Long‐term records of dissolved organic carbon flux from peat‐covered catchments: evidence for a drought effect?

This study considers three long records of dissolved organic carbon (DOC) flux from two catchments with peat‐covered headwaters. The catchments vary in size from 11 to 818 km² and the records are at least 12 years old, with one record going back to 1965. The study compares both annual and monthly DOC flux records with a range of hydroclimatic indicators in order to test which component of droughts may contribute to increasing DOC flux. The study found that: (1) there was no significant correlation between any of the proposed drought variables and DOC flux in any of the study catchments over periods of up to 34 years; (2) the most important variable for explaining the DOC flux was the runoff from the catchments overlying a seasonal cycle and an underlying upward trend was present in some records; (3) the residual time‐series, after removal of the best‐fit models, showed no evidence of increased production after times of severe drought. The lack of any evidence for any additional biogeochemical reactions associated with drought supports evidence that DOC loss from peat is limited by its solubility and that its production is fast on the time‐scale of runoff events. Copyright © 2008 John Wiley & Sons, Ltd.     

Adjacent catchments with similar patterns of land use and climate have markedly different dissolved organic carbon concentration and runoff dynamics

Temporal patterns in specific runoff, dissolved organic carbon concentrations [DOC] and fluxes were examined during two periods: 1994–1997 (period 1) and 2007–2009 (period 2) in five adjacent tributary catchments of Lake Simcoe, the largest lake in southern Ontario, Canada. The catchments displayed similar patterns of land use change with increases in urbanization (5–16%) and forest cover (0.2–4%) and declines in agriculture (4–8%) between 1994 and 2008. Climate in the catchments was similar; temperature increased slightly, but no significant change in precipitation was observed. Despite similar pattern of climate and land use, runoff responses and tributary [DOC] were different across the catchments. Following a very dry year (i.e. 1999), runoff increased steadily until the end of record. We observed increased variability in tributary [DOC] and higher DOC exports in period 2. This led to ~10% increase in [DOC] and a 13% increase in flux between the two study periods. Between the two periods, [DOC] increased by 15% in spring and 25% in summer, whereas flux increased by 17% in spring and 48% in summer. [DOC] was consistently higher in the growing (summer + autumn) than the dormant (winter + spring, minus spring melt months) seasons, but no unique pattern or simple linear flow/concentrations relationships existed. This suggests complex spatial and temporal pattern to runoff controls on DOC and flow dynamics in adjacent catchments. We therefore caution against extrapolating from monitored to unmonitored catchments. Copyright © 2012 John Wiley & Sons, Ltd.     

Spatial and temporal patterns of dissolved organic matter quantity and quality in the Mississippi River Basin, 1997–2013

Recent studies have found insignificant or decreasing trends in time‐series dissolved organic carbon (DOC) datasets, questioning the assumption that long‐term DOC concentrations in surface waters are increasing in response to anthropogenic forcing, including climate change, land use, and atmospheric acid deposition. We used the weighted regressions on time, discharge, and season (WRTDS) model to estimate annual flow‐normalized concentrations and fluxes to determine if changes in DOC quantity and quality signal anthropogenic forcing at 10 locations in the Mississippi River Basin. Despite increases in agriculture and urban development throughout the basin, net increases in DOC concentration and flux were significant at only 3 of 10 sites from 1997 to 2013 and ranged between ?3.5% to +18% and ?0.1 to 19%, respectively. Positive shifts in DOC quality, characterized by increasing specific ultraviolet absorbance at 254 nm, ranged between +8% and +45%, but only occurred at one of the sites with significant DOC quantity increases. Basinwide reductions in atmospheric sulfate deposition did not result in large increases in DOC either, likely because of the high buffering capacity of the soil. Hydroclimatic factors including annual discharge, precipitation, and temperature did not significantly change during the 17‐year timespan of this study, which contrasts with results from previous studies showing significant increases in precipitation and discharge over a century time scale. Our study also contrasts with those from smaller catchments, which have shown stronger DOC responses to climate, land use, and acidic deposition. This temporal and spatial analysis indicated that there was a potential change in DOC sources in the Mississippi River Basin between 1997 and 2013. However, the overall magnitude of DOC trends was not large, and the pattern in quantity and quality increases for the 10 study sites was not consistent throughout the basin.     

Seasonal controls on DOC dynamics in nested upland catchments in NE Scotland

Spatial and temporal variability of hydrological responses affecting surface water dissolved organic carbon (DOC) concentrations are important for determining upscaling patterns of DOC export within larger catchments. Annual and intra‐annual variations in DOC concentrations and fluxes were assessed over 2 years at 12 sites (3·40–1837 km²) within the River Dee basin in NE Scotland. Mean annual DOC fluxes, primarily correlated with catchment soil coverage, ranged from 3·41 to 9·48 g m^?2 yr^?1. Periods of seasonal (summer–autumn and winter–spring) DOC concentrations (production) were delineated and related to discharge. Although antecedent temperature mainly determined the timing of switchover between periods of high DOC in the summer‐autumn and low DOC in winter‐spring, inter‐annual variability of export within the same season was largely dependent on its associated water flux. DOC fluxes ranged from 1·39 to 4·80 g m^?2 season^?1 during summer–autumn and 1·43 to 4·15 g m^?2 season^?1 in winter–spring.Relationships between DOC areal fluxes and catchment scale indicated that mainstem fluxes reflect the averaging of highly heterogeneous inputs from contrasting headwater catchments, leading to convergent DOC fluxes at catchment sizes of ca 100 km². However, during summer–autumn periods, in contrast to winter–spring, longitudinal mainstem DOC fluxes continue to decrease, most likely because of increasing biological processes. This highlights the importance of considering seasonal as well as annual changes in DOC fluxes with catchment scale. This study increases our understanding of the temporal variability of DOC upscaling patterns reflecting cumulative changes across different catchment scales and aids modelling of carbon budgets at different stages of riverine systems. Copyright © 2010 John Wiley & Sons, Ltd.     

Acidification recovery in a changing climate: Observations from thirty-five years of stream chemistry monitoring in forested headwater catchments at the Turkey Lakes watershed,Ontario

Long-term ecosystem studies are valuable for understanding integrated ecosystem response to global changes in atmospheric deposition and climate. We examined trends for a 35-year period (1982/83–2017/18) in concentrations of a range of solutes in precipitation and stream water from nine headwater catchments spanning elevation and surficial geology gradients at the Turkey Lakes watershed (TLW) in northeastern Ontario, Canada. Average annual water year (WY, October to September) concentrations in precipitation significantly declined over the period for sulphate (SO₄²⁻), nitrate (NO₃⁻) and chloride (Cl⁻), while calcium (Ca²⁺) and potassium (K⁺) concentrations increased, resulting in a significant pH increase from 4.2 to 5.7. Trends in stream chemistry through time are generally consistent with expectations associated with acidification recovery. Concentration of many stream water solutes (SO₄²⁻, Cl⁻, calcium [Ca²⁺], magnesium [Mg²⁺] and NH₄⁺ generally decreased, while others (silica [SiO₂] and dissolved organic carbon [DOC]) generally increased. Increases were also observed for alkalinity (six of nine catchments), acid neutralizing capacity ([ANC]; six of nine catchments) and pH (eight of nine catchments), while conductivity declined (six of nine catchments). Variability in trends among catchments are associated with differences in surficial geology and wetland cover. While absolute solute concentrations were generally lower at bedrock dominated high-elevation catchments compared to till dominated lower elevation catchments, the rate of change of concentration was often greater for high elevation catchments. This study confirms continued, but non-linear stream chemistry recovery from acidification, particularly at the less buffered high and moderate elevation sites. The heterogeneity of responses among catchments highlights our incomplete understanding of the relative importance of different mechanisms influencing stream chemistry and the consequences for downstream ecosystems.     

DOC budgets of drained peat catchments: implications for DOC production in peat soils

The blocking of drainage ditches in peat has been proposed as a possible mitigation strategy for the widely observed increases in dissolved organic carbon (DOC) concentrations from northern peatlands. This study tested the hypothesis that drain‐blocking could lead to lower DOC concentrations by measuring the DOC export from a series of small peat‐covered catchments over a period of 2 years. Six catchments were chosen: two were pristine that had never been drained; three where drains had been blocked (one in 1995, and two in 2003); and a control peat drain catchment where the drain was left unblocked throughout the study. In the case where drains were blocked as part of thus study, the drains were observed for 2 months before blocking and 2 years after blocking. The results show that: (i) high concentrations of DOC can come from water ponded in the drain; (ii) the DOC export (flux of DOC per area of catchment) from the six study catchments shows a high degree of positive correlation with both catchment size and water yield; (iii) distinctly lower DOC export with water yield was observed for the catchments containing higher‐order channels (>27 500 m²) as opposed to single drain catchments (>7500 m²); (iv) drain‐blocking resulted in a statistically significant decrease in DOC export (average was 39%) but the effect upon DOC concentration explained only 1% of the variance in the data. The results suggest that drain blocking works by decreasing the flow from the drain, not by changing the production of DOC in the peat. The change in export with catchment size implies a considerable removal of DOC from large catchments. Copyright © 2009 John Wiley & Sons, Ltd.     

Dissolved organic carbon dynamics in two streams draining forested catchments at loch ard,Scotland

Dissolved organic carbon (DOC) was measured at four or eight hour intervals between mid-1989 and mid-1991 in two catchments in west central Scotland. The experimental catchment had been recently clear-felled and the control remained under forest. The amount of DOC varied during individual storm events following the stream hydro-graph. Maximum variations were found in the summer half-year and in the clear-felled catchment. There was also evidence of the exhaustion of DOC in the later events of a sequence. Differences between the catchments were related to catchment characteristics and to land-use change. The reduced magnitude of variation in DOC with discharge in the control stream was due to the influence of a wetland area through which the stream flowed. The mean DOC concentrations were similar in the two streams and annual exports were 15 g m^?2 from the control and 16g m^?2 from the felled catchment. The stream draining the clear-felled catchment had greater high flow DOC concentrations in the summer half-year, probably due to the effect of greater mean summer temperatures on DOC release and of the greater supply of organic debris in the stream channel.     

Variation in riverine phosphorus between 1994 and 2003 as affected by land-use and loading reductions in six medium-sized to large German rivers

Land use and in-stream transformation exert great influence on concentrations and loads of phosphorus (P) in rivers. We aimed to display differences in the courses of total P (TP) and soluble reactive P (SRP) concentrations and loads in six medium-sized to large rivers in the central region of Germany, and to identify the reasons for different long-term trends. Therefore, we applied multivariate statistics to 10-year-time series (1994–2003) of TP, SRP, discharge (Q), water temperature (T), pH, dissolved organic carbon (DOC), total organic carbon (TOC), dissolved oxygen (DO), total iron (Fe), and total manganese (Mn). Statistical results were related to land use in the catchments of the rivers. TP concentrations ranged between 0.02 and 0.78 mg l⁻¹, and SRP concentrations ranged between 0.01 and 0.44 mg l⁻¹. Q correlated negatively with TP and SRP concentrations over the entire year. Furthermore, Fe correlated significantly and positively to TP and SRP and therefore, ferric hydroxides likely were the major P sorption sites. DOC showed significant positive correlation to SRP particularly in spring, indicating manure exposure in early spring as a major source of both, DOC and SRP. Significant negative correlations between DO and SRP in summer hint at internal P loading in rivers or in flushed lakes. Different forms of land use were the reasons of enhanced or retarded recovering from previous increases in P concentrations. High portions of arable land within some of the catchments impeded the process of decreases since 1996 because of remaining high-diffuse emissions from fertilized soils. Agricultural practices, exposing fertilizer to soils within the river catchments and high Q in early spring caused high TP and SRP loads to downstream systems, and evoked risks for downstream river reaches.     

Seasonal,hydrological, and land management factors controlling dissolved organic carbon concentrations in the loch fleet catchments,Southwest Scotland

Dissolved organic carbon (DOC) was measured at hourly or two-hourly intervals during more than 30 events in one forested and two moorland subcatchments of the Loch Fleet catchment in southwest Scotland. The dominantly peaty soils in the catchments resulted in small discharge-related DOC variations within individual events, with a maximum range of about 2 mg 1^?1. Seasonal variations were larger with an amplitude of 8-9 mg 1^?1 and maximum concentrations in the summer months. The forested stream had the highest mean DOC, twice as large as the comparable moorland stream in the preliming phase. Applications of lime to the catchments increased stream DOC concentrations, with the largest increases in the moorland catchments.     

Hydrological and catchment controls on event-scale dissolved organic carbon dynamics in boreal headwater streams

Hydrological events transport large proportions of annual or seasonal dissolved organic carbon (DOC) loads from catchments to streams. The timing, magnitude and intensity of these events are very sensitive to changes in temperature and precipitation patterns, particularly across the boreal region where snowpacks are declining and summer droughts are increasing. It is important to understand how landscape characteristics modulate event-scale DOC dynamics in order to scale up predictions from sites across regions, and to understand how climatic changes will influence DOC dynamics across the boreal forest. The goal of this study was to assess variability in DOC concentrations in boreal headwater streams across catchments with varying physiographic characteristics (e.g., size, proportion of wetland) during a range of hydrological events (e.g., spring snowmelt, summer/fall storm events). From 2016 to 2017, continuous discharge and sub-daily chemistry grab samples were collected from three adjacent study catchments located at the International Institute for Sustainable Development-Experimental Lakes Area in northwestern Ontario, Canada. Catchment differences were more apparent in summer and fall events and less apparent during early spring melt events. Hysteresis analysis suggested that DOC sources were proximal to the stream for all events at a catchment dominated by a large wetland near the outlet, but distal from the stream at the catchments that lacked significant wetland coverage during the summer and fall. Wetland coverage also influenced responses of DOC export to antecedent moisture; at the wetland-dominated catchment, there were consistent negative relationships between DOC concentrations and antecedent moisture, while at the catchments without large wetlands, the relationships were positive or not significant. These results emphasize the utility of sub-daily sampling for inferring catchment DOC transport processes, and the importance of considering catchment-specific factors when predicting event-scale DOC behaviour.     

Spatial variation of wetlands and flux of dissolved organic carbon in boreal headwater streams

In order to investigate the relation between water chemistry and functional landscape elements, spatial data sets of characteristics for 68 small (0·2–1·5 km²) boreal forest catchments in western central Sweden were analysed in a geographical information system (GIS). The geographic data used were extracted from official topographic maps. Water sampled four times at different flow situations was analysed chemically. This paper focuses on one phenomenon that has an important influence on headwater quality in boreal, coniferous forest streams: generation and export of dissolved organic carbon (DOC). It is known that wetland cover (bogs and fens) in the catchment is a major source of DOC. In this study, a comparison was made between a large number of headwater catchments with varying spatial locations and areas of wetlands. How this variation, together with a number of other spatial variables, influences the DOC flux in the streamwater was analysed by statistical methods. There were significant, but not strong, correlations between the total percentages of wetland area and DOC flux measured at a medium flow situation, but not at high flow. Neither were there any significant correlations between the percentage of wetland area connected to streams, nor the percentage of wetland area within a zone 50 m from the stream and the DOC flux. There were, however, correlations between catchment mean slope and the DOC flux in all but one flow situations. This study showed that, considering geographical data retrieved from official sources, the topography of a catchment better explains the variation in DOC flux than the percentage and locations of distinct wetland areas. This emphasizes the need for high‐resolution elevation models accurate enough to reveal the sources of DOC found in headwater streams. Copyright © 2007 John Wiley & Sons, Ltd.     

EXPORTS OF ORGANIC CARBON IN BRITISH RIVERS

This study provides the first detailed estimate of riverine organic carbon fluxes in British rivers, as well as highlighting major gaps in organic carbon data in national archives. Existing data on organic carbon and suspended solids concentrations collected between 1989 and 1993, during routine monitoring by the River Purification Boards (RPBs) in Scotland and the National River Authorities (NRAs) in England and Wales, were used with annual mean flows to estimate fluxes of dissolved and particulate organic carbon (DOC and POC) in British rivers. Riverine DOC exports during 1993 varied from 7·7–103·5 kg ha⁻¹ year⁻¹, with a median flux of 31·9 kg ha⁻¹ year⁻¹ in the 85 rivers for which data were available. There was a trend for DOC fluxes to increase from the south and east to the north and west. A predictive model based on mean soil carbon storage in 17 catchments, together with regional precipitation totals, explained 94% of the variation in the riverine DOC exports in 1993. This model was used to predict riverine DOC fluxes in regions where no organic carbon data were available. Calculated and predicted fluxes were combined to produce an estimate for exports of DOC to tidal waters in British rivers during 1993 of 0·68±0·07 Mt. Of this total, rivers in Scotland accounted for 53%, England 38% and Wales 9%. Scottish blanket peats would appear to be the largest single source of DOC exports in British rivers. An additional 0·20 Mt of organic carbon were estimated to have been exported in particulate form in 1993, approximately two–thirds of which was contributed by English rivers. It is suggested that riverine losses of organic carbon have the potential to affect the long-term dynamics of terrestrial organic carbon pools in Britain and that rivers may regulate increases in soil carbon pools brought about by climate change. © 1997 by John Wiley & Sons, Ltd.     

Flowpath controls on high-spatial-resolution water-chemistry profiles in headwater streams

Stream water chemistry is routinely measured over time at fixed and sparse sites, which provides a coarse image of spatial variability. Here, we measured nitrate, dissolved organic carbon (DOC) and several chemical proxies for water flowpaths, catchment residence time and biogeochemical transformations, every 50–100 m along 13 km of streams in six agricultural headwater catchments (1.1–3.5km²). The objective was to examine controls on longitudinal nitrate profiles at a high spatial resolution during four seasons: rewetting of the catchments in autumn, winter high-flow, spring recession and summer low-flow. Our results showed monotonic trends in longitudinal profiles for nitrate and DOC, which were opposite for the two solutes. Spatial trends in water-chemistry profiles persisted across seasons, which suggests time-invariant controls on the spatial variations in concentrations. Four catchments exhibited decreasing nitrate and increasing DOC from upstream to downstream, while two catchments exhibited increasing nitrate and decreasing DOC. These smooth gradients did not reflect a longitudinal land-use gradient, but rather an increase in the proportion of groundwater inflows when moving downstream, as suggested by the chemical proxies and punctual discharge measurements. Water chemistry also changed abruptly at confluences, at a farm point source and at a localized groundwater inflow zone.     

Dissolved organic matter dynamics during the spring snowmelt at a boreal river valley mire complex in Northwest Russia

Boreal mire landscapes are rich in soil carbon and significantly contribute to the carbon input of aquatic ecosystems. They are composed of different mesoscale ecohydrological subunits, whose individual contributions to the water and carbon export of mire catchments are not well understood. The spring snowmelt period is the major hydrological event in the annual water cycle of the boreal regions and strongly influences the carbon flux between the terrestrial and aquatic systems. The aim of this study was (1) to provide a conceptual understanding of the spatial and temporal dynamics of the surface water chemistry along a swamp forest‐fen‐bog gradient during the snowmelt period, (2) to quantify the exported dissolved organic carbon (DOC) content in the runoff and (3) to identify the ecohydrological landscape unit that contributes most to DOC export during the snowmelt period in a heterogeneous mire complex in Northwest Russia. The highest DOC concentrations were detected in the swamp forest, and the lowest concentrations were observed at the treeless bog by the end of the snowmelt period (swamp forest: 37–43 mg l^?1, bog: 13–17 mg l^?1). During the spring snowmelt period, a significant amount (~1.7 g C m^?2) of DOC was transferred by the ~74 mm of runoff from the catchment into the river. Variability in the thawing periods led to differences in the relative contributions of each ecohydrological zone to the carbon export measured at a stream channel draining the studied part of the mire complex. An increased understanding of the variation in DOC concentrations and contributions from the mesoscale ecohydrological subunits to carbon export can help to predict the potential regional loss of DOC based on land cover type under climate change. Copyright © 2015 John Wiley & Sons, Ltd.     

Fluxes of inorganic carbon from two forested catchments in the Appalachian mountains

This study uses long‐term records of stream chemistry, discharge and air temperature from two neighbouring forested catchments in the southern Appalachians in order to calculate production of dissolved CO₂ and dissolved inorganic carbon (DIC). One of the pair of catchments was clear‐felled during the period of the study. The study shows that: (1) areal production rates of both dissolved CO₂ and DIC are similar between the two catchments even during and immediately after the period of clear‐felling; (2) flux of total inorganic carbon (dissolved CO₂+ DIC) rises dramatically in response to a catchment‐wide acidification event; (3) DIC and dissolved CO₂ are dominantly released on the old water portion of the discharge and concentrations peak in the early autumn when flows in the study catchments are at their lowest; (4) total fluvial carbon flux from the clear‐felled catchment is 11·6 t km⁻² year⁻¹ and for the control catchment is 11·4 t km⁻² year⁻¹. The total inorganic carbon flux represents 69% of the total fluvial carbon flux. The method presented in the study provides a useful way of estimating inorganic carbon flux from a catchment without detailed gas monitoring. The time series of dissolved CO₂ at emergence to the stream can also be a proxy for the soil flux of CO₂. Copyright © 2005 John Wiley & Sons, Ltd.     

Reconstructing long‐term records of dissolved CO2

The dissolved CO₂ concentration of stream waters is an important component of the terrestrial carbon cycle. This study reconstructs long‐term records of dissolved CO₂ concentration for the outlets of two large catchments (818 and 586 km²) in northern England. The study shows that:

• 1. The flux of dissolved CO₂ from the catchments (as carbon per catchment area), when adjusted for that which would be carried by the river water at equilibrium with the atmosphere, is between 0 and 0·39 t km⁻² year⁻¹ for the River Tees and between 0 and 0·65 t km⁻² year⁻¹ for the River Coquet.
• 2. The flux of dissolved CO₂ is closely correlated with dissolved organic carbon (DOC) export and is unrelated to dissolved CO₂ export from the headwaters of the study catchments.
• 3. The evasion rate of CO₂ from the rivers (as carbon per stream area) is between 0·0 and 1·49 kg m⁻² year⁻¹, and calculated in‐stream productions of CO₂ are estimated as between 0·5 and 2·5% of the stream evasion rate.
• 4. By mass balance, it is estimated that 8% of the annual flux of DOC is lost within the streams of the catchment.

The study shows that the loss of CO₂ from the streams of the Tees catchment is between 3·1 and 7·5 kt year⁻¹ (as carbon) for the River Tees, which is the same order as annual CH₄ flux from peats within the catchment and approximately 50% of the net CO₂ exchange to the peats of the catchment. Copyright © 2005 John Wiley & Sons, Ltd.     

Hydrological control of dissolved organic carbon dynamics in a forested headwater catchment,Kiryu Experimental Watershed,Japan

To evaluate the influence of hydrological processes on dissolved organic carbon (DOC) dynamics in a forested headwater catchment, DOC concentration was observed along the flow path from rainfall to stream water via throughfall, soil water, groundwater, and spring water for 4 years, and DOC flux through the catchment was calculated. The spatial and temporal variations in DOC concentration and flux were compared with physical hydrological observations and the mean residence time of water. In the upslope soil layer, DOC concentrations were not significantly correlated with water fluxes, suggesting that DOC concentrations were not strictly controlled by water fluxes. In the upslope perennial groundwater, DOC concentration was affected by the change in the amount of microbial degradation of DOC produced by changes in the mean residence time of water. In stream water, the temporal variation in DOC concentration was usually affected by changes in DOC concentration of the inflow component via vertical infiltration from above the perennial groundwater. During dry periods, however, the component from inflow via vertical infiltration was negligible and DOC in the upslope perennial groundwater became the major component of stream water DOC. The temporal variation in stream water DOC concentration during baseflow was affected by rainfall patterns over several preceding months. Therefore, records of rainfall over several preceding months are one of the most important factors for predicting changes in DOC concentration on a catchment scale. Copyright © 2007 John Wiley & Sons, Ltd.     

An evaluation of the hysteresis in chemical concentration–discharge (C–Q) relationships from drained,intensively managed grasslands in southwest England

This study examined the hysteresis exhibited in concentration–discharge (C–Q) relationships in the runoff from four hydrologically separated fields (catchments) at an intensively managed grassland. The objectives were to examine C–Q relationships constructed from high-resolution time series of flow, temperature, pH, conductivity, nitrate and turbidity, and their implications for hydrological processes. High-resolution datasets from the quality assured records of the Rothamsted Research North Wyke Farm Platform in the UK were examined using a graphical method and cross-correlation statistics. The study found that storm events based C–Q hysteresis reflects the cross-correlation that is generally hidden in time series analysis of large datasets, and that although Q and water quality variables can be effectively influenced by catchment size, the C–Q relationship is less significantly influenced. The dominant C–Q relationships of the water variables in the study area reflect that saturated overland flow was prevalent during the study period in the catchments, while the CCF results indicate coupled transfer of sediments and solute in the area at lag ≥ 0.

EDITOR D. Koutsoyiannis

ASSOCIATE EDITOR M. D. Fidelibus     

Effects of bedrock groundwater discharge on spatial variability of dissolved carbon,nitrogen, and phosphorous concentrations in stream water within a forest headwater catchment

The quantitative evaluation of the effects of bedrock groundwater discharge on spatial variability of stream dissolved organic carbon (DOC), dissolved inorganic nitrogen (DIN) and dissolved inorganic phosphorous (DIP) concentrations has still been insufficient. We examined the relationships between stream DOC, DIN and DIP concentrations and bedrock groundwater contribution to stream water in forest headwater catchments in warm-humid climate zones. We sampled stream water and bedrock springs at multiple points in September and December 2013 in a 5 km² forest headwater catchment in Japan and sampled groundwater in soil layer in small hillslopes. We assumed that stream water consisted of four end members, groundwater in soil layer and three types of bedrock groundwater, and calculated the contributions of each end member to stream water from mineral-derived solute concentrations. DOC, DIN and DIP concentrations in stream water were compared with the calculated bedrock groundwater contribution. The bedrock groundwater contribution had significant negative linear correlation with stream DOC concentration, no significant correlation with stream DIN concentration, and significant positive linear correlation with stream DIP concentration. These results highlighted the importance of bedrock groundwater discharge in establishing stream DOC and DIP concentrations. In addition, stream DOC and DIP concentrations were higher and lower, respectively, than those expected from end member mixing of groundwater in soil layer and bedrock springs. Spatial heterogeneity of DOC and DIP concentrations in groundwater and/or in-stream DOC production and DIP uptake were the probable reasons for these discrepancies. Our results indicate that the relationships between spatial variability of stream DOC, DIN and DIP concentrations and bedrock groundwater contribution are useful for comparing the processes that affect stream DOC, DIN and DIP concentrations among catchments beyond the spatial heterogeneity of hydrological and biogeochemical processes within a catchment.