Ecological Archives E095-245-A1

Alexander D. Huryn, Jonathan P. Benstead, Stephanie M. Parker. 2014. Seasonal changes in light availability modify the temperature dependence of ecosystem metabolism in an arctic stream. Ecology 95:2826–2839.

Appendix A. Supporting material including a detailed description of methods; one table summarizing relevant variables used to calculate metabolism and nutrient uptake; and five figures showing location of Ivishak Spring; ER and GPP vs. day of the year; summary of daily PAR, air temperature, and water temperature; GPP, ER, NEP vs. water temperature; and seasonal patterns of organic matter biomass and DOC concentration.


• Field Methods

o Dissolved oxygen

o Reaeration and stream depth

o Groundwater dilution

o Light

o Processing of organic matter samples

• Literature cited in Appendix A

• Table 1

• Figures A1 to A5

Field Methods

Dissolved Oxygen

On each semi-monthly visit to the field site, dissolved oxygen concentration (mg O2 L-1) was recorded at 3-min intervals for ~30–72 hrs using Hydrolab MS5 Minisondes (Hach Hydromet, Loveland, CO, USA) equipped with luminescent dissolved oxygen (LDO) sensors. The LDO sensors were calibrated in the field using DO-saturated stream water. Following calibration, sondes were placed together for 30+ min in a 20-L bucket containing continuously aerated 1stream-water to later assess sensor correspondence. Following this procedure the sondes were positioned upstream (0-m) and downstream (200-m) of a 200-m study reach. At the end of the deployment the sondes were again placed together in aerated stream water to obtain data allowing the assessment of sensor correspondence and drift. This procedure was used for all dates except during December 2007 and 2008 when helicopter access to the study site was not available. Sondes deployed in November were activated during December using an autostart procedure.

Reaeration and stream depth

Reaeration was measured using a modification of the method of Tobias et al. (2009), which is based on the evasion rate of sulfur hexafluoride (SF6) gas from water as a proxy for O2 (Wanninkhof et al. 1990, Hall and Tank 2005). In summary, a 16-L solution of rhodamine WT (RWT; Turner Designs, Sunnyvale, CA, USA; 3 g 20% RWT solution/L water) was added to a 20-L gas-tight Tedlar bag (SKC, Eighty Four, PA, USA). The remaining headspace was filled with SF6 and the bag was shaken to ensure saturation of the RWT solution with SF6. The gas-tight bag was then transported to the field site where its contained solution was injected into the stream at a rate of 50 mL min-1 (FMI Lab Pump Model QGB, Fluid Metering, Syosset, NY, USA) for a period sufficient to saturate the stream (a minimum of 180 min as indicated by RWT concentration plateaus at 200 m). Once the stream was saturated, 20 50-mL water samples were taken at even intervals across the width of the stream at 0-m, 50-m, 100-m, 160-m and 200-m (n = 100 samples total). Immediately after sampling, each sample was used to completely fill a 10-mL screw-cap gas-sampling vial (Labco Exetainer, High Wycombe, Buckinghamshire, UK). The remaining 40-mL sample of water was used for measurement of RWT concentration. The 10-mL samples in gas-sampling vials were returned to the University of Alabama, Tuscaloosa, where SF6 concentration was measured.

The procedure for measuring SF6 concentration consisted of equilibrating the dissolved SF6 contained in 8 mL of water with air contained in a 2-mL headspace, sampling a known volume of the headspace with a gas-tight syringe, and then measuring SF6 concentration by mass spectrometry (Agilent gas chromatograph, model 6890N, coupled to a mass selective detector, model 5973 inert, Santa Clara, CA, USA). Relative concentrations of SF6 between samples were corrected for dilution using the concentrations of RWT measured for each sample to correct for the local effects of groundwater infiltration at each sampling location. The decline of dilution-corrected, relative SF6 concentrations was used to estimate the rate of SF6 evasion (kSF6, min-1), which was converted to DO evasion (kDO) using the ratio of their Schmidt numbers (Wanninkhof et al. 1990, Hall and Tank 2005). kDO was measured over a range of discharge conditions and a regression equation relating discharge to the reaeration coefficient was used to estimate metabolism for different dates. Water travel time, required to estimate kSF6, was estimated by releasing a slug containing 3 kg of NaCl dissolved in 16 L of stream water approximately 40 m above 0 m. The concentration of NaCl in the water passing 0 m and 200 m was estimated using a standard curve relating changes in stream water conductivity (μS cm-1, recorded at 3-min intervals simultaneously at 0 m and 200 m using Hydrolab MS5 Minisondes) to NaCl concentration (mg/L). The area beneath the concentration profile (above baseline) was integrated and the time when 50% of the NaCl contained in the slug passed the beginning and end of the reach was estimated. Travel time was calculated as the difference between the times estimated for 50% of the slug to pass 0 m and 200 m.

Estimates of stream depth are required for converting volumetric estimates of GPP and ER rates (m-3) to areal rates (m-2). Stream depth was measured at 5 equidistant locations across the width of the stream at 5-m intervals from 0 m to 200 m. Mean reach depth on each date was regressed against stage height to produce a continuous (15-min interval) estimate of stream depth. Stream stage height and temperature was recorded at 15-min intervals using a digital recording pressure-transducer (Hobo U20 Water Level Data Logger, Onset Corporation, Pocasset, MA, USA) deployed in a PVC stilling well in a pool 40 m downstream of the 200-m station.

Groundwater dilution

The quantity of groundwater entering the study reach was assessed using the longitudinal decline in RWT concentration (measured at 5-m intervals) following plateau. The effect of groundwater dilution on stream DO requires knowledge of groundwater DO (Hall and Tank 2005). Groundwater DO was measured on two dates (1 July and 2 August 2009) using a handheld dissolved-oxygen meter equipped with an LDO probe (YSI ProODO, Yellow Springs, OH, USA) in ten 30-cm PVC wells (n = 10) placed in the riparian zone in locations that were likely sites of groundwater infiltration into the channel.


PAR (μmol·m-2·s-1) was measured using underwater quantum sensors (LI-190SA, LI-COR, Lincoln, NE, USA) interfaced with digital recorders. Sensors were placed ~10 cm below the surface of the water (approximate average reach depth) at two locations. The first location was approximately 20 m upstream of 0 m; the second was approximately 30 m below 200 m. PAR was measured every 5 s and data were recorded as 5-min means. The field site was not visited during December 2007 and December 2008. PAR was measured during December by deploying light meters from November through January. Data were averaged over the two PAR sensors.

Processing of organic matter samples

After coarse organic matter samples were collected using a Surber sampler, cobble and pebble-size particles were removed from the sampler frame and scrubbed into a bucket using a nylon brush. The remaining particles within the sample frame were then agitated to entrain organic particles into the capture net. The contents of the net were then combined with organic matter sieved from the bucket and preserved with 4% formaldehyde until processing, which consisted of separating organic materials into identifiable categories (e.g., bryophytes, willow leaf detritus, wood) followed by oven drying at 55ºC, weighing (= dry mass), combusting at 500ºC, and reweighing (= ash mass). The difference between dry and ash mass was used to estimate ash-free dry mass (AFDM).

During sampling for epilithon, the partitioned area was scrubbed with a wire bush, rinsed with stream water, and the resulting slurry brought to a volume of 125 mL in an amber sample bottle. In the laboratory, a 10-mL subsample of each sample was filtered onto a GF/F glass fiber filter and chlorophyll was extracted for 24 hours (Strickland and Parsons, 1972, Slavik et al. 2004). Total chlorophyll was measured using a Turner Designs Aquafluor fluorometer and expressed as mg/m² stream bottom. An additional subsample was filtered onto a GF/F filter to assess biofilm AFDM (mg AFDM m-2).

Water samples for dissolved organic carbon (DOC) were filtered (GF/F) and placed in 125-mL acid-washed amber bottles. The pH was then adjusted to ~2 with concentrated HCl and the samples frozen or stored on ice until transport to the laboratory. DOC concentrations were measured using an automated TCN analyzer (Shimadzu TOC-VCPN, Shimadzu Scientific Instruments, Columbia, MD, USA).

Literature cited

Hall, R. O., Jr., and J. L. Tank. 2005. Correcting whole-stream estimates of metabolism for groundwater input. Limnology and Oceanography: Methods 3:222–229.

Slavik, K. A., B. J. Peterson, L. A. Deegan, W. B. Bowden, A. E. Hershey, and J. E. Hobbie. 2004. Long-term responses of the Kuparuk River ecosystem to phosphorus fertilization. Ecology 85:939–954.

Strickland, J. D. H., and T. R. Parsons. 1972. A practical handbook of seawater analysis, Second edition. Fisheries Research Board of Canada, Bulletin 167:1–310.

Tobias, C. R., J. K. Böhlke, J. W. Harvey, and E. Busenberg. 2009. A simple technique for continuous measurement of time-variable gas transfer in surface waters. Limnology and Oceanography: Methods 7:185–195.

Wanninkhof, R., P. J. Mulholland, and J. W. Elwood. 1990. Gas exchange rates for a first- order stream determined with deliberate and natural tracers. Water Resources Research 26: 1621–1630.

Table 1. Summary of relevant variable statistics used to calculate metabolism and nutrient uptake for Ivishak Spring, AK (March 2007–August 2009). "US" refers to mean daily water temperature measured (3-min intervals) at the upstream station during measurements of DO for calculations of metabolism. "DS" refers to mean daily water temperatures measured 200-m downstream of the upstream station. Water temperature was also measured continuously at 10-min intervals (see Fig. 6). n refers to the number of dates for which measurements were made, except for channel width, channel depth, and groundwater DOsat. For these variables, n = the number of dates × the number of measurements made at different point locations on each date.

Variable Units Mean SE n Max Min
Channel width
m 2.7 0.3 10 × 15 4.6 1.3
Channel depth
m 0.12 0.01 8 × 150 0.16 0.08
Water velocity
m/s 0.31 0.02 16 0.40 0.21
L/s 122 7 20 177 70
Groundwater input
% 28.7 1.4 19 40.3 10.3
Groundwater DOsat
% 70.5 3.1 2 × 10 97.9 49.4
Reaeration (kDO)
d-1 227 18 5 271 174
Water temperature - US
°C 5.9 0.1 23 7.3 4.9
Water temperature - DS
°C 5.6 0.2 23 7.5 4.2



Fig. A1. Location of Ivishak Spring ("Ivishak Hot Spring", 69° 01' 24.1"N, 147° 43' 21.1"W), a tributary of the Ivishak River in the Arctic National Wildlife Refuge on the North Slope of Alaska.


Fig. A2. ER and GPP (g C·m-2·d-1) vs. day of the year for Ivishak Spring, AK, showing annually repeated pattern of highest rates during summer. GPP and ER was estimated semi-monthly from March 2007 to August 2009.



Fig. A3. Seasonal patterns of daily mean PAR (μmol·m-2·s-1, upper), air temperature (ºC, center) and water temperature (ºC, lower) measured at Ivishak Spring, AK, from March 2007 to August 2009. Gaps in the data indicate periods of instrument malfunction.



Fig. A4. GPP (top), ER (center) and NEP (bottom) versus temperature (ºC, left) and PAR (μmol·m-2·s-1, right) for Ivishak Spring, AK. GPP, ER, and NEP were estimated semi-monthly from March 2007 to August 2009.



Fig. A5. Seasonal patterns of organic matter biomass (g AFDM m-2, upper) and DOC (mg C L-1 ± SE, lower) measured semi-monthly (March 2007–August 2009) for Ivishak Spring. In upper panel, filled circles = bryophytes, open circles = willow leaf litter, triangles = FPOM.

[Back to E095-245]