Soil Solution DOC Dynamics During Winter in Boreal Hillslopes Paulina Bastviken

Soil Solution DOC Dynamics During Winter in Boreal Hillslopes Paulina Bastviken
Soil Solution DOC Dynamics
During Winter in Boreal Hillslopes
Paulina Bastviken
When plants and animals die they are decomposed into microscopic
particles of organic carbon. In the ground, these carbon particles are
dissolved in the soil water and eventually transported to the stream
channel with the flow of the groundwater. Today the quantities of
dissolved organic carbon (DOC) have been observed to increase in
many lakes and streams around the world, which constitute a threat
against the water quality and ecologic environment of these surface
waters. The amount of organic carbon that is dissolved and
transported in the soil water is mainly controlled by processes related
to temperature and hydrology, two factors which vary seasonally.
Because of difficulties to sample soil water at temperatures below
0°C studies of DOC transport between soil and water during the
winter season are limited. This study therefore conducted a winter
sampling of soil water, with the focus on DOC. Samples were
collected in March 2014 at sites along three hillslopes, orthogonal to
two streams, in a typical Swedish boreal forest northwest of Umeå.
The soil water was extracted with the help of suction lysimeters
installed at different depths in the soil, and heating equipment
powered by batteries. The collected samples were analyzed for DOC
concentration and absorbance after which the results were grouped
together with results from previous sampling campaigns, conducted
in the summer and autumn of 2013. Parallel to this, data representing
a longer time series (2009 to 2012) at another hillslope was
processed. During the summer and autumn an increase in DOC
concentration was observed. The increase was assumed to be caused
by high production and effective degradation of organic matter in the
soil during this warm period. Generally, a decrease in the DOC
concentration then followed during the winter season. One possible
reason for this decrease could be that the bacterial degradation in the
soil continued, during the winter, and transformed the dissolved
carbon into CO2 and CH4. Another possibility is that the DOC was
flushed into the streams by autumn rain events. The study also found
differences concerning the DOC concentration and character in the
soil water, as well as the seasonal variation of these parameters, with
soil depth and distance from the stream along the hillslope profile.
These differences could be correlated to the organic content of the
soil, from which the soil water had been extracted.
Uppsala universitet, Institutionen för geovetenskaper
Kandidatexamen i Geovetenskap, 180 hp
Självständigt arbete i geovetenskap, 15 hp
Tryckt hos Institutionen för geovetenskaper
Geotryckeriet, Uppsala universitet, Uppsala, 2014.
Självständigt arbete Nr 104
Soil Solution DOC Dynamics
During Winter in Boreal Hillslopes
Paulina Bastviken
Självständigt arbete Nr 104
Soil Solution DOC Dynamics
During Winter in Boreal Hillslopes
Paulina Bastviken
Supervisors: José Ledesma and Kevin Bishop
Sammanfattning
När växter och djur dör bryts de ner till mikroskopiska organiska kolpartiklar. I jorden
löser sig dessa kolpartiklar i markvattnet och förs så småningom ut i bäckfåran med
grundvattenströmningen. Idag har mängden upplöst organiskt kol (DOC) observerats
öka i många sjöar och vattendrag runt om i världen, vilket utgör ett hot mot den
vattenkvalitén och den ekologiska miljön i dessa ytvatten. Hur mycket organiskt kol
som löser sig och transporteras i markvattnet kontrolleras i huvudsak av processer
relaterade till temperatur och hydrologi, två faktorer som varierar med årstiderna. På
grund av svårigheter att provta markvatten vid temperaturer under 0°C är kolflödet
mellan mark och vatten under vintersäsongen endast undersökt i begränsad
utsträckning. För att råda bot på detta gjordes i denna studie en vinterprovtagning av
markvatten, med avseende på DOC. Prov samlades in i mars månad 2014 från
provpunker längs tre backprofiler, ortogonala till två bäckar, i en typisk svensk
barrskog nordväst om Umeå. Markvattnet extraherades med hjälp av suglysimetrar,
installerade på olika djup i marken, och batteridriven värmeutrustning. I de insamlade
proverna analyserades koncentrationen och karaktären av DOC varefter resultatet
sammanfördes med analysresultat från tidigare provtagningar, utförda sommaren och
hösten 2013. Parallellt bearbetades även data, insamlat under en längre tidsperiod
(2009 till 2012), från en annan backprofil. Under sommar- och höstsäsongen
observerades en koncentrationsökning av DOC i markvattnet. Ökningen antogs bero
på hög produktion och effektiv nedbrytning av organiskt material i marken under
denna varma period. Under vinterhalvåret följde sedan, generellt sett, en minskning i
koncentrationen av DOC. En möjlig anledning till minskningen skulle kunna vara att
den bakteriella nedbrytningen i marken fortsatte under vinternsäsongen och
omvandlar det upplösta kolet till CO2 och CH4. En annan möjlighet är att
kolpartiklarna spolades ut i vattendragen av regnskurar under hösten. I studien
observerades även skillnader gällande koncentration och karaktär av DOC i
markvattnet, samt den säsongsmässiga variationen av dessa parametrar, med
markdjupet och avståndet från bäckfåran längs backprofilen. Dessa skillnader kunde
kopplas till mängden organiskt material i jorden, varifrån markvattnet provtagits.
Abstract
When plants and animals die they are decomposed into microscopic particles of
organic carbon. In the ground, these carbon particles are dissolved in the soil water
and eventually transported to the stream channel with the flow of the groundwater.
Today the quantities of dissolved organic carbon (DOC) have been observed to
increase in many lakes and streams around the world, which constitute a threat
against the water quality and ecologic environment of these surface waters. The
amount of organic carbon that is dissolved and transported in the soil water is mainly
controlled by processes related to temperature and hydrology, two factors which vary
seasonally. Because of difficulties to sample soil water at temperatures below 0°C
studies of DOC transport between soil and water during the winter season are
limited. This study therefore conducted a winter sampling of soil water, with the focus
on DOC. Samples were collected in March 2014 at sites along three hillslopes,
orthogonal to two streams, in a typical Swedish boreal forest northwest of Umeå. The
soil water was extracted with the help of suction lysimeters installed at different
depths in the soil, and heating equipment powered by batteries. The collected
samples were analyzed for DOC concentration and absorbance after which the
results were grouped together with results from previous sampling campaigns,
conducted in the summer and autumn of 2013. Parallel to this, data representing a
longer time series (2009 to 2012) at another hillslope was processed. During the
summer and autumn an increase in DOC concentration was observed. The increase
was assumed to be caused by high production and effective degradation of organic
matter in the soil during this warm period. Generally, a decrease in the DOC
concentration then followed during the winter season. One possible reason for this
decrease could be that the bacterial degradation in the soil continued, during the
winter, and transformed the dissolved carbon into CO2 and CH4. Another possibility is
that the DOC was flushed into the streams by autumn rain events. The study also
found differences concerning the DOC concentration and character in the soil water,
as well as the seasonal variation of these parameters, with soil depth and distance
from the stream along the hillslope profile. These differences could be correlated to
the organic content of the soil, from which the soil water had been extracted.
Table of content
1. Introduction
2. Background
1
1
2.1. DOC
2.2. Riparian zone
2.3. Absorbance
2
3
4
3. Objectives and hypotheses
4. Material and methods
5
5
4.1. General overview
4.1. Study area
4.2. Sampling procedure
5. Results
5
6
10
13
5.1. DOC concentrations
5.1.1. ROK-transects
5.1.2. S-transect
5.2. DOC character
5.2.1. ROK-transects
5.2.2. S-transect
6. Discussion
13
13
17
20
20
23
25
6.1. DOC concentrations
6.1.1 General trends
6.1.2. ROK-transects
6.1.3. S-transect
6.2. DOC character
6.2.1. ROK-transects and the S-transect
7. Conclusion
8. Acknowledgment
9. References
25
25
25
26
26
26
27
28
28
1. Introduction
Water of good quality is a resource of indisputable importance. In human society,
water is needed both in everyday life, for drinking, growing food, maintaining hygiene
etc., and in the form of a chemical substance in many various industrial processes.
The state of the water is also greatly important from the ecological perspective, since
it plays a key role in maintaining an environment in good condition.
Today, one of the threats against good water quality in the northern
hemisphere is leaching of large, and increasing, quantities of organic carbon into
surface waters (Evans et. al., 2005). High concentrations of dissolved organic carbon
(DOC) in streams and lakes have many environmental impacts. Altered light climate
(Jones et. al., 1992), microorganism growth (Jones et. al, 1992; Löfgren et. al., 2003)
and mobilization of toxic substances (Bergkvist et. al., 1989) are examples of such
impacts, which all may disturb the balance in the aquatic ecosystem. In addition, as a
consequence of these disturbances, DOC may become the source of drinking water
contamination (Löfgren et. al., 2003; Lavonen et. al., 2013).
These problems have led to increasing efforts to gain more knowledge of DOC
origin and transport (Haei et. al., 2010). Research in this field has revealed the soil
immediately adjacent to the water, only a few meters wide from the shoreline, as one
key morphological factor controlling the DOC export to surface waters (Bishop et. al.,
2004). This near-stream soil is referred to as the riparian zone. High organic content
deep down into the soil profile, in combination with the hydrological mechanisms in
this region, located at the base of the slope, make the riparian zone contribute with
substantial amounts of the total DOC ending up in the stream (Dosskey & Bertsch,
1994; Hinton et. al., 1998).
Besides the hydrological control of DOC release, temperature is an important
factor, governing primary production and degradation of organic matter (Hope et. al.,
1994). Consequently, the DOC concentration in soil water and streams varies
seasonally as product of these two controlling factors. Many studies have examined
the seasonality (Eckhardt & Moore, 1990; Köhler et. al., 2008; Haei et. al., 2010;
Ågren et. al., 2012) but it is still unclear how the winter affects DOC transport in high
latitude boreal landscape with long duration of snow cover, (Ågren et. al., 2012). One
reason for the lack in temporal coverage during the winter season is that snow cover
and soil frost complicate sampling of soil water. Challenging these difficulties, this
study aimed at investigating the effect of winter conditions on soil water DOC
dynamics.
To address this aim, sampling was conducted at different locations in a boreal
catchment during a winter campaign and analyzed together with previously collected
data. The concentration and character of the DOC were examined.
2. Background
DOC in soil water and the factors controlling DOC export to streams including
hydrology and various soil characteristics of the boreal catchment are the main focus
of this study. This section therefore contains background information concerning
origin, chemical properties and environmental impacts of DOC, followed by a
description of the riparian zone influence on DOC transport. In the end of the section
the parameter absorbance, used for estimating DOC character, is also introduced.
1
2.1. DOC
Since water is a common compound on earth, living organisms need to be
hydrophobic to not to get dissolved by the environment surrounding them. When
these organisms eventually die and, through the oxidizing work of bacteria and fungi,
become decomposed organic matter the solubility in water increases (Kleber &
Johnson, 2010). This increase in solubility is due to the gain of oxygen-containing
functional groups, like -OH and -COOH, which by their polar character can interact
with the polar water molecules. The process by which the organic matter becomes
soluble is highly dependent on the structure of the carbon compounds and the pH of
the surroundings (Kleber & Johnson, 2010). Other processes that result in carbon in
solution, besides microbial degradation, are root exudation and chemical leakage
from plant litter (Kalbitz et. al., 2000).
DOC is defined as the amount of carbon that remains in a water sample after
letting the water pass through a filter of size 0,45 micrometer (Ledesma, 2012). Total
organic carbon (TOC), on the other hand, also includes the carbon particles trapped
in the filter. Usually, in boreal catchments the particle fraction is so small, in
comparison to the dissolved fraction, that one can consider DOC and TOC to be
equal (Laudon et. al., 2004; Ågren et. al., 2007). This generalization is used in this
study, after verification in the laboratory analysis (data not shown), and DOC is
chosen as main term.
High concentration of DOC in surface waters can have many environmental
impacts. For example, large quantities of DOC cause the color of the water in
streams and lakes to turn into a nuance of yellow to brown (Löfgren et. al., 2003), a
phenomena called brownification. This color can disturb food web interactions by
changing the light climate in the water environment, which affects organisms
depending on photosynthesis (Jones et. al., 1992), and by being a substrate for
bacteria and fungi (Jones et. al., 1992; Löfgren et. al., 2003). In other words, DOC
can cause microorganism growth which in its turn can put the aquatic ecosystem out
of balance. A secondary consequent of this, problematic for human society, is that
increased amounts of microorganisms in surface water increase the amount of
microorganisms in the drinking water supply. This contamination may give the water
an unwanted taste and odor or in the worst case become the source of disease
(Löfgren et. al., 2003; Lavonen et. al., 2013).
DOC can also cause chemical disturbances in the water environment. Even
though not toxic in itself, DOC can easily form complexes with heavy metals
(Bergkvist et. al., 1989) and organic pollutants (Dawson et. al., 2009), a chemical
reaction which mobilizes these toxic substances in the water. In addition, the -COOHgroups of the DOC can release protons and thereby contribute to acidification
(Eshleman & Hemond, 1985).
DOC compounds can have many various chemical structures which give the
carbon molecules different properties. This is referred to as the DOC character or
quality. Degradation rate and bacterial growth are for example affected by the
character of the DOC, since the chemical structure of the carbon compounds
regulates the bioavailability (Ågren et. al., 2008). Smaller (low-weight) and less
complex (aliphatic) DOC molecules are easier for bacteria to consume in comparison
to larger (high-weight) and more complex (aromatic) molecules (Ågren et. al., 2008).
Vegetation type and level of soil saturation are major controls of DOC character,
which can be used to investigate DOC origin (Ågren et. al., 2008). How the character
is used when examining DOC is further explained in the absorbance section (2.3).
2
One process that reduces the export of DOC to surface waters is dissolved organic
matter (DOM) sorption, a process of adsorption of DOM to the surfaces of mineral
particles (Dahm, 1981). Especially, aluminum- and iron oxides (Dahm, 1981; Kaiser
& Guggenberger, 2000) in clay minerals tend to effectively remove the DOC from
solution. In this process the functional groups of the DOC form complexes with the
metals through ligand exchange (Gu et. al. 1995; Kaiser & Guggenberger, 2000), a
reaction which place and hold the DOC in solid phase. DOC can be chemically
removed from soil solution by mineralization (Glazel et. al., 2004). During this
process the DOC is decomposed by microorganisms and transformed into inorganic
compounds, such as CO2 and CH4. The third process, by which the concentration of
DOC in the soil water is reduced, is lateral flushing through the aquatic conduit
(Dyson et. al., 2011). This last process will be further explained in the next section,
Riparian zone.
2.2. Riparian zone
When precipitation lands on the ground surface in a catchment, it infiltrates the soil
and percolates down to the groundwater table. After reaching this level, the water
flow downslope according to the pressure gradient, until eventually reaching the
stream channel. During this part of the hydrological cycle, the water is constantly
affected by the chemical characteristics of the soil it is travelling through (Bishop et
al., 2004). The quantity and composition of dissolved substances in the soil solution
therefore change along the way.
Postglacial processes have developed a soil cover in boreal catchments of
Sweden which can be classified according to the slope (Bishop et al., 2004).
Upslope soils, covering the mayor part of the catchment, consist of podzols while
more organic and saturated soils are found, just adjacent to the stream, in the so
called riparian zone (Bishop et al. 2004).
In the riparian zone, organic carbon is accumulated all through the soil profile
due to the level of saturation and the slow degradation rates (Haei, 2010). This large
amount of organic carbon, in combination with the fact that all precipitation falling
within the watershed needs to pass through this narrow body of soil before entering
the stream, make the riparian zone a local hotspot for DOC export to surface waters
(Dosskey & Bertsch, 1994). In addition to the soil characteristics of the riparian zone,
a hydrological factor, referred to as the transmissivity feedback mechanism (Rodhe,
1989; Bishop et. al. 2004), has a central role in the DOC release. This mechanism
occurs due to the increase in conductivity upward in the vertical soil profile. During
rain fall events or snow melt, the groundwater table rises in the soil profile and enters
soil layers both richer in organic matter and more hydraulically conductive (Fig. 1).
The high flow of water through this conductive and organic layer flushes the DOC
from the soil water, increasing the export of carbon to the stream (Bishop et. al.,
2004).
3
Figure 1.Schematic illustration of the riparian zone, modified after Winterdahl et. al. (2011).
The grey color display soil rich in organic matter and blue arrows display the infiltration and
lateral flow through the soil. Increasing size of arrows indicates higher conductivity.
2.3. Absorbance
One way of characterizing DOC is to measure its spectroscopic properties (Chin et.
al., 1994). By investigating how a water sample absorbs different wavelengths,
information concerning the size and structure of the carbon compounds within the
sample can be gained. This information can then be used to derive the mobility (Chin
et. al., 1994; Löfgren et. al., 2003) and bioavailability of the carbon molecules
(Perdue, 1998; Löfgren et. al., 2003), determine the tendency for the carbon to react
with pollutants (Chin et. al., 1994; Löfgren et. al., 2003) or trace the environment
which the carbon originates from (Ågren et. al., 2008).
The general principle of the method is that radiation of a known wavelength
and intensity is sent through one side of a cuvette containing the sample and that the
remaining intensity is measured on the opposite side. In mathematical terms,
absorbance (A) is the transmitted intensity (l1) divided by the incident intensity (l0) of
a wavelength, presented in the form of a logarithm (A=log(I1/I0)) (Fuwa & Vallee,
1963). The unit of absorbance is cm-1, which refers to the thickness of the sample
cuvette, through which the radiation is sent.
Regarding characterization of DOC, two particular wavelengths (254 nm and
365 nm) have been empirically proven to be useful as a ratio (A254/A365), to
investigate the relative amount of low-weight and high-weight carbon compounds
(Ågren et. al., 2008; Maehder, 2012). A high value of the ratio A 254/A365 corresponds
to a high content of aliphatic compounds in the DOC, which are compounds more
mobile and available for microorganisms (Ågren et. al., 2008). If to consider origin, a
high A254/A365 corresponds to DOC likely to have been derived from forest areas,
where the soils are of mineral character. In opposite, a low A254/A365 indicates a high
content of high-weight carbon compounds, with low mobility and bioavailability,
usually derived from organic soils in wetlands (Ågren et. al., 2008). How the molecule
character of the DOC affects the binding capacity between the DOC and toxic
4
substances is more complex. This depends, among other things, on the pH of the
surroundings and on the type of contaminant (Kordel et. al., 1997).
Together with the absorbance ratio a useful spectrometic parameter, describing the
character of the DOC, is Specific Ultraviolet Absorbance (SUVA) (Weishaar et. al.,
2003). The SUVA is defined as the absorbance of ultra violet radiation at a given
wavelength after the absorbance has been normalized for DOC concentration. If the
wavelength of 254 nm is used, the SUVA will correlate positively to a high fraction of
aromatic carbon (Weishaar et. al., 2003).
3. Objectives and hypotheses
The project objectives were to investigate seasonal variability of DOC concentration
and character in the soil water of a boreal catchment, and to examine how the
seasonal trend may vary vertically in the soil or between different locations relative to
the stream channel. Within the seasonality, the study especially aimed at
investigating the variability during winter conditions.
The differences between locations and horizons, regarding percentage of
organic matter and the level of saturation, were hypothesized to affect the DOC
concentration and character of the soil water. Organic horizons were, in comparison
to mineral horizons, expected to contain soil water with high concentration of DOC
(Michalzik & Matzner, 1999; Kalbitz et. al., 2000, Ågren et. al., 2008, Grabs et. al.,
2012), characterized by high-weight (Gu et. al., 1995; Ågren et. al., 2008) and
complex molecules (McKnight et. al., 1992; Ågren et. al., 2008)
In addition, the DOC concentrations were thought to be more variable in
organic horizons, than in mineral horizons, as DOC was expected to be adsorbed to
the surface of mineral particles (Dahm, 1981; McDowell, 1985; Hope et. al., 1994).
A third hypothesis was that the seasonal variation of DOC concentration would
follow a cyclic pattern governed by lateral flow rates and primary production. The
DOC concentration in the soil water was expected to increases during the summer
season to reach its maximum value in the autumn (Köhler et. al., 2008), as a
consequence of high temperatures generating vegetation growth and intensive
bacterial activity (Christ & David, 1996). During the winter, the DOC concentration
was then thought to remain high, because of low lateral flow rates deep down in the
soil profile, until decreasing rapidly in the spring due to snow melt flushing (Ågren et.
al., 2007; Buffram et. al., 2007; Köhler et. al., 2009).
4. Material and methods
4.1. General overview
During a winter field campaign in 2014, soil water samples were collected at three
riparian and four upslope monitoring sites, together forming three lines orthogonal to
stream water channels (transects), in a boreal catchment. At each site the samples
were collected along a depth profile with the use of previously installed suction
lysimeters. Heating boxes were placed above the ground where the lysimeters are
placed in order to decrease the risk of water freezing during the sampling procedure.
After collection, the soil water samples were weighed and frozen before being further
analyzed in the water laboratory at SLU Umeå.
5
To be able to investigate seasonality, data from previous campaigns, in the summer
and autumn of 2013, were considered together with the data collected in the present
study. A longer time series and a broader spatial coverage were obtained by using
data from three other monitoring sites, one riparian and two upslope (from now on
called the S-transect). The data available from the S-transect had been collected
regularly, from the spring 2009 to the end of the autumn 2012. Under the following
headings, the study site and the sampling method are described in more detail.
4.2. Study area
The winter campaign in March 2014 was conducted in the 68 km2 boreal Krycklan
catchment (Figure 2), located approximately 50 km northwest of Umeå, in the north
of Sweden (Ledesma et. al., 2013).
The bedrock of the area consists of Svecofennian gneissic rock, to the
greatest extent of sedimentary origin (Ågren et. al., 2007). Overlaying the bedrock
glacial till and glacio-fluvial deposits form a sediment cover, up to tens of meters thick
(Ledesma et. al., 2013). The soil profiles in the upslope regions display welldeveloped iron podzols, while more organic soils are found in the riparian zones
(Ledesma et. al., 2013). In general, the topography forms a gentle slope toward the
southeast with the elevation ranging between 372 and 126 m a.m.s.l (Ledesma et.
al., 2013).
The catchment are mainly covered by forest (88%), with some elements of
wetland (8%), agricultural landscape (3%) and lakes (1%) (Grabs et. al., 2012). In the
forest the tree volume consists of Scots pine, Norway spruce and Deciduas forest,
dominated by Birch (Ledesma et. at, 2013).
Climatic data (1981-2010) display an annual mean daily temperature of 1,8 °C
and an annual mean precipitation of 640 mm (Ledesma et. al., 2013). Half of the
precipitation falls as snow and the duration of the snow cover is from October to May,
on average 168 days per year (Laudon et. al., 2011). Mean annual runoff and
evapotranspiration is 321 mm/year and 319 mm/year, respectively (1981-2009)
(Ledesma et. al., 2013).
Forestry and alternated drainage paths are activities which have affected the
area, but the anthropological impacts on the catchment are considered to be rather
low (Ledesma et al., 2013). Today, less than 1% of the catchment is harvested yearly
and the central part is left with no management (Ledesma et al., 2013).
The Krycklan catchment is a part of the Svartberget field research infrastructure
which is in the regime of Swedish University of Agricultural Science (SLU) (Laudon,
2012). The catchment has been the object of several hundred scientific publications
(Laudon, 2012) since it became an area of research in 1923 (Ledesma, 2012)
One of the present monitoring systems, installed in 2007, is the Riparian
Observatory of Krycklan (ROK) (Grabs et. al., 2012). The observatory consists of 13
instrumented sites riparian soil profiles (sites), spatially spread over the catchment in
sedimentary and till deposits (Grabs et. al., 2012). The equipment installed at the
sites will be further describes under the section Sampling procedure (4.3).
The seven sites, sampled June to March (2013-2014), are together forming
three transects, orthogonal to three different streams located in the middle of the
Krycklan catchment (figures 2 and 3). At each transect one riparian site from the ROK
are located at the base of the slope, approximately 1-2 m from the stream. Further
from the stream one or two upslope sites make the transects complete. The upslope
sites, installed in 2012, were sampled for the first time in June 2013.
6
For practical reasons these transects will in the following text be referred to as the
ROK-transects. Separately the names of the transects will also include the numbers
of the riparian sites, as in the example of the ROK2-transect which include the
riparian site number 2 (R2) and the upslope sites number 34 and 35 (U34 and U35).
Details about the ROK-transects are presented in table 1 and the observation
equipment, installed at all the sites, will be described under the heading sampling
procedure.
The S-transect was established in 1992 (Seibert et. al., 2009) and consists of
three observation sites along a line following the direction of the lateral groundwater
flow, adjacent to the same stream as the ROK5-transect. The sites, in the transect,
have been sampled monthly since 1996 (Seibert et. al., 2009). The dataset used in
this study was sampled once a month to every second month, from April 2009 to
December 2012. The details of the observation sites along the S-transect are
presented in table 2.
7
Table 1. Description of ROK-transects.
Transect
Site
Distance from
stream [m]
Profile
depth [cm]
Mean
groundwater
level [cm]
Soil
Description
Vegetation
ROK2
R2
1-2
75
4
Peat
-
ROK2
U34
26-27
75
0
Peat to 0-20 cm depth,
uneven thickness,
then Gley soil
Birch
ROK2
U35
53-54
75
>47*
Gley podzol
Pine
ROK5
R5
1-2
75
11
Peat
-
ROK5
U36
24-25
90
>37*
Podzol
Pine
ROK10
R10
1-2
75
16
Peat
-
ROK10
U37
16-17
75
>14*
Podzol
Spruce
*The groundwater level was below recognizable depth during several campaigns. At these
occasions the maximum measurable depth are used.
Table 2. Description of the S-transect, reconstructed after information in Nyberg et. al. (2001)
Stählin et. al. (2001) and Seibert et. al. (2009).
Transect
S
Site
S4
Distance from
stream [m]
4
Profile
depth [cm]
Groundwater
level [cm]
Soil
Description
Vegetation
30-40
Peat to 30 cm depth,
organic-rich mineral
soil to 60 cm depth,
then mineral soil
-
Spruce
Pine
65
S
S12
12
70
30-40
Peat to 20 cm depth,
organic-rich mineral
soil to 50 cm depth,
then mineral soil
S
S22
22
90
55
Podzol
8
Figure 2. Krycklan catchment with sampling sites positions. Map: José Ledesma (2014).
Figure 3. Sampling sites positions in close-up view. Map: José Ledesma (2014).
9
4.3 Sampling procedure
The soil water samples were collected using suction cup lysimeters. A suction
lysimeter is a small ceramic cylinder which, similar to the soil texture, has a porous
structure able to accommodate soil water. The lysimeters are buried at different
depths in the soil, and connected to thin tubes which can transport soil water to the
surface. During sampling, sterile needles are attached to the end of the tubes, at the
surface, and extraction of water is then enabled by inserting these needles into the
rubber caps of previously vacuumed sampling bottles.
The lysimeter setup in the ROK-transects consists of a depth profile with a pair
of suction lysimeters installed 15, 30, 45, 60 and 75 cm below the ground surface. An
extra pair of lysimeters is installed, at 90 cm depth, at U36. In order to monitor the
position of the groundwater table in the vicinity of the lysimeters, all observation sites
are also equipped with a perforated PVC tube hosting an automatic water logging
device (Ledesma, 2012). Figure 4 displays the sampling equipment at an observation
site in a cross section.
Figure 4. Illustration of lysimeter setup. (Grabs et. al., 2012).
To increase the probability of extracting water from the suction lysimeters in
potentially freezing condition during the campaign in March 2014, heating boxes were
installed at all the sites before sampling (fig. 5). The heating equipment consisted of a
car battery which, through a transformer and a thermostat, was connected to a
heating cable. The heating box also contained an inner box of isolating styrofoam,
10
with space for the sampling bottles. When installing a box, the heating cable was
inserted as far down as possible down the pipe where the lysimeters tubes come to
the surface, which in most cases was approximately one to two decimeters. The
effect and importance of the heating box was difficult to evaluate as generally the
batteries stopped functioning during the time of sampling, or did not function at all.
Luckily, mild temperatures contributed to enough water being sampled at all sites
except R2 and U34.
Figure 5. Heatingbox. Photo: Paulina Bastviken (2014).
11
The soil water samples were collected in acid-washed highly-density polyethylene
bottles (250 mL) (Ledesma, 2012). Prior to the sampling, the bottles were vacuumed
in the laboratory at SLU Umeå (fig. 6). This was done so that the pressure difference
between the bottles and the lysimeters could drive the water up the lysimeter tubes.
In order to extract the water left in the lysimeter tubes since the previous sampling
campaign, smaller rinsing bottles (50 mL) were connected to the lysimeters before
the sampling bottles.
Figure 6. Sampling bottles being vacuumed. Photo: Paulina Bastviken (2014).
When sampling at all sites had been completed, rinsing bottles and soil water
samples were transported to the laboratory at SLU Umeå where they were weighted.
After checked for possible remaining vacuum, the rinsing bottles were poured away
while the soil water samples were placed in a cooling room. Approximately 2-4 days
after collection the water samples were subsampled and analyzed for TOC
concentration, pH and absorbance, at SLU Umeå. During the data processing, worth
mentioning is that a quality check was done on the A254/A365 data, removing all
extreme values likely to be misleading. These extreme values are the result of
absorbance ratios derived from samples with very low DOC concentration, where the
potential errors in the absorbance measurements are larger than the actual values of
absorbance.
Except for the heating box, the data from all previous campaigns, at the ROKtransects and the S-transect, had been collected according to the same procedure as
the sampling in March 2014.
12
5. Results
In this section, the result of the study will be presented and described. The DOC
concentration and DOC character will be addressed under separate headings.
Graphs containing the same type of data are plotted in the same scale to make them
easy to compare.
5.1. DOC concentration
5.1.1. ROK-transects
During summer and autumn 2013 (June to October) all riparian sites in the ROKtransects displayed an increase in soil water DOC concentration. Also site U35
exhibited this tendency, but with a temporary decrease in the DOC concentration
during August and September. At the remaining upslope observational sites the DOC
concentration remained relatively stable throughout the summer and autumn. In the
following winter (October to March) the DOC concentration decreased at site R5 and
R10, as well as at site U35. The concentration, at site U37, remained stable, while
the samples at site U36 revealed a small increase in DOC concentration during the
season. At site R2 and U34, soil frost prevented collection of samples during the
winter campaign. A persistent tendency, throughout the sampling period and at all
sites, is that the changes in DOC concentrations seem to become less evident
deeper down in the soil profile. Figure 7 illustrates the seasonality of the DOC
concentration at all ROK-transects.
To investigate general relationships, the organic content of the soil at all
lysimeter depths were plotted against the corresponding annual mean DOC
concentration (fig. 8). This revealed that higher DOC concentrations are found in
horizons containing more organic material. In figure 9, the mean annual DOC
concentrations are visualized along the depth profile at each site.
The relationship between the DOC concentration and the organic content of
the soil was further investigated by classifying all lysimeter horizons into three
categories according to the percentage of carbon (organic, organo-mineral and
mineral), and then by examining the seasonality of the DOC concentration for each
category separately (fig. 10). The DOC concentration in the mineral horizons was
stable at 5 mg/L throughout the sampling period. This was also the case for the
organo-mineral horizons during summer and autumn. During the winter season, the
DOC concentrations in the organo-mineral horizons then increased, but it should be
noted that this increase only was based on one single sample. In the organic
horizons, an increase in soil water DOC concentration occurred during summer and
autumn, followed by a decrease during winter.
13
Figure 7. Seasonal variability (June 2013 to March 2014) of soil water DOC concentration at
all sites in the ROK-transects. The colors illustrate the depth of the lysimeter position which is
written in centimeters in the legend. The black line and the blue triangle display the average
groundwater table during the period of sampling, relative to the lysimeters.
14
Figure 8. Annual mean DOC concentration plotted against carbon content of the soil, in the
horizons of the lysimeter positions. The legend indicates which ROK-sites the lysimeters
belong to (brown color for lysimeters in the riparian zone, green color for lysimeters upslope
and different shapes on the symbols for different sites).
Figure 9. Annual mean DOC concentration plotted against depth at each ROK-site (the
lysimeters are illustrated with the same symbols as in figure 8).
15
Figure 10. DOC concentration in horizons classified as organic (>15% C), organo-mineral (315% C) and mineral (<3% C). Numbers beside campaigns indicate how many samples that
were successfully extracted out of the total amount of available lysimeters in each category.
16
5.1.2. S-transect
In the graphs of figure 11 and 12, based on samples from the S-transect, the data
cover three years (April 2009 to December 2012) of DOC concentration variability. To
make the winter trends in these figures clear, the winter seasons according to the
climatic definition (first of December to last of February), were marked with a grey
background.
At site S4 (fig. 11) the DOC concentrations varied between 20 to 90 mg/L
during the sampling period and, similar to the riparian sites sampled at the ROKtransects, lower and less varied concentrations can be correlated with increasing
depth. A cyclic pattern is also visible at S4. The seasonal cycle displays an increase
in DOC concentration, from approximately may to November, followed decrease,
ending in may the year after.
The winter trends in 2009 and 2010 look similar taking into account the
difference in sampling frequency during September to March. This would suggest a
slow decrease in DOC concentration during these two winters, followed by a more
rapid decrease in the beginning of the spring. The rapid decrease in spring, which
ends with local minimum values of DOC concentration, is recurrent during the period
and correlates well with data of discharge (fig. 11b).
Along the depth profile at site S4 the pattern of variation changes. The
lysimeters at 10 and 25 cm depth display quick drops in DOC concentration during
summer, a drop which is displayed earlier further down in the profile in 2009 and
2011 while not displayed at all in these deeper horizons in 2010. After plotting the
DOC concentrations against precipitation data these drops seemed to be correspond
to great rainfall events (fig. 11a).
In the end of the winter between 2011 and 2012 a break in the cyclic trend at
S4 is displayed. The DOC concentrations in the lower horizons start to increase at
the same time as a rapid decrease occurred in the horizons from the top two
lysimeters. Meteorological records from the period show a temperature drop, down to
-30°C, at point of time for this break in the seasonal trend (fig. 11c).
At the upslope site S12 (fig. 12) the soil water contained less DOC, with
concentrations ranging from 5 to 40 mg/L. The concentrations seem more stable
throughout the period than at S4, but very few samples had been collected from the
top three lysimeters at S12. The lack of data could hide eventual variation in the
shallower horizons. The frequently mentioned tendency, of DOC concentration
decreasing with depth, is visible at site S12 and the lysimeter at 20 cm depth display
the same respond to precipitation events, as the shallower lysimeters at site S4.
The samples from the upslope site S22 displayed low DOC concentrations
from 1 to 5 mg/L independent on depth and with no visible seasonality (fig. 12).
17
Figure 11. Seasonal variability (April 2009 to December 2012) of soil water DOC
concentration at the riparian site S4 in the S-transect, including precipitation (a), discharge
(b) and temperature (c) (all with a green, dotted line). As in figure 6, the colors illustrate the
depth of the lysimeter position. The lysimeter depth in centimeter and the relative position of
the groundwater table is displayed in the legend. The winter seasons are marked with grey
areas.
18
Figure 12. Seasonal variability (April 2009 to December 2012) of soil water DOC
concentration at the sites upslope, S12 and S22, in the S-transect. Precipitation data (green
dotted line) is included in the graph displaying DOC concentrations at site S12. The winter
seasons are marked with a grey area, otherwise the symbols are the same as in figure 7.
19
5.2. DOC character
5.2.1. ROK-transects
At all riparian sites, both the A254/A365 and the SUVA254 stayed relatively stable
throughout the whole sampling period, with values between 3 and 6 in general. The
DOC in the soil water samples collected at 15 cm depth in the riparian profiles (the
horizon richest in organic matter) display a vague tendency to coincide with the
lowest A254/A365 and highest SUVA254. As an exception, the SUVA254 at site R5
exhibits the reverse of this tendency, correlating positively with increasing depth.
The absorbance ratios were generally more varied in the samples from the
upslope sites, than in the samples from the riparian sites. A clear depth-dependent
trend is seen in both the A254/A365 and the SUVA254 of site U35, where the shallower,
more organic horizons correlate with values indicating higher molecular weight and
higher aromaticity. The A254/A365 at site U35 also became more varied with depth, a
trend which as well can be seen at site U34 and U36 where one of the lysimeters at
15 cm depth extracted water with more seasonally stable absorbance ratio than other
lysimeters. In figure 13 and 14 the results from the character analysis, absorbance
and SUVA, for all ROK-transects are illustrated.
20
Figure 13. Seasonal variability (June 2013 to March 2014) of A254/A365 at all sites in the ROKtransects. Illustration symbols are the same as in figure 7.
21
Figure 14. Seasonal variability (June 2013 to March 2014) of SUVA254 at all sites in the ROKtransects. Illustration symbols are the same as in figure 7.
22
5.2.2. S-transect
The absorbance ratio from the sites in the S-transect followed the same pattern as
the absorbance ratio from the ROK-transects. S4, the riparian site, (fig. 15) displays
ratios temporally stable around 5 with small variations along the depth profile. Going
upslope to S12 (fig. 16), a depth trend is visible with shallower horizons correlating to
lower absorbance ratio. The ratio range at S12 is approximately 4-7 and more
variability seems to be present in comparison to S4. Differences with depth are hard
to distinguish since the shallower lysimeters, 5, 10 and 20 cm depth, were sampled
much less frequently than the deeper lysimeters. At the upslope site S22 (fig. 16) the
variability in time and along the depth profile is high, with absorbance ratios shifting
from around 0 to 15.
Figure 15. Seasonal variability (April 2009 to December 2012) of A254/A365 at the riparian site,
S4, in the S-transect. The colors illustrate the depth of the lysimeter position. The lysimeter
depth in centimeter and the relative position of the groundwater table is displayed in the
legend. The winter seasons are marked with grey areas.
23
Figure 16. Seasonal variability (April 2009 to December 2012) of A254/A365 at the upslope
sites, S12 and S22, in the S-transect. The symbol system is the same as in figure 15.
24
6. Discussion
Following the same structure as the presentation of the result, the DOC
concentrations and character will be discussed separately.
6.1. DOC concentration
6.1.1. General trends
The result did reflect the fact that organic soil is a source material for DOC (Kleber &
Johnson, 2010) by displaying a clear correlation between the DOC concentration in
the soil water and the carbon content of the soil. Organic horizons did, in general,
accommodate soil water with higher DOC concentration in comparison to more
mineral horizons. As a consequence the riparian sites, with high content of organic
matter all through the soil profile, displayed higher levels of DOC in the soil water
than the upslope mineral sites. An exception from this tendency was site U35. At this
site samples revealed a low carbon content in the soil while the collected soil water,
contained high levels of DOC. Explanations for this are not obvious. One hypothesis
is that spatial variation in vegetation (on the ground surface or in the form of roots)
provided carbon easily dissolved at this site. Local variation in pH, promoting the
dissolving processes (Kleber & Johnson, 2010), could also perhaps have led or
contributed to this result. The dataset available for this study was not sufficient to
better understand the relatively high DOC concentration at U35.
Another difference, between organic and mineral soils, was the level of
variability in the DOC concentration. The DOC concentration in the water from
mineral horizons displayed more stability. Together with the concentration of DOC
being low in these horizons, this supports the theory of DOM sorption in mineral soils
removing carbon from solution (Dahm, 1981; Kaiser & Guggenberger, 2000).
6.1.2. ROK-transects
The seasonal trend during summer and autumn (2013), at all the riparian sites in all
the ROK- transects, was that the concentration of DOC increased in the soil water.
This followed the hypothesis that primary production and high bacterial activity are
the factors controlling the level of soil water DOC during high temperature seasons
(Christ & David, 1996).
During the following winter, the main trend (displayed at site R5, R10 and U35)
was that the concentration of soil water DOC decreased. The reason for this
decrease could be low DOC input in combination with continued DOC consumption
by bacteria. If this is the case, the DOC in the soil under the snow cover is
mineralized into CO2 or CH4 (Glazel et. al., 2004) that either stays in the solution or
escape to the atmosphere. Another possibility is that the DOC was flushed in late
autumn rain events (Dyson et. al., 2011), which the long period with no data during
the winter season would hide. To be able to rule out one or the other of the suggested
processes more frequent sampling would have been required. Analysis of CO2 in the
samples would also provide helpful information.
Showing an opposite trend during the winter season, the lysimeter at 15 cm
depth at site U36 is displaying an increase in DOC concentration. Similar to site U35,
this result could reflect the influence of spatial variation, regarding vegetation and pH,
on DOC concentration. It should also be mentioned that it is this measurement which
25
causes the DOC increase during winter in the organo-mineral horizons of the ROKtransects (fig. 10), as this was the only sample extracted from this category during
the winter campaign.
Site U37 displayed no change in DOC concentration by season or by depth.
The high mineral content in the soil is probably making the DOM sorption the
dominant factor controlling DOC concentration at this site.
The results from the ROK-transects reflect how DOC concentrations in the
horizons of boreal hillslope profiles change during a warm winter condition. Likely is
that the trend would be different during other temperature conditions (Haei et. al.,
2010), with for example more or less extensive soil frost (Ågren et. al, 2012). In
addition to climatic setting, global warming is therefore an important factor to consider
in the quest for understanding the present and future winter seasonal dynamics of
soil water DOC in different geographical regimes.
6.1.3. S-transect
Site S4 at the S-transect displayed a general cyclic pattern consistent with the
hypothesis of DOC fluctuations during summer, autumn and spring. The changes in
DOC concentration in combination with the precipitation and discharge data suggest
effects of lateral flushing and dilution (Rodhe et. al., 1989; Bishop et. al., 2004; Dyson
et. al., 2011), both related to snow melt and summer/autumn rain events. The
difference in the response along depth profile is likely an effect of the position of the
groundwater level and the amount of infiltrating water.
During the winter seasons 2009 and 2010 the general decrease in DOC at S4
could, similar to the riparian zones at the ROK-transects, be related to DOC
mineralization and flushing events. The fluctuation observed in the winter 2011, when
the shallower lysimeter horizons display a rapid decrease in DOC concentration at
the same time as the lower horizon accumulate DOC in the soil water, is possibly
related to another process. In the study conducted by Ågren et. al. (2012), DOC
freeze out due to soil frost is described as a process increasing DOC concentrations
in the deeper parts of the soil profile. The ice developing from the top of the soil
profile transport the DOC downward, as the organized structure in the ice crystals are
unwilling to accommodate these organic molecules. This process may explain the
fluctuations in the winter 2011 at site S4, since the variation in the DOC concentration
is accompanied by a significant temperature decrease and potentially a significant
soil frost. Unfortunately no data of soil frost depth was available for the sites and time
of this study.
According to the data available at the upslope sites S12 and S22 the DOM
sorption seemed to be dominating over the seasonal trends due to the mineral
composition of the soil. Low frequency in successful extraction of samples from the
shallower horizons at site S12 is though possibly hiding larger variability.
6.2. DOC character
6.2.1. ROK-transects and the S-transect
A clear trend of low absorbance ratio and high SUVA was observed in the organic
horizons of all transects while the opposite was observed for the mineral horizons.
This was an expected result which indicated higher content of high-weight and
26
aromatic molecules in the soil solution of more organic soil (Weishaar et. al., 2003;
Ågren et. al., 2008).
That no change in absorbance was displayed, along with decreases in DOC
concentration possibly related to bacterial mineralization, suggests that different
fractions may be limited and that there is organisms present in the soil specialized in
consuming the carbon of the available character.
The high variability in the DOC character in soil horizons with high mineral
content could be an indication of the high temporal and spatial variability of the
organic carbon in these soil bodies. On the contrary, the peat at the riparian sites
forms a more stable source of organic carbon molecules. It is though important to
take into account that the accuracy of absorbance depends on the concentration of
DOC. The measurement errors, in the data from the upslope sites, are probably
enhancing the variability to a great extent due to the low DOC concentration at these
locations (especially site U36 and U37).
A chemical process affecting the A254/A365 and the SUVA254, suggested by
Maloney et. al. (2005) and Köhler et. al. (2013), is mobilization of aluminum and iron.
This could also have an influence on the data, increasing the absorbance in the soil
water samples.
7. Conclusion
The organic content of the soil was reflected in the DOC concentration and character
of the soil water. Soil water from horizons with high percentage of organic matter
contained higher concentrations of DOC, with a higher fraction of high-weight and
complex molecules (Weishaar et. al., 2003; Ågren et. al., 2008), compared to soil
horizons with low organic content. The concentrations of DOC were also observed to
be more varied in the soil water from organic horizons than in the soil water from
mineral horizons.
Regarding the variability of the DOC character, the absorbance analysis
revealed higher variability in the mineral horizons than in the organic horizons. Since
the potential error in the absorbance measurement increases with lower DOC
concentration, the true variability of the DOC character was though probably lower
than shown in the figures of this study.
At the riparian sites, the DOC concentration seasonality was dominated by an
increase in the summer/autumn and a decrease in the spring, likely being the result
of increased primary production in combination with bacterial activity (Christ &
David, 1996) and lateral flushing during snow melt (Ågren et. al., 2007; Buffram et.
al., 2007; Köhler et. al., 2009). Rain fall events during summer and autumn were
effecting the DOC concentrations intra-seasonally, by diluting and flushing the DOC
in the soil horizons (Rodhe, 1989; Bishop et. al. 2004; Dyson et. al., 2011). The main
seasonal trend during winter was that the DOC levels had decreased. The reasons
for this decrease were in this study thought to be bacterial mineralization (Glazel et.
al., 2004) or late autumn rain events. Fluctuations in the DOC concentration, likely to
be related to DOC freeze out (Ågren et. al., 2012), were also observed. The sites
upslope and the mineral horizons remained relatively stable throughout all seasons,
supporting theories of DOM sorption (Dahm, 1981; McDowell, 1985; Hope et. al.,
1994) as a strong control of DOC concentration.
27
8. Acknowledgement
Special thanks are given to José Ledesma for being the main source of knowledge
and feedback during the development of this thesis. I would also like to thank Kevin
Bishop for introducing me to the subject of the project, Magnus Hellqvist for giving
structure to the process of writing and Joachim Place for help with figures. Gratitude
is also given to staff at the Svartberget Field Station and the department of Forest
Ecology and Management at SLU Umeå for help during sampling and laboratory
analysis.
9. References
Bergkvist, B., Folkeson, L., Berggren, D. (1989). Fluxes of Cu, Zn, Pb, Cd, Cr,
and Ni in temperate forest ecosystems. A literature review. Water, Air, and Soil
Pollution, vol. 47, pp. 217–286.
Bishop, K., Seiber, J., Köhler, S., Laudon, H. (2004). Resolving the Double
Paradox of rapidly mobilized old water with highly variable responses in runoff
chemistry. Hydrological processes, vol.18, pp.185-189.
Buffam, I., Laudon, H., Temnerud, J., Mörth, C.M., Bishop, K. (2007). Landscapescale variability of acidity and dissolved organic carbon during spring flood in a
boreal stream network. Journal of Geophysical Research, vol. 112.
Chin, Y., Alken, G., O´Loughlin, E. (1994). Molecular Weight, Polydispersity, and
Spectroscopic Properties of Aquatic Humic Substances. Environmental
Science and Technology, vol 28, pp. 1853-1858.
Christ, M.J., David, M.B. (1996). Temperature and moisture effects on the
production of dissolved organic carbon in a spodosol. Soil Biology and
Biochemistry, vol. 28, pp. 1191-1199.
Dahm, C.N. (1981). Pathways and Mechanisms for Removal of Dissolved Organic
Carbon from Leaf Leachate in Streams. Canadian Journal of Fisheries and
Aquatic Science, vol. 38, pp. 68-76.
Dawson, J.J.C., Malcolm, I.A., Middlemas, S.J., Tetzlaff, D., Soulsby, C. (2009). Is the
Composition of Dissolved Organic Carbon Changing in Upland Acidic
Streams? Environmental Science and Technology, vol. 43, pp. 7748–7753.
Dosskey, M.G., Bertsch, P.M. (1994) Forest Sources and Pathways of OrganicMatter Transport to a Blackwater Stream – A Hydrological Approach.
Biogeochemistry, vol. 24, pp.1-19.
Dyson, K.E., Billett, M.F., Dinsmore, K.J., Harvey, F., Thomson, A.M., Piirainen S.,
Kortelainen, P. (2011). Release of aquatic carbon from two peatland
catchments in E. Finland during the spring snowmelt period. Biogeochemistry,
vol. 103, pp. 125-142.
Eckhardt, B.W., Moore, T.R. (1990). Controls on Dissolved Organic Carbon
Concentrations in Streams, Southern Québec. Canadian Journal of Fisheries
and Aquatic Sciences, vol. 47, pp. 1537-1544.
Eshleman, K.N., Hemond, H.F. (1985). The role of organic acids in the acid-base
status of surface waters at Bickford Watershed, Massachussetts. Water
Resources Research, vol. 21, pp. 1503–1510.
Evans, C.D., Monteith, D.T., Cooper, D.M. (2005). Long-term increases in surface
water dissolved organic carbon: Observations, possible causes and
environmental impacts. Environmental Pollution, vol. 137, pp. 55–71.
28
Fuwa, K., Vallee, B.L. (1963). The Physical Basis Absorption Spectrometry of
Analytical
Atomic - The Pertinence of the Beer-Lambert Law. Analytic
Chemistry, vol. 35, pp. 942–946.
Glatzel, S., Basiliko, N., Moore, T. (2004). Carbon dioxide and methane production
potentials of peats from natural, harvested, and restored sites, eastern
Quebec, Canada. Wetlands, vol. 24, pp. 261–267.
Grabs, T., Bishop, K.H., Laudon, H., Lyon, S.W., J. Seibert, J. (2012). Riparian zone
processes and soil water total organic carbon (TOC): implications for spatial
variability, upscaling and carbon exports. Biogeosciences, vol. 9, pp. 3031–
3069.
Gu, B., Schmitt, J., Chen, Z., Liang, L., McCarthy, J.F. (1995). Adsorption and
desorption of differend organic matter fraction on iron oxide. Geochimica et
Cosmochimica Acta, vol. 59, pp. 219-229.
Haei, M., Öquist, M.G., Buffam, I., Ågren, A., Blomkvist, P., Bishop, K., Löfvenius,
M.O., Laudon, H. (2010). Cold winter soils enhance dissolved organic carbon
concentrations in soil and stream water. Geophysical Research Letters, vol.
37.
Hope, D., Billett, M.F., Cresser, M.S. (1994). A Review of the Export of Carbon in
River Water: Fluxes and Processes. Environmental Pollution, vol. 84, pp. 301324.
Hinton, M.J., Schiff, S.L., English, M.C. (1998) Sources and flowpaths of dissolved
organic carbon during storms in two forested watersheds of the Precambrian
Shield. Biogeochemistry, vol. 41, pp. 175-197.
Jones, R.I. (1992).The influence of humic substances on lacustrine planktonic food
chains. Hydrobiologia, vol. 229, pp. 73-91.
Kaiser, K., Guggenberger, G. (2000). The role of DOM sorption to mineral surfaces in
the preservation of organoc matter in soils. Organic Geochemistry, vol. 31, pp.
711-725.
Kalbitz, K., Solinger, S., Park, J.H., Michalzik, B., Matzner, E. (2000). Controls on the
dynamics of dissolves organic matter in soils: A reveiw. Soil Science, vol. 165,
pp. 277-304.
Kleber, M., Johnson, M.G. (2010). Chapter three - Advances in understanding the
molecular structure of soil organic matter: Implications for interactions in the
environment. Advances in Agronomy, vol. 106, pp. 77-142.
Kordel, W., Dassenakis, M., Lintelmann, J., Padberg, S. (1997). The importance of
natural organic material for environmental processes in waters and soilsTechnical Report. Pure and Applied Chemistry, vol. 69, pp. 1571-1600.
Köhler, S.J., Buffam, I., Laudon, H., Bishop, K.H. (2008). Climate’s control of
intra-annual and interannual variability of total organic carbon concentration
and flux in two contrasting boreal landscape elements. Journal of
Geophysical Research, vol. 113.
Köhler, S.J., Buffam, I., Seibert, J., Bishop, K.H., Laudon, H. (2009) Dynamics of
stream water TOC concentrations in a boreal headwater catchment:
Controlling factors and implications for climate scenarios. Journal of
Hydrology, vol. 373, pp. 44-56.
Köhler, S.J., Kothawala, D., Futter, M.N., Liungman, O., Tranvik, L.J. (2013) In-Lake
Processes Offset Increased Terrestrial Inputs of Dissolved Organic Carbon
and Color to Lakes. Plos One, vol. 8.
Laudon, H., Köhler, S., Buffam, I. (2004). Seasonal TOC export from seven boreal
catchments in northern Sweden. Aquatic Sciences, vol. 66, pp. 223-230.
29
Laudon, H., Berggren, M., Ågren, A., Buffam, I., Bishop, K., Grabs, T., Jansson, M.,
Köhler, S. (2011). Patterns and Dynamics of Dissolved Organic Carbon (DOC)
in Boreal Streams: The Role of Processes, Connectivity, and Scaling.
Ecosystems, vol. 14, pp. 880-893.
Laudon, H. (2012). A Fieldguide to Krycklan Catchment Study. Swedish University
of Agricultural Sciences. Faculty of Forest Sciences. Department of Forest
Ecology and Management. Umeå. [Brochure]
Lavonen, E.E., Gonsior, M., Tranvik, L.J., Schmitt-Kopplin, P., Köhler, S.J. (2013).
Selective Chlorination of Natural Organic Matter: Identification of Previously
Unknown Disinfection Byproducts. Environmetal Science and Technology, vol.
47, pp. 2264–2271.
Ledesma, J.L.J. (2012). Riparian zone functioning in forested boreal landscapes
– Literature review. Swedish University of Agricultural Sciences. Faculty of
Natural Resources and Agricultural Sciences. Department of Aquatic Sciences
and Assessment. Uppsala. (PhD Literature review, 2010).
Ledesma, J.L.J., Grabs, T., Futter, M.N., Bishop, K.H., Laudon, H., Köhler, S.J.
(2013). Riparian zone control on base cation concentration in streams.
Biosciences, vol. 10, pp. 3846-3868.
Löfgren, S., Forsius, M., Andersen, T. (2003). The color of water – climate induced
water color increase in Nordic lakes and streams due to humus. Nordic council
of Ministers brochure, 12 pp. [Brochure].
Maehder, S. (2012). Organic carbon fluxes in a boreal catchment – spatial variability
of bioavailability and importance of the riparian zone. Swedish University of
Agricultural Sciences. Faculty of Natural Resources and Agricultural Sciences.
Department of Aquatic Sciences and Assessment . Uppsala. (Master’s thesis:
Independent Project in Environmental Science, 2012).
Maloney, K.O., Morris, D.P., Moses, C.O., Osburn, C.L. (2005). The role of iron and
dissolved organic carbon in the absorption of ultraviolet radiation in humic lake
water. Biogeochemistry, vol. 75, pp. 393-407.
McDowell, W.H. (1985). Kinetics and mechanisms of dissolved organic carbon
retention in a headwater stream. Biogeochemistry, vol. 1, pp. 329-352.
McKnight, D.M., Bencala, K.E., Zellweger, G.W., Aiken, G.R., Feder, G.L., Thorn,
K.A. (1992). Sorption of dissolved organic carbon by hydrous aluminium and
iron oxides occurring at the confluence of Deer Creek with the Snake river,
Summit country, Colorado. Environmental Science and Technology, vol. 26,
pp. 1388-1396.
Michalzik, B., Matzner, E. (1999). Dynamics of dissolved organic nitrogen and
carbon in a Central European Norway spruce ecosystem. European Journal of
Soil Science, vol. 50, pp. 579-590.
Nyberg, L., Stähli, M., Mellander, P.E., and Bishop, K.H. (2001). Soil frost effects on
soil water and runoff dynamics along a boreal forest transect: 1. Field
investigations. Hydrological Processes, vol. 15, pp. 909–926.
Perdue. E.M. (1998). Chemical Composition, Structure, and Metal Binding
Properties. Aquatic Humic Substances, vol. 133, pp. 41-61.
Rodhe, A. (1989). On the generation of stream runoff in till soils. Nordic Hydrology,
vol. 20, pp. 1–8.
Seibert, J., Grabs, T., Köhler, S., Laudon, H., Winterdahl, M., Bishop, K. (2009).
Linking soil- and stream-water chemistry based on a Riparian FlowConcentration Integration Model. Hydrology and Earth System Science, vol.
13, pp. 2287–2297.
30
Stähli, M., Nyberg, L., Mellander, P.-E., Jansson, P.E., and Bishop, K.H.
(2001). Soil frost effects on soil water and runoff dynamics along a boreal
transect: 2. Simulations. Hydrological Processes, vol. 15, pp. 927–941.
Weishaar, J.L., Aiken, G.R., Bergamaschi, B.A., Fram, M.S., Fujii, R., Mopper, K.
(2003). Evaluation of Specific Ultraviolet Absorbance as an Indicator of the
Chemical Composition and Reactivity of Dissolved Organic Carbon.
Environmental Science and Technology, vol. 37, pp. 4702-4708.
Winterdahl, M., Futter, M., Köhler, S., Laudon, H., Seibert, J., Bishop, K. (2011).
Riparian soil temperature modification of the relationship between flow and
dissolved organic carbon concentration in a boreal stream. Water Resources
Research, vol. 47.
Ågren, A., Buffam, I., Jansson, M., Laudon, H. (2007). Importance of seasonality and
small streams for the landscape regulation of dissolved organic carbon export.
Journal of Geophysical Research, vol. 112.
Ågren, A., Buffram, I., Berggren, M., Bishop, K., Jansson, M., Laudon, H. (2008).
Dissolved organic carbon characteristics in boreal streams in a forest-wetland
gradient during the transition between winter and summer. Journal of
Geophysical Research, vol. 113.
Ågren, A.M., Haei, M., Blomkvist, P., Nilsson, M.B., Laudon, H. (2012). Soil frost
enhances stream dissolved organic carbon concentrations during episodic
spring snow melt from boreal mires. Global Change Biology, vol. 18, pp. 18951903.
31
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Related manuals

Download PDF

advertisement