Phenology and Taxonomic Composition of Lotic Chironomidae (Diptera) Communities in Contrasting Thermal Regimes

Phenology and taxonomic composition of lotic Chironomidae (Diptera) communities in contrasting thermal regimes

A DISSERTATION SUBMITTED TO THE FACULTY OF THE GRADUATE SCHOOL OF THE UNIVERSITY OF MINNESOTA BY

Raymond William Bouchard, Jr.

IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY

Advisor: L.C. Ferrington, Jr.

December 2007

UMI Number: 3289167

UMI Microform 3289167 Copyright 2008 by ProQuest Information and Learning Company. All rights reserved. This microform edition is protected against unauthorized copying under Title 17, United States Code.

ProQuest Information and Learning Company 300 North Zeeb Road P.O. Box 1346 Ann Arbor, MI 48106-1346

Permission for CHAPTER VII, reproduced from Aquatic Insects 28: 57-66:

Dear Will Bouchard,

Thank you for your correspondence requesting permission to reproduce the following material from our Journal in your thesis.

Bouchard, R.W., Jr., M.A. Carrillo & L.C. Ferrington, Jr, "Lower lethal temperature for adult male Diamesa mendotae Muttkowski (Diptera: Chironomidae) a winter-emerging aquatic insect" Aquatic Insects, vol. 28 (2006) 57-66.

We will be pleased to grant entirely free permission on the condition that you acknowledge the original source of publication and insert a reference to the Journal’s web site:

http://www.informaworld.com

Thank you for your interest in our Journal.

Yours sincerely

Robert Blackmore Permissions Administrator (UK) Journals

i Permission for CHAPTER VIII, reproduced from Hydrobiologia 568: 403-416:

Dear Sir,

With reference to your request (copy herewith) to reprint material on which Springer Science and Business Media controls the copyright, our permission is granted, free of charge, for the use indicated in your enquiry.

This permission - allows you non-exclusive reproduction rights throughout the World. - permission includes use in an electronic form, provided that content is * password protected; * at intranet; - excludes use in any other electronic form. Should you have a specific project in mind, please reapply for permission. - requires a full credit (Springer/Kluwer Academic Publishers book/journal title, volume, year of publication, page, chapter/article title, name(s) of author(s), figure number(s), original copyright notice) to the publication in which the material was originally published, by adding: with kind permission of Springer Science and Business Media.

The material can only be used for the purpose of defending your dissertation, and with a maximum of 100 extra copies in paper.

Permission free of charge on this occasion does not prejudice any rights we might have to charge for reproduction of our copyrighted material in the future.

Nel van der Werf Assistant Rights and Permissions/Springer

ii ACKNOWLEDGEMENTS

During the course of this study many people provided me with help and support. First, I must thank my advisor Len Ferrington for the help, guidance, support, and advice I have received from him over the last eight years. In addition, I thank my thesis committee, David Andow, Ralph Holzenthal, and Bruce Vondracek, for their help and useful guidance to keep me on track.

I would also like to thank members of the Chironomid Research Group (CRG), whose advice and humor helped me through this process: Brenda Asmus, Giana Gelsey, Moriya Rufer, Brian Schuetz, Adam Sealock, and Claire Serieyssol. Our meetings, both in formal lab meetings and informal discussions, greatly helped me focus and improve my research. In addition, I am appreciative of help I received from other members of the University of Minnesota community including friends and colleagues within and without the Department of Entomology: Stanley Asah, Mario Carrillo, Andy Graves, Byron Karns, Steve Kells, Henrique Paprocki, and Desi Robertson. Many people outside the University have also contributed directly to this thesis and more generally to the paths that I have taken in my studies: Jon Gelhaus, Susan Gresens, Barbara Hayford, and Don Huggins. Identifications of Stempellinella were provided by Torbjørn Ekrem. Working with all of these folks has been a pleasure and I hope we are able to continue collaboration into the future.

This work was partially funded by the University of Minnesota (UMN) Doctoral Dissertation Fellowship, several Bell Museum of Natural History Dayton and Wilkie Fellowships, and several Chiang Student Travel Fellowships, Department of Entomology, UMN.

I am also appreciative for the help and permission to sample some great streams in Minnesota. Specifically, I thank the Belwin Foundation and Janice Odegaard for the

iii opportunity to work in Valley Creek. In addition, I thank Dakota County Parks and the many property owners that allowed me to sample sites on their property.

Finally, but perhaps most importantly, I thank my family. My parents, Ray and Judy are responsible for my interest in aquatic systems and natural history in general. I was fortunate to have the support of my parents and brother, Durell, through my many years of study. Most of all I thank my wife Erin for supporting me through my studies as well as being willing to proofread my thesis. Of course I also have to thank Gus for letting me get work done once in a while and giving me plenty of excuses to take needed breaks.

iv Raymond William Bouchard, Jr. 334 Words

PHENOLOGY AND TAXONOMIC COMPOSITION OF LOTIC CHIRONOMIDAE (DIPTERA) COMMUNITIES IN CONTRASTING THERMAL REGIMES

ABSTRACT

Temperature is generally assumed to be the most important factor influencing aquatic insects. However, for most aquatic insect species sufficient information is not available to predict how temperature affects their life history and distribution. To improve the understanding of this relationship, the influence of temperature on the phenology, composition, and community dynamics of chironomid communities in Minnesota streams was performed followed by a detailed study on a common winter-emerging species, Diamesa mendotae Muttkowski. The phenology and composition of chironomid communities was measured using biweekly collections of surface floating pupal exuviae from six thermally stable and six thermally variable streams in east-central Minnesota. The results of this assessment indicate that chironomid taxa richness is significantly influenced by thermal variability with thermally variable streams in this region supporting many more taxa. This pattern was driven by the high summer temperatures in thermally variable streams which support a diverse fauna of warm-adapted taxa (i.e., Chironominae and Tanypodinae) that were largely excluded from more thermally stable streams. Differences in thermal preferences among congeneric taxa did not appear to increase temporal partitioning of the habitat to permit a greater number of species to coexist. These patterns indicated that temporal shifts in the community were therefore largely manifest at the subfamily or tribe level. Autecology research on D. mendotae indicated that this species is well adapted to winter activity, with a supercooling point in the adults of -21.5 °C and freeze tolerance in the larvae. Life history studies indicate that the growth of this species is largely limited to habitats with temperatures below 10 °C. In

v thermally buffered streams, this species was bivoltine or multivoltine with the bulk of larval development occurring during winter. Both community-level and species-level research provides a better understanding of how temperature regulates chironomid species and communities in small streams of the upper Midwest. This knowledge is important if these aquatic communities are to be used more effectively in biological monitoring and will improve predictions of the impacts of climate change on aquatic species and communities.

vi TABLE OF CONTENTS

COPYRIGHT PERMISSIONS ...... i ACKNOWLEDGEMENTS ...... iii ABSTRACT ...... v LIST OF TABLES ...... xv LIST OF FIGURES ...... xvii

CHAPTER I: INTRODUCTION, METHODS, AND STUDY SITE DESCRIPTIONS FOR “PHENOLOGY AND TAXONOMIC COMPOSITION OF LOTIC CHIRONOMIDAE (DIPTERA) COMMUNITIES IN CONTRASTING THERMAL REGIMES” ...... 1 INTRODUCTION ...... 1 STUDY SITES...... 3

SITE DESCRIPTIONS ...... 3 METHODS ...... 6

COLLECTION OF LAND USE/LAND COVER, SUBSTRATE, DISCHARGE, TEMPERATURE,

AND WATER QUALITY DATA ...... 6

SAMPLING SFPE ...... 10

LABORATORY PROCESSING ...... 10 Sample Picking...... 10 Subsampling ...... 10 Slide mounting and identification ...... 11 CHARACTERIZATION OF STUDY SITES ...... 12

LAND USE/LAND COVER ...... 12

SUBSTRATE ...... 13

DISCHARGE ...... 15

THERMAL REGIMES ...... 15

DISSOLVED OXYGEN ...... 20

WATER QUALITY ...... 20

vii MULTIPLE FACTORS ...... 23

STUDY SITE SUMMARY ...... 23

CHAPTER II: EMERGENCE PATTERNS AND FACTORS CONTROLLING PHENOLOGY AND EMERGENCE TIMING OF CHIRONOMIDAE (DIPTERA) ...... 26 INTRODUCTION ...... 26 MEASURING EMERGENCE IN CHIRONOMIDAE ...... 29 FACTORS INFLUENCING EMERGENCE IN CHIRONOMIDAE ...... 32

DURATION OF CHIRONOMIDAE LIFE STAGES ...... 34 Egg Stage ...... 34 Larval Stage ...... 36 Pupal stage...... 37 Adult stage and oviposition ...... 37

TEMPERATURE ...... 38

NUTRITION ...... 45

PHOTOPERIOD ...... 47

LUNAR AND TIDAL PERIODS ...... 49

INTERMITTENCY AND DISCHARGE ...... 51

SUBSTRATE ...... 52

DISSOLVED OXYGEN ...... 53

SALINITY ...... 54

POLLUTION ...... 54 Heavy Metal Pollution ...... 55 Pesticides ...... 56 Thermal effluents ...... 56

BIOTIC INTERACTIONS ...... 56 Intraspecific interactions ...... 56 Interspecific interactions ...... 57 Competition ...... 57 Predators...... 58

viii Parasites & diseases ...... 58

SEX ...... 58

INTERACTIONS OF FACTORS ...... 59 EMERGENCE PATTERNS AND STRATEGIES ...... 61

VOLTINISM ...... 62

EMERGENCE TIMING ...... 65

PROTANDRY ...... 67

SEASONAL SYNCHRONOUS EMERGENCE ...... 68 Mechanisms of Synchronous Emergence ...... 68 Advantages of Synchronized Emergence ...... 70 Predator satiation ...... 70 Mate finding and oviposition in appropriate habitats ...... 70 Reproductive isolation ...... 71

ASYNCHRONOUS EMERGENCE ...... 72 Mechanisms of Asynchronous Emergence ...... 72 Advantages of Asynchronous Emergence ...... 72 Risk reduction ...... 72 Intraspecific resource partitioning ...... 73 EMERGENCE VARIABILITY ...... 73

EMERGENCE TIMING VARIABILITY ...... 74

VOLTINISM VARIABILITY ...... 75

POSSIBLE GENETIC DIFFERENCES ...... 77 SUMMARY & RECOMMENDATIONS FOR FUTURE RESEARCH ...... 77

CHAPTER III: USE OF SURFACE FLOATING PUPAL EXUVIAE TO MEASURE AND ASSESS CHIRONOMIDAE (DIPTERA) COMMUNITIES IN MID-ORDER NORTHERN TEMPERATE STREAMS ...... 81 ABSTRACT ...... 81 INTRODUCTION ...... 82 METHODS ...... 88

ix STUDY SITES AND SAMPLING OF SFPE ...... 88

DATA ANALYSIS ...... 88 Subsampling ...... 89 Sampling frequency ...... 90 Subsampling and sampling frequency ...... 92 RESULTS ...... 92

SUBSAMPLING ...... 93

SAMPLING FREQUENCY ...... 96

SUBSAMPLING AND SAMPLING FREQUENCY ...... 100 DISCUSSION ...... 102

SUBSAMPLING ...... 102

SAMPLING FREQUENCY ...... 105

SUBSAMPLING AND SAMPLING FREQUENCY ...... 107 CONCLUSIONS...... 109

CHAPTER IV: RICHNESS, TAXONOMIC COMPOSITION, AND PHENOLOGY OF CHIRONOMIDAE IN NORTHERN TEMPERATE STREAMS WITH CONTRASTING THERMAL REGIMES ...... 111 ABSTRACT ...... 111 INTRODUCTION ...... 113 METHODS ...... 119

STUDY SITE, SAMPLING, AND SAMPLE PROCESSING ...... 119

DATA ANALYSIS ...... 119 Richness ...... 119 Taxonomic composition and subfamily/tribe emergence patterns...... 120 Emergence patterns ...... 122 RESULTS ...... 123

RICHNESS ...... 123

TAXONOMIC COMPOSITION ...... 124

SUBFAMILY/TRIBE EMERGENCE PATTERNS ...... 129

x COMMUNITY EMERGENCE PATTERNS ...... 133 Estimated Emergence Duration ...... 133 Community Turnover ...... 134 DISCUSSION ...... 137

RICHNESS ...... 137

TAXONOMIC COMPOSITION AND SUBFAMILY/TRIBE EMERGENCE PATTERNS ...... 145

COMMUNITY EMERGENCE PATTERNS ...... 151 CONCLUSIONS...... 153

CHAPTER V: THERMAL PREFERENCES AND PARTITIONING OF CHIRONOMIDAE (DIPTERA) COMMUNITIES IN NORTHERN TEMPERATE STREAMS ...... 155 ABSTRACT ...... 155 INTRODUCTION ...... 156 METHODS ...... 161

STUDY SITES, SAMPLING, AND LABORATORY PROCESSING ...... 161

DATA ANALYSIS ...... 162 Thermal preferences ...... 162 Thermal partitioning ...... 163 RESULTS ...... 164

THERMAL PREFERENCES ...... 164

THERMAL PARTITIONING ...... 171 DISCUSSION ...... 172

THERMAL PREFERENCES ...... 172

THERMAL PARTITIONING ...... 175 CONCLUSIONS...... 183

CHAPTER VI: WINTER GROWTH, DEVELOPMENT, AND EMERGENCE OF DIAMESA MENDOTAE MUTTKOWSKI (DIPTERA: CHIRONOMIDAE) ...... 185 ABSTRACT ...... 185

xi INTRODUCTION ...... 186 METHODS ...... 188

STUDY SITES ...... 188

ESTIMATION OF EMERGENCE USING SURFACE FLOATING PUPAL EXUVIAE ...... 189

ESTIMATION OF LARVAL GROWTH AND DEVELOPMENT ...... 189 In situ enclosures ...... 189 Egg masses ...... 190 Sampling and sample processing ...... 191 Measurement of larvae ...... 192

DATA ANALYSIS ...... 192 RESULTS ...... 194

EMERGENCE PATTERNS ...... 194

IN-STREAM ASSESSMENT OF LARVAL DEVELOPMENT AND GROWTH ...... 197 Temperature and dissolved oxygen ...... 197 Development ...... 197 Growth ...... 200 DISCUSSION ...... 202

EMERGENCE PATTERNS ...... 202

DEVELOPMENT...... 204

GROWTH ...... 206 CONCLUSIONS...... 208

CHAPTER VII: LOWER LETHAL TEMPERATURE FOR ADULT MALE DIAMESA MENDOTAE MUTTKOWSKI (DIPTERA: CHIRONOMIDAE), A WINTER- EMERGING DIAMESINAE...... 210 ABSTRACT ...... 210 INTRODUCTION ...... 211 MATERIALS AND METHODS ...... 212

COLLECTION AND TREATMENT OF TEST SPECIMENS ...... 212

DETERMINATION OF SCPS ...... 212

xii DETERMINATION OF LLT50 ...... 213

CALCULATIONS OF AIR TEMPERATURE METRICS ...... 214

DATA ANALYSIS ...... 214 RESULTS AND DISCUSSION ...... 215 ACKNOWLEDGEMENTS ...... 222

CHAPTER VIII: FREEZE TOLERANCE IN LARVAE OF THE WINTER-ACTIVE DIAMESA MENDOTAE MUTTKOWSKI (DIPTERA: CHIRONOMIDAE): A CONTRAST TO ADULT STRATEGY FOR SURVIVAL AT LOW TEMPERATURES ...... 223 ABSTRACT ...... 223 INTRODUCTION ...... 224 MATERIALS AND METHODS ...... 227

COLLECTION AND TREATMENT OF TEST SPECIMENS ...... 227

DETERMINATION OF SCPS FOR LARVAE, PUPAE, AND ADULTS ...... 227

MEASUREMENT OF LARVAL INSTAR AND PHASE ...... 228

EFFECT OF SHORT-TERM EXPOSURE TO SUBZERO TEMPERATURES ON LARVAL

MORTALITY ...... 229

LONG-TERM MORTALITY AFTER SHORT-TERM EXPOSURE TO SUBZERO

TEMPERATURES ...... 231

ANALYSIS OF DATA ...... 231 SCPs ...... 231 Mortality ...... 231 RESULTS ...... 232

LARVAL INSTAR AND PHASE ...... 233

SCPS ...... 234

EFFECT OF SHORT-TERM EXPOSURE TO SUBZERO TEMPERATURES ON LARVAL

MORTALITY ...... 234

LONG-TERM MORTALITY AFTER SHORT-TERM EXPOSURE TO SUBZERO

TEMPERATURES ...... 235

xiii DISCUSSION ...... 237

SCPS ...... 237

EVIDENCE FOR LARVAL FREEZE TOLERANCE ...... 238

LONG-TERM MORTALITY AFTER SHORT-TERM EXPOSURE TO SUBZERO

TEMPERATURES ...... 239

ECOLOGICAL IMPLICATIONS OF FREEZE TOLERANCE IN D. MENDOTAE LARVAE .....240 Winter activity ...... 241 Diapause ...... 241

LARVA TO ADULT: CHANGING COLD-HARDINESS STRATEGIES ...... 242 ACKNOWLEDGEMENTS ...... 244

APPENDICES ...... 245 Appendix A: Abundance of individual Chironomidae taxa for each study site and study totals ...... 246 Appendix B: Ranked mean site abundance for ground-water dominated (GWD) (n=6) and surface-water dominated (SWD) streams (n=6)...... 260 Appendix C: Taxa with the greater absolute abundance and relative abundance in either GWD or SWD sites...... 268

APPENDIX D: Individual species emergence patterns of chironomids for each study stream ...... 276

APPENDIX F: Violin plots of mean daily water temperature at emergence (i.e., thermal preferences) for chironomid genera in ascending order of mean emergence temperature ...... 349

REFERENCES CITED ...... 352

xiv LIST OF TABLES

Chapter I

Table 1.2: Area above sample site and percent land use/land cover for study sites derived from MPCA monitoring sites which correspond to study sites or near the study sites. The location column provides the site location from which the LULC data was derived in relation to the sites in this study (Ag = Agriculture, Fo = Forest, Ra = Range, Ur = Urban, Wa = Water, We = Wetland; † data not available for study site; * indicates a significant difference between stream classes at the α <0.05 level) ...... 12

Chapter IV

Table 4.1: Taxa richness for chironomid subfamilies/tribes with stream class means (% taxa richness) and p-values derived from a Mann-Whitney U-test comparing GWD and SWD stream classes (TP = Tanypodinae, DI = Diamesinae, PR = Prodiamesinae, OR = Orthocladiinae, CH = Chironomini, TT = Tanytarsini; thermal range is the range of mean daily water temperatures recorded in each stream during one year; * indicates a significant difference between stream classes at the α <0.05 level) ...... 124

Table 4.2: Total taxa richness for chironomid subfamilies/tribes from ground-water dominated (GWD) and surface-water dominated (SWD) sites with these taxa separated into unique and shared taxa between the two stream classes ...... 125

xv Table 4.3a: Thirty taxa with the greatest abundance in GWD streams (GA = GWD site taxon relative abundance, SA = SWD site taxon relative abundance; abundance was calculated as log([GA-SA]+1) and is sensitive to the difference in raw abundance between GWD and SWD sites; relative abundance was calculated as log([GA+1]/[SA+1]) and is sensitive to relative differences in abundance between GWD and SWD sites; the complete table is presented in Appendix C) ...... 127

Chapter VII

xvi LIST OF FIGURES

Chapter I

Figure 1.1: Sample site locations for ground water dominated (GWD) and surface water dominated streams (SWD) ...... 5

Figure 1.2a: Example of a regression of mean daily air temperature 3-day average and mean daily water temperatures used to develop a model to predict mean daily water temperature for a SWD stream (Cedar Creek; the provided regression equation is derived from the relationship of water and air temperatures when 3-day average air temperatures were above -3.5°C) ...... 8

Figure 1.2b: Example of a regression of mean daily air temperature 3-day average and mean daily water temperatures used to develop a model to predict mean daily water temperature for a GWD streams (Valley Creek) ...... 9

Figure 1.3: Box plots of land use/land cover from GWD and SWD stream classes (open box = mean, middle line represents median, upper and lower bounds of box plots represent 75th and 25th percentiles; * indicates a significant difference between stream types at the α <0.05 level) ...... 13

Figure 1.4: Box plots of substrate composition from GWD and SWD stream types (open box = mean, middle line represents median, upper and lower bounds of box plots represent 75th and 25th percentiles; * indicates a significant difference between stream types at the α <0.05 level) ...... 14

Figure 1.5: Biweekly measurements of discharge from six study sites. Gray lines and points represent GWD streams and black lines and points represent SWD streams ...... 15

Figure 1.6: Annual pattern of daily mean temperatures during the sampling period from six GWD (left) and six SWD (right) streams ...... 17

Figure 1.7: Thermal regimes for all sites for 2002, 2003, and 2004 measured as mean daily water temperatures (black line = data recorded by temperature logger, grey lines = modelled thermal data using air temperatures) ...... 19

Figure 1.8: Biweekly spot measurements of dissolved oxygen from six study sites (gray lines and points represent GWD streams; black lines and points represent SWD streams) ...... 20

Figure 1.9: Box plots of mean Total Phosphorus and Total Kjeldahl Nitrogen from the two stream classes (open box = mean, middle line represents median, upper and lower bounds of box plots represent 75th and 25th percentiles; * indicates a significant difference between stream types at the α <0.05 level) ...... 22

xvii

Figure 1.11: Non-metric multidimensional scaling of LULC, substrate, and water quality for the 12 study sites (black points = GWD sites, grey points = SWD sites) ...... 23

Chapter III Figure 3.1: Distribution of the number of exuviae picked from samples collected in ground-water dominated (GWD) and surface-water dominated (SWD) streams ...... 93

Figure 3.2: Percent of total sample richness at each 50 count subsample for samples with >50 specimens in ground-water dominated (GWD) and surface-water dominated (SWD) streams (error bars represent standard error; GWD n=44, SWD n=56) ...... 94

Figure 3.3: Percent error for composition metrics at each 50 count subsample for samples with >50 specimens in ground-water dominated (GWD) and surface-water dominated (SWD) streams (error bars represent standard error; GWD n=44, SWD n=56) ...... 95

Figure 3.4: Percent error for diversity and biotic index metrics at each 50 count subsample for samples with >50 specimens in ground-water dominated (GWD) and surface-water dominated (SWD) streams (error bars represent standard error; GWD n=44, SWD n=56) ...... 96

Figure 3.5: Mean percent of total community identified in single samples from ground- water dominated (GWD) and surface-water dominated (SWD) sites for biweekly sampling periods (error bars represent standard error; n=3-10) ...... 97

Figure 3.6: Mean maximum number of taxa collected for combinations of 1, 2, 3, 4, 5, and 6 samples for ground-water dominated (GWD) and surface-water dominated (SWD) streams (error bars represent standard error; n=6) ...... 98

Figure 3.7: Temporal distribution of individual samples comprising 1, 2, 3, 4, 5, and 6 sample combinations which produced the greatest taxa richness in ground-water dominated (GWD) and surface-water dominated (SWD) streams ...... 99

Figure 3.8: Mean percent of taxa collected from ground-water dominated (GWD) and surface-water dominated (SWD) streams during the period of April to September for different sampling regime intervals using a 1000 count subsample (error bars represent standard error; n = 6) ...... 100

xviii Figure 3.9: Mean percent of taxa collected from April to September under different sampling regime intervals and subsample sizes for both ground-water (GWD) and surface-water dominated (SWD) streams (error bars represent standard error; n = 6) ....101

Chapter IV

Figure 4.1a: Biweekly chironomid subfamily/tribe richness and daily mean water temperature from ground-water dominated (GWD; left column) and surface-water dominated (SWD; right column) streams ...... 130

Figure 4.1b: Biweekly chironomid subfamily/tribe richness and daily mean water temperature from ground-water dominated (GWD; left column) and surface-water dominated (SWD; right column) streams ...... 131

Figure 4.2: Community turnover measured as differences in the Sorenson Similarity Coefficient between sample pairs at different sample intervals for ground-water dominated (GWD; left columns) and surface-water dominated (SWD; right columns) streams ...... 135

Figure 4.3: Mean similarity of adjacent samples for ground-water dominated (GWD) and surface-water dominated (SWD) streams (error bars represent standard error, n =6) .....136

Figure 4.4: Mean values for the Sorenson Similarity Coefficient of adjacent samples versus the mean daily change in temperature over the period between sample pairs for ground-water dominated (GWD) and surface-water dominated (SWD) streams ...... 137

Chapter V

Figure 5.1: Violin plots of subfamily/tribe weighted thermal preferences in ascending order of mean emergence temperature based on the abundance of emergence at different temperatures (solid circle = mean; line = median; upper and lower bounds = range; the width of the plots represent relative abundance and not absolute abundance; n: Diamesinae [4,393], Prodiamesinae [1,409], Orthocladiinae [56,765], Tanytarsini [13,542], Chironomini [6,104], Tanypodinae [1,111]) ...... 165

Figure 5.2: Subfamily/tribe richness at 2 °C intervals for the calculated thermal preferences of 261 taxa from 12 streams ...... 166

Figure 5.3: Phylogenetic patterns of thermal preferences for chironomid subfamilies and tribes from Sæther (2000) (taxa in grey text are “cool-adapted”, genera in black text are “warm-adapted”; black nodes designate a presumed shift from cool-adapted to warm- adapted; taxa in boxes were collected in the present study; the placement of the Telmatogetoninae is uncertain and this subfamily is largely marine so no assumptions about its thermal preferences are included) ...... 167

xix Figure 5.4a: Phylogenetic patterns of thermal preferences for genera in the Tanypodinae, Diamesinae, and Prodiamesinae with subfamily/tribe relationships for the Orthocladiinae and Chironominae from a phylogeny is modified from Sæther (1977) (genera in grey text are “cool-adapted” with thermal preferences below 15 °C, genera in black text are “warm-adapted” with thermal preferences above 15 °C) ...... 168

Figure 5.4b: Phylogenetic patterns of thermal preferences for genera in the Orthocladiinae from a phylogeny is modified from Sæther (1977) (genera in grey text are “cool-adapted” with thermal preferences below 15 °C, genera in black text are “warm- adapted” with thermal preferences above 15 °C; dashed lines indicate uncertain placement) ...... 169

Figure 5.4c: Phylogenetic patterns of thermal preferences for genera in the Chironominae from a phylogeny is modified from Sæther (1977) (genera in grey text are “cool-adapted” with thermal preferences below 15 °C, genera in black text are “warm-adapted” with thermal preferences above 15 °C; dashed lines indicate uncertain placement) ...... 170

Figure 5.5: Thermal preference ranges for multi-taxa genera (open boxes represent mean; middle line represents median; upper and lower bounds of box plots represent 75th and 25th percentiles; whisker caps represent the 10th and 90th percentiles; filled circles represent outliers; filled circles represent outliers; n: 2[98], 3[61], 4[26], 5[15], 6[4], 7[9], 8[5], 9[8], 10[1], 11[4], 15[1]) ...... 171

Figure 5.6: Thermal preference ranges for multi-taxa genera (open boxes represent mean; middle line represents median; upper and lower bounds of box plots represent 75th and 25th percentiles; whisker caps represent the 10th and 90th percentiles; filled circles represent outliers; filled circles represent outliers; n: 2[54], 3[11], 4[6], 5[7], 6[2], 9[1]) ...... 172

Chapter VI

Figure 6.1: In-stream enclosures for Diamesa mendotae in Valley Creek above riffle ..190

Figure 6.2: Instar separation for Diamesa mendotae based on head width and head length ...... 193

Figure 6.3: Emergence patterns for D. mendotae in Valley Creek as measured by SFPE compared with mean daily temperature ...... 195

Figure 6.4: Emergence patterns for D. mendotae in Trout Brook as measured by SFPE compared with mean daily temperature ...... 196

Figure 6.5: Emergence patterns for D. mendotae in Pine Creek as measured by SFPE compared with mean daily temperature ...... 196

xx Figure 6.6: Weekly spot measurements of dissolved oxygen (mg/l) from within the enclosures and outside the enclosures in the stream ...... 197

Figure 6.7: Instar composition and daily mean temperature for Trial I (error bars represent standard error; n=3) ...... 199

Figure 6.8: Instar composition and daily mean temperature for Trial II (error bars represent standard error; n=2) ...... 199

Figure 6.9: Instar composition and daily mean temperature for Trial III (error bars represent standard error; n=3) ...... 200

Figure 6.10: Change in mean larval AFDM (µg) following the start of experiments for all Diamesa mendotae specimens present in samples (error bars represent standard error; Trials I and III: n=3, Trial II: n =2) ...... 201

Figure 6.11: Mean larval AFDM (µg) versus cumulative degree days for all Diamesa mendotae specimens present in samples (error bars represent standard error; Trials I and III: n=3, Trial II: n =2) ...... 202

Chapter VII

Figure 7.1: Mean (± SE) proportion of mortality (one-minute exposure) and cumulative proportion (± SE) of individual adult male Diamesa mendotae freezing (CPIF) at different subzero temperatures. Each mean proportion of mortality represents 29-30 individuals in 3 independent replicates. The CPIF curve represents 29 observations .....217

Figure 7.2: Daily minimum air temperatures from River Falls (Wisconsin, USA) and daily minimum water temperatures from Trout Brook (Minnesota, USA) during 2003. Both locations are within ≈32 km from the study site on the Kinnickinnic River (Wisconsin, USA). Both Trout Brook and Kinnickinnic River are groundwater- dominated streams ...... 219

Figure 7.3: Daily average minimum air temperature (1940-2004) and seven-day running average of the probability of temperature dropping below the LLT50 of adult male D. mendotae (-21.5ºC) at River Falls, Wisconsin, USA. The probability was calculated by summing the number of years for a given date where the temperature dropped below -21.5ºC and dividing the resulting sum by the total number of years (n=65) ...... 220

Chapter VIII

Figure 8.1: Diamesa mendotae larval developmental phases. msth lsh = developing mesothoracic leg sheaths; L1: mesothoracic leg sheaths not obvious or only beginning to differentiate, L2: large, obvious mesothoracic leg sheaths, and L3: mesothoracic leg sheaths touching or nearly touching ventrally with thorax greatly swollen ...... 229

xxi

Figure 8.2: Relationship between head width and length in larvae (n = 81) of Diamesa mendotae. Two larvae from which SCPs were measured were not included because the head capsule was damaged. Circles represent 61 larvae, with 20 individuals overlapping in measurement ...... 232

Figure 8.3: Effect of developmental stage on the supercooling point (SCP) of Diamesa mendotae. Larval developmental phases as in Figure 8.1. The center bars of the box plots represent the median, the upper and lower ends of the boxes represent the 25th and 75th percentiles, the whiskers represent the 10th and 90th percentiles, circles represent outliers, and the filled squares represent the mean. Developmental stages with significantly different (P < 0.05) mean SCPs are indicated by different letters below each box plot as determined by Tukey’s Studentized Range Test. Numbers in parentheses represent sample size. One larva from which SCPs were obtained (n = 83) is not included in the graph because it was too damaged to determine developmental phase ..233

Figure 8.4: Corrected mean proportion of mortality (± SE) (closed circles) and cumulative proportion (± SE) of Diamesa mendotae larvae freezing (open circles, CPIF) after short-term exposure (1 min) to subzero temperatures. Each mean proportion of mortality represents 31 to 50 individuals in 3 to 5 independent replicates. Mean proportions of mortality for all stages followed by different lowercase letters are significantly different (P < 0.05) as determined by Tukey’s Studentized Range Test ....235

Figure 8.5: Long-term survivorship of Diamesa mendotae larvae after surviving short- term exposure to subzero temperatures. Survivorship curves for -25 and -30°C treatments were not included due to low post-experiment survival (0-2.5%). Each treatment consists of 7 to 50 surviving individuals combined from 2 to 5 replicates .....236

Figure 8.6: Pupation and emergence proportions for Diamesa mendotae larvae surviving short-term exposure to subzero temperatures. Treatments consist of a total of 7 to 50 individuals from 2 to 5 replicates ...... 236

xxii CHAPTER I

INTRODUCTION, METHODS, AND STUDY SITE DESCRIPTIONS FOR “PHENOLOGY AND TAXONOMIC COMPOSITION OF LOTIC CHIRONOMIDAE (DIPTERA) COMMUNITIES IN CONTRASTING THERMAL REGIMES”

INTRODUCTION

Despite the importance of Chironomidae in aquatic systems, little is understood about the life histories and ecologies of most species and the factors that control their distribution and abundance. Although temperature is generally assumed to be the most important factor influencing aquatic insects, sufficient information is not available for most aquatic insect species, especially chironomids, to predict how temperature controls their life history and distribution. The study of chironomids and methods to assess the factors that influence their life histories, abundance, and distribution are important if these aquatic communities are to be understood and more effectively used in biological assessment and monitoring. Specifically the ability to determine how temperature influences species and communities is important in the face of climate change, which will have an especially large impact on aquatic systems and their inhabitants. Needed research includes both community-level and autecological assessments to understand how anthropogenic impacts and a changing environment will influence these communities and these aquatic systems in general. Methods such as the use of surface floating pupal exuviae (SFPE) provide an additional and effective tool for studying the effect of temperature on the life history and ecology chironomids. To address these issues, the following chapters are an assessment of the influence of temperature on the phenology, composition, and community dynamics of chironomid communities in streams of the upper Midwest, followed by detailed research on a winter-emerging species common to the region.

1 In this chapter, the methods that were used to collect data for chapters III through VI are presented to reduce repetition in subsequent chapters. Chapter I also includes a description of the study sites used in the following chapters. To better characterize these streams, results are presented on the land use/land cover (LULC), discharge, temperature, water quality, and substrate at study sites. Specifically this information permits comparison of the characteristics of the array of streams used in this study and contrasts the two stream classes (i.e., thermally stable and thermally variable streams) which are the foundation to this research. Chapter II is a review of the literature on the emergence patterns of Chironomids and the factors that influence these patterns. This review paper is important for the remaining chapters as it links emergence, the community measure used in most chapters, to the factors that control this endpoint. Chapters IV, V, and VI use data derived from the collection of chironomid SFPE. A review of this method and an assessment of how it performs in estimating species richness and community metrics are addressed in chapter III. Chapter IV is an assessment of how temperature influences the phenology and composition of chironomid communities by comparing these measures between and among thermally stable and thermally variable streams on a small spatial scale. Chapter V is a further investigation of how these communities are shaped through individual species’ thermal preferences and community interactions. Chapters VI, VII, and VIII focus on a single species of winter-emerging chironomid, Diamesa mendotae Muttkowski. This species is a common and important constituent of many ground-water influenced streams in Minnesota and Wisconsin and is particularly fascinating as a result of its winter activity. Chapter VI is an assessment of how temperature influences the life history of D. mendotae and chapters VII and VIII deal with the cold hardiness of the larvae, pupae, and adults. The totality of these chapters provides a better understanding of how temperature influences both chironomid species and communities in small streams of the upper Midwest.

2 STUDY SITES

SITE DESCRIPTIONS The streams in this study are divided into two classes: ground-water dominated (GWD) and surface-water dominated (SWD) streams. For the purposes of this study these classes are important as GWD streams are thermally stable due to inputs of ground-water, whereas SWD streams lack this buffering and are thermally variable. This nomenclature is used because the commonly used terms such as warm-water and cold-water streams are misleading as these terms are only accurate for part of the year in temperate regions. For example, cold-water streams are relatively warm during the winter compared to warm- water streams which are often ice covered in Minnesota. Therefore, GWD is used to describe streams typically termed cold-water streams and SWD is used for warm-water streams. The GWD streams are fed by thermally buffered spring-fed streams and/or by direct ground-water inputs. The ground-water feeding GWD streams and their tributaries enters the streams near the average mean air temperature (van der Kamp 1995), which in this region is ≈7 °C. As a result of the input of ground-water, these streams can be easily identified because they never fully freeze over in winter whereas SWD streams have ice cover for at least part of the winter. This criterion was used to select the streams in this study. However, it should be cautioned that this criterion to separate stream classes will vary at different latitudes, altitudes, and for different habitat types.

The data from chapters I, III, IV, V, VI, and VIII were collected from twelve stream sites located in eastern Minnesota, USA (Anoka, Scott, Chisago, Dakota, Washington Co.), consisting of six SWD and six GWD streams (Table 1.1; Figure 1.1). The data used in chapter VII were collected at an additional GWD site in western Wisconsin, USA (Pierce Co.; 44.8312°N, 92.7327°W) located on the Kinnickinnic River. Only limited data characterizing the Kinnickinnic River site are available so a comprehensive analysis of this site is not provided. The study sites are all located on relatively small, wadeable streams. Streams with minimal anthropogenic impact were selected for study, but some degree of human influence is unavoidable in this region where agriculture and residential

3 development are prominent characteristics of the landscape. Study sites consisted of reaches upstream of any anomalous features (e.g., bridges) that might artificially influence stream conditions. In some cases sampling took place downstream of bridges, but in these cases the sample site was located >200 m downstream of the bridge (i.e., Rush Creek, Valley Creek).

Table 1.1: Study stream type, locations, and year sampled with the location of sites from which water quality (WQ) and land use/land cover (LC) data were derived from other sources (GWD = ground-water dominated, SWD = surface water dominated). Sample WQ/LC Site Type County Coordinates Year Location 44.7752°N, Eagle Creek GWD Scott 93.3855°W 2003 same 44.5435°N, Pine Creek GWD Dakota 92.9003°W 2003 same 44.5453°N, Trout Brook GWD Dakota 92.8057°W 2003 ≈0.25 km down 45.0762°N, Browns Creek GWD Washington 92.8101°W 2002 same 45.1973°N, Mill Stream GWD Washington 92.7709°W 2002 ≈0.25 km up 44.9188°N, Valley Creek GWD Washington 92.8006°W 2002 ≈0.25 km up 45.4031°N, Cedar Creek SWD Anoka 93.2094°W 2004 ≈2 km down 44.5216°N, Chub Creek SWD Dakota 93.1811°W 2003 ≈4 km down 44.7608°N, Credit River SWD Scott 93.3424°W 2004 ≈2 km down 45.7143°N, Rock Creek SWD Chisago 92.8749°W 2003 ≈4 km up 45.6576°N, Rush Creek SWD Chisago 92.8967°W 2003 ≈0.5 km up 45.5584°N, Sunrise River SWD Chisago 92.8575°W 2003 ≈3 km up

Figure 1.1: Sample site locations for ground water dominated (GWD) and surface water dominated streams (SWD).

5 METHODS

COLLECTION OF LAND USE/LAND COVER, SUBSTRATE, DISCHARGE, TEMPERATURE, AND

WATER QUALITY DATA At six sites (three SWD [Cedar, Chub, Credit] and three GWD [Eagle, Pine, and Trout]) dissolved oxygen (DO), discharge, and qualitative substrate assessments were performed biweekly during each sampling event. The measurement of DO was performed using a handheld YSI 85 DO and conductivity meter. Substrate cover was determined following rapid bioassessment protocol methods (Barbour et al. 1999) and involved visual estimation of the percent cover of each of five substrate types (i.e., boulder [>256 mm], cobble [64-256 mm], gravel [2-64 mm], sand [0.06-2 mm], and silt [0.004-0.06 mm]). Although substrate estimations did not change greatly over the course of the study year, annual average values of substrate were used. A qualitative habitat assessment for the remaining six sites (Browns, Mill, Rock, Rush, Sunrise, Valley) was performed once following the same methods.

Discharge was determined by measuring flow 4/10 from the bottom of the stream using a Marsh-McBirney Flo-MateTM 2000 at a minimum of eight locations along a cross- sectional transect. Flow was only measured every other sample event when possible, but stream cross-section area was measured during most sample events. When ice cover was present or during periods of very high flow, discharge or cross-sectional area could not be measured. To determine discharge for all sample events, a model was developed to permit estimation of discharge the cross sectional area of the stream for each of the six sites. A first-order regression was used to estimate discharge from stream cross sectional areas for Eagle, Pine, Trout, and Cedar sites. A second-order regression was found to fit better for the Chub and Credit sites due to periods of low flow during the summer and autumn. The discharge was better predicted from cross-sectional area for streams with variable discharge compared to the stable streams. For example the r2 for the model from Credit River was 0.999 and 0.991 for Chub Creek. The lower predictability in streams

6 with stable discharge is likely related to the fact that due to the low discharge variability, measurement errors are more obvious.

Nutrient data and LULC was available online for most study sites or near study sites through the Minnesota Pollution Control Agency (MPCA; http://www.pca.state.mn.us/) and the Metropolitan Council (http://www.metrocouncil.org/). Four water quality variables (total phosphorus, total Kjeldahl nitrogen, total suspended solids [TSS], and turbidity) were selected based on their importance and the presence of data for most of the study sites. Data from a single year that most closely corresponded to the sample year for a given site were used. In some cases, only averages of water quality measures were available, so the data presented represents the average of all available readings during the sample year. All four water quality measures were not available for Mill Stream and Rock Creek and total Kjeldahl nitrogen was not available for Cedar Creek. Data on LULC was available for ten sites excluding Credit River and Sunrise River.

An Onset StowAway TidbiT® temperature logger set to measure water temperature every 15 minutes was deployed in each stream during the sample period. Data were downloaded from the loggers every 4 weeks to avoid loss of data resulting from the loss of a logger or logger malfunction. Using the 15 minute interval readings, mean daily temperatures were determined by averaging all temperature readings during each 24 hour period. The minimum, maximum, and the thermal range for each site was determined from these mean daily temperatures. Total degree days for each site were determined by summing the daily mean temperatures for all days where mean temperatures were above 0º C.

Some malfunctions occurred resulting in the loss of short periods of data. In these cases, models were developed from the available water temperature data from each stream and concurrent air temperature data derived from the Midwest Regional Climate Center for River Falls, Wisconsin, USA. Linear regressions of the three-day average of air temperature were determined to provide the best prediction of water temperature (Figures

7 1.2a & 1.2b). This relationship alone was sufficient to predict water temperatures in GWD streams (Figures 1.2b), but in SWD streams this relationship was complicated due to the effect of ice cover during the winter (Figures 1.2a). For SWD streams, the linear regression was used above an air temperature threshold which ranged from 0-3.5°C depending on the site. Below this air temperature threshold water temperatures were predicted to be 0°C. The r2 for the linear regressions were relatively high ranging from 0.76 to 0.92 and the models were deemed sufficient to predict mean daily water temperatures. In addition to assessing yearly differences in the thermal regimes over the course of the study (i.e., 2002-2004), the models developed were used to predict water temperatures in each stream for the two years from which temperatures were not deployed.

8 Figure 1.2b: Example of a regression of mean daily air temperature 3-day average and mean daily water temperatures used to develop a model to predict mean daily water temperature for a GWD streams (Valley Creek).

Results from LULC, substrate, DO, discharge, temperature, and water quality were compared graphically using box plots and line plots created in SigmaPlot©. Differences between the two stream classes for LULC, substrate, and water quality were tested using a Mann-Whitney U-test in the NCSS© program (Hintze 2001). The Mann-Whitney U- test is a nonparametric test used in place of a two-sample t-test when the assumptions for normality are not met and is useful when there are few items in each sample (Sokal & Rohlf 1995, Hintze 2001). One of the assumptions of the Mann-Whitney U-test is that there are no ties; however, this assumption was not met for several comparisons so the Approximation With Correction in NCSS© was used if ties were present. Due to the small sample sizes this nonparametric test was used to test for significant differences in LULC, substrate, temperature, and water quality between the stream classes. To assess differences between individual streams and stream classes resulting from interactions of multiple factors, a non-metric multidimensional scaling using LULC, substrate, and water quality data was performed using NCSS©.

9 SAMPLING SFPE Chironomidae SFPE samples were collected on a bi-weekly basis for the duration of one year (i.e., 26 sampling events), following the methods of Ferrington et al. (1991). Samples were collected for one year during 2002, 2003, and 2004 depending on the site (Table 1.1). Samples consisted of a timed 10-minute sampling period within the sample reach. Working upstream, SFPE were sampled by scooping exuviae into a pan from areas where they collect (e.g., snags, in vegetation, back eddies) and pouring this material through a 125-µm sieve (i.e., standard testing sieve no. 120). Samples were then transferred to 118-ml jars and preserved with 75% ethanol.

LABORATORY PROCESSING Sample Picking In the laboratory, samples were sieved and rinsed in a 125-µm sieve to remove the preservative. A small portion of the sample was placed in a picking tray, and under a dissecting microscope, SFPE were picked from the sample into 3.7-ml vials with 75% ethanol. After a complete pass of the sample tray, the sample was swirled and scanned again. This was repeated until two successive passes did not recover any additional SFPE. Picking of specimens was halted when the entire sample was picked or a total of 1,000 specimens were picked. A subsample of 1,000 is greater than many studies using SFPE and this sample size was deemed to be sufficient to collect an adequate proportion of the community (Chapter III). Specimens that were broken, covered in extensive fungal hyphae, or dried and compressed were not picked or counted to avoid identification problems or uncertainty regarding the age of the SFPE. Whole pupae or SFPE with adults still attached to the skin in some way were also not counted; however, this material including adults was picked from the sample to aid with identification of SFPE.

Subsampling For six streams (Cedar Creek, Chub Creek, Credit River, Eagle Creek, Pine Creek, Trout Brook), samples were picked from each collection in 50-count subsamples. After

10 removal of preservative, the detritus and SFPE were transferred to a 1,000-ml beaker and enough water was added to suspend the sample. The sample was swirled to suspend the collection material and a small amount was quickly poured into a picking tray. Beginning from one side of the tray, SFPE were picked into 3.7-ml vials containing 75% ethanol. Fifty SFPE were picked into each vial and this was repeated until the sample was completed or 1,000 specimens were picked.

Slide mounting and identification Before slide mounting, specimens were divided into morphotaxa under a dissecting microscope. Exuviae were dehydrated in 95% ethanol, dissected, and slide mounted in Euparol. Five vouchers for each morphotaxon were slide mounted individually using 18- mm round cover glasses. Any remaining specimens of each morphotaxon were mounted with 1-30 specimens per slide using an 18-mm square cover glass. Specimens could be identified to morphotype under a dissecting microscope and in cases where a particular morphotype was abundant (>40 specimens), only 25% of these specimens were slide mounted and identified. The identity of the remaining unmounted specimens was estimated based on the proportions of each taxon in the slide mounted material.

Identifications were made under a compound microscope at 100-400X. Genus-level and some species group identifications were made using Coffman & Ferrington (1996) and Wiederholm (1986). For species-level identifications and species-group identifications the following sources were used: Ablabesmyia (Roback 1985), Brillia (Oliver & Roussel 1983), Cardiocladius (Oliver & Bode 1985), Cladotanytarsus (Bilyj & Davies 1989), Conchapelopia (Roback 1981), Cricotopus (Simpson et al. 1983), Cryptochironomus (Curry 1958, Mason 1986), Dicrotendipes (Epler 1987, Epler 1988), Eukiefferiella (Lehmann 1972), Helopelopia (Roback 1981), Heterotrissocladius (Sæther 1975), Hydrobaenus (Sæther 1976), Labrundinia (Roback 1987), Limnophyes (Sæther 1990), Lopescladius (Sæther 1983, Coffman & Roback 1984), Meropelopia (Roback 1981), Micropsectra (Oliver & Dillon 1994), Nanocladius (Sæther 1977), Nilotanypus (Roback 1986), Odontomesa (Sæther 1985a), Orthocladius (Soponis 1977, Soponis 1987, Soponis

11 1990, Sæther 2003), Parachaetocladius (Sæther & Sublette 1983), Paracladopelma (Jackson 1977), Parakiefferiella (Tuiskunen 1986), Paraphaenocladius (Sæther & Wang 1995), Paratanytarsus (Reiss & Säwedal 1981), Paratendipes (Hayford 1998), Polypedilum (Maschwitz & Cook 2000), Procladius (Roback 1980), Pseudochironomus (Sæther 1977), Rheocricotopus (Sæther 1985b), Saetheria (Jackson 1977), Stenochironomus (Borkent 1984), Tanytarsus (Ekrem et al. 2003), Thienemanniella (Hestenes & Sæther 2000), and Tribelos (Grodhaus 1987). Additional taxonomic and distributional information was obtained from Sæther (1969) and Oliver et al. (1990). Voucher specimens are deposited in the University of Minnesota Insect Collection.

CHARACTERIZATION OF STUDY SITES

LAND USE/LAND COVER Table 1.2: Area above sample site and percent land use/land cover for study sites derived from MPCA monitoring sites which correspond to study sites or near the study sites. The location column provides the site location from which the LULC data was derived in relation to the sites in this study (Ag = Agriculture, Fo = Forest, Ra = Range, Ur = Urban, Wa = Water, We = Wetland; † data not available for study site; * indicates a significant difference between stream classes at the α <0.05 level). Area* Ag Fo Ra Ur Wa We* Site (km2) (%) (%) (%) (%) (%) (%) GWD Streams Eagle 2.8 31.6 30.8 30.2 0.4 0.5 6.3 Pine 51.8 57.4 4.9 35.5 1.4 0.0 0.8 Trout 45.6 57.2 5.9 36.5 0.1 0.0 0.3 Browns 88.3 25.9 13.3 46.7 5.1 5.3 3.8 Mill 19.9 32.0 22.5 37.8 0.2 1.9 5.6 Valley 20.1 32.5 16.6 50.2 0.3 0.0 0.4 Mean 43.8 36.9 14.6 42.8 1.4 1.8 2.5 SWD Streams Cedar 71.7 29.7 19.5 19.1 1.3 0.2 30.1 Chub 109.6 43.3 10.3 29.8 0.7 1.1 14.7 Credit† 133.1 ------Rock 167.3 24.0 12.7 45.5 1.3 0.4 16.0 Rush 146.6 19.2 16.4 32.4 2.7 8.9 20.3 Sunrise† ------Mean 125.7 29.1 14.7 31.7 1.5 2.7 20.3 p-value 0.0106 0.17140.9143 0.1714 0.3359 0.4500 0.0095 12 On average GWD site watersheds were smaller than SWD site watersheds with both stream class watersheds dominated by agriculture and rangeland (Table 1.2). Forest land cover also contributed considerably to the LULC in most of the study site watersheds. Urban and water land cover was only a small percentage of the watersheds in both stream classes. In SWD site watersheds only, wetland land cover contributed considerably to the total land cover, and was the only significant (α < 0.05) LULC difference between the two stream classes (Figure 1.3).

SUBSTRATE Substrates at most study sites consisted of a mix of cobble, gravel, sand, and silt (Table 1.3). Exceptions to this were Cedar Creek which was dominated by silt and Rush Creek which had a large percentage of boulder (i.e., 50%). Gravel and sand on average were more common in GWD streams, whereas boulder, cobble, and silt were more common in SWD streams. However, there were no significant differences (α < 0.05) in substrate composition between GWD and SWD stream sites (Figure 1.4).

13 Figure 1.4: Box plots of substrate composition from GWD and SWD stream types (open box = mean, middle line represents median, upper and lower bounds of box plots represent 75th and 25th percentiles; * indicates a significant difference between stream types at the α <0.05 level).

Table 1.3: Average percent substrate cover estimated from field observations for study sites, and average of values for the two stream classes (* indicates a significant difference between stream classes at the α <0.05 level). Site % Boulder % Cobble % Gravel % Sand % Silt GWD Streams Eagle 0 0 12 62 25 Pine 0 0 10 56 34 Trout 1 11 24 54 9 Browns 0 20 30 30 20 Mill 5 20 55 15 5 Valley 5 30 35 20 10 Mean 2 14 28 39 17 SWD Streams Cedar 0 0 0 0 100 Chub 0 5 23 49 23 Credit 4 26 28 23 19 Rock 10 10 20 40 10 Rush 50 20 10 20 0 Sunrise 0 30 40 20 10 Mean 11 15 20 25 27 p-value 0.7319 0.8707 0.4225 0.2946 1.0000

14 DISCHARGE Average discharge for the six streams from which discharge data was available ranged from 0.22 to 0.64 m3/s. Average discharge was greater for SWD streams and, with the exception of Cedar Creek, annual discharge patterns in SWD streams were more variable compared to GWD streams (Figure 1.5). The discharge in these variable SWD streams fluctuated most during spring and early summer (i.e., April through June), resulting from several spates including a large spate in Credit River of 3.27 m3/s. Credit River and Chub Creek had low discharge during much of the summer and autumn including a lengthy period of near zero flow in Chub Creek. This was related to a drier than usual summer and autumn in 2003, which was not recorded for Credit River as it was sampled in 2004.

Figure 1.5: Biweekly measurements of discharge from six study sites. Gray lines and points represent GWD streams and black lines and points represent SWD streams.

THERMAL REGIMES As expected, annual and diel thermal variability was significantly greater in SWD streams compared to GWD streams (Figure 1.6; Table 1.4), with an average range of only 14.28 ºC in GWD streams versus 24.77 ºC in SWD streams. Overall, SWD streams were colder in the winter and warmer in the summer resulting on average in a 10.50 ºC difference in the annual range of mean daily water temperatures. During winter (i.e.,

15 December-February), temperatures in SWD streams were at or near 0 ºC when ice was present on these streams. There is little fluctuation in water temperatures from 0 ºC during January and February in SWD streams despite low air temperatures. In contrast, winter water temperatures in GWD streams were warmer although temperatures ranged from ≈1-9º C among the six sites during this period. In addition, within GWD streams there was considerable variation in mean daily water temperature through the winter with fluctuations of ≈4º C occurring every few weeks. Air temperature had a considerable effect on water temperature (see Figures 1.2a & 1.2b). This relationship was further evidenced by similar patterns in the highs and lows of the thermal regimes in streams sampled during the same year (Figure 1.6). In addition to different thermal patterns, the total DD accumulated in each stream class was significantly different with an average of 443 DD fewer accumulated in GWD streams (Table 1.4). Although the total DD accumulated are significantly different, in a biological sense these differences are not large and would not be expected to have a large effect on the biota in these streams. Differences in DD are also reflected by mean temperatures, which were also not significantly different, although on average GWD streams were cooler (8.79º C) than SWD streams (10.79º C).

Despite thermal differences in the two stream classes, mean daily water temperatures in both classes could be predicted using a 3-day average of mean daily air temperature. Several other studies have also had good results in estimating water temperature from air temperature (e.g., Crisp & Howson. 1982, Mackey & Berrie 1991). The only major difference in estimating water temperature between the two stream classes is that the use of second-order regression equation as opposed to a first-order regression was more important in SWD streams. This was a result of nearly constant temperatures of 0 ºC recorded during the winter when air temperatures were considerably below 0 ºC. Models developed from air temperature could be used to reasonably predict the influence of increasing air temperatures on water temperatures in these streams and subsequently the impact on the organisms in these streams.

16 from six GWD (left) and SWD (right) temperatures during the sampling period streams. streams. Figure 1.6: Annual pattern of daily mean

Table 1.4: Water temperature (°C) measures for the 12 study sites with average values for the two stream classes (* indicates a significant difference between stream classes at the α <0.05 level). Site Range* Mean* Minimum* Maximum* Degree Days* GWD Streams Eagle 9.47 10.31 5.38 14.84 3765 Pine 15.51 9.01 2.52 18.03 3288 Trout 12.92 8.32 1.05 13.97 3036 Browns 16.59 8.34 0.55 17.15 3044 Mill 17.20 8.20 0.30 17.51 2993 Valley 13.96 8.53 1.70 15.67 3115 Mean 14.28 8.79 1.92 16.19 3207 SWD Streams Cedar 24.67 9.67 0.01 24.67 3541 Chub 25.43 10.03 0.13 25.56 3661 Credit 23.04 10.34 -0.05 23.00 3786 Rock 24.72 9.72 0.05 24.77 3549 Rush 24.75 10.03 0.36 25.11 3662 Sunrise 25.94 9.83 -0.44 25.50 3589 Mean 24.77 10.79 0.01 24.79 3649 p-value 0.0022 0.0152 0.0043 0.0022 0.0411

The thermal regime patterns between the two stream classes appear to be consistent despite the year the temperature was recorded. To assess this further, comparisons of the thermal regimes for each stream in 2002, 2003, and 2004 were compared to identify any large differences in the thermal regime between years that could influence the chironomid community. Regardless of the year, thermal regimes were similar within each site (Figure 1.7) and differences between stream classes where subsequently independent of the year chironomids were sampled.

mperatures (black line = data ured as mean daily water te data using air temperatures). all sites for 2002, 2003, and 2004 meas , grey lines = modelled thermal thermal lines = modelled , grey r recorded by temperature logge Figure 1.7: Thermal regimes for

19 DISSOLVED OXYGEN In general, DO levels in GWD streams were relatively constant throughout the year (Figure 1.8). Unlike Eagle Creek and Pine Creek, there was some change in DO levels during the year in Trout Brook with higher levels present from January through mid March. In contrast, DO levels were more variable in SWD streams, both annually and between sites. Although more variable than the GWD streams, Credit River had the most stable levels of DO among the SWD streams. The levels of DO were quite low in Cedar Creek through much of the summer and into early autumn. These low DO levels are possibly the result of one or more factors including stream profile, large amounts of decaying allochthonous matter, and limited primary production. Chub Creek had an opposite pattern with high levels of DO including supersaturation from mid summer through early winter. In Chub Creek there were large amounts of algae in the stream during this period, which likely produced these high daytime levels of DO.

Figure 1.8: Biweekly spot measurements of dissolved oxygen from six study sites (gray lines and points represent GWD streams; black lines and points represent SWD streams).

WATER QUALITY Average total phosphorus and total Kjeldahl nitrogen were in general higher in most SWD sites (Table 1.5). However, levels of these nutrients also were relatively high in

20 one GWD stream, Browns Creek, which had the highest values of nitrogen for streams in the study (Table 1.5). Levels of TSS and turbidity were similar between the two stream classes with the exception of Browns Creek which again had high levels of these measures (Table 1.5). Despite apparent differences in total phosphorus, total Kjeldahl nitrogen, TSS, and turbidity, there were no significant differences (α < 0.05) between the two stream classes for these water quality measures (Figures 1.9 & 1.10). In general the SWD streams had lower variability compared to the GWD streams, but this was the result of relatively high levels of nutrients and suspended material in Browns Creek. High levels of nutrients and suspended solids in Browns Creek may be related to a large runoff event in July 2002 which resulted in very high levels of the water quality parameters presented here. However, even with the removal of Browns Creek there are no significant differences in the four water quality measures between the two stream types.

Table 1.5: Mean (n) water quality measures for study sites derived from multiple sampling events during one year collected by the MPCA and Metropolitan Council (Phosphorus = Total Phosphorus, Nitrogen = Total Kjeldahl Nitrogen, TSS = Total Suspended Solids; † data not available for study site; * indicates a significant difference between stream classes at the α <0.05 level). Phosphorus Nitrogen TSS Turbidity Site Year (mg/l) (mg/l) (mg/l) (NTU) GWD Streams Eagle 2003 0.08 (22) 0.31 (22) 19 (20) 6 (20) Pine 2006 0.05 (7) 0.28 (8) 5 (5) 2 (8) Trout 2006 0.04 (7) 0.20 (7) 3 (1) 2 (7) Browns 2002 0.25 (20) 1.73 (22) 207 (20) 29 (20) Mill† - - - - - Valley 2002 0.14 (36) 0.66 (36) 61 (32) 12 (27) Mean 0.11 0.64 59 10 SWD Streams Cedar 2006 0.27 (27) 1.20 (27) 50 (27) 13 (26) Chub 2004 0.22 (10) 1.12 (10) 20 (10) 6 (10) Credit 2004 0.13 (12) 9 (9) 8 (12) Rock† - - - - - Rush 2006 0.13 (21) 0.93 (22) 11 (20) 7 (22) Sunrise 2006 0.09 (21) 0.82 (22) 11 (20) 9 (22) Mean 0.17 1.02 20 9 p-value 0.2948 0.1905 1.0000 0.5993

21 Figure 1.9: Box plots of mean Total Phosphorus and Total Kjeldahl Nitrogen from the two stream classes (open box = mean, middle line represents median, upper and lower bounds of box plots represent 75th and 25th percentiles; * indicates a significant difference between stream types at the α <0.05 level).

Figure 1.10: Box plots of mean Total Suspended Solids and Turbidity from the two stream classes (open box = mean, middle line represents median, upper and lower bounds of box plots represent 75th and 25th percentiles; * indicates a significant difference between stream types at the α <0.05 level). 22 MULTIPLE FACTORS The use of non-metric multidimensional scaling indicated that based on multiple measures, the stream classes did not form separate clusters (Figure 1.11). Two sites did appear to have less similarity with the other ten sites: Rush Creek and Cedar Creek. The lower similarity in these sites appears to be related to substrate differences where Rush Creek had a large percentage of boulder and Cedar Creek had a large percentage of silt. However, because the different stream classes did not form separate clusters, LULC, water quality, and substrate are not consistently different between these classes and subsequently would not be expected to be responsible for differences in the chironomid communities between thermally stable and variable streams.

Figure 1.11: Non-metric multidimensional scaling of LULC, substrate, and water quality for the 12 study sites (black points = GWD sites, grey points = SWD sites).

STUDY SITE SUMMARY Consistent differences in measures of temperature, discharge, and DO were observed between GWD and SWD streams. These three parameters are interrelated and reflect differences in the geology in these watersheds which influence the amount of ground-

23 water input. For example, temperature and discharge are directly influenced by the amount of ground-water entering a stream because ground-water inputs result in the stability of both of these variables. However, discharge was less predictable because factors such as the amount of riparian wetlands in a watershed can result in greater discharge stability (e.g., Cedar Creek), but because this water is at or near the surface it is not buffered. As a result, some thermally variable streams with extensive interconnected riparian wetlands have an annual discharge that can be quite stable. Dissolved oxygen is less directly impacted by ground-water input, but it is linked to temperature as well as a number of other factors. However, these other factors can override the effect of temperature. For example, despite similar thermal regimes among the SWD streams, DO patterns were presumably different as a result of primary production, decomposition, and stream profile. Therefore, thermal variability does not always indicate high variability in DO levels. As a result of differences in ground-water influence, the variability in temperature, discharge, and DO is often greater in SWD streams compared to GWD streams. However, discharge and DO are considerably variable because they can be influenced by a number of other factors beside ground-water input. Therefore among these three variables, temperature is the most consistently predictable among the sites when the influence of ground-water and air temperatures is understood in each site.

Water quality, LULC, and substrate composition were not consistently different between the two stream types. The water quality in the stream classes were expected to differ as a result of the source of water in each class. For example, SWD streams receive water that runs off the land and collects potential pollutants, whereas the water entering GWD streams would be expected to receive some filtering as a result of passing through the soil. Surprisingly, there were no significant differences between water quality measures in the two stream classes. The only significant differences in LULC measures between the two stream classes were the percent wetlands. The differences in the percent wetlands are a reflection of differences in geology in these two stream types that also account for the differences in the degree of ground-water influence. Although substrate is a reflection of the geology in a watershed, the geologic differences that determine the

24 degree of ground-water influence do not appear to have a large impact on the substrate composition in the sites selected for this study. Differences in the watershed area between stream classes were significant (p = 0.0106) with the watershed area for most SWD streams larger than GWD streams. This difference could potentially have an impact on the communities at the study sites since a larger watershed could result in greater taxa richness due to drift or colonization from a larger number and potentially more diverse upstream habitats. However, this pattern is not consistent as Browns Creek (GWD) has a larger watershed area than Cedar Creek (SWD). This difference permits the influence of watershed area to be tested because if watershed area has an important influence on taxa richness then Browns Creek would be predicted to have greater taxa richness than the other GWD streams and Cedar Creek. Interactions of multiple factors also did not appear to result in consistent differences between the stream classes based on non-metric multidimensional scaling. As a result, the effects of LULC, substrate, and water quality would not be expected to result in consistent differences in the chironomid communities between these stream types.

The streams used in this study were selected as minimally impacted streams in the region, but as a result of considerable anthropogenic activity reflected by the LULC, there are subtle human impacts to these streams. In much of the Midwest, minimally impacted streams are uncommon and these streams conform to the model condition of wadeable lotic habitats for the region. Perhaps more important is the fact that in most cases there are no significant differences between the stream classes, indicating that community patterns in these streams would be predicted to be largely impacted by the factors that do differ (i.e., temperature, discharge, DO). This permits the assessment of the influence of temperature on these communities, although these other covarying factors can not be completely ruled out. Although temperature does vary with discharge and DO, among the array of study streams temperature is the most consistent and predictable parameter. The consistency in temperature between the two stream classes therefore permits an assessment of the impact of thermal regime on the chironomid communities in these streams.

CHAPTER II

EMERGENCE PATTERNS AND FACTORS CONTROLLING PHENOLOGY AND EMERGENCE TIMING OF CHIRONOMIDAE (DIPTERA)

INTRODUCTION

Adult emergence is commonly used as an endpoint in many aquatic insect studies that collect adults or in some cases the exuviae or cast skins (e.g., Chironomidae, Odonata) as a measure of taxon presence, life history, population size, and productivity. The measurement of emergence in aquatic insects has been used for taxonomic, ecological, productivity, and biological assessment studies. Collections of emerging aquatic insects are commonly used because the taxonomy tends to be better for adults than for immatures. Better taxonomic resolution permits generation of more comprehensive taxa lists and reduces the problems associated with drawing conclusions from results that include lumped taxa with potentially different tolerances and life histories. The use of adult emergence is also beneficial, as the events and the environment experienced by an insect between hatching and emergence are fundamental to the use of insects as bioindicators and measures of ecosystem and population parameters. Therefore, emerging aquatic insects represent the endpoint of the successful utilization of a habitat by the immature stages and the resulting data can be informative of the habitat type and water quality. This is particularly important as chironomids are r-strategists where only a small fraction of the eggs produced develop into viable adults capable of reproduction. Although magnitude of emergence and phenology has been used to effectively study chironomids, there is sometimes little consideration given to what emergence represents. The use of emerging chironomids as measures of taxa richness, community composition, and phenology should include an understanding of the life history dynamics of chironomid species and factors that influence their emergence.

The family Chironomidae is ubiquitous, occurring on all continents, in nearly every aquatic habitat, and under a diverse range of conditions (Cranston 1995). In many cases chironomids are the most abundant and/or taxa rich group in collections of emerging aquatic insects (e.g., Abundance: 71-91% - Morgan 1958, 70-90% - Morgan & Waddell 1961, 81% - Judd 1953, 93% - Judd 1957, 80% - Judd 1960, 99% - Judd 1962, 58% - Judd 1964, 74-85% - Ringe 1974, >76% - Paasivirta 1974, 84% - Nordlie & Arthur 1981, 85% - Jónsson 1987, 75% - Smukalla & Meyer 1988, Richness: 66% – Mason & Lehmkuhl 1983). In fact, chironomids probably dominate richness in most aquatic habitats (Coffman 1973); however, chironomids are not commonly identified species and there are few studies of emergence where both chironomids and other aquatic insects are consistently identified to the species level. Unless a study is narrowly focused on another taxonomic group, collections of emerging aquatic insects will usually include a considerable proportion of chironomids. Despite their importance, there is a lack of information on the autecologies of chironomid species or a comprehensive understanding of the factors that influence their distributions, phenologies, and community dynamics.

Among aquatic insects, timing of emergence is largely determined by exogenous and endogenous cues and environmental constraints operating on the larva and pupa (Corbet 1964). In general, the development of chironomids is considered to be largely determined by temperature and nutrition (Berg & Hellenthal 1992, Tokeshi 1995). However, there are many other factors (e.g., photoperiod, pollution) that can potentially influence developmental rates and timing, although they are generally not assessed because they are often considered to have minimal impact on natural populations. As a result, most research on chironomid life histories has not addressed these factors. Consequently, any discussion is hypothetical without empirical evidence demonstrating that the influence of these factors is generally less important.

Several reviews address the influence of factors such as temperature, nutrition, and photoperiod on the emergence of chironomids and other aquatic insects, as well as their general life history patterns (e.g., Corbet 1964, Ward & Stanford 1982, Butler 1984,

Sweeney 1984, Pinder 1986, Tokeshi 1995). This review provides an update and synthesis of recent research addressing factors that influence the seasonal emergence patterns and timing in Chironomidae, as well as seasonal emergence patterns observed in this family. Diel emergence patterns and their underlying factors are not covered in this review. Although abundance of Chironomidae can vary depending on a wide variety of factors, including many of the factors discussed here, this review only focuses on how factors affect the timing of development of chironomids as they relate to timing of emergence. Furthermore, how body size and fecundity is influenced in chironomids is not discussed, nor are the factors that determine the geographic distribution of chironomid species. These topics are too broad and could not be dealt with in detail in a single review. Many studies have identified a number of factors that impact the growth (i.e., the increase of body size or mass) of chironomids, however most do not explicitly measure development (i.e., the progression through developmental stages which terminate with the adult) (e.g., Wilcox et al. 2005, Walther et al. 2006). Although these processes are generally correlated, in some cases growth and development can be in disequilibrium, where low or high temperatures result in smaller less fecund individuals because larvae do not reach optimal size before metamorphosis (Sweeney & Vannote 1978). Therefore, these studies are avoided in this review as they do not deal explicitly with developmental rates and timing, although it should be mentioned that in many cases growth rates will be correlated to developmental rates.

This review begins with a brief introduction to the methods used to collect emerging chironomids and some of the advantages and disadvantages of various methods. More detailed review of these methods are provided in other sources; however, an overview is needed due to the influence methodology has on measures of chironomid emergence. The largest portion of this review deals with the factors that influence emergence timing in chironomids. These factors include temperature, nutrition, photoperiod, lunar period, tidal period, substrate, discharge, intermittency, pollution, and biotic interactions. Although most studies do not deal with the interaction of factors, multiple factors are probably important in determining emerging timing in chironomids. Therefore, a section

is specifically devoted to the effect of multiple factors although the literature is currently limited in this regard. Many studies have identified emergence patterns for chironomid communities or individual taxa ranging from subfamilies to populations. An overview of these patterns and an assessment of some of the underlying causes of these patterns are assessed. Although emergence timing can be predicted, there is also considerable variability in the emergence timing of chironomids both spatially and temporally, and to address this, a section is devoted to emergence variability and some of the causes of this variability when it is known. Finally, the review is concluded with an overview of topics and questions that require further study or those that remain to be assessed.

MEASURING EMERGENCE IN CHIRONOMIDAE

There are a variety of methods and trap types used to collect emerging chironomids for the estimation of emergence patterns, densities, or the collection of specimens for taxonomic purposes. In general, methods to collect Chironomidae adults or surface floating pupal exuviae (SFPE) can be used to estimate activity periods, as most chironomids are relatively short lived. In addition, the measurement of aquatic insect emergence can also provide information regarding the requirements (e.g., optimal temperatures for growth) of the immature stages, although these interpretations need to be made cautiously. The numbers, taxa, and source of insects collected by different trap types and trap locations vary and need to be considered when data are interpreted. Trap types have advantages and disadvantages and their selection must be appropriate for the purpose and goals of a study. Even for similar trap types, different designs and environmental factors can result in considerable differences in estimation of chironomid emergence (e.g., Kimerle & Anderson 1967, Daniel et al. 1985). Detailed descriptions of the methods and their attributes are not discussed in this review, but comprehensive reviews of the methods to sample aquatic insect emergence can be found in Mundie (1956) and Davies (1984).

Traps and methods used to collect or measure emerging chironomids can be divided into three broad types: 1) traps that collect emerging chironomids from a known area (e.g., emergence/tent traps), 2) traps that collect insects from a wide area but with the exact emergence location unknown (e.g., light traps, flight intercept/window traps), and 3) sampling that collects SFPE from an unknown area but within a single water body (e.g., surface drift nets, hand netting of SFPE). The collection of SFPE obviously differs from the other two trap types in that only the exuviae, as well as some adults, larvae, pupae, and associated specimens are collected. In contrast, the other two methods which employ emergence traps primarily only collect adults. With some traps or methods the placement or specific design of the trap can affect the specimens collected (e.g., source of emerging chironomids). As a result of design and trap placement differences between methods, the spatial scale from which chironomids are collected vary. The sampling methods that collect emerging chironomids from a defined area allow estimation of density, but limit the types and numbers of habitats that can be sampled. In contrast, methods that collect chironomids from a wide area will result in a more comprehensive estimate of species richness of the taxa in that area, but less precision in knowing where specimens originate. For example with window traps, the material collected would be expected to originate from a wide potential area, and therefore can be used as a technique to survey large areas and would be more effective obtaining a more comprehensive survey of large habitats such as lakes (Jónsson et al. 1986). The spatial scale of the collection area for SFPE samples falls between the other two collection types as it is site or water body specific, but the actual emergence location within the habitat or areal density of a species generally can not be determined.

Sampling methods also have different errors and bias. For example, there are some sources of error associated with light trapping to assess emergence patterns. Lewis et al. (1954) determined that increases in the abundance of chironomids in light traps were correlated with increased wind and likely not with emergence abundance. Light trap collection may also have bias toward crepuscular and nocturnal species. For example, Coffman (1974) noted that Corynoneura and Thienemanniella were not collected by light

traps, despite being collected by SFPE sampling. Because these genera were more likely to emerge during the day, the lack of these taxa at light traps reflects a diurnal activity pattern that could result in error in the assessment of such a community. The placement of an emergence trap is important as its location will collect material emerging from below the trap and may provide only a limited assessment of the water body or even habitat.

Chironomid life history parameters (e.g., voltinism) can be difficult to assess from emergence data without complementary benthic sampling to determine instar composition or development rates (Learner & Potter 1974, Butler 1984, Butler & Anderson 1990, Wrubleski & Rosenberg 1990, Berg & Hellenthal 1992, Tokeshi 1995). Life history estimation errors will be especially prevalent for multivoltine species with overlapping generations (i.e., species with life cycle duration exceeding recruitment frequency). Misinterpretation can also occur in univoltine populations that appear bivoltine because part of the cohort emerges while a portion is interrupted by suboptimal conditions (Butler 1984). For example, cohort splitting can occur when part of a population emerges in the autumn and the remainder emerges in the spring due to a halt in emergence caused by low winter temperatures. Even when both larval and imagine/SFPE samples are collected, infrequent sampling can also result in incorrect assessment of voltinism and phenology (Butler 1984, Tokeshi 1995, Boothroyd 1999). To avoid some of these problems, measures of life cycle duration and potential voltinism are sometimes performed in the laboratory. However, the number of potential generations determined in the laboratory often does not correspond to the number observed in nature (e.g., Mackay 1977a). Rossaro (1991) suggests that this apparent discrepancy may be the result of the difficulty in identifying small generations in the field. Other sources of error may also result from the difference between field and laboratory conditions. For example, field populations may be regulated by many more factors than are typically tested in the laboratory (e.g., competition, fluctuating temperatures). To determine voltinism, both larval and adult/SFPE should be collected

frequently enough to monitor cohorts as they develop through the year (e.g., weekly or biweekly intervals depending on the life histories of the community).

FACTORS INFLUENCING TIMING OF EMERGENCE IN CHIRONOMIDAE

The timing of development and emergence in chironomids is determined by environmental conditions and the specific responses of species and individuals to these factors. In general, temperature and food are considered to be the major exogenous determinants of the developmental rate and life history timing in chironomids (Berg & Hellenthal 1992, Tokeshi 1995). Other important factors that can influence the rate of development include dissolved oxygen, substrate, habitat intermittency, pollution, and biotic interactions. All of these factors influence developmental rates by affecting enzymatic, hormonal, or molecular processes, which in turn affect the rate of energy assimilation and the duration that is required to obtain sufficient energy to produce an adult capable of reproduction. Aquatic insect development, and therefore emergence timing, is limited by a combination of the rate of energy assimilation and metabolic costs (Johannsson 1980). Increased developmental times are therefore related to a reduction in optimal conditions or factors that either reduce energy assimilation or increase metabolic costs. Under severe conditions, usually when temperatures are below or exceed physiological thresholds, development will be halted and dormancy will occur until conditions improve or death of the insect occurs.

Periodic factors such as thermoperiod, photoperiod, and tidal period can influence life history timing in chironomids through the initiation or breaking of developmental halts (i.e., dormancy). These exogenous cues generally maintain an internal clock that controls development and emergence timing (Corbet 1964). Although often not specifically addressed in many studies, the interaction of multiple factors is probably responsible for timing of emergence in most chironomid species.

Before proceeding it is necessary to define the term dormancy and the types of dormancy, as the use of this term often varies among authors. For the purposes of this review, dormancy is defined as a period during which the progression of development is halted. Dormancy can be the result of different factors such as temperatures below optimum (i.e., hibernation), temperatures above optimum (i.e., aestivation), or non-thermal factors such as nutrition, photoperiod, and water body intermittency (i.e., athermopause) (Mansingh 1971). Quiescence, oligopause, and diapause are the three degrees of dormancy defined by Mansingh (1971) that will be followed in further discussions of dormancy in chironomids. In most papers, the term diapause is applied loosely to include most forms of dormancy. Here non-feeding, high-intensity dormancy is referred to as diapause and a less strict dormancy during which periods of feeding may take place as oligopause. Diapause generally involves a number of additional behavioral and physiological modifications (e.g., winter-case construction, gut evacuation, burrowing, dehydration), as well as the cessation of feeding. In oligopause there are fewer behavioral changes because activity may be maintained during periods of warmer conditions, although developmental progression is generally halted. This form of dormancy can result in larger adults because energy assimilation can continue, provided that periods of temperatures are high enough for feeding activity to take place. Typically, oligopause is more common in taxa occupying milder habitats that permit some activity during periods of adverse conditions. Quiescence is short-term dormancy as a response to mild adverse conditions, but is not dealt with in this review as this form of dormancy does not commonly appear in studies of chironomids.

Although temperature, nutrition, and numerous other factors appear to affect timing of emergence of all chironomid taxa, the influence of these factors differentially impacts species and populations. This is the result of the different requirements and optima that are a function of different physiologies and behaviors observed in the many species in the family Chironomidae. Most obvious is the ability of some species to develop at high temperatures, whereas other species require low temperatures. These different traits may also include the ability for rapid development and may reflect the type of habitat the

immatures occupy. Typically understanding these species specific traits are at least part of the goal of studies on chironomids, as this information permits the prediction of environmental conditions. To determine these traits, the ecological variables (e.g., temperature, food resources, photoperiod, substrate) must be measured and correlated with the presence and life histories of chironomid species. Therefore, a comprehensive understanding of the traits of chironomid species, including how environmental variables affect the life histories of these insects, is necessary to understand the ecology of aquatic habitats and to use these species in biological assessment.

DURATION OF CHIRONOMIDAE LIFE STAGES Developmental rates for some Chironomidae species have been determined to be fast with emergence occurring in only 7 d for Apedilum elachistus (Nolte 1995), 12 d for Paratanytarsus dissimilis at 28 °C (Nebeker 1973), 14 d for Chironomus kiiensis (Surakarn & Yano 1995), 15 d for Chironomus sancticaroli (Strixino & Strixino 1982), 10-12 d for Chironomus strenzkei (Syrjämäki 1965), and 9-16 days for several species in a desert stream (Gray 1981). Many chironomids are small and fast growing which are potentially adaptations for survival in habitats where mortality rates are high (Huryn & Wallace 2000) or conditions are highly unpredictable. Others have very long life cycles lasting two or more years including arctic species (Sendstad et al. 1976, Welch 1976, Butler 1982, Hershey 1985) and northern temperate species (Butler & McMillan 1990). Typically species with long generation times occupy habitats where growth is limited to a short period each year most commonly due to low temperatures. For example, Butler & McMillan (1990) attributed a three-year life cycle in Chironomus cucini to low profundal temperatures.

Egg Stage The duration of the egg stage in chironomids appears to be primarily influenced by temperature with a decrease in the length of the egg stage as temperatures increase and an increase in hatch failure at both low and high temperatures (Tokeshi 1995). There is a large range (<2-44 days) in the duration of the egg stage under laboratory conditions both

within and between a species (Pinder 1995; e.g., Miall & Hammond 1900, Sadler 1935, Hilsenhoff 1966, Kokkinn 1990, Yano et al. 1991, Jackson & Sweeney 1995, Frouz et al. 2002, Olsen et al. 2003). Salinity has also been determined to have an impact on the duration of the egg stage (Kokkinn 1990). However, at temperatures at which low failure of egg hatch is observed, the duration of the egg stage is generally short (<2-7 days) (Pinder 1995a; e.g., Zavřel 1926, Thienemann 1954, Hilsenhoff 1966, Dejoux 1971, Nebeker 1973, Gray 1981, Jackson & Sweeney 1995, Frouz et al. 2002, Olsen et al. 2003). As a result of the short length of the egg stage, it would be expected to have a minimal impact on the overall life cycle duration and timing of emergence in chironomids. However, the vast majority of studies that assess the egg stage are based on laboratory research so it is not clear what dynamics may occur under natural conditions.

Dormancy in the egg stage is common in aquatic insects including Diptera (Hynes 1970), and egg dormancy by chironomids is mentioned or suggested in several papers (e.g., Tokeshi 1995, Williams 1998). There are some examples where evidence suggests the possibility of egg dormancy (e.g., Heuschele 1969, Mozley 1970, Tokeshi 1986a), but dormancy in this stage is difficult to prove due to sampling limitations in the field (Tokeshi 1995). There are few studies that provide what could be considered direct evidence of dormancy of chironomid eggs. Kettisch (1938) provides evidence for overwintering in the egg stage by Cricotopus trifascia based on observations of modified egg masses on Potamogeton stems during October that could be held for several weeks at 1ºC, but when brought to 20ºC hatched in 2-3 days. Williams & Hynes (1976) provide evidence for several taxa (e.g., Trissocladius, Diplocladius, Orthocladius) utilizing eggs resistant to desiccation to survive through dry periods in intermittent streams. Lack of evidence of dormancy in the egg stage is possibly the result of sampling methods (e.g., large mesh sizes) or sample processing methods which do not collect this stage. Further work is needed to establish whether in some cases chironomids diapause in the egg stage or whether delayed hatching is simply a result of low temperatures slowing development and not true diapause where development is halted. However, the dormancy of the egg stage may simply be unusual in chironomids because the pupal and adult stages are

generally short and oviposition and hatching is likely to take place during periods when pupation and emergence can occur (i.e., during periods of optimal temperatures).

Larval Stage The majority of the life cycle in chironomids is spent in the larval stage as this is the stage where all or nearly all of the energy required for reproduction is acquired (Tokeshi 1995). As a result, the majority of the timing of life history is regulated in the larval stage. The first instar tends to be short in duration (e.g., Danks 1971a), although there are examples of extended development times for this stage (e.g., Hydrobaenus kondoi - Kondo 1996). In most taxa, the greatest amount of time when active development is occurring is spent as a 4th instar, especially in species with long generation times (Ward & Cummins 1978, Ineichen et al. 1979, Rasmussen 1984, Berg & Hellenthal 1992, Tokeshi 1995, Frouz et al. 2002). Although maximal growth occurs in the 4th instar, considerable amounts of biomass can also be accumulated by the 3rd instar (Berg & Hellenthal 1992).

As result of the larva requiring the greatest portion of a chironomid’s life history, much of the timing of development is determined by factors acting on the larval stage. The larval stage is the major life stage in chironomids that undergoes dormancy. The most common larval instars cited for undergoing diapause are the 2nd, 3rd, and 4th instars (e.g., Hilsenhoff 1966, Anderson & Hitchcock 1968, Danks 1971a, Sokolova 1971, Danks & Jones 1978, Matěna 1979, Rasmussen 1984, Lindegaard & Jónsson 1987; Goddeeris 1990, Goddeeris 1991, Groenendijk et al. 1998, Goddeeris et al. 2001, Werle et al. 2004). Less commonly the 1st instar has been identified as overwintering or diapausing (e.g., Shiozawa & Barnes 1977, Ward & Cummins 1978, Sugg et al. 1983), but the relatively small number of studies that have identified diapause in the 1st instar may be the result of sampling methods which usually miss small larvae in other studies (Storey & Pinder 1985, Pinder 1995). However, the 1st instar or larvula is generally considered to be short- lived and is primarily a stage of dispersal (Pinder 1995) and would therefore not be expected to be commonly used as a dormant stage.

The overwintering cohort can be comprised almost entirely of one instar or combinations of instars, although one instar usually dominates (Danks 1971a). There also appears to be relative plasticity in the composition of overwintering instars as a result of different environmental conditions (Danks 1971a).

Pupal stage The pupal stage of chironomids is usually considered to be quite short and can require as little as 4 hrs to several days (e.g., Sadler 1935, Morgan 1949, Hilsenhoff 1966, Gray 1981, Sankarperumal & Pandian 1991, Nolte 1995, Frouz et al. 2002, Olsen et al. 2003, McKie et al. 2004). Because the pupal stage is generally short and is a non-feeding stage, there are a limited number of factors that influence the duration of this stage. Temperature appears to be the major factor determining the duration of the pupal stage, as several experiments have shortened the duration of this stage by increasing temperature (e.g., Sankarperumal & Pandian 1991, Olsen et al. 2003). Although there are examples of chironomids overwintering as prepupae (e.g., Oliver 1968), there is not strong evidence that chironomids undergo dormancy in the pupal stage. However, Morley & Ring (1972) observed that in some marine midges the pupal stage lasted 46 days, which potentially is evidence of pupal dormancy.

Adult stage and oviposition There is little research on the duration of the adult stage, but in general it is considered to be short, usually only a few days (e.g., Syrjämäki 1965, Hilsenhoff 1966) and primarily directed to reproduction (Armitage 1995). As with other stages, temperature is likely the most important factor influencing the duration of the adult stage, although other factors may also be important, such as humidity and sex (Hilsenhoff 1966). As a result of the short duration of the adult state, the time from emergence to oviposition likely has little influence on the phenology of chironomid species. In some cases, adults are relatively long lived and can survive for several weeks under laboratory conditions (Diamesa mendotae: Ferrington et al. submitted) and in the field under controlled conditions (L.C.

Ferrington Jr. unpublished data). If in nature females of Diamesa mendotae are capable of mating and oviposition after several weeks, it could contribute to the length of the period from one generation to another. In several species, the female emerges with eggs incompletely developed and oviposition may not take place for 1-3 days (Oliver 1971, e.g., Sadler 1935, Andersen & Hitchcock 1968, Oliver 1968, Fischer 1969). This could cause some delay in oviposition, but it would generally be short. In other species the female emerges with fully developed eggs (Oliver 1971, e.g., Hashimoto 1957, Oliver 1968) and additional generation time would only be limited to the time required to find a mate, copulate, and oviposit. This period would be especially short in parthenogenetic species (e.g., Edward 1963) and in many marine species where the period from emergence to oviposition can be very short and last only 15 min to 2 h (Neumann 1976). Although several other aquatic insect taxa can be dormant as an imago (e.g., Culicidae, Trichoptera), there is no evidence that Chironomidae undergo dormancy as an imago (Tokeshi 1995).

TEMPERATURE Temperature is generally considered to be the primary factor determining developmental rates in chironomids due to its influence on development, respiration, and energy assimilation (Huryn 1990). The patterns of development regulated by temperature are complicated as this factor can have variable impacts on the growth rates, developmental rates, dormancy cues, and emergence cues of different life stages. In addition, the affects of temperature on chironomids are complex, because they can reflect daily or seasonal patterns as well as absolute values or amplitudes and rates of change (Ward & Stanford 1982). These diverse thermal patterns in aquatic habitats are influenced by a variety of factors including latitude, altitude, hydrology, typography, meteorology, and habitat type (Macan 1958, Ward & Stanford 1982, Ward 1985, Wrubleski & Rosenberg 1990).

Chironomidae are particularly interesting study subjects for the effect of temperature on aquatic organisms because of the wide range of thermal preferences observed for the many species within this family. Some chironomids are well adapted to surviving and

even developing or emerging at low temperatures (Oliver 1968, Sæther 1968, Danks 1971b, Mackey 1976a, Welch 1976, Chapter VI). For example, some species can be observed emerging in winter or early spring when water temperature is near 0 °C and ice may be present (Oliver 1968, Welch 1973, Sherk & Rau 1996, Ferrington 2000, Bouchard pers. obs.). In general as latitude increases, the proportion of chironomids in aquatic habitats increases (Oswood 1989), and in artic aquatic habitats the insect community is usually dominated by chironomids (Oliver 1968, Welch 1976). In stark contrast, chironomids also inhabit streams receiving heated waters (e.g., geothermally heated waters, thermal effluents from human activities) at temperatures well exceeding 35 °C (e.g., Winterbourn 1969 - Chironomus cylindricus at 39-41 °C, Hayford et al. 1995 – Paratendipes thermophilus at 39 °C, Brues 1932 – Chironomus tentans at 44.8 °C, Cricotopus sp. at 39 °C) and pools (e.g., Hinton 1951, Hinton 1960a, Hinton 1960b – Polypedilum vanderplanki at 41 °C, Surakarn & Yano 1995 – Chironomus kiiensis at 38- 39 °C).

The effect temperature has on the time required for a chironomid to develop from egg to imago can mainly be attributed to two factors: 1) the biology of a species (and to a lesser degree the individual) and 2) the thermal regime to which the insect is exposed. Chironomids require sufficient heat energy to complete their life cycle and the amount required for each species varies as well as the range of temperatures at which development can occur. As with most insects, an estimate of the time required for larval development at different thermal regimes can be predicted using estimates of a species’ thermal constant (i.e., the amount of heat energy required for development). Such an estimate requires knowledge of a species’ lower threshold or developmental zero (i.e., the temperature at which development is zero), the total number of degree days required for development (i.e., thermal constant), and the thermal regime to which it is exposed. The number of degree days required for larval development is generally consistent in many species independent of the temperature at which they are reared (Sweeney 1984); however, there is variability in the thermal constant between species and in their thermal thresholds. Some species utilize warm habitats (i.e., warm stenothermic), whereas other

species are dependent on cold habitats (i.e., cool stenothermic) and others are more facultative (i.e., eurythermic) (Rossaro 1991). Generally these thermal preferences can be identified by determining the relationship between temperature and developmental rate (e.g., McKie et al. 2004) or the pattern of abundance of taxa from habitats with different thermal regimes (e.g., Rossaro 1991). For example, the relationship between temperature and development for cool and warm stenothermic species is generally parabolic with a narrow thermal range, whereas in eurythermic species this curve is usually broadly parabolic or asymptotic (McKie et al. 2004). However, Rossaro (1991) concluded that many warm adapted taxa are somewhat eurythermic. In many species there is a more gradual increase in developmental rates at lower temperatures and a rapid decrease at high temperatures once some threshold has been reached (e.g., Olsen et al. 2003). It should also be noted that this pattern refers to developmental rates and not growth rates which are most commonly presented in life history studies. For example, growth rates will likely be greatest at a lower temperature and decrease from this peak to a slower rate when compared to developmental rates. However, due to laboratory limitations, many studies do not expose test specimens to temperatures high enough to identify the upper thermal limit at which successful development can occur.

Hauer & Benke (1991) estimated that in general growth rates for chironomids are zero at ≈4-7 °C and maximum growths rates occur at 21-24 °C. However, there are many examples of species capable of growth and development at low temperatures where developmental zeros may be close to 0ºC with maximal developmental rates at <10ºC (e.g., Diamesa incallida – Nolte & Hoffman 1992a, Hydrobaenus kondoi – Kondo 1996, Diamesa mendotae - Chapter VI). Several species have also been shown to continue growing during the winter in deep lakes (e.g., Mozley 1970). In warm-adapted species, developmental zeros may be considerably higher (e.g., 11ºC for Procladius choreus – Mackey 1976b, 9.0 °C for Glyptotendipes paripes – Lobinske et al. 2002) with maximal growth rates also occurring at higher temperatures (e.g., 32.5 °C for Glyptotendipes paripes – Lobinske et al. 2002).

Many studies have shown experimentally in the laboratory and in controlled field experiments that temperature strongly affects developmental rate and generation time in chironomids (e.g., Konstantinov 1958a, Neumann 1961, Dejoux 1971, Nebeker 1971, Menzie 1981, Strixino & Strixino 1982, Reist & Fischer 1987, Rempel & Carter 1987, Kokkinn 1990, Maier et al. 1990, Okazaki & Yano 1990, Yano et al. 1991, Sankarperumal & Pandian 1991, Stevens 1993). Temperature is also implicated in many field studies to determine the length of development, although under natural conditions it is more difficult to separate temperature from other factors (e.g., Borutzky 1939a, Borutzky 1939b, Miller 1941, Mundie 1957, Hall 1960, Palmén 1962, Hilsenhoff 1966, Koskinen 1968a, Danks & Oliver 1972, Welch 1973, Titmus 1979, Butler 1980, LeSage & Harrison 1980a, Rossaro & Cironi 1987, Wrubleski & Rosenberg 1990). Much of the evidence for the effect of temperature under natural conditions is the variation observed in the timing of emergence which correlates to annual temperature differences. Differences in timing of emerging can also often be related to thermal differences between habitats or within habitats. For example, Welch (1973) determined in an arctic pond that chironomids emerged earlier from shallow portions of the pond. However, as many of these studies are based on field data, it is difficult to separate the effect of temperature on development from other factors.

Temperature also has an impact on the feeding rates, digestion, rate of assimilation, and respiration, which can influence growth and development rates (e.g., Janković 1974, Sankarperumal & Pandian 1991, Frouz et al. 2002). In Chironomus plumosus, Hilsenhoff (1966) determined that feeding was halted below temperatures of 5° C. Johannson (1980) determined that maximum feeding rates occurred between 22 and 24° C for Chironomus plumosus. In Corynoneura scutellata, Kesler (1981) determined that maximal feeding rates occurred at 30° C. However, increased ingestion rates resulting from higher temperatures do not directly cause greater growth rates, as higher temperatures also increase respiration. Therefore, the greatest growth rates for some chironomids are associated with intermediate temperatures (Mackey 1977a, Tokeshi 1995, Frouz et al. 2002).

In many chironomids, low or high temperatures are suboptimal for development and some form of dormancy is used to pass these unfavorable periods. For example, in Clunio marinus, pupation and emergence is halted at temperatures around 8-10 °C (Neumann 1966, Koskinen 1968b, Neumann 1975, Neumann 1976). Development halted for Chironomus plumosus f. semireductus when temperatures dropped to 10° C (Johannsson 1980). MacRae & Ring (1993) determined that emergence of Cricotopus myriophylli was prevented when temperature was below 16° C and above 25° C. In many temperate and arctic regions dormancy is most commonly regulated by low winter temperatures (Oliver 1971). Evidence of the overwintering capabilities in chironomids is the fact that the larvae of many species are tolerant to freezing (Andersen 1946, Crisp & Lloyd 1954, Danks 1971b, Paterson 1971, Andrews & Rigler 1985, Danell 1981, Lencioni 2004, Bouchard et al. 2006b). Slightly suboptimal temperatures may be more likely to induce oligopause, whereas very low (including freezing temperatures) or high temperatures would be expected to induce diapause. During oligopause, activity and growth may continue when feeding is possible, but during this dormancy pupation and emergence is halted.

Dormancy induced by temperature is usually associated with one or more behavioral and physiological changes that improve survival during unfavorable periods. For example, many species of Chironomidae cease feeding, evacuate their gut, undergo some dehydration, burrow deeper into the substrate, build cocoons, or modify existing cases to prepare for dormancy and protect from freezing (e.g., Sæther 1962, Hilsenhoff 1966, Sokolova 1966, Armitage 1970, Danks 1971a, Danks 1971b, Madder et al. 1977, Danks & Jones 1978). In some Chironomidae, temperature is identified as a major determinant of when an organism ceases or begins feeding (Moore 1979). For the few species where this has been studied, it appears that temperature regulates feeding cessation or resumption and photoperiod has little effect (Danks 1971a). For example, Hilsenhoff (1966) determined that feeding of Chironomus plumosus did not take place at temperatures of 5° C or less and reduced feeding occurred at temperatures between 6 and

12º C. The pitcher plant midge, Metriocnemus knabi halts feeding at about 4° C and gut clearing occurs before the larval habitat freezes (Paterson 1971, Paterson & Cameron 1982). MacRae & Ring (1993) determined that feeding of Cricotopus myriophylli was halted at low temperatures and overwintering individuals were identified with both ends of their larval case sealed. Armitage (1968) determined that Tanytarsini and Limnochironomus pulsus ceased feeding and built winter cocoons when temperatures dropped in the autumn and early winter. Danks & Jones (1978) reported that cocoon formation was observed at 2º C and maximal proportions of cocoons occurred at 1º C. Some species also build cocoons to prepare for aestivation and provide protection from warm conditions (Kondo 1996). Spaniotoma akamusi larvae burrowed deeply into the mud (20-70 cm) when temperatures rose in April (Yamagishi & Fukuhara 1970, Yamagishi & Fukuhara 1971). Hilsenhoff (1966) observed non-feeding larvae of Chironomus plumosus borrowing up to 51 cm into the mud. Empty guts suggested that individuals recovered from deep in the mud were aestivating and with falling water temperature, these larvae returned to the surface layer of the mud (Hilsenhoff 1966).

In many cases, a change in temperature is also the cue for the breaking of dormancy and a return to activity and feeding. In most cases, an increase in water temperature is the cue for resumption of activity, although for cold adapted taxa the opposite is true. For example, Danks (1971a) determined that in Einfeldia synchrona, most larvae left their winter cocoons to build summer tubes and began feeding and development at temperatures above 4° C. Conversely, Kondo (1996) determined that when temperatures dropped below 15° C, 2nd instar larvae broke aestivation and began to feed again. Similarly Mozley (1970) observed that Trissocladius grandis only resumed activity in late autumn when temperatures decreased and was assumed to aestivate during the summer. In some cases, it is possible that chironomids will undergo two periods of dormancy during the winter and the summer. For example, Danks (1971a) determined that feeding and development was halted during winter and summer in Einfeldia synchrona. In addition, temperature can influence feeding rates in aquatic insects (Ward & Stanford 1982), although this needs more study in chironomids.

In many chironomid species, pupation and/or emergence are usually related to increasing temperatures (e.g., Rempel 1936, Jónasson 1961, Jónasson 1965, Palmén & Aho 1966, Danks & Oliver 1972, Coffman 1973, Boerger 1981). The effect of temperature on the initiation of pupation may be especially important for spring emerging individuals, which are dormant or exhibit slow growth during cold periods. Danks & Oliver (1972) estimated that pupation for arctic chironomids from shallow ponds have a threshold of ~4-5 °C and pupal ecdysis is at ~7 °C. Laville (1971a) determined that emergence began when temperatures reached 7 °C. LeSage & Harrison (1980a) determined that the emergence of Cricotopus was limited to temperatures between 5-27° C with only very low emergence at 5-10° C and 25-27° C. Morgan & Waddell (1961) determined that temperature thresholds for shallow water species were 7.5° C and 6.0° C for deep water species. Other examples include estimated pupation thresholds of 8-10° C for Chironomus tentans (Sadler 1935), pupation and emergence at 7° C for Einfeldia synchrona (Danks 1971a), and emergence at 8-10° C for some Scottish species (Morgan 1958). The initiation of pupation can also be a response to decreasing temperatures as in Tokunagayusurika (=Propsilocerus) akamusi (Iwakuma & Yasuno 1983, Iwakuma et al. 1984, Iwakuma 1992).

Timing of emergence may be controlled by different thresholds or cues acting on the developmental timing of pupation, development of the pupa, and emergence of the adult (Thienemann 1954, Danks & Oliver 1972). For example, Danks & Oliver (1972) estimated thresholds for pupation for some arctic chironomids to be about 4-5° C and 7° C for adult emergence. Jónasson (1972) determined that the spring emergence period of Chironomus anthracinus is short and occurs between 6.2-7.2º C when temperatures are increasing, but the lake has yet to stratify. Further experimentation determined that 4th instar larvae can pupate at temperatures as low as 2.5º C, but apparently higher temperatures are needed for emergence. In addition, the cue for emergence and halting emergence may be different. Morgan & Waddell (1961) determined that temperatures of 6-7.5° C in spring initiated emergence, but emergence declined despite higher

temperatures in late summer. However, it was not possible to separate other effects such as decreasing photoperiod or changing food availability.

Ice cover can mechanically restrict emergence of chironomids. In many temperate, arctic, and alpine habitats, emergence begins shortly after thaws (e.g., Miller 1941, Oliver 1968, Laville 1971a, Wiederholm et al. 1977, Welch et al. 1988, Sherk & Rau 1992). In the arctic, emergence from small habitats follows a week or more after the ice has cleared, whereas in larger lentic or lotic habitats chironomids usually emerge shortly after open water appears (Oliver 1968, Danks & Oliver 1972, Welch 1973, Hayes & Murray 1987). Some chironomid species emerge from lakes before ice has completely melted (Oliver 1968, Welch 1973, Sherk & Rau 1996) and some species are able to utilize cracks in ice to emerge (Oliver 1968, Lindegaard & Maehl 1992). In fact, the emergence of chironomids under ice results in mortality of a large percentage of these individuals (Welch 1973). In addition, Jónasson (1972) determined that growth of Chironomus anthracinus was halted under ice cover, whereas it continued during winter when the lake was ice free. Sæther (1968) suggested that slower rates of development for Diamesa valkanovi and Diamesa davisi at sites close to glaciers and snow banks were the result of shorter snow- and ice-free periods. The Antarctic species, Belgica antarctica emerges shortly after snow melts from the larval habitat (Sugg et al. 1983).

NUTRITION Compared to temperature, fewer studies are available that directly assess the influence of nutrition, either quality and quantity, on life cycle timing in Chironomidae (Storey 1987). In many cases, the absolute impact of nutrition may be obscured due to study methods or that it is correlated with other factors (e.g., temperature). However, the quality, quantity, and type of food has been determined to affect developmental rates and impact life histories in chironomid larvae (e.g., Darby 1962, Karunakaran 1971, Sokolova 1971, Mackey 1977a, Ward & Cummins 1979, Drake 1982, Storey 1987, Mattingly et al. 1981, Pinder 1992, Gresens 1997). However, most studies only assess the effect of nutrition on

growth rates and not developmental rates, and therefore these studies are not incorporated into this review.

Nutrition is a potentially limiting resource for aquatic insects that can affect the timing of development and ultimately emergence. Both the quality and quantity of food resources can have an effect on the development of chironomids. Biever (1971) determined that both food amount and type influenced the developmental time for a species of Chironomus by increasing the developmental time at both low and high amounts of food. Stanko-Mishic et al. (1999) determined that both food type and particle size influenced the development time of Chironomus riparius. Frouz et al. (2004) determined that the developmental rates of Glyptotendipes paripes were influenced by the species of alga consumed. Johannsson (1980) determined that the growth of Chironomus plumosus f. semireductus was positively influenced by increased food quality as determined by its assimilation efficiency. In fact, growth is greatly reduced in autumn when food quality decreases despite temperatures remaining relatively high. The developmental rates of Pseudochironomus richardsoni were greater on a diet of diatoms compared to detritus (Gresens 1997). Postma et al. (1994) determined that food limitation increased larval development time in Chironomus riparius. Although much of the research on the impact of food on the life histories of chironomids has been focused on collector/gatherers or herbivores, an influence on food quantity and quality has also been identified for predatory chironomids. For example, Kajak & Dusoge (1970) determined that the developmental rate of Procladius choreus increased when the quantity of easily captured prey was increased. The developmental rates of Ablabesmyia monilis are dependent on the type of food consumed (Mackey 1977a).

The life histories of some chironomids may be timed to correspond to periods when food resources are available or most abundant (e.g., autumn leaf fall, spring macrophyte growth) (Pinder 1977, Johannsson 1980, Rasmussen 1984, Tokeshi 1986a, Pinder 1992). Increased temperatures can also increase decomposition rates and the amount of microbial growth on substrates, thereby increasing the amounts of fine particulate organic

matter (FPOM) (e.g., Hildrew et al. 1984, Rempel & Carter 1986), or by increasing growth of biofilms, algae, and macrophytes (e.g., Wright 1978, Lamberti & Resh 1983, Swanson & Hammer 1983, Lamberti & Resh 1985, Johnson & Pejler 1987). The initiation and suspension of growth in Chironomus anthracinus was determined by Jónasson & Kristiansen (1967) to be partially influenced by primary production. Food sources may also be influenced by discharge as low flows may increase accumulation of diatoms and particulate matter (Wright 1978). Kaufman & King (1987) determined that the phenology of Xylotopus par was dependent on the age of wood colonized by larvae of this species. The presence of appropriate food sources can also determine when a habitat will be colonized by chironomids (Pinder 1992). Still other species may be constrained by the phenologies of host plants or animals such as Myriophyllum, Nuphar, Potamogeton, Ephemeroptera, Odonata, Plecoptera, Megaloptera, freshwater sponges, or bryozoans (e.g., Berg 1949, Kangasniemi & Oliver 1983, van der Velde & Hiddink 1987, de la Rosa 1992, Tokeshi 1993, Epler & de la Rosa 1995, Jacobsen 1998).

PHOTOPERIOD Although the timing of development for chironomids is considered to be largely determined by temperature and nutrition, photoperiod may also be important in regulating the timing of life stages. Photoperiod is more important in the regulation of dormancy and less important in mediating growth and development rates. However, Nolte & Hoffman (1992b) provides an example of photoperiod influencing the timing of development in larvae. In this example, the duration from egg to pupa in Pseudodiamesa branickii was influenced by photoperiod with shorter generation times occurring under long-day photoperiods.

Most studies of the impact of photoperiod on the timing of chironomid life cycles are associated with the initiation or ending of dormancy. Presumably, the control of pupation by photoperiod prevents emergence of adults and hatching of eggs at low temperatures, which may not provide sufficient time to advance to a 2nd or greater instar capable of overwintering (Danks 1971a). In Metriocnemus knabi (Paris & Jenner 1959) and

Chironomus tentans (Englemann & Shappirio 1965) dormancy is apparently controlled by photoperiod, with short photophases inducing and maintaining diapause and long photophases terminating diapause. Matěna (1989) assumed that dormancy of Chironomus plumosus in the autumn was the result of a shorter photoperiod. Koskinen (1968a) hypothesized that photoperiod limited Chironomus salinarius to one generation per year as a result of dormancy control. Danks (1971a, 1978) determined that an 8 hr photophase inhibited pupation for most individuals from several species of Chironomini, while a 16 h photophase permitted pupation. Rychen Bangerter & Fischer (1989) determined that the photoperiod threshold for inducing dormancy was a 12 h photophase in Chironomus nuditarsis and a 13 h photophase for Chironomus plumosus. There are also a number of examples where species emerge at the same time during different years and at different depths despite different thermal regimes. In these cases photoperiod is often implicated as synchronizing emergence of these species (e.g., Brundin 1949, Aagaard 1978b). For example, Stahl (1986) determined that although thermal regimes in a reservoir receiving cooling water from a power plant were altered annually, generations maintained synchrony. Because photoperiod was the only variable that remained constant year after year, Stahl (1986) hypothesized that photoperiod was responsible for maintaining synchrony within cohorts.

In some cases, photoperiod can cause a developmental halt although activity (e.g., feeding) in the larva can continue (i.e., oligopause). Pupation in natural populations of Chironomus riparius is halted in October-November despite higher temperatures than those observed in early spring when pupation and emergence occurs, suggesting that photoperiod controls diapause (Goddeeris et al. 2001). This was further supported by laboratory experiments that determined an 8 h photoperiod induced dormancy in 4th and 3rd instar larvae in Chironomus riparius, whereas a 16 h photoperiod did not. Although development was halted under these conditions, feeding apparently continued and allowed these individuals to continue gaining biomass (Goddeeris et al. 2001). Oligopause is also observed in open water populations of Clunio marinus (later described as Clunio balticus by Heimbach [1978]), where photoperiods below LD 12:12 result in

extended growth periods (Neumann 1976). Fischer (1974) and Ineichen et al. (1979) also observed that feeding and growth continued under short photoperiods when development was halted in the 4th instar of Chironomus species. This oligopause could then be quickly terminated by the induction of long day conditions (Ineichen et al. 1979).

LUNAR AND TIDAL PERIODS The influence of lunar period on life history timing of chironomids is largely derived from studies on marine chironomids. The timing of emergence of many of these species is linked to tidal period, which is related to lunar period. In most populations of Clunio marinus, pupation is timed to the lunar period and is controlled by an endogenous semilunar rhythm (Neumann 1966, Neumann 1971, Neumann 1975a, Neumann 1976). This also appears to be the case for Clunio tsushimensis (Saigusa & Akiyama 1995). The pupal stage only lasts 3-5 d for Clunio marinus and the eggs of intertidal populations are deposited directly on the substrates, which is only exposed during the low-water levels of spring tides. As a result, synchronization of pupation and emergence in this species is important (Neumann 1966, Neumann 1971, Oliver 1971). Based on field observations, Soong et al. (1999) suggested that similar emergence cues may also occur in Pontomyia oceana, because emergence is timed to low tides when the larval habitat in the littoral zone is exposed. This is further supported by the fact that another co-occurring species, Pontomyia natans, whose larvae utilize the sublittoral zone, lacks a strict timing of emergence to tidal periods. Depending on the population of Clunio and the reliable cues available to these populations, the endogenous rhythm in Clunio marinus is synchronized by the periodicity of moonlight or by tidal action (Neumann 1971, Neumann 1975a). Typically, pupation in populations from southern and middle latitude coasts can be synchronized using a moonlight pattern, whereas northern latitude populations do not respond to moonlight, but can be synchronized by diel turbulence patterns which mimic tidal action (Neumann 1975, Neumann 1976). The exception to lunar period as a cue in Clunio marinus occurs in arctic populations that do not emerge on a semilunar period, but emerge during periods when temperatures rise sufficiently for overwintering larvae to complete their development (Neumann 1975). In Clunio takahashii, emergence

corresponds to tidal period and occurs twice daily regardless of the time of day (Hashimoto 1965). Furthermore, there does not appear to be any link to lunar period and emergence, and therefore emergence likely corresponds to the exposure of the habitat (Hashimoto 1965).

There are also studies that indicate some impact of lunar period on freshwater chironomid species, although the reasoning behind why these species undergo these emergence patterns is not clear. Some chironomid species in Lake Victoria have emergence periods that are linked to lunar periods, with maximum emergence occurring near new and full moons (MacDonald 1956, Corbet 1958, Fryer 1959, Corbet 1964). In some cases, this apparent pattern has been explained by the possible effect of a full moon decreasing the relative brightness of a light trap (Corbet 1958). However, Corbet (1958) determined that peaks occurred shortly after the new moon and a lunar emergence pattern was only observed for some species. The distinct lunar emergence patterns observed by MacDonald (1956) were also manifest in the larval stage where there appeared to be two overlapping but distinct generations at any given time. Tjønneland (1962) identified lunar periodicity in the midge Conochironomus acutistilus corresponding to maximal emergence occurring at both the first quarter and third quarter of the moon, with the lowest emergence occurring during new and full moons. The midge Chironomus brevibucca has mass emergence a few days after the full moon in Lake Bangweulu, Zambia (Fryer 1959).

Although the causes and reasons for lunar synchrony in freshwater midges are not always clear, it is possible that lunar periodicity may serve as a means to synchronize emergence (Corbet 1964). This may be true as the few examples of lunar synchrony in freshwater chironomids occurs in tropical lakes where lunar period may be the best cue in an environment and other factors such as temperature and photoperiod are more stable than in temperate habitats.

INTERMITTENCY AND DISCHARGE Many Chironomidae species are tolerant to desiccation and can survive in dry or moist habitats (e.g., Hinton 1960a, Edward 1968, Miller 1970, Jones 1975, Grodhaus 1980, McLachlan & Cantrell 1980, Adams 1983, Okazaki & Yano 1990, Cranston & Nolte 1996, Williams 1998, Chou et al. 1999, Suemoto et al. 2005). Dry or moist conditions will halt or slow development of larvae and cause potential life history changes that alter the timing of emergence. For example, some chironomids that inhabit vernal habitats undergo diapause during dry periods (e.g., Grodhaus 1980, Williams 1998), which therefore lengthens the life history. Similar to hibernating species, chironomids diapausing in dry habitats often construct protective cases for protection from desiccation (e.g., Grodhaus 1980).

The availability of wetted habitats will also determine the timing of chironomid life cycles (McLachlan 1983, Wotton et al. 1992, Stevens et al. 2006). Therefore, rainfall may be an important factor in determining emergence patterns (e.g., Sonoda & Trivinho- Strixino 2000). For example, Karunakaran (1971) determined from light trapping in Singapore that emergence peaks were correlated to rainfall patterns. Similarly, Clement et al. (1977) determined that predictable emergence patterns were a result of regular flooding patterns in rice fields, which made the habitats available for colonization. Rosenberg et al. (1988) hypothesized that most of the chironomid emergence from poor fens occurred in spring because sufficient growth and development must take place before the habitat dries in summer and before the arrival of winter. As a result the most abundant taxa tend to be univoltine with synchronous emergence. Human induced changes in habitats, such as reservoirs that are drawn down, may also influence the life history of chironomids (e.g., Sephton & Paterson 1986, Suemoto et al. 2005). Rainfall is also important for chironomid species that dwell in rock pools, as the drying and rewetting of these habitats determine when development can take place (Nolte 1995). As a result, many of the species that inhabit ephemeral habitats have very short life spans (e.g., MacLachlan & Cantrell 1980, Gray 1981, McLachlan 1983, Ladle et al. 1985, Wotton et al. 1992, Nolte 1995).

Even in permanent streams and rivers, discharge may have an impact on the life histories of chironomids. For example, spates can result in aquatic insects seeking refuge in the hyporheic zone during which time these organisms would be expected to decrease activities such as feeding and subsequently cause slower development. Spates can also result in changes in the quantity of food through mechanical disruption of substrates that could influence feeding in chironomids. Reduced emergence has also been observed during periods of high discharge (e.g., LeSage & Harrison 1980a, Rossaro & Cironi 1987) that may reflect a delay in emergence due to unsuitable conditions. However, the effect of spates would only be expected to delay emergence for a short period of time and patterns at greater temporal scales would be minimally affected.

SUBSTRATE The availability of appropriate substrates can potentially influence developmental times for chironomids by affecting other attributes such as feeding rates and feeding efficiency. For example, Ristola et al. (1999) hypothesized that sand substrates could increase feeding efficiency as larvae could selectively feed on food particles, whereas larvae grown on a finer substrate were forced to consume both food and inorganic materials. Beattie (1978) determined that Pentapedilum uncinatum was bivoltine in mud substrates and univoltine in sand substrates. However, Sibley et al. (1998) determined there was little affect of particle size on the life cycle timing of Chironomus tentans, although emergence success was reduced and timing increased for individuals grown in a clay substrate. Some species utilize macrophytes as substrates or food sources (e.g., Mackey 1977b, Menzie 1981, Storey 1987, Nolte 1991), and therefore the seasonal fluctuations in the presence of macrophytes may in turn be connected to the life history of epiphytic chironomids. The life history of Metriocnemus knabi is linked to the phenology of the pitcher plant that it inhabits as the plant grows and senesces (Paterson & Cameron 1982).

DISSOLVED OXYGEN Dissolved oxygen (DO) in aquatic habitats can vary as a result of natural processes (e.g., lake stratification) and human impacts (e.g., nutrient enrichment). Although many species are able to survive and even remain active at low levels of DO (Lindeman 1942, Nagell 1978, Heinis & Crommentuijn 1989), periods of low oxygen can affect the life history of chironomids. Generally, low levels of oxygen would be expected to slow developmental rates because behaviors or physiological adjustments can require energy and low DO levels can induce dormancy or reduce feeding rates (e.g., Brundin 1951, Nagell 1978, Heinis & Crommentuijn 1989, Heinis et al. 1989). Jónasson (1965), Jónasson & Kristiansen (1967), and Jónasson (1972) determined that the growth of Chironomus anthracinus was limited in late summer during a period of low dissolved oxygen content despite high levels of primary production and was partially responsible for increasing the length of generations. In addition, Jónasson & Kristiansen (1967) determined that growth resumed 3-4 weeks after autumn overturn, despite a decrease in primary production. Butler & Anderson (1990) suggested that the long life cycle for a merovoltine Chironomus was the result of anoxic conditions in the bog inhabited by the larvae that halted activity and feeding in the larvae. Ikeshoji (1974) determined that oxygen deficiencies increased the time from hatching to emergence. Phillipp (1938) determined that developmental time of Chironomus thummi increased in aquaria without aeration compared to aerated aquaria. Learner & Potter (1974) suggested that low levels of DO were possibly responsible for a shift for several chironomid species from two to one generation per year. Yamagishi & Fukuhara (1971) hypothesized that three generations of Chironomus plumosus were present in the littoral zone, whereas only two generations were identified from the profundal as a result of low DO levels during the summer. Miller (1941) hypothesized that emergence of Tanytarsus (=Micropsectra) during the summer was the result of changing DO and carbon dioxide levels as temperatures remained relatively constant from spring to autumn. Prat & Rieradevall (1995) provided evidence to support a reduction in the number of generations at sites with low dissolved oxygen.

Although less well supported, low levels of DO may also have an impact on other life stages besides the larva. Rempel (1936) hypothesized that oxygen levels in lake dwelling chironomids may be responsible for the initiation of pupation; however there is no experimental evidence of this relationship. Pinder (1992) attributed differences in hatching from 6-58 days to gradients in oxygen level in composite egg masses of Thienemanniella vittata.

SALINITY There is only limited information available on the effect of salinity on chironomid life cycles, particularly information on how this variable impacts developmental rates and life history timing. Neumann (1976) noted that in general, euryhaline species require longer periods for development than freshwater species. Kokkinn (1990) determined that the developmental rate of Tanytarsus barbitarsis was affected by salinity, where increasing salinity increased the generation time

POLLUTION In addition to natural factors that influence the timing of chironomids, anthropogenic influences can also have an effect on the developmental rates and phenology of chironomids. Most of the research on the effects of pollutants on the life history of chironomids comes from toxicity studies most commonly on Chironomus riparius and Chironomus tentans. However, there is a lack of in situ or field studies that assess how pollution influences the phenology of chironomids. Measurement of end points such as the life cycle duration are commonly used in toxicity studies, in many cases providing a more sensitive test than acute toxicity tests (Kosalwat & Knight 1987). Although nutrient enrichment is an important anthropogenic impact on chironomids, the effect of this type of pollution on developmental rates has not been assessed. Experimental studies where nutrient and DO levels are varied show that these variables can affect the developmental rates of chironomids (see DISSOLVED OXYGEN), and therefore it is likely that nutrient pollution can accelerate or slow development.

Heavy Metal Pollution In most cases, heavy metal pollution simply results in death of chironomid larvae (e.g., Hatakeyama 1988) and does not impact life cycle dynamics by lengthening or shortening the life cycle. However, as with other stressors, sublethal levels have been tested and when developmental endpoints are measured, they often identify a lengthening of chironomid life histories. This differs from studies of terrestrial insects which have indicated that heavy metal exposure decreases developmental time (Postma et al. 1995). The increased development time is potentially linked to the physiological costs of heavy metal tolerance such as the synthesis of metallothionein proteins. Studies on the sublethal levels of heavy metal pollution (e.g., zinc, lead, copper, cadmium) have provided evidence of decreased development rates resulting in delayed emergence and an increase in the duration of the emergence period in Chironomus riparius and Chironomus tentans (Wentsel et al. 1978, Pascoe et al. 1989, Timmermans et al. 1992, Postma & Davids 1995, Watts & Pascoe 1996, Watts & Pascoe 2000, Servia et al. 2006). The time to emergence in Paratanytarsus parthenogeneticus was influenced by copper concentrations in the water and it was hypothesized that changes in developmental rates were related to a decrease in food uptake. Kosalwat & Knight (1987) determined that exposure to copper in sediments resulted in delayed emergence of Chironomus decorus. Postma et al. (1994) determined that cadmium exposure increased larval development time. However, Hatakeyama (1987) determined that, on average, Polypedilum nubifer exposed to cadmium emerged 2-4 d faster than the control individuals.

Most studies that have identified an impact of heavy metals on the development and life history of chironomids have been performed in the laboratory. One field study (Groenendijk et al. 1998) of the effect of heavy metals on Chironomus riparius at impacted and non-impacted sites determined that there was no difference in the voltinism between polluted and unpolluted sites, possibly because drift from non-impacted sites maintained populations in impacted sites.

Pesticides The insecticide lindane has been shown to delay emergence in Chironomus tentans (Macek et al. 1976) and Chironomus riparius (Taylor et al. 1993, Hirthe et al. 2001). The herbicide atrazine resulted in several herbivorous chironomid taxa emerging earlier than those emerging from untreated ponds (Dewey 1986). However, it was not clear what exact mechanism resulted in this pattern, whether the loss of food caused a truncation of the population or if development was accelerated.

Thermal effluents Modified temperatures in aquatic habitats caused by geothermal input, power station cooling waters, and reservoir deep releases can result in altered emergence patterns in aquatic insects (Raddum 1985, Stahl 1986, Hayford et al. 1995). The impact of increased temperatures can be severe enough to result in emergence of chironomids into air temperatures that could be fatal (Nebeker 1971, Raddum 1985, Hayford et al. 1995). Thermal effluents can alter the voltinism of chironomid species, generally increasing the number of generations that can be completed due to an increase in degree days. For example, Parkin & Stahl (1981) determined that Tanypus stellatus increased the number of generations completed per year from 2 to 3-4 at stations receiving thermal effluent from a power plant. In fact, four generations were estimated from a site receiving nearly 1000 additional degree days than the site where only three generations were estimated. In contrast to thermal releases, bottom release dams can lower water temperatures by discharging water from the bottom of a reservoir. For example, Mason & Lehmkuhl (1983) determined that the release of water from a dam resulted in a delay of chironomid emergence of about 2 weeks.

BIOTIC INTERACTIONS Intraspecific interactions Intraspecific interactions can influence the growth and development of chironomids through competition for resources such as food or substrate (Kajak & Warda. 1968, Hooper et al. 2003a). Reist & Fischer (1987) determined that increased larval density

slowed development in Chironomus plumosus. Biever (1971) determined that greater intraspecific density increased the developmental time for Chironomus sp., Goeldichironomus holoprasinus, and Tanypus grodhausi. High densities increase generation development time in Chironomus riparius (Stanko-Mishic et al. 1999, Silver et al. 2000, Hooper et al. 2003b), Glyptotendipes tokunagai (Yano et al. 1991), and Tanytarsus oyamai (Okazaki & Yano 1990). Rasmussen (1985) determined that increased intraspecific densities decreased developmental rates for both Chironomus riparius and Glyptotendipes paripes and could potentially alter voltinism.

High larval densities can also result in production of large amounts of waste products that can impact the developmental rates of chironomids. Ikeshoji (1974) determined that increased densities resulted in greater amounts of waste products that decreased DO levels and increased the time from hatching to emergence.

Competition within a species has also been associated with cannibalism in chironomids where late instar larvae of one cohort consume the eggs of later generations. Jónasson (1965) and Jónasson (1972) observed a lack of recruitment for Chironomus anthracinus every other year. These years that lacked recruitment corresponded with years when a high density of larvae were present and it was hypothesized that larva were feeding on eggs from a different generation, resulting in the termination of the new generation. As a result, during years when high densities were present, this species was assumed to have a two year life cycle, whereas only one year was required when densities were low. Similarly, Matěna (1989) observed that eggs laid early in the spring apparently did not survive as a result of high competition and predation from other chironomids.

Interspecific interactions Competition. Presumably, the competition of different species that utilize similar resources can result in increased developmental time. However, in many cases there may be some form of resource partitioning or dynamic processes in aquatic habitats that reduce this competition. For example, Rasmussen (1985) determined that increased

interspecific densities did not have a strong effect on growth rates suggesting, that resource partitioning allows coexistence of two species. In addition, Cannings & Scudder (1978) determined that species of Chironomus emerged at different times from different lakes depending on the coexistence of other species in the lake.

Predators. The impact of predation on chironomid developmental rates has not been widely studied. In particular, there is little research on the impact of invertebrate predators on chironomid life histories. Ball & Baker (1995, 1996) determined that the presence of fish predators increased developmental time of Chironomus tentans. Macchiusi & Baker (1992) determined that the presence of a fish predator decreased the proportion of 3rd instar larvae molting into 4th instars over a 7-day period. Impacts on the growth and development may be a response to the increased energy required to build protective tubes or limitations in feeding activity.

Parasites & diseases. Infestations of parasites and disease have been identified in Chironomidae which commonly lead the death, sterilization, or a reduction in the size of individuals (e.g., Hilsenhoff 1966, LeSage & Harrison 1980b, Titmus & Badcock 1981). Presumably, a parasite sequesters energy that is then unavailable to the host organism, which could prolong the development of chironomids. However, I have found no research with empirical evidence that these organisms can alter the timing of emergence in chironomids.

SEX Several studies on chironomids have identified differences in the developmental rates of males and females within the same species (e.g., Stevens 1998). Differences in the timing of emergence are commonly observed with males on average emerging earlier than females (i.e., protandry), apparently the result of the larger females requiring greater time to acquire sufficient biomass. The occurrence of protandry in chironomids is generally explained by the greater amount of time and degree days required for the larger females to develop eggs (Stevens 1998, Frouz et al. 2002). This is supported by the

results of Palmén (1962) who determined that the ratio of males to females decreased through the season.

INTERACTIONS OF FACTORS In most chironomids, the timing of emergence is most likely a result of many interacting factors. In many cases, there may be interaction between temperature and photoperiod. For example, Danks (1978) determined that an 8 hr photophase inhibited pupation of Chironomus decorus larvae at 15 °C, but not at 20 °C. In addition, individuals of Endochironomus nigricans that failed to emerge under an 8 h photoperiod at 15 °C emerged at 20 °C. Similarly, Fischer (1974) determined that diapause in Chironomus nuditarsis was initiated by a 6 h photophase at 10 and 15 °C, but not at 20 °C. However, this interaction does not appear to be universal in chironomids as Fischer (1974) was unable to identify a temperature-photoperiod interaction in Chironomus plumosus. Neumann & Krüger (1985) determined that Clunio marinus only entered oligopause under both low temperatures and short day conditions. The interactions between temperature and photoperiod may be further complicated by changes in the response of individuals to these cues. Ineichen et al. (1979) hypothesized that the sensitivity of dormant larvae to photoperiod diminishes over time, but in species experiencing low temperatures, the dormancy is then maintained by low temperatures. This may be supported by the fact that emergence of most Chironomus nuditarsis individuals will eventually take place even under short-day photoperiods (Rychen Bangerter & Fischer 1989). Some species undergo two periods of dormancy (i.e., hibernation and aestivation) that may be mediated by different factors or combinations of factors such as temperature and photoperiod (e.g., Tanytarsus sylvaticus – Goddeeris 1990). Photoperiod and temperature may also function differently across a community. For example, Aagaard (1978b) hypothesized that in a northern temperate lake, taxa emerging in the spring were primarily regulated by the spring temperature rise, whereas summer emerging species are affected more by photoperiod.

Perhaps the most important and sometimes the most difficult interaction to separate is that of temperature and nutrition. For some species, maximal growth occurs during periods when temperature and food are greatest (e.g., Jónasson & Kristiansen 1967, Johannsson 1980). The combination of temperature and food has been studied or implicated in several studies of chironomid life histories (e.g., Ward & Cummins 1979, Boothroyd 1987, Storey 1987, Gresens 1997, Boothroyd 1999). Developmental patterns can be very complex. For example, at low temperatures growth and development may be largely limited (low metabolic rates), but at high temperatures, food quality may become more important (Gresens 1997). Ward & Cummins (1979) determined that growth rates in Paratendipes albimanus were determined by an interaction of food quality and temperature. Johannsson (1980) discussed the interaction of multiple factors where temperature regulates gut filling and ingestion rates, whereas energy assimilation is more dependent on food quality. The impact of temperature and nutrition is further complicated by the many pathways through which they influence chironomids. For example, temperature can affect the feeding rates and energy assimilation in chironomids while also affecting the growth rates of food resources (e.g., algae, biofilm). There may also be interactions with food, temperature, oxygen, and photoperiod (Jónasson 1972, Danks 1978, Storey 1987, Gresens 1997) and in many cases these interactions may be more than additive (Gresens 1997). Often these factors may be difficult to separate without performing controlled laboratory experiments (Sweeney 1984, Pinder 1992).

Postma et al. (1994) determined that independently cadmium exposure and food limitation increased larval development time, but when Chironomus riparius larvae were simultaneously exposed to both cadmium and food limitation, there was no impact of increasing cadmium concentrations. Postma et al. (1994) hypothesized that this pattern was the result of high 1st and 2nd instar mortality, which provided the surviving larvae with greater amounts of food. Hooper et al. (2003b) and Hooper et al. (2005) determined that there was an interaction between larval density of Chironomus riparius and the effect of 14C-cypermethrin (synthetic pyrethroid) and lufenuron (synthetic insect growth

regulator), respectively, where higher densities moderated the effects of the pesticide on the time to emergence.

EMERGENCE PATTERNS AND STRATEGIES

Corbet (1964) identified four emergence patterns in aquatic insects: 1) continuous, 2) rhythmic, 3) sporadic, and 4) seasonal, all four of which have been identified in chironomids. Continuous emergence is steady emergence throughout the year. This type of emergence is most common in thermally stable habitats where physical factors vary little and allow chironomids to emergence throughout the year (Corbet 1964; e.g., Lehmann 1979, Boothroyd 1988, Sonoda & Trivinho-Strixino 2000). In general, this includes tropical and ground-water influenced habitats (Hynes 1970); although, exceptions to this generalization include Southern Hemisphere streams, where it has been hypothesized that many chironomids from this region possess non-seasonal emergence or poorly synchronized life histories patterns as a result of climatic variation or reduced seasonality of allochthonous inputs (Towns 1981).

Rhythmic emergence is emergence timed to some regular reoccurring cue such as photoperiod or lunar period. The diel emergence patterns of many chironomids are linked to photoperiod. There are fewer examples of rhythmic emergence at greater temporal scales and these examples are generally linked to lunar period influence. The marine midge Clunio is a good example of a rhythmic emergence pattern. Sporadic emergence is less understood than other emergence patterns because it is likely controlled by a number of factors that are difficult to assess in the field. Some cases of apparently sporadic emergence may be the result of ineffective sampling methods such as widely spaced sampling intervals. However, there may be cases where the emergence of a chironomid species is sensitive to more or less arrhythmic environmental factors such as precipitation or discharge (e.g., Schmid 1992).

Seasonal emergence patterns are the most commonly identified and discussed type of emergence in the Chironomidae. Seasonal emergence is largely influenced by two factors depending on the region or habitat: 1) temperature in temperate and arctic habitats and 2) rainfall in tropical habitats and intermittent habitats (Corbet 1964). Chironomids from arctic and alpine aquatic habitats tend to be regulated by seasonal patterns in temperature through the control of developmental and dormancy periods (Hynes 1970, Danks & Oliver 1972, Butler 1982, Butler 1984, Mihuc & Toetz 1996). There are fewer studies on tropical and subtropical chironomid communities (e.g., Ferrington et al. 1993, Sonoda & Trivinho-Strixino 2000), but these communities may be regulated by other factors such as environmental conditions (e.g., dissolved oxygen, food quality, and biotic interactions) (Cowell & Vodopich 1981). In temperate habitats, emergence of chironomids is largely limited to warmer months (Corbet 1964), with the exception of species that occupy habitats that do not become completely ice covered in winter (e.g., ground water influenced habitats). The development of species in intermittent habitats are constrained to wetted periods and warm periods in temperate and arctic regions. However, species-level studies of intermittent habitats are not common (e.g., Grodhaus 1980, Chou et al. 1999).

VOLTINISM A large range of voltinisms can be identified in chironomid species ranging from multivoltine to merovoltine. The fewest known number of generations completed per year is 1/7 for Chironomus species in arctic ponds (Butler 1982). There are also several examples of chironomids that require 3 years to complete a generation (e.g., Sendstad et al. 1976, Welch 1976, Hershey 1985, Butler & McMillan 1990, Lindegaard & Mæhl 1992). The greatest number of generations completed within a single year is less clear due to the difficulty of separating cohorts in multivoltine populations. Completion of five generations in a year is not infrequent (e.g., LeSage & Harrison 1980a, Drake 1982, Pinder 1983, Tokeshi 1986a, Drake & Arias 1995, Goddeeris et al. 2001) and some of these studies suggest that greater numbers of generations are possible. MacDonald (1956) estimated that three chironomid species in Lake Victoria had life cycle durations

of 2 months and, due to the continuous growing season, could complete 6 continuous generations per year. However, because overlapping generations were present, a total of 12 generations per year could be completed. Seven generations were also estimated for Tanytarsus barbitarsis in an Australian saline lake (Paterson & Walker 1974); however, due to a long sampling interval this estimate needs to be confirmed (Tokeshi 1995). In general, voltinism would be expected to decrease for taxa from low to high latitudes (Borutsky 1963, Tokeshi 1995). Therefore, future research will likely determine that chironomids with more than five generations are more common than appear in the literature, due to the difficultly of separating generations in multivoltine populations and the lack of studies in warmer regions where many generations would be expected to be more common (Tokeshi 1995).

Based on available literature, most chironomids undergo one to five generations per year, with univoltine and bivoltine life histories most common (Learner & Potter 1974, Potter & Learner 1974, Boerger 1981, Pinder 1983, Singh & Harrison 1984, Tokeshi 1986a, Lenat 1987, Berg & Hellenthal 1991). Tokeshi (1995) determined from 125 taxa that 33% were univoltine, 44% bivoltine, and 18% multivoltine. This is similar to a review of literature by Berg & Hellenthal (1991) that presented the proportions of 30-50% univoltine and 50-70% bivoltine and multivoltine. Proportions of voltinism in a community are apparently partly determined by the habitat. Tokeshi (1995) provides examples of lotic and lentic communities from the same latitude where, on average, lotic communities have a higher voltinism than those from lentic habitats. In contrast to this pattern, Ringe (1974) determined that most chironomids from two streams in Germany had only 1-2 generations. Large, warm rivers may also contain a prevalence of multivoltine species compared to small, cool streams which contain more univoltine species (Berg & Hellenthal 1992). Although most studies in temperate streams identify a considerable number of univoltine taxa, Berg & Hellenthal (1991) identified only one univoltine chironomid species from a lotic habitat. However, the authors suggested the lack of univoltine species may be a result of the buffered thermal regime in this stream.

As a result of seasonal patterns in temperate and arctic regions, univoltine species undergo periods of warm and cool water (Berg & Hellenthal 1992). As a result, most univoltine species have a synchronized emergence in the spring or early summer (Harper & Cloutier 1979, Tokeshi 1995). Much less common are univoltine species with an autumn emergence (Tokeshi 1995). In temperate and arctic habitats, growth and development of most chironomids are limited to the spring, summer, and fall, commonly resulting in multi-year generations (e.g., Sendstad et al. 1976, Welch 1976, Butler 1982a). Semivoltine (i.e., 2 year generation) or merovoltine (i.e., >2 year generation) chironomid species are also common, especially in high latitude habitats (e.g., Rempel 1936, Sendstad et al. 1976, Welch 1976, Butler 1982, Hershey 1985, Butler & McMillan 1990). In general, multivoltine species possess a cold growing generation and one to several warm growing generations (Harper & Cloutier 1979, Shiozawa & Barnes 1977, Berg & Hellenthal 1992, Prat & Rieradevall 1992). Due to year round warm temperatures that permit continuous growth and development, asynchronous, multivoltine species appear to be more common in tropical streams (e.g., Lehmann 1979, Sonoda & Trivinho-Strixino 2000).

There also appear to be taxonomic patterns of voltinism at the subfamily level, with the Orthocladiinae having greater voltinism compared to the Chironominae and Tanypodinae (Drake 1982), which is unusual considering the general thermal preferences of these species (Tokeshi 1995). However, this may simply be a reflection of the bias in studies conducted in temperate habitats where the more cool-adapted Orthocladiinae are capable of more generations.

In general, smaller chironomids undergo more generations per year than larger species (Jónsson 1985, Lindegaard & Mortensen 1988, Tokeshi 1995), and growth rates in chironomids have been linked to the size of larvae (Konstantinov 1958b, Mackey 1977a, Huryn & Wallace 1986). In many chironomid taxa, the terminal size of the 4th instar larva, as well as the thermal regime, are predictors of the voltinism for that species (Huryn 1990). However, caution must be used because size is not always a good

predictor of voltinism (Tokeshi 1995). Exceptions to this pattern can be observed in species where the phenology is linked to seasonal foods or a particular habitat, such as sand (Huryn 1990).

Less commonly, overlapping univoltine generations have been proposed or identified (e.g., MacDonald 1956, Gerstmeier 1989). For example, Lindegaard & Mæhl (1992) determined that summer and autumn/spring emergence peaks of Heterotrissocladius changi and Heterotrissocladius oliveri were not the result of a bivoltine population, but rather two overlapping univoltine generations. This form of voltinism raises questions regarding genetic flow among cohorts because it would essentially limit mating between the two cohorts.

EMERGENCE TIMING The timing of emergence for aquatic insects determines the environmental and biological factors to which the adult will be exposed and ultimately determines the success of the generation (Corbet 1964). In general, the greatest peak in emergence in temperate habitats is observed in the spring as water temperatures rise (e.g., Miller 1941, Mundie 1957, Buckley & Sublette 1964, Sandberg 1969, Learner & Potter 1974, Titmus 1979, Soponis 1980, Boerger 1981, Williams 1982, Jónsson 1987, Rempel & Harrison 1987, Wrubleski & Rosenberg 1990, Schmid 1992, Stagliano et al. 1998, Chapter IV) and not the period during maximum temperatures. In some communities, there is also commonly a smaller emergence peak in the autumn. However, some studies of lentic habitats have reported that the greatest emergence occurs at the summer during maximal temperatures (e.g., Brundin 1949, Miller 1941, Judd 1953, Sandberg 1969). Emergence in most temperate and arctic habitats slows or completely stops in the fall or late winter. However, in habitats receiving large amounts of groundwater, emergence can continue throughout the year (e.g., Chapter IV) with the magnitude of winter emergence determined by the regional climate and the degree of groundwater influence. It has been hypothesized that the cause of maximal emergence during spring and early summer is a

response of late-instar, overwintering larvae completing development when temperature increases the following spring (Butler 1980).

The duration of chironomid life cycles appears to place some constraints on the timing of emergence. Univoltine, merovoltine, and semivoltine species are more likely to have spring emergence as they often utilize the change in temperature as a cue to complete development (Iwakuma 1992, Tokeshi 1995). In contrast, multivoltine species are possibly regulated by the availability of food (Iwakuma 1992). Some bivoltine, trivoltine, and multivoltine species also have a large emergence in the spring, possibly as a result of greater synchrony in this cohort compared to those later in the year (Tokeshi 1995). Less commonly, there are also examples of winter-emerging taxa that are bivoltine or more (Berg & Hellenthal 1992, Chapter VI). Autumn emergence usually consists of the final generation of the year for bivoltine, trivoltine, and multivoltine species, although there are some univoltine species with autumn emergence (Tokeshi 1995). In tropical and subtropical streams where temperature is more consistent and low winter temperatures are not present to synchronize emergence, chironomid emergence tends to be more asynchronous and lack distinct peaks (e.g., Frommer & Sublette 1971). However, many tropical streams appear to be influenced by some seasonal factors (e.g., rainfall) that do serve to synchronize some species (Butler 1984).

Although many chironomid species attain maximal emergence in the spring, the timing of peak emergence for individual species varies. For example, some winter emerging taxa such as Diamesa have emergence peaks in winter, late autumn, or late winter (Herrmann et al. 1987, Chapter VI). In addition, as a result of seasonal patterns in temperature, nutrition, and other factors, the duration of separate generations within bivoltine, trivoltine, and multivoltine species can vary considerably.

Corbet (1964) divided seasonal emerging taxa in temperate regions into spring emerging and summer emerging, where spring emerging taxa diapause in the final instar, but summer emerging taxa do not. Danks & Oliver (197a) modified this classification by

including “absolute spring species” which refer to species that diapause as larvae that are ready to emerge as soon as temperatures increase in the spring. Many chironomid species in temperate habitats have a spring/early summer emergence (Singh & Harrison 1984, Wrubleski & Rosenberg 1990).

Different emergence patterns or classifications have been identified/proposed for the emergence or flight activity of chironomids (e.g., Sandeberg 1969, Aagaard 1978a, Jónasson 1987). For example, Aagaard (1978a) divided the chironomid community into three emergence groups based on the timing and duration of the emergence of common species: spring short, long, and summer/autumn short. The cause of these groups were hypothesized to be related to emergence cues and thermal preferences of the different groups, with emergence of spring short taxa timed with rising spring temperatures and autumn/summer short synchronized by photoperiod. The cause of the emergence pattern for taxa with long emergence periods was less clear, although it may have been the result of a thermal threshold above which emergence could occur continuously. Jónasson (1987) based groups on the flight activity of different taxa with different species characteristic of the spring, summer, and autumn periods.

PROTANDRY Protandry appears to be wide spread in Chironomidae, as it has been observed by numerous researchers in the field (Palmén 1956, Hashimoto 1962, Palmén 1962, Danks 1971a, Yamagishi & Fukuhara 1971, Danks & Oliver 1972, Learner & Potter 1974, Ringe 1974, Wiederholm et al. 1977, Cloutier & Harper 1978, Butler 1980, Boerger 1981, Sugg et al. 1983, Okazaki & Yano 1990, MacRae & Ring 1993) and under laboratory conditions (e.g., Hatakeyama 1987, Frouz et al. 2002, Soong et al. 1999, Stevens 1998). However, protandry is not universal in the Chironomidae as it does not occur in some taxa (e.g., Palmén 1956, LeSage & Harrison 1980a). The cause or benefits of protandry in chironomids has not been directly studied and it is not clear whether it is driven by natural or sexual selection or both. Protandry may increase the likelihood of mating or assist in outbreeding (Danks & Oliver 1972, Armitage 1995). For example,

mating success would be increased in short-lived chironomids if there were males ready to mate when females emerge (Armitage 1995). Palmén (1956) hypothesized that early emergence may be a behavioral adaptation that permits males to establish swarms to await females. In the case of marine chironomids, protandry may be related to the fact that males assist with the eclosion of females (Hashimoto 1957, Neumann 1976). It is also possible that protandry is controlled by natural selection if the growth rates of males and females are similar, but there is a benefit to larger females that is not realized by the males (Bradshaw et al. 1997). This has been suggested in chironomids (e.g., Danks & Oliver 1972) and different growth rates of males and females have been determined by Fisher & Rosin (1969) and Fischer (1974). Servia et al. (2006) determined that copper exposure reduced the difference between the development of male and female Chironomus riparius by slowing development of males more than females and subsequently causing a decrease in protandry.

SEASONAL SYNCHRONOUS EMERGENCE Mechanisms of Synchronous Emergence Synchronous emergence is common in many chironomid species especially in habitats with strong seasonality. The duration of a species’ emergence period is generally determined by environmental factors and the biology of individual species. For example, Olander & Palmén (1968) hypothesized that the decreased emergence period for Clunio marinus was shortened by low water temperatures, although voltinism was apparently unaffected. For most species, the emergence period becomes shorter at higher latitudes and altitudes as the growing season is shortened (Corbet 1964, Oliver 1971). However, the opposite would be true of cold-adapted taxa in temperate environments whose emergence period may decrease in populations at lower latitudes. Therefore, as the period of suitable conditions decreases, emergence is forced into a shorter period.

The most important mechanism for synchronizing chironomid emergence is probably the initiation and breaking of dormancy. Seasonal synchronous emergence is common in many overwintering chironomids (e.g., Danks 1971a) and may be the result of dormancy

in these species (Goddeeris 1987). Therefore, depending on the thermal tolerance of the species, low winter or high summer temperatures in temperate habitats may be a synchronizing factor when the final instar is dormant (Corbet 1964, Butler 1984, Singh & Harrison 1984). This can be apparent in multivoltine taxa where summer generations are asynchronous, but the spring generation is synchronized (e.g., Lloyd 1941). Chironomid emergence in arctic and sub-arctic lakes and ponds tends to be especially highly synchronized with the majority of individuals within a species or even an entire community, emerging within days or weeks of each other (Brundin 1949, Oliver 1968, Oliver 1971, Danks & Oliver 1972, Wiederholm et al. 1977, Butler 1980). A study on life history of Einfeldia synchrona in small pond near Ottawa, Canada by Danks (1971a) is a good example of this form of synchronization. This species enters winter dormancy as a 3rd or 4th instar and rapidly completes development in the spring when temperatures rise. Pupation is initiated when temperature exceeds a threshold and long day lengths occur resulting in a synchronized emergence pattern. In addition, the arctic pond communities studied by Butler (1980) had highly synchronized emergence that generally emerged in the same order from year to year. High synchronization has also been identified in alpine and subalpine lakes with most emergence occurring within 10-15 days (Laville 1971a, Laville 1971b). In contrast, chironomid communities in arctic streams and rivers may be more asynchronous (Hayes & Murray 1987). Synchrony of emergence can even vary within a habitat. Emergence synchrony was greater in deeper arctic habitats compared to more shallow habitats with more variable temperature regimes (Danks & Oliver 1972).

Different thermal thresholds for various life stages and phases in aquatic insects may also serve to synchronize development and ultimately emergence (Butler 1984). Lloyd (1941) suggested that in the midge Spaniotoma minima (=Limnophyes minimus) different stages possess different thermal thresholds for development. In general, the thresholds tend to be lower for larval development than for adult emergence (Danks & Oliver 1972). Danks & Oliver (1972) suggest that the pattern of higher thermal thresholds for pupation and emergence compared to larval development may serve to synchronize emergence, as

emergence would be postponed until temperatures increased, during which time lagging individuals could catch up. In some species, emergence may be synchronized by the halting of development at a specific stage that can allow other members of the cohort to catch up (e.g., Paris & Jenner 1959, Goddeeris 1990, Goddeeris 1991, van de Bund & Davids 1994). For example, photoperiodic control of dormancy synchronizes the larval development in Chironomus riparius and subsequently the emergence of adults (Goddeeris et al. 2001). Both temperature and photoperiod was hypothesized to control dormancy in Tanytarsus debilis and synchronized emergence of this species in the spring (Goddeeris 1991). In fact, this species underwent 1-2 periods of dormancy, with both an overwintering oligopause strictly in the 3rd instar and in some cases a short prepupal oligopause in the spring. It was hypothesized that the continuation of feeding and development for individuals that had not attained the prepupal phase allows these individuals to catch up and thereby synchronize emergence of the population.

Advantages of Synchronized Emergence Predator satiation. Although increased emergence can potentially attract a greater number of predators, it has been hypothesized that by concentrating emergence over a short period of time, predators will find ample prey during that period, but become satiated.

Mate finding and oviposition in appropriate habitats. In general, most chironomids are considered to possess short life spans, and therefore it is important that emergence is synchronized to maximize mating success (Butler 1984). For some species emerging in harsh environments (e.g., marine, alpine, arctic), synchronous emergence may be a means to enhance mate finding in an environment where winds or low temperatures could result in the displacement from the habitat or death of adults before they complete mating and oviposition. Some parthenogenetic species have low synchrony compared to sympatric bisexual populations, supporting the hypothesis that synchronous emergence is important to maximize mating success (Butler 1984). Open water marine chironomids do not need to time their emergence to the exposure of the larval habitats, but instead face the threat

of conditions that can displace adults from larval habits before mating and oviposition can occur (Neumann 1976). The open water Clunio marinus populations appear to be suited to these conditions, as Neumann (1966) determined that more than 80% of individuals from laboratory stocks emerged within 1 h of each other. Many species, especially smaller taxa that emerge at low temperatures or under windy conditions, do not aerially swarm but mate on the substrate (e.g., Butler 1980). Therefore, there may be an increased need of synchronization to increase the chances of mating as a result of reduced dispersal abilities and a lack of mating swarms that otherwise helps to aggregate individuals.

In addition, synchronous emergence can also help improve the chance that emergence will coincide with favorable conditions. For some species, optimal conditions for emergence, mating, and oviposition may be ephemeral requiring synchronous emergence and rapid oviposition. In some marine chironomids, the oviposition site is only available for a short period of time and synchronization of emergence is required for successful mating (Neumann 1976). This may also be the case for the antarctic species, Belgica antarctica that has a synchronized emergence and the females typically oviposit within a day (Sugg et al. 1983). This species can be confronted with windy conditions or low temperatures that could lead to adult mortality or displacement from oviposition sites before mating or oviposition can occur.

Reproductive isolation. Although the impacts of congeneric interbreeding in chironomids are not well studied, it would be expected to be detrimental as reproductive effort would be wasted. Butler (1980) hypothesized that the observed staggered emergence from arctic ponds may reduce interactions between individuals that will result in unsuccessful mating especially among congeners. Boerger (1981) identified a relationship between the size of mating swarms and the co-emergence of congeneric species. For example, congeneric species that formed large mating swarms were more likely to emerge at different times. In contrast, species that formed small swarms

emerged at the same time. This suggests that synchronous emergence is a means to reduce multispecies swarms of congeners and decreases the chances of interbreeding.

ASYNCHRONOUS EMERGENCE Mechanisms of Asynchronous Emergence Asynchronous emergence has been observed for a number of chironomid taxa and is generally considered to be the response to long periods when optimal or sufficient temperatures for growth, development, and emergence are present (i.e., habitats that lack strong seasonal patterns). Asynchronous emergence would also be expected to be more common in taxa with broad environmental preferences or plasticity in their preferences. In general, chironomid taxa from tropical habitats or habitats receiving large amounts of groundwater, and therefore with relatively constant temperatures, are expected to emerge throughout the year with relatively asynchronous emergence (Oliver 1971; e.g., MacDonald 1956, Corbet 1964, Sonoda & Trivinho-Strixino 2000). Summer emergence of multivoltine species in temperate habitats will also commonly be asynchronous. Food abundance or other resources may also alter emergence patterns by reducing synchronization under conditions when food is scarce and emergence of some members of a population is delayed (Danks 1978).

Advantages of Asynchronous Emergence Risk reduction. Although synchronous emergence benefits mate finding, the risk to such a strategy is that a catastrophic disturbance (e.g., hurricane, flood) could lead to the loss of all or most of a cohort or population. Therefore, a continuous or protracted emergence will reduce the likelihood of a catastrophic loss. Asynchronous emergence has been identified in several southern hemisphere chironomid communities where seasonality is reduced (e.g., New Zealand – Boothroyd 1987, Boothroyd 1999; Australia – Cranston 1997) and may have evolved as a response to variable climatic conditions (Towns 1981, Boothroyd 1987). More asynchronous emergence would also be less likely to attract large number of predators (e.g., Odonata, birds, bats). For example, Tait Bowman (1980) suggested that swifts and swallows flock over a specific English lake in the spring due to

the availability of large chironomid prey (e.g., Chironomus anthracinus, Chironomus plumosus).

Intraspecific resource partitioning. Asynchronous emergence could also serve to partition resources. By staggering emergence and subsequently oviposition, intraspecific competition may be reduced, thus increasing the overall success of the species. As a result of different body sizes and gape limitations, it would be expected that different instars would utilize different microhabitats and food sources (e.g., Tokeshi 1991b, Berg 1995, Ingvason et al. 2002). For example, Frouz et al. (2004) determined that different algal species had different effects on the developmental rates of different instars of Glyptotendipes paripes. To maximize growth and development, different instars may therefore, selectively feed on different food sources.

VARIABILITY OF EMERGENCE

Many chironomid species and populations possess considerable flexibility in their life histories and phenologies that allow them to utilize different habitats or survive unusual conditions (Hilsenhoff 1966, Danks 1971b, Lindegaard & Jónsson 1979, Butler 1982). This includes differences in voltinism, timing of emergence, and duration of emergence that can vary on an annual basis or can differ between regions, habitat types, and even within habitats. In some species and habitats there is considerable annual and seasonal variability in phenology and emergence of chironomids (e.g., Mundie 1957, Harper & Cloutier 1979, Carter 1980, Rossaro 1987, Wrubleski & Rosenberg 1990, Boothroyd 1999). This variability may be the response to fluctuations in some limiting factor or factors (Fischer 1974, Danks 1978, Rossaro 1987), although genetics may also play a role in life history variability. There also exists a possibility that biological interactions (e.g., competition, predation) could also have different impacts at different locations.

Variation in timing of emergence and voltinism may be manifested as a reduction in developmental rates that cause a delay in emergence and subsequent lengthening of the

life histories as a result of differences in environmental factors (Oliver 1968). The factors that regulate dormancy and the duration of dormancy in chironomids may also vary at different locations. For example, northern populations have longer generation times compared to southern populations as a result of slower developmental rates and longer developmental halts. Life history variation can also be observed in different populations at the same latitude due the influence of elevation, groundwater influence, and thermal effluents. Unfortunately, most studies of seasonal emergence are only one year in duration and do not account for annual differences. Furthermore, in many multiyear studies the parameters measured may not be sufficient to identify the factors that result in annual differences.

VARIABILITY OF EMERGENCE TIMING Although many species emerge during a predictable period or season, the date of first

emergence or the date where 50% of the generation has emerged (EM50) typically varies annually. For example, timing of emergence for Allochironomus crassiforceps was determined to vary annually as a result of differences in temperature from year to year Palmén (1962). Koskinen (1968a) suggested that differences in the emergence time of Chironomus salinarius from year-to-year were the result of temperature variation where the required total degree days occurred at different times between two years. The annual variability of food resources can be responsible for altering the voltinism of chironomids from year-to-year (e.g., Carter 1980). Although flight activity may not be directly related to emergence patterns, Gislason et al. (1995) suggested that annual variation in flight activity along the Laxa River in Iceland was the result of differences in food conditions. Habitat differences can also cause considerable emergence variation even in populations in close proximity to each other. In artic ponds (Danks & Oliver 1972, Butler 1980) and alpine lakes (Laville 1971a), timing of emergence for chironomid species can vary from pond to pond even within a relatively small area. In addition to variation in the timing of emergence, the duration of emergence can also vary on an annual basis. For example, Palmén (1962) determined that the emergence duration of Allochironomus crassiforceps

was shorter when spring temperatures increased rapidly and longer when the spring warm-up was more gradual.

VOLTINISM VARIABILITY Voltinism is not considered to be a fixed character in aquatic insects, but may vary on an annual basis depending on environmental factors (Palmén & Aho 1966, Butler 1984, Sherk & Rau 1996). Generally as conditions become less favorable, developmental rate slows and dormancy may occur, thereby increasing the generation time and reducing the number of generations that can be completed. Different degrees of voltinism within a species is not uncommon as a result of environmental differences between regions, habitats, and variation within a habitat. Only a few examples of intraspecific variation in voltinism patterns are included here due to the large number reported in the literature. Chironomus riparius shows a wide range of voltinism, from one generation (Rasmussen 1984) to at least 5 generations (Gower & Buckland 1978, Goddeeris et al. 2001), with other studies reporting 3-4 generations per year (Learner & Potter 1974, Groenendijk et al. 1998). Similar ranges of 1 to 5 generations per year have also been identified in Chironomus salinarius, Cricotopus annulator, Cricotopus bicinctus, and Cricotopus sylvestris. In marine midges, differences in voltinism have been determined for populations of Clunio marinus at different latitudes and temperatures (Neumann 1976). Differences in elevation can also produce similar effects to latitudinal changes. Sæther (1968) hypothesized that Diamesa valkanovi was semivoltine although some individuals may be univoltine at lower reaches where the snow and ice free period is longer. Differences in voltinism within a species is also common at similar latitudes but different habitats. Learner & Potter (1974) documented annual differences in emergence patterns and voltinism from two ponds for several species of chironomids. Cloutier & Harper (1978) determined considerable variation in the emergence patterns of the same species of Tanypodinae from nearby streams.

A species can also have different voltinism as a result of different thermal regimes within the habitat. In lentic habitats emergence patterns can be complicated due to the gradient

of temperature from shallow to deeper portions of a lake or pond (e.g., Borutzky 1939a, Borutzky 1939b, Brundin 1949, Mundie 1957, Jónasson 1965, Laville 1971a, Yamagishi & Fukuhara 1971, Danks & Oliver 1972, Maitland et al. 1972, Laville & Giani 1974, Titmus 1979, Moore 1980, Uutala 1981, Růžičková 1987, Prat & Rieradevall 1995). For example, Sokolova (1971) determined that Chironomus plumosus and Procladius ferrugineus were univoltine in the profundal of a reservoir but bivoltine in shallower regions of the reservoir. Jónasson (1972) determined that the voltinism of Chironomus anthracinus shifted from univoltine to semivoltine from 11 to 20 m depth. Añón Suárez (2002) determined that voltinism of Ablabesmyia reissi shifted from bivoltine in the shallower regions of a lake to univoltine in the deeper portions, although food quantity differences may also be responsible for these shifts. Stahl (1986) determined that the voltinism of Tanypus stellatus changed from bivoltine to trivoltine and multivoltine as degree days increased in a cooling reservoir. Most examples of variation of voltinism within a habitat come from lentic studies most likely due to more distinct and greater thermal gradients within short spatial scales. However, it would be expected that voltinism of lotic chironomids would change along a longitudinal gradient especially in streams where groundwater influence decreases downstream.

When more extreme individual differences are present in a generation, differences in the voltinism within a cohort can occur. This is termed cohort splitting where part of a cohort may emerge during one season, whereas another portion of the cohort is forced to emerge in the following year (Butler 1984). For example, Jónasson (1972) and Carter (1980) determined that in some cases a portion of a cohort Chironomus anthracinus did not complete development in spring and were forced to complete development and emerge the following spring. Rosenberg et al. (1977) determined that part of the population of Cricotopus bicinctus completed three generations whereas other individuals completed only two generations.

POSSIBLE GENETIC DIFFERENCES Even when studied under controlled laboratory conditions, considerable ranges in development times and responses to dormancy cues are observed for individuals within a cohort. For example, Goddeeris et al. (2001) determined that the duration of dormancy varied, with half of a laboratory population of Chironomus riparius undergoing short diapause and the other half undergoing a long diapause in response to a short-day photoperiod. Different responses to environmental variables and dormancy cues within a population or cohort may be the result of genetic polymorphism or the result of differential responses by different larval instars, phases, or subphases (Danks 1978, Rychen Bangerter & Fischer 1989, Goddeeris et al. 2001; larval phases and subphases in Chironomidae are described in Wülker & Götz (1968), Ineichen et al. (1983), and Goddeeris et al. 2001). Koskinen (1968a) hypothesized that differences between the minimum emergence temperature thresholds of Chironomus salinarius from Norway and those determined by Neumann (1961) in Germany were the result of physiological differences in populations from different latitudes. Danks (1978) detected and discussed the implications of a small fraction of the population of several Chironomini species which emerged under different environmental conditions than their siblings. Danks (1978) hypothesized that rather than constituting leakage or experimental variation, individuals that respond differently than their siblings may essentially provide insurance against environmental changes. It is important to understand the variability within a population as in some cases it is possible that voltinism could vary within a cohort depending on the environmental factors (e.g., cohort splitting). Although genetic polymorphism has been implied in a number of studies on chironomids, there are not any studies that have isolated genetic factors responsible for differences in the timing of emergence in chironomids.

SUMMARY & RECOMMENDATIONS FOR FUTURE RESEARCH

The phenology and emergence timing in chironomids is largely a result of a balance of multiple factors that effect developmental rates and regulate the timing and duration of

dormancy. The most important factor controlling the phenologies in chironomids is usually temperature with other factors such as nutrition, photoperiod, DO, pollution, and biotic interactions also having an influence. However, the importance of these various factors depends on the species and the environmental conditions of the habitat. As a result of multiple factors that can vary annual and between habitats, there can be considerable variation in the timing of emergence and voltinism among species and populations. These patterns can be further complicated by interaction of factors that regulate phenology in chironomids (Ward & Cummins 1979, Storey 1987, Gresens 1997). In fact, these interactions are likely the norm, although they are not commonly addressed in studies.

As a result of the threat of a changing environment (e.g., climate change, habitat fragmentation) faced by chironomids and aquatic insects, it is important that we assess how altered temperatures will impact aquatic species. However, this is impossible without understanding the thermal requirements of chironomids, such as the minimal and maximal temperatures, as well as thermal regimes required for successful development into a reproductive adult. In addition, there is a need to assess genetic variation in thermal requirements and tolerances of aquatic insect species to predict the impact of climate change on these species (Hogg et al. 1995). The ability to model the thermal requirements of chironomids, in conjunction with an understanding of the dispersal and colonization abilities, will allow predictions of the effects of a changing environment on these species and communities.

There is still a great deal that is not understood regarding the life histories and the factors that control these patterns in chironomids. Much of the available research is based on field studies that provide invaluable results, but due to the uncontrolled nature of most field studies, the many variables that control the life histories of chironomids can not be easily separated. For example, many studies have identified a relationship between timing of emergence and thermoperiod, however, these studies are generally observational and do not result from controlled experiments. Photoperiod, thermoperiod,

and autochthonous food resources are often correlated to temperature making it difficult to separate the influences of these factors. Future research should address timing of emergence under different thermoperiods with different photoperiods to separate these two factors. For example, additional research is needed to understand the influence of diel fluctuating temperatures and photoperiods on chironomids in a laboratory setting. The effect of increasing or decreasing photoperiods has also not been assessed in chironomids and may have an effect on the timing of chironomid life histories. In addition, for many studies it is not clear what factors are working on chironomids and what mechanisms are affected. For example, rising temperatures could initiate feeding in chironomid larvae, allow growth and development to resume, or more likely a combination mechanisms.

Some other questions that remain unanswered in the Chironomidae include whether or not true dormancy occurs in the egg and pupal stages and how long can these stages remain viable during dormancy. Does the number of degree days necessary for development and emergence vary depending on the temperature at which the larvae are exposed? In aquatic insects in general, Sweeney (1984) noted that there is little research on the enzymatic, hormonal, or molecular mechanisms regulating life histories. The effect of behavior on development in chironomids is also not well studied. For example, when do chironomid larvae feed? If they only feed at night or day then photoperiod may have a considerable effect on the growth and development rates of chironomid larvae as the duration of feeding could be altered at different latitudes. This might serve to equalize generation times between different latitudes since species that feed during the day will feed longer at high latitudes and balance out the shorter season resulting from lower temperatures (Sweeney 1984).

Since Oliver’s review paper on the life history of Chironomidae in 1971, there has been increased work on the life history of chironomids from subfamilies other than Chironominae. However, the use of Chironominae in controlled laboratory studies still dominate the literature presumably due to the ease of laboratory culture. As a result,

there is a need perform studies on chironomids from other subfamilies including species that are more difficult to culture due to their lotic preferences. The bulk of research on chironomids also has been based on temperate and arctic habitats. Therefore, greater attention is needed in tropical regions as life history dynamics can be different and will help elucidate patterns in the Chironomidae overall.

An understanding of the factors determining developmental rates and the life history patterns of Chironomidae is important in studies using emergence as a measure of a chironomid community. As a result of the huge number of species in the family Chironomidae, there is a poor understanding of the autecology of most species. Unfortunately, there is also much that is not understood of some fundamental aspects of chironomid life histories. Studies are needed to assess both development and emergence in chironomids, because an understanding of emergence patterns and the factors that control the timing of emergence are fundamental to the use of chironomids in ecological and biological assessments.

CHAPTER III

USE OF SURFACE FLOATING PUPAL EXUVIAE TO MEASURE AND ASSESS CHIRONOMIDAE (DIPTERA) COMMUNITIES IN MID-ORDER NORTHERN TEMPERATE STREAMS

ABSTRACT

The ability of different subsample sizes and sampling frequencies to assess various community, diversity, and biotic index metrics and to measure site taxa richness of chironomid surface floating pupal exuviae (SFPE) was tested in six ground-water dominated (GWD) and six surface-water dominated (SWD) streams in eastern Minnesota, USA. A subsample of 300 individuals was sufficient to collect on average 85% of total taxa richness and to estimate most metrics with an error of about 1%. Estimates of taxa richness were less clear and each study will need to assess the proportion of the community that should be collected to achieve specific goals. I recommend that in most cases where collection of winter emerging taxa is not important, sampling should be limited to the period from April to September, as 97% of chironomid emergence was determined to take place during this period. This study provides guidelines to aid future studies to tailor the SFPE method to specific goals and resources depending on the proportion of the community that needs to be evaluated.

INTRODUCTION

Despite the importance of Chironomidae in aquatic systems, in many ecological or biological assessment studies, these insects are often only identified to the family or subfamily level due to the difficulty associated with processing and identifying the larvae (Wilson & McGill 1977, Berg & Hellenthal 1992, Rosenberg 1992). The collection of chironomid surface floating pupal exuviae (SFPE) is an effective method that permits the rapid measure and assessment of chironomid communities. The use of chironomid SFPE also offers an approach with several advantages over other methods commonly used to assess chironomid communities (Coffman 1973). Most attractive is the improved taxonomic resolution, ease of identification, and the greater number of taxa collected by the SFPE method compared to the identification of larvae from benthic samples. In fact most SFPE can be identified to the genus level and some to the species level without slide mounting. The process of emergence and the subsequent accumulation of chironomid exuviae also permits collection of taxa from a wide range of habitats, including those difficult to sample with a kick net (e.g., wood, hyporheic, deep water), and allows a more comprehensive assessment of the community (Wilson 1994). Collections of SFPE can also be considered a better measure of the characteristics of a stream as they represent the end point of the life cycle (i.e., emergence) of a chironomid, and therefore, represent an organism that successfully utilized the habitat. Furthermore, most benthic sampling methods use nets with mesh sizes that are insufficient to collect small larvae (Storey & Pinder 1985, Hudson & Adams 1998); however, the use of SFPE generally does not suffer from this problem when sufficiently fine mesh in nets are used.

The use of SFPE to assess composition of chironomids from water bodies was first proposed by Thienemann (1910). A variety of studies have used SFPE including taxonomic surveys (e.g., Humphries 1938, Brundin 1949, Brundin 1966, Reiss 1968, Lehmann 1971, Murray 1974), ecological assessments (e.g., Coffman 1973, Coffman 1974, Rossaro 1978, Wartinbee 1979, Laville 1981, Soponis 1983, Rieradevall & Prat 1986, Vilchez-Quero & Lavandier 1986, Boothroyd 1988, Casas & Vilchez-Quero 1989,

Coffman et al. 1992, Gendron & Laville 1992, Ruse 1992, Casas & Vilchez-Quero 1993, Blackwood et al. 1995, Fend & Carter 1995, Ferrington et al. 1995, Gendron & Laville 1995, Hardwick et al. 1995, Ruse 1995a, Ruse 1995b, Cranston et al. 1997, Coffman & de la Rosa 1998, Boothroyd 1999, Chou et al. 1999, Ferrington 2000, Wright & Cranston 2000, Hayford et al. 2006, Ferrington 2007), and biological assessment (e.g., Wilson & Bright 1973, Wilson 1977, Wilson & McGill 1977, Wilson 1980, Mason & Lehmkuhl 1983, Wilson & Wilson 1983, Ruse & Wilson 1984, Wilson & Wilson 1984, Laville & Viaud-Chauvet 1985, Frank 1987, Wilson 1987, Wilson 1988, Bazerque et al. 1989, Buskens 1989, Ferrington & Crisp 1989, Wilson 1989, Frantzen 1992, Wilson 1992, Wilson 1994, Ruse & Wilson 1995, Kownacki 1995, Wright et al. 1996, Ruse & Davison 2000, Ruse et al. 2000, Hayford & Ferrington 2005, Calle-Martínez & Casas 2006, Raunio et al. 2007, Sealock & Ferrington submitted). However, the information obtained through collection of SFPE provides a variety of information as many studies present multiple types of results (e.g., biological assessment studies contain ecological information). Some studies have addressed methodological aspects of the SFPE technique (e.g., Wilson & Bright 1973, Wilson 1977, Hayes & Murray 1988, Ferrington et al. 1991, Gendron & Laville 1995, Raunio & Muotka 2005, Kavanaugh et al. submitted), including the physical properties of SFPE, differences in techniques, and effort associated with the approach. However, there is still a need to evaluate this technique to determine how best to apply the technique to maximize its effectiveness while minimizing cost.

One of the more important issues in designing a sampling protocol for studies using SFPE is to determine how to apply subsampling or whether subsampling is necessary. The intention of subsampling is generally to reduce cost and effort, as there is a diminishing return in information provided from most large samples. In addition, subsampling can help to standardize samples and improve comparison between samples. For example, species richness is affected by sample size and through the use of fixed- count samples, the need for rarefaction can be avoided (Vinson & Hawkins 1996, Sovell & Vondracek 1999). In many studies using SFPE, entire samples can be sorted and

identified with relatively low effort. The ability to economically process entire samples is an advantage of the SFPE method, but depending on the collection method, time of collection, habitat type, and enrichment status of a water body, subsampling may be necessary. For example, during periods of synchronous spring emergence or in enriched habitats below waste water treatment plants, samples of SFPE can commonly include thousands of exuviae. Brundin (1966) provides an example from a stream with apparently low enrichment where 12 drift nets set out in November collected approximately 40,000 exuviae in 4 h. Samples of SFPE collected by Coffman & de la Rosa (1998) usually contained at least 1000 exuviae and up to 70,000. Therefore, a subsampling protocol may need to be in place before sample processing begins, to minimize the effort required to process large samples where the processing of thousands of exuviae may only provide limited additional information.

Of the 72 SFPE studies cited in an earlier paragraph, the majority (49) do not describe subsampling. Of the remaining studies, 18 used a fixed count sampling procedure and five studies processed a sample fraction, although this was not always a fixed fraction. Fixed-count subsamples from 100 to 500 are most commonly used in biological assessment of benthic macroinvertebrate samples (Barbour & Gerritsen 1996); however, there are a variety of opinions regarding the size of subsamples that minimizes cost while minimizing the reduction of information to acceptable levels (Doberstein et al. 2000). Similar to benthic collections, subsample sizes for SFPE ranged from 100-600. Of the 18 studies using a fixed-count method, a 200-count subsample was the most common (9) although 100 (1), 300 (2), 500 (4), 600 (2) count subsamples are also used. The use of subsampling is generally performed to allow a reasonable estimate of taxa richness while minimizing effort. For example, Ruse & Wilson (1984) determined that three 200-count samples collected between May and September would detect 80% of a site’s chironomid community as assessed by a 12 collection regime. Wilson (1980) suggested that a single SFPE sample of at least 500 individuals during the summer would provide a “reasonably comprehensive species list”.

The majority of SFPE studies used accumulated SFPE (69%). Most studies using fixed- count subsampling (16 of 17 studies) collected accumulated SFPE (i.e., hand net or pan/sieve). Many of the studies that used collections of accumulated SFPE often reported that a large number of individuals were collected. The collection of relatively large samples containing many SFPE can be advantageous, as small samples are less likely to accurately characterize the community and make comparisons between samples difficult, especially when measures that are sensitive to sample size (e.g., taxa richness) are used. Due to problems with comparing taxa richness between collections with different SFPE abundances, a sample with many SFPE is advantageous as it allows fixed-count subsampling or rarefaction to generate similarly sized samples.

Collections of accumulated SFPE have several other advantages compared to the use of drift net samples. For example, the use of drift nets requires both placement and retrieval of nets which increases sampling effort and time. During the sampling interval, drift nets can be vandalized or damaged by fisherman or boaters. The collection of accumulated SFPE provides an average of the chironomids emerging over a 24-48 hr period, whereas the period of emergence for drift nets is limited to the period the nets are deployed. While this ability to control the sampling duration can be advantageous (e.g., studies on diel emergence patterns), it generally reduces the ability to make comparisons between studies that use different deployment times. Despite these problems with drift nets, neither drift nets nor the collection of accumulated SFPE provides areal measurements of density or richness. There are examples of quantitative sampling of SFPE described by Wartinbee & Coffman (1976) and Coler & Kondratieff (1989) where SFPE were collected from a known area of stream. However, these quantitative methods limit the types of habitats that can be sampled and do not collect midges emerging from a wide range of habitats, which eliminates one of the major advantages of the SFPE method. Coffman (1974) described the use of artificial barriers to prevent SFPE from entering or leaving a sample reach. By collecting SFPE from above the downstream blockage, areal estimates can be made; however, Wilson & Bright (1973) provide evidence that at least some exuviae sink after only 2 hrs and may not be collected by surface collections.

Although areal estimates of chironomid emergence density can not be made, the use of a consistent sampling procedure (i.e., fixed time, fixed volume) can provide semi- quantitative data. If applied consistently, a fixed-count subsample will allow an estimate of the relative abundance when using either accumulated SFPE or drift nets. However, the degree of accuracy for relative abundance measures using accumulated SFPE collections has not been assessed. The use of a timed collection of accumulated SFPE should help to reduce error in the estimation of the relative abundance of chironomids, but the magnitude of this error is unknown and these estimates should therefore be treated cautiously. The use of richness measures should be less prone to sampling error, although this form of measure has also not been sufficiently tested.

A second important issue when designing a study is how often samples should be collected. The goal of most studies will be to minimize the amount of community overlap in samples while maximizing the proportion of the community sampled. Many studies do this or assume that they do this by sampling seasonally so that different spring, summer, and autumn communities can be collected. In many cases, winter samples are not collected and it is assumed that there is no or very little emergence. It is true that in northern temperate streams, the period of chironomid emergence is generally limited to ice-free periods, although some chironomids have been observed emerging through cracks or small openings in ice (Chapter II). The determination of the ice free period can be complicated to estimate as it is dependent on climate and the characteristics of water bodies (e.g., size, ground-water influence). For example, ground-water dominated (GWD) streams in temperate regions generally do not freeze and emergence can occur year round. However, the degree of thermal buffering varies between streams in the same region and often among stretches within a given stream, such that there can be gradients of ground-water influence. In surface-water dominated (SWD) streams, openings in the ice may persist during the winter and emergence can be observed from these openings. However, in the winter both GWD and SWD streams in Minnesota diversity tends to be relatively low, consisting primarily only of Diamesa and some Orthocladius (Chapter IV). In contrast, at lower latitudes winter emergence of

chironomids may be relatively taxa rich. For example, Ferrington (2000) and Ferrington (2007) determined that winter communities (i.e., December through February) of chironomids in Kansas consisted of an estimated 48-52 species. Therefore, based on region and stream type, a researcher must assess whether sampling during the winter is necessary.

Determination of optimal sampling frequency for the collection of SFPE is often not assessed or mentioned in papers that use the method. When recommendations are provided they must be considered cautiously as the methods vary between studies (e.g., drift net versus accumulated SFPE collections) and vary across regions and habitat types sampled. Therefore, there is a need to test the SFPE approach for different regions and habitats to determine the methods that maximize accurate estimation of the attributes of interest while minimizing effort and cost. Specifically, work is needed to determine the fixed-count size sufficient for ecological and biological assessment studies if the fixed- count approach is appropriate. The effect of sample size on the number of taxa collected and on several common indices is determined to assess the sample size necessary to achieve a sufficient number of species and to minimize index error. The frequency and timing of a sampling regime should also be assessed to maximize the use of SFPE in various habitats and regions. Different combinations of samples and sampling intervals were tested to identify when and how often SFPE samples should be collected in small, northern temperature streams to measure taxa richness. In addition, assessments of how different combinations of sample size and sampling frequency affect estimates of taxa richness were also evaluated. Specifically this research provides results on the number of collections and the frequency of these collections that are necessary to obtain a desired percentage of the community as determined through a biweekly sampling program. Furthermore, the percentage of the community that is identified from different subsample sizes is presented to provide recommendations on the use of this method and whether or not it is a tool that is appropriate for SFPE. The objectives of this research were to develop recommendations for the sampling of SFPE from wadeable, northern temperate

streams to maximize the information that can be obtained from chironomid communities while reducing the effort and cost of performing these studies.

METHODS

STUDY SITES AND SAMPLING OF SFPE See Chapter I for a description of study sites and the methods used to collect SFPE. In this research two sets of data were used for analyses. From the 12 sites that were sampled, the SFPE from six sites were picked as 50-count subsamples to permit the generation of subsamples at 50-count intervals from 50-1000 depending on the size of the sample (Chapter I). Analyses that involved determination of the effect of subsample size on various measures were derived from these six streams only. Analyses that assessed the effect of the full sample on taxa richness (i.e., sample frequency and timing) used data from all 12 study sites.

DATA ANALYSIS Before proceeding it is important to define how two terms are specifically used throughout this study: total richness and subsample. Unless explicitly stated, total site richness will refer to the total number of taxa collected from a site using a biweekly sampling program (i.e. 26 collections) over the course of a year. For example, measures of percent total taxa will refer to the number of taxa collected by a particular effort level in relation to the total number of taxa collected in that site during a one year, biweekly sampling program. As defined here total richness does not refer to the total number of species occurring in the stream, as a much more comprehensive sampling regime would be necessary to exhaustively sample these habitats. In addition, because some samples contained more than an estimated 10,000 SFPE, a 1000-count subsample maximum was used for all samples from all 12 sites, although not all samples contained 1000 or more SFPE. As a result, in this study a whole or full sample refers to 1000-count maximum sample derived from a 10-minute collection period. Collections referred to as subsamples are samples with a maximum of 50-950 specimens depending on the

subsample size. See Chapter I for the methods used to generate subsamples. The use of a 1000-count maximum sample size will often result in some taxa not being detected, which will reduce the number of taxa identified from the true richness. However, a subsample size of 1000 individuals was considered to be larger than the subsample size used in most ecological and biological assessment studies and therefore sufficient for most purposes.

Subsampling Only samples with more than 50 specimens were used to determine how subsampling affects richness and metric estimates as it can be assumed that in most cases samples of 50 or fewer specimens will be completely processed. Cumulative species richness curves were calculated for total taxa and sensitive taxa using 50-count subsamples from the six sites where these data were available. Sensitive taxa were identified as taxa with tolerance values of 4 or lower in Ferrington et al. (2007). The following 10 community composition metrics and indices were calculated: % dominant taxon, % Tanypodinae, % Orthocladiinae, % Chironomini, % Tanytarsini, ratio of Orthocladiinae to Chironominae, Shannon’s Diversity Index, Margalef’s Diversity Index, Simpson’s Diversity Index, and Biotic Index (BI). The % Diamesinae and % Prodiamesinae were not calculated due to the small number of taxa and specimens collected from many sites. The equations for the last four metrics are as follows. Shannon’s Diversity Index was calculated as:  n   n  H '= −∑ i  ln i   N   N  th where ni = the number of specimens for the i species; N = the total number of specimens. Simpson’s index of diversity was calculated as 1-D where:

 ni (ni −1) D = ∑   N()N −1  th where ni = the number of specimens for the i species; N = the total number of specimens. Margalef’s diversity index was calculated as: (S −1) D = MG ln N

where S = the total number of taxa in the sample; N = the total number of specimens. The biotic index was calculated as: n T BI= ∑ i i N th th where ni = the number of specimens for the i species; Ti = the tolerance value for the i species; N = the total number of specimens. The tolerance values used in the biotic index were derived from Ferrington et al. (2007). When possible, values for the Upper Midwest were used, but for cases where these values were not available, tolerance values were used from the following regions in this order of preference: Midwest > Mid Atlantic > Northwest. One exception to this guideline was Polypedilum which in the Midwest has an estimated tolerance value of 3.1. However, this value was considered low for most lotic Polypedilum taxa in this region so the value of 6 for the Mid Atlantic was used. Taxa where tolerance values were not available were not included in the analyses, but these taxa were generally uncommon and many are likely semiaquatic.

The two richness measures and the 10 community metrics and indices were calculated at each 50 count subsample step (i.e., 50, 100, 150, 200, ..., 1000) and the difference between the value for the total sample and each successive subsample were calculated. This value was then divided by the range for that metric and multiplied by 100 to determine the percent error from the full sample. For metrics where there are no upper bounds on values (i.e., the ratio of Chironominae to Orthocladiinae and Shannon’s Index) the range was determined as the spread of the returned values. The percent error from the total sample for each successive subsample was plotted to assess how the estimated values for each index changes as the size of the sample increased.

Sampling frequency Mean total site richness was determined for two week intervals for both stream types and plotted as a line graph to illustrate temporal patterns in taxa richness. The percent of total taxa collected (i.e., the number of taxa identified from all 26 collections of SFPE at each site during a one year sampling regime) at each date for each of the 12 sites was

determined to assess when richness was greatest and when sampling would yield the greatest richness. For each site, taxa richness was determined for single samples and composites of two or more samples, to a maximum of six samples. The cumulative taxa richness was determined as the total number of taxa identified by the combined samples (i.e., the cumulative taxa was not simply the addition of the total number of taxa identified from each sample). This cumulative richness could be calculated as the sum of the number of shared taxa among the samples plus the number of unique taxa in each sample. From these combinations, the sample(s) that generated the greatest taxa richness were determined and plotted as the number of collections versus richness, to assess the accumulation of species as the number of samples was increased. The assessment of temporal patterns of mean taxa richness generated from single collections (see above) is useful in determining when the greatest richness occurred for individual sampling events; however, additional analysis was performed to assess how combinations of multiple collections affected these patterns. The month of each sample from the combinations which obtained the greatest diversity was also determined. In cases where there were multiple combinations of samples that produced equal cumulative taxa richness, the samples that were separated most temporally were selected. For example, from the combination of two samples that produced the highest combined richness, it was determined from which months these two samples were collected. This was done for all combinations (i.e., 1, 2, 3, 4, 5 and 6 sample combinations) to determine the distribution of months that generated the greatest cumulative taxa richness. This was done because most sample regimes spread out the sampling period by season or sample interval to minimize community overlap and thereby increase the proportion of the community collected. The temporal distribution of the collections that produced the highest taxa richness values were plotted in a histogram to determine when sampling provided maximum resolution of the community known to occur in each stream.

The percent of total site richness was determined for 2, 4, 6, 8, 10, and 12 week sampling intervals at each site to determine the frequency of sampling required to collect a

different proportion of the total community. The accumulation curves for increasing interval size and sample number were assessed for both GWD and SWD streams.

Subsampling and sampling frequency The combination of sampling frequency and subsample was then analyzed concurrently using data from the six streams where 50-count subsample data were available. The average percent of the community collected at each site was determined for 100, 200, 300, 500, and 1000 count subsamples collected at 2, 4, 6, 8, 10, and 12 week sampling intervals. Due to the smaller number of streams in this comparison (n=6) and because earlier analysis determined there was little difference between the proportions of taxa collected at different collection intervals for GWD and SWD streams (see RESULTS;

SAMPLING FREQUENCY), sites from both classes were combined. The proportion of the total richness identified from different subsample sizes and collection intervals were plotted with a bar chart of the number of collections made in each collection interval.

RESULTS

A total of 82,889 exuviae were picked, identified and enumerated. From this material, 261 taxa were identified (Appendix A). The distribution of sample sizes reveals that 47% of the samples had 100 or fewer specimens including 21% with no specimens (Figure 3.1). All of the samples without specimens occurred during collection events at low water temperatures or when ice cover was present (i.e., October through March). About 59% of samples had 200 or fewer specimens and 71% of the samples had 300 or fewer specimens. Nearly 10% of the samples contained more than 1000 exuviae. Of these samples with >1000 individuals there was no determination of the total number of exuviae in the sample, but in some samples the total count likely exceeded 10,000 specimens. There was some difference in the distribution of the size of samples between GWD and SWD streams. Samples lacking specimens and large samples (i.e., >1000) were more frequently collected from SWD streams (0 specimens - 31%; >1000 specimens - 15%) than GWD streams (0 specimens - 11%; >1000 specimens - 4%). In

contrast 70% of GWD samples had 1-400 specimens compared to only 40% in samples from SWD streams (Figure 3.1).

Figure 3.1: Distribution of the number of exuviae picked from samples collected in ground-water dominated (GWD) and surface-water dominated (SWD) streams.

SUBSAMPLING Based on the richness estimates made from whole samples or 1000 count subsamples, there was a rapid decrease in the return from picking additional specimens (Figure 3.2). A single 100 count subsample was sufficient to collect on average (±SE) 68% (±2.55) in GWD streams and 59% (±2.75) of the total sample richness. An increase from 100 to 200 specimens increased the percent collected by 13-16%. The addition of another 100 specimens (i.e., 200 to 300 individuals) only added an additional 6-8% of the taxa in the sample. Patterns for sensitive taxa richness were similar although slightly larger samples were needed to collect a larger proportion of the sensitive taxa for SWD streams. For both richness measures there was a difference between GWD and SWD streams with accumulation curves lagging in SWD streams.

The percent error of subsamples as calculated as the difference from whole samples for community composition measures were on average less than 5% at relatively small 93

subsample sizes including 50-count subsamples (Figure 3.3). For both streams types and all six community composition metrics, a subsample size of 300 specimens reduced the error to about 1% or less. Below a subsample size of 300 specimens, the percent error was not only greater, but the curves were more unstable with greater standard error for most metrics. There was also greater error observed for SWD stream samples for most of the metrics, with the exception of % Orthocladiinae and % Dominant taxon.

The three diversity indices had larger errors than the community composition metrics especially Margalef’s Diversity Index (Figure 3.4). As with community composition metrics, these errors for diversity indices decreased rapidly and for Shannon’s Index of Diversity and Simpson’s Index of Diversity the error on average was reduced to <2% in 300 count subsamples. Error for Margalef’s Diversity Index was near 7% for SWD streams at 300 individuals, but it was considerably lower for GWD streams (2%). In both Shannon’s Index of Diversity and Margalef’s Index of Diversity, there was greater error in SWD stream samples. The Biotic Index had relatively low error with less than 0.5% error for a 300 count subsample (Figure 3.4). Differences between the estimated BI for SWD and GWD streams are somewhat difficult to interpret, but once the curves stabilized at about 400 specimens it appeared that there is slightly greater error in the SWD streams.

SAMPLING FREQUENCY The percent of total taxa per site collected on each date were similar for both GWD and SWD streams although the percent collected increased earlier and decreased later in the year for GWD streams (Figure 3.5). The time of maximum average percent richness was similar for both stream types with maximum emergence in GWD streams occurring in early September (43%), whereas maximum average percent richness for SWD streams occurred in late August (42%). From early May through early September, the percent collected for both GWD and SWD streams was similar, although there was a small midsummer decrease for both stream types (Figure 3.5).

During the period from March 1 through October 31, 97% of the total exuviae were collected. During this period every taxon collected in this study was represented (i.e., there were no unique taxa collected during the months of November, December, January, and February). The sampling period from April through September collected 84% of the total exuviae and all but three of the total taxa collected. All three of these taxa were from the genus Chaetocladius. Although most taxa were collected during April- September, the relative abundances of some cool-adapted taxa may not be accurately measured (e.g., Diamesa, Chaetocladius, some Hydrobaenus, some Orthocladius, Stilocladius, and Synorthocladius), as a majority of the specimens of these taxa were collected during October through March. In addition, although only three taxa were missed during April-September from the cumulative data, there were a greater number of taxa missed at individual sites (0-10%) with an average of 96% of total site taxa collected during this period.

The maximum percentage of taxa collected for combinations of 1, 2, 3, 4, 5, and 6 samples were slightly higher for GWD streams (Figure 3.6). As the number of samples increased, there was a decreasing return in the additional percentage recovered. On average, the addition of a sample to 1, 2, 3, 4, and 5 samples for both stream types resulted in an increase of the percentage of the taxa richness collected of 15.63%, 12.49%, 6.07%, 4.28%, and 3.45% respectively.

An examination of the distribution of the time of collection for samples comprising the sample combinations which produced the greatest taxa richness revealed that within combinations individual collections were often separated by a period of a month or more (i.e., there were few combinations that consisted of temporally adjacent samples). However, the time of collection was not consistent across sites and did not allow the selection of specific and consistent sampling times that maximized the taxa collected across all sites. For example, the three samples that generated the greatest taxa richness for Trout Brook were collected in April, May, and July, whereas the samples for Rush 98

Creek were collected in May, August, and September. In general, samples were spread from March through October with peaks in the distribution of samples in spring (i.e., May, June) and late summer (i.e., August) (Figure 3.7). The distribution of these samples generally supported was similar to the patterns observed for the percent richness patterns where the greatest percentages of total richness from individual samples was collected in May-June and August-September (Figure 3.5).

The use of a consistent sampling interval collected on average about 15% less of the total richness (Figure 3.8) than the combinations of samples that were selected systematically to produce the combinations with the greatest richness (Figure 3.6). The reduction in the number of collections reduced the percent of the community sampled, with a greater decrease in the proportion of taxa collected associated with larger decreases in the number of samples (Figures 3.8). For example, the greatest decreases were observed at the shifts from 2 to 4 and 10 to 12 week sampling intervals where the proportion of taxa on average decreased by ≈16-18%. In both of these shifts, the number of collection 99

events were reduced by half. Conversely, the lowest decrease (≈7-8%) in the percentage of taxa collected is at the shift from an 8 to 6 week sampling interval where the number of collections is only decreased by 25%. In addition, there is little difference in the patterns between SWD and GWD streams, although there is a slight separation between the two curves as the sampling interval increased.

SUBSAMPLING AND SAMPLING FREQUENCY In general, the shape of the curves for the proportion of taxa identified at different collection intervals for 1000–count samples was similar to patterns derived both from 12 sites (Figure 3.8) and six sites (Figure 3.9). For subsampled data only, the pattern of a decreasing proportion of the community sampled by fewer collections is similar for all subsample sizes assessed (Figure 3.9). The reduction in proportion of taxa collected as the subsample size is reduced was relatively consistent with a drop of about 4-5% at each 100-specimen decrease, with the exception of the shift from a 200- to 100-count

100

subsample where the percent collected decreased by 9%. The results from the testing of subsampling alone on 12 richness, community metrics, and indices (see SUBSAMPLING) suggested that a 300 count subsample was effective in estimating community measures, therefore the remaining results in this section will focus on this subsample size. The use of a 300 count subsample and a 2 week interval sampling regime between April and September collected an average of 85% of the total taxa at a site. Doubling the interval period to 4 weeks decreased the proportion of species identified to 71% of the total richness, a 14% drop. Although there is nearly a 30% drop from the total taxa collected when only 6 samples were collected, from 42 to 88 taxa (mean = 67 taxa) were still collected at each site using a 4 week collection interval.

Figure 3.9: Mean percent of taxa collected from April to September under different sampling regime intervals and subsample sizes for both ground-water (GWD) and surface-water dominated (SWD) streams (error bars represent standard error; n = 6).

101

DISCUSSION

SUBSAMPLING For all analyses of the effect of subsampling on the 12 metrics and indices, a subsample size of 300 specimens identified a majority of the taxa and estimated metrics and indices with a minimal amount of error. A 100 count subsample also on average identified a majority of taxa (64%), but for estimating metrics and indices there was considerable error and variability in this error. A subsample size of 300 specimens was on average sufficient to identify a large proportion (85%) of the taxa collected in these SFPE samples. There was also limited variability and error in the estimation for most of the community composition metrics tested using a 300 count subsample. Furthermore, there was low error and variability in the estimation of the biotic index at a subsample size of 300 specimens. The error for the estimated BI from SWD streams stabilized at a subsample size of about 300-350 specimens. The requirement for at least 300 individuals may be the result of the fact that a biotic index is responsive to abundance of individual taxa and a large subsample size is needed to estimate the relative abundance of individual taxa.

As would be expected, increasing the number of samples results the collection of a greater proportion of the community; however, there is an effect of diminishing returns where a greater sample size provides a smaller number of new taxa. Based on the diminishing returns at larger subsample sizes, a 300 specimen sample size would be expected to collect most of the common species, with rare taxa increasingly more likely to be missed. In addition, most of the samples collected (71%) had 300 or fewer SFPE specimens which indicates that a 300 count subsample protocol would result in only 29% of samples requiring subsampling when a 10 minute collection of accumulated SFPE is used. However, the subsample size would need to be tailored to the goals of the study based on the importance of rare taxa (Cao et al. 2001). For example, some biological assessment protocols ignore rare taxa, and therefore for the purposes of biological assessment a 300 count subsample may be sufficient. However, the impact of rare taxa

102

on biological assessment using chironomid SFPE needs to be examined. In particular, comparisons should be made between SFPE and benthic samples as the SFPE method collects a larger proportion of the chironomid community compared to benthic samples (Ferrington et al. 1991, Ruse 1995a, Sealock & Ferrington submitted). As a result we would expect that SFPE samples are more likely to collect a greater proportion of uncommon and rare taxa. This suggests that the use of SFPE would not require the processing of as many specimens as needed in benthic samples to collect the abundant and important taxa to characterize the composition of the community. However, the effect of subsampling on the estimation of relative abundance of individual taxa was not assessed in this study and should be considered if such measures are part of a study.

In general, there was higher error and variability for diversity indices at lower subsample sizes than for community composition measures. Despite this greater error, two of the three diversity indices evaluated had low errors (<2%) at subsamples of 300 individuals. The exception was Margalef’s Diversity Index where a high error was the result of how this index is calculated, which takes into account both taxa richness and total abundance. As the number of specimens in a subsample increases, the number of new taxa identified from each additional 50-specimen subsample declines. In Margalef’s Diversity Index, a consistently increasing sample size, but a leveling off or asymptote in the taxa richness, results in a decrease of the index because this index is largely a ratio of richness and abundance. As a result of these attributes, Margalef’s Diversity Index may not be appropriate for use with SFPE samples unless subsample size is consistent or rarefaction is used.

Some additional consideration should also be given to the genus-level tolerance values which were used to calculate the BI. Some of the genera collected in this study included a large number of species, including four with more than 10 species (i.e., Cricotopus, Orthocladius, Polypedilum, Tanytarsus). Despite the likely possibility that congeners possess different tolerance values to different anthropogenic impacts, all of these congeners are given the same tolerance value (Ferrington et al. 2007). Therefore, the use

103

of genus-level tolerance values could have a stabilizing effect on the estimation of the BI for communities that contain a large number of congeners with differences in tolerances. In some cases, these values had to be derived from other regions and for some genera no tolerance values were available. These deficiencies including a lack of tolerance values for many taxa at the species level and regional tolerance values suggest that more work is needed to improve tolerance values for chironomids. Therefore, an assessment of the use of a BI based on SFPE requires further testing.

Many studies have used SFPE to effectively assess the biological condition of aquatic habitat using community metrics (e.g., richness, composition, diversity indices, biotic indices) to assess the condition of lotic habitats (e.g., Wright et al. 1996, Cranston et al. 1997, Hayford & Ferrington 2005). In many cases, community metrics are also used to describe chironomid communities as part of ecological studies (e.g., Blackwood et al. 1995). These metrics include many of the measures of taxa richness, taxonomic composition, diversity indices, and biotic indices that were assessed in this study. Some studies using SFPE have also employed multivariate measures, cluster analyses, and indicator taxa to identify the impact of anthropogenic disturbance (e.g., Cranston et al. 1997, Calle-Martínez & Casas 2006, Raunio et al. 2007). The use of subsampling of SFPE samples when data are analyzed with multivariate methods requires further study as there are few assessments of how subsampling effects outcomes of multivariate analyses in biological assessment (Walsh 1997).

There was commonly a difference between the GWD and SWD curves, with a smaller proportion of the community richness for a site and a greater error for most metrics in SWD streams. These differences between GWD and SWD streams were likely the result of the greater taxa richness and taxa turnover in SWD streams (Chapter IV). The difference between stream types indicates that for many measures less effort is needed to sample GWD streams due to the greater stability and lower taxa richness of the communities. However, the differences between these stream types was not large enough to recommend the use of different subsampling strategies for these two stream types.

104

SAMPLING FREQUENCY The greatest diversity and abundance of emerging chironomids were collected between April and September with peaks during June and August. A decrease in richness was generally observed in July in both GWD and SWD streams, perhaps the result of high temperatures or low levels of dissolved oxygen. The total number of taxa collected could therefore be maximized by focusing sampling to periods when the greatest emergence occurs (i.e., April through September). However, such an approach will miss all or the bulk of the emergence of some cold stenothermic species (e.g., Diamesa, Chaetocladius, some Orthocladius, and Micropsectra) which emerge in the winter, early spring, and late autumn. Several studies have observed considerable winter emergence by chironomids (Hågvar & Østbye 1973, Ferrington 2000, Ferrington 2007, Bouchard et al. 2006a) and these winter-emerging taxa may be more species rich and abundant than is usually assumed. Some of these winter-emerging taxa may be very important segments of the community with a large influence on the productivity of the habitat. For example, Berg & Hellenthal (1991) determined that Diamesa nivoriunda contributed 33.9% to the total production of chironomids in Juday Creek in northern Indiana. The bulk of the emergence of Diamesa in the current study occurred from October through March and would therefore be largely missed in an April through September sampling protocol. Although in most cases the relative abundance of these winter-emerging taxa will not be quantified, many of these species continue to emerge into April and early May, so they often will be represented in estimates of taxa richness. Therefore, a study design should consider whether estimates of relative abundance of winter-emerging taxa is important. However, this consideration will generally be most important for GWD streams in the northern temperate region. Most of the SWD streams are frozen during cold months and large emergences generally begin within weeks of ice-off in these streams during March or April, with the majority of emergence completed by October.

The combinations of samples that contained the most taxa collected a considerable portion of the community and suggest that most of the community could be collected

105

with only a few samples (Figure 3.6). However, it is not possible to know a priori exactly which samples or combinations of samples will detect the greatest proportion of the community. Samples comprising combinations that produced the greatest richness were spread across the months March through October with the greatest diversities collected from April through September. Detailed examination of the distribution of these samples did not reveal a consistent pattern across all sites, although sample timing was generally spread out. For example, August samples for some sites contributed to more to richness than other samples, whereas in other sites the greater contribution to richness was from June collections. Such a pattern is expected as temporally adjacent collection events are more likely to produce samples with shared taxa and subsequently smaller total taxa richnesses. Thus a consistent sampling interval that separates collection events as much as possible is probably the best strategy for sampling SFPE. Such a sampling strategy supports the results of Raunio & Muotka (2005) who determined that to best discriminate sites along a disturbance gradient, there was not a consistent combination of sample times. The authors therefore recommended that a sample regime should consist of multiple samples taken from across seasons.

Many SFPE sampling regimes use a consistent sampling interval (e.g., biweekly) during periods of open water. This study indicated that the number of samples in a sampling regime had a considerable impact on the proportion of the community collected. As a result, greater decreases in the proportion of taxa collected were observed as the interval was increased from 2 to 4 and 10 to 12 weeks and a smaller drop was detected at the shift from a 6 to 8 week sampling interval (Figures 3.8). Individual studies will have to assess which sampling interval provides the desired proportion of the community to be collected. Because a relatively large sampling frequency (e.g., 8 weeks) will likely collect most of the common taxa, selection of a sample interval will need to be based on the importance of collecting rare taxa.

106

SUBSAMPLING AND SAMPLING FREQUENCY Both subsampling and sampling frequency are not commonly assessed simultaneously, possibly because of the multiplicative number of permutations of these two factors that are possible. However, it is necessary to determine the effects of both of these aspects of sample design on estimates of taxa richness. In general, overall patterns in accumulation curves were similar with a steady decrease of about 4-5% of the proportion of taxa collected between 1000, 500, 300, and 200 count subsamples. There was a greater decrease in the proportion of taxa collected (9%) when the subsample size was decreased from 200 to 100 specimens (Figure 3.9). The accumulation curves for sampling frequency was similar to patterns observed when sampling frequency was considered alone, with decreases in the proportion of taxa collected ranging from ≈6-16% at the sampling intervals tested. However, when subsampling is also considered, the decrease in the estimate of taxa collected is compounded by reductions in the sample size. For example, on average only 29% of total taxa were collected with a single sample using a 300 count subsample (Figure 3.9). Thus using a 300 count subsample, six samples (i.e., 4 week sampling interval) were required to collect on average more than 2/3 of the community.

The recommendations from other studies for the number of samples required to estimate the richness at a site are as low as a single sample collected between June and September (e.g., Wilson 1980). In Finland however, Raunio & Muotka (2005) estimated that in a large river and four smaller rivers, three-sample combinations identified 57-71% and five-sample combinations identified 72-82% of the community estimated from a first- order jackknife analysis. A three-sample regime was determined to collect 80% of the community in English rivers by Ruse & Wilson (1984). This study showed that chironomid communities in mid-sized (i.e., 3rd to 4th order) northern temperate streams in North America are relatively diverse and a correspondingly greater number of samples is needed to estimate the number of species present. The results of the present study are similar to those of Gendron & Laville (1995) who determined that 7-8 samples collected

107

every 4-5 weeks were sufficient to characterize the chironomid community in a temperate, 4th order river.

As a result of resource limitations, the goals of most studies are not to collect and identify every species that is present at a site, but rather to obtain estimates of community attributes that can be used to compare sites or site types. Such assessment may include measures of taxa richness, acknowledging that only a fraction of the community is collected. For example in biological assessment and most ecological studies, the use of consistent sampling approach and sample processing protocols should provide an estimate of the relative taxa richness based on some standard unit of effort that can be used to compare sites. Other measures of community attributes are also commonly used which often are largely driven by the most abundant species. Despite a reduction in the number of samples and sample size, a relatively small number of subsampled samples will still collect most of the common species, with increasingly rare taxa being more likely to be missed. This statement is easy to reconcile with subsampling because as long as subsamples are removed from a sample randomly, abundant taxa will usually be present. However, unlike subsampling, by increasing the sampling interval, there is a greater chance of missing taxa that while abundant, have highly synchronous emergences.

In contrast to biological assessment, studies of phenology will require a greater sampling frequency to identify emergence patterns of the taxa within a stream. To determine the voltinism of chironomid taxa using SFPE, samples need to be collected frequently enough to identify patterns for multivoltine species (Boothroyd 1999). However, it should be recognized that there are limits to the use of SFPE to assess voltinism, especially for species with overlapping generations. If study goals are to explicitly determine the voltinism of chironomid species, then the use of benthic samples collected over very short temporal scales would be more appropriate than collecting exclusively emerging specimens (Learner & Potter 1974, Butler & Anderson 1990, Wrubleski & Rosenberg 1990, Berg & Hellenthal 1992).

108

CONCLUSIONS

Some studies have identified considerable variability in the response of chironomids to anthropogenic stressors (e.g., Lenat & Folley 1983, Rabeni & Wang 2001). However, because most biological assessment studies use techniques that primarily collect larvae, taxonomic resolution is generally limited to the genus level. Thus, increased variability or lack of a response by chironomid communities to anthropogenic stressors could be the result of insufficient resolution and a limited understanding of the ecological requirements of these species. Techniques using SFPE can achieve much better taxonomic resolution and provide a more comprehensive collection of the chironomid community at a site. Many studies have used SFPE to effectively assess anthropogenic impact, however there have been few assessments of how the results of benthic and SFPE sampling methods compare. Direct comparisons of benthic and SFPE methods need to be made to determine how well the SFPE method performs for the commonly used ecological and biological assessment techniques such as multimetric and multivariate methods. In addition, it is also fundamentally important to assess how changing sampling intervals and subsampling large samples will influence common biomonitoring metrics, and then to devise protocols that maximize information and effectiveness while minimizing cost.

The results of this study corroborate the subsample size used by other studies (i.e., 200- 500) to estimate community measures. The determination of a universal target for subsample size and sample frequency for the estimation of taxa richness are less clear and decisions should be based on the goals of individual studies. For example, the goals of biological assessment studies are generally to characterize community structure by determining the presence or absence of important taxa and often the relative abundance of these taxa. Subsampling SFPE samples to 300 individuals from three samples should be sufficient to collect the majority of the most abundant, and therefore the taxa that contribute most to the structure and function of aquatic communities. Estimations of total taxa richness or phenological studies will generally require a larger subsample from

109

samples collected more frequently, as the chironomid communities in many 3rd to 4th order northern temperate streams (especially SWD streams) can harbor more than 100 species of Chironomidae (Chapter IV). There was some difference between GWD and SWD streams due to the greater richness and emergence turnover in SWD streams. However, the differences were not sufficient to propose that different sampling or processing methods should be used for these stream types, although consideration should be given in the analysis of data from GWD and SWD stream types as the communities are quite different. It is hoped that the results of this study will provide guidelines to enhance future research using chironomid SFPE to assess and study streams in this region.

110 CHAPTER IV

RICHNESS, TAXONOMIC COMPOSITION, AND PHENOLOGY OF CHIRONOMIDAE IN NORTHERN TEMPERATE STREAMS WITH CONTRASTING THERMAL REGIMES

ABSTRACT

Water temperature is a major factor affecting the life histories of aquatic organisms and subsequently influences the composition of aquatic insect communities in lotic habitats. The effects of thermal regime on aquatic communities must be understood if these habitats are to be protected from anthropogenic stressors and environmental change. Collections of chironomid surface floating pupal exuviae from six thermally variable (surface-water dominated [SWD]) and six thermally stable (ground-water dominated [GWD]) streams were made over one year to assess the influence of thermal regime on the taxa (i.e., operational taxonomic units) richness and community composition on this family in northern temperate streams. The total taxa richness in SWD streams (mean = 117 taxa) was significantly greater than total taxa richness in GWD streams (mean = 69 taxa). The composition of chironomid subfamilies and tribes was also significantly different between the two stream classes with more taxa from the subfamilies/tribes Chironomini, Tanytarsini, and Tanypodinae present in SWD streams and only the Prodiamesinae more taxa rich in GWD streams. In contrast, richness of Orthocladiinae and Diamesinae were not significantly different between the two stream classes. Greater taxa richness in SWD streams is related to the high summer water temperatures that permit the utilization of the habitat by warm-adapted taxa (e.g., Chironomini, Tanytarsini, and Tanypodinae) and low autumn, winter, and spring temperatures which are suitable to cool-adapted taxa (e.g., Orthocladiinae and Diamesinae). Patterns of richness can be explained by greater thermal variability which permits the coexistence of a greater number of taxa utilizing the same habitat at different times of the year. In addition, it is possible that greater variability reduces biotic interactions that would benefit a smaller

111 number of taxa that are better competitors. These results suggest that in these temperate streams, a greater number of species can utilize a habitat due to differences in their thermal preferences which permit a temporally staggered utilization of the resources in the habitat. This study has implications for biological assessment that often relies on the taxa richness and composition to monitor the biological health of a habitat. If there are significant differences in the communities between relatively unimpaired habitats, biological assessment methods will need to compensate for these differences to avoid misclassification of stream conditions. In addition, these results have implications for climate change where increases in temperature may result in an increase in overall taxa richness, but could cause the loss of cool-adapted taxa.

112 INTRODUCTION

Thermal regime is a major determinant of the life history, phenology, taxonomic composition and structure of macroinvertebrate communities in lotic waters. Aquatic insects display a wide range of thermal preferences including cold stenothermic, warm stenothermic, and eurythermic (Rossaro 1991). These thermal preferences determine survival at different temperatures and influence life history characteristics such as growth, development, timing of emergence, voltinism, and dormancy (Macan 1961, Ward & Stanford 1982). Specifically, aquatic insects require sufficient cumulative heat energy (i.e., degree days) above a minimum threshold to develop into adults capable of reproduction. These thermal requirements largely determine the range of habitats that are suitable and the broadscale spatial distribution of the insect. In many cases, there is also a temporal aspect to the development and activity of most aquatic insects due to the dynamic nature of the thermal regime of most aquatic habitats. The annual variation in water temperature limits the period in most temperate habitats when sufficient heat energy occurs for growth, development, emergence, and reproduction. Therefore, the timing and pattern of temperature (i.e., diel and seasonal amplitudes) in aquatic habitats determine the suitability of a stream and the life history timing within that habitat. However, distributional and seasonal patterns of an aquatic insect species can be complicated as a result of a number of factors that modify the effect of local conditions such as ground water influence, stream size, stream shading, and altitude. In addition, many aquatic insect species exhibit life history changes and behaviors (e.g., modified voltinism, dormancy, burrowing, case building) to maximize exposure to optimal temperatures. The thermal conditions in lotic habitats combined with the thermal preferences, thermal optima, and biology of the species present in a regional species pool have a strong influence on the composition of the communities for the aquatic habitats in that region. Many other factors beside water temperature (e.g., substrate, discharge, food resources) are also important in determining the success of a species in an aquatic habitat (Macan 1961). In addition, biotic interactions can also be important where species that have the least amount of niche overlap are more likely to coexist, although the

113 importance of these interactions in chironomid communities has come into question (Tokeshi & Townsend 1987, Townsend 1989, Tokeshi 1992a).

Although temperature and other factors establish the suitability of a habitat, more complex patterns determine the actual composition of an aquatic insect community at any given time. In general, insect communities in lotic systems are dynamic at seasonal and greater temporal scales due to the variable nature of environmental conditions in these habitats and biotic interactions. Many studies have identified environmental variability and disturbance as major factors driving richness and community composition in lotic systems (e.g., Vannote et al. 1980, Resh et al. 1988, Coffman 1989, Reice et al. 1990, Reice 1994). However, for the purposes of this study, environmental variability in a stream must be separated from other confounding factors. Disturbance is generally considered an event outside the normal range of conditions that alters the physical environment and changes resources, thereby disrupting communities and populations (Resh et al. 1988). Present theory regarding disturbance as it relates to taxa richness is that intermediate disturbance leads to the greatest richness, with lower richness at high and low levels of disturbance as explained either through the intermediate disturbance hypothesis (Connell 1978) or the dynamic equilibrium model (Huston 1979) (see Resh et al. [1988] for an overview of these hypotheses and their applicability to aquatic ecosystems). In contrast, environmental variability or heterogeneity is different from disturbance, because it is usually predictable and in many cases aquatic insects have evolved mechanisms to survive these changes (Stanford & Ward 1983, Reice et al. 1990). The relationship between taxa richness and environmental variability would be expected to be linear with increased variability correlated with greater taxa richness. The line between disturbance and variability is not clear (Resh et al. 1988) and it could be argued that temporal forms of environmental variability are mild disturbance. However, for the purposes of this study, environmental variability or heterogeneity is defined as change that is predictable rather than stochastic and results in a positive relationship with taxa richness.

114 Many studies have assessed community and richness patterns for aquatic insects and chironomids as a function of longitudinal, latitudinal, and altitudinal gradients (e.g., Ide 1935, Sprules 1947, Zaćwilichowska 1970, Vannote et al. 1980, Gray et al. 1983, Prat et al. 1983, Ward & Williams 1986, Lindegaard & Brodersen 1995, Jacobsen et al. 1997, Coffman & de la Rosa 1998). From this research, the pattern emerges that increased taxa richness is influenced by the heterogeneity and variability (e.g., substrate, temperature, food resources) of the habitat. In many studies of aquatic insect communities along natural gradients as well as in many other aquatic studies, temperature is often considered to have a large if not the largest influence on the abundance, composition, and richness of aquatic macroinvertebrates (e.g., Ide 1935, Ward 1976, Hildrew & Edington 1979, Ross & Wallace 1982, Gray et al. 1983, Thorp & Chesser 1983, Ziser 1985, Ward & Williams 1986, Coffman 1989, France 1991, Lindegaard & Brodersen 1995, Ferrington 2000, Jacobsen et al. 1997, Milner et al. 2001). In addition, several studies on the Chironomidae have determined or hypothesized that chironomid community composition and taxa richness is correlated to thermal diversity (e.g., Miller 1941, Coffman 1989, Lindegaard & Brodersen 1995). For example, Coffman (1989) proposed the hypothesis that tropical chironomid communities should be less diverse due to reduced annual environmental variability. This suggests that in seasonally stable habitats (e.g., tropical streams, ground-water influenced streams, sites downstream of bottom-release dams) temperature is more constant and the subset of species that are best suited to the conditions present will dominate (Ward 1976). In contrast, more seasonally variable temperature regimes (i.e., surface-water dominated temperate streams) have greater niche space as a result of greater thermal heterogeneity, which results in increased taxa richness (Ward & Stanford 1982). In other words, species in seasonally variable habitats can partition the habitat temporally with cool-adapted taxa utilizing the habitat when temperatures are predictably low and warm-adapted taxa occurring when water temperatures are predictably high. To employ such a life history, many species have developed a variety of mechanisms to survive through periods when the thermal regime is unsuitable or competition for resources is too high (e.g., dormancy, rapid colonization) (Ward & Stanford 1982).

115

Thermal preferences in the Chironomidae are often apparent at the subfamily/tribe level with most members in the subfamilies Diamesinae and Orthocladiinae emerging at lower temperatures than those in the subfamilies/tribes Tanypodinae, Tanytarsini, and Chironomini (Oliver 1971, Coffman 1973, Ferrington et al. 1993, Berg & Hellenthal 1992, Armitage 1995). The thermal preferences of Prodiamesinae are less clear as they are not taxa rich at high elevations as are the Diamesinae, but reach their greatest richness in GWD lowland streams (Lindegaard & Brodersen 1995). As a result, the Diamesinae and Orthocladiinae are generally considered more cool-adapted, whereas the Tanypodinae, Tanytarsini, and Chironomini are primarily warm-adapted. These patterns of shifting taxonomic compositions are apparent along latitudinal, longitudinal, and altitudinal gradients where temperature is a major variable. For example, in tropical and subtropical habitats (e.g., Freeman 1955, Lehmann 1979, Soponis 1980, Ferrington et al. 1993) temperatures may not be low enough during the winter to accommodate cold- adapted species and these taxa are generally less species rich in these regions. In contrast, taxa such as the Orthocladiinae tend to consist of a greater proportion of the community in temperate and arctic habitats. As a result, the timing of emergence for these subfamilies/tribes provides insight into the community dynamics in lotic systems, as well as information on broadscale patterns of community composition in aquatic insects. Although these broad taxonomic generalizations appear to hold, within subfamilies/tribes there may be taxa (i.e., genera or species) that apparently have divergent thermal preferences in relation to the majority of the subfamily/tribe (Rossaro 1991). Unfortunately, little is understood regarding evolutionary histories in chironomids as they relate to thermal preferences. However, it is important that thermal preferences and other requirements of aquatic insects are understood if they are to be used effectively in biological assessment studies or to understand the effect of climate change on these ecosystems.

Several studies have assessed the impact of altitude, latitude, and longitude on chironomid communities, with thermal regimes often identified as a major factor in

116 explaining richness and composition patterns; however, there are many other factors in these comparisons that can impact chironomid community patterns (e.g., climate, stream size, productivity, regional species pool). For example, the predicted richness patterns in Coffman & de la Rosa (1998) were obscured by differences in the regional species pools between temperate and tropical streams. To minimize the influence of these other factors, this study assessed chironomid community patterns from streams with substantially different thermal regimes, but within the same geographic region. Three major questions are addressed by this research:

Question 1: Are thermally variable streams (i.e., SWD streams) more taxonomically rich than more thermally stable streams (i.e., GWD streams)?

H0: Chironomidae total richness does not vary between thermally variable and thermally stable streams.

HA: The chironomid taxa richness in streams with more variable thermal regimes will be greater compared less thermally variable streams due to the ability of both warm-adapted and cool-adapted taxa to utilize the habitat at different times in thermally heterogeneous streams.

Question 2: How does thermal regime affect the taxonomic composition of Chironomidae communities?

H0: The taxonomic composition of Chironomidae subfamilies/tribes do not differ between GWD and SWD streams.

HA: The subfamilies Diamesinae, Prodiamesinae, and Orthocladiinae will be more taxonomically rich and abundant in GWD streams. In contrast, the subfamilies/tribes Chironomini, Tanytarsini, and Tanypodinae will be more taxonomically rich and abundant in SWD streams. Additionally, within both GWD and SWD streams, there will be seasonal patterns for these subfamilies with Chironomini, Tanytarsini, and Tanypodinae more taxonomically rich and abundant during summer and Diamesinae and Orthocladiinae more taxonomically rich and abundant during fall, winter, and spring.

117

Question 3: How is the duration of emergence for chironomids and community turnover influenced by thermal regimes?

H0: The duration of emergence and turnover of chironomid communities do not differ between GWD and SWD streams.

HA: The length of emergence for chironomids will be, on average, longer in thermally stable environments (i.e., GWD streams). In contrast, chironomids in streams with greater thermal variability (i.e., SWD streams) will, on average, have shorter periods of emergence due to increased turnover in the community resulting from greater seasonality.

These hypotheses were tested by an assessment of the influence of different temporal (i.e., seasonal) and spatial (i.e., between stream types) patterns of thermal regime on the emergence of lotic chironomid communities. The emergence of chironomid communities was assessed during one year using surface floating pupal exuviae (SFPE) from 12 sites including six thermally stable streams (i.e., ground-water dominated [GWD]) and six thermally variable streams (i.e., surface-water dominated [SWD]). Due to the large diversity of chironomids and the wide range of thermal diversity of the species within this family, this group is a good choice to assess the effect of water temperature on aquatic insect communities. In fact, chironomid fossils, often only identified to genus-level, are commonly used to reconstruct past climate conditions including past temperatures in ancient waterbodies (e.g., Walker et al. 1991, Lotter et al. 1999, Olander et al. 1999, Brooks & Birks 2000, Brodersen & Anderson 2002, Ilyashuk & Ilyashuk 2007). The use of collections of accumulated SFPE is also predicted to maximize the proportion of the community collected while minimizing the effort required (Coffman 1973, Ferrington et al. 1991). Although this method does not provide areal measures of abundance, at sufficient sample sizes and sampling frequencies, the collection of SFPE should provide a reasonable estimate of the taxa emerging during the 48 h preceding the collection event (Chapter III). Furthermore, use of a standardized sampling procedure (i.e., 10-minute timed sample period) will allow comparisons between sites in this study. Due to the large

118 number of species and relatively large number of sample sites, results and discussion of species-specific patterns within and between SWD and GWD streams could not be thoroughly presented. Rather, this study focused on broadscale patterns by assessing how community and subfamily/tribe level patterns relate to different thermal regimes in wadeable, northern temperate streams.

METHODS

STUDY SITE, SAMPLING, AND SAMPLE PROCESSING A total of 12 sites consisting of six GWD (Browns Creek, Eagle Creek, Mill Stream, Pine Creek, Trout Brook, and Valley Creek) and six SWD streams (Cedar Creek, Chub Creek, Credit River, Rock Creek, Rush Creek, and Sunrise River) were assessed in this study. See chapter I for a description of the study sites and for the methods used to measure thermal regimes, collect SFPE, and process samples.

DATA ANALYSIS Richness Richness estimates were derived from collections made during 2002, 2003, and 2004. There was minimal variation in the thermal regime of these sites from year to year and more importantly, the thermal regimes between the two stream classes were consistently different regardless of the year (Chapter I). This indicates that potential effects of thermal regimes on these communities were consistent during the three years sampling was performed. However, there could naturally be considerable variability in taxa richness from year to year as a result of other factors. Multiyear studies of chironomid communities in the literature were also assessed to identify annual taxa richness variability within lotic sites; however, there are relatively few studies that include a multiyear sampling regime and many of these studies do not report richnesses for individual years. One of the few examples of a multiyear study on a chironomid community was performed by Seibert (1980) using the collection of adults in a GWD stream. Taxa richness numbers for four years of sampling had a mean (±SD) of 95.75

119 (±10.53) taxa. In another example from a GWD stream, Ruse (1995b) collected chironomids using SFPE and determined that the mean (±SD) taxa richness in this site was 67.33 (±1.53) taxa. Based on limited thermal differences between 2002-2004 in the study sites and evidence from other studies that annual variation in richness within a site is not large (particularly for SFPE sampling programs), year to year variation in estimations of chironomid richness should be minimal. Therefore, estimations of taxa richness from a site using collections of SFPE would not be expected to vary greater from year to year within a three year period, which indicates that comparisons of taxa richnesses between sites sampled during different years should be valid.

The total richness and subfamily/tribe richness was determined for each site and for the two stream classes. The tribe Pseudochironomini comprised only a single taxon and is phylogenetically closely-related to the tribe Chironomini. Therefore the single species of Pseudochironomini was included with Chironomini for convenience in the results and analyses presented. Differences in taxa richness for the six subfamilies/tribes and total richness between the two stream classes were tested using a Mann-Whitney U-test in the NCSS© program (Hintze 2001). The Mann-Whitney U-test is a nonparametric test used in place of the t-test when the assumptions for normality are not met and is useful when there are few items in each sample (Sokal & Rohlf 1995, Hintze 2001). Due to the small sample sizes, and because the assumption of normality was not met for several of the variables tested, this nonparametric test was used. One of the assumptions of the Mann- Whitney U-test is that there are no ties; however, this assumption was not met for several comparisons so the Approximation With Correction in NCSS© was used when ties were present. The relationship between chironomid taxa richness and the range of daily mean water (DMT) temperatures was assessed using a linear regression in NCSS© and graphically portrayed using SigmaPlot©.

Taxonomic composition and subfamily/tribe emergence patterns The use of collections of accumulated SFPE does not provide measures of areal diversity (Chapter III), and subsequently there is some uncertainty regarding the magnitude of

120 error in abundance estimates associated with this method. A log transformation was used on abundance data to reduce misclassification of taxa abundance through the use of a measure that does not provide areal estimates of abundance. The raw and relative abundances of taxa between stream classes were compared using methods modified from Hawkins et al. (2000) to identify chironomid taxa that were most associated with each stream class. To compare raw abundance between GWD streams (GA) and SWD streams (SA), the logarithm (base 10) of the difference between the two stream classes plus 1 (i.e., log[(GA-SA)+1]) was calculated for each taxon. Relative abundance between classes was compared by calculating the logarithm (base 10) of GA+1 divided by SA+1 (i.e., log[(GA+1)/(SA+1)]). However, each of these measures suffer from different problems.

The measure of raw abundance differences can be misleading for very abundant taxa. For example, if 10,000 individuals are collected on average in one stream class and 11,000 in the other class, a difference of 1,000 individuals is large compared to another taxon where fewer than 1,000 total individuals were collected. For measures of relative abundance, taxa with low total numbers, but a relatively large difference between classes will be given a high value. However, interpretations based on taxa with small numbers can be misleading and should be made cautiously. To remedy these short comings, the ratio of the raw abundance difference to the relative abundance difference was calculated to identify taxa that had large raw abundance differences but low relative abundance differences. Arbitrarily, taxa with a ratio of less than 0.30 were eliminated and taxa were ordered by raw abundance differences. The thirty taxa most abundant within each stream class are presented in this chapter and the complete list is available in Appendix C. Only the thirty taxa with the greatest difference between stream classes is presented in this chapter due to space restrictions and because these taxa represent more important differences, as they are taxa with greater total abundance and greater difference between the classes. To assess seasonal richness and subfamily/tribe emergence patterns, the number of subfamily/tribe taxa identified from each sample was plotted as bar graphs with line plots of the thermal regime for each site using SigmaPlot©.

121

Emergence patterns The estimated emergence duration (EED) for subfamilies/tribes was calculated following the “estimation method” presented by Coffman & de la Rosa (1998). The EED was calculated by determining the proportion of samples from a site in which a taxon occurs and multiplying this proportion by the sample interval (i.e., 14.04 days). This method assumes that the proportion of samples from which a taxon is collected is equal to the proportion of the year that it emerges. This assumption is influenced by several factors including how effectively the taxon is sampled and the sampling frequency (Coffman & de la Rosa 1998). Although the EED may not be an exact measure of a taxon’s duration of emergence it should be at least an approximation for abundant taxa if the sampling frequency is relatively short (e.g., 2 weeks). Furthermore, because sampling effort and frequency was consistent between sites and samples, it can be assumed that meaningful EED comparisons can be made between the sites in this study. Subfamily/tribe and average site EED values were compared using a Mann-Whitney U-test in NCSS© with the Approximation With Correction when ties were present.

Turnover was determined by calculating similarity in relation to the interval between samples. Similarity was calculated for all possible combinations between all samples in a site as Sorenson Similarity Coefficient. The Sorenson Similarity Coefficient was calculated as: 2a C = jk 2a + b + c where a = the number of taxa in common, b = the number of taxa present in sample 1 only, and c = the number of taxa present in sample 2 only. The similarity coefficient is sensitive to differences in shared taxa and taxa unique to the sample; however, it does not take into account taxa that are absent from both samples. For this coefficient, unity equals complete similarity (i.e., all taxa are shared) and 0 if the samples are completely dissimilar (i.e., no shared taxa). In this experimental design, there was 26 sample events for each site, and therefore at total of 325 sample comparisons were made for each site.

122 Turnover differences between sites were compared by plotting similarity against the interval between samples using SigmaPlot©. The mean Sorenson Similarity Coefficient of only adjacent samples (i.e., 25 sample pairs per site) was also compared between the two stream classes using a Mann-Whitney U-test in NCSS©. The mean similarities between adjacent samples from the two stream classes were also assessed graphically over time using SigmaPlot©.

RESULTS

RICHNESS A total of 82,889 SFPE were picked from the samples collected at the 12 sites. From this material a total of 261 chironomid taxa were identified from 92 genera (Appendix A). The taxa collected included members from seven subfamilies/tribes: Tanypodinae, Diamesinae, Prodiamesinae, Orthocladiinae, Chironomini, Pseudochironomini, and Tanytarsini. This includes all of the subfamilies known from Minnesota with the exception of Podonominae. However, one whole pupa from the genus Lasiodiamesa (Podonominae) was collected from Cedar Creek, but was not counted in the totals because it did not represent an emerged individual. As discussed in Chapter I, the range of mean daily water temperatures were significantly different between GWD and SWD streams with a thermal range of 14.28 °C in GWD and 24.77 °C in SWD streams (Table 4.1). The greatest number of total taxa collected at an individual site was 131 identified from Chub Creek (SWD) and the lowest was 54 taxa in Eagle Creek (GWD) (Table 4.1). On average 69 taxa were identified from individual GWD streams compared to 117 in SWD streams. This difference in the average total number of taxa between the two stream classes was significant (p = 0.0050) (Table 4.1). Across all streams in each of the two stream classes, 148 total taxa were collected in GWD streams and 238 total taxa were collected in SWD streams (Table 4.2). From these taxa, a total of 113 taxa were unique to SWD streams only, compared to only 23 taxa unique to GWD streams.

123 Table 4.1: Taxa richness for chironomid subfamilies/tribes with stream class means (% taxa richness) and p-values derived from a Mann-Whitney U-test comparing GWD and SWD stream classes (TP = Tanypodinae, DI = Diamesinae, PR = Prodiamesinae, OR = Orthocladiinae, CH = Chironomini, TT = Tanytarsini; thermal range is the range of mean daily water temperatures recorded in each stream during one year; * indicates a significant difference between stream classes at the α <0.05 level). Thermal Site Range* DI PR* OR CH* TT* TP* Total* GWD Streams Eagle 9.47 1 2 32 8 8 3 54 Pine 15.51 1 2 40 19 15 7 84 Trout 12.92 2 2 57 9 10 2 82 Browns 16.59 1 2 48 12 11 4 78 Mill 17.20 1 2 37 7 9 2 58 Valley 13.96 3 2 37 7 7 1 57 Mean 14.28 2(2) 2(3) 42(61) 10(15) 10(14) 3(4) 69 SWD Streams Cedar 24.67 1 0 34 39 28 16 118 Chub 25.43 1 0 48 38 27 17 131 Credit 23.04 1 1 50 29 17 9 107 Rock 24.72 2 1 54 27 29 7 120 Rush 24.75 3 2 58 23 26 7 119 Sunrise 25.94 1 0 57 22 20 7 107 Mean 24.77 2(1) 1(1) 50(43) 30(25) 25(21) 11(9) 117 p-value 0.0022 1.0000 0.0089 0.1712 0.0050 0.0051 0.0090 0.0050

TAXONOMIC COMPOSITION On average, the subfamilies/tribes Orthocladiinae, Chironomini, Tanytarsini, and Tanypodinae were more taxonomically rich in SWD streams (Table 4.1). Prodiamesinae richness was on average greater in GWD streams and there was no difference in the average richness of Diamesinae between the two stream classes. Differences in species richness between the two stream classes was significant (α <0.05 level) for the subfamilies/tribes Prodiamesinae, Chironomini, Tanytarsini, and Tanypodinae (Table 4.1). The number of Orthocladiinae species was not significantly different between GWD and SWD streams (p = 0.1712). Although the Orthocladiinae dominated both SWD and GWD stream communities, this subfamily made up a much larger percent of the total community in GWD streams compared to SWD streams (Table 4.1). Conversely, the percent Chironomini, Tanytarsini, and Tanypodinae were considerably greater in SWD streams compared to the GWD streams. 124 Table 4.2: Total taxa richness for chironomid subfamilies/tribes from ground- water dominated (GWD) and surface-water dominated (SWD) sites with these taxa separated into unique and shared taxa between the two stream classes. GWD SWD Subfamily/Tribe GWD SWD Total only only Shared Diamesinae 3 3 4 1 1 2 Prodiamesinae 2 2 2 0 0 2 Orthocladiinae 86 89 107 17 20 69 Chironomini 27 75 78 3 51 24 Tanytarsini 21 46 46 0 25 21 Tanypodinae 9 23 25 2 16 7 Total 148 238 261 23 113 125

Much of the difference in the total richness between the two stream classes was driven by the significantly greater number of taxa from the subfamilies/tribes Chironomini, Tanytarsini, and Tanypodinae in SWD streams. Specifically, 92 taxa from the subfamilies/tribes Chironomini, Tanytarsini, and Tanypodinae were only collected in SWD streams compared to only five taxa from these subfamilies/tribes found in GWD streams (Table 4.2). Similar numbers of Orthocladiinae taxa were unique to GWD (17 taxa) and SWD (20 taxa) streams and this subfamily had the largest number of shared taxa (69). Differences in the number of taxa from each subfamily/tribe were also reflected in the percentages of these groups from GWD and SWD streams. In most cases, taxa that were unique to one stream class or the other were uncommon, generally consisting of only a small number of specimens and/or only occurring at a single site. However, some of these taxa were abundant, and therefore merit some discussion. For example, in the GWD streams the relatively abundant unique taxa (i.e., >50 total specimens present in >1 site), Cricotopus sp. 1 (Orthocladiinae), Cricotopus sp. 6 (Orthocladiinae), Orthocladius vaillanti (Orthocladiinae), Parachaetocladius sp. (Orthocladiinae), and Stictochironomus sp. 2 (Chironomini) were only collected in GWD streams. In the SWD streams, 23 unique taxa were collected at >1 site with >50 total specimens collected including taxa from Tanypodinae, Orthocladiinae, Chironomini, and Tanytarsini. The largest number of these taxa were from the subfamily Orthocladiinae (i.e., Corynoneura sp. 2, Cricotopus sp. 8, Cricotopus sp. 9, Cricotopus sp. 12, Lopescladius sp. 1, Lopescladius sp. 2, Orthocladius nigritus, Rheosmittia sp., 125 Thienemanniella similis, and Tvetenia sp. 3) followed by the Tanytarsini (i.e., Cladotanytarsus sp. 2, Cladotanytarsus sp. 4, Tanytarsus lobiger, Tanytarsus sp. 19, Tanytarsus sp. 21, Tanytarsus sp. 22, and Tanytarsus sp. 23), the Tanypodinae (Ablabesmyia monilis, Larsia sp., Nilotanypus fimbriatus, and Procladius sp. 1), and the Chironomini (Chironomus sp. 2 and Paralauterborniella nigrohalterale).

Five of the ten most abundant taxa for both GWD and SWD streams (Corynoneura sp. 1, Cricotopus (C.) bicinctus, Parakiefferiella sp. 3, Thienemanniella sp. 1, Tvetenia sp. 1), all Orthocladiinae, were shared between the two stream classes (Appendix B). From the 60 taxa that had the greatest difference in raw abundance between stream classes, two Diamesinae (Diamesa sp. and Pagastia orthogonia) and both Prodiamesinae (Odontomesa fulva and Prodiamesa oliveri) were more abundant in GWD streams (Table 4.3a). The number of abundant Orthocladiinae and Chironomini between the two stream classes was similar, but there were considerably more Tanytarsini taxa in SWD streams (Table 4.3a & 4.3b). Different species from the genera Cricotopus, Orthocladius, Parametriocnemus, and Thienemanniella were more abundant in the two stream classes. In GWD streams, six Orthocladiinae taxa were more abundant than in SWD streams (Eukiefferiella claripennis, Limnophyes sp. 1, Parachaetocladius sp., Parakiefferiella sp. 1, Stilocladius sp., and Tvetenia sp. 2). By contrast there were seven Orthocladiinae taxa that were more abundant in SWD streams (Corynoneura sp. 3, Diplocladius sp., Lopescladius sp. 1, Lopescladius sp. 2, Nanocladius rectinervis, Rheocricotopus (P.) sp. 1, and Rheosmittia sp.). Four Chironomini were more abundant in GWD streams (Paratendipes albimanus, Polypedilum aviceps, Polypedilum laetum, Stictochironomus sp. 2). Similarly four Chironomini taxa among the top 30 taxa with the greatest differences in abundance between stream classes from SWD streams were two species of Dicrotendipes as well as Microtendipes sp. 2 and Polypedilum (U.) obtusum. Although there were several Tanytarsini taxa that were more abundant in GWD streams, they tended to be taxa that are considered cool-adapted (e.g., Micropsectra, Paratanytarsus penicillatus group). In contrast, the more abundant Tanytarsini in SWD streams were of the genus Tanytarsus and the Paratanytarsus inopertus group as well as one species of

126 Cladotanytarsus, all generally considered as lentic or common in warmer habitats. Among the most abundant 60 taxa with differences between GWD and SWD streams, the only Tanypodinae taxon, Nilotanypus fimbriatus, was more abundant in SWD streams.

Table 4.3a: Thirty taxa with the greatest abundance in GWD streams (GA = GWD site taxon relative abundance, SA = SWD site taxon relative abundance; abundance was calculated as log([GA-SA]+1) and is sensitive to the difference in raw abundance between GWD and SWD sites; relative abundance was calculated as log([GA+1]/[SA+1]) and is sensitive to relative differences in abundance between GWD and SWD sites; the complete table is presented in Appendix C). Subfamily/ Relative Tribe Taxon Abundance Abundance Diamesinae Diamesa sp. 2.74 0.93 Orthocladiinae Orthocladius (O.) obumbratus 2.69 0.98 Orthocladiinae Eukiefferiella claripennis 2.35 1.14 Orthocladiinae Cricotopus (C.) sp. 6 2.20 2.21 Prodiamesinae Odontomesa fulva 2.19 1.51 Orthocladiinae Thienemanniella xena 2.13 1.43 Tanytarsini Micropsectra polita 2.08 1.15 Orthocladiinae Parametriocnemus sp. 2 2.05 1.99 Orthocladiinae Cricotopus (C.) sp. 1 2.05 2.05 Chironomini Polypedilum (U.) aviceps 1.99 0.76 Tanytarsini Paratanytarsus pen. gr. sp. 2 1.96 1.90 Orthocladiinae Cricotopus (C.) sp. 4 1.94 0.60 Prodiamesinae Prodiamesa olivacea 1.86 1.75 Tanytarsini Micropsectra sp. 2 1.84 1.42 Orthocladiinae Parachaetocladius sp. 1.79 1.79 Orthocladiinae Tvetenia sp. 2 1.77 1.66 Orthocladiinae Orthocladius (O.) frigidus 1.76 1.52 Tanytarsini Paratanytarsus pen. gr. sp. 1 1.71 1.37 Orthocladiinae Orthocladius (O.) robacki 1.70 1.24 Chironomini Stictochironomus sp. 2 1.61 1.62 Orthocladiinae Stilocladius sp. 1.59 1.06 Orthocladiinae Parakiefferiella sp. 1 1.59 1.60 Tanytarsini Micropsectra nigripila 1.55 0.48 Orthocladiinae Limnophyes sp. 1 1.54 0.93 Orthocladiinae Cricotopus (C.) vierriensis 1.42 0.50 Orthocladiinae Orthocladius (E.) rivicola 1.40 0.42 Chironomini Polypedilum (P.) laetum 1.36 0.77 Orthocladiinae Orthocladius (S.) lignicola 1.34 0.97 Chironomini Paratendipes albimanus 1.33 0.43 Diamesinae Pagastia orthogonia 1.29 1.25

127

Table 4.3b: Thirty taxa with the greatest abundance in SWD streams (GA = GWD site taxon relative abundance, SA = SWD site taxon relative abundance; abundance was calculated as log([SA-GA]+1) and is sensitive to the difference in raw abundance between GWD and SWD sites; relative abundance was calculated as log([SA+1]/[GA+1]) and is sensitive to relative differences in abundance between GWD and SWD sites; the complete table is presented in Appendix C). Subfamily/ Relative Tribe Taxon Abundance Abundance Orthocladiinae Lopescladius sp. 2 2.48 2.48 Orthocladiinae Thienemanniella taurocapita 2.45 2.19 Orthocladiinae Orthocladius (O.) oliveri 2.41 2.11 Orthocladiinae Rheocricotopus (P.) sp. 1 2.27 2.15 Tanytarsini Paratanytarsus inopertus gr. sp. 2 2.22 2.16 Tanytarsini Tanytarsus sp. 22 2.20 2.20 Chironomini Dicrotendipes modestus/neomod. 2.13 2.07 Orthocladiinae Orthocladius (O.) carlatus 2.13 0.84 Tanytarsini Tanytarsus sp. 1 2.10 1.98 Chironomini Dicrotendipes fumidus 2.07 0.96 Orthocladiinae Orthocladius (O.) nigritus 2.05 2.05 Tanytarsini Cladotanytarsus sp. 3 2.03 0.93 Tanytarsini Paratanytarsus inopertus gr. sp. 1 1.97 1.42 Orthocladiinae Parametriocnemus sp. 4 1.92 1.80 Orthocladiinae Thienemanniella lobapodema 1.92 1.71 Orthocladiinae Diplocladius sp. 1.89 1.77 Tanytarsini Tanytarsus sp. 10 1.88 0.98 Orthocladiinae Nanocladius rectinervis 1.86 1.04 Tanytarsini Tanytarsus sp. 15 1.86 1.87 Tanytarsini Tanytarsus sp. 21 1.83 1.83 Tanytarsini Tanytarsus sp. 23 1.81 1.81 Tanytarsini Tanytarsus sp. 12 1.80 1.69 Orthocladiinae Corynoneura sp. 3 1.79 1.04 Orthocladiinae Rheosmittia sp. 1.79 1.80 Orthocladiinae Orthocladius (O.) mallochi 1.72 0.83 Orthocladiinae Cricotopus (C.) sp. 7 1.66 1.30 Chironomini Microtendipes sp. 2 1.66 1.61 Orthocladiinae Lopescladius sp. 1 1.65 1.66 Tanypodinae Nilotanypus fimbriatus 1.62 1.64 Chironomini Polypedilum (U.) obtusum 1.62 1.52

128 SUBFAMILY/TRIBE EMERGENCE PATTERNS The greatest number of taxa emerging during a single sample event was 63 taxa in Chub Creek (SWD stream) in August. This contrasts to Mill Stream where the greatest number of taxa emerging from this GWD stream during a single collection event was 22. The dates when the maximum number of taxa were emerging during a single sample event in each stream ranged widely from April through September with August the most frequent month (5 of 12 sites). Typically, emergence was greater in GWD streams during late autumn, winter, and early spring from approximately October through March (Figures 4.1a & 4.1b). From approximately April through September, the number of taxa emerging from each of the SWD streams was greater than the total from GWD streams. Much of the richness in late spring through early autumn was the result of a greater number of taxa from the subfamilies/tribes Chironomini, Tanytarsini, and Tanypodinae. The number of Orthocladiinae taxa emerging on a single sample event was variable between GWD and SWD streams although there tended to be more or as many species in SWD streams compared to GWD streams. Although there was no difference in the average number of Diamesinae taxa between the two stream classes, Diamesinae emerged for considerably longer in GWD streams. Prodiamesinae were nearly absent from SWD streams, but typically emerged through much of the spring, summer, and fall in all six GWD sites.

The durations of annual emergence periods in GWD streams were longer than SWD streams, with emergence occurring through most of the year from January/February through November/December (Figures 4.1a & 4.1b). In SWD streams emergence was generally limited to the period from March through October. In both stream types, taxa richness began to steadily increase in March with peaks in richness occurring between May and August. In eight sites, there was a mid summer decrease in richness that followed a spring peak. In all sites, the autumn decrease in richness began in August. Much of the increase in taxa richness in the spring from both stream classes corresponded to an increase in water temperatures. Similarly the decrease in richness in the autumn corresponded to a drop in temperatures.

129 Figure 4.1a: Biweekly chironomid subfamily/tribe richness and daily mean water temperature from ground-water dominated (GWD; left column) and surface-water dominated (SWD; right column) streams.

130 Figure 4.1b: Biweekly chironomid subfamily/tribe richness and daily mean water temperature from ground-water dominated (GWD; left column) and surface-water dominated (SWD; right column) streams.

131 There was variability in the time at which different subfamilies/tribes began and ended emergence at different sites; however, some generalizations could be made. The emergence of the Diamesinae was primarily limited to the period from October through May in both stream classes, with no emergence from SWD and limited emergence in some GWD streams from December through February. As a subfamily, the Orthocladiinae had the longest emergence duration in the GWD streams with emergence beginning in February and continuing through December. The emergence period for this subfamily in SWD streams was also long, beginning in April and continuing until November. The Orthocladiinae emergence period in SWD streams generally corresponded to the period from ice off to the refreezing of the stream. The Tanytarsini were the next to emerge, in March or April in GWD sites, but April in SWD sites. Most of the emergence of Prodiamesinae in GWD streams occurred from March through November. Prodiamesinae were uncommon in SWD streams and were only collected sporadically in some SWD sites from April through October. In GWD streams, Chironomini and Tanypodinae usually began to emerge in April or May. Emergence for these two groups in SWD streams started later in May from nearly all sites. In both stream types, the Chironomini and Tanypodinae generally ceased to emerge in September or October, with emergence of the Tanypodinae stopping somewhat earlier than the Chironomini. An exception to the timing for the end of emergence in Chironomini was Chub Creek where Dicrotendipes fumidus continued to emerge until December.

It should be noted that one of the SWD streams, Credit River, was somewhat unusual in that it displayed some characteristics of a GWD stream. Although Credit River froze during periods of the winter, it remained open longer than the other SWD streams sites and had the lowest thermal range (23.04 °C) among SWD streams. As a result the timing of emergence for the Orthocladiinae and Diamesinae at this site was similar to that of the GWD streams. In addition, the subfamily Prodiamesinae was more abundant in this site than in any other SWD stream. The total taxa richness in this site was also tied for the lowest richness among the SWD streams.

132 Although the dates at which emergence was initiated or terminated for the different subfamilies/tribes varied between stream classes, the DMT at which emergence took place was more similar between classes. Diamesinae emergence generally did not begin in the autumn until DMT dropped below 12 °C and did not end until DMT exceeded 12 °C the following spring. The emergence of Prodiamesinae began at about 8 °C and ended when temperatures dropped back to 8 °C. The temperatures at which Orthocladiinae began emergence in early spring differed between streams. In most streams, Orthocladiinae typically began emerging when temperatures reached 7 °C; however, in Cedar Creek and Credit River, emergence of Orthocladiinae began at temperatures of 2 °C. In both stream classes, the Orthocladiinae halted emergence at about 7 °C in the autumn with the exception of Rock Creek, Rush Creek, and Sunrise River where emergence continued until temperatures reached about 2 °C. Tanypodinae and Chironomini generally both started emergence at temperatures of about 11-12 °C, but the temperatures at which emergence was halted differed between these groups. In the Chironomini and Tanypodinae in GWD streams, emergence ended near 12 °C, whereas emergence of Tanypodinae ended earlier at relatively high temperatures of about 18 °C in SWD streams. Tanytarsini generally started emergence earlier than the Chironomini and Tanypodinae at temperatures of 6-10 °C and completed most emergence by the autumn when temperatures reached 7 °C.

COMMUNITY EMERGENCE PATTERNS Estimated Duration of Emergence Numerically the estimated emergence duration EED of taxa was on average longer in GWD streams for all subfamilies/tribes and for the total community, with the exception of the Tanypodinae (Table 4.4). However, differences between stream classes are only significant for Diamesinae (p = 0.0091) and Prodiamesinae (p = 0.0058). Differences in the total community average EED was also not significant (p = 0.1075). Diamesinae and Prodiamesinae had on average the longest mean EED in GWD streams, usually exceeding 100 days and reaching 281 days in Pine Creek. Diamesinae in SWD streams

133 and the other subfamilies/tribes from both stream classes had shorter average EEDs of about 40-80 days.

Table 4.4: Estimated emergence duration with stream class means and p- values derived from a Mann-Whitney U-test comparing GWD and SWD stream classes (DI = Diamesinae, PR = Prodiamesinae, OR = Orthocladiinae, CH = Chironomini, TT = Tanytarsini, TP = Tanypodinae; * indicates a significant difference between stream classes at the α <0.05 level). Site DI* PR* OR CH TT TP Total GWD Streams Eagle 112 190 87 51 89 66 85 Pine 281 197 87 61 73 42 80 Trout 140 190 88 41 80 28 84 Browns 112 84 59 39 37 28 53 Mill 112 105 66 48 58 21 63 Valley 122 119 115 48 110 56 106 Mean 146.62 147.40 83.67 47.94 74.59 40.17 78.50 SWD Streams Cedar 14 0 46 37 65 52 48 Chub 42 0 63 57 77 53 63 Credit 112 84 99 65 102 42 85 Rock 35 14 62 34 53 42 52 Rush 23 14 68 32 49 32 53 Sunrise 84 0 72 29 44 34 56 Mean 51.86 18.72 68.56 42.42 64.80 42.50 59.39 p-value 0.0091 0.0058 0.2971 0.2971 0.4848 0.6863 0.1075

Community Turnover In general, there was a tendency for lower similarity with increasing sample interval in SWD streams (Figure 4.2). At an interval of about 215 days most sample pairs were completely dissimilar (i.e., no shared taxa). The pattern of sample comparisons for GWD streams in many cases was U-shaped indicating that there is an initial decrease in similarity followed by an increase in similarity at longer intervals (Figure 4.2).

134

ght columns) streams. streams. ght columns) een sample pairs at different -water dominated (SWD; ri on Similarity Coefficient betw ed as differences in the Sorens -water dominated (GWD; left columns) and surface (GWD; -water dominated d Figure 4.2: Community turnover measur sample intervals for groun 135 When only adjacent samples are compared, there was not a significant difference between GWD (0.61 ± 0.03) and SWD (0.60 ± 0.02) streams (p = 1.0000). However, when the pattern of similarity for the two stream classes was compared over the course of the year, different patterns were apparent. Similarity between adjacent samples was consistent over most of the year with a drop in similarity in the autumn. Similarity was more variable in SWD streams with higher similarity during the winter and lower similarity during the spring and late summer/autumn. During the summer, both stream classes had, on average, comparable similarities between adjacent samples. When the mean similarities for adjacent samples were compared to the daily change in temperature during the period between these samples, there was a significant relationship between these measures for SWD streams (p = 0.0000) (Figure 4.4). However, this relationship was not significant for GWD streams (p = 0.3358).

Figure 4.3: Mean similarity of adjacent samples for ground-water dominated (GWD) and surface-water dominated (SWD) streams (error bars represent standard error, n =6).

136 Figure 4.4: Mean values for the Sorenson Similarity Coefficient of adjacent samples versus the mean daily change in temperature over the period between sample pairs for ground-water dominated (GWD) and surface-water dominated (SWD) streams.

DISCUSSION

RICHNESS The total taxa richness for the SWD streams (107-131 taxa) was higher than many other estimates of taxa richness for similar North American temperate streams (e.g., 84 taxa – Lenat 1987, 72 taxa - Rempel & Harrison 1987) and European temperate streams (e.g., 71 taxa – Gendron & Laville 1995). Chub Creek in particular had a large number of taxa (131), although this richness did not exceed the 143 taxa identified by Coffman (1973) in Linesville Creek. The average number of chironomid taxa collected (117 taxa) was similar to the number identified by Boerger (1981) in a Canadian brown-water stream (112 taxa). Individual site richnesses in the current study exceeded the richness for the European summer-warm lowland streams assessed by Lindegaard & Brodersen (1995) that ranged from 35-87 taxa. The taxa richness for GWD streams in the current study (54-84 taxa) was similar to that of other northern temperate GWD streams (e.g., 62 taxa - Singh & Harrison 1984, 48 taxa – Berg & Hellenthal 1992). Total richness for GWD 137 streams was also similar to the numbers compiled by Lindegaard & Brodersen (1995) for summer-cold, lowland (i.e., GWD) streams, primarily European, with most streams falling within a range of 32-81 taxa. One exception to the generalization that GWD are relatively species poor (<100 taxa) was a study cited by Lindegaard & Brodersen (1995) on the Breitenbach by Siebert (1980) which identified 134 chironomid species. However, the large number of taxa may be the result of the methodology since the study by Siebert (1980) studied emergence over the course of five years. There was also a much larger difference in the total richness between GWD and SWD streams in the present study (48 taxa) than in the review by Lindegaard & Brodersen (1995) where there was only an 18 taxa difference between these stream types. Similarly in a comparison of a SWD stream (Rohrwiesenbach) and a GWD stream (Breitenbach) in Germany, Ringe (1974) determined that there was little difference in total chironomid richness with 81 taxa in the GWD stream and 85 taxa in the SWD stream.

It is not clear what is responsible for the consistently high taxa richness in all six SWD streams compared to most other studies of similar streams. The relatively high richness values for the total chironomid community and the subfamilies/tribes may be related to the method used to measure chironomid emergence in this study. Many of the studies, although not all, based their taxa richness totals on collections of adults or larvae. Two factors permit a greater proportion of the community to be identified using the SFPE method: 1) improved taxonomic resolution and 2) a greater proportion of the community can be sampled (Coffman & de la Rosa 1998; Chapter III). Use of SFPE for biodiversity and estimates of community structure has been recommended because it provides as good as or better resolution than methods used for adults and because congeners are easier to separate than in the adults (Coffman & de la Rosa 1998). Compared to other methods such as kick netting for larvae or using emergence traps for the adults, the use of the SFPE method collects chironomids emerging from all available habitats within a study site including those that are difficult to sample (e.g., hyporheic, wood) (Coffman 1973, Ferrington et al. 1991, Chapter III). However, methodological effects do not explain why the number of taxa for SWD streams was high, whereas the corresponding totals for

138 GWD streams fit within the range of similar streams in the literature. Although this could not be assessed, Coffman (1989) argued that ecological heterogeneity, stream size, altitude, latitude, and biogeographical potential are major factors that influence chironomid richness in lotic habitats. The effect of ecological heterogeneity (e.g. Ferrington 2007) and stream size (e.g. Coffman 1989) on chironomid taxa richness has been documented.

The effects of several of the factors influencing chironomid richness discussed by Coffman (1989) are difficult to isolate (e.g., latitude, altitude) due to the interaction of multiple factors. For example, the effect of latitude can be obscured by differences in climate and biogeographic potential between sites along these gradients. As a result, latitudinal gradient assessments of aquatic insects and chironomids have been mixed, with some suggesting that taxa richness is higher than similar temperate streams and others suggesting that the richness between these regions is similar (e.g. Arthington 1990, Coffman et al. 1992, Coffman & de la Rosa 1998). The question of the influence of latitude can not be assessed without including another factor outlined by Coffman (1989): biogeographical potential or the regional species pool. This factor determines which species are present regionally that can potentially colonize a habitat and is a reflection of the biogeography and history of the group (e.g., dispersal patterns, speciation). To this list of factors that potentially influence the taxa richness in aquatic habitats should be added the effect of disturbance regime which may also play an important role in structuring communities.

The stream size, altitude, and latitude for the SWD sites in this study are similar to other studies from the northern temperate regions in North America and Europe from comparable streams (e.g. Rempel & Harrison 1987, Gendron & Laville 1995, Lindegaard & Brodersen 1995). This suggests that differences in regional species pools or ecological heterogeneity differences could be influencing the variable number of taxa in these streams; however, without comparable collecting techniques and sample regimes, comparison can only be made cautiously. Determining the role of these different factors

139 in structuring aquatic communities on a global scale remains an interesting, important, and complex question that has not been fully addressed.

There was a clear difference in the total richness of chironomid taxa between GWD and SWD streams with a significantly larger number of taxa present in SWD streams. The differences in taxa richness between GWD and SWD streams are related to the ability of warm-adapted taxa such as Chironomini, Tanytarsini, and Tanypodinae to utilize SWD habitats that still support a rich community of Orthocladiinae and some Diamesinae, several of which are timed to winter development and emergence in early spring. By contrast, the water temperatures in GWD streams are apparently not high enough to support these warm-adapted taxa and they are eliminated from the community or they do not colonize these habitats. Taxa generally considered cool-adapted such as Diamesinae and Orthocladiinae are able to utilize both SWD and GWD habitats. Despite the fact that SWD streams become very warm in the summer (i.e., >25ºC), temperatures are low during the autumn, winter, and spring which permit these cool-adapted taxa to utilize these habitats. Presumably, these taxa inhabiting thermally diverse environments that exceed thermal thresholds possess mechanisms (e.g., dormancy, rapid colonization) to survive these suboptimal conditions. The increase in taxa richness in the SWD streams was largely driven by the presence of taxa that were unique to this stream class (113 taxa) compared to the number only collected in GWD streams (23 taxa). Predictably, the majority (81%) of the taxa unique to SWD streams were from the Chironomini, Tanytarsini, and Tanypodinae. This pattern suggests that the majority of lotic chironomid biodiversity in this region could be protected by SWD streams. However, there were many taxa that occurred in both stream types but were much more abundant or restricted to GWD streams (e.g. Diamesa sp. Cricotopus sp. 1, Cricotopus sp. 6, Eukiefferiella claripennis, Orthocladius frigidus, Orthocladius vaillanti, Parachaetocladius sp., Parametriocnemus sp. 2, Tvetenia sp. 2, Stictochironomus sp. 2, Micropsectra sp. 2; see also Table 4.3a, Table 4.3b, & Appendix A). It is not clear how stable these apparently cool-adapted taxa are in SWD streams, and therefore, without additional research it is difficult to know whether these species would be supported by SWD streams only.

140

Interestingly, when unique taxa that consisted of >50 total specimens from >1 site were examined, the majority of these taxa from both GWD and SWD streams were in the subfamily Orthocladiinae. This was perhaps due to the low abundance of many of the warm-adapted, unique Chironomini, Tanytarsini, and Tanypodinae taxa in the SWD streams. Some of these warm-adapted taxa are often considered to be more lentic and their low abundances may indicate that they are more marginal in lotic habitats despite high summer water temperatures. Differences in the occurrences of these unique abundant taxa may indicate specific thermal preferences for these taxa. Although, less common and widespread taxa may also be indicative of the differences in the thermal preferences of these taxa, these abundant taxa are more likely to indicate actual thermal preferences. For example, the abundant taxa unique to GWD streams may indicate that these taxa require long periods of relatively cool water for development, whereas abundant taxa in SWD streams may indicate that these taxa require high temperatures.

Taxa richness was related to the range of annual DMT in the 12 streams studied. This relationship can be explained by the hypothesis proposed by Coffman (1989) that states that increased thermal variability or heterogeneity will result in greater taxa richness due to increased thermal niche partitioning. Ward & Dufford (1979) identified a pattern of increasing macroinvertebrate richness at sites downstream of a spring and attributed this pattern to greater thermal stability at upstream sites. A number of other studies have identified a similar pattern, although different measures of thermal regime are often used (e.g., maximum temperature, mean temperature). In a study of Ephemeroptera, Plecoptera, and Trichoptera along a longitudinal gradient, Sprules (1947) identified a correlation between the mean summer water temperature and the number of species. In the present study, although there was a significant difference between mean water temperatures in SWD and GWD streams, this difference was only 2 °C (Chapter I; Table 1.4) and it did not produce a clear relationship with taxa richness. In a study of altitudinal zonation, Ward (1986) determined that the richness of Diptera increased from upstream to downstream sites, a pattern that followed changes of increasing thermal

141 range at these sites. Jacobsen et al. (1997) determined that aquatic insect family richness increased linearly with maximum stream temperatures for a number of studies from around the world. However, when viewed at a global scale, many more factors such as biogeography and regional species pools (Coffman 1989) influence taxa richness and may explain why this pattern was observed. For example, although tropical streams have limited thermal heterogeneity, the evolutionary history of the regional species pool is different than temperate streams and permits a greater number of taxa to occupy a minimally variable habitat (Coffman & de la Rosa 1998). An impact of greater predation in tropical streams has also been proposed to sustain greater richness by maintaining low densities of aquatic insects (Stout & Vandermeer 1975, Jacobsen et al. 1997, Coffman & de la Rosa 1998). In contrast, the taxa richness in temperate streams are predicted to be driven more by thermal heterogeneity where increased annual variability in water temperature increases taxa richness rather than thermal maxima. The relationship between maximum temperature and taxa richness also breaks down for warm, stable temperate streams (e.g., geothermally influenced streams) that do not support higher richness than temperate thermally variable streams (e.g., Hayford et al. 1995).

There was some variability in taxa richness within stream classes; however, it was not clear which factors may lead to differences in taxa richness between sites. Differences in thermal range within the classes could be important; however in the GWD streams, the stream with the greatest thermal range (i.e., Mill Streams) only had four more taxa than the most homogenous stream (i.e., Eagle Creek). There are many factors beside temperature that can impact the habitat suitability for a species (e.g., substrate, discharge, food resources). There also was no consistent evidence that other factors (e.g., nutrients, substrate, and discharge) were responsible for patterns in taxa richness. Measures of nutrients, suspended solids, and substrates were not significantly different between the two stream classes (Chapter I). Some factors did differ between the classes, such as discharge and DO patterns. For example, the frequency of spates was higher in SWD streams and it is reasonable to predict that this disturbance would have an impact on the communities (Chapter I). However, one SWD site had relatively stable discharge (Cedar

142 Creek) and the taxa richness of this site fell within the range for SWD streams. Furthermore, it is not clear whether the high flows that occur at some sites represent disturbance to the communities or if the taxa present in these sites possess coping mechanisms (e.g., burrowing) that prevent impact to their populations. Patterns of DO were also different between the two stream classes with GWD streams possessing a more stable regime (Chapter I). However, DO regimes were not consistent in SWD streams and predictions regarding the relationship between DO and taxa richness were not clear. Disturbance (e.g., spates) also has an impact on the community composition of aquatic habitats by disrupting the community or populations and permitting the colonization of new species and the presence of poorer competitors (Resh et al. 1988). The effect of disturbance regimes (i.e., prone disturbance) or the disturbance history of each stream was not assessed, but they could have affected the community composition and richness in these sites. Watershed size may also influence these communities and the watershed for the SWD streams were significantly larger than the GWD streams; however, when specific examples are compared, watershed size does not appear to be a good predictor of total taxa richness. For example, the watershed size of Browns Creek (GWD) was greater than Cedar Creek (SWD) and close to Chub Creek (SWD) (Chapter I). Although there was overlap in watershed size between these streams, taxa richnesses were very different with only 78 taxa collected in Browns Creek compared to 118 taxa in Cedar Creek and 131 taxa in Chub Creek. In contrast, there is no overlap between the stream types for both thermal range and taxa richness. Furthermore, the upstream reaches of the Browns Creek watershed is SWD and it only becomes GWD as it approaches the sample site. Therefore, this would argue against watershed scale effects on patterns of local chironomid diversity, but the effect of watershed size deserves additional study. It is likely that multiple factors contributed in some way to the composition and richness of chironomids in these communities, but there was a greater, more consistent pattern between richness and thermal regime.

As predicted, taxa richness was greater in thermally variable streams; however, it would be interesting to determine whether this pattern holds when more thermally stable streams

143 (e.g., sites near spring sources, geothermally influenced streams) are included. Although the GWD streams where much less thermally variable than SWD streams, the GWD streams still had an annual range of mean daily water temperature of ≈14 °C. In general, springs are thermally stable and have low chironomid richness, and therefore would be expected to fit the model derived from the sites in this study. For example, in a review of European springs Lindegaard (1995) determined that chironomid richness for most of these springs was within 10-40 taxa. Webb et al. (1995) assessed the diversity of seven Illinois springs with thermal ranges of only ≈1-5 °C and only collected a total of five chironomid taxa. In a study of springs in the Central High Plains of the United States, Blackwood et al. (1995) determined that species diversity of chironomids was low (≤20 taxa were identfied from 92% of the springs sampled) compared to streams in Kansas. However, it possible that other factors such as substrate and regional species pools influence richness, and therefore to test the influence of thermal regime on these habitats, springs from the same region would need to be assessed. In addition, the sites selected clustered at the two ends of the model and it would be interesting to fill in the model with additional sites with intermediate thermal regimes.

In 4-5 streams there was a depression in taxa richness in the midsummer generally following a large spring peak. The reduction was especially prominent in Eagle Creek, the site with the lowest thermal range, which experienced a drop from 33 taxa in June to four taxa in July. A drop in the abundance of emerging chironomids was also observed by Ringe (1974) in a SWD stream, but a similar midsummer decrease was not observed in a GWD stream that was concurrently studied. This pattern may be related to high water or air temperatures that occur during the midsummer that exceed the thermal tolerances of many taxa and result in some form of dormancy. However, the pattern is variable and occurs at most sites in both streams types making determination of its cause difficult to ascertain in this study.

144 TAXONOMIC COMPOSITION AND SUBFAMILY/TRIBE EMERGENCE PATTERNS Five of the ten most abundant taxa, all orthoclads, were shared between the two stream classes. In comparing a GWD and a SWD stream, Rempel & Harrison (1987) determined that in the top ten abundant taxa from each stream, Corynoneura and Thienemanniella were present in both streams. Similarly in the present study, the same taxon from each of these two genera is on average abundant in both GWD and SWD streams. However, no other taxa were shared among the top ten abundant taxa in the comparison of Rempel & Harrison (1987), whereas in the present study an additional three taxa were shared (5 total) in the top ten abundant between the two stream classes. Presumably, these five taxa are eurythermic or eurytopic which would explain their abundance in both stream types and their nearly continuous emergence from spring through autumn in many of the study sites. Although Rossaro (1991) determined that eurythermic species were uncommon, it is expected that eurythermic species would be more widespread due to their ability to utilize a wider range of habitats during a greater portion of the year in temperate zones.

Overall, the taxa that were more abundant in either stream class accounted for similar proportions of the major subfamilies/tribes. Exceptions to this were the Diamesinae and Prodiamesinae species that were only represented in the top 30 most abundant taxa in GWD streams. The Tanytarsini were also better represented in the SWD streams due to the presence of several taxa in the genus Tanytarsus. The Orthocladiinae were the dominant group among the most abundant taxa for both stream classes. This suggests that there is considerable diversity in thermal preference within this subfamily that allows the maintenance of high richness in temperate streams with contrasting thermal regimes or that they are able to utilize low temperatures when they occur in SWD streams. Tanypodinae were poorly represented in both stream classes, indicating that this subfamily is not abundant or diverse in temperate, lotic habitats.

The Diamesinae and Prodiamesinae were each only represented by a few taxa. The Diamesinae had similar richness between the two stream classes, but members of this

145 subfamily had greater abundance and emerged for a longer period of time in GWD streams. Diamesinae were present in SWD streams, but with the exception of some sporadic winter emergence through holes in the ice, they were generally limited to a short period of emergence in the spring. One of the common members of this subfamily, Diamesa mendotae, requires temperatures below ≈10 °C for larval growth and development (Chapter VI) that would explain the shorter emergence period and lower abundance of this subfamily in SWD streams. Although the subfamily Prodiamesinae was largely limited to GWD streams, this taxon emerged from spring through autumn. It is not clear what lead to this pattern unless long periods of moderate temperatures are required for the growth and development of larvae.

As in many other studies of temperate lotic habitats, the Orthocladiinae were the dominate subfamily/tribe in this study (e.g., Coffman 1973, Boerger 1981, Drake 1982, Singh & Harrison 1984, Rempel & Harrison 1987, Berg & Hellenthal 1992). Orthocladiinae dominated total taxa richness in all sites from the present study with the exception of Cedar Creek. In contrast to lotic temperate habitats, this subfamily is generally less abundant or diverse in lentic and low latitude habitats (e.g., Freeman 1955, MacDonald 1956, Iovino & Miner 1970, Petr 1970, Lehmann 1979, Ferrington et al. 1993, Coffman & de la Rosa 1998). Members of Orthocladiinae are considered to be cool-water adapted (Oliver 1971, Berg & Hellenthal 1992, Ferrington 2000), and therefore were expected to be more rich and abundant during periods of low water temperature in SWD streams and to dominate in GWD streams. In this study, species of Orthocladiinae were among the earliest taxa to emerge in both GWD and SWD streams which fit this prediction. However, this subfamily had similar taxa richness between the two stream types and did not have a drop in richness in some SWD streams during the summer even when DMT exceeded 20ºC. The equal or greater number of Orthocladiinae in SWD streams compared to GWD streams is somewhat surprising, as they would be expected to have lower richness in SWD streams and especially to have reduced emergence during warm periods. Although orthoclads are considered cool-adapted, it is probably common that they continue to dominate temperate, lotic habitats during the

146 summer. For example, Coffman (1973) determined that in a stream in Pennsylvania, USA, the proportion of Orthocladiinae in individual samples was almost always at least 50%. In the present study this was true of the GWD streams, but in several of the SWD streams (e.g., Cedar Creek and Chub Creek), the Orthocladiinae were less than 50% of collection taxa richness for much of the summer. The similar orthoclad taxa richness between the stream classes is at least partly related to the fact that low temperatures are present in both stream classes for at least part of the year. In addition, Orthocladiinae occupy the widest range of habitats in the Chironomidae, including dominating arctic habitats and most temperate habitats as well as maintaining relatively high diversities in tropical habitats (Oliver 1971). Therefore, individual species within this subfamily have a wide range of thermal preferences that allow it to inhabit both cool and warm habitats, although it is much more successful at low temperatures.

In the present study, the Chironomini (+ Pseudochironomini) was the second most diverse subfamily/tribe. In general, Chironomini, including Pseudochironomini, are generally species poor in small lotic habitats, especially those at high latitudes (Harper & Cloutier 1979). As expected, the number of Chironomini collected from GWD streams were less then the number in SWD streams. The emergence timing of the Chironomini also fits predictions, with emergence starting later than most other subfamilies/tribes and with their greatest emergence occurring during the summer. However, the Chironomini are a widespread, common group that is usually diverse in temperate habitats. A total of 78 Chironomini taxa were collected in the present study with individual SWD streams supported from 22-39 Chironomini taxa. These values are similar to the number of taxa collected in some other studies. Harper & Cloutier (1979) identified 50 species from these tribes from several sites within a relatively small area of a single stream. Most of the other examples of numbers of Chironomini from streams reviewed by Harper & Cloutier (1979) are similar for other temperate lotic habitats (17-41 species). In contrast, the richness of Chironomini in SWD streams of this study is greater than most of the studies on SWD streams compiled by Lindegaard & Brodersen (1995); however, the 7-19

147 Chironomini taxa identified from GWD streams was similar to the values presented by Lindegaard & Brodersen (1995) for a number of GWD streams.

Tanytarsini had similar taxa richness to the Chironomini in both GWD and SWD streams, although overall, the tribe Tanytarsini was slightly less taxa rich than the Chironomini. This tribe is generally considered to be a warm-adapted group, although there are some Tanytarsini, such as Micropsectra, that are cold-stenothermic (Lindegaard & Brodersen 1995). This was apparent as in most streams of this study where Micropsectra, emerged early in the spring before most Chironomini, Tanytarsini, and Tanypodinae. Between the two stream classes, the Tanytarsini were on average more than twice as rich in SWD streams. As a result, richness and emergence patterns for the Tanytarsini fit predictions fairly well with lower diversity in GWD streams and the bulk of emergence occurring during summer. As with the Chironomini, there was a greater number of Tanytarsini in the SWD streams compared to similar streams compiled by Lindegaard & Brodersen (1995). Again, the richness of this group was similar in GWD streams to the values reported for other similar streams in Lindegaard & Brodersen (1995).

The subfamily Tanypodinae had emergence patterns similar to the Chironomini; however, this subfamily was considerably less taxa rich than the other warm-adapted groups, representing only 4% of GWD taxa and 9% of SWD taxa. This percent composition is low compared to the studies compiled by Lindegaard & Brodersen (1995) where mean percentages for GWD streams was 11% and 15% for SWD streams. Most of the studies reviewed by Boerger (1981) also contained a greater proportion of this subfamily. However, the taxa richness for the streams in the current study were similar to the studies compiled by Lindegaard & Brodersen (1995). Tanypodinae are most abundant and diverse in warm habitats or during periods of warm water (e.g., Stahl 1986) and tend to be taxa poor and with low densities in temperate streams. Therefore, the current research supports these patterns, as this taxon was more rich in SWD streams, species emerged at higher DMT, and the emergence of its members were largely limited to the summer. 148

Although many thermal preference patterns based on subfamily/tribe appear to hold, there are several exceptions to this generalization. For example, Micropsectra, Paratanytarsus penicillatus gr., and Stictochironomus taxa emerged at lower temperatures than the majority of taxa from the Tanytarsini and Chironomini. In the Tanypodinae, the Zavrelimyia were collected in GWD streams, and therefore at lower water temperatures. In the typically cool-adapted subfamily Orthocladiinae, the genera Rheocricotopus and Lopescladius emerged at higher temperatures than most other members of this subfamily. In addition, there were taxa within genera that differed in their thermal preference compared to other congeners. For example, in the cool-adapted genus Orthocladius, the species Orthocladius carlatus emerges on average at warm temperatures, whereas most of the other members of this genus emerge at low temperatures. Interestingly, although in Chub Creek Dicrotendipes fumidus began emergence during mid-summer at temperatures near 24 °C, emergence continued into fall and winter at water temperatures near 0 °C and did not stop until the stream froze. It would be interesting to assess the life history dynamics of some of these cool-adapted species in SWD streams to determine whether they aestivate and whether they are able to utilize the near 0 ºC water temperatures during the winter for growth and development. Conversely, the mechanisms used by warm-adapted taxa to survive during periods of low temperatures in these lotic systems also would be interesting to assess.

As with total richness values, the subfamily/tribe richness values from the current study for SWD streams were different than the studies reported by Lindegaard & Brodersen (1995) from seven SWD, lowland streams in Europe. There were two studies, both in North America, where taxa richness for subfamilies/tribes was similar to the current study (Coffman 1973, Boerger 1981). It is possible that methodological differences resulted in the discrepancy between the current study and many others in the literature. Both of the studies with similar subfamily/tribe richness values used relatively thorough methods to assess the chironomid community. Coffman (1973) used the SFPE method and the study by Boerger (1981) used a series of 16 emergence traps that were checked

149 frequently. Again, there may also be differences in the stream heterogeneity, disturbance, or differences in the regional species pools between these various studies. Although total richness values differed from some other studies, the hierarchy of richnesses of the subfamilies/tribes was similar to that of other studies in lotic habitats where Orthocladiinae > Chironomini > Tanytarsini > Tanypodinae > Diamesinae > Prodiamesinae (e.g., Lehmann 1971, Coffman 1973, Boerger 1981, Lindegaard & Brodersen 1995).

Differences in the composition at the subfamily/tribe level in the present study were related to the thermal regime. The Chironomini, Tanytarsini, and Tanypodinae are most rich at sites with greater thermal variability (i.e., SWD streams). Specifically, the water temperatures in these streams become warm enough to support these warm-adapted taxa. Conversely, the Orthocladiinae are more cool-adapted and as a result the GWD streams consist of a greater proportion of this taxon compared to SWD streams. Williams & Hogg (1988) also determined that the chironomid fauna was dominated by Orthocladiinae near the source of a spring with an annual thermal variation of 11 °C and that Tanytarsini and Tanypodinae were more abundant at downstream sites where thermal variation was greater (16 °C). Similarly, Ward & Williams (1986) determined that cooler upper reach sites along a river course were dominated by Orthocladiinae, whereas the lower, warmer reaches were dominated by Chironomini.

The differences in the composition at the subfamily/tribe level between the two stream classes further support the hypothesis that temperature has an important effect on taxa richness in these streams. In particular, the presence of a greater number of warm- adapted taxa in SWD streams was responsible for the increased taxa richness, suggesting that variation in temperature enables greater taxa richness. There is no evidence that members of the Chironomini, Tanytarsini, and Tanypodinae are more tolerant of natural factors such as flow regime variability. Stream order influences the number of chironomid taxa present in a habitat (Coffman 1989), but the streams in this study were similar in size (Chapter I), so size can be eliminated as a confounding factor. Potential

150 differences in regional species pool, which can also be an important factor determining taxa richness (Coffman & de la Rosa 1998), can be ruled out as a complicating factor in the current study.

COMMUNITY EMERGENCE PATTERNS There clearly appeared to be different thermal thresholds for the initiation and halting to emergence in both stream classes. A similar pattern also was noted by Morgan & Waddell (1961). This pattern may be related to the life history patterns of the chironomid taxa in these communities. For many taxa, low water temperatures and the initiation of dormancy may have a synchronizing effect on the cohort. For example, many spring emerging taxa may continue to grow and develop, although they are prevented from emergence in the autumn (Danks 1971a, Butler 1984, Singh & Harrison 1984). This can result from lower thermal thresholds for larval activity compared to those required to initiate pupation and emergence (Danks & Oliver 1972). Therefore, the decline in emergence in the autumn at relatively high temperatures, especially in SWD streams, may be the result of warm-adapted taxa becoming dormant. In contrast, cool-adapted taxa break dormancy at these temperatures, but do not reach maturity until the following spring. However, the effect of photoperiod or food resource patterns can not be ruled out as potentially influencing the emergence patterns of these chironomid communities.

Although the EED for most subfamilies/tribes was greater in GWD streams as predicted, these differences were only statistically significant for the subfamilies Diamesinae and Prodiamesinae. The Prodiamesinae were almost exclusively limited to GWD sites, so a greater EED between the two stream classes would be expected for this subfamily. Although Diamesinae were present in most SWD streams, this subfamily was limited to short periods of emergence in the spring and autumn, whereas in GWD streams this subfamily emerged for much of the autumn, winter, and spring. The lack of a significant difference between the other subfamilies/tribes in GWD and SWD sites may be a reflection of the considerable seasonality in GWD streams or life history characteristics of the species. Although GWD streams are relatively stable, they still undergo a

151 considerable annual swing in temperatures of on average 16 °C that could affect the growth, development, dormancy, and emergence of the taxa within these habitats. Other factors could be the presence of taxa existing under suboptimal conditions. For example, in GWD streams, there are many taxa considered to be warm-adapted, but in general, these taxa have shorter emergence periods when compared to the emergence periods in SWD streams.

The increase in taxa richness in SWD streams resulting from greater community turnover is supported by analysis of similarity between samples within a site. The decrease in similarity followed by an increase at long intervals in GWD streams is related to the fact that the late winter/early spring community is similar to the late autumn/early winter community in the same year. The patterns in turnover for both stream classes assessed in this study are different from the pattern for tropical streams from Coffman & de la Rosa (1998). This indicates that the turnover for temperate streams, even those that are buffered, do not approach the community stability of tropical streams. This is in part a reflection of the fact that water temperatures in the tropical streams studied by Coffman & de la Rosa (1998) had annual thermal ranges near 0 ºC compared to the GWD streams which had an average thermal range of 14.28ºC. Although temperature typically varies very little in tropical streams, the fluctuation of other factors can be considerable (e.g., precipitation) and some taxa may have seasonal emergence patterns (e.g., McElravy et al. 1982, Wolda & Flowers 1985, Sweeney et al. 1995).

Although the mean Sorenson Similarity Coefficient values between adjacent samples were not significantly different between stream classes, patterns in the similarity of adjacent samples over the course of a year were different between the two stream classes. Two distinct decreases in similarities were apparent in the SWD streams in the spring and late summer/autumn. In SWD streams, these decreases were significantly related to the change in temperature. As a result, the increased turnover which appears to influence total taxa richness in SWD streams, is limited to relatively short periods (i.e., spring, late summer/autumn) of the year when the rate of temperature change is greatest.

152

CONCLUSIONS

There were significant differences in the total chironomid richness between GWD and SWD streams. In addition, the number of subfamilies and tribes in these habitats differed with significantly more warm-water adapted taxa (i.e., Chironomini, Tanytarsini, and Tanypodinae) occurring in SWD streams. However, there was not a greater number of cool-water adapted taxa (i.e., Orthocladiinae and Diamesinae) present in GWD streams. Consequently, much of the difference in chironomid richness between SWD and GWD streams is a result of seasonally higher temperatures in SWD streams that permit colonization by warm-adapted taxa such as Chironomini, Tanytarsini, and Tanypodinae. The results of community emergence patterns were not as clear because the EED was not statistically significantly different between the stream classes, although there was evidence of greater community turnover in SWD streams. The lack of a statistical difference in EED between the two stream classes may be a reflection of the residual seasonality in GWD streams despite thermal buffering in these streams. The increased turnover in SWD streams provides evidence that in these more thermally variable streams, the annually fluctuating temperatures provides additional niche space which is temporary, but predictable annually, and allows more taxa to utilize the habitat at different times during the course of a year. It is likely that other factors could influence the composition and richness of chironomids in these streams, but it appears that thermal regime is the most important and consistent factor influencing community structure. Based on these conclusions, the null hypotheses are rejected, and alternative hypotheses accepted, for the first and second questions posed in this study. However, the results for the third question were not as clear because the results of two tests conflicted. There were different turnover patterns between the stream classes, with greater turnover in SWD streams during periods of higher thermal change. These results support acceptance of the alternative hypothesis; however, the EEDs for the entire community and most subfamilies/tribes was not significantly lower in SWD streams. Although differences in the EED tests were not significant, they did match the pattern as stated by the alternative

153 hypothesis. Therefore, the alternative hypothesis is tentatively accepted due to a lack of significant difference between the community EEDs for the two stream classes.

The considerable difference in mean taxa richness between GWD and SWD streams raises several implications for the impact of these natural differences on biological assessment and monitoring. Considering that these streams were selected because they represented minimally impacted sites for the region, these differences have considerable implications for the development of models, such as an Index of Biotic Integrity, to predict and assess anthropogenic impacts in this region. Although GWD streams are more stable, there is still considerable seasonality in northern temperate habitats that influences and synchronizes the emergence of chironomids from these habitats. At a broader level, the use of aquatic organisms in biological monitoring depends on an understanding of the relationships between the biota and their environment. The effects of natural factors, such as thermal regime, must be identified to separate anthropogenic stressors from environmental effects for the generation of accurate stream condition assessments. Research on how temperature and thermal regime influences aquatic communities in northern temperate streams is also important as it will improve our understanding of the effects of climate change and provide tools to model these impacts on aquatic communities. For example, increases in average temperatures and thermal ranges in GWD streams as a consequence of global climate change could result in increased diversity, which by many standards would be considered desirable. However, increased temperatures will have a negative impact on cool-adapted taxa that may rely on periods of sustained low or moderate water temperatures. Furthermore, the consequences of the loss of cool-adapted species would be an undesirable side-effect that could change food resources to trout populations in GWD streams. A better understanding of the effects of thermal regime, and specifically a changing thermal regime, on these species rich communities in conjunction with the influence of anthropogenic stressors is important if we are to conserve aquatic biological diversity and water quality for future use.

154 CHAPTER V

THERMAL PREFERENCES AND SEASONAL PARTITIONING OF CHIRONOMIDAE (DIPTERA) COMMUNITIES IN NORTHERN TEMPERATE STREAMS

ABSTRACT

Communities of Chironomidae in lotic systems tend to be highly species rich, often supporting more than 100 species within a single stream reach; however, the mechanisms that allow the coexistence of such large communities are not well understood. Temperature is considered to be an important, if not the most important, factor in regulating the richness and composition of aquatic insect communities. To assess the importance of chironomid thermal preferences in shaping these communities, the thermal preferences for 261 chironomid taxa distributed among 92 genera, and seven subfamilies or tribes were determined. Thermal preference patterns within subfamilies/tribes and genera indicated that thermal preference is relatively well conserved within higher taxonomic groupings; however, within these taxonomic groupings, there were many genera or species that did not possess thermal requirements similar to other members within their respective higher taxonomic assemblage. Taxa that differed in their thermal preference could represent a mechanism to permit the coexistence of closely related taxa. This was tested by assessing differences in thermal preferences between co-occurring congeneric taxa from small to large genera; however, there was no significant influence of the number of taxa within a genus on this measure suggesting that thermal preference differences do not reduce competition between congeneric taxa. Based on this research and other research on the community dynamics in chironomids, it appears that the large richnesses of these communities are permitted by seasonal partitioning the habitat at the subfamily/tribe level and at lower taxonomic levels by density-independent factors, such as disturbance and biotic and abiotic fluctuations that result in the community remaining in a state of nonequilibrium.

155 INTRODUCTION

Chironomid communities are commonly one of the most species rich groups of aquatic insects; however, the mechanisms that permit the coexistence of many species from the same family are not well understood. A wide range of mechanisms allowing coexistence of ecologically similar and/or closely related taxa have been suggested, including both density-independent factors (e.g., substrate, thermal, food resource heterogeneity) and density-dependent factors (e.g., competition) (Hart 1983, Coffman 1989). Density- independent factors provide a greater number of potential niche spaces within a given habitat that can be used by a greater number of species possessing different traits and requirements that allow them to fill the various niches. Factors considered density- independent can also include disturbance or environmental fluctuations that cause reductions in population size, and thereby reduce competitive exclusion and/or potentially allows the colonization of new species (Huston 1979). The influence of density-dependent factors are more difficult to measure in aquatic habitats, but when they operate as defined by Gause’s Law, they are predicted to result in the competitive exclusion of species that utilize the same resources, and thereby decreases taxa richness. However, the factors that influence community composition are probably a combination of these types of factors, such that the dimensions of a species’ niche is both determined by the physical environment and interactions between species (Connell & Orias 1964). The challenge is then to determine the relative importance of different abiotic and biotic factors that influence the structure, composition, and taxa richness in aquatic insect communities, particularly for groups such as the Chironomidae where there are often many co-occurring species present.

Temporal, spatial, and behavioral separation of the utilization of shared resources is commonly suggested as a mechanism to aid in the coexistence of ecologically similar aquatic insect species including taxa that are systematically closely and distantly related taxa (e.g., Ide 1935, Britt 1962, Cummins 1964, Sheldon & Jewett 1967, Madsen 1968, Grant & Mackay 1969, Coleman & Hynes 1970, Harper & Pilon 1970, Mackay 1972,

156 Benke & Benke 1975, Cather & Gaufin 1976, Oswood 1976, Iversen 1978, Johannsson 1978, Merritt et al. 1978, Hildrew & Edington 1979, Vannote & Sweeney 1980, Sweeney & Vannote 1981, Brittain 1982, McElravy et al. 1982, Georgian & Wallace 1983, Teague et al. 1985, Morin & Harper 1986, Elliott 1987, Haro & Wiley 1992, Keiper et al. 2002). Some papers have also attributed observed differences in population sizes among coexisting taxa to competition effects (e.g., Istock 1966, Istock 1973, Benke 1978). In many cases, temperature tolerance and preference is implicated in causing the temporal and spatial separation of ecologically similar taxa although other differences such as substrate, flow, and food resource preferences also are commonly implicated.

There are several lines of reasoning that may explain temporal and spatial differences in emergence or larval activity between ecologically similar taxa if competition is responsible for these relationships and patterns (Butler 1984). First, these differences may be the result of coevolution where competition over the evolutionary history of these taxa has lead to shifts in their life cycle or preferences which reduce competition or niche overlap. Second, the differences in life histories may simply be the result of the evolutionary histories of species. In this case, species have evolved there own requirements and preferences independently and their coexistence is simply fortuitous as it permits their coexistence. Taken a step further, this could indicate that communities are shaped by the colonization of species from the regional species pool that is then filtered to a subset of taxa which minimize niche overlap. Third, these taxa may be relatively labile in their life histories or preferences that results in shifts of these characteristics when the taxa co-occur. A final reason pointed out by Butler (1984) is that these patterns could simply be random, and therefore the result of biologists attempting to attribute reasons to observed patterns. This problem argues for the need to perform rigorous manipulation and statistical hypothesis testing to assess coexistence among aquatic insect (Butler 1984, Tokeshi 1986b).

Other studies have determined that considerable niche overlap was present in some coexisting congeneric taxa but no hypothesis regarding their coexistence was obvious

157 (e.g., Hildrew & Edington 1979, Tavares-Cromar & Williams 1997, Alverson et al. 2001). The lack of apparent differences in the niche space or considerable niche space overlap does not necessarily indicate that there is none, but that these differences may have been missed or are too subtle for human observation to detect (Hynes 1970). However, there is another explanation for the apparent ability of ecologically similar taxa to coexist, which emanates from the question of whether or not competitive exclusion or Gause’s Law is a major or even significant driver of the composition of lotic communities. Streams and rivers can be extremely variable in a number of factors that have a strong impact on aquatic insects such as temperature, discharge, and food resources. This variability can theoretically provide differing levels of disturbance (e.g., high flows) that would be expected to result in the reduction of population densities of many species. Periodic reductions in densities of populations due to environmental fluctuations and disturbance would be expected to prevent or reduce competitive exclusion. Therefore, due to the dynamic nature of streams these lotic communities may be in a state of nonequilibrium that would allow species that compete for the same resource to coexist.

Among studies on aquatic insect coexistence, there is a prevalence of research on Ephemeroptera, Odonata, Plecoptera, and Trichoptera with a smaller number of studies on aquatic Coleoptera, Hemiptera, and Diptera. Compared to the importance of chironomids in aquatic systems, only a disproportionately small number of studies have been carried out for this group. Many chironomid communities are very taxa rich with >100 taxa commonly occurring within a single stream reach (Chapter IV). The number of chironomid taxa emerging over a short period of time (i.e., ≈2 days, as determined with SFPE samples) from streams in Minnesota often exceed 30 taxa during portions of the year, but can as high as 63 taxa on a single collection date (Chapter IV). Collections of larvae by de la Rosa (1985) determined that the number of larvae present during a collection event in Linesville Creek ranged from 71-84 taxa. Although the number of potentially interacting species in temperate streams can be high, the number of co- occurring taxa may be considerably higher in tropical streams as a result of reduced

158 seasonality. For example, Coffman et al. (1992) determined that 174 chironomid taxa emerged over a 12 hour period from a river in Guinea.

Although large, coexisting communities of chironomids are often present in a single stream reach, there are few studies assessing how species rich communities, often containing many abundant species, are supported. As a result of the large number of co- occurring chironomid taxa, there are a number of resources in which chironomids could potentially be in competition (e.g., food, habitat). Many chironomids appear to be generalists (i.e., collector-gatherers) (Pinder 1986, Tokeshi 1986b, Tokeshi 1992a, Tavares-Cromar & Williams 1997, Ferrington et al. in press) and peaks in the abundance of chironomids are often associated with periods of increased food resources (McLachlan et al. 1978, Tokeshi 1986b, Pinder 1992, Tokeshi 1992a). Chironomids may also compete for habitats or substrates. In the present study, some sites had relatively homogeneous substrates, although a large number of taxa were supported were supported in these habitats. For example, a site dominated by soft sediment (Cedar Creek) had the 4th highest taxa richness. Without assessing specifically the location of the larvae for taxa within these habitats, the degree of habitat overlap can not be determined; however, the fact that taxa rich communities occur in relatively homogenous stream reaches suggests that considerable habitat overlap could occur in this community. Due to the common occurrence of a large number of coexisting chironomid species within sample reaches and similarities in feeding habits, large proportions of chironomid communities could be in competition for the same food resources and habitats. However, if the competition for these resources is strong enough and these resources are limiting, shifts in thermal preferences could allow taxa to utilize habitats at different times, thereby reducing competition and allowing a greater number of taxa to utilize the same habitat.

Temporal separation has been identified in the Chironomidae suggesting that this mechanism may allow some species to coexist (e.g., Heuschele 1969, Cannings & Scudder 1978, Butler 1980, Boerger 1981, Lenat & Foley 1983, Goddeeris 1987, Johnson & Pejler 1987, Berg & Hellenthal 1992). Some research on chironomids has assessed

159 competition among species in relatively depauperate communities or subsets of communities (e.g., Tokeshi & Townsend 1987, Tokeshi 1992a, Tavares-Cromar & Williams 1997). Studies on the taxonomically rich chironomid community in Linesville Creek are an exception to this pattern (e.g., 154 taxa - de la Rosa 1985). This research determined that when the trophic role, temporal distribution, and spatial distribution were determined for species and pairs of these species were compared, fewer than 1% of species pairs were determined to be in potential competition (de la Rosa 1985, Coffman & de la Rosa 1998).

The thermal regime to which an aquatic insect species has been exposed over its evolutionary history determines its response to its environment and these differential responses by the species in a regional species pool determine dynamics at the population and community levels (Ward & Stanford 1982). Therefore, a mechanism to allow different species to utilize similar resources in a temperate stream may be the evolution of different thermal preferences or emergence optima. Evolutionary changes in the physiology of an aquatic insect (i.e., changes in enzymatic activation thresholds) could allow a species to utilize a habitat or resource that is underused. In lotic habitats within temperate regions, a greater diversity of chironomids is achieved in more thermally variable streams (e.g., surface water dominated streams [SWD]), because habitats with greater thermal heterogeneity can support species with different thermal optima at different times of the year (Chapter IV). Thus temporal partitioning would be expected to limit competition among ecologically similar taxa if they possess different thermal requirements. Closely related taxa (i.e., congeners) are more likely to be ecologically similar, and therefore in direct competition. Therefore, if competition plays a role in structuring the diverse communities of chironomid in lotic systems, it can be predicted that through evolutionary changes, a species could alter its thermal optima (i.e., the trait is plastic or labile) such that species from the regional species pool with lower overlap of thermal preference would be more likely to coexist. If competition occurs on a temporal scale then it would be expected that congeneric taxa would have greater thermal differentiation between taxa in genera with a greater number of co-occurring congeners.

160 In the present study, sample sites contained many co-occurring taxa within the same genus (e.g. >3 congeneric taxa), including Cedar Creek which supported 15 Tanytarsus taxa. The occurrence of genera with many co-occurring taxa raises the question of how these taxa occupy the same habitat.

In a previous chapter (Chapter IV), it was determined that chironomid taxa richness was positively related to increased thermal variability. To explain and test a possible mechanism behind these community richness patterns the following two questions were asked: (1) Are thermal preferences of chironomid taxa conserved among higher taxonomic groupings or are they flexible, and (2) Are closely related species (i.e., congeners) in large genera more likely to be temporally separated through emergence at different temperatures than genera where only a small number of species co-occur? These questions are relevant because temperature is considered to be one of the most important factors determining the timing of the life history of aquatic insects. Furthermore, chironomids provide an excellent opportunity to test this hypothesis, because they possess a wide range of thermal preferences and are usually represented by many species, including differing numbers congeneric taxa, within a single habitat. Some studies have also determined that chironomid communities are one of the taxa most affected by modified thermal regime (e.g., Saltveit et al. 1994, Hogg & Williams 1996).

METHODS

STUDY SITES, SAMPLING, AND LABORATORY PROCESSING See Chapter I for a description of the 12 study sites from which the data in this chapter were derived, the methods used to collect and process SFPE samples, and the methods used to measure thermal regime in these sites.

161 DATA ANALYSIS Thermal preferences The mean daily water temperature for the days that coincided with sample events when SFPE were collected (i.e., 312 samples) from all 12 sites was determined. Thermal preferences for each operational taxonomic unit (OTU; i.e., species or morphospecies), genus, and subfamily/tribe were estimated by determining the weighted mean daily water temperature at which taxa were collected. This measure can also be referred to as the

cumulative temperature at which 50% the individuals from a taxon emerged (ET50).

Thermal preferences or ET50 for each taxon were calculated using the following equation from Rossaro (1991):

∑ni ⋅Ti ET50 = ∑ni

where ni is the abundance of a taxon in a sample and Ti is the mean daily water temperature for the day that sample was collected. Violin plots of temperature versus relative abundance were created for subfamilies/tribes and genera using NCSS© (Hintze 2001) to show relative emergence abundance at different temperatures. A violin plot is essentially the combination of a box plot and a histogram that allows examination and comparison of the mean, spread, and distribution of the data for a number of groups (Hintze 2001). For comparisons at the subfamily/tribe level, the tribe Pseudochironomini was included with the Chironomini because only a single species was collected in this study.

The thermal preferences of subfamilies and tribes were assessed using a phylogeny proposed by Sæther (2000). The placement of the Telmatogetoninae is uncertain and this subfamily is also unusual as it is a primarily marine taxon. Therefore, the thermal preferences of the Telmatogetoninae are not considered. The thermal preferences of chironomid genera were also assessed phylogenetically by identifying cool-adapted and warm-adapted taxa from a phylogeny of the Chironomidae proposed by Sæther (1977). Phylogenies within the Tanypodinae, Diamesinae, and Prodiamesinae were not available, so these taxa are plotted without their relationships resolved. Taxa with thermal

162 preferences below 15 °C were considered cool-adapted taxa and taxa with thermal preferences 15 °C or higher were considered to be warm-adapted. Only genera with more than 10 individuals collected in the study were included in the phylogenetic trees to avoid misclassification of the thermal preferences based on only a few specimens. With taxa that make up only a small percent of the collections, it is difficult to determine whether they are truly rare, or whether the sites from which they were collected represent marginal environments, and therefore are not characteristic of their thermal preferences.

Thermal partitioning Differences in thermal preference ranges for co-occurring taxa in genera of different sizes were tested by determining mean thermal preference ranges for all genera with >2 co- occurring taxa. By comparing mean thermal preference ranges for differently sized genera, it can be determined whether in larger genera there was increased thermal variability or separation between taxa. This analysis was performed by determining the thermal preference range for all taxa pairs within a genus of co-occurring taxa. For example, a genus of two co-occurring taxa, only one pair was compared, but in a genus of five co-occurring taxa, ten pairs were compared. The ranges for all taxa pairs within a genus were then averaged. The ranges of thermal preferences for differently sized genera were also calculated using only taxa that contributed at least 0.5% of a site’s total abundance. Box plots were created to visualize differences in thermal preference ranges for genera of different sizes in SigmaPlot©. To test for differences between the thermal ranges of differently sized genera, a one-way analysis of variance (ANOVA) was used in NCSS© after applying a log(x+1) transformation. If significant differences were identified, a Tukey-Kramer multiple comparison test was applied using NCSS© to test for differences in the mean thermal preference ranges between genera

163 RESULTS

THERMAL PREFERENCES The thermal preferences for a total of 261 OTUs, 92 genera, and six subfamilies/tribes were determined. From these 92 genera, 42 possessed two or more congeneric taxa ranging from 2-20 taxa although, many congeneric taxa did not co-occur at the same locality. The mean emergence temperature for the subfamilies and tribes indicated a steady increase of mean thermal preference from more primitive to more derived taxa with the exception of the Tanypodinae which are considered to be the most primitive subfamily/tribe collected in this study (Figure 5.1). Diamesinae on average emerged at the lowest temperatures (6.32 °C) followed by the Prodiamesinae (11.22 °C) and the Orthocladiinae (13.71 °C). The Tanytarsini and Chironomini had similar thermal preferences of 15.78 °C and 16.03 °C respectively. The Tanypodinae had the highest thermal preference of 18.37°C. Diamesinae did not emerge when temperatures exceeded ≈17 °C. In contrast, Tanytarsini and Tanypodinae did not emerge at temperatures below ≈4 °C and ≈7 °C respectively. The Orthocladiinae and Chironomini were collected at temperatures ranging from 0-26 °C, which represents the range of mean daily water temperatures measured during sample events in this study.

Although there was considerable overlap in the thermal ranges of subfamilies/tribes that were collected, there was separation between many subfamilies/tribes where the greatest relative abundance (i.e., the widest portions in the violin plots) occurred (Figure 5.1). For example, Diamesinae and Tanypodinae separated from the other groups on either end of the thermal scale. Prodiamesinae and Orthocladiinae were similar in the temperatures where the bulk of their emergence occurred as were the Tanytarsini and Chironomini; however these two groups differed from each other.

164

An examination of thermal preferences at the species level indicated that the largest number of taxa had thermal preferences or ET50 values between 16 and 18 °C, whereas only a few taxa had thermal preferences below 8 °C or above 22 °C (Figure 5.2). Both species of Prodiamesinae had thermal preferences between 10 and 12 °C whereas the four Diamesinae taxa were spread out between 6 and 18 °C. The maximum number of Orthocladiinae (24 taxa) had thermal preferences between 10 and 12 °C. Tanytarsini thermal preferences were higher than the Orthocladiinae with maximum richness (22 taxa) between 16 and 18 °C. Both the Chironomini (23 taxa) and Tanypodinae (10 taxa) had maximum richnesses between the temperatures of 18 and 20 °C.

165 Figure 5.2: Subfamily/tribe richness at 2 °C intervals for the calculated thermal preferences of 261 taxa from 12 streams.

Based on Brundin (1966) the ancestor of the Chironomidae evolved in cool, flowing waters. Assumptions of thermal preferences at the subfamily/tribe level for the Chironomidae indicate that most (i.e., 8 of 10 subfamilies excluding the Telmatogetoninae) of these higher taxa are also cool-adapted (Figure 5.3). At the subfamily level there appears to be two major shifts from cool-adapted to warm adapted with the common ancestors of the Tanypodinae and the Chironominae.

Using 15 °C as a threshold for cool-adapted and warm-adapted taxa, 31%, 0%, 0%, 27%, 12%, and 17% of genera differed from the thermal preferences of their respective subfamily/tribe for the Tanypodinae, Diamesinae, Prodiamesinae, Orthocladiinae, Tanytarsini, and Chironomini, respectively. For species, 24%, 25%, 0%, 35%, 22%, and 21% of taxa differed from the thermal preferences of their respective subfamily/tribe for the Tanypodinae, Diamesinae, Prodiamesinae, Orthocladiinae, Tanytarsini, and Chironomini, respectively. On average, 22% of genera differed from the subfamily/tribe thermal preference and 27% of species deviated from the subfamily/tribe thermal

166 preference. The number of taxa within genera with two or more taxa that differed from other members of the genus comprised 38 out of 212 taxa or ≈18% of congeneric taxa.

In general, genera that differed in their thermal preferences from most other taxa in their subfamily/tribe were scattered among other groups (Figures 5.4a, 5.4b, 5.4c; Appendix E, Appendix F) revealing no phylogenetically consistent patterns to explain these preferences. For example, genera that differed in their thermal preferences from their higher groupings were scattered and in general were not considered to be closely related by Sæther (1977). Patterns within genera were also similar with the majority of congeneric taxa possessing similar thermal preferences, but taxa that differed in their thermal preference were not uncommon (Appendix D, Appendix E).

167 Figure 5.4a: Phylogenetic patterns of thermal preferences for genera in the Tanypodinae, Diamesinae, and Prodiamesinae with subfamily/tribe relationships for the Orthocladiinae and Chironominae from a phylogeny is modified from Sæther (1977) (genera in grey text are “cool-adapted” with thermal preferences below 15 °C, genera in black text are “warm- adapted” with thermal preferences above 15 °C).

168 Figure 5.4b: Phylogenetic patterns of thermal preferences for genera in the Orthocladiinae from a phylogeny is modified from Sæther (1977) (genera in grey text are “cool-adapted” with thermal preferences below 15 °C, genera in black text are “warm-adapted” with thermal preferences above 15 °C; dashed lines indicate uncertain placement). 169 Figure 5.4c: Phylogenetic patterns of thermal preferences for genera in the Chironominae from a phylogeny is modified from Sæther (1977) (genera in grey text are “cool-adapted” with thermal preferences below 15 °C, genera in black text are “warm-adapted” with thermal preferences above 15 °C; dashed lines indicate uncertain placement).

170 THERMAL PARTITIONING Genus sizes for the total study ranged from a single taxon to 20 taxa. The number of co- occurring congeneric taxa ranged from 2-15 taxa with 183 genera or ≈80% of the multi- taxa genera containing only 2-4 taxa. However, there were still 47 genera with five or more co-occurring taxa including 14 genera with nine or more congeners. Most of these large genera consisted of groupings of taxa in the genera Cricotopus, Orthocladius, Polypedilum, and Tanytarsus. Thermal preference ranges of co-occurring congeneric taxa were as low as 0 °C, whereas the largest was 15.05 °C.

Mean thermal preference ranges for different genus sizes were not significantly different when all taxa were included (F = 0.62; df = 10, 221; p = 0.7992) (Figure 5.5). Similarly, when rare taxa (i.e., <0.5% of total site abundance) were removed, mean thermal preference range was not significantly difference between genus sizes (F = 1.54; df = 5,

171 75; p = 0.1889) (Figure 5.6). There was possibly a slight trend of increasing mean thermal preference range as temperature increased, particularly in the treatment where rare taxa were removed; however, for both treatments, there was considerable variation in this measure in genera with few taxa. In addition, due to a lack in thermal preference differences in co-occurring congeneric taxa, temporal separation of these closely related taxa was not common (Appendix D).

DISCUSSION

THERMAL PREFERENCES The Chironomidae are hypothesized to have originated in cool, running waters (Brundin 1966), but have radiated into a wide range of environments that required adaptation to different thermal regimes. The order of thermal preference for the most common subfamilies/tribes from cool-adapted to warm-adapted is: Diamesinae → Prodiamesinae

172 → Orthocladiinae → Tanytarsini → Chironomini → Tanypodinae, which fits with most predictions of the thermal preferences in these groups (Lindegaard & Brodersen 1995). Thermal preferences of most chironomid taxa at lower taxonomic levels (i.e., species and genus) are similar to preferences at subfamily and tribe levels. Within higher taxonomic groupings there are scattered genera and species that do not conform to the thermal preferences of their respective higher taxa. For example, the orthoclads, Lopescladius and Thienemanniella, have thermal preferences above 15 °C and, similarly, Stictochironomus and Micropsectra from the Chironomini and Tanytarsini have thermal preferences below 15 °C.

Thermal preference patterns within genera were also similar with most congeneric taxa (82%) possessing similar thermal preferences to congeners; however, there are many examples of species or morphospecies with different thermal preferences than other co-

occurring congeners. For example, Orthocladius carlatus has a higher ET50 than other members of the genus. Although apparent differences in thermal preferences were apparent, assessment of co-occurring congeneric taxa generally do not reveal considerable differences in timing of emergence that would suggest that the emergence of these closely related taxa is partitioned temporally (Appendix D). In addition, species groups of Paratanytarsus emerge at different temperatures; however, they do not experience a large degree of site overlap (i.e., P. penicillatus gr. species are generally found in GWD streams, whereas P. inopertus gr. species are founds in SWD streams). Rossaro (1991) also determined that chironomid species within a genus or genera within a subfamily can have different thermal preferences than other members from their higher taxonomic grouping. Due to differences in spatial and temporal occurrence of some taxa, it is not possible to determine the influence of other factors that may be driving the occurrence and abundance of these taxa (e.g., substrate or food resource differences between sites).

However, there is considerable evidence that thermal preferences at the subfamily/tribe level are responsible for differences in emergence times and spatial distribution of

173 chironomid taxa. This study and many others (e.g., Coffman 1973, Berg & Hellenthal 1992, Lindegaard & Brodersen 1995, Coffman & de la Rosa 1998) have identified patterns of shifting subfamily/tribe composition both seasonally and spatially (e.g., along the river continuum) in the Chironomidae that are correlated to changes in temperature. Therefore, it does appear that differences in the thermal preferences of Chironomidae at higher taxonomic levels do allow taxa to utilize these habitats at different times and permits a greater number of taxa to utilize these habitats. In contrast, at lower taxonomic levels (i.e., genus and species) thermal preferences are more conserved, suggesting that there is less opportunity to reduce competition through modification of life histories or by possessing different thermal optima at these taxonomic levels.

The ability of species rich communities of a single taxon to co-occur is a reflection of changes that have occurred to the common ancestor of this group (Webb et al. 2002). As a result, the combination of phylogenetic and ecological information can provide insights into both the evolution and ecology of these taxa. In the Chironomidae, many of the genus- and species-level thermal preferences appear to have existed for a long period of time and are therefore relatively conserved; however, at lower taxonomic levels there is still a considerable subset of taxa that differ in their thermal preferences with their respective subfamily/tribe. In an analysis of the evolutionary liability of 20 traits in North American aquatic insects, Poff et al. (2006) determined that thermal preference was the second most labile trait measured among all aquatic insect orders. Although in the present study thermal preferences were measured differently than in Poff et al. (2006) and thermal preference plasticity was not compared with other traits, the results of the present study indicate that this may also be true for the Chironomidae. There also generally were not consistent patterns in thermal preferences in the phylogeny of the Chironomidae (Figures 5.3, 5.4a, 5.4b, 5.4c), which suggests that shifts in thermal preferences are relatively common. Care must be exercised when interpreting phylogenetic and ecological relations as a phylogenetic tree only represents a hypothesis that requires further testing (Tokeshi 1991a). Furthermore, the estimates of thermal preferences are derived from only a small subset of the total chironomid diversity. For

174 example, the genus-level patterns determined in the present study are only representative of the species that occurred in these 12 streams, which only includes a small fraction of the global richness. There are also a number of taxa for which species-level identifications could not be made or taxa where a species name could not attached. In both cases, it is possible that these taxa represent multiple taxa that could potentially possess different thermal preferences that in turn would lead to errors in the determination of thermal preferences. However, it is assumed that in general the use of SFPE provide as good or better resolution than the larvae and adults (Coffman 1973) that should reduce these errors even if species-level names can not yet be placed with many taxa.

THERMAL PARTITIONING The mechanisms that permit the coexistence of many chironomid species in lotic habitats are not well understood. For example, the number of taxa at individual sites in this study ranged from 54 to 131 emerging over the course of a year (Chapter I). There appears to be considerable temporal separation in the emergence of these communities as a result of the seasonality in these systems, which allow taxa with different thermal optima to utilize the stream at different times of the year. Seasonal shifts in the composition of the community of emerging chironomids suggest that differences in thermal optima or preferences plays a role in allowing different taxa to utilize the same habitat at different times in the year (Chapter IV). Because closely related taxa are expected to be more likely to compete for similar resources, temporal separation through different thermal preferences could help to reduce competition. Although thermal preferences within lower taxonomic units were relatively well conserved compared to higher taxonomic groupings, there were differences in the preferences of many species/morphospecies and genera. These differences may enhance the ability of closely related genera to coexist as these thermally divergent taxa would be predicted to have reduced competition. However, there was not a significant difference in the thermal preferences of co- occurring congeneric taxa within small versus large genera. In addition, there was considerable overlap in the emergence periods of many of the closely related taxa which

175 occurred in the same stream. In most cases multi-taxa genera included rare taxa (i.e., <0.5% of the total abundance for a site). As a result of their low abundance, these taxa may represent marginal or rare taxa that do not reach abundances large enough to exert competitive pressure on congeners. It is also possible that their low abundance is a result of considerable ecological overlap with another taxon that is a competitor. The use of rare taxa in this analysis may result in a reduced difference in the measured thermal separation between co-occurring congeners due to their limited influence on congeners. Therefore, comparisons of the thermal ranges within co-occurring congeneric taxa was repeated with taxa comprising >0.5% of a site’s total abundance. Although a pattern of increasing thermal range with an increasing number of congeneric taxa was suggested, this relationship was also not significant. This lack of difference in thermal preferences and timing of emergence suggests that thermal preferences do not have a large impact on reducing competition between closely related taxa.

Although it can be argued that the use of emergence data may be misleading, because the dynamics of the larvae are not assessed, it can be assumed that the emergence of aquatic insects generally corresponds to the greatest drain on food resources by the late instar larvae (Crowley & Johnson 1982). Although the Chironomidae possess mechanisms for dormancy, there is little evidence that dormancy in the pupal stage is widespread. In addition, because most growth in chironomids occurs during the 4th instar, the period near emergence is likely to be the time when these larvae are exerting the greatest impact on the environment and community. Therefore, in general the conditions that are appropriate for the development of the larvae are probably similar to those necessary for the initiation of pupation and emergence. Some species show different thresholds for larval development, pupation, and emergence, usually with higher temperatures required for pupation and emergence (Chapter II), but these differences are typically not large. Photoperiod can also influence the development of the various life stages differently (Chapter II). However, the effects of thermal and photoperiod induced halts most likely occur during periods when conditions are suboptimal for the development of the larvae. As a result, we can assume that the emergence of chironomids most likely corresponds to

176 periods of development for the larvae and specifically periods when late instar larvae are exerting the greatest consumptive pressure on food resources. Therefore, the use of emergence data probably does not overestimate the number of potentially interacting larvae. In fact, the use of emergence data is probably a conservative measure of the number of species that are present as larvae and potentially interacting at the time of sampling. Therefore, because SFPE samples commonly contain more than 30 chironomid taxa, it can assumed that in these streams it is not uncommon for the larvae, and specifically the late instar larvae, for 30+ taxa to be utilizing resources at the same time. This was especially true in SWD streams that tended to have the greatest richness during the summer and the greatest single sample richnesses (Chapter IV). This large number of co-occurring taxa raises questions regarding the mechanisms that allow many species, including closely related species, from the same family to utilize the same habitat concurrently.

There are a number of studies that have identified differences in the emergence patterns of closely related chironomid species. For example, Boerger (1981) reported that species of Tanytarsus (3 species), Cricotopus (3 species), and Polypedilum (2 species) displayed distinct and separate emergence times; however, Limnophyes (2 species) and Stempellinella (3 species) had overlapping emergence periods. To explain this pattern, Boerger (1981) identified a relationship between the size of mating swarms and the co- emergence of congeneric species where those taxa that formed large mating swarms emerged at different times and those with small mating swarms co-occurred. Therefore, Boerger (1981) hypothesized that sexual isolation was the cause of different emergence timing rather than interactions of the larvae. Butler (1980) hypothesized that differences in emergence times for synchronously emerging, surface-mating chironomids in arctic ponds was the result of coevolution to increase reproductive isolation. A number of other studies have suggested that differences in the timing of larval development and/or adult emergence were the result of competitive interactions of the larvae (e.g., Heuschele 1969, Lenat & Foley 1983, Goddeeris 1987, Johnson & Pejler 1987, Berg & Hellenthal 1992). However, the influence of competition among larvae is difficult to determine and it was

177 not directly measured in these studies. In some cases, there was evidence that competitive interactions impact species because differences or shifts in life histories occur between sites with and without the presence of one or the other taxon. For example, in British Columbia, Cannings & Scudder (1978) determined that several Chironomus species from saline lakes had different emergence patterns depending on the presence of other congeneric species. Cloutier & Harper (1978) also determined that the emergence patterns of Arctopelopia flavifrons and A. americana differed between sites where they coexisted and sites where only one of these congeners was present.

Studies directly assessing the competition of chironomids are more common in lentic habitats. In contrast to lotic systems, permanent lentic habitats are more stable and homogeneous that would be predicted to permit greater competition and competitive exclusion among chironomid species. For example, Ramcharan & Paterson (1978) concluded that most of the abundant taxa in a bog lake were ecologically isolated. Cantrell & McLachlan (1977) determined that competitive interactions between Chironomus plumosus and Tanytarsus gregarius combined with the environmental preferences of the two species were responsible for the distribution of Tanytarsus gregarius in a lake. Despite a lack of spatial separation, Rasmussen (1984, 1985) determined that Chironomus riparius and Glyptotendipes paripes apparently avoided competition through different feeding modes where Chironomus riparius fed on the substrate surface, whereas Glyptotendipes paripes filtered particles from the water column from within its tubes.

Coexistence between ecologically similar species in lakes may also be the result of co- occurring taxa possessing different advantages. Rychen Bangerter & Fischer (1989) hypothesized that the coexistence of Chironomus plumosus and Chironomus nuditarsis was the result of a greater competitive advantage for larval Chironomus plumosus that was offset by the longer reproductive period due to the shorter photoperiods necessary to initiate dormancy in Chironomus nuditarsis. Johnson & Pejler (1987) also hypothesized that in Lake Ekrem, the life histories for populations of Chironomus plumosus and

178 Chironomus anthracinus were altered through competition and that coexistence between these sympatric species was a result of predation pressure. Pool size was determined to control biotic interactions between two chironomid species inhabiting rock pools in Australia (Jones 1974). In small pools where food resources were limited, predation of Allotrissocladius sp. on early instar larvae of Paraborniella tonnoiri was hypothesized to be responsible for the absence of the later from small pools.

An explanation for the coexistence of sympatric taxa in some studies of aquatic insects is the utilization of different food resources (e.g., Mackay 1972, Keiper & Foote 2000); however, this mechanism has not commonly been identified in chironomids (however see Rasmussen [1984] and Rasmussen [1985]). Perhaps this is a result of the fact that many chironomids appear to be generalists, utilizing similar resources and possessing flexible feeding habitats (Pinder 1986, Tokeshi 1986b, Henriques-Oliveira et al. 2003). For example, in an investigation of the diets of a lotic chironomid community, Tavares- Cromar & Williams (1997) determined that detritus was the dominant food in the environment and in the guts of chironomids, including taxa considered to be grazers. Tokeshi (1986b) identified considerable overlap in the utilization of available food resources by epiphytic chironomids possibly resulting from the occurrence of high food availability. McLachlan et al. (1978) also determined that coexisting species fed on similar food types, although it was determined that different species and instars selected food differently. Some partitioning many occur as a result of varied food preferences between the different larval instars that reduce both interspecific and intraspecific competition (McLachlan et al. 1978, Pinder 1986). There is probably some partitioning of food resources that promotes species coexistence among taxa with very different food resource preferences and feeding methods (e.g., collector-gatherers versus collector- filterers) (Tavares-Cromar & Williams 1997). A species’ feeding guild may also change as a result of differences in the food sources available (de la Rosa 1985). It has also been shown that the flexible feeding strategies allow some coexisting and ecologically similar species to utilize similar resources when food is abundant while shifting to different food sources when resources are scarce (Townsend & Hildrew 1979). In addition, the patch

179 arrangement of resources may also influence the amount of competition. Silver et al. (2000) determined that the dispersion of patches reduced local densities of Chironomus riparius and potentially reduced intraspecific competition. Differences in substrate preferences have also been implicated in the separation of niche space for aquatic insects with similar ecological requirements (Rae 1985, Rae 1987, Schmid 1992, Crosa & Buffagni 2002). Discharge can also be a factor related to the number of chironomid species present in streams (Rae 1990).

A number of studies have identified overlapping emergence periods or apparent niche overlap among co-occurring congeners (e.g., Hildrew & Edington 1979, Soponis 1983, Tavares-Cromar & Williams 1997, Alverson et al. 2001). Similarly in the present study, there was considerable overlap in the times of emergence of closely related taxa (Appendix D). In addition, studies that cite the effect of competition or some other biotic interaction between closely related species or ecologically similar taxa, rarely measure these interactions. Although it is often suggested that life history differences may be responsible for the coexistence of closely related taxa, there could also be differences in other factors that were not measured or factors that are difficult measure (Hynes 1970). A number of studies have concluded that competition or predation does not exert a strong force on the chironomid communities in lotic habitats (e.g., Tokeshi & Townsend 1987, Townsend 1989, Tokeshi 1992a). In contrast, it appears that chironomid communities are largely shaped by density-independent processes as defined by the nonequilibrium dynamics model of Huston (1979). In part, this model assumes that most natural communities are in a state of nonequilibrium as a result of density-independent factors that result in periodic reductions in population size. These density-independent factors include disturbance, as well as a number of other factors, such as fluctuations in the biotic and abiotic environment and some forms of predation (Huston 1979). The unequal size of genera present were hypothesized by Tokeshi (1991a) to indicate a lack of interspecific competition, because this interaction would be predicted to limit the number of congeneric species, and therefore competitors, occurring in lotic habitats. This dissimilarity was also prevalent in this study with genera, such as Cricotopus,

180 Orthocladius, Polypedilum, and, Tanytarsus, contributing many more co-occurring congeners than other genera. However, the effect of physical variables and biotic interactions were not directly assessed by Tokeshi (1991a) so other explanations for this pattern is possible.

The importance of disturbance (e.g., spates, drought) shaping many lotic communities has received more attention recently (e.g., Reice 1985, Resh et al. 1988, Reice 1990, Reice 1994, Townsend et al. 1997). Disturbance is often defined as an event that is unpredictable or stochastic (Resh et al. 1988, Reice 1990) and is hypothesized to be important in structuring aquatic communities because these events result in population decreases. In contrast, aquatic insects can adapt to predictable or moderate changes, and would be expected to reduce the effect of these changes on a population (Reice 1990). However, it is likely that predictable fluctuations in lotic environments (e.g., spring spates, seasonal changes in temperature) can also result in greater richness in a stream because environmental fluctuations generate greater heterogeneity that can be exploited by different species. Thus streams with greater thermal variability support significantly more species rich chironomid communities than thermally stable streams (Chapter IV).

Richness of chironomids also appears to be associated with increased food resources (McLachlan et al. 1978, Tokeshi 1986b, Pinder 1992, Tokeshi 1992a). For example, McLachlan et al. (1978) suggested that the richness of species is determined by the availability of appropriate food resources, but that competition for those resources does not led to competitive exclusion. Observationally, this also may have occurred in the present study as the site with the greatest richness (Chub Creek – 131 taxa) also had considerable algal blooms during the summer when richness was highest. This seasonal pattern of increasing food availability coupled with greater richness suggests that an increase in food sources each year opens this resource to taxa that have either remained present in the habitat (e.g., present with limited growth/development or dormant) or have colonized the habitat (e.g., drift or oviposition). Because this resource is temporally

181 limited, taxa that co-occur and utilize this resource do not persist in an active state long enough for competitive exclusion to take place.

The pattern of changing food abundance and other factors (e.g., temperature) coupled with the life histories of chironomids appear to be important in shaping these communities and permitting many species to utilize the same habitat (Tokeshi 1986b, Pinder 1992). In particular, the ability of many chironomids to undergo dormancy, quickly colonize habitats, and undergo rapid growth and development is important in shaping chironomid communities. These characteristics allow many chironomid taxa to rapidly exploit seasonal and temporary resources during periods of optimal conditions. Furthermore, disturbance and the loss through mortality or emergence would indicate that habitats or resources will always be available to new colonizers. This would be expected to result in patchiness in species aggregations in habitats that may appear homogeneous (Tokeshi & Townsend 1987).

Townsend (1989) raised the idea that some rare taxa may represent species that are early colonists or opportunists which are poor competitors under established conditions, but when new habitats are available they quickly fill these habitats and undergo rapid development. This idea is supported by Pinder (1985) and Ladle et al. (1985) who identified a new species of chironomid that rapidly colonized an artificial channel, dominated the community, and disappeared as other species of chironomids colonized the habitat. It is interesting that many animal communities often consist of a large number of rare taxa represented by only a few specimens (Tokeshi 1992b) and the chironomid communities in this study are no exception. The prevalence of rare taxa raises a number of questions regarding their importance in aquatic systems and what factors result in their rareness.

182 CONCLUSIONS

A large number of chironomid species (>100 taxa) are supported in aquatic insect communities in SWD streams located in the Upper Midwest. There are differences in both the thermal preferences and times of emergence for different subfamilies, tribes, and some genera. Thus, thermal partitioning functions at higher taxonomic levels, such as the subfamily, tribe, and to a lesser degree, genus levels to allow a greater number of taxa to utilize the same habitat at different times of the year. In contrast, genera and species have relatively rigid thermal preferences as most taxa at these levels possess similar thermal preferences and times of emergence compared to their respective higher taxonomic groupings. Specifically, thermal preferences between co-occurring congeneric taxa from small versus large genera did not indicate that a greater difference in thermal preference was present between taxa with more co-occurring congeneric taxa. Based on current and existing research, it can be concluded that temperature does not facilitate density- dependent interactions by sorting ecologically similar congeneric species from the regional species pool into a subset of taxa that possess greater differences in thermal preferences. There are arguments for a strong effect of competition in governing the composition and abundance of aquatic organisms (e.g., Hart 1983); however, most current research indicates that aquatic systems, and particularly lotic systems, density- independent interactions are the primary driver of composition and abundance of aquatic insects (e.g., Tokeshi & Townsend 1987, Townsend 1989, Pinder 1992, Tokeshi 1992a). The effect of either density-dependent and density-independent can not be ruled out and a variety of factors are probably important in shaping communities of aquatic insects and allowing coexistence of ecologically similar taxa in lotic habitats (Tokeshi & Townsend 1987, Townsend 1989, Tavares-Cromar & Williams 1997). These factors include microhabitat preferences, feeding preferences, biological interactions (e.g., competition, predation), environmental variability, and stochastic events. Therefore, the coexistence in some streams with more than ten congeneric species appear to reflect differences in resource preferences between taxa and the dynamic nature of lotic systems, which

183 maintains communities in a state of nonequilibrium due to environmental instability that reduces competitive exclusion.

The calculated thermal preferences provided in this study for 261 chironomid taxa provides autecological information for taxa in a species-rich family where relatively little is known about the ecologies for most species. An understanding of a species’ tolerances, requirements, and preferences can help to enhance their use in ecological and biological assessment studies, as it is important to first understand their responses to natural conditions, before they can be used to assess the impact of altered conditions or anthropogenic stressors. In addition, it is important to understand the factors, both abiotic and biotic, that regulate the composition of communities and the abundance of taxa within those communities. Considerably more work needs to be done on aquatic communities, especially chironomids, to determine the mechanisms that shape these communities so that they can be used effectively in biological assessment. Many of the chironomid taxa from the streams in the present study appear to coexist, but without additional research it can not be stated if these congeneric taxa utilize similar resources and are subsequently in competition with each other. Due to the large diversity of chironomids and some difficulties with identification of the larvae, much of what is known about chironomids and their requirements and preferences is limited. Future research will need to assess these characteristics in this family, as well as characteristics such as dispersal abilities of the larvae and adults. For example, if as it appears, stochastic processes dominate the structuring of these systems, the ability for species to disperse into these habitats will largely determine their presence and subsequently the species richness. The use of null model tests or other ecological statistics testing was beyond the scope of this study; however, in the future these methods may aid in the determination of the mechanisms that permit the co-existence of many chironomid species in some lotic systems.

184

CHAPTER VI

WINTER GROWTH, DEVELOPMENT, AND EMERGENCE OF DIAMESA MENDOTAE MUTTKOWSKI (DIPTERA: CHIRONOMIDAE)

ABSTRACT

The emergence, growth, and development patterns of the winter-emerging species, Diamesa mendotae Muttkowski, were assessed using both collections of surface floating pupal exuviae (SFPE) and in situ enclosures. Emergence data indicated that this species emerges from October through May, when water temperatures are below ≈10 °C, and with peaks in emergence in the fall and late winter. Although emergence occurs through the winter, there was lower emergence during December and January; however, there was no clear thermal threshold associated with this lull. Development of larvae from field enclosures supported the emergence data and indicated that the reduction and halt of emergence in the spring was related to water temperatures that became unsuitable for the growth or survival of the larvae. Development continued through January when water temperatures were at their lowest for the study stream, indicating that the mid-winter decrease in emergence could have been related to low air temperatures fatal to the adults. Growth rates of D. mendotae were not considerably greater than other chironomid taxa that have been tested at similar temperatures indicating that lower developmental zeros may be more important in allowing this species to dominate ground-water dominated streams in Minnesota. The results of this study demonstrate that D. mendotae is well suited for growth and development at low temperatures and provides an assessment of important factors that regulate this species at low water and air temperatures.

185

INTRODUCTION

The winter-active aquatic insect community of temperate zones is often ignored because it is generally considered to have little importance in the structure and function of the overall community. This is largely because the winter environment is considered harsh and marginal for poikilothermic animals, and therefore, these species only comprise a small subset of the community, that does not contribute substantially to the for much total biomass, secondary production, or matter processing. In many northern temperate aquatic habitats, this may be true (e.g., surface-water dominated [SWD] streams), because temperatures in these streams are often near 0 °C for much of the year (Chapter I); however, there are examples of chironomids emerging at these water temperatures (e.g., Oliver 1968, Welch 1973, Sherk & Rau 1996, Ferrington 2000, Bouchard pers. obs). The ability of aquatic insects to utilize these habitats in winter is the result of the physical properties of water that buffer the organisms in these habitats (Moore & Lee 1991). Additional protection from low temperatures is found in streams receiving considerable amounts of ground-water (e.g., ground-water dominated [GWD] streams), which would be expected to allow insect activity to be maintained even during the coldest months. As a consequence, many aquatic insects are capable of growth and development during periods when most other terrestrial poikilothermic organisms are dormant.

In addition to larval development in this buffered environment, emergence of chironomids also occurs despite low air temperatures (e.g., Bouchard et al. 2006a). However, the occurrence of winter emergence in northern temperate habitats is sometimes considered to be leakage from a fall or spring emerging species and therefore, of little consequence to the population as a whole. Variability in the emergence times for chironomids has been hypothesized to be the result of individual genetic variability, which may buffer a population against unusual environmental conditions (Danks 1978). Chironomids can be observed emerging throughout the winter in Minnesota and they are often joined by Plecoptera, Trichoptera, and Tipulidae during late autumn and late winter; however, in Minnesota the most common component of winter-emerging communities,

186

especially during midwinter, are the Chironomidae (Bouchard pers. obs.). In some cases, chironomid winter emergences are large suggesting that they are not leakage of a small portion of the population. The emergence, activity, swarming, mating, and oviposition of Diamesa at low air temperatures has been well documented (e.g., Young 1969, Hågvar & Østbye 1973, Hansen & Cook, 1976, Kohshima 1984, Herrmann et al. 1987, Nolte & Hoffman 1992a, Ferrington 2000, Ferrington 2007). In Minnesota and Wisconsin, D. mendotae is a common inhabitant of many streams and this species can be collected as adults from September through May (Hansen & Cook 1976). This species can be observed walking on snow at temperatures, at least as low as -7 ºC, and commonly has large emergences during days with maximum temperatures near 0 ºC (Bouchard pers. obs.). This species, therefore, appears to be able to utilize the cold winter months in Minnesota as a period when considerable development, mating, and oviposition occurs.

Although providing some insight into an aquatic insect’s biology, observations of emergence will generally only provide a limited view of the life history of D. mendotae and most other winter-active species. Several life history parameters (e.g., voltinism, dormancy) can be difficult or impossible to determine using emergence data only, especially for taxa that are multivoltine, asynchronous, or undergo cohort splitting (Butler 1984, Butler & Anderson 1990, Wrubleski & Rosenberg 1990, Berg & Hellenthal 1992a, Tokeshi 1995). Even the use of frequent larval collections can be problematic for multivoltine species with continuous emergence and overlapping generations. Therefore, growth and development rates may need to be determined experimentally in the field or laboratory. Several studies have used in situ enclosures to assess the growth or development rates of chironomids (e.g., Huryn & Wallace 1986, Nolte & Hoffmann 1992a, Gresens 1997). It is assumed that the use of field enclosures will provide a better assessment of the developmental times of chironomid larvae under natural thermal regimes. Although laboratory estimations permit many conditions to be controlled, this means that in almost all cases a stable thermal regime is used to estimate life history parameters that are then used to corroborate field results. However, this may lead to misleading interpretations as most aquatic insects are exposed to both diel fluctuations in

187

temperatures and seasonally changing temperatures. Some studies on other aquatic insects have established that diel temperature fluctuations can alter hatch time or the rate of growth and development (Sweeney 1984). Among aquatic Diptera, one example is the pitcher-plant mosquito, Wyeomyia smithii (Coquillett) that had a reduced growth rate under fluctuating thermal conditions (Bradshaw 1980). Many taxa will also be exposed to different thermal regimes over the course of their life cycle that may have an impact on life cycle timing. For example, in spring emerging taxa, early instars could have lower threshold temperatures or developmental cues than later instars that are exposed to higher temperatures. Therefore, in situ experimental work is needed to determine developmental and growth rates or to support laboratory results (Howe 1967).

Little is known about the life histories of most winter-active aquatic insects. In this study, the emergence patterns of a winter-active chironomid were assessed from three streams. However, emergence data can only provide a limited view of the life history of a chironomid since the majority of the life cycle is spent in the water. Field enclosures were used to measure the developmental and growth rates of the larvae of D. mendotae under different natural thermal regimes over the course of the emergence period of the adults. The use of in situ enclosures permitted the determination of the duration of the different life stages and therefore an estimation of the voltinism of D. mendotae and the thresholds for development and emergence. By determining emergence patterns and larval growth and developmental rates, a better assessment of the factors and thresholds that affect both the adult and immature stages and subsequently regulate the life history of this species, could be inferred.

METHODS

STUDY SITES The emergence patterns of D. mendotae were assessed from three GWD streams with abundant populations of this species: Valley Creek (Washington Co., MN), Trout Brook (Dakota Co., MN), and Pine Creek (Dakota Co., MN). Growth and development patterns

188

of this species were assessed in Valley Creek using in-stream enclosures. See Chapter I for a detailed description of these streams and the locations of these sites.

ESTIMATION OF EMERGENCE USING SURFACE FLOATING PUPAL EXUVIAE See Chapter I for details on the methods used to collect SFPE and measure mean daily water temperatures. Collections of SFPE were made biweekly in 2002 for Valley Creek and 2003 for Trout Brook and Pine Creek. An Onset StowAway TidbiT temperature logger was deployed in each stream during the sample period to measure water temperature every 15 minutes. Although the SFPE could not be verified to be D. mendotae, extensive collections of adult Diamesa at Trout Brook and Valley Creek consisted almost exclusively of D. mendotae. Fewer adults were collected from Pine Creek, but these specimens were also D. mendotae. Emergence patterns were assessed graphically by plotting the number of SFPE collected on each date versus the mean daily water temperatures from the stream in which the SFPE were collected. The duration of emergence for D mendotae was also estimated for each stream using the estimation method from Coffman & de la Rosa (1998). This method requires the assumption that the proportion of samples in which the species is collected corresponds to the proportion of the year in which the species is emerging. Therefore, the duration of emergence is calculated by multiplying the proportion of samples in which D. mendotae SFPE occur by the sampling interval (i.e., 14.04 d for a biweekly sampling interval).

ESTIMATION OF LARVAL GROWTH AND DEVELOPMENT In situ enclosures Four cages measuring 20 cm3 were constructed of acrylic glass with 63-µm Nitex mesh windows measuring 14 cm2 (Figure 6.1). These enclosures were placed in Valley Creek at the head of a riffle where the current was relatively swift but still laminar. The cages were filled with ≈6 cm of clean gravel substrate and allowed to be colonized by stream algae and microorganisms that could pass through the screen. A 200-µm Nitex mesh screen and a 1.5-cm wire mesh were placed upstream of the enclosures to reduce the amount of large and potentially damaging debris from collecting on the cages. An Onset

189

StowAway TidbiT® temperature logger was randomly placed into one of the cages and a second logger was deployed outside the enclosures by attaching it to the wire mesh upstream of the enclosures.

Figure 6.1: In-stream enclosures for Diamesa mendotae in Valley Creek above riffle.

Egg masses To obtain egg masses, pairs of adults were collected from the snow or from an emergence trap when there was no snow cover. Adults were collected directly into 7.39-ml snap top vials along with a small amount of stream water or melted snow. The adults were maintained in a lab refrigerator at ≈5 °C and monitored daily. When an egg mass was detected in the vial, the adults were removed from the vial and preserved for later species confirmation. Egg masses were maintained at ≈5 °C until enough egg masses were obtained to place into the cages. When a sufficient number was achieved, the egg masses were placed into three of the four cages. The fourth cage was maintained as a control to ensure that larvae from the stream were not able to enter the cages and to observe differences in algal growth between the control cage and those with larvae. To ensure a sufficient number of larvae in the cages, nine egg masses were placed into each of the 190

three cages in Trial I. This number of egg masses proved to be sufficient and in subsequent trials (i.e., Trials II, III, & IV) only three egg masses were placed into each of the three cages. For unknown reasons, no larvae were collected from one test enclosure in Trial II so data are only available from two enclosures during this trial. A trial consisted of the period from the placement of the egg masses in the cages to the first emergence of adults or cohort failure. Trial I was run from January 3, 2006 through March 7, 2006, Trial II from March 28, 2006 through May 8, 2006, and Trial III from October 27, 2006 though January 26, 2007. Trial IV was run from February 2, 2007 through March 14, 2007; however, data from Trial IV are not presented in this study due to a large spate from snow melt that damaged the cages, resulting in premature termination of the experiment before any emergence.

Egg masses that were collected in preparation for Trial III, but were not used in the enclosures, were maintained at ≈5 ºC and checked daily. This was done to estimate the duration of the egg stage since this could not be done in the field. Larvae within an individual egg mass hatched at different rates, usually with the embryos at the ends of the egg mass maturing and hatching first, but most of the larvae hatched within ≈24 hrs. As a result of the different hatching times, the estimate of the duration of the egg stage was measured as the time from when the egg mass was first detected to the time when >50% of the larvae in the egg mass had hatched.

Sampling and sample processing The cages were monitored on a weekly basis at which time samples were taken from each cage and water quality was measured. On each sample date enclosures were first inspected for emerged adults or SFPE. Samples consisting of four stones from the gravel substrate placed in the cages at the beginning of the experiment were then carefully removed from each cage and placed individually into wide-mouth 118-ml Nalgene jars with water from the cage. During each weekly sampling event dissolved oxygen (DO) was measured outside the enclosures and inside of one test enclosure using a handheld YSI 85 meter. In addition, the outside of the enclosure screens were scrubbed to reduce

191

clogging of the screen by algae and biofilms. Samples were kept in a cooler with snow and/or stream water and returned to the laboratory for processing within ≈3 hrs of sample collection. In the laboratory, the rocks were scrubbed and rinsed and all live larvae were picked and preserved in 70% ethanol. Larvae that did not move when prodded were assumed to be dead and were not picked. All larvae from the same collection (i.e., all four rocks) from each cage were combined and were treated as a single sample for the analyses.

Measurement of larvae The specimens from cage 1 in Trial I (n=179) were slide mounted. From these specimens, the head capsule length and width could be measured for 107 specimens using an ocular micrometer in a compound microscope. Specimens from which measurements could not be taken were mostly first instars that were difficult to obtain good slide preparations with the ventral side of the head capsule facing upward. The head capsule length was measured as the distance from the postocciput to the antennal base and head width was measured as the head width at its widest point. A plot generated from the head length and width data from cage 1 in Trial I indicated that instars were separable based on the size of the head capsule (Figure 6.2). The instar was determined for all remaining larvae by measuring the head capsule under a dissecting microscope. The body length, measured as the length from the antennal base to the procercus, was determined for all larvae from all trials using a dissecting microscope.

DATA ANALYSIS The duration of the first, second, and third instars were calculated by estimating the day when each instar was the largest proportion of the cohort and determining the difference between these times. The duration of the egg stage could not be estimated in the field, but, observations of the egg masses not used in the in-stream enclosures were made in the laboratory under constant 5 °C temperatures and used as an estimate of the duration of this stage. Although stream temperatures had diel fluctuations, average daily means of 4- 7° C when the egg masses were placed in the cages were near the laboratory 5 °C

192

temperature, with the exception of Trial II when temperatures in the stream ranged from ≈8-9 °C. This estimate of egg stage duration was needed to estimate the duration of the first instar. The mean duration of the fourth instar could not be estimated because the trials were terminated when the first emergence was recorded and not when all larvae had either emerged or died.

Figure 6.2: Instar separation for Diamesa mendotae based on head width and head length.

The body length measurements were used to estimate larval body mass. From length measurements, the ash free dry mass (AFDM) of each larva was estimated using the following equation from Nolte (1990):

ln weight = −2.631+ 2.602lnlength where the only information needed is the length of a larva measured as the distance from the procercus to the antennal base. Although this regression was developed using different Diamesa species than the species in this study, D. mendotae has a similar body

193

shape to those used in the development of the model. Furthermore, the ability of this equation to predict body weight for multiple Diamesa species had a correlation coefficient of 0.989. Using the estimates of AFDM, daily instantaneous growth rates for D. mendotae from the stream enclosures could be made. Daily instantaneous growth rates were calculated using the following equation:

g = ln(M f / M i ) × 1/ t

where Mf is the final mean AFDM of larvae in the enclosure, Mi is the initial mean AFDM of the larvae when they were first collected, and t is the number of days from first collection to first emergence (Huryn & Wallace 1986, Hauer & Benke 1991, Reynolds & Benke 2005).

RESULTS

EMERGENCE PATTERNS In all three streams, D. mendotae has a similar emergence pattern with emergence occurring from October through May. There was also sporadic emergence in June (Trout Brook, Pine Creek), July (Pine Creek), and August (Trout Brook) depending on the site (Figures 6.3, 6.4, & 6.5); however these collections never consisted of more than six specimens. The only month from which no emergence was recorded was September. Diamesa SFPE were collected in nearly every sampling event from middle October through the end of May. Exceptions to this were a lack of specimens collected in early January and early May in Valley Creek. Although biweekly collections of SFPE from these three streams did not identify large January emergence peaks, large January emergences were observed from Valley Creek in 2006 and 2007 when maximum air temperatures warmed above 0 ºC (Bouchard pers obs.). The duration of emergence was estimated to be 211, 267, and 281 d for Valley Creek, Trout Brook, and Pine Creek, respectively. The emergence pattern was bimodal at all sites, although in Valley Creek and Trout Brook there were additional peaks in the autumn. Emergence in the spring

194

nearly ceased when average daily water temperatures reached ≈10 °C. Similarly, emergence of D. mendotae did not begin in the autumn until average daily water temperatures dropped below ≈10 °C. The relationships between water temperature and the peaks in emergence are less clear as there is no obvious increase or decrease in the temperature that corresponds to these peaks. In December and January, there is also a reduction in the number of D. mendotae emerging. During this period average daily water temperatures were lowest (i.e., ≈2 °C) in some streams for short periods. However, there is not a clear water temperature threshold associated with decreased emergence during the middle of winter because there are considerable fluctuations in water temperature during the winter.

Figure 6.3: Emergence patterns for D. mendotae in Valley Creek as measured by SFPE compared with mean daily temperature.

195

Figure 6.4: Emergence patterns for D. mendotae in Trout Brook as measured by SFPE compared with mean daily temperature.

Figure 6.5: Emergence patterns for D. mendotae in Pine Creek as measured by SFPE compared with mean daily temperature.

196

IN-STREAM ASSESSMENT OF LARVAL DEVELOPMENT AND GROWTH Temperature and dissolved oxygen A comparison of the loggers from inside and outside the enclosures determined that the temperatures were on average 0.35 ºC warmer within the enclosures with the maximum temperature difference recorded 0.64 ºC. Levels of DO were generally lower and more variable in the enclosures (Figure 6.6). There was a peak in the levels of DO shortly after the egg masses were placed into the cages in Trials I and III and again near the time when the trials were terminated for all three trials. During Trial I, mean daily temperatures were estimated to be 4.17 °C with a minimum of 2.35 °C and maximum of 6.20 °C. On average the daily mean temperature in Trial II was 10.00 °C with a minimum of 8.16 °C and maximum of 11.69°C. Mean daily water temperatures in Trial III was 5.54 °C with a minimum of 2.63 °C and a maximum of 8.21 °C.

Figure 6.6: Weekly spot measurements of dissolved oxygen (mg/l) from within the enclosures and outside the enclosures in the stream.

Development Observations of adult pairs in the laboratory indicated that after nine days, most of the females finished ovipositing, as no further egg masses were recorded after this time, although egg masses were only monitored for 14 d and additional egg masses may have 197

been detected. When held for 14 d with a male, 62-64% of females oviposited, allowing for easy collection of sufficient numbers of egg masses for use in the stream enclosures. Observations of embryogenesis and hatching of larvae from six egg masses held in the lab at ≈5 °C had a mean (±SE) duration of 7.33 (±0.92) d. After hatching, many of the larvulae remained within the egg mass for ≈24 hrs. Egg masses were held on average 3.28 d before they were placed into field enclosures. The duration of the egg masses before larval hatching in the enclosures was subsequently estimated to be 4.06 d.

As with other chironomids, D. mendotae has four instars. Using a plot of larval head width and head length, the four instars of D. mendotae larvae could be easily separated (Figure 6.2). In all three trials, no larvae were collected until the second sampling period, 14 d after the egg masses were placed in the cages. Lack of larvae in the collection after seven days was probably the result of the larvae not traveling far from the egg mass after hatching and not an extended egg stage. During all trials, the number of larvae collected in each sample dropped as the experiment proceeded. This decrease was due to the removal of specimens each week and natural mortality in the enclosures indicated by the dead larvae.

In Trial I, the mean duration of the first, second, and third instars were estimated to be 10.97, 7.68, and 15.68 d, respectively. In Trial II the mean duration for the first, second, and third instar were 9.47, 8.49, and 4.64 d, respectively. However, the short duration of the third instar in Trial II should be also treated with caution as many of the third and fourth instar larvae appeared to have died as a result of the warming water. In Trial III, on average the first, second, and third instars had durations of 15.66, 14.58, and 13.40 d, respectively. Fourth instar larvae were first collected after 35, 28, and 42 d for Trials I, II, and III, respectively. First emergence occurred 63 d after the eggs were placed in the cages in Trial I (Figure 6.7) and 91 d for Trial III (Figure 6.9). In Trial II the number of dead larvae became more numerous than observed in the other trials beginning April 17, 2006. On May 8, 2006, no live larvae were collected and the trial was terminated and therefore no emergence was recorded for Trial II (Figure 6.8).

198

Figure 6.7: Instar composition and daily mean temperature for Trial I (error bars represent standard error; n=3).

Figure 6.8: Instar composition and daily mean temperature for Trial II (error bars represent standard error; n=2).

199

Figure 6.9: Instar composition and daily mean temperature for Trial III (error bars represent standard error; n=3).

Growth Mean daily instantaneous growth rates (±SE) for Trial I were estimated to be 0.086 (±0.004) µg µg-1 d-1 and 0.064 (±0.001) µg µg-1 d-1 for Trial III. Mean AFDM (±SE) of the final sample at first emergence from the enclosures was 201.43 (±45.44) and 253.57 (±28.34) µg for Trials I and III, respectively. The increase in mean larval mass for Trial III lagged behind Trials I and II (Figure 6.10). This was largely a reflection of slow development and accumulation of mass for the early instars at the beginning of the Trial III. After this approximately two week delay, mass was accumulated at similar rates to Trials I and III. However, this does not account for the differences in the emergence times, because the first emergence of Trial III lagged 4 weeks behind that of Trial I. In fact, similar average weights were estimated for Trials I and III after 63 d, but first emergence for Trial III was not recorded until 91 d after the egg masses were placed in the cages.

200

A total of 267 and 553 DD were accumulated before first emergence during Trials I and III, respectively. In Trial II, 420 DD were accumulated before this trial was terminated. Greater separation in the accumulation of mass between trials was observed when mean AFDM was compared to cumulative degree days (Figure 6.11). Specifically, a greater number of DD were accumulated before similar mean larval masses were attained in Trials II and III compared to Trial I. For example, larval mean AFDM was 201 µg after 254 DD in Trial I and only 12 µg after 267 DD in Trial III. This indicated a substantial delay in development in Trial III where considerable mass gain did not begin until December when temperatures were consistently below 6 °C (Figure 6.9).

201

Figure 6.11: Mean larval AFDM (µg) versus cumulative degree days for all Diamesa mendotae specimens present in samples (error bars represent standard error; Trials I and III: n=3, Trial II: n =2).

DISCUSSION

EMERGENCE PATTERNS Collections of Diamesa from Minnesota streams suggest that D. mendotae is largely limited to GWD streams in the region. This species also has a relatively long estimated period of emergence (i.e., 211-281 d) in the three streams studied. Furthermore, in these streams D. mendotae is often the dominant species of Diamesa, although other species are known to co-occur with D. mendotae (e.g., D. nivoriunda; Hansen & Cook 1976, Bouchard unpublished data). In fact, the larvae of this species appear to dominate the chironomid communities in the rocky substrates of these GWD streams from late fall through early spring. In SWD and intermittent streams in this region, other species such as D. nivoriunda and D. cheimatophila are more commonly collected. In these habitats, Diamesa emergence is generally limited to a short period in the fall and spring due to winter ice cover; however, if openings in the ice are present, emergence of these species can occur even when water temperatures are at or near 0 °C (Bouchard pers obs). It is 202

possible that D. mendotae can emerge at similarly low water temperatures; however, in this study the daily mean temperatures do not drop below ≈1 °C. When mean daily water temperatures in the study streams drop below 4 ºC minimum air temperatures generally are very low (e.g., <-25 °C) and are potentially fatal to emerging adults. Bouchard et al. (2006a) determined that 50% of adult male D. mendotae die when temperatures drop to -

21.5 °C (i.e., lower lethal temperature [LLT50]). Based on historical temperature records from the region, the probability of daily minimum air temperatures in January exceeding the LLT50 of the species in this region can be as high as 35%. The low emergence or lull in emergence observed during January from all three sites is perhaps related to the relatively high probability of occurrence of air temperatures that are lethal to the adults during this period. However, it is not clear which mechanisms regulate the pupation and emergence of D. mendotae that would permit emergence at optimal air temperatures.

The ability of this species to time emergence to short periods in midwinter when relatively warm air temperatures occur is particularly interesting. This timing is complicated, because in chironomids formation of adult tissues in the pupa requires several days. Consequently, there is a lag between the time pupation is initiated to adult emergence (Chapter II). For example, Nolte & Hoffmann (1992a) determined that D. incallida required ≈72 hrs to complete the pupal stage at 8 ºC. Winter water temperatures in Valley Creek are only ≈4-6 ºC and therefore the pupal stage of D. mendotae would be expected to be 72 hrs or more during winter. If pupation cues for D. mendotae are simply a change in water temperature, then there would be potential high mortality if the duration of a warm spell was short. Appropriate timing of emergence can also be complicated by the fact that high air temperatures melt snow cover and cause decreases in water temperature as melt water near 0 °C enters the stream. As a result, additional work is needed to determine the cue(s) for pupation and emergence in D. mendotae.

As water temperatures warmed in the spring, the emergence of D. mendotae declined until the species was only sporadically collected from late spring through early autumn. This decline in emergence occurred at water temperatures of ≈10 °C and higher, and

203

could reflect developmental thresholds in the immature stages. Without a better understanding of the factors influencing the development of the eggs, larvae, and pupae it is difficult to conclude which factors are affecting which stages. Due to the close linkage of temperature and photoperiod, a photoperiodic cue can not be ruled out for initiating pupation and emergence. In addition, from emergence data it is not possible to determine whether these factors regulate larval growth and development, initiate dormancy, or initiate pupation and emergence. In addition, the upper lethal threshold and longevity of the adults of this species to high temperatures have not been tested and could be responsible for the decline in emergence during the warmer months.

DEVELOPMENT The minimum development time of 63 d determined for D. mendotae is relatively rapid during cold winter months in Minnesota. Based on the duration of emergence duration of this species from several sites, this species can potentially undergo three or more generations. Emergence data indicate that a minimum of two generations were completed. However, further study is needed to determine whether emergence peaks represent 2-3 synchronized cohorts or whether they are emergence peaks of an asynchronous emerging species during optimal conditions.

Growth and development of D. mendotae larvae continued throughout the winter when relatively low water temperatures of ≈2-6º C occurred. Therefore, in Valley Creek this species is able to continue development through the entire winter. As a result, lower thresholds could not be determined because temperatures in this stream did not become low enough to halt development or trigger dormancy. In contrast, temperatures in Valley Creek presumably become high enough to result in dormancy. This was indicated by the lack of emergence, high mortality and the eventual failure of the cohort in the stream enclosures when temperatures exceeded ≈10 ºC. In the field, D. mendotae larvae were not collected when stream water temperatures exceed ≈10 °C (Bouchard unpublished data), and presumably the larvae enter the hyporheic for aestivation during periods of warm temperatures in summer.

204

It is difficult to identify factors directly lethal to Diamesa larvae at temperatures above ≈10 °C without more detailed physiological studies. High temperatures may have inactivated control enzymes in metabolic pathways or that metabolic needs at high temperatures exceeded the ability of the larvae of D. mendotae to assimilate energy. It is also possible that temperature was not directly lethal, but that low DO levels that are fatal or suboptimal for the larvae. Additional work is needed to determine the DO and upper thermal thresholds of this species and the physiological mechanisms that control these tolerances.

The development time of 63-91 d for this species is not the fastest recorded for Diamesa. Nolte & Hoffmann (1992a) determined that D. incallida could complete development in 39-43 d. Although considerably faster than D. mendotae, D. incallida was studied near a spring source where the water temperatures did not fluctuate greatly from 8 °C. In the current study, stream temperatures were ≈4-6 °C during the majority of the period when development of D. mendotae occurred. Development also occurred when temperatures were 6-10 °C, but in Valley Creek these temperatures largely occur during spring and fall and are short in duration. Therefore, if D. mendotae and D. incallida have similar thermal thresholds, D. mendotae would be expected to require greater developmental time due to the thermal conditions of Valley Creek. This is supported by the accelerated development in Trial II when temperatures were above 8 °C.

Trial III required a considerably longer period to first emergence compared to Trial I. Second only to temperature, nutrition is considered to often have an impact on the life history of chironomids (Chapter II; e.g., Sokolova 1971, Mackey 1977a, Ward & Cummins 1979, Storey 1987, Mattingly et al. 1981, Pinder 1992, Gresens 1997). Based on qualitative observations of algal growth in the test and control enclosures, it may be likely that reduced amounts of food were responsible for the extended developmental time. Although the screens on the cages were scrubbed during each visit, algae and biofilms accumulated on the screens and may have reduced the amount of water, and

205

therefore nutrients and microorganisms entering the cages. Additional evidence of potential food limitation may also be supported by measurements of DO where lower levels of DO may indicate lower primary production and higher larval respiration (Figure 6.6). Levels of DO were consistently lower in the enclosures across trials compared to measurements taken in the stream. DO levels in the enclosures were similar or greater than that of the stream for a few periods including shortly after the egg masses were placed in the cage and toward the end of the trials. It is possible that the first peak in the trials represents an increase in algal activity that is rapidly controlled by the grazing larvae. The second peak in DO before the termination of the trials was possibly the result of a second increase in algal growth due to decreased grazing pressure because of pupation or a decrease in respiration by the larvae resulting from the emergence, mortality, and removal of larvae.

In chironomids, photoperiod has also been identified as a factor that controls the development period (Chapter II). For example, Nolte & Hoffman (1992b) provide evidence for an extended developmental time in Pseudodiamesa branickii resulted from the photoperiod to which the eggs were exposed. Trial III began in October when the photoperiod was decreasing, whereas the other trials largely took place during periods of increasing photoperiod. The fact that the estimated average mass of the larvae was nearly identical for Trials I and III, suggests that some factor other than nutrition was responsible for the delayed emergence in Trial III. However, the effects of photoperiod and nutrition could not be separated to determine the cause of the extended development time in Trial III.

GROWTH However, in Trial III there was reduced accumulation of mean larval mass at the beginning of the trial when temperatures were above 6 °C that may indicate slowed growth due to suboptimal temperatures.

206

The daily instantaneous growth rates for D. mendotae are not very rapid when compared to maximal growth rates determined for other taxa at higher temperatures (e.g., Huryn & Wallace 1986, Hauer & Benke 1991, Reynolds & Benke 2005). However, comparisons must be made cautiously because experiments were performed on mixed species chironomid assemblages in these literature studies. The growth rates are similar to the rates predicted by these studies at low temperatures; however many of the taxa tested had growth rates of 0 µg µg-1 d-1 when temperatures approached 5 ºC (Hauer & Benke (1991, Reynolds & Benke 2005). In contrast, Huryn & Wallace (1986), working in a stream with limited thermal variation, determined that growth rates for the largest larvae (2-3 mm) was 0 µg µg-1 d-1 at 2.3 ºC. At 2.9 ºC, growth rates were 0.031 µg µg-1 d-1 and at 5.5 ºC growth rates were 0.55 µg µg-1 d-1. The chironomids assessed by Huryn & Wallace (1986) were largely comprised of Orthocladiinae, which as a subfamily are more cold- adapted (Oliver 1971), and helps to explain why developmental zeros were lower than those determined by Hauer & Benke (1991) and Reynolds & Benke (2005). Therefore, it appears that the growth rates of D. mendotae and other chironomids are consistently slow at low temperatures, but that the developmental zeros of these taxa differ. Based on available results, it appears that D. mendotae has at least a lower thermal threshold compared to the Chironominae tested thus far, but further work is needed compare the developmental zeros of several cool-adapted species such, as Diamesinae and many Orthocladiinae.

Emergence and therefore, development of other Diamesa species (Bouchard pers. obs.) and Orthocladiinae at water temperatures near 0 ºC (Oliver 1968, Welch 1973, Sherk & Rau 1996, Ferrington 2000), confirms that these cold-adapted taxa are capable of growth and development at very low temperatures. In fact, Diamesa often dominate glacial streams where temperatures often do not exceed 2 ºC (Burgheer & Ward 2001, Milner & Petts 1994, Ward 1994, Milner et al. 2001), including a stream with maximum temperatures of 0.92 ºC (Burgheer & Ward, 2001). Therefore, the growth and development of larvae, followed by pupation and emergence occurs at very low water

207

temperatures for species within the genus; however, it is not clear how lower thresholds may vary among species.

It has been determined that within the genus Diamesa optimal thermal conditions vary among species. For example, Rossaro (1991) estimated optimum temperatures for many chironomid species, including 12 species of Diamesa, in Italy ranged from 4.23 to 9.80 ºC. This would place D. mendotae at the low end of this range, as much of the growth of this species occurred near 4 ºC. This similar to the pattern observed by Ferrington (2000) for Diamesa in Kansas where this taxon was collected at temperatures near 3 °C.

Although not directly measured in this study, D. mendotae may also be responsible for large amounts of secondary production in GWD streams. For example, Berg & Hellenthal (1991) determined that D. nivoriunda was responsible for 33% of the chironomid secondary production with 10,100 mg dry mass m-2 in a thermally buffered northern temperate stream in Indiana. It is likely that D. mendotae has similar importance in GWD streams in Minnesota as it is the dominant chironomid larva collected from coarse substrates from late fall through early spring (Chapter IV; Appendix D). This species is also large for a lotic chironomid, with the largest larvae collected from the enclosures reaching 10.53 mm and 900 µg. Thus, this species and other winter-active species (e.g., Chaetocladius, Orthocladius, Micropsectra) can contribute considerably to the dynamics of these systems as a result of sustained winter development. This is particularly true when GWD streams are compared to SWD streams since the winter community in SWD streams is assumed to be largely dormant due to ice and snow cover.

CONCLUSIONS

The results of this study and others confirm that the species Diamesa mendotae is well suited to life at low temperatures. This species emerges throughout the winter with peaks in emergence in late autumn and late winter with an apparent midwinter reduction in

208

emergence. In the GWD streams in Minnesota studied mean daily water temperatures reach 2 °C during the winter, but there was no clear threshold in winter water temperatures associated with the lull in emergence of D. mendotae. As a result of relatively warm winter water temperatures, growth and development of larvae occurs through the winter. Therefore, the reduced emergence of this species during January is potentially related to predictably low air temperatures that may be lethal to the adults. In contrast, the decline and halt of emergence in the spring appeared to be associated with water temperatures that influenced a developmental thermal threshold in the larvae. When temperatures increased above ≈10 °C emergence nearly halted and the use of in- stream enclosures demonstrated increased larval mortality above this threshold.

The genus Diamesa is often a dominant taxon in high altitude streams particularly in streams near glaciers (e.g., Sæther 1968, Brittain & Milner 2001, Burgheer & Ward 2001, Milner et al. 2001). This genus is not commonly considered to be dominant in low altitude, temperate streams; however, this study and others (e.g., Berg & Hellenthal 1991) indicate that this genus can be extremely important in streams with sustained low temperatures that do not freeze in the upper Midwest. Although Diamesa species also occur at low altitudes in lower latitude streams (e.g., Blackwood et al. 1995, Ferrington 2000) and in SWD streams in Minnesota (Chapter IV), other low altitude Diamesa species may be better suited for habitats with limited periods of low temperatures (i.e., autumn and spring). In addition, some Diamesa species may have higher thermal thresholds or be more tolerant to pollution than D. mendotae, allowing them species to inhabit SWD streams and have broader distributions. As a result of large, winter-active populations of Diamesa in many GWD streams in Minnesota, the winter community dynamics of these streams are very different compared to SWD streams in the same region.

209 CHAPTER VII

LOWER LETHAL TEMPERATURE FOR ADULT MALE DIAMESA MENDOTAE MUTTKOWSKI (DIPTERA: CHIRONOMIDAE), A WINTER- EMERGING DIAMESINAE

R.W. Bouchard, Jr., M.A. Carrillo, & L.C. Ferrington, Jr.

ABSTRACT

The lower lethal temperature (LLT50) and supercooling point (SCP) of male adult Diamesa mendotae Muttkowski (Diptera: Chironomidae) were determined from field-

collected individuals. The LLT50 was determined by exposing individuals to -10, -15, -20, -25, and -30°C for one minute. Supercooling points were determined from individuals from the same collection date using surface-contact thermometry. Mean survivorship was 96.7, 91.2, and 71.2% for tests at -10, -15, and -20°C, respectively. Exposure to -25 and -30°C resulted in 100% mortality in all tests. Results show that adult males of D. mendotae are freeze-intolerant with a mean SCP of -21.6ºC which is only 0.1°C lower

than its linear interpolated estimated LLT50 of -21.5°C. Our results confirm that a low SCP, rather than freeze tolerance, is a mechanism that facilitates emergence and adult activity of this species during winter conditions. The implications of SCP and LLT50 with regard to adult survivorship and potential voltinism are discussed.

Bouchard, R.W. Jr., M.A. Carrillo & L.C. Ferrington Jr. 2006. Lower lethal temperature for adult male Diamesa mendotae Muttkowski (Diptera: Chironomidae), a winter- emerging aquatic insect. Aquatic Insects 28:57-66.

210 INTRODUCTION

Adults of Diamesa mendotae Muttkowski (Diptera: Chironomidae) commonly emerge during winter from groundwater-dominated streams in Minnesota. Adults are active on snow and can mate and fly at ambient air temperatures below 0ºC, indicating physiological mechanisms to prevent freezing. Adults are not known to feed, but have average longevities at 6ºC ranging from 12.1 to 33.0 days depending on population and time of emergence (Ferrington et al. submitted). In an earlier study, Carrillo et al. (2004a) determined the supercooling point (SCP) (i.e., the temperature at which spontaneous freezing of the body fluids occurs) of field-collected adults as an initial attempt to understand the cold-adapted strategies of this species.

Organisms that can survive exposure to low temperatures require protection from the lethal effects of the freezing of body fluids and can be classified as either freeze tolerant or freeze intolerant (Salt 1961). Freeze-tolerant organisms are able to withstand freezing of extracellular body fluids while freeze-intolerant organisms can not survive freezing and use mechanisms to prevent ice formation (Baust & Rojas 1985). Many freeze- intolerant organisms lower the temperature at which their body fluids freeze (i.e., supercool), thereby preventing crystallization and tissue injury (Lee 1991). Studies on the cold hardiness of other chironomid species have reported both supercooling (Kohshima 1984, Ring 1989) and freeze tolerance (Scholander et al. 1953, Hinton 1960b, Danks 1971b, Baust & Edwards 1979, Block 1982, Lencioni 2004) as mechanisms to survive subfreezing temperatures. However, the significance of the SCP as a mechanism to survive subfreezing temperatures in freeze-intolerant individuals has been recently questioned by Renault et al. (2002), who point out that freeze-intolerant species may experience a high percentage of mortality before SCP temperatures are reached, whereas freeze-tolerant species, by definition, have low mortality when exposed to temperatures at or well below their SCP.

In our earlier experiments to determine SCP of D. mendotae, all test specimens died during the experiments, suggesting freeze intolerance (Carrillo et al. 2004a). However, 211 we cautioned that our test design did not allow us to distinguish between mortality resulting from freezing versus handling during the test procedure. In addition, our previous experiments were always run to temperatures equal to or lower than -30ºC to ensure that a SCP value could be attained for each test specimen. Consequently, we had

no way of estimating the lower lethal temperatures (LLT50) (i.e., the temperature at which 50% of the population die when exposed for a defined period of time) for this species. In the present study, we repeated our experimental methods for determining SCP on randomly-chosen subsets of field-collected adult males, but used an alternative test

protocol to provide estimates of the LLT50 for other random subsets of the same batch of field-collected adult males.

MATERIALS AND METHODS

COLLECTION AND TREATMENT OF TEST SPECIMENS Adult males of D. mendotae (n=265) were collected on 13 February 2004 from snow banks along the Kinnickinnic River south of River Falls, Wisconsin, USA. Air temperatures were below freezing when individuals were collected and during several days prior to the day of collection. Individual male D. mendotae were located on snow banks, scooped individually into one-dram vials with a small amount of snow, and maintained at 6ºC until tested. No food sources were provided but the water from melted snow prevented dehydration as individuals aged. Batches of 10 males were randomly selected and their SCP determined at three, five, and seven days post-collection. One male was lost from the second batch; therefore, the SCP of this batch was calculated from nine individuals. Additional batches of 10 males were randomly selected for LLT50 determinations on three, five, seven and 10 days post-collection.

DETERMINATION OF SCPS Supercooling points of adult males were recorded using surface-contact thermometry as described by Carrillo et al. (2004b). Adults were attached to a 24-gauge copper- constantan thermocouple using high-vacuum grease (Dow Corning®, Dow Corning

212 Corporation, Midland, MI USA). Insects with thermocouples were placed at the center of 19 × 19 × 19-cm polystyrene containers with a starting temperature of ≈0ºC. Containers were closed with rubber stoppers and transferred to a -80ºC freezer (Revco Scientific Inc., Asheville, NC) to cool insects at a rate of ≈1ºC min-1 (Carrillo et al. 2004). Body temperatures were recorded with a multi-channel data logger (Personal Daq/56 data acquisition system, Iotech, Inc., Cleveland, Ohio, USA) that reported data every one second through a USB cable directly to a computer, where they were stored, graphed, and the SCP read as the lowest temperature reached before freezing. Freezing was identified by small rise in temperature (i.e., exotherm) during an otherwise decreasing trend, indicating the release of latent heat of fusion (Lee 1991).

DETERMINATION OF LLT50 Batches of 10 adult males were placed into 16 x 150-mm glass test tubes and closed with a small piece of foam. A 24-gauge copper-constantan thermocouple was placed inside each test tube in close proximity to the test insects to monitor temperature. Because insects were maintained at 6ºC after collection, we started our experiments at this temperature. To achieve this temperature, a 35 × 35 × 35-cm polystyrene container with an initial temperature of ≈27ºC was first introduced to a -80ºC freezer and allowed to drop to ≈6ºC. Subsequently, test tubes were placed inside this container at ≈6ºC and placed again into the -80ºC freezer. The container size and freezer temperature allowed us to cool insects at a rate of ≈0.3ºC min-1. Individual test tubes with 10 insects were then cooled to -10, -15, -20, -25, or -30ºC and removed from the container one minute (i.e., minimum exposure; Bale et al., 1988) after the designated temperature had been reached. These treatment temperatures were selected because they spanned the SCP range for males of D. mendotae previously published (Carrillo et al. 2004a). After exposure, test tubes were removed from the container and immediately placed into a programmable growth chamber (Percival Scientific, Inc., Perry, IA) at 6ºC. Insects were then quickly removed from the test tubes, transferred to individual 95 x 15-mm plastic Petri dishes, and placed again at 6ºC.

213 Mortality was assessed 24 h after treatment, and death was defined as lack of movement when individuals were prodded with soft forceps. Control groups were handled similarly as those used for low-temperature exposure but were held at 6ºC to correct for natural mortality (Abbott 1925). Experiments were independently replicated three times for each treatment temperature, but not all treatment temperatures were run during the same cooling period.

Supercooling points from adult males were used to calculate a cumulative proportion of individuals freezing (CPIF) for comparison with mortality. The CPIF was calculated by summing the number of individuals that froze at or above each one-degree temperature step and dividing each resulting sum by the total number of individuals measured (Koch et al. 2004, Carrillo et al. 2005a,b).

CALCULATIONS OF AIR TEMPERATURE METRICS Files of daily minimum and maximum temperatures from 1940 to 2004 were obtained from the Midwest Regional Climate Center for River Falls, Wisconsin, USA, at ≈10 km from our sample site. Averages of daily minimum were calculated as arithmetic means for the 65 years of data available. In addition, the probability of the daily minimum

temperature being less than the LLT50 of adult male D. mendotae was calculated for each day of the year. Furthermore, a seven-day running average was calculated for probability data to smooth temporal variation and to better illustrate trends.

DATA ANALYSIS A one-way analysis of variance (ANOVA) was preformed using NCSS (Hintze 2001) to test for significant differences in SCP among batches of males. The original data for batches of males deviated significantly from the assumptions of ANOVA (i.e., skewness, kurtosis, non-normality of residuals, and unequal variances) and standard transformations did not correct all deviations. However, eliminating outliers successfully remedied all deviations. Means SCPs of batches for the present study were also compared to the mean

214 SCPs of batches reported in our earlier publication (Carrillo et al. 2004a) using ANOVA to test for potential differences in SCP among years.

The mean percentage of mortality for the three tests at each predetermined subzero temperature was used to calculate the LLT50. Linear and curvilinear interpolation of the

percentage of mortality at -20°C and -25°C were used to estimate the LLT50. In addition, both mortality and CPIF curves were plotted to determine the relationship between SCP and adult mortality.

RESULTS AND DISCUSSION

The overall mean (±SE) SCP for adult males of D. mendotae was -21.60 ± 0.48ºC. The SCP of individual males within batches varied from -25.55 to -14.01°C (Table 7.1). Although the maximum SCP determination in this study for a male in batch #2 (- 14.01°C) exceeded the range of SCPs recorded for individual males in our previous study (-26.71 to -16.78°C), all other determinations were within the range of values previously observed. Lacking additional evidence, we do not know if it is proper to interpret these higher values as a result of natural variation in D. mendotae or to methodological error (e.g., handling injury). However, with the outliers removed from the analysis, the estimated mean SCP for batch #2 dropped to -20.50°C (Table 7.1). The mean SCP for batches of males varied from ≈-23 to -19°C (Table 7.1), and were not significantly different (F = 3.21; df = 2, 26; P = 0.059). Similarly, the mean SCPs after removal of outliers were not significantly different from the mean SCP of test batches in our earlier determinations (Carrillo et al. 2004a) (F = 3.65; df = 1, 6; P = 0.104).

The low mean SCPs of adult males of D. mendotae reported here (-21.60ºC) and in Carrillo et al. (2004a) (-23.11ºC) indicates that this species may be able to withstand low winter temperatures without freezing. Our previous results also showed that the SCP of males and females were not significantly different (P < 0.05), and were invariant over a period of 17 days when test individuals were aged at 6°C. Furthermore, results from the

215 present study have shown that SCPs of D. mendotae adults of different generations from the same habitat are also not significantly different. However, comparative SCP determinations among populations in streams with differing thermal regimes and among species are necessary to understand better the mechanisms of cold tolerance within this genus and other genera of cold-tolerant Diamesinae.

Table 7.1: Supercooling point (°C) (mean ± SE) of test batches of adult male Diamesa mendotae from raw data (A) and data with outliers removed (B). Values in parentheses represents minimum and maximum supercooling point values. Analysis Batch 1 Batch 2 Batch 3 -22.17 ± 0.34 -19.41 ± 0.82 -22.99 ± 0.83 A (-23.40, -20.14) (-22.38, -14.01) (-25.55, -18.52) n=10 n=9 n=10 -20.50 ± 0.45 -22.70 ± 0.87 a B − (-22.38, -19.18) (-25.36, -18.52) n=7 n=9 a No outliers were removed.

A low SCP has been interpreted as a primary mechanism for freeze-intolerant insect species to survive subfreezing temperatures in highly fluctuating environments. However, some insects experience substantial cold-induced injury at temperatures above the SCP, such as diapausing pupae of Mamestra configurata Walker (Turnock et al. 1983), which leads to reductions in survivorship. By contrast, other species that are freeze tolerant may have moderate to extreme capacity to supercool (Miller 1982, Ring 1982). Thus, it appears that it is insufficient to only measure the SCP when attempting to understand the mechanism or mechanisms that a species uses to survive subfreezing temperatures (Renault et al. 2002). Consequently, the ecological and fitness implications of SCP as a

measure of LLT50 have been questioned, and Renault et al. (2002) recommended that

SCP estimates should be coupled to determinations of LLT50.

In the present study, mean percentages of mortality were below 29% at temperatures above the mean SCP. However, test temperatures below the mean SCP (i.e., -25 and - 30ºC) resulted in 100% mortality (i.e., a 71% increase in mortality) (Fig. 7.1). In addition, 216 mortality and CPIF curves appeared to follow a similar pattern suggesting that mortality

occurred as a result of freezing (Fig. 7.1). The estimated LLT50 based on linear interpolation was -21.50°C. Curvilinear estimates ranged from -22.8°C to 21.9°C, depending on assumptions and curve fitting, but were not substantially different from the estimate based on linear assumptions. None of these estimates deviated markedly from the mean SCP of -21.60°C for adult males of D. mendotae. Therefore, based on the

concordance of the mean SCP with estimated LLT50, and the similar patterns of the mortality and CPIF curves we conclude that adult males of the population of D. mendotae in the Kinnickinnic River in Wisconsin, USA are freeze intolerant.

Studies on the cold hardiness of chironomids are few; however, freeze tolerance and freeze intolerance (e.g., supercooling and/or seeking physical protection) have been reported as strategies for surviving subfreezing temperatures (Irons et al. 1993). For example, larval populations of Belgica antarctica Jacobs have SCPs ranging from -6.2 to 217 -5.7°C, but are freeze tolerant to -15°C (Baust & Edwards 1979, Block 1982). In contrast, Baust and Edwards (1979) reported similar adult SCPs of -5.3°C, but demonstrated that field-collected adults are freeze intolerant. Similarly, larvae of Eretmoptera murphyi Schaeffer are freeze-tolerant but pupae and adults are freeze intolerant (Harrison & Block 1988, Sømme & Block 1991).

In the present study, we demonstrate that adult males of D. mendotae are freeze intolerant (Fig. 7.1); however, the strategy of survival for larvae of this species is unknown. Nevertheless, wrinkled, partially dehydrated, and presumably frozen larvae of Diamesa zernyi Edwards collected from frozen substrates of the Noce Bianco glacial stream in Italy (Lencioni 2004), suggest that freeze tolerance occurs among larvae in at least one species of Diamesa and could be a strategy used by larvae of D. mendotae. Other reports of freeze tolerance among larvae of several chironomid genera are in Scholander et al. (1953), Hinton (1960b), Leader (1962), Danks (1971b, 1981), and Olsson (1981). However, some of these reports focused on recovery of larvae from frozen sediments and it is unclear if larvae were frozen. For instance, Olsson (1981) experimentally recovered live larvae from sediments frozen in the laboratory at -4°C for five months, a temperature that matched the lowest recorded from frozen river sediments in northern Sweden, but is warmer than SCP of B. antarctica and E. murphyi.

Freeze tolerance is not commonly seen in adults of winter-active insects probably because these organisms depend on remaining active to emerge, copulate, and oviposit; therefore, a low SCP is likely advantageous for winter activity. For example, adult activity among species of Diamesa at temperatures below freezing includes emergence (Hermann et al. 1987, Nolte & Hoffman 1992a, Ferrington 2000), flying (Hågvar & Østbye 1973), swarming (Young 1969), mating and walking on snow (Hansen & Cook 1976) or within small cavities of glacial ice (Kohshima 1984, Boothroyd & Cranston 1999), and aggregating under overhanging snow along undercut stream banks (LCF personal observations). Despite their freeze intolerance, these observations spanning a breadth of activities at subfreezing air temperatures suggest that chill tolerance (i.e., the

218 ability to survive subfreezing temperatures above the SCP for extended periods of time; Bale 1996) may also be common among species of Diamesa.

In Minnesota and Wisconsin we have observed moderate to abundant densities of D. mendotae adults on snow adjacent to several groundwater-dominated streams and rivers sporadically from November through March. Water temperatures in these streams (e.g., Trout Brook, Minnesota, USA) typically vary between ≈2ºC to as much as 9°C through winter, even when minimum air temperature drops considerably below freezing (Fig. 7.2). Although we have not determined larval growth patterns and do not have quantitative emergence, it seems likely that this species may be at least bivoltine in some streams. Our assumption is further supported by Nolte and Hoffman (1992a) for Diamesa incallida (Walker) in Breitenbach where they concluded that as many as 8-10 generations could theoretically be completed annually in the thermally-buffered spring run, including four or more generations during cooler water periods. Their conclusions resulted from

219 larval growth studies and continuous oviposition, as indicated by the observation of egg masses throughout the entire duration of their 13-month field studies. They also concluded that continuous oviposition, short embryogenesis, and rapid larval growth rates in the cool, but thermally-buffered, habitat facilitates the multivoltine life cycle with emergence through winter months.

Embryogenesis and larval growth and development are largely governed by water temperature and can constrain the theoretical number of generations a species can produce in a given thermal setting. Therefore, it seems that from an evolutionary perspective the actual number of generations produced will represent a balance between embryogenesis and larval growth rates and the probability of successful mating and oviposition of adults. At the latitude of our sample site in the Kinnickinnic River, minimum air temperatures through winter fluctuate more than water temperatures of groundwater-dominated streams, and air temperatures drop to values well below zero and

220 even below the estimated LLT50 (Fig. 7.2). In the last 65 years near our sample site, the average minimum air temperatures from November to April were less than 0°C for 162 days of the year (Fig. 7.3). Therefore, for a freeze-intolerant species that is exposed to subfreezing temperatures, inability to increase its cold hardiness (e.g., by depressing its SCP) or to find a protective microhabitat will greatly increase the probability of mortality resulting from freezing of its body fluids. Thus, it is unlikely that there would be sufficient reproductive success through this interval from November to April to sustain the production of several winter maturing generations without such mechanisms. In

contrast, in a freeze-intolerant species (e.g., D. mendotae) where the adult SCP and LLT50 are low, the probability of survival for extended periods should be increased, resulting in improved opportunity to successfully mate and oviposit. Thus, a balance between growth rates as a function of water temperature and the ability of adults to withstand subfreezing air temperatures as a consequence of SCP depression will contribute to the number of successive cohorts through winter.

From Carrillo et al. (2004a) we know that adults of D. mendotae maintain constant SCPs for at least 17 days after emergence. Assuming that adults can survive short-term exposure to temperatures approaching their SCP, adults of D. mendotae can emerge and survive the 65-year average daily minimum temperature during winter months at our

sample site (Fig. 7.3). However, temperatures commonly drop below the LLT50 of adult male D. mendotae during the winter, fall, and spring (e.g., Fig. 7.2) resulting in potentially lethal conditions. Moreover, the probability of the actual minimum air temperature being less than the LLT50 for adult males of D. mendotae on any given date increase to ≈35% in January (Fig. 7.3). At higher latitudes, the daily probability of temperatures dropping below their LLT50 increases, subsequently increasing the probability of death and possibly reducing the number of generations produced through

winter. Despite a relatively high probability of temperatures dropping below the LLT50 for adult males of D. mendotae, these organisms emerge from groundwater-dominated streams at this latitude throughout the winter (LFC & RWB, personal observations). Therefore, it is possible that mortality on individuals active when temperatures are lower

221 than the LLT50 does not have an important effect on the population dynamics of this species.

Despite the clear connection between low SCPs and LLT50 and a higher probability of winter survival, several freeze-intolerant winter-active species with high SCPs have been reported to survive winter conditions (Sømme & Østbye 1969). The mechanism to survive conditions for these species may result from their ability to seek physical protection in microhabitats (e.g., under rocks and cut banks near groundwater-dominated streams) and not from their capacity to develop a physiological response to overcome the negative effects of low temperatures. Therefore, a combination of a low SCP, low LLT50, and utilization of protective microhabitats could improve the probability of this species to successfully reproduce continuously through cold months as a winter-active and multivoltine species.

ACKNOWLEDGEMENTS

We would like to thank two anonymous reviewers for critical comments on an earlier version of this manuscript. This work was partially funded by a University of Minnesota Doctoral Dissertation Fellowship (MAC) and the Minnesota Agricultural Experiment Station.

222 CHAPTER VIII

FREEZE TOLERANCE IN LARVAE OF THE WINTER-ACTIVE DIAMESA MENDOTAE MUTTKOWSKI (DIPTERA: CHIRONOMIDAE): A CONTRAST TO ADULT STRATEGY FOR SURVIVAL AT LOW TEMPERATURES

R.W. Bouchard Jr., M.A. Carrillo, S.A. Kells, & L.C. Ferrington Jr.

ABSTRACT

The winter-active Diamesa mendotae Muttkowski (Diptera: Chironomidae) is freeze intolerant in the adult stage with a low mean supercooling point (SCP) of ≈-20°C. However, cold-hardiness strategies for immatures of this species are unknown. In this study, we measured SCP values for D. mendotae larvae, pupae, and adults using surface- contact thermometry. In addition, the lower lethal temperature (LLT) was determined for the larval stage. The mean SCPs for larvae (-7.4ºC) and pupae (-9.1ºC) were relatively high compared to adults (-19.7ºC). Our results indicate that the larvae of D. mendotae

are freeze tolerant with a LLT99 (-25.4°C), ≈10°C lower than their minimum SCP (-15.6°C). Freeze tolerance in these larvae may be a strategy to provide protection from short-term exposures to ice crystals or to permit diapause within frozen substrates. The change in cold-hardiness strategy from freeze tolerant to freeze intolerant between the larval and adult stages of this species is likely a result of the different habitats occupied by these two life stages.

Bouchard, R.W. Jr., M.A. Carrillo, S.A. Kells & L.C. Ferrington Jr. 2006. Freeze tolerance in larvae of the winter-active Diamesa mendotae Muttkowski (Diptera: Chironomidae): a contrast to adult strategy for survival at low temperatures. Hydrobiologia 568:403-416.

223 INTRODUCTION

In cold-temperate regions such as Minnesota, overwintering insects must have the ability to survive periods of low temperatures. Some species avoid subzero temperatures (<0ºC) by moving to protected habitats or by migrating long distances (Lee 1989, Oswood et al. 1991, Irons et al. 1993). However, many insect species exposed to subfreezing temperatures must have physiological mechanisms to survive these conditions. There are two commonly recognized cold-hardiness strategies used by insects for surviving subzero temperatures: freeze avoidance (or intolerance) and freeze tolerance (Salt 1961, Baust & Rojas 1985, Zachariassen 1985, Bale 1989). Freeze-avoidant species can not survive freezing of body tissues and fluids, and generally get protection from freezing by supercooling and/or by seeking protection in sheltered microhabitats (Bale 1989). Supercooling is the ability to maintain fluids in the liquid state below their normal freezing point, and the supercooling point (SCP) is the temperature at which spontaneous freezing occurs (Sømme 1982, Zachariassen & Kristiansen 2000). Freeze-avoidant species usually lower their SCP by synthesizing cryoprotectants (e.g. polyols, sugars, and antifreeze proteins) or by removing ice nucleating agents (INAs) through mechanisms such as gut clearing (Zachariassen 1985, Bale 1989, Lee 1989). In contrast, freeze- tolerant species can survive extracellular ice formation by using INAs that increase the SCP to promote a controlled freezing process at high subzero temperatures (Zachariassen 1985, Lee 1989, Block 1991). Slow and controlled extracellular ice formation prevents internal cell damage and provides enough time to make metabolic adjustments allowing these species to withstand freezing (Block 1991). Similar to freeze-avoidant species, freeze-tolerant species also use cryoprotectants such as polyols and sugars; however, the function of these compounds is not to depress the SCP, but to prevent intracellular freezing, decrease intracellular fluid loss, and to protect macromolecules (Storey & Storey 1997).

Most cold-hardiness studies involve terrestrial insects or the terrestrial stages of aquatic insects (e.g., Carrillo et al. 2004a, Bouchard et al. 2006a). Limited studies on aquatic

224 stages are likely a result of the assumption that insects in an aqueous environment face less extreme conditions because of thermal buffering from groundwater inputs, the high specific heat of water, and insulation from surface ice (Frisbie & Lee 1997, Lencioni 2004). In addition, most studies on the freeze tolerance of aquatic insects have focused on lentic rather than lotic habitats (Irons et al. 1993). This is probably because insects in lentic habitats are more likely to experience predictable freezing. However, many lotic habitats also freeze at the surface or along margins, although the extent may be less predictable. Aquatic insects may be subjected to supercooled flowing waters resulting in exposure to ice-crystal formation (i.e. frazil ice) within the supercooled fluid (Ashton 1979, Oswood et al. 1991). Invertebrates in intermittent habitats may also be exposed to subzero temperatures and external ice (Frisbie & Lee 1997).

Most cold-hardy aquatic insects are inactive at subzero temperatures, but some species are able to maintain activity (e.g. crawling, flying, mating, and ovipositing) under these conditions. There are several examples of activity at subzero temperatures in the genus Diamesa Meigen (Diptera: Chironomidae). Diamesa arctica (Boheman) has been observed swarming at temperatures of -0.4°C (Young 1969). Kohshima (1984) reported an unknown Diamesa species in the Nepal Himalayas that was active at -16°C. This species was subsequently determined to be two species, Diamesa kohshimai Sæther & Willassen and Diamesa yalavia Sæther & Willassen. In Minnesota, Diamesa mendotae Muttkowski is a common winter-active chironomid that can be observed crawling across the snow along stream banks at subzero temperatures (Carrillo et al. 2004a). Previous studies have estimated that D. mendotae adults have a mean SCP of ≈-20°C, a value lower than that reported for other winter-active species (Sømme & Østbye 1969, Baust & Edwards 1979). Mortality of D. mendotae adults occurs approximately at its SCP (Bouchard et al. 2006a), indicating that this species is freeze intolerant and low SCPs likely confer enough protection to maintain winter activity.

In contrast to adults, the few cold hardiness studies involving larvae of other chironomid species suggest that this life stage uses freeze tolerance as a strategy for survival at

225 subzero temperatures (e.g. Scholander et al. 1953, Danks 1971b, Andrews & Rigler 1985, Lencioni 2004). The ability to tolerate freezing is often based on the premise that these larvae have been extracted from frozen sediments, although the state of their body fluids was not directly determined (Frisbie & Lee 1997). Determinations of SCPs in chironomid larvae indicate that most freeze at temperatures between ≈-11 and -5°C (Danks 1971b, Baust & Edwards 1979). In some cases, the sediments where larvae are extracted can be buffered and may not reach temperatures lower than the SCP of the larvae. However, some studies have identified viable chironomid larvae recovered from lake sediments where temperatures as low as -18 to -20°C were recorded (Scholander et al. 1953, Andrews & Rigler 1985). In addition, viable larvae were obtained from sediments frozen to -18°C in the laboratory (Danks 1971b). Therefore, it is reasonable to assume that some of these larvae were freezing, but research is needed to provide direct evidence for the survival mechanisms of chironomid larvae in response to freezing conditions.

In the present study, we hypothesized that D. mendotae larvae are freeze tolerant and have relatively high SCPs. Because the pupal stage is non-feeding and transitional, we also hypothesized that pupal SCPs are lower than larval SCPs and closer to values reported for adults (Carrillo et al. 2004a, Bouchard et al. 2006a). To test these hypotheses, we estimated the SCPs of larvae, pupae, and adults of D. mendotae. Adult SCPs were determined to confirm previously published estimates, and to directly compare them with values for larvae and pupae collected from the same location at the same time. In addition to SCPs, we determined the lower lethal temperatures (LLT) for larvae. However, LLT experiments for pupae were not conducted as a result of the low number of individuals collected. To our knowledge, this is first time SCPs are reported for aquatic pupae and the first time SCPs and LLTs are reported for larvae in the subfamily Diamesinae. Our results were compared to those obtained from previous experiments with D. mendotae adults (Carrillo et al. 2004a, Bouchard et al. 2006a). Finally, we discuss the ecological implications of cold-hardiness strategies used by this species to survive exposure to low temperatures.

226 MATERIALS AND METHODS

COLLECTION AND TREATMENT OF TEST SPECIMENS Larvae, pupae, and adults of D. mendotae were collected on 17, 23, and 26 February 2005 from Trout Brook northeast of Cannon Falls, Minnesota, USA (44.5453°N, 92.8057°W). Fourth-instar larvae and pupae were collected from riffles using a D-frame dip net with 500-µm mesh. Detritus and macroinvertebrates were placed in freezer bags and stored in a cooler containing snow until returned to the laboratory. Once in the laboratory, material was poured out into a white plastic pan and D. mendotae larvae and pupae were picked from the detritus using soft forceps. Attempts were made to collect only large, active larvae to insure that healthy, late instars were tested. Specimens were picked into 118-ml snap-top jars with damp paper towels and held in a programmable growth chamber (Percival Scientific, Inc., Perry, IA) at 6°C until needed. Adults were located on the snow alongside the stream and scooped into individual 3.7-ml snap-top vials with a small amount of snow to prevent dehydration. The vials were stored in a cooler containing snow until they arrived at the laboratory and were held in a growth chamber at 6°C until needed. All experiments were run within 72 h of specimen collection, but most were tested <24 h post collection.

DETERMINATION OF SCPS FOR LARVAE, PUPAE, AND ADULTS For larvae and pupae, specimens were removed from the 118-ml snap-top jars using soft forceps, transferred into a white plastic pan with a small amount of water, and then placed briefly on a paper towel to remove excess external moisture. For adults, specimens were removed from each 3.7-ml snap-top vial by upturning the vial on a piece of paper towel. Although vials contained small amounts of water from thawed snow, there was less concern about removing external moisture from the adults due to their hydrophobic cuticle. Methods for determining SCPs followed Carrillo et al. (2004b). Specimens were attached to 24-gauge copper-constantan thermocouples using a thin layer of high-vacuum grease (Dow Corning®, Dow Corning Corporation, Midland, MI, USA). Two insect-thermocouple arrangements were placed at the center of a polystyrene

227 container (0.19 × 0.19 × 0.19 m) with a starting temperature of ≈0°C and closed with a rubber stopper. This was repeated for a total of five containers (i.e. 10 total insects per trial) and placed simultaneously into a -80°C freezer (Revco Scientific Inc., Asheville, NC, USA) to cool insects at a rate of ≈1°C min-1. Temperatures were recorded at 1-s intervals with a multichannel datalogger (Personal Daq/56 data acquisition system IOtech Inc., Cleveland OH). Specimens were cooled to at least -30°C before the containers were removed from the freezer. Specimens were removed from the thermocouple and preserved in 70% ETOH in individual 3.7-ml snap-top vials and labeled such that each individual specimen could be later matched with their corresponding SCP. The SCPs were estimated as the lowest temperature reached before the release of the latent heat of fusion resulting from the freezing of the insect’s body fluids (Lee 1991). The SCPs of larvae and pupae were measured together during the same trials; however adults were tested in separate trials. The SCPs were obtained for a total of 83 larvae, 24 pupae, and 46 adults (153 total) by repeating this experiment 18 times. Supercooling points were not obtained for all 180 specimens tested as thermocouples did not maintain good contact with ≈15% of the insects.

MEASUREMENT OF LARVAL INSTAR AND PHASE Instar and developmental phase were determined for larvae preserved from the SCP experiments. To determine instar, an ocular micrometer (Nikon) in a dissecting microscope (Olympus SZX12) was used to measure larval head capsule width and length to the nearest 0.01 mm. Head length was measured from the post occiput to the frontoclypeal margin. Head width was measured at the widest part of the head. Larval development was estimated with a dissecting microscope for fourth instar larvae using phases simplified from Wülker & Götz (1968). Each individual larva was assigned to one of three categories: 1) small imaginal discs without well-defined mesothoracic leg sheaths (leg sheath differentiation may be beginning or completely absent) (L1), 2) well- defined mesothoracic leg sheaths widely separated ventrally (L2), and 3) thorax greatly swollen with large, obvious mesothoracic leg sheaths meeting or nearly meeting ventrally (L3) (Figure 8.1).

228 Figure 8.1: Diamesa mendotae larval developmental phases. msth lsh = developing mesothoracic leg sheaths; L1: mesothoracic leg sheaths not obvious or only beginning to differentiate, L2: large, obvious mesothoracic leg sheaths, and L3: mesothoracic leg sheaths touching or nearly touching ventrally with thorax greatly swollen.

EFFECT OF SHORT-TERM EXPOSURE TO SUBZERO TEMPERATURES ON LARVAL

MORTALITY A method modified from Koch et al. (2004) was used to determine LLTs. Foam plugs were placed midway down 16 x 150-mm glass test tubes and a filter paper (Whatman grade 1 55-mm circle) was inserted into each test tube so that they rested against the foam insert. Larvae were removed from the 6ºC chamber and placed in a white plastic pan with a small amount of water. Larvae were picked using soft forceps and placed briefly on a dry paper towel to remove as much external water as possible. Prior testing indicated no mortality when larvae (n = 10) were placed on dry filter paper, maintained at 6°C for 1 h, and then rewetted. Temperatures selected for treatment were -5, -10, -15, -20, -25, and -30°C. For each temperature treatment, 10 larvae were placed into each test tube and a 24-gauge copper-constantan thermocouple was positioned in close proximity to the larvae. An additional 10 larvae were added to another test tube and maintained at 6°C as a control during the period the other treatments were tested. Any mortality in the 229 control treatments was used as a correction factor for the subzero temperature treatments (Abbott 1925). The test tubes were then placed into a 0.35 x 0.35 x 0.35-m polystyrene container with an initial temperature of ≈6°C. This initial temperature was attained by placing the container with a starting temperature of ≈25°C into a -80°C freezer until the center of the container reached ≈6ºC. The container with test tubes inside was plugged and then returned to the -80°C freezer. The size of the container maintained a rate of cooling of ≈0.3-0.4°C min-1. Individual test tubes were cooled to designated temperatures of -5, -10, -15, -20, -25, and -30°C and removed from the container 1 min after the target temperature had been achieved. This experiment was repeated five times although not all six temperature treatments were tested in each experiment, but each temperature was tested at least three times. After the larvae were removed from the container, the filter paper was transferred to a 55-mm Petri dish containing ≈2 ml of stream water, and placed into the 6°C chamber. At this time, the control specimens were treated similarly. The condition of larvae was noted immediately after removal from the container and then 24 h after removal. Survivorship (i.e. lack of moribundity) was assessed as the number of larvae active or moving when prodded with forceps 24 h post treatment.

To understand the effect of ice formation on larval mortality, the cumulative proportion of individuals freezing (CPIF) was directly compared to the mean proportion of mortality that resulted after short-term exposure to subzero temperatures. For a valid comparison between parameters, specimens from the same collection were used in both experiments and larval SCP experiments (n = 83; see “Determination of SCPs for larvae, pupae, and adults”) were performed shortly before or after mortality experiments. From larval SCPs, the CPIF was determined by calculating the number of individuals freezing at or above each one-degree temperature step and dividing this sum by the total number of individuals tested (Koch et al. 2004, Carrillo et al. 2005a, Bouchard et al. 2006a). The CPIF was then plotted against the mean proportion of mortality of the six treatments from the short-term exposure experiments.

230 LONG-TERM MORTALITY AFTER SHORT-TERM EXPOSURE TO SUBZERO TEMPERATURES To determine if freeze or prefreeze injury occurred and caused long-term mortality, larvae that survived short-term exposure to subzero temperatures and the control treatments were maintained in Petri dishes in a growth chamber at 6°C. Survivorship and life stage were monitored daily. Emerging adults were removed and placed in individual 3.7-ml vials with a small amount of stream water and monitored daily. Specimens were assessed as alive as previously explained. Any larvae, pupae, or adults that died overnight were removed and preserved in 70% ETOH. The experiment was terminated after 147 d.

ANALYSIS OF DATA SCPs Box plots were created to visualize the distribution of SCPs at different developmental stages (i.e. L1, L2, L3, pupa, and adult) (SigmaPlot 2001). Differences among the SCPs for the three larval phases and the pupae were tested using a one-way analysis of variance (ANOVA) (PROC GLM; SAS Institute, 1998) after applying an x0.25 transformation as recommended by the Box-Cox procedure (MacAnova 2002). When significant differences were observed at P ≤ 0.05, Tukey’s Studentized Range Test was used to separate the means (SAS Institute, 1998). The same procedures and transformation were used for a second time to identify any differences in SCPs among the three different larval phases, pupae, and adults despite the fact that SCPs for adults were measured in different trials. Because adult SCPs were measured in different trials, a statistical comparison may not be considered to be valid by some readers; however, previous experiments showed similar SCP values for this developmental stage (Carrillo et al. 2004a, Bouchard et al. 2006a). Therefore, it is unlikely that the adult SCPs were an artifact of the procedure or time of measurement.

Mortality An ANOVA (PROC GLM; SAS Institute 1998) on the arcsine-square-root transformed proportion of mortality (Southwood & Henderson 2000) was performed to test for

231 significant differences in mortality among temperature treatments for the experiments on short-term exposure to subzero temperatures. When significant differences were identified (P ≤ 0.05), Tukey’s Studentized Range Test was used to separate the means. Probit analysis (PROC PROBIT; SAS Institute 1998) was used to estimate the temperature (95% fiducial limits, FL) at which 50 and 99% of mortality occurred (i.e.,

LLT50 and LLT99). For specimens surviving short-term exposure to subzero temperatures, the arcsine-square-root transformed proportion of larvae pupating and emerging were tested for significance between treatments using an ANOVA (PROC GLM; SAS Institute 1998).

RESULTS

232 LARVAL INSTAR AND PHASE The relationship between larval head capsule width and length formed a single cluster suggesting that all larvae were the same instar (Figure 8.2). One specimen had a short head length compared to the other larvae, but this specimen was shriveled and had relatively large imaginal discs and therefore, was not likely to be an earlier instar. We can conclude that all of the larvae tested for SCPs and likely all of the larvae tested for LLTs were in the fourth and final instar. The number of specimens in each larval phase was similar with 28.1, 39.0, and 32.9% for the L1, L2, and L3 phases, respectively (Figure 8.3).

233 SCPS Overall, D. mendotae larvae (n = 83) had a SCP (mean ± SE) of -7.4 ± 0.26°C with values ranging from -15.6 to -2.3ºC (Figure 8.3). Diamesa mendotae pupae (n = 24) had a comparable SCP (mean ± SE) of -9.1 ± 0.58°C with values ranging from -15.7 to -4.3ºC (Figure 8.3). Adults (n = 46) had a SCP (mean ± SE) of -19.7 ± 0.58°C with values ranging from -25.8 to -10.5ºC (Figure 8.3). Significant differences were found in the mean SCP among phases of development when including larvae and pupae (F = 7.49; df = 3, 102; P = 0.0001) and when including all developmental stages (F = 114.54; df = 4, 147; P < 0.0001) (Figure 8.3). Mean SCPs for immature stages were relatively high with a steady decrease in the mean as the insect matures and then a large decrease in SCP in the adults (Figure 8.3). Mean SCPs for phase L1 and L2 larvae were significantly higher than mean values for L3 larvae and pupae (Figure 8.3). Adult mean SCPs were significantly lower than those of all other developmental stages. There was also a greater amount of variability in the SCP of phase L3 larvae, pupae, and adults (Figure 8.3).

EFFECT OF SHORT-TERM EXPOSURE TO SUBZERO TEMPERATURES ON LARVAL

MORTALITY Significant differences in the proportion of larval mortality among subzero temperatures were observed (F = 40.90; df = 5, 19; P < 0.0001) (Figure 8.4). A significant increase in mortality of ≈61% occurred when the temperature dropped from -15 to -20ºC. In contrast, larval mortality was not significantly different at temperatures above -15ºC and remained below ≈22%, despite a mean SCP of only ≈-7ºC (see SCPs results). By plotting the proportion of mortality with the CPIF, a separation between these curves was apparent (Figure 8.4), indicating that D. mendotae larvae froze at higher temperatures

than those causing mortality. The larval LLT50, and LLT99, and their respective 95% FL

were -17.4°C (-18.7, -16.2) and -25.4°C (-30.3, -23.0), respectively. The LLT50 occurred

10ºC below the mean SCP for larvae (-7.4°C). The LLT99 occurred ≈10ºC below the minimum SCP recorded for D. mendotae larvae (i.e. -15.6ºC).

234 Figure 8.4: Corrected mean proportion of mortality (± SE) (closed circles) and cumulative proportion (± SE) of Diamesa mendotae larvae freezing (open circles, CPIF) after short- term exposure (1 min) to subzero temperatures. Each mean proportion of mortality represents 31 to 50 individuals in 3 to 5 independent replicates. Mean proportions of mortality for all stages followed by different lowercase letters are significantly different (P < 0.05) as determined by Tukey’s Studentized Range Test.

LONG-TERM MORTALITY AFTER SHORT-TERM EXPOSURE TO SUBZERO TEMPERATURES Survivorship curves were similar for the 6, -5, -10, -15, and -20°C treatments with most post-experiment mortality occurring within 30 d after the experiment (Figure 8.5). However, two larvae survived 143 and 147 days until the experiment was terminated in the -15 and -5°C treatments, respectively. Of specimens surviving the short-term exposure to subzero temperature experiments, including the control group (n = 185), 71% pupated and 16% emerged as adults. Numerical differences in pupation rates were observed; with treatments 6 and -5°C ranging from 62 to 64% compared to the -10, -15, and -20°C treatments which ranged from 75 to 82% pupation. However, no significant differences (F = 1.36; df = 4, 16; P = 0.2905) were observed between the proportion of larvae pupating for specimens cooled below (i.e. -10, -15, -20°C) and above (i.e. 6, -5°C) the mean SCP for larvae (Figure 8.6). In contrast, significant differences were observed for adult emergence rate across treatment groups (F = 6.02; df = 4, 16; P = 0.0038); however, no particular pattern among treatments was apparent (Figure 8.6). 235 Figure 8.5: Long-term survivorship of Diamesa mendotae larvae after surviving short- term exposure to subzero temperatures. Survivorship curves for -25 and -30°C treatments were not included due to low post-experiment survival (0-2.5%). Each treatment consists of 7 to 50 surviving individuals combined from 2 to 5 replicates.

236 DISCUSSION

SCPS The mean SCP for adults in this study was similar to values previously reported for D. mendotae (Carrillo et al. 2004a, Bouchard et al. 2006a). A mean SCP of -19.7°C for the Trout Brook population was numerically similar to the mean SCP of -21.6°C recorded from another population of D. mendotae in Wisconsin. These SCPs were lower than values determined for other winter-active insects (Sømme & Østbye 1969). Bouchard et al. (2006a) reported freeze intolerance in D. mendotae adults and hypothesized that low SCPs allowed these insects to survive and remain active at subzero temperatures.

The estimated mean SCP for larvae (-7.4ºC) was significantly higher (≈12ºC) than values determined for adults of this species. In aquatic insects, this pattern of higher SCPs in larvae compared to the adults has been identified previously (Moore & Lee 1991). These larval SCPs are comparable to SCPs of Chironomidae larvae using different experimental approaches (Danks 1971b, Baust & Edwards 1979, Oswood et al. 1991, Frisbie & Lee 1997). One exception to this is the marine midge, Paraclunio alaskensis Coquillett, where mean SCPs as low as -14.2ºC were reported by Ring (1989). For the pupal stage, the relatively high mean SCP (-9.1ºC) contradicted our original hypothesis, being closer to the larval stage than to the mean SCP for the adult stage. This result contrasts with reports of pupal SCPs for the majority of terrestrial insects, which generally have SCPs below -20ºC (Sømme 1982, Bale 1989). The low pupal SCPs are generally attributed to the fact that the pupa is a non-feeding stage with fewer INAs compared to the feeding stages. To our knowledge, this is the first measurement of SCPs in an aquatic pupa, and the lack of information for this stage may be a result of the fact that most aquatic insects overwinter as eggs or larvae (Moore & Lee 1991); therefore, the generally short-lived pupal stage does not experience long-term exposure to low temperatures. However, in the case of D. mendotae, pupae may be exposed to ice crystals and/or subzero temperatures as a result of the winter emergence of this species.

237 The determination of SCPs for different developmental stages of D. mendotae indicates this insect begins to decrease its mean SCP late in the larval stage into the pupal stage (Figure 8.3). The mean SCP drops abruptly from -9.1 to -19.7ºC between the pupal and adult stages; however, it is not clear at which stage this large decrease in the SCP occurs. The lower mean and minimum SCP values through development suggest that there could be physiological preparation late in the fourth instar and pupa for the transition to increased supercooling capabilities in the adult stage.

EVIDENCE FOR LARVAL FREEZE TOLERANCE According to Bale (1991), freeze tolerance or intolerance for an organism can be determined by exposing insects to temperatures above and below the mean SCP for a given species. By comparing the cumulative proportion at which individuals freeze (i.e., CPIF) to the rate of mortality at different subzero temperatures, one can determine if a species is tolerant or intolerant of freezing. For D. mendotae, comparison of the proportion of mortality and CPIF curves (Figure 8.4), indicated that mortality occurred well below larval SCP temperatures and that these larvae were freeze tolerant. However, this relationship between mortality and CPIF alone is not a determinant of freeze tolerance (Koch et al. 2004, Carrillo et al. 2005a). For example, in Koch et al. (2004) and Carrillo et al. (2005a) a similar gap between these two curves was observed for the freeze-intolerant Harmonia axyridis (Pallas) (Coleoptera: Coccinellidae) and Plodia interpunctella (Hübner) (Lepidoptera: Pyralidae). However, in these studies, differences between the two curves were attributed to short exposures (i.e. 1 min) to subzero temperatures and the relatively large mass of the specimens resulting in some specimens not fully freezing. Nevertheless, these insects generally had 100% mortality close to their 100% CPIF despite their large mass. According to Salt (1953) survival of freeze- intolerant species after minimum exposure (e.g. 1 min) to temperatures corresponding to their SCP may be an artifact of laboratory experiments. In addition, it has been demonstrated for freeze-intolerant insects that the proportion of mortality increases with time at their SCP (Block et al. 1988, Koch et al. 2004, Carrillo et al. 2005a).

238 For smaller, freeze-intolerant organisms such as D. mendotae adults (Bouchard et al. 2006a) and other species like the adults of the parasitoid Habrobracon hebetor (Say) (Hymenoptera: Braconidae) (Carrillo et al. 2005b), the mortality and CPIF curves were much closer compared to either H. axyridis or P. interpunctella. For D. mendotae larvae, the fourth instar is presumably similar in mass to the adult and we would have expected closer mortality and CPIF curves if the stage was freeze intolerant. Even for freeze- intolerant insects with greater body mass, like H. axyridis and P. interpunctella, their

LLT99 occurred within only 2 to 5ºC below the temperature at which the CPIF curve reached 100% (Koch et al. 2004, Carrillo et al. 2005a). In contrast, the LLT99 for D. mendotae larvae occurred at ≈10ºC below the temperature at which the CPIF curve reached 100%, and therefore we conclude that D. mendotae larvae are freeze tolerant and not an artifact of the methodology.

LONG-TERM MORTALITY AFTER SHORT-TERM EXPOSURE TO SUBZERO TEMPERATURES Although insects may be freeze tolerant, mortality will still occur at temperatures below the SCP (Sinclair 1999). Most freeze-tolerant organisms can convert up to 65% of the body water into ice without negative effects (Storey & Storey 1997). An excessive and uncontrolled water loss in the body causes an increase in solute concentration and cell shrinkage that may result in cell injury and death (Lee 1989). In general, the probability of mortality increases at lower temperatures below the SCP. Although we observed a similar pattern of increasing mortality at lower temperatures, larvae surviving a short exposure to different temperatures above or below their SCP exhibited a similar long- term survival (Figure 8.5). If irreversible freeze injury occurred, the effect was measurable within 24 h post exposure. Therefore, we can conclude that the assessment of mortality 24 h post treatment was sufficient to determine mortality rates resulting from subzero-temperature exposure.

An unexpected result, although not significant, was the numerically greater pupation rates of specimens cooled below the mean SCP compared to those cooled above this temperature (Figure 8.6). However, there was no pattern for adult emergence. These

239 results are difficult to explain without further investigation. In addition, we must note that these specimens were not maintained under natural conditions in the laboratory and lacked photoperiod and thermal cues; which potentially altered their behavior. However, relatively high pupation and emergence rates in a laboratory setting are interesting as they may prove to be an effective method for rearing specimens to obtain associated material for cold-adapted chironomid species. A similar methodology is described for rearing Simuliidae from mature larvae (Adler et al. 2004).

ECOLOGICAL IMPLICATIONS OF FREEZE TOLERANCE IN D. MENDOTAE LARVAE The larvae of most aquatic insects are not generally exposed to low subzero temperatures as a result of the thermal buffering from water or the ability to find safe refugia (e.g. hyporheic zone) during periods when the surrounding water may freeze (Moore & Lee 1991, Irons et al. 1993, Lencioni 2004). Therefore, the ability of D. mendotae to survive the freezing of body fluids may be an evolutionarily conserved characteristic widespread in many chironomid subfamilies, including the Diamesinae, which are considered to have diversified in cold-water habitats (Brundin 1966). In fact, Baust & Lee (1981) concluded that freeze tolerance in Belgica antarctica Jacobs was evolutionarily conserved because the larvae do not appear to be exposed to low subzero temperatures. Other species of Diamesa seem to be freeze tolerant (e.g. D. zernyi Edwards; Lencioni 2004) and this trait has not been lost in D. mendotae although it may not be as critically important in thermally-buffered habitats such as groundwater-influenced streams. In surface-water dominated or intermittent streams experiencing ice-crystal formation, D. mendotae may migrate into the buffered hyporheic zone to avoid ice inoculation. The ability to tolerate freezing conditions in D. mendotae larvae may improve their ability to survive in variable or unpredictable habitats. In D. mendotae larvae, freeze tolerance could be a mechanism to 1) remain active in habitats that can undergo short periods of freezing temperatures or 2) enter diapause during long periods of low temperatures.

240 Winter activity The ability to tolerate freezing may be advantageous for organisms active during cold periods in an aquatic environment. Evidence suggests that some winter-active species occurring in streams can undergo development during the winter. Berg & Hellenthal (1991) reported that Diamesa nivoriunda (Fitch), a species developing from autumn to spring, contributed 33.9% to the total chironomid production. Nolte & Hoffman (1992a) determined that Diamesa incallida (Walker) can undergo 7 to 8 generations per year despite low water temperatures. In Trout Brook, D. mendotae may also be multivoltine and contribute greatly to secondary production as water temperatures are low in the winter, but remain above freezing year round (Bouchard et al. 2006a) due to groundwater inputs. However, D. mendotae has also been collected from streams that are not greatly buffered by groundwater (Bouchard unpublished data) and may experience supercooling conditions increasing the risk of spontaneous ice inoculation in unprotected aquatic insects. Supercooling of water occurs in areas of swift flow or turbulence and can result in the formation of ice crystals called frazil and anchor ice (Oswood et al. 1991; Lencioni 2004). Frazil and anchor ice can be dangerous to stream invertebrates as it might cause ice inoculation even if water temperatures are above the SCP of the organism. In laboratory experiments, chironomid larvae exposed to external water (e.g. in contact with wet filter paper) have much higher SCPs than those free of external water (Danks 1971b, Frisbie & Lee 1997); therefore, the ability to withstand these short-term exposures to frazil or anchor ice may be beneficial. This capability would permit species to survive and remain active during cold periods without requiring migration into the hyporheic zone or the construction of winter cocoons. Such mechanisms may allow these species to be flexible and take advantage of these habitats earlier in the spring or during warm periods in the winter.

Diapause Freeze tolerance can permit chironomid larvae to diapause during relatively long periods of low temperatures when conditions are not suitable for activity. Many chironomid species have been determined to use diapause in arctic ponds (Scholander et al. 1953,

241 Danks 1971b) and glacial streams (Lencioni 2004). For example, Lencioni (2004) collected apparently-frozen D. zernyi from frozen glacial streams which could then be thawed and revived in the laboratory. Despite surface or edge ice, liquid water is usually present in most perennial, surface-water dominated streams in temperature zones, but the shallow margins of these habitats can experience freezing (Olsson et al. 1981). Intermittent streams may experience more widespread freezing conditions depending on the depth of moisture and freezing in hyporheic during the winter. In Minnesota, D. mendotae has been collected from surface-water dominated streams and other Diamesa species have been collected from intermittent streams. In these habitats, Diamesa larvae could be exposed to freezing substrates (Bouchard unpublished data) and experience long-term freezing. In these conditions, the ability to overwinter by diapausing as larvae could allow survival in freezing conditions and increase the range of habitats available to D. mendotae. However, further research is needed to determine if D. mendotae larvae occupy such marginal habitats during the winter and if diapause is a strategy for winter survival.

LARVA TO ADULT: CHANGING COLD-HARDINESS STRATEGIES Based on previous and current results, it is clear that D. mendotae uses two different cold hardiness strategies through its life cycle. According to Sømme (1982), so far all arthropod eggs tested have been freeze intolerant; therefore, we presume this to be the case for D. mendotae eggs. From our current results we classify larvae of D. mendotae as freeze tolerant, whereas Bouchard et al. (2006a) classified the adult stage as freeze intolerant. A similar shift in strategies across life stages has also been observed in the Antarctic chironomid, B. antarctica (Baust & Edwards 1979); however adult D. mendotae have a much lower SCP. In D. mendotae, both the larval and adult life stages appear to be well-adapted to cold temperatures, but shifts from one strategy to another seem to be the result of the different environments in which these life stages inhabit.

The larval stage presumably remains in contact with water and potentially ice crystals and is more likely to experience freezing as a result of ice inoculation. A freeze-avoidant

242 strategy using lower SCPs would be less effective in an aqueous environment where inoculation from external ice crystals could result in the freezing of body fluids and consequently death (Moore & Lee 1991, Frisbie & Lee 1997). Therefore, a freeze- tolerance strategy with high SCPs may allow these larvae to survive short-term exposures to ice crystals or diapause for longer periods during these conditions. The physiological mechanisms increasing SCPs in this stage is unknown; however, as a feeding stage it is more likely to possess greater amounts of INAs (Salt 1953, Leather et al. 1993). A possible production of cryoprotectants, as reported for the larvae of B. antarctica (Baust & Edwards 1979), could aid in protecting the cells during the freezing of extracellular fluids.

All of the literature on winter diapause in the Chironomidae focuses on the larvae and there is no evidence that Chironomidae diapause as pupae or adults. In general, the pupal stage lasts less than 72 h (Langton 1995), although pupal stage duration has not been tested for D. mendotae under natural conditions and may require longer periods as a result of low water temperatures during the winter. Nolte & Hoffman (1992a) determined that D. incallida pupae required 72 h at 8°C for eclosion. The short pupal stage does not appear to be utilized as an overwintering stage in Chironomidae as it is in many terrestrial insects. We had hypothesized, based on the literature available for terrestrial insects, that the pupa would have SCPs closer to the adults. Whether the pupa stage is freeze tolerant or intolerant is still unknown, but the high SCPs and the aquatic habitat shared with the larvae suggests that freeze tolerance may be a better strategy due to the threat of freezing body fluids through ice nucleation.

For D. mendotae adults, the ability to greatly lower their SCP allows them to be active in subzero conditions, a requirement for mating and ovipositing during the winter (Carrillo et al. 2004a). The physiological mechanisms underlying the ability of D. mendotae adults to depress SCP to temperatures of ≈-20ºC is also unknown. However, these adults are presumably non-feeding and therefore could potentially have fewer INAs in their bodies. According to Zachariassen et al. (2004), the removal of all INAs in an insect’s

243 body can reduce their SCP by as much as 10ºC. The production of antifreeze compounds such as polyols, sugars, thermal hysteresis proteins has also been implicated in lowering SCPs in many freeze-avoidant species (Leather et al. 1993) and can not be excluded as a mechanism in D. mendotae adults.

The present study increases our understanding of the cold hardiness of D. mendotae. The ability of D. mendotae larvae to tolerate freezing may increase the range of habitats this species can occupy (e.g. groundwater dominated and surface-water dominated streams). More research is needed to determine how the distribution and success of this species in different habitats is regulated by cold-hardiness mechanisms in all life stages of this insect. Our results, together with those of Bouchard et al. (2006a), provide strong evidence for a change in strategies between the larval and adult stages. However, it is not clear which physiological mechanisms are used to acquire and maintain cold hardiness in this species, or when the change in strategies occurs between life stages. Future research needs to determine the presence, effect, and timing of cryoprotectants, INAs, and gut contents for larvae, pupae, and adults to assess if these mechanisms play a role in cold hardiness of this species.

ACKNOWLEDGEMENTS

244 APPENDICES 245

Creek, TR=TroutBrook,BRBrownsMIMillStream, (Ground-water dominatedstreams- Creek,CR=CreditRiver,RORock Cedar Creek,CH=Chub VA =ValleyCreek;Surface-water

Appendix A:

Creek, RU=RushSUSunriseRiver).

for eachstudysiteandtotals.

Abundance ofindividual

246

EA =EagleCreek,PNPine dominated streams-CE=

Chironomidaetaxa

TAXON EA PN TR BR MI VA GWD CE CH CR RO RU SU SWD ALL TANYPODINAE Ablabesmyia idei 1 1 1 Ablabesmyia mallochi 1 1 117 67 1 8 10 4 207 208 Ablabesmyia monilis 65 34 1 100 100 Ablabesmyia peleensis 1 1 1 Conchapelopia fasciata 2 4 2 5 13 12 25 2 6 1 46 59 Conchapelopia rurika 4 11 2 17 26 9 6 1 42 59 Conchapelopia telema 5 1 1 7 7 Hayesomyia senata 8 3 11 11 Helopelopia cornuticaudata 6 2 1 9 9

247 Labrundinia maculata 2 1 3 3 Labrundinia neopilosella 1 1 2 3 5 6 Labrundinia pilosella 1 1 23 66 6 3 2 100 101 Larsia sp. 40 10 1 51 51 Meropelopia sp. 1 10 3 13 13 Natarsia sp. 1 7 8 8 Nilotanypus fimbriatus 39 43 86 52 19 239 239 Paramerina sp. 5 1 6 10 1 1 1 13 19 Pentaneura sp. 9 3 2 14 14 Procladius bellus 4 4 4 Procladius sp. 1 147 8 5 2 162 162 Procladius sp. 2 3 3 3 Psectrotanypus sp. 1 1 1

TAXON EA PN TR BR MI VA GWD CE CH CR RO RU SU SWD ALL TANYPODINAE (Cont.) Radotanypus sp. 2 2 2 Tanypus sp. 1 1 1 Zavrelimyia sp. 14 8 6 2 30 1 1 31 DIAMESINAE Diamesa sp. 82 619 2070 113 133 756 3773 1 4 376 15 12 84 492 4265 Pagastia orthogonia 5 106 111 1 1 112 Potthastia sp. 1 9 9 9 Potthastia sp. 2 4 3 7 7 PRODIAMESINAE

248 Odontomesa fulva 70 464 320 14 38 50 956 22 2 1 25 981 Prodiamesa olivacea 88 288 24 4 15 12 431 2 2 433 ORTHOCLADIINAE Acricotopus sp. 2 1 3 36 2 38 41 Brillia flavifrons 6 56 40 9 1 4 116 5 43 7 4 6 65 181 Bryophaenocladius sp. 3 3 3 Cardiocladius albiplumus 10 27 37 1 1 2 4 41 Cardiocladius sp. 1 1 4 7 12 9 3 100 112 124 Chaetocladius piger gr. sp. 2 2 15 17 54 54 71 Chaetocladius dentiforceps gr. sp. 3 11 1 2 14 2 2 16 Chaetocladius sp. 5 1 1 27 27 28 Chaetocladius sp. 6 1 4 5 5 Corynoneura sp. 1 166 1991 866 118 560 89 3790 117 196 1141 216 95 52 1817 5607

TAXON EA PN TR BR MI VA GWD CE CH CR RO RU SU SWD ALL ORTHOCLADIINAE (Cont.) Corynoneura sp. 2 26 67 144 237 237 Corynoneura sp. 3 34 34 138 46 165 34 6 389 423 Corynoneura sp. 4 5 4 339 21987 654 29 368 53 149 3 602 1256 Corynoneura sp. 5 13 13 225 8 233 246 Corynoneura sp. 6 2 4 6 6 Cricotopus (C.) bicinctus 16 184 112 57 827 1196 35 481 548 129 51 118 1362 2558 Cricotopus (C.) trifascia 1 291 64 7 2 40 405 1 59 260 132 42 62 556 961 Cricotopus (C.) vierriensis 38 44 24 171 277 6 111 7 124 401 Cricotopus (C.) sp. 1 16 646 662 662

249 Cricotopus (C.) sp. 4 22 119 192 30 14 391 768 4 7 131 43 33 34 252 1020 Cricotopus (C.) sp. 5 3 46 283 4 1 9 346 1 271 70 92 12 207 653 999 Cricotopus (C.) sp. 6 32 2 919 953 953 Cricotopus (C.) sp. 7 2 7 9 27 7 13 11 219 277 286 Cricotopus (C.) sp. 8 13 20 102 135 135 Cricotopus (C.) sp. 9 98 30 2 130 130 Cricotopus (C.) sp. 10 1 1 1 Cricotopus (C.) sp. 12 45 68 2 115 115 Cricotopus (I.) sylvestris 4 4 8 126 27 2 155 163 Cricotopus (I.) sp. 1 12 12 18 22 40 52 Diplocladius sp. 2 2 442 4 15 1 462 464 Doncricotopus sp. 43 1 13 2 59 19 12 83 1 1 116 175 Epoicocladius sp. 1 1 1

TAXON EA PN TR BR MI VA GWD CE CH CR RO RU SU SWD ALL ORTHOCLADIINAE (Cont.) Eukiefferiella brehmi gr. sp. 1 1 3 3 242 249 42 7 150 1 200 449 Eukiefferiella sp. 2 1 1 14 78 37 129 130 Eukiefferiella sp. 3 3 3 3 Eukiefferiella claripennis 441 36 623 61 146 151 1458 1 22 1 82 2 108 1566 Eukiefferiella ilkleyensis 2 177 48 84 75 386 111 105 71 101 388 774 Heterotrissocladius changi 17 6 2 3 28 1 1 29 Heterotrissocladius marcidus 4 2 2 20 28 3 3 4 10 38 Hydrobaenus johannseni 1 1 10 15 38 1 1 8 73 74 Hydrobaenus pillipes 12 1 1 14 14

250 Krenosmittia sp. 42 1 43 59 6 65 108 Limnophyes sp. 1 37 136 48 2 2 225 2 1 22 25 250 Limnophyes sp. 2 10 4 15 4 3 36 1 3 7 11 47 Limnophyes sp. 3 5 7 2 2 5 4 25 38 2 3 5 1 49 74 Limnophyes sp. 4 1 1 1 Lopescladius sp. 1 36 224 260 260 Lopescladius sp. 2 401 673 668 1742 1742 Lopescladius sp. 3 1 1 1 Nanocladius sp. 1 2 2 27 3 4 34 36 Nanocladius crassicornis 1 3 4 1 79 21 4 23 9 137 141 Nanocladius distinctus 1 1 2 2 3 Nanocladius rectinervis 3 21 8 9 41 143 156 92 9 22 45 467 508 Nanocladius spiniplenus 79 440 271 120 86 14 1010 101 98 67 64 104 434 1444

TAXON EA PN TR BR MI VA GWD CE CH CR RO RU SU SWD ALL ORTHOCLADIINAE (Cont.) Orthocladius (Eud.) sp. 2 2 2 Orthocladius (E.) rivicola 376 1 377 110 10 112 232 609 Orthocladius (E.) rivulorum 5 5 49 63 127 239 244 Orthocladius (O.) carlatus 7 150 157 42 503 118 51 249 963 1120 Orthocladius (O.) clarkei 1 1 2 6 9 2 17 19 Orthocladius (O.) dorenus 3 20 23 3 15 29 5 52 75 Orthocladius (O.) frigidus 2 169 1 174 346 2 3 5 351 Orthocladius (O.) mallochi 1 28 31 60 4 73 271 19 5 372 432 Orthocladius (O.) nigritus 11 9 173 473 666 666

251 Orthocladius (O.) obumbratus 2468 4 731 10 121 3334 54 22 106 148 24 33 387 3721 Orthocladius (O.) oliveri 3 3 6 224 990 30 4 206 71 1525 1531 Orthocladius (O.) robacki 3 16 2 286 307 12 1 13 320 Orthocladius (O.) vaillanti 56 2 58 58 Orthocladius (S.) lignicola 27 2 18 23 37 28 135 2 1 1 7 11 146 Parachaetocladius sp. 15 85 167 94 361 361 Paracladius sp. 3 3 3 Paracricotopus sp. 3 3 3 Parakiefferiella sp. 1 227 227 227 Parakiefferiella sp. 3 9 1411 67 8 1 135 1631 10 838 1907 17 14 11 2797 4428 Parakiefferiella sp. 4 1 1 1 7 2 10 11 Parakiefferiella sp. 6 12 12 12 Parametriocnemus sp. 1 204 11 56 259 39 35 604 105 28 139 18 10 22 322 926

TAXON EA PN TR BR MI VA GWD CE CH CR RO RU SU SWD ALL ORTHOCLADIINAE (Cont.) Parametriocnemus sp. 2 306 175 6 19 169 675 1 1 676 Parametriocnemus sp. 3 28 16 16 3 63 1 8 1 1 11 74 Parametriocnemus sp. 4 1 1 2 3 79 280 60 70 492 494 Paraphaenocladius exagitans 3 1 1 5 21 1 1 23 28 Paraphaenocladius nasthecus 2 2 2 Paraphaenocladius sp. 1 2 2 2 Psectrocladius (P.) psilopterus gr. sp. 1 44 2 1 47 47 Pseudosmittia sp. 1 1 2 2 3 Psilometriocnemus sp. 7 3 1 5 16 16

252 Rheocricotopus (P.) sp. 1 1 1 2 1 5 366 166258 301 1097 1099 Rheocricotopus (R.) sp. 1 21 3 24 1 1 2 26 Rheocricotopus (R.) effusus 1 1 2 2 3 Rheosmittia sp. 8 356 2 366 366 Smittia sp. 1 1 1 Stictocladius? sp. 6 6 6 Stilocladius sp. 5 242 247 13 5 18 265 Synorthocladius sp. 1 1 1 6 39 46 47 Thienemanniella lobapodema 4 4 14 439 31 13 1 498 502 Thienemanniella similis 14 1 7 193 215 215 Thienemanniella taurocapita 3 1 1 5 541 837 202 63 22 1665 1670 Thienemanniella xena 332 108 159 9 9 214 831 5 3 18 26 857 Thienemanniella sp. 1 381 843 213 168 61 34 1700 103 1523 1471 275 209 130 3711 5411

TAXON EA PN TR BR MI VA GWD CE CH CR RO RU SU SWD ALL ORTHOCLADIINAE (Cont.) Thienemanniella sp. 2 1 1 1 Tvetenia sp. 1 23 66 242 698 179 392 1600 6 1900 284 309 125 2624 4224 Tvetenia sp. 2 1 4 345 350 2 2 352 Tvetenia sp. 3 1 12 7 29 78 127 127 Tvetenia sp. 4 3 7 10 1 1 11 Xylotopus sp. 1 1 1 1 2 Zalutschia sp. 1 1 1 Orthocladiinae Genus A 1 1 1 CHIRONOMINI

253 Chironomus sp. 1 3 3 1 1 2 1 5 8 Chironomus sp. 2 6 194 27 19 246 246 Chironomus sp. 3 11 11 11 Cladopelma sp. 33 33 33 Cryptochironomus eminentia 2 2 2 Cryptochironomus ponderosus 2 2 2 6 6 Cryptochironomus sp. 1 6 3 1 10 4 12 25 4 3 1 49 59 Cryptochironomus sp. 2 31 31 31 Cryptochironomus sp. 3 4 4 4 Cryptochironomus sp. 4 2 15 4 21 21 Cryptochironomus sp. 5 1 1 2 2 Cryptochironomus sp. 7 5 5 5 Cryptochironomus sp. 8 1 1 1

TAXON EA PN TR BR MI VA GWD CE CH CR RO RU SU SWD ALL CHIRONOMINI (Cont.) Cryptochironomus sp. 9 2 2 2 Cryptotendipes sp. 1 1 39 2 2 43 44 Demicryptochironomus (I.) sp. 1 1 1 1 Demicryptochironomus (D.) sp. 2 3 3 3 Dicrotendipes fumidus 17 1 74 92 793 4 797 889 Dicrotendipes modestus 1 1 1 799 2 1 1 804 805 Dicrotendipes tritomus 5 5 5 Dicrotendipes sp. 1 1 1 1 Dicrotendipes sp. 2 19 3 22 22

254 Dicrotendipes sp. 4 21 25 46 46 Dicrotendipes sp. 5 3 3 3 Dicrotendipes sp. 6 1 1 1 Dicrotendipes sp. 7 2 2 2 Dicrotendipes sp. 8 1 1 1 Einfeldia sp. 1 1 1 1 Einfeldia sp. 2 1 1 1 Endochironomus nigricans 4 2 6 6 Glyptotendipes sp. 1 11 2 13 13 Harnischia sp. 1 2 3 3 Microtendipes sp. 1 35 5 40 119 26 33 178 218 Microtendipes sp. 2 1 1 1 23 226 15 5 270 271 Microtendipes sp. 3 1 1 2 2

TAXON EA PN TR BR MI VA GWD CE CH CR RO RU SU SWD ALL CHIRONOMINI (Cont.) Parachironomus sp. 1 2 1 3 3 Parachironomus sp. 2 2 2 2 Parachironomus sp. 3 1 1 1 Parachironomus sp. 4 4 4 4 Parachironomus sp. 5 2 2 2 Parachironomus sp. 6 2 2 2 Paracladopelma nais 10 4 14 14 Paracladopelma nereis 2 2 10 9 1 2 22 24 Paracladopelma undine 29 17 5 51 1 20 2 23 74

255 Paracladopelma sp. 1 1 1 1 3 3 Paralauterborniella nigrohalterale 24 64 10 19 1 118 118 Paratendipes albimanus 137 107 3 19 32 298 66 29 40 26 11 4 176 474 Paratendipes sp. 2 2 2 4 4 Phaenopsectra sp. 1 4 4 42 68 61 15 13 199 203 Phaenopsectra sp. 2 3 10 13 11 3 3 6 23 36 Phaenopsectra sp. 3 5 1 9 15 15 Phaenopsectra sp. 4 1 1 1 Polypedilum (P.) bergi 1 1 1 Polypedilum (P.) fallax 1 3 6 10 1 9 4 5 6 6 31 41 Polypedilum (P.) illinoense/angulum 20 2 22 13 163 18 17 5 8 224 246 Polypedilum (P.) laetum 8 59 6 26 44 15 158 8 11 3 2 4 28 186 Polypedilum (P.) trigonus 6 6 6

TAXON EA PN TR BR MI VA GWD CE CH CR RO RU SU SWD ALL CHIRONOMINI (Cont.) Polypedilum (U.) aviceps 1 450 83 36 141 12 723 2 131 11 3 147 870 Polypedilum (U.) flavum 1 2 3 3 Polypedilum (U.) obtusum 2 2 234 4 2 7 1 248 250 Polypedilum sp. 1 4 6 10 4 41 137 29 7 9 227 237 Polypedilum sp. 3 1 1 22 22 23 Polypedilum sp. 4 1 1 1 Polypedilum sp. 5 1 1 1 Polypedilum sp. 7 2 35 37 37 Polypedilum sp. 8 3 1 2 2 8 8

256 Robackia sp. 9 9 9 Saetheria tylus 12 11 1 24 11 4 2 17 41 Saetheria sp. 1 1 1 1 1 2 3 Stenochironomus sp. 1 4 1 5 5 Stenochironomus sp. 2 1 1 1 1 2 Stenochironomus sp. 3 3 3 6 9 9 7 1 32 35 Stictochironomus sp. 1 52 2 2 4 60 1 1 2 62 Stictochironomus sp. 2 238 3 241 241 Tribelos sp. 47 47 47 Xenochironomus sp. 36 1 1 38 38 Chironomini Genus B 1 1 2 2 PSEUDOCHIRONOMINI Pseudochironomus richardsoni 2 2 2

TAXON EA PN TR BR MI VA GWD CE CH CR RO RU SU SWD ALL TANYTARSINI Cladotanytarsus sp. 1 211 1 21 3 236 148 93 169 52 9 471 707 Cladotanytarsus sp. 2 90 6 11 107 107 Cladotanytarsus sp. 3 92 2 94 333 311 53 18 18 733 827 Cladotanytarsus sp. 4 157 32 1 190 190 Cladotanytarsus sp. 5 2 2 2 Cladotanytarsus sp. 6 2 2 2 Cladotanytarsus sp. 7 1 1 1 Micropsectra geminata 4 4 1 2 7 18 2 2 2 6 24 Micropsectra nigripila 188 190 20 4 2 404 27 122 44 1 1 1 196 600

257 Micropsectra polita 93 404 164 6 108 2 777 5 15 19 10 4 1 54 831 Micropsectra sp. 2 1 8 114 17 283 423 8 3 11 434 Paratanytarsus sp. 2 25 5 30 30 Paratanytarsus inopertus. gr. sp. 1 2 8 7 17 106 327 18 86 32 569 586 Paratanytarsus inopertus gr. sp. 2 1 1 30 946 12 5 993 994 Paratanytarsus inopertus gr. sp. 3 1 1 9 9 10 Paratanytarsus laccophilus 7 7 7 Paratanytarsus penicillatus gr. sp. 1 6 83 3 219 311 8 8 319 Paratanytarsus penicillatus gr. sp. 2 3 534 537 1 1 538 Rheotanytarsus sp. 1 216 47 14 15 69 59 420 43 728 442 47 37 19 1316 1736 Rheotanytarsus sp. 2 1 22 7 10 9 49 49 Rheotanytarsus distinctissimus 6 6 22 98 13 2 14 42 191 197 Stempellina sp. 3 2 5 5

TAXON EA PN TR BR MI VA GWD CE CH CR RO RU SU SWD ALL TANYTARSINI (Cont.) Stempellinella sp. 1 1 1 13 109 8 34 22 186 187 Stempellinella fimbriata 1 1 23 29 47 19 118 119 Stempellinella leptocelloides 1 1 42 3 21 66 67 Sublettea sp. 5 1 17 23 23 Tanytarsus confusus 1 1 2 2 Tanytarsus lobiger 9 122 211 29 53 5 429 429 Tanytarsus neoflavellus 31 31 31 Tanytarsus sepp 1 1 26 61 16 42 8 4 157 158 Tanytarsus wirthi 38 3 1 42 42

258 Tanytarsus sp. 1 1 1 2 661 32 6 35 9 743 745 Tanytarsus sp. 2 1 1 2 2 Tanytarsus sp. 10 14 20 3 3 1 12 53 45 387 24 24 22 502 555 Tanytarsus sp. 11 11 5 16 16 Tanytarsus sp. 12 2 2 39 137 160 25 13 1 375 377 Tanytarsus sp. 13 1 3 1 5 5 Tanytarsus sp. 16 1 3 4 4 Tanytarsus sp. 17 32 2 34 34 Tanytarsus sp. 18 14 39 53 1 1 54 Tanytarsus sp. 19 176 1 5 1 183 183 Tanytarsus sp. 21 382 10 5 397 397 Tanytarsus sp. 22 936 1 4 1 942 942 Tanytarsus sp. 23 367 11 1 379 379

TAXON EA PN TR BR MI VA GWD CE CH CR RO RU SU SWD ALL TANYTARSINI (Cont.) Tanytarsus sp. 24 5 5 5 Tanytarsus sp. 27 24 24 24

Total Abundance 5907 9248 9587 2466 2936 6505 36649 5798 11476 15162 4700 4322 4782 46240 82889 Total Richness54 84 82 78 58 57 149 118 131 107 120 119 106 238 262

259

Appendix B: Ranked mean site abundance for ground-water dominated (GWD) (n=6) and surface-water dominated (SWD) streams (n=6).

260 GWD SWD Taxon Mean Taxon Mean 1 Corynoneura sp. 1 631.7 Thienemanniella sp. 1 618.5 2 Diamesa sp. 628.8 Parakiefferiella sp. 3 466.2 3 Orthocladius (O.) obumbratus 555.7 Tvetenia sp. 1 437.3 4 Thienemanniella sp. 1 283.3 Corynoneura sp. 1 302.8 5 Parakiefferiella sp. 3 271.8 Lopescladius sp. 2 290.3 6 Tvetenia sp. 1 266.7 Thienemanniella taurocapita 277.5 7 Eukiefferiella claripennis 243.0 Orthocladius (O.) oliveri 254.2 8 Cricotopus (C.) bicinctus 199.3 Cricotopus (C.) bicinctus 227.0 9 Nanocladius spiniplenus 168.3 Rheotanytarsus sp. 1 219.3 10 Odontomesa fulva 159.3 Rheocricotopus (P.) sp. 1 182.8 11 Cricotopus (C.) sp. 6 158.8 Paratanytarsus inop. gr. sp. 2 165.5 12 Thienemanniella xena 138.5 Orthocladius (O.) carlatus 160.5 13 Micropsectra polita 129.5 Tanytarsus sp. 22 157.0 14 Cricotopus (C.) sp. 4 128.0 Dicrotendipes modestus 134.0 15 Polypedilum (U.) aviceps 120.5 Dicrotendipes fumidus 132.8 16 Parametriocnemus sp. 2 112.5 Tanytarsus sp. 1 123.8 17 Cricotopus (C.) sp. 1 110.3 Cladotanytarsus sp. 3 122.2 18 Corynoneura sp. 4 109.0 Orthocladius (O.) nigritus 111.0 19 Parametriocnemus sp. 1 100.7 Cricotopus (C.) sp. 5 108.8 20 Paratanytarsus pen. gr. sp. 2 89.5 Corynoneura sp. 4 100.3 21 Prodiamesa olivacea 71.8 Paratanytarsus inop. gr. sp. 1 94.8 22 Micropsectra sp. 2 70.5 Cricotopus (C.) trifascia 92.7 23 Rheotanytarsus sp. 1 70.0 Tanytarsus sp. 10 83.7 24 Cricotopus (C.) trifascia 67.5 Thienemanniella lobapodema 83.0 25 Micropsectra nigripila 67.3 Diamesa sp. 82.0 26 Eukiefferiella ilkleyensis 64.3 Parametriocnemus sp. 4 82.0 27 Orthocladius (E.) rivicola 62.8 Cladotanytarsus sp. 1 78.5 28 Parachaetocladius sp. 60.2 Nanocladius rectinervis 77.8 29 Tvetenia sp. 2 58.3 Diplocladius sp. 77.0 30 Cricotopus (C.) sp. 5 57.7 Nanocladius spiniplenus 72.3 31 Orthocladius (O.) frigidus 57.7 Tanytarsus lobiger 71.5 32 Paratanytarsus pen. gr. sp. 1 51.8 Tanytarsus sp. 21 66.2 33 Orthocladius (O.) robacki 51.2 Corynoneura sp. 3 64.8 34 Paratendipes albimanus 49.7 Eukiefferiella ilkleyensis 64.7 35 Cricotopus (C.) vierriensis 46.2 Orthocladius (O.) obumbratus 64.5 36 Eukiefferiella brehmi gr. sp. 1 41.5 Tanytarsus sp. 23 63.2 37 Stilocladius sp. 41.2 Tanytarsus sp. 12 62.5 38 Stictochironomus sp. 2 40.2 Orthocladius (O.) mallochi 62.0 261 GWD SWD Taxon Mean Taxon Mean 39 Cladotanytarsus sp. 1 39.3 Rheosmittia sp. 61.0 40 Parakiefferiella sp. 1 37.8 Parametriocnemus sp. 1 53.7 41 Limnophyes sp. 1 37.5 Cricotopus (C.) sp. 7 46.2 42 Polypedilum (P.) laetum 26.3 Microtendipes sp. 2 45.0 43 Orthocladius (O.) carlatus 26.2 Lopescladius sp. 1 43.3 44 Orthocladius (S.) lignicola 22.5 Cricotopus (C.) sp. 4 42.0 45 Brillia flavifrons 19.3 Polypedilum (U.) obtusum 41.3 46 Pagastia orthogonia 18.5 Chironomus sp. 2 41.0 47 Cladotanytarsus sp. 3 15.7 Nilotanypus fimbriatus 39.8 48 Dicrotendipes fumidus 15.3 Orthocladius (E.) rivulorum 39.8 49 Parametriocnemus sp. 3 10.5 Corynoneura sp. 2 39.5 50 Orthocladius (O.) mallochi 10.0 Corynoneura sp. 5 38.8 51 Stictochironomus sp. 1 10.0 Orthocladius (E.) rivicola 38.7 52 Doncricotopus sp. 9.8 Polypedilum sp. 1 37.8 53 Orthocladius (O.) vaillanti 9.7 Polypedilum (P.) illinoense 37.3 54 Tanytarsus sp. 10 8.8 Thienemanniella similis 35.8 55 Tanytarsus sp. 18 8.8 Ablabesmyia mallochi 34.5 56 Paracladopelma undine 8.5 Eukiefferiella brehmi gr. sp. 1 33.3 57 Krenosmittia sp. 7.2 Phaenopsectra sp. 1 33.2 58 Nanocladius rectinervis 6.8 Micropsectra nigripila 32.7 59 Microtendipes sp. 1 6.7 Rheotanytarsus distinctissimus 31.8 60 Cardiocladius albiplumus 6.2 Cladotanytarsus sp. 4 31.7 61 Limnophyes sp. 2 6.0 Stempellinella sp. 1 31.0 62 Corynoneura sp. 3 5.7 Tanytarsus sp. 19 30.5 63 Zavrelimyia sp. 5.0 Microtendipes sp. 1 29.7 64 Heterotrissocladius changi 4.7 Paratendipes albimanus 29.3 65 Heterotrissocladius marcidus 4.7 Procladius sp. 1 27.0 66 Limnophyes sp. 3 4.2 Tanytarsus sepp 26.2 67 Rheocricotopus (R.) effusoides 4.0 Cricotopus (I.) sylvestris 25.8 68 Saetheria tylus 4.0 Polypedilum (U.) aviceps 24.5 69 Orthocladius (O.) dorenus 3.8 Nanocladius crassicornis 22.8 70 Polypedilum (P.) illinoense 3.7 Cricotopus (C.) sp. 8 22.5 71 Micropsectra geminata 3.0 Cricotopus (C.) sp. 9 21.7 72 Conchapelopia rurika 2.8 Eukiefferiella sp. 2 21.5 73 Paratanytarsus inop. gr. sp. 1 2.8 Tvetenia sp. 3 21.2 74 Chaetocladius piger gr. sp. 2 2.8 Cricotopus (C.) vierriensis 20.7 75 Psilometriocnemus sp. 2.7 Paralauterborniella nigrohalt. 19.7 76 Phaenopsectra sp. 3 2.5 Stempellinella fimbriata 19.7 262 GWD SWD Taxon Mean Taxon Mean 77 Chaetocladius dent. gr. sp. 3 2.3 Doncricotopus sp. 19.3 78 Paracladopelma nais 2.3 Cricotopus (C.) sp. 12 19.2 79 Conchapelopia fasciata 2.2 Cardiocladius sp. 1 18.7 80 Meropelopia sp. 1 2.2 Eukiefferiella claripennis 18.0 81 Corynoneura sp. 5 2.2 Cladotanytarsus sp. 2 17.8 82 Phaenopsectra sp. 2 2.2 Ablabesmyia monilis 16.7 83 Cardiocladius sp. 1 2.0 Labrundinia pilosella 16.7 84 Cricotopus (I.) sp. 1 2.0 Hydrobaenus johannseni 12.2 85 Tvetenia sp. 4 1.7 Stempellinella leptocelloides 11.0 86 Cryptochironomus sp. 1 1.7 Brillia flavifrons 10.8 87 Polypedilum (P.) fallax 1.7 Krenosmittia sp. 10.8 88 Polypedilum sp. 1 1.7 Chaetocladius piger gr. sp. 2 9.0 89 Potthastia sp. 1 1.5 Micropsectra polita 9.0 90 Cricotopus (C.) sp. 7 1.5 Orthocladius (O.) dorenus 8.7 91 Cricotopus (I.) sylvestris 1.3 Larsia sp. 8.5 92 Paramerina sp. 1.0 Limnophyes sp. 3 8.2 93 Orthocladius (O.) oliveri 1.0 Cryptochironomus sp. 1 8.2 94 Stictocladius? sp. 1.0 Rheotanytarsus sp. 2 8.2 95 Rheotanytarsus distinctissimus 1.0 Psectrocladius (P.) ps. gr. sp. 1 7.8 96 Chaetocladius sp. 6 0.8 Tribelos sp. 7.8 97 Orthocladius (E.) rivulorum 0.8 Conchapelopia fasciata 7.7 98 Paraphaenocladius exagitans 0.8 Synorthocladius sp. 7.7 99 Thienemanniella taurocapita 0.8 Dicrotendipes sp. 4 7.7 100 Nanocladius crassicornis 0.7 Cryptotendipes sp. 7.2 101 Thienemanniella lobapodema 0.7 Conchapelopia rurika 7.0 102 Phaenopsectra sp. 1 0.7 Tanytarsus wirthi 7.0 103 Acricotopus sp. 0.5 Cricotopus (I.) sp. 1 6.7 104 Bryophaenocladius sp. 0.5 Acricotopus sp. 6.3 105 Paracladius sp. 0.5 Xenochironomus sp. 6.3 106 Chironomus sp. 1 0.5 Polypedilum sp. 7 6.2 107 Stenochironomus sp. 3 0.5 Nanocladius sp. 1 5.7 108 Radotanypus sp. 0.3 Tanytarsus sp. 17 5.7 109 Diplocladius sp. 0.3 Cladopelma sp. 5.5 110 Nanocladius sp. 1 0.3 Stenochironomus sp. 3 5.3 111 Orthocladius (Eud.) sp. 0.3 Cryptochironomus sp. 2 5.2 112 Orthocladius (O.) clarkei 0.3 Polypedilum (P.) fallax 5.2 113 Parametriocnemus sp. 4 0.3 Tanytarsus neoflavellus 5.2 114 Paraphaenocladius nasthecus 0.3 Paratanytarsus sp. 2 5.0 263 GWD SWD Taxon Mean Taxon Mean 115 Paraphaenocladius sp. 1 0.3 Polypedilum (P.) laetum 4.7 116 Rheocricotopus (P.) sp. 1 0.3 Chaetocladius sp. 5 4.5 117 Paracladopelma nereis 0.3 Thienemanniella xena 4.3 118 Polypedilum (U.) obtusum 0.3 Odontomesa fulva 4.2 119 Tanytarsus sp. 1 0.3 Limnophyes sp. 1 4.2 120 Tanytarsus sp. 12 0.3 Tanytarsus sp. 27 4.0 121 Ablabesmyia mallochi 0.2 Paraphaenocladius exagitans 3.8 122 Labrundinia neopilosella 0.2 Paracladopelma undine 3.8 123 Labrundinia pilosella 0.2 Phaenopsectra sp. 2 3.8 124 Chaetocladius sp. 5 0.2 Sublettea sp. 3.8 125 Eukiefferiella sp. 2 0.2 Dicrotendipes sp. 2 3.7 126 Hydrobaenus johannseni 0.2 Paracladopelma nereis 3.7 127 Limnophyes sp. 4 0.2 Polypedilum sp. 3 3.7 128 Nanocladius distinctus 0.2 Cryptochironomus sp. 4 3.5 129 Parakiefferiella sp. 4 0.2 Stilocladius sp. 3.0 130 Pseudosmittia sp. 0.2 Orthocladius (O.) clarkei 2.8 131 Rheocricotopus (R.) effusus 0.2 Saetheria tylus 2.8 132 Smittia sp. 0.2 Tanytarsus sp. 11 2.7 133 Synorthocladius sp. 0.2 Pentaneura sp. 2.3 134 Thienemanniella sp. 2 0.2 Hydrobaenus pillipes 2.3 135 Xylotopus sp. 0.2 Paramerina sp. 2.2 136 Orthocladiinae Genus A 0.2 Orthocladius (O.) robacki 2.2 137 Cryptotendipes sp. 0.2 Glyptotendipes sp. 1 2.2 138 Dicrotendipes modestus 0.2 Parakiefferiella sp. 6 2.0 139 Microtendipes sp. 2 0.2 Hayesomyia senata 1.8 140 Polypedilum sp. 3 0.2 Limnophyes sp. 2 1.8 141 Saetheria sp. 1 0.2 Orthocladius (S.) lignicola 1.8 142 Stenochironomus sp. 2 0.2 Parametriocnemus sp. 3 1.8 143 Paratanytarsus inop. gr. sp. 2 0.2 Chironomus sp. 3 1.8 144 Paratanytarsus inop. gr. sp. 3 0.2 Micropsectra sp. 2 1.8 145 Stempellinella sp. 1 0.2 Heterotrissocladius marcidus 1.7 146 Stempellinella fimbriata 0.2 Parakiefferiella sp. 4 1.7 147 Stempellinella leptocelloides 0.2 Helopelopia cornuticaudata 1.5 148 Tanytarsus sepp 0.2 Robackia sp. 1.5 149 Paratanytarsus inop. gr. sp. 3 1.5 150 Natarsia sp. 1.3 151 Polypedilum sp. 8 1.3 152 Paratanytarsus pen. gr. sp. 1 264 GWD SWD Taxon Mean Taxon Mean 153 Conchapelopia telema 1.2 154 Potthastia sp. 2 1.2 155 Paratanytarsus cf. laccophilus 1.2 156 Corynoneura sp. 6 1.0 157 Cryptochironomus ponderosus 1.0 158 Endochironomus nigricans 1.0 159 Polypedilum (P.) trigonus 1.0 160 Micropsectra geminata 1.0 161 Labrundinia neopilosella 0.8 162 Orthocladius (O.) frigidus 0.8 163 Chironomus sp. 1 0.8 164 Cryptochironomus sp. 7 0.8 165 Dicrotendipes tritomus 0.8 166 Stenochironomus sp. 1 0.8 167 Stempellina sp. 0.8 168 Tanytarsus sp. 13 0.8 169 Tanytarsus sp. 24 0.8 170 Procladius bellus 0.7 171 Cardiocladius albiplumus 0.7 172 Cryptochironomus sp. 3 0.7 173 Parachironomus sp. 4 0.7 174 Paratendipes sp. 2 0.7 175 Tanytarsus sp. 16 0.7 176 Labrundinia maculata 0.5 177 Procladius sp. 2 0.5 178 Eukiefferiella sp. 3 0.5 179 Paracricotopus sp. 0.5 180 Demicryptochironomus sp. 2 0.5 181 Dicrotendipes sp. 5 0.5 182 Harnischia sp. 0.5 183 Parachironomus sp. 1 0.5 184 Paracladopelma sp. 1 0.5 185 Polypedilum (U.) flavum 0.5 186 Prodiamesa olivacea 0.3 187 Chaetocladius dent. gr. sp. 3 0.3 188 Nanocladius distinctus 0.3 189 Pseudosmittia sp. 0.3 190 Rheocricotopus (R.) effusoides 0.3 265 GWD SWD Taxon Mean Taxon Mean 191 Rheocricotopus (R.) effusus 0.3 192 Tvetenia sp. 2 0.3 193 Cryptochironomus eminentia 0.3 194 Cryptochironomus sp. 5 0.3 195 Cryptochironomus sp. 9 0.3 196 Dicrotendipes sp. 7 0.3 197 Microtendipes sp. 3 0.3 198 Parachironomus sp. 2 0.3 199 Parachironomus sp. 5 0.3 200 Parachironomus sp. 6 0.3 201 Saetheria sp. 1 0.3 202 Stictochironomus sp. 1 0.3 203 Chironomini Genus B 0.3 204 Pseudochironomus richardsoni 0.3 205 Cladotanytarsus sp. 5 0.3 206 Cladotanytarsus sp. 6 0.3 207 Tanytarsus confusus 0.3 208 Tanytarsus sp. 2 0.3 209 Ablabesmyia idei 0.2 210 Ablabesmyia peleensis 0.2 211 Psectrotanypus sp. 0.2 212 Tanypus sp. 0.2 213 Zavrelimyia sp. 0.2 214 Pagastia orthogonia 0.2 215 Cricotopus (C.) sp. 10 0.2 216 Epoicocladius sp. 0.2 217 Heterotrissocladius changi 0.2 218 Lopescladius sp. 3 0.2 219 Parametriocnemus sp. 2 0.2 220 Tvetenia sp. 4 0.2 221 Xylotopus sp. 0.2 222 Zalutschia sp. 0.2 223 Cryptochironomus sp. 8 0.2 224 Demicryptochironomus sp. 1 0.2 225 Dicrotendipes sp. 1 0.2 226 Dicrotendipes sp. 6 0.2 227 Dicrotendipes sp. 8 0.2 228 Einfeldia sp. 1 0.2 266 GWD SWD Taxon Mean Taxon Mean 229 Einfeldia sp. 2 0.2 230 Parachironomus sp. 3 0.2 231 Phaenopsectra sp. 4 0.2 232 Polypedilum (P.) bergi 0.2 233 Polypedilum sp. 4 0.2 234 Polypedilum sp. 5 0.2 235 Stenochironomus sp. 2 0.2 236 Cladotanytarsus sp. 7 0.2 237 Paratanytarsus pen. gr. sp. 2 0.2 238 Tanytarsus sp. 18 0.2

267

Appendix C: Taxa with the greater absolute abundance and relative abundance in either GWD or SWD sites; Taxa are ranked from those with the greatest abundance in GWD (positive values) to those with the greatest abundance in SWD streams (negative values); GA = mean taxon abundance from GWD sites, SA = mean taxon abundance from SWD sites; absolute abundance differences were calculated as log([GA-SA]+1) and is sensitive to the absolute difference in abundance between GWD and SWD sites; relative abundance differences were calculated as log([GA+1]/[SA+1]) and is sensitive to relative differences in abundance between GWD and SWD sites.

268

 GA + 1  Subfamily/ log  Tribe Taxon log(GA-SA)  SA + 1  Diamesinae Diamesa sp. 2.74 0.93 Orthocladiinae Orthocladius (O.) obumbratus 2.69 0.98 Orthocladiinae Corynoneura sp. 1 2.52 0.49 Orthocladiinae Eukiefferiella claripennis 2.35 1.14 Orthocladiinae Cricotopus (C.) sp. 6 2.20 2.21 Prodiamesinae Odontomesa fulva 2.19 1.51 Orthocladiinae Thienemanniella xena 2.13 1.43 Tanytarsini Micropsectra polita 2.08 1.15 Orthocladiinae Parametriocnemus sp. 2 2.05 1.99 Orthocladiinae Cricotopus (C.) sp. 1 2.05 2.05 Chironomini Polypedilum (U.) aviceps 1.99 0.76 Orthocladiinae Nanocladius spiniplenus 1.99 0.52 Tanytarsini Paratanytarsus penicillatus gr. sp. 2 1.96 1.90 Orthocladiinae Cricotopus (C.) sp. 4 1.94 0.60 Prodiamesinae Prodiamesa olivacea 1.86 1.75 Tanytarsini Micropsectra sp. 2 1.84 1.42 Orthocladiinae Parachaetocladius sp. 1.79 1.79 Orthocladiinae Tvetenia sp. 2 1.77 1.66 Orthocladiinae Orthocladius (O.) frigidus 1.76 1.52 Tanytarsini Paratanytarsus penicillatus gr. sp. 1 1.71 1.37 Orthocladiinae Orthocladius (O.) robacki 1.70 1.24 Orthocladiinae Parametriocnemus sp. 1 1.68 0.46 Chironomini Stictochironomus sp. 2 1.61 1.62 Orthocladiinae Stilocladius sp. 1.59 1.06 Orthocladiinae Parakiefferiella sp. 1 1.59 1.60 Tanytarsini Micropsectra nigripila 1.55 0.48 Orthocladiinae Limnophyes sp. 1 1.54 0.93 Orthocladiinae Cricotopus (C.) vierriensis 1.42 0.50 Orthocladiinae Orthocladius (E.) rivicola 1.40 0.42 Chironomini Polypedilum (P.) laetum 1.36 0.77 Orthocladiinae Orthocladius (S.) lignicola 1.34 0.97 Chironomini Paratendipes albimanus 1.33 0.43 Diamesinae Pagastia orthogonia 1.29 1.25 Orthocladiinae Orthocladius (O.) sp. 3 1.03 1.07 Chironomini Stictochironomus sp. 1 1.03 0.97 Tanytarsini Tanytarsus sp. 18 0.99 0.97 Orthocladiinae Parametriocnemus sp. 3 0.99 0.70 269  GA + 1  Subfamily/ log  Tribe Taxon log(GA-SA)  SA + 1  Orthocladiinae Brillia flavifrons 0.98 0.43 Orthocladiinae Eukiefferiella brehmi gr. sp. 1 0.96 0.35 Orthocladiinae Corynoneura sp. 4 0.95 0.32 Orthocladiinae Cardiocladius albiplumus 0.81 0.72 Tanypodinae Zavrelimyia sp. 0.77 0.79 Chironomini Paracladopelma undine 0.75 0.47 Orthocladiinae Heterotrissocladius changi 0.74 0.77 Orthocladiinae Limnophyes sp. 2 0.71 0.54 Orthocladiinae Rheocricotopus (R.) effusoides 0.67 0.68 Orthocladiinae Heterotrissocladius marcidus 0.60 0.49 Orthocladiinae Psilometriocnemus sp. 0.56 0.67 Chironomini Phaenopsectra sp. 3 0.54 0.65 Chironomini Paracladopelma nais 0.52 0.64 Tanypodinae Meropelopia sp. 1 0.50 0.62 Orthocladiinae Chaetocladius dentiforceps gr. sp. 3 0.48 0.54 Tanytarsini Micropsectra geminata 0.48 0.48 Diamesinae Potthastia sp. 1 0.40 0.54 Orthocladiinae Tvetenia sp. 4 0.40 0.52 Chironomini Saetheria tylus 0.34 0.36 Orthocladiinae Stictocladius? sp. 0.30 0.48 Orthocladiinae Chaetocladius sp. 6 0.26 0.45 Orthocladiinae Bryophaenocladius sp. 0.18 0.40 Orthocladiinae Paracladius sp. 0.18 0.40 Tanypodinae Radotanypus sp. 0.12 0.37 Orthocladiinae Orthocladius (Eud.) sp. 0.12 0.37 Orthocladiinae Paraphaenocladius nasthecus 0.12 0.37 Orthocladiinae Paraphaenocladius sp. 1 0.12 0.37 Orthocladiinae Limnophyes sp. 4 0.07 0.34 Orthocladiinae Smittia sp. 0.07 0.34 Orthocladiinae Thienemanniella sp. 2 0.07 0.34 Orthocladiinae Orthocladiinae Genus A 0.07 0.34 Orthocladiinae Xylotopus sp. 0.00 0.30 Chironomini Stenochironomus sp. 2 0.00 0.30 Tanypodinae Ablabesmyia idei -0.07 -0.34 Tanypodinae Ablabesmyia peleensis -0.07 -0.34 Tanypodinae Psectrotanypus sp. -0.07 -0.34 Tanypodinae Tanypus sp. -0.07 -0.34 Orthocladiinae Cricotopus (C.) sp. 10 -0.07 -0.34 270  GA + 1  Subfamily/ log  Tribe Taxon log(GA-SA)  SA + 1  Orthocladiinae Epoicocladius sp. -0.07 -0.34 Orthocladiinae Lopescladius sp. 3 -0.07 -0.34 Orthocladiinae Nanocladius distinctus -0.07 -0.33 Orthocladiinae Pseudosmittia sp. -0.07 -0.33 Orthocladiinae Rheocricotopus (R.) effusus -0.07 -0.33 Orthocladiinae Zalutschia sp. -0.07 -0.34 Chironomini Cryptochironomus sp. 8 -0.07 -0.34 Chironomini Demicryptochironomus (I.) sp. 1 -0.07 -0.34 Chironomini Dicrotendipes sp. 1 -0.07 -0.34 Chironomini Dicrotendipes sp. 6 -0.07 -0.34 Chironomini Dicrotendipes sp. 8 -0.07 -0.34 Chironomini Einfeldia sp. 1 -0.07 -0.34 Chironomini Einfeldia sp. 2 -0.07 -0.34 Chironomini Parachironomus sp. 3 -0.07 -0.34 Chironomini Phaenopsectra sp. 4 -0.07 -0.34 Chironomini Polypedilum (P.) bergi -0.07 -0.34 Chironomini Polypedilum sp. 4 -0.07 -0.34 Chironomini Polypedilum sp. 5 -0.07 -0.34 Chironomini Saetheria sp. 1 -0.07 -0.33 Tanytarsini Cladotanytarsus sp. 7 -0.07 -0.34 Chironomini Cryptochironomus eminentia -0.12 -0.37 Chironomini Cryptochironomus sp. 5 -0.12 -0.37 Chironomini Cryptochironomus sp. 9 -0.12 -0.37 Chironomini Dicrotendipes sp. 7 -0.12 -0.37 Chironomini Microtendipes sp. 3 -0.12 -0.37 Chironomini Parachironomus sp. 2 -0.12 -0.37 Chironomini Parachironomus sp. 5 -0.12 -0.37 Chironomini Parachironomus sp. 6 -0.12 -0.37 Chironomini Chironomini Genus B -0.12 -0.37 Chironomini Pseudochironomus richardsoni -0.12 -0.37 Tanytarsini Cladotanytarsus sp. 5 -0.12 -0.37 Tanytarsini Cladotanytarsus sp. 6 -0.12 -0.37 Tanytarsini Tanytarsus confusus -0.12 -0.37 Tanytarsini Tanytarsus sp. 2 -0.12 -0.37 Chironomini Chironomus sp. 1 -0.12 -0.35 Tanypodinae Labrundinia maculata -0.18 -0.40 Tanypodinae Procladius sp. 2 -0.18 -0.40 Orthocladiinae Paracricotopus sp. -0.18 -0.40 271  GA + 1  Subfamily/ log  Tribe Taxon log(GA-SA)  SA + 1  Chironomini Demicryptochironomus (D.) sp. 2 -0.18 -0.40 Chironomini Dicrotendipes sp. 5 -0.18 -0.40 Chironomini Harnischia sp. -0.18 -0.40 Chironomini Parachironomus sp. 1 -0.18 -0.40 Chironomini Paracladopelma sp. 1 -0.18 -0.40 Chironomini Polypedilum (U.) flavum -0.18 -0.40 Tanypodinae Procladius bellus -0.22 -0.43 Orthocladiinae Eukiefferiella sp. 3 -0.22 -0.43 Chironomini Cryptochironomus sp. 3 -0.22 -0.43 Chironomini Parachironomus sp. 4 -0.22 -0.43 Chironomini Paratendipes sp. 2 -0.22 -0.43 Tanytarsini Tanytarsus sp. 16 -0.22 -0.43 Tanypodinae Labrundinia neopilosella -0.22 -0.41 Chironomini Cryptochironomus sp. 7 -0.26 -0.45 Chironomini Dicrotendipes tritomus -0.26 -0.45 Chironomini Stenochironomus sp. 1 -0.26 -0.45 Tanytarsini Stempellina sp. -0.26 -0.45 Tanytarsini Tanytarsus sp. 13 -0.26 -0.45 Tanytarsini Tanytarsus sp. 24 -0.26 -0.45 Orthocladiinae Corynoneura sp. 6 -0.30 -0.48 Chironomini Cryptochironomus ponderosus -0.30 -0.48 Chironomini Endochironomus nigricans -0.30 -0.48 Chironomini Polypedilum (P.) trigonus -0.30 -0.48 Tanypodinae Paramerina sp. -0.34 -0.41 Tanypodinae Conchapelopia telema -0.34 -0.50 Diamesinae Potthastia sp. 2 -0.34 -0.50 Tanytarsini Paratanytarsus laccophilus -0.34 -0.50 Tanypodinae Natarsia sp. -0.37 -0.52 Chironomini Polypedilum sp. 8 -0.37 -0.52 Tanytarsini Paratanytarsus inopertus gr. sp. 3 -0.37 -0.50 Orthocladiinae Eukiefferiella ilkleyensis -0.37 -0.31 Tanypodinae Helopelopia cornuticaudata -0.40 -0.54 Orthocladiinae Parakiefferiella sp. 4 -0.40 -0.52 Tanytarsini Robackia sp. -0.40 -0.54 Chironomini Phaenopsectra sp. 2 -0.43 -0.40 Tanypodinae Hayesomyia senata -0.45 -0.58 Chironomini Chironomus sp. 3 -0.45 -0.58 Orthocladiinae Parakiefferiella sp. 6 -0.48 -0.60 272  GA + 1  Subfamily/ log  Tribe Taxon log(GA-SA)  SA + 1  Chironomini Glyptotendipes sp. 1 -0.50 -0.62 Tanypodinae Pentaneura sp. -0.52 -0.64 Orthocladiinae Hydrobaenus pillipes -0.52 -0.64 Orthocladiinae Orthocladius (O.) clarkei -0.54 -0.59 Tanytarsini Tanytarsus sp. 11 -0.56 -0.67 Orthocladiinae Paraphaenocladius exagitans -0.60 -0.56 Chironomini Paracladopelma nereis -0.64 -0.65 Chironomini Cryptochironomus sp. 4 -0.65 -0.74 Chironomini Polypedilum (P.) fallax -0.65 -0.52 Chironomini Polypedilum sp. 3 -0.65 -0.70 Chironomini Dicrotendipes sp. 2 -0.67 -0.75 Orthocladiinae Krenosmittia sp. -0.68 -0.39 Chironomini Sublettea sp. -0.68 -0.77 Orthocladiinae Limnophyes sp. 3 -0.70 -0.44 Tanytarsini Tanytarsus sp. 27 -0.70 -0.78 Tanypodinae Conchapelopia rurika -0.71 -0.49 Orthocladiinae Chaetocladius sp. 5 -0.73 -0.76 Orthocladiinae Cricotopus (I.) sp. 1 -0.75 -0.55 Orthocladiinae Orthocladius (O.) dorenus -0.77 -0.48 Chironomini Stenochironomus sp. 3 -0.77 -0.72 Tanytarsini Paratanytarsus sp. 2 -0.78 -0.85 Chironomini Cryptochironomus sp. 2 -0.79 -0.86 Tanytarsini Tanytarsus neoflavellus -0.79 -0.86 Orthocladiinae Nanocladius sp. 1 -0.80 -0.78 Tanypodinae Conchapelopia fasciata -0.81 -0.57 Chironomini Cladopelma sp. -0.81 -0.88 Tanytarsini Tanytarsus sp. 17 -0.82 -0.88 Orthocladiinae Acricotopus sp. -0.83 -0.77 Chironomini Polypedilum sp. 7 -0.86 -0.91 Orthocladiinae Chaetocladius piger gr. sp. 2 -0.86 -0.56 Chironomini Xenochironomus sp. -0.87 -0.92 Chironomini Cryptochironomus sp. 1 -0.88 -0.65 Chironomini Cryptotendipes sp. -0.90 -0.90 Tanytarsini Tanytarsus wirthi -0.90 -0.95 Orthocladiinae Synorthocladius sp. -0.93 -0.93 Chironomini Dicrotendipes sp. 4 -0.94 -0.99 Orthocladiinae Psectrocladius (P.) psilop. gr. sp. 1 -0.95 -0.99 Chironomini Tribelos sp. -0.95 -0.99 273  GA + 1  Subfamily/ log  Tribe Taxon log(GA-SA)  SA + 1  Tanytarsini Rheotanytarsus sp. 2 -0.96 -1.01 Tanypodinae Larsia sp. -0.98 -1.02 Orthocladiinae Doncricotopus sp. -1.02 -0.46 Tanytarsini Stempellinella leptocelloides -1.07 -1.05 Orthocladiinae Hydrobaenus johannseni -1.11 -1.09 Tanypodinae Labrundinia pilosella -1.24 -1.21 Tanypodinae Ablabesmyia monilis -1.25 -1.27 Orthocladiinae Cardiocladius sp. 1 -1.25 -0.88 Tanytarsini Cladotanytarsus sp. 2 -1.28 -1.30 Orthocladiinae Cricotopus (C.) sp. 12 -1.30 -1.33 Tanytarsini Stempellinella fimbriata -1.32 -1.28 Chironomini Paralauterborniella nigrohalterale -1.32 -1.34 Orthocladiinae Eukiefferiella sp. 2 -1.35 -1.31 Orthocladiinae Tvetenia sp. 3 -1.35 -1.37 Orthocladiinae Nanocladius crassicornis -1.36 -1.18 Orthocladiinae Cricotopus (C.) sp. 8 -1.37 -1.39 Chironomini Microtendipes sp. 1 -1.38 -0.70 Orthocladiinae Cricotopus (I.) sylvestris -1.41 -1.10 Orthocladiinae Cricotopus (C.) sp. 9 -1.42 -1.44 Tanytarsini Tanytarsus sepp -1.43 -1.39 Orthocladiinae Cricotopus (C.) trifascia -1.43 -0.38 Tanypodinae Procladius sp. 1 -1.45 -1.46 Orthocladiinae Cricotopus (C.) bicinctus -1.48 -0.33 Tanytarsini Tanytarsus sp. 19 -1.50 -1.51 Tanytarsini Rheotanytarsus distinctissimus -1.50 -1.24 Tanytarsini Stempellinella sp. 1 -1.51 -1.46 Tanytarsini Cladotanytarsus sp. 4 -1.51 -1.53 Chironomini Phaenopsectra sp. 1 -1.53 -1.33 Chironomini Polypedilum (P.) illinoense -1.54 -0.96 Tanypodinae Ablabesmyia mallochi -1.55 -1.50 Orthocladiinae Thienemanniella similis -1.57 -1.58 Orthocladiinae Corynoneura sp. 5 -1.58 -1.13 Chironomini Polypedilum sp. 1 -1.58 -1.20 Orthocladiinae Orthocladius (E.) rivulorum -1.60 -1.37 Tanytarsini Cladotanytarsus sp. 1 -1.61 -0.48 Orthocladiinae Corynoneura sp. 2 -1.62 -1.63 Chironomini Chironomus sp. 2 -1.62 -1.63 Chironomini Polypedilum (U.) obtusum -1.62 -1.52 274  GA + 1  Subfamily/ log  Tribe Taxon log(GA-SA)  SA + 1  Tanypodinae Nilotanypus fimbriatus -1.62 -1.64 Orthocladiinae Lopescladius sp. 1 -1.65 -1.66 Orthocladiinae Cricotopus (C.) sp. 7 -1.66 -1.30 Chironomini Microtendipes sp. 2 -1.66 -1.61 Orthocladiinae Cricotopus (C.) sp. 5 -1.72 -0.46 Orthocladiinae Orthocladius (O.) mallochi -1.72 -0.83 Orthocladiinae Corynoneura sp. 3 -1.79 -1.04 Orthocladiinae Rheosmittia sp. -1.79 -1.80 Tanytarsini Tanytarsus sp. 12 -1.80 -1.69 Tanytarsini Tanytarsus sp. 23 -1.81 -1.81 Tanytarsini Tanytarsus sp. 21 -1.83 -1.83 Orthocladiinae Nanocladius rectinervis -1.86 -1.04 Tanytarsini Tanytarsus lobiger -1.86 -1.87 Tanytarsini Tanytarsus sp. 10 -1.88 -0.98 Orthocladiinae Diplocladius sp. -1.89 -1.77 Orthocladiinae Parametriocnemus sp. 4 -1.92 -1.80 Orthocladiinae Thienemanniella lobapodema -1.92 -1.71 Tanytarsini Paratanytarsus inopertus gr. sp. 1 -1.97 -1.42 Tanytarsini Cladotanytarsus sp. 3 -2.03 -0.93 Orthocladiinae Orthocladius (O.) nigritus -2.05 -2.05 Chironomini Dicrotendipes fumidus -2.07 -0.96 Tanytarsini Tanytarsus sp. 1 -2.10 -1.98 Chironomini Dicrotendipes modestus/neomodestus -2.13 -2.07 Orthocladiinae Orthocladius (O.) carlatus -2.13 -0.84 Tanytarsini Rheotanytarsus sp. 1 -2.18 -0.61 Tanytarsini Tanytarsus sp. 22 -2.20 -2.20 Tanytarsini Paratanytarsus inopertus gr. sp. 2 -2.22 -2.16 Orthocladiinae Tvetenia sp. 1 -2.24 -0.42 Orthocladiinae Rheocricotopus (P.) sp. 1 -2.27 -2.15 Orthocladiinae Parakiefferiella sp. 3 -2.29 -0.43 Orthocladiinae Orthocladius (O.) oliveri -2.41 -2.11 Orthocladiinae Thienemanniella taurocapita -2.45 -2.19 Orthocladiinae Lopescladius sp. 2 -2.48 -2.48 Orthocladiinae Thienemanniella sp. 1 -2.53 -0.50

275

=>0and<10individuals/sample =>10and<50individuals/sample =>50individuals/sample A PPENDIX

of chironomids foreachstudystream.

Individual speciesemergencepatterns

276

EAGLE CREEK (2003) Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec 17 30 13 27 27 10 24 08 22 05 19 03 17 01 15 28 10 24 08 22 05 19 03 17 31 TANYPODINAE Conchapelopia rurika Meropelopia sp. 1 Zavrelimyia sp. DIAMESINAE Diamesa sp. PRODIAMESINAE Odontomesa fulva Prodiamesa olivacea ORTHOCLADIINAE 277 Brillia flavifrons Cardiocladius sp. 1 Chaetocladius dentiforceps gr. sp. 3 Corynoneura sp. 1 Corynoneura sp. 4 Cricotopus (C.) bicinctus Cricotopus (C.) trifascia Cricotopus (C.) vierriensis Cricotopus (C.) sp. 4 Cricotopus (C.) sp. 5 Eukiefferiella claripennis Heterotrissocladius marcidus

EAGLE CREEK (2003) Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec ORTHOCLADIINAE (Continued) 17 30 13 27 13 27 10 24 08 22 05 19 03 17 01 15 28 10 24 08 22 05 19 03 17 31 Limnophyes sp. 1 Limnophyes sp. 2 Limnophyes sp. 3 Nanocladius rectinervis Nanocladius spiniplenus Orthocladius (O.) mallochi Orthocladius (O.) obumbratus Orthocladius (S.) lignicola Parachaetocladius sp. Parakiefferiella sp. 3 278 Parametriocnemus sp. 1 Parametriocnemus sp. 2 Parametriocnemus sp. 3 Psilometriocnemus sp. Rheocricotopus (P.) sp. 1 Rheocricotopus (R.) effusoides Thienemanniella xena Thienemanniella sp. 1 Tvetenia sp. 1 Tvetenia sp. 2 CHIRONOMINI Paratendipes albimanus

EAGLE CREEK (2003) Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec CHIRONOMINI (Continued) 17 13 27 13 27 10 24 08 22 05 19 03 17 01 15 28 10 24 08 22 05 19 03 17 31 Phaenopsectra sp. 2 Phaenopsectra sp. 3 Polypedilum (P.) laetum Polypedilum (U.) aviceps Polypedilum sp. 1 Stictochironomus sp. 1 Stictochironomus sp. 2 TANYTARSINI Micropsectra geminata Micropsectra nigripila 279 Micropsectra polita Micropsectra sp. 2 Paratanytarsus inopertus gr. sp. 1 Rheotanytarsus sp. 1 Tanytarsus sp. 1 Tanytarsus sp. 10

PINE CREEK (2003) Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec 17 30 13 27 13 27 10 24 08 22 05 19 03 17 01 15 28 10 24 08 22 05 19 03 17 31 TANYPODINAE Conchapelopia fasciata Conchapelopia rurika Labrundinia pilosella Meropelopia sp. 1 Paramerina sp. Radotanypus sp. Zavrelimyia sp. DIAMESINAE Diamesa sp. 280 PRODIAMESINAE Odontomesa fulva Prodiamesa olivacea ORTHOCLADIINAE Acricotopus sp. Brillia flavifrons Corynoneura sp. 1 Corynoneura sp. 5 Cricotopus (C.) bicinctus Cricotopus (C.) trifascia Cricotopus (C.) vierriensis Cricotopus (C.) sp. 1

PINE CREEK (2003) Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec ORTHOCLADIINAE (Continued) 17 30 13 27 13 27 10 24 08 22 05 19 03 17 01 15 28 10 24 08 22 05 19 03 17 31 Cricotopus (C.) sp. 4 Cricotopus (C.) sp. 5 Cricotopus (C.) sp. 6 Cricotopus (C.) sp. 7 Cricotopus (I.) sylvestris Doncricotopus sp. Eukiefferiella claripennis Eukiefferiella ilkleyensis Heterotrissocladius changi Heterotrissocladius marcidus 281 Limnophyes sp. 1 Limnophyes sp. 2 Limnophyes sp. 3 Nanocladius sp. 1 Nanocladius crassicornis Nanocladius rectinervis Nanocladius spiniplenus Orthocladius (O.) carlatus Orthocladius (O.) frigidus Orthocladius (O.) obumbratus Orthocladius (O.) oliveri Orthocladius (O.) robacki

PINE CREEK (2003) Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec ORTHOCLADIINAE (Continued) 17 30 13 27 13 27 10 24 08 22 05 19 03 17 01 15 28 10 24 08 22 05 19 03 17 31 Orthocladius (S.) lignicola Parachaetocladius sp. Parakiefferiella sp. 3 Parametriocnemus sp. 1 Paraphaenocladius exagitans Thienemanniella taurocapita Thienemanniella xena Thienemanniella sp. 1 Tvetenia sp. 1 Tvetenia sp. 2 282 CHIRONOMINI Chironomus sp. 1 Cryptochironomus sp. 1 Cryptotendipes sp. Dicrotendipes fumidus Dicrotendipes modestus Microtendipes sp. 1 Paracladopelma nais Paracladopelma nereis Paracladopelma undine Paratendipes albimanus Phaenopsectra sp. 1

PINE CREEK (2003) Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec CHIRONOMINI (Continued) 17 30 13 27 13 27 10 24 08 22 05 19 03 17 01 15 28 10 24 08 22 05 19 03 17 31 Phaenopsectra sp. 2 Phaenopsectra sp. 3 Polypedilum (P.) fallax Polypedilum (P.) illinoense/angulum Polypedilum (P.) laetum Polypedilum (U.) aviceps Saetheria tylus Stictochironomus sp. 1 TANYTARSINI Cladotanytarsus sp. 1 283 Cladotanytarsus sp. 3 Micropsectra geminata Micropsectra nigripila Micropsectra polita Micropsectra sp. 2 Paratanytarsus inopertus gr. sp. 1 Paratanytarsus penicillatus gr. sp. 1 Rheotanytarsus sp. 1 Stempellinella sp. 1 Tanytarsus sepp Tanytarsus sp. 1 Tanytarsus sp. 10

PINE CREEK (2003) Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec TANYTARSINI (Continued) 17 30 13 27 13 27 10 24 08 22 05 19 03 17 01 15 28 10 24 08 22 05 19 03 17 31 Tanytarsus sp. 12 Tanytarsus sp. 18 284

TROUT BROOK (2003) Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec 17 30 13 27 13 27 10 24 08 22 05 19 03 17 01 15 28 10 24 08 22 05 19 03 17 31 TANYPODINAE Paramerina sp. Zavrelimyia sp. DIAMESINAE Diamesa sp. Pagastia orthogonia PRODIAMESINAE Odontomesa fulva

285 Prodiamesa olivacea ORTHOCLADIINAE Acricotopus sp. Brillia flavifrons Bryophaenocladius sp. Cardiocladius sp. 1 Chaetocladius piger gr. sp. 2 Chaetocladius dentiforceps gr. sp. 3 Corynoneura sp. 1 Corynoneura sp. 4 Cricotopus (C.) bicinctus Cricotopus (C.) trifascia Cricotopus (C.) vierriensis

TROUT BROOK (2003) Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec ORTHOCLADIINAE (continued) 17 30 13 27 13 27 10 24 08 22 05 19 03 17 01 15 28 10 24 08 22 05 19 03 17 31 Cricotopus (C.) sp. 1 Cricotopus (C.) sp. 4 Cricotopus (C.) sp. 5 Cricotopus (C.) sp. 6 Cricotopus (C.) sp. 7 Doncricotopus sp. Eukiefferiella brehmi gr. sp. 1 Eukiefferiella claripennis Eukiefferiella ilkleyensis Heterotrissocladius changi 286 Hydrobaenus johannseni Limnophyes sp. 1 Limnophyes sp. 2 Limnophyes sp. 3 Limnophyes sp. 4 Nanocladius rectinervis Nanocladius spiniplenus Orthocladius (E.) rivicola Orthocladius (E.) rivulorum Orthocladius (O.) carlatus Orthocladius (O.) clarkei Orthocladius (O.) dorenus

TROUT BROOK (2003) Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec ORTHOCLADIINAE (continued) 17 30 13 27 13 27 10 24 08 22 05 19 03 17 01 15 28 10 24 08 22 05 19 03 17 31 Orthocladius (O.) frigidus Orthocladius (O.) obumbratus Orthocladius (O.) oliveri Orthocladius (O.) robacki Orthocladius (O.) sp. 3 Orthocladius (S.) lignicola Parachaetocladius sp. Parakiefferiella sp. 1 Parakiefferiella sp. 3 Parakiefferiella sp. 4 287 Parametriocnemus sp. 1 Parametriocnemus sp. 2 Parametriocnemus sp. 3 Parametriocnemus sp. 4 Psilometriocnemus sp. Smittia sp. Stictocladius? sp. Thienemanniella lobapodema Thienemanniella taurocapita Thienemanniella xena Thienemanniella sp. 1 Thienemanniella sp. 2

TROUT BROOK (2003) Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec ORTHOCLADIINAE (continued) 17 30 13 27 13 27 10 24 08 22 05 19 03 17 01 15 28 10 24 08 22 05 19 03 17 31 Tvetenia sp. 1 Tvetenia sp. 2 CHIRONOMINI Dicrotendipes fumidus Paracladopelma undine Paratendipes albimanus Phaenopsectra sp. 3 Polypedilum (P.) laetum Polypedilum (U.) aviceps Saetheria tylus 288 Stictochironomus sp. 1 Stictochironomus sp. 2 TANYTARSINI Cladotanytarsus sp. 1 Micropsectra geminata Micropsectra nigripila Micropsectra polita Micropsectra sp. 2 Paratanytarsus inopertus gr. sp. 1 Paratanytarsus inopertus gr. sp. 2 Paratanytarsus penicillatus gr. sp. 1 Rheotanytarsus sp. 1

TROUT BROOK (2003) Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec ORTHOCLADIINAE (continued) 17 30 13 27 13 27 10 24 08 22 19 03 17 01 15 28 10 24 08 22 05 19 03 17 31 Tanytarsus sp. 10

289

BROWNS CREEK (2002) Jan F Mar Apr May Jun Jul Aug Sep Oct Nov Dec J ? 31 15 1 15 29 10 24 9 22 6 26 1121 7 20 6 19 3 19 31 13 27 12 26 ? TANYPODINAE Ablabesmyia mallochi Conchapelopia fasciata Conchapelopia rurika Zavrelimyia sp. DIAMESINAE Diamesa sp. PRODIAMESINAE Odontomesa fulva Prodiamesa olivacea 290 ORTHOCLADIINAE Brillia flavifrons Cardiocladius albiplumus Cardiocladius sp. 1 Chaetocladius sp. 6 Corynoneura sp. 1 Corynoneura sp. 3 Corynoneura sp. 4 Cricotopus (C.) bicinctus Cricotopus (C.) trifascia Cricotopus (C.) sp. 4 Cricotopus (C.) sp. 5

BROWNS CREEK (2002) Jan F Mar Apr May Jun Jul Aug Sep Oct Nov Dec J ORTHOCLADIINAE (Continued) ? 31 15 1 15 29 10 24 9 22 6 26 11 21 7 20 6 19 3 19 31 13 27 12 26 ? Cricotopus (I.) sylvestris Doncricotopus sp. Eukiefferiella brehmi gr. sp. 1 Eukiefferiella sp. 2 Eukiefferiella claripennis Eukiefferiella ilkleyensis Heterotrissocladius marcidus Krenosmittia sp. Limnophyes sp. 1 Limnophyes sp. 2 291 Limnophyes sp. 3 Nanocladius crassicornis Nanocladius distinctus Nanocladius rectinervis Nanocladius spiniplenus Orthocladius (Eud.) sp. Orthocladius (E.) rivicola Orthocladius (O.) clarkei Orthocladius (O.) frigidus Orthocladius (O.) mallochi Orthocladius (O.) obumbratus Orthocladius (O.) robacki

BROWNS CREEK (2002) Jan F Mar Apr May Jun Jul Aug Sep Oct Nov Dec J ORTHOCLADIINAE (Continued) ? 31 15 1 15 29 10 24 9 22 6 26 11 21 7 20 6 19 3 19 31 13 27 12 26 ? Orthocladius (S.) lignicola Parakiefferiella sp. 3 Parametriocnemus sp. 1 Parametriocnemus sp. 2 Parametriocnemus sp. 3 Parametriocnemus sp. 4 Pseudosmittia sp. Rheocricotopus (P.) sp. 1 Rheocricotopus (R.) effusus Stilocladius sp. 292 Thienemanniella taurocapita Thienemanniella xena Thienemanniella sp. 1 Tvetenia sp. 1 Tvetenia sp. 4 CHIRONOMINI Cryptochironomus sp. 1 Paracladopelma undine Polypedilum (P.) fallax Polypedilum (P.) illinoense/angulum Polypedilum (P.) laetum Polypedilum (U.) aviceps

BROWNS CREEK (2002) Jan F Mar Apr May Jun Jul Aug Sep Oct Nov Dec J CHIRONOMINI (Continued) ? 31 15 1 15 29 10 24 9 22 6 26 11 21 7 20 6 19 3 19 31 13 27 12 26 ? Polypedilum (U.) obtusum Polypedilum sp. 1 Saetheria tylus Saetheria sp. 1 Stenochironomus sp. 2 Stenochironomus sp. 3 TANYTARSINI Cladotanytarsus sp. 1 Cladotanytarsus sp. 3 Micropsectra geminata 293 Micropsectra polita Paratanytarsus inopertus gr. sp. 3 Paratanytarsus penicillatus gr. sp. 2 Rheotanytarsus sp. 1 Rheotanytarsus distinctissimus Stempellinella fimbriata Stempellinella leptocelloides Tanytarsus sp. 10

MILL STREAM (2002) Jan F Mar Apr May Jun Jul Aug Sep Oct Nov Dec J ? 31 15 1 15 29 10 24 9 22 6 26 1123 7 20 6 19 3 17 31 13 27 12 26 ? TANYPODINAE Conchapelopia fasciata Labrundinia neopilosella DIAMESINAE Diamesa sp. PRODIAMESINAE Odontomesa fulva Prodiamesa olivacea ORTHOCLADIINAE Brillia flavifrons 294 Cardiocladius albiplumus Chaetocladius piger gr. sp. 2 Chaetocladius sp. 5 Chaetocladius sp. 6 Corynoneura sp. 1 Corynoneura sp. 4 Cricotopus (C.) trifascia Cricotopus (C.) sp. 4 Cricotopus (C.) sp. 5 Diplocladius sp. Eukiefferiella brehmi gr. sp. 1 Eukiefferiella claripennis

MILL STREAM (2002) Jan F Mar Apr May Jun Jul Aug Sep Oct Nov Dec J ORTHOCLADIINAE (Continued) ? 31 15 1 15 29 10 24 9 22 6 26 11 23 7 20 6 19 3 17 31 13 27 12 26 ? Eukiefferiella ilkleyensis Heterotrissocladius changi Krenosmittia sp. Limnophyes sp. 1 Limnophyes sp. 3 Nanocladius spiniplenus Orthocladius (S.) lignicola Parakiefferiella sp. 3 Parametriocnemus sp. 1 Parametriocnemus sp. 2 295 Parametriocnemus sp. 3 Paraphaenocladius exagitans Paraphaenocladius nasthecus Paraphaenocladius sp. 1 Psilometriocnemus sp. Rheocricotopus (R.) effusoides Stilocladius sp. Synorthocladius sp. Thienemanniella xena Thienemanniella sp. 1 Tvetenia sp. 1 Tvetenia sp. 4

MILL STREAM (2002) Jan F Mar Apr May Jun Jul Aug Sep Oct Nov Dec J ORTHOCLADIINAE (Continued) ? 31 15 1 15 29 10 24 9 22 6 26 11 23 7 20 6 19 3 17 31 13 27 12 26 ? Xylotopus sp. Orthocladiinae Genus A CHIRONOMINI Cryptochironomus sp. 1 Microtendipes sp. 1 Paratendipes albimanus Polypedilum (P.) fallax Polypedilum (P.) laetum Polypedilum (U.) aviceps Polypedilum sp. 3 296 TANYPODINAE Cladotanytarsus sp. 1 Micropsectra geminata Micropsectra nigripila Micropsectra polita Micropsectra sp. 2 Paratanytarsus penicillatus gr. sp. 1 Paratanytarsus penicillatus gr. sp. 2 Rheotanytarsus sp. 1 Tanytarsus sp. 10

VALLEY CREEK (2002) Jan F Mar Apr May Jun Jul Aug Sep Oct Nov Dec J ? 31 15 1 15 29 10 24 9 22 6 26 1126 7 20 6 19 3 17 31 13 27 12 26 ? TANYPODINAE Conchapelopia fasciata DIAMESINAE Diamesa sp. Pagastia orthogonia Potthastia sp. 1 PRODIAMESINAE Odontomesa fulva Prodiamesa olivacea

297 ORTHOCLADIINAE Brillia flavifrons Chaetocladius dentiforceps gr. sp. 3 Corynoneura sp. 1 Corynoneura sp. 4 Cricotopus (C.) bicinctus Cricotopus (C.) trifascia Cricotopus (C.) vierriensis Cricotopus (C.) sp. 4 Cricotopus (C.) sp. 5 Cricotopus (C.) sp. 6 Cricotopus (I.) sp. 1 Doncricotopus sp.

VALLEY CREEK (2002) Jan F Mar Apr May Jun Jul Aug Sep Oct Nov Dec J ORTHOCLADIINAE (Continued) ? 31 15 1 15 29 10 24 9 22 6 26 11 26 7 20 6 19 3 17 31 13 27 12 26 ? Eukiefferiella brehmi gr. sp. 1 Eukiefferiella claripennis Eukiefferiella ilkleyensis Heterotrissocladius changi Heterotrissocladius marcidus Limnophyes sp. 2 Limnophyes sp. 3 Nanocladius spiniplenus Orthocladius (O.) dorenus Orthocladius (O.) frigidus 298 Orthocladius (O.) mallochi Orthocladius (O.) obumbratus Orthocladius (O.) robacki Orthocladius (O.) sp. 3 Orthocladius (S.) lignicola Parachaetocladius sp. Paracladius sp. Parakiefferiella sp. 3 Parametriocnemus sp. 1 Parametriocnemus sp. 2 Paraphaenocladius exagitans Psilometriocnemus sp.

VALLEY CREEK (2002) Jan F Mar Apr May Jun Jul Aug Sep Oct Nov Dec J ORTHOCLADIINAE (Continued) ? 31 15 1 15 29 10 24 9 22 6 26 11 26 7 20 6 19 3 17 31 13 27 12 26 ? Thienemanniella xena Thienemanniella sp. 1 Tvetenia sp. 1 CHIRONOMINI Dicrotendipes fumidus Microtendipes sp. 2 Paracladopelma nais Paratendipes albimanus Polypedilum (P.) laetum Polypedilum (U.) aviceps 299 Stictochironomus sp. 1 TANYTARSINI Micropsectra nigripila Micropsectra polita Micropsectra sp. 2 Paratanytarsus penicillatus gr. sp. 1 Rheotanytarsus sp. 1 Tanytarsus sp. 10 Tanytarsus sp. 18

CEDAR CREEK (2004) Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec 14 28 11 25 9 24 7 21 5 19 2 16 3014 28 11 25 8 22 6 20 3 17 1 16 31 TANYPODINAE Ablabesmyia idei Ablabesmyia mallochi Ablabesmyia monilis Conchapelopia rurika Labrundinia maculata Labrundinia neopilosella Labrundinia pilosella Larsia sp. Natarsia sp. 300 Paramerina sp. Pentaneura sp. Procladius bellus Procladius sp. 1 Procladius sp. 2 Psectrotanypus sp. Zavrelimyia sp. DIAMESINAE Diamesa sp. ORTHOCLADIINAE Acricotopus sp. Corynoneura sp. 1

CEDAR CREEK (2004) Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec ORTHOCLADIINAE (Continued) 14 28 11 25 9 24 7 21 5 19 2 16 30 14 28 11 25 8 22 6 20 3 17 1 16 31 Corynoneura sp. 5 Cricotopus (C.) bicinctus Cricotopus (C.) trifascia Cricotopus (C.) vierriensis Cricotopus (C.) sp. 4 Cricotopus (C.) sp. 5 Cricotopus (I.) sylvestris Cricotopus (I.) sp. 1 Diplocladius sp. Doncricotopus sp. 301 Heterotrissocladius marcidus Hydrobaenus johannseni Limnophyes sp. 1 Limnophyes sp. 2 Limnophyes sp. 3 Nanocladius crassicornis Nanocladius rectinervis Orthocladius (O.) dorenus Orthocladius (O.) obumbratus Orthocladius (O.) oliveri Parakiefferiella sp. 3 Parakiefferiella sp. 4

CEDAR CREEK (2004) Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec ORTHOCLADIINAE (Continued) 14 28 11 25 9 24 7 21 5 19 2 16 30 14 28 11 25 8 22 6 20 3 17 1 16 31 Parametriocnemus sp. 1 Parametriocnemus sp. 2 Paraphaenocladius exagitans Psectrocladius (P.) psilop. gr. sp. 1 Pseudosmittia sp. Rheocricotopus (P.) sp. 1 Rheocricotopus (R.) effusoides Thienemanniella lobapodema Thienemanniella sp. 1 Zalutschia sp. 302 CHIRONOMINI Chironomus sp. 1 Chironomus sp. 3 Cladopelma sp. Cryptochironomus sp. 1 Cryptochironomus sp. 2 Dicrotendipes modestus Dicrotendipes tritomus Dicrotendipes sp. 4 Dicrotendipes sp. 5 Dicrotendipes sp. 6 Dicrotendipes sp. 7

CEDAR CREEK (2004) Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec CHIRONOMINI (Continued) 14 28 11 25 9 24 7 21 5 19 2 16 30 14 28 11 25 8 22 6 20 3 17 1 16 31 Dicrotendipes sp. 8 Einfeldia sp. 1 Einfeldia sp. 2 Endochironomus nigricans Glyptotendipes sp. 1 Microtendipes sp. 1 Microtendipes sp. 2 Parachironomus sp. 1 Parachironomus sp. 2 Parachironomus sp. 3 303 Parachironomus sp. 4 Parachironomus sp. 5 Parachironomus sp. 6 Paracladopelma sp. 1 Paratendipes albimanus Paratendipes sp. 2 Phaenopsectra sp. 2 Phaenopsectra sp. 4 Polypedilum (P.) bergi Polypedilum (P.) fallax Polypedilum (P.) illinoense/angulum Polypedilum (P.) trigonus

CEDAR CREEK (2004) Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec CHIRONOMINI (Continued) 14 28 11 25 9 24 7 21 5 19 2 16 30 14 28 11 25 8 22 6 20 3 17 1 16 31 Polypedilum sp. 1 Polypedilum sp. 3 Polypedilum sp. 5 Stenochironomus sp. 3 Tribelos sp. Xenochironomus sp. TANYTARSINI Cladotanytarsus sp. 7 Micropsectra geminata Micropsectra nigripila 304 Micropsectra polita Micropsectra sp. 2 Paratanytarsus sp. 2 Paratanytarsus inopertus gr. sp. 1 Paratanytarsus cf. laccophilus Rheotanytarsus sp. 1 Rheotanytarsus sp. 2 Rheotanytarsus distinctissimus Stempellinella fimbriata Stempellinella leptocelloides Tanytarsus neoflavellus Tanytarsus sepp

CEDAR CREEK (2004) Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec TANYTARSINI (Continued) 14 28 11 25 9 24 7 21 5 19 2 16 30 14 28 11 25 8 22 6 20 3 17 1 16 31 Tanytarsus wirthi Tanytarsus sp. 1 Tanytarsus sp. 11 Tanytarsus sp. 12 Tanytarsus lobiger Tanytarsus sp. 17 Tanytarsus sp. 18 Tanytarsus sp. 19 Tanytarsus sp. 21 Tanytarsus sp. 22 305 Tanytarsus sp. 23 Tanytarsus sp. 24 Tanytarsus sp. 27

CHUB CREEK (2003) Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec 17 30 13 27 13 27 10 24 08 22 05 19 03 17 01 15 28 10 24 08 22 05 19 03 17 31 TANYPODINAE Ablabesmyia mallochi Ablabesmyia monilis Ablabesmyia peleensis Conchapelopia fasciata Conchapelopia rurika Conchapelopia telema Hayesomyia senata Helopelopia cornuticaudata Labrundinia maculata 306 Labrundinia neopilosella Labrundinia pilosella Larsia sp. Nilotanypus fimbriatus Paramerina sp. Pentaneura sp. Procladius sp. 1 Tanypus sp. DIAMESINAE Diamesa sp. ORTHOCLADIINAE Acricotopus sp.

CHUB CREEK (2003) Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec ORTHOCLADIINAE (Continued) 17 30 13 27 13 27 10 24 08 22 05 19 03 17 01 15 28 10 24 08 22 05 19 03 17 31 Brillia flavifrons Corynoneura sp. 1 Corynoneura sp. 3 Corynoneura sp. 4 Corynoneura sp. 5 Cricotopus (C.) bicinctus Cricotopus (C.) trifascia Cricotopus (C.) vierriensis Cricotopus (C.) sp. 4 Cricotopus (C.) sp. 5 307 Cricotopus (C.) sp. 7 Cricotopus (C.) sp. 12 Cricotopus (I.) sylvestris Cricotopus (I.) sp. 1 Diplocladius sp. Doncricotopus sp. Epoicocladius sp. Eukiefferiella claripennis Hydrobaenus johannseni Hydrobaenus pillipes Limnophyes sp. 1 Limnophyes sp. 2

CHUB CREEK (2003) Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec ORTHOCLADIINAE (Continued) 17 30 13 27 13 27 10 24 08 22 05 19 03 17 01 15 28 10 24 08 22 05 19 03 17 31 Limnophyes sp. 3 Nanocladius sp. 1 Nanocladius crassicornis Nanocladius rectinervis Nanocladius spiniplenus Orthocladius (O.) carlatus Orthocladius (O.) clarkei Orthocladius (O.) dorenus Orthocladius (O.) mallochi Orthocladius (O.) obumbratus 308 Orthocladius (O.) oliveri Orthocladius (O.) robacki Parakiefferiella sp. 3 Parametriocnemus sp. 1 Parametriocnemus sp. 3 Parametriocnemus sp. 4 Paraphaenocladius exagitans Psectrocladius (P.) psilop. gr. sp. 1 Rheocricotopus (P.) sp. 1 Thienemanniella lobapodema Thienemanniella similis Thienemanniella taurocapita

CHUB CREEK (2003) Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec ORTHOCLADIINAE (Continued) 17 30 13 27 13 27 10 24 08 22 05 19 03 17 01 15 28 10 24 08 22 05 19 03 17 31 Thienemanniella sp. 1 Tvetenia sp. 1 Tvetenia sp. 3 CHIRONOMINI Chironomus sp. 1 Chironomus sp. 2 Cryptochironomus ponderosus Cryptochironomus sp. 1 Cryptochironomus sp. 4 Cryptochironomus sp. 9 309 Cryptotendipes sp. Dicrotendipes fumidus Dicrotendipes modestus Dicrotendipes sp. 1 Dicrotendipes sp. 2 Dicrotendipes sp. 4 Endochironomus nigricans Glyptotendipes sp. 1 Harnischia sp. Microtendipes sp. 1 Microtendipes sp. 2 Parachironomus sp. 1

CHUB CREEK (2003) Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec CHIRONOMINI (Continued) 17 30 13 27 13 27 10 24 08 22 05 19 03 17 01 15 28 10 24 08 22 05 19 03 17 31 Paracladopelma nereis Paracladopelma undine Paralauterborniella nigrohalterale Paratendipes albimanus Phaenopsectra sp. 1 Phaenopsectra sp. 2 Polypedilum (P.) fallax Polypedilum (P.) illinoense/angulum Polypedilum (P.) laetum Polypedilum (U.) aviceps 310 Polypedilum (U.) flavum Polypedilum (U.) obtusum Polypedilum sp. 1 Polypedilum sp. 4 Polypedilum sp. 7 Saetheria tylus Stenochironomus sp. 1 Stenochironomus sp. 3 Xenochironomus sp. PSEUDOCHIRONOMINI Pseudochironomus richardsoni

CHUB CREEK (2003) Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec 17 30 13 27 13 27 10 24 08 22 05 19 03 17 01 15 28 10 24 08 22 05 19 03 17 31 TANYTARSINI Cladotanytarsus sp. 1 Cladotanytarsus sp. 2 Cladotanytarsus sp. 3 Cladotanytarsus sp. 4 Cladotanytarsus sp. 5 Micropsectra geminata Micropsectra nigripila Micropsectra polita Paratanytarsus sp. 2 311 Paratanytarsus inopertus gr. sp. 1 Paratanytarsus inopertus gr. sp. 2 Paratanytarsus inopertus gr. sp. 3 Rheotanytarsus sp. 1 Rheotanytarsus sp. 2 Rheotanytarsus distinctissimus Stempellinella sp. 1 Tanytarsus confusus Tanytarsus sepp Tanytarsus wirthi Tanytarsus sp. 1 Tanytarsus sp. 2

CHUB CREEK (2003) Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec TANYTARSINI (Continued) 17 30 13 27 13 27 10 24 08 22 05 19 03 17 01 15 28 10 24 08 22 05 19 03 17 31 Tanytarsus sp. 10 Tanytarsus sp. 11 Tanytarsus sp. 12 Tanytarsus lobiger Tanytarsus sp. 16 Tanytarsus sp. 19

312

CREDIT CREEK (2004) Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec 14 28 11 25 9 24 7 21 5 19 2 16 3014 28 11 25 8 22 6 20 3 17 1 16 31 TANYPODINAE Ablabesmyia mallochi Conchapelopia fasciata Conchapelopia rurika Conchapelopia telema Helopelopia cornuticaudata Labrundinia pilosella Larsia sp. Nilotanypus fimbriatus Procladius sp. 1 313 DIAMESINAE Diamesa sp. PRODIAMESINAE Odontomesa fulva ORTHOCLADIINAE Brillia flavifrons Chaetocladius piger gr. sp. 2 Chaetocladius dentiforceps gr. sp. 3 Chaetocladius sp. 5 Corynoneura sp. 1 Corynoneura sp. 3 Corynoneura sp. 4

CREDIT CREEK (2004) Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec ORTHOCLADIINAE (Continued) 14 28 11 25 9 24 7 21 5 19 2 16 30 14 28 11 25 8 22 6 20 3 17 1 16 31 Cricotopus (C.) bicinctus Cricotopus (C.) trifascia Cricotopus (C.) vierriensis Cricotopus (C.) sp. 4 Cricotopus (C.) sp. 5 Cricotopus (C.) sp. 7 Cricotopus (C.) sp. 12 Cricotopus (I.) sylvestris Diplocladius sp. Doncricotopus sp. 314 Eukiefferiella brehmi gr. sp. 1 Eukiefferiella claripennis Eukiefferiella ilkleyensis Hydrobaenus johannseni Hydrobaenus pillipes Limnophyes sp. 1 Limnophyes sp. 2 Limnophyes sp. 3 Nanocladius sp. 1 Nanocladius crassicornis Nanocladius distinctus Nanocladius rectinervis

CREDIT CREEK (2004) Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec ORTHOCLADIINAE (Continued) 14 28 11 25 9 24 7 21 5 19 2 16 30 14 28 11 25 8 22 6 20 3 17 1 16 31 Nanocladius spiniplenus Orthocladius (O.) carlatus Orthocladius (O.) clarkei Orthocladius (O.) dorenus Orthocladius (O.) mallochi Orthocladius (O.) nigritus Orthocladius (O.) obumbratus Orthocladius (O.) oliveri Orthocladius (O.) robacki Orthocladius (S.) lignicola 315 Parakiefferiella sp. 3 Parametriocnemus sp. 1 Parametriocnemus sp. 3 Parametriocnemus sp. 4 Paraphaenocladius exagitans Rheocricotopus (P.) sp. 1 Thienemanniella taurocapita Thienemanniella xena Thienemanniella sp. 1 Tvetenia sp. 1 Tvetenia sp. 3

CREDIT CREEK (2004) Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec 14 28 11 25 9 24 7 21 5 19 2 16 3014 28 11 25 8 22 6 20 3 17 1 16 31 CHIRONOMINI Chironomus sp. 2 Cryptochironomus eminentia Cryptochironomus ponderosus Cryptochironomus sp. 1 Cryptochironomus sp. 4 Cryptotendipes sp. Dicrotendipes fumidus Dicrotendipes modestus Dicrotendipes sp. 2 316 Microtendipes sp. 1 Microtendipes sp. 2 Paracladopelma nereis Paracladopelma undine Paralauterborniella nigrohalterale Paratendipes albimanus Phaenopsectra sp. 1 Phaenopsectra sp. 2 Polypedilum (P.) fallax Polypedilum (P.) illinoense/angulum Polypedilum (P.) laetum Polypedilum (U.) aviceps

CREDIT CREEK (2004) Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec CHIRONOMINI (Continued) 14 28 11 25 9 24 7 21 5 19 2 16 30 14 28 11 25 8 22 6 20 3 17 1 16 31 Polypedilum (U.) obtusum Polypedilum sp. 1 Polypedilum sp. 8 Saetheria tylus Saetheria sp. 1 Stenochironomus sp. 1 Stenochironomus sp. 3 Stictochironomus sp. 1 TANYTARSINI Cladotanytarsus sp. 1 317 Cladotanytarsus sp. 2 Cladotanytarsus sp. 3 Cladotanytarsus sp. 4 Micropsectra geminata Micropsectra nigripila Micropsectra polita Paratanytarsus inopertus gr. sp. 1 Paratanytarsus inopertus gr. sp. 2 Rheotanytarsus sp. 1 Rheotanytarsus distinctissimus Stempellinella sp. 1 Tanytarsus sepp

CREDIT CREEK (2004) Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec TANYTARSINI (Continued) 14 28 11 25 9 24 7 21 5 19 2 16 30 14 28 11 25 8 22 6 20 3 17 1 16 31 Tanytarsus sp. 1 Tanytarsus sp. 10 Tanytarsus sp. 12 Tanytarsus lobiger

318

ROCK CREEK (2003) Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec ??????1225102306 20 05190320021803163013???? TANYPODINAE Ablabesmyia mallochi Ablabesmyia monilis Conchapelopia fasciata Labrundinia pilosella Natarsia sp. Nilotanypus fimbriatus Paramerina sp. DIAMESINAE Diamesa sp. 319 Potthastia sp. 2 PRODIAMESINAE Odontomesa fulva ORTHOCLADIINAE Brillia flavifrons Cardiocladius albiplumus Cardiocladius sp. 1 Corynoneura sp. 1 Corynoneura sp. 2 Corynoneura sp. 3 Corynoneura sp. 4 Corynoneura sp. 6

ROCK CREEK (2003) Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec ORTHOCLADIINAE (Continued) ??????1225102306 20 05190320021803163013???? Cricotopus (C.) bicinctus Cricotopus (C.) trifascia Cricotopus (C.) sp. 4 Cricotopus (C.) sp. 5 Cricotopus (C.) sp. 7 Cricotopus (C.) sp. 8 Cricotopus (C.) sp. 9 Cricotopus (C.) sp. 10 Doncricotopus sp. Eukiefferiella brehmi gr. sp. 1 320 Eukiefferiella sp. 2 Eukiefferiella sp. 3 Eukiefferiella claripennis Eukiefferiella ilkleyensis Hydrobaenus johannseni Krenosmittia sp. Limnophyes sp. 3 Lopescladius sp. 2 Nanocladius crassicornis Nanocladius rectinervis Nanocladius spiniplenus Orthocladius (E.) rivicola

ROCK CREEK (2003) Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec ORTHOCLADIINAE (Continued) ??????1225102306 20 05190320021803163013???? Orthocladius (E.) rivulorum Orthocladius (O.) carlatus Orthocladius (O.) dorenus Orthocladius (O.) frigidus Orthocladius (O.) mallochi Orthocladius (O.) nigritus Orthocladius (O.) obumbratus Orthocladius (O.) oliveri Orthocladius (S.) lignicola Paracricotopus sp. 321 Parakiefferiella sp. 3 Parakiefferiella sp. 4 Parametriocnemus sp. 1 Parametriocnemus sp. 4 Rheocricotopus (P.) sp. 1 Rheosmittia sp. Synorthocladius sp. Thienemanniella lobapodema Thienemanniella similis Thienemanniella taurocapita Thienemanniella xena Thienemanniella sp. 1

ROCK CREEK (2003) Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec ORTHOCLADIINAE (Continued) ??????1225102306 20 05190320021803163013???? Tvetenia sp. 1 Tvetenia sp. 3 CHIRONOMINI Chironomus sp. 1 Chironomus sp. 2 Cryptochironomus ponderosus Cryptochironomus sp. 1 Cryptochironomus sp. 4 Demicryptochironomus (D.) sp. 2 Dicrotendipes modestus 322 Microtendipes sp. 2 Paracladopelma nereis Paracladopelma undine Paracladopelma sp. 1 Paralauterborniella nigrohalterale Paratendipes albimanus Paratendipes sp. 2 Phaenopsectra sp. 1 Phaenopsectra sp. 2 Polypedilum (P.) fallax Polypedilum (P.) illinoense/angulum Polypedilum (P.) laetum

ROCK CREEK (2003) Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec CHIRONOMINI (Continued) ??????1225102306 20 05190320021803163013???? Polypedilum (U.) aviceps Polypedilum (U.) obtusum Polypedilum sp. 1 Polypedilum sp. 8 Saetheria tylus Stenochironomus sp. 2 Stictochironomus sp. 1 Xenochironomus sp. TANYTARSINI Cladotanytarsus sp. 1 323 Cladotanytarsus sp. 2 Cladotanytarsus sp. 3 Cladotanytarsus sp. 4 Cladotanytarsus sp. 6 Micropsectra nigripila Micropsectra polita Paratanytarsus inopertus gr. sp. 1 Paratanytarsus inopertus gr. sp. 2 Paratanytarsus penicillatus gr. sp. 2 Rheotanytarsus sp. 1 Rheotanytarsus sp. 2 Rheotanytarsus distinctissimus

ROCK CREEK (2003) Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec TANYTARSINI (Continued) ??????1225102306 20 05190320021803163013???? Stempellina sp. Stempellinella sp. 1 Stempellinella fimbriata Stempellinella leptocelloides Sublettea sp. Tanytarsus confusus Tanytarsus sepp Tanytarsus sp. 1 Tanytarsus sp. 10 Tanytarsus sp. 12 324 Tanytarsus sp. 13 Tanytarsus lobiger Tanytarsus sp. 19 Tanytarsus sp. 21 Tanytarsus sp. 22 Tanytarsus sp. 23

RUSH CREEK (2003) Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec ??????1225102306 20 05190320021803163013???? TANYPODINAE Ablabesmyia mallochi Conchapelopia fasciata Conchapelopia telema Labrundinia pilosella Nilotanypus fimbriatus Paramerina sp. Procladius sp. 1 DIAMESINAE Diamesa sp. 325 Pagastia orthogonia Potthastia sp. 2 PRODIAMESINAE Odontomesa fulva Prodiamesa olivacea ORTHOCLADIINAE Brillia flavifrons Cardiocladius albiplumus Cardiocladius sp. 1 Corynoneura sp. 1 Corynoneura sp. 2 Corynoneura sp. 3

RUSH CREEK (2003) Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec ORTHOCLADIINAE (Continued) ??????1225102306 20 05190320021803163013???? Corynoneura sp. 4 Corynoneura sp. 6 Cricotopus (C.) bicinctus Cricotopus (C.) trifascia Cricotopus (C.) sp. 4 Cricotopus (C.) sp. 5 Cricotopus (C.) sp. 7 Cricotopus (C.) sp. 8 Cricotopus (C.) sp. 9 Eukiefferiella brehmi gr. sp. 1 326 Eukiefferiella sp. 2 Eukiefferiella claripennis Eukiefferiella ilkleyensis Heterotrissocladius changi Heterotrissocladius marcidus Hydrobaenus johannseni Krenosmittia sp. Lopescladius sp. 1 Lopescladius sp. 2 Nanocladius crassicornis Nanocladius rectinervis Nanocladius spiniplenus

RUSH CREEK (2003) Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec ORTHOCLADIINAE (Continued) ??????1225102306 20 05190320021803163013???? Orthocladius (E.) rivicola Orthocladius (E.) rivulorum Orthocladius (O.) carlatus Orthocladius (O.) frigidus Orthocladius (O.) mallochi Orthocladius (O.) nigritus Orthocladius (O.) obumbratus Orthocladius (O.) oliveri Orthocladius (S.) lignicola Parakiefferiella sp. 3 327 Parakiefferiella sp. 4 Parametriocnemus sp. 1 Parametriocnemus sp. 3 Parametriocnemus sp. 4 Psectrocladius (P.) psilop. gr. sp. 1 Rheocricotopus (P.) sp. 1 Rheocricotopus (R.) effusoides Rheocricotopus (R.) effusus Rheosmittia sp. Stilocladius sp. Synorthocladius sp. Thienemanniella lobapodema

RUSH CREEK (2003) Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec ORTHOCLADIINAE (Continued) ??????1225102306 20 05190320021803163013???? Thienemanniella similis Thienemanniella taurocapita Thienemanniella xena Thienemanniella sp. 1 Tvetenia sp. 1 Tvetenia sp. 2 Tvetenia sp. 3 Tvetenia sp. 4 CHIRONOMINI Chironomus sp. 2 328 Cryptochironomus sp. 1 Cryptochironomus sp. 3 Cryptochironomus sp. 5 Cryptochironomus sp. 7 Cryptochironomus sp. 8 Cryptotendipes sp. Dicrotendipes modestus Harnischia sp. Microtendipes sp. 2 Microtendipes sp. 3 Paracladopelma sp. 1 Paralauterborniella nigrohalterale

RUSH CREEK (2003) Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec CHIRONOMINI (Continued) ??????1225102306 20 05190320021803163013???? Paratendipes albimanus Phaenopsectra sp. 1 Polypedilum (P.) fallax Polypedilum (P.) illinoense/angulum Polypedilum (P.) laetum Polypedilum (U.) obtusum Polypedilum sp. 1 Polypedilum sp. 8 Stenochironomus sp. 3 Chironomini Genus B 329 TANYTARSINI Cladotanytarsus sp. 1 Cladotanytarsus sp. 3 Micropsectra nigripila Micropsectra polita Paratanytarsus inopertus gr. sp. 1 Paratanytarsus inopertus gr. sp. 2 Paratanytarsus penicillatus gr. sp. 1 Rheotanytarsus sp. 1 Rheotanytarsus sp. 2 Rheotanytarsus distinctissimus Stempellina sp.

RUSH CREEK (2003) Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec TANYTARSINI (Continued) ??????1225102306 20 05190320021803163013???? Stempellinella sp. 1 Stempellinella fimbriata Stempellinella leptocelloides Sublettea sp. Tanytarsus sepp Tanytarsus wirthi Tanytarsus sp. 1 Tanytarsus sp. 10 Tanytarsus sp. 12 Tanytarsus sp. 13 330 Tanytarsus lobiger Tanytarsus sp. 16 Tanytarsus sp. 17 Tanytarsus sp. 21 Tanytarsus sp. 22

SUNRISE RIVER (2003) Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec ? ? ? ? ? ? 12 25 10 23 06 2005 19 03 20 02 18 03 16 30 13 ? ? ? ? TANYPODINAE Ablabesmyia mallochi Conchapelopia fasciata Conchapelopia rurika Hayesomyia senata Helopelopia cornuticaudata Nilotanypus fimbriatus Pentaneura sp. DIAMESINAE Diamesa sp. 331 ORTHOCLADIINAE Brillia flavifrons Cardiocladius albiplumus Cardiocladius sp. 1 Corynoneura sp. 1 Corynoneura sp. 2 Corynoneura sp. 3 Corynoneura sp. 4 Cricotopus (C.) bicinctus Cricotopus (C.) trifascia Cricotopus (C.) sp. 4 Cricotopus (C.) sp. 5

SUNRISE RIVER (2003) Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec ORTHOCLADIINAE (Continued) ? ? ? ? ? ? 12 25 10 23 06 20 05 19 03 20 02 18 03 16 30 13 ? ? ? ? Cricotopus (C.) sp. 7 Cricotopus (C.) sp. 8 Cricotopus (C.) sp. 9 Cricotopus (C.) sp. 12 Diplocladius sp. Doncricotopus sp. Eukiefferiella brehmi gr. sp. 1 Eukiefferiella sp. 2 Eukiefferiella claripennis Eukiefferiella ilkleyensis 332 Heterotrissocladius marcidus Hydrobaenus johannseni Hydrobaenus pillipes Limnophyes sp. 3 Lopescladius sp. 1 Lopescladius sp. 2 Lopescladius sp. 3 Nanocladius sp. 1 Nanocladius crassicornis Nanocladius rectinervis Nanocladius spiniplenus Orthocladius (E.) rivicola

SUNRISE RIVER (2003) Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec ORTHOCLADIINAE (Continued) ? ? ? ? ? ? 12 25 10 23 06 20 05 19 03 20 02 18 03 16 30 13 ? ? ? ? Orthocladius (E.) rivulorum Orthocladius (O.) carlatus Orthocladius (O.) clarkei Orthocladius (O.) mallochi Orthocladius (O.) nigritus Orthocladius (O.) obumbratus Orthocladius (O.) oliveri Orthocladius (S.) lignicola Parakiefferiella sp. 3 Parakiefferiella sp. 6 333 Parametriocnemus sp. 1 Parametriocnemus sp. 3 Parametriocnemus sp. 4 Rheocricotopus (P.) sp. 1 Rheosmittia sp. Stilocladius sp. Synorthocladius sp. Thienemanniella lobapodema Thienemanniella similis Thienemanniella taurocapita Thienemanniella sp. 1 Tvetenia sp. 1

SUNRISE RIVER (2003) Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec ORTHOCLADIINAE (Continued) ? ? ? ? ? ? 12 25 10 23 06 20 05 19 03 20 02 18 03 16 30 13 ? ? ? ? Tvetenia sp. 3 Xylotopus sp. CHIRONOMINI Chironomus sp. 1 Cryptochironomus sp. 1 Cryptochironomus sp. 5 Demicryptochironomus (Irmakia) sp. 1 Microtendipes sp. 3 Paracladopelma nereis Paralauterborniella nigrohalterale 334 Paratendipes albimanus Phaenopsectra sp. 1 Polypedilum (P.) fallax Polypedilum (P.) illinoense/angulum Polypedilum (P.) laetum Polypedilum (U.) aviceps Polypedilum (U.) flavum Polypedilum (U.) obtusum Polypedilum sp. 1 Polypedilum sp. 7 Polypedilum sp. 8 Robackia sp.

SUNRISE RIVER (2003) Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec CHIRONOMINI (Continued) ? ? ? ? ? ? 12 25 10 23 06 20 05 19 03 20 02 18 03 16 30 13 ? ? ? ? Saetheria sp. 1 Stenochironomus sp. 3 Chironomini Genus B TANYTARSINI Cladotanytarsus sp. 1 Cladotanytarsus sp. 3 Micropsectra nigripila Micropsectra polita Micropsectra sp. 2 Rheotanytarsus sp. 1 335 Rheotanytarsus sp. 2 Rheotanytarsus distinctissimus Stempellinella sp. 1 Stempellinella fimbriata Sublettea sp. Tanytarsus sepp Tanytarsus sp. 2 Tanytarsus sp. 10 Tanytarsus sp. 12 Tanytarsus sp. 13 Tanytarsus lobiger Tanytarsus sp. 19

SUNRISE RIVER (2003) Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec TANYTARSINI (Continued) ? ? ? ? ? ? 12 25 10 23 06 20 05 19 03 20 02 18 03 16 30 13 ? ? ? ? Tanytarsus sp. 22 Tanytarsus sp. 23

336

Appendix E: Estimated thermal preferences (ET50 = temperature at which 50% of emergence occurred) and total study abundance for chironomid subfamilies/tribes, genera, and species ordered from cool to warm thermal preferences; The subfamily/tribe genus tables also include the number of taxa within each higher taxonomic grouping and the range of thermal preferences within higher groupings for taxa with more than one taxa.

337

Subfamily/Tribe ET50 Range Abundance #Taxa #Genera Diamesinae 6.32 10.09 4393 4 3 Prodiamesinae 11.22 0.42 1414 2 2 Orthocladiinae 13.71 23.23 56847 107 38 Tanytarsini 15.78 11.57 12976 46 8 Chironomini 16.03 16.26 6146 77 24 Pseudochironomini 17.27 - 2 1 1 Tanypodinae 18.37 14.00 1113 25 16

338 Subfamily/Tribe Genus Abundance #Taxa ET50 Range Orthocladiinae Chaetocladius 120 5 3.11 9.81 Orthocladiinae Stilocladius 265 1 5.28 Diamesinae Diamesa 4265 1 6.19 Orthocladiinae Hydrobaenus 88 2 6.22 2.11 Tanypodinae Psectrotanypus 1 1 7.08 Orthocladiinae Orthocladiinae Genus A 1 1 7.71 Orthocladiinae Bryophaenocladius 3 1 8.53 Orthocladiinae Diplocladius 464 1 9.16 Orthocladiinae Parachaetocladius 361 1 9.55 Orthocladiinae Stictocladius? 6 1 9.73 Orthocladiinae Orthocladius 9294 14 10.22 5.58 Diamesinae Pagastia 112 1 10.28 Prodiamesinae Prodiamesa 433 1 10.92 Tanypodinae Zavrelimyia 31 1 11.04 Chironomini Stictochironomus 303 2 11.06 1.24 Orthocladiinae Heterotrissocladius 67 2 11.15 0.39 Tanypodinae Meropelopia 13 1 11.18 Tanypodinae Natarsia 8 1 11.25 Prodiamesinae Odontomesa 981 1 11.34 Orthocladiinae Synorthocladius 47 1 11.43 Orthocladiinae Limnophyes 372 4 11.57 5.01 Orthocladiinae Psilometriocnemus 16 1 11.57 Orthocladiinae Pseudosmittia 3 1 11.67 Orthocladiinae Eukiefferiella 2922 5 11.96 4.47 Tanytarsini Micropsectra 1889 4 12.07 2.48 Orthocladiinae Paracladius 3 1 12.49 Orthocladiinae Xylotopus 2 1 12.86 Orthocladiinae Parakiefferiella 4678 4 12.92 11.70 Orthocladiinae Parametriocnemus 2170 4 13.16 6.30 Orthocladiinae Brillia 181 1 13.25 Orthocladiinae Smittia 1 1 13.53 Orthocladiinae Corynoneura 7775 6 13.55 9.41 Orthocladiinae Paraphaenocladius 32 3 13.74 2.01 Diamesinae Potthastia 16 2 13.97 4.11 Orthocladiinae Nanocladius 2132 5 13.98 7.17 Orthocladiinae Zalutschia 1 1 14.06 Chironomini Paracladopelma 115 4 14.27 6.76 Orthocladiinae Cricotopus 8436 14 14.33 10.75

339 Subfamily/Tribe Genus Abundance #Taxa ET50 Range Orthocladiinae Doncricotopus 175 1 14.48 Chironomini Paratendipes 478 2 14.48 4.31 Tanypodinae Radotanypus 2 1 14.53 Orthocladiinae Tvetenia 4714 4 14.60 10.26 Chironomini Polypedilum 1910 14 14.78 12.14 Chironomini Saetheria 44 2 15.04 1.10 Chironomini Microtendipes 491 3 15.51 2.55 Tanypodinae Paramerina 19 1 15.54 Tanytarsini Paratanytarsus 2484 7 15.59 9.54 Orthocladiinae Krenosmittia 108 1 15.78 Tanytarsini Sublettea 23 1 16.08 Tanytarsini Tanytarsus 4384 20 16.14 7.85 Tanypodinae Conchapelopia 125 3 16.26 4.47 Orthocladiinae Thienemanniella 8656 6 16.39 11.73 Tanytarsini Rheotanytarsus 1982 3 16.45 1.60 Orthocladiinae Cardiocladius 165 2 16.63 2.94 Chironomini Chironomus 265 3 16.68 7.78 Chironomini Chironomini Genus B 2 1 16.71 Chironomini Tribelos 47 1 16.78 Orthocladiinae Psectrocladius 47 1 16.82 Orthocladiinae Rheosmittia 366 1 17.05 Tanytarsini Stempellina 5 1 17.20 Orthocladiinae Acricotopus 41 1 17.22 Pseudochironomini Pseudochironomus 2 1 17.27 Tanytarsini Stempellinella 373 3 17.53 1.48 Tanypodinae Procladius 169 3 17.63 10.69 Chironomini Glyptotendipes 13 1 17.67 Chironomini Dicrotendipes 1775 10 17.77 7.82 Chironomini Phaenopsectra 255 4 17.83 9.72 Tanytarsini Cladotanytarsus 1836 7 17.89 5.67 Tanypodinae Pentaneura 14 1 18.14 Chironomini Parachironomus 14 6 18.16 2.10 Chironomini Einfeldia 2 2 18.29 8.46 Tanypodinae Larsia 51 1 18.30 Chironomini Cryptochironomus 133 10 18.40 9.17 Orthocladiinae Rheocricotopus 1128 3 18.42 7.69 Chironomini Endochironomus 6 1 18.60 Chironomini Cladopelma 33 1 18.72 Tanypodinae Helopelopia 9 1 18.74 340 Subfamily/Tribe Genus Abundance #Taxa ET50 Range Chironomini Xenochironomus 38 1 18.75 Chironomini Stenochironomus 42 3 18.75 6.85 Chironomini Paralauterborniella 118 1 18.85 Tanypodinae Ablabesmyia 310 4 19.09 3.79 Tanypodinae Hayesomyia 11 1 19.68 Orthocladiinae Lopescladius 2003 3 19.71 2.41 Tanypodinae Labrundinia 110 3 19.98 1.71 Tanypodinae Nilotanypus 239 1 20.15 Tanypodinae Tanypus 1 1 21.08 Chironomini Cryptotendipes 44 1 21.55 Chironomini Robackia 9 1 21.67 Chironomini Harnischia 3 1 22.19 Chironomini Demicryptochironomus 4 2 23.81 2.26 Orthocladiinae Epoicocladius 1 1 24.52 Orthocladiinae Paracricotopus 3 1 24.68

341 Subfamily/Tribe Taxon Abundance ET50 Orthocladiinae Chaetocladius sp. 5 28 1.44 Orthocladiinae Chaetocladius piger gr. sp. 2 71 2.17 Orthocladiinae Stilocladius sp. 265 5.28 Orthocladiinae Chaetocladius sp. 6 5 5.38 Orthocladiinae Hydrobaenus johannseni 74 5.89 Diamesinae Diamesa sp. 4265 6.19 Tanypodinae Psectrotanypus sp. 1 7.08 Orthocladiinae Orthocladiinae Genus A 1 7.71 Orthocladiinae Hydrobaenus pillipes 14 8.00 Orthocladiinae Orthocladius (O.) frigidus 351 8.22 Orthocladiinae Orthocladius (O.) oliveri 1531 8.38 Tanypodinae Procladius sp. 2 3 8.46 Orthocladiinae Bryophaenocladius sp. 3 8.53 Orthocladiinae Limnophyes sp. 4 1 8.53 Orthocladiinae Orthocladius (E.) rivulorum 244 8.72 Orthocladiinae Chaetocladius dentiforceps gr. sp. 3 16 8.81 Orthocladiinae Tvetenia sp. 4 11 8.81 Orthocladiinae Parakiefferiella sp. 1 227 9.03 Orthocladiinae Diplocladius sp. 464 9.16 Chironomini Chironomus sp. 3 11 9.24 Orthocladiinae Parachaetocladius sp. 361 9.55 Orthocladiinae Orthocladius (O.) obumbratus 3721 9.70 Orthocladiinae Stictocladius? sp. 6 9.73 Orthocladiinae Thienemanniella sp. 2 1 9.88 Tanytarsini Micropsectra geminata 24 9.90 Orthocladiinae Orthocladius (Eud.) sp. 2 10.00 Orthocladiinae Cricotopus (C.) sp. 1 662 10.02 Chironomini Phaenopsectra sp. 3 15 10.09 Orthocladiinae Orthocladius (O.) nigritus 666 10.09 Orthocladiinae Orthocladius (O.) sp. 3 58 10.15 Diamesinae Pagastia orthogonia 112 10.28 Orthocladiinae Orthocladius (E.) rivicola 609 10.49 Orthocladiinae Tvetenia sp. 2 352 10.65 Orthocladiinae Parakiefferiella sp. 6 12 10.66 Chironomini Stictochironomus sp. 2 241 10.80 Orthocladiinae Rheocricotopus (R.) effusoides 26 10.92 Prodiamesinae Prodiamesa olivacea 433 10.92 Orthocladiinae Heterotrissocladius marcidus 38 10.98 Tanytarsini Paratanytarsus penicillatus gr. sp. 1 319 11.01 342 Subfamily/Tribe Taxon Abundance ET50 Orthocladiinae Parametriocnemus sp. 2 676 11.02 Tanypodinae Zavrelimyia sp. 31 11.04 Orthocladiinae Limnophyes sp. 1 250 11.05 Orthocladiinae Thienemanniella xena 857 11.10 Tanypodinae Meropelopia sp. 1 13 11.18 Orthocladiinae Eukiefferiella claripennis 1566 11.18 Tanypodinae Natarsia sp. 8 11.25 Orthocladiinae Limnophyes sp. 2 47 11.28 Tanytarsini Micropsectra sp. 2 434 11.33 Prodiamesinae Odontomesa fulva 981 11.34 Orthocladiinae Heterotrissocladius changi 29 11.37 Orthocladiinae Cricotopus (C.) sp. 6 953 11.39 Orthocladiinae Synorthocladius sp. 47 11.43 Orthocladiinae Orthocladius (O.) dorenus 75 11.43 Orthocladiinae Eukiefferiella sp. 2 130 11.46 Orthocladiinae Psilometriocnemus sp. 16 11.57 Orthocladiinae Orthocladius (O.) robacki 320 11.64 Orthocladiinae Pseudosmittia sp. 3 11.67 Orthocladiinae Orthocladius (O.) clarkei 19 11.81 Orthocladiinae Orthocladius (S.) lignicola 146 11.82 Orthocladiinae Corynoneura sp. 1 5607 12.01 Chironomini Stictochironomus sp. 1 62 12.05 Diamesinae Potthastia sp. 1 9 12.17 Tanytarsini Micropsectra polita 831 12.30 Tanytarsini Micropsectra nigripila 600 12.37 Orthocladiinae Nanocladius spiniplenus 1444 12.38 Chironomini Polypedilum (U.) aviceps 870 12.38 Orthocladiinae Orthocladius (O.) mallochi 432 12.39 Orthocladiinae Cricotopus (C.) vierriensis 401 12.43 Orthocladiinae Paracladius sp. 3 12.49 Tanytarsini Tanytarsus sp. 1 745 12.52 Orthocladiinae Parametriocnemus sp. 3 74 12.52 Orthocladiinae Parametriocnemus sp. 1 926 12.55 Tanytarsini Tanytarsus sp. 18 54 12.62 Orthocladiinae Cricotopus (C.) sp. 4 1020 12.62 Orthocladiinae Eukiefferiella brehmi gr. sp. 1 449 12.66 Chironomini Paracladopelma undine 74 12.68 Orthocladiinae Xylotopus sp. 2 12.86 Chironomini Paracladopelma nais 14 13.02 343 Subfamily/Tribe Taxon Abundance ET50 Orthocladiinae Cricotopus (C.) sp. 8 135 13.05 Orthocladiinae Parakiefferiella sp. 3 4428 13.11 Orthocladiinae Eukiefferiella ilkleyensis 774 13.19 Tanytarsini Tanytarsus sp. 2 2 13.25 Orthocladiinae Brillia flavifrons 181 13.25 Chironomini Polypedilum (P.) laetum 186 13.40 Orthocladiinae Smittia sp. 1 13.53 Orthocladiinae Limnophyes sp. 3 74 13.54 Orthocladiinae Paraphaenocladius exagitans 28 13.56 Orthocladiinae Orthocladius (O.) carlatus 1120 13.80 Orthocladiinae Rheocricotopus (R.) effusus 3 13.83 Tanytarsini Paratanytarsus penicillatus gr. sp. 2 538 13.83 Orthocladiinae Zalutschia sp. 1 14.06 Chironomini Einfeldia sp. 1 1 14.06 Chironomini Microtendipes sp. 1 218 14.30 Orthocladiinae Cardiocladius albiplumus 41 14.42 Orthocladiinae Paraphaenocladius sp. 1 2 14.44 Chironomini Paratendipes albimanus 474 14.45 Orthocladiinae Doncricotopus sp. 175 14.48 Tanypodinae Radotanypus sp. 2 14.53 Chironomini Cryptochironomus sp. 5 2 14.65 Tanytarsini Tanytarsus sp. 19 183 14.67 Chironomini Dicrotendipes fumidus 889 14.69 Chironomini Stenochironomus sp. 2 2 14.70 Orthocladiinae Tvetenia sp. 1 4224 14.81 Orthocladiinae Cricotopus (C.) trifascia 961 14.94 Chironomini Cryptochironomus sp. 9 2 14.97 Chironomini Saetheria tylus 41 14.97 Tanytarsini Tanytarsus sp. 21 397 15.12 Chironomini Dicrotendipes sp. 5 3 15.16 Orthocladiinae Cricotopus (C.) bicinctus 2558 15.46 Chironomini Polypedilum (U.) flavum 3 15.54 Tanypodinae Paramerina sp. 19 15.54 Tanypodinae Conchapelopia rurika 59 15.56 Orthocladiinae Paraphaenocladius nasthecus 2 15.56 Orthocladiinae Eukiefferiella sp. 3 3 15.65 Tanytarsini Tanytarsus sp. 10 555 15.70 Orthocladiinae Cricotopus (C.) sp. 5 999 15.70 Orthocladiinae Corynoneura sp. 4 1256 15.77 344 Subfamily/Tribe Taxon Abundance ET50 Orthocladiinae Krenosmittia sp. 108 15.78 Tanytarsini Cladotanytarsus sp. 1 707 15.80 Chironomini Phaenopsectra sp. 2 36 15.90 Orthocladiinae Thienemanniella sp. 1 5411 15.92 Chironomini Saetheria sp. 1 3 16.07 Tanytarsini Sublettea sp. 23 16.08 Chironomini Polypedilum (P.) illinoense/angulum 246 16.09 Tanypodinae Ablabesmyia idei 1 16.14 Chironomini Dicrotendipes sp. 7 2 16.14 Chironomini Dicrotendipes sp. 8 1 16.14 Orthocladiinae Cricotopus (C.) sp. 10 1 16.24 Tanytarsini Cladotanytarsus sp. 6 2 16.24 Tanytarsini Rheotanytarsus sp. 1 1736 16.28 Diamesinae Potthastia sp. 2 7 16.28 Chironomini Cryptochironomus eminentia 2 16.32 Chironomini Chironomus sp. 1 8 16.40 Chironomini Microtendipes sp. 2 271 16.48 Orthocladiinae Cricotopus (C.) sp. 12 115 16.50 Tanypodinae Conchapelopia fasciata 59 16.52 Tanytarsini Tanytarsus sp. 13 5 16.55 Tanytarsini Rheotanytarsus sp. 2 49 16.65 Orthocladiinae Nanocladius rectinervis 508 16.67 Chironomini Chironomini Genus B 2 16.71 Tanytarsini Paratanytarsus inopertus gr. sp. 2 994 16.73 Chironomini Tribelos sp. 47 16.78 Orthocladiinae Psectrocladius (P.) psilopterus gr. sp. 1 47 16.82 Chironomini Microtendipes sp. 3 2 16.84 Chironomini Parachironomus sp. 1 3 16.85 Orthocladiinae Cricotopus (I.) sp. 1 52 16.91 Tanytarsini Tanytarsus sp. 27 24 16.96 Chironomini Chironomus sp. 2 246 17.02 Tanytarsini Tanytarsus sp. 22 942 17.02 Tanytarsini Stempellinella sp. 1 187 17.03 Orthocladiinae Corynoneura sp. 5 246 17.04 Orthocladiinae Rheosmittia sp. 366 17.05 Tanytarsini Tanytarsus lobiger 429 17.06 Chironomini Polypedilum (P.) fallax 41 17.11 Tanytarsini Stempellina sp. 5 17.20 Orthocladiinae Acricotopus sp. 41 17.22 345 Subfamily/Tribe Taxon Abundance ET50 Pseudochironomini Pseudochironomus richardsoni 2 17.27 Orthocladiinae Parametriocnemus sp. 4 494 17.32 Chironomini Parachironomus sp. 6 2 17.35 Chironomini Polypedilum (P.) bergi 1 17.35 Orthocladiinae Cardiocladius sp. 1 124 17.36 Chironomini Dicrotendipes tritomus 5 17.43 Tanytarsini Paratanytarsus sp. 2 30 17.51 Tanytarsini Tanytarsus neoflavellus 31 17.55 Tanytarsini Paratanytarsus inopertus gr. sp. 1 586 17.56 Tanytarsini Tanytarsus sp. 16 4 17.59 Chironomini Glyptotendipes sp. 1 13 17.67 Chironomini Polypedilum sp. 8 8 17.71 Chironomini Cryptochironomus sp. 1 59 17.75 Tanypodinae Procladius sp. 1 162 17.76 Tanytarsini Stempellinella fimbriata 119 17.77 Chironomini Polypedilum (U.) obtusum 250 17.88 Tanytarsini Rheotanytarsus distinctissimus 197 17.88 Tanytarsini Paratanytarsus laccophilus 7 17.97 Chironomini Polypedilum sp. 1 237 17.97 Tanytarsini Tanytarsus wirthi 42 17.99 Chironomini Dicrotendipes sp. 4 46 18.07 Orthocladiinae Nanocladius distinctus 3 18.08 Tanytarsini Tanytarsus sp. 17 34 18.09 Tanytarsini Tanytarsus sp. 12 377 18.12 Tanypodinae Pentaneura sp. 14 18.14 Chironomini Paracladopelma sp. 1 3 18.14 Tanypodinae Larsia sp. 51 18.30 Tanypodinae Labrundinia neopilosella 6 18.40 Tanytarsini Tanytarsus sp. 24 5 18.41 Tanytarsini Tanytarsus sp. 23 379 18.45 Tanytarsini Stempellinella leptocelloides 67 18.51 Orthocladiinae Thienemanniella taurocapita 1670 18.52 Chironomini Parachironomus sp. 5 2 18.58 Chironomini Stenochironomus sp. 3 35 18.58 Tanytarsini Cladotanytarsus sp. 3 827 18.58 Tanypodinae Ablabesmyia monilis 100 18.58 Chironomini Endochironomus nigricans 6 18.60 Orthocladiinae Cricotopus (C.) sp. 9 130 18.60 Orthocladiinae Rheocricotopus (P.) sp. 1 1099 18.61 346 Subfamily/Tribe Taxon Abundance ET50 Chironomini Cryptochironomus sp. 4 21 18.64 Chironomini Parachironomus sp. 2 2 18.67 Chironomini Parachironomus sp. 3 1 18.67 Tanytarsini Cladotanytarsus sp. 7 1 18.67 Chironomini Cladopelma sp. 33 18.72 Tanypodinae Labrundinia maculata 3 18.72 Tanypodinae Helopelopia cornuticaudata 9 18.74 Chironomini Phaenopsectra sp. 1 203 18.74 Chironomini Xenochironomus sp. 38 18.75 Chironomini Cryptochironomus sp. 2 31 18.75 Chironomini Paratendipes sp. 2 4 18.76 Chironomini Dicrotendipes sp. 2 22 18.77 Chironomini Paralauterborniella nigrohalterale 118 18.85 Chironomini Parachironomus sp. 4 4 18.95 Orthocladiinae Lopescladius sp. 3 1 19.01 Orthocladiinae Tvetenia sp. 3 127 19.07 Orthocladiinae Nanocladius crassicornis 141 19.10 Chironomini Polypedilum sp. 3 23 19.11 Tanypodinae Procladius bellus 4 19.15 Chironomini Polypedilum (P.) trigonus 6 19.28 Tanypodinae Ablabesmyia mallochi 208 19.35 Chironomini Paracladopelma nereis 24 19.44 Orthocladiinae Lopescladius sp. 2 1742 19.46 Orthocladiinae Nanocladius sp. 1 36 19.55 Tanytarsini Tanytarsus sp. 11 16 19.59 Chironomini Cryptochironomus sp. 7 5 19.63 Tanypodinae Hayesomyia senata 11 19.68 Orthocladiinae Cricotopus (C.) sp. 7 286 19.69 Tanytarsini Tanytarsus confusus 2 19.71 Chironomini Phaenopsectra sp. 4 1 19.80 Chironomini Polypedilum sp. 5 1 19.80 Tanypodinae Ablabesmyia peleensis 1 19.92 Chironomini Cryptochironomus ponderosus 6 19.97 Tanypodinae Conchapelopia telema 7 20.03 Tanypodinae Labrundinia pilosella 101 20.11 Tanypodinae Nilotanypus fimbriatus 239 20.15 Orthocladiinae Corynoneura sp. 2 237 20.28 Tanytarsini Tanytarsus sepp 158 20.37 Tanytarsini Paratanytarsus inopertus gr. sp. 3 10 20.55 347 Subfamily/Tribe Taxon Abundance ET50 Orthocladiinae Parakiefferiella sp. 4 11 20.73 Tanytarsini Cladotanytarsus sp. 4 190 20.74 Orthocladiinae Cricotopus (I.) sylvestris 163 20.77 Orthocladiinae Thienemanniella similis 215 20.91 Chironomini Polypedilum sp. 7 37 21.01 Tanypodinae Tanypus sp. 1 21.08 Chironomini Dicrotendipes modestus 805 21.14 Orthocladiinae Corynoneura sp. 6 6 21.26 Tanytarsini Cladotanytarsus sp. 2 107 21.27 Orthocladiinae Lopescladius sp. 1 260 21.42 Orthocladiinae Corynoneura sp. 3 423 21.42 Tanytarsini Cladotanytarsus sp. 5 2 21.47 Chironomini Cryptotendipes sp. 44 21.55 Chironomini Stenochironomus sp. 1 5 21.55 Orthocladiinae Thienemanniella lobapodema 502 21.60 Chironomini Robackia sp. 9 21.67 Chironomini Harnischia sp. 3 22.19 Chironomini Dicrotendipes sp. 1 1 22.48 Chironomini Dicrotendipes sp. 6 1 22.51 Chironomini Einfeldia sp. 2 1 22.51 Chironomini Cryptochironomus sp. 8 1 22.84 Chironomini Demicryptochironomus (D.) sp. 2 3 23.24 Chironomini Cryptochironomus sp. 3 4 23.82 Orthocladiinae Epoicocladius sp. 1 24.52 Chironomini Polypedilum sp. 4 1 24.52 Orthocladiinae Paracricotopus sp. 3 24.68 Chironomini Demicryptochironomus (Irmakia) sp. 1 1 25.50

348

median; upperandlowerbounds=range;thewidthof temperature atemergence(i

emergence temperature(solidcircle=mean; line = for chironomidgenerainascendingorderofmean A

the plotsrepresentrelativeabundanceandnot PPENDIX

F: absolute abundance).

Violin plotsofmeandailywater

349

.e., thermalpreferences)

350

351

REFERENCES CITED

Aagaard, K. 1978a. The Chironomidae of the exposed zone of Øvre Heimdalsvatn. Holarctic Ecology 1:261-265. Aagaard, K. 1978b. The chironomids of Lake Målsjøen. A phenological, diversity, and production study. Norwegian Journal of Entomology 25:21-37. Abbott, W.S. 1925. A method of computing the effectiveness of an insecticide. Journal of Economic Entomology 18:265-267. Adams, A. 1983. Cryptobiosis in Chironomidae (Diptera) - two decades on. Antenna 8:58-61. Adler, P.H., D.C. Currie, & D.M. Wood. 2004. The black flies (Simuliidae) of North America. Cornell University Press, Ithaca. Alverson, A.J., G.W. Courtney & M.R. Luttenton. 2001. Niche overlap of sympatric Blepharicera larvae (Diptera: Blephariceridae) from the southern Appalachian Mountains. Journal of the North American Benthological Society 20:564-581. Andersen, F.S. 1946. East Greenland lakes as habitats for chironomid larvae. Meddelelser om Grønland 100:1-65. Anderson, J.F. & S.W. Hitchcock. 1968. Biology of Chironomus atrella in a tidal cove. Annals of the Entomological Society of America 61:1597-1603. Andrews, D. & F.H. Rigler. 1985. The effects of and Arctic winter on benthic invertebrates in the littoral zone of Char Lake, Northwest Territories. Canadian Journal of Zoology 63:2825-2834. Añón Suárez, D.A. 2002. Life history and secondary production of Ablabesmyia reissi (Diptera: Chironomidae) from Lake Escondido, Bariloche, Argentina. Journal of the North American Benthological Society 21:414-429. Armitage, P.D. 1968. Some notes on the food of the chironomid larvae of a shallow woodland lake in South Finland. Annales Zoologici Fennici 5:6-13. Armitage, P.D. 1970. The Tanytarsini (Diptera, Chironomidae) of a shallow woodland lake in South Finland, with species reference to the effect of winter conditions on the larvae. Annales Zoologici Fennici 7:313-322.

352 Armitage, P.D. 1995. Behavior and ecology of adults, p. 194-224. In: The Chironomidae: the biology and ecology of non-biting midges. P.D. Armitage, P.S. Cranston, and L. C.V. Pinder (eds.). Chapman & Hall, London, UK. Arthington, A.H. 1990. Latitudinal gradients in insect species richness of Australian lotic systems: a selective review. Tropical Freshwater Biology 2:179-196. Ashton, G.D. 1979. River ice. American Scientist 67:38-45. Bale, J.S. 1989. Cold hardiness and overwintering of insects. Agricultural Zoology Reviews 3:157-192. Bale, J.S. 1991. Implications of cold hardiness for pest management. In Lee R. E., Jr. & D. L. Denlinger (eds), Insects at low temperature. Chapman & Hall, New York: 461-498. Bale, J.S. 1996. Insect cold hardiness: a matter of life and death. European Journal of Entomology 93:369-382. Bale, J.S., R. Harrington, & M.S. Clough. 1988. Low temperature mortality of the peach- potato aphid Myzus persicae. Ecological Entomology 13:121−129. Ball, S.L. & R.L. Baker. 1995. The non-lethal effects of predators and the influence of food availability on life history of adult Chironomus tentans (Diptera: Chironomidae). Freshwater Biology 34:1-12. Ball, S.L. & R.L. Baker. 1996. Predator-induced life history changes: antipredator behavior costs or facultative life history shifts. Ecology 77:1116-1124. Barbour, M.T. & J. Gerritsen. 1996. Subsampling of benthic samples: a defense of the fixed-count method. Journal of the North American Benthological Society 15:386-391. Barbour, M.T., J. Gerritsen, B.D. Snyder & J.B. Stribling. 1999. Rapid bioassessment protocols for use in streams and wadeable rivers: periphyton, benthic macroinvertebrates and fish, second edition. U.S. Environmental Protection Agency, Washington, D.C. Baust, J.G. & J.S. Edwards. 1979. Mechanisms of freezing tolerance in an Antarctic midge, Belgica antarctica. Physiological Entomology 4:1-5.

353 Baust, J.G. & R.E. Lee. 1981. Environmental "homeothermy" in an Antarctic insect. Antarctic Journal of the United States 15:170-172. Baust, J.G. & R.R. Rojas. 1985. Review - insect cold hardiness: facts and fancy. Journal of Insect Physiology 31:755-759. Bazerque, M.F., H. Laville, & Y. Brouquet. 1989. Biological quality assessment in two rivers of the northern plain of France (Picardie) with special reference to chironomid and diatom indices. Acta Biologica Debrecen Oecologica Hungarica 3:29-39. Beattie, D.M. 1978. Life-cycle and changes in carbohydrates, proteins and lipids of Polypedilum uncinatum Goet. (Diptera; Chironomidae). Freshwater Biology 8:109-113. Benke, A.C. 1978. Interactions among coexisting predators - a field experiment with dragonfly larvae. Journal of Animal Ecology 47:335-350. Benke, A.C. & S.S. Benke. 1975. Comparative dynamics and life histories of coexisting dragonfly populations. Ecology 56:302-317. Berg, C.O. 1949. Limnological relations of insects to plants of the genus Potamogeton. Transactions of the American Microscopical Society 68:279-291. Berg, M.B. 1995. Larval food and feeding behavior, p. 136-168. In: The Chironomidae: the biology and ecology of non-biting midges. P.D. Armitage, P.S. Cranston, & L.C.V. Pinder (eds.). Chapman & Hall, London, UK. Berg M.B. & R.A. Hellenthal. 1991. Secondary production of Chironomidae (Diptera) in a north temperate stream. Freshwater Biology 25:497-505. Berg, M.B. & R.A. Hellenthal. 1992. The role of Chironomidae in energy flow of a lotic ecosystem. Netherlands Journal of Aquatic Ecology 26:471-476. Biever, K.D. 1971. Effect of diet and competition in laboratory rearing of chironomid midges. Annals of the Entomological Society of America 64:1166-1169. Bilyj, B. & I.J. Davies. 1989. Descriptions and ecological notes on seven new species of Cladotanytarsus (Chironomidae: Diptera) collected from an experimentally acidified lake. Canadian Journal of Zoology 67:948-962.

354 Blackwood, M.A., S.H. Hall, & Ferrington, L.C., Jr. 1995. Emergence of Chironomidae from springs in the Central High Plains regions of the United States. Journal of the Kansas Entomological Society 68:132-151. Block, W. 1982. Supercooling points of insects and mites on the Antarctic Peninsula. Ecological Entomology 7:1-8. Block, W. 1991. To freeze or not to freeze? invertebrate survival of sub-zero temperatures. Functional Ecology 5:284-290. Block, W., Y.Z. Erzinclioglu, & M.R. Worland. 1988. Survival of freezing in Calliphora larvae. CryoLetters 9:86-93. Boerger, H. 1981. Species composition, abundance and emergence phenology of midges (Diptera: Chironomidae) in a brown-water stream of West-Central Alberta, Canada. Hydrobiologia 80:7-30. Boothroyd, I.K.G. 1987. Taxonomic composition and life cycles of Chironomidae (Diptera) in a northern New Zealand stream. Entomologica Scandinavica Supplementum 29:15-22. Boothroyd, I.K.G. 1988. Temporal and diel emergence of Chironomidae (Diptera: Insecta) from a New Zealand stream. Verhandlungen der Internationale Vereinigung für Theoretische und Angewandte Limnologie 23:1399-1404. Boothroyd, I.K.G. 1999. Life history of Kaniwhaniwhanus chapmani Boothroyd (Chironomidae: Orthocladiinae): population dynamics, emergence, and drift. New Zealand Journal of Marine and Freshwater Research 33:351-359. Borkent, A. 1984. The systematics and phylogeny of the Stenochironomus complex (Xestochironomus, Harrisius, and Stenochironomus) (Diptera: Chironomidae). Memoirs of the Entomological Society of Canada 128:1-269. Borutzky, E.V. 1939a. Dynamics of the biomass of Chironomus plumosus in the profundal of Lake Beloie. Trudy Limnologicheskoi Stantsii v Kosine 22:156-95. Borutzky, E.V. 1939b. Dynamics of the total benthic biomass in the profundal of Lake Beloie. Trudy Limnologicheskoi Stantsii v Kosine 22:196-218.

355 Borutsky, E.V. 1963. Emergence of Chironomidae (Diptera) imagines from continental waterbodies of different climatic belts as a factor of food supply of fishes. Zoologicheskii zhurnal 42:233-247. Bouchard, R.W., Jr., M.A. Carrillo, & L.C. Ferrington, Jr. 2006a. Lower lethal temperature for adult male Diamesa mendotae Muttkowski (Diptera: Chironomidae), a winter-emerging aquatic insect. Aquatic Insects 28:57-66. Bouchard, R.W., Jr., M.A. Carrillo, S.A. Kells & L.C. Ferrington, Jr. 2006b. Freeze tolerance in larvae of the winter-active Diamesa mendotae Muttkowski (Diptera: Chironomidae): a contrast to adult strategy for survival at low temperatures. Hydrobiologia 568:403-416. Bradshaw, W.E. 1980. Thermoperiodism and the thermal environment of the pitcher- plant mosquito, Wyeomyia smithii. Oecologia 46:13-17. Bradshaw, W.E., C.M. Holzapfel, C.A. Kleckner & J.J. Hard. 1997. Heritability of development time and protandry in the pitcher-plant mosquito, Wyeomyia smithii. Ecology 78:969-976. Britt, N.W. 1962. Biology of two species of Lake Erie mayflies, Ephoron album (Say) and Ephemera simulans Walker. Bulletin of the Ohio Biological Survey New Series 1:1-70. Brittain, J.E. 1982. Biology of mayflies. Annual Review of Entomology 27:119-147. Brittain, J.E. & A.M. Milner. 2001. Ecology of glacier fed rivers: current status and concepts. Freshwater Biology 46:1571-1578. Brodersen, K.P. & N.J. Anderson. 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. & H.J.B. Birks. 2000. Chironomid-inferred late-glacial and early-Holocene mean July air temperatures for Krakenes Lake, western Norway. Journal of Paleolimnology 23:77-89. Brues, C.T. 1932. Further studies on the fauna of North American hot springs. Proceedings of the American Academy of Arts and Sciences 67:185-303.

356 Brundin, L. 1949. Chironomiden und andere Bodentiere de sudschwedischen Urgebirgseen. Ein Beitrag zur Kenntnis der bodenfaunistischen Charakterzuge schwedischer oligotropher Seen. Report of the Institute of Freshwater Research, Drottningholm 30:1-914. Brundin, L. 1951. The relation of O2-microstratification at the mud surface to the ecology of the profundal bottom fauna. Report of the Institute of Freshwater Research, Drottningholm 32:32-42. Brundin, L. 1966. Transantarctic 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 Handlingar. 11:1-472 pp + 30 plates. Buckley, B.R. & J.E. Sublette. 1964. Chironomidae (Diptera) of Louisiana II. The limnology of the upper part of Cane River Lake, Natchitoches Parish, Louisiana, with particular reference to the emergence of Chironomidae. Tulane Studies in Zoology 11:151-166. Burgheer, P. & J.V. Ward. 2001. Longitudinal and seasonal distribution patterns of the benthic fauna of an alpine glacial stream (Val Roseg, Swiss Alps). Freshwater Biology 46:1705-1721. Buskens, R.F.M. 1989. Monitoring of chironomid larvae and exuviae in the Beuven, a soft water pool in the Netherlands, and comparisons with palaeolimnological data. Acta Biologica Debrecen Oecologica Hungarica 3:41-50. Butler, M.G. 1980. Emergence phenologies of some Arctic Alaskan Chironomidae, p. 307-314. In: Chironomidae - ecology, systematics, cytology and physiology. D.A. Murray (ed.). Pergamon Press, Oxford, UK. Butler, M.G. 1982. A 7-year life cycle for two Chironomus species in arctic Alaskan tundra ponds (Diptera: Chironomidae). Canadian Journal of Zoology 60:58-70. Butler, M.G. 1984. Life histories of aquatic insects, p. 24-55. In: Ecology of aquatic insects. V.H. Resh & D. Rosenberg (eds.). Praeger Scientific Publishers, New York, NY.

357 Butler, M.G. & D.H. Anderson. 1990. Cohort structure, biomass, and production of a merovoltine Chironomus population in a Wisconsin bog lake. Journal of the North American Benthological Society 9:180-192. Calle-Martínez, D. & J.J. Casas. 2006. Chironomid species, stream classification, and water-quality assessment: the case of 2 Iberian Mediterranean mountain regions. Journal of the North American Benthological Society 25:465-476. Cannings, R.A. & G.G.E. Scudder. 1978. Phenology of Chironomus spp. in saline lakes of central British Columbia. Verhandlungen der Internationale Vereinigung für Theoretische und Angewandte Limnologie 20:2641-2646. Cantrell, M.A. & A.J. McLachlan. 1977. Competition and chironomid distribution patterns in a newly flooded lake. Oikos 29:429-433. Cao, Y., D.P. Larsen, & R.St-J. Thorne. 2001. Rare species in multivariate analysis for bioassessment: some considerations. Journal of the North American Benthological Society 20:144-153. Carrillo, M.A., C.A. Cannon, & L.C. Ferrington Jr. 2004a. Effect of sex and age on the supercooling point of the winter-active Diamesa mendotae Muttkowski (Diptera: Chironomidae). Aquatic Insects 26:243−251. Carrillo M.A., N. Kaliyan, C.A. Cannon, R.V. Morey, & W.F. Wilcke. 2004b. A simple method to adjust cooling rates for supercooling point determination. CryoLetters 25:155-160. Carrillo, M.A., C.A. Cannon, W.F. Wilcke, R.V. Morey, N. Kaliyan, & W.D. Hutchison. 2005a. Relationship between supercooling point and mortality at low temperatures in Indianmeal moth (Lepidoptera: Pyralidae). Journal of Economic Entomology 98:618−625. Carrillo, M.A., G.E. Heimpel, R.D. Moon, C.A. Cannon, & W.D. Hutchison. 2005b. Cold hardiness of Habrobracon hebetor (Say) (Hymenoptera: Braconidae), a parasitoid of pyralid moths. Journal of Insect Physiology 51:759-768. Carter, C.E. 1980. The life cycle of Chironomus anthracinus in Lough Neagh. Holarctic Ecology 3:214-217.

358 Casas, J.J. & A. Vilchez-Quero. 1989. A faunistic study of the lotic chironomids (Diptera) of the Sierra Nevada (S.E. of Spain): changes in the structure and composition of the populations between spring and summer. Acta Biologica Debrecen Oecologica Hungarica 3:83-93. Casas, J.J. & A. Vilchez-Quero. 1993. Altitudinal distribution of lotic chironomid (Diptera) communities in the Sierra Nevada Mountains (Southern Spain). Annales de Limnologie 29:175-187. Cather, M.R. & A.R. Gaufin. 1976. Comparative ecology of three Zapada species of Mill Creek, Wasatch Mountains, Utah (Plecoptera: Nemouridae). American Midland Naturalist 95:464-471. Chou, R.Y.M., L.C. Ferrington, Jr., B.L. Hayford & H.M. Smith. 1999. Composition and phenology of Chironomidae (Diptera) from an intermittent stream in Kansas. Archiv für Hydrobiologie 147:35-64. Clement, S.L., A.A. Grigarick & M.O. Way. 1977. The colonization of California rice patties by chironomid midges. Journal of Applied Ecology 14:379-389. Cloutier, L. & P.P. Harper. 1978. Phénologie de Tanypodinae de ruisseaux des Laurentides (Diptera: Chironomidae). Canadian Journal of Zoology 56:1129- 1139. Coffman, W.P. 1973. Energy flow in a woodland stream ecosystem. II. The taxonomic composition and phenology of the Chironomidae as determined by the collection of pupal exuviae. Archiv für Hydrobiologie 71:281-322. Coffman, W.P. 1974. Seasonal differences in the diel emergence of a lotic chironomid community. Entomologisk Tidskrift Supplement 95:42-48. Coffman, W.P. 1989. Factors that determine the species richness of lotic communities of Chironomidae. Acta Biologica Debrecen Oecologica Hungarica 3:95-100. Coffman, W.P., C. de la Rosa, K.W. Cummins & M.A. Wilzbach. 1992. Species richness in some Neotropical (Costa Rica) and Afrotropical (West Africa) lotic communities of Chironomidae (Diptera). Netherlands Journal of Aquatic Ecology 26:229-237.

359 Coffman, W.P. & C.L. de la Rosa. 1998. Taxonomic composition and temporal organization of tropical and temperate species assemblages of lotic Chironomidae. Journal of the Kansas Entomological Society 71:388-406. Coffman, W.P. & L.C. Ferrington, Jr. 1996. Chironomidae, p. 635-754. In: An Introduction to the Aquatic Insects of North America. R.W. Merritt & K.W. Cummins (eds.). Kendall/Hunt Publishing Co., Dubuque, IA. Coffman, W.P. & S.S. Roback. 1984. Lopescladius (Cordiella) hyporheicus. A new subgenus and species (Diptera: Chironomidae: Orthocladiinae). Proceedings of the Academy of Natural Sciences of Philadelphia 136:130-144. Coleman, M.J. & H.B.N. Hynes. 1970. The life histories of some Plecoptera and Ephemeroptera in a southern Ontario stream. Canadian Journal of Zoology 48:1333-1339. Coler, B.G. & B.C. Kondratieff. 1989. Emergence of Chironomidae (Diptera) from a delta-swamp receiving thermal effluent. Hydrobiologia 174:67-77. Connell, J.H. 1978. Diversity in tropical rain forests and coral reefs. Science 199:1302- 1310. Connell, J.H. & E. Orias. 1964. The ecological regulation of species diversity. American Naturalist 98:399-414. Corbet, P.S. 1958. Lunar periodicity of aquatic insects in Lake Victoria. Nature 182:330- 331. Corbet, P.S. 1964. Temporal patterns of emergence in aquatic insects. Canadian Entomologist 96:264-279. Cowell, B.C. & D.S. Vodopich. 1981. Distribution and seasonal abundance of benthic macroinvertebrates in a subtropical Florida lake. Hydrobiologia 78:97-105. Cranston, P.S. 1995. Chap 1: Introduction to the Chironomidae, p. 1-7. In: The Chironomidae: Biology and ecology of non-biting midges. P.D. Armitage, P.S. Cranston, and L.C.V. Pinder (eds.). Chapman and Hall. Cranston, P.S. 1997. Revision of Australian Rheotanytarsus Thienemann & Bause (Diptera: Chironomidae) with emphasis on immature stages. Invertebrate Taxonomy 11:705-734.

360 Cranston, P.S., P.D. Copper, R.A. Hardwick, C.L. Humphrey, & P.L. Dostine. 1997. Tropical acid streams - the chironomid (Diptera) response in northern Australia. Freshwater Biology 37:473-483. Cranston, P.S. & U. Nolte. 1996. Fissimentum, a new genus of drought-tolerant Chironomini (Diptera: Chironomidae) from the Americas and Australia. Entomological News 107:1-15. Crisp, D.T. & G. Howson. 1982. Effect of air temperature upon mean water temperature on streams in the north Pennines and English Lake District. Freshwater Biology 12:359-367. Crisp, G. & L. Lloyd. 1954. The community of insects in a patch of woodland mud. Transactions of the Royal Entomological Society of London 105:269-313. Crosa, G. & A. Buffagni. 2002. Spatial and temporal niche overlap of two mayfly species (Ephemeroptera): the role of substratum roughness and body size. Hydrobiologia 474:107-115. Crowley, P.H. & D.M. Johnson. 1982. Habitat and seasonality as niche axes in an odonate community. Ecology 63:1064-1077. Cummins, K.W. 1964. Factors limiting the microdistribution of larvae of the caddisflies Pycnopsyche lepida (Hagen) and Pycnopsyche guttifer (Walker) in a Michigan stream (Trichoptera: Limnephilidae). Ecological Monographs 34:271-295. Curry, L.L. 1958. Larvae and pupae of the species of Cryptochironomus (Diptera) in Michigan. Limnology and Oceanography 3:427-442. Danell, K. 1981. Overwintering of invertebrates in a shallow northern Swedish lake. Internationale Revue der Gesamten Hydrobiologie 66:837-845. Daniel, P.M. 1985. A comparison of clear and opaque funnel traps for emerging insects in a southwestern Ohio pond. Ohio Journal of Science 85:199-202. Danks, H.V. 1971a. Life history and biology of Einfeldia synchrona (Diptera: Chironomidae). Canadian Entomologist 103:1597-1606. Danks, H.V. 1971b. Overwintering of some north temperate and arctic Chironomidae I. The winter environment. The Canadian Entomologist 103:589-604.

361 Danks, H.V. 1978. Some effects of photoperiod, temperature, and food on emergence in three species of Chironomidae (Diptera). Canadian Entomologist 110:289-300. Danks, H.V. 1981. Arctic arthropods: A review of systematics and ecology with particular reference to the North American fauna. Entomological Society of Canada 608 p. Danks, H.V. & J.W. Jones. 1978. Further observations on winter cocoons in Chironomidae (Diptera). Canadian Entomologist 110:667-669. Danks, H.V. & D.R. Oliver. 1972. Seasonal emergence of some high arctic Chironomidae (Diptera). Canadian Entomologist 104:661-686. Darby, R.E. 1962. Midges associated with California rice fields, with special reference to their ecology (Diptera: Chironomidae). Hilgardia 32:1-206. Davies, I.J. 1984. Sampling aquatic insect emergence, p. 161-227. In: A Manual on methods for the assessment of secondary productivity in fresh waters. Vol. 501. J.A. Downing and F.H. Rigler (eds.). Blackwell Scientific Publications, London, UK. de la Rosa, C. 1985. Resource utilization by Chironomidae (Diptera) in a woodland stream ecosystem, Linesville Creek, Crawford County, Pennsylvania. Ph.D. Dissertation. University of Pittsburg, Pittsburg, PA. de la Rosa, C.L. 1992. Phoretic associations of Chironomidae (Diptera) on Corydalidae (Megaloptera) in northwestern Costa Rican streams. Journal of the North American Benthological Society 11:316-323. Dejoux, C. 1971. Recherches sur le cycle de développement de Chironomus pulcher (Diptera: Chironomidae). Canadian Entomologist 103:465-476. Dewey, S.L. 1986. Effects of the herbicide atrazine on aquatic insect community structure and emergence. Ecology 67:148-162. Doberstein, C.P., J.R. Karr, & L.L. Conquest 2000. The effect of fixed-count subsampling on macroinvertebrate biomonitoring in small streams. Freshwater Biology 44:355-371. Drake, C.M. 1982. Seasonal dynamics of Chironomidae (Diptera) on the Bulrush Schoenoplectus lacustris in a chalk stream. Freshwater Biology 12:225-240.

362 Drake, P. & A.M. Arias. 1995. Distribution and production of Chironomus salinarius (Diptera: Chironomidae) in a shallow coastal lagoon in the Bay of Cádiz. Hydrobiologia 299:195-206. Edgar, W.D. & P.S. Meadows. 1969. Case construction, movement, spatial distribution and substrate selection in the larva of Chironomus riparius Meigen. Journal of Experimental Biology 50:247-253. Edward, D.H.D. 1963. The biology of a parthenogenetic species of Lundstroemia (Diptera: Chironomidae) with descriptions of the immature stages. Proceedings of the Royal Entomological Society of London Series A 38:165-170. Edward, D.H. 1968. Chironomidae in temporary freshwaters. Australian Society for Limnology Newsletter 6:3-5. Ekrem, T., M.F. Sublette & J.E. Sublette. 2003. North American Tanytarsus I. Descriptions and Keys to Species in the eminulus, gregarius, lugens and mendax Species Groups (Diptera: Chironomidae). Annals of the Entomological Society of America 96:265-328. Elliott, J.M. 1987. Egg hatching and resource partitioning in stoneflies: the six British Leuctra spp. (Plecoptera: Leuctridae). Journal of Animal Ecology 56:415-426. Englemann, W. & D.G. Shappirio. 1965. Photoperiod control of the maintenance and termination of larval diapause in Chironomus tentans. Nature 207:548-549. Epler, J.H. 1987. Revision of the Nearctic Dicrotendipes Kieffer, 1913 (Diptera: Chironomidae). Evolutionary Monographs 9:1-102 + 241 figures. Epler, J.H. 1988. Biosystematics of the genus Dicrotendipes Kieffer, 1913 (Diptera: Chironomidae) of the world, Philadelphia, PA. 214 pp. Epler, J.H. & C.L. de la Rosa. 1995. Tempisquitoneura, a new genus of Neotropical Orthocladiinae (Diptera: Chironomidae) symphoretic on Corydalus (Megaloptera: Corydalidae). Journal of the North American Benthological Society 14:50-60.

Fend, S.V. & J.L. Carter. 1995. The relationship of habitat characteristics to the distribution of Chironomidae (Diptera) as measured by pupal exuviae collections in a large river system. Journal of Freshwater Ecology 10:343-359.

363 Ferrington, L.C. Jr. 2000. Hibernal emergence patterns of Chironomidae in lotic habitats of Kansas versus ambient air and water temperatures. In: Hoffrichter O, editor. Late 20th Century Research on Chironomidae. Aachen, Germany: Shaker Verlag. p. 375−382 Ferrington, L.C., Jr. 2007. Hibernal emergence patterns of Chironomidae in lotic habitats of Kansas versus substrate composition, p. 99-105. In: Contributions to the systematics and ecology of aquatic Diptera - A tribute to Ole A. Sæther. T. Andersen (ed.). Caddis Press, Columbus, OH. Ferrington, L.C., Jr., M.A. Blackwood, C.A. Wright, N.H. Crisp, J.L. Kavanaugh & F.J. Schmidt. 1991. A protocol for using surface-floating pupal exuviae of Chironomidae for rapid bioassessment of changing water quality, p. 181-190. In: Sediment and stream water quality in a changing environment: trends and explanations. Vol. 203. N.E. Peters & D.E. Walling (eds.). IAHS Press, Oxfordshire, UK. Ferrington, L.C., Jr., K.M. Buzby & E.C. Masteller. 1993. Composition and temporal abundance of Chironomidae emergence from a tropical rainforest stream at El Verde, Puerto Rico. Journal of the Kansas Entomological Society 66:167-180. Ferrington, L.C., Jr., W.P. Coffman, & M.B. Berg. in press. Chironomidae. In: An introduction to the aquatic insects of North America. R.W. Merritt, K.W. Cummins, & M.B. Berg (eds.). Kendall/Hunt Publishing Co., Dubuque, IA. Ferrington, L.C., Jr. & N.H. Crisp. 1989. Water chemistry characteristics of receiving streams and the occurrence of Chironomus riparius and other Chironomidae in Kansas. Acta Biologica Debrecen Oecologica Hungarica 3:115-126. Ferrington, L.C. Jr., D. Hansen, B. Karns, & R.W. Bouchard Jr. Submitted. Longevities of Diamesa mendotae Muttkowski, a hibernal emerging species of Diamesinae (Diptera: Chironomidae). Proceedings of the XV International Symposium on Chironomidae. Ferrington, L.C., Jr., R.G. Kavanaugh, F.J. Schmidt & J.L. Kavanaugh. 1995. Habitat separation among Chironomidae (Diptera) in Big Springs. Journal of the Kansas Entomological Society 62:152-165.

364 Fischer, J. 1969. Zur Fortpflanzungsbiologie von Chironomus nuditarsis Str. Revue Suisse de Zoologie 76:23-55. Fischer, J. 1974. Experimentelle Beiträge zur Ökologie von Chironomus (Diptera). I. Dormanz bei Chironomus nuditarsis und Ch. plumosus. Oecologia 16:73-95. Fischer, J. & S. Rosin. 1969. Das larvale Wachstum von Chironomus nuditarsis Str. Revue Suisse de Zoologie 76:727-734. France, R.L. 1991. Empirical methodology for predicting changes in species range extension and richness associated with climate warming. International Journal of Biometeorology 34:211-216. Frank, C. 1987. A comparative study of chironomid (Diptera) emergence data from 14 lakes in the urban region of West Berlin. Entomologica Scandinavica Supplementum 29:211-216. Frantzen, N.M.L.H.F. 1992. Water quality changes of the river Meuse assessed by chironomid pupal exuviae. Netherlands Journal of Aquatic Ecology 26:543-549. Freeman, P. 1955. A study of the Chironomidae (Diptera) of Africa south of the Sahara, Part I. Bulletin of the British Museum (Natural History), Entomology 4:1-67. Frisbie, M.P. & R.E. Lee Jr. 1997. Inoculative freezing and the problem of winter survival for freshwater macroinvertebrates. Journal of the North American Benthological Society 16:635-650. Frommer, S.I. & J.E. Sublette. 1971. The Chironomidae (Diptera) of the Philip L. Boyd Deep Canyon Desert Research Center, Riverside Co., California. Canadian Entomologist 103:414-423. Frouz, J., A. Ali & R.J. Lobinske. 2002. Influence of temperature on developmental rate, wing length, and larval head capsule size of pestiferous midge Chironomus crassicaudatus (Diptera: Chironomidae). Journal of Economical Entomology 95:699-705. Frouz, J., A. Ali & R.J. Lobinske. 2004. Laboratory evaluation of six algal species for larval nutritional suitability of the pestiferous midge Glyptotendipes paripes (Diptera: Chironomidae). Journal of Economic Entomology 97:1884-1890.

365 Fryer, G. 1959. Lunar rhythm of emergence, differential behaviour of the sexes, and other phenomena in the African midge, Chironomus brevibucca (Kieff.). Bulletin of Entomological Research 50:1-8. Gendron, J.M. & H. Laville. 1992. Diel emergence patterns of drifting chironomid (Diptera) pupal exuviae in the Aude River (Eastern Pyrenees, France). Netherlands Journal of Aquatic Ecology 26:273-279. Gendron, J. & H. Laville. 1995. Biodiversity and sampling frequency of the pupal exuviae of Chironomidae (Diptera) in a 4th order river. Archiv für Hydrobiologie 135:243-257. Georgian, T. & J.B. Wallace. 1983. Seasonal production dynamics in a guild of periphyton-grazing insects in a southern Appalachian stream. Ecology 64:1236- 1248. Gerstmeier, R. 1989. Phenology and bathymetric distribution of the profundal chironomid fauna in Starnberger See (F.R. Germany). Hydrobiologia 184:29-42. Gislason, G.M., T. Hrafnsdottir & A. Gardarsson. 1995. Flight periods of midges (Chironomidae and Simuliidae) in the River Laxa, N-Iceland, p. 177-184. In: Chironomids: from genes to ecosystems. Vol. 496. P.S. Cranston (ed.). CSIRO Publications, East Melbourne, Australia. Goddeeris, B.R. 1987. The time factor in the niche space of Tanytarsus-species in two ponds in the Belgian Ardennes (Diptera: Chironomidae). Entomologica Scandinavica Supplementum 29:281-288. Goddeeris, B.J. 1990. Life cycle characteristics in Tanytarsus sylvaticus (van der Wulp, 1859) (Chironomidae, Diptera). Annales de Limnologie 26:51-64. Goddeeris, B.R. 1991. Life cycle characteristics in Tanytarsus debilis (Meigen, 1830) (Diptera, Chironomidae). Annales de Limnologie 27:141-156. Goddeeris, B.J., A.C. Vermeulen, E.D. Geest, H. Jacobs, B. Baert & F. Ollevier. 2001. Diapause induction in the third and fourth instar of Chironomus riparius (Diptera) from Belgian lowland brooks. Archiv für Hydrobiologie 150:307-327.

366 Gower, A.M. & P.J. Buckland. 1978. Water quality and the occurrence of Chironomus riparius Meigen (Diptera: Chironomidae) in a stream receiving sewage effluent. Freshwater Biology 8:153-164. Grant, P.R. & R.J. Mackay. 1969. Ecological segregation of systematically related stream insects. Canadian Journal of Zoology 47:691-694. Gray, L.J. 1981. Species composition and life histories of aquatic insects in a lowland Sonoran Desert stream. American Midland Naturalist 106:229-242. Gray, L.J., J.V. Ward, R. Martinson & E. Bergey. 1983. Aquatic macroinvertebrates of the Piceance Basin, Colorado: community response along spatial and temporal gradients of environmental conditions. Southwestern Naturalist 28:125-135. Gresens, S.E. 1997. Interactive effects of diet and thermal regime on growth of the midge Pseudochironomus richardsoni Malloch. Freshwater Biology 38:365-373. Grodhaus, G. 1980. Aestivating chironomid larvae associated with vernal pools, p. 315- 322. In: Chironomidae - ecology, systematics, cytology and physiology. D.A. Murray (ed.). Pergamon Press, Oxford, UK. Grodhaus, G. 1987. Endochironomus Kieffer, Tribelos Townes, Synendotendipes, n. gen., and Endotribelos, n. gen. (Diptera: Chironomidae) of the Nearctic Region. Journal of the Kansas Entomological Society 60:167-247. Groenendijk, D., J.F. Postma, M.H.S. Kraak & W. Admiraal. 1998. Seasonal dynamics and larval drift of Chironomus riparius (Diptera) in a metal contaminated lowland river. Aquatic Ecology 32:341-351. Hågvar, S. & E. Østbye. 1973. Notes on some winter-active Chironomidae. Norsk Entomologisk Tidsskrift 20:253-257. Hall, R.E. 1960. The Chironomidae of three chalk streams in southern England. Proceedings of the International Congress of Entomology 11:178-181. Hansen, D.C. & E.F. Cook. 1976. The systematics and morphology of the Nearctic species of Diamesa Meigen, 1835 (Diptera: Chironomidae). Memoirs of the American Entomological Society 30:1−203. Hardwick, R.A., P.D. Cooper, P.S. Cranston, C.L. Humphrey & P.L. Dostine. 1995. Spatial and temporal distribution patterns of drifting pupal exuviae of

367 Chironomidae (Diptera) in streams of tropical northern Australia. Freshwater Biology 34:569-578. Haro, R.J. & M.J. Wiley. 1992. Secondary consumers and the thermal equilibrium hypothesis: insights from Michigan spring brooks, p. 179-188. In: Proceedings of the First International Conference on Ground Water Ecology. Vol. 419. J.A. Stanford & J.J. Simons (eds.). American Water Resources Association, Bethesda, MD. Harper, P.P. & L. Cloutier. 1979. Chironomini and Pseudochironomini of a Québec highland stream (Diptera: Chironomidae). Entomologica Scandinavica Supplement 10:81-94. Harper, P.P. & J. Pilon. 1970. Annual patterns of emergence of some Quebec stoneflies (Insecta: Plecoptera). Canadian Journal of Zoology 48:681-694. Harrisson, P.M. & W. Block.1988. Survival of freezing in a midge introduced to the Antarctic. CryoLetters 9:433. (abstract only) Hart, D.D. 1983. The importance of competitive interactions within stream populations and communities, p. 99-136. In: Stream ecology. J.R. Barnes & G.W. Minshall (eds.). Plenum Publishing Corporation. Hashimoto, H. 1957. Peculiar mode of emergence in the marine chironomid Clunio (Diptera, Chironomidae). Science reports of the Tokyo Kyoiku Daigaku Section B 8:177-226. Hashimoto, H. 1962. Ecological significance of the sexual dimorphism in marine chironomids. Science reports of the Tokyo Kyoiku Daigaku Section B 10:15-46. Hashimoto, H. 1965. Discovery of Clunio takahashii Tokunaga from Japan. Japanese Journal of Zoology 14:13-29. Hatakeyama, S. 1987. Chronic effects of Cd on reproduction of Polypedilum nubifer (Chironomidae) through water and food. Environmental Pollution 48:249-261. Hatakeyama, S. 1988. Chronic effects of Cu on reproduction of Polypedilum nubifer (Chironomidae) through water and food. Ecotoxicology and Environmental Safety 16:1-10.

368 Hauer, F.R. & A.C. Benke. 1991. Rapid growth of snag-dwelling chironomids in a blackwater river: the influence of temperature and discharge. Journal of the North American Benthological Society 10:154-164. Hawkins, C.P., R.H. Norris, J.N. Hogue & J.W. Feminella. 2000. Development and evaluation of predictive models for measuring the biological integrity of streams. Ecological Applications 10:1456-1477. Hayes, B.P. & D.A. Murray. 1987. Species composition and emergence of Chironomidae (Diptera) from three high arctic streams on Bathurst Island, Northwest Territories, Canada. Entomologica Scandinavica Supplementum 29:355-360. Hayes, B.P. & D.A. Murray. 1988. IX. Running waters (continued) diel variation in chironomid emergence and implications for the use of pupal exuviae in river classification. Verhandlungen der Internationale Vereinigung für Theoretische und Angewandte Limnologie 23:1261-1266. Hayford, B.L. 1998. A systematic revision of Paratendipes (Diptera: Chironomidae) with special emphasis on the evolution of thermophily. Unpublished M.S. thesis. 274 pp. Hayford, B., J. Bachmann & M. Gotov. 2006. A comparison between communities of Chironomidae (Insecta: Diptera) from lake bays and affluent streams of the Lake Hovsgol watershed, Mongolia. Proceedings of the Academy of Natural Sciences of Philadelphia 155:13-23. Hayford, B.L. & L.C. Ferrington, Jr. 2005. Biological assessment of Cannon Creek, Missouri by use of emerging Chironomidae (Insecta: Diptera). Journal of the Kansas Entomological Society 78:89-99. Hayford, B.L., J.E. Sublette & S.J. Herrmann. 1995. Distribution of chironomids (Diptera: Chironomidae) and Ceratopogonids (Diptera: Ceratopogonidae) along a Colorado thermal spring effluent. Journal of the Kansas Entomological Society 68:77-92. Heimbach, F. 1978. Sympatric species, Clunio marinus Hal. and Cl. balticus n. sp. (Dipt., Chironomidae), isolated by differences in diel emergence time. Oecologia 32:195- 202.

369 Heinis, F. & T. Crommentuijn. 1989. The natural habitat of the deposit feeding chironomid larvae Stictochironomus histrio (Fabricius) and Chironomus anthracinus Zett. in relation to their responses to changing oxygen concentrations. Acta Biologica Debrecen Oecologica Hungarica 3:135-140. Heinis, F., W.J. van de Bund & C. Davids. 1989. Avoidance of low oxygen and food concentrations by the larvae of Tanytarsus sp. Acta Biologica Debrecen Oecologica Hungarica 3:141-145. Henriques-Oliveira, A.L., J.L. Nessimian & L.F.M. Dorvillé. 2003. Feeding habits of chironomid larvae (Insecta: Diptera) from a stream in the Floresta da Tijuca, Rio de Janeiro, Brazil. Brazilian Journal of Biology 63:269-281. Herrmann, S.J., J.E. Sublette & M. Sublette. 1987. Midwinter emergence of Diamesa leona Roback in the Upper Arkansas River, Colorado, with notes on other diamesines (Diptera: Chironomidae). Entomologica Scandinavica Supplement 29:309-322. Hershey, A.E. 1985. Littoral chironomid communities in an arctic Alaskan lake. Holarctic Ecology 8:39-48. Hestenes, T.C. & O.A. Sæther. 2000. Three new Nearctic Thienemanniella Kieffer species with a review of the Nearctic species, p. 103-127. In: Late 20th century research on Chironomidae: an anthology from the 13th International Symposium on Chironomidae, Freiburg, 5-9 September 1997. O. Hoffrichter (ed.). Shaker Verlag, Aachen. Heuschele, A.S. 1969. Invertebrate life cycle patterns in the benthos of a floodplain lake in Minnesota. Ecology 50:998-1011. Hildrew, A.G. & J.M. Edington. 1979. Factors facilitating the coexistence of hydropsychid caddis larvae (Trichoptera) in the same river system. Journal of Animal Ecology 48:557-576. Hildrew, A.G., C.R. Townsend & J. Francis. 1984. Cellulolytic decomposition in streams of contrasting pH and its relationship with invertebrate community structure. Freshwater Biology 14:323-328.

370 Hilsenhoff, W.L. 1966. The biology of Chironomus plumosus (Diptera: Chironomidae) in Lake Winnebago, Wisconsin. Annals of the Entomological Society of America 59:465-473. Hinton, H.E. 1951. A new chironomid from Africa, the larva of which can be dehydrated without injury. Proceedings of the Zoological Society of London 121:371-380. Hinton, H.E. 1960a. Cryptobiosis in the larva of Polypedilum vanderplanki Hint. (Chironomidae). Journal of Insect Physiology 5:286-300. Hinton, H.E. 1960b. A fly larva that tolerates dehydration and temperatures of -270 °C to +102 °C. Nature 188:336-337. Hintze, J. 2001. NCSS and PASS. Kaysville, UT: Number Cruncher Statistical Systems. Hirthe, G., T.C. Fisher, M. Crane & A. Callaghan. 2001. Short-term exposure to sublethal doses of lindane affects developmental parameters in Chironomus riparius Meigen, but has no effect on larval glutathione-S-transferase activity. Chemosphere 44:583-589. Hogg, I.D., D.D. Williams, J.M. Eadie & S.A. Butt. 1995. The consequences of global warming for stream invertebrates: a field simulation. Journal of Thermal Biology 20:199-206. Hogg, I.D. & D.D. Williams. 1996. Response of stream invertebrates to a global- warming thermal regime: an ecosystem-level manipulation. Ecology 77:395-407. Hooper, H.L., R.M. Sibly, T.H. Hutchinson & S.J. Maund. 2003a. The influence of larval density, food availability and habitat longevity on the life history and population growth rate of the midge Chironomus riparius. Oikos 102:515-524. Hooper, H.L., R.M. Sibly, S.J. Maund & T.H. Hutchinson. 2003b. The joint effects of larval density and 14C-cypermethrin on the life history and population growth rate of the midge Chironomus riparius. Journal of Applied Ecology 40:1049-1059. Hooper, H.L., R.M. Sibly, T.H. Hutchinson & S.J. Maund. 2005. Joint effects of density and a growth inhibitor on the life history and population growth rate of the midge Chironomus riparius. Environmental Toxicology and Chemistry 24:1140-1145. Howe, R.W. 1967. Temperature effects on embryonic development in insects. Annual Review of Entomology 12:15-42.

371 Hudson, P.L. & J.V. Adams. 1998. Sieve efficiency in benthic sampling as related to chironomid head capsule width. Journal of the Kansas Entomological Society 71:456-468. Humphries, C.F. 1938. The chironomid fauna of the Großer Plöner See, the relative density of its members and their emergence period. Archiv für Hydrobiologie 33:535-584. Huryn, A.D. 1990. Growth and voltinism of lotic midge larvae: patterns across an Appalachian Mountain basin. Limnology and Oceanography 32:339-351. Huryn, A.D. & J.B. Wallace. 1986. A method for obtaining in situ growth rates of larval Chironomidae (Diptera) and its application to studies on secondary production. Limnology and Oceanography 31:216-222. Huryn, A.D. & J.B. Wallace. 2000. Life history and production of stream insects. Annual Review of Entomology 45:83-110. Huston, M. 1979. A general hypothesis of species diversity. American Naturalist 113:81- 101. Hynes, H.B.N. 1970. The ecology of stream insects. Annual Review of Entomology 15:25-42. Ide, F.P. 1935. The effect of temperature on the distribution of the mayfly fauna of a stream. University of Toronto Studies Biological Series 39:1-76. Ikeshoji, T. 1974. Isolation and identification of the overcrowding factors of chironomid larvae. Japanese Journal of Sanitary Zoology 24:201-206. Ilyashuk, B.P. & E.A. Ilyashuk. 2007. Chironomid record of Late Quaternary climatic and environmental changes from two sites in Central Asia (Tuva Republic, Russia) - local, regional or global causes. Quaternary Science Reviews 26:705- 731. Ineichen, H., B. Meyer & M. Lezzi. 1983. Determination of the developmental stage of living fourth instar larvae of Chironomus tentans. Developmental Biology 98:278-286. Ineichen, H., U. Riesen-Willi & J. Fischer. 1979. Experimental contributions to the ecology of Chironomus (Diptera) II: the influence of the photoperiod on the

372 development of Chironomus plumosus in the 4th larval instar. Oecologia 39:161- 183. Ingvason, H.R., J.S. Olafsson & A. Gardarsson. 2002. Temporal pattern in resource utilization of Tanytarsus gracilentus larvae. Verhandlungen der Internationale Vereinigung für Theoretische und Angewandte Limnologie 28:1041-1045. Ingvason, H.R., J.S. Olafsson, A. Gardarsson & R. Jonsdottir. 2006. Diapause and fat condition of Tanytarsus gracilentus larvae in a sub-arctic lake. Verhandlungen der Internationale Vereinigung für Theoretische und Angewandte Limnologie 29:1316-1320. Iovino, A.J. & F.D. Miner. 1970. Seasonal abundance and emergence of Chironomidae of Beaver Reservoir, Arkansas. Journal of the Kansas Entomological Society 43:197-216. Irons, J.G., III, L.K. Miller, & M.W. Oswood. 1993. Ecological adaptations of aquatic macroinvertebrates to overwintering in interior Alaska (U.S.A.) subarctic streams. Canadian Journal of Zoology 71:98-108. Istock, C.A. 1966. Distribution, coexistence, and competition of whirligig beetles. Evolution 20:211-234. Istock, C.A. 1973. Population characteristics of a species ensemble of water boatmen (Corixidae). Ecology 54:535-544. Iversen, T.M. 1978. Life cycle and growth of three species of Plecoptera in a Danish spring. Entomologiske Meddelelser 46:57-62. Iwakuma, T. 1992. Emergence of Chironomidae from the shallow eutrophic Lake Kasumigaura, Japan. Hydrobiologia 245:27-40. Iwakuma, T. & M. Yasuno. 1981. Chironomid populations in highly eutrophic Lake Kasumigaura. Verhandlungen der Internationale Vereinigung für Theoretische und Angewandte Limnologie 21:664-674. Iwakuma, T. & M. Yasuno. 1983. Fate of the univoltine chironomid, Tokunagayusurika akamusi (Diptera: Chironomidae), at emergence in Lake Kasumigaura, Japan. Archiv für Hydrobiologie 99:37-59.

373 Iwakuma, T., M. Yasuno & Y. Sugaya. 1984. Chironomid production in relation to phytoplankton primary production in Lake Kasumigaura, Japan. Verhandlungen der Internationale Vereinigung für Theoretische und Angewandte Limnologie 22:1150-1159. Jackson, G.A. 1977. Nearctic and Palaearctic Paracladopelma Harnisch and Saetheria n. gen. (Diptera: Chironomidae). Journal of the Fisheries Research Board of Canada 34:1321-1359. Jackson, J.K. & B.W. Sweeney. 1995. Egg and larval development times for 35 species of tropical stream insects from Costa Rica. Journal of the North American Benthological Society 14:115-130. Jacobsen, R.E. 1998. The symbiotic relationship of a chironomid with its ephemeropteran hosts in an Arizona mountain stream. Journal of the Kansas Entomological Society 71:426-438. Jacobsen, D.J., R. Schultz & A. Encalada. 1997. Structure and diversity of stream invertebrate assemblages: the influence of temperature with altitude and latitude. Freshwater Biology 38:247-261. Janković, M. 1974. Feeding and food assimilation in larvae of Prodiamesa olivacea. Entomologisk Tidskrift Supplement 95:116-119. Johannsson, O.E. 1978. Co-existence of larval Zygoptera (Odonata) common to the Norfolk Broads (U.K.). Oecologia 32:303-321. Johannsson, O.E. 1980. Energy dynamics of the eutrophic chironomid Chironomus plumosus f. semireductus from the Bay of Quinte, Lake Ontario. Canadian Journal of Fisheries and Aquatic Sciences 37:1254-1265. Johnson, R.K. & B. Pejler. 1987. Life histories and coexistence of the two profundal Chironomus species in Lake Erken, Sweden. Entomologica Scandinavica Supplementum 29:233-238. Jónasson, P.M. 1961. Population dynamics in Chironomus anthracinus Zett. in the profundal zone of Lake Esrom. Verhandlungen der Internationale Vereinigung für Theoretische und Angewandte Limnologie 14:196-203.

374 Jónasson, P.M. 1965. Factors determining population size of Chironomus anthracinus in Lake Esrom. Mitteilungen Internationale Vereinigung fur Theoretische und Angewandte Limnologie 13:139-162. Jónasson, P.M. 1972. Ecology and production of the profundal benthos in relation to phytoplankton in Lake Esrom. Oikos Supplementum 14:1-148. Jónasson, P.M. & J. Kristiansen. 1967. Primary and secondary production in Lake Esrom. growth of Chironomus anthracinus in relation to seasonal cycles of phytoplankton and dissolved oxygen. Internationale Revue der Gesamten Hydrobiologie 52:163- 217. Jones, R.E. 1975. Dehydration in an Australian rockpool chironomid larva, (Paraborniella tonnoiri). Journal of Entomology (Series A) 49:111-119. Jónsson, E. 1985. Population dynamics and production of Chironomidae (Diptera) at 2 m depth in Lake Esrom, Denmark. Archiv für Hydrobiologie Supplementband 70:239-278. Jónsson, E. 1987. Flight periods of aquatic insects at Lake Esrom, Denmark. Archiv für Hydrobiologie 110:259-274. Jónsson, E., A. Gardarsson & G.M. Gíslason. 1986. A new window trap used in the assessment of the flight periods of Chironomidae and Simuliidae. Freshwater Biology 16:711-719. Judd, W.W. 1953. A study of the population of insects emerging as adults from the Dundas Marsh, Hamilton, Ontario, during 1948. American Midland Naturalist 49:801-824. Judd, W.W. 1957. A study of the population of emerging and littoral insects trapped as adults from tributary water of the Thames River at London, Ontario. American Midland Naturalist 58:394-412. Judd, W.W. 1960. A study of the population of insects emerging as adults from South Walker Pond at London Ontario. American Midland Naturalist 63:194-210. Judd, W.W. 1962. A study of the population of insects emerging as adults from Medway Creek at Arva, Ontario. American Midland Naturalist 68:463-473.

375 Judd, W.W. 1964. A study of the population of insects emerging as adults from Saunders Pond at London, Ontario. American Midland Naturalist 71:402-414. Kajak, Z. & K. Dusoge. 1970. Production efficiency of Procladius choreus Mg (Chironomidae, Diptera) and its dependence on the trophic conditions. Polskie Archiwum Hydrobiologii 17:217-224. Kajak, Z. & J. Warda. 1968. Feeding of benthic non-predatory Chironomidae in lakes. Annales Zoologici Fennici 5:57-64. Kangasniemi, B.J. & D.R. Oliver. 1983. Chironomidae (Diptera) associated with Myriophyllum spicatum in Okanagan Valley lakes, British Columbia. Canadian Entomologist 115:1545-1546. Karunakaran, L. 1971. Studies on the fluctuation in numbers of certain Chironomidae (Diptera, Nematocera) taken at a light trap in Singapore. Limnologica 8:173-182. Kaufman, M.G. & R.H. King. 1987. Colonization of wood substrates by the aquatic xylophage Xylotopus par (Diptera: Chironomidae) and a description of its life history. Canadian Journal of Zoology 65:2280-2286. Kavanaugh, J.L., R.W. Bouchard, Jr., & L.C. Ferrington, Jr. (submitted) Efficacy of surface floating pupal exuviae collection for estimating emergence frequencies of abundant Chironomidae (Diptera). Submitted to the Proceedings of the XV International Symposium on Chironomidae. Keiper, J.B. & B.A. Foote. 2000. Biology and Larval Feeding Habits of Coexisting Hydroptilidae (Trichoptera) from a Small Woodland Stream in Northeastern Ohio. Annals of the Entomological Society of America 93:225-234. Keiper, J.B., W.E. Walton & B.A. Foote. 2002. Biology and ecology of higher Diptera from freshwater wetlands. Annual Review of Entomology 47:207-232. Kesler, D.H. 1981. Grazing rate determination of Corynoneura scutellata Winnertz (Chironomidae: Diptera). Hydrobiologia 80:63-66. Kettisch, J. 1937. Zur Kenntnis der Morphologie und Ökologie der Larve von Cricotopus trifasciatus. Konowia 16:193-204. Kimerle, R.A. & N.H. Anderson. 1967. Evaluation of aquatic insect emergence traps. Journal of Economic Entomology 60:1255-1259.

376 Kimerle, R.A. & N.H. Anderson. 1971. Production and bioenergetic role of the midge Glyptotendipes barbipes (Staeger) in a waste stabilization lagoon. Limnology and Oceanography 16:646-659. Koch, R.L., M.A. Carrillo, R.C. Venette, C.A. Cannon, & W.D. Hutchison. 2004. Cold hardiness of the multicolored Asian lady beetle (Coleoptera: Coccinellidae). Environmental Entomology 33:815-822. Kohshima, S. 1984. A novel cold-tolerant insect found in a Himalayan glacier. Nature 310:225-227. Kokkinn, M.J. 1990. Is the rate of embryonic development a predictor of overall development rate in Tanytarsus barbitarsis Freeman (Diptera: Chironomidae). Australian Journal of Marine and Freshwater Research 41:575-579. Kokkinn, M.J. & W.D. Williams. 1988. Adaptations to life in a hypersaline water-body: adaptations at the egg and early embryonic stage of Tanytarsus barbitarsis Freeman (Diptera, Chironomidae). Aquatic Insects 10:205-214. Kondo, S. 1996. Life cycle of Hydrobaenus kondoi Sæther (Chironomidae) at the middle reaches of the Kiso River, Japan. Hydrobiologia 318:79-84. Konstantinov, A.C. 1958a. The influence of temperature on the rate of growth and development of chironomid larvae. Doklady Akademii nauk SSSR 120:1362- 1365. Konstantinov, A.C. 1958b. On the type of growth of Chironomidae larvae. Doklady Akademii Nauk SSSR 120:1151-1154. Kosalwat, P. & A.W. Knight. 1987. Chronic toxicity of copper to a partial life cycle of the midge, Chironomus decorus. Archive of Environmental Contamination and Toxicology 16:283-290. Koskinen, R. 1968a. Seasonal and diel emergence of Chironomus salinarius Kieff. (Dipt., Chironomidae) near Bergen, Western Norway. Annales Zoologici Fennici 5:65-70. Koskinen, R. 1968b. Seasonal emergence of Clunio marinus Haliday (Dipt., Chironomidae) in Western Norway. Annales Zoologici Fennici 5:71-75.

377 Kownacki, A. 1995. The use of chironomid pupal exuviae for ecological characterization of the Upper Vistula (southern Poland). Acta Hydrobiologica 37:41-50. Kureck, A. 1978. Alternative Schlüpfzeiten der Mücke Chironomus thummi und deren ökologische Bedeutung. Verhandlungen der Gesellschaft für Ökologie Kiel 1977:207-210. Kureck, A. 1980. Circadian eclosion rhythm in Chironomus thummi: ecological adjustment to different temperature levels and the role of temperature cycles, p. 73-79. In: Chironomidae - ecology, systematics, cytology and physiology. D.A. Murray (ed.). Pergamon Press, Oxford, UK. Ladle, M., D.A. Cooling, J.S. Welton & J.A.B. Bass. 1985. Studies on Chironomidae in experimental recirculating stream systems. II. The growth, development and production of a spring generation of Orthocladius (Euorthocladius) calvus Pinder. Freshwater Biology 15:243-255. Lamberti, G.A. & V.H. Resh. 1983. Geothermal effects on stream benthos: separate influences of thermal and chemical components on periphyton and macroinvertebrates. Canadian Journal of Fisheries and Aquatic Sciences 40:1995- 2009. Lamberti, G.A. & V.H. Resh. 1985. Distribution of benthic algae and macroinvertebrates along a thermal stream gradient. Hydrobiologia 128:13-21. Langton, P.H. 1995. Chap 8: The pupa and events leading to eclosion, p. 169-193. In: The Chironomidae: Biology and ecology of non-biting midges. P.D. Armitage, P.S. Cranston, & L.C.V. Pinder (eds.). Chapman and Hall, London, UK. Laville, H. 1971a. Recherches sur les Chironomides (Diptera) lacustres du Massif de Néouvielle (Haute-Pyrénées). Première partie. systématique, écologie, phénologie. Annales de Limnologie 7:173-332. Laville, H. 1971b. Recherches sur les Chironomides (Diptera) lacustres du Massif de Néouvielle (Haute-Pyrénées). Deuxiéme partie. Communautés et production. Annales de Limnologie 7:335-414.

378 Laville, H. 1981. Récoltes d'exuvies nymphales de Chironomides (Diptera) dans le Haut- Lot, de la source (1 295 m) au confluent de la Truyère (223 m). Annales de Limnologie 17:255-289. Laville, H. & N. Giani. 1974. Phénologie et cycles biologiques des Chironomides de la zone littorale (0-7 m) du lac de Port-Bielh (Pyrénées centrales). Entomologisk Tidskrift Supplement 95:139-155. Laville, H. & M. Viaud-Chauvet. 1985. Etude comparée de la structure des peuplements de Chironomidés dans cinq rivières du Massif Central: relation entre cette structure et la qualité des eaux. Verhandlungen der Internationale Vereinigung für Theoretische und Angewandte Limnologie 22:2261-2269. Leader, J.P. 1962. Tolerance to freezing of hydrated and partially hydrated larvae of Polypedilum (Chironomidae). Journal of Insect Physiology 8:155−163. Learner, M.A. & D.W.B. Potter. 1974. The seasonal periodicity of emergence of insects from two ponds in Hertfordshire, England, with special reference to the Chironomidae (Diptera: Nematocera). Hydrobiologia 44:495-510. Learner, M.A. & R.W. Edwards. 1966. The distribution of the midge Chironomus riparius in a polluted river system and its environs. International Journal of Air and Water Pollution 10:757-768. Leather, S.R., K.F.A. Walters, & J.S. Bale. 1993 The Ecology of Insect Overwintering. Cambridge University Press, Cambridge. Lee, R.E. Jr. 1989. Insect cold-hardiness: to freeze or not to freeze. BioScience 39:308- 313. Lee, R.E. Jr. 1991. Principles of insect low temperature tolerance. In: RE Lee Jr, DL Denlinger, editors. Insects at low temperature. New York: Chapman and Hall. p. 17−46. Lehmann, J. 1971. The chironomids of the Fulda (systematic, ecological and faunistic investigations). Archiv für Hydrobiologie Supplementband 37:466-555. Lehmann, J. 1972. Revision der europpäischen Arten (Puppen and Imagines) der Gattung Eukiefferiella Thienemann. Beiträge zur Entomologie 22:347-405.

379 Lehmann, J. 1979. Chironomidae (Diptera) aus Fließgewässern Zentralafrikas Tiel I: Kivu-Gebiet, Ostzaire. Spixiana Supplement 3:1-144. Lenat, D.R. 1987. The macroinvertebrate fauna of the Little River, North Carolina: taxa list and seasonal trends. Archiv für Hydrobiologie 110:19-43. Lenat, D.R. & D.R. Folley. 1983. Lotic chironomids of the North Carolina mountains. Memoirs of the American Entomological Society 34:145-164. Lencioni, V. 2004. Survival strategies of freshwater insects in cold environments. Journal of Limnology 63:45-55. LeSage, L. & A.D. Harrison. 1980a. The biology of Cricotopus (Chironomidae: Orthocladiinae) in an algal-enriched stream: part I. Normal biology. Archiv für Hydrobiologie Supplementband 57:375-418. LeSage, L. & A.D. Harrison. 1980b. The biology of Cricotopus (Chironomidae: Orthocladiinae) in an algal-enriched stream: part II. Effects of parasitism. Archiv für Hydrobiologie Supplementband 58:1-25. Lewis, D.J., A.J. Henry & D.N. Grindley. 1954. Daily changes in the numbers of chironomid midges at Khartoum. Proceedings of the Royal Entomological Society of London Series A 29:124-128. Lindegaard, C. 1995. Chironomidae (Diptera) of European cold springs and factors influencing their distribution. Journal of the Kansas Entomological Society 68:108-131. Lindegaard, C. & K.P. Brodersen. 1995. Distribution of Chironomidae (Diptera) in the river continuum, p. 257-271. In: Chironomids: from genes to ecosystems. P.S. Cranston (ed.). CRISO, East Melbourne, Australia. Lindegaard, C. & P.M. Jónasson. 1979. Abundance, population dynamics and production of zoobenthos in Lake Mývatn, Iceland. Oikos 32:202-227. Lindegaard, C. & P. Mæhl. 1992. Abundance, population dynamics and production of Chironomidae (Diptera) in an ultraoligotrophic lake in south Greenland. Netherlands Journal of Aquatic Ecology 26:297-308.

380 Lindegaard, C. & E. Mortensen. 1988. Abundance, life history and production of Chironomidae (Diptera) in a Danish lowland stream. Archiv für Hydrobiologie Supplementband 81:563-587. Lindeman, R.L. 1942. Seasonal distribution of midge larvae in a senescent lake. The American Midland Naturalist 27:428-444. Lloyd, L. 1941. The seasonal rhythm of a fly (Spaniotoma minima) and some theoretical considerations. Transactions of the Royal Society of Tropical Medicine and Hygiene 35:93-104. Lobinske, R.J., A. Ali & J. Frouz. 2002. Laboratory estimation of degree-day developmental requirements of Glyptotendipes paripes (Diptera: Chironomidae). Environmental Entomology 31:608-611. Lotter, A.F., I.R. Walker, S.J. Brooks & W. Hofmann. 1999. An intercontinental comparison of chironomid paleotemperature inference models: Europe vs. North America. Quaternary Science Reviews 18:717-735. Macan, T.T. 1958. The temperature of a small stony stream. Hydrobiologia 12:89-106. Macan, T.T. 1961. Factors that limit the range of freshwater animals. Biological Reviews of the Cambridge Philosophical Society 36:151-198. MacAnova. 2002. MacAnova for Windows, version 4.12. University of Minnesota, School of Statistics. Available online at: http://www.stat.umn.edu/macanova/download.html. Macchiusi, F. & R.L. Baker. 1992. Effects of predators and food availability on activity and growth of Chironomus tentans (Chironomidae: Diptera). Freshwater Biology 28:207-216. MacDonald, W.W. 1956. Observations on the Biology of Chaoborids and Chironomids in Lake Victoria and on the Feeding Habits of the Elephant-Snout Fish' (Mormyrus kannume Forsk.). Journal of Animal Ecology 25:36-53. Macek K.J., K.S. Buxton, S.K. Derr, J.W. Dean & S. Sauter. 1976. Chronic toxicity of lindane to selected aquatic invertebrates and fishes. EPA-60013-76-047. U.S. Environmental Protection Agency, Washington, DC.

381 Mackay, R.J. 1972. Temporal patterns in life history and flight behavior of Pycnopsyche gentilis, P. luculenta, and P. scabripennis (Trichoptera: Limnephilidae). Canadian Entomologist 104:1819-1835. Mackey, A.P. & A.D. Berrie. 1991. The prediction of water temperatures in chalk streams from air temperatures. Hydrobiologia 210:183-189. Mackey, A.P. 1976a. Quantitative studies on the Chironomidae (Diptera) of the rivers Thames and Kennet. I. The Acorus zone. Archiv für Hydrobiologie 78:240-267. Mackey, A.P. 1976b. Quantitative studies on the Chironomidae (Diptera) of the rivers Thames and Kennet. II. The Thames flint zone. Archiv für Hydrobiologie 78:310- 318. Mackey, A.P. 1977a. Growth and development of larval Chironomidae. Oikos 28:270- 275. Mackey, A.P. 1977b. Quantitative studies on the Chironomidae (Diptera) of the rivers Thames and Kennet. III. The Nuphar zone. Archiv für Hydrobiologie 79:62-102. MacRae, I.V. & R.A. Ring. 1993. Life history of Cricotopus myriophylli Oliver (Diptera: Chironomidae) in the Okanagan Valley, British Columbia. Canadian Entomologist 125:979-985. Madder, M.C.A., D.M. Rosenberg & A.P. Wiens. 1977. Larval cocoons in Eukiefferiella claripennis (Diptera: Chironomidae). Canadian Entomologist 109:891-892. Madsen, B.L. 1968. A comparative ecological investigation of two related mayfly nymphs. Hydrobiologia 31:37-49. Maier, K.J., P. Kosalwat & A.W. Knight. 1990. Culture of Chironomus decorus (Diptera: Chironomidae) and the effect of temperature on its life history. Environmental Entomology 19:1681-1688. Maitland, P.S., N.W. Charles, N.C. Morgan, K. East & M.C. Gray. 1972. Preliminary research on the production of Chironomidae in Loch Leven, Scotland, p. 795-812. In: Productivity Problems of Freshwaters. Z. Kajak and A. Hillbricht-Ilkowska (eds.). Polish Scientific Publishers, Kraków, Poland. Mansingh, A. 1971. Physiological classification of dormancies in insects. Canadian Entomologist 103:983-1009.

382 Maschwitz, D.E. & E.F. Cook. 2000. Revision of the Nearctic species of the genus Polypedilum Kieffer (Diptera: Chironomidae) in the subgenera P. (Polypedilum) Kieffer and P. (Uresipedilum) Oyewo and Sæther. Bulletin of the Ohio Biological Survey 12:135. Mason, P.G. 1986. Four new species of the Cryptochironomus fulvus (Johannsen) species complex (Diptera: Chironomidae). Entomologica Scandinavica 16:399-413. Mason, P.G. & D.M. Lehmkuhl. 1983. Effects of the Squaw Rapids hydroelectric development on Saskatchewan River Chironomidae (Diptera). Memoirs of the American Entomological Society 34:187-210. Matěna, J. 1989. Seasonal dynamics of a Chironomus plumosus (L.) (Diptera, Chironomidae) population from a fish pond in Southern Bohemia. Internationale Revue der Gesamten Hydrobiologie 74:599-610. Mattingly, R.L., K.W. Cummins & R.H. King. 1981. The influence of substrate organic content on the growth of a stream chironomid. Hydrobiologia 77:161-165. McElravy, E.P., H. Wolda & V.H. Resh. 1982. Seasonality and annual variability of caddisfly adults (Trichoptera) in a "non-seasonal" tropical environment. Archiv für Hydrobiologie 94:302-317. McKie, B.G., P.S. Cranston & R.G. Pearson. 2004. Gondwanan mesotherms and cosmopolitan eurytherms: effects of temperature on the development and survival of Australian Chironomidae (Diptera) from tropical and temperate populations. Marine and Freshwater Research 55:759-768. McLachlan, A. 1983. Life-history tactics of rain-pool dwellers. Journal of Animal Ecology 52:545-561. McLachlan, A.J., A. Brennan & R.S. Wotton. 1978. Particle size and chironomid (Diptera) food in an upland river. Oikos 31:247-252. McLachlan, A.J. & M.A. Cantrell. 1980. Survival strategies in tropical rain pools. Oecologia 47:344-351. Menzie, C.A. 1981. Production ecology of Cricotopus sylvestris (Fabricius) (Diptera: Chironomidae) in a shallow estuarine cove. Limnology and Oceanography 26:467-481.

383 Merritt, R.W., D.H. Ross & B.V. Peterson. 1978. Larval ecology of some lower Michigan black flies (Diptera: Simuliidae) with keys to the immature stages. Great Lakes Entomologist 11:177-208. Miall, L.C. & A.R. Hammond. 1900. The structure and life-history of the harlequin fly (Chironomus). Clarendon Press, Oxford, UK. 196 pp. Mihuc, T.B. & D.W. Toetz. 1996. Phenology of aquatic macroinvertebrates in an alpine wetland. Hydrobiologia 330:131-136. Miller, L.K. 1982. Cold-hardiness strategies of some adult and immature insects overwintering in interior Alaska. Comparative Biochemistry and Physiology 73:595−604. Miller, P.L. 1970. On the occurrence and some characteristics of Cyrtopus fastuous Bigot (Dipt., Stratiomyidae) and Polypedilum sp. (Dipt., Chironomidae) from temporary habitats in Western Nigeria. Entomologist's Monthly Magazine 105:233-238. Miller, R.B. 1941. A contribution to the ecology of the Chironomidae of Costello Lake, Algonquin Park, Ontario. The University of Toronto Press, Toronto, Canada. 63 pp. Milner, A.M., J.E. Brittain, E. Castella & G.E. Petts. 2001. Trends of macroinvertebrate community structure in glacier-fed rivers in relation to environmental conditions: a synthesis. Freshwater Biology 46:1833-1847. Milner, A.M. & G.E. Petts. 1994. Glacial rivers: physical habitat and ecology. Freshwater Biology 32:295-307. Moore, J.W. 1979. Some factors influencing the distribution, seasonal abundance and feeding of subarctic Chironomidae (Diptera). Archiv für Hydrobiologie 85:302- 325. Moore, J.W. 1980. Factors influencing the composition, structure and density of a population of benthic invertebrates. Archiv für Hydrobiologie 88:202-218. Moore, M.V. & R.E. Lee Jr. 1991. Surviving the big chill: overwintering strategies of aquatic and terrestrial insects. American Entomologist 37:111-118. Morgan, M.J. 1949. The metamorphosis and ecology of some species of Tanypodinae (Dipt., Chironomidae). Entomologist's Monthly Magazine 85:119-126.

384 Morgan, N.C. 1958. Insect emergence from a small Scottish loch. Verhandlungen der Internationale Vereinigung für Theoretische und Angewandte Limnologie 13:823- 825. Morgan, N.C. & A.B. Waddell. 1961. Insect emergence from a small trout loch and its bearing on the food supply of fish. Freshwater and Salmon Fisheries Research 25:1-39. Morin, A. & P.P. Harper. 1986. Phénologie et microdistribution des adultes et des larves de Trichoptères filtreurs dans un ruisseau des Basses Laurentides (Québec). Archiv für Hydrobiologie 108:167-183. Morley, R.L. & R.A. Ring. 1972. The intertidal Chironomidae (Diptera) of British Columbia II. life history and population dynamics. Canadian Entomologist 104:1099-1121. Mozley, S.C. 1970. Morphology and ecology of the larva of Trissocladius grandis (Kieffer) (Diptera, Chironomidae), a common species in the lakes and rivers of Northern Europe. Archiv für Hydrobiologie 67:433-451. Mundie, J.H. 1956. Emergence traps for aquatic insects. Mitteilungen Internationale Vereinigung fur Theoretische und Angewandte Limnologie 7:1-13. Mundie, J.H. 1957. The ecology of Chironomidae in storage reservoirs. Transactions of the Royal Entomological Society of London 109:149-233. Murray, D.A. 1974. Notes on some Chironomidae (Diptera) from the Killarney Area, Ireland. Entomologisk Tidskrift Supplement 95:177-181. Nagell, B. 1978. Resistance to anoxia of Chironomus plumosus and Chironomus anthracinus (Diptera) larvae. Holarctic Ecology 1:333-336. Nebeker, A.V. 1971. Effect of high winter water temperatures on adult emergence of aquatic insects. Water Research 5:777-783. Nebeker, A.V. 1973. Temperature requirements and life cycle of the midge Tanytarsus dissimilis (Diptera: Chironomidae). Journal of the Kansas Entomological Society 46:160-165. Neumann, D. 1961. Osmotische Resistenz und Osmoregulation aquatischer Chironomidenlarven. Biologisches Zentralblatt 80:693-715.

385 Neumann, D. 1966. Die lunare und tägliche Schlüpfperiodik der Mücke Clunio. Steuerung und Abstimmung auf die Gezeitenperiodik. Zeitschrift fur vergleichende Physiologie 53:1-61. Neumann, D. 1971. The temporal programming of development in the intertidal chironomid Clunio marinus (Diptera: Chironomidae). Canadian Entomologist 103:315-318. Neumann, D. 1975. Lunar and tidal rhythms in the development and reproduction of an intertidal organism, p. 451-463. In: Physiological adaptation to the environment : proceedings of a symposium held at the 1973 meeting of the American Institute of Biological Sciences. F.J. Vernberg (ed.). Intext Educational Publishers, New York, NY. Neumann, D. 1976. Adaptations of chironomids to intertidal environments. Annual Review of Entomology 21:387-414. Neumann, D. & M. Krüger. 1985. Combined effects of photoperiod and temperature on the diapause of an intertidal chironomid. Oecologia 67:154-156. Nielsen, E.T. 1962. Contributions to the ethology of Glyptotendipes (Phytotendipes) paripes Edwards. Oikos 13:48-75. Nolte, U. 1990. Chironomid biomass determination from larval shape. Freshwater Biology 24:443-451. Nolte, U. 1991. Seasonal dynamics of moss-dwelling chironomid communities. Hydrobiologia 222:197-211. Nolte, U. 1995. From egg to imago in less than seven days: Apedilum elachistus (Chironomidae), p. 177-184. In: Chironomids: from genes to ecosystems. Vol. 496. P.S. Cranston (ed.). CSIRO Publications, East Melbourne, Australia. Nolte, U. & T. Hoffmann. 1992a. Fast life in cold water: Diamesa incallida (Chironomidae). Ecography 15:25-30. Nolte, U. & T. Hoffmann. 1992b. Life cycle of Pseudodiamesa branickii (Chironomidae) in a small upland stream. Netherlands Journal of Aquatic Ecology 26:309-314.

386 Nordlie, K.J. & J.W. Arthur. 1981. Effect of elevated water temperature on insect emergence in outdoor experimental channels. Environmental Pollution (Series A) 25:53-65. Okazaki, A. & K. Yano. 1990. Biology of Tanytarsus oyamai Sasa (Diptera: Chironomidae). Transactions of the Shikoku Entomological Society 19:89-99. Olander, H., H.J.B. Birks, A. Korhola & T. Blom. 1999. An expanded calibration model for inferring lakewater and air temperatures from fossil chironomid assemblages in Northern Fennoscandia. The Holocene 9:279-294. Olander, R. & E. Palmén. 1968. Taxonomy, ecology and behavior of the northern Baltic Clunio marinus Halid. (Dipt., Chironomidae). Annales Zoologici Fennici 5:97- 110. Oliver, D.R. 1968. Adaptations of Arctic Chironomidae. Annales Entomologici Fennici 5:111-118. Oliver, D.R. 1971. Life history of the Chironomidae. Annual Reviews of Entomology 16:211-230. Oliver, D.R. & R.W. Bode. 1985. Description of the larva and pupa of Cardiocladius albiplumus Sæther (Diptera: Chironomidae). Canadian Entomologist 117:803- 809. Oliver, D.R. & M.E. Dillon. 1994. Systematics of some species of Micropsectra (Diptera: Chironomidae) living in low-order streams in Southern Ontario, Canada. Canadian Entomologist 126:199-217. Oliver, D.R., M.E. Dillon & P.S. Cranston. 1990. A catalog of Nearctic Chironomidae. Canadian Government Publishing Centre, Ottawa, CA. 90 pp. Oliver, D.R. & M.E. Roussel. 1983. Redescription of Brillia Kieffer (Diptera: Chironomidae) with descriptions of Nearctic species. Canadian Entomologist 115:257-279. Olsen, A., J.S. Bale, B.S.C. Leadbeater, M.E. Callow & J.B. Holden. 2003. Developmental thresholds and day-degree requirements of Paratanytarsus grimmii and Corynoneura scutellata (Diptera: Chironomidae): two midges associated with potable water treatment. Physiological Entomology 28:315-322.

387 Olsson, T.I. 1981. Overwintering of benthic macroinvertebrates in ice and frozen sediment in a North Swedish stream. Holarctic Ecology 4:161−166. Oswood, M.W. 1976. Comparative life histories of the Hydropsychidae (Trichoptera) in a Montana lake outlet. American Midland Naturalist 96:493-497. Oswood, M.W. 1989. Community structure of benthic invertebrates in interior Alaskan (USA) streams and rivers. Hydrobiologia 172:97-110. Oswood, M.W., L.K. Miller, & J.G. Irons III. 1991. Overwintering of freshwater benthic macroinvertebrates. In Lee, R.E. Jr. & D.L. Denlinger (eds), Insects at low temperature. Chapman & Hall, New York: 360-375. Paasivirta, L. 1974. Insect emergence and output of incorporated energy and nutrients from the oligotrophic lake Pääjärvi, southern Finland. Annales Zoologici Fennici 11:126-140. Palmén, E. 1956. Periodic emergence in some chironomids - an adaptation to nocturnalism, p. 248-256. In: Bertil Hanström: Zoological Papers in honour of his sixty-fifth birthday November 20th, 1956. K.G. Wingstrand (ed.). Zoological Institute, Lund, Sweden. Palmén, E. 1962. Studies on the ecology and phenology of the chironomids (Dipt.) of the Northern Baltic. 1. Allochironomus crassiforceps K. Annales Entomologici Fennici 28:137-168. Palmén, E. & L. Aho. 1966. Studies on the ecology and phenology of the Chironomidae (Dipt.) of the Northern Baltic. Annales Zoologici Fennici 3:217-244. Paris, O.H., Jr. & C.H. Jenner. 1959. Photoperiodic control of diapause in the pitcher- plant midge, Metriocnemus knabi, p. 601-624. In: Photoperiodism and related phenomena in plants and animals. Vol. 908. R.B. Withrow (ed.). American Association for the Advancement of Science Publication, Washington, D.C. Parkin, R.B. & J.B. Stahl. 1981. Chironomidae (Diptera) of Baldwin Lake, Illinois, a cooling reservoir. Hydrobiologia 76:119-128. Pascoe, D., K.A. Williams & D.W.J. Green. 1989. Chronic toxicity of cadmium to Chironomus riparius Meigen - effects upon larval development and adult emergence. Hydrobiologia 175:109-115.

388 Paterson, C.G. 1971. Overwintering ecology of the aquatic fauna associated with the pitcher plant Sarracenia purpurea L. Canadian Journal of Zoology 49:1455-1459. Paterson, C.G. & C.J. Cameron. 1982. Seasonal dynamics and ecological strategies of the pitcher plant chironomid, Metriocnemus knabi Coq. (Diptera: Chironomidae), in southeast New Brunswick. Canadian Journal of Zoology 60:3075-3083. Paterson, C.G. & K.F. Walker. 1974. Seasonal dynamics and productivity of Tanytarsus barbitarsis Freeman (Diptera: Chironomidae) in the benthos of a shallow, saline lake. Australian Journal of Marine and Freshwater Research 25:151-165. Petr, T. 1970. Chironomidae (Diptera) from light catches on the man-made Volta Lake in Ghana. Hydrobiologia 35:449-468. Phillipp, V.P. 1938. Experimentelle Studien zur Ökologie von Chironomus thummi Kieffer. Zoologischer Anzeiger 122:237-245. Pinder, L.C.V. 1977. The Chironomidae and their ecology in chalk streams. Freshwater Biological Association Annual Report 45:62-69. Pinder, L.C.V. 1983. Observations on the life-cycles of some Chironomidae in Southern England. Memoirs of the American Entomological Society 34:249-265. Pinder, L.C.V. 1985. Studies on Chironomidae in experimental recirculating stream systems. I. Orthocladius (Euorthocladius) calvus sp. nov. Freshwater Biology 15:235-241. Pinder, L.C.V. 1986. Biology of freshwater Chironomidae. Annual Reviews of Entomology 31:1-23. Pinder, L.C.V. 1992. Biology of epiphytic Chironomidae (Diptera: Nematocera) in chalk streams. Hydrobiologia 248:39-51. Pinder, L.C.V. 1995. Biology of the eggs and first-instar larvae, p. 107-135. In: The Chironomidae: the biology and ecology of non-biting midges. P.D. Armitage, P.S. Cranston, & L.C.V. Pinder (eds.). Chapman & Hall, London, UK. Pinder, L.C.V. & P.C. Clare. 1980. The history and production of Rheotanytarsus curtistylus in a small stream in southern England. Acta Universitatis Carolinae - Biologica 12:163-170.

389 Poff, N.L., J.D. Olden, N.K.M. Vieira, D.S. Finn, M.P. Simmons & B.C. Kondratieff. 2006. Functional trait niches of North American lotic insects: traits-based ecological applications in light of phylogenetic relationships. Journal of the North American Benthological Society 25:730-755. Postma, J.F. & C. Davids. 1995. Tolerance induction and life cycle changes in cadmium- exposed Chironomus riparius (Diptera) during consecutive generations. Ecotoxicology and Environmental Safety 30:195-202. Postma, J.F., M.C. Buckert-de Jong, N. Staats & C. Davids. 1994. Chronic toxicity of cadmium to Chironomus riparius (Diptera: Chironomidae) at different food levels. Archives of Environmental Contamination and Toxicology 26:143-148. Postma, J.F., A. van Kleunen & W. Admiraal. 1995. Alterations in life-history traits of Chironomus riparius (Diptera) obtained from metal contaminated rivers. Archives of Environmental Contamination and Toxicology 29:469-475. Potter, D.W.B. & M.A. Learner. 1974. A study of the benthic macro-invertebrates of a shallow eutrophic reservoir in South Wales with emphasis on the Chironomidae (Diptera): their life-histories and production. Archiv für Hydrobiologie 74:186- 226. Prat, N., M. Puig, G. Gonzalez & X. Millet. 1983. Chironomid longitudinal distribution and macroinvertebrate diversity along the Llobregat River (NE Spain). Memoirs of the American Entomological Society 34:267-277. Prat, N. & M. Rieradevall. 1992. Life cycle and production of Cladopelma virescens (Mg.) (Diptera: Chironomidae) in Lake Banyoles (NE Spain). Netherlands Journal of Aquatic Ecology 26:315-320. Prat, N. & M. Rieradevall. 1995. Life cycle and production of Chironomidae (Diptera) from Lake Banyoles (NE Spain). Freshwater Biology 33:511-524. Rabeni, C.F. & N. Wang. 2001. Bioassessment of streams using macroinvertebrates: are the Chironomidae necessary? Environmental Monitoring and Assessment 71:177- 185. Raddum, G.G. 1985. Effects of winter warm reservoir release on benthic stream invertebrates. Hydrobiologia 122:105-111.

390 Rae, J.G. 1985. A multivariate study of resource partitioning in soft bottom lotic Chironomidae. Hydrobiologia 126:275-285. Rae, J.G. 1990. The effect of naturally varying discharge rates on the species richness of lotic midges. Hydrobiologia 196:209-216. Rasmussen, J.B. 1984. Comparison of gut contents and assimilation efficiency of fourth instar larvae of two coexisting chironomid, Chironomus riparius Meigen and Glyptotendipes paripes (Edwards). Canadian Journal of Zoology 62:1022-1026. Rasmussen, J.B. 1984. The life-history, distribution, and production of Chironomus riparius and Glyptotendipes paripes in a prairie pond. Hydrobiologia 119:65-72. Rasmussen, J.B. 1985. Effects of density and microdetritus enrichment on the growth of chironomid larvae in a small pond. Canadian Journal of Fisheries and Aquatic Sciences 42:1418-1422. Raunio, J. & T. Muotka. 2005. The use of chironomid pupal exuviae in river biomonitoring: the importance of sampling strategy. Archiv für Hydrobiologie 164:529-545. Raunio, J., R. Paavola & T. Muotka. 2007. Effects of emergence phenology, taxa tolerances and taxonomic resolution on the use of chironomid pupal exuvial technique in river biomonitoring. Freshwater Biology 52:165-176. Reice, S.R. 1994. Nonequilibrium determinants of biological community structure. American Scientist 82:424-435. Reice, S.R., R.C. Wissmar & R.J. Naiman. 1990. Disturbance regimes, resilience, and recovery of animal communities and habitats in lotic ecosystems. Environmental Management 14:647-659. Reiss, F. 1968. Okologische und systematische Untersuchungen an Chironomiden (Diptera) des Bodensees. Ein Beitrag zur lakustrischen Chironomidenfauna des nordlichen Alpenvorlandes. Archiv für Hydrobiologie 64:176-323. Reist, A. & J. Fischer. 1987. Experimentelle Untersuchungen zur Einwirkung von Temperatur und Besiedlungsdichte auf die Entwicklung der Chironomus-Arten Ch. plumosus, Ch. nuditarsis und Ch. bernensis (Diptera). Zoologische Jahrbücher, Abteilung Systematik, Ökologie und Geographie der Tiere 114:1-13.

391 Reiss, F. & L. Säwedal. 1981. Keys to males and pupae of the Palaearctic (excl. Japan) Paratanytarsus Thienemann & Bause, 1913, n. comb., with descriptions of three new species (Diptera: Chironomidae). Entomologica Scandinavica Supplement 15:73-104. Rempel, J.G. 1936. The life-history and morphology of Chironomus hyperboreus. Journal of the Biological Board of Canada 2:209-221. Rempel, R.S. & J.C.H. Carter. 1987. Temperature influences on adult size, development, and reproductive potential of aquatic Diptera. Canadian Journal of Fisheries and Aquatic Sciences 44:1743-1752. Rempel, R.S. & A.D. Harrison. 1987. Structural and functional composition of the community of Chironomidae (Diptera) in a Canadian Shield stream. Canadian Journal of Zoology 65:2545-2554. Renault, D., C. Salin, G. Vannier, & P. Vernon. 2002. Survival at low temperatures in insects: What is the ecological significance of the supercooling point? CryoLetters 23:217−228. Resh, V.H., A.V. Brown, A.P. Covich, M.E. Gurtz, H.W. Li, G.W. Minshall, S.R. Reice, A.L. Sheldon, J.B. Wallace & R.C. Wissmar. 1988. The role of disturbance in stream ecology. Journal of the North American Benthological Society 7:433-455. Reynolds, S.K. & A.C. Benke. 2005. Temperature-dependent growth rates of larval midges (Diptera: Chironomidae) from a southeastern U.S. stream. Hydrobiologia 544:69-75. Rieradevall, M. & N. Prat. 1986. Diel pattern of chironomid drift in the Llobregat River (N.E. Spain). Oecologia Aquatica 8:61-70. Ring, R.A. 1982. Freezing-tolerant insects with low supercooling points. Comparative Biochemistry and Physiology 73:605−612. Ring, R.A. 1989. Intertidal Chironomidae of B.C., Canada. Acta Biologica Debrecina Supplementum Oecologica Hungarica 3:275−288. Ring, R.A. 1989. Intertidal Chironomidae of B.C., Canada. Acta Biologica Debrecen Oecologica Hungarica 3:275-288.

392 Ringe, V.F. 1974. Chironomiden-Emergenz 1970 in Breitenbach und Rohrwiesenbach Schlitzer Produktionsbiologische Studien (10). Archiv für Hydrobiologie Supplementband 45:212-304. Ristola, T., J. Pellinen, M. Ruokolainen, A. Kostamo & J.V.K. Kukkonen. 1999. Effect of sediment type, feeding level, and larval density on growth and development of a midge (Chironomus riparius). Environmental Toxicology and Chemistry 18:756- 764. Roback, S.S. 1980. The immature chironomids of the eastern United States IV. Tanypodinae - Procladiinae. Proceedings of the Academy of Natural Sciences of Philadelphia 132:1-63. Roback, S.S. 1981. The immature chironomids of the eastern United States V. Pentaneurini - Thienemannimyia Group. Proceedings of the Academy of Natural Sciences of Philadelphia 133:73-128. Roback, S.S. 1985. The immature chironomids of the eastern United States VI. Pentaneurini - Genus Ablabesmyia. Proceedings of the Academy of Natural Sciences of Philadelphia 137(2):153-212. Roback, S.S. 1986. The immature chironomids of the eastern United States VIII. Pentaneurini - Genus Nilotanypus, with the description of a new species from Kansas. Proceedings of the Academy of Natural Sciences of Philadelphia 138:443-465. Roback, S.S. 1987. The immature chironomids of the eastern United States IX. Pentaneurini - genus Labrundinia with the Description of some Neotropical material. Proceedings of the Academy of Natural Sciences of Philadelphia 139:159-209. Rosenberg, D.M. 1992. Freshwater biomonitoring and Chironomidae. Netherlands Journal of Aquatic Ecology 26:101-122. Rosenberg, D.M., A.P. Wiens & B. Bilyj. 1988. Chironomidae (Diptera) of peatlands in northwestern Ontario, Canada. Holarctic Ecology 11:19-31. Rosenberg, D.M., A.P. Wiens & O.A. Sæther. 1977. Life histories of Cricotopus (Cricotopus) bicinctus and C. (C.) mackenziensis (Diptera: Chironomidae) in the

393 Fort Simpson Area, Northwest Territories. Journal of the Fisheries Research Board of Canada 34:247-253. Ross, D.H. & J.B. Wallace. 1982. Factors influencing the longitudinal distribution of larval Hydropsychidae (Trichoptera) in a southern Appalachian stream system (U.S.A.). Hydrobiologia 96:185-199. Rossaro, B. 1978. Composizione tassonomica e fenologia delle Orthocladiinae (Diptera, Chironomidae) nel Po a Caorso (piacenza), determinate mediante analisi delle exuviae delle pupe. Rivista di Idrobiologia 17:287-300. Rossaro, B. 1987. Chironomid emergence in the Po River (Italy) near a nuclear power plant. Entomologica Scandinavica Supplementum 29:331-338. Rossaro, B. 1991. Chironomids and water temperature. Aquatic Insects 13:87-98. Rossaro, B. & R. Cironi. 1987. Insect emergence in a stretch of the Po River (Italy). Archiv für Hydrobiologie 109:245-256. Ruse, L.P. 1992. Correlations between chironomid pupal skin collections and habitats recorded from a chalk river. Netherlands Journal of Aquatic Ecology 26:411-417. Ruse, L.P. 1995a. Chironomid community structure deduced from larvae and pupal exuviae of a chalk stream. Hydrobiologia 315:135-142. Ruse, L. 1995b. Chironomid emergence from an English chalk stream during a three year study. Archiv für Hydrobiologie 133:223-244. Ruse, L. & M. Davison. 2000. Long-term data assessment of chironomid taxa structure and function in the River Thames. Regulated Rivers: Research & Management 16:113-126. Ruse, L.P., S.J. Herrmann & J.E. Sublette. 2000. Chironomidae (Diptera) species distribution related to environmental characteristics of the metal-polluted Arkansas River, Colorado. Western North American Naturalist 60:34-56. Ruse, L.P. & R.S. Wilson. 1984. The monitoring of river water quality within the Great Ouse Basin using the chironomid exuvial analysis technique. Water Pollution Control 83:116-135. Ruse, L.P. & R.S. Wilson. 1995. Long-term assessment of water and sediment quality of the River Thames using chironomid pupal skins, p. 113-123. In: Chironomids:

394 from genes to ecosystems. P.S. Cranston (ed.). CRISO, East Melbourne, Australia. Růžičková, J. 1987. Cohort analysis and production estimate of Chironomus larvae (Diptera, Chironomidae) in a carp pond in Southwest Bohemia. Vestník Ceskoslovenské Spolecnosti Zoologické 51:140-151. Rychen Bangerter, B. & J. Fischer. 1989. Different dormancy response in the sympatric Chironomus species Ch. plumosus and Ch. nuditarsis. Zoologische Jahrbücher, Abteilung Systematik, Ökologie und Geographie der Tiere 116:145-150. Sadler, W.O. 1935. Biology of the midge Chironomus tentans Fabricius, and methods for its propagation. Memoirs of the New York State Agricultural Experiment Station 173:1-25. Sæther, O.A. 1962. Larval overwintering cocoons in Endochironomus tendens Fabricius. Hydrobiologia 20:377-381. Sæther, O.A. 1968. Chironomids of the Finse area, Norway, with special reference to their distribution in a glacier brook. Archiv für Hydrobiologie 64:426-483. Sæther, O.A. 1969. Some Nearctic Podonominae, Diamesinae and Orthocladiinae (Diptera: Chironomidae). Fisheries Research Board of Canada, Ottawa, Canada. 154 pp. Sæther, O.A. 1975. Nearctic and Palaearctic Heterotrissocladius (Diptera: Chironomidae). Department of Fisheries and the Environment, Fisheries and Marine Service, Ottawa, Canada. 67 pp. Sæther, O.A. 1976. Revision of Hydrobaenus, Trissocladius, Zalutschia, Paratrissocladius, and some related genera (Diptera: Chironomidae). Department of Fisheries and the Environment, Fisheries and Marine Service, Ottawa, Canada. 287 pp. Sæther, O.A. 1977. Female genitalia in Chironomidae and other Nematocera: morphology, phylogenies, keys. Department of Fisheries and the Environment, Fisheries and Marine Service, Ottawa, Canada. 209 pp.

395 Sæther, O.A. 1977. Taxonomic studies on Chironomidae: Nanocladius, Pseudochironomus, and the Harnischia complex. Department of Fisheries and the Environment, Ottawa, Canada. 144 pp. Sæther, O.A. 1983. Three new species of Lopescladius Oliveira, 1967 (syn. "Cordites" Brundin, 1966, n. syn.), with a Phylogeny of the Parakiefferiella group. Memoirs of the American Entomological Society 34:279-298. Sæther, O.A. 1985a. A review of Odontomesa, Pagast, 1947 (Diptera, Chironomidae, Prodiamesinae). Spixiana Supplement 11:15-29. Sæther, O.A. 1985b. A review of the genus Rheocricotopus Thienemann & Harnisch, 1932 (Diptera: Chironomidae), with the description of three new species. Spixiana Supplement 11:59-108. Sæther, O.A. 1990. A review of the genus Limnophyes Eaton from the Holarctic and Afrotropical regions (Diptera: Chironomidae, Orthocladiinae). Entomologica Scandinavica Supplement 35:1-139. Sæther, O.A. 2000. Phylogeny of the subfamilies of Chironomidae (Diptera). Systematic Entomology 25:393-403. Sæther, O.A. 2003. A Review of Orthocladius subgen. Symposiocladius Cranston (Diptera: Chironomidae). Aquatic Insects 25:281-317. Sæther, O.A. & J.E. Sublette. 1983. A review of the genera Doithrix n. gen., Georthocladius Strenzke, Parachaetocladius Wülker and Pseudorthocladius Goetghebuer (Diptera: Chironomidae, Orthocladiinae). Entomologica Scandinavica Supplement 20:1-100. Sæther, O.A. & X. Wang. 1995. Revision of the genus Paraphaenocladius Thienemann, 1924 of the world (Diptera: Chironomidae, Orthocladiinae). Entomologica Scandinavica Supplement 48:1-69. Saigusa, M. & T. Akiyama. 1995. The tidal rhythm of emergence, and the seasonal variation of this synchrony, in an intertidal midge. Biological Bulletin 188:166- 178. Salt, R.W. 1953. The influence of food on cold hardiness of insects. The Canadian Entomologist 85:261-269.

396 Salt, R.W. 1961. Principles of insect cold-hardiness. Annual Review of Entomology 6:55−74. Saltveit, S.J., T. Bremnes & J.E. Brittain. 1994. Effect of a changed temperature regime on the benthos of a Norwegian regulated river. Regulated Rivers: Research & Management 9:93-102. Sandberg, G. 1969. A quantitative study of chironomid distribution and emergence in Lake Erken. Archiv für Hydrobiologie Supplementband 35:119-201. Sankarperumal, G. & T.J. Pandian. 1991. Effect of temperature and Chlorella density on growth and metamorphosis of Chironomus circumdatus (Kieffer) (Diptera). Aquatic Insects 13:167-177. SAS Institute Inc. 1998. Statistical Analysis System, version 6.12. SAS Institute, Cary. Schmid, P.E. 1992. Habitat preferences as patch selection of larval and emerging chironomids (Diptera) in a gravel brook. Netherlands Journal of Aquatic Ecology 26:419-429. Scholander, P.F., W. Flagg, R.J. Hock, & L. Irving. 1953. Studies on the physiology of frozen plants and animals in the arctic. Journal of Cellular and Comparative Physiology 42:1−56. Sealock, A.W. & L.C. Ferrington, Jr. (submitted) Sampling efficiency of Chironomidae (Diptera) across disturbance gradients. Boletim do Museu Municipal do Funchal. Sendstad, E., J.O. Solem & K. Aagaard. 1976. Studies of terrestrial chironomids (Diptera) from Spitsbergen. Norwegian Journal of Entomology 24:91-98. Sephton, T. & C.G. Paterson. 1986. Production of the chironomid Procladius bellus in an annual drawdown reservoir. Freshwater Biology 16:721-733. Servia, M.J., A.R.R. Péry, M. Heydorff, J. Garric & L. Lagadic. 2006. Effects of copper on energy metabolism and larval development in the midges Chironomus riparius. Ecotoxicology 15:229-240. Sheldon, A.L. & S.G. Jewett, Jr. 1967. Stonefly emergence in a Sierra Nevada stream. Pan-Pacific Entomologist 43:1-8.

397 Sherk, T. & G. Rau. 1992. Emergence of Chironomidae from Findley Lake and two ponds in the Cascade Mountains, U.S.A. Netherlands Journal of Aquatic Ecology 26:321-330. Sherk, T. & G. Rau. 1996. Emergence of Chironomidae from Findley Lake in the coniferous forest of the Cascade Mountains after early and late thaws. Hydrobiologia 318:85-101. Shiozawa, D.K. & J.R. Barnes. 1977. The microdistribution and population trends of larval Tanypus stellatus Coquillett and Chironomus frommeri Atchley and Martin (Diptera: Chironomidae) in Utah Lake, Utah. Ecology 58:610-618. Sibley, P.K., D.A. Benoit & G.T. Ankley. 1998. Life cycle and behavioral assessments of the influence of substrate particle size on Chironomus tentans (Diptera: Chironomidae) in laboratory studies. Hydrobiologia 361:1-9. Siebert, M. 1980. Die Emergenz der Chironomiden im Breitenbach 1969-1973. Archiv für Hydrobiologie Supplement 58:310-355. SigmaPlot. 2001. SigmaPlot 2001 for Windows, version 7.0. SPSS Inc. Chicago. Silver, P., J.K. Cooper, M.A. Palmer & E.J. Davis. 2000. The arrangement of resources in patchy landscapes: effect on distribution, survival, and resource acquisition of chironomids. Oecologia 124:216-224. Simpson, K.W., R.W. Bode & P. Albu. 1983. Keys for the genus Cricotopus adapted from "Revision der Gattung Cricotopus van der Wulp und ihrer Verwandten (Diptera, Chironomidae)" by M. Hirvenoja. University of the State of New York, Albany, NY. 134 pp. Sinclair, B.J. 1999. Insect cold tolerance: how many kinds of frozen? European Journal of Entomology 96:157-164. Singh, M.P. & A.D. Harrison. 1984. The chironomid community (Diptera: Chironomidae) in a Southern Ontario stream and the annual emergence patterns of common species. Archiv für Hydrobiologie 99:221-253. Singh, M.P. & A.D. Harrison. 1984. The chironomid community (Diptera: Chironomidae) in a Southern Ontario stream and the annual emergence patterns of common species. Archiv für Hydrobiologie 99:221-253.

398 Smukalla, R. & E. Meyer. 1988. Insect emergence from a shallow southern West German lake, with special reference to the parasitic host-associated water mite larvae. Hydrobiologia 169:149-166. Sokal, R.R. & F.J. Rohlf. 1995. Biometry. W.H. Freeman and Co., New York, NY. 887 pp. Sokolova, G.A. 1966. On the hibernation mode of larvae of Limnochironomus ex. gr. nervosus Staeg. (Diptera, Chironomidae). Zoologicheskii zhurnal 45:140. Sokolova, N.Y. 1971. Life cycles of chironomids in the Uchinskoye Reservoir. Limnologica 8:151-155. Soluk, D.A. 1985. Macroinvertebrate abundance and production of psammophilous Chironomidae in shifting sand areas of a lowland river. Canadian Journal of Fisheries and Aquatic Sciences 42:1296-1302. Sømme, L. 1982. Supercooling and winter survival in terrestrial arthropods. Comparative Biochemistry and Physiology A: Physiology 73:519-543. Sømme, L. & W. Block. 1991. Adaptations to alpine and polar environments in insects and other terrestrial arthropods. In RE Lee Jr, DL Denlinger, editors. Insects at low temperature. New York: Chapman and Hall. p. 318−359. Sømme, L. & E. Østbye. 1969. Cold hardiness in some winter active insects. Norsk Entomologisk Tidskrift 16:45−48. Sonoda, K.C. & S. Trivinho-Strixino. 2000. Dinâmica da emergência de Chironomidae (Diptera) da fitofauna de Cabomba piauhyensis Gardney, 1844, na Lagoa do infernão (Estação Ecológica de Jataí, Luiz Antônio, SP), p. 743-754. In: Estação Ecológica de Jataí. Vol. 900. J.E. Santos & J.S.R. Pires (eds.). Rima, São Carlos. Soong, K., G. Chen & J. Cao. 1999. Life history studies of the flightless marine midges Pontomyia spp. (Diptera: Chironomidae). Zoological Studies 38:466-473. Soponis, A.R. 1977. A revision of the Nearctic species of Orthocladius (Orthocladius) van der Wulp (Diptera: Chironomidae). Memoirs of the Entomological Society of Canada 102:1-188. Soponis, A.R. 1980. Taxonomic composition of Chironomidae (Diptera) in a sand- bottomed stream of northern Florida, p. 163-169. In: Chironomidae - ecology,

399 systematics, cytology and physiology. D.A. Murray (ed.). Pergamon Press, Oxford, UK. Soponis, A.R. 1983. Emergence of Polypedilum (Chironomidae) in a sand-bottomed stream of northern Florida. Memoirs of the American Entomological Society 34:309-313. Soponis, A.R. 1987. Notes on Orthocladius (Orthocladius) frigidus (Zetterstedt) with a redescription of the species (Diptera: Chironomidae). Entomologica Scandinavica Supplements 29:123-131. Soponis, A.R. 1990. A revision of the Holarctic species of Orthocladius (Euorthocladius) (Diptera: Chironomidae). Spixiana Supplement 13:1-56. Southwood T.R.E. & P.A. Henderson. 2000. Ecological Methods. Blackwell Ltd., Oxford. Sovell, L.A. & B. Vondracek. 1999. Evaluation of the fixed-count method for Rapid Bioassessment Protocol III with benthic macroinvertebrate metrics. Journal of the North American Benthological Society 18:420-426. Sprules, W.M. 1947. An ecological investigation of stream insects in Algonquin Park, Ontario. University of Toronto Studies Biological Series 56:1-81. Stagliano, D.M., A.C. Benke & D.H. Anderson. 1998. Emergence of aquatic insects from 2 habitats in a small wetland of the southeastern USA: temporal patterns of numbers and biomass. Journal of the North American Benthological Society 17:37-53. Stahl, J.B. 1986. A six-year study of abundance and voltinism of Chironomidae (Diptera) in an Illinois cooling reservoir. Hydrobiologia 134:67-79. Stanford, J.A. & J.V. Ward. 1983. Insect species diversity as a function of environmental variability and disturbance in stream systems, p. 265-278. In: Stream ecology: application and testing of general ecological theory. J.R. Barnes & G.W. Minshall (eds.). Plenum Press, New York, NY. Stanko-Mishic, S., J.K. Cooper & P. Silver. 1999. Manipulation of habitat quality: effects on chironomid life history traits. Freshwater Biology 41:719-727.

400 Stevens, M.M. 1993. Larval development in Chironomus tepperi (Diptera: Chironomidae) under laboratory conditions. Population Ecology 22:776-780. Stevens, M.M. 1998. Development and survival of Chironomus tepperi Skuse (Diptera: Chironomidae) at a range of constant temperatures. Aquatic Insects 3:181-188. Stevens, M.M., S. Helliwell & P.S. Cranston. 2006. Larval chironomid communities (Diptera: Chironomidae) associated with establishing rice crops in southern New South Wales, Australia. Hydrobiologia 556:317-325. Stites, D.L. & A.C. Benke. 1989. Rapid growth rates of chironomids in three habitats of a subtropical blackwater river and their implications for P:B ratios. Limnology and Oceanography 34:1278-1289. Storey, A.W. & L.C.V. Pinder. 1985. Mesh-size and efficiency of sampling of larval Chironomidae. Hydrobiologia 124:193-197. Storey, A.W. 1987. Influence of temperature and food quality on the life history of an epiphytic chironomid. Entomologica Scandinavica Supplementum 29:339-347. Storey, K.B. & J.M. Storey. 1997. To freeze or not to freeze − the dilemma for life below 0ºC. The Biochemist 19: 8-13. Stout, J. & J. Vandermeer. 1975. Comparison of species richness for stream-inhabiting insects in tropical and mid-latitude streams. American Naturalist 109:263-280. Strixino, S.T. & G. Strixino. 1982. Ciclo de vida de Chironomus sancticaroli Strixino & Strixino, (Diptera, Chironomidae). Revista Brasileira de Entomologia 26:183-189. Suemoto, T., K. Kawai & H. Imabayashi. 2005. Dried-up zone as a temporal stoke of chironomid larvae: survival periods and density in a reservoir bank. Hydrobiologia 545:145-152. Sugg, P., J.S. Edwards & J. Baust. 1983. Phenology and life history of Belgica antarctica, an Antarctic midge (Diptera: Chironomidae). Ecological Entomology 8:105-113. Surakarn, R. & K. Yano. 1995. Development of a paddy-dwelling chironomid, Chironomus kiiensis (Diptera, Chironomidae) under different temperatures. Japanese Journal of Entomology 63:389-398.

401 Swanson, S.M. & U.T. Hammer. 1983. Production of Cricotopus ornatus (Meigen) (Diptera: Chironomidae) in Waldsea Lake, Saskatchewan. Hydrobiologia 105:155-164. Sweeney, B.W. 1984. Factors influencing life history patterns of aquatic insects, p. 56- 100. In: Ecology of aquatic insects. V.H. Resh & D. Rosenberg (eds.). Praeger Scientific Publishers, New York, NY. Sweeney, B.W., J.K. Jackson & D.H. Funk. 1995. Semivoltinism, seasonal emergence, and adult size variation in a tropical stream mayfly (Euthyplocia hecuba). Journal of the North American Benthological Society 14:131-146. Sweeney, B.W. & R.L. Vannote. 1978. Size variation and the distribution of hemimetabolous aquatic insects: Two thermal equilibrium hypotheses. Science 200:444-446. Sweeney, B.W. & R.L. Vannote. 1981. Ephemerella mayflies of White Clay Creek: bioenergetic and ecological requirements among six coexisting species. Ecology 62:1353-1369. Syrjämäki, J. 1965. Laboratory studies on the swarming behavior of Chironomus strenzkei Fittkau in litt. (Dipt., Chironomidae) I. Mechanism of swarming and mating. Annales Zoologici Fennici 2:145-152. Tait Bowman, C.M. 1980. Emergence of chironomids from Rostherne Mere, England, p. 291-295. In: Chironomidae - ecology, systematics, cytology and physiology. D.A. Murray (ed.). Pergamon Press, Oxford, UK. Tavares-Cromar, A.F. & D.D. Williams. 1997. Dietary overlap and coexistence of chironomid larvae in a detritus-based stream. Hydrobiologia 354:67-81. Taylor, E.J., S.J. Blockwell, S.J. Maund & D. Pascoe. 1993. Effects of lindane on the life-cycle of a freshwater macroinvertebrate Chironomus riparius Meigen (Insecta: Diptera). Archives of Environmental Contamination and Toxicology 24:145-150. Teague, S.A., A.W. Knight & B.N. Teague. 1985. Stream microhabitat selectivity, resource partitioning, and niche shifts in grazing caddisfly larvae. Hydrobiologia 128:3-12.

402 Thienemann, A. 1910. Das sammeln von Puppenhäuten der chironomiden. Eine bitte um Mitarbeit. Archive für Hydrobiologie 6:213-214. Thienemann, A. 1954. Chironomus. Leben, Verbreitung und wirtshaftliche Bedeutung der Chironomiden. Binnengewasser 20:1-834. Thienemann, A. & K. Strenzke. 1940. Terrestrische Chironomiden III-IV: Zwei parthenogenetische Formen. Zoologischer Anzeiger 132:24-40. Thorp, J.H. & R.K. Chesser. 1983. Seasonal responses of lentic midge assemblages to environmental gradients. Holarctic Ecology 6:123-132. Timmermans, K.R., W. Peeters & M. Tonkes. 1992. Cadmium, zinc, lead and copper in Chironomus riparius (Meigen) larvae (Diptera, Chironomidae): uptake and effects. Hydrobiologia 241:119-134. Titmus, G. 1979. The emergence of midges (Diptera: Chironomidae) from a wet gravel- pit. Freshwater Biology 9:165-179. Titmus, G. & R.M. Badcock. 1981. A morphometric study of the effects of a mermithid parasite on its host, the chironomid midge Einfeldia dissidens (Walker) (Diptera). Zeitschrift für Parasitenkunde 65:353-357. Tjønneland, A. 1962. The nocturnal flight activity and the lunar rhythm of emergence in the African midge Conochironomus acutistilus (Freeman). Contributions from the Faculty of Science, University College of Addis Ababa, Series Zoology 4:1-21. Tokeshi, M. & C.R. Townsend. 1987. Random patch formation and weak competition: coexistence in an epiphytic chironomid community. Journal of Animal Ecology 56:833-845. Tokeshi, M. 1986a. Population dynamics, life histories and species richness in an epiphytic chironomid community. Freshwater Biology 16:431-441. Tokeshi, M. 1986b. Resource utilization, overlap and temporal community dynamics: a null model analysis of an epiphytic chironomid community. Journal of Animal Ecology 55:491-506. Tokeshi, M. 1991a. Faunal assembly in chironomids (Diptera): generic association and spread. Biological Journal of the Linnean Society 44:353-367.

403 Tokeshi, M. 1991b. On the feeding habits of Thienemannimyia festiva (Diptera: Chironomidae). Aquatic Insects 13:9-16. Tokeshi, M. 1992a. The structure of diversity in an epiphytic chironomid community. Netherlands Journal of Aquatic Ecology 26:461-470. Tokeshi, M. 1992b. Dynamics of distribution in animal communities: theory and analysis. Researches on Population Ecology 34:249-273. Tokeshi, M. 1993. On the evolution of commensalism in the Chironomidae. Freshwater Biology 29:481-489. Tokeshi, M. 1995. Life cycles and development, p. 224-268. In: The Chironomidae: the biology and ecology of non-biting midges. P.D. Armitage, P.S. Cranston, & L.C.V. Pinder (eds.). Chapman & Hall, London, UK. Towns, D.R. 1981. Life histories of benthic invertebrates in a kauri forest stream in northern New Zealand. Australian Journal of Marine and Freshwater Research 32:191-211. Townsend, C.R. 1989. The patch dynamics concept of stream community ecology. Journal of the North American Benthological Society 8:36-50. Townsend, C.R. & A.G. Hildrew. 1979. Resource partitioning by two freshwater invertebrate predators with contrasting foraging strategies. Journal of Animal Ecology 48:909-920. Townsend, C.R., M.R. Scarsbrook & S. Dolédec. 1997. The intermediate disturbance hypothesis, refugia, and biodiversity in streams. Limnology and Oceanography 42:938-949. Tuiskunen, J. 1986. The Fennoscandian species of Parakiefferiella Thienemann (Diptera: Chironomidae, Orthocladiinae). Annales Zoologici Fennici 23:175-196. Turnock, W.J., R.J. Lamb, & R.P. Bodnaryk. 1983. The effects of cold stress during pupal diapause on the survival of Mamestra configurata (Lepidoptera: Noctuidae). Oecologia 56:185−192. Uutala, A.J. 1981. Composition and secondary production of the chironomid (Diptera) communities in two lakes in the Adirondack mountain region, New York, p. 139- 154. In: Effects of Acidic Precipitation on Benthos, Proceedings of the

404 Symposium on the effects of Acidic Precipitation on Benthos, 1980. R. Singer (ed.). North American Benthological Society, Hamilton, N.Y. van de Bund, W.J. & C. Davids. 1994. The influence of biotic factors on life-history parameters of a littoral chironomid species. Verhandlungen der Internationale Vereinigung für Theoretische und Angewandte Limnologie 25:2482-2484. van der Kamp, G. 1995. The hydrogeology of springs in relation to the biodiversity of spring fauna: a review. Journal of the Kansas Entomological Society 68:4-17. van der Velde, G. & R. Hiddink. 1987. Chironomidae mining in Nuphar lutea (L.) Sm. (Nymphaeaceae). Entomologica Scandinavica 29:255-264. Vannote, R.L., G.W. Minshall, K.W. Cummins, J.R. Sedell & C.E. Cushing. 1980. The river continuum concept. Canadian Journal of Fisheries and Aquatic Sciences 37:130-137. Vannote, R.L. & B.W. Sweeney. 1980. Geographic analysis of thermal equilibria: a conceptual model for evaluation the effect of natural and modified thermal regimes on aquatic insect communities. The American Naturalist 115:667-695. Vilchez-Quero, A. & P. Lavandier. 1986. Composition et rythme journalier de la dérive des exuvies nymphales de Chironomidés dans le Guadalquivir (Sierra de Cazorla - Espagne). Annales de Limnologie 22:253-260. Vinson, M.R. & C.P. Hawkins. 1996. Effects of sampling area and subsampling procedure in comparisons of taxa richness among streams. Journal of the North American Benthological Society 15:392-399. Walker, I.R., J.P. Smol, D.R. Engstrom & H.J.B. Birks. 1991. An assessment of Chironomidae as quantitative indicators of past climatic change. Canadian Journal of Fisheries and Aquatic Sciences 48:975-987. Walsh, C.J. 1997. A multivariate method for determining optimal subsample size in the analysis of macroinvertebrate samples. Marine and Freshwater Research 48:241- 248. Walther, D.A., M.R. Whiles, M.B. Flinn & D.W. Butler. 2006. Assemblage-level estimation of nontanypodine chironomid growth and production in a southern Illinois stream. Journal of the North American Benthological Society 25:444-452.

405 Ward, A.F. & D.D. Williams. 1986. Longitudinal zonation and food of larval chironomids (Insecta: Diptera) along the course of a river in temperate Canada. Holarctic Ecology 9:48-57. Ward, G.M. & K.W. Cummins. 1978. Life history and growth pattern of Paratendipes albimanus in a Michigan headwater stream. Annals of the Entomological Society of America 71:272-284. Ward, G.M. & K.W. Cummins. 1979. Effects of food quality on growth of a stream detritivore, Paratendipes albimanus (Meigen) (Diptera: Chironomidae). Ecology 60:57-64. Ward, J.V. 1976. Effects of thermal constancy and seasonal temperature displacement on community structure of stream macroinvertebrates, p. 302-307. In: Thermal Ecology II. Vol. 404. G.W. Esch & R.W. McFarlane (eds.). Technical Information Center, Energy Research and Development Administration, Oak Ridge, TN. Ward, J.V. 1985. Thermal characteristics of running waters. Hydrobiologia 125:31-46. Ward, J.V. 1986. Altitudinal zonation in a Rocky Mountain stream. Archiv für Hydrobiologie Supplementband 74:133-199. Ward, J.V. 1994. Ecology of alpine streams. Freshwater Biology 32:277-294. Ward, J.V. & R.G. Dufford. 1979. Longitudinal and seasonal distribution of macroinvertebrates and epilithic algae in a Colorado springbrook-pond system. Archiv für Hydrobiologie 86:284-321. Ward, J.V. & J.A. Stanford. 1982. Thermal responses in the evolutionary ecology of aquatic insects. Annual Review of Entomology 27:97-117. Wartinbee, D.C. 1979. Diel emergence patterns of lotic Chironomidae. Freshwater Biology 9:147-156. Wartinbee, D.C. & W.P. Coffman. 1976. Quantitative determination of chironomid emergence from enclosed channels in a small lotic ecosystem. American Midland Naturalist 95:479-485.

406 Watts, M.M. & D. Pascoe. 1996. Use of the freshwater macroinvertebrate Chironomus riparius (Diptera: Chironomidae) in the assessment of sediment toxicity. Water Science and Technology 34:101-107. Watts, M.M. & D. Pascoe. 2000. Comparison of Chironomus riparius Meigen and Chironomus tentans Fabricius (Diptera: Chironomidae) for assessing the toxicity of sediments. Environmental Toxicology and Chemistry 19:1885-1892. Webb, C.O., D.D. Ackerly, M.A. McPeek & M.J. Donoghue. 2002. Phylogenies and community ecology. Annual Review of Ecology and Systematics 33:475-505. Webb, D.W., M.J. Wetzel, P.C. Reed, L.R. Phillippe & M.A. Harris. 1995. Aquatic biodiversity of Illinois springs. Journal of the Kansas Entomological Society 68:93-107. Welch, H.E., Jr. 1973. Emergence of Chironomidae (Diptera) from Char Lake, Resolute, Northwest Territories. Canadian Journal of Zoology 51:1113-1123. Welch, H.E. 1976. Ecology of Chironomidae (Diptera) in a polar lake. Journal of the Fisheries Research Board of Canada 33:227-247. Welch, H.E., J.K. Jorgenson & M.F. Curtis. 1988. Emergence of Chironomidae (Diptera) in fertilized and natural lakes at Saqvaqjuac, N.W.T. Canadian Journal of Fisheries and Aquatic Sciences 45:731-737. Wentsel, R., A. McIntosh & W.P. McCafferty. 1978. Emergence of the midge Chironomus tentans when exposed to heavy metal contaminated sediment. Hydrobiologia 57:195-196. Werle, S.F., D.G. Smith & E. Klekowski. 2004. Life in crumbling clay: the biology of Axarus species (Diptera: Chironomidae) in the Connecticut River. Northeastern Naturalist 11:443-458. Wiederholm, T. 1986. Chironomidae of the Holarctic region: Part 2. Pupae. Vol. 28. Entomologica Scandinavica, Sandby, Sweden. pp. 482. Wiederholm, T., K. Danell & K. Sjöberg. 1977. Emergence of chironomids from a small man-made lake in northern Sweden. Norwegian Journal of Entomology 24:99- 105.

407 Wilcox, H.S., J.B. Wallace, J.L. Meyer & J.P. Benstead. 2005. Effects of labile carbon addition on a headwater stream food web. Limnology and Oceanography 50:1300-1312. Williams, D.D. & H.B.N. Hynes. 1976. The ecology of temporary streams I. the faunas of two Canadian streams. Internationale Revue der Gesamten Hydrobiologie 61:761-787. Williams, D.D. 1982. Emergence pathways of adult insects in the upper reaches of a stream. Internationale Revue der Gesamten Hydrobiologie 67:223-234. Williams, D.D. 1998. The role of dormancy in the evolution and structure of temporary water invertebrate communities. Advances in Limnology 52:109-124. Williams, D.D. & I.D. Hogg. 1988. Ecology and production of invertebrate in a Canadian coldwater spring-springbrook system. Holarctic Ecology 11:41-54. Wilson, R.S. 1977. Chironomid pupal exuviae in the River Chew. Freshwater Biology 7:9-17. Wilson, R.S. 1980. Classifying rivers using chironomid pupal exuviae, p. 209-216. In: Chironomidae - ecology, systematics, cytology and physiology. D.A. Murray (ed.). Pergamon Press, Oxford, UK. Wilson, R.S. 1987. Chironomid communities in the River Trent in relation to water chemistry. Entomologica Scandinavica Supplementum 29:387-393. Wilson, R.S. 1988. A survey of the zinc-polluted River Nent (Cumbria) and the East and West Allen (Northumberland), England, using chironomid pupal exuviae. Spixiana 14:167-174. Wilson, R.S. 1989. The modification of chironomid pupal exuvial assemblages by sewage effluent in rivers within the Bristol Avon catchment, England. Acta Biologica Debrecen Oecologica Hungarica 3:367-376. Wilson, R.S. 1992. Monitoring organic enrichment of rivers using chironomid pupal exuvial assemblages. Netherlands Journal of Aquatic Ecology 26:521-525. Wilson, R.S. 1994. Monitoring the effect of sewage effluent on the Oxford Canal using chironomid pupal exuviae. Water and Environmental Management 8:171-182.

408 Wilson, R.S. & J.D. McGill. 1977. A new method of monitoring water quality in a stream receiving sewage effluent using chironomid pupal exuviae. Water Research 11:959-962. Wilson, R.S. & P.L. Bright. 1973. The use of chironomid pupal exuviae for characterizing streams. Freshwater Biology 3:283-302. Wilson, R.S. & S.E. Wilson 1983. A reconnaissance of the River Rhine using Chironomidae pupal exuviae (Insecta: Diptera). Memoirs of the American Entomological Society 34:361-385. Wilson, R.S. & S.E. Wilson. 1984. A survey of the distribution of Chironomidae (Diptera, Insecta) of the River Rhine by sampling pupal exuviae. Hydrobiological Bulletin 18:119-132. Winterbourn, M.J. 1969. The distribution of algae and insects in hot spring thermal gradients at Waimangu, New Zealand. New Zealand Journal of Marine and Freshwater Research 3:459-465. Wolda, H. & R.W. Flowers. 1985. Seasonality and diversity of mayfly adults (Ephemeroptera) in a "nonseasonal" tropical environment. Biotropica 17:330-335. Wotton, R.S., P.D. Armitage, K. Aston, J.H. Blackburn, M. Hamburger & C.A. Woodward. 1992. Colonization and emergence of midges (Chironomidae: Diptera) in slow sand filter beds. Netherlands Journal of Aquatic Ecology 26:331- 339. Wright, C.A., L.C., Ferrington, Jr., & N.H. Crisp. 1996. Analysis of chlordane-impacted stream using chironomid pupal exuviae (Diptera: Chironomidae). Hydrobiologia 318:69-77. Wright, I. & P. Cranston. 2000. Are Australian lakes different? - chironomid and chaoborid exuviae from Lake McKenzie, a coastal temperate dune lake. Verhandlungen der Internationale Vereinigung für Theoretische und Angewandte Limnologie 27:303-308. Wright, J.F. 1978. Seasonal and between year variation in the chironomid larvae of a chalk stream. Verhandlungen der Internationale Vereinigung für Theoretische und Angewandte Limnologie 20:2647-2651.

409 Wrubleski, D.A. & D.M. Rosenberg. 1990. The Chironomidae (Diptera) of Bone Pond, Delta Marsh, Manitoba, Canada. Wetlands 10:243-275. Wülker, W. & P. Götz. 1968. Die Verwendung der Imaginalscheiben zur Bestimmung des Entwicklungszustandes von Chironomus-Larven (Dipt.). Zeitschrift für Morphologie der Tiere 62:363-388. Yamagishi, H. & H. Fukuhara. 1970. On the summer burrowing of the chironomid larvae, Spaniotoma akamusi Tokunaga into the deeper mud layer in Lake Suwa. Zoological Magazine 79:89-91. Yamagishi, H. & H. Fukuhara. 1971. Ecological studies on Chironomidae in Lake Suwa. I. Population dynamics of two large chironomids, Chironomus plumosus L. and Spaniotoma akamusi Tokunaga. Oecologia 7:309-327. Yano, K., M. Takayama & T. Ueda. 1991. Biology of Glyptotendipes tokunagai Sasa (Diptera: Chironomidae) 1. development of immature stages. Transactions of the Shikoku Entomological Society 19:177-188. Young, R.M. 1969. Field observations on a midwinter breeding flight of Diamesa arctica (Diptera: Chironomidae). Annals of the Entomological Society of America 6:1204. Zachariassen, K.E. 1985. Physiology of cold tolerance in insects. Physiological Reviews 65:799-832. Zachariassen, K.E. & E. Kristiansen. 2000. Ice nucleation and antinucleation in nature. Cryobiology 41:257-279. Zachariassen, K.E., E. Kristiansen, & S.A. Pedersen. 2004. Inorganic ions in cold- hardiness. Cryobiology 48:126-133. Zaćwilichowska, K. 1970. Diptera larvae in the River San and in some of its tributaries. Acta Hydrobiologica 12:197-208. Zaderer, P. 1982. Emergenz der Chironomiden im Gossenköllesee im Sommer 1981: Unterschiede der Fängigkeit zweier verschiedener Fallensysteme. Jahresbericht der Abteilutig Limnologie Innsbruck 8:135-144. Zavrel, J. 1926. "Tanytarsus connectens". Spisy Vydavane Prirodovedeckov Fakultou Masarykovy University 65:1-47.

410 Zbarakh, T.I. 1971. On the biology of Chironomus behningi Goetgh. of Lake Biili-Kul (River Talas Basin). Limnologica 8:141-142. Ziser, S.W. 1985. The effects of a small reservoir on the seasonality and stability of physicochemical parameters and macrobenthic community structure in a Rocky Mountain stream. Freshwater Invertebrate Biology 4:160-177.

411