- •• .["_
J1 1 •.u * . •« L0
jjjj-llil llMtyg. •^-^jjttlfe. l.l.-ll.. I"
J •^.
1 ... . .fe.ftBI$».- IM.. Illllll.I.ll.lll
1 • 11. • I • [|.ft I. . ].lLixii,..ii.i$'ifo.iii.. MIL nJil •». _ .. I.i,.: MM IflllMjIpM' *<&> 1 -I.- . l*.'IS .•» ] vir..<.n.in •ii.i.n.ll.ir' v A NSC L llUlilllJiilljE! ] . c: ] i...i.,i..i..i.i..r9 Fig 3 150 s s 5 4 \W \> \, 3+ -•-j.—i>„Utt.+ i i ..+ i.J„L„jJl...L:I V % J. ...n.iC^.,.1.111 i i . i i„. i I ill • :: \ \J' ul . li.rlll I • i i . i I I lii I ;; \\, 31 I.i^a^i^feitMl i I 11 i i 11 in I ::; •^ X"^ J I,,UgJliaUBifeh i J l I..I I .-i in x.|„ I I I. l.-fa. . ••: • .4 . +4 . 4,4- | \ \ V -a & Millillnliii A \ ^t. 3 • *" tAiM*j>j*> i . i i + I » i .»u t-.E \/•^ \XJ j 3J-l I »wv »*•*!, •^jU^MMStjMm afc^M • I • II.1 • 1 . ± l_Q \S\ 3litijy*hl E: UUJ I * i I i I + * \ \ J i t.iMdl»»«»*i l i i l 11 11 ti. t_L: \\\ 3l^f%HP!tfBiHi i + + ^L: ""Afc> 31 iJfcjfeEr • m + i + i i + 11 ,ii i E0 V^C\ \ J • .-fetEim 11.111 • i .1,1 % *\> 3 +i-hAt„Ml..l:„„+ i,.t.t.t I » I i |Fo + % \ \> 3. +YMtAa*H 4 + 4+4, +|+, \°\ 3 • i i_. 11.- Mi I I i I I i I . ++I % V*5 \ %> \ S, 3—- ll + + 1 . + . 4 •4 |+| + \W\. 3 uu^ifeUifiI • I• • • ii >•• I %, \. 31 i.u^ij.Lii^ 11111 + + + ii* •«• ^5/ mrr-ri "i" i i • i «i1'"*' *& ° ** ", w * •|..,..wr..—y.,,,1, ( , ,1,1,,,. ^ 8 Fig. 4 lO BY18 • „BYI6 BY10 #• , BY17 BY01 • RB01 RB02 ./!, WV02 ' BYU QB15 RB04 AV CO | ti \ BY15 RB.1^Avtl^V05 V CK02 Q807 - BL35 AV0VQ02»% /Q QS06 RAJ7. Dil/V, RS07 ••J • •/orM RA76* . • .^ # /C303* /W^RAZfjr1 CH03 BL02 SLt6 • AV03* . ^ |CHOf CR0' • RA05 CH07 CH05 AV10 CH02' *CHOf mCH0 6 CH09 • %Jww* /WJ2# -1.0 PCA AXIS 1 2.0 & f Q& ^ & <<* /> }P r*?^ ^ / ^ i& ^ ^ 153 i i i i i i i i i iiITIiiii i iTiiiiii r -4-2 0 2 4 8 9 10 11 12 8 9 10 11 12 0 Fig 7 154 CHAPTER 5: CONCLUSIONS Arctic systems are considered to be especially sensitive to environmental change due to narrow ecological thresholds limited by temperature, nutrients, precipitation, and light availability. These environmental constraints have multiple influences on the chemical, physical, and biological composition of these systems. This research aimed to identify and predict the influence of environmental change, such as increased temperatures, to Arctic aquatic ecosystems with the use of chironomids as paleoindicators. The determination of the primary environmental gradients that direct both limnological and biological variation at both the local and regional scale allowed for a robust interpretation of trends through a paleo-environmental perspective. This was achieved through sampling multiple lakes and ponds with differing local physical and landscape characteristics (e.g., depth, surface area, elevation), as well as across a regional scale where broader climatic variables were the primary determinant of invertebrate community structure. The examination of the limnology of 57 lakes and 56 ponds allowed for the analysis of gradients in both physical and chemical attributes of aquatic systems that might directly influence biological communities. The influence of these physical characteristics was found to be pronounced in the limnology of ponds, with significantly higher variation of several limnological variables compared to deeper lake systems. The relationship between nitrogen and temperature may be particularly important, as nitrogen concentrations were found to be higher in warmer, larger, shallow systems and phosphorus concentrations lower in lakes. The relationship found between temperature 155 and nutrients within and between regions may have direct influence on invertebrate communities, especially under continued climate warming. Our approach of sampling lakes and ponds from both multiple regions of varying latitude, as well as multiple lakes from within the same geographical area, was to examine both the 'pure' and interactive influences of nutrients / productivity and temperature on Chironomidae communities. The association of these cold-water adapted taxa and temperature are often used as a proxy indicator of periods of cold climate in paleolimnological studies. Our results indicated, as expected, deep lakes in colder regions of higher latitude were found to be primarily represented by chironomid communties of the cold-water adapted Diamesinae and Orthocladiinae sub-families. In contrast, lakes and ponds within the warmer and more productive areas contained species that are considered warm-water adapted taxa of the Tribe Chironomini, many of which have yet to be documented north of tree-line. These sites also had an absence of cold-water indicator taxa. While temperature was the primary gradient found to influence chironomid distributions in our dataset, several taxa were found to have higher abundances in lakes and ponds with corresponding high MSSWT as well as TN (e.g., Cricotopus intersectus and Cladotanytarsus mancus gr.). In contrast others were primarily found in colder lakes with lower TN. This nested secondary relationship with nitrogen may be of particular interest as taxa are likely influenced by coincident changes in both temperature and nutrients/productivity. These relationships may offer new insights into the ecological response of aquatic communities to changes in both these variables as a result of possible future climate change. 156 The establishment of the baseline ecological conditions for chironomid taxa currently represented across a large primary temperature gradient allowed for a surface training-set of chironomid community structure to be created. The analysis of the distribution of chironomids along this temperature gradient resulted in the development of a robust inference model of MSSWT with comparable performance statistics to existing regional models. The application of this midge based paleo-climate model to a sediment core from Baker Lake found that summer water temperatures increased 3-4 °C over the last century. This followed the marked decline of several cold-water Orthocladiinae indicator taxa after -1940 AD, which intensified with the complete extirpation of several taxa around -1990 AD. The pronounced, but gradual, reduction of cold-water taxa suggests that conditions have changed within Baker Lake to no longer favour cold-water stenotherms. This is consistent with other paleolimnological investigations of midges in Arctic systems, but at a much higher resolution for post-1800 AD time period (Anthropocene era). The reduction of these cold-water indicators corresponded to the introduction of several other chironomid taxa generally reflective of warmer temperate systems during the same period, including a marked introduction of Cladotanytarsus mancus gr. around 1985 AD. This taxon is indicative of a warmer system with higher concentrations of nitrogen. Thus, it is highly noteworthy that the shift towards warm-water indicator taxa, the introduction and expansion of Cladotanytarsus mancus gr., and the extirpation of cold-water Orthocladiinae all occurred at a time when the mean annual temperature of Baker Lake increased 1.5 °C beyond the 1971-2000 climate normals. Likewise, the paleo- 157 temperature model also indicated an increase in reconstructed MSSWT during this same period that corresponded well to that of the climate record from the Baker Lake weather station. Brooks (2006) highlighted the importance, and rarity, of significant late-Holocene paleo-climate reconstructions using chironomid indicators. While several other studies have found significant environmental change in Arctic lakes over the Quaternary (e.g., Thomas et al. 2008; Axeford et al. 2009), the shifts in the chironomid community detected for the last century were much more abrupt, and confined to the upper-most intervals of the cores. Larocque-Tobler (2009) produced a high-resolution record of the paleo-history of Lake Silvaplana, Switzerland, but found fluctuations in the temperature record over the last 400 years and a less pronounced warming signal for recent years. Reconstructions that provid evidence of late-Holocene temperature warming are also less pronounced in other areas of the Arctic (Solovieva et al. 2005; Thomas et al. 2008) than our findings of a 3 - 4 °C increase over the last century. Agreement between the instrumental record and chironomid-inferred temperature has also been shown to be strong (Larocque and Hall 2003; Larocque et al. 2009b), but the combination of significant late-Holocene climate warming matching equally pronounced warming obtained from the instrumental record has not yet been demonstrated (Brooks 2006). The high resolution and sensitivity of Baker Lake to long-term temperature changes may be more detectable due to the large size and volume of Baker Lake. Lakes with larger water volumes are less susceptible to water level fluctuations, small-scale influences from weather, and changes in solute concentrations significant enough to affect organisms, 158 resulting in a clearer response to direct temperature effects (Abrosetti and Barbatini 1999). Thus, the pronounced warming of Baker Lake over the last century demonstrated by shifts in chironomid assemblages that agrees with similar warming indicated by the instrumental record is a fundamental marker of climate warming for the Canadian Arctic. There are several future research avenues that arise from my PhD work. The further application of this model could be used for down-core reconstructions of selected lakes within other bioregions in the eastern Canadian Arctic to examine both regional temperature changes, as well as potential changes to localized environmental conditions, such as nutrient regimes. In addition, the range and introduction of several warm-water adapted Tribe Chironomini species can be examined further by targeting areas where we have identified Chironomini species within the surface sediment calibration-set that were previously unknown to occur north of tree-line. The examination of when these species were introduced could be used as a fundamental marker of the influence of warming systems on the thresholds for biological community responses. Further ecological modeling of species responses along environmental gradients could also be done using HOF modeling, segmented linear regressions, Generalized Linear Models, etc. In addition, the discovery of Chaoborus in cores on Baffin Island could be further researched to determine whether these populations are new introductions or if our knowledge of Chaoborus ecology is poorly understood. 159 References Abrosetti W, Barbatini L (1999) Deep water warming in lakes: an indicator of climate change. J Limnol 58:1-9 Axeford Y, Briner JP, Cooke CA, Francis DR, Michelutti N, Miller GH, Smol JP, Thomas EK, Wilson CR, and Wolfe AP (2009) Recent changes in a remote Arctic lake are unique within the past 200,000 years. PNAS 106:18431-18432 Brooks SJ (2006) Fossil midges (Diptera: Chironomidae) as palaeoclimatic indicators for the Eurasian region. Quaternary Sci Rev 25:1894-1910. Larocque I, Hall RI (2003) Chironomids as quantitative indicators of mean July air temperature: validation by comparison with century-long meteorological records from northern Sweden. J Paleolimnol 29:475^-93. Larocque-Tobler I, Grosjean M, Heiri O, and Trachsel M (2009a) High-resolution chironomid-inferred temperature history since AD 1580 from varved Lake Silvaplana, Switzerland: comparison with local and regional reconstructions. Holocene 19:1201-1212. Larocque I, Grosjean M, Heiri O, Bigler C, Blass A (2009b) Comparison between chironomid-inferred July temperatures and meteorological data AD 1850-2001 from varved Lake Silvaplana, Switzerland. J Paleolimnol 41:329-342. Solovieva N, Jones VJ, Nazarova L, Brooks SJ, Birks HJB, Grytnes J-A, Appleby PG, Kauppila T, Kondratenok B, Renberg I, and Ponomarev V (2009) Palaeolimnological evidence for recent climatic change in lakes from the northern Urals, arctic Russia. J Paleolimnol 33:463^182. Thomas EK, Axford Y, and Briner JP (2008) Rapid 20th century environmental change on northeastern Baffin Island, Arctic Canada inferred from a multi-proxy lacustrine record. J Paleolimnol 40:507-517. 160 APPENDIX 1: Benthic biomonitoring in Arctic tundra streams; a community based approach in Iqaluit, Nunavut, Canada. Medeiros, A.1, Luszczek, C.1, Shirley, J.2, and Quinlan, R.1 1. York University, Department of Biology, 4700 Keele St, Toronto, Ontario, M3J1P3. 2. Nunavut Research Institute, Box 1720, Iqaluit, Nunavut XOA OHO Corresponding Author: [email protected] SHORT TITLE: Biomonitoring of Arctic Streams Keywords: Arctic, streams, benthos, benthic invertebrates, biomonitoring, Nunavut, aquatic systems. Published in the journal Arctic, Vol. 64, No. 1 (March 2011) p. 59-72 Abstract Recent residential, commercial, and industrial development in the catchments of several Arctic streams has heightened the need to assess these freshwater systems accurately. It was imperative to develop methods that would be both effective at judging ecological condition of tundra streams and suitable for use by local groups. An investigation of two streams influenced by urbanization in Iqaluit, Nunavut, was carried out between July and August each year in 2007 - 09. Simple summary metrics (e.g., Shannon Index) and multivariate analysis (DCA, RD A) both demonstrated biological impairment in the benthic community at site locations downstream of urbanized portions of a local stream. This impairment was characterized by a loss of diversity and a dramatic shift of the benthic community to one dominated by chironomids from the subfamily Orthocladiinae. Elevated levels of total nitrogen (TN) and total phosphorus (TP) and several metals (Zn, Sr, Rb, Al, Co, Fe) were also found to be significantly related to benthic assemblages within these disturbed areas. This investigation also addressed taxonomic sufficiency, indicating that while family-level taxonomic identifications were sensitive enough to differentiate between pristine and impacted stream sites, a more precise taxonomic identification of the dominant benthos taxa (Insecta: Diptera: Chironomidae) to sub-family/tribe level identified a significant shift towards pollution- tolerant taxa. This higher taxonomic resolution will allow for the adaptation of protocols and the use of simple summary metrics to be effective for a community-based biomonitoring program in Arctic tundra streams. 162 Introduction Freshwater ecosystems of the Canadian Arctic are expected to be substantially altered by changes in runoff, water levels, river-ice regimes, and biogeochemistry resulting from climate warming (Oswood, 1992; Rouse et al., 1997; Wrona et al., 2005). These systems are characterized by precipitous shifts in seasonality, with productive summers and non-productive winters. Temperature, precipitation, light availability, nutrient cycling, abiotic processes, and the geographic distribution of biological communities are highly governed by this seasonality. Across the Arctic, from Alaska to eastern Nunavut, residents have observed changes in river ice conditions, run-off, flow regimes, and water levels, which have impeded access to important fishing areas and made travel more hazardous (Fox and Huntington, 2005). While these environmental changes have led to renewed focus examining the effects of climate change on Arctic aquatic systems (Rouse et al., 1997; Vincent, 2005), these systems have also been exposed to a range of other anthropogenic stressors. Residues from industrial contaminants (e.g., pesticides, and metals) transported from global sources via long-range atmospheric pathways have been detected in Arctic freshwater fish for at least the past 10 years, often at concentrations that exceed known thresholds for human health impairment (Indian Affairs and Northern Development, 2003). Increased resource exploration justifies renewed interest in aquatic systems, which are vulnerable to tailings and effluents from a growing number of mining exploration and development activities (Bailey et al., 1998; Clements et al., 2000). 163 Rapid municipal development in many Arctic communities results in a variety of disturbances, including run-off and leaching from municipal landfill sites and sewage containment areas, hydrocarbon and chemical spills (waste oil, fuel, lubricants, de-icing liquids), industrial activity, residential waste, stream channel diversion (that often accompanies road construction), and increased sedimentation from gravel haul operations. The residents of many Arctic communities drink water from local streams and rivers, as well as harvest sea-run and land-locked char (Salvelinus alpinus). As the health of community residents depends on ecosystem condition much more directly than is typical of North American settlement areas, monitoring these systems is fundamentally important for community well-being. However, our ability to accurately evaluate the combined affects of climate change, contaminants, and local development on Arctic freshwater ecosystems is made difficult by our lack of knowledge of the biodiversity and natural variability of these systems in their undisturbed state. Bailey et al. (1998) were able to successfully distinguish disturbed streams (due to mining contamination) from pristine environments in the Yukon using a predictive model and a reference condition approach. Although this approach has the potential to detect major benthic invertebrate responses, less extreme disturbances are expected to be much more difficult to quantify with simple summary metrics, and may require more robust multivariate analysis to quantify ecological condition (Reynoldson et al., 1997; Bowman and Somers, 2005). In addition, Yoder and Rankin (1998) demonstrated that errors in overly simplistic estimates of ecological condition are reduced when incorporating a 164 robust biological assessment of stream condition with an evaluation of biocriteria in conjunction with water chemistry and toxicological parameters. Although we recognize that it may be impossible to find Arctic stream sites that are truly undisturbed by global human activity, bioassessment can be conducted by comparing test sites with known or perceived disturbance to "control" sites that are less affected by local pollution or physical disturbance (Stoddard et al., 2006). Thus, biological impairment will be defined as a significant difference in the biological and chemical condition of sampling locations between areas of perceived disturbance and control locations. Several standard Canadian protocols exist to conduct this type of analysis (e.g., Canadian Aquatic Biomonitoring Network), however, many focus on family-level identifications and subsequent calculated quantitative indices that are used to assess the health of streams. Several standard Canadian protocols for conducting this type of analysis exist, such as the Canadian Aquatic Biomonitoring Network. However, many focus on using family-level identifications with quantitative indices to assess the health of streams. Such assessments, while common for temperate systems, may be problematic for Arctic tundra streams. Jones (2008) highlights the tradeoff between higher taxonomic resolution and information content and higher costs, logistics, and data quality. The problem inherent in working with reduced taxonomic detail is that detection of responses depends on sufficient taxonomic richness (Jones, 2008). The naturally low diversity in Arctic streams provides less information for distinguishing sites than in temperate systems; therefore, low taxonomic resolution (family level) may result in unacceptable loss of critical information necessary for the proper identification of stream 165 impairment. For example, one of the dominant, and most diverse, groups in the Arctic is the Family Chironomidae (Oliver and Dillon, 1997). Within this family there are several species that are specially adapted for specific habitats and environmental conditions (Pinder, 1995), and also have wide differences in their tolerance to anthropogenic pollution (Clements et al., 2000; Mousavi, Primicerio, and Amundsen, 2003). Thus, within-group variability of key taxa may provide sufficient information to be applied to a biomonitoring program within a tundra stream environment. In addition, without an understanding of the baseline biodiversity and community structure of benthic invertebrates within undisturbed Arctic systems, it may be difficult to determine ecological condition using a biological indicator approach. The difficulty, expertise, and cost involved in using species level identifications for any taxa may limit the applicability for most community-based biomonitoring programs based in the Arctic. As such, any methodology and identification of taxa must be geared towards residents with limited expertise and training, yet be powerful enough to distinguish between control sites and impacted areas. Thus, the objectives of this study were to investigate whether shifts in the composition of benthic invertebrates in Arctic streams could used to detect and quantify impairment downstream of known areas of point-source contamination, and to adapt and/or modify methods for the application of a local community-based biomonitoring program for Arctic streams. In order to address these objectives we sought to answer the following questions; (1) Is it possible to detect a response in the benthic invertebrate community downstream from sources of contamination adjacent to Arctic streams? 166 (2) What is the taxonomic precision necessary to detect responses in the benthic community with the use of simple summary metrics? (3) Can the results from these community-based methods and summary metrics be confirmed with a robust interpretation of the biological and chemical data to quantify ecological condition? During the summer months (early July - early August for 2007-2009) two local streams (Airport Creek, and the Apex River) that flow through the city of Iqaluit, Nunavut were investigated. In addition to the quantification of the ecological condition within the urbanized portions of these streams, this investigation also provides a framework for the further development of a benthic biomonitoring program that is suitable for assessing Nunavut's streams and can be carried out by locally-trained residents. Study Area The City of Iqaluit (63° 45' 8" N, 68° 33' 50" W) is located on the southern tip of Baffin Island (Fig. 1), and characterized by a large number of small shallow ponds, lakes, and high gradient streams due to the impermeability of permafrost and the underlying geology, dominated by Precambrian bedrock and glacial deposition. Catchment vegetation is sparse, and primarily consists of dwarf prostrate shrubs (e.g., Salix spp.), grasses, sedges, and a variety of tundra forbes. 167 Airport Creek Airport Creek (also known as Carney Creek) drains a small basin of approximately 8 km , and its lower reaches are exposed to industrial, military (historic), and urban impacts (Fig. 2). The municipality mines both sides of the stream banks for gravel at the northern edge of the city's industrial zone, which contributes a considerable amount of sediment to the stream. The city has also dug a trench approximately 200m long to divert the river flow around industrial areas. An abandoned military landfill and scrap yard is located close to a gravel hauling operation, and could be affecting stream water quality during the melt-season as water flows from the dumpsite into small ephemeral streams that discharge into the main channel. Airport creek has numerous inputs from additional industrial sources, including the Iqaluit airport that discharges de-icing fluids to the stream. In addition, waste oil and chemicals have reportedly been dumped into the main channel, where the creek flows through the city's industrial park. During one such incident in 2007, approximately 170 L of crankcase oil were dumped into the creek over the course of 48 hours (Neary, 2006). A commercial greenhouse (built in 2008) is situated along the stream bank upstream of the airport (site ACGMid), and several teams of sled dogs are tethered along the stream bank (Fig. 2B) at two downstream locations (sites ACDMid and ACMouth). Elevated concentrations of dissolved lead, aluminum, manganese, and iron in water samples have been reported for several years, which may be due to the numerous metal contamination points along the urbanized portion of the stream (Permaki and Decker, 2000; INAC, 2008). High concentrations of short-and medium-chained chlorinated paraffins in 168 sediment and water from several locations along Airport Creek downstream of the military landfill and gravel haul operation have also been reported (Dick, Gallaghar, and Tomy, 2010). The city has recently also started building a road to a new gravel source in Airport Creek's headwaters, which may further affect this stream's ecosystem in the future. Apex River The Apex River is located on the eastern side of Iqaluit and flows into the community of Apex. Apex is a subdivision of the city of Iqaluit and home to approximately 1500 people, many of whom hunt, fish and trap. The Apex river catchment area has been estimated at approximately 60 km and elevations within the watershed reach as high as 365 m.a.s.l. at the headwaters (Obradovic, 1986). The river flows through two gorges, which cover 4 km of the creek's 8 km length. During the open water period sea-run Arctic char (Salvelinus alpinus) feed in the estuarine areas near the river mouth. Several local families fish these char with gillnets throughout the summer period. Additionally, many residents drink Apex River water during the ice-free period (approximately June to October). The city of Iqaluit has also identified several reaches of the Apex River as the preferred alternative withdrawal site for freshwater to recharge the city's drinking water reservoir during times of low water availability (Trow Associates, 2004). Unlike Airport Creek, the Apex River does not have any known contamination points along its reaches. The main inputs to the river are limited to areas adjacent to gravel haul operations, approximately two km upstream from the river mouth, where the 169 city of Iqaluit has excavated gravel for the last -10 - 12 years to produce material for local development. The gravel excavation is primarily concentrated along the east bank of the main channel, resulting in sediment loading downstream. Stream modification is limited to an area of channelization to support a bridge that connects a road from the city to these gravel excavation sites. The city of Iqaluit also announced in late 2009 that the west bank of the river, approximately three km upstream of its mouth, will be an area for future development, including the city's new cemetery that is currently under construction. These plans for the west bank of the Apex River underscore the need to document baseline benthic biodiversity before the development occurs, which would also allow comparison between the local Apex River and Airport Creek biota Methods Benthic Invertebrate Sampling Benthic invertebrates were sampled at several locations along Airport Creek and the Apex River in Iqaluit, Nunavut Territory, Canada. Sampling was conducted between early July and August, in each of the years from 2007 to 2009. A control versus impact (C/I) sampling regime was implemented with collections occurring at sites along the mouth of each stream, at the headwaters/source, and at a midpoint location (in between the headwaters and the mouth). All sites were sampled several times during the ice-free season (from peak to base-flow periods) in order to address issues of seasonality and flow that may affect abundances and community composition. Benthic invertebrates were sampled with a kick and sweep method with the use of a 30x30 cm dip-net with a 500 um aperture mesh. One 'pool' and two riffle/run samples 170 (5-minute sampling effort each) were collected along the reach at each site. During benthic sampling, water samples were collected in pre-cleaned polyethylene bottles at 0.5 m below the water surface at all sites and treated immediately in the field, following the protocols outlined by Environment Canada (1994). Samples were then sorted in the lab for benthic invertebrates with the use of a stereomicroscope and sorting trays. The total number of individuals obtained per individual sample (pool vs. riffle/run) was often below 300, therefore the use of sub-sampling was deemed unnecessary and each 5- minute sampling effort sample was sorted and counted in its entirety. Laboratory methods Invertebrates were identified to Family using Pennak (1978) and Merrit and Cummins (1996), and placed into scintillation vials containing 95% ethanol. Ephemeroptera, Plecoptera, Trichoptera (EPT), and Chironomidae were further identified to sub-family, tribe, or genus (where possible) using a Nikon SMZ1500 stereomicroscope and identified with Morihara and McCafferty (1979), Oliver and Roussel (1983), Wiederholm (1983), Epler (2001), Rieradevall and Brooks (2001), and Brooks Langdon, and Heiri (2007). In addition, a proportion (10%) of Chironomidae individuals from these samples were mounted permanently on glass microscope slides in Euparal® mounting medium and then identified to genus. These specimen identifications were made at 400x magnification to the highest taxonomic resolution possible. As one of our main objectives is to develop a community-based biomonitoring program within small Arctic towns, some samples were identified by local personnel, including students enrolled in the Environmental Technology Program at Nunavut Arctic 171 College, Iqaluit. However, because of inconsistency noted between specimen identifications done by locally trained personnel and those done by professional researchers, this initial identification of Chironomidae to genus was conducted on only a portion of each sample. Subsequent analyses were conducted with finer-resolution Chironomidae identifications (to sub-family, genus, or species) via "obvious" diagnostic structures that the authors believe would be readily apparent to local personnel with minimal training and equipment. These identifications make the procedure suitable for a community-based approach to monitoring. Water samples collected at each sampling point for each stream were analyzed by the National Laboratory for Environmental Testing (NLET) at the Canadian Centre for Inland Waters (CCIW), Burlington, Canada (Table 2). A cooler containing the water samples for a number of sites for 2007 was lost in transit, and therefore several sites did not have corresponding water chemistry available. For these sites, we substituted water quality data from samples collected by Indian and Northern Affairs Canada (INAC) and analyzed by the Taiga Environmental Laboratory in Yellowknife, NT. Statistical Analysis Several commonly used indices of taxonomic composition (e.g., CABIN; Reynoldson et al., 2006) were calculated in order to compare between upstream, 'un disturbed' locations and those downstream of perceived urban influence. These included; Shannon Index of diversity (H'), Pielou's Evenness (J'), Simpsons Reciprocal Index (1/D'), and % EPT. Patterns observed in the benthic community were similar for all of the simple summary metrics; therefore analysis has been illustrated with the Shannon Index. Composition was summarized as mean index values of the riffle/run and pool samples from each site, and the standard error of this mean was calculated for 2008 and 2009 samples. Samples analyzed in 2007 were aggregates of the three transect samples (two riffles and one pool) from each site location; therefore composition was summarized as index values calculated from the pooled sample. The mean index values of coarse (family-level) vs. fine (higher-taxonomic-resolution) identifications for all study factors were then compared using one-way analysis of variance (ANOVA), or the Kruskal- Wallis test if a variable was not normally distributed. Multivariate Analysis A detrended correspondence analysis (DCA) was preformed in CANOCO v4.53 (ter Braak and Smilauer, 1998) on samples, with taxa identified to recommended levels of taxonomic resolution (Table 1). In order to give taxa equal weights for the DCA analysis, the percent abundance was calculated as the percent of total identifiable individuals, and square root transformed. This allowed for the observation of any intra- site comparisons (riffle/run vs. pool) as well as any gradients among taxa at the different sampling locations. Rare taxa, identified as those occurring at a relative abundance of less than 1% and in no more than two samples, were excluded from ordination analysis. Total phosphorus (TP), Total Nitrogen (TN), Pb, Mn, Mo, Rb were log transformed to normality, while nitrite-nitrate (NO2NO3), dissolved inorganic carbon (DIC), Ca, K, Na, Cr, Al, Co, Cu, Fe, Ni, Sr, and Zn were square root transformed for subsequent analysis. Individual correlations of water chemistry variables and DCA axes were then tested by the Spearman rank correlation coefficients with leave-one-out jack-knifing (rs\ack)- The 20 sampled site locations from 2007-09 were classified by Two-Way INdicator SPecies ANalysis (TWINSPAN; Hill, 1979), with pseudo-species cut levels defined as 0%, 2%, 5%, 10%, and 20% for relative abundance data of benthic invertebrates. An ANOVA was conducted on the species abundance and water chemistry variables among groups. Significant differences (P < 0.05) between TWINSPAN divisions were also tested with the use of an independent Mest of DCA sample scores from the first two axes from each site. These significant groups were used to delineate 'disturbed' sites from 'undisturbed' sites for further analysis. Direct gradient analysis was then used to determine environmental variables that explained a significant (P < 0.05) direction of variation in benthic invertebrate assemblages. The significance of each environmental variable was tested using redundancy analysis (RDA; linear models) in an ordination constrained to each water chemistry variable, performed with CANOCO using 999 unrestricted Monte Carlo permutations (reduced model). Backwards selection was conducted and included only environmental variables that were non-collinear (variance inflation factors < 10) and significant (P < 0.05) in an ordination constrained to one environmental variable at a time. Results Total abundances of retained specimens in 5-minute collections generally ranged from 110 to 1465 individuals, with the exception of the ACGMid location of Airport Creek sampled in 2008 (downstream from a private greenhouse) that contained over 4000 174 specimens (Fig. 3). Taxonomic richness ranged from 4 -16 identifiable groups at each sampling location (Fig. 4). In total, 27 taxa were identified (Table 1). Samples collected during 2007 contained fewer individuals and fewer taxa compared to 2008 and 2009 samples, likely due to colder summer temperatures that year (Figs. 3 and 4; Environment Canada, 2010). While the upstream locations of both the Apex River and Airport Creek contained similar taxonomic richness, locations downstream of the area or urban influence along the reaches of Airport Creek contained fewer taxa in samples collected in 2008 and 2009 (Fig. 4A). Total abundances also declined in these Airport Creek downstream site locations compared to the upstream site, with the exception of the ACGMid location sampled in 2008 (Fig. 3). In contrast total abundance and taxonomic richness differed for the Apex River samples for each of the three years (Figs. 3,4). Taxonomic Resolution While family-level identification of taxa did distinguish between 'disturbed' and 'control' portions of Airport Creek (Fig. 5), the increased taxonomic resolution of identifications increased the significance of comparisons for all metrics examined (Table 3) and elucidated patterns that allowed for of the identification of several sensitive taxa that were extirpated from sites in Airport Creek downstream of urbanized zones (Fig. 6). This also allowed for a comparison of intra-family differences in the dominant family for each sampling location in both Airport Creek (Fig. 6A) and the Apex River (Fig. 6B). The upstream site locations for both rivers contained a higher taxonomic richness of Chironomidae than locations within the urbanized areas. In addition, the absence of several taxa, indicating disturbances to the ecosystem, would not have been elucidated using a lower taxonomic resolution. The interpretation of metrics was also more complicated using family-level identifications, where the low evenness of dominant groups skewed the community descriptors for the headwater sites (Fig. 7B). The increased taxonomic resolution from further identification increased the ability to distinguish the extirpation of numerous chironomid taxa downstream of the numerous point-sources of pollution along the reaches of Airport Creek (Fig. 6A). Several taxa were absent in samples downstream of the urbanized zones along Airport Creek. These included; Tanytarsini, Trichotanypus spp. (Diptera:Podonominae), Pagastia, Procladius, Protanypus, Arctopelopia, Thienemanniella, Corynoneura spp., and Psectrocladius (Mesopsectrocladius) gr. As a result, the chironomid assemblages downstream, within urbanized areas, were almost entirely represented by species of the sub-family Orthocladiinae (Fig. 6A). These included Hydrobaenus spp., Cricotopus bicinctus, Cricotopus (Cricotopus) tremulus gr., Orthocladius type S, and Orthocladius (Orthocladius) gr. Several Krenosmittia spp. were also identified at the ACGMid sampling location of Airport Creek, which was characterized by a heavily modified gravel/cobblestone substrate. While samples from downstream sites along the Apex River did not contain differences in benthic assemblages as pronounced as Airport Creek locations, some taxonomic differences were observed at the midpoint sampling location, where the relative abundance of Chironomini, and Simuliidae were higher (Figs. 5B, 6B). The Tanypodinae comprised upwards of 70% of samples from the upstream sampling sites of the Apex River (Fig. 6B), but made up less than 10% of the upstream community from 176 Airport Creek (Fig. 6A). Several taxa were also exclusively found in the Apex River, including Chironomus (Diptera: Chironomidae, Tribe Chironomini), Derotanypus (Diptera: Chironomidae, Tribe Macropelopiini) and Ameletus inopinatus (Ephemeroptera: Ameletidae). Assessment Metrics The diversity metrics (Shannon Index, Simpsons Reciprocal, Pileou's evenness) all indicated that sampling sites downstream of the urbanized zone (Fig. 1) had significantly lower diversity, and were more uneven than the upstream 'control' location for all years (Fig. 7). In contrast, little difference in the metrics was observed in control versus downstream locations for the Apex River (Fig. 8). Chironomids represented the largest overall taxonomic group from 2007-2009 (Fig. 5). The upstream 'control' location of Airport Creek also contained more chironomid taxa than downstream locations (Fig. 6A), thus the Shannon Index scores were higher when the metric took this increased taxonomic resolution into account (Fig. 7A). Similarly, the upstream locations contained higher relative abundances of EPT than were found downstream for Airport Creek (Fig. 5A) and the Apex River (Fig. 5B). Although % EPT was similar for the two rivers, Airport Creek generally contained higher abundances of Plecoptera, whereas the Apex River had more Ephemeroptera. The approximate hatching and emergence dates for baetids were deduced based on the week larvae began to be represented in samples, which was observed to occur rapidly in early July (approximately July 10-15). Adult emergence was observed starting in early August during all three sampling years. Ameletus (Ameletidae) larvae were also observed in the Apex River immediately following ice melt (late June - early July), but were rarely found thereafter. Low abundances of Trichoptera and predaceous water beetle larvae (Coleoptera: Dytiscidae) were found in both rivers, comprising less than 2% of the total samples. Multivariate Analysis A robust examination of differences between site locations based on the relative abundances of benthic invertebrates was conducted in order to support the results and interpretation of the simple assessment metrics. A Two-Way INdicator SPecies ANalysis (TWINSPAN; Hill 1979) was conducted to detect major differences in the samples based on major indicator taxa. The analysis ordinates the relative abundance values of all taxa in each sample using reciprocal averaging, and creates a dichotomy using a calculated centroid line to make binomial divisions to create a hierarchical classification of sites (Hill 1979). The TWINSPAN analysis of benthic invertebrate assemblages resulted in the clear separation of Airport Creek sampling sites that were downstream of urbanized influence from the headwater control site and Apex River locations (Fig. 9). Relationships between benthic community composition and samples were then examined with the use of ordinations performed using CANOCO v4.53 (ter Braak and Smilauer 1998). The detrended correspondence analysis (DCA), a variation of reciprocal averaging (correspondence analysis), was conducted (with detrending by segments, square-root transformation of species abundance, and down-weighting of rare taxa) in order to observe the relationship between samples in ordinal space. The DCA sample scores were then used to statistically delineate groups of sites that represented specific 178 environmental relationships with the use of a one-way ANOVA of the TWINSPAN classified division. The first TWINSPAN division produced groups that had significantly different DCA Axis 1 scores (t - - 10.9, P < 0.05), while the second division produced groups that had significantly different DCA Axis 2 scores (C2/3 t = 4.58, P < 0.01; D2/3 t = -2.8, P < 0.05). Nine taxa had significantly different relative abundances based on ANOVAs of TWINSPAN divisions, with the first division representing significant differences in Orthocladiinae and Tanypodinae (Fig. 9). Subsequent ANOVAs conducted on TWINSPAN divisions, DCA axes, and benthos metrics indicated significant differences between control versus impacted sites, significant within-stream variation, and potential differences between sampling years (Table 3). When limited to a specific year's samples (2009 only), separated by habitat type, axis 1 of the DCA (Fig. 10) distinguished downstream impacted sites of Airport Creek from the up-stream 'un-disturbed' location, and from the Apex River sites. DCA axis 2 corresponded to the habitat-specific differences within the three individual transects conducted at each site location, separating 'riffle/run' transects from the 'pool' samples. Water chemistry samples indicated that concentrations of metals and nutrients were elevated downstream of the urbanized portions of Airport Creek (Table 2). These included total phosphorus (TP), total nitrogen (TN), sodium, iron, chromium, manganese, rubidium, strontium, and zinc. Water samples collected from the ACGMid site in 2008, in particular, contained much higher concentrations of N02/N03 (nitrite+nitrate), sodium, potassium, molybdenum, and strontium than the upstream samples (Table 2). 179 The concentrations of all water chemistry variables examined generally remained consistent throughout all samples of the Apex River (Table 2). Spearman correlations of water chemistry to DCA axes indicated that potassium (r = 0.78), sodium (r = 0.71), molybdenum (r = 0.73), and strontium (r = 0.74) were significantly correlated (P < 0.05) to DCA axis 1, and TP (r = 0.59), calcium (r = 0.75), magnesium (r = 0.70), TN (r = 0.54), molybdenum (r - 0.64), and strontium (r = 0.65) were significantly correlated (P < 0.05) to axis 2. A redundancy analysis (RDA) allowed for an examination of the relationship between assemblage data and water chemistry variables for each sample. The gradient length of the DCA ordination of all pooled 2007-2009 sites was 1.74 SD for axis 1, indicating species-environment relationships would be best described by linear models. Ordinations constrained to each normalized water chemistry variable in our dataset determined that variables that influenced significant (P < 0.05) amounts of variation in the species dataset were N02N03, TP, Al, Ca, Co, Fe, K, Mg, Mo, Na, Rb, Sr, and Zn. A backwards elimination process, which sequentially removed collinear variables, eliminated Ca, Na, Mg, and Sr as variables retained in the final canonical ordination. The first two RDA axes accounted for 39.7% of the variance between benthic invertebrate assemblages (Table 4), and the first two RDA axes were each found to explain significant portions of canonical variation (P < 0.001). The eigenvalues of the first axis (0.397) and second axis (0.108) explained 50.5% of the species variation (Table 4). The RDA also separated sites located downstream of the urbanized portion of Airport Creek from control sites and Apex River sites along the first axis (Fig. 11). Relationships 180 between water chemistry variables and the benthic invertebrate dataset were examined through regression coefficients of variables of the first two axes, and the interset correlations of the water chemistry variables (Table 5). The correlations indicated that RDA Axis 1 reflects the influence of a gradient in metal concentrations, while RDA Axis 2 may reflect differences between the sampling years. Discussion Biological Response The comparison of downstream sampling locations and those upstream of urbanized areas of Airport Creek indicated the severe impact from urban pollution sources on this Arctic stream. Substantial streambank and catchment modifications of several portions of Airport Creek are evident as a casual observer travels from the mouth of the river to the headwaters (Fig. 2). Disturbances from several commercial, industrial and legacy sources (closed metals landfill, abandoned military dump) along multiple portions of the river are likely responsible for shifts in the benthic community. The loss of several major taxa from assemblages (e.g., Tanypodinae, Tanytarsini, Diamesinae, Podonominae) at downstream sampling locations (Fig. 6A) is likely due to inputs of heavy metals to the stream from the local metals landfill. Mousavi et al. (2003) found that the relative abundance of Orthocladiinae was highest at sampling locations in sub arctic Norwegian lakes that contained high concentrations of heavy metal pollution. Likewise, chironomid richness has been shown to be significantly reduced in proportion to heavy metal pollution (Winner, Boesel, and Farrel, 1980; Clements et al., 2000; Mousavi et al., 2003). Stream modification, channelization, and substrate modification are also likely to have altered the benthic community within downstream locations. For example, larvae of the chironomid genus Krenosmittia, which primarily occur in streams with a gravel bed (Brooks et al, 2007), are not found at any sampling locations other than at a midpoint site where gravel inputs have altered the streambed composition. In addition to impacts that may result from contamination from the military landfill and scrap yard, Airport Creek has several other point-sources of potential contamination and impairment of the benthic community. The particularly high abundance of chironomids observed at the AC8GMid sampling site is likely due to the elevated concentration of nutrients that were found at this location (Fig. 3, Table 2). This site is directly downstream from a small private greenhouse situated on the north bank of the stream, approximately 5 m from the water's edge. The operator of this greenhouse also diverts water from the stream through a crude hose and pump to water plants. It is likely that runoff enriched with nutrients from plant fertilizers used in the greenhouse has leached into the stream, as there were elevated concentrations of nutrients commonly found in plant fertilizers (e.g., molybdenum, and nitrogen) within water samples from this location in 2008 compared to 2007 (before the construction of the greenhouse). Benthic samples conducted directly upstream from the greenhouse contained normal abundances of all taxa compared to other samples from Airport Creek, and did not have elevated concentrations of any nutrients (Table 2). Since downstream samples did not contain high specimen abundances (Fig. 3), the largest impact of the greenhouse may be localized to a small portion downstream from the greenhouse where available nutrients are readily mobilized by the localized benthic and microbial communities. 182 Taxonomic Sufficiency and Simple Summary Indices Several species of chironomids are known to be sensitive to the types of stressors found along the urbanized portions of the streams in Iqaluit (Clements et al., 2000; Porinchu and MacDonald, 2003; Mousavi et al., 2003), and are often the dominant taxa within these systems. Although it is a time-consuming process, the further identification of the Chironomidae to higher taxonomic resolutions may be necessary in order to describe specific differences in the benthic community response to point-sources of anthropogenic pollution that would account for the extirpation of several taxa (e.g., Tanytarsini, Tanypodinae). Jones (2008) indicates that dominant high-information taxa often warrant more taxonomic precision; therefore, change within chironomid assemblages could be the primary indicator for the disturbances within the urbanized portions of both Iqaluit streams. As future economic development in Nunavut may influence water quality in other communities, a pan-Nunavut biomonitoring program should include further identification of Chironomidae specimens. Diversity and abundance indices (e.g., the Shannon Index, Simpson's Reciprocal Index) are attractive to community-based biomonitoring programs because they allow minimally trained personnel to manipulate complex ecological data and generate values that can be mathematically compared to reference sites (Jones et al., 2005). Although these metrics indicated a disturbance to the benthic community at downstream locations of Airport Creek within urbanized areas (Fig. 3), the power of these metrics was substantially increased when the taxonomic resolution was increased (Fig. 7; Table 3). For example, identifications made at a higher taxonomic resolution (Table 1) indicated 183 specific impacts to the benthic community composition, as samples were shown to skew towards pollution-tolerant Orthocladiinae chironomids (Fig. 6). This shift in benthic composition is only discernable when the diversity metrics are of sufficient taxonomic resolution that would reflect the lower diversity, and lower evenness in sampling sites downstream of the urbanized zone of Airport Creek. Multivariate Analysis While the large shift in benthic taxa downstream of urbanized areas in Airport Creek was observed with the use of simple summary metrics (Figs. 5-8), the use multivariate techniques allowed for a robust examination of significant relationships between biological and chemical data. The Apex River and Airport Creek are different systems with their own unique disturbances, however, both the TWINSPAN analysis (Fig. 9) and the DCA ordination of sampling sites indicated that the benthic communities of both rivers responded to a gradient of disturbance (DCA axis 1). The DCA analysis indicated that there was a significant difference in DCA axis 2 scores when comparing 2007 and 2009 sampling years (Table 3). This is consistent with Milner et al. (2006), who also found that annual and inter-annual variability made biomonitoring of Alaskan streams difficult to interpret with current protocols, therefore making it difficult to compare samples conducted over subsequent sampling years. For example, when restricting the analysis to only 2009 samples, the second axis of the DCA indicated a separation of habitat types, with the upstream 'un-disturbed' sampling location of Airport Creek grouping alongside sampling locations of the Apex River (Fig. 10). Downstream 184 samples from Airport Creek did not separate along DCA axis 2, indicating insensitivity to localized habitat characteristics (riffles vs. pools). The combination of the TWINSPAN analysis and ordination techniques (DCA, RDA) successfully separated the site locations where a loss of diversity, and dramatic shift to a benthic community dominated by Orthocladiinae at locations downstream of urbanized portions of Airport Creek have occurred. Thus, our analysis indicated that the identification of Chironomidae to sub-family/tribe allowed for the separation of anthropogenically-impacted reaches of Airport Creek from minimally disturbed areas. In addition, the RDA of sites suggests that portions of Airport Creek within the urbanized zone of Iqaluit have significantly different water quality than the 'control' sites and the Apex River for several key parameters that may be influencing benthic invertebrate assemblages (Fig. 11). This separation of sampling locations in these disturbed areas with the use of TWINSPAN and DCA analysis, in combination with constrained ordinations indicating the relationship with elevated metal and nutrient concentrations, confirm the relationships seen with the simple summary metrics that will be targeted to local community groups to conduct and monitor local resources. Conclusions and Recommendations The core objective of this project was to be able to evaluate the ability to conduct biomonitoring within Arctic tundra streams that may be influenced by localized point- source pollution. Our results indicated that simple taxonomic summary metrics are able to detect changes in stream assemblages with small adjustments to widely used 185 biomonitoring protocols (e.g., CABIN). The power of these biotic metrics to discriminate sites, and specific relationships to contamination, is improved by more precise taxonomic identifications of the Chironomidae (Table 3, Fig. 6). Specific intra-family benthic- invertebrate responses to point-source pollution were observed (shifts towards pollution tolerant Orthocladiinae species of chironomids). As such, we have recommended a level of taxonomic precision that should both be statistically significant for the quantification of stream health, and feasible for local groups to carry out with limited training and expertise on Arctic benthos (Table 1). The examination of these data through the use of multivariate assessment methods also supported the results and interpretation of the simple assessment metrics that would be applicable to localized biomonitoring. The application of these multivariate analyses allows for a robust examination of the data for further scientific review and confirmation of the results beyond the target audience of local community members. While it will likely be necessary to further revise the field and laboratory protocols, and to develop additional reference materials based on how consistently and accurately the protocols can be employed by community members, it would be worthwhile to proceed with wider scale sampling of stream benthic invertebrates across several streams that may be influenced by local development. In addition, further additions to the baseline knowledge of benthic assemblages in Arctic tundra systems, and their potential response to anthropogenic disturbances, would allow for a broader examination of the biocriteria for indices used in future bioassessment studies. 186 Although the scope of this study did not include ecotoxicological approaches, the relationship between biotic degradation and increased concentrations of metals and nutrients indicate that there is a strong association between benthic invertebrates and anthropogenic pollution along the urbanized reaches of Airport Creek (Fig. 11). It would be worthwhile to conduct further toxicological analysis in the future to determine whether additional compounds may also be influencing the benthic invertebrate community at sites along the urbanized portions of Airport Creek. Additionally, it would also be interesting to explore the possibility that these species may suffer detrimental effects from contaminants at concentrations lower than current water quality guidelines (largely based on temperate zone data) due to being at the northern threshold of their ranges in the extreme environmental conditions that represent this region of the Arctic. Acknowledgements This project was funded by a NSERC Discovery Grant and NSERC Northern Research Supplement held by RQ, NSERC Northern Research Internships (NRINT) held by AM & CL, the Northern Scientific Training Program (NSTP), and additional York University funding for graduate student research. We are grateful to Dr. Derek Muir, Xiaowa Wang, and colleagues at the National Laboratory for Environmental Testing (NLET) for water chemistry analysis. We also thank the Department of Indian and Northern Affairs Canada (INAC) for providing INAC Iqaluit water chemistry data. Fieldwork assistance and support was provided by the staff of the Nunavut Research Institute, Nunavut Arctic College, Milissa Elliott, Andrew Dunford, and Roger Ell. We 187 would also like to thank anonymous reviewers, David Barton and Donna Giberson for providing comments that improved the manuscript, and Donna Giberson for aiding in some of our specimen identifications. References BAILEY, R.C., KENNEDY, M.G., DERVISH, M.Z., and TAYLOR, R.M. 1998. Biological assessment of freshwater ecosystems using a reference condition approach: comparing predicted and actual benthic invertebrate communities in Yukon streams. Freshwater Biology 39(4):765-774. BOWMAN, M. and SOMERS, K. 2005. Considerations when using the reference condition approach for bioassessment of freshwater ecosystems. Water Quality Research Journal of Canada 40(3):347-360. BROOKS, S.J., LANGDON, P.G. and HEIRI, O. 2007. The identification and use of Palaearctic Chironomidae larvae in palaeoecology. QRA Technical Guide No. 10, Quaternary Research Association, London. CLEMENTS, W.H., CARLISLE, D.M., LAZORCHAK, J.M., and JOHNSON, P.C. 2000. Heavy metals structure benthic communities in Colorado mountain streams. Ecological Applications 10(2):626-638. ENVIRONMENT CANADA. 1994. Manual of analytical methods. Canadian Centre for Inland Waters, Burlington. ENVIRONMENT CANADA. 2010. 2010. National climate data and information archive. http://climate.weatheroffice.gc.ca/climateData/canada_e.html DICK, T.A., GALLAGHER, C.P. and TOMY, G.T. 2010. Short- and Medium-Chain Chlorinated Paraffins in fish, water and soils from the Iqaluit, Nunavut (Canada), area. World Review of Science, Technology, and Sustainable Development 7(4):387-401. FOX, S., and HUNTINGTON, H. 2005. The Changing Arctic: Indigenous Perspectives. pp62-95. In Arctic Climate Impact Assessment. Cambridge University Press, 1042pp. EPLER, J.H. 2001. Identification Manual for the larval Chironomidae (Diptera) of North and South Carolina. A guide to the taxonomy of the midges of the southeastern United States, including Florida. Special Publication SJ2001-SP13. North Carolina Department of Environment and Natural Resources, Raleigh, NC, and St. Johns River Water Management District, Palatka, FL. 526 pp. HILL, M.O. 1979. TWINSPAN. A Fortran program for arranging multivariate data in an ordered two-way table by classification of the individuals and attributes. Cornell University, Ithaca, NY. 188 INDIAN AFFAIRS AND NORTHERN DEVELOPMENT. 2003. Canadian Arctic Contaminants Assessment Report II. Highlights. Ministry of Public Works and Government Services, Ottawa, Ontario. 118pp. INAC. 2008. Iqaluit baseline water quality monitoring program data 2005/2006; 2007/2008. Indian and Northern Affairs Canada, Iqaluit, Nunavut. JONES, C, SOMERS, K.M. CRAIG, B., and REYNOLDSON, T. 2004. Ontario Benthic Biomonitoring Network Protocol Manual. Dorset, Ontario. Ontario Ministry of Environment and Environment Canada. JONES, F.C. 2008. Taxonomic sufficiency: The influence of taxonomic resolution on freshwater bioassessments using benthic macroinvertebrates. Environmental Reviews 16(l):45-69. MERRIT R.W. and CUMMINS K.W. (eds). 1996. An Introduction to the Aquatic Insects of North America, 3rd edn. Kendall/Hunt Publishing Co., Dubuque, Iowa. MILNER, A.M., CONN, S.C., and BROWN, L.E. 2006. Persistence and stability of macroinvertebrate communities in streams of Denali National Park, Alaska: Implications for biological monitoring. Freshwater Biology 51(2):373-387. MORIHARA, D.K. and MCCAFFERTY, W.P. 1979. The Baetis larvae of North America (Ephemeroptera:Baetidae). Transactions of the Entomological Society 105(1):139-221. MOUSAVI, S.K. PRIMICERIO, R., and AMUNDSEN, P-A 2003. Diversity and structure of Chironomidae (Diptera) communities along a gradient of heavy metal contamination in a subarctic watercourse. The Science of the Total Environment 3O7(l-3):93-110. NEARY, D. 2006. On the lookout for polluters. News North, June 12, 2006. OBRADOVIC, M. 1986. An Isotopic and Geochemical Study of Runoff in the Apex River Watershed, Baffin Isalnd, NWT. MSc. Thesis: University of Windsor. Windsor Ontario, Canada. OLIVER, D.R. and ROUSSEL, M.E. 1983. The insects and arachnids of Canada, Part 11: The genera of larval midges of Canada - Diptera: Chironomidae. Agriculture Canada Publication. 1746, 263pp. OLIVER, D.R. and DILLON, M.E. 1997. Chironomids (Diptera: Chironomidae) of the Yukon Arctic North Slope and Herschel Island, pp. 615 - 635 in H.V. Danks and J.A. Downes (Eds.), Insects of the Yukon. Biological Survey of Canada (Terrestrial Arthropods), Ottawa. OSWOOD, M.W., MILNER, A.M., and IRONS, J.G. 1992. Climate change and Alaskan rivers and streams, In: Firth, P. and Fisher, S.G., eds. Global change and Freshwater Ecosystems. Springer-Verlag, New York. Pp. 192-210. PENNAK, R.W. 1978. Fresh-water Invertebrates of the United States. Second Edition. John Wiley & Sons. 803pp. PERMAKI, L.A. and DECKER, J.F. 2000. Lead in soil and sediment in Iqaluit, Nunavut, Canada, and links with human health. Environmental Monitoring and Assessment 63(2):329-339. 189 PORINCHU, D.F., and MACDONALD, G.M. 2003. The use and application of freshwater midges (Chironomidae: Insecta: Diptera) in geographical research. Progress in Physical Geography 27(3):378-422. PINDER, L.C.V. 1995. The habitats of Chironomidae larvae. In: Armitage, P.D., Cranston, P.S., and Pinder, L.C.V (eds). The Chironomidae: Biology and ecology of non-biting midges. Chapman and Hall, London, pp. 107-133. REYNOLDSON, T.B., LOGAN, C, PASCOE, T., and THOMPSON, S.P. 2006. CABIN (Canadian Aquatic Biomonitoring Network) Invertebrate Biomonitoring Field and Laboratory Manual. National Water Research Institute, Environment Canada, Ottawa. REYNOLDSON, T.B., NORRIS, R.H., RESH, V.H., DAY, K.E., and ROSENBERG, D.M. 1997. The Reference Condition: A Comparison of Multimetric and Multivariate Approaches to Assess Water-Quality Impairment Using Benthic Macroinvertebrates. Journal of the North American Benthological Society 16(4):833-852. RIERADEVALL, M., and BROOKS, S.J. 2001. An identification guide to subfossil Tanypodinae larvae (Insecta:Diptera:Chironomidae) based on cephalic setation. Journal of Paleolimnology 25(l):81-99. ROUSE, W., DOUGLAS, M.S.V., HECKY, R., KLING, G., LESACK, L., MARSH, P., MCDONALD, M., NICHOLSON, B., ROULET, N. and SMOL, J. P. 1997. Effects of climate change on the freshwaters of Arctic and sub Arctic North America. Hydrological Processes, ll(8):873-902. STODDARD, J., LARSEN, D., HAWKINS, C, JOHNSON, R. and NORRIS, R. 2006. Setting Expectations for the Ecological Condition of Streams: The Concept of Reference Condition. Ecological Applications, 16(4): 1267-1276. TER BRAAK, C.J.F. and SMILAUER, P. 1998. CANOCO reference manual and user's guide to CANOCO for windows: Software for canonical community ordination (version 4). Microcomputer Power, New York. WINNER, R.W., BOESEL, M.V. and FARRELL, M.P. 1980. Insect community structure as an index of heavy metal pollution in lotic ecosystems. Canadian Journal of Fisheries and Aquatic Sciences 37(4):647-655. WIEDERHOLM, T. (Ed). 1983. Chironomidae of the Holarctic region Keys and diagnosis. Part 1. Larvae. Entomologica Scandinavica Suppl. No. 19. WRONA, F.J., PROWSE, T.D., REIST, J.D., HOBBIE, J.E., LEVESQUE, L.M.J., and VINCENT, W.F. 2005. Climate Change Effects on Aquatic Biota, Ecosystem Structure and Function. Ambio 35(7):359-369. YODER, CO., and RANKIN, E.T. 1998. The role of biological indicators in a state water quality management process. Environmental Monitoring and Assessment 51(l):61-88. 190 Tables Table 1 List of taxa found in Airport Creek and the Apex River from June-September, 2007-2009. Taxa that were retained for our analysis (rare taxa were removed) are indicated, including the recommended taxonomic precision for community-based biomonitoring. Recommended taxonomic Taxon Retained precision Athericidae Family Athericidae Ameletus inopinatus Family Ameletidae Acerpenna pygmaea 191 Table 2 Water Chemistry for Airport Creek (AC) and the Apex ] (AP) 2007 - 2009. * indicates substituted INAC data (see methods for details). NO2NO3 = nitrate + nitrite, TP = total shorus, TN = total nitrogen, DIC = dissolved inorganic carbon. 2 "X TP TN DIC Ca MgB K Na Al Cr Co Cu Fe Pb Mn Mo Ni Rb Sr Zn N03 Sites Ug/L ug/L fig/L mg/L mg/L mg/L mg/L mg/L ug/L ftg/L ug/L ug/L fjg/L ug/L fig/L ug/L ug/L ug/L ug/L ug/L AC9Head 71 1.9 175 9.1 14.8 1.82 0.16 0.86 13.2 0.05 0.02 1.69 45.3 0.31 1.5 0.24 0.33 0.29 22.2 2.83 AC9GMid 37 2.8 104 8.5 12.2 1.67 0.51 1.89 15.4 0.10 0.02 1.10 15.2 0.02 1.3 0.41 0.13 0.39 26.4 2.22 AC9DMid 38 18.8 118 7.7 9.9 1.25 0.30 1.16 52.9 0.18 0.08 3.09 110.0 0.13 8.7 0.25 0.23 0.38 17.9 4.40 AC9Mouth 43 7.4 184 11.9 17.2 2.13 0.56 2.51 29.1 0.16 0.12 1.21 161.0 0.07 41.9 0.46 0.22 0.54 32.5 7.77 AC8Head 120 5.0 120 0.5 31.0 3.10 0.20 1.40 4.7 0.01 0.20 0.50 5.0 0.01 0.3 0.20 0.10 0.30 28.5 2.30 AC8GMid 217 6.2 319 17.1 24.5 3.77 1.46 8.32 27.5 0.18 0.04 1.74 39.9 0.08 1.9 1.32 0.18 0.86 60.1 1.63 AC8DMid 140 20.0 290 3.3 20.6 1.80 0.20 1.30 25.0 0.01 0.20 0.90 428.0 0.20 26.4 0.30 0.60 0.50 31.0 3.60 AC8Mouth 310 5.0 360 2.0 38.9 5.10 1.00 6.40 19.7 0.20 0.30 1.50 425.0 0.01 132 0.60 0.50 0.80 70.8 16.3 *AC7Head 50 0.5 130 0.2 10.1 1.30 0.10 0.60 9.7 0.01 0.10 0.70 50.0 0.01 0.7 0.20 0.20 0.20 12.8 1.10 AC7DMid 17 1.4 107 3.6 5.7 0.71 0.21 0.71 26.4 0.20 0.05 1.80 44.0 0.01 1.8 0.25 0.20 0.77 27.6 1.99 *AC7Mouth 31 0.5 36 1.4 17.4 2.40 0.40 2.10 44.1 0.01 0.10 1.20 209.0 0.01 49.0 0.40 0.30 0.40 27.1 5.60 AP9Head 18 0.7 95 4.8 6.9 0.84 0.16 0.70 18.4 0.06 0.03 0.42 16.4 0.01 2.3 0.08 0.25 0.34 11.6 0.60 AP9Mid 15 1.2 103 4.1 6.0 0.77 0.13 0.52 8.5 0.02 0.01 0.52 13.7 0.01 2.6 0.12 0.15 0.30 10.5 0.47 AP9Mouth 14 1.2 84 4.2 6.0 0.75 0.14 0.54 12.6 0.05 0.01 0.49 19.4 0.01 1.1 0.09 0.17 0.31 10.4 0.29 AP8Head 58 3.6 191 5.5 8.6 1.10 0.20 0.82 19.4 0.08 0.08 0.93 65.2 0.03 4.9 0.13 0.24 0.45 15.1 1.90 AP8Mid 74 1.1 166 6.2 9.2 1.26 0.18 0.81 11.4 0.06 0.03 2.17 42.5 0.05 5.9 0.19 0.18 0.42 16.5 0.35 AP8Mouth 47 4.1 116 6.0 9.2 1.21 0.16 0.84 17.1 0.12 0.02 1.04 45.0 0.03 1.4 0.16 0.23 0.45 16.7 2.18 AP7Head 89 3.1 166 5.2 8.1 0.96 0.18 1.16 42.5 0.11 0.15 2.70 76.3 0.21 9.7 0.08 0.29 0.37 12.9 3.79 AP7Mid 59 2.4 134 5.3 7.9 0.92 0.17 0.98 29.5 0.01 0.07 3.33 68.4 0.09 7.3 0.09 0.20 0.38 13.0 1.46 *AP7Mouth 10 0.5 90 0.4 7.0 1.00 0.10 0.60 30.0 0.80 0.10 0.60 50.0 0.50 1.6 0.10 0.20 0.20 10.2 10.0 Table 3 Summary of one-way ANOVA analysis of metrics derived from Airport Creek and Apex River samples in 2007 - 09. Fl and PI denote higher taxonomic resolution of identifications; F2 and P2 denote lower (family) taxonomic resolution of identifications. Abbreviations; C/I = Control versus impacted sites, Stream = difference between Airport Creek and the Apex River sites, Years = difference between 2007, 2008, and 2009 sampling years, Habitat = difference between pool vs. riffle/run samples. Bold type indicates statistically significant values (p < 0.05). An asterisk (*) indicates a significant difference between 2009 and 2007 (Tukey post hoc test). Factor Metric Fj Pi F2 P2 C/I DCA AXl 118.99 < 0.001 23.04 < 0.001 DCA AX2 2.70 0.12 2.94 0.10 H' 50.44 < 0.001 6.94 0.02 D' 15.59 < 0.001 12.30 < 0.001 Stream DCA AXl 26.12 < 0.001 7.73 0.01 DCA AX2 5.44 0.03 2.94 0.10 H' 15.97 < 0.001 5.46 0.03 D" 15.92 < 0.001 6.34 0.02 Years DCA AXl 0.37 0.70 2.23 0.14 DCA AX2 6.59 *0.01 3.28 0.06 H' 0.16 0.85 0.87 0.44 D' 0.62 0.55 1.12 0.35 Habitat DCA AXl 0.87 0.36 0.64 0.43 DCAAX2 0.36 0.55 0.27 0.61 H' 0.01 0.95 0.01 0.94 D' 0.02 0.90 0.09 0.76 193 Table 4 Eigenvalues, species-environment correlations, and cumulative % species variance of each Redundancy Analysis (RDA) axis. RDA RDA Total Axis 1 Axis 2 Variance Eigenvalues 0.397 0.108 1.0 Species-environment correlations 0.976 0.733 Cumulative percentage variance of species data 39.7 50.5 of species-environment relation 62.8 79.8 Sum of all eigenvalues 1.0 Sum of all canonical eigenvalues 0.633 Table 5 Regression coefficients of the first two RDA axes and interest correlations for backwards-selected environmental variables. * Indicates P < 0.01, ** indicates P < 0.05. Regression Interset Env. Coefficients Correlations Variable Axis 1 Axis 2 Axis 1 Axis 2 Mo -1.00* 0.68 -0.84 0.12 Rb -0.37* -0.11 -0.66 0.01 Zn -0.19 -0.85** -0.62 -0.21 K 0.50* -0.67 -0.60 -0.06 TP -0.15 0.07 -0.56 0.12 Fe 0.19 0.46 -0.52 0.05 N02N03 0.19 0.35 -0.46 0.24 Al -0.21** -0.54 -0.41 -0.46 Co -0.20 0.45 -0.43 0.15 Figures Figure 1 Location of benthic sampling locations for Airport Creek and the Apex River, Iqaluit, Nunavut Territory. Figure 2 Airport Creek, Iqaluit, Nunavut. (a) Midpoint sampling location near dog teams; (b) Downstream of midpoint sampling location. Figure 3 Total number of invertebrates collected at each location for riffle/run and pool samples for; (a) 2007 Airport Creek, (b) 2007 Apex River, (c) 2008 Airport Creek, (d) 2008 Apex River, (e) 2009 Airport Creek, and (f) 2009 Apex river. Figure 4 The total number of taxa identified in each sample for 2007-2009 for (a) Airport Creek, and (b) The Apex River. Figure 5 Mean relative abundance (%) of major taxonomic groups of benthic invertebrates from upstream (APHead), midpoint (APMid), and mouth (APMouth) sampling locations within A) Airport Creek and B) Apex River. Standard error bars of means are indicated. Figure 6 Mean relative abundance (%) of Chironomidae and Podonominae from upstream (ACHead), midpoint (ACGMid and ACDMid), and mouth (ACMouth) sampling locations within A) Airport Creek and B) Apex River. Standard error bars of means are indicated. Figure 7 Mean Shannon Index (H') values for sampling locations within Airport Creek A) with tribe/sub-family taxonomic identification of Chironomidae B) and Family level 196 taxonomic identification. Standard error bars of means are indicated for 2008 and 2009 samples. Figure 8 Mean Shannon Index (H') values for sampling locations within the Apex River A) with tribe/sub-family taxonomic identification of Chironomidae B) and Family level taxonomic identification. Standard error bars of means are indicated for 2008 and 2009 samples. Figure 9 TWINSPAN classification of river benthos assemblages from Airport Creek and the Apex River from 2007-2009. Taxa listed are the TWINSPAN indicator taxa. Environmental variables showing a significant difference at a division are given: (*) P < 0.05; (**) P < 0.01; and (***) P < 0.001. Significant differences (P < 0.05) between TWINSPAN divisions via ANOVA of DCA Axis 1 (bold boxes) and DCA Axis 2 (checkered boxes) sample scores are indicated. Figure 10 Detrended correspondence analysis of 2009 samples indicating sample locations relative to taxonomic groupings. A) kick-sweep samples from 'pool' transects, B) kick-sweep samples from riffle/run transects, and C) impaired site locations. Sites; • APHead, • APMid, • APMouth, a ACHead, A ACGMid, VACDMid, o ACMouth. Figure 11 Redundancy analysis indicating sample locations relative to significant (p = < 0.05) water chemistry variables. Abbrv; • TWINSPAN group D2, ^TWINSPAN group D3, • TWINSPAN Group C3 - Midpoint, • TWINSPAN Group C2, 1. Acentrella, 2. Acerpenna, 3. Dytiscidae, 4. Empididae, 5. Hydrachnidae, 6. Oligochaeta, 7. Plecoptera, 8. Simuliidae, 9. Tipulidae, 10. Trichoptera, 11. Diamesinae, 12. Orthocladiinae, 13. Tanypodinae, 14. Chironomini, 15. Tanytarsini, 16. Trichotanypus, 17. Corynoneura. 197 Baffin Region •^ Sample Location Rivers Landfill •i Airport I I Urban Zone 0 O.S 1 2 Km _j i i i * ^ i i ^ Fig.l 198 Fig. 2A Fig. 2B 199 a. 2007 b.2007 1000 1000 £ 800 £ 600 1 400 •g £ 200 f <# & J? 1? 6* v* / c. 2008 d. 2008 4087 1000 • Pool 1000 DPool J 800 E | 600 1 400 200 •> »** > ^ c? ^r DPool £ & 4? J? *° Fig. 3 200 a. 20 -i -2009 - 2008 2007 S15 CO fc10 E 3 5H 0 *>6 ^6 vOw # F 0 r^ b. 20 l —D—2009 - A- • 2008 *•"•— 2007 E i 5 0 • / S Fig. 4 201 Q) Relative Abundance (%) Relative Abundance (%) ore _-k|>0W^010>-vl00CDO OOOOOOOOOOO \ ° SO O I O 5! ?! * C =? =! OJ 22 ° 3F 5 Q. Q. Q. Relative Abundance (%) Relative Abundance (%) ->Mu^oiaNaoioo X .JMU»01fl)S00fflO OOOOOOOOOOO OOOOOOOOOOO • • • 2 2 £ =f Q. 3 Shannon Index (H') 03 ere* o > o X CD 01 Q. 5 o 5! > O D > O <: ^ N3 M M o OOO ooo N 00 IO Shannon Index (H') o -*. c5 Ol —^ ui to 5 X CD 01 a. « i » '' > # y aO Q.' v » > o \ • o 2 S 1 /* '/ * 1 ' i S* >^ t 2 rim b •*9 LJ*J •• ro• l>0' NIi o ooo ooo O c ~j oo APHead APMid APMouth Fig. 8 TP*, Ca*, Mg*, K**, Na**, Al*, Fe*, Mn*, Mo***, Rb***, Sr***, Zn* Tanypodinae Orthocladiinae NOj/NO,*, Ca*, TN** Corynoneura Acentrella Acentrella Tanypodinae Plecoptera Acerpenna Trichotanypus L. ; C2 : J C3 ! ! D2 ! D3 , i N=7 i i N=5 i I N=4 I I N=4 i i i i i i • AP9Mid AC9Head AC9GMid AC9Mouth AC7Head AP9Head AC9DMid AC8GMid AP7Head AP9Mouth AC7DMid AC8DMid AC8Head AP7Mid AC7Mouth AC8Mouth AP8Head AP7Mouth AP8Mid AP8Mouth Fig. 9 Trichoptera + + Oligochaeta Dytiscidae + + Chironomini Tanytarsini + Trichotanypus sp. + B Tanypodinae** A#d| Tipulidae Empididae V^QD Orthocladiinae A ** VC^ - A . ... , + Hydrachnidae _. + Plecoptera Ephemeroptera Diamesinae Simuliidae^ ^^rynoneura sp. -2 DCA AXIS 1:31.9% .10 ACSDMid* AC8Head« AC7Head 00 NO„NO AC8Mouth Co CM Mo TP AP9Mld W AS^»^ . AP9mm < Q AC9Mouth 4 ^ ACSGMtf A 13 APSHead * AP7Hmd ACBDUid Al APTMouth I -1.0 RDAAXIS 1:39.7% 1.0 Fig. 11 206 APPENDIX 2: Raw water chemistry and environmental data "1'° PCA axis 1=26.1% 1° "10 PCA axis 1=32.5% 1 Fig. 1 (Appendix) Principal components analysis for inter-regional differences among the 57 lake (solid circles) and 56 ponds (gray circles). Appendix Table 1: Limnogical and field data for lakes sampled. Site Lat Long Area Elev Zmax MSSWT PH COND ORP CI Ca Mg K Na DD° DD° ha m m °C uS cm1 mV mgL1 mgL1 mgL1 mgL1 mgL1 CH03 59 27 -95 86 2214 5 105 45 12 7 86 17 216 13 13 06 03 12 CHOI 59 41 -95 41 690 0 65 45 12 2 77 1 277 07 13 06 01 1 1 CH02 59 67 95 43 1195 1 70 26 15 1 82 3 253 22 13 07 03 16 CH15 59 73 -96 97 1256 6 221 23 189 65 9 233 04 07 04 02 06 CH09 59 79 -96 49 515 2 192 29 129 72 13 269 02 09 04 02 07 CH18 59 92 -96 51 3142 181 88 17 7 69 12 75 06 1 1 05 03 08 CH08 60 03 97 28 282 1 260 27 12 6 75 8 244 06 13 08 03 08 CH06 60 21 -96 14 1064 2 140 23 12 2 72 10 255 08 09 04 03 09 CH04 60 27 -95 62 1798 3 100 50 122 75 13 269 15 12 06 03 12 CH05 60 44 95 69 226 2 100 40 120 73 12 257 25 09 03 03 14 AV03 6124 -9419 5917 14 30 16 2 84 83 115 22 9 35 21 08 115 AV05 6126 -94 18 45 4 21 20 167 83 74 111 13 7 63 17 10 71 AV06 6127 94 24 179 0 13 20 17 3 78 144 137 24 8 41 24 10 124 AV09 6128 94 32 350 1 21 60 183 71 69 121 164 34 16 07 84 AV19 6129 -94 13 23610 10 50 194 78 116 113 23 9 49 24 13 15 0 AV18 6131 -94 18 575 6 6 25 188 79 205 93 513 58 46 20 27 4 AV17 6131 9418 4620 1 14 58 19 4 75 99 98 22 7 40 24 10 118 AV21 6132 94 46 390 8 33 40 17 9 86 16 52 20 16 04 03 13 RA17 62 84 92 22 60 6 14 55 12 3 81 255 38 30 6 27 5 43 30 16 8 RA19 62 85 -92 25 618 16 93 12 4 72 180 62 28 3 13 5 32 21 17 2 RA20 62 86 -92 29 37 4 14 25 15 5 73 92 135 10 1 14 3 19 17 61 RA15 62 86 -92 23 32 6 18 22 15 1 78 73 47 88 68 13 12 52 RA08 62 87 92 17 86 7 31 74 13 9 73 101 94 17 0 74 18 14 97 RA13 62 87 -92 21 90 8 16 87 13 6 70 74 107 102 66 13 1 1 56 RA16 62 87 92 25 30 9 18 29 15 1 76 85 96 107 80 15 13 60 RA03 62 89 -92 18 1 1 28 50 12 4 82 66 63 50 62 13 19 44 RA05 62 90 -92 21 29 0 23 35 13 3 74 85 56 83 94 12 15 55 RA06 62 90 -92 28 64 9 54 42 126 76 40 47 43 38 08 08 28 RA04 62 91 -92 22 96 0 23 30 13 1 76 78 51 85 84 13 13 48 RAH 62 91 -92 16 24 6 34 32 14 2 72 111 95 180 93 20 17 89 RA07 62 92 92 18 32 6 32 20 13 6 71 159 66 315 10 6 31 20 15 0 RA02 62 92 -92 23 29 8 13 40 15 6 71 74 133 96 96 13 13 51 RA12 62 92 92 16 262 8 32 60 15 1 71 45 94 52 44 07 08 32 IQ10 63 59 68 75 08 40 40 17 1 86 49 144 77 33 18 03 43 IQ04 63 76 68 45 88 158 90 102 79 43 165 1 1 96 10 02 08 IQ11 63 77 68 53 22 133 50 115 80 24 91 1 1 35 1 1 02 08 IQ05 63 78 -68 44 106 3 196 17 2 82 82 35 114 22 67 07 02 06 IQ06 63 78 68 54 47 134 58 10 5 83 39 146 08 96 14 01 09 IQ01 63 78 -68 55 18 175 40 10 3 80 19 123 08 44 10 01 08 IQ13 63 78 -68 56 16 125 30 85 81 50 228 10 10 3 20 02 08 IQ03 63 79 -68 53 19 155 30 114 84 27 169 07 52 07 01 07 IQ20 63 80 -68 53 98 175 25 129 84 24 147 08 45 06 01 06 IQ14 63 81 -68 56 61 161 40 105 86 47 144 10 93 08 02 07 BL01 64 24 -95 99 188700 0 13 5 10 5 72 20 144 13 24 10 04 06 BL02 64 34 -95 97 49 7 61 12 0 10 6 77 32 130 43 40 09 06 25 BL03 64 34 -95 94 15 6 70 25 110 84 48 170 10 1 50 14 07 51 BL36 64 34 -95 91 115 71 81 118 82 39 16 26 66 1 1 07 19 BL37 64 35 95 93 74 77 28 13 8 84 53 200 48 82 13 07 28 BL06 64 35 96 01 12 5 73 27 120 82 49 134 51 88 12 06 19 BL38 64 35 95 93 69 74 29 144 87 56 193 35 90 14 09 24 BL16 64 51 -96 14 300 1 77 13 0 82 80 23 252 18 38 08 04 1 1 BL29 64 58 96 27 1 1 134 50 85 88 11 59 04 17 06 04 04 BL31 64 61 -96 28 514 150 40 10 7 80 19 216 05 25 09 04 06 RB07 66 55 86 26 79 58 72 10 3 85 83 113 50 18 2 37 12 33 CR01 70 46 -68 51 715 20 45 86 81 14 144 53 05 04 03 24 CR02 70 47 -68 47 50 18 20 105 76 6 169 15 02 02 01 1 1 CR04 7048 -68 63 86 5 9 123 56 80 17 181 22 04 03 03 14 208 CO — ( O =S: CO TJ- :> CO T|- O ^D —' O OO ON O v: < N O H ^O H yj 00 —« O — O — OOOOOCSOOOO — O ^HCS — O — O— OO — — — OO — OOCS — OO — OO o J CS 60 fl 3. cocsoocooooNr>c\cs- cnoo^iNcoO'^frco __ — — OO — OOO — OOOOOO— OO o o OOOOOOOOOOOOOOOOOOOO — ooo -1 oooooooooooooooooo D 60 'coinr^eNOTtocoi>r-co^ov-iocO'^r'^D U"1 ON O'O'OCO'^-OCS^CSOOcO^O'^'^-ONCSOOOO^HCOcOr^OOO ir^ooTtoor--\ov-)Oomoo — cor^csoocoov cs cs -=:-=: oo <; in ^- — o oo — Ttcoovcor-oo- coTf-csocor^cst>eo- co — cs ^ CO CN co ^- CO -J MI a. h iri H o h cocooooNooooo\r>»or^inONO incsoocsvoinr^coooor-r^Tj-csON- _ o c- o r> a\ r--\oooaNr>oo — inovi — — rl-cncscocscscscs- csr>i/")inm>nr> -J — o — o o OO-—< — O — — CS —• CS CS CS — oooooooooooooooo oooo — ooo 60 3. oolna^ON^^v)no^oocl^SM^^^ i- oo — oooooo — ooo — oo ^ • ooooooooooooooooo -1 oooooooooooooooooooooooo .0 60 a. a. in ON r^ cs oo — o — mo-^^oocoorsom^o „ ON : ^j- cs —• — cs cs — — o — fscocococo — cor^r^ J oooooooooooooooooo Tf O o OOOOOOOOOOOOOOOOOOOo o o o o 60 z a. O^O — t^'OOOOOT}-- r^^cOCTsOOCSCS oococoONOor^oomr-fSfNr>incscsooTt'^-csoocoiocsoo ^ o^csoocsr>ONOOcoTj-»noo^o\ocS'-'CS c 60 2 r^cO^OCO^cOO^COOOCS^OONO- CSO ONHaNcno\^\oviinoin^o\ONN ^J i 'Tt- CS CS CO fri-HcS — cs - **cs 60 ft. a. — r-oocsincsomcoooeoooooooocOTj- cs — cscoo->o^ooNTj-ocot-^\ooocoiooocscnoooo — rt-r- .-.J moo- o — oooov-iooocs- — — - cscso — csoocooooo — oo — — — ocsoo 3 u >csr>cov)CScotsooooov-)inooo»nou") ooo — r>cNoocs 210 Appendix Table 4: Limnological and field data for ponds (< 2m depth) Site Lat Long Area Elev Zmax MSSWT PH COND ORP CI Ca Mg K Na DD° DD° ha m m °C nS cm] mV mg L[ mg L1 mg L1 mg L1 mgL1 CH16 59 23 -97 89 28 3 246 10 20 8 63 20 201 0 21 1 89 0 84 0 56 16 CH19 59 34 -97 52 126 263 13 20 1 66 13 147 044 134 0 63 0 19 09 CH14 59 73 -96 96 18 224 19 183 64 8 148 0 26 0 65 0 47 0 12 06 CH17 59 93 96 51 12 8 181 15 19 6 63 17 195 104 156 0 94 0 25 1 1 CH07 60 15 -96 57 159 2 194 14 122 75 9 261 0 34 1 11 044 0 25 06 CH13 60 16 -96 21 283 5 132 17 20 2 68 16 204 101 128 0 69 0 27 10 CH11 60 45 -95 70 00 96 08 20 8 71 25 30 3 37 168 0 74 0 07 24 AV01 61 15 -94 12 2518 4 15 17 9 76 1006 112 235 00 12 10 18 30 4 83 1210 AV26 6124 -94 11 69 23 15 20 4 71 70 177 9 01 2 27 108 0 63 55 AV23 6125 -94 12 35 0 20 07 187 80 65 170 10 60 5 29 178 101 57 AV25 6125 -94 12 28 9 19 10 20 1 82 182 //// 38 80 7 09 4 33 168 23 1 AV02 6125 94 16 57 0 11 05 14 1 82 90 91 21 10 4 80 2 36 126 12 2 AV15 6125 94 14 88 1 15 10 22 0 70 73 86 14 60 3 17 178 091 87 AV04 6126 -94 14 20 5 24 1 1 15 2 73 43 137 109 1150 120 011 08 AV13 6126 -9415 57 4 9 10 193 64 95 215 14 50 8 77 2 19 1 16 78 AV14 6126 94 15 512 11 08 22 0 76 120 19 18 60 9 82 2 59 106 98 AV24 6127 -94 23 54 1 20 15 197 77 138 //// 20 80 13 30 2 98 157 116 AV12 6128 -94 11 118 11 08 19 0 68 146 133 3140 9 53 3 40 108 160 AV11 6128 94 23 42 0 20 05 15 5 84 121 120 20 60 9 92 2 58 150 107 AV07 6128 -94 27 448 3 11 17 157 85 240 124 69 90 6 47 5 68 185 34 6 AV16 6129 -94 35 56 4 32 15 217 74 75 152 8 01 6 28 2 09 154 56 AV08 6129 -94 38 184 23 15 183 71 87 121 85 80 8 40 7 03 197 410 AV22 6132 94 49 72 9 19 19 187 79 77 84 15 30 3 87 195 101 92 AV20 6132 94 47 105 9 18 18 16 8 80 73 90 15 30 4 16 198 0 94 87 RA18 62 84 -92 18 14 42 04 13 0 81 179 85 12 90 26 30 2 21 2 86 72 RAH 62 87 92 19 16 8 02 119 82 256 26 44 40 17 10 413 2 52 24 5 RA10 62 87 -92 18 312 12 10 157 78 135 73 17 00 10 80 2 18 2 09 98 RA09 62 89 92 20 32 11 02 17 8 83 479 70 109 00 19 10 9 25 351 58 2 IQ18 63 79 68 54 01 159 05 12 4 86 58 149 0 90 12 30 146 0 14 09 IQ02 63 79 68 55 26 159 15 14 5 70 27 111 0 87 3 94 0 57 0 16 06 IQ17 63 80 68 55 34 172 15 86 83 22 224 0 63 451 0 73 0 09 06 IQ12 63 80 -68 56 37 171 10 10 1 86 40 97 101 8 57 1 12 0 16 08 IQ15 63 81 -68 58 69 136 12 98 86 64 183 102 14 90 133 0 18 07 BL35 64 33 -95 90 50 53 16 13 1 84 32 134 194 5 13 0 84 0 55 14 BL32 64 34 95 90 19 75 09 10 1 84 20 27 142 2 96 0 68 0 39 1 1 BL04 64 34 95 99 08 81 05 102 88 22 37 0 95 2 90 044 0 54 07 BL33 64 34 -95 91 32 71 09 78 82 18 143 150 3 01 0 59 0 37 10 BL34 64 35 -95 91 09 68 08 115 81 23 117 162 3 58 0 68 0 45 1 1 BL08 64 40 96 01 58 67 08 113 83 36 147 1 11 6 87 109 0 89 12 BL09 64 40 96 03 26 67 09 104 84 25 194 0 68 4 86 0 69 0 58 09 BL11 64 41 -96 03 20 55 15 13 9 84 36 43 2 67 6 02 0 92 0 63 17 BL13 64 45 96 07 14 8 50 10 118 84 39 117 5 75 5 17 0 98 0 55 33 BL15 64 46 96 08 50 59 12 114 81 23 155 121 4 72 0 67 0 59 10 BL14 64 46 96 09 20 1 56 12 10 0 82 22 181 173 3 72 0 79 0 43 1 1 BL17 64 52 96 16 19 81 1 1 14 6 83 77 91 17 10 6 80 177 0 53 67 BL23 64 55 -96 21 118 98 1 1 12 8 80 53 73 12 90 4 00 161 0 36 56 BL26 64 56 96 22 13 2 101 10 75 80 37 161 8 42 4 50 144 0 43 34 BL25 64 56 -96 22 3 1 106 10 146 80 41 135 7 98 4 12 132 0 41 30 BL24 64 57 96 22 11 1 97 10 113 83 66 148 17 30 6 13 2 25 0 59 62 BL28 64 57 96 25 613 113 1 1 113 81 17 196 100 2 45 0 79 0 37 06 BL27 64 57 96 22 32 103 12 72 82 37 185 8 74 4 02 181 0 36 33 RB01 66 53 86 22 58 34 05 12 8 85 142 115 9 64 24 40 5 67 162 67 RB02 66 55 86 28 17 17 10 112 84 66 132 5 83 1160 2 34 0 80 35 RB04 66 55 86 31 68 150 1 1 13 7 89 56 99 197 971 2 40 0 85 13 RB14 66 56 86 28 08 200 05 116 82 102 130 5 21 17 50 4 59 125 30 RB17 66 58 86 32 22 76 05 12 0 85 95 116 3 32 17 10 4 55 125 21 211 OOOONOOOOCOOO^O — OOOO00r--Ot>OO: oooor>inONcoTj-csoNCNco- cocscscocoinvoo-^-NOcscsaNcso cocsinooin — h-oint^-ONr^TtoNcsooococOTfNOt^ooo: or>-oo — \oino\No- r^r>ONincs^csooTtONONCocsoinoNO^or> 60 ooocscsocsincsoooocsocscs — csoo — o — — — oor>ooocs- — — oo — o — — omo — — — o — cs — o & 3 o3 Tt CS CS CO CS CI o r> cs NO <0 — OOOOO^^: o o cs cs o cs ^ cs J oooo •* o o 60 OOOOOOOOOOOOOOOOOOOOOOOO oooooooooooooooo o o o o o o o CO — o — — p 3 o — oooooocsooooocsoooooooooo: OOOO ON ON ONNONO*3-OOOONOON NO NO O — ONONcoco-^tooTf-comooooNo: ON Tf- csoo^OTT-oocSr^t^coco^cocs — ONONinoor-ooooooo r- m o cs _, ,_, — vo — coincocscSininr>in^Or>cocscsco O ON 60 3 COCSNOCOrJ-^Or>OOCSCSOCSr>— COt-^t^OOiTl-r>OCSOO: TJ- OO o o OooooooocScocscs^ooooooNoooooNOTfNOcoONor^-in: o i> in in in in 60 ^-o — OO — OCO — CSCOCS — — cscscscscscscs — cscs oooooooooo ooooooooooooo cs — cs — 0* 3 i> m t^ ^o OOOO' 60 OOOOOOOOOOOOOOOOOOOOOOOO oooooooooooooooooooooooooooo £ 3 OO^O — 0\ONf-ONcOoOirr' — CO ON NO COOONOCSNO— ONOOOOOOO — — — co —i co r> co "* •<* ^t CO CO t> r> t^ r> m m — — co cs cs — 60 OOOOOO — OOOOOOOOOOOOOOOOO oooooooooo o o o o ooooo — ooooo z 3 c» TtcS — O— NOOO^ coO'O — ONC-^ocNOTt-oinNooo^oooooooQinoN^-oin T3 or--inr^^ooNin— : •* — COCOO—'OONONONOOOcSNONOOCScSCOCOCOOOTj-^-cSOCSinON 60 c — ONOOcoo^t-^csoNcocsinNONocooorfr^inr^coin^ Tj-, 'a COONO — OOOOOOOOOTJ-O — ON — Ot^' NO in t—i TJ- ON oo incsoNt>-in- o^-- OCONONOOO — ^toooinco cs tt in NO Tt r> CS 60 NOin^ O — ^j- r-^ r- TJ- -rt- — '^oooinm — mONeS" t/1 60 OOOOOOOOOOOOOOOOOOOOOOOO oooooooooooooooooooooooooooo 3 CS < NOCOONOONCOONCOCOCOONOO- co; ^ ^ ^ oo r- •<*• in H o in NO o in cs o NO o ** i •^•OTj-ooNoONOONinooNOcs—iNOin — ONOOCSCSCSCOON _ _ .r>r>cscsONOO ^-NO' 213 Appendix 3: Raw counts of chironomid head capsules in surface sediments Appendix Table 7: Orthocladiinae of the tree-line region Site CHOI CH02 CH03 CH04 CH05 CH06 CH07 CH08 CH09 Taxon Code Abiskomyia Abisk Hydrobaenus Hydrob 1 Heterotanytarsus Hetero 05 05 1 Synorthocladius Synth 05 05 Psectrocladius (Allopsectrocladius) ParaAl 15 Psectrocladius (Mesopsectrocladius) Psecme Psectrocladius (Monopsectrocladius) Psecmo 1 85 15 14 5 25 3 4 Psectrocladius (p) sordidellus Psecp 1 25 1 4 25 4 5 4 6 Parakiefferiella type A ParakA 2 Parakiefferiella type B ParakB 1 2 1 1 Parakiefferiella nigra ParakN 1 1 I 1 Parakiefferiella triquetra ParaTQ 2 2 1 1 Pseudodiamesa Pseudo Zalutschia type C (Barley type B) ZalutC Zalutschia lugulata pauca ZalutL 1 15 1 05 1 Zalutchia mlutschicola ZalutZ 05 2 55 2 15 05 2 Zalutschia mucronata ZalutM 1 Heterotrissocladius grimshawi-type Hgnm 5 35 8 8 6 45 25 5 1 Heterotrissocladius marcidus-type Hmarc 4 3 13 9 1 1 35 Heterotrissocladius maeaeri-type 1 Hmarl 05 4 05 1 Heterotrissocladius maeaeri-type 2 Hmar2 1 2 Heterotrissocladius sp 2 Hsp2 1 Heterotrissocladius subpilosus-type Hsub Pseudodiamesa larvula Hlarv Mesocricotopus Mesoc 05 Nanocladius Nanoc 65 Geoorthocladius Geoor 05 Tnssocladius Tnssoc Cricotopus mtersectus Cinter 2 2 25 45 2 05 Orthocladius type I Orthoi 25 15 3 0 0 05 0 Cricotopus tremulus CncTr 15 3 05 Cricotopus trifasiatus-type Cnctnf Cricotopus type P CncP Cricotopus sylvestris-type CncSy Cricotopus cyhndraceus-type CncCy 3 Cricotopus bicinctus type CncBi 1 Paracladius Paracl 05 Smittia Smittia Paracricotopus Paracn Chaetocladius Chaet 1 05 Corynoneura arcttca CorArC 2 Corynoneura type A CoryA 1 1 1 2 Euhefferiella gr Eukif 1 05 Litnnophyes Limno 05 05 Orthocladius trigonolabis-type Orthotr 1 Orthocladius type S OrthoS Orthocladius sp 2 (Walker) Ortho2 Orthocladius consobrinus-type Orthoc 1 1 Orthocladius hgnicola OrthoL Orthocladius rivulorum-type OrthoR Diplocladius Diplo Metnocnemus gr Metroc 1 05 Thienemanniella type E ThienE 1 1 214 Appendix Table 8: Tanytarsini, Diamesinae, and Chironomini of the tree-line region. Site CHOI CH02 CH03 CH04 CH05 CH06 CH07 CH08 CH09 Taxon Code Tanytarsus lactescens-type Tanylac 1 1 2 1 1 Tanytarsini no pedastule unident TanyNopod 1 1 2 3 9 10 5 2 3 35 Tanytarsus no spur unident TanyNoSp 1 15 3 05 1 4 Tanytarsus mendax-type TanyMend 3 3 5 2 1 3 2 Tanytarsus lugens-type TanyLug 1 9 5 7 7 2 6 5 2 Tanytarsus palhdicomis-type Tanypall 4 3 3 2 4 1 Tanytarsus chinyensis-type Tanychy 5 4 1 Tanytarsus glabrescens-type Tanyglab Micropsectra contracta-type Mcont 3 1 9 11 4 1 Tanytarsus nemorosus-type Tanynum Cladotanytarsus mancus Cladman 6 7 3 2 2 45 1 Cladotanytarsus type A CladA 1 1 Micropsectra pallidula Mpadu 2 2 Micropsectra insignilobus Minsig 4 45 45 18 4 05 5 Micropsectra type R MicroR Micropsectra type A MicroA 5 Paratanytarsus penicillatus type Parap 4 35 3 5 2 05 Paratanytarsus type A ParaA 1 Paratanytarsus austnacus-type ParaAUS 2 5 4 3 5 6 Corynocera ambigua Cambig 25 57 51 22 19 5 165 Corynocera oliveri Coryohv 25 1 Zavrelia /Stempellinella ZavStemp 35 1 1 05 15 05 1 Constempellina Constmol 25 05 15 0 0 0 0 0 Stempallina Stmpina 2 1 Protanypus Protany 1 Diamesa Diamesa Monodiamesa Monodia 2 Cladopelma Cpelma 15 05 2 25 Dicotendipes Dicrot 5 2 1 15 2 3 Microtendipes Microt 6 105 15 65 25 55 65 65 Paratendipes paraten 1 1 Sergentia Serg 4 2 1 2 6 55 15 Cryptochironomus Crypto 15 1 1 Endochironomus Endo 25 Pseudochironomus Psedoc 05 Chironomus anthracinus-type ChirANT 1 2 1 Chironomus plumosus-type ChirPL Parachironomus Parach Pagastiella Pagast 3 2 Glyptotendipes Glycot Stictochironomus Sticto 25 05 Polypedilum Polyp 1 1 1 1 2 Harnischia /Paracladopelma Harnesh Chironomini larvula Clarvula 1 10 1 Arctopelopia / Thienemannimyia ArcT 7 2 1 1 4 1 5 Derotanypus Dero Procladius Proclad 1 3 2 7 25 1 5 1 Chaoborus tnvittatus Chaob Total HC 93 5 50 67 5 1135 179 162 5 92 5 87 82 215 Appendix Table 9: Orthocladiinae of the Arviat area (Hudson Bay region) Site AVOl AV03 AV02 AV04 AV05 AV06 AVll AV09 AV07 AVI Taxon Code Abiskomyia Abisk Hydrobaenus Hydrob 5 Heterotanytarsus Hetero Synorthocladius Synth P (Allopsectrocladius) ParaAl P (Mesopsectrocladius) Psecme P (Monopsectrocladius) Psecmo 1 1 P (p ) sordidellus Psecp 13 3 35 5 12 8 35 8 34 45 5 Parakiefferiella type A ParakA Parakiefferiella type B ParakB 1 1 1 1 Parakiefferiella nigra ParakN 1 Parakiefferiella triquetra ParaTQ 1 Pseudodiamesa Pseudo Zalutschia type C (Barley type B) ZalutC 05 25 05 Zalutschia lugulata pauca ZalutL 35 10 5 Zalutchia zalutschicola ZalutZ 1 2 15 05 05 Zalutschia mucronata ZalutM H gnmshawi-type Hgnm H marcidus type Hmarc 4 H maeaeri-type 1 Hmarl 1 H maeaeri-type 2 Hmar2 15 H sp2 Hsp2 H subpilosus type Hsub Pseudodiamesa larvula Hlarv Mesocncotopus Mesoc 05 Nanocladius Nanoc 1 15 Geoorthocladius Geoor 1 Trissocladius Tnssoc Cricotopus intersectus Cinter 35 05 75 25 25 2 6 55 2 05 Orthocladius type 1 Orthoi 0 0 0 0 0 0 0 0 0 0 Cricotopus tremulus CncTr Cricotopus trifasiatus-type Cnctnf 1 05 Cricotopus type P CncP 0 15 2 2 1 15 05 25 1 0 Cricotopus sylvestris-type CncSy 05 1 Cricotopus cylindraceus-type CncCy 05 05 4 Cricotopus bicinctus-iype CncBi 05 Paracladius Paracl 1 Smittia Smittia 1 05 Paracricotopus Paracn Chaetocladius Chaet Corynoneura arctica CorArC 1 1 1 2 2 1 1 4 Corynoneura type A CoryA Eukiefferiella gr Eukif 05 Limnophyes Limno 13 1 05 1 Orthocladius tngonolabis-type Orthotr Orthocladius type S OrthoS 1 15 2 05 3 Orthocladius sp 2 (Walker) Ortho2 1 2 Orthocladius consobrinus-type Orthoc 05 1 Orthocladius lignicola OrthoL Orthocladius rivulorum-type OrthoR Diplocladius Diplo Memocnemus gr Metroc 1 Thienemanniella type E ThienE 3 1 1 216 Appendix Table 10: Tanytarsini, Diamesinae, and Chironomini of the Arviat area Site AVOl AV03 AV02 AV04 AV05 AV06 AVll AV09 AV07 AVK Taxon Code Tanytarsus lactescens type Tanylac Tanytarsini no pedastule TanyNp 11 45 2 2 2 45 10 1 15 Tanytarsus no spur TanyNS 15 15 1 45 9 Tanytarsus mendax-type TanyMe 1 1 75 2 10 5 1 Tanytarsus lugens-type TanyLu 85 2 0 1 1 14 5 6 13 4 0 Tanytarsus pallidicomis-type Tanypal 1 25 1 2 1 Tanytarsus chinyensis-type Tanychy 1 Tanytarsus glabrescens-type Tanygla Micropsectra contracta type Mcont 0 0 0 0 0 0 0 0 0 0 Tanytarsus nemorosus type Tanynu Cladotanytarsus mancus Cladmn 10 4 2 1 10 5 2 5 6 3 Cladotanytarsus type A CladA 1 Micropsectra pallidula Mpadu 1 Micropsectra insignilobus Minsig 13 05 Micropsectra type R MicroR Micropsectra type A MicroA Paratanytarsus penicillatus Parap 45 75 45 3 1 105 Paratanytarsus type A ParaA 1 95 1 Paratanytarsus austnacus ParaAUS 2 1 1 2 Corynocera ambigua Cambig 145 152 5 8 47 Corynocera oliveri Coryohv Zavrelia /Stempellinella ZavStem 05 1 2 1 05 Constempellina Consthio 0 1 0 0 0 1 0 05 0 0 Stempallina Stmpina Protanypus Protany Diamesa Diamesa Monodiamesa Monodia 05 Cladopelma Cpelma 1 2 1 15 1 45 1 Dicotendipes Dicrot 2 1 15 05 2 3 1 Microtendipes Microt 2 15 17 5 10 Paratendipes Paraten Sergentia Serg 6 35 15 5 45 15 5 05 115 9 1 Cryptochironomus Crypto 05 1 Endochironomus Endo 1 1 1 1 1 Pseudochironomus Psedoc Chironomus anthracinus-type ChirAN 45 3 55 4 4 45 4 Chironomus plumosus-type ChirPL 1 15 1 1 1 Parachironomus Parach 2 1 1 Pagastiella Pagast Glyptotendipes Glycot 1 05 1 Stictochironomus Sticto 2 05 1 1 2 2 1 Polypedilum Polyp 2 1 25 2 3 1 Harnischia /Paracladopelma Harnesh Chironomini larvula Clarvula 3 2 3 1 1 2 5 Arctopelopia / Thieneman ArcT 1 16 5 Derotanypus Dero 2 Procladius Proclad 1 1 25 2 7 35 8 1 3 Chaoborus tnvittatus Chaob Total 81 50 99 5 80 49 5 84 5 63 370 49 84 5 217 Appendix Table 11: Orthocladiinae of the Rankin Inlet area (Hudson Bay region) Site RA17 RA19 RA19 RA20 RA15 RA08 RA13 RA16 RA10 Taxon Code (2006) (2008) Abiskomyia Abisk 3 Hydrobaenus Hydrob 1 05 1 Heterotanytarsus Hetero Synorthocladius Synth Psectrocladius (Allopsectrocladius) ParaAllo Psectrocladius (Mesopsectrocladius) Psecmeso Psectrocladius (Monopsectrocladius) Psecmon 05 Psectrocladius (p) sordidellus Psecp 45 4 75 55 85 25 Parakiefferiella type A ParakA Parakiefferiella type B ParakB 1 1 Parakiefferiella nigra ParakN Parakiefferiella triquetra ParakTQ 1 2 Pseudodiamesa Pseudo Zalutschia type C (Barley type B) ZalutC 1 1 05 05 05 Zalutschia lugulata pauca ZalutL 15 15 25 Zalutchia zalutschicola ZalutZ Zalutschia mucronata ZalutM Heterotrissocladius gnmshawi-type Hgnm 2 1 05 Heterotrissocladius marcidus-type Hmarc 05 7 3 Heterotrissocladius maeaeri-type 1 Hmarl 05 05 1 Heterotrissocladius maeaeri-type 2 Hmar2 1 Heterotrissocladius sp 2 Hsp2 Heterotrissocladius subpilosus-type Hsub Pseudodiamesa larvula Hlarv Mesoc ncotopus Mesoc Nanocladius Nanoc 05 Geoorthocladius Geoor Trissocladius Tnssocl Cricotopus intersectus Cinter 2 7 55 4 2 25 25 25 Orthocladius type I orthoi 05 15 0 0 1 0 0 0 Cricotopus tremulus CncTrem 1 Cricotopus trifasiatus type Cnctnf Cricotopus type P CncP 0 0 0 2 0 05 0 0 Cricotopus sylvestns type CncSyl Cricotopus cylindraceus-type CncCly 1 05 1 05 Cricotopus bicinctus type CncBi 1 Paracladius Paracl 25 6 9 15 05 Smittia Smittia Paracncotopus Paracnc Chaetocladius Chaet 05 Corynoneura arctica Cory Arc 2 Corynoneura type A CoryA Eukiefferiella gr Eukif Limnophyes Limno 05 Orthocladius trigonolabis-type Orthotn Orthocladius type S OrthoS 1 05 Orthocladius sp 2 (Walker) Ortho2 05 05 Orthocladius consobnnus-type Orthocon Orthocladius lignicola OrthoLig Orthocladius rivulorum-type OrthoRiv 1 05 Diplocladius Diplo Metnocnemus gr Metrocn Thienemanniella type E ThienE 218 Appendix Table lib: Orthocladiinae of the Rankin Inlet area (Hudson Bay region) Site RA05 RA06 RA04 RAll RA07 RA02 RA: Taxon Code Abiskomyia Abisk Hydrobaenus Hydrob 05 Heterotanytarsus Hetero Synorthocladius Synth 1 Psectrocladius (Allopsectrocladius) ParaAllo Psectrocladius (Mesopsectrocladius) Psecmeso Psectrocladius (Monopsectrocladius) Psecmon 2 Psectrocladius (p ) sordidellus Psecp 15 15 10 5 18 45 20 5 51 Parakiefferiella type A ParakA Parakiefferiella type B ParakB 4 Parakiefferiella nigra ParakN 2 Parakiefferiella triquetra ParakTQ 1 Pseudodiamesa Pseudo Zalutschia type C (Barley type B) ZalutC 05 15 Zalutschia lugulata pauca ZalutL 1 1 Zalutchia zalutschicola ZalutZ Zalutschia mucronata ZalutM Heterotrissocladius gnmshawi-type Hgnm 1 o; Heterotrissocladius marcidus type Hmarc Heterotrissocladius maeaeri-type 1 Hmarl Heterotrissocladius maeaeri-type 2 Hmar2 Heterotrissocladius sp 2 Hsp2 Heterotrissocladius subpilosus-type Hsub Pseudodiamesa larvula Hlarv Mesocncotopus Mesoc 05 1 Nanocladius Nanoc Geoorthocladius Geoor Trissocladius Tnssocl Cricotopus intersectus Cmter 1 35 2 2 10 1 Orthocladius type I orthoi 1 2 0 0 0 25 0 Cricotopus tremulus CncTrem 05 Cricotopus trtfasiatus-type Cnctaf Cricotopus type P CncP 0 0 1 0 0 05 0 Cricotopus sylvestris-type CncSyl 1 Cricotopus cyhndraceus-type CncCly 1 2 3 Cricotopus bicinctus-type CncBi 15 1 1 Paracladius Paracl 1 Smittia Smittia Paracncotopus Paracnc Chaetocladius Chaet 05 05 Corynoneura arctica Cory Arc 1 4 4 3 Corynoneura type A CoryA Euheffenella gr Eukif 1 Limnophyes Limno 05 1 Orthocladius trigonolabis-type Orthotn Orthocladius type S OrthoS Orthocladius sp 2 (Walker) Ortho2 1 Orthocladius consobrinus-type Orthocon Orthocladius lignicola OrthoLig Orthocladius rivulorum-type OrthoRiv Diplocladius Diplo Metnocnemus gr Metrocn Thienemanmella type E ThienE Appendix Table 12a: Tanytarsini, Diamesinae, and Chironomini of the Rankin Inlet area Site RA17 RAI9 RAW RA20 RA15 RA08 RA13 RA16 RAIO Taxon Code (2006) (2008) Tanytarsus lactescens type Tanylac Tanytarsini no pedastule - unident TanyNopod 1 45 55 45 3 3 15 05 1 Tanytarsus no spur - unident TanyNoSp 1 1 115 3 1 45 1 3 Tanytarsus mendax type TanyMend 1 2 Tanytarsus lugens-type TanyLug 2 1 1 0 1 4 5 0 4 Tanytarsus pallidicornis-type Tanypall 5 1 5 1 3 2 Tanytarsus chinyensis type Tanychy Tanytarsus glabrescens type Tanyglab Micropsectra contracta type Mcont 0 0 2 0 0 2 1 0 0 Tanytarsus nemorosus type Tanynum Cladotanytarsus mancus Cladman 4 2 6 15 1 8 Cladotanytarsus type A CladA 1 05 Micropsectra pallidula Mpadu 1 1 Micropsectra insignilobus Minsig 7 16 1 12 5 15 5 Micropsectra type R MicroR Micropsectra type A MicroA Paratanytarsus penicillatus-type Parap 2 5 28 5 6 05 25 3 15 Paratanytarsus type A ParaA Paratanytarsus austnacus-type ParaAUS 1 5 4 Corynocera ambigua Cambig 185 6 27 5 75 86 5 35 7 47 Corynocera oliveri Coryoliv 1 3 3 Zavrelia / Stempellinella ZavStemp 2 1 6 05 35 1 Constempellina Consthiol 0 0 0 0 0 0 0 0 0 Stempallina Stmpina Protanypus Protany 05 05 Diamesa Diamesa Monodiamesa Monodia Cladopelma Cpelma 25 15 5 1 05 1 Dicotendipes Dicrot 1 1 4 1 15 3 Microtendipes Microt 1 35 1 Paratendipes paraten Sergentia Serg 05 45 35 65 4 15 1 Cryptochironomus Crypto 05 05 15 Endochironomus Endo 1 Pseudochironomus Psedoc Chironomus anthracinus-type ChirANT 55 4 45 3 4 2 1 25 Chironomus plumosus-type ChirPL Parachironomus Parach Pagastiella Pagast Glyptotendipes Glycot 1 1 Stictochironomus Sticto 1 35 125 35 1 4 Polypedilum Polyp 2 1 Harmschia /Paracladopelma Harnesh 05 Chironomini larvula Clarvula 1 1 1 Arctopelopia / Thienemannimyia ArcT 1 1 3 2 2 Derotanypus Dero Procladius Proclad 1 35 1 25 2 35 Chaoborus tnvittatus Chaob Total 49 5 58 177 125 134 5 80 79 5 75 57 220 Appendix Table 12b: Tanytarsini, Diamesinae, and Chironomini of the Rankin area Site RA05 RA06 RA04 RAll RA07 RA02 RA12 Taxon Code Tanytarsus lactescens-type Tanylac Tanytarsini no pedastule - unident TanyNopod 2 25 6 25 45 29 45 Tanytarsus no spur - unident TanyNoSp 1 0 6 5 7 23 Tanytarsus mendax-type TanyMend 1 4 1 Tanytarsus lugens type TanyLug 1 3 5 3 6 42 1 Tanytarsus pallidicornis type Tanypall 4 1 35 1 Tanytarsus clunyensis type Tanychy Tanytarsus glabrescens-type Tanyglab Micropsectra contracta-type Mcont 0 1 4 0 2 0 0 Tanytarsus nemorosus-type Tanynum Cladotanytarsus mancus Cladman 35 3 1 39 Cladotanytarsus type A CladA 1 1 Micropsectra pallidula Mpadu Micropsectra insignilobus Minsig 1 0 1 1 4 3 Micropsectra type R MicroR Micropsectra type A MicroA Paratanytarsus penicillatus-type Parap 25 2 18 2 3 24 3 Paratanytarsus type A ParaA 1 Paratanytarsus austnacus type ParaAUS 25 Corynocera ambigua Cambig 55 5 35 5 74 1 28 5 Corynocera oliveri Coryohv Zavrelia /Stempellinella ZavStemp Constempellina Consthiol 0 0 05 0 0 1 0 Stempallina Stmpina Protanypus Protany Diamesa Diamesa Monodiamesa Monodia 1 Cladopelma Cpelma 25 15 65 Dicotendipes Dicrot 05 1 3 15 12 Microtendipes Microt 2 Paratendipes paraten Sergentia Serg 45 115 3 4 5 1 Cryptochironomus Crypto 1 05 1 Endochironomus Endo Pseudochironomus Psedoc Chironomus anthracinus-type ChirANT 1 1 15 7 10 Chironomus plumosus-type ChirPL Parachironomus Parach 05 Pagastiella Pagast Glyptotendipes Glycot 1 1 2 Stictochironomus Sticto 1 15 45 Polypedilum Polyp Harnischia /Paracladopelma Harnesh Chironomini larvula Clarvula 1 Arctopelopia / Thienemannimyia ArcT 2 1 2 Derotanypus Dero Procladius Proclad 1 2 3 13 5 Chaoborus tnvittatus Chaob Total 84 5 66 165 5 58 50 5 264 5 55 221 Appendix Table 13: Orthocladiinae of the Repulse Bay area (Hudson Bay region) Site RBOl RB07 RB02 RB04 RB14 RB: Taxon Code Abiskomyia Abisk Hydrobaenus Hydrob 25 Heterotanytarsus Hetero Synorthocladius Synth 05 0! Psectrocladius (Allopsectrocladius) ParaAllo Psectrocladius (Mesopsectrocladius) Psecmeso Psectrocladius (Monopsectrocladius) Psecmon Psectrocladius (p) sordidellus Psecp 10 10 5 85 25 3 Parakiefferiella type A ParakA 1 Parakiefferiella type B ParakB 1 1 Parakiefferiella nigra ParakN Parakiefferiella triquetra ParakTQ Pseudodiamesa Pseudo Zalutschia type C (Barley type B) ZalutC 1 Zalutschia lugulata pauca ZalutL 1 Zalutchia zatutschicola ZalutZ Zalutschia mucronata ZalutM Heterotrissocladius gnmshawi type Hgnm 1 Heterotrissocladius marcidus-type Hmarc 1 Heterotrissocladius maeaeri type 1 Hmarl 35 Heterotrissocladius maeaeri-type 2 Hmar2 Heterotrissocladius sp 2 Hsp2 05 Heterotrissocladius subpilosus type Hsub Pseudodiamesa larvula Hlarv Mesocncotopus Mesoc 05 Nanocladius Nanoc Geoorthocladius Geoor 05 Tnssocladius Tnssocl Cricotopus mtersectus Cinter 6 5 4 5 1 6 Orthocladius type 1 orthoi 0 0 05 0 0 1 Cricotopus tremulus CncTrem 1 Cricotopus trifasiatus-type Cncrnf Cricotopus type P CncP 05 0 05 0 1 0 Cricotopus sylvestris-type CncSyl 1 15 05 Cricotopus cylindraceus type CncCly Cricotopus bicinctus type CncBi Paracladius Paracl 15 1 Smittia Smittia 2 Paracricotopus Paracnc Chaetocladius Chaet 1 05 Corynoneura arcttca Cory Arc 1 8 1 5 3 3 Corynoneura type A CoryA Eukiejfenella gr Eukif Limnophyes Limno 2 65 1 Orthocladius trigonolabis-type Orthotn Orthocladius type S OrthoS 1 1 Orthocladius sp 2 (Walker) Ortho2 1 1 1 Orthocladius consobrinus-type Orthocon Orthocladius lignicola OrthoLig Orthocladius nvulorum type OrthoRiv Diplocladius Diplo Metnocnemus gr Metrocn 1 Thtenemanniella type E ThienE 3 222 Appendix Table 14: Tanytarsini, Diamesinae, and Chironomini of the Repulse Bay area Site RBOl RB07 RB02 RB04 RB14 RB17 Taxon Code Tanytarsus lactescens-type Tanylac Tanytarsini no pedastule - unident TanyNopod 1 2 35 2 05 15 Tanytarsus no spur - unident TanyNoSp 15 1 3 2 35 3 Tanytarsus mendax-type TanyMend 5 7 35 10 3 22 Tanytarsus lugens type TanyLug 7 6 5 0 2 2 Tanytarsus palhdicornis type Tanypall 1 1 Tanytarsus chinyensis type Tanychy Tanytarsus glabrescens-type Tanyglab Micropsectra contracta-type Mcont 0 1 0 0 0 0 Tanytarsus nemorosus-type Tanynum Cladotanytarsus mancus Cladman 9 05 4 3 8 3 Cladotanytarsus type A CladA Micropsectra palhdula Mpadu Micropsectra insignilobus Minsig 20 Micropsectra type R MicroR Micropsectra type A MicroA Paratanytarsus penicillatus-type Parap 75 14 15 1 5 3 Paratanytarsus type A ParaA 4 6 1 2 3 Paratanytarsus austnacus type ParaAUS Corynocera ambigua Cambig 109 Corynocera oliveri Coryohv 2 15 Zavrelia/ Stempellinella ZavStemp Constempellina Consthiol 0 0 0 0 0 0 Stempallina Stmpina Protanypus Protany Diamesa Diamesa Monodiamesa Monodia Cladopelma Cpelma 15 05 1 Dicotendipes Dicrot 14 5 12 5 15 5 14 5 13 Microtendipes Microt Paratendipes paraten Sergentia Serg 4 7 Cryptochironomus Crypto Endochironomus Endo Pseudochironomus Psedoc Chironomus anthracinus-type ChirANT 45 35 5 5 45 Chironomus plumosus type ChirPL 1 1 Parachironomus Parach Pagastiella Pagast Glyptotendipes Glycot Stictochironomus Sticto 1 05 Polypedilum Polyp Harnischia /Paracladopelma Harnesh Chironomini larvula Clarvula 4 1 1 1 1 3 Arctopelopia / Thienemannimyia ArcT 6 Derotanypus Dero Procladius Proclad 45 75 1 85 7 4 Chaoborus trivittatus Chaob Total 815 229 5 76 5 67 5 615 81 223 Appendix Table 15: Orthocladiinae of the Baker Lake area (Central region) Site BLOI BL02 BLll BL17 BL3: BL06 BL16 Taxon Code Abiskomyia Abisk 4 35 Hydrobaenus Hydrob Heterotanytarsus Hetero Synorthocladius Synth Psectrocladius (Allopsectrocladius) ParaAllo Psectrocladius (Mesopsectrocladius) Psecmeso 6 Psectrocladius (Monopsectrocladius) Psecmon 1 1 Psectrocladius (p) sordidellus Psecp 35 65 8 85 Parakiefferiella type A ParakA Parakiefferiella type B ParakB 2 3 Parakiefferiella nigra ParakN 2 3 Parakiefferiella triquetra ParakTQ 1 15 Pseudodiamesa Pseudo Zalutschia type C (Barley type B) ZalutC 35 6 Zalutschia lugulata pauca ZalutL 25 2 05 Zalutchia zalutschicola ZalutZ Zalutschia mucronata ZalutM Heterotrissocladius grimshawi-type Hgnm 7 3 6 Heterotrissocladius marcidus type Hmarc 1 15 5 9 Heterotrissocladius maeaeri-type 1 Hmarl 45 2 05 Heterotrissocladius maeaeri-type 2 Hmar2 1 05 05 Heterotrissocladius sp 2 Hsp2 Heterotrissocladius subpilosus type Hsub Pseudodiamesa larvula Hlarv Mesocricotopus Mesoc 1 2 Nanocladius Nanoc Geoorthocladius Geoor 05 Tnssocladius Tnssocl 1 Cricotopus intersectus Cinter 2 1 15 Orthocladius type l orthoi 0 0 0 0 0 05 Cricotopus tremulus CncTrem Cricotopus trifasiatus-type Cnctnf Cricotopus type P CncP 0 0 Cricotopus sylvestns-type CncSyl 1 Cricotopus cylindraceus type CncCly 2 25 Cricotopus bicinctus-type CncBi Paracladius Paracl 95 2 1 05 Smittia Smittia Paracricotopus Paracnc Chaetocladius Chaet Corynoneura arctica Cory Arc 1 2 1 25 Corynoneura type A CoryA Eukiefferiella gr Eukif 1 05 Limnophyes Limno 1 Orthocladius tngonolabis-type Orthotn Orthocladius type S OrthoS Orthocladius sp 2 (Walker) Ortho2 Orthocladius consobnnus type Orthocon Orthocladius lignicola OrthoLig Orthocladius nvulorum type OrthoRiv Diplocladius Diplo Metnocnemus gr Metrocn Thienemanniella type E ThienE 3 Appendix Table 16: Tanytarsini, Diamesinae, and Chironomini of the Baker Lake area Site BLOI BL02 BLll BL17 BL35 BL06 BL16 Taxon Code Tanytarsus lactescens-type Tanylac Tanytarsini no pedastule - unident TanyNopod 4 1 1 Tanytarsus no spur - unident TanyNoSp 1 1 Tanytarsus mendax-type TanyMend 21 24 22 Tanytarsus lugens-type TanyLug 1 1 1 1 Tanytarsus pallidicomis-type Tanypall 1 Tanytarsus chinyensis-type Tanychy Tanytarsus glabrescens-type Tanyglab 5 12 5 Micropsectra contracta-type Mcont 1 2 25 Tanytarsus nemorosus-type Tanynum 4 35 25 Cladotanytarsus mancus Cladman 6 1 75 1 Cladotanytarsus type A CladA Micropsectra pallidula Mpadu 1 1 Micropsectra insignilobus Minsig 65 85 12 5 Micropsectra type R MicroR Micropsectra type A MicroA Paratanytarsus penicillatus-type Parap 1 05 2 25 4 45 2 Paratanytarsus type A ParaA Paratanytarsus austriacus-type ParaAUS 1 Corynocera ambigua Cambig 18 25 1 Corynocera oliveri Coryohv 1 2 5 1 Zavrelia /Stempellinella ZavStemp 15 2 2 Constempellina Consthiol 0 0 0 0 0 0 0 Stempallina Stmpina Protanypus Protany 2 25 05 Diamesa Diamesa Monodiamesa Monodia Cladopelma Cpelma 1 2 1 Dicotendipes Dicrot 1 35 Microtendipes Microt Paratendipes paraten Sergentia Serg 45 Cryptochironomus Crypto 1 Endochironomus Endo Pseudochironomus Psedoc Chironomus anthracinus-type ChirANT 4 1 1 Chironomus plumosus-type ChirPL 1 1 Parachironomus Parach Pagastiella Pagast Glyptotendipes Glycot Stictochironomus Sticto Polypedilum Polyp 1 1 Harnischia /Paracladopelma Harnesh 1 Chironomini larvula Clarvula 1 1 Arctopelopia / Thienemannimyia ArcT 1 2 Derotanypus Dero 2 Procladius Proclad 1 2 5 1 1 1 Chaoborus trivittatus Chaob Total 61 90 5 58 5 70 5 57 52 53 5 225 Appendix Table 17: Orthocladiinae of the Iqaluit and Clyde River areas (Baffin region) Site IQ04 IQ05 IQ06 IQOl IQO: IQ02 CROl CR02 CR04 Taxon Code Abiskomyia Abisk 1 85 25 2 Hydrobaenus Hydrob 15 05 Heterotanytarsus Hetero Synorthocladius Synth 05 Psectrocladius (Allopsectrocladius) ParaAllo Psectrocladius (Mesopsectrocladius) Psecmeso Psectrocladius (Monopsectrocladius) Psecmon Psectrocladius (p ) sordidellus Psecp 25 5 12 5 3 18 5 05 Parakiefferiella type A ParakA Parakiefferiella type B ParakB 1 15 Parakiefferiella nigra ParakN Parakiefferiella triquetra ParakTQ 1 Pseudodiamesa Pseudo 2 Zalutschia type C (Barley type B) ZalutC 25 35 15 5 23 5 3 35 Zalutschia lugulata pauca ZalutL 2 Zalutchia zalutschicola ZalutZ Zalutschia mucronata ZalutM Heterotrissocladius gnmshawi type Hgnm 16 5 105 15 1 55 17 5 15 7 Heterotrissocladius marcidus-type Hmarc 45 35 1 4 15 13 5 3 Heterotrissocladius maeaeri type 1 Hmarl 16 5 6 25 25 26 17 5 9 Heterotrissocladius maeaeri type 2 Hmar2 55 8 1 35 12 5 2 3 Heterotrissocladius sp 2 Hsp2 05 4 Heterotrissocladius subpilosus type Hsub 35 Pseudodiamesa larvula Hlarv Mesocncotopus Mesoc 2 1 1 1 Nanocladius Nanoc Geoorthocladius Geoor 05 05 Tnssocladius Tnssocl Cricotopus mtersectus Cinter 1 1 3 1 Orthocladius type 1 orthoi 0 0 0 0 0 0 0 0 Cricotopus tremulus CncTrem Cricotopus trifasiatus type Cnctnf Cricotopus type P CncP 1 0 05 0 0 0 0 0 Cricotopus sylvestris type CncSyl 1 1 Cricotopus cyhndraceus-type CncCly Cricotopus bicinctus-type CncBi Paracladius Paracl 1 Smittia Smittia Paracncotopus Paracnc Chaetocladius Chaet Corynoneura arctica Cory Arc 1 4 1 Corynoneura type A CoryA Euhefferiella gr Eukif 75 Limnophyes Limno 25 1 Orthocladius tngonolabis-type Orthotn Orthocladius type S OrthoS 1 10 5 Orthocladius sp 2 (Walker) Ortho2 1 2 Orthocladius consobnnus-type Orthocon Orthocladius lignicola OrthoLig Orthocladius rivulorum-type OrthoRiv Diplocladius Diplo Metnocnemus gr Metrocn 1 Thienemanniella type E ThienE 226 Appendix Table 18: Tanytarsini, Diamesinae, and Chironomini of the Baffin region Site IQ04 IQ05 IQ06 IQOl IQ03 IQ02 CROl CR02 CR04 Taxon Code Tanytarsus lactescens-type Tanylac 1 05 Tanytarsini no pedastule - unident TanyNopod 2 25 8 2 4 2 3 1 Tanytarsus no spur - unident TanyNoSp Tanytarsus mendax-type TanyMend 1 Tanytarsus lugens type TanyLug 1 0 3 2 25 13 0 0 0 Tanytarsus pallidicornis type Tanypall 1 1 1 1 Tanytarsus chinyensis type Tanychy Tanytarsus glabrescens type Tanyglab Micropsectra contracta-type Mcont 0 0 1 1 15 0 5 2 0 Tanytarsus nemorosus-type Tanynum Cladotanytarsus mancus Cladman 1 Cladotanytarsus type A CladA Micropsectra pallidula Mpadu 1 2 45 1 Micropsectra insigmlobus Minsig 4 4 05 14 2 15 5 10 13 Micropsectra type R MicroR Micropsectra type A MicroA Paratanytarsus pemcillatus-type Parap 1 65 Paratanytarsus type A ParaA 15 05 1 1 1 2 Paratanytarsus austnacus-type ParaAUS Corynocera ambigua Cambig Corynocera oliveri Coryohv 105 1 Zavrelia /Stempellinella ZavStemp Constempellina Consthiol 0 0 0 0 0 0 0 0 0 Stempalhna Stmpina Protanypus Protany 1 2 15 05 6 15 Diamesa Diamesa Monodiamesa Monodia Cladopelma Cpelma 2 Dicotendipes Dicrot 2 Microtendipes Microt Paratendipes paraten Sergentta Serg 1 35 2 65 5 1 Cryptochironomus Crypto Endochironomus Endo Pseudochironomus Psedoc Chironomus anthracinus-type CmrANT 05 1 Chironomus plumosus type ChirPL 1 Parachironomus Parach Pagastiella Pagast Glyptotendipes Glycot Stictochironomus Sticto 2 2 15 1 Polypedilum Polyp Hamischia /Paracladopelma Harnesh 1 Chironomini larvula Clarvula 2 1 Arctopelopia / Thienemannimyia ArcT Derotanypus Dero Procladius Proclad 1 2 5 4 65 4 Chaoborus trivittatus Chaob 2 2 1 3 2 1 Total 60 5 63 51 915 50 68 5 129 104 5 51 227 Appendix Table 19: Orthocladiinae of the northern Baffin and Bylot Island areas Site QB15 QBOl QB06 QB07 BYIO BY18 BYOl BY17 BY15 BY14 BY16 Taxon Code Abiskomyia Abisk 6 8 6 18 5 20 2 15 44 5 15 Hydrobaenus Hydrob 2 2 1 1 15 1 4 Heterotanytarsus Hetero Synorthocladius Synth P (Allopsectrocladms) ParaAllo P (Mesopsectrocladius) Psecmeso P (Monopsectrocladius) Psecmon P (p) sordidellus Psecp 75 65 23 5 Parakiefferiella type A ParakA Parakiefferiella type B ParakB Parakiefferiella nigra ParakN Parakiefferiella triquetra ParakTQ 15 Pseudodiamesa Pseudo 3 4 25 Z type C (Barley type B) ZalutC 25 15 3 05 45 25 Z lugulata pauca ZalutL 05 05 15 29 5 115 23 102 Z zalutschicola ZalutZ Z mucronata ZalutM H grimshawi type Hgnm 2 15 3 1 H marcidus-type Hmarc 15 95 H maeaeri-type 1 Hmarl 2 05 15 05 H maeaeri-type 2 Hmar2 05 05 05 H sp2 Hsp2 1 1 05 H subpilosus-type Hsub Pseudodiamesa larvula Hlarv 2 Mesocncotopus Mesoc 25 Nanocladius Nanoc Geoorthocladius Geoor 05 Tnssocladius Tnssocl Cricotopus intersectus Cinter 05 05 1 25 3 15 35 Orthocladius type 1 orthoi 0 0 0 0 0 0 0 Cricotopus tremulus CricTrem Cricotopus trifasiatus Cncrnf Cricotopus type P CncP 05 0 05 0 1 05 Cricotopus sylvestns CncSyl Cricotopus cylindraceus CncCly 15 2 Cricotopus bicinctus CncBi Paracladius Paracl 15 3 Smittia Smittia 05 15 1 05 Paracncotopus Paracnc 1 Chaetocladius Chaet 15 05 Corynoneura arctica Cory Arc 1 2 25 10 Corynoneura type A CoryA Eukiefferiella gr Eukif 10 15 05 1 05 Limnophyes Limno 1 10 5 19 5 1 45 Orthocladius trigonolabis Orthotn 55 Orthocladius type S OrthoS 05 5 12 5 37 5 75 65 87 Orthocladius sp 2 Ortho2 16 1 1 1 Orthocladius consobrinus Orthocon 05 05 15 35 1 1 Orthocladius lignicola OrthoLig 05 05 Orthocladius rivulorum OrthoRiv Diplocladius Diplo 15 Metnocnemus gr Metrocn 2 05 35 25 1 Thienemanniella type E ThienE 65 1 228 Appendix Table 20: Tanytarsini, Diamesinae, and Chironomini of the northern Baffin and Bylot Island areas Site QB15 QBOl QB06 QB07 BY10 BY18 BYOl BY17 BY15 BY14 BY16 Taxon Code Tanytarsus lactescens Tanylac 1 Tanytarsini no pedastule TanyNopod 2 5 1 05 4 4 2 85 35 115 Tanytarsus no spur TanyNoSp 1 35 15 4 4 45 15 Tanytarsus mendax-type TanyMend 2 4 2 5 1 Tanytarsus lugens-type TanyLug 0 12 16 23 24 95 Tanytarsus pallidicornis Tanypall 2 15 Tanytarsus chinyensis Tanychy Tanytarsus glabrescens Tanyglab Micropsectra contracta Mcont 05 0 Tanytarsus nemorosus Tanynum Cladotanytarsus mancus Cladman 15 Cladotanytarsus type A CladA Micropsectra pallidula Mpadu 14 5 05 75 1 1 Micropsectra insignil Minsig 19 5 13 55 2 1 22 5 1 Micropsectra type R MicroR 4 1 1 4 Micropsectra type A MicroA Paratanytar penicillatus Parap 25 55 1 Paratanytarsus type A ParaA 115 3 4 6 22 14 Paratanytar austnacus ParaAUS 1 1 4 Corynocera ambigua Cambig 5 1 27 5 15 Corynocera oliveri Coryohv 1 12 14 5 35 6 5 1 Zavrelia / Stempellinella ZavStemp Constempellina Consthiol 0 0 0 0 0 0 0 Stempallina Stmpina Protanypus Protany 15 05 1 05 05 Diamesa Diamesa 25 1 Monodiamesa Monodia Cladopelma Cpelma Dicotendipes Dicrot 05 05 Microtendipes Microt Paratendipes paraten Sergentia Serg 13 5 25 18 5 34 14 Cryptochironomus Crypto Endochironomus Endo Pseudochironomus Psedoc Chironomus anthracinus ChirANT 2 05 1 05 Chironomus plumosus ChirPL 15 25 15 15 Parachironomus Parach Pagastiella Pagast Glyptotendipes Glycot Stictochironomus Sticto Polypedilum Polyp Harnischia /Paraclad Harnesh Chironomini larvula Clarvula Arctopelopia / Thiene 1 5 1 2 ArcT Derotanypus 1 1 1 2 Dero Procladius Proclad Chaoborus trivittatus 1 1 1 3 45 5 23 5 65 Chaob Total 102 52 67 77 615 85 5 1415 151 148 3165 302 5 229 Appendix 4: Raw counts of chironomid head capsules in a sediment core from Baker Lake, Nunvut Appendix Table 21: Orthocladiinae of Baker Lake (0 - 5.5cm downcore) Taxon Depth 05 1 15 2 25 3 35 4 45 5 55 Code Abiskomyia Abisk 15 1 2 2 7 55 45 35 3 Hydrobaenus Hydrob 15 25 15 25 2 Oliveridia ohverd 1 35 05 1 15 05 Pseudodiamesa Pseudo 25 05 5 2 15 P (Mesopsectrocladius) Psecmeso 1 1 P (Monopsectrocladius) Psecmon 15 1 P (p) sordidellus Psecp 15 05 15 35 2 2 1 1 35 Parakiefferiella type B ParakB 15 2 4 7 8 19 5 65 5 7 8 Parakiefferiella nigra ParakN 4 2 5 3 3 95 95 35 7 Parakiefferiella triquetra ParakTQ 1 4 1 1 1 35 Zalutschia type C (Barley type B) ZalutB 05 3 15 1 15 05 2 1 Zalutschia lugulata pauca ZalutL 1 1 05 H grimshawi type Hgnm 05 19 32 14 5 45 5 35 5 48 43 36 34 30 5 H marcidus type Hmarc 05 05 3 15 4 45 55 H maeaeri type 1 Hmarl 75 10 75 18 19 5 19 25 17 5 14 5 16 19 5 H maeaeri-type 2 Hmar2 1 1 1 1 5 2 7 H sp2 Hsp2 1 1 3 2 05 15 H subpilosus type Hsub 1 1 2 1 1 Pseudodiamesa larvula Hlarv Mesocricotopus Mesoc 2 55 3 65 8 8 5 35 3 65 Nanocladius Nanoc 1 1 3 05 Trissocladius Tnssocl 1 2 Cricotopus intersectus Cmter 2 1 Orthocladius oliveri Orthohv 55 Cricotopus trifasiatus-type Cnctnf 05 05 Cricotopus type P CncP 05 1 2 Cricotopus obnixus-type CncOb Cricotopus sylvestns type CncSyl 1 2 Cricotopus cylindraceus type CncCly 1 2 25 Paracladius Paracl 11 55 15 115 21 17 5 18 17 21 15 19 Smittia Smittia 2 15 Chaetocladius Chaet 1 4 05 Corynoneura arctwa Cory Arc 1 25 3 4 35 4 2 3 3 Eukieffenella gr / Tventia Eukif 2 25 3 35 8 5 65 8 45 75 20 Limnophyes Limno 35 3 Orthocladius trigonolabis-type Orthotn 2 3 3 Orthocladius type S OrthoS 15 2 35 25 15 2 35 1 115 Orthocladius sp 2 (Walker) Ortho2 Orthocladius consobnnus-type Orthocon 2 15 15 2 35 5 1 25 Orthocladius rivulorum-type OrthoRiv Orthocladius type I Orthol 1 1 5 25 1 1 Metnocnemus gr Metrocn Thienemanniella type E ThienE 1 2 15 2 4 2 3 3 230 Appendix Table 22: Tanytarsini, Diamesinae, and Chironomini of Baker Lake (0 - 5.5 cm downcore) Taxon Depth 05 I 15 225 335 4 45 555 Code Tanytarsini no pedastule - unident TanyNopod 15 3 1 35 2 1 25 3 25 2 Tanytarsus no spur unident TanyNoSp 1 Tanytarsus lugens type TanyLug 2 1 1 14 13 5 22 12 4 5 6 Tanytarsus pallidicomis type Tanypall 05 Tanytarsus chinyensis type Tanychy 2 Micropsectra contracta type Mcont 25 10 13 35 13 15 5 8 25 13 5 65 4 Cladotanytarsus mancus Cladman 6 10 8 15 5 10 9 115 3 3 1 Micropsectra pallidula Mpadu 3 Micropsectra insignilobus Minsig 4 10 9 16 33 34 5 48 5 16 7 6 55 Micropsectra type R MicroR 05 Paratanytarsus penicillatus-type Parap 1 1 1 105 75 4 4 45 7 11 Paratanytarsus austnacus type ParaAUS 1 2 1 Corynocera oliveri Coryohv 1 2 4 4 1 5 3 Zavrelia / Stempellinella ZavStemp 2 2 1 2 15 1 Constempellina Constem 1 35 2 55 65 65 7 105 35 3 3 Protanypus Protany 1 2 3 05 55 15 8 35 15 2 Monodiamesa Monodia 1 15 25 7 55 5 15 2 2 1 Sergentia Serg 05 05 5 1 Chironomus anthracinus type ChirANT 3 9 115 14 18 25 19 17 5 9 115 18 Chironomus plumosus type ChirPL 1 Stictochironomus type B StictoB 1 05 1 1 Stictochironomus StictoR 3 3 1 2 4 11 10 55 8 5 Harmschia /Paracladopelma Harnesh 2 15 05 45 1 1 Chironomini larvula Clarvula 1 1 3 2 1 2 1 Arctopelopia / Thienemannimyia ArcT 1 2 3 1 1 1 Procladius Proclad 25 4 5 6 55 4 3 3 2 105 total 56 5 1185 167 5 143 277 277 325 248 196 5 173 5 234 231 Appendix Table 23: Orthocladiinae of Baker Lake (6-11 cm downcore) Taxon Depth 6 65 7 75 8 85 9 95 10 10 5 11 Code Abiskomyia Abisk 7 6 7 2 5 75 75 5 75 10 5 24 Hydrobaenus Hydrob 4 5 5 05 2 4 25 25 7 35 Oliveridia oliverd 25 25 35 45 1 3 1 2 5 35 10 Pseudodiamesa Pseudo 1 15 2 5 3 1 15 05 15 2 P (Mesopsectrocladius) Psecmeso P (Monopsectrocladius) Psecmon 1 P (p) sordidellus Psecp 1 35 1 35 1 15 05 3 Parakiefferiella type B ParakB 4 10 75 15 1 3 1 6 1 5 3 Parakiefferiella nigra ParakN 8 7 11 75 2 75 10 5 85 10 6 18 Parakiefferiella triquetra ParakTQ 3 3 1 1 1 2 4 3 Zalutschia type C (Barley type B) ZalutB 1 45 15 05 Zalutschia lugulata pauca ZalutL 05 1 H gnmshawi-type Hgnm 43 57 64 5 40 35 5 49 29 5 32 19 30 54 H marcidus-type Hmarc 12 5 2 4 05 25 2 15 15 25 H maeaeri-type 1 Hmarl 15 39 27 5 12 95 15 5 19 115 13 5 18 5 27 5 H maeaeri type 2 Hmar2 7 12 115 65 4 65 3 55 75 7 13 H sp2 Hsp2 1 15 05 05 1 05 H subpilosus-type Hsub 1 45 2 1 1 2 25 2 Pseudodiamesa larvula Hlarv 1 2 1 Mesocncotopus Mesoc 7 16 5 11 75 4 55 3 15 55 5 115 Nanocladius Nanoc 05 Trissocladius Tnssocl 1 Cricotopus intersectus Cinter 1 Orthocladius oliveri Orthohv 7 1 05 1 1 Cricotopus trifasiatus-type Cnctnf 05 1 1 15 05 05 Cricotopus type P CncP 1 Cricotopus obnixus-type CncOb Cricotopus sylvestns type CncSyl Cricotopus cyhndraceus type CncCly 2 2 15 05 3 Paracladius Paracl 24 29 30 11 13 15 5 10 12 5 15 16 18 Smittia Smittia 05 05 1 05 1 1 Chaetocladius Chaet 1 05 1 1 Corynoneura arctica Cory Arc 6 25 1 1 1 1 1 2 2 3 Eukieffenella gr / Tventia Eukif 9 20 20 5 55 6 7 7 15 12 5 29 215 Limnophyes Limno Orthocladius tngonolabis-type Orthotri 1 1 05 15 05 1 1 Orthocladius type S OrthoS 4 85 85 4 3 7 3 3 25 8 75 Orthocladius sp 2 (Walker) Ortho2 1 05 Orthocladius consobrinus-type Orthocon 05 05 15 1 Orthocladius rivulorum type OrthoRiv Orthocladius type I Orthol 35 3 25 3 1 15 4 5 45 2 Metnocnemus gr Metrocn Thienemanniella type E ThienE 1 15 2 1 1 Appendix Table 24: Tanytarsini, Diamesinae, and Chironomini of Baker Lake (6-11 cm downcore) Taxon Depth 6 65 7 75 8 85 9 95 10 10 5 11 Code Thienemanniella type E ThienE 1 15 2 1 1 Tanytarsini no pedastule TanyNopod 1 9 8 15 3 35 3 05 1 3 45 Tanytarsus no spur TanyNoSp Tanytarsus lugens TanyLug 6 2 1 5 1 2 1 1 4 Tanytarsus pallidicornis Tanypall 1 Tanytarsus chinyensis Tanychy Micropsectra contracta Mcont 115 15 5 7 6 55 14 5 7 13 5 85 20 5 10 5 Cladotanytarsus mancus Cladman 1 Micropsectra pallidula Mpadu 1 Micropsectra insignilobus Minsig 6 65 7 1 1 5 6 9 11 14 14 Micropsectra type R MicroR Paratanytarsus penicillatus Parap 8 22 11 7 35 45 55 4 75 2 15 5 Paratanytarsus austnacus ParaAUS 05 1 1 1 Corynocera oliveri Coryoliv 1 1 5 1 15 1 1 4 3 Zavrelia / Stempellinella ZavStemp 1 1 Constempellina Constem 25 3 05 25 15 2 35 15 1 25 05 Protanypus Protany 2 65 1 15 35 15 45 I 2 Monodiamesa Monodia 2 1 35 1 05 Sergentia Serg 15 05 1 15 05 2 1 3 Chironomus anthracinus ChirANT 22 32 5 95 125 55 19 5 22 11 7 05 3 Chironomus plumosus ChirPL 1 1 Stictochironomus type B StictoB 3 05 1 1 1 1 25 3 Stictochironomus StictoR 5 25 7 3 1 2 1 1 1 3 Harnischia /Paraclad Harnesh 2 1 2 1 1 25 1 Chironomini larvula Clarvula 2 6 1 25 2 1 3 2 1 1 2 Arctopelopia / Thieneman ArcT 1 1 3 Procladius Proclad 5 65 3 3 2 1 1 1 1 total 236 5 370 290 5 182 5 126 5 205 172 5 162 172 5 220 306 233 Appendix Table 25: Orthocladiinae of Baker Lake (11.5- 15 cm downcore) Taxon Depth 115 12 12 5 13 13 5 14 145 15 Code Abiskomyia Abisk 11 13 5 215 25 5 15 5 12 5 16 13 Hydrobaenus Hydrob 3 2 25 45 15 25 Oliveridia oliverd 8 5 5 1 2 9 05 Pseudodiamesa Pseudo 75 6 6 45 95 2 15 1 P (Mesopsectrocladius) Psecmeso P (Monopsectrocladius) Psecmon P (p) sordidellus Psecp 15 05 1 Parakiefferiella type B ParakB 3 25 2 25 1 25 Parakiefferiella nigra ParakN 4 13 5 13 13 5 5 75 9 75 Parakiefferiella triquetra ParakTQ Zalutschia type C (Barley type B) ZalutB 05 05 1 1 Zalutschia lugulata pauca ZalutL 1 05 1 H gnmshawi-type Hgnm 25 5 27 35 30 5 18 6 45 35 H marcidus-type Hmarc 10 5 4 12 5 15 05 3 05 H maeaeri-type 1 Hmarl 105 115 115 11 25 3 55 35 H maeaeri type 2 Hmar2 3 95 14 5 10 5 75 4 10 5 H sp2 Hsp2 05 15 1 3 H subpilosus-type Hsub 13 5 115 75 3 25 15 Pseudodiamesa larvula Hlarv 1 1 4 1 2 1 3 Mesocncotopus Mesoc 2 3 1 45 3 3 6 Nanocladius Nanoc 05 Trissocladius Tnssocl Cricotopus intersectus Cinter 05 1 15 05 15 Orthocladius oliveri Orthohv 3 05 15 25 1 Cricotopus trifasiatus-type Cnctnf Cricotopus type P CncP 05 05 Cricotopus obnixus-type CncOb 05 Cricotopus sylvestns type CncSyl Cricotopus cylindraceus-type CncCly 05 35 2 1 25 Paracladius Paracl 14 5 12 5 55 7 4 45 3 5 Smittia Smittia 2 1 05 1 Chaetocladius Chaet 05 05 1 Corynoneura arctica Cory Arc 2 4 15 25 1 1 3 Eukiefferiella gr / Tventia Eukif 11 16 15 16 5 17 5 11 30 65 Limnophyes Limno Orthocladius trigonolabis-type Orthotn 05 1 Orthocladius type S OrthoS 3 5 45 75 55 11 12 5 45 Orthocladius sp 2 (Walker) Ortho2 Orthocladius consobrinus-type Orthocon 1 05 05 25 1 05 Orthocladius rivulorum type OrthoRiv Orthocladius type I Orthol 2 5 85 2 1 1 4 15 Metnocnemus gr Metrocn 1 Thienemanniella type E ThienE 1 2 15 2 2 2 234 Appendix Table 26: Tanytarsini, Diamesinae, and Chironomini of Baker Lake (11.5 15 cm downcore) Taxon Depth 115 12 12 5 13 13 5 14 14 5 15 Code Tanytarsini no pedastule - unident TanyNopod 25 45 2 25 Tanytarsus no spur - unident TanyNoSp Tanytarsus lugens type TanyLug 3 3 3 6 3 1 35 3 Tanytarsus palhdicornis-type Tanypall Tanytarsus chinyensis type Tanychy Micropsectra contracta type Mcont 2 6 5 15 45 35 3 1 Cladotanytarsus mancus Cladman Micropsectra palhdula Mpadu Micropsectra insignilobus Minsig 45 4 3 3 2 05 7 15 Micropsectra type R MicroR 05 15 05 Paratanytarsus penicillatus-type Parap 55 75 7 25 4 3 2 Paratanytarsus austnacus type ParaAUS 1 2 1 Corynocera oliveri Coryoliv 1 1 1 Zavrelia /Stempellinella ZavStemp 05 Constempellina Constem 2 1 Protanypus Protany 75 3 2 1 25 Monodiamesa Monodia 1 Sergentia Serg 2 2 25 1 15 Chironomus anthracinus type ChirANT 2 1 15 Chironomus plumosus type ChirPL Stictochironomus type B StictoB 1 05 Stictochironomus StictoR 2 1 4 4 1 Harnischia /Paracladopelma Harnesh Chironomini larvula Clarvula 1 Arctopelopia / Thienemannimyia ArcT 1 Procladius Proclad 1 2 1 total 136 5 2115 200 209 5 140 5 94 5 149 5 73 235 Appendix Table 27: Orthocladiinae of Baker Lake (15.5 - 20 cm downcore) Taxon Depth 15 5 16 16 5 17 17 5 18 18 5 19 19 5 20 Code Abiskomyia Abisk 6 28 5 25 19 5 39 5 5 45 2 16 215 Hydrobaenus Hydrob 25 1 15 1 1 15 2 Oliveridia oliverd 75 4 5 115 3 1 25 7 Pseudodiamesa Pseudo 15 75 65 5 9 35 25 2 15 45 P (Mesopsectrocladius) Psecmeso P (Monopsectrocladius) Psecmon 05 P (p) sordidellus Psecp 1 05 Parakieffenella type B ParakB 3 3 1 3 45 1 3 Parakieffenella nigra ParakN 8 15 95 75 105 45 3 2 35 Parakieffenella triquetra ParakTQ Zalutschia type C (Barley type B) ZalutB 15 2 1 1 3 Zalutschia lugulata pauca ZalutL 05 1 25 H gnmshawi-type Hgrim 8 15 5 13 5 10 5 12 05 75 18 7 H marcidus-type Hmarc 55 8 15 85 05 45 85 H maeaeri-type 1 Hmarl 115 185 22 18 5 23 13 5 1 10 5 1 11 H maeaeri-type 2 Hmar2 65 24 11 13 42 5 14 5 19 5 10 75 12 H sp2 Hsp2 1 H subpilosus-type Hsub Pseudodiamesa larvula Hlarv 1 1 1 1 25 1 1 Mesocncotopus Mesoc 1 05 05 1 1 1 2 Nanocladius Nanoc 05 Trissocladius Tnssocl 1 Cricotopus intersectus Cinter 35 05 1 Orthocladius oliveri Orthohv 05 05 2 35 Cricotopus trifasiatus type Cnctnf 1 Cricotopus type P CncP 05 05 1 Cricotopus obnixus type CncOb 05 1 05 Cricotopus sylvestns-type CncSyl Cricotopus cyhndraceus-type CncCly 1 2 1 Paracladius Paracl 85 11 45 35 6 1 1 1 2 Smittia Smittia 1 05 05 1 05 Chaetocladius Chaet Corynoneura arctica Cory Arc 1 2 2 1 35 2 1 1 Eukieffenella gr / Tventia Eukif 65 23 9 14 5 20 05 4 15 125 13 Limnophyes Limno Orthocladius tngonolabis-type Orthotn 1 Orthocladius type S OrthoS 8 12 5 55 8 8 4 35 05 4 75 Orthocladius sp 2 (Walker) Ortho2 Orthocladius consobrmus-type Orthocon 05 Orthocladius nvulorum-type OrthoRiv 1 Orthocladius type I Orthol 1 2 Metnocnemus gr Metrocn Thienemanmella type E ThienE 1 1 3 236 Appendix Table 28: Tanytarsini, Diamesinae, and Chironomini of Baker Lake (15.5 20 cm downcore) Taxon Depth 15 5 16 165 17 17 5 18 185 19 19 5 20 Code Tanytarsini no pedastule - unident TanyNopod 2 1 25 1 1 25 Tanytarsus no spur unident TanyNoSp 1 Tanytarsus lugens-type TanyLug 1 4 25 3 4 1 3 Tanytarsus pallidicomis-type Tanypall 1 Tanytarsus chmyensis-type Tanychy Micropsectra contracta type Mcont 25 4 8 9 85 2 Cladotanytarsus mancus Cladman Micropsectra pallidula Mpadu Micropsectra insignilobus Minsig 6 65 35 11 12 5 1 05 1 25 Micropsectra type R MicroR 1 4 05 6 Paratanytarsus penicillatus-type Parap 3 4 6 55 1 15 Paratanytarsus austnacus-type ParaAUS 4 1 Corynocera oliveri Coryoliv 1 Zavrelia / Stempellinella ZavStemp Constempellina Constem 05 1 05 05 Protanypus Protany 05 3 2 35 4 1 2 15 1 Monodiamesa Monodia 1 05 1 Sergentia Serg 15 2 1 Chironomus anthracinus-type ChirANT 1 Chironomus plumosus-type ChirPL Stictochironomus type B StictoB Stictochironomus StictoR 2 Harnischia /Paracladopelma Harnesh Chironomini larvula Clarvula Arctopelopia / Thienemannimyia ArcT 1 Procladius Proclad total 93 5 212 1515 155 255 5 50 5 57 5 137 5 237