Nitrous Oxide and Dinitrogen: The Missing Flux in ... - ACS Publications

Subscriber access provided by UB + Fachbibliothek Chemie | (FU-Bibliothekssystem)

Article

Nitrous Oxide and Dinitrogen: The Missing Flux in Nitrogen Budgets of Forested Catchments? Eric M. Enanga, Nora J. Casson, Tarrah A. Fairweather, and Irena F. Creed Environ. Sci. Technol., Just Accepted Manuscript • Publication Date (Web): 03 May 2017 Downloaded from http://pubs.acs.org on May 4, 2017

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 free 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 accessible to all readers and 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.

Environmental Science & Technology 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 28

Environmental Science & Technology

1

Nitrous Oxide and Dinitrogen:

2

The Missing Flux in Nitrogen Budgets of Forested Catchments?

3 4

Eric M. Enanga1, Nora J. Casson1,2, Tarrah A. Fairweather1, Irena F. Creed1*

5 6

1

7

N6A 5B7

8

2

9

Portage Avenue, Winnipeg, Manitoba, Canada, R3B 2E9

10

Department of Biology, Western University, 1151 Richmond Street, London, Ontario, Canada,

Current Affiliation: Nora Casson, University of Winnipeg, Department of Geography, 515

*Corresponding Author (telephone 1-519-661-4265; fax 1-519-661-3935)

11 12

Emails: [email protected], [email protected], [email protected], [email protected]

13 14

Abstract

15

Most forest nitrogen budgets are imbalanced, with nitrogen inputs exceeding nitrogen outputs.

16

The denitrification products nitrous oxide (N2O) and dinitrogen (N2) represent often-unmeasured

17

fluxes that may close the gap between explained nitrogen inputs and outputs. Gaseous N2O and

18

N2 effluxes, dissolved N2O flux, and traditionally measured dissolved nitrogen species (i.e.,

19

nitrate, ammonium, and dissolved organic nitrogen) were estimated to account for the annual

20

nitrogen output along hillslope gradients from two catchments in a temperate forest. Adding N2O

21

and N2 effluxes to catchment nitrogen output not only reduced the discrepancy between nitrogen

22

inputs and outputs (9.9 kg ha-1 yr-1 and 6.5 or 6.3 kg ha-1 yr-1, respectively), but also between

23

nitrogen outputs from two catchments with different topographies (6.5 kg ha-1 yr-1 for the

ACS Paragon Plus Environment

1


24

catchment with a large wetland, 6.3 kg ha-1 yr-1 for the catchment with a very small wetland).

25

Dissolved N2O comprised a very small portion of the annual nitrogen outputs. Nitrogen inputs

26

exceeded nitrogen outputs throughout the year except during spring runoff, and also during

27

autumn storms in the catchment with the large wetland. Failing to account for denitrification

28

products, especially during summer rainfall events, may lead to underestimation of annual

29

nitrogen losses.

Page 2 of 28

30 31

Keywords: Temperate, Forest, Topography, Hydrology, Biogeochemistry, Nitrogen, Budget

32

33 34

Figure for Abstract


2

Page 3 of 28

35 36


Introduction The global nitrogen (N) cycle has undergone great alterations during the past century that

37

have resulted in an increase in global reactive N.1-3 Atmospherically deposited reactive N is

38

currently estimated at 280 Tg of N yr-1,4. Several potential fates await this atmospherically

39

deposited N in forest ecosystems (Figure 1): (1) it may be assimilated by organisms

40

(microorganisms and higher); (2) it may undergo transformations that include nitrification and

41

denitrification, with nitrous oxide (N2O) as an obligate intermediate and dinitrogen (N2) as the

42

final product;5,6 or (3) it may be lost to aquatic systems in hydrologic flows. The potential fate of

43

atmospherically deposited N is important because of its potential to become an ecosystem

44

disservice or service.7 For example, on the one hand, denitrification of reactive N to N2O is

45

linked to global warming8 and ozone depletion9 (an ecosystem disservice), but on the other hand,

46

it is linked to removal of dissolved nitrate (NO3-) from water exiting catchments, thereby

47

shielding receiving aquatic ecosystems from the effects of NO3- loading10 (an ecosystem

48

service).

49

A predictive understanding of catchment N transformations at different spatial and

50

temporal scales remains a major scientific challenge.2,11-13 Partitioning N within ecosystems into

51

its different species based on inputs and outputs is an essential first step towards quantifying the

52

different forms of N in the system.14 However, our ability to account for the fate of N in

53

catchments is incomplete.2,13 There remains a portion of atmospherically deposited N that cannot

54

be accounted for that is often referred to as the “missing N” in catchment N budgets.2,15

55

Catchment N budgets can be improved by considering topography, which can play an

56

important role in dissolved N output to streams.16 Catchments with large, flat, and intermittently

57

wet topographic features generated less NO3- and ammonium (NH4+) output, but more dissolved


3


Page 4 of 28

58

organic nitrogen (DON) output than catchments with little or no wetlands.17 However, gaseous N

59

output to the atmosphere provides another important pathway through which N can leave the

60

catchment.11,18 There have been relatively few catchment-scale determinations of N gas outputs

61

in forests as N gas effluxes are not routinely integrated into forest N budgets. Recent studies

62

have observed that rain induced bursts of N2O and N2 efflux during the snow-free period19 and

63

N2O efflux beneath a snow pack20 are significant sources of N gas in forests. Not considering

64

this pathway may contribute to the failure to close N input-output budgets in catchments.15

65

The central premise of this paper is that improved estimates of soil N2O and N2 efflux

66

from forested landscapes require consideration of topographic controls on N cycling and routing

67

processes. Topography influences N transformations by controlling soil temperature, moisture,

68

and redox conditions, and regulates dissolved organic carbon and NO3- availability while helping

69

create redox conditions required for N transformations including denitrification and nitrification

70

to occur. We test the following hypotheses: (1) the major difference in N output among

71

catchments is due to variations in the flux of gaseous N (N2O + N2); and (2) factoring gaseous

72

and dissolved N outputs narrows the gap in catchment input-output N budgets. We test these

73

hypotheses in catchments located in a temperate forest, and expect that the findings will help

74

assess the ability of temperate forests to process the reactive loads of N, understand the relative

75

importance of dissolved vs. gaseous effluxes from catchments N budgets, and close the N input

76

vs. output N budget of catchments.

77


4

Page 5 of 28


78

Materials and Methods

79

Study Area. The Turkey Lakes Watershed is a 10.5 km2 watershed located near the eastern

80

shore of Lake Superior in the Algoma Highlands of Central Ontario (47°03ʹ00ʺN, 84°25ʹ00ʺW),

81

60 km north of Sault Ste. Marie in the Great Lakes-St. Lawrence forest region (Figure 2).

82

The watershed receives an average annual precipitation of 1189 mm with an average

83

annual temperature of 4.6°C (1981-2010). The bedrock of the watershed is greenstone with small

84

outcrops of granite.21 A thin discontinuous till of varying depth overlays the bedrock with depth

85

ranging from < 1 m at high elevations, 1-2 m at lower elevations, and occasionally up to 65 m in

86

depressions. Dispersed pockets of Ferric Humisols are found in bedrock-controlled depressions

87

and adjacent to streams and lakes, and Orthic Ferro-Humic and Humo-Ferric podzolic soils are

88

dominant.22 The watershed contains uneven-aged, mature- to over-mature, old-growth forest.

89

The temperate forest is dominated (90%) by sugar maple (Acer saccharum Marsh.), with yellow

90

birch (Betula alleghaniensis Britton), white pine (Pinus strobus L.), white spruce (Picea glauca

91

Moench Voss.), ironwood (Ostrya virginiana (Mill.) K. Koch), and red oak (Quercus rubra L.)

92

in the upland areas. Sugar maple also dominates in the wetland areas where it is mixed with

93

eastern white cedar (Thuja occidentalis L.), black ash (Fraxinus nigra Marsh.), and balsam fir

94

(Abies balsamea (L.) Mill.).23

95

Watershed monitoring was established in 1980 to study the effects of atmospheric

96

deposition on both terrestrial and aquatic ecosystems.24,25 While there has been a significant

97

decline in sulphur deposition since the 1980s, N deposition has remained relatively stable up to

98

and including the period of this study.26 Catchment 38 (C38) and catchment 35 (C35) are two

99

north-facing catchments in the Turkey Lakes Watershed in close proximity (~1 km apart) that

100

receive similar N inputs. Topography is the main difference between the two catchments: a large


5


101

wetland (1.58 ha) covers 25% of the 6.33 ha area of the more gently sloped C38 (13.5° average

102

slope), while a very small wetland (0.03 ha) covers < 1% of the 3.12 ha area of the relatively

103

steeper C35 (19.1° average slope) (Figure 2). This work builds on previous studies that reported

104

differences in dissolved N output among catchments in the forested landscape of the Great

105

Lakes-St. Lawrence forest region at the northern edge of the temperate forest biome of North

106

America.16,27-29

Page 6 of 28

107 108

Experimental Design. Total (wet and dry) N inputs were collected at the Canadian Air and

109

Precipitation Monitoring Network (CAPMoN) site located closest to the catchments. Total

110

(gaseous and dissolved) N outputs were sampled in C35 and C38 from 2005 to 2010 water years

111

(June 1 to May 30). In c38, soil N2O efflux measurements were collected along hillslope

112

gradients of inner wetland (IW), outer wetland (OW), lowland (LOW), and upland (UP)

113

positions derived from a LiDAR-based 5 m digital elevation model using digital terrain analysis

114

methods.30 Soil N2O efflux measurements were collected between 64 and 72 times per position

115

during the snow-free season (June to September) from 2006 to 2010 using ground-based static

116

chambers.19 Soil N2O efflux measurements were collected between 36 to 40 times per position

117

during the snow season (October to May) in 2006/2007 from sampling ports attached at different

118

depths along 1.25 m PVC tubes inserted into the snowpack.20 Daily soil N2 efflux was estimated

119

using the acetylene-inhibition technique in the IW and OW positions for one day during the

120

snow-free season of 2010. Dissolved NH4+, NO3-, and DON concentrations were measured from

121

water samples collected at weirs at the outlets of both C38 and C35 in the 2005 to 2010 water

122

years, and dissolved N2O concentrations were measured from water samples collected at weirs at

123

the outlets of both catchments in the 2006 water year. The median and 25th and 75th percentiles


6

Page 7 of 28


124

of estimates of N2O and N2 in the snow and snow-free seasons in C38 are given in Table 1.

125

Dissolved NH4+, NO3-, DON, and N2O fluxes were calculated by multiplying concentrations by

126

stream discharge. Sampling and analytical methods have been described fully in other

127

publications19,20,31 and are summarized in the Supporting Information.

128 129

Data Analyses. Daily outputs estimated from flux measurements taken at irregular intervals

130

were summed to generate monthly and annual N outputs for the 2005 to 2010 water years. Daily

131

soil N2O effluxes were measured in the IW, OW, LOW, and UP positions in C38 at daily to

132

monthly intervals during the snow-free (June to September) season and at daily to bimonthly

133

intervals during snow season (October to May) in the 2005 to 2010 water years. Dissolved

134

stream NO3-, NH4+, and DON fluxes were measured in C35 and C38 at daily to biweekly

135

intervals in the 2005 to 2010 water years.

136 137 138

Several methods were applied to derive daily soil and dissolved N flux estimates on all days for the 2005 to 2010 water years from measured values. Regressions using daily precipitation as the independent variable were used to model

139

daily soil N2O effluxes for rain days during the snow-free season [i.e., when (1) effective

140

precipitation (same day plus previous day, in order to capture rainfall that may have fallen during

141

the night) exceeded 3 mm32, and (2) when the water table depth was less than 10 mm] at the IW

142

and OW positions. Separate models were developed for each of the IW and OW positions based

143

on 14 rain day observations of N2O efflux from each position.19

144

No significant relationships were observed between environmental drivers (temperature

145

or precipitation) and soil N2O effluxes at the IW or OW positions on non-rain days or at the

146

LOW or UP positions. Soil N2O effluxes at the IW and OW positions on non-rain days in the


7


Page 8 of 28

147

snow-free season were infilled on days with no measurements using median values from the

148

distributions of measured snow-free season N2O effluxes on non-rain days at each position. Soil

149

N2O effluxes at the LOW and UP positions on both rain and non-rain days in the snow-free

150

season were infilled on days with no measurements using median values from the distributions of

151

measured snow-free season N2O effluxes on both rain and non-rain days at each position. Soil

152

N2O effluxes at all positions in the snow season were infilled on days with no measurements by

153

using median values from the distributions of measured snow season N2O effluxes at each

154

position.

155

Daily N2 effluxes at the IW and OW positions on rain days in the snow-free season were

156

calculated using measured snow-free season rain day N2:N2O ratios at each position which were

157

then multiplied by the daily estimated N2O efflux (if estimated N2O efflux was negative, N2

158

efflux was given as zero). N2 efflux was not estimated during non-rain days in the IW and OW or

159

in the UP and LOW as minimal denitrification was assumed under the dry conditions that these

160

days and positions represent.19

161

Daily dissolved N2O fluxes were estimated using median values derived from the

162

distribution of measured dissolved N2O concentrations multiplied by daily discharge. Missing

163

dissolved NH4+, NO3-, and DON were infilled using linear interpolation. Daily soil N2O and N2

164

efflux estimates were multiplied by the respective areas of the IW, OW, LOW, and UP positions

165

in each catchment (Figure 2) and summed to estimate daily catchment N2O and N2 output

166

estimates. Monthly and annual soil N2O and N2 outputs were calculated by summing daily output

167

estimates. Monthly and annual total N2O outputs were calculated by summing annual soil and

168

dissolved N2O output estimates. Monthly and annual dissolved N outputs were calculated by

169

summing the measurement-based fluxes for dissolved inorganic nitrogen (DIN, NO3- + NH4+)


8

Page 9 of 28


170

and DON for each year. Differences in measured daily N2O and total gaseous N (N2O + N2)

171

effluxes at different positions were assessed using ANOVAs on ranks with Post-Hoc Dunn’s

172

tests. Data analyses were performed using SigmaPlot 12 (Systat Software, San Jose, CA).

173 174 175

Results Atmospheric N inputs were much greater than N outputs, particularly when only DIN and

176

DON output sources were considered (Figure 3). Dissolved N outputs were smaller in C38 than

177

in C35, primarily due to less DIN output from C38 (Figure 3). Dissolved N2O export in the

178

stream was extremely small during both the snow-free and snow seasons and therefore not

179

considered further (Figure 4), and the differences in dissolved N outputs were maintained when

180

gaseous N2O + N2 efflux was included for the IW and OW positions (Figure 4, bottom).

181

Average monthly sums of daily efflux estimates revealed peaks in DIN and DON outputs

182

in spring and, to a lesser extent, autumn (Figure 5). These peaks coincided with periods of high

183

catchment discharge during spring snowmelt and autumn storms. In contrast, peaks in N2O and

184

N2 outputs were revealed in the summer (Figure 5). Catchment 38 generated more dissolved

185

organic N and gaseous N2O + N2 output, whereas C35 generated more inorganic N output,

186

dominated by NO3-.

187

Total soil N2O efflux during the snow-free season was similar for the two catchments

188

(0.8 kg N ha-1 yr-1 in C38 vs. 0.6 kg N ha-1 yr-1 in C35), while N2 efflux estimates were much

189

greater in C38 (1.5 kg N ha-1 yr-1) than in C35 (0.06 kg N ha-1 yr-1). Similarly, total soil N2O

190

efflux during the snow season was similar for the two catchments (0.3 kg N ha-1 yr-1 in C38 and

191

C35), whereas total soil N2 efflux was greater in C38 (1.0 kg N ha-1 yr-1) than in C35 (0.04 kg N

192

ha-1 yr-1). There was considerably more output of DON and DIN in the snow season compared to


9


193

the snow-free season. Furthermore, there was considerably more output of DON, less output of

194

NO3-, but similar output of NH4+ in C38 compared to C35.

Page 10 of 28

195

The average annual N output from the two catchments was similar when gaseous N2 and

196

N2O were added to the dissolved N output (Figure 3). The average annual N input was 9.9 kg N

197

ha-1 yr-1, and the average annual N output was 6.5 kg N ha-1 yr-1 from C38 (66% of N input) vs.

198

6.3 kg ha-1 yr-1 from C35 (64% of N input). The composition of the average annual N output

199

differed, with N2O + N2 comprising 55% of N output from C38, but only 16% from C35. The

200

NO3-, NH4+, and DON comprised the remaining 45% of N output from C38 and 84% of N output

201

from C35.

202 203

Discussion

204

Estimates of catchment N inputs vs. N outputs are important because they provide

205

insights on the functioning of a forest ecosystem. This is particularly important when considering

206

the ecosystem effects of atmospheric N deposition that has been elevated since the industrial

207

revolution12,34 and the potential fate of this elevated atmospheric N deposition on forests35 and

208

associated downstream waters.7,36,37 Yet most studies have reported catchment N inputs that have

209

often far exceeded N outputs, leaving in question the fate of the “missing” N in these catchment

210

N budgets.

211

This paper examined the role of gaseous N2O and N2 effluxes in N budget estimates in a

212

temperate forest. Nitrogen output was compared in two catchments, one with 25% wetland (C38)

213

and one with < 1% wetland (C35). When considering DIN and DON output only, 2.9 kg N ha-1

214

yr-1 was exported from C38, whereas 5.3 kg N ha-1 yr-1 was exported from C35. This represented

215

only 30% (C38) and 54% (C35) of the 9.9 kg N ha-1 yr-1 that was atmospherically deposited onto


10

Page 11 of 28


216

the catchment. However, when gaseous N2O and N2 effluxes were incorporated in the N budgets,

217

the N outputs were similar; this represents a reduction of 90% of the difference between the two

218

catchments. Although the difference in N output between the catchments was diminished, the

219

proportion of different N species that comprised these N budgets was different due to the

220

different area proportions of wetlands, suggesting differences in the predominance of N

221

transformation processes within the catchments.

222 223

The Missing Nitrogen of Nitrogen Budgets. Incorporating gaseous and dissolved N species had

224

two effects on the N budgets: (1) it closed the gap in N output between C38 and C35 as indicated

225

above; and (2) it narrowed the gap between N input and N output in both catchments.

226

Gaseous N2O efflux contributed 17% in C38 and 14% in C35 to the total N output, and

227

gaseous N2 efflux contributed an additional 38% in C38 and 2% in C35 to the total N output. The

228

combined gaseous N2O + N2 output represented 36% in C38 and 10% in C35 of the total

229

atmospherically deposited N, within reasonable range of Bouwman et al.4 who estimated N2O +

230

N2 output as 40% of atmospherically deposited N. Wetlands are a considerable source of N2O

231

and N2, highlighting the importance of wetlands in mitigating the accumulation of

232

atmospherically deposited N during late summer, but not during winter. Dissolved N2O efflux

233

was less than 1% of the total N2O efflux in both catchments, consistent with observations made

234

by Davidson and Swank,38 suggesting that dissolved N2O efflux to streams may not be a

235

substantial pathway of N output in forested catchments.

236

There was less NO3- output from C38 (15% of total N output) than C35 (76% of total N

237

output), which suggests that NO3- is not being hydrologically flushed from C38,28 but rather is

238

stored in the wetland and then transformed to other forms of N.17 The NO3- may be converted to


11


Page 12 of 28

239

DON (e.g., ferrous wheel hypothesis of Davidson et al.39) or to N2O or N2 through

240

denitrification.5,40 There was similarly small NH4+ output from both catchments (0.06 kg ha-1 yr-1

241

or 1% of total N output in C38 and 0.07 kg ha-1 yr-1 or 1% of total N output in C35). There was

242

more DON output in C38 (29% of the total N output) than C35 (8% of the total N output). These

243

observations are consistent with those made in other catchments of the Turkey Lakes

244

Watershed.14 While most of the dissolved N output coincided with spring melt, interception of

245

the drainage waters by the wetland in C38 likely resulted in the retention and conversion of N to

246

other forms (e.g., DON).41 Dissolved inorganic and organic N that accumulates during the winter

247

can then be converted to gaseous N2O and N2 during warmer conditions in the summer.41,42

248

Topography influences the atmospheric vs. aquatic fate of N. The catchment with the

249

large wetland (C38) had much greater gaseous N2O and N2 output than the catchment with a

250

small wetland. During the snow season, N2O and N2 efflux occurred continuously, while during

251

the snow free season, bursts of N2O efflux occurred in response to rain events.19 However, these

252

differences between the catchments were mitigated by considerable N2O efflux beneath the

253

snowpack, where there was no observed difference in N2O efflux among the different

254

positions.20 Therefore, both C38 and C35 produced substantial amounts of N2O during the snow

255

season (0.5 kg ha-1 yr-1). Total annual gaseous N2O efflux estimates from the two catchments for

256

the years 2005 to 2010 ranged from 0.8 to 1.7 kg N ha-1 yr-1 in C38 and from 0.5 to 1.6 kg N ha-1

257

yr-1 in C35; these estimates are reasonably similar to the values reported for a temperate forest in

258

the United States (0.3 to 1.4 kg N ha-1 yr-1)15 and a temperate forest in Germany (0.2 to 7.1 kg N

259

ha1 yr-1).43 Previous studies showed that topographic indicators of hydrologic storage potential

260

(i.e., effects of topographic flats and depressions on water storage) were the best predictors of

261

DON output in catchments of the Turkey Lakes Watershed, and topographic indicators of both


12

Page 13 of 28


262

hydrologic storage potential and hydrologic flushing potential (i.e., the effects of topographic

263

slopes on the potential for variable source areas to expand and tap into previously untapped

264

areas) were the best predictors of NO3- output.16,17 Hydrologic flushing potential may not be a

265

good predictor of dissolved N2O and N2 output, because large precipitation events promote the

266

formation of N2O and N2 and rapid off-gassing of N before it can reach the catchment outlet.19

267

However, hydrologic storage potential might also be a good predictor of gaseous N2O and N2

268

output.

269 270

Remaining Missing Nitrogen in Nitrogen Budgets. Incorporating gaseous N2O and N2

271

narrowed the gap between N inputs and outputs, but the average annual N input was still

272

approximately 3.5 kg N ha-1 yr-1 greater than average annual N output.

273

Neither the shallow soil, the partially decomposed leaf matter litter nor the tree biomass

274

is likely to be a significant long-term sink for N in this temperate forest ecosystem. For shallow

275

soil, repeated sampling of the upper soil horizons over the past 25 years has shown that, while

276

there is significant variability both year-to-year and across the landscape, total N content in these

277

pools has not increased significantly.44-46 For litter, Morrison and Foster47 observed an increase

278

in litter N pools over a 15-year period, but attributed it to immobilization from upper soil layers

279

and therefore not an absolute increase in the litter-soil pool. The low carbon:nitrogen (C:N) and

280

high nitrification rates in soils associated with sugar maple forests mean that most of the N is

281

converted to NO3- which is a mobile form of N that is either exported to the stream or converted

282

to gaseous forms through denitrification,14 reducing likelihood of N retention in these soils. Tree

283

biomass may also be a significant long-term sink for N. However, while N pools in tree biomass

284

have not been reported at the Turkey Lakes Watershed since the 1980s, reports from mature


13


Page 14 of 28

285

temperate forests in eastern North America suggest that N pools in forest biomass are not likely

286

to be significant sinks at this late stage of forest development.15,48

287

There are several uncertainties in our N budget calculations. First, dry deposited N that

288

falls on leaf surfaces may not reach the forest floor, but may be processed on the leaf surface.49

289

Second, gaseous N might evade to the atmosphere via a pathway not measured, including

290

transpiration of dissolved N2O in soil water,50 volatilization of NH4+, or off-gassing of dissolved

291

N2O from the stream.51 Third, N might be sequestered in deeper soil horizons not captured in the

292

soil surveys.52 Soils in the Turkey Lakes Watershed are typically shallow, but pockets of deeper

293

soils exist, particularly in the wetlands, where anaerobic conditions promote accumulation of

294

organic N and NH4+, which may act as a long-term sink for N. There are also factors that could

295

increase the discrepancy between inputs and outputs in our calculations. We did not account for

296

N fixation as a possible input. Published rates of N fixation rates in temperate forests suggest that

297

this is a minor input,53 but measurements are sparse and uncertain.52 Furthermore, we did not

298

estimate N2 efflux from upland positions or from wetland positions on non-rain days during the

299

growing season. While N2O efflux numbers for these positions during these time periods suggest

300

that denitrification rates are likely low, this could mean that our estimate of N2 efflux is

301

conservative. Finally, we did not account for significant N storage in the soil. However, if N

302

storage did occur, it likely occurred at a similar rate within each of the two catchments, and

303

therefore the effect of N storage on the N budget would have been similar in the two catchments.

304

While there remains some missing portion of atmospherically deposited N, accounting

305

for N2O and N2 narrowed the observed differences in N outputs between catchments with and

306

without wetlands. The presence of wetlands contributes gaseous N products (N2O and N2) to the

307

composition of N outputs from catchments due to the presence of reducing conditions (lower


14

Page 15 of 28


308

redox potentials). These gaseous N products account for substantial amounts of both

309

atmospherically deposited N (10 to 36% of N input) and N in discharge waters (16 to 55% of N

310

output). While these results demonstrate the importance of gaseous N products to N budgets in

311

temperate forests, scaling this process up across the region remains a significant challenge.12 A

312

process-based understanding that considers topographic influences on N transformations is key

313

to estimating these elusive fluxes. Future work on resolving N budgets in forested catchments

314

can use the topographic framework presented here to help constrain this highly heterogeneous

315

process.

316 317

Acknowledgements

318

This research was funded by an NSERC Discovery grant to Irena Creed (217053-2009

319

RGPIN). Data are available by contacting Irena Creed ([email protected]). The Canadian Forest

320

Service, Natural Resources Canada, provided discharge and chemistry data (Fred Beall), and

321

Environment Canada and Climate Change, provided meteorological data (Dean Jeffries) and

322

atmospheric nitrogen deposition data (Robert Vet).

323 324 325

Supporting Information Available Detailed methods for atmospheric N input and gaseous and dissolved N output

326

measurements are provided in Supporting Information. This information is available free of

327

charge via the Internet at http://pubs.acs.org.

328


15


329 330

References (1) Vitousek, P. M.; Aber, J. D.; Howarth, R. W.; Likens, G. E.; Matson, P. A.; Schindler,

331

D. W.; Schlesinger, W. H.; Tilman, D. G. Technical report: Human alteration of the global

332

nitrogen cycle: Sources and consequences. Ecol. Appl. 1997, 7, 737-750.

333

(2) Galloway, J. N.; Dentener, F. J.; Capone, D. G.; Boyer, E. W.; Howarth, R. W.;

334

Seitzinger, S. P.; Asner, G. P.; Cleveland, C. C.; Green, P. A.; Holland, E. A. et al., Nitrogen

335

cycles: past, present, and future. Biogeochemistry 2004, 70, 153-226.

336

(3) Rockström, J.; Steffen, W.; Noone, K.; Persson, Å.; Chapin III, F. S.; Lambin, E. F.;

337

Lenton, T. M.; Scheffer, M.; Folke, C.; Schellnhuber, H. J.; et al., A safe operating space for

338

humanity. Nature, 2009, 461, 472-475.

339

Page 16 of 28

(4) Bouwman, A. F.; Beusen, A. H. W.; Griffioen, J.; Van Groenigen, J. W.; Hefting, M.

340

M.; Oenema, O. Van Puijenbroek, P. J. T. M.; Seitzinger, S.; Slomp, C. P.; Stehfest E. Global

341

trends and uncertainties in terrestrial denitrification and N2O emissions. Phil. Trans. Royal Soc.

342

B 2013, 368 20130122.

343

(5) Gambrell, R. P.; Patrick Jr., W. H. Chemical and microbiological properties of

344

anaerobic soils and sediments, In Plant Life in Anaerobic Environments; Hook, D. D., Crawford,

345

R. M. Eds.; Ann Arbor Sci. Pub. Inc.: Ann Arbor, Michigan, 1978.

346

(6) Fowler, D.; Pilegaard, K.; Sutton, M. A.; Ambus, P.; Raivonen, M.; Duyzer, J.;

347

Simpson, D.; Fagerli, H.; Fuzzi, S.; Schjoerring, J. K.; et al., Atmospheric composition change:

348

Ecosystems-atmosphere interactions. Atmos. Environ. 2009, 43, 5193-5267.

349

(7) Burgin, A. M.; Lazaar, J. G.; Groffman, P. M.; Gold, A. J.; Kellog, D. Q. Balancing

350

nitrogen retention ecosystem services and greenhouse gas disservice at the landscape scale. Ecol.

351

Eng. 2013, 56, 26-35.


16

Page 17 of 28


(8) IPCC. Climate Change 2013: The Physical Science Basis. Contribution of Working

352 353

Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change;

354

Stocker, T.F. et al., Eds.; Cambridge University Press: New York, 2013. (9) Ravishankara, A. R.; Daniel, J. S.; Portmann, R. W. Nitrous Oxide (N2O): The

355 356

dominant Ozone-depleting substance emitted in the 21st century. Science 2009, 326, 126. (10) Mosier, A. R.; Kroeze, C.; Nevison, C.; Oenema, O; Seitzinger, S.; van Cleemput,

357 358

O. Closing the global N2O budget: nitrous oxide emissions through the agricultural nitrogen

359

cycle. Nut. Cycling Agroecosyst. 1998, 52, 225-248. (11) Groffman, P. M.; Butterbach-Bahl, K.; Fulweiler, R. W.; Gold, A. J.; Morse, J. L.;

360 361

Stander, E. K.; Tague, C.; Tonitto, C.; Vidon, P. Challenges to incorporating spatially and

362

temporally explicit phenomena (hotspots and hot moments) in denitrification models.

363

Biogeochemistry 2009, 93, 49-77. (12) Davidson, E. A.; David, M. B.; Galloway, J. N.; Goodale, C. L.; Haeuber, R.;

364 365

Harrison, J. A.; Howarth, R. W.; Jaynes, D. B.; Lowrance, R. R.; Nolan, B. T.; et al., Excess

366

nitrogen in the U.S. environment: Trends, risks, and solutions. ESA Issues Ecol. 2012, 15, 1-16.

367

(13) Duncan, J. M.; Groffman, P. M.; Band, L. E. Towards closing the watershed

368

nitrogen budget: Spatial and temporal scaling of denitrification. Biogeosciences 2013, 118, 1105-

369

1119.

370

(14) Foster, N.; Spoelstra, J.; Hazlett, P.; Schiff, S.; Beall, F. D.; Creed, I. F.; David C.

371

Heterogeneity in soil nitrogen within first-order forested catchments at the Turkey Lakes

372

Watershed. Can. J. For. Resour. 2005, 35, 797.

373

(15) Yanai, R. D.; Vadeboncoeur, M. A.; Hamburg, S. P.; Arthur, M. A.; Fuss, C. B.;

374

Groffman, P. M.; Siccama, T. G.; Driscoll, C. T. From missing source to missing sink: Long-


17


Page 18 of 28

375

term changes in the nitrogen budget of a northern hardwood forest. Environ. Sci. Technol. 2013,

376

47, 11440-11448. (16) Creed, I. F.; Beall, F. D. Distributed topographic indicators for predicting nitrogen

377 378

export from headwater catchments. Water Resour. Res. 2009, 45, 1-12. (17) Mengistu, S. G.; Creed, I.F.; Webster, K. L.; Enanga, E.; Beall, F. D. Searching for

379 380

similarity in topographic controls on carbon, nitrogen and phosphorus export from forested

381

headwater catchments. Hydrol. Process. 2014, 28, 3201-3216. (18) Goodale, C.L.; Fredriksen, G.; Weiss, M. S.; McCalley, C. K.; Sparks, J.P.; Thomas,

382 383

S.A. Soil processes drive seasonal variation in retention of 15N tracers in a deciduous forested

384

catchment. Ecology 2015, 96, 2653-2668. (19) Enanga, E. M.; Creed, I. F.; Casson, N. J.; Beall, F. D. Summer storms trigger soil

385 386

N2O efflux episodes in forested catchments. J. Geophys. Res. Biogeosci. 2016, 121, 95-108. (20) Enanga, E. M.; Creed, I. F.; Fairweather, T.; Casson, N. J.; Beall, F. D. Snow

387 388

covered soils produce N2O that is lost from forested catchments. J. Geophys. Res. Biogeosci.

389

2016, 121, 2356-2368. (21) Jeffries D. S.; Kelso, J. R. M.; Morrison, I. K. Physical, chemical, and biological

390 391

characteristics of the Turkey Lakes Watershed, central Ontario, Canada. Can. J. Fish. Aq. Sci.

392

1988, 45, 3-13. (22) Canada Soil Survey Committee. Canadian System of Soil Classification, Department

393 394

of Agriculture, Ottawa, Ontario, Canada, 1978. (23) Wickware, G. M.; Cowell, D. W. Forest ecosystem classification of the Turkey

395 396

Lakes Watershed, Ontario, Environ. Cons. Serv., Lands Dir., Eco. Land. Class. Ser. 1985, 18,

397

33.


18

Page 19 of 28


398

(24) Kelso, J. R. M. Preface. Can. J. Fish. Aq. Sci. 1988, 45, 2.

399

(25) Jeffries, D. S. Foreword, The Turkey Lakes Watershed Study after two decades.

400 401

Water, Air and Soil Pollut. Focus 2002, 2, 1-3. (26) Mengistu, S. G.; Quick, C. G.; Creed, I.F. Nutrient export from catchments on

402

forested landscapes reveals complex nonstationary and stationary climate signals. Water Resour.

403

Res. 2013, 49, 1-18.

404

(27) Creed, I. F.; Band L. E.; Foster, N. W.; Morrison, I. K.; Nicolson, J. A.; Semkin, R.

405

S.; Jeffries, D. S. Regulation of nitrate-N release from temperate forests: A test of the N flushing

406

hypothesis. Water Resour. Res. 1996, 32, 3337-3354.

407

(28) Creed, I. F.; Band L. E. Exploring functional similarity in the export of nitrate-N

408

from forested catchments: A mechanistic modeling approach. Water Resour. Res. 1998. 34,

409

3079-3093.

410

(29) Creed, I. F.; Band L. E. Export of nitrogen from catchments within a temperate

411

forest: evidence for a unifying mechanism regulated by variable source area dynamics. Water

412

Resour. Res. 1998, 34, 3105-3120.

413 414 415 416

(30) Webster, K.L.; Creed, I.F.; Beall, F.D.; Bourbonniere, R.A. A topographic template for estimating soil carbon pools in forested catchments, Geoderma 2011, 160, 457-467 (31) Sirois, A.; Vet, R. Detailed analysis of sulphate and nitrate atmospheric deposition estimates at the Turkey Lakes Watershed. Can. J. Fish. Aq. Sci. 1988, 45, 14-25.

417

(32) Kulkarni, M.V.; Groffman, P.M.; Yavitt, J.B.; Goodale, C.L. Complex controls of

418

denitrification at ecosystem, landscape and regional scales in northern hardwood forests. Ecol.

419

Model. 2015, 298, 39-52.


19


420 421 422

Page 20 of 28

(33) Snedecor GW, Cochran WG. 1989. Statistical methods. 8th edn. Ames, IA: Iowa State University Press. 503 p. (34) Galloway, J. N.; Townsend, A. R.; Erisman, J. W.; Bekunda, M.; Cai, Z.; Freney, J.

423

R.; Martinelli, L. A.; Seitzinger, S. P.; Sutton, M. A. Transformation of the nitrogen cycle: recent

424

trends, questions, and potential solutions. Science 2008, 320, 889-892.

425

(35) Yanai, R. D.; Tokuchi, N.; Campbell, J. L.; Green, M. B.; Matsuzaki, E.; Laseter, S.

426

N.; Brown, C. L.; Bailey, A. S.; Lyons, P.; Levine, C. R.; Buso, D. C.; Likens, G. E.; Knoepp, J.;

427

Fukushima, K. Sources of uncertainty in estimating stream solute export from headwater

428

catchments at three sites. Hydrol. Process. 2015, 29, 1793-1805.

429 430 431 432

(36) Kliewer. B. A.; Gilliam, J. W. Water table management effects on denitrification and nitrous oxide evolution. Soil Sci. Soc. Am. J. 1995, 59, 1694-1701. (37) Vidon, P.; Hubbard, L. E.; Soyeux, E. Seasonal solute dynamics across land uses during storms in glaciated landscape of the US Midwest. J. Hydrol. 2009, 376, 34-47.

433

(38) Davidson, E. A.; Swank W. T. Nitrous oxide dissolved in soil solution: An

434

insignificant pathway of nitrogen loss from a southeastern hardwood forest. Water Resour. Res.

435

1990, 26, 1687-1690.

436

(39) Davidson, E. A.; Chorover, J.; Dail D. B. A mechanism of abiotic immobilization of

437

nitrate in forest ecosystems: the ferrous wheel hypothesis, Glob. Change Biol. 2003, 9, 228-236.

438

(40) Wrage, N.; Velthof, G. L.; van Beusichem, M. L.; Oenema, O. Role of nitrifier

439

denitrification in the production of nitrous oxide. Soil Biol. Biochem. 2001, 33, 1723-1732.

440 441

(41) Farquharson, R.; Baldock, J. Concepts in modelling N2O emissions from land use. Plant Soil, 2008, 309, 147-167.


20

Page 21 of 28

442


(42) Holtan-Hartwig, L.; Dorsch, P.; Bakken, L. R. Low temperature control of soil

443

denitrifying communities: kinetics of N2O production and reduction. Soil Biol. Biochem. 2002,

444

34, 1797-1806.

445 446 447 448 449

(43) Brumme, R.; Borken, W.; Finke, S.; Hierarchical control on nitrous oxide emission in forest ecosystems. Glob. Biogeochem. Cycles, 1999, 13, 1137-1148. (44) Morrison, I. K. Organic matter and mineral distribution in an old-growth Acer saccharum forest near the northern limit of its range. Can. J. For. Res. 1990, 20, 1332-1342. (45) Creed, I. F.; Trick, C.G.; Band L.; Morrison, I. K. Characterizing the spatial pattern

450

of soil carbon and nitrogen pools in the Turkey Lakes Watershed: a comparison of regression

451

techniques. Water, Air Soil Pollut. Focus 2002, 2, 81-102.

452

(46) Webster, K. L.; Creed, I.F.; Bourbonniere, R.A.; Beall, F.D. Controls on the

453

heterogeneity of soil respiration in a tolerant hardwood forest, J. Geophys. Res. 2008, 113, 1-15.

454

(47) Morrison, I. K.; Foster, N. W. Fifteen-year change in forest floor organic and

455 456 457 458

element content and cycling at the Turkey Lakes Watershed. Ecosystems 2001, 4, 545-554. (48) Watmough, S. A.; Dillon, P. J. Major element fluxes from a coniferous catchment in central Ontario, 1983–1999. Biogeochemistry 2004, 67, 369-399. (49) Uscola, M.; Villa-Salvador, P.; Oliet, J.; Warren, C. R. Foliar absorption and root

459

translocation of nitrogen from different chemical forms in seedlings of two Mediterranean trees,

460

Environ. Exp. Botany, 2014, 104, 34-43.

461

(50) Mosier, A. R.; Mohanty, S. K.; Bhadrachalam, A.; Chakravorti, S. P. Evolution of

462

dinitrogen and nitrous oxide from the soil to the atmosphere through rice plants. Biol. Fertile

463

Soils 1990, 9, 61-67.


21


464 465 466 467 468 469

Page 22 of 28

(51) Baulch, H. M.; Schiff, S.L.; Maranger, R.; Dillon, P.J. Nitrogen enrichment and the emission of nitrous oxide from streams, Glob. Biogeochem. Cycles, 2011 25, GB4013. (52) Johnson, D. W.; Turner, J. Nitrogen budgets of forest ecosystems: A review. For. Ecol. Manag. 2014, 318, 370-379. (53) Roskoski, Nitrogen fixation in hardwood forests of the northeastern United States. Plant Soil 1980, 54, 33-44.


22

Page 23 of 28


470

Table 1. Medians and the 25th and 75th percentiles of distributions of soil N2O and N2 (mg ha-1 d-1) effluxes across a hillslope gradient

471

and stream N2O fluxes at the catchment outlet in C38. Snow-free Season (rain > 3 mm)

Snow-free Season (no rain) th

472

Soil N2O IW OW LOW UP Soil N2 IW OW Stream N2O .

25 Percentile 3,889 5,685 3,262 2,350

0.00003

th

Median

75th Percentile

6,058 8,807 4,982 4,847

9,322 14,404 9,518 6,838

0.00018

0.00425

25 Percentile 3,710 9,827 3,262 2,350 72,675 98,862 0.00001

Snow Season th

Median 6,959 17,874 4,982 4,847 136,327 179,809 0.00140


75 Percentile 12,222 30,535 9,518 6,838 239,428 307,182 0.00266

th

25 Percentile 1,519 803 1,196 1,584 29,753 8,080 2.95

Median

75th Percentile

1,139 1,293 1,214 1,315 22,315 13,007 7.56

5,524 2,038 8,105 4,091 108,205 20,499 35.1

23


Page 24 of 28

473

474 475

Figure 1. Conceptual model showing the different fates of atmospherically deposited nitrogen:

476

(1) assimilated by organisms (microorganisms and higher); (2) nitrification and denitrification,

477

with N2O as an obligate intermediate and N2 as the final product; or (3) lost to aquatic systems

478

via hydrologic flows.


24

Page 25 of 28


479 480

Figure 2. Map of the Turkey Lakes Watershed in Central Ontario, Canada with the two

481

catchments used in the present study (C38 and C35). (Maps created in Creed laboratory)


25


Page 26 of 28

482 483

Figure 3. Average annual N budget for 2005 to 2010 in the C38 and C35 catchments in the

484

Turkey Lakes Watershed showing traditionally-measured N species, including dissolved

485

inorganic nitrogen (DIN, nitrate-N + ammonium-N) and dissolved organic nitrogen (DON), as

486

well as gaseous and dissolved nitrous oxide (N2O) and dinitrogen (N2) efflux. Error bars

487

represent the standard deviations of N outputs with the grey shaded area representing the

488

standard deviation of N inputs.


26

Page 27 of 28


489 490

Figure 4. Daily measured N2O (top), N2 (middle) and total gaseous N (N2O + N2) (bottom)

491

efflux estimates across a hillslope gradient in C38 during the snow-free season (June to

492

September) from 2005 to 2010 (left) and during the snow season in October 2006 to May 2007

493

(right). IW and OW positions include estimates of N2 derived from experimentally determined

494

N2:N2O ratios; N2 was not estimated at LOW and UP positions. Different letters indicate

495

significant differences among topographic positions (p < 0.05).


27


Page 28 of 28

496 497

Figure 5. Monthly total N outputs for the C38 and C35 catchments from 2005-2010 in the

498

Turkey Lakes Watershed.


28

Nitrous Oxide and Dinitrogen: The Missing Flux in ... - ACS Publications

Recommend Documents