BGBiogeosciencesBGBiogeosciences1726-4189Copernicus PublicationsGöttingen, Germany10.5194/bg-15-821-2018Wet–dry cycles impact DOM retention in subsurface soilsOlshanskyYanivyanivo@email.arizona.eduRootRobert A.https://orcid.org/0000-0003-2094-807XChoroverJonDepartment of Soil, Water and Environmental Science, University of Arizona, Tucson 85721, USAYaniv Olshansky (yanivo@email.arizona.edu)9February201815382183223June201724August201714December201715December2017This work is licensed under the Creative Commons Attribution 4.0 International License. To view a copy of this licence, visit https://creativecommons.org/licenses/by/4.0/This article is available from https://bg.copernicus.org/articles/15/821/2018/bg-15-821-2018.htmlThe full text article is available as a PDF file from https://bg.copernicus.org/articles/15/821/2018/bg-15-821-2018.pdf
Transport and reactivity of carbon in the critical zone are highly controlled
by reactions of dissolved organic matter (DOM) with subsurface soils,
including adsorption, transformation and exchange. These reactions are
dependent on frequent wet–dry cycles common to the unsaturated zone,
particularly in semi-arid regions. To test for an effect of wet–dry cycles on
DOM interaction and stabilization in subsoils, samples were collected from
subsurface (Bw) horizons of an Entisol and an Alfisol from the Catalina-Jemez
Critical Zone Observatory and sequentially reacted (four batch steps) with
DOM extracted from the corresponding soil litter layers. Between each
reaction step, soils either were allowed to air dry (“wet–dry” treatment)
before introduction of the following DOM solution or were maintained under
constant wetness (“continually wet” treatment). Microbial degradation was
the dominant mechanism of DOM loss from solution for the Entisol subsoil,
which had higher initial organic C content, whereas sorptive retention
predominated in the lower C Alfisol subsoil. For a given soil, bulk dissolved
organic C losses from solution were similar across treatments. However, a
combination of Fourier transform infrared (FTIR) and near-edge X-ray
absorption fine structure (NEXAFS) spectroscopic analyses revealed that
wet–dry treatments enhanced the interactions between carboxyl functional
groups and soil particle surfaces. Scanning transmission X-ray microscopy
(STXM) data suggested that cation bridging by Ca2+ was the primary
mechanism for carboxyl association with soil surfaces. STXM data also showed
that spatial fractionation of adsorbed OM on soil organo-mineral surfaces was
diminished relative to what might be inferred from previously published
observations pertaining to DOM fractionation on reaction with specimen
mineral phases. This study provides direct evidence of the role of wet–dry
cycles in affecting sorption reactions of DOM to a complex soil matrix. In
the soil environment, where wet–dry cycles occur at different frequencies
from site to site and along the soil profile, different interactions between
DOM and soil surfaces are expected and need to be considered for the overall
assessment of carbon dynamics.
Introduction
Dissolved organic matter (DOM) is the main vehicle of organic carbon (OC)
and nutrient transport to the subsoil (Kaiser and Kalbitz, 2012; Kalbitz et
al., 2000). There it stimulates key biogeochemical processes including
heterotrophic microbial activity (Fontaine et al., 2007), mineral
transformation and organic and inorganic nutrient and contaminant
mobilization (Chorover et al., 2007; Polubesova and Chefetz, 2014; Zhao et
al., 2011). Interactions with subsoil surfaces act to stabilize DOM against
advective transport and microbial degradation (Eusterhues et al., 2014;
Kalbitz et al., 2000; Lutzow et al., 2006). Furthermore, prior studies have
shown that DOM generated in the surface litter layers can be transported
preferentially to clay-enriched subsoils via macropore flow paths that
bypass the intervening matrix (Rumpel and Kögel-Knabner, 2010).
Particularly in semi-arid vadose zones, these DOM–subsoil interactions occur
in a context of frequent wet–dry cycles. Although such cyclic conditions
likely impact C dynamics, the nature of their effects on micro- to
molecular-scale organo-mineral associations remains poorly known.
The principal chemical mechanisms affecting DOM retention at soil particle
surfaces, including ligand exchange with surface hydroxyl groups,
ion exchange of organic moieties at charged sites, cation bridging, hydrogen
bonding and van der Waals interactions, depend on both DOM molecular
composition and mineral surface chemistry (Chorover and Amistadi, 2001; Gu
et al., 1994; Kleber et al., 2007, 2014). Interactions of DOM with dissolved
polyvalent cations (e.g., Fe3+ and Al3+) may also result in its
coagulation and co-precipitation with nucleating metal (oxy)hydroxides (Chen
et al., 2014a; Eusterhues et al., 2011). Drying of OM–mineral complexes can
affect the mode of interaction. These effects may include changing of
adsorption mode and product surface chemistry. For example, drying can
convert OM adsorbate from outer- to inner-sphere coordination (Kang et al.,
2008), promote exposure of hydrophobic functional groups of the adsorbed
species and increase surface catalyzed transformation reactions (Olshansky
et al., 2014). For systems where cation bridging plays a prominent role in
DOM adsorption (e.g., to the siloxane surfaces of 2:1 layer type clay
minerals), cation charge and valence effects are important, with increasing
exchangeable Ca2+ relative to Na+, resulting in greater DOM
retention (Setia et al., 2013).
Due to the heterogeneous nature of both DOM and soil mineral constituents,
fractionation of DOM occurs as a result of a gradient of interaction
affinities between the DOM components and various soil particle surfaces
(Kaiser et al., 1997; Oren and Chefetz, 2012a). DOM fractionation has been
studied extensively on single mineral phases (Chorover and Amistadi, 2001;
Vazquez-Ortega et al., 2014) and on bulk soils (Guo and Chorover, 2003;
Kaiser et al., 1997; Oren and Chefetz, 2012b). Metal (oxy)hydroxides have
been suggested as a dominant adsorbent for DOM with the result being
preferential retention of high molar mass aromatic and carboxylated moieties
(Chorover and Amistadi, 2001; Vazquez-Ortega et al., 2014). Conversely,
layered silicates (e.g., smectites, kaolinite) were reported to adsorb
mainly low molar mass and aliphatic DOM fractions (Chorover and Amistadi,
2001; Polubesova et al., 2008). While the use of specimen mineral phases in
adsorption experiments facilitates elucidation of molecular mechanisms of
DOM interaction, it does not account for the complexity of competitive
interactions associated with heterogeneous assemblies of weathered surfaces
as found in natural soils. Conversely, using whole soils in adsorption
experiments has traditionally hindered mechanistic interpretations of DOM
uptake results. However, increased spatial resolution of spectroscopic
methods has helped to overcome these shortcomings by providing micro- and
nanoscale information on both soil–mineral phases and associated organic
molecules (Chen et al., 2014b).
The current study aimed to utilize such methodological advances to
elucidate (i) how wet–dry cycles affect the reactions between DOM and
subsoil particle surfaces and (ii) whether spatial fractionation of DOM is
detectable with nanoscale resolution spectroscopic methods. We hypothesized
that discontinuous wet–dry cycling during DOM reaction with subsoils would
increase complexation of carboxyl groups with metal (oxy)hydroxide surfaces
or hydroxylated edge surfaces of aluminosilicate clays and promote
association of hydrophobic fractions with pre-adsorbed and desiccated DOM
components relative to continuously wet conditions. Such wetting–drying
episodes have been hypothesized to affect OC dynamics in water-limited
portions of the critical zone, such as those that occur in the semi-arid
southwestern USA (Miller et al., 2005; Perdrial et al., 2014), but they have
not been previously investigated in controlled laboratory experiments.
Materials and methodsSoil samples
Soils were sampled from below a mixed conifer forest in the Santa Catalina
Mountains (SCM) and Jemez River Basin (JRB) critical zone observatories
(CZOs) in Arizona and New Mexico, respectively (Chorover et al., 2011). The
JRB soil was collected from the south slope of San Antonio Mountain
(35∘55′10′′ N, 106∘36′52′′ W) at an elevation of 2750 m. The SCM
soil was collected from the northeast slope of the zero-order basin located
in the Marshall Gulch experimental site (32∘25′44′′ N,
110∘46′14′′ W) at elevation of 2600 m. The mean annual temperature is 6
and 10.4 ∘C for the JRB and SCM sites, respectively. Both sites are
subjected to bimodal annual precipitation patterns with averages of 850 and
940 mm yr-1. Parent rock is igneous felsic at both sites, and granitic in
the SCM and rhyolitic in the JRB. Therefore, the soils used in experiments
developed under similar vegetation and climatic conditions but in different
parent materials. The SCM and JRB soils are classified as Typic Ustorthents
and mixed Psammentic Cryoboralfs, respectively (Soil Survey Staff, 2010;
USDA-NRCS, 1999). Soils were collected from the litter layer (0–2 cm) and
Bw3 horizon (80–100 cm), from pedons excavated (one in each site) in
April 2012 and October 2015 for SCM and JRB, respectively. The samples were
collected from different locations within each pit and composited to one
representative local sample. The SCM litter layer was collected in October
2015. Soils were air dried and sieved to obtain the fine earth (< 2 mm) fraction and stored in a closed container.
Table 1 presents the bulk
properties of the studied subsoils as measured using standard methods
(Sparks, 1996). The mineral assemblages of both soils were dominated by
quartz, feldspars and aluminosilicate clays (Table S1 in the Supplement). The SCM soil had
higher OM content (1.1 ± 0.5 mg C mg-1) and lower pH
(6.1 ± 0.04) than the JRB soil (0.17 ± 0.2 mg C mg-1 and
7.05 ± 0.11).
Physicochemical characteristics of the study soils.
a BET-N2 specific surface area.
b Cation exchange capacity.
c Organic carbon.
d Obtained in soil aqueous extract (1:10 with 8.2 MΩ, Barnstead
water).
e Humification index.
f Fluorescence index.
Dissolved organic matter extraction
The extraction of DOM was achieved by mixing the air-dried and sieved JRB or
SCM litter with ultrapure water (1:5 g g-1), and placing the suspension on a
reciprocal shaker at 150 rpm for 24 h. Suspensions were centrifuged at
15 000 g for 30 min to separate the solids, using polypropylene copolymer
(PPCO) centrifuge bottles. Adsorption or contamination of DOM from these
bottles was measured to be negligible (Vazquez-Ortega et al., 2014). The
supernatant solution was transferred into 50 mL PPCO centrifuge tubes and
centrifuged again at 40 000 g for 20 min to remove colloidal organic material
and the inorganic clay fraction. Supernatant solutions were filtered through
pre-combusted and cleaned 0.7 µm glass fiber filters. Total organic carbon (TOC) was
measured immediately after extraction (Shimadzu TOC-VCSH, Columbia, MD) and
solutions were diluted using ultrapure water to give initial dissolved
organic carbon (DOC) concentrations of 45 mg L-1 (Table 1). DOM
solutions were stored at 4 ∘C prior to use.
Sequential batch experiments
To model the effect of sequential hydrologic events delivering litter
leachate to subsoils in the two CZO sites, subsoils were reacted in a set of
four steps with DOM extracted from the litter layer of the corresponding
profile. Thirty mL aliquots of DOM (DOC = 45 mg L-1) solution were
mixed with 3.0 g of soil in 50 mL PPCO centrifuge tubes and agitated (150 rpm,
orbital shaker) at room temperature, in the dark. Preliminary kinetic
experiments indicated an apparent equilibration time of 98 h, and this was
chosen as the equilibration time for each reactor vessel. Suspensions were
centrifuged for 30 min at 40 000 g and 28 mL was removed by careful
pipetting just below the surface to avoid loss of solids, filtered through
precombusted 0.7 µm glass fiber filters and the solutions were stored
at 4 ∘C for a maximum of 24 h prior to analysis, as discussed
below. For continually wet treatments, a fresh 28 mL aliquot of DOM solution was added to
each tube and suspensions agitated for an additional 98 h (28 mL was used
because ca. 2 mL remained as entrained solution in the wet soil paste). For
wet–dry treatments, the soil pastes were air dried for 24 h (drying was
accomplished by directing a low-flow circulating dry-air stream to promote
desiccation), then an aliquot of 30 mL DOM solution was added to each tube
and suspensions were re-agitated for 98 h, for a total of four sequential
reaction cycles. Three replicates were prepared for each soil and treatment
combination. After the four sequential reaction cycles, soils were
freeze-dried and total organic carbon and total nitrogen (TN) were
measured using an ECS 4010 CHNSO analyzer (Costech, MI, Italy). During the
experiment, samples were maintained under oxic conditions by equilibration
with oxygenated headspace. It is important to note that microbial activity
was not suppressed throughout the reaction steps.
Characterization of DOM solutions before and after reaction
Reacted and unreacted DOM solutions were characterized by the following
suite of complementary analytical methods: soluble TOC and TN were
determined by total elemental analyzer (Shimadzu TOC-L and TNM-L, Columbia,
MD), absorbance spectra (190 to 655 nm) were collected using a UV–Vis
spectrometer (Shimadzu Scientific Instruments UV-2501PC, Columbia, MD, USA),
fluorescence excitation–emission matrices (EEMs) were obtained with a
FluoroMax-4 equipped with a 150 W Xe-arc lamp source (Horiba Jobin Yvon,
Irvine CA, USA) and Fourier transform infrared (FTIR) spectra were
collected using a Nicolet NEXUS 670 IR spectrometer (Madison, WI). The EEMs
were acquired with excitation (Ex) from 200 to 450 nm and emission (Em) from
250 to 650 nm in 5 nm increments. Spectra were collected with Ex and Em
slits at 5 and 2 nm bandwidths, respectively, and an integration time of
100 ms. Ultrapure water blank EEMs were subtracted and fluorescence
intensities were normalized to the area under the water Raman peak,
collected at excitation 350 nm. Additionally, an inner-filter correction was
performed based on the corresponding UV–Vis scans (Murphy et al., 2013).
Transmission FTIR spectra were collected with a KBr beam splitter and a
deuterated triglycine sulfate (DTGS) detector. Aliquots of 2 mL of JRB DOM
solutions were transferred onto IR transmissive Ge windows and dried under
vacuum for 19 h; spectra were collected in transmission mode. For SCM DOM, 2 mL
aliquots were freeze-dried and mixed with IR-grade KBr, then compressed
into pellets. For each sample, 120 scans were collected over the spectral
range of 400–4000 cm-1 at a resolution of 4 cm-1. Clean Ge
windows and KBr pellets were used as background.
Scanning transmission X-ray microscopy and near-edge X-ray adsorption
fine structure (STXM–NEXAFS) analysis of soils
Scanning transmission X-ray microscopy (STXM) and near-edge X-ray absorption fine structure (NEXAFS) analyses were conducted on clay-size isolates to avoid
particulate organic matter and to overcome possible alteration of C
speciation during preparation of thin sections (Chen et al., 2014b). Clay
size fractions (< 2 µm) of the reacted and unreacted JRB soils
were separated by sedimentation after dispersion in ultrapure water using a
sonication bath. Samples for STXM analysis were prepared by depositing 5 µL
of diluted aqueous suspension onto a Si3N4 window (75 nm
thick) and air-dried. The samples were analyzed by STXM on beamline 10ID-1
at the Canadian Light Source (CLS), a 2.9 GeV third-generation synchrotron
source. The microscope set up used a 25 nm Fresnel zone plate, which
provided a maximum spatial resolution of ca. 30 nm. Samples were kept under
1/6 atm of He during measurement.
Spatially resolved spectra were obtained by collecting stacks of images at
energies below and above C 1s, Ca 2p, Fe 2p, element edges. The dwell time
was set to 1 ms and pixel sizes of 150 nm. Incident energy was calibrated
with CO2 at 290.74 eV.
The aXis2000 software package (Hitchcock et al., 2012) was used for STXM
image and spectral processing. Stacks were aligned and converted to optical
density using a clean area of the Si3N4 window for normalization.
Regions of interest (ROI) of C, Ca and Fe were extracted from each stack by
subtracting below the edge from the optical density (OD) maps. C NEXAFS
spectra were extracted by averaging the pixels from the ROI. NEXAFS spectra
were normalized and peak deconvolutions were performed using the ATHENA
software package (Ravel and Newville, 2005). Peak assignments were based on
Cody et al. (1998, 2008), Myneni (2002) and Urquhart et al. (1997).
Data analysis
Statistical analyses were performed using R software packages (Mangiafico,
2016). Data were checked for normality and equal variance. Means were tested
using one-way ANOVA for parametric analysis and Kruskal–Wallis for non-parametric
analysis. The differences between means were examined using Tukey's honest significant difference (HSD) or
Dunn tests for parametric or non-parametric analyses, respectively.
Parametric tests used to evaluate the difference of TOC, TN and C-to-N ratio
between treatments, while nonparametric tests were used to evaluate UV–Vis and
fluorescence data. The specific UV absorbance (SUVA254) was calculated
by normalizing absorbance at incident wavelength 254 nm by the cell path
length (1 cm) and DOC concentration (M). Fluorescence index (FI, Eq. 1) and
humification index (HIX, Eq. 2) values were calculated from the corrected
EEMs (McKnight et al., 2001; Ohno, 2002) as follows:
FIEx370=I450I500,HIXEx255=∑(I435→480)∑(I300→345),
where Ex is the excitation wavelength (nanometers) and I is the fluorescence
intensity at each wavelength.
Spectra collected by FTIR were background corrected using KBr pellets or the
Ge transmission window as blanks and baseline corrected using the spline
function in the OMNIC 8 software program (Thermo Nicolet Co., Madison, WI).
Peak positions were determined using the second-order Savitzky–Golay
method. The Voigt line shape (a convolution between mixed Gaussian and
Lorentzian line shapes) was fitted to the peaks in the 850–1850 cm-1
region using Grams/AI 8.0 spectroscopy software (Thermo Electron
Corporation). Changes in DOM molecular composition were evaluated by
quantifying peak intensity ratios. Peak assignments were based on Socrates (2004),
Mayo et al. (2004), Omoike and Chorover et al. (2004) and Abdulla et al. (2010).
The organic carbon (a, b), nitrogen (c, d) and C:N(e, f), for
equilibrated solutions (a, c, e) and solid phases after four reaction steps
(b, d, f). Values for equilibrated solution OC and N represent cumulative
removal from solution per soil mass. Dashed lines in OC and N plots show
continuously wet treatments, dotted lines in the C:N plot represent values of
unreacted DOM solutions, error bars are the standard deviation and letters
indicate significant difference (ANOVA and Tukey's HSD p< 0.05) from
unreacted control.
ResultsTotal OC and nitrogen
The loss of DOC from solution per unit mass of soil was largely independent
of reaction step and treatment. The mass loss of DOC upon reaction with SCM
soil was 156 ± 5, 217 ± 3, 167 ± 17 and 192 ± 10 mg kg-1
for steps 1–4, respectively, in the wet–dry treatment, and 163 ± 3, 222 ± 4, 217 ± 2.5 and 214 ± 6 mg kg-1 in
the continuously wet treatment. The mass loss of DOC upon reaction with JRB
soil was 248 ± 19, 257 ± 1, 197 ± 5 and 200 ± 12 mg kg-1
for steps 1–4, respectively, in the wet–dry treatment, and 256 ± 7, 236 ± 26, 176 ± 44 and 208 ± 2 mg kg-1 in
the continuously wet treatment. Hence, the mean fraction of OC removed from
DOM solution was 58 ± 5 % (SD) after each reaction step with JRB
soil, and OC uptake values were not significantly different between the
continuously wet and wet–dry treatments. In the SCM soil, the mean fraction
of OC removed was 41 ± 4 % of the total after each reaction step in
the wet–dry treatment. In contrast to the other three treatments, the
continually wet SCM treatment indicated increasing amounts of OC removed in
each step, with 39 ± 0.8 % in the first step, 48 ± 1 % in
the second and 56 ± 1 % in the third and fourth steps (Fig. 1).
At the end of four reaction steps the TOC of JRB soils increased from 1700 ± 74 mg OC kg-1
for the unreacted soil to 2750 ± 87 and 2840 ± 99 mg OC kg-1 for the wet–dry and
continuously wet treatments, respectively (Fig. 1). For the JRB soil,
increases in solid-phase OC were not significantly different (Student
t test, p> 0.95) from the cumulative amounts of DOC removed from
reacted solutions (902 ± 26 and 876 ± 34 mg OC kg-1 for
wet–dry and continuously wet treatments, respectively) and represent a 60 %
increase in soil TOC. Conversely, for the SCM soil, despite comparable
cumulative losses from solution (733 ± 29 and 817 ± 2 mg OC kg-1
for wet–dry and continuously wet treatments, respectively), solid-phase analyses indicated that the OC content of the reacted SCM (11 200 ± 380 and 11 200 ± 290 mg OC kg-1 soil for wet–dry and
continuously wet treatments, respectively) soils were effectively unchanged
relative to the unreacted control (11 800 ± 180 mg OC kg-1). We
then tested for differences between the mean change in OC in the reacted
soils and the mean amount of OC removed from solution using the Student
t test. Results demonstrate a significant mass loss of OC in the SCM soil
(p≤ 0.05), amounting to 1370 ± 840 and 1440 ± 680 mg OC kg-1
soil (for wet–dry and continuously wet treatments, respectively).
These values represents 11 ± 7 and 11 ± 5 % of the total
carbon in the wet–dry and continually wet systems.
Patterns in the removal of total N from the DOM solutions showed similar
trends for both soils. In the first two wet–dry steps, a higher proportion
of TN was removed from the solution (65–70 and 50–66 % for SCM and
JRB soils, respectively) than in the third and fourth steps (31–44 % for
both soils). The measured increase in soil TN by the end of the experiment
were 63 and 143 mg N kg soil-1 for SCM and JRB soils, respectively.
These values are slightly higher than the sum of TN removed from the
solution (51 and 88 mg N kg soil-1 for SCM and JRB soils, respectively; Fig. 1).
The C:N ratio for all reacted DOM solutions decreased from step 1 to step 4,
indicating preferential loss of C from solution, with no significant
difference between the continually wet and wet–dry treatments. However,
after the first reaction with the SCM soil, the C:N ratio was 22.0 ± 1.3,
which was higher than the unreacted DOM (14.1 ± 0.8). It is
important to note that DOM extracted from unreacted soil had a C:N ratio of
23.7 ± 0.9, and C:N of DOM decreased during the sequential reaction
steps. After the fourth reaction step, ratios of 11.1 ± 0.8 and 9.6 ± 0.8 were observed for the wet–dry and the continually wet
treatments,
respectively. The C:N of the reacted DOM solution with JRB soil decreased,
from 10 ± 1.0 after the first reaction step to 4.6 ± 0.5 after
the fourth reaction step. The C:N ratio of unreacted DOM solution was 8.4 ± 0.8. The overall change in soil C:N ratio was evaluated by the
differences between unreacted soil and soils reacted four times with DOM
solutions (Fig. 1). Reacted SCM soils had significantly lower C:N (24.2 ± 1) than unreacted SCM soil (30.5 ± 1.8). However, no change in
C:N was detected for reacted versus unreacted JRB soils.
The fluorescence Index (FI), humification index (HIX) and specific
UV absorbance at 245 nm (SUVA254), for equilibrated solutions reacted
with JRB and SCM soils. The solid lines are wet–dry series, dashed lines are
continuously wet and dotted lines are unreacted DOM; error bars are the
standard deviation.
UV–Vis and fluorescence spectroscopy
Reaction with subsoils altered spectroscopic properties of the
litter-derived DOM solutions as reflected in UV–Vis (SUVA254) and
fluorescence indices (HIX and FI), and there was relatively little variation
between continually wet and wet–dry treatments (Fig. 2). For both JRB and
SCM the SUVA254 values of DOM decreased (relative to unreacted DOM)
upon contact with soil (Fig. 2), with the exception of the fourth step in
wet–dry treatment of SCM soil (Fig. 2). This effect of contact with soil
on SUVA254 was larger for JRB than SCM, although it decreased with
progressive reaction steps even for JRB soils from ca. 200 L mol-1 cm -1 in the
first step to ca. 50 L mol-1 cm-1 by the fourth
step. High SUVA254 (905 ± 35 L mol-1 cm-1) was
measured for DOM extracted from unreacted JRB soil (Table 1). We note that
SUVA254 values of unreacted DOM also decreased between the first (393 L mol-1 cm-1)
and subsequent steps (∼ 350 L mol-1 cm-1),
indicating some alteration of DOM chromophores in
the stock DOM solution during the experiment. Although this was a small
change relative to soil reaction effects, alteration was also evident in the
HIX of unreacted JRB DOM. Therefore, treatment effects (continuously wet and
wet–dry) were evaluated on the basis of differences between reacted and
unreacted solutions for the same reaction step. The effect of reaction with
soil on SUVA254 values were less pronounced for SCM relative to JRB
soils. In the wet–dry treatments of SCM soil, SUVA254 values of
the first three steps were generally consistent at
ca. 330 ± 13 (L mol-1 cm-1),
and in the fourth step the SUVA254 increased
to 530 ± 2 (L mol-1 cm-1). Conversely, SUVA254
values increased slightly over the course of the experiment from 324 ± 10 to 410 ± 16 L mol-1 cm-1 for the continually wet SCM
treatment.
Humification index (HIX) values for the reacted DOM were generally higher or
similar to the unreacted DOM (Fig. 2). As with the SUVA254 index, the
fourth step of SCM wet–dry treatment was the exception (Fig. 2), giving a
lower HIX for reacted compared to unreacted DOM. The HIX values for DOM
that reacted with JRB soil were similar for continually wet and wet–dry
treatments. Conversely, with SCM soil, values for the wet–dry treatments were
lower than for continually wet treatments. The relative differences between
reacted and unreacted DOM were lower for the JRB system than for the SCM
system. For both JRB and SCM soils, higher fluorescence index (FI) values
were observed for reacted relative to unreacted DOM (Fig. 2) whereas
wet–dry versus wet-only treatment effects were negligible. For JRB, FI
values increased from 1.31 ± 0.04 (unreacted DOM) to 1.53 ± 0.04,
whereas corresponding values for SCM were 1.34 ± 0.04 and 1.42 ± 0.02, respectively. All FI values are in close agreement with the value of
DOM associated with predominantly plant material (ca. 1.4), as opposed to
microbial-derived DOM (ca. 1.9; McKnight et al., 2001).
Transmission FTIR spectra of the dried DOM solution reacted with
JRB soils from steps 1 to 4 for continuously wet (a) and wet–dry
cycles
(b) and the unreacted JRB DOM solution (bottom black line). For color
rendering of this image please refer to the online version.
Transmission FTIR spectra of the dried DOM solution reacted with
SCM soils from steps 1 to 4 for continuously wet (a) and wet–dry
cycles
(b) and the unreacted SCM DOM solution (bottom line). For color
rendering of this image please refer to the online version.
FTIR
Transmission FTIR spectra of reacted and unreacted DOM for the JRB and SCM
systems are shown in Figs. 3 and 4, respectively. The most prevalent peaks
in the spectra were associated with amide I and II (1636 and 1560 cm-1,
respectively), carboxylate (asymmetric and symmetric stretches at 1592 and
1417 cm-1, respectively), alkyl (CH2 and CH3 bending
vibrations at 1455 and 1380 cm-1, respectively) and aromatic moieties
(C = C ring vibration at 1500 cm-1, phenol O–H bend 1370 cm-1) and
O-alkyl (CO- stretch at 1030–1150 cm-1).
For JRB soil, the first reaction step in both continually wet and wet–dry
treatments was accompanied by a decrease in peak intensities of carboxylate
(1592 and 1417 cm-1) and amide (1636 and 1560) relative to
O-alkyl (1150–1030 cm-1). Additionally, primary alcohol (1035 cm-1)
peak intensity decreased relative to secondary alcohol (1100 cm-1). This trend persisted in the second step with JRB soil for both
treatments, although the pattern was less pronounced and differed by
treatment. Specifically, the wet–dry treatment showed a larger decrease in
the asymmetric carboxylate stretch (1592 cm-1), whereas the
continuously wet treatment showed a larger decrease in the amide I peak (1636 cm-1). In the third step, the decrease in amide and carboxyl peaks
relative to O-alkyl was not as pronounced for the wet–dry treatment as it was in the
continually wet treatment. Finally, in the fourth step of the wet–dry
system, a pronounced decrease in amide and carboxyl peaks relative to
O-alkyl was again observed, whereas it was not in the continually wet
treatment (Fig. 3).
Figure 4 shows the spectra of reacted and unreacted DOM in the SCM system.
The SCM DOM spectra show similar peaks as the JRB with the addition of
carboxyl (C = O stretch at 1720 cm-1) and ester (C = O stretch 1770 cm-1 and
C–O stretch 1265 cm-1). Similar to the JRB system, after
reaction with soil, the peaks associated with carboxyl, carboxylate and
amide decreased relative to the O-alkyl peaks, and this trend was more
pronounced in the first step than in the subsequent steps. Similar to the
JRB system, in the fourth step of the wet–dry treatment, a pronounced
decrease in carboxyl, carboxylate and amide peaks was again observed
relative to the O-alkyl peaks.
JRB soil reacted with DOM under wet–dry cycling. (a) C NEXAFS
spectra extracted from C, Ca and Fe regions of STXM map. Spectra of
unreacted soil (top) and DOM solution (bottom) are presented. Dashed
vertical lines point out C species. (b) Tri-colored STXM map of fine
fraction from JRB soil reacted four times with DOM under wet–dry cycling; Fe (red),
Ca (blue) and C (green). Image size 25 µm × 25 µm. For color
rendering of this image please refer to the online version.
STXM–NEXAFS
Given limitations in beam time, synchrotron analyses were focused on the JRB
soil because it showed larger OC accumulation over the course of the
experiment. Scanning transmission X-ray microscopy (STXM) images of C, Fe
and Ca obtained for the isolated fine fraction of JRB soils reacted four
times with DOM in wet–dry and continually wet treatments are shown in
Figs. 5 and 6, respectively. The OC signal was observed over all particle
surfaces, from continually wet and wet–dry treatments after four reaction
steps. Locations of higher Fe and Ca content were observed for both
treatments. Near-edge X-ray absorption fine structure (NEXAFS) spectra
extracted from C, Ca and Fe-rich regions of interest (ROI) of the STXM maps
and C NEXAFS spectra of bulk unreacted soil and DOM are included in Figs. 5
and 6. Spectra of the unreacted DOM consist of peaks representing aromatic
(1 s →π∗ at 285.1 eV), alkyl (1 s → 3 p/σ∗ at 287.5 eV),
amide (1 s →π∗ at 288 eV), carboxyl
(1s →π∗ at 288.5 and 290 eV) and O-alkyl (1 s →π∗ at 289.5 eV)
moieties. The C NEXAFS spectra of unreacted soil show no strong peaks of
amide, carboxyl and O-alkyl, similar to the unreacted DOM spectra. However,
after four steps of reaction with DOM, soil from both continually wet and
wet–dry treatments exhibited greatly enhanced carboxyl and O-alkyl peaks
relative to the unreacted soil. In the wet–dry treatment, the aromatic peak
was absent. The O-alkyl peak was more pronounced for the continually wet
than for the wet–dry treatment. Additionally, the amide peak was suppressed
in the reacted soil compared to the unreacted DOM, and for the wet–dry
treatment this peak was absent and was not included in the fitted spectra
(Supplement). The C NEXAFS spectra of Ca- and Fe-enriched ROIs
are similar to the average whole image spectra. However in the Ca ROI, the
carboxyl peak intensity was enhanced relative to Fe ROI and the averaged
whole image spectra. This carboxyl enhancement, which was absent in the
unreacted soil, was most pronounced in the wet–dry treatment.
Variations in the C NEXAFS spectra of the reacted soils following each
reaction step are displayed in Fig. 7. After the first reaction step,
intensities of the carboxyl and O-alkyl peaks were relatively increased. For
the continually wet treatment, spectra collected following the second and
third steps show an increase in alkyl and O-alkyl peaks, whereas this trend
was less evident in the wet–dry treatment.
JRB soil reacted with DOM under continuously wet conditions. (a) C
NEXAFS spectra extracted from C, Ca and Fe regions of STXM map. Spectra of
unreacted soil (top) and DOM solution (bottom) are presented. Dashed
vertical lines point out C species. (b) Tri-colored STXM map of fine
fraction from JRB soil reacted four times with DOM during the continuously wet
treatment. Fe (red), Ca (blue) and C (green). Image size 25 µm × 25 µm.
For color rendering of this image please refer to the online version.
C NEXAFS extracted from C (red in Fig. 6) regions of STXM map for
the second step of the continuously wet treatment (a) and from all four steps
of the wet–dry treatment (b). For color rendering of this image please
refer to the online version.
Discussion
Specific surface area (SSA) and OC content are dominant factors controlling
sorption of DOM to soil. For comparable mineralogy, higher SSA tends to
increase DOM sorption, while higher solid-phase OC content suppresses it
(Kaiser et al., 1997; Oren and Chefetz, 2012b). In addition, solution
chemistry can control DOM–soil interactions. For example, low pH can
neutralize weakly acidic OM functionalities, thereby decreasing
electrostatic repulsion from negatively charged surfaces, whereas bivalent
cations such as Ca2+ can form bridging complexes between
negatively charged surface and DOM sites (e.g., Setia et al., 2013).
Further, the presence of polyvalent metal cations in solution can promote
precipitation of (meta-)stable OM–metal complexes (Kleber et al., 2014).
Gradual drying of pore water changes the ionic strength of the solution, and
can potentially promote interactions with metal cations in solution and at
organo-mineral surfaces. In the current study, in spite of differences in
soil constituents and DOM compositions deriving from the two distinct CZO
sites, similar amounts of DOM were removed from solution with both JRB and
SCM soils. The fact that OC did not accumulate in the solid-phase SCM soil
despite significant removal from solution suggests that decomposition and
mineralization are dominant factors indicated in the removal of OC from the
reacted SCM DOM solutions. Since microbial activity was not suppressed in
this study, an active microbial community was presumably present in the
soils. Therefore, addition of labile OC in the form of DOM may have resulted
in microbial growth and biotransformation of pre-existing soil OC. Indeed,
the pronounced decrease in C:N ratio of the reacted soil is consistent with
microbial transformation of organic matter (German et al., 2011). Higher HIX
for all SCM reacted samples, with the exception of the last step in the
wet–dry treatment, further support OM transformation. Enhanced
mineralization in the SCM relative to JRB soil may be related to its
substantially higher native OC content (Table 1), which would preclude
surface stabilizing interactions (Kaiser et al., 1997; Oren and Chefetz,
2012b). Moreover, higher OC content makes the SCM soil more susceptible to
the priming effect of the added labile OC as DOM (Blagodatsky et al., 2010).
The relatively lower HIX value for the last step of wet–dry treatment
coincides with higher SUVA254. Since SUVA254 index is correlated
with sample aromaticity (Weishaar et al., 2003), an increase in the aromatic
peak in the FTIR spectra was expected. However, FTIR spectra show a relative
increase in O-alkyl rather than the aromatic vibrations. It is possible that
the relative decrease observed in the 1550 to 1700 cm-1 region of the
FTIR spectra is mainly due to a decrease in carboxyl associated peaks rather
than increased aromaticity. It is unclear if the removed fraction was
exchanged with previously adsorbed OM or preferentially decomposed in the
solution. Additional study using isotopically labeled material may provide
additional information regarding decomposition and exchange reactions in
similar systems.
Conversely, significant DOM or soil organic matter decomposition was not
observed for the JRB soil experiments, as evidenced from the C mass balance.
Therefore, changes in reacted DOM composition can be attributed to
preferential adsorption and exchange reactions. The increased FI value of
the reacted DOM further suggests preferential adsorption of plant-relative OM
over
microbial-derived OM. The slight decrease in SUVA254 values is also
consistent with this observation, since polyphenols derived from lignin
account for most of the aromaticity in DOM.
Spectra from C-NEXAFS obtained for the JRB soil fine fraction corroborate
the solution data obtained by FTIR. A pronounced increase in the carboxyl
peak (288.5 eV) after the first reaction step (Fig. 7) is consistent with
the decreased intensity of carboxyl in the reacted DOM solutions (Fig. 3).
NEXAFS spectra collected after the second and third steps of both treatments
show additional increases in the O-alkyl (289.5 eV) and alkyl (287.5 eV)
that corroborate the relative decrease in FTIR peak intensities for these
functionalities. The fact that the NEXAFS of the reacted JRB soils clearly
shows a relative increase in the carboxyl peak from the third to the fourth
step in the wet–dry treatment (Fig. 7) suggests that preferential
adsorption of the carboxylic component was facilitated by the pre-existing
soil–DOM phases of the dried soil. Prior work has shown that soil drying may
promote conformational changes in pre-adsorbed DOM that promotes
preferential desorption of O-alkyl relative to further inner-sphere
coordination of carboxyl components (Kang et al., 2008; Kang and Xing,
2007). Additional support for the formation of inner-sphere carboxyl
complexes is from the higher preferential adsorption of carboxyl over amide
as observed in FTIR spectra of wet–dry compared to continuously wet treatments
(Fig. 3).
Due to the heterogeneous composition of soil surfaces and DOM, spatial
fractionation of the adsorbed OC moieties was expected. Figures 5 and 6 show
that in both wet–dry and continuously wet treatments, regions containing
higher content of Fe and Ca can be distinguished. Interestingly, the C
NEXAFS spectra of these distinct locations are generally similar. It is
important to note that low Fe spectral signals were detected over all of the
particle surfaces images with STXM. This observation contradicts our initial
hypothesis, and previous observations (Chorover and Amistadi, 2001; Kaiser
et al., 1997; Oren and Chefetz, 2012b; Vazquez-Ortega et al., 2014) that
iron (oxy)hydroxides will preferentially adsorb carboxyl-containing
moieties. These results suggest that weathered particle surfaces,
potentially already coated with a thin layer of metal (Fe) oxides and
co-associated organic matter, may smear out what might otherwise be observed
as a spatial fractionation at this scale (nanometers).
However, close inspection of the C spectra extracted from Fe- and Ca-enriched
zones and whole particle regions reveal that in samples treated with wet–dry
steps, the amplitude of the carboxyl peak shows a relative increase
preferentially in the Ca-enriched regions (Fig. 5 and Supplement). This finding suggests that cation bridging interactions are
pronounced in stabilizing the carboxyl component in the studied soil. It is
important to note that the solution pH was close to 7, and therefore
deprotonated carboxylate species were predominant in the suspension. Regions
of high Ca are likely associated with charged aluminosilicate surfaces
hosting exchangeable cations. The enhancement effect of drying on
Ca–carboxylate complex formation can be related to the tendency of the
Ca2+ hydration shell to become more acidic upon drying (Sposito, 1984).
As water molecules are gradually removed during air drying, polarizing
forces of the Ca2+ cation increases, enhancing the tendency of
hydration water to donate protons (Dowding et al., 2005). Therefore, upon
drying, protonation of the carboxylate functionality is expected.
Protonation of carboxylate decreases the electrostatic repulsion from
negatively charged clay surfaces and increases the overall interaction with
clays. It is important to note that our studied soils are predominantly
composed of silicate and aluminosilicate minerals and are relatively
depleted in crystalline and short-range-order metal oxides.
Conclusions
Results of this study show that wet–dry cycles affect interactions between
DOM and subsurface soils, in this case by enhancing the interactions between
carboxyl functional group and soil surfaces. Interactions of these
functionalities were dominated by Ca2+ bridging to soil surfaces. The
data also demonstrate that nanoscale spatial fractionation of DOM on soil
organo-mineral surfaces was diminished relative to what might be inferred
from previous observations pertaining to DOM fractionation on specimen
mineral phases. This is likely due to the heterogeneous composition of the
weathered soil surfaces and passivation of the underlying mineralogy by
metal oxide and OM films. Expanding the experiment to include soils with a
higher proportion of short-range-order (oxy)hydroxides may result in more
pronounced nanoscale spatial fractionation of DOM, but that is unknown at
present. Fractionation of DOM in solution under wet–dry
conditions for a soil that presented measurable decomposition of the DOM
(SCM) was similar to that for a soil that did not show any detectable decomposition
(JRB).
This study provides direct evidence of the role of wet–dry cycles in the
sorption reactions of DOM to a complex soil matrix. In the soil environment,
where wet–dry cycles occur at variable frequencies from site to site and
along the soil profile, different interactions between DOM and soil surfaces
are expected. This wet–dry effect can partially explain the observation that
carbohydrates predominate in subsoil horizons, where soil is less subjected
to drying, whereas aromatic and carboxylic compounds are more prevalent in
top soils, where wet–dry cycles are more frequent (Kaiser and Kalbitz,
2012). Our findings demonstrate the need to consider the effect of wet–dry
cycles in studying the interactions between DOM and soil surfaces.
STXM data and FITR spectra are available via the following link:
10.4211/hs.eb20c1fa74ad44a2a51834985dbf4481 (Olshansky, 2018).
The Supplement related to this article is available online at https://doi.org/10.5194/bg-15-821-2018-supplement.
The authors declare that they have no conflict of interest.
Acknowledgements
This research was funded by the Binational Agricultural Research and
Development (BARD) program, postdoctoral fellowship to Yaniv Olshansky grant
no. FI-534-2015, and the National Science Foundation, grant no. EAR 13-31408,
which supports the Catalina-Jemez Critical Zone Observatory. The STXM
analysis described in this paper was performed at the Canadian Light Source
beamline 10ID-1, which is supported by the Canadian Foundation for
Innovation, Natural Sciences and Engineering Research Council of Canada, the
University of Saskatchewan, the Government of Saskatchewan, Western Economic
Diversification Canada, the National Research Council Canada, and the
Canadian Institutes of Health Research. Thanks to Mary Kay Amistadi, Rachel
Nadine Burnett and Prakash Dhakal for assistance with
analysis.
Edited by: Michael Weintraub
Reviewed by: two anonymous referees
References
Abdulla, H. A. N., Minor, E. C., Dias, R. F., and Hatcher, P. G.: Changes in
the compound classes of dissolved organic matter along an estuarine
transect: A study using FTIR and 13C NMR, Geochim. Cosmochim. Acta, 74,
3815–3838, 2010.
Blagodatsky, S., Blagodatskaya, E., Yuyukina, T., and Kuzyakov, Y.: Model of
apparent and real priming effects: Linking microbial activity with soil
organic matter decomposition, Soil Biol. Biochem., 42, 1275–1283, 2010.
Chen, C., Dynes, J. J., Wang, J., and Sparks, D. L.: Properties of Fe-Organic
Matter Associations via Coprecipitation versus Adsorption, Environ. Sci.
Technol., 48, 13751–13759, 2014a.
Chen, C., Dynes, J. J., Wang, J., Karunakaran, C., and Sparks, D. L.: Soft
X-ray spectromicroscopy study of mineral-organic matter associations in
pasture soil clay fractions, Environ. Sci. Technol., 48, 6678–6686, 2014b.
Chorover, J. and Amistadi, M. K.: Reaction of forest floor organic matter at
goethite, birnessite and smectite surfaces, Geochim. Cosmochim. Acta, 65,
95–109, 2001.
Chorover, J., Kretzschmar, R., Garica-Pichel, F., and Sparks, D. L.: Soil
biogeochemicial processes within the critical zone, Elements, 3,
321–326, 2007.
Chorover, J., Troch, P. A., Rasmussen, C., Brooks, P. D., Pelletier, J. D.,
Breshears, D. D., Huxman, T. E., Kurc, S. A., Lohse, K. A., McIntosh, J. C.,
Meixner, T., Schaap, M. G., Litvak, M. E., Perdrial, J., Harpold, A., and
Durcik, M.: How Water, Carbon, and Energy Drive Critical Zone Evolution: The
Jemez–Santa Catalina Critical Zone Observatory, Vadose Zone J., 10, 884–899, 2011.
Cody, G. D., Ade, H., Wirick, S., Mitchell, G. D., and Davis, A.:
Determination of chemical-structural changes in vitrinite accompanying
luminescence alteration using C-NEXAFS analysis, Org. Geochem., 28,
441–455, 1998.Cody, G. D., Ade, H., Alexander, C. M. O. D., Araki, T., Butterworth, A.,
Fleckenstein, H., Flynn, G., Gilles, M. K., Jacobsen, C., Kilcoyne, A. L.
D., Messenger, K., Sandford, S. A., Tyliszczak, T., Westphal, A. J., Wirick,
S., and Yabuta, H.: Quantitative organic and light-element analysis of comet
81P/Wild 2 particles using C-, N-, and O-µ-XANES, Meteorit.
Planet. Sci., 43, 353–365, 2008.
Dowding, C. E., Borda, M. J., Fey, M. V., and Sparks, D. L.: A new method for
gaining insight into the chemistry of drying mineral surfaces using
ATR-FTIR, J. Colloid Interface Sci., 292, 148–151, 2005.
Eusterhues, K., Rennert, T., Knicker, H., Kogel-Knabner, I., Totsche, K. U.,
and Schwertmann, U.: Fractionation of organic matter due to reaction with
ferrihydrite: Coprecipitation versus adsorption, Environ. Sci. Technol.,
45, 527–533, 2011.
Eusterhues, K., Neidhardt, J., Hädrich, A., Küsel, K., and Totsche,
K. U.: Biodegradation of ferrihydrite-associated organic matter,
Biogeochemistry, 119, 45–50, 2014.
Fontaine, S., Barot, S., Barré, P., Bdioui, N., Mary, B., and Rumpel, C.:
Stability of organic carbon in deep soil layers controlled by fresh carbon
supply, Nature, 450, 277–280, 2007.
German, D. P., Weintraub, M. N., Grandy, A. S., Lauber, C. L., Rinkes, Z. L.,
and Allison, S. D.: Optimization of hydrolytic and oxidative enzyme methods
for ecosystem studies, Soil Biol. Biochem., 43, 1387–1397, 2011.
Gu, B., Schmitt, J., Chen, Z., Liang, L., and McCarthy, J. F.: Adsorption and
desorption of natural organic matter on iron oxide: mechanisms and models,
Environ. Sci. Technol., 28, 38–46, 1994.
Guo, M. and Chorover, J.: Transport and fractionation of dissolved organic
matter in soil columns, Soil Sci., 168, 108–118, 2003.Hitchcock, A., Hitchcock, P., Jacobsen, C., Zimba, C., Loo, B., Rotenberg,
E., Denlinger, J., and Kneedler, R.: aXis 2000-Analysis of X-ray Images and
Spectra, available at: http://unicorn.mcmaster.ca/aXis2000.html (last access: June 2016), 2012.
Kaiser, K. and Kalbitz, K.: Cycling downwards – dissolved organic matter in
soils, Soil Biol. Biochem., 52, 29–32, 2012.
Kaiser, K., Guggenberger, G., Haumaier, L., and Zech, W.: Dissolved organic
matter sorption on subsoils and minerals studied by 13C-NMR and DRIFT
spectroscopy, Eur. J. Soil Sci., 48, 301–310, 1997.
Kalbitz, K., Solinger, S., Park, J.-H., Michalzik, B., and Matzner, E.:
Controls on the dynamics of dissolved organic matter in soils: A review,
Soil Sci., 165, 277–304, 2000.
Kang, S. and Xing, B.: Adsorption of dicarboxylic acids by clay minerals as
examined by in situ ATR-FTIR and ex situ DRIFT, Langmuir, 23,
7024–7031, 2007.
Kang, S., Amarasiriwardena, D., and Xing, B.: Effect of dehydration on
dicarboxylic acid coordination at goethite/water interface, Colloid.
Surface A, 318, 275–284, 2008.
Kleber, M., Sollins, P., and Sutton, R. K.: A conceptual model of
organo-mineral interactions in soils: self-assembly of organic molecular
fragments into zonal structures on mineral surfaces, Biogeochemistry, 85,
9–24, 2007.
Kleber, M., Eusterhues, K., and Keiluweit, M.: Mineral–Organic associations:
formation, properties, and relevance in soil environments, edited by:
Sparks, D., Adv. Agron., 130, 1–140, 2014.
Lutzow, M. V., Kogel-Knabner, I., Ekschmitt, K., Matzner, E., Guggenberger,
G., Marschner, B., and Flessa, H.: Stabilization of organic matter in
temperate soils: mechanisms and their relevance under different soil
conditions – a review, Eur. J. Soil Sci., 57, 426–445, 2006.Mangiafico, S.: Summary and Analysis of Extension Program Evaluation in R,
version 1.10.0, available at: http://rcompanion.org/handbook/, last access: December, 2016.
Mayo, D. W., Miller, F. A., and Hannah, R. W.: Course Notes on the
Interpretation of Infrared and Raman Spectra, John Wiley & Sons, Inc.,
Hoboken, NJ, USA, 2004.
McKnight, D. M., Boyer, E. W., Westerhoff, P. K., Doran, P. T., Kulbe, T., and
Anderson, D. T.: Spectroflourometric characterization of dissolved organic
matter for indication of precursor organic material and aromaticity, L&O,
46, 38–48, 2001.
Miller, A. E., Schimel, J. P., Meixner, T., Sickman, J. O., and Melack, J. M.:
Episodic rewetting enhances carbon and nitrogen release from chaparral soils,
Soil Biol. Biochem., 37, 2195–2204, 2005.
Murphy, K. R., Stedmon, C. A., Graeber, D., and Bro, R.: Fluorescence
spectroscopy and multi-way techniques. PARAFAC, Anal. Methods, 5, 6557,
2013.
Myneni, S. C.: Soft X-ray spectroscopy and spectromicroscopy studies of
organic molecules in the environment, Rev. Mineral. Geochemistry, 49,
485–579, 2002.
Ohno, T.: Fluorescence inner-filtering correction for determining the
humification index of dissolved organic matter, Environ. Sci. Technol.,
36, 742–746, 2002.
Olshansky, Y., Polubesova, T., and Chefetz, B.: Reconstitution of cutin
monomers on smectite surfaces?: adsorption and esterification, Geoderma,
232–234, 406–413, 2014.Olshansky, Y.: Wet-dry cycles impact DOM retention in subsurface
soils, HydroShare, available at:
10.4211/hs.eb20c1fa74ad44a2a51834985dbf4481, 2018.
Omoike, A. and Chorover, J.: Spectroscopic study of extracellular polymeric
substances from Bacillus subtilis: aqueous chemistry and adsorption
effects, Biomacromolecules, 5, 1219–1230, 2004.
Oren, A. and Chefetz, B.: Sorptive and desorptive fractionation of dissolved
organic matter by mineral soil matrices, J. Environ. Qual., 41, 526–533,
2012a.
Oren, A. and Chefetz, B.: Successive sorption-desorption cycles of dissolved
organic matter in mineral soil matrices, Geoderma, 189–190, 108–115,
2012b.
Perdrial, J. N., McIntosh, J., Harpold, A., Brooks, P. D., Zapata-Rios, X.,
Ray, J., Meixner, T., Kanduc, T., Litvak, M., Troch, P. A., and Chorover, J.:
Stream water carbon controls in seasonally snow-covered mountain catchments:
impact of inter-annual variability of water fluxes, catchment aspect and
seasonal processes, Biogeochemistry, 118, 273–290, 2014.
Polubesova, T. and Chefetz, B.: DOM-Affected Transformation of Contaminants
on Mineral Surfaces: A Review, Crit. Rev. Environ. Sci. Technol., 44,
223–254, 2014.Polubesova, T., Chen, Y., Navon, R., and Chefetz, B.: Interactions of
hydrophobic fractions of dissolved organic matter with Fe(3+) – and
Cu(2+)-montmorillonite, Environ. Sci. Technol., 42, 4797–4803, 2008.
Ravel, B. and Newville, M.: ATHENA, ARTEMIS, HEPHAESTUS: Data analysis for
X-ray absorption spectroscopy using IFEFFIT, J. Synchrotron Radiat., 12,
537–541, 2005.
Rumpel, C. and Kögel-Knabner, I.: Deep soil organic matter – a key but
poorly understood component of terrestrial C cycle, Plant Soil, 338,
143–158, 2010.
Setia, R., Rengasmy, P., and Marschner, P.: Effect of exchangeable cation
concentration on sorption and desorption of dissolved organic carbon in
saline soils, Sci. Total Environ., 465, 226–232, 2013.
Socrates, G.: Infrared and Raman characteristic group frequencies: Tables
and charts, 3rd Edn., John Wiley & Sons, Ltd, Chichester, UK, 2004.
Soil Survey Staff: Keys to Soil Taxonomy, 11 Edn., USDA- Natural
Resources Conservation Service, Washington, DC, 2010.
Sparks, D. L.: Methods of soil analysis, in: Part 3: chemical methods, Soil
Science Society of America, Madison, WI, 1996.
Sposito, G.: The structure of water near clay mineral surfaces, in The
surface chemistry of soils, 47–77, Oxford University Press, New York,
1984.
Urquhart, S. G., Hitchcock, A. P., Smith, A. P., Ade, H., and Rightor, E. G.:
Inner-Shell Excitation Spectroscopy of Polymer and Monomer Isomers of
Dimethyl Phthalate, J. Phys. Chem. B, 101, 2267–2276, 1997.
USDA-NRCS: Soil survey geographic (SSURGO) database for Sandoval
County area, New Mexico (Includes parts of Los Alamos and Rio Arriba
Counties), Fort Worth, TX, 1999.Vázquez-Ortega, A., Hernandez-Ruiz, S., Amistadi, M. K., Rasmussen, C., and
Chorover, J.: Fractionation of Dissolved Organic Matter by
(Oxy)Hydroxide-Coated Sands: Competitive Sorbate Displacement during
Reactive Transport, Vadose Zone J., 13, 2–13, 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.
Zhao, P., Zavarin, M., Leif, R. N., Powell, B. A., Singleton, M. J.,
Lindvall, R. E., and Kersting, A. B.: Mobilization of actinides by dissolved
organic compounds at the Nevada Test Site, Appl. Geochem., 26,
308–318, 2011.