DOC export is exceeded by C fixation in May Creek: A late-successional watershed of the Copper River Basin, Alaska.

Understanding the entirety of basin-scale C cycling (DOC fluxes and CO2 exchanges) are central to a holistic perspective of boreal forest biogeochemistry today. Shifts in the timing and magnitude of dissolved organic carbon (DOC) delivery in streams and eventually into oceans can be expected, while simultaneously CO2 emission may exceed CO2 fixation, leading to forests becoming stronger CO2 sources than sinks amplifying rising trace gases in the atmosphere. At May Creek, a representative late-successional boreal forest watershed at the headwaters of the Copper River Basin, Alaska, we quantified the seasonality of DOC flux and landscape-scale CO2 exchange (eddy covariance) over two seasonal cycles. We deployed in situ fDOM and conductivity sensors, performed campaign sampling for water quality (DOC and water isotopes), and used fluorescence spectroscopy to ascertain DOC character. Simultaneously, we quantified net CO2 exchange using a 100 ft eddy covariance tower. Results indicate DOC exports were pulse-driven and mediated by precipitation events. Both frequency and magnitude of pulse-driven DOC events diminished as the seasonal thaw depth deepened, with inputs from terrestrial sources becoming major contributors to the DOC pool with decreasing snowmelt contribution to the hydrograph. A three-component parallel factorial analysis (PARAFAC) model indicated DOC liberated in late-season may be bioavailable (tyrosine-like). Combining Net Ecosystem Exchange (NEE) measurements indicate that the May Creek watershed fixes 142-220 g C m-2 yr-1 and only 0.40-0.57 g C m-2 yr-1 is leached out as DOC. Thus, the May Creek watershed and similar mature spruce forest dominated watersheds in the Copper River Basin are currently large ecosystem C sinks and exceeding C conservative. An understanding of DOC fluxes from Gulf of Alaska watersheds is important for characterizing future climate change-induced seasonal shifts.

Introduction

The boreal forest is an important biome because of its large role in the terrestrial carbon (C) cycle [1]. The boreal forest fixes up to 20% of the earth’s C [2], almost entirely controls the seasonal fluctuations in the northern hemispheric CO2 concentration [3], and retains over 30% of the planet’s total soil C, much of it in the form of permafrost [4, 5]. This pool of soil C is twice that of the contemporary atmosphere [6] and may have a series of fates as the global climate changes. The boreal forest C cycle is comprised of both gaseous fixation and losses (i.e. CO2 and CH4) and dissolved fluxes from soils into streams, rivers and long-distance oceanic sinks as DOC [7, 8]. In order to fully appreciate the relative magnitudes of trace gas C and DOC we require studies that are comprehensive in boreal forest C measurements and include detailed studies of the nature of DOC, as the quality of DOC may have important consequences for stream, river and near-shore marine food webs [9].

The boreal forest region of coastal Gulf of Alaska (GoA) has received much attention recently due to observations of massive glacier melt and permafrost thaw as manifested by rapid and accelerating climate change [10, 11]. Climate models predict up to a 40% increase in runoff from Alaska rivers by 2050 due to glacial melt [12], and over the coming decades an increase in river discharge is predicted to taper as glacial melt and permafrost thaw rates near completion [11]. There is however, considerable uncertainty over how this region and its glacial-soil-aquatic-marine system will change in the next 40 years. Some predict new habitats for salmon as the headwater streams become increasingly suitable for spawning as deglaciated landscapes undergo succession [13, 14], while conversely and/or simultaneously, permafrost and decreased river flows may lead to the loss of habitat as freshwater sources dry seasonally or permanently.

Changes in the source and magnitude of freshwater discharge in headwater streams are likely to alter downstream nutrient biogeochemistry in the coastal Gulf of Alaska (GoA). Recent studies have shown that glacial melt-derived nutrients fuel heterotrophic growth in the nearshore GOA [15-18]. Nutrient flux in catchments with significant glacial coverage is dominated by melting processes of these glaciers and the associated DOC pulse that is released from previously cryogenic- C, which is mostly lysed, readily biodegradable microbial cellular matter [16]. At locations where glaciers have recently receded, vegetation is characterized by nitrogen-fixing plants Dryas spp. and Shepherdia spp. and have little/no organic horizon to adsorb/sequester meltwater C. Over time as melting nears completion, C flux in these sites will less resemble glacial melt and shift to being dominated by atmospherically-fixed C with DOC character more resembling the new vegetation from spruce, birch, cottonwood, and willow.

Terrestrial vegetation contributes to stream water DOC through litter fall and/or exudation, and also generates soil organic carbon (SOC), which in turn can modify throughflow and groundwater storage characteristics. Litter inputs relevant to these processes consist of coniferous needles, twigs, stems, and logs, and waxy ericaceous leaves [19]. The understory is covered with thick mats of Sphagnum that have been shown to significantly contribute to DOC in the form of exudates [20]. Vegetation-derived SOC accumulates in the mineral layer as vegetation succession proceeds. Near-surface SOC in the organic horizon accumulates at faster initial rate, but reaches a maximum quickly, and is readily mobilized by disturbances, i.e. fire [21].

Northern ecosystems have carbon cycles that can be biogeochemically complex, whereby aqueous and gaseous C fluxes can determine the fate of land surface exchanges with the surrounding atmosphere and adjoining rivers and subsequent marine systems. Recently, for instance, [22] have indicated that the boreal system of Eurasia is contributing over 1.75 Tg C to the Arctic Ocean via DOC efflux and that over the course of the summer, the source of this C becomes progressively dominated by ancient permafrost C. The Yukon River, like the Ob, Yenesi, Ingidin, and Mackenzie is also a major catchment which transports C from interior Alaska to the Bering Sea [23], while smaller watersheds in south-central and southeastern Alaska contribute to the C cycle of near shore marine systems in Prince William Sound and the Gulf of Alaska, running North-South. However, while these DOC studies have quantified one form of C export, they are seldom accompanied by measurements which quantify the net CO2 exchange of the same ecosystem, nor the standing pool of C in trees [24, 25], making it difficult to appreciate the relative role of DOC export as a fraction of C sequestration, a more holistic assessment of watershed C cycling.

Fluorescence spectroscopy is a tool that is commonly utilized to characterize the chemical constituency of dissolved organic matter in freshwater systems. Recent efforts have been directed at characterizing other Alaskan freshwater aquatic ecosystems in the Tongass [15, 16] and Yukon [26-28], among others in the boreal [29-31]. To date, no studies have attempted to classify seasonal fluorescence characteristics anywhere in the Copper River Basin or the Wrangell-St. Elias region of Alaska.

At May Creek, a representative climax spruce forest at the headwaters of the Copper River, we examined the physical, biogeochemical, and hydrologic processes that control dissolved organic C export to the river and relate the export to CO2 flux values generated via eddy covariance [25]. A combination of expedition sampling and deployed in-situ sensors was used to observe dissolved components, and an eddy covariance flux tower was installed at the field site to observe Net Ecosystem Exchange (NEE) flux values over the same timeframe. The goal of this study was to determine the seasonal and inter-annual variability in the timing and magnitude of dissolved organic C export, determine how the chemical characteristics of dissolved C fluctuated seasonally and between water source types, and to relate DOC export to C fixation. Understanding these terrestrial-aquatic interactions is an important step in our depiction of these processes, and is useful for furthering upscaling efforts to determine basin-wide nutrient fluxes.

Materials and methods

Study site

The Copper River Basin watershed contains landscape types which have been previously delineated [17] as glacial, proglacial lake (early successional), non-glacial Boreal montane, and boreal lowland (late successional). To employ a space-for-time assumption for successional processes, we selected a non-glacial montane watershed for our study site. We hypothesized that atmospherically-fixed C accumulates rapidly in the standing biomass and accounts for the majority of the total C budget in headwater streams, as evidenced by annual Net Ecosystem Exchange (NEE) and DOC export masses.

May Creek is a climax spruce forested watershed (31.5 km2) at the headwaters of the Copper River (). It is underlain with discontinuous permafrost at depth of 50 cm. The basin ranges in elevation from 450–1400 m. The vegetation consists primarily of mixed white/black spruce, cottonwoods, peat, lichens, mosses, and willow shrubs (Salix), and the forest floor is overlain with a thick organic mat up to ~0.3 m. The soil consists of an organic layer underlain by clay sediment. The site is accessible only by fixed-wing aircraft from McCarthy, the nearest road-accessible town. To account for the spectrum of water types in the region, our sampling sites () included the following: 1) May Creek (MC, 61.3485667°, -142.6967833°), 2) a nearby subsurface-fed spring within the watershed (MC Spring, 61.3493000°, -142.6926333°), 3) Young Creek (61.3497500°, -142.7266000°), a neighboring creek that drains a larger (289 km2) non-glacial higher elevation (450–2,609 m,) watershed that includes both alpine grassland tundra and spruce forest at lower elevations, and 4) MC Wetland (61.3476333°, -142.7220833°), a lowland site that receives a combination of subsurface thaw and terrestrial throughflow.

Collection and analysis of water samples

Individual water samples were collected during the 2012 and 2013 field seasons at approximately biweekly intervals from May to September. Each sample was collected in an acid-rinsed 1 L borosilicate amber vial, triple rinsed with river water, submerged until filled, and capped with no headspace. Samples were filtered by passing water through a 0.45 um Acrodisk® syringe-driven unit. Any leachable DOC on the filter was flushed by passing 100 mL DI water through the filter and discarding this filtrate prior to collecting in sample vials. Samples were filtered into 40-mL VOC TraceClean amber vials for DOC analysis, 2-mL autosampler vials (National Scientific) for isotope analysis, and acid-rinsed PTFE bottles for fluorescence analysis, each of which were rinsed once with filtered stream water. Samples were transported to the laboratory on ice. DOC was analyzed using a Shimadzu TOC-VCHN analyzer at the University of Alaska Anchorage ASET laboratory. Water isotopic values (δ18O SMOW) were determined at the UAA Stable Isotope Laboratory with a Picarro WS-CRDS system utilizing a CTC Analytics PAL autosampler to deliver a liquid water sample via injection into a vaporization chamber. The vaporized water sample is fed into a closed cavity for analysis, during which a discrete quantity of laser light is introduced into the cavity and the absorption curve of the light is used to determine the isotopic ratio (18O/16O and 2H/1H) of the water sample.

Hydrological and in-situ sensor data

A Hach FH950 portable velocity meter and a top-end wading rod were used to estimate stream discharge during sampling campaigns at a reach of 200 cm. The stream was partitioned into 10-cm intervals and discharge calculated with two-point velocity measurements (0.2 and 0.8 depth) for stage (h) > 30 cm and a 1-point (0.6 depth) measurement for h< 30 cm. An Onset HOBO U20 pressure transducer was deployed in situ and stage calculated at 5-min intervals using atmospheric pressure readings taken at the eddy covariance tower. The stage-discharge relationship was log-log transformed and linearly regressed to yield a ratings curve. Discharge (Q) was determined by regressing stage to discharge.

Specific conductance was measured at 5-minute intervals with an Onset HOBO U24 conductivity data logger. The unit was calibrated several times throughout the season with a 1-point calibration standard. Fluorescent dissolved organic matter (fDOM) is the fluorescent component of chromophoric dissolved organic matter (CDOM), a component of the DOC pool that can be monitored in real time using in-situ devices. We took fDOM readings in May Creek at 1-minute intervals with a Cyclops7 CDOM sensor (Turner Designs, Sunnyvale, California) connected to a Campbell data logger. The unit was suspended on a rod at 30 cm depth. All fDOM data (collected in mV) correspond to fluorescence at excitation/emission wavelengths 325/470 nm calibrated to ppb PTSA (1,3,6,8-pyrenetetrasulfonic acid) with serially-diluted standards, per manufacturer recommendation. Data were time-averaged to 5-minute intervals to synchronize with the U24 and U20 sensor data. All units deployed in May Creek were checked for fouling and debris at each sampling campaign and cleared as necessary.

Spectrofluorometric data

Fluorescence readings were taken on an Aqualog spectrofluorometer (Horiba Scientific, Edison, NJ) equipped with an internal UV spectrophotometer and analyzed according to the procedures of [32]. Briefly, acid-rinsed 1-cm cuvettes were triple rinsed with Millipure water and then filtered sample prior to loading in the instrument. Excitation emission matrices (EEMs) were generated over excitation 240–450 nm (5 nm increments) and emission 300–600 nm (2 nm increments). Millipure water blanks were collected every 5 samples in addition to a daily Raman blank that was used for normalization under the water Raman curve to adjust for instrument-specific bias. Concentrated samples (>5 ppm DOC) were diluted 10x and corrected by dilution factor. Integration time was 0.1 seconds. All samples were collected in S/R ratio mode, blank subtracted, inner-filter corrected using the manufacturer’s supplied software, and Raman-normalized. Method blank checks consisting of DI water transported to the field revealed no contamination.

Fluorescence Index (FI) values were calculated according to the procedures of [33]: For Excitation = 370 nm. FI is commonly used as a terrestrial vs. biogenic index, with lower values between 1.2–1.5 indicating presence of highly terrestrial-derived fulvic acids and higher values indicating a more microbial-derived source [34]. Humification Index (HIX) values were calculated according to [35] as For Excitation = 255 nm. HIX values range from 0–1, with values closer to 1 indicating a highly humified character and lower values indicating a highly proteinaceous (tyrosine or tryptophan) character [35]. Parallel factor analysis (PARAFAC) was conducted on water samples using the drEEM toolbox for Matlab (Mathworks, Natick MA) [36, 37]. A three-component model was validated by split half analysis and through visible inspection of spectral loadings [38]. The percent relative contribution of each component calculated to the sum of total fluorescence is reported for each water sample.

Eddy covariance flux measurements

Tower-based ecosystem exchange measurements were quantified using the eddy covariance (EC) technique [39, 40] on a 30-meter-tall scaffold tower. Measurements of the three-dimensional wind and sonic temperature were done with an ultrasonic anemometer (CSAT3, Campbell Scientific, Inc., Logan, UT, USA) while high-frequency fluctuations of CO2 and water vapor were done with a closed-path infrared gas analyzer (IRGA; LI-7200, LI-COR, Inc., Lincoln, NE). A 0.95 cm outer diameter and 100 cm long insulated copper intake tubing was used, with the air intake cup 15 cm behind and 1 cm below the center control volume of the sonic anemometer. The sampling flow rate through the IRGA was maintained at 15 LPM. The IRGA was calibrated twice monthly with a CO2-free air and 450 ppm CO2 gas standards, and a portable dew point generator (LI-610, LI-COR, Inc.). The turbulence parameters and CO2 and water vapor fluctuations were sampled at 20 Hz with the raw time series data stored on a data logger (LI-7550, LI-COR, Inc.).

Measurements at the top of the tower included incoming and reflected short- and long-wave radiation (NR01, Hukseflux, Delft, Netherlands), photosynthetically active radiation (PAR; LI-190SB, LI-COR, Inc.), air temperature and relative humidity (HMP155, Vaisala, Helsinki, Finland), wind vector (034B, Met One Instruments, Grants Pass, OR), and barometric pressure (PTB101B, Vaisala). Wind profile measurements (014A, Met One Instruments) were done above the ground, while CO2 and water vapor concentration profiles were measured within the canopy (AP200, Campbell Scientific, Inc.). Ground measurements included soil temperature (Type-T thermocouples, Omega Engineering, Stamford, CT, USA), ground heat flux (HFP01, Hukseflux), and soil moisture (EC-10, Decagon Devices, Pullman, WA, USA). The slow response meteorological sensors were sampled at 10-s intervals and stored as half-hourly averages. Precipitation data were obtained from the National Resource Conservation Service (NRCS) SNOTEL site located at May Creek.

Raw data processing was done using EddyPro® open source software (v5.1.1; LI-COR, Inc.) which included spike detection and removal [41], and physical thresholds for out-of-range values. Fluxes of CO2, H, and λE were calculated as half-hourly block averages with a double coordinate rotation [42] and time lag compensation using a covariance maximization method. Since a closed-path IRGA was used, the gas concentrations were expressed as mixing ratios and therefore was not necessary to apply the WPL [43] corrections to compensate for air density fluctuations [44, 45]. Spectral corrections were performed for both high-pass [46] and low-pass [47] filtering effects. We used the crosswind-integrated flux footprint model of [48] to estimate the upwind sampling area. Additional quality checks on the processed fluxes were evaluated using the overall quality flags described in [49]. Calculated fluxes were u*-filtered [50, 51] with a threshold of 0.1 before gap-filling [52, 53] was performed with the Marginal Distribution sampling method [54] using REddyProcWeb (https://www.bgc-jena.mpg.de/bgi/index.php/Services/REddyProcWeb).

Statistical analysis

Statistical tests were performed using JMP 11.0 (SAS institute, Cary, North Carolina). Unless otherwise stated, group-wise mean comparisons were performed via Student’s t-test, while regression analysis is reported as analysis of variance (ANOVA) test of significance for a linear model. Significances are noted at 95% confidence.

Results and discussion

The Copper River basin of interior Alaska is one of the larger in all of Alaska and North America, being in the top 10 and on par with those such as the Yukon, McKenzie, Columbia, Mississippi and Missouri. It is ~ 7.3 million hectares (about the size of West Virginia and is the largest single freshwater source to the GoA, with an annual discharge of ~65 km3; that is second in Alaska only to the Yukon River (11). At this scale, biogeochemical processes are thus exceedingly important while having complex hydrologic and abiotic factors that govern dissolved C export from surface and subsurface flowpaths and atmospheric C exchanges between this boreal forest and the atmosphere. Here, we discuss and present a perspective on the C cycle of this boreal forest watershed that combines dissolved organic C export with fixed C inputs.

DOC export and water isotopes

DOC concentrations at May Creek were highest during the spring freshet when stored C in the snowpack was liberated to the neighboring flow path (). For 2012 and 2013, DOC levels were 10.1 and 19.6 mg/L on May 14th, the first sampling date for each year. These values decreased seasonally as samples collected on August 20th were 4.5 and 0.94 mg/L for 2012 and 2013. For Young Creek, the 2012 and 2013 values were 6.7 and 11.2 mg/L on May 14th decreasing to 1.02 and 1.62 mg/L on August 20th. MC Spring DOC values increased slowly during 2012 (p<0.0122) from 0.68 mg/L on May 14th to 0.83 mg/L on August 20th. For 2013, MC Spring DOC did not significantly increase seasonally (p<0.1152). The reported seasonal mean value is 1.07 mg/L, higher than the 2012 seasonal mean of 0.81 mg/L (p<0.0091).

Seasonal and interannual trends.

(A) DOC concentration and (B) water δ18O for Creek (MAYC and YNGC), subsurface fed spring (MCSP), and mixed source (MCWT) sites.

Riverine C stocks and fluxes may change over time as plant succession generates higher pools of soil organic carbon (SOC) in the organic horizon and releases these pools seasonally. Export of these pools in regions of frozen soil as DOC () is facilitated by seasonal weather patterns. During the spring freshet, winter-accumulated snow melts quickly and water runs overland where it dissolves the abundant, less-degraded carbon present in shallow soil layers before entering streams, releasing stored nutrients and flushing the terrestrial landscape. This process does not involve a deep penetration because soil is still frozen, which constitutes an impermeable barrier prior to thawing [30]. With an increase air temperature and thaw depth, subsurface flowpaths through soils migrate deeper and the surface water nutrient composition changes when soil organic C and porewater DOC is flushed. Leaching occurs laterally along seasonal frost or discontinuous permafrost boundaries until pooling or draining into a stream/river basin.

Water δ18O at May Creek () increased seasonally in 2012 (p<0.0001), from -22.68 ppm on May 14th to -21.99 ppm on August 20th. For 2013, water isotope values did not increase seasonally and expressed a mean of -21.64 ppm. A similar trend was present for Young Creek, with 2012 values increasing from -23.09 ppm on May 14th to -22.69 ppm on August 20th; the first two seasonal values in the month of May were high. No seasonal enrichment was observed in 2013; a mean value of -21.69 ppm is reported. Water source may be inferred from, among others, shifts in the δ18O and δ 2H values for stream and river water [55-58]. Typically, when snow melt is the dominant source for river systems, δ18O and δ2H values are depleted compared to seasonal transitions into rainfall driven hydrographs that exhibit enriched δ18O and δ2H values, following the classic seasonality of precipitation isotopes [57, 59]. River water δ18O values in our study region increased seasonally in 2012 due to a shift in source from snowmelt to an active layer thaw and precipitation event-driven hydrograph (). In 2013 however, this temporal pattern was not observed with surface water values being relatively consistent at the seasonal mean of -21.6 ‰ from May to September. This consistency in the surface water isotope values is likely the result of a rapid snowmelt and warm air temperatures from May-July () preventing seasonally-thawed soil porewater from being sufficiently flushed. Our observations from 2012 indicate a snowpack with melt water input being more ephemeral and isotopically enriched rainfall becoming a larger fraction of river water sources as the growing season progressed.

Hydrological characteristics and in-situ observations

Peak flow was observed between late May and early June, with annual maxima of 0.54 cms on June 12th (2012) and 0.63 cms on May 28th (2013). Discharge tapered through the end of August (), with minima of 0.13 cms for August 23rd (2012) and 0.08 cms for August 16th (2013). Discharge tended to occur in pulses and correlated strongly with precipitation events in 2012. For 2013, the hydrograph was driven by one large pulse in late-May and with several minor pulses throughout the season. The major nutrient pulse for 2013 differed from others in 2012 and later in the season in that the pulse was accompanied by a decrease in specific conductance and fDOM (). Snow water equivalency reached 0 mm on April 28 and May 12 for the 2012 and 2013 water years, respectively (). Soil Temperature (5 cm) values recorded during 2013 revealed a seasonal increase from 0 ºC on May 17th to a maximum of 12.5 ºC on July 29th.

Real-time in-situ hydrologic data collected from sensors deployed in May Creek over 2012 and 2013 at 30 cm depth.

Data are 5-minute averages of (A) discharge (in m3 s-1), (B) fDOM (in ppb PTSA standard), and (C) specific conductance (in μS-cm).

Specific conductance (SpCond), a measurement of the electric current in the water carried by ionized substances, increased steadily throughout both seasons at May Creek (). We noted minima of 298 μS-cm on June 12th (2012) and 193 μS-cm on May 28th (2013). Maxima were observed at 523 μS-cm on August 26th (2012) and 559 μS-cm on August 16th (2013). The release of such ionized compounds were diluted during precipitation events and negatively correlated with discharge spikes.

Seasonal fDOM levels correlated with DOC values obtained from the grab samples (Figs ). Coupled with episodic pulsing events, fDOM values decreased seasonally both years and correlated strongly with discharge. For both seasons, fDOM values fluctuated diurnally with pronounced amplification in mid-May tapering off to low amplification in late July. For May 17th, a representative spring freshet date, mean values of 281 ppb PTSA and 0.34 cms for 2012 and 346 ppb PTSA and 0.48 cms for 2013 were observed. As the growing season progressed, both fDOM and discharge decreased. On July 29th, a representative summer date at maximum soil T, fDOM and discharge were observed at 144 ppb PTSA and 0.24 cms for 2012 and 116 ppb PTSA and 0.11 cms for 2013, respectively. These values correlated (both positively and negatively) with episodic precipitation events; For 2012, both fDOM and discharge increased following precipitation as nutrients stored in a relatively shallower snowpack were released following multiple rain events (). For 2013, these values negatively correlated during one large multi-day pulse event from May 23rd to June 2nd, when a deeper snowpack on top of a delayed snowmelt () was observed. fDOM values at this time decreased in response to increasing discharge and decreasing specific conductance.

Seasons with early snowmelt, as in 2012, create highly efficient conditions to leach extractable DOC from soil mineral layers. In 2013, a combination of delayed, rapid snowmelt with an onset of low precipitation and warm air temperatures from May-July () may have led to low soil moisture with the occurrence of small, isolated pockets of flow paths that allowed the infrequent precipitation to advect into streams without interacting with soil C. Our findings for 2013 support this hypothesis, since the major hydrograph pulse event involved a drop in fDOM values, an enrichment δ18O values, and a drop in SpCond. Seasonal flow path structure is known to have an impact on runoff rates and DOC export [60, 61]. Extreme events (i.e. outburst floods, slope failure from permafrost melt, precipitation pulses, etc.) play a critical role in stream ecosystems, and extreme values may be more important than mean values since the absence of continuously-recording instruments collecting such data would leave the detection of extreme events to chance. Consequently, what may matter most is to be able to measure changes in the frequencies and magnitudes of these events, rather than changes in mean values. Much of the variability observed in water chemistry parameters in the Wrangell-St. Elias National Park has been noted as pulse-dependent [62], and indicates for the first time this real-time interdependency between fDOM, discharge, and specific conductance. The fluorescent fraction of the CDOM pool, fDOM has been implicated recently as a proxy for DOC concentrations in the marine environment [63], and recent efforts [64-66] have been directed at determining high-resolution continuous DOC flux for freshwater systems. The fDOM data from this study reveals pulse events that can both increase and decrease DOC following precipitation, and that this is presumably dependent on snowmelt timing and seasonal flowpath structure.

Seasonal fluorescence characteristics

May Creek FI values () increased seasonally for both 2012 (p<0.0071) and 2013 (p<0.0049), from 1.54 on May 14th to 1.64 on Sept 21st (2012) and from 1.51 on May 14th to 1.65 on August 20th (2013). Humification Index (HIX) values at May Creek ( did not decrease seasonally, with mean seasonal values of 0.901 and 0.907 for 2012 and 2013. For Young Creek, the same seasonal trend was observed, with an increase in FI values for both 2012 (p<0.0339) and 2013 (p<0.0002) from 1.61 to 1.66 (2012) and from 1.56 to 1.77 (2013). HIX values at Young Creek did not decrease in 2012 (seasonal mean value 0.8427) but did decrease significantly in 2013 (p<0.0004) from 0.8994 on May 14th to 0.7456 on August 20th. Both creek systems exhibited lower FI values and higher HIX values than the sites receiving predominantly subsurface-derived water. MC Spring exhibited mean FI values of 1.79 and 1.83 for 2012 and 2013, with no seasonal increase (p<0.6288 and p<0.5892 respectively). HIX values at MC Spring did not change seasonally (p<0.9530) in 2012, with a seasonal mean value of 0.5167. However, HIX did increase significantly in 2013 (p<0.0116) from 0.6140 to 0.6770. Thus, both creeks provide a mostly humic terrestrial contribution of DOC with a seasonally decreasing terrestrial contribution later in the fall, presumably due to an increasing contribution from groundwater. MC Spring exhibited consistently more biogenic DOC and less humified character than all other sites, with MC Wetland showing similar MC Spring-like character in 2012 and a mixture of sources in 2013.

Seasonal and interannual fluorescence characteristics by site for 2012 and 2013.

(A) Fluorescence index and (B) Humification index.

The fluorescence characteristics (FI and HIX) of filtered water are commonly used to infer the relative composition of terrestrial vs. biogenic composition (FI) as well as the relative extent of highly aromatic, humified C to proteinaceous C (HIX). The seasonal values of each of these indices by site, as presented in , suggest subsurface-dominated flowpath regions export DOC of a different origin from highly terrestrial-draining surface water sources. From May-September FI values increased at both May Creek and Young Creek, indicating that DOC becomes increasingly microbially-derived as opposed to terrestrially-derived [67]. This is consistent with the seasonality observed traits of the Yukon River Basin [27] and this DOC presumably becomes increasingly ancient, as inferred by [26]. DOC output in May Creek is more terrestrially-derived because microbial activity is low due to cold temperatures. The subsurface source MC Spring revealed significantly lower HIX values than May Creek and Young Creek, which indicates high H:C ratios as humification is not fully progressed [67] and derived from extracellular release and leachate from bacteria. MC Spring is also higher in ionic strength as confirmed by specific conductance readings (). Interestingly, HIX values for Young Creek decreased seasonally in 2013, while MC Spring increased with values converging towards the end of the season. DOC values from 2013 season revealed a rapid mobilization of dissolved C stocks in May, with lessening snowmelt and shallow active layer contribution as summer progressed. Young Creek is a less forested and a larger non-glacial fed watershed compared to May Creek with steep banks eroding sites of discontinuous permafrost, which when thawed, pulse into the creek. Young Creek HIX values indicates DOC contributions are less-humified than May Creek and that they increase seasonally, a pattern which is not observed at the highly-forested May Creek. This is presumably due to the larger standing forest biomass and relative contribution to throughfall C in May Creek. Young Creek drains a higher elevation basin, much of which is above treeline. Although humic-like components still dominate the DOC pool in May and Young Creeks, a deepening active layer in a warmer boreal forest could potentially release additional biogenic-C in to the riverine system and potentially into the downstream rivers such as the Copper River and into the GoA with cascading consequences for in-stream biology and oceanic food webs.

We obtained a validated 3-component PARAFAC model for the EEMs data from both sampling years (. C1 is generally referred to as Peaks A and C and are described as regions found in freshwater ecosystems and contain fluorescence signatures related to highly degraded, aromatic DOM [68-71]. C2 is described in the literature as Peak M, or microbial derived humic peaks, which are comprised of relatively aliphatic and low molecular weight DOM [68, 69]. C3 region is generally described as tyrosine-like fluorescence [72, 73]. Plotting HIX vs C1 () shows that as HIX increases, so does the relative contribution of C3, which makes sense given that higher HIX values describe compounds that highly aromatic with high oxygen, hence the longer emission wavelengths. Plotting HIX vs C3 () shows that as HIX decreases, the relative contribution of C3 increases. Again, this makes sense given that low HIX values are associated with newly formed compounds that are highly aliphatic and contain low oxygen, hence the shorter wavelengths. The strong positive correlation with HIX and C3 from May to August for both years suggests that aliphatic compound are produced at the end of the season. Similar fluorescence signatures found from earlier studies (similarity score > 0.98 in OpenFluor), overlapped with identified components from this study [74]. Component 1 matched fluorescence signatures from 26 studies, C2 3 studies [75-77] and C3, 2 studies [78, 79]. Most notable were the spectral properties of component 1 at ex max 255–260 nm and em max at 448–480 nm are humic-like signatures found in forested systems and matched C2 from [80, 67]. Component 2 with spectral properties of Ex325/Em396 nm was compared to C6 from [81], C2 from [75] and C2 from [76] to an extent. This fluorescence signature is often referred to as the “M Peak” and is ubiquitous in a wide range of environments [68, 69]. Component 3, referred to as protein-like fluorescence at ex max 270–275 nm and emission 304–312 nm, overlapped with spectral signatures of C3 from [78] and C3 from [79]. Comparing the spectral properties with other studies in boreal forested systems, this component matched protein-like fluorescence component C13 from [80]. Consistent with [26, 32, 82, 83], we identify the fluorescence components in this watershed as phenols, lignin, tannins, gallic acid and amino acid tyrosine.

Parallel Factorial Analysis (PARAFAC) of fluorescence dataset.

(A) PARAFAC loadings for a validated 3-component model. (B) Strong correlations to humification index (HIX) are presented.

Our EEMs exhibited excitation/emission fluorescence in regions corresponding to UVA (ex 290–325 nm, em 370–430 nm) and UVC (ex 320–360 nm, em 420–460 nm) noted in [67]. Using the strong correlation and inverse correlation of C1 and C3 respectively to HIX () combined with seasonal trends shown in , we can identify that May Creek and Young Creek were dominated by this fluorescence character, with Young Creek showing decreased humification by August and September. MC Spring however, displayed a protein-like signal (tyrosine-like ex 270–275 nm, em 304–312 or tryptophan-like ex 270–280 nm, em 330–368 nm) throughout the season, with an increase in UVA/UVC DOC character in fall. The mixed source MC Wetland exhibited MC Spring-like (low humification, high FI) DOC in 2012, but was creek-like (high humification, lower FI) in 2013. As also purported by FI and HIX indices, Young creek exported less UVA and UVC-like character in late summer, due to less relative tree cover and more contribution from mobilized thawing processes. As air temperatures and thus thaw depth increase, ionized chemical species associated with specific conductance () and of biogenic origin () sequestered at lower depths are mobilized following precipitation events and subsurface advection. With warming temperatures and a deepening active layer, we would expect an increased export resembling this chemical character.

Net ecosystem exchange, CO2 flux, and correlations to DOC

NEE values for this boreal forest show consistency in the diurnal patterns being a net source of CO2 to the atmosphere during the nighttime periods, and a strong sink during mid-day, regardless of month (). The magnitudes of nighttime respiration changed from 2.2 μmol m-2 s-1 in May 2012 to 6.0 μmol m-2 s-1 in July. The mid-day peaks of NEE are lowest in autumn (September) at -4.1 μmol m-2 s-1 and highest in July at -11.9 μmol m-2 s-1. The sharp diurnal switch from source to sink of the monthly averages also reflect the seasonal transition from shorter to longer days at the higher latitudes where by the transition becomes very abrupt in June and July near the summer solstice and occurs hours earlier than in May and again in September.

During the two years of study, the rates and magnitudes of C accumulation (), were lowest and slowest in 2012, with the maximum amount in 2012 being 142 g C m-2. In 2013 however, the C accumulation rate began earlier and never plateaued during the measurement period reaching a maximum of 220 g C m-2. This supports the hypothesis that atmospherically-fixed C is accumulating in the standing biomass and accounts for the majority of the total C budget in headwater streams in this south central Alaskan system. Taking mean annual DOC and discharge values into account, our data indicates DOC export for May Creek from May to September to be 0.57 g C/m2 for 2012 and 0.40 g C/m2 for 2013. These values compare with other late successional catchment-derived C of 1.64 g C/m2 for the Yukon River Basin [84], 1.5–8.3 g C/m2 for a bog of Ontario, Canada [85] and 4.6–17.6 g C/m2 for a central Siberia site underlain with continuous permafrost [86]. According to [84], DOC export values vary considerably by catchment properties, and large values in excess of 20 g C/m2 yr are typically seen when precipitation greatly exceeds evapotranspiration and/or upland area exceeds wetland area. Conversely, DOC export is low, < 10 g C/m2 yr, when catchment is dominated by peatland, relief is negligible, and/or precipitation and evapotranspiration values are similar [84, 85]. In May Creek, it is believed that low relief and dense standing biomass contribute to the low values compared with other studies.

Net ecosystem exchange of may creek watershed for 2012 and 2013.

Data are half-hourly averages, expressed as g C/m2, collected from eddy covariance flux tower installed on-site.

CO2 exchange in the May Creek Watershed exhibited diurnal and seasonal patterns that are consistent with ecosystems globally and in forests throughout the northern hemisphere [24, 87, 88]. Irradiance levels fluctuate diurnally from being zero for up to 8 hrs per day to maximum values of >1500 μmol m-2 s-1 during mid-day; driving in large part the variation in GPP and thus NEE we observe daily as reported for other boreal and temperate forests [87-89]. Carbon sequestration over the course of the growing season we observed are within the ranges of other North American boreal forests, including the sites of the BOREAS program and other Canadian boreal sites [87, 90, 91]. However, none of these programs have placed C sequestration in the context of both the gaseous and aqueous exchanges of C in a headwater stream; in our system the gaseous exchange dominates the budget. However, if boreal watersheds are less forested and more dominated by bogs and fens, the relative contribution of gaseous to aqueous C cycling maybe significantly different than what we observed for May Creek, in the interior of south central Alaska [91]. C fixation may be lower, but DOC efflux may also be lower.

Understanding the magnitude of DOC export from boreal watersheds in the context of the watershed C net CO2 exchange is paramount to recognizing the scope and scale of basin-wide C biogeochemical cycling; but seldom is this accomplished as it requires an interdisciplinary approach and a holistic perspective. Our results indicate that this boreal forest ecosystem is exceedingly C conservative during the growing season and acting as a very strong overall C sink. Seasonal weather patterns appear to be a greater driver of the observed variability in DOC quantity and composition than does cumulative C fixation, with DOC export constituting less than 5% of summer NEE. This exceedingly tight C cycle is reflective of the storage of C in standing biomass, in forest understory species and in the formation of soil C pools that are relatively recalcitrant. This strong C-sink capacity of these watersheds are however venerable to catastrophic events that may alter DOC export and C fixation such as fires, which are know to very rapidly oxidize forest biomass, remove understory protection of soil C pools and result in large C source traits over short and longer time periods until successional processes restore the forest ecosystem and the thermal stability of the underlying soil c pools [92, 93].

Conclusion

In this study, we examined the seasonal hydrologic and atmospheric processes that control C exports and storage in a late-successional forest ecosystem and anticipate such mobilization in currently glaciated areas of the basin to be similar to May Creek in the coming years as the significant glacial coverage (18% of the basin, [94] melts, permafrost thaws, and vegetation advances). Pulse events and seasonal weather patterns in this region have a major effect on the timing, magnitude, and chemical characteristics of exported DOC, and understanding these effects is paramount to predicting future C flux.

Snow water equivalent (SWE), Precipitation accumulation, and air temperature data for periods of spring freshet.

Data are expressed for (A) 2012 and (B) 2013 field seasons. Reproduced from SNOTEL site 1096: May Creek.

(TIF)

Click here for additional data file.

Seasonal cumulative precipitation, snow water equivalent, and air temperature values for May Creek.

Data reproduced from SNOTEL site 1096: May Creek.

(TIF)

Click here for additional data file.

Diurnal CO2 flux (μmol m-2 s-1 CO2) characteristics at May Creek for 2012 field season.

(TIF)

Click here for additional data file.

May creek basin grab sample data.

(XLSX)

Click here for additional data file.

May creek sensor data.

(XLSX)

Click here for additional data file.

May creek basin fluorescence data.

(XLSX)

Click here for additional data file.

May creek net ecosystem exchange data.

(XLSX)

Click here for additional data file.

May creek snotel data.

(XLSX)

Click here for additional data file.

10 Sep 2019

PONE-D-19-14655

Pulse driven DOC inputs control the C exports in May Creek: a late-successional headwater boreal forest watershed of the Copper River Basin, Alaska

PLOS ONE

Dear Dr. Tomco,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

We would appreciate receiving your revised manuscript by 11 October 2019.  When you are ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter.

To enhance the reproducibility of your results, we recommend that if applicable you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). This letter should be uploaded as separate file and labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. This file should be uploaded as separate file and labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. This file should be uploaded as separate file and labeled 'Manuscript'.

Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.

We look forward to receiving your revised manuscript.

Kind regards,

Robert L. Bradley

Academic Editor

PLOS ONE

Journal Requirements:

1. When submitting your revision, we need you to address these additional requirements.

Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

http://www.journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and http://www.journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. Please include in the Methods section the names of the locations, and provide geographic coordinates for the data set.

3. We note that Figure  [1] in your submission contain [map/satellite] images which may be copyrighted. All PLOS content is published under the Creative Commons Attribution License (CC BY 4.0), which means that the manuscript, images, and Supporting Information files will be freely available online, and any third party is permitted to access, download, copy, distribute, and use these materials in any way, even commercially, with proper attribution. For these reasons, we cannot publish previously copyrighted maps or satellite images created using proprietary data, such as Google software (Google Maps, Street View, and Earth). For more information, see our copyright guidelines: http://journals.plos.org/plosone/s/licenses-and-copyright.

We require you to either (1) present written permission from the copyright holder to publish these figures specifically under the CC BY 4.0 license, or (2) remove the figures from your submission:

You may seek permission from the original copyright holder of Figure [1] to publish the content specifically under the CC BY 4.0 license.

We recommend that you contact the original copyright holder with the Content Permission Form (http://journals.plos.org/plosone/s/file?id=7c09/content-permission-form.pdf) and the following text:

“I request permission for the open-access journal PLOS ONE to publish XXX under the Creative Commons Attribution License (CCAL) CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). Please be aware that this license allows unrestricted use and distribution, even commercially, by third parties. Please reply and provide explicit written permission to publish XXX under a CC BY license and complete the attached form.”

Please upload the completed Content Permission Form or other proof of granted permissions as an "Other" file with your submission.

In the figure caption of the copyrighted figure, please include the following text: “Reprinted from [ref] under a CC BY license, with permission from [name of publisher], original copyright [original copyright year].”

If you are unable to obtain permission from the original copyright holder to publish these figures under the CC BY 4.0 license or if the copyright holder’s requirements are incompatible with the CC BY 4.0 license, please either i) remove the figure or ii) supply a replacement figure that complies with the CC BY 4.0 license. Please check copyright information on all replacement figures and update the figure caption with source information. If applicable, please specify in the figure caption text when a figure is similar but not identical to the original image and is therefore for illustrative purposes only.

The following resources for replacing copyrighted map figures may be helpful:

USGS National Map Viewer (public domain): http://viewer.nationalmap.gov/viewer/

The Gateway to Astronaut Photography of Earth (public domain): http://eol.jsc.nasa.gov/sseop/clickmap/

Maps at the CIA (public domain): https://www.cia.gov/library/publications/the-world-factbook/index.html and https://www.cia.gov/library/publications/cia-maps-publications/index.html

NASA Earth Observatory (public domain): http://earthobservatory.nasa.gov/

Landsat: http://landsat.visibleearth.nasa.gov/

USGS EROS (Earth Resources Observatory and Science (EROS) Center) (public domain): http://eros.usgs.gov/#

Natural Earth (public domain): http://www.naturalearthdata.com/

4. Please include captions for your Supporting Information files at the end of your manuscript, and update any in-text citations to match accordingly. Please see our Supporting Information guidelines for more information: http://journals.plos.org/plosone/s/supporting-information.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Partly

Reviewer #2: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: I Don't Know

Reviewer #2: I Don't Know

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: No

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: GENERAL COMMENTS:

The manuscript provides a useful carbon dataset that helps link forest/overland carbon flux with stream export of dissolved organic carbon (DOC). These data and findings will be of interest to a broad audience working on global to boreal carbon cycles. The authors present an interesting story around seasonal drivers of carbon flux and help highlight how growing condition (light and temperature) and water (precipitation and snowmelt) influence differently dissolved and gaseous carbon flux. The manuscript has many positives and I think a few modifications are needed to help the manuscript more fully explore the data and bring together the full story. I have three main suggestions that I think would make the manuscript more useful.

1) Make full use of the EEMs --- Given that you have the full EEM, why only focus on FI and HIX. Peak picking (Paulo Coble's work) or PARAFAC might provide more useful insight when tested through a multivariate DOM method. SUVA and spectral slope would also be useful measure to understand how the complexity of DOM changes across the growing season. The discussion starts to exam the EEMs visually, but I think it would be useful to take the analysis farther by using the full set of DOM parameters that optical chemistry can generate. If these data are then analyzed through multivariate analysis, one can speak about the DOM as a pool and detailed univariate explanations of all the variables could be avoided.

2) Statistics, questions, and hypothesis tested needs more detail and justification --- It is unclear to me what test and question were used to generate p-values in this study. More detail is needed for the reader to understand what data are being compared, why, and how. For example, how did you define season? How are you testing rate of change across time? I think better defining the studies hypothesis and question with the link statistics will make the results easier to follow with respect to your broader story.

3) Combining results and discussion sections will make it easier for your reader to follow the data and the story --- Overall the results are written clearly and as individual statements I understood the pattern and result being described. I think what I struggled to understand was how all the variables and different temporal scales that where tested fit together as one story. The discussion and abstract did a fantastic job describing the overall story but at times I struggled to link the broad story to the specific results. I think to help the reader move through your complex data set, combining the results and discussion section would communicate more clearly the big picture story and the result that supports that big idea. This might shorten then manuscript a little as well better highlight the interpretation of the result alongside the result.

SPECIFIC COMMENTS

Abstract: The abstract is clear and provides a nice snapshot of the study and its relevance. I did not fully understand the last sentence of the abstract. I am not certain what shift is being referenced and I am not certain in what respect this carbon mass is integrated and important. Could the sentences be revised to be more exact?

Key Points:

1) Does this include the frozen period? It seems odd that CO2 fixation wouldn't change between growing and dormant seasons. This point might need to be qualified with the timer period of the study

3) Browning has multiple means. Since your paper is about DOC, could you use another word to describe the drought impact on the forest?

4) Did you mean carbon or DOC rather than nutrient?

Introduction: the introduction is framed well and does a nice job setting up the study and the need for the combined study of carbon gas and DOC flux. The introduction was a little long. You could likely remove the paragraph about DOM fluorescence without loss to your overall story. To me this approach is now very common and it's OK to not focus on how DOM was characterized but focus instead on what knowledge around the DOM characteristics can tell us about C cycles in this study

L150-L155: I think MC needs to be define in text. MC could also be used when describing the actually creek in text as it was in the figures

L189: Did you mean fluorescence rather than absorbance?

L338-L339: The July value is nearly 3x higher. This seems a much larger change in magnitude than the water isotopes and fDOM, FI, and HIX values above, which were interpreted as significantly different. Why is this considered stable?

L444: This should be a statistical inspection using multivariate methods to show how the DOM pool changes. These visuals help show the pattern, but I think the data need to be analyzed statistically, in order to show the magnitude and direction of change.

L459-L462: This hypothesis needs to be stated in the methods with statistical test explaining how the hypothesis was tested

Figure 5: Could error bars be added to each point to show how much day to day or night to night variation is displayed in each monthly mean at each time point?

Figure 7: These are useful plots. I think the data within the EEMs needs to be explored in the results section through PARAFAC or peak picking. Otherwise, this rich data set isn't explored to its fullest. In terms of the plot, the xyz axis font is to small and not a very good resolution. I zoomed in and the text was fuzzy and difficult to read. You should make a note that the z-scales vary from plot to plot and specify the units (RU, QSU, etc...). Conversely, you could standardize each plot by the max value and this would allow all plots to be displayed on the same relative scale. The second option would draw the reader to changes in characteristics of DOM between spring and fall rather than highlight changes "concentration"

Reviewer #2: Review PlosOne.

This manuscript presents the results of a two-year study conducted in a boreal watershed of Alaska. The authors describe the fluxes (magnitude, seasonal and annual variations) and chemical characteristics of DOC in four stream types of watersheds characterized by different vegetation cover and soil characteristics. They also measured CO2 fluxes for one growing season at one of the sites to assess the significance of DOC exportations as compared to C fixation through photosynthesis. This study was motivated by the fact that global warming may increase DOC outputs and reduce CO2 fixation in these ecosystems, which may turn them into C sources. This is an interesting and important topic that researchers must tackle. The main conclusion is that the watershed does not export large amounts of DOC as compared to C fixation. Therefore, in contrast with their hypothesis the watershed is still a C sink and highly C conservative. They also found that most DOC efflux occurred as a pulse during snow melt in May.

The amount of work, the sampling effort (four streams), the methods (eddy covariance, fluorescence spectroscopy) and analyses (stable isotopes, DOC concentrations) deployed for this study are considerable. The authors described the materials and methods thoroughly and clearly, and the bibliography is thorough.

The results are interesting and deserve publication. However, I was disappointed by the way the authors presented, interpreted and discussed their results. First, the figures could be of better quality. I think for instance that the use of dots instead of line prevents from visualizing the seasonal trends. The resolution of Fig 7 is too low for the x- and y-labs to be readable. I think that the figures should also be reorganized to facilitate the discussion. For instance, why not grouping results with similar trends in a single figure (e.g., fDOM with DOC flux, and delta 18O with Q)?.

The authors insist on the importance of precipitation and temperature on DOC fluxes in this type of ecosystem but did not include the data with the ms and didn't discuss their results properly. I would bring some of these data from the supplementary materials to the main text and maybe remove the Fig 5 which does not provide particularly insightful information. The fact that the vegetation is a source of CO2 at night and a sink during the day is quite obvious.

In addition, I found the discussion section not very well written and organized. I think the interpretation of the results could be clearer and more convincing. They, in my opinion, do not do a satisfactory job at answering basic questions that come to mind after reading the results section (see in comments below).

The abstract as well as the introduction describe the study sites and the context thoroughly, but the abstract does not emphasize the results and the main conclusions of their work, which in my opinion should be clarified. I also think that the results could have been discussed more thoroughly and in a clearer way. For instance, the fDOM results are barely discussed.

Here are my comments for each section of the manuscript.

Abstract:

The first paragraph could be shortened (in my opinion, only the first and last sentences should be conserved) and the authors should put more emphasis on their results and conclusions. Several sentences are very long and could be split to clarify the message.

The writing could be improved.

Some statements are not entirely supported by the results or not explained well enough.

L. 18-21: Both aqueous and gaseous C processes are key drivers of food webs and climate feedbacks in these ecosystems, regardless permafrost or glaciers are melting or not. This sentence is clumsy. The second part of the sentence (“as northern…”) is redundant with the first part. It is excessively long for the message it conveys.

L. 30. DOC “inputs” or “outputs”?

L. 31-35. This sentence is too long. Split before “and fluorescence…”

L. 31. “depended on seasonal subsurface flowpath structure”. Not very clear. What do you mean exactly?

L. 35-38. “DOC fluxes…”. This sentence is too long and not clear. If there is a positive relationship between NEE and DOC outputs state it clearly. “revealing” is not appropriate. The first part of the sentence does not reveal anything like that. Of course the C that is fixed by the vegetation is internally processed. How couldn’t it be?

L. 40-42. Not supported by the data.

Introduction

Overall, I don’t think the introduction is very well organized and written. Some sentences are too long (be careful to not overuse “as” - e.g., L.79-80) and ambiguous.

There should be a logical flow between the paragraphs, which is not always the case here.

The authors should insist on the importance of the study on a climate change context.

L. 61-62. No capital letters for boreal forest. I don’t really like the way you use “most important” to compare biomes. What about “the boreal forest plays a large role in global C cycle due to …”

L. 64. Planet’s

L. 66. All ecosystems have gaseous and aqueous fluxes. Boreal forests are characterized by large DOC fluxes. Rewrite the sentence.

L.71. “for” instead of “in”.

L. 80. “increasingly”

L.81. Why “decreased river flows”? Precipitation is not expected to decrease in the area and permafrost and glaciers thaw may increase river flow. I may have missed something. Clarify.

L. 91-93. This sentence is not clear. I think you should describe the consequences of the previous sentences.

L. 94-95. This sentence is ambiguous. Do you mean that terrestrial vegetation provides stream water with C directly (without passing through the soil)? What is the substrate you’re talking about? SOC or plant litter?

L. 96. Litter inputs to what? Rivers or SOC?

L. 98. “to” DOC

L. 102. Why are gaseous C fluxes particularly more complex than in other ecosystems? Because of CH4 emissions due to anoxic conditions in the soil? explain.

L. 117. It partitions…

L. 120. Briefly explain the nature and origin of fulvic acids; that FA differ depending on their origin and biogeochemical processes; that FA coming from plant litter and the soil have a higher degree of aromaticity as compared to microbially-derived FA; that we can estimate the relative contributions of FA sources by measuring FI. Some of this info is in your M&M section.

L. 127 and L. 132. Dissolved “organic” C

Materials and methods

Sampling was well performed. This part is clear and well described.

A short explanation of the meaning of fDOM values or why they performed this type of analysis could be added.

What kind of stats was performed on DOC seasonal trends? Mann-Kendall tests are commonly used.

Results

Fig 2 and 4. These figures are not very clear. I would rather use lines than dots. It would help visualizing seasonal trends.

Fig. 2B. Why isn’t delta 18O expressed in per mil?

Fig. 7. The x- and y-axes are not readable

Why is there no title for the first paragraph?

Results from MCWT are not presented in the text.

L. 277. The first two d18O values of May are high at YC. Mention it.

L. 283. Why cms and not m3 s-1?

L. 285. It would be insightful to display precipitation and temperature data to verify whether stream flow correlated with precipitation and/or snow melt resulting from high temperatures.

L. 298. Negatively correlated.

L. 299. The trends are indeed similar, i.e. decrease from May to June, but the correlation is not obvious.

L. 301. I agree that it correlates well in 2012 but not in 2013. It seems that there is a time lag.

L. 301-302. The diurnally variation in fDOM is not visible from this figure.

L. 307-308. When does it correlated positively and negatively?

L. 316-317. FI values are always >1.5. which -according to what you write in the M&M section- means that all DOC is from microbial origin.

L. 328. Does “this” refer to “the combination of low FI and HIX”?

L. 337. “a” strong “sink”

L. 338. “changed”

L. 341. There are additional spaces after “at” and “s-1”.

L. 341-342. Where are the monthly averages?

Discussion

The section “seasonality of DOC export” is poorly written and not clear. The interpretation of the results is weak and no references are cited. The links between the different analyses (e.g. DOC, d18O, FI, HIX and conductance) should be discussed more deeply. One way to do that could be a reduction of the number of sections. The questions that come to mind after reading the results section are: Why is there a DOC concentration peak in May? Why is it higher at MAYC than at YNGC? Where does this DOC come from? Why is fDOM positively correlated with DOC flux? Why is fDOM positively correlated with Q in 2012 and negatively in 2013? Etc. The discussion should clearly answer these questions.

L. 353. “North” America. The first sentence could be split after “North America”.

L. 357. What do you mean by “patterns”?

L. 364. Riverine C what? Stocks? Fluxes?

L. 367. Is the amount of nutrients stored in the snow cover significant? What kind of nutrients are you talking about? Please, clarify this point.

L. 369. What about removing “however” and replace it by “which constitutes an impermeable …”. Do you mean that most DOC leached in May (during snow melt) comes entirely from the snow pack? Where does this DOC come from? Develop.

L. 371-373. This sentence is too long and contains too many “as”.

L. 374. What are these nutrients with high specific conductance?

L. 375. The influence of precipitation events and subsurface mechanisms you’re talking about here is not clear.

L. 386-390. The cause of the absence of trend in d18O in 2013 is not clear.

L. 393-394. Explain why. On average, the DOC export seems however as high in 2013 as in 2012.

L. 398. The high Q in May 2013 is due to a rapid snow melt. It was associated with low conductivity, low d18O and fDOM values, suggesting that DOC comes almost entirely from the snow pack. Can you explain why the snow pack contains so much DOC and where it comes from?

L. 420. You could provide more information regarding biogenic and terrestrial DOC.

L. 424. Explain that the DOC output in May is more terrestrially-derived because microbial activity is low due to low temperatures. The DOC that’s released from the snow pack is not recycled by soil microorganisms.

L. 437. Why does the throughfall C result in more humified DOC?

L. 441-443. I agree with this statement but it is not clear how your data support it. Develop.

L. 444-455. This paragraph is very descriptive. It should be moved to the results section. As mentioned earlier, the resolution of the figure is too low for the x and y labels to be readable.

L. 469. Your values are therefore very low as compared to those reported in the literature. How do you explain that?

L. 481-484. C fixation would be indeed lower but as you mentioned before (L. 470), the DOC efflux would also be lower.

L. 485-498. The last paragraph is good. In my opinion, the conclusion mentioned L.488, i.e. the watershed is C conservative, it is the main take home message and should appear in the title. The high seasonality of nutrients fluxes in boreal catchment is not new.

Conclusion

The conclusion is good but these points were in my opinion not developed enough in the discussion.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: Yes: Clayton J Williams

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files to be viewed.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org. Please note that Supporting Information files do not need this step.

23 Oct 2019

Uploaded as separate document.

Submitted filename: Response to Reviewers.docx

Click here for additional data file.

1 Nov 2019

DOC export is exceeded by C fixation in May Creek: a late-successional watershed of the Copper River Basin, Alaska

PONE-D-19-14655R1

Dear Dr. Tomco,

We are pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it complies with all outstanding technical requirements.

Within one week, you will receive an e-mail containing information on the amendments required prior to publication. When all required modifications have been addressed, you will receive a formal acceptance letter and your manuscript will proceed to our production department and be scheduled for publication.

Shortly after the formal acceptance letter is sent, an invoice for payment will follow. To ensure an efficient production and billing process, please log into Editorial Manager at https://www.editorialmanager.com/pone/, click the "Update My Information" link at the top of the page, and update your user information. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to enable them to help maximize its impact. If they will be preparing press materials for this manuscript, you must inform our press team as soon as possible and no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

With kind regards,

Robert L. Bradley

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

12 Nov 2019

PONE-D-19-14655R1

DOC export is exceeded by C fixation in May Creek: a late-successional watershed of the Copper River Basin, Alaska

Dear Dr. Tomco:

I am pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please notify them about your upcoming paper at this point, to enable them to help maximize its impact. If they will be preparing press materials for this manuscript, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

For any other questions or concerns, please email plosone@plos.org.

Thank you for submitting your work to PLOS ONE.

With kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Robert L. Bradley

Academic Editor

PLOS ONE