BGBiogeosciencesBGBiogeosciences1726-4189Copernicus PublicationsGöttingen, Germany10.5194/bg-13-4697-2016Fast-freezing with liquid nitrogen preserves bulk dissolved organic matter
concentrations, but not its compositionThiemeLisal.thieme@campus.tu-berlin.deGraeberDanielhttps://orcid.org/0000-0001-8331-9639KaupenjohannMartinSiemensJanChair of Soil Science, Department of Ecology, Technical University
of Berlin, Berlin, GermanyChair of Soil Resources, Institute of Soil Science and Soil
Conservation, iFZ Research Centre for Biosystems, Land Use and Nutrition,
Justus-Liebig University Giessen, Giessen, GermanyDepartment of Bioscience, Catchment Science and Environmental
Management, Aarhus University, Silkeborg, DenmarkLisa Thieme (l.thieme@campus.tu-berlin.de)22August201613164697470515March201622March201619July201620July2016This work is licensed under a Creative Commons Attribution 3.0 Unported License. To view a copy of this license, visit http://creativecommons.org/licenses/by/3.0/This article is available from https://bg.copernicus.org/articles/13/4697/2016/bg-13-4697-2016.htmlThe full text article is available as a PDF file from https://bg.copernicus.org/articles/13/4697/2016/bg-13-4697-2016.pdf
Freezing can affect concentrations and spectroscopic properties
of dissolved organic matter (DOM) in water samples. Nevertheless, water
samples are regularly frozen for sample preservation. In this study we tested
the effect of different freezing methods (standard freezing at
-18 ∘C and fast-freezing with liquid nitrogen) on DOM
concentrations measured as organic carbon (DOC) concentrations and on
spectroscopic properties of DOM from different terrestrial ecosystems (forest
and grassland). Fresh and differently frozen throughfall, stemflow, litter
leachate and soil solution samples were analyzed for DOC concentrations,
UV-vis absorption and fluorescence excitation–emission matrices combined with
parallel factor analysis (PARAFAC). Fast-freezing with liquid nitrogen
prevented a significant decrease of DOC concentrations observed after
freezing at -18 ∘C. Nonetheless, the share of PARAFAC components 1
(EXmax < 250 nm (340 nm), EXmax: 480 nm) and
2 (EXmax: 335 nm, EXmax: 408 nm) to total fluorescence
and the humification index (HIX) decreased after both freezing treatments,
while the shares of component 3 (EXmax: < 250 nm
(305 nm), EXmax: 438 nm) as well as SUVA254 increased. The
contribution of PARAFAC component 4 (EXmax: 280 nm,
EXmax: 328 nm) to total fluorescence was not affected by freezing.
We recommend fast-freezing with liquid nitrogen for preservation of bulk DOC
concentrations of samples from terrestrial sources, whereas immediate
measuring is preferable to preserve spectroscopic properties of DOM.
Introduction
In addition to dissolved organic carbon (DOC) concentrations, properties of
dissolved organic matter (DOM) are crucial for its role in biogeochemical
cycles of carbon and nutrients as well as for its effect on pollutant
dynamics (Bolan et al., 2011). Spectroscopic methods like UV-vis absorption
and fluorescence spectroscopy used as single excitation–emission scans,
synchronous scans and excitation–emission matrices (EEMs) in combination with
different indices and/or parallel factor analysis (PARAFAC) are increasingly
applied to characterize chromophoric dissolved organic matter (cDOM) in
various environments (e.g., Murphy et al., 2008; Yamashita et al., 2010;
Stedmon and Markager, 2005; Graeber et al., 2012; Otero et al., 2007,
Traversa et al., 2014; Kalbitz et al., 1999).
The applicability of optical methods for characterizing DOM and the
comparability of results in multidisciplinary studies relies on the
preservation of samples prior to their analysis. DOM properties depend on
many physicochemical and biological boundary conditions, so that artifacts
caused by sample storage or sample pre-treatment may be produced easily. For
these reasons it is recommended to directly filter samples after collection
and store them in the cold and dark prior to measurement as soon as possible
(Santos et al., 2010; Spencer and Coble, 2014). However, immediate
measurement is often not possible for practical reasons such as a large
number of samples or remote or separated sampling sites, so that freezing of
filtered DOM samples is often the selected storage method (Murphy et al.,
2008; Yamashita et al., 2010; Graeber et al., 2012). Freezing can affect the
physicochemical composition of samples (Edward and Cresser, 1992), so that
improved conservation techniques, which avoid or minimize potential
artifacts of freezing, are required. During the freezing process, DOM is
preferentially excluded from the ice phase and enriched in the remaining
liquid phase (Belzile et al., 2002; Xue et al., 2015). The increasing solute
concentrations and changing physical conditions in the remaining liquid
phase during the freezing process could promote conformational and
configurational changes of DOM molecules as well as particle and complex
formation depending on DOM composition and sample type (Zaritzky, 2006;
Edward and Cresser, 1992). One potential technique for minimizing these
effects could be fast freezing with liquid N2, by radically reducing
the freezing time.
Whereas studies on sample preservation of marine waters (Del Castillo and
Coble, 2000; Yamashita et al., 2010a; Conmy et al., 2009) showed only small
freezing effects on DOM fluorescence characteristics, research with a variety
of freshwater samples produced inconsistent results. Fellman et al. (2008)
measured DOC concentrations and UV absorption in fresh and frozen and/or thawed
Alaska stream water samples and reported a significant decrease of DOC
concentration and specific ultraviolet absorption at 254 nm (SUVA254).
They recommended freezing as an acceptable storage method for freshwater
samples with low DOC concentration and/or low SUVA254 values. In
contrast, Yamashita et al. (2010) observed only minor changes in absorption-based indices after freezing and thawing of Venezuela river water but
significant alterations (decrease and increase) for PARAFAC component
intensities. A freeze–thaw experiment with water samples from a large number
of UK locations conducted by Spencer et al. (2007) showed large and variable changes (decreasing and
increasing) in DOM fluorescence intensity and absorbance after freezing and
thawing. Likewise Peakock et al. (2015) found strong and inconsistent effects of freezing and thawing on
absorbance properties of cDOM in water from bog pools, fen ditches and lakes.
In a study of sample preservation on rainwater cDOM fluorescence, Santos et
al. (2010) found a decrease of protein-like fluorescence intensity due to
freezing.
While many studies investigated the influence of different soil sample
pre-treatments on DOC concentrations and DOM composition (e.g., Christ and
David, 1994; Sun et al., 2015) only few studies focused on the influence on
these properties when using different preservation methods for the extracted
soil solutions. Otero et al. (2007) conducted freeze–thaw experiments on
salt marsh pore water and found no changes in characteristics of synchronous
fluorescence scans.
The impact of sample preservation like freezing seems highly variable
depending on sample and DOM characteristics. While most studies focused on
samples from marine or freshwater ecosystems, there is a lack of information
on sample pre-treatment effects on cDOM properties of water samples from
terrestrial ecosystems, especially soil solution. Due to different sources of
DOM in land and water environments (Bolan et al., 2011) and therefore
different chemical characteristics, it is unlikely that insights regarding
the alterations of samples during storage can be transferred from one sample
type to another. To help close this
gap, we investigate in this study the influence of freezing and thawing on
DOC concentration, spectral absorption and fluorescence properties for a wide
range of water samples (throughfall, litter leachate and soil solution) from
different terrestrial ecosystems (grasslands and forests). We tested in how
far fast-freezing with liquid nitrogen might prevent concentration and
partitioning effects and minimize structural changes of DOM. We hypothesized
(i) that sample type affects freeze–thaw effects on DOC concentrations and
DOM properties, because of different physical and chemical DOM
characteristics and therefore different response to changing conditions
during freezing and (ii) that fast-freezing with liquid nitrogen reduces
these freeze–thaw effects, because it minimizes the freezing time and thus
prevents partitioning effects and their physical consequences.
Material and methodsStudy sites
The study was conducted on experimental plots in the Schorfheide Chorin
Exploratory of the German “Biodiversity Exploratories”, which were
established as platform for large-scale and long-term functional biodiversity
research (Fischer et al., 2010). The experimental plots are located in a
young glacial landscape in NE Germany with an annual mean temperature of 8 to
8.5 ∘C and an annual mean precipitation of 500 to 600 mm. The
forest plots are dominated either by pine (Pinus sylvestris L.) or
beech (Fagus sylvatica L.) on Cambisols (IUSS working group WRB,
2014). The grassland plots are meadows, pastures and mown pastures on
Histosols, Gleysols and Cambisols.
Sampling and sample preparation
For the experiments, we collected solution samples from five forest and three
grassland plots on 17 and 18 June 2014 within a bi-weekly 2-day sampling
routine of above and below-ground water samples in the DFG priority program
“Biodiversity Exploratories”. Together we collected 27 samples for the
freezing experiment including six throughfall (TF), five stemflow (SF), five
forest litter leachate (LL) as well as six top- and five subsoil solution
samples. Volume-weighted composite samples for the experiment were produced
from replicated samplers of the same type (e.g., throughfall collectors,
shallow suction cups) of one plot in “aged” 500 mL PE bottles. The bottles
were bi-weekly used in the field for the same samples, after washing in the
dishwasher and with deionized water. TF was sampled with funnel-type
collectors (diameter 0.12 m, polyethylene) 0.3 m above soil surface. We
pooled five replicates at grassland and 20 replicates arranged in two lines
of 10 samplers in a cross-shaped
form at forest sites. To minimize alterations of the sample and contamination
such as evaporation, photo chemical reactions and algae growth, the sampling
bottles were wrapped with aluminium foil and closed with a 1.6 mm polyester
mesh and a table-tennis ball. SF was sampled with sliced polyurethane hoses
(diameter: 0.04 m) as a collar sealed with a polyurethane-based glue to the
bark of three trees per site at approximately 1.5 m height and connected
with a polypropylene (PP) or polyethylene (PE) barrel via a PE tube. LL was
collected with three zero-tension lysimeters per site (280 cm2 sampling
area) consisting of polyvinyl chloride plates covered with a PE net (mesh
width 0.5 mm) connected with PE hoses to 2 L PE bottles stored in a box
below ground. We sampled soil solution with nylon membrane
(0.45 µm) suction cups (ecoTech, Germany). Three samplers were
installed beneath the A horizon (Top) at approximately 10 cm depth. Another
three were installed in the B horizon (Sub) in approximately 50 cm depth in
the forest plots and 60 or 70 cm depth in the grassland sites. Suction cups
were connected to 2 L PE bottles in an insulated aluminium box placed into a
soil pit. Soil water was extracted by applying a vacuum of 50 kPa to the PE
bottles with an electric pump after each sampling.
After mixing, the samples were transported on ice to the laboratory and
stored overnight at 5 ∘C. We measured pH (Knick, Germany) and
electrical conductivity (WTW, Germany) in all samples prior to filtration
through ∼ 0.7 µm glass microfiber filters (Whatman GF/F). The
filters were washed with 100 mL deionized water and 10 mL of sample before
sample filtration. The filtered sample was split in three aliquots for
different preservation treatments: (i) no preservation (fresh) for which
samples were stored at 5 ∘C in the dark and DOC concentrations were
measured 24 h after sampling while fluorescence as well as absorbance were
measured within 48 h; (ii) preservation by freezing for which the samples
were stored at -18 ∘C for 4 weeks, and (iii) fast-freezing with
liquid nitrogen (N2), for which 12 mL sample aliquots were filled in
pre-rinsed 15 mL (5 mL sample) PP falcon tubes, dipped in liquid nitrogen
for 30 s and then stored at -18 ∘C for 42 days. Fresh samples and
samples frozen at -18 ∘C were stored in 20 mL PE scintillation
vials (NeoLab) that were pre-rinsed with 5 mL sample before filling.
Fluorescence, absorbance and DOC concentration from all frozen samples were
measured after defrosting over night at 5 ∘C in the dark. For all
preparation steps and treatments control samples of ultrapure water (EVOQUA,
Germany) were analyzed, showing no release of DOM (DOC concentration and DOM
fluorescence) from laboratory equipment.
Laboratory analysis
We measured the concentration of DOC as non-purgeable organic carbon on a
Shimadzu TOC-5050A (Duisburg, Germany) with a limit of quantification of
2 mg C L-1. Absorption spectra of DOM were scanned at wavelength from
400 to 600 nm using a Lambda 20 UV-vis spectrometer (Perkin Elmer, USA) and
a 1 cm quartz cuvette. Absorbance measurements were baseline corrected using
ultrapure water. All fluorescence EEMs were measured on a Hitachi F-4500
fluorescence spectrometer (Hitachi, Japan) directly after absorption
measurement in the same cuvette. We measured excitation from 240 to 450 nm
(5 nm steps) and emission from 300 to 600 nm (2 nm steps) with a slit
width of 5 nm and scan speed 12000 nm min-1. We corrected our EEMs
according to the protocol from Murphy (2010) with the fdomcorrect function in
the drEEM toolbox (version 2.0) of Murphy et al. (2013) using Matlab (Version
Matlab2011b, The MathWorks Inc.). We used the correction curves supplied by the
manufacturer for the excitation and
emission correction factors. We measured ultrapure water fluorescence spectra
for blank correction and to convert EEMs to Raman units by normalizing them
to the area under the Raman peak at 350 nm excitation wavelength (Lawaetz
and Stedmon, 2009). In order to apply the inner-filter correction of
Lakowicz (2006) integrated in the drEEM toolbox, all aliquots were diluted
with ultrapure water to achieve an absorption of < 0.3 at 254 nm
(Ohno, 2002). For this reason, not all treatments of one sample were diluted
with the same dilution factor. To test the possible influence of different
dilutions on the pH-related changes in fluorescence (Patel-Sorrentino et al.,
2002; Baker et al., 2007), dilution series with samples (n=14) from the
same plots and same sample types but with different sampling dates were
measured for pH, absorption and fluorescence according to the protocol
described above. We compared the differences of 31 dilutions and calculated
the mean absolute deviation (MAD). These were compared to the MAD of
measurement precision, determined by analyzing 11 samples in three
replications. For the PARAFAC components %C1, %C2 and %C3 and
SUVA254 the MAD caused by dilution were less or equal than the precision
MAD, so that there was no influence of dilution on the three humic-like
components and the specific UV absorbance at 254 nm. For %C4 and HIX the
effect of dilution could exceed the precision of fluorescence measurements.
For detailed information see supporting information.
Spectroscopic indices and PARAFAC modeling
Based on the absorbance spectra, we calculated specific ultraviolet
absorbance (SUVA254) as the absorbance at 254 nm divided by the DOC
concentration. The SUVA254 is reported in L mg-1 m-1, and
is associated with bulk aromaticity (Weishaar et al., 2003). Moreover, we
calculated the humification index (HIX) from fluorescence EEMs (Ohno, 2002).
The HIX ranges from 0 to 1 and allows characterizing samples based on their
degree of DOM humification.
In addition to the calculation of indices, we used parallel factor analysis
(PARAFAC) to mathematically decompose the trilinear data of the EEMs into
fluorescence components of DOM (Stedmon et al., 2003). Further pre-processing
steps of EEMs (smoothing of Rayleigh and Raman scatter and sample
normalization), as well as the PARAFAC analysis, were conducted with the
drEEM toolbox (version 2.0, Murphy et al., 2013). We chose a four component
PARAFAC model (components referred as C1 to C4), visually checked the
randomness of residuals and the component spectral loadings, split-half
validated the model and generated the best fit by random initialization. For
comparison in statistical analysis we used the relative percentage
distribution of the four PARAFAC components (% of the sum of total peak
fluorescence of all PARAFAC components), so that percentage values for the
components will be given as %C1 to %C4.
Statistical analysis
The DOM composition variables used for statistical analysis were the PARAFAC
components %C1 to %C4, the spectroscopic indices HIX and SUVA254,
as well as the DOC concentration. For all statistical analysis the variables
were scaled and centered. We conducted a pair-wise (samples as strata)
permutational multivariate analysis of variance (PERMANOVA) with DOC
concentrations of the fresh samples as factor based on Euclidean distances in
R (Oksanen et al., 2015; R core team, 2015). The adonis function was used to
assess the influence of sample preparation (fresh, frozen, fast-freezing) and
of the initial DOC concentration on DOM variables. To investigate
preservation effects on single variables we conducted linear mixed-effect
models (sometimes called multi-level models, lme function, Linear and
Nonlinear Mixed Effects Models package for R, Pinheiro et al., 2015) with
samples as random intercept on each of the DOM composition variables. These
were used instead of simple linear models or ANOVAs, since we could not
expect the same intercept for all samples due to different sample
concentrations. To test the influence of the initial DOC concentration on
single preservation treatments we performed Spearman Rank Order Correlation.
To assess the influence of sample type (TF, SF, LL, Top or Sub) on the
relative change of DOM composition due to fast-freezing with liquid nitrogen
or freezing at -18 ∘C in relation to the measurement of fresh,
cooled samples, we used an ANOVA with the sample type as fixed factor (aov
function in R). To remove sample concentration-related effects and to
calculate relative changes, the differences between the two preservations
(either fast-freezing or freezing at -18 ∘C) relative to the
measurements of fresh samples were calculated for each sample before the
ANOVA. This was only done for variables, for which we found strong,
significant effects with the linear mixed-effect models.
ResultsDOM concentrations
The samples covered a wide range of DOC concentrations (Fig. 1a, b). Fresh TF
samples showed the lowest concentrations ranging from 5 to 17 mg C L-1,
SF samples had the highest DOC concentrations ranging from 12 to
138 mg C L-1 (Fig. 1b). High concentrations up to 75 mg C L-1
were also found for LL samples, but average values were smaller than for SF
(Fig. 1b). In the mineral soil, concentrations decreased from 13 to
124 mg C L-1 in topsoil samples to 9 to 47 mg C L-1 in subsoil
samples.
Absolute DOC concentrations (measured in fresh samples) and changes
in DOC concentrations after freezing (-18 ∘C) and fast-freezing
with liquid nitrogen; (a, c, e) all samples (n=27); (b, d, f) ordered by sample type (throughfall (TF) n=6, stemflow (SF) n=5,
litter leachate (LL) n=5, top soil solution (Top) n=6, sub-soil solution
(Sub) n=5); gray dashed line: analytical reproducibility; ***
significant changes (linear mixed models (lme), p< 0.05);
boxplots: solid line: median, dashed line: mean.
We found a significant treatment effect (linear mixed-effect models (lme),
p< 0.05) on DOC concentration when comparing the fresh and frozen
samples (Fig. 1c). In 24 of 27 samples DOC concentrations decreased after
freezing at -18 ∘C and subsequent thawing, with an average change
of -1.6 mg C L-1 or -6 % respectively. The maximum decrease
that was found equalled -6 mg C L-1 and -25 %, respectively. In
contrast to freezing at -18 ∘C, fast-freezing with liquid nitrogen
did not result in significant changes (lme, p>0.05) of DOC
concentrations (Fig. 1c). This different behavior between normal freezing
and fast-freezing was also found for the influence of the initial DOC
concentration on changes of DOM properties. Only the -18 ∘C
treatment showed a significant correlation (Spearmans rank r=-0.447,
p= 0.0194), indicating a larger decrease of DOC concentrations due to
freezing for samples with higher initial DOC concentrations.
PARAFAC fluorescence components
The analysis of fluorescence spectra using PARAFAC resulted in four
components that were characterized according to the review of Fellman et
al. (2010) (Table 1). C1 exhibited its main excitation maximum at
< 250 nm, a secondary maximum at 340 nm and an emission maximum at
480 nm and was described as UVA humic-like fluorophore with a terrestrial
source and a high molecular weight (Murphy et al., 2006; Stedmon et al.,
2003; Shutova et al., 2014; Fellman et al., 2010). C2 had a maximum
excitation at 335 nm and an emission maximum at 408 nm and was named also
UVA humic-like, but associated with low molecular weight (Murphy et al.,
2006; Fellman et al., 2010; Stedmon et al., 2003). C3 was defined by an
excitation maximum at < 250 nm, a secondary maximum at 305 nm and
an emission maximum at 438 nm. This component dominated fulvic acid
fractions of humic substances (Santín et al., 2009; He et al., 2006).
Finally, C4 was characterized by its excitation maximum at 280 nm and an
emission maximum at 328 nm and was classified as tryptophan-like, as its
fluorescence resembles free tryptophan. Therefore, this component was
associated with free or bound proteins (Fellman et al., 2010).
Characteristics of PARAFAC components based on Fellman et al. (2010).
We found different distributions of PARAFAC components for different sample
types (Fig. 2). The contribution of %C1 to the total fluorescence
increased from TF over SF to LL and then decreased again from LL to Sub
(Fig. 2), while %C2 showed just the opposite trend. In contrast, %C3
tended to increase from TF to Sub, whereas %C4 showed a decreasing trend
(Fig. 2).
Mean distribution of PARAFAC components %C1–%C4 for
different sample types.
The conducted PERMANOVA was highly significant (p< 0.001),
indicating that the preservation significantly affects the DOM composition.
The interaction between treatment and initial DOC concentration of the fresh
treatment explains a reasonable part of the variance (R2=0.14) and is
highly significant (p< 0.001). Therefore the original DOC
concentration of the fresh sample well explains the variable strength of the
treatment effect.
Similar changes in component distribution were found as a consequence of
freezing at -18 ∘C and fast-freezing with liquid nitrogen
(Fig. 3). We observed a significant (lme, p< 0.05) decrease in all
samples for the relative fraction of the humic-like components %C1 and
%C2 after freezing at -18 ∘C and fast-freezing compared to the
fresh control samples (Fig. 3a, b). The contribution of %C1 to the total
fluorescence decreased on average by -3 % with maximum changes of
-5 % for freezing at -18 ∘C and -6 % for fast-freezing
with liquid nitrogen. The average decrease of %C2 was -3 % and the
maximum -8 % for both treatments.
Changes of relative distribution of PARAFAC components after
freezing (-18 ∘C) and fast-freezing with liquid nitrogen;
(a, c, e, g) all samples (n=27); (b, d, f, h) ordered by
sample type (throughfall (TF) n=6, stemflow (SF) n=5, litter leachate
(LL) n=5, top soil solution (Top) n=6, sub-soil solution (Sub) n=5);
gray dashed line: analytical reproducibility; *** significant changes
(linear mixed models (lme), p< 0.05) ; boxplots: solid line:
median, dashed line: mean.
In contrast to %C1 and %C2, the share of %C3 to the total
fluorescence intensity increased upon freezing (Fig. 3e, f). All samples
frozen at -18 ∘C showed an increase in the relative intensity of
the %C3 signal, with an average increase of +6 % for both
treatments. The maximum increase was 10 % (freezing at -18 ∘C)
and 12 % (freezing with liquid N2). No significant effects of sample
preservation (lme, p>0.05) were found for %C4, the
protein-like component (Fig. 3g, h).
Aromaticity and humification index
We found SUVA254-values ranging from 1.1 L mg-1 m-1 up to
4.5 L mg-1 m-1 for fresh samples (Fig. 4a, b). Samples frozen at
-18 ∘C and fast-frozen samples showed a significant increase (lme,
p< 0.05) of their SUVA254 (Fig. 4c). The average change was
+0.4 L mg-1 m-1 equivalent to +20 % for samples frozen
at -18 ∘C and +0.5 L mg-1 m-1 equivalent to
+24% for samples that were fast-frozen with liquid nitrogen.
Absolute values (measured in fresh samples) and changes of SUVA254
after freezing (-18 ∘C) and fast-freezing with liquid nitrogen; (a, c, e) all samples (n=27); (b, d, f) ordered by sample type (throughfall
(TF) n=6, stemflow (SF) n=5, litter leachate (LL) n=5, top soil
solution (Top) n=6, sub-soil solution (Sub) n=5); gray dashed line:
analytical reproducibility; *** significant changes (linear mixed models
(lme), p< 0.05); boxplots: solid line: median, dashed line: mean.
The humification index of the freshly measured samples ranged from 0.806 to
0.931 in TF and SF samples and from 0.849 to 0.975 for Sub, Top and LL
samples (Fig. 5a, b). We found a significant decrease (lme,
p< 0.05) of the HIX when comparing the freshly measured samples
with the frozen and the fast-frozen samples (Fig. 5c). The average change was
-0.016 or -2 % for samples frozen at -18 ∘C and -0.020
or -2 % for samples fast-frozen with liquid nitrogen. The maximum
decrease was -0.128 or -15 % for -18 ∘C samples and
-0.076 or -8 % for liquid nitrogen samples (Fig. 5 c, d, e, f).
Absolute values (measured in fresh samples) and changes of HIX
after freezing (-18 ∘C) and fast-freezing with liquid nitrogen; (a, c, e) all samples (n=27); (b, d, f) ordered by sample type (throughfall
(TF) n=6, stemflow (SF) n=5, litter leachate (LL) n=5, top soil
solution (Top) n=6, sub-soil solution (Sub) n=5); gray dashed line:
analytical reproducibility; *** significant changes (linear mixed models
(lme), p< 0.05); boxplots: solid line: median, dashed line: mean.
Discussion
We found that freezing at -18 ∘C significantly reduced DOC
concentrations across all sample types and that the effect is higher with
higher initial DOC concentrations. This is in line with results of Fellman et
al. (2008) investigating the effect of freezing and thawing on Alaskan stream
water samples. This loss of DOC concentration might be due to aggregation and
irreversible particle formation (Giesy and Briese, 1978) induced by
partitioning and concentration effects during the freezing process (Belzile
et al., 2002; Xue et al., 2015). Indeed, our results indicated that
fast-freezing with liquid nitrogen can prevent significant reductions of bulk
DOC for samples with a large range of DOM concentrations. In contrast to
effects on DOC concentrations, we found similar significant effects of
fast-freezing as well as freezing at -18 ∘C on the chromophoric
humic fraction of DOM (PARAFAC components, HIX and SUVA254). The
increase of aromaticity as indicated by higher SUVA254 values indicates
a stronger removal of non-aromatic DOM during freezing and thawing. On the
other hand, the decrease in the HIX suggests a preferential removal of
humified cDOM. One potential explanation for the fact that fast-freezing in
liquid nitrogen resulted in significant changes of DOM fluorescence
properties, but only small changes of bulk DOC concentrations, is that cDOM
reacted stronger to freezing and thawing than the remaining DOM so that
spectroscopic properties were affected, but bulk DOC concentrations were not.
Fast freezing may have failed to prevent changes of cDOM composition because
(i) cDOM changes occurred not only during the freezing process (-18 or
-196 ∘C in liquid nitrogen), but also in frozen state at
-18 ∘C in the freezer during storage or (ii) cDOM was affected by
the thawing process that was identical for both freezing treatments. The
former might be supported by a re-crystallization of ice crystals in frozen
state (Luyet, 1967; Meryman, 2007).
No significant changes of protein-like fluorescence (%C4) due to freezing
and thawing were observed. This is in contrast to the results of Spencer et
al. (2007) and Santos et al. (2010), which could be related to similar
fluorescence characteristics, but different chemical composition of
proteinaceous fluorescence material from aquatic sources and the solutions
from terrestrial ecosystems tested in this study.
In our experiment we used relative small sample volumes (fresh,
-18 ∘C: 20 mL, N2: 12 mL) because we commonly keep the
volume that is stored frozen as small as possible due to space limitations in
deep freezers. We think that increasing the volume of samples that are
subjected to freezing also increases the risk of artifacts, because of
increasing concentration effects due to extended freezing time.
Conclusions
Freezing and thawing affected the DOC concentration, spectral absorption and
fluorescence properties of water samples (throughfall, litter leachate and
soil solution) from different terrestrial ecosystems (grasslands and
forests). In contrast, fast-freezing with liquid nitrogen minimized the
changes of bulk DOC concentrations but not the changes of spectroscopic cDOM
properties. Different thawing protocols for minimizing sample storage
effects on DOM should be tested in future studies. We suggest the use of
fast-freezing for preservation of bulk DOC concentrations, especially for
highly concentrated samples, when the increased effort and cost of using
liquid nitrogen in the field is justified by advantages regarding the
minimization of freezing artefacts. To preserve cDOM characteristics of
samples from terrestrial sources normal freezing or fast-freezing should be
avoided. Instead, filtration, cooling and measurements soon after the
sampling should be the method of choice, if possible.
Information about the Supplement
The data related to statistical analysis and for generating the figures are available in the
Supplement.
The Supplement related to this article is available online at doi:10.5194/bg-13-4697-2016-supplement.
Lisa Thieme, Martin Kaupenjohann, and Jan Siemens designed the experiment, Lisa Thieme performed
the experiments. All authors analyzed the data and wrote the manuscript.
Acknowledgements
We thank the managers of the three Exploratories, Kirsten Reichel-Jung, Swen Renner, Katrin Hartwich, Sonja Gockel, Kerstin Wiesner, and Martin Gorke for
their work in maintaining the plot and project infrastructure; Christiane
Fischer and Simone Pfeiffer for giving support through the central office,
Michael Owonibi for managing the central data base, and Markus Fischer,
Eduard Linsenmair, Dominik Hessenmöller, Jens Nieschulze, Daniel Prati,
Ingo Schöning, François Buscot, Ernst-Detlef Schulze, Wolfgang W. Weisser and the late Elisabeth Kalko for their role in setting up the
Biodiversity Exploratories project.
The work has been (partly) funded by the DFG Priority Program 1374
“Infrastructure-Biodiversity-Exploratories” (SI 1106/4-1,2). D. Graeber was
supported by a grant from the Danish Centre for Environment and Energy,
Aarhus University.
Field work permits were issued by the responsible state environmental
offices of Baden-Württemberg, Thüringen, and Brandenburg (according
to § 72 BbgNatSchG). We thank Sabine Dumke and Robert Jonov for
sample measurement. And we thank three anonymous referees whose comments helped to improve this manuscript.
Edited by: T. Treude
Reviewed by: three anonymous referees
References
Baker, A., Elliott, S., and Lead, J. R.: Effects of filtration and pH
perturbation on freshwater organic matter fluorescence, Chemosphere, 67,
2035–2043, 2007.
Belzile, C., Gibson, J. A. E., and Vincent, W. F.: Colored dissolved organic
matter and dissolved organic carbon exclusion from lake ice: Implications
for irradiance transmission and carbon cycling, Limnol. Oceanogr., 47,
1283–1293, 2002.
Bolan, N. S., Adriano, D. C., Kunhikrishnan, A., James, T., McDowell, R.,
and Senesi, N.: Dissolved Organic Matter: Biogeochemistry, Dynamics, and
Environmental Significance in Soils, Adv. Agron., 110, 1–75, 2011.
Christ, M. and David, M. B.: Fractionation of dissolved organic carbon in
soil water: Effects of extraction and storage methods, Commun.
Soil Sci. Plant, 25, 3305–3319, 1994.Conmy, R. N., Coble, P. G., Cannizzaro, J. P., and Heil, C. A.: Influence of
extreme storm events on West Florida Shelf CDOM distributions, J. Geophys.
Res.-Biogeo., 114, 1–17, 10.1029/2009JG000981, 2009.
Del Castillo, C. E. and Coble, P. G.: Seasonal variability of the colored
dissolved organic matter during the 1994–95 5NE6 and 5SW6 Monsoons in the
Arabian Sea, Deep-Sea Res. Pt. II, 47, 1563–1579, 2000.
Edwards, A. C. and Cresser, M. S.: Freezing and Its Effect on Chemical and
Biological Properties of Soil: Advances in Soil Science, edited by: Stewart, B. A., Springer New York, New York, NY, 59–79, 1992.
Fellman, J. B., D'Amore, D. V., and Hood, E.: An evaluation of freezing as a
preservation technique for analyzing dissolved organic C, N and P in surface
water samples, Sci. Total Enciron., 392, 305–312,
2008.
Fellman, J. B., Hood, E., and Spencer, R. G. M.: Fluorescence spectroscopy
opens new windows into dissolved organic matter dynamics in freshwater
ecosystems: A review, Limnol. Oceangr., 55, 2452–2462,
2010.
Fischer, M., Bossdorf, O., Gockel, S., Hänsel, F., Hemp, A.,
Hessenmöller, D., Korte, G., Nieschulze, J., Pfeiffer, S., Prati, D.,
Renner, S., Schöning, I., Schumacher, U., Wells, K., Buscot, F., Kalko,
E. K. V., Linsenmair, K. E., Schulze, E.-D., and Weisser, W. W.:
Implementing large-scale and long-term functional biodiversity research: The
Biodiversity Exploratories, Basic Appl. Ecol., 11, 473–485,
2010.
Giesy, J. P. and Briese, L. A.: Particulate formation due to freezing humic
waters, Water Resour. Res., 14, 542–544, 1978.
Graeber, D., Gelbrecht, J., Pusch, M. T., Anlanger, C., and Schiller, D.
von: Agriculture has changed the amount and composition of dissolved organic
matter in Central European headwater streams, Sci. Total Environ., 438,
435–446, 2012.He, Z., Ohno, T., Cade-Menun, B. J., Erich, M. S., and Honeycutt, C. W.:
Spectral and Chemical Characterization of Phosphates Associated with Humic
Substances, Soil Sci. Soc. Am. J., 70, 1741, 10.2136/sssaj2006.0030,
2006.
IUSS working group WRB: World reference base for soil resources 2014:
International soil classification system for naming soils and creating
legends for soil maps, World Soil Resources Reports No. 106, Rome, 2014.
Kalbitz, K., Geyer, W., and Geyer, S.: Spectroscopic properties of dissolved
humic substances – a reflection of land use history in a fen area,
Biogeochemistry, 47, 219–238, 1999.Lakowicz, J. R.: Principles of Fluorescence Spectroscopy, 3rd Edn.,
Springer Science + Business Media, 2006.
Lawaetz, A. J. and Stedmon C. A.: Fluorescence Intensity Calibration Using
the Raman Scatter Peak of Water, Appl. Spectrosc., 63, 936–940, 2009.
Luyet, B.: Various Modes of Recrystallization of Ice, Physics of Snow and
Ice proceedings, 1, 51–70, 1967.
Meryman, H. T.: Cryopreservation of living cells: principles and practice,
Transfusion, 47, 935–945, 2007.
Murphy, K. R., Ruiz, G. M., Dunsmuir, William T. M., and Waite, T. D.:
Optimized Parameters for Fluorescence-Based Verification of Ballast Water
Exchange by Ships, Environ. Sci. Technol., 40, 2357–2362,
2006.
Murphy, K. R., Stedmon, C. A., Waite, T. D., and Ruiz, G. M.: Distinguishing
between terrestrial and autochthonous organic matter sources in marine
environments using fluorescence spectroscopy, Mar. Chem., 108, 40–58,
2008.
Murphy, K. R., Butler, K. D., Spencer, R. G. M., and Stedmon, C. A., et al.:
Measurement of Dissolved Organic Matter Fluorescence in Aquatic
Environments: An Interlaboratory Comparison, Environ. Sci. Technol., 44,
9405–9412, 2010.Murphy, K. R., Stedmon, C. A., Graeber, D., and Bro, R.: Fluorescence
spectroscopy and multi-way techniques. PARAFAC, Anal. Methods, 5, 6557,
10.1039/c3ay41160e, 2013.
Ohno, T.: Fluorescence Inner-Filtering Correction for Determining the
Humification Index of Dissolved Organic Matter, Environ. Sci. Technol., 36,
742–746, 2002.Oksanen, J., Blanchet, F. G., Kindt, R., Legendre, P., Minchin, P. R., O `
Hara, R. B., Simpson, G. L., Solymos, P., Stevens, M. H. H., and Wagner, H.:
vegan: Community Ecology Package, R package version 2.2-1:
http://CRAN.R-project.org/package=vegan, last access: 12 November 2015.
Otero, M., Mendonça, A., Válega, M., Santos, E. B. H., Pereira, E.,
Esteves, V., and Duarte, A.: Fluorescence and DOC contents of estuarine pore
waters from colonized and non-colonized sediments: Effects of sampling
preservation, Chemosphere, 67, 211–220,
2007.
Patel-Sorrentino, N., Mounier, S., and Benaim, J. Y.: Excitation–emission
fluorescence matrix to study pH influence on organic matter fluorescence in
the Amazon basin rivers, Water Res., 36, 2571–2581, 2002.
Peacock, M., Freeman, C., Gauci, V., Lebron, I., and Evans, C. D.:
Investigations of freezing and cold storage for the analysis of peatland
dissolved organic carbon (DOC) and absorbance properties, Environm. Sci., 17,
1290–1301, 2015.Pinheiro, J., Bates, D., DebRoy, S., Sarkar, D., and R core team:
(nlme): Linear and Nonlinear Mixed Effects
Models, R package version 3.1-120: http://CRAN.R-project.org/package=nlme,
last access: 12 November 2015.
R core team: R: A language and environment for, R Foundation for Statistical
Computing, Vienna, Austria, 2015.
Santín, C., Yamashita, Y., Otero, X. L., Álvarez, M., and Jaffé,
R.: Characterizing humic substances from estuarine soils and sediments by
excitation-emission matrix spectroscopy and parallel factor analysis,
Biogeochemistry, 96, 131–147, 2009.
Santos, P., Otero, M., Santos, E., and Duarte, A. C.: Molecular fluorescence
analysis of rainwater: Effects of sample preservation, Talanta, 82,
1616–1621, 2010.
Shutova, Y., Baker, A., Bridgeman, J., and Henderson, R. K.: Spectroscopic
characterisation of dissolved organic matter changes in drinking water
treatment: From 5PARAFAC6 analysis to online monitoring wavelengths, Water
Res., 54, 159–169, 2014.
Spencer, R. G., Bolton, L., and Baker, A.: Freeze/thaw and pH effects on
freshwater dissolved organic matter fluorescence and absorbance properties
from a number of UK locations, Water Res., 41, 2941–2950,
2007.
Spencer, R. G. and Coble, P. G.: Sampling Design for Organic Matter
Fluorescence Analysis, in: Aquatic Organic Matter Fluorescence, 1st
Edn., edited by: Coble, P. G., Lead, J. R., Baker, A., Reynolds, D. M., and Spencer, R.
G., Cambridge environmental chemistry series, Cambridge University
Perss, 125–146, 2014.
Stedmon, C. A. and Markager, S.: Tracing the production and degradation of
autochthonous fractions of dissolved organic matter using fluorescence
analysis, Limnol. Oceangr., 50, 1415–1426,
2005.
Stedmon, C. A., Markager, S., and Bro, R.: Tracing dissolved organic matter
in aquatic environments using a new approach to fluorescence spectroscopy,
Mar. Chem., 82, 239–254, 2003.Sun, S.-Q., Cai, H.-Y., Chang, S. X., and Bhatti, J. S.: Sample
storage-induced changes in the quantity and quality of soil labile organic
carbon, Scientific Reports, 5, 17496, 10.1038/srep17496, 2015.
Traversa, A., D'Orazio, V., Mezzapesa, G. N., Bonifacio, E., Farrag, K.,
Senesi, N., and Brunetti, G.: Chemical and spectroscopic characteristics of
humic acids and dissolved organic matter along two Alfisol profiles,
Chemosphere, 111, 184–194, 2014.
Weishaar, J. L., Aiken, G. R., Bergamaschi, B. A., Fram, M. S., Fujii, R.,
and Mopper, K.: Evaluation of Specific Ultraviolet Absorbance as an Indicator
of the Chemical Composition and Reactivity of Dissolved Organic Carbon,
Environ. Sci. Technol., 37, 4702–4708, 2003.
Xue, S., Wen, Y., Hui, X., Zhang, L., Zhang Z., Wang, J., and Zhang, Y..: The
migration and transformation of dissolved organic matter during the freezing
processes of water, J. Environ. Sci., 27, 168–178,
2015.
Yamashita, Y., Cory, R. M., Nishioka, J., Kuma, K., Tanoue, E., and
Jaffé, R.: Fluorescence characteristics of dissolved organic matter in
the deep waters of the Okhotsk Sea and the northwestern North Pacific Ocean
Deep-Sea Res. Pt. II, 57, 1478–1485, 2010.
Zaritzky, N.: Physical-Chemical Principles in Freezing, in: Handbook of
frozen food processing and packaging, edited by: Sun, D.-W., CRC Press, 3–32,
2006.