Controls on the Respiration of Ancient Carbon Draining From Permafrost Soils Into Sunlit Arctic Surface Waters

Abstract

The thawing of ancient organic carbon stored in arctic permafrost soils, and its oxidation to carbon dioxide (CO2, a greenhouse gas), is predicted to amplify global warming. However, the extent to which organic carbon in thawing permafrost soils will be released as CO2 is uncertain. A critical unknown is the extent to which dissolved organic carbon (DOC) from thawing permafrost soils is respired to CO2 by microbes upon export of freshly thawed DOC to both dark bottom waters and sunlit surface waters. In this study, we quantified the radiocarbon age and ¹³C composition of CO2 produced by microbial respiration of DOC that was leached from permafrost soils and either kept in the dark or exposed to ultraviolet and visible wavelengths of light. We show that permafrost DOC most labile to microbial respiration was as old or older (ages 4,000–11,000 a BP) and more ¹³C‐depleted than the bulk DOC in both dark and light‐exposed treatments, likely indicating respiration of old, ¹³C‐depleted lignin and lipid fractions of the permafrost DOC pool. Light exposure either increased, decreased, or had no effect on the magnitude of microbial respiration of old permafrost DOC relative to respiration in the dark, depending on both the extent of DOC oxidation during exposure to light and the wavelength of light. Together, these findings suggest that photochemical changes affecting the lability of permafrost DOC during sunlight exposure are an important control on the magnitude of microbial respiration of permafrost DOC in arctic surface waters.

Full text

Controls on the Respiration of Ancient Carbon Draining
From Permafrost Soils Into Sunlit Arctic Surface Waters
E. C. Rieb
1
, C. A. Polik
1,2
, C. P. Ward
3
, G. W. Kling
4
, and R. M. Cory
1
1
Department of Earth and Environmental Sciences, University of Michigan, Ann Arbor, MI, USA,
2
Now at Department of
Ecology, Evolution, and Behavior, University of Minnesota, St. Paul, MN, USA,
3
Department of Marine Chemistry and
Geochemistry, Woods Hole Oceanographic Institution, Woods Hole, MA, USA,
4
Department of Ecology and Evolutionary
Biology, University of Michigan, Ann Arbor, MI, USA
Abstract The thawing of ancient organic carbon stored in arctic permafrost soils, and its oxidation to carbon
dioxide (CO
2
, a greenhouse gas), is predicted to amplify global warming. However, the extent to which organic
carbon in thawing permafrost soils will be released as CO
2
is uncertain. A critical unknown is the extent to
which dissolved organic carbon (DOC) from thawing permafrost soils is respired to CO
2
by microbes upon
export of freshly thawed DOC to both dark bottom waters and sunlit surface waters. In this study, we quantified
the radiocarbon age and
13
C composition of CO
2
produced by microbial respiration of DOC that was leached
from permafrost soils and either kept in the dark or exposed to ultraviolet and visible wavelengths of light. We
show that permafrost DOC most labile to microbial respiration was as old or older (ages 4,000–11,000 a BP) and
more
13
C‐depleted than the bulk DOC in both dark and light‐exposed treatments, likely indicating respiration of
old,
13
C‐depleted lignin and lipid fractions of the permafrost DOC pool. Light exposure either increased,
decreased, or had no effect on the magnitude of microbial respiration of old permafrost DOC relative to
respiration in the dark, depending on both the extent of DOC oxidation during exposure to light and the
wavelength of light. Together, these findings suggest that photochemical changes affecting the lability of
permafrost DOC during sunlight exposure are an important control on the magnitude of microbial respiration of
permafrost DOC in arctic surface waters.
Plain Language Summary Organic carbon has been frozen in arctic permafrost soils for thousands
of years. As these soils warm and thaw, the vast stores of ancient organic carbon they contain can amplify global
warming if the carbon is converted to carbon dioxide (a greenhouse gas) in both dark soils and when drained into
sunlit lakes and streams. However, quantifying the carbon dioxide produced from thawed permafrost organic
carbon requires understanding the controls on this conversion. Here we show that native permafrost microbes
can respire permafrost organic carbon to carbon dioxide both in the dark and after carbon exposure to different
wavelengths of light. The carbon dioxide produced was as old as the organic carbon in the soil, suggesting that
even ancient carbon can be respired by microbes. This study suggests that respiration by microbes will
contribute to the rapid conversion of ancient permafrost organic carbon to carbon dioxide in sunlit arctic surface
waters, further amplifying global warming.
1. Introduction
Permafrost soils are thawing in many regions of the Arctic (e.g., Jorgenson et al., 2006; Osterkamp, 2007; Smith
et al., 2022; Zhao et al., 2020). Once thawed, the ancient organic carbon in permafrost soils can be respired to
carbon dioxide (CO
2
, a greenhouse gas; Schuur et al., 2015; Vaughn & Torn, 2019), contributing to the arctic
amplification of climate change (McGuire et al., 2018). Permafrost thaw and thaw slump disturbances (ther-
mokarsts; Kokelj & Jorgenson, 2013; Olefeldt et al., 2016) may also increase the lateral export of old, previously‐
frozen dissolved organic carbon (DOC) from permafrost soils to surface waters (Frey & McClelland, 2009;
McFarlane et al., 2022; Plaza et al., 2019). DOC draining from permafrost soils to arctic streams is labile to
microbial respiration (Drake et al., 2015; Mann et al., 2015; Spencer et al., 2015), particularly after the
composition of this DOC has been altered by sunlight exposure (i.e., coupled photochemical and microbial
degradation of DOC; Cory et al., 2013; Ward et al., 2017).
In arctic freshwaters, coupled photochemical and microbial degradation of DOC may account for over 90% of the
total DOC processed in the water column (Cory et al., 2014). Sunlight has a strong influence on DOC composition
and thus on its respiration to CO
2
in arctic waters due to high concentrations of sunlight‐absorbing
RESEARCH ARTICLE
10.1029/2023JG007853
Key Points:
•Microbes respire ancient,
13
C‐depleted
dissolved organic carbon (DOC) from
permafrost soils to carbon dioxide
•Effects of sunlight exposure on the
amount of permafrost DOC respired
depend on the extent of DOC oxidation
by each wavelength of light
•Sunlight exposure of permafrost DOC
can increase or decrease microbial
respiration in arctic lakes and streams
Supporting Information:
Supporting Information may be found in
the online version of this article.
Correspondence to:
R. M. Cory,
rmcory@umich.edu
Citation:
Rieb, E. C., Polik, C. A., Ward, C. P.,
Kling, G. W., & Cory, R. M. (2024).
Controls on the respiration of ancient
carbon draining from permafrost soils into
sunlit arctic surface waters. Journal of
Geophysical Research: Biogeosciences,
129, e2023JG007853. https://doi.org/10.
1029/2023JG007853
Received 24 OCT 2023
Accepted 10 APR 2024
Author Contributions:
Conceptualization: E. C. Rieb,
C. A. Polik, C. P. Ward, G. W. Kling,
R. M. Cory
Data curation: E. C. Rieb
Formal analysis: E. C. Rieb
Funding acquisition: E. C. Rieb,
C. P. Ward, G. W. Kling, R. M. Cory
Investigation: E. C. Rieb, C. A. Polik,
C. P. Ward
Methodology: E. C. Rieb, C. A. Polik,
C. P. Ward, G. W. Kling, R. M. Cory
Resources: C. P. Ward, G. W. Kling,
R. M. Cory
Software: E. C. Rieb, C. A. Polik
Supervision: C. P. Ward, G. W. Kling,
R. M. Cory
Validation: E. C. Rieb, C. P. Ward
Visualization: E. C. Rieb, R. M. Cory
© 2024. The Authors.
This is an open access article under the
terms of the Creative Commons
Attribution License, which permits use,
distribution and reproduction in any
medium, provided the original work is
properly cited.
RIEB ET AL. 1 of 19

(chromophoric) DOC (CDOM) that absorb all or nearly all incoming ultraviolet (UV) and visible sunlight in the
water column (Cory et al., 2014,2015). Rates of sunlight degradation of DOC are not strongly influenced by
water column depth, mixing, or turbidity in arctic surface waters (Cory et al., 2015; Cory & Kling, 2018; Li
et al., 2019). Further, permafrost DOC is readily degraded by even the less energetic visible wavelengths of
sunlight (Bowen et al., 2020b; Ward & Cory, 2016).
Most of the permafrost DOC degraded by UV and visible sunlight remains in the DOC pool as partially oxidized
or chemically‐altered compounds (Stubbins et al., 2017; Ward & Cory, 2016,2020). Thus, even after sunlight
degradation, permafrost DOC may be detected by its radiocarbon age in surface waters, as reported in headwater
streams impacted by thawing permafrost (Mann et al., 2015; Neff et al., 2006; Spencer et al., 2015; Vonk
et al., 2013) and in one major arctic river (Schwab et al., 2020). As permafrost DOC is increasingly exposed to
sunlight during transport from streams to rivers, its lability to microbes should increase (Cory et al., 2013; Ward
et al., 2017), promoting its respiration to CO
2
. Consistent with this expected loss of permafrost DOC by coupled
photochemical and microbial degradation, most studies have reported little to no old DOC in major arctic rivers
receiving DOC from streams impacted by permafrost thaw (Aiken et al., 2014; Neff et al., 2006; Rogers
et al., 2021; Wild et al., 2019).
In addition to loss of permafrost DOC by coupled photochemical and microbial degradation, preferential
degradation of the oldest fractions of the permafrost DOC by these processes could mask the presence of ancient
DOC in streams. Permafrost DOC is a mixture of thousands of compounds that span a range of ages (e.g., Rogers
et al., 2021). There is evidence that older DOC is selectively respired relative to modern DOC in rivers and lakes
(Mann et al., 2015; McCallister & del Giorgio, 2012). McCallister and del Giorgio (2012) hypothesized that
selective respiration of relatively old DOC was due to its degradation by sunlight. For the mixture of DOC ages
found in permafrost DOC, this hypothesis is supported by evidence that sunlight degradation of permafrost DOC
produced the same low molecular weight, aliphatic DOC compounds most labile to microbial respiration in the
dark (Nalven et al., 2020; Ward et al., 2017). However, it is not known whether sunlight exposure makes rela-
tively older or younger compounds within the old permafrost DOC pool more labile to microbes.
The
13
C composition of the permafrost DOC respired by microbes, in addition to the age, may characterize the
chemical composition of permafrost DOC most labile to microbes. Prior work showed that the fraction of
permafrost DOC respired by microbes in the dark has a
13
C composition that is similar to the bulk DOC (Mann
et al., 2015; Spencer et al., 2015). A null interpretation of these results is that all
13
C compositions of permafrost
DOC are similarly labile to microbial respiration. Another interpretation is that the relatively large amount of
DOC respired masked any preferential respiration of labile fractions of DOC differing in their
13
C composition
compared to the bulk permafrost DOC. For example, prior studies used warm incubations (20°C) supporting high
respiration rates and consumption of up to 50% of the DOC. As the amount of DOC consumed increases, the
isotopic signature of the CO
2
produced should increasingly match the average isotopic signature of the bulk DOC.
Therefore, the similarity between the
13
C composition of the bulk and respired permafrost DOC in prior work
does not necessarily imply that all fractions of the bulk permafrost DOC are similarly labile to microbial
respiration. In contrast, in arctic surface waters impacted by permafrost thaw, such as small headwater streams,
conditions including ice‐free average water temperatures of 3–12°C (Adams et al., 2010; Cory et al., 2014;
Docherty et al., 2019), short residence times, and rapid replenishment with fresh, terrestrial DOC (Cory
et al., 2015; Neilson et al., 2018) may favor respiration of a smaller percentage of the DOC pool. Under these
conditions, it is not known whether microbes selectively respire compound classes that are depleted or enriched in
13
C compared to the bulk DOC.
The respiratory quotient of DOC respired may also provide information on fractions of permafrost DOC most
labile to microbes. This quotient of CO
2
produced to O
2
consumed during respiration is thought to depend on the
oxidation state of the substrate respired (Masiello et al., 2008). Substrates with a lower average oxidation state of
carbon are hypothesized to yield relatively less CO
2
per mol O
2
consumed compared to substrates having a higher
average oxidation state of the carbon (Masiello et al., 2008; Pries et al., 2020). This hypothesis is based on
respiratory quotients for glucose, oxalic acid, and other small organic compounds (Dilly, 2001; Theenhaus
et al., 1997). While the respiratory quotients for larger or more heterogeneous DOC compounds like lipids or
lignin‐like DOC have not been tested, this hypothesis predicts that more reduced lipid and lignin‐like compounds
are respired with a lower respiratory quotient than more oxidized carbohydrates and organic acids (Dilly, 2001;
Romero‐Kutzner et al., 2015).
Writing – original draft: E. C. Rieb,
R. M. Cory
Writing – review & editing: E. C. Rieb,
C. A. Polik, C. P. Ward, G. W. Kling,
R. M. Cory
Journal of Geophysical Research: Biogeosciences
10.1029/2023JG007853
RIEB ET AL. 2 of 19

Finally, another major uncertainty preventing a quantitative assessment of the effects of sunlight exposure on
respiration of permafrost DOC is the wavelength dependence of this process. The one study that has isolated the
effects of individual wavelengths of light on DOC degradation reports substantial variability in both the direction
and magnitude of the effects of ultraviolet (UV) and visible wavelengths of light on microbial respiration (Reader
& Miller, 2014). Because more visible than UV photons of sunlight are absorbed by DOC, incorrect assumptions
about the relative importance of UV and visible light for producing DOC labile to microbes can lead to a sys-
tematic under‐ or overestimation of CO
2
from microbial respiration in arctic surface waters.
To address these knowledge gaps, permafrost DOC spanning a range of ages and compositions was collected from
younger and older glacial surfaces and from under two common vegetation types in the Alaskan Arctic.
Permafrost DOC was exposed to UV and visible wavelengths of LED‐generated light, alongside dark controls.
Following light exposure, biological incubations at water temperatures of arctic surface waters were conducted to
make the first direct quantifications of the radiocarbon age of CO
2
produced by microbial respiration of ancient
permafrost DOC, both in the dark and after light exposure. The
13
C composition of CO
2
from respiration and
respiratory quotients were also quantified to characterize the composition of permafrost DOC most labile to
microbes.
2. Materials and Methods
2.1. Soil Collection
Permafrost soil cores were collected on the North Slope of Alaska during the ice‐free summer months of June‐
August 2018 near the Toolik Field Station (Table S1 in Supporting Information S1). In the foothills of this region,
mountain glaciations have produced land surfaces of different ages from relatively younger (∼14,000 years BP
since last glaciation) to relatively older (>250,000 years BP since last glaciation; Hamilton, 2003). Soil cores
were collected from within the permafrost layer (at 85 cm below the surface and ∼10–30 cm below the maximum
summer thaw depth; see Romanowicz & Kling, 2022) of Imnavait Creek wet sedge (wet sedge) and Toolik Lake
tussock tundra (tussock tundra) soils, which represent the dominant landscape ages and vegetation types of the
low Arctic (Figure S1 and Table S1 in Supporting Information S1; Ping et al., 1998; Trusiak et al., 2018a; Walker
et al., 2005; Walker & Maier, 2008). DOC leached from these soils has a range of chemical compositions and ages
(Bowen et al., 2020b; Trusiak et al., 2018a; Ward et al., 2017). In June 2022, soil was sampled from a thermokarst
failure on the shore of Lake LTER 395 on the North Slope of Alaska (referred to as “thermokarst”), where an
abrupt collapse of thawing soil exposed deeper permafrost soil (Figure S1 and Table S1 in Supporting Infor-
mation S1). Soil was sampled from the permafrost layer on a freshly‐cleaned soil profile in the headwall of the
thermokarst failure (>80 cm below the surface and ∼10–30 cm below the maximum summer thaw depth) using
MilliQ‐rinsed pickaxes.
The permafrost and thermokarst soil samples were collected as previously described in detail (Bowen
et al., 2020b), including precautions to minimize radiocarbon (
14
C) contamination by rinsing gloves and tools
with deionized water prior to soil collection and storing soil samples in
14
C‐free facilities and freezers. These
protocols were shown to result in no detectable
14
C contamination of soils (Bowen et al., 2020b). All soils were
stored in freezers at the Toolik Field Station until overnight shipment to Woods Hole Oceanographic Institution
(WHOI), where soils were stored in
14
C‐free freezers until further use.
2.2. Soil Leachate Preparation and Characterization
DOC was leached from the permafrost and thermokarst soils as previously described (Bowen et al., 2020b).
Briefly, frozen soil and UVC‐oxidized MilliQ water (Table S2 in Supporting Information S1) were mixed in 5‐
gallon, MilliQ‐rinsed HDPE buckets and allowed to leach in the dark for 1–4 days at 4°C. Both the soil‐to‐water
ratio of soil leachates and the leaching time were adjusted to achieve a final concentration of ∼500–1,500 μM
DOC in the leachates, as estimated from the absorbance of chromophoric dissolved organic matter at 305 nm
(a
305
).
All leachates were passed through MilliQ‐rinsed 60 μm mesh screens to remove the largest particulates and then
through MilliQ‐rinsed 5 μm high‐capacity Whatman cartridge filters. Subsamples of each leachate for light
exposure experiments were then filtered through 0.2 μm high‐capacity cartridge filters to minimize microbial
activity (Step A in Figure S2 in Supporting Information S1; Ward et al., 2017). Additional subsamples of each
Journal of Geophysical Research: Biogeosciences
10.1029/2023JG007853
RIEB ET AL. 3 of 19

leachate were prepared as inoculum for biological incubations by passing the 5‐μm‐filtered water through 1.2 μm
glass‐fiber filters (hereafter referred to as the inoculum; Step B in Figure S2 in Supporting Information S1). All
filters used to prepare leachates were rinsed with 5 L of MilliQ water before use. Leach tests for DOC
contamination from the filters found that less than 5 μM DOC leached into MilliQ rinses from the Whatman and
Sterivex filters. Thus, any
14
C contamination from filtering permafrost leachates was within the instrumental
precision of the radiocarbon analyses (≤6‰; see Bowen et al., 2020b). Prior work following the same protocols
as in this study for leachate preparation concluded that potential
14
C contamination during the soil leaching
process before filtration was also within the instrumental precision of the radiocarbon analyses (Bowen
et al., 2020b). Soil leachates for light exposure experiments and biological incubations were stored at 4°C until
further use (less than 48 hr for light exposure experiments, and less than 1 week for biological incubations).
Supporting soil leachate chemistry analyses (pH, specific conductivity, iron, DOC, and chromophoric and
fluorescent dissolved organic matter) were performed as previously described (Bowen et al., 2020b; Cory
et al., 2013,2014; Kling et al., 2000).
Due to sample loss during experiments and
14
C analyses, duplicate leachates were prepared from the thermokarst
and tussock tundra soils. From the thermokarst soil, the first leachate prepared was used for both UV and visible
light treatments and the second leachate was used for the dark control. From the tussock tundra soil, there was
sample loss from the first leachate (labeled A) for some of the dark and visible light treatment samples (Table 1).
Thus, the second leachate prepared from the tussock tundra soil (labeled B) replaced the samples lost from
leachate A and provided experimental duplicates for the samples analyzed from leachate A (Table 1). See Section
S1.1 in Supporting Information S1 for details.
2.3. LED Light Exposure Experiments
Each 0.2‐μm‐filtered soil leachate was allowed to warm from the 4°C storage temperature to room temperature for
12–24 hr prior to the start of dark or light treatments. Each soil leachate was then placed in six precombusted,
500 mL quartz flasks with ground glass stoppers without headspace. For each soil leachate, duplicate quartz flasks
Table 1
Δ
14
C and δ
13
C of Permafrost DOC and the CO
2
From Respiration of Permafrost DOC During Biological Incubations
Treatment Imnavait wet sedge tundra Toolik tussock tundra A Toolik tussock tundra B LTER 395 thermokarst
Δ
14
C‐DOC (‰) Dark 594 ±2502 ±2441 ±2433 ±1
UV 589 ±2502 ±2432 ±1440 ±1
Visible 582 ±2502 ±2439 ±2437 ±1
Δ
14
C‐CO
2
respired (‰) Dark 689 ±52 668 ±22 448 ±6
UV 572 ±2740 ±40 597 ±2415 ±1
Visible 627 ±16 624 ±4405 ±0
14
C age of DOC (a BP) Dark 7,170 ±35 5,540 ±30 4,610 ±25 4,500 ±20
UV 7,080 ±35 5,540 ±30 4,470 ±20 4,600 ±20
Visible 6,940 ±35 5,540 ±30 4,580 ±20 4,560 ±20
14
C age of CO
2
respired (a BP) Dark 9,439 ±1,360 8,815 ±541 4,711 ±81
UV 6,755 ±40 10,835 ±1,240 7,229 ±35 4,247 ±10
Visible 7,853 ±339 7,793 ±94 4,100 ±5
δ
13
C‐DOC (‰) Dark 28.8 25.3 25.9 26.3
UV 28.1 25.6 25.5 25.9
Visible 28.2 25.3 25.7 26.0
δ
13
C‐CO
2
respired (‰) Dark 40.2 ±2.7 31.9 ±1.3 32.2 ±0.7
UV 37.3 ±0.5 44.9 ±3.0 29.9 ±1.0 28.5 ±0.3
Visible 32.8 ±0.6 28.1 ±0.2
Note. The Δ
14
C and δ
13
C of permafrost DOC are reported following dark treatment, UV light treatment (305 nm), or visible light treatment (405 nm), and subsequent
inoculation. The Δ
14
C and δ
13
C of the CO
2
from respiration of permafrost DOC are reported following biological incubations. Values for DOC are reported with the
instrumental error, and values for CO
2
, when available, are reported as the average ±1 SE of experimental replicates (n=2).
Journal of Geophysical Research: Biogeosciences
10.1029/2023JG007853
RIEB ET AL. 4 of 19

were exposed to 305 and 405 nm LED light treatments using custom‐built 10 ×1 LED chip arrays maintained at
30°C using heat sinks and cooling fans (Bowen et al., 2020b; Ward et al., 2021). These wavelengths representing
UV (305 nm) and visible (405 nm) sunlight were chosen because studies suggest that production of DOC photo‐
products labile to microbial respiration is highest near 305 nm (Miller et al., 2002) and differs by wavelength
(Reader & Miller, 2014; Wetzel et al., 1995). Duplicate dark controls were run alongside light treatments at room
temperature (23°C) in the dark (Step C in Figure S2 in Supporting Information S1).
Changes in water chemistry and photochemical O
2
consumption during light exposure were measured as in prior
work (Table S3 in Supporting Information S1, Ward & Cory, 2016). The amount of DOC oxidized to CO
2
during
LED light exposures was estimated from photochemical O
2
consumption, assuming 1 mol DOC completely
oxidized to CO
2
per mol O
2
consumed (Cory et al., 2014; Ward & Cory, 2020). This is a conservative estimate of
the amount of DOC oxidized because the ratio of CO
2
produced per O
2
consumed is often >1 in high‐DOC and
high‐iron waters similar to permafrost leachates (Bowen et al., 2020b; Ward & Cory, 2020). The durations of light
exposures were chosen to achieve complete oxidation of ∼5%‐10% of the initial DOC from all leachates and
ranged from 20 to 120 hr depending on the soil site and wavelength (Table S3 in Supporting Information S1).
Oxidation of 5%–10% of the DOC was chosen because prior work showed substantial microbial response (in both
respiration and bacterial production) upon oxidation of <5% of the DOC (Nalven et al., 2020; Ward et al., 2017).
In addition, short residence times in headwater streams impacted by permafrost thaw likely mean that only a small
percent of the DOC is oxidized before export downstream (e.g., Cory & Kaplan, 2012). For each soil site, the
same amount of DOC was oxidized by both UV and visible light (Table S3 in Supporting Information S1).
The duplicates of soil leachates from each dark or light treatment were composited in precombusted glass bottles
(Step D in Figure S2 in Supporting Information S1) and stored overnight in the dark at 4°C before further use in
biological incubations. This storage period allowed for decay of reactive oxygen species produced upon exposure
of DOC to UV and visible light (Andrews et al., 2000; Cory et al., 2010; Page et al., 2014; White et al., 2003).
2.4. Biological Incubations
Each soil leachate treatment (dark, UV, visible) was mixed with the respective inoculum from each site to achieve
20% inoculum by volume (Cory et al., 2013; Step D in Figure S2 in Supporting Information S1). From each
inoculated dark and light‐exposed leachate, replicates for
14
C and
13
C analysis of the CO
2
produced by microbial
respiration were filled in precombusted 125 mL borosilicate bottles with greased glass stoppers and no headspace
(Step E in Figure S2 in Supporting Information S1). For each dark or light‐exposed leachate, there were duplicate
viable and killed treatments to quantify the C isotopic signatures of the CO
2
produced during respiration. The
viable treatment was unamended, and the killed treatment was amended with saturated mercuric chloride (all
preservations had 1% HgCl
2
by sample volume).
The amount of DOC respired from each dark and light‐exposed leachate was quantified as O
2
consumption and
CO
2
production by microbial respiration (Table S4 in Supporting Information S1). O
2
consumed and CO
2
pro-
duced by respiration were quantified as the difference in dissolved O
2
and dissolved inorganic carbon (DIC),
respectively, between viable and killed treatments from a split of each inoculated dark and light‐exposed leachate
placed in precombusted, gas‐tight 12 mL soda glass exetainers with no headspace (Step E in Figure S2 in
Supporting Information S1). For each dark and light‐exposed leachate, there were triplicate viable and killed
treatments for analysis of O
2
and DIC. The viable treatment was unamended, and the killed treatment was
amended with saturated mercuric chloride.
All viable and killed inoculated soil leachates were incubated in the dark at 10°C for between 17 and 30 days (Step
E in Figure S2 and Table S4 in Supporting Information S1). This incubation temperature is representative of a
typical, annual average temperature for non‐frozen lakes and streams near soil sampling sites (Cory et al., 2014).
The incubation duration was chosen for each set of dark and light‐exposed soil leachates from a soil site so that
respiration produced at least a 10% increase in the total DIC of the water compared to the start of the incubation
(Table S4 in Supporting Information S1). A 10% increase in the total DIC was chosen so that the
14
C and
13
C
isotopic compositions of the CO
2
produced by microbial respiration were detectable against the isotopic com-
positions of the background DIC in the soil leachates. At the end of the incubations, all viable leachate treatments
in 125 mL borosilicate bottles and 12 mL exetainers were preserved by addition of saturated mercuric chloride.
These preserved samples were stored at 4°C in the dark until
14
C and
13
C analysis (up to 2 months) or until
analysis of O
2
consumption and CO
2
production by respiration (less than 2 weeks) as previously described
Journal of Geophysical Research: Biogeosciences
10.1029/2023JG007853
RIEB ET AL. 5 of 19

(Bowen et al., 2020b; Cory et al., 2014). Respiratory quotients were calculated as the ratio of CO
2
production to
O
2
consumption by microbial respiration.
2.5. Δ
14
C and δ
13
C of DOC
A 75 mL sample of each soil leachate after dark or light treatment and inoculation (Step D in Figure S2 in
Supporting Information S1) was 0.22‐μm Sterivex filtered and frozen for
14
C and
13
C analysis of the DOC present
at the start of the biological incubations (Table 1and Table S5 in Supporting Information S1). The Δ
14
C and δ
13
C
of the DOC were quantified at the National Ocean Sciences Accelerator Mass Spectrometry (NOSAMS) Facility
at WHOI, using previously described methods (Xu et al., 2021). Briefly, soil leachates were acidified to pH <2
with UVC‐oxidized trace‐metal grade phosphoric acid (85%) and stripped of dissolved inorganic carbon (DIC)
with high‐purity helium gas in the dark. The DOC was then oxidized with UVC light to DIC, and the resultant
CO
2
was extracted cryogenically. A subsample of the CO
2
was analyzed for
13
C using a VG Prism‐II or Optima
stable isotope ratio mass spectrometer, and the δ
13
C (‰) was calculated as follows:
δ13C=13 Rsample /13Rstandard – 1))
where
13
R is the isotope ratio of a sample or standard (VPDB), as defined by:
13R=13 C/12C)
The remaining CO
2
was reduced to graphite with H
2
and an iron catalyst, and then analyzed for
14
C isotopic
composition using an accelerator mass spectrometer at the NOSAMS facility (Longworth et al., 2015). The Δ
14
C
(‰) and radiocarbon age of DOC were calculated from the fraction modern using the oxalic acid I standard
(NIST‐SRM 4990).
2.6. Δ
14
C and δ
13
C of CO
2
From Microbial Respiration
To characterize the isotopic composition of the DOC respired by microbes, the Δ
14
C and δ
13
C of dissolved
inorganic carbon (DIC) were quantified in duplicate at NOSAMS from the viable and killed treatments of each
dark and light‐exposed leachate at the end of the incubation (after Step E in Figure S2 in Supporting Informa-
tion S1), following procedures previously described for quantification of the Δ
14
C and δ
13
C of CO
2
produced
from photomineralization of permafrost DOC (Bowen et al., 2020b). Water samples were acidified as described in
Section 2.5 and stripped of DIC using high‐purity nitrogen gas. The
14
C and
13
C of the resultant, trapped and
purified CO
2
were analyzed at the NOSAMS facility and converted to Δ
14
C and δ
13
C values as described in
Section 2.5.
The Δ
14
C and δ
13
C of CO
2
produced by microbial respiration of permafrost DOC (Δ
14
C
resp
and δ
13
C
resp
) were
calculated as follows:
∆14Cresp =∆14 CViable ×[DIC]Viable)∆14 CKill ×[DIC]Kill)
[DIC]Viable [DIC]Kill
δ13Cresp =δ13 CViable ×[DIC]Viable)δ13 CKill ×[DIC]Kill)
[DIC]Viable [DIC]Kill
The Δ
14
C and δ
13
C of CO
2
produced by microbial respiration of permafrost DOC are reported as the average ±1
standard error (SE) of duplicate viable treatments relative to duplicate killed controls (Table 1and Table S5 in
Supporting Information S1). See Section S1.2 in Supporting Information S1 for details about statistical analyses.
3. Results
3.1. DOC and Leachate Characterization
All soil leachates were relatively dilute (specific conductivity 8 to 24 µS cm
1
; Table S2 in Supporting Infor-
mation S1), mildly acidic (pH 5.5 to 6.8, Table S2 in Supporting Information S1), and relatively high in both DOC
Journal of Geophysical Research: Biogeosciences
10.1029/2023JG007853
RIEB ET AL. 6 of 19

concentration (609–1,726 μM C; Table S2 in Supporting Information S1) and total dissolved iron concentration
(3.1–22.2 μM iron) compared to most arctic surface waters that are not directly impacted by permafrost thaw
(Aiken et al., 2014; Cory et al., 2013,2014). The composition of the DOC in all leachates was characteristic of
DOC draining from permafrost soils based on the relatively low SUVA
254
and high fluorescence index (Abbott
et al., 2014; Cory et al., 2013; Mann et al., 2014; Spencer et al., 2015; Stubbins et al., 2017). Compared to pore
waters from the upper, thawed soil layer at the same sites in the Alaskan Arctic, the permafrost soil leachates had
lower conductivity but similar pH and DOC concentrations (Page et al., 2013,2014; Trusiak et al., 2018a,2018b).
Thus, the composition of the permafrost leachates in this study was similar to the composition of both streams and
soil pore waters in permafrost tundra.
3.2. Respiration of DOC
Respiration produced from 8 ±1 to 93 ±7 μM CO
2
(mean ±1 SE) over the 17 to 30‐day incubations at 10°C for
all dark and light‐exposed soil leachates (Figure 1, Table S4 in Supporting Information S1), which was from
0.7 ±0.1 to 5.4 ±0.4% of the initial DOC. The rate of CO
2
production was significantly, positively correlated
with the DOC concentration at the start of the incubation (Figure S4 in Supporting Information S1,p<0.01).
Light exposure increased, decreased, or did not change the amount of DOC respired to CO
2
compared to
respiration of DOC kept in the dark, with the effect of light differing by wavelength and by soil site. For the wet
sedge, tussock tundra B, and thermokarst soil leachates, microbial respiration was 76%, 16%, and 219% higher,
respectively, in the UV light treatment than in the dark treatment (Figure 1, Figure S3 in Supporting Informa-
tion S1). Respiration from the tussock tundra A leachate was 44% lower in the UV treatment than in the dark
treatment. For the thermokarst and tussock tundra A leachates, respiration was 153% and 14% higher, respec-
tively, in the visible light treatments than in the dark treatment (Figure 1, Figure S3 in Supporting Information S1).
Respiration from the wet sedge leachate was 25% lower in the visible light treatment than in the dark treatment.
For the tussock tundra B leachate, there was no significant difference in respiration between the UV and visible
treatments and the dark treatment (Figure 1, Figure S3 in Supporting Information S1).
The respiratory quotient (the ratio of CO
2
production to O
2
consumption by microbial respiration) for respiration
of DOC from the dark and light‐exposed soil leachates ranged from 0.61 ±0.16 to 1.47 ±0.15 (Figure S3 and
Table S4 in Supporting Information S1). Microbes respired UV‐exposed wet sedge DOC with a significantly
higher respiratory quotient compared to the dark DOC from this soil site. In contrast, microbes respired UV‐
exposed tussock tundra A DOC with a lower respiratory quotient compared to the dark DOC. For the tussock
tundra B and thermokarst leachates, the respiratory quotient was not significantly different between dark and UV‐
Figure 1. CO
2
produced versus O
2
consumed by microbial respiration of DOC kept in the dark (gray), exposed to ultraviolet
light (UV; 305 nm, teal), and exposed to visible light (405 nm, orange) for: Imnavait wet sedge tundra (circle symbols),
Toolik tussock tundra A (square symbols), Toolik tussock tundra B (diamond symbols), and LTER 395 thermokarst (triangle
symbols). Panel (a) shows all of the data, and the region outlined in red is enlarged in panel (b). Data are plotted with the 1:1
line (dashed). Values for CO
2
production and O
2
consumption are shown as the average ±1 SE of experimental replicates
(n=3). When no error bars are visible, they are smaller than the data point symbols.
Journal of Geophysical Research: Biogeosciences
10.1029/2023JG007853
RIEB ET AL. 7 of 19

light exposed DOC. Unlike UV light, visible light exposure resulted in no significant difference in respiratory
quotient compared to the dark DOC for any soil sites.
3.3. Δ
14
C of Initial and Respired DOC
The age of the DOC in the soil leachates after dark or light treatment and inoculation (Step D in Figure S2 in
Supporting Information S1) ranged from ∼4,500 a BP for thermokarst and tussock tundra A and B DOC to
∼7,000 a BP for wet sedge DOC (Table 1). DOC in leachates of soils collected from the younger‐age landscape
(tussock tundra A, tussock tundra B, and thermokarst; Table S1 in Supporting Information S1) was younger than
the DOC in the wet sedge leachate, which was collected from the older‐age landscape. The CO
2
from respiration
in all dark and light‐exposed leachates had a wider range in age than did the DOC, from 4,100 to 10,835 a BP
(Table 1). There was no significant correlation between the Δ
14
C of the DOC and the Δ
14
C of the CO
2
from
respiration (Figure 2a, Table 1).
For all of the soil leachates paired with a dark control, CO
2
from respiration of light‐exposed DOC was younger
than the CO
2
from respiration of DOC kept in the dark. The degree to which CO
2
from respiration of light‐
exposed DOC was enriched in
14
C relative to respiration of DOC kept in the dark ranged from 33 ±6‰ for
UV‐treated thermokarst leachate to 117 ±54‰ for UV‐treated wet sedge leachate (Figure 2a).
For the majority of the soil leachates that were exposed to both UV and visible light, there was no wavelength‐
dependent effect of light on the
14
C isotopic composition of the CO
2
from respiration. The exception to this result
was the tussock tundra A leachate, for which the CO
2
from the visible light treatment was significantly enriched in
14
C compared to the CO
2
from the UV light treatment (Figure 2a). For the thermokarst and tussock tundra B
leachates, there was little to no significant difference in the
14
C isotopic composition of the CO
2
from respiration
between the UV and visible light treatments (Figure 2a).
3.4. δ
13
C of Initial and Respired DOC
The
13
C isotopic composition of the DOC in the leachates after dark or light treatment and inoculation (Step D in
Figure S2 in Supporting Information S1) ranged from 28.8‰ to 25.5‰ (Table 1). Wet sedge DOC was the
most
13
C‐depleted (28.4 ±0.2‰), while thermokarst and tussock tundra DOC were relatively more
13
C‐
enriched (26.0 ±0.1 and 25.7 ±0.1‰, respectively; Table 1). The CO
2
produced by microbial respiration
of permafrost DOC had a substantially wider range of
13
C isotopic compositions than did the DOC, from 44.9‰
to 28.1‰, and was always depleted relative to the DOC by 2–19‰ (Table 1). There was no significant cor-
relation between the δ
13
C of the DOC and the δ
13
C of CO
2
from respiration (Figure 2b, Table 1). The δ
13
C of CO
2
Figure 2. (a) Δ
14
C composition and (b) δ
13
C composition of CO
2
produced from microbial respiration of DOC kept in the
dark (circle symbols), exposed to ultraviolet light (UV; 305 nm, diamond symbols), and exposed to visible light (405 nm,
square symbols) versus the composition of the initial DOC of dark or light‐exposed permafrost DOC for: Imnavait wet sedge
tundra (orange symbols), Toolik tussock tundra A (dark blue symbols), Toolik tussock tundra B (yellow symbols), and LTER
395 thermokarst (teal symbols). Data are plotted with the 1:1 line (dashed). Values for Δ
14
C‐CO
2
and δ
13
C‐CO
2
are shown as
the average ±1 SE of experimental replicates (n=2).
Journal of Geophysical Research: Biogeosciences
10.1029/2023JG007853
RIEB ET AL. 8 of 19

from respiration in dark and light‐exposed soil leachates was significantly, positively correlated with both the
respiratory quotient (Figure 3a) and the Δ
14
C of the CO
2
from respiration (Figure 3b).
For the wet sedge and tussock tundra B soil leachates, light exposure had no significant effect on the
13
C isotopic
composition of the CO
2
from respiration compared to respiration of DOC kept in the dark (Figure 2b). For the
thermokarst leachate, the CO
2
from respiration in UV and visible light treatments was significantly enriched in
13
C compared to the CO
2
from respiration in the dark treatment. For all of the soil leachates that were exposed to
both UV and visible light, there was little to no wavelength‐dependent effect of light on the
13
C isotopic
composition of the CO
2
from respiration (Figure 2b).
4. Discussion
These first quantifications of the age of CO
2
produced by microbial respiration of permafrost DOC (Figure 2a)
show that microbes can respire ancient permafrost DOC to CO
2
upon export of this carbon to sunlit surface
waters, as expected from prior work (Cory et al., 2013). This result is consistent with prior studies that
quantified the age of permafrost DOC respired in the dark based on changes in the bulk DOC age (Mann
et al., 2015; Spencer et al., 2015). Together, these results and prior work demonstrate that permafrost DOC is
labile to microbial respiration in arctic surface waters, whether it is freshly thawed from dark soils (this study;
Mann et al., 2015; Melchert et al., 2022; Spencer et al., 2015) or has been exposed to UV and visible sunlight
(this study).
While the permafrost DOC respired by microbes was always old (>4,000 a BP), microbial respiration consistently
produced younger CO
2
from light‐exposed DOC than from dark controls. There are several potential explanations
for this result. First, sunlight might completely oxidize relatively older DOC, leaving behind relatively younger
DOC for microbes to respire. Studies exposing DOC from marine and lower‐latitude freshwaters to high doses of
UV light have found that younger DOC is preferentially oxidized to CO
2
by sunlight (Beaupré & Druffel, 2012;
Beaupré et al., 2007; Ishikawa et al., 2019). However, for permafrost DOC exposed to environmentally‐relevant,
lower doses of UV and visible light, the age of the DOC completely oxidized to CO
2
by light has been shown to be
similar to the age of the bulk permafrost DOC (≤70‰, or ≤930 years, different from the bulk DOC; Bowen
et al., 2020b). This result, combined with the small amount of DOC completely oxidized to CO
2
by sunlight prior
Figure 3. δ
13
C‐CO
2
produced from microbial respiration of DOC kept in the dark (circle symbols), exposed to ultraviolet
light (UV; 305 nm, diamond symbols), and exposed to visible light (405 nm, square symbols) increased with (a) increasing
respiratory quotient and (b) decreasing age of the CO
2
from: Imnavait wet sedge tundra (orange symbols), Toolik tussock
tundra A (dark blue symbols), Toolik tussock tundra B (yellow symbols), and LTER 395 thermokarst (teal symbols). In panel
(a), data were fit using a least‐squares regression where R
2
=0.94 and p<0.01. All values on the x‐axis are shown as the
average ±1 SE (n=3). All values on the y‐axis are shown as the average ±1 SE (n=2). In panel (b), data were fit using a
least‐squares regression where R
2
=0.57 and p<0.05. All values on the x‐ and y‐axes are shown as the average ±1
SE (n=2).
Journal of Geophysical Research: Biogeosciences
10.1029/2023JG007853
RIEB ET AL. 9 of 19

to the incubation (Table S3 in Supporting Information S1), suggests that complete oxidation of DOC during light
exposure should not cause substantial changes in the average age of the bulk DOC available for microbes to
respire.
Second, light might cause microbes to respire younger DOC by altering the composition of the DOC remaining
after light exposure (Bowen et al., 2020a; Cory et al., 2010; Ward & Cory, 2016; Wetzel et al., 1995), either by
making younger DOC more labile or older DOC less labile compared to the same age fractions kept in the dark.
These changes in the lability of different DOC age fractions would represent an increase or decrease in the amount
of total labile DOC available for microbes to respire and might correspond to light treatments in this study that
increased or decreased the magnitude of microbial respiration, respectively. Consistent with the former, prior
work demonstrated that a major effect of sunlight exposure of permafrost DOC was to produce DOC labile to
microbial respiration (Nalven et al., 2020; Ward et al., 2017). Results from this study suggest that the labile DOC
produced by sunlight is relatively younger than the DOC respired by microbes in the dark.
These results confirming microbial respiration of old permafrost DOC both in the dark and after light exposure
should apply to permafrost‐thaw impacted waters across the Arctic for the following reasons. First, the
composition of the DOC from permafrost soils in this study, as given by SUVA
254
and fluorescence index, is
within the range previously reported for permafrost organic carbon from other arctic sites (Bowen et al., 2020b;
Cory et al., 2013; Mann et al., 2015; Spencer et al., 2015; Stubbins et al., 2017). Second, the
13
C and
14
C of the
permafrost DOC is within the range for soil organic carbon in general (Hoefs, 2015; Trumbore & Druffel, 1995)
and specifically for permafrost DOC (Bowen et al., 2020b; Mann et al., 2015; Rogers et al., 2021; Spencer
et al., 2015; Stubbins et al., 2017; Vonk et al., 2013). Third, the respiratory quotients for permafrost DOC are
within the range previously reported (Berggren et al., 2012; Cory et al., 2014), suggesting that microbes in this
study respired terrestrially‐derived DOC by similar respiratory pathways as in other freshwaters.
While the age and composition of the bulk, permafrost DOC is consistent with prior work, the age and
composition of the DOC respired both in the dark and after light exposure differs from prior work. For example,
prior studies showed that microbes respire fractions of permafrost DOC that are similar in both age and
13
C
composition to the bulk DOC (δ
13
C of respired DOC of 26 to 23‰; Mann et al., 2015; Melchert et al., 2022).
In contrast, results from this study show that the CO
2
from respiration is often older and substantially more
13
C‐
depleted (44.9‰ to 28.1‰) than the bulk DOC.
Differences in the incubation conditions and amount of DOC consumed might explain why respiration of rela-
tively old and
13
C‐depleted permafrost DOC fractions was observed here but not in prior studies. In prior work
performed at 20°C, between 35% and 50% of permafrost DOC was respired during incubations (Mann et al., 2015;
Rogers et al., 2021; Spencer et al., 2015; Vonk et al., 2013). In contrast, in this study, incubations were performed
at 10°C to be representative of temperatures in arctic surface waters (Adams et al., 2010; Cory et al., 2014;
Docherty et al., 2019), and respiration consumed only 1%–5% of the DOC. The lower respiration rates produced
by the relatively cold incubation temperature in this study are likely more representative of respiration rates in
arctic surface waters (O’Donnell et al., 2016), in comparison to incubations at higher temperatures. As the amount
of DOC consumed increases, the isotopic signature of the DOC consumed becomes closer to the isotopic
signature of the bulk DOC. Conversely, when only small amounts of the total DOC are consumed, there might be
selective respiration of isotopically lighter fractions of the DOC pool. Thus, the C isotopic composition of CO
2
from respiration in this study might reflect the C isotopic composition of the most labile fractions of the DOC
pool. However, the age and
13
C composition of the CO
2
from respiration is insufficient on its own to interpret the
DOC substrates most labile to microbes. Thus, here we interpret the C isotopic composition of CO
2
from
respiration alongside other proxies for substrate composition from this study and the literature.
4.1. Microbes Respire
13
C‐Depleted Fractions of Permafrost DOC to CO
2
In this study, the δ
13
C of CO
2
(Figure 2b) likely reflects the selective respiration of a range of
13
C‐depleted
fractions within the bulk DOC from each soil site. For example, microbial respiration of wet sedge and tus-
sock tundra A DOC produced CO
2
that was highly
13
C‐depleted relative to the expected range of δ
13
C for bulk
soil organic carbon (Drake et al., 2015; Hoefs, 2015; Mann et al., 2015; Melchert et al., 2022; Trumbore &
Druffel, 1995). Microbial respiration of predominantly lipid or lignin‐like fractions of the DOC pool could
produce
13
C‐depleted CO
2
because these compound classes are more depleted in
13
C compared to other fractions
of DOC (δ
13
C ranges of 27 to 38‰ for lipids and 25 to 32‰ for lignin‐like DOC; Benner et al., 1987;
Journal of Geophysical Research: Biogeosciences
10.1029/2023JG007853
RIEB ET AL. 10 of 19

Bertoldi et al., 2014; Kling & Fry, 1992; McCallister & del Giorgio, 2008). From other soil sites (tussock tundra B
and thermokarst), microbial respiration produced CO
2
that, while still
13
C‐depleted relative to the DOC, was
much closer to the
13
C composition of the bulk DOC. At these sites, microbes might have respired a range of the
fractions of the DOC pool, including relatively
13
C‐depleted lipid and lignin‐like compounds as well as relatively
13
C‐enriched carbohydrates and organic acids (Bowling et al., 2008; Whelan et al., 1970). In addition to pref-
erential microbial respiration of
13
C‐depleted sources within the DOC pool, other processes might have
contributed to the respiration of a range of
13
C‐depleted CO
2
(Section S2.2 in Supporting Information S1).
The respiratory quotient of CO
2
production to O
2
consumption during respiration provides further evidence in
support of microbial respiration of a range of fractions of DOC (Figure 3a). The range of respiratory quotients for
respiration of DOC in the literature (∼0.5–4; Allesson et al., 2016; Berggren et al., 2012; Cory et al., 2014; del
Giorgio et al., 2006) is hypothesized to depend on the composition of the DOC compounds respired (Masiello
et al., 2008). Relatively more reduced (and relatively
13
C‐depleted) lipid and lignin‐like compounds are predicted
to be respired with a lower respiratory quotient than are more oxidized (and relatively
13
C‐enriched) carbohy-
drates and organic acids within the DOC pool (Dilly, 2001; Masiello et al., 2008; Romero‐Kutzner et al., 2015).
Assuming the hypothesis above explains variability in the respiratory quotient for DOC, then the strong, positive
correlation between the respiratory quotient and the
13
C of the CO
2
from respiration (Figure 3a) is consistent with
the expected
13
C composition of different fractions of the DOC pool.
Results in this study suggesting that old,
13
C‐depleted DOC can constitute the majority of the DOC respired by
microbes from the wet sedge and tussock tundra A permafrost soil sites are consistent with prior work. Rogers
et al. (2021) provided evidence that the oldest, most
13
C‐depleted fractions of DOC from Yedoma permafrost soil
separate out at the lowest energies by ramped‐pyrolysis oxidation and thus might be the most labile to microbial
respiration (Leifeld & von Lützow, 2014), as observed for the wet sedge and tussock tundra A permafrost soils in
this study (Figure 3b).
There are also several lines of evidence from prior work for respiration of lignin‐like permafrost DOC, which is
proposed to contribute to the highly
13
C‐depleted signature of the most labile permafrost DOC in this study. For
example, lignin‐like formulas constituted a substantial percentage of the DOC respired by microbes from
permafrost soils in previous studies (Nalven et al., 2020; Rogers et al., 2021; Ward et al., 2017). Additionally,
microbial communities associated with permafrost soils across the Arctic have been shown to have a relatively
high abundance of genes related to the metabolism of a wide range of organic carbon fractions including aromatic
DOC associated with lignin (Nalven et al., 2020; Waldrop et al., 2023). These findings suggest that permafrost
microbial communities are equipped to respire the abundant lignin‐like fractions of the DOC pool (Ward
et al., 2017).
Unlike lignin‐like DOC, lipid fractions have not been shown to be consumed by microbes from permafrost DOC.
However, lipid fractions of soil organic matter are relatively labile to microbial respiration (e.g., Reynolds
et al., 2018). The lack of evidence for respiration of lipid fractions of permafrost DOC could be due to limitations
of the electrospray ionization (ESI) FT‐ICR MS analysis that has frequently been used to characterize the
composition of DOC (Rogers et al., 2021; Spencer et al., 2015; Ward et al., 2017). Lipid fractions of DOC are
known to ionize poorly via ESI, making it difficult to assess changes in the lipid content of permafrost DOC
(Hockaday et al., 2009).
4.2. Light Exposure Had Wavelength‐Dependent Effects on the Magnitude of DOC Respired
Given that the most common effect of light treatment was to increase respiration relative to respiration in the dark
(Figure 1, Figure S3 and Table S4 in Supporting Information S1), results from this study are generally consistent
with the prior conclusion that exposure of permafrost DOC to broadband sunlight stimulates microbial respiration
(Cory et al., 2013; Nalven et al., 2020; Ward et al., 2017). However, results from this study show that different
wavelengths of light have a broader range of effects on respiration than previously observed, ranging from
increasing to decreasing respiration relative to respiration of DOC kept in the dark. These effects are likely due to
changes in the DOC pool size and the DOC chemical composition.
Light exposure of DOC has been proposed to increase or decrease respiration by increasing or decreasing the
amount of DOC labile to microbes, respectively (Bowen et al., 2020a; Ward et al., 2017; Wetzel et al., 1995).
Light exposure of DOC can change the amount of biologically labile DOC by completely oxidizing some of this
Journal of Geophysical Research: Biogeosciences
10.1029/2023JG007853
RIEB ET AL. 11 of 19

DOC to CO
2
(reducing the total amount of DOC labile to respiration), and by
concurrently changing the chemical composition of the remaining DOC to
make it more or less labile to microbes (Bowen et al., 2020a; Cory et al., 2010;
Ward & Cory, 2016; Wetzel et al., 1995; Xie et al., 2004). Therefore, changes
in the magnitude of respiration after DOC light exposure reflect the net effect
of light exposure on the size of the labile DOC pool.
The net effect of light exposure on the amount of labile DOC—and, thus, on
whether respiration increases or decreases after light exposure relative to
respiration of dark DOC—likely depends on the extent to which DOC was
oxidized during light exposure in this study. For the two soil sites with less
than 5% of the DOC pool oxidized to CO
2
by UV or visible light (tussock
tundra B and thermokarst; Figure 4), both UV and visible light had either no
net effect or increased respiration relative to respiration in the dark. For the
two soil sites where 5% or more of the DOC pool was oxidized to CO
2
by UV
or visible light (wet sedge and tussock tundra A), UV and visible light either
increased or decreased respiration relative to respiration in the dark
(Figure 4). These results are consistent with prior work demonstrating that
increasing oxidation of DOC by broadband sunlight generally increases
respiration when a relatively small amount (<5%) of the DOC is oxidized to
CO
2
(Bowen et al., 2020a; Reader & Miller, 2014; Ward et al., 2017) but
might have diminishing returns or even decrease respiration relative to
respiration in the dark at greater amounts of DOC oxidized (Bowen
et al., 2020a; Reader & Miller, 2014).
For DOC from a given soil site, UV and visible light doses were chosen to
achieve the same extent of DOC oxidation by light of both wavelengths
(Table S3 in Supporting Information S1). Thus, the differences in the microbial response to UV versus visible
light‐exposed DOC for a given soil site should be due to wavelength‐dependent differences in the production and
removal of labile DOC. UV wavelengths of light have been proposed to be more efficient at yielding low mo-
lecular weight acids and aldehydes than are visible wavelengths of light (Miller et al., 2002; Wetzel et al., 1995).
In turn, low molecular weight acid and aldehyde photo‐products are thought to be highly biologically labile
(Drake et al., 2015; Miller et al., 2002) and might account for increases in respiration after light exposure of DOC.
For the wet sedge and thermokarst soil sites, for which UV light exposure of DOC increased respiration (Figure 1,
Table S4 in Supporting Information S1), the relatively high respiratory quotients for DOC exposed to UV light
(≥1) are consistent with respiration of organic acids (Table S4 in Supporting Information S1; Masiello
et al., 2008). For the UV light treatment of the thermokarst site, production of CO
2
with a
13
C composition similar
to the bulk DOC provides further evidence in support of respiration of organic acids, which are expected to have a
similar
13
C composition to that of the bulk DOC (Figure 2b, Table 1; Bowling et al., 2008). These results suggest
that UV light was relatively more efficient at producing labile DOC than was visible light for DOC from the wet
sedge and thermokarst soils.
However, higher‐energy UV light might also be relatively more efficient at removing biologically labile DOC
than is lower‐energy visible light (Reader & Miller, 2014). When this is the case, net production of labile DOC
might be higher under visible light than UV light. For example, for the tussock tundra A soil site, UV light
decreased respiration while visible light increased respiration relative to respiration in the dark (Figure 1, Figure
S3 and Table S4 in Supporting Information S1). For this soil site, the relatively high respiratory quotient for DOC
exposed to visible light (RQ of 0.96; Figure S3 and Table S4 in Supporting Information S1), compared to the
relatively lower respiratory quotient for DOC from the same site exposed to UV light, is consistent with respi-
ration of labile organic acids (Masiello et al., 2008) following exposure of DOC to visible light.
However, there are conflicting conclusions in the literature about whether changes in DOC composition during
light exposure could affect the respiratory quotient. Some studies report increases in the respiratory quotient after
light exposure, interpreted as respiration of more highly oxidized, low molecular weight DOC photoproducts
from light‐exposed DOC than from dark DOC (Allesson et al., 2016; Berggren et al., 2012). In contrast, other
studies provide evidence that microbes respire the same types of DOC after light exposure that they are already
Figure 4. The difference between CO
2
production from microbial respiration
in light and dark DOC treatments versus the percentage of DOC that was
oxidized by light prior to biological incubations for DOC exposed to
ultraviolet light (UV; 305 nm, diamond symbols) and visible light (405 nm,
square symbols). Values for the difference between CO
2
production from
respiration in light and dark DOC treatments are shown as the average ±1 SE
of experimental replicates (n=3).
Journal of Geophysical Research: Biogeosciences
10.1029/2023JG007853
RIEB ET AL. 12 of 19

genomically‐equipped to respire in the dark (Cory et al., 2014; Nalven et al., 2020; Ward et al., 2017). Results of
this study provide support for the latter interpretation by demonstrating that microbes respired fractions of the
DOC pool with similar
13
C compositions and respiratory quotients both after light exposure and in the dark for
DOC from most soils and wavelengths of light (Figure 2b, Figure S3 and Table S4 in Supporting Information S1).
Given that both the δ
13
C of CO
2
from respiration and the respiratory quotient are thought to depend on the
composition of the DOC (Berggren et al., 2012; Crow et al., 2006; Masiello et al., 2008; Pries et al., 2020), these
results suggest that light exposure of DOC did not substantially change the fractions of DOC respired or their
respiratory pathways compared to respiratory pathways in the dark for most soil sites. Thus, changes in the
magnitude of respiration after light exposure are most likely a response to changes in the size of the labile DOC
pool, rather than to major shifts in the composition of the most labile DOC available to microbes.
4.3. Rates of Microbial Respiration of Permafrost DOC in Arctic Surface Waters
Water‐column production rates for any product of DOC photodegradation are sensitive to the wavelength‐
dependent yield of the product per mole of light absorbed by chromophoric dissolved organic matter (CDOM,
the light‐absorbing fraction of DOC). In the case of respiration increased or decreased by DOC light exposure, the
product is more or less CO
2
from respiration upon absorption of light by CDOM, respectively, compared to
respiration in the dark. In prior studies where exposure of DOC to broadband (natural) sunlight was found to
increase respiration of the DOC relative to dark controls (Cory et al., 2013,2014), the assumptions were that (a)
all wavelengths of sunlight increased respiration of DOC, and (b) exposure of DOC to UV wavelengths of
sunlight was more efficient at increasing respiration than exposure to visible wavelengths of sunlight (e.g., Cory
et al., 2013,2014). In contrast, this study shows that both UV and visible wavelengths of light can also decrease
respiration relative to respiration in the dark. Additionally, this study shows that exposure to UV light can result in
similar or lower amounts of CO
2
from respiration than does exposure to visible light for some permafrost DOC.
Thus, water‐column rates of respiration of sunlight‐exposed DOC (e.g., Cory et al., 2014) must be revised in
response to the invalidated assumptions.
A quantitative revision of water‐column rates of respiration of sunlight‐exposed DOC requires measurements of
the wavelength‐dependent yield of more or less CO
2
from respiration per mole of light absorbed by CDOM,
which was not possible based on the experimental design in this study (see Section S1.3 in Supporting Infor-
mation S1). However, results from this study enable a qualitative assessment of the net effect of sunlight exposure
of DOC on water‐column respiration rates. An important factor modulating the wavelength‐dependent effect of
DOC light exposure on water column respiration rates is the amount of UV versus visible light absorbed by DOC
in the water column. Although absorbance of light by DOC is much higher in the UV than in the visible, more total
visible than UV light reaches the surface, and thus more visible than UV light is absorbed by DOC in the water
column (Figure S5 in Supporting Information S1). Thus, relatively small yields of more or less CO
2
from
respiration of DOC exposed to visible light can have a disproportionately large impact on water‐column rates of
CO
2
production compared to the yields of CO
2
from respiration of DOC exposed to UV light.
For example, for the tussock tundra A DOC, the extra CO
2
from respiration of tussock tundra DOC exposed to
visible light might completely or partially offset the reduction in CO
2
from respiration of tussock tundra DOC
exposed to UV wavelengths of light, relative to respiration in the dark. Conversely, for wet sedge DOC, the
reduction in CO
2
from respiration of wet sedge DOC exposed to visible wavelengths of light might completely or
partially offset the extra CO
2
from respiration of wet sedge DOC exposed to UV wavelengths of light.
However, the experimental light doses in this study were chosen to achieve equal oxidation of DOC by both UV
and visible wavelengths of light for each soil site (Table S3 in Supporting Information S1) and are not repre-
sentative of the ratio of UV to visible photon doses reaching arctic surface waters. To oxidize the same amount of
DOC with visible as with UV light, the ratio of visible to UV light received by DOC from LEDs in this study
(∼56:1; Table S3 in Supporting Information S1) was ∼5 times greater than the ratio of visible to UV light received
by DOC from natural sunlight in arctic surface waters (∼11:1; Bowen et al., 2020b). This difference in the ratio of
visible to UV between this study and broadband sunlight in prior work might explain why prior work has reported
that sunlight generally increases microbial respiration relative to respiration in the dark (Cory et al., 2013,2014;
Ward et al., 2017). For example, consider the DOC from the wet sedge soil site, for which this study showed that
UV light exposure increases respiration and visible light exposure decreases respiration relative to respiration in
the dark (Table S4 in Supporting Information S1). In arctic surface waters, wet sedge DOC would receive ∼5
Journal of Geophysical Research: Biogeosciences
10.1029/2023JG007853
RIEB ET AL. 13 of 19

times less visible light than in this study over the period of time required to receive the same amount of UV light as
in this study. As discussed previously, small doses of light frequently increase microbial respiration relative to
respiration in the dark, while larger doses of light can decrease microbial respiration because labile DOC is
increasingly altered or removed during photo‐oxidation. Therefore, while this study showed that relatively large
doses of visible light decrease respiration of wet sedge DOC, it is possible that the lower doses of visible light
from natural sunlight would have a weaker effect.
These results demonstrate two knowledge gaps to be addressed to quantify water‐column rates of CO
2
production
from respiration of DOC in sunlit surface waters. First, the wavelength dependence of the yield of more or less
CO
2
from respiration of sunlight‐exposed DOC must be quantified. Second, the wavelength‐dependent yield of
any product of DOC photodegradation (including CO
2
from respiration) should be quantified using doses of UV
and visible light similar to what is absorbed by DOC during its residence times in sunlit surface waters.
5. Conclusions
By measuring the age of CO
2
produced by respiration of light‐exposed permafrost DOC, we show that ancient
permafrost DOC is respired upon exposure to sunlight in arctic surface waters. The CO
2
produced by microbes in
light treatments and in dark controls was as old or older than the bulk DOC. It is already established that the
coupled oxidation by sunlight and microbial respiration of DOC is an important control on the oxidation of
modern DOC to CO
2
in arctic surface waters (Cory et al., 2013,2014; Stubbins et al., 2017; Ward et al., 2017). As
the annual thaw depth increases, or as thermokarst failures expose permafrost soils to surface conditions, the
modern DOC exported from land to water will be increasingly mixed with old organic compounds ranging in age
from a few thousands of years up to ∼45,000 years before present (Schwab et al., 2020; Strauss et al., 2017).
Therefore, sunlight exposure will continue to be an important control on the magnitude of CO
2
emitted from arctic
freshwaters as the DOC in these waters is increasingly comprised of ancient permafrost carbon in a warming
Arctic.
Additionally, this study showed that the permafrost DOC respired by microbes both in the dark and after light
exposure was substantially
13
C‐depleted relative to the bulk DOC. This result is interpreted as evidence that
13
C‐
depleted fractions of the permafrost DOC pool, including a mixture of lipid and lignin‐like DOC, are more
biologically labile than
13
C‐enriched DOC fractions, both when freshly thawed from dark soil and after exposure
to sunlight. Microbial respiration of permafrost DOC in small headwater streams that are most strongly impacted
by permafrost thaw (Aiken et al., 2014; Neff et al., 2006; Vonk et al., 2013) is likely fueled primarily by these
most labile,
13
C‐depleted fractions of the permafrost DOC pool. In these waters, labile permafrost DOC that is
consumed by coupled oxidation by sunlight and microbial respiration is expected to be rapidly replenished by
fresh DOC inputs from soils (Cory et al., 2015; Neilson et al., 2018). However, as water residence times increase,
the DOC remaining that is exported downstream might be composed of a greater percentage of
13
C‐enriched
fractions than DOC received from permafrost soils.
Results of this study and prior work (Bowen et al., 2020a; Cory et al., 2013; Ward et al., 2017) suggest that the
magnitude of permafrost DOC respired in sunlit arctic surface waters depends in part on the extent to which DOC
oxidation by sunlight changes the amount of labile DOC. Sunlight exposure is expected to consistently increase
respiration of permafrost DOC (relative to respiration in the dark) in arctic headwater streams, where short
residence times of days to weeks mean that a small percentage of permafrost DOC will be oxidized by sunlight
prior to export downstream (Cory et al., 2015). As permafrost DOC moves from headwater streams to larger
streams, both the water residence time and stream surface area increase (e.g., Neilson et al., 2018). With
increasing time spent in sunlit waters, the percentage of permafrost DOC oxidized by both sunlight and microbes
in larger streams is generally expected to increase (Cory et al., 2015). Greater oxidation of DOC by sunlight in
larger streams might increasingly remove DOC labile to microbes, offsetting some of the increases in oxidation of
DOC driven by the longer water residence time. However, the reported lack of old permafrost DOC in major
arctic rivers (Aiken et al., 2014; Neff et al., 2006; Rogers et al., 2021; Vonk et al., 2013; Wild et al., 2019)
suggests that more of the permafrost DOC reaching surface waters is oxidized to CO
2
than is exported down-
stream to larger rivers. Results from this study implicate increased respiration of sunlight‐exposed DOC as a
contributor to this loss of permafrost DOC in the smaller, shallow, unshaded surface waters of the Arctic (Cory
et al., 2014) that are increasingly ice‐free during peak solar radiation (Cooley et al., 2019; Zhang et al., 2021).
Journal of Geophysical Research: Biogeosciences
10.1029/2023JG007853
RIEB ET AL. 14 of 19

Conflict of Interest
The authors declare no conflicts of interest relevant to this study.
Data Availability Statement
All data are available in the manuscript or the Supporting Information S1, and data are available at the
Arctic LTER (https://arc‐lter.ecosystems.mbl.edu/arctic‐data‐catalog) (Cory et al., 2023a,2023b,2023c,
2023d,2023e).
References
Abbott, B. W., Larouche, J. R., Jones, J. B., Bowden, W. B., & Balser, A. W. (2014). Elevated dissolved organic carbon biodegradability from
thawing and collapsing permafrost. Journal of Geophysical Research: Biogeosciences,119(10), 2049–2063. https://doi.org/10.1002/
2014JG002678
Adams, H. E., Crump, B. C., & Kling, G. W. (2010). Temperature controls on aquatic bacterial production and community dynamics in arctic
lakes and streams. Environmental Microbiology,12(5), 1319–1333. https://doi.org/10.1111/j.1462‐2920.2010.02176.x
Aiken, G. R., Spencer, R. G. M., Striegl, R. G., Schuster, P. F., & Raymond, P. A. (2014). Influences of glacier melt and permafrost thaw on the
age of dissolved organic carbon in the Yukon River basin. Global Biogeochemical Cycles,28(5), 525–537. https://doi.org/10.1002/
2013GB004764
Allesson, L., Ström, L., & Berggren, M. (2016). Impact of photochemical processing of DOC on the bacterioplankton respiratory quotient in
aquatic ecosystems. Geophysical Research Letters,43(14), 7538–7545. https://doi.org/10.1002/2016GL069621
Andrews, S. S., Caron, S., & Zafiriou, O. C. (2000). Photochemical oxygen consumption in marine waters: A major sink for colored dissolved
organic matter? Limnology & Oceanography,45(2), 267–277. https://doi.org/10.4319/lo.2000.45.2.0267
Beaupré, S. R., & Druffel, E. R. M. (2012). Photochemical reactivity of ancient marine dissolved organic carbon. Geophysical Research Letters,
39(18), L18602. https://doi.org/10.1029/2012GL052974
Beaupré, S. R., Druffel, E. R. M., & Griffin, S. (2007). A low‐blank photochemical extraction system for concentration and isotopic analyses of
marine dissolved organic carbon. Limnology and Oceanography: Methods,5(6), 174–184. https://doi.org/10.4319/lom.2007.5.174
Benner, R., Fogel, M. L., Sprague, E. K., & Hodson, R. E. (1987). Depletion of
13
C in lignin and its implications for stable carbon isotope studies.
Nature,329(6141), 708–710. https://doi.org/10.1038/329708a0
Berggren, M., Lapierre, J. F., & del Giorgio, P. A. (2012). Magnitude and regulation of bacterioplankton respiratory quotient across freshwater
environmental gradients. The ISME Journal,6(5), 984–993. https://doi.org/10.1038/ismej.2011.157
Bertoldi, D., Santato, A., Paolini, M., Barbero, A., Camin, F., Nicolini, G., & Larcher, R. (2014). Botanical traceability of commercial tannins
using the mineral profile and stable isotopes. Journal of Mass Spectrometry,49(9), 792–801. https://doi.org/10.1002/jms.3457
Bowen, J. C., Kaplan, L. A., & Cory, R. M. (2020a). Photodegradation disproportionately impacts biodegradation of semi‐labile DOM in streams.
Limnology & Oceanography,65(1), 13–26. https://doi.org/10.1002/lno.11244
Bowen, J. C., Ward, C. P., Kling, G. W., & Cory, R. M. (2020b). Arctic amplification of global warming strengthened by sunlight oxidation of
permafrost carbon to CO
2
.Geophysical Research Letters,47(12), 0–3. https://doi.org/10.1029/2020GL087085
Bowling, D. R., Pataki, D. E., & Randerson, J. T. (2008). Carbon isotopes in terrestrial ecosystem pools and CO
2
fluxes. New Phytologist,178(1),
24–40. https://doi.org/10.1111/j.1469‐8137.2007.02342.x
Cooley, S. W., Smith, L. C., Ryan, J. C., Pitcher, L. H., & Pavelsky, T. M. (2019). Arctic‐Boreal lake dynamics revealed using CubeSat imagery.
Geophysical Research Letters,46(4), 2111–2120. https://doi.org/10.1029/2018GL081584
Cory, R. M., Crump, B. C., Dobkowski, J. A., & Kling, G. W. (2013). Surface exposure to sunlight stimulates CO
2
release from permafrost soil
carbon in the Arctic. Proceedings of the National Academy of Sciences of the United States of America,110(9), 3429–3434. https://doi.org/10.
1073/pnas.1214104110
Cory, R. M., Harrold, K. H., Neilson, B. T., & Kling, G. W. (2015). Controls on dissolved organic matter (DOM) degradation in a headwater
stream: The influence of photochemical and hydrological conditions in determining light‐limitation or substrate‐limitation of photo‐
degradation. Biogeosciences,12(22), 6669–6685. https://doi.org/10.5194/bg‐12‐6669‐2015
Cory, R. M., & Kaplan, L. A. (2012). Biological lability of streamwater fluorescent dissolved organic matter. Limnology & Oceanography,57(5),
1347–1360. https://doi.org/10.4319/lo.2012.57.5.1347
Cory, R. M., & Kling, G. W. (2018). Interactions between sunlight and microorganisms influence dissolved organic matter degradation along the
aquatic continuum. Limnology and Oceanography Letters,3(3), 102–116. https://doi.org/10.1002/lol2.10060
Cory, R. M., McNeill, K., Cotner, J. P., Amado, A., Purcell, J. M., & Marshall, A. G. (2010). Singlet oxygen in the coupled photochemical and
biochemical oxidation of dissolved organic matter. Environmental Science & Technology,44(10), 3683–3689. https://doi.org/10.1021/
es902989y
Cory, R. M., Rieb, E. C., Polik, C. A., Ward, C. P., & Kling, G. W. (2023a). Accession numbers for radiocarbon and stable carbon isotopes of
dissolved organic carbon (DOC) and dissolved inorganic carbon (DIC) in soil leachates from permafrost soils collected from the North Slope of
Alaska in the summers of 2018 and 2022 (Version 1) [Dataset]. Environmental Data Initiative.https://doi.org/10.6073/pasta /
f7a565d6b8b8ca1b5fa2d8f4d28dee4b
Cory, R. M., Rieb, E. C., Polik, C. A., Ward, C. P., & Kling, G. W. (2023b). Daily water‐column rates of sunlight absorption by chromophoric
dissolved organic matter (CDOM) leached from permafrost soils collected from the North Slope of Alaska in the summers of 2018 and 2022
(Version 1) [Dataset]. Environmental Data Initiative.https://doi.org/10.6073/pasta/3a34daab0f8bb4e59ef39068f311fa94
Cory, R. M., Rieb, E. C., Polik, C. A., Ward, C. P., & Kling, G. W. (2023c). Methane concentrations in dissolved organic carbon (DOC) leachates
from permafrost soils collected from the North Slope of Alaska in the summers of 2018 and 2019 (Version 1) [Dataset]. Environmental Data
Initiative.https://doi.org/10.6073/pasta/e386e272d73577e42b2ae1a18fecf6a0
Cory, R. M., Rieb, E. C., Polik, C. A., Ward, C. P., & Kling, G. W. (2023d). Preparation of dissolved organic carbon (DOC) leachates from
permafrost soils collected from the North Slope of Alaska in the summers of 2018 and 2022 (Version 1) [Dataset]. Environmental Data
Initiative.https://doi.org/10.6073/pasta/1bf2a9fcbd47f8af1c789cabe02322d6
Acknowledgments
We thank K. Clippinger, N. LaFramboise,
J. Bowen, J. Dobkowski, K. Romanowicz,
N. Christman, B. Crump, E. Peterson, and
researchers, technicians, and support staff
of the Arctic LTER and Toolik Field
Station for assistance with field and
laboratory work. We additionally thank K.
Romanowicz for providing methane
concentrations for permafrost soil
leachates. We thank J. Burton, L. Xu, and
researchers and technicians at the National
Ocean Sciences AMS (NOSAMS) facility
at WHOI for training and assistance with
carbon isotope measurements. Research
was supported by NSF DDRIG 2228992
(R.M.C. and E.C.R.), NSF DEB 1754835
and 2224743 (G.W.K. and R.M.C.), NSF
DEB 1637459 and OPP 1936769 (G.W.K),
and NSF OCE 2219660 (C.P.W.). All
authors contributed to the study design,
sample collection, laboratory experiments,
data analysis, and manuscript preparation.
Journal of Geophysical Research: Biogeosciences
10.1029/2023JG007853
RIEB ET AL. 15 of 19

Cory, R. M., Rieb, E. C., Polik, C. A., Ward, C. P., & Kling, G. W. (2023e). Radiocarbon and stable carbon isotopes of carbon dioxide produced by
respiration of dissolved organic carbon (DOC) leached from permafrost soils collected from the North Slope of Alaska in the summers of 2018
and 2022 (Version 1) [Dataset]. Environmental Data Initiative.https://doi.org/10.6073/pasta/50e5f1dbff90130bd40658e9f00a14d3
Cory, R. M., Ward, C. P., Crump, B. C., & Kling, G. W. (2014). Sunlight controls water column processing of carbon in arctic fresh waters.
Science,345(6199), 925–928. https://doi.org/10.1126/science.1253119
Crow, S. E., Sulzman, E. W., Rugh, W. D., Bowden, R. D., & Lajtha, K. (2006). Isotopic analysis of respired CO
2
during decomposition of
separated soil organic matter pools. Soil Biology and Biochemistry,38(11), 3279–3291. https://doi.org/10.1016/j.soilbio.2006.04.007
Del Giorgio, P. A., Pace, M. L., & Fischer, D. (2006). Relationship of bacterial growth efficiency to spatial variation in bacterial activity in the
Hudson River. Aquatic Microbial Ecology,45(1), 55–67. https://doi.org/10.3354/ame045055
Dilly, O. (2001). Microbial respiratory quotient during basal metabolism and after glucose amendment in soils and litter. Soil Biology and
Biochemistry,33(1), 117–127. https://doi.org/10.1016/S0038‐0717(00)00123‐1
Docherty, C. L., Dugdale, S. J., Milner, A. M., Abermann, J., Lund, M., & Hannah, D. M. (2019). Arctic river temperature dynamics in a changing
climate. River Research and Applications,35(8), 1212–1227. https://doi.org/10.1002/rra.3537
Drake, T. W., Wickland, K. P., Spencer, R. G. M., & Striegl, R. G. (2015). Ancient low‐molecular‐weight organic acids in permafrost fuel rapid
carbon dioxide production upon thaw. Proceedings of the National Academy of Sciences of the United States of America,112(45), 13946–-
13951. https://doi.org/10.1073/pnas.1511705112
Frey, K. E., & McClelland, J. W. (2009). Impacts of permafrost degradation on arctic river biogeochemistry. Hydrological Processes,23(1),
169–182. https://doi.org/10.1002/hyp.7196
Hamilton, T. D. (2003). Glacial geology of the Toolik Lake and Upper Kuparuk River Regions. In Biological papers of the University of Alaska
(pp. 1–24). University of Alaska.
Hockaday, W. C., Purcell, J. M., Marshall, A. G., Baldock, J. A., & Hatcher, P. G. (2009). Electrospray and photoionization mass spectrometry for
the characterization of organic matter in natural waters: A qualitative assessment. Limnology and Oceanography: Methods,7(1), 81–95. https://
doi.org/10.4319/lom.2009.7.81
Hoefs, J. (2015). Stable isotope geochemistry (6th ed.). Springer‐Verlag. https://doi.org/10.1007/978‐3‐540‐70708‐0
Ishikawa, N. F., Butman, D., & Raymond, P. A. (2019). Radiocarbon age of different photoreactive fractions of freshwater dissolved organic
matter. Organic Geochemistry,135, 11–15. https://doi.org/10.1016/j.orggeochem.2019.06.006
Jorgenson, M. T., Shur, Y. L., & Pullman, E. R. (2006). Abrupt increase in permafrost degradation in Arctic Alaska. Geophysical Research
Letters,33(2), L02503. https://doi.org/10.1029/2005GL024960
Kling, G. W., Fry, B., & O'Brien, W. J. (1992). Stable isotopes and planktonic trophic structure in Arctic Lakes. Ecology,73(2), 561–566. https://
doi.org/10.2307/1940762
Kling, G. W., Kipphut, G. W., Miller, M. M., & O’Brien, W. J. (2000). Integration of lakes and streams in a landscape perspective: The importance
of material processing on spatial patterns and temporal coherence. Freshwater Biology,43(4), 477–497. https://doi.org/10.1046/j.1365‐2427.
2000.00515.x
Kokelj, S. V., & Jorgenson, M. T. (2013). Advances in Thermokarst Research. Permafrost and Periglacial Processes,24(2), 108–119. https://doi.
org/10.1002/ppp.1779
Leifeld, J., & von Lützow, M. (2014). Chemical and microbial activation energies of soil organic matter decomposition. Biology and Fertility of
Soils,50(1), 147–153. https://doi.org/10.1007/s00374‐013‐0822‐6
Li, A., Aubeneau, A. F., King, T., Cory, R. M., Neison, B. T., Bolster, D., & Packman, A. I. (2019). Effects of vertical hydrodynamic mixing on
photomineralization of dissolved organic carbon in arctic surface waters. Environmental Sciences: Processes & Impacts,21(4), 748–760.
https://doi.org/10.1039/C8EM00455B
Longworth, B. E., Von Reden, K. F., Long, P., & Roberts, M. L. (2015). A high output, large acceptance injector for the NOSAMS Tandetron
AMS system. Nuclear Instruments and Methods in Physics Research Section B: Beam Interactions with Materials and Atoms,361, 211–216.
https://doi.org/10.1016/j.nimb.2015.04.005
Mann, P. J., Eglinton, T. I., McIntyre, C. P., Zimov, N., Davydova, A., Vonk, J. E., et al. (2015). Utilization of ancient permafrost carbon in
headwaters of Arctic fluvial networks. Nature Communications,6(1), 7856. https://doi.org/10.1038/ncomms8856
Mann, P. J., Sobczak, W. V., Larue, M. M., Bulygina, E., Davydova, A., Vonk, J. E., et al. (2014). Evidence for key enzymatic controls on
metabolism of Arctic river organic matter. Global Change Biology,20(4), 1089–1100. https://doi.org/10.1111/gcb.12416
Masiello, C. A., Gallagher, M. E., Randerson, J. T., Deco, R. M., & Chadwick, O. A. (2008). Evaluating two experimental approaches for
measuring ecosystem carbon oxidation state and oxidative ratio. Journal of Geophysical Research,113(G3), G03010. https://doi.org/10.1029/
2007JG000534
McCallister, S. L., & del Giorgio, P. A. (2008). Direct measurement of the δ
13
C signature of carbon respired by bacteria in lakes: Linkages to
potential carbon sources, ecosystem baseline metabolism, and CO
2
fluxes. Limnology & Oceanography,53(4), 1204–1216. https://doi.org/10.
4319/lo.2008.53.4.1204
McCallister, S. L., & del Giorgio, P. A. (2012). Evidence for the respiration of ancient terrestrial organic C in northern temperate lakes and
streams. Proceedings of the National Academy of Sciences of the United States of America,109(42), 16963–16968. https://doi.org/10.1073/
pnas.1207305109
McFarlane, K. J., Throckmorton, H. M., Heikoop, J. M., Newman, B. D., Hedgpeth, A. L., Repasch, M. N., et al. (2022). Age and chemistry of
dissolved organic carbon reveal enhanced leaching of ancient labile carbon at the permafrost thaw zone. Biogeosciences,19(4), 1211–1223.
https://doi.org/10.5194/bg‐19‐1211‐2022
McGuire, A. D., Lawrence, D. M., Koven, C., Clein, J. S., Burke, E., Chen, G., et al. (2018). Dependence of the evolution of carbon dynamics in
the northern permafrost region on the trajectory of climate change. Proceedings of the National Academy of Sciences of the United States of
America,115(15), 3882–3887. https://doi.org/10.1073/pnas.1719903115
Melchert, J. O., Wischhöfer, P., Knoblauch, C., Eckhardt, T., Liebner, S., & Rethemeyer, J. (2022). Sources of CO
2
produced in freshly thawed
pleistocene‐age Yedoma permafrost. Frontiers in Earth Science,9.https://doi.org/10.3389/feart.2021.737237
Miller, W. L., Moran, M., Sheldon, W. M., Zepp, R. G., & Opsahl, S. (2002). Determination of apparent quantum yield spectra for the formation of
biologically labile photoproducts. Limnology & Oceanography,47(2), 343–352. https://doi.org/10.4319/lo.2002.47.2.0343
Nalven, S. G., Ward, C. P., Payet, J. P., Cory, R. M., Kling, G. W., Sharpton, T. J., et al. (2020). Experimental metatranscriptomics reveals the
costs and benefits of dissolved organic matter photo‐alteration for freshwater microbes. Environmental Microbiology,22(8), 3505–3521.
https://doi.org/10.1111/1462‐2920.15121
Neff, J. C., Finlay, J. C., Zimov, S. A., Davydov, S. P., Carrasco, J. J., Schuur, E. A. G., & Davydova, A. I. (2006). Seasonal changes in the age and
structure of dissolved organic carbon in Siberian rivers and streams. Geophysical Research Letters,33(23), L23401. https://doi.org/10.1029/
2006GL028222
Journal of Geophysical Research: Biogeosciences
10.1029/2023JG007853
RIEB ET AL. 16 of 19