- Email: [email protected]
Factors controlling cadmium and lead activities in different parent material-derived soils from the Pearl River Basin
Accepted Manuscript Factors controlling cadmium and lead activities in different parent material-derived soils from the Pearl River Basin Shuran He, Qin Lu, Wenyan Li, Zongling Ren, Zhen Zhou, Xiao Feng, Yulong Zhang, Yongtao Li PII:
S0045-6535(17)30701-4
DOI:
10.1016/j.chemosphere.2017.05.007
Reference:
CHEM 19221
To appear in:
ECSN
Received Date: 19 February 2017 Revised Date:
9 April 2017
Accepted Date: 1 May 2017
Please cite this article as: He, S., Lu, Q., Li, W., Ren, Z., Zhou, Z., Feng, X., Zhang, Y., Li, Y., Factors controlling cadmium and lead activities in different parent material-derived soils from the Pearl River Basin, Chemosphere (2017), doi: 10.1016/j.chemosphere.2017.05.007. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT 1
Factors Controlling Cadmium and Lead Activities in Different
2
Parent Material-Derived Soils from the Pearl River Basin
3
Shuran Hea, Qin Lua, Wenyan Lia, Zongling Rena, Zhen Zhoua, Xiao Fenga, Yulong
5
Zhanga*, Yongtao Lib*
6
a
7
University, Guangzhou, 510642, P.R. China
8
b
9
University, Guangzhou, 510642, P.R. China
RI PT
4
SC
College of Natural Resources and Environment, South China Agricultural
10 11
M AN U
Key Laboratory of Arable Land Conservation, MOA, South China Agricultural
Corresponding authors:
13
Dr. Yulong Zhang, [email protected]
14
Dr. Yongtao Li, [email protected],
15
South China Agricultural University, No. 483, Wushan Road, Guangzhou 510642, P.R.
16
China
17
Phone: +86-20-38297890
18
Fax: +86-20-85280292
20
EP
AC C
19
TE D
12
1
ACCEPTED MANUSCRIPT Abstract
22
Labile metals in agricultural soils are available to crops and thus pose a great health
23
risk for human beings. Therefore, factors influencing heavy metal activity are of
24
interest to researchers. In this study, a total of 142 soil samples representing 5 typical
25
parent materials in the Pearl River Basin (PRB), China were collected to investigate
26
factors impacting the distribution of labile Cd and Pb in the soils. The results showed
27
that the labile fractions accounted for 0.03%-14.7% for Cd and 0.01%-0.39% for Pb
28
of the total metals, and the labile fractions were linearly correlated to their
29
corresponding total contents. The step regression analyses suggested that the key
30
factors impacting labile Cd and Pb varied in different parent material soils. Pb activity
31
was highly sensitive to pH in alkaline limestone soils. The quartz sand remained in
32
granite-produced soils enhanced Cd activity. And dissolved organic matter (DOM)
33
compositions considerably influenced Cd and Pb activities in sand shale, diluvium,
34
and alluvium soils. Land use impacts heavy metal activities. The labile Cd and Pb in
35
paddy soils were higher than those in non-paddy soils, although total metals in the
36
soils were comparable. It could be ascribed to the long-term equilibrium of metals
37
between the solution and solid phases of the paddy soils. The results provide a
38
theoretical basis for preliminary prediction of heavy metal activity and provide a
39
technical support for heavy metal activity management and pollution control based on
40
soil parent materials.
AC C
EP
TE D
M AN U
SC
RI PT
21
41 42
Keywords: metal activity; agricultural soil; land use; parent material; DMT 2
ACCEPTED MANUSCRIPT 43
1. Introduction Heavy metal contamination in the environment, especially the agricultural soils,
45
is drawing a great attention due to its high risk to human beings (Yang et al., 2013;
46
Qayyum et al., 2017). Relative to the total metal, the activity of heavy metal in soils is
47
considerably associated with the uptake of crop and thus human health (Sastre et al.,
48
2004; Tai et al., 2013). Therefore, illustrating the factors affecting the heavy metal
49
activity in agricultural soils is significant for the evaluation of their environmental
50
risks.
SC
M AN U
51
RI PT
44
The activities of heavy metals are influenced by soil properties and compositions, such as pH, organic matter, and iron oxides (Sauve et al., 2000a, 2000b; Lu et al.,
53
2003; Chen et al., 2008; Yu et al., 2016a, 2016b). Sauve et al. (2000a) reported that
54
pH, soil organic matter (SOM), and total contents of Ni, Cu, Cd, Zn can be used to
55
predict the percentages of labile fraction based on a big dataset. In the heavily
56
contaminated soils, the metals cannot be effectively fixed due to limited sorption
57
capacities, resulting in high activities of ions (Tsang et al., 2006). However, the heavy
58
metals tightly bound to varying matrices of soil have lower risks for transportation to
59
plants (Jacquat et al., 2009). Parent rocks substantially influence both the properties
60
and trace metal remaining of soils. Weathering of parent rocks produces the soil
61
primary and secondary minerals such as, quartz, kaoline, and montmorillonite. The
62
minerals are different in particle size, cation exchange capacity, metal species, etc.
63
which endows soils with corresponding properties. For some areas without any
64
pollution source, the metal compositions and contents are determined by the parent
AC C
EP
TE D
52
3
ACCEPTED MANUSCRIPT materials (Mico et al., 2006). Besides, the behaviors of the remaining metals can be
66
notably influenced by the parent material-related features. For instance, calcareous
67
soil is originated from limestone or dolomite and contains high carbonates and
68
organic matter (Chen et al., 2000). As carbonates form precipitates with heavy metals
69
and organic matters fix heavy metals with functional groups such carboxyl and
70
hydroxyl, they would enhance the accumulation of heavy metals. Considering the
71
important and complex role of parent materials on soil properties and heavy metals, it
72
is significant to understand the effects of parent materials on the activities of heavy
73
metals, as well as the factors impacting metal activities in different parent
74
material-derived soils.
M AN U
SC
RI PT
65
In this work, the Pearl River Basin (PRB) in China is selected for investigating
76
the heavy metal distributions in different parent material-derived soils of different
77
pollution levels and related impacting factors. On one hand, many mines and related
78
industries, such as, Dabaoshan Mine, Fankou Mine and Dongguan industrial district,
79
are located in the study area (Wong et al., 2002; Yin et al., 2016). Consequently,
80
heavy metal pollution in the agricultural soils, especially Cd and Pb, is severe in its
81
tributaries due to wastewater irrigation, atmospheric deposition, etc. On the other
82
hand, various parent rocks (residual and transported) including limestone, granite,
83
sand-shale, diluvium, and alluvium are extensively distributed in this area (Chen et al.,
84
2000).
AC C
EP
TE D
75
85
The aims of the present study were to investigate the distributions of labile Cd
86
and Pb in five representative parent material soils from the PRB; and more 4
ACCEPTED MANUSCRIPT 87
importantly, to unveil the factors affecting Cd and Pb activities in the soils.
88
2. Materials and methods
89
2.1. Sampling sites The catchment of PRB is up to 44.21×104 km2. One hundred forty-two (142)
91
surface soil (0-20 cm) samples from the PRB, including 54 samples in Xijiang River
92
Basin, 64 samples in Beijiang River Basin, and 24 samples in Pearl River Delta (Fig.
93
1) were collected. The sampling sites were so selected to include five representative
94
parent materials (limestone (L), sand shale (S), diluvium (D), alluvium (A), and
95
granite (G)) and reflect different pollution levels (e.g., background level and pollution
96
levels). Three zones including L1 (a background site), L2 polluted by a non-ferrous
97
metals smelter in Hechi City, and L3 slightly polluted zone, were located in the
98
limestone area. Four zones including S1 (a background site), S2 and S3 polluted by
99
the Dachang tin-polymetallic mining and located in the upper and lower catchment of
100
the Diaojiang River, and S4 slightly polluted, were in the sand shale area. Seven zones
101
including D1 and D2 polluted by the Fankou lead-zinc deposit, D3 and D4 polluted
102
by the Danxia lead-zinc smelter, D5 (a background site), D6 and D7 polluted by the
103
Maba lead-zinc smelter, were in the diluvium area. Three zones including A1 polluted
104
by the Dabaoshan polymetallic sulfide meso-hypothermal deposit, A2 (a background
105
site), A3 slightly polluted, were in the alluvium area. Two zones including G1 and G2
106
polluted by the industrial activity in Zhongshan City, were in the granite area. In each
107
zone, 6-12 sampling sites were chosen and each site was ~300 m away from the each
108
other. Surface samples (~5 kg) from the paddy (P) and non-paddy (N) soil were
AC C
EP
TE D
M AN U
SC
RI PT
90
5
ACCEPTED MANUSCRIPT collected from May to July, 2015. The samples were termed a combination with
110
parent material, land use type, and number (Table 1). The samples were stored by
111
plastic bags and carried back to the laboratory. The air-dried samples were grounded
112
to pass through 2 mm and 0.15 mm sieves for the different analyses.
113
2.2. Soil characterization
RI PT
109
Soil pH was measured from a 2.5:1 (v/m) water to soil suspension using a pH
115
detector (PHS-3C, China). Cation exchange capacity (CEC) was analyzed using the
116
ammonium acetate saturation methods (Tan et al., 2017). Particle-size analysis was
117
performed using sedimentation and pipette method. Sands (50‒2000 µm), silts (2‒50
118
µm), and clays (<2 µm) were measured (Zhang et al., 2017). Fe and Mn oxides
119
including crystalline and amorphous Fe/Mn were separated and further quantitatively
120
determined by colorimetry (Gleyzes et al., 2002). Soil organic carbon (SOC) was
121
determined by oxidation with dichromate (Zhang et al., 2017). Dissolved organic
122
matter (DOM) in soils was water-extracted (water to solid ratio = 5:1; 1 h) by a shaker.
123
Dissolved organic carbon (DOC) content was detected by a total organic carbon (TOC)
124
analyzer (Elementar vario TOC, Germany). The DOM compositions were
125
characterized by an ultraviolet-visible (UV) spectrophotometer. The absorbances at
126
250, 254, 365, 400, 436, 465, 600, and 665 nm of the equilibrium solution were
127
measured. Specific UV-vis absorptions at 254 nm (SUVA254, L·g-1 cm-1) and ∆logK
128
were calculated by the Eq. 1 and Eq. 2, respectively (Kumada, 1988).
AC C
EP
TE D
M AN U
SC
114
129
SUVA =
130
ΔlogK= logA400-logA600
(1) (2) 6
ACCEPTED MANUSCRIPT where, Ai is the absorbance at i nm, l (cm) is the cell path length and DOC is the
132
dissolved organic carbon content (mg L-1) of the soil solution. SUVA254 and ∆logK are
133
frequently used as the indicators for aromaticity and humification of DOM (Ni et al.,
134
2016). The E2/E3, E2/E4, and E4/E6 are the ratios of the absorbance at 250 nm over
135
365 nm, 254 nm over 436 nm, 465 nm over 665 nm (Table S1, Supplementary Data).
136
The former two ratios reflect the humification of DOM, meanwhile the latter ratio
137
represents aromaticity of DOM (Helms 2006).
138
2.3. Heavy metal analysis
M AN U
SC
RI PT
131
The total concentrations of Cd and Pb in soils were measured by the
140
HCl-HNO3-HClO4-HF digestion method (Li et al., 2009). The digestion solution were
141
analyzed by a flame atomic absorption spectrometry (AAS, Hitachi Z-2300, Japan). A
142
six step sequential extraction of heavy metals in soils was carried out based on the
143
modified Tessier method (Ma and Rao, 1997). The extraction methods were detailed
144
in the Table S2 (Supplementary Data). The Cd and Pb concentrations in the different
145
fractions were also determined by an AAS (Hitachi Z-2300, Japan).
EP
TE D
139
Soil Column Donnan membrane technology (SC-DMT) is a recently developed
147
tool to measure the free and complex metals in soil solution. It decreases the
148
disturbance of the soil chemical compositions and to a great extent simulates the
149
activities of heavy metals under water-soil equilibrium (Li et al., 2009; Ren et al.,
150
2015; Jones et al., 2016). Furthermore, the free metal ions together with ligand
151
complexes can be determined using the SC-DMT system. A soil of 5 g was loaded in
152
the system and the equilibrium time for the solid-water phases was obtained after 48 h.
AC C
146
7
ACCEPTED MANUSCRIPT The measurement procedure was detailed in Li et al. (2009). Since the Cd and Pb
154
concentrations in the equilibrium solutions were low, the metals in the donor and
155
acceptor solutions were determined by an inductively coupled plasma mass
156
spectrometry (ICP-MS, XSeries 2, Thermo Scientific, USA).
157
2.4. Statistical analyses
RI PT
153
Data analysis was performed using the SPSS statistics 20.0 and R software
159
combined with ADE-4. Statistical significance of the differences in heavy metals and
160
properties of soils was tested by one-way ANOVA. Stepwise regression analyses and
161
principle component analyses were conducted to extract factors affecting the
162
distribution of reactive metals in soils.
163
3. Results and discussion
164
3.1. Soil physicochemical properties
TE D
M AN U
SC
158
All the physicochemical properties of the PRB soils are shown in Table 1. TOC
166
and DOC of the soils ranged from 10.7 ± 3.7 to 70.0 ± 34.6 g kg-1 and from 29.0 ± 2.0
167
to 247 ± 80.9 mg kg-1, respectively. The TOC and DOC in limestone and sand-shale
168
soils were higher (P < 0.05) than the other parent material soils. The majority of pH in
169
soils was lower than 7, indicating an acidic soil environment. This could lead to a
170
potential of increasing the Cd and Pb activities (Sauve et al., 2000b). The limestone
171
soils was slightly alkaline, and the pH of L1 (avg. 7.15 and 7.52) was highest among
172
all the samples. Besides, both the relatively high TOC and DOC concentrations were
173
also observed in L1. These soil properties were consistent with their parent material of
174
limestone (Chen et al., 2000). The Fe and Mn oxides are key constituents for Cd and
AC C
EP
165
8
ACCEPTED MANUSCRIPT Pb accumulation in soils. Amorphous Fe oxides are one of the important factors
176
decreasing Cd activity in red soils, based on an investigation about Cd uptake of rice
177
(Yu et al., 2016a). The crystalline and amorphous Fe contents in soils were in the
178
range from 4.39 ± 0.20 to 36.1 ± 1.90 g kg-1 and from 1.36 ± 0.70 to 9.56 ± 2.50 g
179
kg-1. The crystalline and amorphous Mn contents in soils ranged from 14.79 ± 4.80 to
180
603.1 ± 59.5 mg kg-1 and from 8.26 ± 0.1 to 458.2 ± 160.8 mg kg-1. The contents of
181
Fe and Mn oxides in the granite soils were generally higher than the other. CEC
182
reflects the sorption capacity of metals to soils. The highest CEC values were
183
observed in GP1 (17.8 ± 4.70 cmol kg-1) and GN1 (18.1 ± 3.20 cmol kg-1). The
184
particle size distribution results indicated that slit (42%-61%, 47%-60%, 22%-48%,
185
39%-55% and 47%-50% for limestone, sand-shale, diluvium, alluvium, and granite,
186
respectively) is generally the most abundant fraction in soils. The sand and clay
187
fractions of limestone, sand-shale, diluvium, alluvium, and granite ranged from
188
15%-34%, 3%-25%, 29%-59%, 20%-42%, 2%-3%, and 18%-28%, 18%-40%,
189
19%-35%, 15%-35%, 47%-50%, respectively.
EP
TE D
M AN U
SC
RI PT
175
The DOM compositions are tightly related to the speciation of metals in soils.
191
Ren et al. (2015) found that fulvic acid is the most important fraction for binding Pb
192
and Cu in soil solution, due in part to the high polar functional groups. A series of
193
parameters derived from the absorbance values to some extent indicate the functional
194
groups in DOM (Helms, 2006). Table S1 shows that SUVA254 of soils ranged from
195
9.86 ± 0.95 to 28.2 ± 9.96 L g-1 cm-1. The ratios of E2/E3, E2/E4, and E4/E6 were in
196
the range of 5.45±0.70 to 16.33±3.72, 15.01±13.00 to 47.36±15.96, 0.47±0.22
AC C
190
9
ACCEPTED MANUSCRIPT to 2.19±1.25, respectively. ∆logK of the soils varied from 0.18±0.05 to 0.83±0.74.
198
The data analysis revealed that there was no significant difference in DOM
199
compositions among varying parent materials or land use types.
200
3.2. Heavy metal distribution
RI PT
197
The total concentrations of Cd and Pb in soils originated from the five parent
202
materials are shown in Fig. 2. Compared with the background agricultural soils, the
203
Cd and Pb concentrations of soils around the pollution sources were significantly
204
higher. This suggested a significant input of metals from the nearby industrial
205
activities. Except for the background sites, the majority of Cd in the studied soils was
206
higher than the threshold level (0.3 mg kg-1) of the China Environmental Quality
207
Standards for Soil (GB15618-1995, grade II for the agriculture land). In contrast,
208
most of the Pb concentrations of the soils were lower than its threshold level (250 mg
209
kg-1). Heavy pollution of Pb was extensively observed in the diluvium soils (Fig. 2).
210
And the Pb concentrations of the diluvium soils were significantly higher than the
211
other parent material soils. The compositions and concentrations of the metals largely
212
depended on the type of pollution sources and the distance from the sources. All the
213
diluvium sites were nearby the lead-zinc mining and its smelters in the Renhua and
214
Qujiang Districts. Cd and Pb concentrations in D1 were higher than those in D2 (not
215
shown), which is consistent with their distance between the Fankou mining and the
216
sampling sites. The above observation indicated a great influence of industrial
217
activities on accumulations of heavy metals in agricultural soils. In the granite area,
218
the metal pollution sources are extensively distributed due to the well-developed
AC C
EP
TE D
M AN U
SC
201
10
ACCEPTED MANUSCRIPT 219
industry in Zhongshan City. However, Cd (0.32-0.56 mg kg-1) and Pb (22.47-43.54
220
mg kg-1) pollutions were not severe as expected, which might be due to low Cd and
221
Pb input from the local industry. The total dissolved metals (free and ligand complex forms) with DMT system
223
represents the most labile fractions and provides more convincing indicators for
224
potential risk and toxicity of metal pollution soils. The dissolved fractions only
225
accounted for a minor percentages (0.03%-14.7% for Cd and 0.01%-0.39% for Pb) of
226
total metal in the soils. The values are comparable with previous studies (Li et al.,
227
2009; Ren et al., 2015). The free and complex Cd concentrations were in the range
228
from 0.05 to 19.7 µg L-1 and from 0.02 to 0.76 µg L-1, respectively (Fig. 3a). And the
229
free and complex Pb concentrations were in the range from 0.15 to 74.1 µg L-1 and
230
from 0.10 to 26.5 µg L-1, respectively (Fig. 3b). The free Cd concentrations were avg.
231
22 times higher than that of the complex Cd, whereas the free Pb was avg. 3 times
232
higher than its complex form. This phenomena indicated the ligands in soil
233
equilibrium solution were preferentially bound to Pb relative to Cd. This is consistent
234
with Ren et al. (2015)’s observation that in soil solution Pb is dominant in complex
235
form, while Cd mainly exists as free ion. Previous studies also have demonstrated that
236
Pb tends to be much more strongly bound to organic matter in soil solution than Cd
237
(Ge et al., 2005).
AC C
EP
TE D
M AN U
SC
RI PT
222
238
It is worth noting that the total dissolved, free, and complex Cd and Pb in paddy
239
soils were extensively higher than their corresponding non-paddy soils (Fig. 3),
240
although the total Cd and Pb concentrations in the two type soils were comparable. 11
ACCEPTED MANUSCRIPT This occurred in all the five parent material areas, indicating a notable effect of land
242
use on distributions of liable metals. The phenomenon could be explained by varying
243
sorption equilibrium patterns of metals in paddy and non-paddy soils. The paddy soils
244
are usually long-term flooded, resulting in a stable equilibrium of metal sorption
245
between water and soil components. In contrast, the dry soils are periodically irrigated.
246
The frequent alternation of wetting and drying would lead to repeated equilibrium
247
between heterogeneous solid components and water. The soil components, such as,
248
lipids, non-hydrolysable carbons, Fe oxides, clay silicate vary in sorption capacity and
249
dynamics for different pollutants (Weng et al., 2001; Zhang et al., 2013, 2014; Liu et
250
al., 2015, Yu et al., 2016b). Consequently, metals tend to preferentially accumulate in
251
strong sorption components of soils, since metal desorption from weak sorption
252
components are quick and easy (Weng et al., 2001). This speculation can be further
253
supported by the metal fractions in soil constituents based on the sequential extraction.
254
The Cd and Pb bound to organic and residual components in soils are considered to be
255
most stable fractions (Fest et al., 2008; Yang et al., 2013). Fig. 4 shows that the
256
percentages of Cd and Pb bound to organic and residual fractions in non-paddy soils
257
are generally higher than those in paddy soils for the five parent materials. Another
258
reason for the difference in metal availability between paddy and non-paddy soils
259
might be the different Eh in the two types of soils. In paddy soils, the low Eh leads to
260
the transformation of Fe(III) to Fe(II). Consequently, Fe(III)-bound Cd and Pb are
261
released, resulting in high Cd and Pb availability (Yu et al., 2016a). This preferential
262
accumulation of metals in the stable fractions may be the main reason for the
AC C
EP
TE D
M AN U
SC
RI PT
241
12
ACCEPTED MANUSCRIPT 263
contrasting distributions of free and complex Cd and Pb in paddy and non-paddy
264
soils.
265
3.3. Factors affecting the activities of Cd and Pb Besides land use, the labile Cd and Pb were largely dependent on the total metal
267
amount in soils. The correlation analyses between total and dissolved concentrations
268
of Cd and Pb were performed (Fig. 5a and 5b). One fitted linear curve with R2 = 0.72
269
and P < 0.001 for Cd was obtained. The fitted equation was y = 30073x + 16.2. If the
270
Cd contents in pollution soils ranged from 0.2 to 6 mg kg-1, the proportions of labile
271
Cd would decrease from 11.1% to 3.2% based on the equation. The dissolved Cd is
272
considered as a multi-equilibrium fraction between solution and soil constituents (Ren
273
et al., 2015). Consequently, the dissolved Cd content is positively correlated to the Cd
274
fixed in soil constituents. And the majority of Cd in soils was bound to the solid
275
constituents and determined the total Cd concentration in the studied soils (Fig. S1).
276
Thus, it is not surprising to obtain the significant correlation between labile and total
277
Cd in soils. Some exceptions were usually presented in heavily polluted sites (> 1.2
278
mg kg-1). The percentages of labile Cd relative to total Cd in DN1, SP1, LP1, LN1,
279
SP3, and SN3 were lower than the modeled level. This could be partly due to soil pH.
280
The pH in DN1 (7.03), LN1 (7.58), LP1 (7.70), SP3 (6.87), and SN3 (5.88) were
281
generally higher than that of the other soils (avg. 5.34). An alkaline condition is
282
known to favor Cd precipitation and thus prevent Cd activity in soils (Sauve et al.,
283
2000a, 2000b). In contrast, the percentages of labile Cd relative to total Cd in DP3
284
and DP4 were higher than the modeled level. This could be related to the contents and
AC C
EP
TE D
M AN U
SC
RI PT
266
13
ACCEPTED MANUSCRIPT compositions of DOM. As shown in the Table S3 (Supplementary Data), the DOC
286
contents in DP3 and DP4 were 144 and 173 mg kg-1, respectively, which was far
287
higher than the average value of 75.7 mg kg-1 for the other samples. It is observed that
288
DOM increased the solubility of ions from soils (Park et al., 2011). Furthermore,
289
lower ∆logK (avg. 0.24) in DP3 and DP4 relative to the other (avg. 0.35) meant more
290
acidic functional groups of DOM with high humification. These acidic groups such as
291
carboxyl and phenolic hydroxyl groups are active ligand for metals and favor the
292
solubility of Cd from soils (Simmler et al., 2013).
M AN U
SC
RI PT
285
The distributions of labile Pb in soils were similar to the Cd pattern (Fig. 5b).
294
The fitted equation was y=236.5x + 0.19 (R2 = 0.78, P < 0.001). Based on the
295
modeled data, the labile Pb accounted for ~0.024% when its total contents were in the
296
range from 50 to 400 mg kg-1. The portions of labile Pb in AP1 and AN1 were higher
297
than that of the modeled level, whereas the portions of Pb in SN3, SP3, DP1, and
298
DN1 were lower than that of the modeled level. The phenomena could be ascribed to
299
the soil pH and components. The lowest pH values were 3.85 for AP1 and 4.21 for
300
AN1. Strong acidic condition enhances the solubility of Pb in soils (Sauve et al.,
301
2000a, 2000b). Besides, the lower CEC (4.58 cmol kg-1 for AP1 and 4.33 cmol kg-1
302
for AN1) relative to the average level (8.36) of the other soils suggested a lower
303
sorption capacity (Qayyum et al., 2016). In contrast to the others, DN1, SP3, and SN3
304
had higher pH, contents of amorphous Fe and Mn oxides, which is benefit to the Pb
305
fixation in soils.
AC C
EP
TE D
293
14
ACCEPTED MANUSCRIPT Step regression analyses were preformed to extract the key variables for
307
influencing the Cd and Pb activities in soils originated from different parent materials.
308
Only the regression equations in significant level (P < 0.05) are shown in Table 2. The
309
key variables extracted in the equations varied greatly for different parent material
310
soils. Generally, pH, organic matter, and size distribution seemed to be the important
311
factors affecting the Cd and Pb activities. In limestone soils, Pb activity is highly
312
sensitive to pH decrease. The limestone soil usually is alkaline due to high content of
313
carbonates. Hydrogen ion increases the reaction with carbonates and releases the
314
carbonate-bounded Pb. In sand shale soils, free Pb was negatively correlated with
315
sand contents. The complex Pb was mostly influenced by E4/E6. Higher E4/E6 values
316
represent lower aromaticity, suggesting the possibility of higher proportions of
317
reactive groups such as carboxyl and sulfydryl groups. These reactive groups are
318
strong ligand for cations and thus considerably affects the solubility of heavy metals
319
in soils (Simmler et al., 2013; Shirvani et al. 2015). The mechanism could be also
320
presented for diluvium soils in which free and complex Pb were positively correlated
321
with E4/E6. In diluvium soils, free Cd is largely impacted by clay content. Clay is a
322
key sorption component for heavy metals in soils. High clay content would increase
323
sorption capacity for Cd. High E2/E3 represents the high humification degree of
324
DOM. The humification process produces the reactive functional groups such as
325
carboxyl, phenolic hydroxyl, alcoholic hydroxyl, and amino groups of DOM. This is
326
consistent with the positive correlation between complex Cd and E2/E3 in diluvium
327
soils. In alluvium soils, the complex Cd was negatively correlated with TOC. This is
AC C
EP
TE D
M AN U
SC
RI PT
306
15
ACCEPTED MANUSCRIPT consistent with the fact that organic matter is an important sorption component for Cd
329
in soils (Sauve et al., 2000b). In granite soils, free and complex Cd is mostly impacted
330
by sand. The sands in granite-derived soils mainly consist of quartz which is less
331
capable of sorbing Cd. Thus, higher portion of sand benefits Cd solubility in soils.
332
Some cases were hardly explained by the current understanding. For instance, the free
333
Pb is negatively correlated with silt content in granite soils. In alluvium soils, free and
334
complex Pb were negatively impacted by DOM humification degree as indicated by
335
∆logK (Table 2).
M AN U
SC
RI PT
328
336 337
4. Conclusions
Reactive Cd and Pb in agricultural soils pose a direct risk for crops and humans,
339
and factors influencing Cd and Pb activities in soils are of interest for people working
340
in soil pollution and remediation, agricultural safe production, and environmental
341
health. A series of soil properties were determined and were analyzed to extract the
342
key variables for understanding the distributions of reactive Cd and Pb in five
343
different parent material soils from the PRB. The results of the present study
344
demonstrated that the controlling factors for different parent material-derived soils
345
differed. For instance, pH was the key factor impacting free and complex Pb in
346
limestone soils. Sand played an important role in the distributions of free and complex
347
Cd in granite soils. In contrast, DOM compositions considerably influenced Cd and
348
Pb activities in sand shale, diluvium, and alluvium soils. Besides, land use greatly
349
influenced heavy metal activity. The reactive Cd and Pb in the paddy soils were
AC C
EP
TE D
338
16
ACCEPTED MANUSCRIPT higher than those in the non-paddy soils. Total metal noticeably impact Cd and Pb
351
activities in most of the soil samples, as evidenced by the significant linear
352
correlations between the total metal and their corresponding reactive fractions. And
353
effects of acidity and DOM of soils on the activities of Cd and Pb explained some
354
exceptions. In summary, in addition to the total concentration, taking effects of parent
355
materials, land use, acidity, and DOM into account is necessary for assessing the Cd
356
and Pb activities in agricultural soils.
AC C
EP
TE D
M AN U
SC
RI PT
350
17
ACCEPTED MANUSCRIPT 357
Acknowledgements This research was supported by the Natural Science Foundation of China
359
(U1401234 and 41601533), National Science & Technology Pillar Program
360
(2015BAD05B05 and 2014BADB14B01), National Key Research and Development
361
Program of China (2016YFD0800300), and China Postdoctoral Science Foundation
362
(2016M602479).
AC C
EP
TE D
M AN U
SC
RI PT
358
18
ACCEPTED MANUSCRIPT 363
References
364
Chen, J.S., Wang, F.Y., Li, X.D., Song, J.J., 2000. Geographical variations of trace
365
elements in sediments of the major rivers in eastern China. Environmental
366
Geology (Berlin) 39, 1334-1340. Chen, M., Li, X.-M., Yang, Q., Zeng, G.-M., Zhang, Y., Liao, D.-M., Liu, J.-J., Hu,
368
J.-M., Guo, L., 2008. Total concentrations and speciation of heavy metals in
369
municipal sludge from Changsha, Zhuzhou and Xiangtan in middle-south region
370
of China. Journal of Hazardous Materials 160, 324-329.
SC
RI PT
367
Fest, E.P.M.J., Temminghoff, E.J.M., Comans, R.A.J., van Riemsdijk, W.H., 2008.
372
Partitioning of organic matter and heavy metals in a sandy soil: Effects of
373
extracting solution, solid to liquid ratio and pH. Geoderma 146, 66-74.
M AN U
371
Ge, Y., MacDonald, D., Sauvé, S., Hendershot, W., 2005. Modeling of Cd and Pb
375
speciation in soil solutions by WinHumicV and NICA-Donnan model.
376
Environmental Modelling & Software 20, 353-359.
TE D
374
Gleyzes, C., Tellier, S., Astruc, M., 2002. Fractionation studies of trace elements in
378
contaminated soils and sediments: a review of sequential extraction procedures.
379
Trac-Trends in Analytical Chemistry 21, 451-467.
381
Helms, J.R., 2006. Spectral shape as an indicator of molecular weight in chromophoric dissolved organic matter. Old Dominion University.
AC C
380
EP
377
382
Jacquat, O., Voegelin, A., Kretzschmar, R., 2009. Soil properties controlling Zn
383
speciation and fractionation in contaminated soils. Geochimica Et Cosmochimica
384
Acta 73, 5256-5272.
385
Jones, A.M., Xue, Y., Kinsela, A.S., Wilcken, K.M., Collins, R.N., 2016. Donnan
386
membrane speciation of Al, Fe, trace metals and REEs in coastal lowland acid
387
sulfate soil-impacted drainage waters. Science of the Total Environment 547,
388
104-113.
389
Kumada, K., 1988. Chemistry of Soil Organic Matter. Elsevier. 19
ACCEPTED MANUSCRIPT 390
Li, Y.-T., Becquer, T., Dai, J., Quantin, C., Benedetti, M.F., 2009. Ion activity and
391
distribution of heavy metals in acid mine drainage polluted subtropical soils.
392
Environmental Pollution 157, 1249-1257. Liu, C., Yu, H.Y., Liu, C., Li, F., Xu, X., Wang, Q., 2015. Arsenic availability in rice
394
from a mining area: Is amorphous iron oxide-bound arsenic a source or sink?
395
Environmental Pollution 199, 95-101.
RI PT
393
Lu, Y., Gong, Z.T., Zhang, G.L., Burghardt, W., 2003. Concentrations and chemical
397
speciations of Cu, Zn, Pb and Cr of urban soils in Nanjing, China. Geoderma 115,
398
101-111.
400
Ma, L.Q., Rao, G.N., 1997. Chemical fractionation of cadmium, copper, nickel, and
M AN U
399
SC
396
zinc in contaminated soils. Journal of Environmental Quality 26, 259-264. Mico, C., Recatala, L., Peris, A., Sanchez, J., 2006. Assessing heavy metal sources in
402
agricultural soils of an European Mediterranean area by multivariate analysis.
403
Chemosphere 65, 863-872.
TE D
401
Ni, X., Yang, W., Tan, B., Li, H., He, J., Xu, L., Wu, F., 2016. Forest gaps slow the
405
sequestration of soil organic matter: a humification experiment with six foliar
406
litters in an alpine forest. Scientific Reports 6, 19744-19744.
EP
404
Park, J.H., Lamb, D., Paneerselvam, P., Choppala, G., Bolan, N., Chung, J.W., 2011.
408
Role of organic amendments on enhanced bioremediation of heavy metal(loid)
409
contaminated soils. Journal of Hazardous Materials 185, 549-574.
AC C
407
410
Qayyum, M.F., ur Rehman, M.Z., Ali, S., Rizwan, M., Naeem, A., Maqsood, M.A.,
411
Khalid, H., Rinklebe, J., Ok, Y.S., 2017. Residual effects of monoammonium
412
phosphate, gypsum and elemental sulfur on cadmium phytoavailability and
413
translocation from soil to wheat in an effluent irrigated field. Chemosphere 174,
414
515-523.
415
Ren, Z.-L., Tella, M., Bravin, M.N., Comans, R.N.J., Dai, J., Garnier, J.-M., Sivry, Y.,
416
Doelsch, E., Straathof, A., Benedetti, M.F., 2015. Effect of dissolved organic 20
ACCEPTED MANUSCRIPT 417
matter composition on metal speciation in soil solutions. Chemical Geology 398,
418
61-69. Sastre, J., Hernandez, E., Rodriguez, R., Alcobe, X., Vidal, M., Rauret, G., 2004. Use
420
of sorption and extraction tests to predict the dynamics of the interaction of trace
421
elements in agricultural soils contaminated by a mine tailing accident. Science of
422
the Total Environment 329, 261-281.
RI PT
419
Sauve, S., Hendershot, W., Allen, H.E., 2000a. Solid-solution partitioning of metals in
424
contaminated soils: Dependence on pH, total metal burden, and organic matter.
425
Environmental Science and Technology 34, 1125-1131.
SC
423
Sauve, S., Norvell, W.A., McBride, M., Hendershot, W., 2000b. Speciation and
427
complexation of cadmium in extracted soil solutions. Environmental Science and
428
Technology 34, 291-296.
M AN U
426
Shirvani, M., Sherkat, Z., Khalili, B., Bakhtiary, S., 2015. Sorption of Pb(II) on
430
palygorskite and sepiolite in the presence of amino acids: Equilibria and kinetics.
431
Geoderma 249, 21-27.
TE D
429
Simmler, M., Ciadamidaro, L., Schulin, R., Madejon, P., Reiser, R., Clucas, L., Weber,
433
P., Robinson, B., 2013. Lignite Reduces the Solubility and Plant Uptake of
434
Cadmium in Pasturelands. Environmental Science & Technology 47, 4497-4504.
EP
432
Tai, Y., McBride, M.B., Li, Z., 2013. Evaluating specificity of sequential extraction
436
for chemical forms of lead in artificially-contaminated and field-contaminated
437
AC C
435
soils. Talanta 107, 183-188.
438
Tan, X., Liu, Y., Yan, K., Wang, Z., Lu, G., He, Y., He, W., 2017. Differences in the
439
response of soil dehydrogenase activity to Cd contamination are determined by
440
the different substrates used for its determination. Chemosphere 169, 324-332.
441
Tsang, D.C.W., Lo, I.M.C., 2006. Competitive Cu and Cd sorption and transport in
442
soils: A combined batch kinetics, column, and sequential extraction study.
443
Environmental Science & Technology 40, 6655-6661. 21
ACCEPTED MANUSCRIPT 444
Weng, L., Temminghoff, E. J.,Van Riemsdijk, W. H., 2001. Contribution of individual
445
sorbents to the control of heavy metal activity in sandy soil. Environmental
446
Science & Technology 35, 4436-4443. Wong, S.C., Li, X.D., Zhang, G., Qi, S.H., Min, Y.S., 2002. Heavy metals in
448
agricultural soils of the Pearl River Delta, South China. Environmental Pollution
449
119, 33-44.
RI PT
447
Yang, S., Zhou, D., Yu, H., Wei, R., Pan, B., 2013. Distribution and speciation of
451
metals (Cu, Zn, Cd, and Pb) in agricultural and non-agricultural soils near a
452
stream upriver from the Pearl River, China. Environmental Pollution 177, 64-70.
SC
450
Yin, H., Tan, N.H., Liu, C.P., Wang, J.J., Liang, X.L., Qu, M.K., Feng, X.H., Qiu,
454
G.H., Tan, W.F., Liu, F., 2016. The associations of heavy metals with crystalline
455
iron oxides in the polluted soils around the mining areas in Guangdong Province,
456
China. Chemosphere 161, 181-189.
M AN U
453
Yu, H.Y., Liu, C., Zhu, J., Li, F., Deng, D.-M., Wang, Q., and Liu, C., 2016a.
458
Cadmium availability in rice paddy fields from a mining area: The effects of soil
459
properties highlighting iron fractions and pH value. Environmental Pollution 209,
460
38-45.
TE D
457
Yu, H. Y., Li, F. B., Liu, C. S., Huang, W., Liu, T. X., and Yu, W. M., 2016b. Chapter
462
Five - Iron Redox Cycling Coupled to Transformation and Immobilization of
463
Heavy Metals: Implications for Paddy Rice Safety in the Red Soil of South China,
AC C
464
EP
461
In Advances in Agronomy (Donald, L. S., Ed.), pp 279-317, Academic Press.
465
Zhang, Y.L., Ma, X.X., Ran, Y., 2014. Sorption of phenanthrene and benzene on
466
differently structural kerogen: Important role of micropore-filling. Environmental
467
Pollution 185, 213-218.
468
Zhang, Y.L., Ran, Y., Mao, J.D., 2013. Role of extractable and residual organic matter
469
fractions on sorption of phenanthrene in sediments. Chemosphere 90, 1973-1979.
470
Zhang, Y.L., Wang, L.Y., Li, W.Y., Xu, H.J., Shi, Y.C., Sun, Y.T., Cheng, X., Chen, 22
ACCEPTED MANUSCRIPT X.Y., Li, Y.T., 2017. Earthworms and phosphate-solubilizing bacteria enhance
472
carbon accumulation in manure-amended soils. Journal of Soils and Sediments 17,
473
220-228.
AC C
EP
TE D
M AN U
SC
RI PT
471
23
ACCEPTED MANUSCRIPT
Table 1.Characteristics of the soil samples from the PRB.
2.52±0.70 1.71±0.60 3.26±0.60 2.27±1.00 3.07±0.80 2.93±0.10 3.09±1.30 2.68±2.00 5.49±1.00 5.63±1.80 9.56±2.50 8.43±1.20 6.55±0.20 4.13±0.10 5.89±1.20 1.36±0.70 2.14±1.30 2.31±0.80 4.53±1.10 3.76±1.20 3.24±0.30 3.18±1.00 2.10±0.20 1.63±0.10 1.98±0.70 2.50±0.30 2.23±0.80 1.65±0.30 4.44±1.60 2.92±0.60 2.30±0.30 2.85±0.50 5.25±2.10 5.01±1.10 7.28±1.60 4.27±0.70 5.51±0.60 3.73±0.40
MnDCB (mg kg-1) 356±187 288±76.6 38.2±20.4 54.0±36.3 20.3±18.1 46.5±39.3 40.2±9.50 56.1±20.0 68.1±46.5 71.4±56.1 500±200 457±245 15.4±0.60 43.3±1.00 62.8±43.7 200±100 15.8±7.70 14.8±4.80 55.2±18.7 305±132 46.8±25.4 84.7±44.1 17.5±0.60 16.65±2.00 23.8±3.40 59.4±26.5 76.1±54.3 226±58.3 80.7±35.3 45.1±11.7 35.6±14.8 33.4±17.2 108±8.60 63.3±41.0 429±89.4 493±31.8 603±59.5 587±151
MnTamm (mg kg-1)
CEC (cmol kg-1)
RI PT
10.9±3.80 14.3±4.70 16.0±0.80 19.6±5.40 8.40±6.50 11.8±7.40 6.79±2.90 8.18±4.10 19.7±3.60 19.5±6.10 19.8±3.30 23.2±4.50 20.6±0.40 28.6±0.60 12.5±1.80 12.4±0.50 6.27±4.70 6.92±2.30 16.9±2.40 27.0±10.7 10.9±1.20 11.9±1.50 4.39±0.20 4.75±0.10 6.47±1.60 8.60±1.30 15.3±8.00 12.6±1.40 27.3±12.3 36.1±1.90 14.5±11.3 11.4±4.60 28.9±14.4 18.1±4.50 25.4±2.30 26.7±2.90 26.1±4.10 25.5±4.20
FeTamm (g kg-1)
200±37.2 173±31.9 22.0±16.9 24.0±13.0 11.3±5.50 25.6±27.7 23.8±7.40 37.6±21.3 24.2±15.1 28.3±28.1 458±161 406±204 4.64±0.30 19.3±0.60 43.0±35.9 174±90.7 9.52±4.10 10.2±3.00 33.7±8.50 221±48.8 38.4±21.2 73.2±49.4 13.7±1.00 8.26±0.10 21.9±2.90 58.0±27.4 68.4±76.2 203±62.2 43.8±26.6 13.6±0.01 27.6±11.0 24.5±8.00 69.6±34.1 44.6±29.1 363±95.8 338±47.6 414±49.4 386±117
SC
7.15±0.80 7.52±0.10 6.26±0.80 5.65±0.90 5.81±0.40 6.06±1.00 6.32±1.00 6.20±1.50 6.36±0.80 5.93±1.00 6.32±0.80 5.85±1.10 5.25±0.20 6.10±0.10 5.17±0.40 6.84±0.90 5.32±0.20 5.05±0.20 5.03±0.30 6.39±1.00 5.79±0.70 5.84±1.00 5.06±0.10 4.74±0.10 5.09±0.20 5.57±0.20 5.74±1.00 6.05±0.50 4.42±0.60 3.83±0.03 5.60±0.10 5.46±0.10 5.59±1.30 5.24±0.30 6.04±0.30 5.12±0.40 5.92±0.10 5.27±0.40
FeDCB (g kg-1)
M AN U
247±80.9 138±34.6 100±38.8 53.3±2.2 102±29.2 92.0±16.3 237±18.2 99.4±4.40 176±4.10 127±25.2 83.0±2.40 29.0±2.00 189±2.00 84.5±2.40 77.6±30.9 43.4±6.80 84.7±31.4 47.2±1.30 139±13.5 60.4±11.1 176±3.90 105±1.10 175±6.30 180±10.7 72.5±0.80 49.5±1.90 83.3±2.90 41.0±0.90 69.4±27.6 60.3±3.60 48.6±1.50 32.4±1.70 77.7±27.9 34.2±1.60 63.8±1.90 50.7±1.50 59.6±4.00 51.6±3.40
pH (H2O)
TE D
70.0±34.6 59.8±25.9 40.5±8.54 26.2±4.12 38.9±4.74 33.9±12.7 53.5±40.2 18.8±26.1 36.9±8.80 41.2±8.30 44.7±7.83 33.2±9.47 28.4±0.50 28.4±1.07 44.8±9.91 23.5±7.77 36.8±3.88 35.4±7.81 41.2±3.80 33.8±2.71 39.8±2.89 29.5±5.68 46.4±3.03 42.7±0.94 46.2±17.2 36.6±12.9 41.2±7.42 20.4±8.66 32.7±14.0 15.8±0.88 10.7±3.66 12.9±7.33 37.4±13.4 20.4±7.77 37.7±15.3 27.7±7.09 24.0±2.21 22.1±1.67
DOC (mg kg-1)
EP
LP1(3) LN1(3) LP2(3) LN2(3) LP3(3) LN3(3) SP1(3) SN1(3) SP2(9) SN2(9) SP3(3) SN3(3) SP4(3) SN4(3) DP1(3) DN1(3) DP2(3) DN2(3) DP3(3) DN3(3) DP4(3) DN4(3) DP5(3) DN5(3) DP6(3) DN6(3) DP7(3) DN7(3) AP1(4) AN1(3) AP2(4) AN2(4) AP3(3) AN3(4) GP1(6) GN1(6) GP2(6) GN2(6)
TOC (g kg-1)
AC C
Samplea (nb)
10.3±2.82 14.4±7.44 6.84±1.53 5.91±3.40 6.25±0.28 7.31±1.96 8.52±1.81 7.61±0.48 9.61±1.70 10.8±2.67 8.95±1.00 7.83±0.65 11.4±0.11 14.3±0.34 7.87±3.25 4.96±1.31 5.72±0.76 5.16±0.55 6.88±0.44 9.91±1.51 7.19±1.06 6.17±0.73 7.54±0.56 5.55±0.07 4.54±0.39 4.05±0.68 6.11±1.31 3.57±0.26 5.09±1.71 4.71±0.39 5.32±0.82 5.51±1.40 6.90±2.81 4.91±0.71 17.8±4.74 18.1±3.23 15.6±3.52 15.4±1.22
Sand
Silt
Clay
0.26±0.06 0.15±0.03 0.29±0.13 0.26±0.15 0.31±0.18 0.34±0.19 0.25±0.08 0.17±0.03 0.20±0.08 0.15±0.05 0.20±0.08 0.21±0.10 0.03±0.01 0.03±0.01 0.33±0.02 0.53±0.02 0.41±0.06 0.45±0.06 0.32±0.03 0.31±0.01 0.31±0.09 0.31±0.04 0.29±0.02 0.31±0.02 0.55±0.05 0.59±0.01 0.36±0.06 0.42±0.03 0.19±0.07 0.24±0.07 0.42±0.07 0.38±0.05 0.21±0.03 0.38±0.02 0.03±0.01 0.03±0.01 0.02±0.01 0.02±0.01
0.56±0.03 0.60±0.03 0.45±0.09 0.46±0.10 0.48±0.14 0.42±0.16 0.57±0.03 0.60±0.06 0.49±0.07 0.47±0.04 0.53±0.05 0.54±0.03 0.58±0.01 0.57±0.01 0.36±0.03 0.23±0.03 0.31±0.04 0.34±0.08 0.34±0.03 0.35±0.09 0.48±0.02 0.46±0.04 0.47±0.01 0.47±0.02 0.25±0.04 0.22±0.01 0.36±0.05 0.37±0.02 0.53±0.06 0.41±0.04 0.39±0.09 0.47±0.07 0.55±0.02 0.42±0.01 0.50±0.06 0.47±0.08 0.48±0.02 0.49±0.01
0.18±0.05 0.25±0.05 0.26±0.06 0.28±0.05 0.21±0.05 0.24±0.03 0.18±0.06 0.23±0.04 0.31±0.06 0.38±0.04 0.27±0.06 0.25±0.11 0.39±0.01 0.40±0.01 0.30±0.02 0.24±0.03 0.28±0.03 0.21±0.09 0.34±0.03 0.34±0.09 0.21±0.09 0.23±0.05 0.24±0.02 0.22±0.01 0.20±0.02 0.19±0.01 0.28±0.02 0.21±0.04 0.28±0.03 0.35±0.10 0.19±0.04 0.15±0.07 0.24±0.02 0.20±0.02 0.47±0.06 0.50±0.08 0.50±0.01 0.49±0.01
a Sample name is consist of parent materials and land use type. L = limestone, S = sand shale, D = diluvium, A = alluvium, G = granite, P = paddy soil, N = non-paddy soil. b The n represents sample number. Value: mean ± standard deviation.
ACCEPTED MANUSCRIPT Table 2 Stepwise regression models of factors affecting free and complex metal in five parent material-derived soils (P < 0.05). Parent materials
Dissolve d metal
Limeston e
Free Cd
No stepwise regression equation
Complex ed Cd Free Pb Complex ed Pb Free Cd
No stepwise regression equation
Alluvium
Complex ed Pb
—
—
—
—
—
RI PT
0.002 0.000
—
—
—
—
—
—
SFPb=0.124-0.520[Sand] SCPb=0.034+0.017[WAT-Pb]-0.004[FeTamm] -0.001[TOC]-0.001[CEC]+0.015[Clay]+0.38 2[E4/E6] DFCd=7.150+0.071[DOC]-34.772[Clay] DCCd=0.082+0.022[E2/E3]-0.019[RES-Cd]
0.889 1.000
0.861 1.000
0.005 0.000
0.767 0.706
0.715 0.640
0.001 0.004
DFPb=-0.129+0.091[E4/E6]+0.001[OXID-Pb] DCPb=-0.030+0.020[E4/E6]+0.001[OXID-Pb ] No stepwise regression equation ACCd=0.415-0.008[TOC]
0.877 0.842
0.850 0.807
0.000 0.000
— 0.990
— 0.984
— 0.005
0.995 1.000
0.993 1.000
0.002 0.000
0.912 1.000
0.868 1.000
0.045 0.000
1.000
1.000
0.000
—
—
—
No stepwise regression equation
SC
No stepwise regression equation
AFPb=-0.692+6.116[∆logK] ACCd=0.029+0.684[∆logK]-0.005[SUVA254]+ 0.001[E2/E4] GFCd=-12.109+530.095[Sand] GCCd=-0.914+44.473[Sand] -0.008[TOC]-0.157 [E4/E6] GFPb=0.181-0.327[Silt]+0.0004[E2/E4] -0.002[SUVA254] No stepwise regression equation
AC C
Granite
Free Cd Complex ed Cd Free Pb Complex ed Pb Free Cd Complex ed Cd Free Pb Complex ed Pb Free Cd Complex ed Cd Free Pb
—
0.995 1.000
EP
Diluvium
p value
0.997 1.000
M AN U
Complex ed Cd Free Pb Complex ed Pb
LFPba =1.239-0.162[pH] LCPbb=0.412-0.049[pH]-0.003[ORG-Pb]
TE D
Sandshale
R2adj
R2
Stepwise regression equations
LFMetal and LCMetal mean free and complex metals concentrations in limestone soils. [WAT-metal] means water soluble metal, [OXID-metal] means Fe-Mn oxides bound metal, [RES-metal] means residual metal, [FeTamm] means Tamm's extractable iron.
ACCEPTED MANUSCRIPT Captions Fig.1. Sampling sites. The pollution sources and parent materials nearby the sites are also shown in the map. Fig.2. Total concentrations of Cd and Pb in paddy and dry soils from the different
RI PT
parent materials. The boxes span from the 25th percentile to the 75th percentile. The solid and dash lines inside the box represent the sample median and mean values, respectively. The whiskers denote the interval between the minimum and maximum value, respectively. The full circles point to outliers. L = limestone, S = sand shale, D
SC
= diluvium, A = alluvium, G = granite, P = paddy soil, N = non-paddy soil, CK =
M AN U
background soil.
Fig.3. Effects of different land-use types on dissolved Cd and Pb in soil solution by SC-DMT. Letters (a, b, c, etc.) above each bars show statistical significance at p<0.05. The capital is for comparison between paddy and non-paddy soils, and the small letter is for comparison between free and complex metals. L = limestone, S = sand shale, D
TE D
= diluvium, A = alluvium, G = granite, P = paddy soil, N = non-paddy soil. Fig.4. Speciation distributions of Cd (A) and Pb (B) of the studied soils. L = limestone, S = sand shale, D = diluvium, A = alluvium, G = granite, P = paddy soil, N = non-paddy soil.
EP
Fig.5. Total and dissolved metals (Cd (a) and Pb (b)) of the soils from the five parent materials. The correlation curves and equations were obtained based on the samples
AC C
without the name markers. A = alluvium, D = diluvium, G = granite, L = limestone, S = sand shale, P = paddy soil, N = non-paddy soil.
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Fig.1
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Fig.2
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Fig.3
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Fig.4
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Fig.5
ACCEPTED MANUSCRIPT
Highlights Cd and Pb in paddy soils are more labile than in dry soils due to stable equilibrium.
RI PT
Pb activity is very sensitive to pH change in limestone soils. High sand content enhances Cd activity in granite soils.
DOM greatly impacts Cd and Pb activities in sand shale, diluvium, and
AC C
EP
TE D
M AN U
SC
alluvium soils.
S0045-6535(17)30701-4
DOI:
10.1016/j.chemosphere.2017.05.007
Reference:
CHEM 19221
To appear in:
ECSN
Received Date: 19 February 2017 Revised Date:
9 April 2017
Accepted Date: 1 May 2017
Please cite this article as: He, S., Lu, Q., Li, W., Ren, Z., Zhou, Z., Feng, X., Zhang, Y., Li, Y., Factors controlling cadmium and lead activities in different parent material-derived soils from the Pearl River Basin, Chemosphere (2017), doi: 10.1016/j.chemosphere.2017.05.007. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT 1
Factors Controlling Cadmium and Lead Activities in Different
2
Parent Material-Derived Soils from the Pearl River Basin
3
Shuran Hea, Qin Lua, Wenyan Lia, Zongling Rena, Zhen Zhoua, Xiao Fenga, Yulong
5
Zhanga*, Yongtao Lib*
6
a
7
University, Guangzhou, 510642, P.R. China
8
b
9
University, Guangzhou, 510642, P.R. China
RI PT
4
SC
College of Natural Resources and Environment, South China Agricultural
10 11
M AN U
Key Laboratory of Arable Land Conservation, MOA, South China Agricultural
Corresponding authors:
13
Dr. Yulong Zhang, [email protected]
14
Dr. Yongtao Li, [email protected],
15
South China Agricultural University, No. 483, Wushan Road, Guangzhou 510642, P.R.
16
China
17
Phone: +86-20-38297890
18
Fax: +86-20-85280292
20
EP
AC C
19
TE D
12
1
ACCEPTED MANUSCRIPT Abstract
22
Labile metals in agricultural soils are available to crops and thus pose a great health
23
risk for human beings. Therefore, factors influencing heavy metal activity are of
24
interest to researchers. In this study, a total of 142 soil samples representing 5 typical
25
parent materials in the Pearl River Basin (PRB), China were collected to investigate
26
factors impacting the distribution of labile Cd and Pb in the soils. The results showed
27
that the labile fractions accounted for 0.03%-14.7% for Cd and 0.01%-0.39% for Pb
28
of the total metals, and the labile fractions were linearly correlated to their
29
corresponding total contents. The step regression analyses suggested that the key
30
factors impacting labile Cd and Pb varied in different parent material soils. Pb activity
31
was highly sensitive to pH in alkaline limestone soils. The quartz sand remained in
32
granite-produced soils enhanced Cd activity. And dissolved organic matter (DOM)
33
compositions considerably influenced Cd and Pb activities in sand shale, diluvium,
34
and alluvium soils. Land use impacts heavy metal activities. The labile Cd and Pb in
35
paddy soils were higher than those in non-paddy soils, although total metals in the
36
soils were comparable. It could be ascribed to the long-term equilibrium of metals
37
between the solution and solid phases of the paddy soils. The results provide a
38
theoretical basis for preliminary prediction of heavy metal activity and provide a
39
technical support for heavy metal activity management and pollution control based on
40
soil parent materials.
AC C
EP
TE D
M AN U
SC
RI PT
21
41 42
Keywords: metal activity; agricultural soil; land use; parent material; DMT 2
ACCEPTED MANUSCRIPT 43
1. Introduction Heavy metal contamination in the environment, especially the agricultural soils,
45
is drawing a great attention due to its high risk to human beings (Yang et al., 2013;
46
Qayyum et al., 2017). Relative to the total metal, the activity of heavy metal in soils is
47
considerably associated with the uptake of crop and thus human health (Sastre et al.,
48
2004; Tai et al., 2013). Therefore, illustrating the factors affecting the heavy metal
49
activity in agricultural soils is significant for the evaluation of their environmental
50
risks.
SC
M AN U
51
RI PT
44
The activities of heavy metals are influenced by soil properties and compositions, such as pH, organic matter, and iron oxides (Sauve et al., 2000a, 2000b; Lu et al.,
53
2003; Chen et al., 2008; Yu et al., 2016a, 2016b). Sauve et al. (2000a) reported that
54
pH, soil organic matter (SOM), and total contents of Ni, Cu, Cd, Zn can be used to
55
predict the percentages of labile fraction based on a big dataset. In the heavily
56
contaminated soils, the metals cannot be effectively fixed due to limited sorption
57
capacities, resulting in high activities of ions (Tsang et al., 2006). However, the heavy
58
metals tightly bound to varying matrices of soil have lower risks for transportation to
59
plants (Jacquat et al., 2009). Parent rocks substantially influence both the properties
60
and trace metal remaining of soils. Weathering of parent rocks produces the soil
61
primary and secondary minerals such as, quartz, kaoline, and montmorillonite. The
62
minerals are different in particle size, cation exchange capacity, metal species, etc.
63
which endows soils with corresponding properties. For some areas without any
64
pollution source, the metal compositions and contents are determined by the parent
AC C
EP
TE D
52
3
ACCEPTED MANUSCRIPT materials (Mico et al., 2006). Besides, the behaviors of the remaining metals can be
66
notably influenced by the parent material-related features. For instance, calcareous
67
soil is originated from limestone or dolomite and contains high carbonates and
68
organic matter (Chen et al., 2000). As carbonates form precipitates with heavy metals
69
and organic matters fix heavy metals with functional groups such carboxyl and
70
hydroxyl, they would enhance the accumulation of heavy metals. Considering the
71
important and complex role of parent materials on soil properties and heavy metals, it
72
is significant to understand the effects of parent materials on the activities of heavy
73
metals, as well as the factors impacting metal activities in different parent
74
material-derived soils.
M AN U
SC
RI PT
65
In this work, the Pearl River Basin (PRB) in China is selected for investigating
76
the heavy metal distributions in different parent material-derived soils of different
77
pollution levels and related impacting factors. On one hand, many mines and related
78
industries, such as, Dabaoshan Mine, Fankou Mine and Dongguan industrial district,
79
are located in the study area (Wong et al., 2002; Yin et al., 2016). Consequently,
80
heavy metal pollution in the agricultural soils, especially Cd and Pb, is severe in its
81
tributaries due to wastewater irrigation, atmospheric deposition, etc. On the other
82
hand, various parent rocks (residual and transported) including limestone, granite,
83
sand-shale, diluvium, and alluvium are extensively distributed in this area (Chen et al.,
84
2000).
AC C
EP
TE D
75
85
The aims of the present study were to investigate the distributions of labile Cd
86
and Pb in five representative parent material soils from the PRB; and more 4
ACCEPTED MANUSCRIPT 87
importantly, to unveil the factors affecting Cd and Pb activities in the soils.
88
2. Materials and methods
89
2.1. Sampling sites The catchment of PRB is up to 44.21×104 km2. One hundred forty-two (142)
91
surface soil (0-20 cm) samples from the PRB, including 54 samples in Xijiang River
92
Basin, 64 samples in Beijiang River Basin, and 24 samples in Pearl River Delta (Fig.
93
1) were collected. The sampling sites were so selected to include five representative
94
parent materials (limestone (L), sand shale (S), diluvium (D), alluvium (A), and
95
granite (G)) and reflect different pollution levels (e.g., background level and pollution
96
levels). Three zones including L1 (a background site), L2 polluted by a non-ferrous
97
metals smelter in Hechi City, and L3 slightly polluted zone, were located in the
98
limestone area. Four zones including S1 (a background site), S2 and S3 polluted by
99
the Dachang tin-polymetallic mining and located in the upper and lower catchment of
100
the Diaojiang River, and S4 slightly polluted, were in the sand shale area. Seven zones
101
including D1 and D2 polluted by the Fankou lead-zinc deposit, D3 and D4 polluted
102
by the Danxia lead-zinc smelter, D5 (a background site), D6 and D7 polluted by the
103
Maba lead-zinc smelter, were in the diluvium area. Three zones including A1 polluted
104
by the Dabaoshan polymetallic sulfide meso-hypothermal deposit, A2 (a background
105
site), A3 slightly polluted, were in the alluvium area. Two zones including G1 and G2
106
polluted by the industrial activity in Zhongshan City, were in the granite area. In each
107
zone, 6-12 sampling sites were chosen and each site was ~300 m away from the each
108
other. Surface samples (~5 kg) from the paddy (P) and non-paddy (N) soil were
AC C
EP
TE D
M AN U
SC
RI PT
90
5
ACCEPTED MANUSCRIPT collected from May to July, 2015. The samples were termed a combination with
110
parent material, land use type, and number (Table 1). The samples were stored by
111
plastic bags and carried back to the laboratory. The air-dried samples were grounded
112
to pass through 2 mm and 0.15 mm sieves for the different analyses.
113
2.2. Soil characterization
RI PT
109
Soil pH was measured from a 2.5:1 (v/m) water to soil suspension using a pH
115
detector (PHS-3C, China). Cation exchange capacity (CEC) was analyzed using the
116
ammonium acetate saturation methods (Tan et al., 2017). Particle-size analysis was
117
performed using sedimentation and pipette method. Sands (50‒2000 µm), silts (2‒50
118
µm), and clays (<2 µm) were measured (Zhang et al., 2017). Fe and Mn oxides
119
including crystalline and amorphous Fe/Mn were separated and further quantitatively
120
determined by colorimetry (Gleyzes et al., 2002). Soil organic carbon (SOC) was
121
determined by oxidation with dichromate (Zhang et al., 2017). Dissolved organic
122
matter (DOM) in soils was water-extracted (water to solid ratio = 5:1; 1 h) by a shaker.
123
Dissolved organic carbon (DOC) content was detected by a total organic carbon (TOC)
124
analyzer (Elementar vario TOC, Germany). The DOM compositions were
125
characterized by an ultraviolet-visible (UV) spectrophotometer. The absorbances at
126
250, 254, 365, 400, 436, 465, 600, and 665 nm of the equilibrium solution were
127
measured. Specific UV-vis absorptions at 254 nm (SUVA254, L·g-1 cm-1) and ∆logK
128
were calculated by the Eq. 1 and Eq. 2, respectively (Kumada, 1988).
AC C
EP
TE D
M AN U
SC
114
129
SUVA =
130
ΔlogK= logA400-logA600
(1) (2) 6
ACCEPTED MANUSCRIPT where, Ai is the absorbance at i nm, l (cm) is the cell path length and DOC is the
132
dissolved organic carbon content (mg L-1) of the soil solution. SUVA254 and ∆logK are
133
frequently used as the indicators for aromaticity and humification of DOM (Ni et al.,
134
2016). The E2/E3, E2/E4, and E4/E6 are the ratios of the absorbance at 250 nm over
135
365 nm, 254 nm over 436 nm, 465 nm over 665 nm (Table S1, Supplementary Data).
136
The former two ratios reflect the humification of DOM, meanwhile the latter ratio
137
represents aromaticity of DOM (Helms 2006).
138
2.3. Heavy metal analysis
M AN U
SC
RI PT
131
The total concentrations of Cd and Pb in soils were measured by the
140
HCl-HNO3-HClO4-HF digestion method (Li et al., 2009). The digestion solution were
141
analyzed by a flame atomic absorption spectrometry (AAS, Hitachi Z-2300, Japan). A
142
six step sequential extraction of heavy metals in soils was carried out based on the
143
modified Tessier method (Ma and Rao, 1997). The extraction methods were detailed
144
in the Table S2 (Supplementary Data). The Cd and Pb concentrations in the different
145
fractions were also determined by an AAS (Hitachi Z-2300, Japan).
EP
TE D
139
Soil Column Donnan membrane technology (SC-DMT) is a recently developed
147
tool to measure the free and complex metals in soil solution. It decreases the
148
disturbance of the soil chemical compositions and to a great extent simulates the
149
activities of heavy metals under water-soil equilibrium (Li et al., 2009; Ren et al.,
150
2015; Jones et al., 2016). Furthermore, the free metal ions together with ligand
151
complexes can be determined using the SC-DMT system. A soil of 5 g was loaded in
152
the system and the equilibrium time for the solid-water phases was obtained after 48 h.
AC C
146
7
ACCEPTED MANUSCRIPT The measurement procedure was detailed in Li et al. (2009). Since the Cd and Pb
154
concentrations in the equilibrium solutions were low, the metals in the donor and
155
acceptor solutions were determined by an inductively coupled plasma mass
156
spectrometry (ICP-MS, XSeries 2, Thermo Scientific, USA).
157
2.4. Statistical analyses
RI PT
153
Data analysis was performed using the SPSS statistics 20.0 and R software
159
combined with ADE-4. Statistical significance of the differences in heavy metals and
160
properties of soils was tested by one-way ANOVA. Stepwise regression analyses and
161
principle component analyses were conducted to extract factors affecting the
162
distribution of reactive metals in soils.
163
3. Results and discussion
164
3.1. Soil physicochemical properties
TE D
M AN U
SC
158
All the physicochemical properties of the PRB soils are shown in Table 1. TOC
166
and DOC of the soils ranged from 10.7 ± 3.7 to 70.0 ± 34.6 g kg-1 and from 29.0 ± 2.0
167
to 247 ± 80.9 mg kg-1, respectively. The TOC and DOC in limestone and sand-shale
168
soils were higher (P < 0.05) than the other parent material soils. The majority of pH in
169
soils was lower than 7, indicating an acidic soil environment. This could lead to a
170
potential of increasing the Cd and Pb activities (Sauve et al., 2000b). The limestone
171
soils was slightly alkaline, and the pH of L1 (avg. 7.15 and 7.52) was highest among
172
all the samples. Besides, both the relatively high TOC and DOC concentrations were
173
also observed in L1. These soil properties were consistent with their parent material of
174
limestone (Chen et al., 2000). The Fe and Mn oxides are key constituents for Cd and
AC C
EP
165
8
ACCEPTED MANUSCRIPT Pb accumulation in soils. Amorphous Fe oxides are one of the important factors
176
decreasing Cd activity in red soils, based on an investigation about Cd uptake of rice
177
(Yu et al., 2016a). The crystalline and amorphous Fe contents in soils were in the
178
range from 4.39 ± 0.20 to 36.1 ± 1.90 g kg-1 and from 1.36 ± 0.70 to 9.56 ± 2.50 g
179
kg-1. The crystalline and amorphous Mn contents in soils ranged from 14.79 ± 4.80 to
180
603.1 ± 59.5 mg kg-1 and from 8.26 ± 0.1 to 458.2 ± 160.8 mg kg-1. The contents of
181
Fe and Mn oxides in the granite soils were generally higher than the other. CEC
182
reflects the sorption capacity of metals to soils. The highest CEC values were
183
observed in GP1 (17.8 ± 4.70 cmol kg-1) and GN1 (18.1 ± 3.20 cmol kg-1). The
184
particle size distribution results indicated that slit (42%-61%, 47%-60%, 22%-48%,
185
39%-55% and 47%-50% for limestone, sand-shale, diluvium, alluvium, and granite,
186
respectively) is generally the most abundant fraction in soils. The sand and clay
187
fractions of limestone, sand-shale, diluvium, alluvium, and granite ranged from
188
15%-34%, 3%-25%, 29%-59%, 20%-42%, 2%-3%, and 18%-28%, 18%-40%,
189
19%-35%, 15%-35%, 47%-50%, respectively.
EP
TE D
M AN U
SC
RI PT
175
The DOM compositions are tightly related to the speciation of metals in soils.
191
Ren et al. (2015) found that fulvic acid is the most important fraction for binding Pb
192
and Cu in soil solution, due in part to the high polar functional groups. A series of
193
parameters derived from the absorbance values to some extent indicate the functional
194
groups in DOM (Helms, 2006). Table S1 shows that SUVA254 of soils ranged from
195
9.86 ± 0.95 to 28.2 ± 9.96 L g-1 cm-1. The ratios of E2/E3, E2/E4, and E4/E6 were in
196
the range of 5.45±0.70 to 16.33±3.72, 15.01±13.00 to 47.36±15.96, 0.47±0.22
AC C
190
9
ACCEPTED MANUSCRIPT to 2.19±1.25, respectively. ∆logK of the soils varied from 0.18±0.05 to 0.83±0.74.
198
The data analysis revealed that there was no significant difference in DOM
199
compositions among varying parent materials or land use types.
200
3.2. Heavy metal distribution
RI PT
197
The total concentrations of Cd and Pb in soils originated from the five parent
202
materials are shown in Fig. 2. Compared with the background agricultural soils, the
203
Cd and Pb concentrations of soils around the pollution sources were significantly
204
higher. This suggested a significant input of metals from the nearby industrial
205
activities. Except for the background sites, the majority of Cd in the studied soils was
206
higher than the threshold level (0.3 mg kg-1) of the China Environmental Quality
207
Standards for Soil (GB15618-1995, grade II for the agriculture land). In contrast,
208
most of the Pb concentrations of the soils were lower than its threshold level (250 mg
209
kg-1). Heavy pollution of Pb was extensively observed in the diluvium soils (Fig. 2).
210
And the Pb concentrations of the diluvium soils were significantly higher than the
211
other parent material soils. The compositions and concentrations of the metals largely
212
depended on the type of pollution sources and the distance from the sources. All the
213
diluvium sites were nearby the lead-zinc mining and its smelters in the Renhua and
214
Qujiang Districts. Cd and Pb concentrations in D1 were higher than those in D2 (not
215
shown), which is consistent with their distance between the Fankou mining and the
216
sampling sites. The above observation indicated a great influence of industrial
217
activities on accumulations of heavy metals in agricultural soils. In the granite area,
218
the metal pollution sources are extensively distributed due to the well-developed
AC C
EP
TE D
M AN U
SC
201
10
ACCEPTED MANUSCRIPT 219
industry in Zhongshan City. However, Cd (0.32-0.56 mg kg-1) and Pb (22.47-43.54
220
mg kg-1) pollutions were not severe as expected, which might be due to low Cd and
221
Pb input from the local industry. The total dissolved metals (free and ligand complex forms) with DMT system
223
represents the most labile fractions and provides more convincing indicators for
224
potential risk and toxicity of metal pollution soils. The dissolved fractions only
225
accounted for a minor percentages (0.03%-14.7% for Cd and 0.01%-0.39% for Pb) of
226
total metal in the soils. The values are comparable with previous studies (Li et al.,
227
2009; Ren et al., 2015). The free and complex Cd concentrations were in the range
228
from 0.05 to 19.7 µg L-1 and from 0.02 to 0.76 µg L-1, respectively (Fig. 3a). And the
229
free and complex Pb concentrations were in the range from 0.15 to 74.1 µg L-1 and
230
from 0.10 to 26.5 µg L-1, respectively (Fig. 3b). The free Cd concentrations were avg.
231
22 times higher than that of the complex Cd, whereas the free Pb was avg. 3 times
232
higher than its complex form. This phenomena indicated the ligands in soil
233
equilibrium solution were preferentially bound to Pb relative to Cd. This is consistent
234
with Ren et al. (2015)’s observation that in soil solution Pb is dominant in complex
235
form, while Cd mainly exists as free ion. Previous studies also have demonstrated that
236
Pb tends to be much more strongly bound to organic matter in soil solution than Cd
237
(Ge et al., 2005).
AC C
EP
TE D
M AN U
SC
RI PT
222
238
It is worth noting that the total dissolved, free, and complex Cd and Pb in paddy
239
soils were extensively higher than their corresponding non-paddy soils (Fig. 3),
240
although the total Cd and Pb concentrations in the two type soils were comparable. 11
ACCEPTED MANUSCRIPT This occurred in all the five parent material areas, indicating a notable effect of land
242
use on distributions of liable metals. The phenomenon could be explained by varying
243
sorption equilibrium patterns of metals in paddy and non-paddy soils. The paddy soils
244
are usually long-term flooded, resulting in a stable equilibrium of metal sorption
245
between water and soil components. In contrast, the dry soils are periodically irrigated.
246
The frequent alternation of wetting and drying would lead to repeated equilibrium
247
between heterogeneous solid components and water. The soil components, such as,
248
lipids, non-hydrolysable carbons, Fe oxides, clay silicate vary in sorption capacity and
249
dynamics for different pollutants (Weng et al., 2001; Zhang et al., 2013, 2014; Liu et
250
al., 2015, Yu et al., 2016b). Consequently, metals tend to preferentially accumulate in
251
strong sorption components of soils, since metal desorption from weak sorption
252
components are quick and easy (Weng et al., 2001). This speculation can be further
253
supported by the metal fractions in soil constituents based on the sequential extraction.
254
The Cd and Pb bound to organic and residual components in soils are considered to be
255
most stable fractions (Fest et al., 2008; Yang et al., 2013). Fig. 4 shows that the
256
percentages of Cd and Pb bound to organic and residual fractions in non-paddy soils
257
are generally higher than those in paddy soils for the five parent materials. Another
258
reason for the difference in metal availability between paddy and non-paddy soils
259
might be the different Eh in the two types of soils. In paddy soils, the low Eh leads to
260
the transformation of Fe(III) to Fe(II). Consequently, Fe(III)-bound Cd and Pb are
261
released, resulting in high Cd and Pb availability (Yu et al., 2016a). This preferential
262
accumulation of metals in the stable fractions may be the main reason for the
AC C
EP
TE D
M AN U
SC
RI PT
241
12
ACCEPTED MANUSCRIPT 263
contrasting distributions of free and complex Cd and Pb in paddy and non-paddy
264
soils.
265
3.3. Factors affecting the activities of Cd and Pb Besides land use, the labile Cd and Pb were largely dependent on the total metal
267
amount in soils. The correlation analyses between total and dissolved concentrations
268
of Cd and Pb were performed (Fig. 5a and 5b). One fitted linear curve with R2 = 0.72
269
and P < 0.001 for Cd was obtained. The fitted equation was y = 30073x + 16.2. If the
270
Cd contents in pollution soils ranged from 0.2 to 6 mg kg-1, the proportions of labile
271
Cd would decrease from 11.1% to 3.2% based on the equation. The dissolved Cd is
272
considered as a multi-equilibrium fraction between solution and soil constituents (Ren
273
et al., 2015). Consequently, the dissolved Cd content is positively correlated to the Cd
274
fixed in soil constituents. And the majority of Cd in soils was bound to the solid
275
constituents and determined the total Cd concentration in the studied soils (Fig. S1).
276
Thus, it is not surprising to obtain the significant correlation between labile and total
277
Cd in soils. Some exceptions were usually presented in heavily polluted sites (> 1.2
278
mg kg-1). The percentages of labile Cd relative to total Cd in DN1, SP1, LP1, LN1,
279
SP3, and SN3 were lower than the modeled level. This could be partly due to soil pH.
280
The pH in DN1 (7.03), LN1 (7.58), LP1 (7.70), SP3 (6.87), and SN3 (5.88) were
281
generally higher than that of the other soils (avg. 5.34). An alkaline condition is
282
known to favor Cd precipitation and thus prevent Cd activity in soils (Sauve et al.,
283
2000a, 2000b). In contrast, the percentages of labile Cd relative to total Cd in DP3
284
and DP4 were higher than the modeled level. This could be related to the contents and
AC C
EP
TE D
M AN U
SC
RI PT
266
13
ACCEPTED MANUSCRIPT compositions of DOM. As shown in the Table S3 (Supplementary Data), the DOC
286
contents in DP3 and DP4 were 144 and 173 mg kg-1, respectively, which was far
287
higher than the average value of 75.7 mg kg-1 for the other samples. It is observed that
288
DOM increased the solubility of ions from soils (Park et al., 2011). Furthermore,
289
lower ∆logK (avg. 0.24) in DP3 and DP4 relative to the other (avg. 0.35) meant more
290
acidic functional groups of DOM with high humification. These acidic groups such as
291
carboxyl and phenolic hydroxyl groups are active ligand for metals and favor the
292
solubility of Cd from soils (Simmler et al., 2013).
M AN U
SC
RI PT
285
The distributions of labile Pb in soils were similar to the Cd pattern (Fig. 5b).
294
The fitted equation was y=236.5x + 0.19 (R2 = 0.78, P < 0.001). Based on the
295
modeled data, the labile Pb accounted for ~0.024% when its total contents were in the
296
range from 50 to 400 mg kg-1. The portions of labile Pb in AP1 and AN1 were higher
297
than that of the modeled level, whereas the portions of Pb in SN3, SP3, DP1, and
298
DN1 were lower than that of the modeled level. The phenomena could be ascribed to
299
the soil pH and components. The lowest pH values were 3.85 for AP1 and 4.21 for
300
AN1. Strong acidic condition enhances the solubility of Pb in soils (Sauve et al.,
301
2000a, 2000b). Besides, the lower CEC (4.58 cmol kg-1 for AP1 and 4.33 cmol kg-1
302
for AN1) relative to the average level (8.36) of the other soils suggested a lower
303
sorption capacity (Qayyum et al., 2016). In contrast to the others, DN1, SP3, and SN3
304
had higher pH, contents of amorphous Fe and Mn oxides, which is benefit to the Pb
305
fixation in soils.
AC C
EP
TE D
293
14
ACCEPTED MANUSCRIPT Step regression analyses were preformed to extract the key variables for
307
influencing the Cd and Pb activities in soils originated from different parent materials.
308
Only the regression equations in significant level (P < 0.05) are shown in Table 2. The
309
key variables extracted in the equations varied greatly for different parent material
310
soils. Generally, pH, organic matter, and size distribution seemed to be the important
311
factors affecting the Cd and Pb activities. In limestone soils, Pb activity is highly
312
sensitive to pH decrease. The limestone soil usually is alkaline due to high content of
313
carbonates. Hydrogen ion increases the reaction with carbonates and releases the
314
carbonate-bounded Pb. In sand shale soils, free Pb was negatively correlated with
315
sand contents. The complex Pb was mostly influenced by E4/E6. Higher E4/E6 values
316
represent lower aromaticity, suggesting the possibility of higher proportions of
317
reactive groups such as carboxyl and sulfydryl groups. These reactive groups are
318
strong ligand for cations and thus considerably affects the solubility of heavy metals
319
in soils (Simmler et al., 2013; Shirvani et al. 2015). The mechanism could be also
320
presented for diluvium soils in which free and complex Pb were positively correlated
321
with E4/E6. In diluvium soils, free Cd is largely impacted by clay content. Clay is a
322
key sorption component for heavy metals in soils. High clay content would increase
323
sorption capacity for Cd. High E2/E3 represents the high humification degree of
324
DOM. The humification process produces the reactive functional groups such as
325
carboxyl, phenolic hydroxyl, alcoholic hydroxyl, and amino groups of DOM. This is
326
consistent with the positive correlation between complex Cd and E2/E3 in diluvium
327
soils. In alluvium soils, the complex Cd was negatively correlated with TOC. This is
AC C
EP
TE D
M AN U
SC
RI PT
306
15
ACCEPTED MANUSCRIPT consistent with the fact that organic matter is an important sorption component for Cd
329
in soils (Sauve et al., 2000b). In granite soils, free and complex Cd is mostly impacted
330
by sand. The sands in granite-derived soils mainly consist of quartz which is less
331
capable of sorbing Cd. Thus, higher portion of sand benefits Cd solubility in soils.
332
Some cases were hardly explained by the current understanding. For instance, the free
333
Pb is negatively correlated with silt content in granite soils. In alluvium soils, free and
334
complex Pb were negatively impacted by DOM humification degree as indicated by
335
∆logK (Table 2).
M AN U
SC
RI PT
328
336 337
4. Conclusions
Reactive Cd and Pb in agricultural soils pose a direct risk for crops and humans,
339
and factors influencing Cd and Pb activities in soils are of interest for people working
340
in soil pollution and remediation, agricultural safe production, and environmental
341
health. A series of soil properties were determined and were analyzed to extract the
342
key variables for understanding the distributions of reactive Cd and Pb in five
343
different parent material soils from the PRB. The results of the present study
344
demonstrated that the controlling factors for different parent material-derived soils
345
differed. For instance, pH was the key factor impacting free and complex Pb in
346
limestone soils. Sand played an important role in the distributions of free and complex
347
Cd in granite soils. In contrast, DOM compositions considerably influenced Cd and
348
Pb activities in sand shale, diluvium, and alluvium soils. Besides, land use greatly
349
influenced heavy metal activity. The reactive Cd and Pb in the paddy soils were
AC C
EP
TE D
338
16
ACCEPTED MANUSCRIPT higher than those in the non-paddy soils. Total metal noticeably impact Cd and Pb
351
activities in most of the soil samples, as evidenced by the significant linear
352
correlations between the total metal and their corresponding reactive fractions. And
353
effects of acidity and DOM of soils on the activities of Cd and Pb explained some
354
exceptions. In summary, in addition to the total concentration, taking effects of parent
355
materials, land use, acidity, and DOM into account is necessary for assessing the Cd
356
and Pb activities in agricultural soils.
AC C
EP
TE D
M AN U
SC
RI PT
350
17
ACCEPTED MANUSCRIPT 357
Acknowledgements This research was supported by the Natural Science Foundation of China
359
(U1401234 and 41601533), National Science & Technology Pillar Program
360
(2015BAD05B05 and 2014BADB14B01), National Key Research and Development
361
Program of China (2016YFD0800300), and China Postdoctoral Science Foundation
362
(2016M602479).
AC C
EP
TE D
M AN U
SC
RI PT
358
18
ACCEPTED MANUSCRIPT 363
References
364
Chen, J.S., Wang, F.Y., Li, X.D., Song, J.J., 2000. Geographical variations of trace
365
elements in sediments of the major rivers in eastern China. Environmental
366
Geology (Berlin) 39, 1334-1340. Chen, M., Li, X.-M., Yang, Q., Zeng, G.-M., Zhang, Y., Liao, D.-M., Liu, J.-J., Hu,
368
J.-M., Guo, L., 2008. Total concentrations and speciation of heavy metals in
369
municipal sludge from Changsha, Zhuzhou and Xiangtan in middle-south region
370
of China. Journal of Hazardous Materials 160, 324-329.
SC
RI PT
367
Fest, E.P.M.J., Temminghoff, E.J.M., Comans, R.A.J., van Riemsdijk, W.H., 2008.
372
Partitioning of organic matter and heavy metals in a sandy soil: Effects of
373
extracting solution, solid to liquid ratio and pH. Geoderma 146, 66-74.
M AN U
371
Ge, Y., MacDonald, D., Sauvé, S., Hendershot, W., 2005. Modeling of Cd and Pb
375
speciation in soil solutions by WinHumicV and NICA-Donnan model.
376
Environmental Modelling & Software 20, 353-359.
TE D
374
Gleyzes, C., Tellier, S., Astruc, M., 2002. Fractionation studies of trace elements in
378
contaminated soils and sediments: a review of sequential extraction procedures.
379
Trac-Trends in Analytical Chemistry 21, 451-467.
381
Helms, J.R., 2006. Spectral shape as an indicator of molecular weight in chromophoric dissolved organic matter. Old Dominion University.
AC C
380
EP
377
382
Jacquat, O., Voegelin, A., Kretzschmar, R., 2009. Soil properties controlling Zn
383
speciation and fractionation in contaminated soils. Geochimica Et Cosmochimica
384
Acta 73, 5256-5272.
385
Jones, A.M., Xue, Y., Kinsela, A.S., Wilcken, K.M., Collins, R.N., 2016. Donnan
386
membrane speciation of Al, Fe, trace metals and REEs in coastal lowland acid
387
sulfate soil-impacted drainage waters. Science of the Total Environment 547,
388
104-113.
389
Kumada, K., 1988. Chemistry of Soil Organic Matter. Elsevier. 19
ACCEPTED MANUSCRIPT 390
Li, Y.-T., Becquer, T., Dai, J., Quantin, C., Benedetti, M.F., 2009. Ion activity and
391
distribution of heavy metals in acid mine drainage polluted subtropical soils.
392
Environmental Pollution 157, 1249-1257. Liu, C., Yu, H.Y., Liu, C., Li, F., Xu, X., Wang, Q., 2015. Arsenic availability in rice
394
from a mining area: Is amorphous iron oxide-bound arsenic a source or sink?
395
Environmental Pollution 199, 95-101.
RI PT
393
Lu, Y., Gong, Z.T., Zhang, G.L., Burghardt, W., 2003. Concentrations and chemical
397
speciations of Cu, Zn, Pb and Cr of urban soils in Nanjing, China. Geoderma 115,
398
101-111.
400
Ma, L.Q., Rao, G.N., 1997. Chemical fractionation of cadmium, copper, nickel, and
M AN U
399
SC
396
zinc in contaminated soils. Journal of Environmental Quality 26, 259-264. Mico, C., Recatala, L., Peris, A., Sanchez, J., 2006. Assessing heavy metal sources in
402
agricultural soils of an European Mediterranean area by multivariate analysis.
403
Chemosphere 65, 863-872.
TE D
401
Ni, X., Yang, W., Tan, B., Li, H., He, J., Xu, L., Wu, F., 2016. Forest gaps slow the
405
sequestration of soil organic matter: a humification experiment with six foliar
406
litters in an alpine forest. Scientific Reports 6, 19744-19744.
EP
404
Park, J.H., Lamb, D., Paneerselvam, P., Choppala, G., Bolan, N., Chung, J.W., 2011.
408
Role of organic amendments on enhanced bioremediation of heavy metal(loid)
409
contaminated soils. Journal of Hazardous Materials 185, 549-574.
AC C
407
410
Qayyum, M.F., ur Rehman, M.Z., Ali, S., Rizwan, M., Naeem, A., Maqsood, M.A.,
411
Khalid, H., Rinklebe, J., Ok, Y.S., 2017. Residual effects of monoammonium
412
phosphate, gypsum and elemental sulfur on cadmium phytoavailability and
413
translocation from soil to wheat in an effluent irrigated field. Chemosphere 174,
414
515-523.
415
Ren, Z.-L., Tella, M., Bravin, M.N., Comans, R.N.J., Dai, J., Garnier, J.-M., Sivry, Y.,
416
Doelsch, E., Straathof, A., Benedetti, M.F., 2015. Effect of dissolved organic 20
ACCEPTED MANUSCRIPT 417
matter composition on metal speciation in soil solutions. Chemical Geology 398,
418
61-69. Sastre, J., Hernandez, E., Rodriguez, R., Alcobe, X., Vidal, M., Rauret, G., 2004. Use
420
of sorption and extraction tests to predict the dynamics of the interaction of trace
421
elements in agricultural soils contaminated by a mine tailing accident. Science of
422
the Total Environment 329, 261-281.
RI PT
419
Sauve, S., Hendershot, W., Allen, H.E., 2000a. Solid-solution partitioning of metals in
424
contaminated soils: Dependence on pH, total metal burden, and organic matter.
425
Environmental Science and Technology 34, 1125-1131.
SC
423
Sauve, S., Norvell, W.A., McBride, M., Hendershot, W., 2000b. Speciation and
427
complexation of cadmium in extracted soil solutions. Environmental Science and
428
Technology 34, 291-296.
M AN U
426
Shirvani, M., Sherkat, Z., Khalili, B., Bakhtiary, S., 2015. Sorption of Pb(II) on
430
palygorskite and sepiolite in the presence of amino acids: Equilibria and kinetics.
431
Geoderma 249, 21-27.
TE D
429
Simmler, M., Ciadamidaro, L., Schulin, R., Madejon, P., Reiser, R., Clucas, L., Weber,
433
P., Robinson, B., 2013. Lignite Reduces the Solubility and Plant Uptake of
434
Cadmium in Pasturelands. Environmental Science & Technology 47, 4497-4504.
EP
432
Tai, Y., McBride, M.B., Li, Z., 2013. Evaluating specificity of sequential extraction
436
for chemical forms of lead in artificially-contaminated and field-contaminated
437
AC C
435
soils. Talanta 107, 183-188.
438
Tan, X., Liu, Y., Yan, K., Wang, Z., Lu, G., He, Y., He, W., 2017. Differences in the
439
response of soil dehydrogenase activity to Cd contamination are determined by
440
the different substrates used for its determination. Chemosphere 169, 324-332.
441
Tsang, D.C.W., Lo, I.M.C., 2006. Competitive Cu and Cd sorption and transport in
442
soils: A combined batch kinetics, column, and sequential extraction study.
443
Environmental Science & Technology 40, 6655-6661. 21
ACCEPTED MANUSCRIPT 444
Weng, L., Temminghoff, E. J.,Van Riemsdijk, W. H., 2001. Contribution of individual
445
sorbents to the control of heavy metal activity in sandy soil. Environmental
446
Science & Technology 35, 4436-4443. Wong, S.C., Li, X.D., Zhang, G., Qi, S.H., Min, Y.S., 2002. Heavy metals in
448
agricultural soils of the Pearl River Delta, South China. Environmental Pollution
449
119, 33-44.
RI PT
447
Yang, S., Zhou, D., Yu, H., Wei, R., Pan, B., 2013. Distribution and speciation of
451
metals (Cu, Zn, Cd, and Pb) in agricultural and non-agricultural soils near a
452
stream upriver from the Pearl River, China. Environmental Pollution 177, 64-70.
SC
450
Yin, H., Tan, N.H., Liu, C.P., Wang, J.J., Liang, X.L., Qu, M.K., Feng, X.H., Qiu,
454
G.H., Tan, W.F., Liu, F., 2016. The associations of heavy metals with crystalline
455
iron oxides in the polluted soils around the mining areas in Guangdong Province,
456
China. Chemosphere 161, 181-189.
M AN U
453
Yu, H.Y., Liu, C., Zhu, J., Li, F., Deng, D.-M., Wang, Q., and Liu, C., 2016a.
458
Cadmium availability in rice paddy fields from a mining area: The effects of soil
459
properties highlighting iron fractions and pH value. Environmental Pollution 209,
460
38-45.
TE D
457
Yu, H. Y., Li, F. B., Liu, C. S., Huang, W., Liu, T. X., and Yu, W. M., 2016b. Chapter
462
Five - Iron Redox Cycling Coupled to Transformation and Immobilization of
463
Heavy Metals: Implications for Paddy Rice Safety in the Red Soil of South China,
AC C
464
EP
461
In Advances in Agronomy (Donald, L. S., Ed.), pp 279-317, Academic Press.
465
Zhang, Y.L., Ma, X.X., Ran, Y., 2014. Sorption of phenanthrene and benzene on
466
differently structural kerogen: Important role of micropore-filling. Environmental
467
Pollution 185, 213-218.
468
Zhang, Y.L., Ran, Y., Mao, J.D., 2013. Role of extractable and residual organic matter
469
fractions on sorption of phenanthrene in sediments. Chemosphere 90, 1973-1979.
470
Zhang, Y.L., Wang, L.Y., Li, W.Y., Xu, H.J., Shi, Y.C., Sun, Y.T., Cheng, X., Chen, 22
ACCEPTED MANUSCRIPT X.Y., Li, Y.T., 2017. Earthworms and phosphate-solubilizing bacteria enhance
472
carbon accumulation in manure-amended soils. Journal of Soils and Sediments 17,
473
220-228.
AC C
EP
TE D
M AN U
SC
RI PT
471
23
ACCEPTED MANUSCRIPT
Table 1.Characteristics of the soil samples from the PRB.
2.52±0.70 1.71±0.60 3.26±0.60 2.27±1.00 3.07±0.80 2.93±0.10 3.09±1.30 2.68±2.00 5.49±1.00 5.63±1.80 9.56±2.50 8.43±1.20 6.55±0.20 4.13±0.10 5.89±1.20 1.36±0.70 2.14±1.30 2.31±0.80 4.53±1.10 3.76±1.20 3.24±0.30 3.18±1.00 2.10±0.20 1.63±0.10 1.98±0.70 2.50±0.30 2.23±0.80 1.65±0.30 4.44±1.60 2.92±0.60 2.30±0.30 2.85±0.50 5.25±2.10 5.01±1.10 7.28±1.60 4.27±0.70 5.51±0.60 3.73±0.40
MnDCB (mg kg-1) 356±187 288±76.6 38.2±20.4 54.0±36.3 20.3±18.1 46.5±39.3 40.2±9.50 56.1±20.0 68.1±46.5 71.4±56.1 500±200 457±245 15.4±0.60 43.3±1.00 62.8±43.7 200±100 15.8±7.70 14.8±4.80 55.2±18.7 305±132 46.8±25.4 84.7±44.1 17.5±0.60 16.65±2.00 23.8±3.40 59.4±26.5 76.1±54.3 226±58.3 80.7±35.3 45.1±11.7 35.6±14.8 33.4±17.2 108±8.60 63.3±41.0 429±89.4 493±31.8 603±59.5 587±151
MnTamm (mg kg-1)
CEC (cmol kg-1)
RI PT
10.9±3.80 14.3±4.70 16.0±0.80 19.6±5.40 8.40±6.50 11.8±7.40 6.79±2.90 8.18±4.10 19.7±3.60 19.5±6.10 19.8±3.30 23.2±4.50 20.6±0.40 28.6±0.60 12.5±1.80 12.4±0.50 6.27±4.70 6.92±2.30 16.9±2.40 27.0±10.7 10.9±1.20 11.9±1.50 4.39±0.20 4.75±0.10 6.47±1.60 8.60±1.30 15.3±8.00 12.6±1.40 27.3±12.3 36.1±1.90 14.5±11.3 11.4±4.60 28.9±14.4 18.1±4.50 25.4±2.30 26.7±2.90 26.1±4.10 25.5±4.20
FeTamm (g kg-1)
200±37.2 173±31.9 22.0±16.9 24.0±13.0 11.3±5.50 25.6±27.7 23.8±7.40 37.6±21.3 24.2±15.1 28.3±28.1 458±161 406±204 4.64±0.30 19.3±0.60 43.0±35.9 174±90.7 9.52±4.10 10.2±3.00 33.7±8.50 221±48.8 38.4±21.2 73.2±49.4 13.7±1.00 8.26±0.10 21.9±2.90 58.0±27.4 68.4±76.2 203±62.2 43.8±26.6 13.6±0.01 27.6±11.0 24.5±8.00 69.6±34.1 44.6±29.1 363±95.8 338±47.6 414±49.4 386±117
SC
7.15±0.80 7.52±0.10 6.26±0.80 5.65±0.90 5.81±0.40 6.06±1.00 6.32±1.00 6.20±1.50 6.36±0.80 5.93±1.00 6.32±0.80 5.85±1.10 5.25±0.20 6.10±0.10 5.17±0.40 6.84±0.90 5.32±0.20 5.05±0.20 5.03±0.30 6.39±1.00 5.79±0.70 5.84±1.00 5.06±0.10 4.74±0.10 5.09±0.20 5.57±0.20 5.74±1.00 6.05±0.50 4.42±0.60 3.83±0.03 5.60±0.10 5.46±0.10 5.59±1.30 5.24±0.30 6.04±0.30 5.12±0.40 5.92±0.10 5.27±0.40
FeDCB (g kg-1)
M AN U
247±80.9 138±34.6 100±38.8 53.3±2.2 102±29.2 92.0±16.3 237±18.2 99.4±4.40 176±4.10 127±25.2 83.0±2.40 29.0±2.00 189±2.00 84.5±2.40 77.6±30.9 43.4±6.80 84.7±31.4 47.2±1.30 139±13.5 60.4±11.1 176±3.90 105±1.10 175±6.30 180±10.7 72.5±0.80 49.5±1.90 83.3±2.90 41.0±0.90 69.4±27.6 60.3±3.60 48.6±1.50 32.4±1.70 77.7±27.9 34.2±1.60 63.8±1.90 50.7±1.50 59.6±4.00 51.6±3.40
pH (H2O)
TE D
70.0±34.6 59.8±25.9 40.5±8.54 26.2±4.12 38.9±4.74 33.9±12.7 53.5±40.2 18.8±26.1 36.9±8.80 41.2±8.30 44.7±7.83 33.2±9.47 28.4±0.50 28.4±1.07 44.8±9.91 23.5±7.77 36.8±3.88 35.4±7.81 41.2±3.80 33.8±2.71 39.8±2.89 29.5±5.68 46.4±3.03 42.7±0.94 46.2±17.2 36.6±12.9 41.2±7.42 20.4±8.66 32.7±14.0 15.8±0.88 10.7±3.66 12.9±7.33 37.4±13.4 20.4±7.77 37.7±15.3 27.7±7.09 24.0±2.21 22.1±1.67
DOC (mg kg-1)
EP
LP1(3) LN1(3) LP2(3) LN2(3) LP3(3) LN3(3) SP1(3) SN1(3) SP2(9) SN2(9) SP3(3) SN3(3) SP4(3) SN4(3) DP1(3) DN1(3) DP2(3) DN2(3) DP3(3) DN3(3) DP4(3) DN4(3) DP5(3) DN5(3) DP6(3) DN6(3) DP7(3) DN7(3) AP1(4) AN1(3) AP2(4) AN2(4) AP3(3) AN3(4) GP1(6) GN1(6) GP2(6) GN2(6)
TOC (g kg-1)
AC C
Samplea (nb)
10.3±2.82 14.4±7.44 6.84±1.53 5.91±3.40 6.25±0.28 7.31±1.96 8.52±1.81 7.61±0.48 9.61±1.70 10.8±2.67 8.95±1.00 7.83±0.65 11.4±0.11 14.3±0.34 7.87±3.25 4.96±1.31 5.72±0.76 5.16±0.55 6.88±0.44 9.91±1.51 7.19±1.06 6.17±0.73 7.54±0.56 5.55±0.07 4.54±0.39 4.05±0.68 6.11±1.31 3.57±0.26 5.09±1.71 4.71±0.39 5.32±0.82 5.51±1.40 6.90±2.81 4.91±0.71 17.8±4.74 18.1±3.23 15.6±3.52 15.4±1.22
Sand
Silt
Clay
0.26±0.06 0.15±0.03 0.29±0.13 0.26±0.15 0.31±0.18 0.34±0.19 0.25±0.08 0.17±0.03 0.20±0.08 0.15±0.05 0.20±0.08 0.21±0.10 0.03±0.01 0.03±0.01 0.33±0.02 0.53±0.02 0.41±0.06 0.45±0.06 0.32±0.03 0.31±0.01 0.31±0.09 0.31±0.04 0.29±0.02 0.31±0.02 0.55±0.05 0.59±0.01 0.36±0.06 0.42±0.03 0.19±0.07 0.24±0.07 0.42±0.07 0.38±0.05 0.21±0.03 0.38±0.02 0.03±0.01 0.03±0.01 0.02±0.01 0.02±0.01
0.56±0.03 0.60±0.03 0.45±0.09 0.46±0.10 0.48±0.14 0.42±0.16 0.57±0.03 0.60±0.06 0.49±0.07 0.47±0.04 0.53±0.05 0.54±0.03 0.58±0.01 0.57±0.01 0.36±0.03 0.23±0.03 0.31±0.04 0.34±0.08 0.34±0.03 0.35±0.09 0.48±0.02 0.46±0.04 0.47±0.01 0.47±0.02 0.25±0.04 0.22±0.01 0.36±0.05 0.37±0.02 0.53±0.06 0.41±0.04 0.39±0.09 0.47±0.07 0.55±0.02 0.42±0.01 0.50±0.06 0.47±0.08 0.48±0.02 0.49±0.01
0.18±0.05 0.25±0.05 0.26±0.06 0.28±0.05 0.21±0.05 0.24±0.03 0.18±0.06 0.23±0.04 0.31±0.06 0.38±0.04 0.27±0.06 0.25±0.11 0.39±0.01 0.40±0.01 0.30±0.02 0.24±0.03 0.28±0.03 0.21±0.09 0.34±0.03 0.34±0.09 0.21±0.09 0.23±0.05 0.24±0.02 0.22±0.01 0.20±0.02 0.19±0.01 0.28±0.02 0.21±0.04 0.28±0.03 0.35±0.10 0.19±0.04 0.15±0.07 0.24±0.02 0.20±0.02 0.47±0.06 0.50±0.08 0.50±0.01 0.49±0.01
a Sample name is consist of parent materials and land use type. L = limestone, S = sand shale, D = diluvium, A = alluvium, G = granite, P = paddy soil, N = non-paddy soil. b The n represents sample number. Value: mean ± standard deviation.
ACCEPTED MANUSCRIPT Table 2 Stepwise regression models of factors affecting free and complex metal in five parent material-derived soils (P < 0.05). Parent materials
Dissolve d metal
Limeston e
Free Cd
No stepwise regression equation
Complex ed Cd Free Pb Complex ed Pb Free Cd
No stepwise regression equation
Alluvium
Complex ed Pb
—
—
—
—
—
RI PT
0.002 0.000
—
—
—
—
—
—
SFPb=0.124-0.520[Sand] SCPb=0.034+0.017[WAT-Pb]-0.004[FeTamm] -0.001[TOC]-0.001[CEC]+0.015[Clay]+0.38 2[E4/E6] DFCd=7.150+0.071[DOC]-34.772[Clay] DCCd=0.082+0.022[E2/E3]-0.019[RES-Cd]
0.889 1.000
0.861 1.000
0.005 0.000
0.767 0.706
0.715 0.640
0.001 0.004
DFPb=-0.129+0.091[E4/E6]+0.001[OXID-Pb] DCPb=-0.030+0.020[E4/E6]+0.001[OXID-Pb ] No stepwise regression equation ACCd=0.415-0.008[TOC]
0.877 0.842
0.850 0.807
0.000 0.000
— 0.990
— 0.984
— 0.005
0.995 1.000
0.993 1.000
0.002 0.000
0.912 1.000
0.868 1.000
0.045 0.000
1.000
1.000
0.000
—
—
—
No stepwise regression equation
SC
No stepwise regression equation
AFPb=-0.692+6.116[∆logK] ACCd=0.029+0.684[∆logK]-0.005[SUVA254]+ 0.001[E2/E4] GFCd=-12.109+530.095[Sand] GCCd=-0.914+44.473[Sand] -0.008[TOC]-0.157 [E4/E6] GFPb=0.181-0.327[Silt]+0.0004[E2/E4] -0.002[SUVA254] No stepwise regression equation
AC C
Granite
Free Cd Complex ed Cd Free Pb Complex ed Pb Free Cd Complex ed Cd Free Pb Complex ed Pb Free Cd Complex ed Cd Free Pb
—
0.995 1.000
EP
Diluvium
p value
0.997 1.000
M AN U
Complex ed Cd Free Pb Complex ed Pb
LFPba =1.239-0.162[pH] LCPbb=0.412-0.049[pH]-0.003[ORG-Pb]
TE D
Sandshale
R2adj
R2
Stepwise regression equations
LFMetal and LCMetal mean free and complex metals concentrations in limestone soils. [WAT-metal] means water soluble metal, [OXID-metal] means Fe-Mn oxides bound metal, [RES-metal] means residual metal, [FeTamm] means Tamm's extractable iron.
ACCEPTED MANUSCRIPT Captions Fig.1. Sampling sites. The pollution sources and parent materials nearby the sites are also shown in the map. Fig.2. Total concentrations of Cd and Pb in paddy and dry soils from the different
RI PT
parent materials. The boxes span from the 25th percentile to the 75th percentile. The solid and dash lines inside the box represent the sample median and mean values, respectively. The whiskers denote the interval between the minimum and maximum value, respectively. The full circles point to outliers. L = limestone, S = sand shale, D
SC
= diluvium, A = alluvium, G = granite, P = paddy soil, N = non-paddy soil, CK =
M AN U
background soil.
Fig.3. Effects of different land-use types on dissolved Cd and Pb in soil solution by SC-DMT. Letters (a, b, c, etc.) above each bars show statistical significance at p<0.05. The capital is for comparison between paddy and non-paddy soils, and the small letter is for comparison between free and complex metals. L = limestone, S = sand shale, D
TE D
= diluvium, A = alluvium, G = granite, P = paddy soil, N = non-paddy soil. Fig.4. Speciation distributions of Cd (A) and Pb (B) of the studied soils. L = limestone, S = sand shale, D = diluvium, A = alluvium, G = granite, P = paddy soil, N = non-paddy soil.
EP
Fig.5. Total and dissolved metals (Cd (a) and Pb (b)) of the soils from the five parent materials. The correlation curves and equations were obtained based on the samples
AC C
without the name markers. A = alluvium, D = diluvium, G = granite, L = limestone, S = sand shale, P = paddy soil, N = non-paddy soil.
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Fig.1
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Fig.2
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Fig.3
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Fig.4
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Fig.5
ACCEPTED MANUSCRIPT
Highlights Cd and Pb in paddy soils are more labile than in dry soils due to stable equilibrium.
RI PT
Pb activity is very sensitive to pH change in limestone soils. High sand content enhances Cd activity in granite soils.
DOM greatly impacts Cd and Pb activities in sand shale, diluvium, and
AC C
EP
TE D
M AN U
SC
alluvium soils.