Background & Rationale
The western Canadian subarctic is a landscape in transition. Annual mean air temperatures in Yellowknife have increased more than 2° C from the 1950-1970 mean of -5.7° C, with warming mainly occurring in winter [1]. Permafrost thaw is a consequence of warming [2]. Permafrost is soil that has been frozen for thousands of years; it holds stores of previously inaccessible carbon, nutrients, and ions that become available for biological and chemical processing upon thaw [2]. Biodegradable carbon tends to be lost from carbon pools first, as it is easier to mineralize by microbes. Permafrost carbon, which has been “locked up” and unavailable for mineralization tends to have a larger pool of biodegradable dissolved organic carbon, or BDOC [3], [4].
Quantifying the release of BDOC to streams is necessary to understand how warming is changing carbon cycling; quantifying this release in winter is especially important as this season has undergone the most change in the past several decades. Why is understanding carbon cycling important? In Arctic and subarctic riverine systems, dissolved organic carbon (DOC) is the dominant form of organic carbon, and rivers play a significant role in the sequestration, transport, and mineralization of DOC [5], [6]. Globally, Arctic watersheds transport an average of 34 Tg C/yr to the Arctic Ocean [7]. However, another 37-84 Tg/yr of terrestrially sourced DOC is lost via mineralization or is buried in inland waters before it reaches the Arctic Ocean [8], indicating relatively rapid processing of this material within streams and rivers. Once terrestrially sourced carbon reaches riverine systems, it can be rapidly mineralized by heterotrophic organisms [5]. Thus, if more BDOC is available in riverine systems, a large proportion of it will be mineralized into CO2 and contribute to climate feedbacks. Streams and rivers in the study area lie within the Mackenzie River basin and connect this landscape to the Beaufort Sea [9]. Thus, local impacts of thaw may extend to the watershed level and across larger geographic scales.
Gradual permafrost thaw can create supra-permafrost taliks, or bodies of unfrozen ground above the permafrost table (See Figure 1) [10]. The active layer also lies above the permafrost table and is subject to annual freeze thaw cycles. Taliks may form due to active layer deepening or a lack of complete wintertime freezeback of the active layer [10]. Taliks provide connections between unfrozen areas of the soil during winter, which alters the source and chemistry of flow to streams[10]. The specific influence of taliks on subarctic rivers is overall poorly understood, with research almost entirely focused on hydrologic impacts [4], [11]–[14]. In the Great Slave region, continued flow through taliks during winter produces icings, or sheet-like masses of layered ice above the ground surface (See Figure 1 and 2). At the end of the winter, these icings represent an archive of subsurface flow throughout the season and can be analyzed to study the nature of winter water chemistry.
Quantifying the release of BDOC to streams is necessary to understand how warming is changing carbon cycling; quantifying this release in winter is especially important as this season has undergone the most change in the past several decades. Why is understanding carbon cycling important? In Arctic and subarctic riverine systems, dissolved organic carbon (DOC) is the dominant form of organic carbon, and rivers play a significant role in the sequestration, transport, and mineralization of DOC [5], [6]. Globally, Arctic watersheds transport an average of 34 Tg C/yr to the Arctic Ocean [7]. However, another 37-84 Tg/yr of terrestrially sourced DOC is lost via mineralization or is buried in inland waters before it reaches the Arctic Ocean [8], indicating relatively rapid processing of this material within streams and rivers. Once terrestrially sourced carbon reaches riverine systems, it can be rapidly mineralized by heterotrophic organisms [5]. Thus, if more BDOC is available in riverine systems, a large proportion of it will be mineralized into CO2 and contribute to climate feedbacks. Streams and rivers in the study area lie within the Mackenzie River basin and connect this landscape to the Beaufort Sea [9]. Thus, local impacts of thaw may extend to the watershed level and across larger geographic scales.
Gradual permafrost thaw can create supra-permafrost taliks, or bodies of unfrozen ground above the permafrost table (See Figure 1) [10]. The active layer also lies above the permafrost table and is subject to annual freeze thaw cycles. Taliks may form due to active layer deepening or a lack of complete wintertime freezeback of the active layer [10]. Taliks provide connections between unfrozen areas of the soil during winter, which alters the source and chemistry of flow to streams[10]. The specific influence of taliks on subarctic rivers is overall poorly understood, with research almost entirely focused on hydrologic impacts [4], [11]–[14]. In the Great Slave region, continued flow through taliks during winter produces icings, or sheet-like masses of layered ice above the ground surface (See Figure 1 and 2). At the end of the winter, these icings represent an archive of subsurface flow throughout the season and can be analyzed to study the nature of winter water chemistry.
Figure 1. Simplified conceptual model of talik flow and icing development, showing sub, intra and supra-permafrost taliks. Taliks develop due to increased thickness or decreased freezeback of the active layer. Water flow through supra-permafrost taliks may pick up materials previously frozen in permafrost and manifest as icings when flow pushes up to the ground surface. Figure adapted from Sandells and Flocco (2014) [15].
Figure 2. Slideshow detailing the process of icing development. Figures adapted from Hu and Pollard (1997). [16].
Talik presence changes flow paths of water, which can fundamentally change the lateral transport of chemical constituents [14]. Deeper flow pathways can increase exports of major ions and trace metals delivered to streams. If taliks enable flow through freshly thawed permafrost soil, this may also release previously frozen constituents, including older carbon previously held in permafrost, and alter chemical fluxes to streams [17], [18].
My MSc research aims to illustrate how thaw and talik flow during winter are influencing the export of chemical constituents to streams in the Yellowknife area, NT. My first data chapter will focus on characterizing differences in the organic carbon pool (via absorbance and fluorescence metrics, Box 1 and 2, respectively) and nutrients (Box 3) between winter and spring/summer flow and their ecological implications (biological processing). The purpose of this chapter is to 1) identify if talik flow is releasing more BDOC to streams, and 2) assess whether this carbon will be preferentially processed into CO2.
My MSc research aims to illustrate how thaw and talik flow during winter are influencing the export of chemical constituents to streams in the Yellowknife area, NT. My first data chapter will focus on characterizing differences in the organic carbon pool (via absorbance and fluorescence metrics, Box 1 and 2, respectively) and nutrients (Box 3) between winter and spring/summer flow and their ecological implications (biological processing). The purpose of this chapter is to 1) identify if talik flow is releasing more BDOC to streams, and 2) assess whether this carbon will be preferentially processed into CO2.
Box 1: Absorbance Characteristics of Carbon Absorbance spectroscopy exposes samples to the UV spectrum and measures their absorbance at different wavelengths. This data can be used to assess the aromaticity, molecular weight and humic contents of DOC, by calculating parameters such as SUVA, A254, and slope ratios [19]. |
Box 2: Fluorescence Characteristics of Carbon Fluorescence spectroscopy relies on the property of dissolved organic matter fluorescing when exciting with ultraviolet light. This property can be used to characterize humic content (recalcitrant) vs. fresh-like (biodegradable) content in DOC. It can also illustrate the degree of humification within a sample (the process of formation of humic substances) [20]. |
Box 3: Nutrients Microbes rely primarily on nitrogen and phosphorus to complete respiration. Nitrogen and phosphorus can be separated into the following fractions: DIN - Dissolved Inorganic Nitrogen - a simple molecule, easily used by microbes DON - Dissolved Organic Nitrogen - a complex molecule, not readily available for use SRP - Soluble Reactive Phosphorus reactive form of phosphorus used by microbes P - Elemental Phosphorus - unavailable for biological use |
My second data chapter will investigate the physical processes of talik flow and icing formation. This will be explored using water isotopes, and major ions and trace metals (See Box 3 and 4, respectively) as well as level logger data and trail camera footage. The purpose of this chapter is to identify the occurrence of deeper flow paths (suggesting permafrost thaw and talik presence).
Box 3: Water Isotopes Water isotopes include isotopes associated with the water molecule, namely oxygen-18 and deuterium. Water has different isotopic composition due to fractionation, which occurs during phase changes (e.g. evaporation). Water isotopes are useful to trace water source (e.g. precipitation vs groundwater, i.e. "new" vs. "old" water). Isotopic composition are reported as δ values, which represent deviation in per mil (%) from the isotopic composition of standard. |
Box 4: Major Ions and Trace Metals Major ions and trace metals are indicators of changing flow pathways of water. Major ions are typically present in high concentrations (mg/L) and trace metals in low concentrations (µg/L) in natural waters. |
Objectives
Objective 1: Is there a difference in carbon concentration and composition, and nutrient concentrations between winter flow (i.e. icing samples) and open water flow (i.e. lake and stream samples) at different icings? If so, do differences have implications for downstream processing of carbon (higher biodegradability of carbon, higher concentrations of inorganic nutrients)?
Objective 2: Investigate spatial (upstream vs. downstream) and temporal (winter, spring, summer) relationships between carbon concentration/composition and biological processing using an incubation experiment.
Objective 3: Determine if there is evidence of flow through deeper flow paths, suggesting permafrost thaw and flow through taliks. Are results from objectives 1) and 2) primarily caused by talik flow or another unknown mechanism?
Expected Results
I expect:
Alternatively, if the winter carbon pool displays a more recalcitrant, terrestrially sourced signature, then I can conclude that winter stream chemistry is not being significantly affected by talik flow in this area. If the carbon pool is distinct, and biological processing is impacted, but there is little evidence for flow through deeper flow paths, then this could indicate that some other mechanism is influencing carbon exports to streams. Another possible mechanism may be the magnitude of flow during winter, which has generally been increasing in this region since the 1970s [21].
Another uncertainty is the relationship between significantly higher loads of DOC observed during spring and biological processing; I anticipate that my highly DOC-concentrated spring samples may show higher rates of biological processing despite having a more recalcitrant DOC pool. While spring DOC is typically terrestrially sourced and largely recalcitrant, DOC itself is highly complex. DOC contains a mixture of low and high molecular weight compounds, with thousands of different molecular formulae which cannot be resolved through absorbance and fluorescence characteristics alone [18]. This is a problem I will address next semester using FT-ICR-MS (Fourier-transform ion cyclotron resonance mass spectrometry) analysis, which is a more detailed characterization method of organic matter. FT-ICR-MS will identify unique molecular compounds in my DOC pool, allowing me to better understand DOC structure and composition, and connect this to observed differences in biological processing.
- The winter flow carbon pool to be characteristically more biodegradable, having low molecular weight, low aromaticity, low humic-like components, and high protein-like components.
- Carbon concentrations to peak during spring, when large volumes of meltwater running over the landscape typically transport large loads of DOC to streams; I expect this carbon to show a terrestrial signature (high molecular weight, high aromaticity, and high humic-like components).
- Winter samples to contain the highest concentrations of inorganic nutrients, because microbial activity is partly limited by temperature, and thus these valuable nutrients will be retained until temperatures rise in late spring/summer.
- Differences in carbon concentration and composition to drive biological processing in the incubation experiment. I predict that higher concentrations of DOC and a characteristically biodegradable carbon pool will result in higher rates O2 loss, and higher percentages of CO2 gain and DOC loss. I predict that carbon biodegradability will be a larger control on biological processing, as mineralization of recalcitrant material typically takes longer even if DOC concentrations are high.
- Winter samples to have higher concentrations of major cations and trace metals indicative of deeper flow paths.
Alternatively, if the winter carbon pool displays a more recalcitrant, terrestrially sourced signature, then I can conclude that winter stream chemistry is not being significantly affected by talik flow in this area. If the carbon pool is distinct, and biological processing is impacted, but there is little evidence for flow through deeper flow paths, then this could indicate that some other mechanism is influencing carbon exports to streams. Another possible mechanism may be the magnitude of flow during winter, which has generally been increasing in this region since the 1970s [21].
Another uncertainty is the relationship between significantly higher loads of DOC observed during spring and biological processing; I anticipate that my highly DOC-concentrated spring samples may show higher rates of biological processing despite having a more recalcitrant DOC pool. While spring DOC is typically terrestrially sourced and largely recalcitrant, DOC itself is highly complex. DOC contains a mixture of low and high molecular weight compounds, with thousands of different molecular formulae which cannot be resolved through absorbance and fluorescence characteristics alone [18]. This is a problem I will address next semester using FT-ICR-MS (Fourier-transform ion cyclotron resonance mass spectrometry) analysis, which is a more detailed characterization method of organic matter. FT-ICR-MS will identify unique molecular compounds in my DOC pool, allowing me to better understand DOC structure and composition, and connect this to observed differences in biological processing.
References
[1] J. R. Paul, S. v. Kokelj, and J. L. Baltzer, “Spatial and stratigraphic variation of near‐surface ground ice in discontinuous permafrost of the taiga shield,” Permafrost and Periglacial Processes, vol. 32, no. 1, pp. 3–18, Jan. 2021, doi: 10.1002/ppp.2085.
[2] S. E. Tank, R. G. Striegl, J. W. McClelland, and S. v Kokelj, “Multi-decadal increases in dissolved organic carbon and alkalinity flux from the Mackenzie drainage basin to the Arctic Ocean,” Environmental Research Letters, vol. 11, no. 5, p. 054015, May 2016, doi: 10.1088/1748-9326/11/5/054015.
[3] K. Kaiser, M. Canedo-Oropeza, R. McMahon, and R. M. W. Amon, “Origins and transformations of dissolved organic matter in large Arctic rivers,” Scientific Reports, vol. 7, no. 1, p. 13064, Dec. 2017, doi: 10.1038/s41598-017-12729-1.
[4] M. S. Schwab et al., “An Abrupt Aging of Dissolved Organic Carbon in Large Arctic Rivers,” Geophysical Research Letters, vol. 47, no. 23, Dec. 2020, doi: 10.1029/2020GL088823.
[5] T. J. Battin, S. Luyssaert, L. A. Kaplan, A. K. Aufdenkampe, A. Richter, and L. J. Tranvik, “The boundless carbon cycle,” Nature Geoscience, vol. 2, no. 9, pp. 598–600, Sep. 2009, doi: 10.1038/ngeo618.
[6] C. A. Littlefair and S. E. Tank, “Biodegradability of Thermokarst Carbon in a Till-Associated, Glacial Margin Landscape: The Case of the Peel Plateau, NWT, Canada,” Journal of Geophysical Research: Biogeosciences, vol. 123, no. 10, pp. 3293–3307, Oct. 2018, doi: 10.1029/2018JG004461.
[7] J. E. Vonk et al., “Biodegradability of dissolved organic carbon in permafrost soils and aquatic systems: a meta-analysis,” Biogeosciences, vol. 12, no. 23, pp. 6915–6930, Dec. 2015, doi: 10.5194/bg-12-6915-2015.
[8] B. W. Abbott, J. R. Larouche, J. B. Jones, W. B. Bowden, and A. W. Balser, “Elevated dissolved organic carbon biodegradability from thawing and collapsing permafrost,” Journal of Geophysical Research: Biogeosciences, vol. 119, no. 10, pp. 2049–2063, Oct. 2014, doi: 10.1002/2014JG002678.
[9] S. E. Tank et al., “A land‐to‐ocean perspective on the magnitude, source and implication of DIC flux from major Arctic rivers to the Arctic Ocean,” Global Biogeochemical Cycles, vol. 26, no. 4, p. 2011GB004192, Dec. 2012, doi: 10.1029/2011GB004192.
[10] M. A. Walvoord and B. L. Kurylyk, “Hydrologic Impacts of Thawing Permafrost-A Review,” Vadose Zone Journal, vol. 15, no. 6, p. vzj2016.01.0010, Jun. 2016, doi: 10.2136/vzj2016.01.0010.
[11] C. Spence et al., “Hydrological resilience to forest fire in the subarctic Canadian shield,” Hydrological Processes, vol. 34, no. 25, pp. 4940–4958, Dec. 2020, doi: 10.1002/hyp.13915.
[12] P. Lamontagne-Hallé, J. M. McKenzie, B. L. Kurylyk, and S. C. Zipper, “Changing groundwater discharge dynamics in permafrost regions,” Environmental Research Letters, vol. 13, no. 8, p. 084017, Aug. 2018, doi: 10.1088/1748-9326/aad404.
[13] V. F. Bense, H. Kooi, G. Ferguson, and T. Read, “Permafrost degradation as a control on hydrogeological regime shifts in a warming climate,” Journal of Geophysical Research: Earth Surface, vol. 117, no. F3, p. n/a-n/a, Sep. 2012, doi: 10.1029/2011JF002143.
[14] J.-M. st. Jacques and D. J. Sauchyn, “Increasing winter baseflow and mean annual streamflow from possible permafrost thawing in the Northwest Territories, Canada,” Geophysical Research Letters, vol. 36, no. 1, p. L01401, Jan. 2009, doi: 10.1029/2008GL035822.
[15] M. Sandells and D. Flocco, Introduction to the Physics of the Cryosphere, 1st ed., vol. 1. 2014.
[16] X. Hu and W. H. Pollard, “The Hydrologic Analysis and Modelling of River Icing Growth, North Fork Pass, Yukon Territory, Canada,” Permafrost and Periglacial Processes, vol. 8, no. 3, pp. 279–294, Sep. 1997, doi: 10.1002/(SICI)1099-1530(199709)8:3<279::AID-PPP260>3.0.CO;2-7.
[17] N. Colombo et al., “Review: Impacts of permafrost degradation on inorganic chemistry of surface fresh water,” Global and Planetary Change, vol. 162, pp. 69–83, Mar. 2018, doi: 10.1016/j.gloplacha.2017.11.017.
[18] J. E. Vonk, S. E. Tank, and M. A. Walvoord, “Integrating hydrology and biogeochemistry across frozen landscapes,” Nature Communications, vol. 10, no. 1, p. 5377, Dec. 2019, doi: 10.1038/s41467-019-13361-5.
[19] J. L. Weishaar, G. R. Aiken, B. A. Bergamaschi, M. S. Fram, R. Fujii, and K. Mopper, “Evaluation of Specific Ultraviolet Absorbance as an Indicator of the Chemical Composition and Reactivity of Dissolved Organic Carbon,” Environmental Science & Technology, vol. 37, no. 20, pp. 4702–4708, Oct. 2003, doi: 10.1021/es030360x.
[20] C. A. Stedmon and R. Bro, “Characterizing dissolved organic matter fluorescence with parallel factor analysis: a tutorial,” Limnology and Oceanography: Methods, vol. 6, no. 11, pp. 572–579, Nov. 2008, doi: 10.4319/lom.2008.6.572.
[21] H. Crites, S. v. Kokelj, and D. Lacelle, “Icings and groundwater conditions in permafrost catchments of northwestern Canada,” Scientific Reports, vol. 10, no. 1, p. 3283, Dec. 2020, doi: 10.1038/s41598-020-60322-w.
[22] R. H. S. Hutchins, P. Aukes, S. L. Schiff, T. Dittmar, Y. T. Prairie, and P. A. del Giorgio, “The Optical, Chemical, and Molecular Dissolved Organic Matter Succession Along a Boreal Soil-Stream-River Continuum,” Journal of Geophysical Research: Biogeosciences, vol. 122, no. 11, pp. 2892–2908, Nov. 2017, doi: 10.1002/2017JG004094.
[2] S. E. Tank, R. G. Striegl, J. W. McClelland, and S. v Kokelj, “Multi-decadal increases in dissolved organic carbon and alkalinity flux from the Mackenzie drainage basin to the Arctic Ocean,” Environmental Research Letters, vol. 11, no. 5, p. 054015, May 2016, doi: 10.1088/1748-9326/11/5/054015.
[3] K. Kaiser, M. Canedo-Oropeza, R. McMahon, and R. M. W. Amon, “Origins and transformations of dissolved organic matter in large Arctic rivers,” Scientific Reports, vol. 7, no. 1, p. 13064, Dec. 2017, doi: 10.1038/s41598-017-12729-1.
[4] M. S. Schwab et al., “An Abrupt Aging of Dissolved Organic Carbon in Large Arctic Rivers,” Geophysical Research Letters, vol. 47, no. 23, Dec. 2020, doi: 10.1029/2020GL088823.
[5] T. J. Battin, S. Luyssaert, L. A. Kaplan, A. K. Aufdenkampe, A. Richter, and L. J. Tranvik, “The boundless carbon cycle,” Nature Geoscience, vol. 2, no. 9, pp. 598–600, Sep. 2009, doi: 10.1038/ngeo618.
[6] C. A. Littlefair and S. E. Tank, “Biodegradability of Thermokarst Carbon in a Till-Associated, Glacial Margin Landscape: The Case of the Peel Plateau, NWT, Canada,” Journal of Geophysical Research: Biogeosciences, vol. 123, no. 10, pp. 3293–3307, Oct. 2018, doi: 10.1029/2018JG004461.
[7] J. E. Vonk et al., “Biodegradability of dissolved organic carbon in permafrost soils and aquatic systems: a meta-analysis,” Biogeosciences, vol. 12, no. 23, pp. 6915–6930, Dec. 2015, doi: 10.5194/bg-12-6915-2015.
[8] B. W. Abbott, J. R. Larouche, J. B. Jones, W. B. Bowden, and A. W. Balser, “Elevated dissolved organic carbon biodegradability from thawing and collapsing permafrost,” Journal of Geophysical Research: Biogeosciences, vol. 119, no. 10, pp. 2049–2063, Oct. 2014, doi: 10.1002/2014JG002678.
[9] S. E. Tank et al., “A land‐to‐ocean perspective on the magnitude, source and implication of DIC flux from major Arctic rivers to the Arctic Ocean,” Global Biogeochemical Cycles, vol. 26, no. 4, p. 2011GB004192, Dec. 2012, doi: 10.1029/2011GB004192.
[10] M. A. Walvoord and B. L. Kurylyk, “Hydrologic Impacts of Thawing Permafrost-A Review,” Vadose Zone Journal, vol. 15, no. 6, p. vzj2016.01.0010, Jun. 2016, doi: 10.2136/vzj2016.01.0010.
[11] C. Spence et al., “Hydrological resilience to forest fire in the subarctic Canadian shield,” Hydrological Processes, vol. 34, no. 25, pp. 4940–4958, Dec. 2020, doi: 10.1002/hyp.13915.
[12] P. Lamontagne-Hallé, J. M. McKenzie, B. L. Kurylyk, and S. C. Zipper, “Changing groundwater discharge dynamics in permafrost regions,” Environmental Research Letters, vol. 13, no. 8, p. 084017, Aug. 2018, doi: 10.1088/1748-9326/aad404.
[13] V. F. Bense, H. Kooi, G. Ferguson, and T. Read, “Permafrost degradation as a control on hydrogeological regime shifts in a warming climate,” Journal of Geophysical Research: Earth Surface, vol. 117, no. F3, p. n/a-n/a, Sep. 2012, doi: 10.1029/2011JF002143.
[14] J.-M. st. Jacques and D. J. Sauchyn, “Increasing winter baseflow and mean annual streamflow from possible permafrost thawing in the Northwest Territories, Canada,” Geophysical Research Letters, vol. 36, no. 1, p. L01401, Jan. 2009, doi: 10.1029/2008GL035822.
[15] M. Sandells and D. Flocco, Introduction to the Physics of the Cryosphere, 1st ed., vol. 1. 2014.
[16] X. Hu and W. H. Pollard, “The Hydrologic Analysis and Modelling of River Icing Growth, North Fork Pass, Yukon Territory, Canada,” Permafrost and Periglacial Processes, vol. 8, no. 3, pp. 279–294, Sep. 1997, doi: 10.1002/(SICI)1099-1530(199709)8:3<279::AID-PPP260>3.0.CO;2-7.
[17] N. Colombo et al., “Review: Impacts of permafrost degradation on inorganic chemistry of surface fresh water,” Global and Planetary Change, vol. 162, pp. 69–83, Mar. 2018, doi: 10.1016/j.gloplacha.2017.11.017.
[18] J. E. Vonk, S. E. Tank, and M. A. Walvoord, “Integrating hydrology and biogeochemistry across frozen landscapes,” Nature Communications, vol. 10, no. 1, p. 5377, Dec. 2019, doi: 10.1038/s41467-019-13361-5.
[19] J. L. Weishaar, G. R. Aiken, B. A. Bergamaschi, M. S. Fram, R. Fujii, and K. Mopper, “Evaluation of Specific Ultraviolet Absorbance as an Indicator of the Chemical Composition and Reactivity of Dissolved Organic Carbon,” Environmental Science & Technology, vol. 37, no. 20, pp. 4702–4708, Oct. 2003, doi: 10.1021/es030360x.
[20] C. A. Stedmon and R. Bro, “Characterizing dissolved organic matter fluorescence with parallel factor analysis: a tutorial,” Limnology and Oceanography: Methods, vol. 6, no. 11, pp. 572–579, Nov. 2008, doi: 10.4319/lom.2008.6.572.
[21] H. Crites, S. v. Kokelj, and D. Lacelle, “Icings and groundwater conditions in permafrost catchments of northwestern Canada,” Scientific Reports, vol. 10, no. 1, p. 3283, Dec. 2020, doi: 10.1038/s41598-020-60322-w.
[22] R. H. S. Hutchins, P. Aukes, S. L. Schiff, T. Dittmar, Y. T. Prairie, and P. A. del Giorgio, “The Optical, Chemical, and Molecular Dissolved Organic Matter Succession Along a Boreal Soil-Stream-River Continuum,” Journal of Geophysical Research: Biogeosciences, vol. 122, no. 11, pp. 2892–2908, Nov. 2017, doi: 10.1002/2017JG004094.