Stream mesocosm experiments show significant differences in

Subscriber access provided by UNIV OF SOUTHERN INDIANA

Ecotoxicology and Human Environmental Health

Stream mesocosm experiments show significant differences in sensitivity of larval and emerging adults to metals Christopher J Kotalik, and William H Clements Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.9b00883 • Publication Date (Web): 11 Jun 2019 Downloaded from http://pubs.acs.org on June 12, 2019

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 31

Environmental Science & Technology

1

Stream mesocosm experiments show significant differences in sensitivity of larval and

2

emerging adults to metals

3 4

Christopher J. Kotalik* and William H. Clements

5

Department of Fish, Wildlife and Conservation Biology,

6

Colorado State University

7

Fort Collins, Colorado 80521, United States

8 9 10

* Corresponding Author e-mail: [email protected]

11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

1 ACS Paragon Plus Environment


26

Abstract

27

Evaluations of aquatic insect responses to contaminants typically use larval life stages to

28

characterize taxa sensitivity, but the effects of contaminants to emerging terrestrial adults has

29

received less attention. We present the results of two stream mesocosm experiments that exposed

30

aquatic insects to mixtures of Cu and Zn. We compared responses of larvae and emerging adults

31

in a single-species experiment with the mayfly Rhithrogena robusta and a benthic community

32

experiment. Results showed that R. robusta larvae and emerging adults were highly tolerant of

33

metals. In the benthic community experiment, larval and emerging adult life stages of the mayfly

34

Baetidae were highly sensitive to metals exposure, with significant alterations in adult sex ratios.

35

In contrast, the emergence of Chironomidae (midge) was unaffected, but larval abundance

36

strongly decreased. Timing of adult emergence was significantly different among treatments and

37

varied among taxa, with emergence stimulation in Chironomidae and delays in emergence in R.

38

robusta and Simuliidae. Our results demonstrate that metal tolerance in aquatic insects is life

39

stage dependent and that taxa sensitivity is influenced by a combination of physiology and

40

phylogeny. Regulatory frameworks would benefit by including test results that account for

41

effects of contaminants on metamorphosis and adult insect emergence for the development of

42

aquatic life standards.

43 44 45 46 47


Page 2 of 31

Page 3 of 31


48

Introduction

49

Evaluations of aquatic insect responses to contaminants predominately use immature life stages

50

(hereafter referred to as “larvae”) to characterize taxa sensitivity, but the morphological

51

transition from larvae to terrestrial adults has received considerably less attention. Most aquatic

52

insects metamorphose to winged-terrestrial adults to reproduce and complete their life cycle.1

53

Emerging adults provide crucial prey subsidies to linked riparian consumers such as birds, bats,

54

and spiders,2,3 and they supply detrital material and nutrients to terrestrial ecosystems.4,5 Adults

55

are also a vector for contaminant transfer from aquatic to terrestrial ecosystems.6-9 Similar to the

56

morphological and physiological development during early life stages of fish and other aquatic

57

vertebrates, invertebrate metamorphosis is a biologically stressful process because of the greater

58

energy requirements and cell differentiation that occur during tissue reorganization.10 Therefore,

59

assessment of insect larval responses alone may not adequately characterize the effects of

60

contaminants on adult emergence.11-14

61

Field and laboratory studies of aquatic insect emergence have shown greater sensitivity of

62

emerging adults to trace metals compared to larvae.12- 13 Energy used by larvae to detoxify metals

63

at sublethal concentrations may decrease the metabolic resources available to successfully

64

complete metamorphosis.15-18 Metal exposure can reduce insect emergence and affect linked

65

terrestrial ecosystems that are dependent on aquatic subsidies.9 However, metal flux to riparian

66

systems by emerging adults is influenced by the strength of the resource linkage (e.g., number of

67

emerging larvae), and the decrease of metal body burden from larvae to adults after

68

metamorphosis.19-22

69

Quantifying the relationship of metal exposure and emergence propensity in the field is

70

complicated by the seasonality of emergence, colonization, and the inability to control for

71

extraneous variables.23 Traditional testing methodology employed in single-species toxicity tests 3 ACS Paragon Plus Environment


72

is also limited in evaluating adult insect emergence. Because organisms commonly used in

73

laboratory toxicity tests (e.g., Daphnia spp., Ceriodaphnia dubia) do not metamorphose into a

74

terrestrial adult life stage, there is no equivalent endpoint in these studies. Full-life cycle tests

75

with the parthenogenetic mayfly Neocloeon triangulifer have shown differences in sensitivity to

76

metals among life stages through metamorphosis, demonstrating the importance of evaluating

77

adult emergence.13-14 Community-level mesocosm experiments provide an opportunity to

78

quantify the relationship between metal exposure and emergence because a diversity of

79

indigenous aquatic insect taxa are exposed through multiple instar sizes and life stages.24

80

To test the effects of metal exposure on larvae and emerging adults, we conducted two

81

stream mesocosm experiments. We compared the life stage responses of a developmentally

82

mature, late-instar mayfly (Ephemeroptera: Heptageniidae) to the responses of three dominant

83

taxa obtained from a naturally colonized stream benthic community. Both experiments exposed

84

organisms to environmentally relevant concentration gradients of dissolved Cu and Zn. The

85

primary goals of this study were to test the following hypotheses: 1) adult insect emergence is

86

correlated with larval responses across a metal-exposure gradient; 2) responses to metals

87

differbetween larval and adult life stages; and 3) exposure to sublethal concentrations of metals

88

alters timing of emergence and sex ratios in adult aquatic insects.

89 90

Materials and Methods

91

Mesocosm Experiments

92

Two mesocosm experiments were conducted at the Stream Research Laboratory (SRL) located at

93

Colorado State University Foothills Campus, Fort Collins, Colorado, USA. One mesocosm

94

experiment used late-instar, developmentally mature Rhithrogena robusta (Heptageniidae:

95

Ephemeroptera) collected from Joe Wright Creek (~2,500 m ASL), Colorado, an undisturbed 4 ACS Paragon Plus Environment

Page 4 of 31

Page 5 of 31


96

2nd-order stream in the Roosevelt National Forest. Individuals were collected from the underside

97

of cobble substrate using a flat-angled, fine-bristled paint brush to minimize organism handling

98

stress. We selected R. robusta that were developmentally mature based on wing-pad

99

development, as well as other morphological features that indicate maturation (e.g., size, color,

100

pigmentation). Individuals were placed in aerated insulated coolers filled with site water and

101

transported to the SRL. Four un-colonized rock trays were placed in each mesocosm to provide

102

habitat, and 32 or 34 individuals were assigned to each replicate mesocosms. To reduce potential

103

density dependent effects, we used relatively low densities of R. robusta compared to the

104

densities of benthic macroinvertebrates typically exposed in the mesocosms.

105

The second mesocosm experiment used naturally colonized stream benthic communities

106

collected from the Arkansas River (AR), Colorado, a high-elevation (~3000 m ASL) 4th-order

107

mountain stream. AR was historically contaminated by acid mine drainage (including Cu and

108

Zn), but remediation in the last 25 years has significantly improved water quality and metal

109

concentrations rarely exceed water quality criteria.25 Communities were collected from the AR

110

using a technique described by Clements et al.26. Trays (10 x 10 x 6 cm) filled with pebble and

111

small cobble were deployed for 35 d and colonized with a diverse benthic assemblage that is

112

similar in abundance and species richness observed on naturally occurring substrate.24 The trays

113

were colonized by a range of life stages of aquatic insects that included early to mid-instar taxa,

114

as well as mature taxa nearing completion of their life cycle as emerged adults. Five trays were

115

randomly collected from the colonization racks, placed in insulated coolers, transported to the

116

SRL, and assigned treatments.

117 118

Dilution water in the SRL is supplied from a deep mesotrophic reservoir. Water quality in the SRL is representative of cold water Rocky Mountain streams with cool temperatures (10-15 5 ACS Paragon Plus Environment


119

ºC), low hardness (30-35 mg/L CaCO3), alkalinity (25-30 mg/L CaCO3), and dissolved organic

120

carbon (2.5-3.0 mg/L), circumneutral pH (7.0-7.8), and low conductivity (60-90 µS/cm).24,27

121

Current in the 18 mesocosms (20 L) was provided by submersible water pumps (EcoPlus®) at a

122

rate of 2233 L per h (i.e., 1.9 hydraulic circulations per min). Each mesocosm was covered with

123

insect netting (220 micron mesh size) to contain emerged adults. To ensure that most emerging

124

insects were removed, daily adult collection was conducted between 17:00 and 19:00 for both

125

experiments. The flow-through mesocosms (76 x 46 x 12 cm) received water at 1.0 L/min, and

126

metal salt solutions were pumped to mesocosms using peristaltic pumps at a rate of 10 mL/min.

127

To prevent emigration by drifting invertebrates, standpipes were covered with fine mesh.

128

Rhithrogena robusta were exposed to a concentration gradient of Cu and Zn at 1:10, with

129

3 replicate mesocosms (6 total treatments including controls) per treatment ranging from 10 to

130

100 µg/L Cu, and 100 to 1000 µg/L Zn. The AR community was exposed to a gradient of Cu and

131

Zn at approximately 1:15, with 3 replicate mesocosms per treatment ranging from 10 to 100

132

µg/L Cu, and 15 to 1500 µg/L Zn. These mass ratios were chosen to represent the Cu and Zn

133

mixtures observed in Colorado mountain streams affected by AMD.28 To confirm dissolved Cu

134

and Zn concentrations, water samples (15 mL) were taken from each mesocosm during the

135

exposure duration (3x for R. robusta; 4x for AR community). Water samples were filtered

136

through a 0.45-µm filter, acidified to a pH < 2.0 using analytical grade nitric acid, and analyzed

137

using flame or furnace atomic absorption spectrophotometry. Cumulative criterion units (CCU),

138

defined as the ratio of the measured dissolved metal concentrations to the United States

139

Environmental Protection Agency (U.S. EPA) hardness-adjusted criterion value and summed for

140

each metal, were used to quantify the metal mixture of Cu and Zn as one value. We chose to use

141

hardness-adjusted criteria because previous experiments have found that the water chemistry of 6 ACS Paragon Plus Environment

Page 6 of 31

Page 7 of 31


142

the dilution water in the SRL is highly consistent. In addition, the U.S. EPA biotic ligand model

143

(BLM) and hardness-based criteria generate very similar values27. Because the BLM requires

144

additional chemical analyses (e.g., DOC) that were unavailable for these experiments, we chose

145

to use the hardness-based criteria to benchmark metal mixture concentrations.

146

Rhithrogena robusta were exposed for 10 d, and surviving larvae that did not emerge

147

were preserved at the end of the experiment. The AR community exposure was 14 d. At the end

148

of the experiment, colonization trays were rinsed through a 350-µm sieve, with organisms and

149

detrital material preserved in 80% ethanol. Benthic community samples were sorted in the

150

laboratory. All larvae were enumerated and identified to the lowest taxonomic resolution

151

possible, typically to genus, except for chironomids that were identified to subfamily or tribe.

152

Emerged adult insects were enumerated and identified to family or subfamily, and sexed. For

153

biomass estimates, larval and adult taxa were dried at 60 ºC for approximately 72 h and then

154

weighed to the nearest 0.00001 g. To test for differences in emerging-adult biomass over time,

155

daily samples were combined for days 1-4, 5-8, 9-13. Daily adult biomass estimates were not

156

possible due to very low biomass in the high-concentration treatments near the end of the

157

experiment. Larval biomass samples were sorted from the benthic community samples that were

158

exposed for 14 d.

159

Statistical Analyses

160

All statistical analyses were conducted using R statistical computing.29 We used a linear

161

model (“LM” function; package ‘car’) to test the responses of R. robusta larvae and emerging

162

adults to metals.30 A linear model was also fit to Rhithrogena spp. larvae abundance in the AR

163

community to compare to R. robusta larval abundance results. For the AR community

164

experiment, linear models were used to test concentration-response relationships between CCU 7 ACS Paragon Plus Environment


165

and life stage abundances, biomass, and adult sex ratios of the three dominant taxa that were

166

present at sufficient densities (i.e., Baetidae, Chironomidae, and Simuliidae). To determine if the

167

concentration-response relationships of larvae and adults were statistically different, an

168

ANCOVA interaction model was used (“LM” function), with CCU as the continuous predictor

169

and life stage as the categorical predictor of either abundance or biomass. Statistical significance

170

of the interaction term indicates that the slopes of larvae and adults are different from one

171

another (i.e., the effect of CCU is dependent on life stage).

172

To test for differences in the timing of adult emergence abundance and biomass, CCU

173

treatments were categorized (6 treatments, 3 replicates each) to support single-factor repeated

174

measures ANOVA analyses. For emergence data, daily observations were successively

175

combined to estimate cumulative emergence, whereby daily emergence abundance was

176

additively summed over the duration of the experiment. We summed the daily results because we

177

wanted to test if the mean difference in the number of emerging individuals over the entire

178

exposure duration differed among treatments. Adult biomass data were analyzed in the three time

179

groups as previously described. A single-factor repeated measures ANOVA (function “lmer,”

180

package ‘lme4”) 31 was fit to the cumulative emergence data to test if the effects of metals on

181

cumulative emergence abundance and biomass varied over the exposure duration. We chose a

182

covariance structure of compound symmetry, which assumes constant correlation between

183

observations on the same sample replicate (i.e., cumulative observations are dependent on the

184

previous observation). Post hoc comparisons with “estimated marginal means of linear trends”

185

function from package ‘emmeans’ were used to compare treatment means of emergence

186

abundance and biomass to control values with Dunnett’s Test.32 We chose to not conduct date-

187

by-date treatment contrasts because we wanted to test for differences in average cumulative 8 ACS Paragon Plus Environment

Page 8 of 31

Page 9 of 31


188

emergence abundance over the 14 d exposure. All data were log transformed to meet the

189

assumptions of parametric statistics and to improve fit for all statistical models (i.e., linear

190

models, ANCOVA, single-factor repeated measures ANOVA). Statistical significance for all

191

models was p < 0.10. The a priori selection of a slightly higher p-value was motivated by the

192

high within-treatment variation observed with naturally colonized communities, relatively low

193

number of replicates (n = 3), and the goal of reducing the risk of Type II errors (i.e., incorrectly

194

determine that there was no effect of metals). Standardized effect sizes were estimated by partial

195

eta-squared values with 90% confidence intervals for all linear and ANCOVA models.33

196

Results

197

Measured concentrations of Cu and Zn were variable (Table S1), but generally approximated

198

target values (Table 1). Measured treatment concentrations of Cu and Zn for the R. robusta

199

experiment ranged from 14-119 µg/L Cu and 48-1068 µg/L Zn (Table 1). In the AR community

200

experiment, concentrations ranged from 10-86 µg/L Cu and 86-1566 µg/L Zn. Water hardness

201

and alkalinity were approximately 30 mg/L CaCO3 (s.d. = 0.52) and 29 mg/L CaCO3 (s.d. =

202

0.89), respectively. The U.S. EPA acute hardness-adjusted criterion values for Cu was 4.3 µg/L

203

and for Zn 42.2 µg/L.34-35 CCU calculations among all mesocosm observations for the R. robusta

204

experiment ranged from 3.6-52.7 CCU, and for the AR community ranged from 3.4-54.5 CCU

205

(Table 1). Temperature, pH, and conductivity were consistent among treatments and between

206

experiments (Table S1).

207 208 209 9 ACS Paragon Plus Environment


210

Table 1. Target exposure concentrations and average measured concentrations (µg/L) (+/- s.d.)

211

of Cu and Zn in the Rhithrogena robusta and the Arkansas River (AR) community exposures.

Low Mid-Low Mid Mid-High High

Rhithrogena robusta Target Measured Target Measured Cu Cu Zn Zn 12.29 49.45 6.3 (2.26) 62.5 (1.52) 18.59 102.58 12.5 (1.78) 125 (14.56) 35.95 178.07 25 (4.71) 250 (14.11) 52.93 378.13 50 500 (3.20) (20.65) 109.85 997.37 100 (13.61) 1000 (112.77)

CCU 4.0 6.7 12.6 21.2 49.1

Page 10 of 31

AR Community Measured Target Measured Target Cu Cu Zn Zn 7.31 95.56 6.3 (2.14) 93.8 (8.69) 13.07 167.03 12.5 (2.10) 187.5 (24.12) 23.74 305.67 25 (2.46) 375 (9.11) 37.33 653.42 50 750 (78.32) (8.42) 78.33 1481.64 100 (6.48) 1500 (73.76)

CCU = the sum of the measured concentration for each metal divided by the U.S. EPA hardnessadjusted criterion values for Cu (4.3 μg/L) and Zn (42.3 μg/L). 212 213

Emergence (Fig. 1) and larval survival (Fig. S1) of R. robusta did not significantly (p-

214

value < 0.10) decrease in any treatment (Table S2) For two suspected outliers with relatively

215

low adult emergence and high larval mortality compared to other treatments, we ran the linear

216

models with and without these points and identified no significant relationships; therefore, we

217

did not remove these data from the analysis. The total number of individuals emerging was

218

greater than 80% for most treatments, and approximately 75% of individuals that did not emerge

219

over the 10 d exposure were alive at the end of the experiment. In contrast, Rhithrogena spp.

220

from the AR community experiment were not developmentally mature to the point of emergence,

221

but Rhithrogena spp. larvae were significantly reduced (Fig. S1).

222 223 10 ACS Paragon Plus Environment

CCU 4.0 7.0 12.7 24.2 53.2

Page 11 of 31


Rhithrogena robusta Timing of Emergence

Total Emergence Cummulative Average Difference in Emergence Compared to Controls

120

Percent Emergence

100 80 60 40 20

F-stat: 0.01 p-value: 0.9378

0 0

224

10

20

30

40

5

0

-5 Control Low (4 CCU) Mid-Low (7 CCU) Mid (13 CCU) Mid-High (21 CCU) High (49 CCU)

-10

-15

F-stat: 3.33 p-value: 0.0067

2

50

4

6

8

Days

CCU

225

Figure 1. Results from the single-species toxicity test conducted with the mayfly Rhithrogena

226

robusta (Ephemeroptera: Heptageniidae). The left panel shows the percentage of total emerging

227

adults after 10 d exposure to Cu and Zn. The right panel shows the cumulative average daily

228

difference in emergence compared to the controls over the exposure duration. CCU is the ratio of

229

measured dissolved metal concentrations to the U.S. EPA hardness-adjusted criterion value and

230

summed for each metal. Asterisks (*) indicate significant differences from controls.

231 232

Although metal exposure had no effect on total emergence of R. robusta, the timing of

233

emergence over the 10 d experiment varied among metal treatments (Fig. 1; Table S3). Average

234

cumulative emergence greater than controls was interpreted as emergence stimulation, while

235

average cumulative emergence less than controls was considered a delay if a significant decrease

236

in treatment abundance was not detected among treatments. Cumulative emergence observed

237

below control values but with significant decreases in overall abundance is the result of larval

238

mortality preventing adult emergence. Because we observed no treatment effect on total 11 ACS Paragon Plus Environment

* *

10


Page 12 of 31

239

emergence of R. robusta, timing of emergence was delayed among metal treatments. Statistically

240

significant post hoc relationships (compared to controls) were observed for the 4- and 13-CCU

241

treatments. The AR benthic community was highly diverse, with control mesocosms averaging

242 243

(±s.e.) 28.3 (±0.8) taxa and 864 (±135) individuals. Number of taxa and abundance of emerging

244

adults in controls averaged 7.3 (±0.3) and 328 (±29), respectively. Three dominant taxa groups

245

(Baetidae, Chironomidae and Simuliidae) were present at sufficiently high larval and adult

246

densities in controls for statistically valid comparisons. Abundance of Baetidae larvae and

247

emerging adults were negatively (p < 0.0001) correlated with exposure concentration (Fig. 2;

248

Table S2). Although abundance of larval Chironomidae was also negatively correlated with

249

metal exposure (p < 0.0001), the number of emerging adults was not. For both Baetidae and

250

Chironomidae, the ANCOVA interaction term was highly significant (p-value < 0.0001),

251

indicating that the response to metals was dependent on life stage (Table S4). For both groups,

252

effects of metals were greater on larval survival than emergence. In contrast to baetids and

253

chironomids, there were no effects of metals on larval survival or emergence of Simuliidae.

254 Chironomidae

Baetidae

Abundance

100

Simuliidae 1000

1000 Larvae: p= < 0.0001 Adult: p= < 0.0001 CCU*Life Stage: p= < 0.0001

1000

Larvae: p= 0.9998 Adult: p= 0.3670 CCU*Life Stage: p= 0.5979

Larvae: p= < 0.0001 Adult: p= 0.1101 CCU*Life Stage: p= < 0.0001

100 100

10

10 1

Larvae Adult 1

10

0.1 0

10

20

30

40

50

0

10

CCU

20

30 CCU

255 12 ACS Paragon Plus Environment

40

50

0

10

20

30 CCU

40

50

Page 13 of 31


256 257

Figure 2. Effects of Cu and Zn on total larval (solid circles) and emerging adult (open circles)

258

abundances of Baetidae (Ephemeroptera), Chironomidae (Diptera), and Simuliidae (Diptera) in

259

the 14 d community exposure. Linear regression for larvae (solid line) and adults (dashed line)

260

was used to determine significant concentration-response relationships for each life stage.

261

ANCOVA model was used to test the hypothesis that the slopes for larvae and adults differed, as

262

indicated by the p-value for the interaction term (CCU*Life Stage).

263 264

In addition to measuring effects of metals on abundance of larval and adult aquatic

265

insects, we assessed effects on biomass, which is a more direct measure of subsidy export to

266

riparian ecosystems. Effects of metals on the biomass of mayflies, chironomids and blackflies

267

were generally similar to those observed for abundance (Fig. 3; Table S2). Across all metal

268

treatments, biomass of adult mayflies, chironomids and simuliids was greater than larval

269

biomass. Ephemeroptera, Chironomidae, and community biomass all displayed significant

270

ANCOVA interaction terms, with greater effects of metals on larval biomass compared to

271

biomass of emerging adults (Table S4). The timing of adult biomass varied among treatments

272

and taxonomic groups (Fig. S2; Table S3). Biomass of adult baetid mayflies was significantly

273

reduced at the 2 highest treatments compared to control observations. In contrast, biomass of

274

adult chironomids increased over time, with no differences among treatments over the three time

275

periods. Maximum biomass of Simuliidae was generally achieved by day 8 for all treatments, but

276

pairwise comparisons showed no treatment differences compared to controls. Biomass of the AR

277

community was significantly different overall, with the 24-CCU treatment significantly different

278

from controls. 13 ACS Paragon Plus Environment


Page 14 of 31

279

Chironomidae

Ephemeroptera

100

100

10

Biomass (Dry Weight (mg))

10 1

0.1

Larvae Adult

Larvae: p= < 0.0001 Adult: p= < 0.0001 CCU*Life Stage: p= 0.0577

0

10

20

1 30

40

Larvae: p= < 0.0001 Adult: p= 0.9220 CCU*Life Stage: p= 0.0016

0

50

Simuliidae

10

20

30

40

50

Community Biomass

100

1000

10 100 1

0.1


0

280

10

20

30

40

50

10


0

CCU

10

20

30

40

50

CCU

281

Figure 3. Effects of Cu and Zn on total larval (solid circles) and emerging adult (open circles)

282

biomass (dry weight (mg)) of Ephemeroptera, Chironomidae, Simuliidae, and total community

283

biomass in the 14 d community exposure. Linear regression for larvae (solid line) and adults

284

(dashed line) was used to determine significant concentration-response relationships for each life

285

stage. ANCOVA model was used to test the hypothesis that the slopes for larvae and adults

286

differed, as indicated by the p-value for the interaction term (CCU*Life Stage). Note that the

287

log-normalized y-axis may affect the visual interpretation of the life stage interaction. 14 ACS Paragon Plus Environment

Page 15 of 31


Cumulative emergence of the 3 dominant groups was altered during the 14 d exposure

288 289

(Fig. 4; Table S3). Relative to controls, abundance of emerging baetids was significantly

290

stimulated at 7-CCU, but decreased in the highest treatments. Abundance of emerging

291

chironomids was stimulated in all treatments from days 6-9, and the lowest metal treatments

292

were statistically different from the controls. Although the timing of chironomid emergence was

293

clearly altered in the 2 highest treatments, because we tested for differences relative to controls

294

over the entire exposure period, these treatments were not statistically different from controls. In

295

contrast to the other groups, we saw no evidence of emergence stimulation by metals for

296

Simuliidae. Emergence of simuliids was consistently less than controls, with emergence in the 7-

297

to 24-CCU treatments significantly decreased. Lastly, the relationship between sex ratio

298

(expressed as female to male adults) and metal concentration was highly significant for Baetidae,

299

with greater abundance of female adults relative to males as metal concentrations increased (Fig.

300

5). Exposure to metals did not alter the sex ratios of Chironomidae and Simuliidae (Fig. S3).

301

Cummulative Average Difference in Emergence Compared to Controls

302

20

20

F-stat: 81.78 p-value: < 0.0001

*

0

* -20

2

4

6

20

F-stat: 5.36 p-value: 0.0001

* *

10

*

*-20

Treatment Day

10

12

0

-40

*-10

8

F-stat: 21.61 p-value: < 0.0001

*-20

0

Control Low (4 CCU) Mid-Low (7 CCU) Mid (13 CCU) Mid-High (24 CCU) High (54 CCU)

-40

Simuliidae

Chironomidae

Baetidae

* *

-60 2

4

6

8

10

12

Treatment Day

2

4

6

8

10

12

Treatment Day

303

Figure 4. Effects of Cu and Zn on the cumulative average daily difference in emergence

304

abundance compared to the controls over the 14 d exposure of Baetidae (Ephemeroptera),

305

Chironomidae (Diptera), and Simuliidae (Diptera) in the community exposure. Single-factor 15 ACS Paragon Plus Environment


Page 16 of 31

306

repeated measures ANOVA was used to test for significant differences in cumulative emergence

307

among treatments over the entire exposure duration. Post-hoc comparisons with Dunnett’s test

308

were used to test for significant relationships between treatment and control responses. Asterisks

309

(*) indicate significant differences from controls.

310

Baetidae

6

Female:Male Adult

5 4 3 2 F-stat: 51.97 p-value: < 0.0001

1 0 0 311

10

20

30

40

50

60

CCU

312

Figure 5. Effects of Cu and Zn on emerging adult sex ratios (female:male) of Baetidae in the

313

stream benthic community experiment. Linear regression was used to determine significant

314

concentration-response relationships for each life stage.

315 316 317 318 16 ACS Paragon Plus Environment

Page 17 of 31


319

Discussion

320

Exposure of aquatic insect communities to metals significantly decreased abundance and

321

biomass of larval and adult aquatic insects. However, effects of metals differed among dominant

322

taxa and life stages. Chironomidae adult abundances were unaffected by the metal treatments,

323

but larval densities were strongly reduced. Differences in metal sensitivity among body sizes and

324

life stages of aquatic insects have been observed with field, 12 mesocosm, 36-37 and laboratory

325

approaches.13-14 One explanation regarding these observed differences in sensitivity is aquatic

326

insect phenology, which influences seasonal rates of development and the life stages that are

327

represented during contaminant exposure. For example, previous field and mesocosm

328

experiments showed that large, developmentally mature individuals of the mayfly Rhithrogena

329

hageni collected in spring were more tolerant to metal exposure compared to summer

330

populations that were dominated by smaller individuals.37 Populations of chironomids in our

331

control mesocosms were dominated by early to mid-instars (Fig. S4), but these individuals were

332

highly reduced from the high metal treatments. In contrast, developmentally mature larvae were

333

highly tolerant and able to complete their life cycle as emerged adults. Similarly,

334

developmentally mature R. robusta from the single-species experiment were tolerant to metals

335

and showed high emergence success, but the abundance of mid-instar Rhithrogena spp. in the

336

community experiment significantly decreased as metal concentration increased (Fig. S1). These

337

differences in sensitivity among life stages may help to explain the discrepancies often reported

338

between laboratory and field responses of aquatic insects to metals.38-39, 27 Laboratory

339

experiments conducted with aquatic insects routinely employ larger, developmentally mature

340

individuals.



341

Baetidae larvae and adults were both decreased by metal exposure, but greater metal

342

effects were observed in larvae. The sensitivity of Baetidae to metals has been documented in

343

laboratory and mesocosm experiments.13, 40-42 The reduction in abundance of emerging adults

344

was likely due to larval mortality that prevented emergence altogether; however, other authors

345

have reported reduced emergence success during metamorphosis when larvae are exposed to

346

sublethal contaminant concentrations.13, 16, 43 Wesner et al.13 reported decreased emergence

347

success of the mayfly N. triangulifer when larvae were exposed to low concentrations of Zn.

348

Because metamorphosis is an energetically demanding process, sublethal exposure to metals may

349

reduce energy available for emergence, resulting in mortality during this life stage transition. We

350

hypothesize that decreased emergence of baetids in our experiment may have been in part due to

351

mortality occurring during metamorphosis. Given that aquatic insects are the dominant

352

organisms in many freshwater ecosystems, and that metamorphosis is required for most aquatic

353

insects to complete their life cycles, more research is needed to understand effects of sublethal

354

exposure to this sensitive life stage.

355

Physiological traits such as rates of uptake, elimination, and detoxification can influence

356

variation in metal bioaccumulation among aquatic insect taxa.44 Diptera may have greater ability

357

to withstand metal exposures compared to mayflies, in part, because of their higher production of

358

metallothionein-like proteins, greater efflux rates, and production of metal granules.39, 45-46

359

Phylogenetic differences in physiology may explain the greater tolerance of adult chironomids

360

compared to baetids, but it does not account for the greater relative sensitivity of larvae of both

361

groups. One hypothesis for this discrepancy may be body size. Buchwalter et al.44 controlled for

362

body size when assessing the influence of phylogeny on sensitivity of aquatic insects to metals.

363

In contrast, our community experiment incorporated a relatively broad size range of baetids and 18 ACS Paragon Plus Environment

Page 18 of 31

Page 19 of 31


364

chironomids, enabling us to make comparisons across developmental sizes. The observed

365

difference in sensitivity among life stages demonstrates the importance of accounting for insect

366

development and strongly suggests that phylogeny and phenology influence metal sensitivity.

367

Rhithrogena robusta in the single-species toxicity test were highly tolerant to metal

368

exposure, with no effects on total emergence or larval survival. In contrast, larvae and adult

369

emergence of the dominant mayfly (Baetidae) in the community experiments were both

370

significantly decreased. One hypothesis to explain this difference in sensitivity is the duration of

371

life-cycle exposure between taxa. R. robusta were exposed for a shorter portion of their total

372

larval life cycle compared to Baetidae because R. robusta is univoltine (i.e., one generation per

373

year) and Baetidae from our community were multivoltine (i.e., multiple generations per year).

374

Therefore, more-sensitive, less-developed instars of Baetidae were exposed, whereas R. robusta

375

were mainly exposed as fully developed late instars. Additionally, R. robusta develop to a much

376

larger overall body size compared to Baetidae, potentially increasing its relative tolerance to

377

metals.36 Metal exposure in the field is spatiotemporally variable40, 47-48 and because aquatic

378

insects exhibit complex life histories,49 our ability to characterize full-life cycle sensitivity with

379

mesocosms is limited by the exposure duration and the phenology of the community. Moreover,

380

our mesocosms are closed systems that do not allow for natural colonization. Open testing

381

systems that allow for population immigration and emigration over longer testing durations (e.g.,

382

exposure of multiple generations across multiple seasons) may better predict exposure outcomes

383

in the field.50

384

Effects of metals on larval and adult biomass were generally similar to those observed for

385

abundance. Biomass of adult Chironomidae and Simuliidae was much greater than larvae, and

386

more than 25% of the total community biomass in control mesocosms was comprised of adults. 19 ACS Paragon Plus Environment


387

The magnitude of emerging adult biomass was unexpected, and represents a significant export of

388

materials and energy to linked aquatic-terrestrial ecosystems. Individual emerging adults have

389

much greater biomass compared to less developed early instars, and they contribute much greater

390

biomass to consumers. Larval mortality in metal contaminated streams reduces secondary

391

production, and it is likely that our mesocosm results underestimate the effects of metals to

392

aquatic biomass in the field.51 Toxicity assessments and field surveys that incorporate emerging

393

adult biomass will more accurately characterize the relationship between contaminant exposure

394

and aquatic subsidy dynamics.

395

Alterations to the timing of adult insect emergence can reflect larval stress due to direct

396

(i.e., aqueous or dietary exposure) and (or) indirect toxicity (i.e., decreased food resources).

397

Aquatic insects may accelerate or delay their rate of development to emergence to decrease their

398

exposure to contaminants or other environmental stressors.52-60 We observed increased rates of

399

emergence in Chironomidae for all treatments from days 6 through 9, and significantly greater

400

cumulative emergence for metal treatments ranging from 4-13 CCU. The pronounced and

401

immediate stimulation of emergence observed in our study suggests that chironomids can

402

increase their rate of development and transition to their adult life stage. Increased rates of

403

development may decrease aquatic exposure duration and increase the probability of successfully

404

emerging and reproducing.61 However, irregular timing of emergence may result in adults

405

encountering unsuitable terrestrial conditions,62 shift the timing of aquatic-terrestrial subsidies

406

that terrestrial consumers are temporally cued to encounter,63-65 and affect adult reproduction.66

407

Sex ratios were significantly altered in Baetidae, with relatively greater abundance of

408

females emerging as metal concentration increased. Female baetids are generally larger than

409

males,67 which may result in relatively greater tolerance to exposure.36 Differences in gonadal 20 ACS Paragon Plus Environment

Page 20 of 31

Page 21 of 31


410

tissues between sexes may also influence rates of metal accumulation, resulting in discrepancies

411

in toxicity.15,20 Wesner et al.22 observed different rates of metal loss between adult male and

412

female mayfly imagos (i.e., sexually mature adults) through metamorphosis, with males retaining

413

greater body burdens of metals. The synchrony of emergence between sexes is necessary for

414

successful reproduction to occur. Adults of some taxa such as Baetidae are short-lived (24-48

415

h),49 and can experience high mortality due to predation.3, 68 Thus, the mismatches in sex ratios

416

observed in our study, coupled with the alterations in timing of emergence, may result in reduced

417

fecundity and population persistence in aquatic ecosystems exposed to sublethal concentrations

418

of metals.

419

Aquatic insects have complex life histories, yet with few exceptions, most invertebrate

420

test species used in aquatic toxicity testing are short-lived and do not metamorphose. Alternative

421

species that can be cultured in the laboratory such as the parthenogenetic mayfly N. triangulifer

422

more accurately reflect the phenology of naturally occurring aquatic insects and should be used

423

more frequently in single-species toxicity testing. Community mesocosm studies are valuable

424

because they evaluate the responses of multiple aquatic insect taxa across a range of instar sizes

425

and life stages under natural conditions, improving the ability to characterize full-life cycle

426

effects. Aquatic insect emergence is a particularly relevant endpoint because it characterizes

427

effects on a sensitive life stage and links aquatic-derived effects to terrestrial ecosystems. Failure

428

to quantify emergence can mischaracterize contaminant effects on aquatic insect population

429

dynamics and aquatic subsidies to terrestrial ecosystems. Regulatory frameworks would benefit

430

by including test results that account for effects of contaminants on metamorphosis and adult

431

insect emergence for the development of aquatic life standards.

432 21 ACS Paragon Plus Environment


433

Supporting Information

434

Table of water quality characteristics in stream mesocosms (Table S1), summary of statistical

435

results of linear models (Table S2), single-factor repeated measures ANOVA models (Table

436

S3), and ANCOVA models (Table S4), and metal responses of R. robusta and Rhithrogena spp.

437

(Figure S1), and cumulative average adult insect emergence biomass (Figure S2), and sex ratios

438

of Chironomidae and Simuliidae (Figure S3), and head capsule size distribution of

439

Chironomidae in control treatments (Figure S4).

440 441

Acknowledgements

442

Funding for this research was provided by the National Institute of Environmental Health

443

Sciences (1R01ES020917-01). We thank Pete Cadmus, Lindsay Bailey, Sam Duggan, Brian

444

Wolff, Dave Rees, Cory Kotalik, Alexandra Mackewich, and numerous work-study students for

445

logistical help and support with this project. Comments by Joseph Meyer, Boris Kondratieff, and

446

three anonymous reviewers on earlier drafts of this manuscript are greatly appreciated.

447 448

References

449

1. Merritt, R. W.; Cummins, K. W.; Berg M.B., An introduction to the aquatic insects of North

450

America. Kendall Hunt: 2004.

451

2. Sabo, J.; Power, M., River–watershed exchange: effects of riverine subsidies on riparian

452

lizards and their terrestrial prey. Ecology 2002, 83, (7), 1860-1869.


Page 22 of 31

Page 23 of 31

453


3. Baxter, C. V.; Fausch, K. D.; Carl Saunders, W., Tangled webs: reciprocal flows of

454

invertebrate prey link streams and riparian zones. Freshwater Biology 2005, 50, (2),

455

201-220.

456

4. Nakano, S.; Murakami, M., Reciprocal subsidies: dynamic interdependence between terrestrial

457

and aquatic food webs. Proceedings of the National Academy of Sciences 2001, 98,

458

(1), 166-170.

459 460 461

5. Gratton, C.; Vander Zanden, M. J., Flux of aquatic insect productivity to land: comparison of lentic and lotic ecosystems. Ecology 2009, 90, (10), 2689-2699. 6. Echols, K. R.; Tillitt, D. E.; Nichols, J. W.; Secord, A. L.; McCarty, J. P., Accumulation of

462

PCB congeners in nestling tree swallows (Tachycineta bicolor) on the Hudson River,

463

New York. Environmental Science & Technology 2004, 38, (23), 6240-6246.

464

7. Walters, D. M.; Fritz, K. M.; Otter, R. R., The dark side of subsidies: adult stream insects

465

export organic contaminants to riparian predators. Ecological Applications 2008, 18,

466

(8), 1835-1841

467

8. Paetzold, A.; Smith, M.; Warren, P. H.; Maltby, L., Environmental impact propagated by

468

cross‐system subsidy: Chronic stream pollution controls riparian spider populations.

469

Ecology 2011, 92, (9), 1711-1716.

470

9. Kraus, J. M.; Schmidt, T. S.; Walters, D. M.; Wanty, R. B.; Zuellig, R. E.; Wolf, R. E.,

471

Cross‐ecosystem impacts of stream pollution reduce resource and contaminant flux to

472

riparian food webs. Ecological Applications 2014a, 24, (2), 235-243.

473

10. Campero, M.; De Block, M.; Ollevier, F.; Stoks, R., Metamorphosis offsets the link between

474

larval stress, adult asymmetry and individual quality. Functional Ecology 2008, 22,

475

(2), 271-277.



476

11. Pascoe, D.; Williams, K. A.; Green, D. W., Chronic toxicity of cadmium to Chironomus

477

riparius Meigen-effects upon larval development and adult emergence.

478

Hydrobiologia 1989, 175, (2), 109-115.

479

12. Schmidt, T. S.; Kraus, J. M.; Walters, D. M.; Wanty, R. B., Emergence flux declines

480

disproportionately to larval density along a stream metals gradient. Environmental

481

Science & Technology 2013, 47, (15), 8784-8792.

482

13. Wesner, J. S.; Kraus, J. M.; Schmidt, T. S.; Walters, D. M.; Clements, W. H., Metamorphosis

483

enhances the effects of metal exposure on the mayfly, Centroptilum triangulifer.

484

Environmental Science & Technology 2014, 48, (17), 10415-10422.

485

14. Soucek, D. J.; Dickinson, A., Full‐life chronic toxicity of sodium salts to the mayfly

486

Neocloeon triangulifer in tests with laboratory cultured food. Environmental

487

Toxicology and Chemistry 2015, 34, (9), 2126-2137.

488 489 490

15. Hare, L., Aquatic insects and trace metals: bioavailability, bioaccumulation, and toxicity. Critical Reviews in Toxicology 1992, 22, (5-6), 327-369. 16. Liber, K.; Schmude, K. L.; Corry, T. D., Effects of the insect growth regulator diflubenzuron

491

on insect emergence within littoral enclosures. Environmental Entomology 1996, 25,

492

(1), 17-24.

493

17. Sibley, P. K.; Benoit, D. A.; Ankley, G. T., The significance of growth in Chironomus

494

tentans sediment toxicity tests: relationship to reproduction and demographic

495

endpoints. Environmental Toxicology and Chemistry 1997, 16, (2), 336-345.

496 497

18. Rainbow, P. S., Trace metal bioaccumulation: models, metabolic availability and toxicity. Environment International 2007, 33, (4), 576-582.


Page 24 of 31

Page 25 of 31

498 499 500


19. Timmermans, K. R.; Walker, P. A., The fate of trace metals during the metamorphosis of chironomids (Diptera, Chironomidae). Environmental Pollution 1989, 62, (1), 73-85. 20. Kim, K.; Funk, D.; Buchwalter, D., Dietary (periphyton) and aqueous Zn bioaccumulation

501

dynamics in the mayfly Centroptilum triangulifer. Ecotoxicology 2012, 21, (8), 2288-

502

2296.

503

21. Kraus, J. M.; Walters, D. M.; Wesner, J. S.; Stricker, C. A.; Schmidt, T. S.; Zuellig, R. E.,

504

Metamorphosis alters contaminants and chemical tracers in insects: implications for

505

food webs. Environmental Science & Technology 2014b, 48, (18), 10957-10965.

506

22. Wesner, J. S.; Walters, D. M.; Schmidt, T. S.; Kraus, J. M.; Stricker, C. A.; Clements, W. H.;

507

Wolf, R. E., Metamorphosis affects metal concentrations and isotopic signatures in a

508

mayfly (Baetis tricaudatus): Implications for the aquatic-terrestrial transfer of metals.

509


510

23. Buchwalter, D. B.; Clements, W. H.; Luoma, S. N., Modernizing water quality criteria in the

511

United States: A need to expand the definition of acceptable data. Environmental

512

Toxicology and Chemistry 2017, 36, (2), 285-291.

513

24. Clements, W. H., Small-scale experiments support causal relationships between metal

514

contamination and macroinvertebrate community responses. Ecological Applications

515

2004, 14, (3), 954-967.

516

25. Clements, W. H.; Vieira, N. K.; Church, S. E., Quantifying restoration success and recovery

517

in a metal‐polluted stream: a 17‐year assessment of physicochemical and biological

518

responses. Journal of Applied Ecology 2010, 47, (4), 899-910.



519

26. Clements, W. H.; Van Hassel, J. H.; Cherry, D. S.; Cairns, J., Colonization, variability, and

520

the use of substratum-filled trays for biomonitoring benthic communities.

521

Hydrobiologia 1989, 173, (1), 45-53.

522

27. Clements, W. H.; Cadmus, P.; Brinkman, S. F., Responses of aquatic insects to Cu and Zn in

523

stream microcosms: understanding differences between single species tests and field

524

responses. Environmental Science & Technology 2013, 47, (13), 7506-7513.

525

28. Schmidt, T. S.; Clements, W. H.; Mitchell, K. A.; Church, S. E.; Wanty, R. B.; Fey, D. L.;

526

Verplanck, P. L.; San Juan, C. A., Development of a new toxic‐unit model for the

527

bioassessment of metals in streams. Environmental Toxicology and Chemistry 2010,

528

29, (11), 2432-2442.

529

29. Team, R. C., R: A language and environment for statistical computing. 2013.

530

30. Fox, J.; Weisberg, S.; Adler, D.; Bates, D.; Baud-Bovy, G.; Ellison, S.; Firth, D.; Friendly,

531

M.; Gorjanc, G.; Graves, S., Package ‘car’. Vienna: R Foundation for Statistical

532

Computing 2012.

533 534 535 536 537 538 539 540

31. Bates, D.; Mächler, M.; Bolker, B.; Walker, S., Fitting linear mixed-effects models using lme4. arXiv preprint arXiv:1406.5823 2014. 32. Lenth, R., Emmeans: Estimated marginal means, aka least-squares means. R Package Version 2018, 1, (2). 33. Cohen, J., Eta-squared and partial eta-squared in fixed factor ANOVA designs. Educational and Psychological Measurement 1973, 33, (1), 107-112. 34. U.S. EPA. Ambient aquatic life criteria for copper, EPA 440‐5‐84‐031: U.S. Environmental Protection Agency, Office of Water Regulations and Standards: 1985


Page 26 of 31

Page 27 of 31

541 542 543


35. U.S. EPA. National Recommended Water Quality Criteria, EPA- 822-R-02-047; U.S. Environmental Protection Agency, Office of Water: Washington, DC, 2002. 36. Kiffney, P. M.; Clements, W. H., Size‐dependent response of macroinvertebrates to metals in

544

experimental streams. Environmental Toxicology and Chemistry 1996, 15, (8), 1352-

545

1356.

546

37. Clark, J. L.; Clements, W. H., The use of in situ and stream microcosm experiments to assess

547

population‐and community‐level responses to metals. Environmental Toxicology and

548

Chemistry 2006, 25, (9), 2306-2312.

549

38. Buchwalter, D. B.; Cain, D. J.; Clements, W. H.; Luoma, S. N., Using biodynamic models to

550

reconcile differences between laboratory toxicity tests and field biomonitoring with

551

aquatic insects. Environmental Science & Technology 2007, 41, (13), 4821-4828.

552

39. Brix, K. V.; DeForest, D. K.; Adams, W. J., The sensitivity of aquatic insects to divalent

553

metals: A comparative analysis of laboratory and field data. Science of the Total

554

Environment 2011, 409, (20), 4187-4197.

555

40. Clements, W. H., Benthic invertebrate community responses to heavy metals in the Upper

556

Arkansas River Basin, Colorado. Journal of the North American Benthological

557

Society 1994, 13, (1), 30-44.

558

41. Fialkowski, W.; Klonowska-Olejnik, M.; Smith, B. D.; Rainbow, P. S., Mayfly larvae (Baetis

559

rhodani and B. vernus) as biomonitors of trace metal pollution in streams of a

560

catchment draining a zinc and lead mining area of Upper Silesia, Poland.

561

Environmental Pollution 2003, 121, (2), 253-267.



562

42. Mebane, C. A.; Schmidt, T. S.; Balistrieri, L. S., Larval aquatic insect responses to cadmium

563

and zinc in experimental streams. Environmental Toxicology and Chemistry 2017, 36,

564

(3), 749-762.

565

43. Brock, T.; Roessink, I.; Belgers, J. D. M.; Bransen, F.; Maund, S. J., Impact of a benzoyl

566

urea insecticide on aquatic macroinvertebrates in ditch mesocosms with and without

567

non‐sprayed sections. Environmental Toxicology and Chemistry 2009, 28, (10), 2191-

568

2205.

569

44. Buchwalter, D. B.; Cain, D. J.; Martin, C. A.; Xie, L.; Luoma, S. N.; Garland, T., Aquatic

570

insect ecophysiological traits reveal phylogenetically based differences in dissolved

571

cadmium susceptibility. Proceedings of the National Academy of Sciences 2008, 105,

572

(24), 8321-8326.

573

45. Postma, J. F.; Davids, C., Tolerance induction and life cycle changes in cadmium-exposed

574

Chironomus riparius (Diptera) during consecutive generations. Ecotoxicology &

575

Environmental Safety 1995, 30, (2), 195-202.

576

46. Buchwalter, D. B.; Luoma, S. N., Differences in dissolved cadmium and zinc uptake among

577

stream insects: mechanistic explanations. Environmental Science & Technology 2005,

578

39, (2), 498-504.

579

47. Mebane, C. A.; Eakins, R. J.; Fraser, B. G.; Adams, W. J., Recovery of a mining-damaged

580

stream ecosystem. Elemental Science of the Anthropocene 2015, 3: 000042

581

48. Herbst, D. B.; Medhurst, R. B.; Black, N. J., Long‐term effects and recovery of streams from

582

acid mine drainage and evaluation of toxic metal threshold ranges for

583

macroinvertebrate community reassembly. Environmental Toxicology and Chemistry

584

2018, 37, (10), 2575-2592.


Page 28 of 31

Page 29 of 31

585 586 587


49. Brittain, J. E., Life history strategies in Ephemeroptera and Plecoptera. In Mayflies and stoneflies: Life histories and biology, Springer, 1990; pp 1-12. 50. Magbanua, F. S.; Townsend, C. R.; Hageman, K. J.; Piggott, J. J.; Matthaei, C. D., Individual

588

and combined effects of fine sediment and glyphosate herbicide on invertebrate drift

589

and insect emergence: a stream mesocosm experiment. Freshwater Science 2016, 35,

590

(1), 139-151.

591

51. Iwasaki, Y.; Schmidt, T. S.; Clements, W. H., Quantifying differences in responses of aquatic

592

insects to trace metal exposure in field studies and short-term stream mesocosm

593

experiments. Environmental Science & Technology 2018, 52, (7), 4378-4384.

594

52. Nordlie, K. J.; Arthur, J. W., Effect of elevated water temperature on insect emergence in

595

outdoor experimental channels. Environmental Pollution Series A, Ecological and

596

Biological 1981, 25, (1), 53-65.

597

53. Connolly, N.; Crossland, M.; Pearson, R., Effect of low dissolved oxygen on survival,

598

emergence, and drift of tropical stream macroinvertebrates. Journal of the North

599

American Benthological Society 2004, 23, (2), 251-270.

600

54. Pettigrove, V.; Hoffmann, A., Effects of long‐chain hydrocarbon‐polluted sediment on

601

freshwater macroinvertebrates. Environmental Toxicology and Chemistry 2005, 24,

602

(10), 2500-2508.

603

55. Greig, H. S.; Kratina, P.; Thompson, P. L.; Palen, W. J.; Richardson, J. S.; Shurin, J. B.,

604

Warming, eutrophication, and predator loss amplify subsidies between aquatic and

605

terrestrial ecosystems. Global Change Biology 2012, 18, (2), 504-514.



606

56. Piggott, J. J.; Salis, R. K.; Lear, G.; Townsend, C. R.; Matthaei, C. D., Climate warming and

607

agricultural stressors interact to determine stream periphyton community

608

composition. Global Change Biology 2015, 21, (1), 206-222.

609

57. Lee, S. S.; Paspalof, A. M.; Snow, D. D.; Richmond, E. K.; Rosi-Marshall, E. J.; Kelly, J. J.,

610

Occurrence and potential biological effects of amphetamine on stream communities.

611


612

58. Richmond, E. K.; Rosi-Marshall, E. J.; Lee, S. S.; Thompson, R. M.; Grace, M. R.,

613

Antidepressants in stream ecosystems: influence of selective serotonin reuptake

614

inhibitors (SSRIs) on algal production and insect emergence. Freshwater Science

615

2016, 35, (3), 845-855.

616

59. Rogers, H. A.; Schmidt, T. S.; Dabney, B. L.; Hladik, M. L.; Mahler, B. J.; Van Metre, P. C.,

617

Bifenthrin causes trophic cascade and altered insect emergence in mesocosms:

618

Implications for small streams. Environmental Science & Technology 2016, 50, (21),

619

11974-11983.

620

60. Joachim, S.; Roussel, H.; Bonzom, J.-M.; Thybaud, E.; Mebane, C. A.; Van den Brink, P.;

621

Gauthier, L., A long-term copper exposure in a freshwater ecosystem using lotic

622

mesocosms: Invertebrate community responses. Environmental Toxicology and

623

Chemistry 2017, 36, (10), 2698-2714.

624 625 626 627

61. Sibly, R.; Calow, P., A life-cycle theory of responses to stress. Biological Journal of the Linnean Society 1989, 37, (1-2), 101-116. 62. Nebeker, A. V., Effect of high winter water temperatures on adult emergence of aquatic insects. Water Research 1971, 5, (9), 777-783.


Page 30 of 31

Page 31 of 31

628


63. Takimoto, G.; Iwata, T.; Murakami, M., Seasonal subsidy stabilizes food web dynamics:

629

balance in a heterogeneous landscape. Ecological Research 2002, 17, (4), 433-439.

630

64. Yang, L. H.; Edwards, K. F.; Byrnes, J. E.; Bastow, J. L.; Wright, A. N.; Spence, K. O., A

631

meta‐analysis of resource pulse–consumer interactions. Ecological Monographs

632

2010, 80, (1), 125-151.

633 634 635

65. Hastings, A., Temporally varying resources amplify the importance of resource input in ecological populations. Biology Letters 2012, rsbl20120669. 66. Sardiña, P.; Beardall, J.; Beringer, J.; Grace, M.; Thompson, R. M., Consequences of altered

636

temperature regimes for emerging freshwater invertebrates. Aquatic Sciences 2017,

637

79, (2), 265-276.

638 639 640 641

67. Caudill, C. C.; Peckarsky, B. L., Lack of appropriate behavioral or developmental responses by mayfly larvae to trout predators. Ecology 2003, 84, (8), 2133-2144. 68. Jackson, J. K.; Fisher, S. G., Secondary production, emergence, and export of aquatic insects of a Sonoran Desert stream. Ecology 1986, 67, (3), 629-638.

642 643

TOC Abstract Art

644 645


Stream mesocosm experiments show significant differences in

Recommend Documents