Sebestyen et al.
Stream N during autumn
1364
Supporting Information Submission for Paper 2013WR013670
Coupled hydrological and biogeochemical processes controlling variability of nitrogen
species in streamflow during autumn in an upland forest.
Stephen D Sebestyen, James B Shanley, Elizabeth W Boyer, Carol Kendall, and Daniel
H. Doctor
(USDA Forest Service Northern Research Station, Grand Rapids, MN, USA; Penn State
University, University Park, PA, USA; US Geological Survey, Montpelier, VT, USA; US
Geological Survey, Menlo Park, CA, USA; US Geological Survey, Reston, VA, USA)
Water Resources Res., xxx,
doi: 10.1002/2013WR013670
1365
Introduction
1355
1356
1357
1358
1359
1360
1361
1362
1363
1366
The Supporting Information contains figures and additional information to
1367
supplement descriptions of the base flow/stormflow separation (S1); and water
1368
sample collection, preparation, and analysis (Supporting Informations S2-S3).
1369
Nitrate concentrations for the tributary streams are presented in Supporting
1370
Information S4. Nitrate and water isotope data variation over time are shown in
1371
Supporting Information S5. Solute concentration and isotopic composition data are
1372
provided in spreadsheets in Supporting Information S6. Streamflow, precipitation
1373
amount, and water level data are provided in spreadsheets in Supporting
1374
Information S7.
1375
Supporting Information S1: Base flow and stormflow separation.
1376
The Nathan and McMahon [1990] base flow recession technique was used to
1377
calculate quick flow (qquick flow) and slow flow fractions from 30-minute streamflow
1378
data. Any sample collected when qquick flow was greater than 10% of streamflow was
1379
considered to have been sampled from stormflow. Mixing diagrams and review of
1380
the hydrograph showed that one sample (from 10/15/03 18:23 EST) was more
1381
appropriately considered as having been collected during stormflow.
1382
Stream nitrate concentration shown relative to the fraction of quick flow.
2/8/2016 12:03
56
Stream N during autumn
week of peak litterfall
[ nitrate ]
( μmol L-1 )
30
1384
1385
1386
1387
100%
50%
15
0
8/29/03
1383
W-9 stormflow
W-9 base flow
quickflow
quick flow
Sebestyen et al.
0%
9/19/03
10/10/03
10/31/03
11/21/03
References
Nathan, R. J., and T. A. McMahon (1990), Evaluation of automated techniques for
base flow and recession analyses, Water Resour. Res., 26(7), 1465-1473.
Supporting Information S2: Long-term stream sampling and analytical methods.
1388
For the long-term monitoring, concentrations of nitrate and ammonium were
1389
measured by automated colorimetry (continuous flow analysis on a Technicon
1390
AutoAnalyzer), TN by catalytic oxidation combustion (Antek 720C
1391
chemiluminescent nitrogen detector coupled with a Shimadzu TOC-5000A, no
1392
measurements before 1996), and dissolved organic carbon (DOC) concentrations
1393
by continuous flow analysis on a Technicon Autoanalyzer before 1996 and by
1394
combustion oxidation on a Shimadzu TOC-5000A thereafter [Bailey et al., 1995].
1395
Table of limits of detection.
Analyze
Ammonium
DOC before 1996
DOC after 1996
Nitrate
TN
Method Detection Limits (μmol L-1)
2.2
17
8
2.9
1.4
1396
References
1397
Bailey, S. W., C. T. Driscoll, and J. W. Hornbeck (1995), Acid-base chemistry and
aluminum transport in an acidic watershed and pond in New Hampshire,
Biogeochemistry, 28(2), 69-91, doi:10.1007/BF02180678.
1398
1399
2/8/2016 12:03
57
Sebestyen et al.
1400
1401
Stream N during autumn
Supporting Information S3: Additional sampling and analytical information for samples
collected during this study.
1402
Samples from ISCO samplers were usually retrieved weekly in 2003 and within
1403
twelve hours during October. Grab and ISCO samples were collected in new half-
1404
liter LDPE bottles that were first tripled rinsed with sample water, then filled, and
1405
stored refrigerated. In October, some stream nitrogen samples were syringe-filtered
1406
in the field. Filters were first flushed with sample. Bottles were then rinsed with
1407
filtered sample and then filled.
1408
In the lab, unfiltered samples were syringe filtered into pre-cleaned sample bottles,
1409
HDPE bottles for nitrogen species and amber bottles with Teflon-lined caps for
1410
DOC analysis. All HDPE bottles were pre-rinsed and leached with deionized water
1411
(resistance exceeding 18.0 Ωohms cm-1).
1412
Aliquots for nitrogen species concentrations were filtered and refrigerated. Nitrate
1413
isotope samples were filtered and then frozen until analyzed. Nitrate concentrations
1414
were measured at the EW Boyer Water Chemistry Lab. Paired unfiltered grab and
1415
field-filtered samples were collected to test for differences of filters and the effects
1416
of holding time on nitrate concentrations. Unfiltered grab and field-filtered samples
1417
were auto-injected into the ion chromatograph through 20 μm filter caps at the time
1418
of analysis. Analysis of sample pairs showed no detectable difference between field
1419
and laboratory filtering and no effect of storage time on nitrate concentrations for up
1420
to 5 years. Total dissolved nitrogen was measured in the WL Silver Ecosystem
1421
Laboratory at the University of California, Berkeley.
1422
Whatman GF/F filters (0.7-μm nominal pore size) used for lab filtering and glass
1423
bottles for DOC samples were pre-baked for six hours at 450˚ Celsius. For field
1424
filtered samples, PURA-DISC GF/F disposable in-line syringe filters were used
2/8/2016 12:03
58
Sebestyen et al.
Stream N during autumn
1425
though these filters could not be baked. Because samples spanned the duration of
1426
several different studies of DOC, concentrations were measured at multiple
1427
laboratories: USGS MPSIL by the persulfate wet oxidation method on an OI
1428
Analytical 1010 TIC/TOC analyzer [Doctor et al., 2008]; USGS, Boulder, CO by the
1429
persulfate wet oxidation method on an OI Analytical 700 TOC Analyzer [Aiken,
1430
1992]; and the MJ Mitchell Biogeochemistry Lab, SUNY-ESF, Syracuse, NY by the
1431
UV-persulfate oxidation method on a Tekmar-Dohrmann Phoenix 8000 TOC
1432
analyzer.
1433
Deionized water blanks were analyzed every ten to twenty samples. No
1434
contamination of samples from collection, storage, or processing was detected from
1435
blanks.
1436
References
1437
Aiken, G. R. (1992), Chloride interference in the analysis of dissolved organic
carbon by the wet oxidation method, Environ. Sci. Technol., 26(12), 2435-2439.
1438
1439
1440
1441
1442
Doctor, D. H., C. Kendall, S. D. Sebestyen, J. B. Shanley, N. Ohte, and E. W. Boyer
(2008), Carbon isotope fractionation of dissolved inorganic carbon (DIC) due to
outgassing of carbon dioxide from a headwater stream, Hydrol. Process., 22(14),
2410-2423, doi:10.1002/hyp.6833.
1443
2/8/2016 12:03
59
Sebestyen et al.
1444
1445
1446
1447
1448
1449
Stream N during autumn
Supporting Information S4: Nitrate concentrations at the three tributary streams.
(a) Streamflow at the W-9 stream gage. At the (b) W-9 stream and (c) the three
upstream tributaries, nitrate concentrations decreased from 7 to 14 October during
the week of peak leaf fall. The circled numbers show the four closely-spaced
stormflow events during October. The inset figure in (b) shows details of slight
nitrate increases during the third and fourth storm events during October.
(a) STREAMFLOW AT W-9
week of peak litterfall
1
0.1
1
2
(b) NITRATE CONCENTRATIONS AT W-9
W-9
1
4
10
15
0
3
0.5
5
0
10/25
2
3
10/27
3
4
10/29
4
(c) NITRATE CONCENTRATIONS IN TRIBUTARY STREAMS
[ nitrate ]
( μmol L-1 )
45
1451
1452
30
A
B
C
15
0
10/1/03
2/8/2016 12:03
10/8/03
10/15/03
10/22/03
60
10/29/03
0.05
10/31
streamflow
( mm h-1 )
0.01
30
[ nitrate ]
( μmol L-1 )
[ nitrate ]
( μmol L-1 )
streamflow
( mm h-1 )
1450
Sebestyen et al.
1453
1454
1455
1456
1457
1458
Stream N during autumn
Supporting Information S5: Nitrate and water isotope in stream water, precipitation, soil
water, and groundwater.
(a) Streamflow at the W-9 stream gage. (b) δ15N, and (c) δ18O of nitrate in rain, soil
water, stream water, and groundwater. (d) Percentages of nitrate in soil and stream
waters that originated from an atmospheric source, calculated using Eq. 2 and 3.
(e) δ18O of water in rain, soil water, stream water, and groundwater samples.
1459
streamflow
( mm h-1 )
(a) STREAMFLOW AT W-9
week of peak litterfall
1
1
2
3
4
0.5
0
δ15N-nitrate ( ‰ )
(b) δ15N-NITRATE OF RAIN, SOIL WATER, STREAM WATER, AND GROUNDWATER
10
R
R
R
0
S
-10
S
S
δ18O-nitrate ( ‰ )
(c) δ18O-NITRATE OF RAIN, SOIL WATER, STREAM WATER, AND GROUNDWATER
110
W-9 base flow
W-9 stormflow
80
50
S
R
S
surficial soil water
riparian groundwater
rain
20
-10
S
R
R
R
S
apportioned direct
atmospheric source
(d) NITRATE FROM AN ATMOSPHERIC SOURCE
70%
S
35%
0%
S
R
R
R
S
δ18O-water ( ‰ )
(e) δ18O-WATER OF RAIN, SOIL WATER, STREAM WATER, AND GROUNDWATER
1460
1461
0
S
S
-10
-20
8/28/03
2/8/2016 12:03
R
R
R
R
R
R
9/18/03
10/9/03
61
R
R
R
R
R
10/30/03
11/20/03
Sebestyen et al.
1462
1463
1464
1465
1466
1467
1468
1469
Stream N during autumn
Supporting Information S6: 2013WR013670-sup-0006-ds01.xlsx and 2013WR013670sup-0006-ds01.pdf
Worksheet “DATA”: Physical, chemical, and isotope data for each sample included
in Figures 2-12 and Auxliary Materials S4 and S5. The worksheet includes samples
of stream, precipitation, soil, and ground water. Solute concentrations are reported
in units of μmol L-1, δ18O and δ15N are reported in units of permil (‰), and times are
Eastern Standard Time (UTC/GMT-5 h). The method detection of limits are
reported for each solute and analytical precisions are reported for isotopes.
1470
1471
1472
1473
Worksheet “MAP”: Map of sampling locations and landscape features. The
numbers identify particular wells and piezometers. Reproduction of Figure 1 from
2013WR013670.
1474
1475
1476
1477
Worksheet “METADATA”: Metadata for chemical and isotope data. This text is
reproduced from the methods section and Supporting Information S3 of
2013WR013670.
1478
1479
1480
1481
1482
1483
1484
1485
Supporting Information S7: 2013WR013670-sup-0006-ds02.xlsx and 2013WR013670sup-0006-ds02.pdf
Worksheet “STREAMFLOW”: Breakpoint streamflow data for the four stream
gages: W-9, 2003 calendar year; W-9A, 9/1/2003 to 11/30/2003; W-9B, 9/1/2003 to
11/30/2003; and W-9C, 9/1/2003 to 11/30/2003. Specific discharge is reported in
units of mm h-1. These data are plotted in Figures 2, 5, 7, 10, and 11 as well as
Supporting Information S5.
1486
1487
1488
1489
Worksheet “PRECIPITATION”: Daily precipitation amounts from the weighing
bucket rain gage at the R-29 meteorological station. Precipitation amounts are
reported in units of mm d-1. These data are plotted in 11.
1490
1491
1492
1493
Worksheet “WATER LEVELS”: Breakpoint water level data for wells (BW-19, UP,
and MID) and piezometers (T-1 and T-3). Water levels are m below the ground
surface.
1494
1495
1496
1497
Worksheet “METADATA”: Metadata for measurement of streamflow, precipitation,
and water level elevations in wells and piezometers. This text is reproduced from
the methods section and Supporting Information S3 of 2013WR013670.
1498
1499
1500
1501
Worksheet “MAP”: Map of sampling locations and landscape features. The
numbers identify particular wells and piezometers. Reproduction of Figure 1 from
2013WR013670.
1502
1503
All times are Eastern Standard Time (UTC/GMT-5 h)
2/8/2016 12:03
62