We examined the spatial and temporal variability of stream and hyporheic partial pressure of carbon dioxide (pCO2) and the drivers of these variations in a headwater catchment. To examine the temporal dynamics of stream and hyporheic pCO2, we measured pCO2 at high temporal resolution (30-minute interval) at seasonal and event scales in the in the stream and in a hyporheic well in the Watershed 1 well network in the HJ Andrews Experimental Forest. Stream and hyporheic pCO2 were observed for 11 months, from 27 August 2013 to 10 July 2014.
To examine spatial variation and drivers, we measured stream pCO2 at monthly intervals from July 2013 through July 2014 at 38 locations across the 6400-ha HJ Andrews Experimental Forest. Synoptic field measurements and water samples were collected 12 times over a 13-month study period from July 2013 to July 2014. Sample sites were established across the Lookout Creek drainage network at sites that captured the diverse natural and managed landscape of the HJA. Sample sites were chosen to ensure they included streams from a range of basin sizes, elevations, discharge regimes, subclimates and forest management histories.Sampling occurred at approximately monthly intervals; no samples were collected during April 2014. Low elevation sites were sampled during all 12 sample periods. High elevation sites were sometimes inaccessible due to snowfall and were therefore not sampled during some winter months. Streams that were designed as longitudinal transects were sampled from high elevation to low. Sites were sampled at approximately the same time of day over the study period.
30-minute pCO2 data were collected GMM220 series CO2 probes following the method in Johnson et al. [2010]. Probes were sealed in gas permeable PTFE sleeves and connected to external data loggers where mole fraction pCO2 (ppmv) was measured at 5-min intervals, averaged and recoded at 30-min intervals. One probe was placed in the WS01 stream attached to a stryofoam float at a depth of 6 cm and another in a riparian well located 0.4 m from the stream edge at a depth of 65 cm below the water table.
Water samples for the analysis of stream alkalinity were collected in 250 mL HDPE (high-density polyethylene) bottles (Nalgene, Inc., United States). Bottles were acid washed according to the Oregon State University and U.S. Forest Service Cooperative Chemical Analytical Laboratory (CCAL) quality assurance plan (CCAL, 2013). Prior to collection, bottles were rinsed three times with stream water by filling the sample container 1/4 full, then shaking vigorously with the lid loosely tightened to the bottle, making sure to rinse the inside walls, threads, and cap of the bottle. Samples were collected from the thalweg in locations deep enough to fully submerge the bottle while the sample collector stood downstream of the sampling location. In larger streams, samples were taken as close to the thalweg as possible and always collected from locations with strong downstream transport, thereby avoiding stagnant pools or lateral cavities. Sample bottle caps were secured underwater to minimize air collected with the sample and to minimize the amount of sediment or other particulates collected with the water sample. Nitrile gloves were worn during sampling to minimize sample contamination. After collection, samples were placed on ice in a dark container until laboratory processing.
Stream pH and temperature were measured at sample sites using a YSI Model 63 handheld pH, conductivity, salinity and temperature system (YSI, Inc., United States). The meter was calibrated for pH at the beginning of each field day with a two point calibration at pH 7.0 and 10.0, bracketing the range of expected field pH values. When making field measurements, the probe was placed in the thalweg of the stream, slightly downstream from the location of water sampling. The meter was given a minimum of 10 minutes to stabilize prior to reading. The probe was positioned so that the pH electrode was suspended in the water column and not resting in mud or sediment.
CCAL (2013), CCAL water analysis quality assurance plan. Oregon State University and United States Forest Service Cooperative Analytical Laboratory.
Johnson, M., M. Billett, and K. Dinsmore (2010), Direct and continuous measurement of dissolved carbon dioxide in freshwater aquatic systems—method and applications, Ecohydrology, 3, 68–78, doi:10.1002/eco.95.
Analytical lab work was conducted at the Institute for Water and Watersheds Collaboratory at Oregon State University. Carbonate alkalinity was determined using a Radiometer TIM840 AutoTitrator (Radiometer Analytical, SAS, France) according to CCAL procedure 10C.0 by titrating 100 mL of unfiltered sample to a pH of 4.5 with 0.02 N H2SO4 (CCAL, 2013). Detection limits associated with this method are 0.2 mg CaCO3 L-1. Samples were processed within 7 days of field collection.
30-minute stream and hyporheic CO2-probe readings were corrected for deviations in temperature, depth and barrometric pressure from calibration conditions following Johnson et al. (2010).
Stream pCO2 was calculated with the CO2SYS program (van Heuven et al., 2011) using field measurements of temperature and pH, and lab measurements of alkalinity. This calculation is based on equilibrium concentrations of the carbonate species in water at a given pressure, temperature, and pH. The dissociation constants of the carbonate species used for this calculation were those from Millero (1979) for freshwater systems.
Millero, F. (1979), The thermodynamics of the carbonate system in seawater, Geochimica Et Cosmochimica Acta, 43, 1651–1661.
van Heuven, S., D. Pierrot, J. W. B. Rae, E. Lewis, and D. W. R. Wallace (2011), MATLAB Program Developed for CO2 System Calculations. ORNL/CDIAC-105b. Carbon Dioxide Information Analysis Center, Oak Ridge National Laboratory, U.S. Department of Energy, Oak Ridge, Tennessee. http://dx.doi.org/10.3334/CDIAC/ otg.CO2SYS_MATLAB_v1.1.
Basin-wide measurtments were collected from across the Lookout creek drainage network in the HJA (44°14’N, 122°11’W), in the Western Cascades of Central Oregon, USA. The boundary of HJA is delineated by the Lookout Creek watershed, a 64 km2 drainage basin ranging in elevation from 410 m to 1630 m. Lookout Creek drains a network of forested headwater streams and is a tributary of Blue River and the McKenzie River. 38 sample sites at were established across the Lookout Creek drainage network were at a variety of sites that captured the diverse natural and managed landscape of the HJA. Sample sites were chosen to ensure they included streams from a range of basin sizes, elevations, discharge regimes, subclimates and forest management histories. Multiple sample sites were placed along Lookout Creek and its major tributaries and were sampled as longitudinal transects from the highest upstream point to the lowest downstream point accessible by truck. Samples were collected at all gauge stations and from all experimental watersheds. Major confluences were included in sampling design, with sample sites placed above and below these locations.
