Advertisement

Environmental Science and Pollution Research

, Volume 26, Issue 3, pp 2559–2568 | Cite as

Divergent patterns of soil phosphorus discharge from water-level fluctuation zone after full impoundment of Three Gorges Reservoir, China

  • Jun ZhouEmail author
  • Yanhong WuEmail author
  • Xiaoxiao Wang
  • Haijian Bing
  • Yang Chen
  • Hongyang Sun
  • Zhilin Zhong
Research Article
  • 243 Downloads

Abstract

Phosphorus (P) discharged from soils in the water-level fluctuation (WLF) zone becomes increasingly important to the water quality control of the Three Gorges Reservoir (TGR) as the decrease in P input from upstream reaches and point-source pollution. To investigate the amount of soil P discharge from the WLF zone since the full impoundment of the TGR in 2010, soil and sediment samples were collected along the altitudinal gradients (140, 150, 160, 170, and 180 m above sea level) in three transects in the middle reaches of the TGR. Soil P composition was determined by a sequential extraction procedure. Different amounts of P discharge from the WLF zone were found among three soil types because of their difference in the initial P content before impoundment, with an order of yellow earth (171.1 g m−2), fluvo-aquic soil (141.7 g m−2), and purple soil (73.8 g m−2). An altitudinal pattern of soil P discharge was observed with the maximum at the 170-m sites. The downward transport of exchangeable P and clay-bound P with runoff was the major path of the soil P discharge at the 170-m sites with a slope gradient > 15°. Considerable P discharge with erosion at the upper section of the WLF zone was facilitated by the longer exposure period compared with that at bottom section (150-m sites) because of the annual anti-seasonal impoundment-exposure cycles of the TGR. The transformation of Al/Fe-P and subsequent release to water was a main mechanism of the soil P discharge during the impoundment period. The altitudinal pattern of P discharge was a result of joint effects of slope gradient, soil P forms, and the anti-seasonal hydrological regime of the TGR. The results highlight the critical role of the upper section (165–175 m) in controlling the P output from the WLF zone into the water of the TGR.

Keywords:

Soil phosphorus discharge Water-level fluctuation zone Three Gorges Reservoir Transformation of phosphorus forms Water quality 

Introduction

Water quality control is a priority issue for the Three Gorges Reservoir (TGR) as algal blooms have occurred in several tributaries every spring since 2004 (Fu et al. 2010). Excessive phosphorus (P) in water was identified as a key factor causing the incidence of the algal blooms (Ye et al. 2006; Ji et al. 2017). P entering the water of the TGR through water-level fluctuation (WLF) zone becomes increasingly important for the water quality of the TGR, because P flux from the upstream reaches into the TGR contributed approximately less than 3% (Huang et al. 2014) and P from point source pollution decreased as a result of the enforcements to alleviate industrial and urban pollution (Gao et al. 2016). The P entering the TGR through the WLF zone was mainly derived from soil itself within the WLF zone and agricultural activity near the WLF zone. The second source has been investigated extensively by many studies (e.g., Bouraima et al. 2016; Ma et al. 2016; Ouyang et al. 2017; Shen et al. 2014). Knowledge about the amounts of P discharge from the soils within the WLF zone and their influence factors are still scarce, although a large number of P stocks were observed in the WLF zone (Wu et al. 2016) and could be the secondary source of pollution deteriorating water quality of the TGR (Bao et al. 2015).

Previous studies in the WLF zone focused on the temporal variation, spatial distribution, sorption-release characteristics, and composition of soil P and discussed the effect of the land use type, microtopography, and the duration of submerging and exposure period (e.g., Guo et al. 2014; Xiao 2010; Ye et al. 2015; Yuan et al. 2008; Zhang et al. 2012; Zhou et al. 2018). However, the temporal variations in soil P pools in the WLF zone are still in dispute. For example, while Li et al. (2014) observed an accumulation of P in two WLF zone transects in 2011, Guo (2012) suggested the soil in the WLF zone was a source of P from 2010 to 2011. In addition, because most of these studies used soil samples collected in 2008–2011 when the water level just reached the maximum (175 m above sea level, asl), it is not clear whether the loss of soil P in the WLF zone has occurred since 2010 when the full operation of the TGR started.

According to the anti-seasonal hydrological regime of the TGR, the mean annual exposure period in the upper section (165–175 m asl) of the WLF zone is about 125 days longer than that in the bottom section (145–155 m asl). And the whole WLF zone is exposed to the precipitation and runoff in wet season. However, it is still unknown that the effect of the anti-seasonal hydrological regime on the amounts and spatial pattern of soil P loss in the WLF zone. Specifically, it is needed to investigate whether the longer submerging period results in a larger P discharge by the transformation of P forms due to the changes in redox in the bottom section, or whether the longer exposure period leads to a greater P loss by the more intensive erosion in the upper section.

Slope gradient potentially influences the soil P discharge by its controls on erosion and deposition. The WLF zone in the TGR was categorized into three classes according to its slope gradients (Bao et al. 2015). Wang et al. (2016) found that the role of slope gradient in the deposition of sediments varied in different sections of the WLF zone. Bao et al. (2018) reported that the soil erosion rate in the WLF zone was positively correlated with the slope gradient in general. In addition, the initial contents of soil P in the WLF zone before the impoundment were related to the soil and land use types in the WLF zone (Xiao 2010; Ye et al. 2015). The composition of soil P forms is another factor of affecting the P discharge since the geochemical reactions of different soil P forms differ in response to the annual impoundment-exposure cycles (Zhou et al. 2018). However, little is known on the joint effects of all the above factors and their interactions on the soil P discharge in the WLF zone.

In the presented study, soil samples were collected along the altitudinal gradients (140, 150, 160, 170, and 180 m asl) in three WLF zone transects with different soil and land use types in the middle reaches of the TGR. The composition of soil P forms was measured by a sequential extraction procedure. The goals are about to answer the following questions:
  1. 1)

    What is the amount of soil P discharge in the WLF zone of the TGR 5 years after its full impoundment?

     
  2. 2)

    Is there any altitudinal pattern of the P discharge along the WLF zone transects?

     
  3. 3)

    What are the key mechanisms regulating the amount and altitudinal pattern of the P discharge in the WLF zone of the TGR?

     

Materials and methods

Study area

The WLF zone represents a physical area along the TGR banks with the lowest and highest boundary at 145 and 175 m asl, respectively. The WLF zone was formed by the annually cyclic impoundment and exposure due to the full impoundment of the TGR since 2010 (Bao and He 2011). Three transects in the middle reaches of the TGR were selected to collect soil samples (Fig. 1), because the most P and the largest area of the WLF zone were located in this area (Bao et al. 2015; Wu et al. 2016). The slope gradient ranged from 5° to 25°. Three kinds of soil types, yellow earth, fluvo-aquic soil, and purple soil, were included in these transects (Table 1). In the area with the slope gradients of 5–10°, soil had been covered by at least 30 cm sediments. The land use types of the background sites (180 m asl) were abandoned cropland, citrus orchard, and shrubs in the LJ, XS, and DZ sites, respectively. About 30–40% of area at the 170 sites was occupied by grass. Areas at the 150–160-m sites were mostly bare in all transects (Table 1).
Fig. 1

The location of sampling sites in the Three Gorges Reservoir, China

Table 1

Description of sites and soil properties along the water-level fluctuation zone gradients in the Three Gorges Reservoir, China

Site

Elevation

(m)

Coordinate

Slope

Soil type/sediment

Land use

pH

SOMa (%)

CO32−b (%)

LJ

180

30° 4′ 0.13″ N

107° 51′ 58.44″ E

10–15°

Yellow earth

Abandoned cropland

7.6 ± 0.3a

5.2 ± 1.0a,b

3.4 ± 1.1a,b

170

15–20°

Yellow earth

Grass

7.7 ± 0.1a,b

4.0 ± 1.0b

2.1 ± 0.9a

160

5–10°

Sediment

Bare

8.1 ± 0.1c

5.5 ± 0.3a

3.7 ± 0.3b

150

5–10°

Sediment

Bare

8.0 ± 0.0b,c

5.3 ± 0.3a,b

3.5 ± 0.3a,b

140

Sediment

Underwater

8.6 ± 0.1d

5.5 ± 0.4a

5.2 ± 0.3c

XS

180

30° 12′ 13.30″ N

107° 56′ 45.71″ E

10–15°

Fluvo-aquic soil

Citrus orchard

7.0 ± 0.1a

4.3 ± 0.2a

1.3 ± 0.1a,b

170

15–20°

Fluvo-aquic soil

Grass

7.6 ± 0.1b

1.9 ± 0.1b

0.6 ± 0.0a

160

10–15°

Fluvo-aquic soil

Bare

7.7 ± 0.5b

2.4 ± 0.2b

1.7 ± 0.9b

150

5–10°

Sediment

Bare

8.4 ± 0.2c

3.9 ± 0.5a

6.3 ± 0.2c

140

Sediment

Underwater

8.7 ± 0.1c

5.3 ± 0.6c

6.2 ± 0.5c

DZ

180

30° 19′ 5.52″ N

108° 5′ 3.55″ E

~ 25°

Purple soil

Shrubs

6.4 ± 0.5a

3.4 ± 0.1a

0.4 ± 0.2a

170

~ 25°

Purple soil

Grass

6.9 ± 0.4a,b

2.9 ± 0.4a

0.7 ± 0.1a,b

160

~ 25°

Purple soil

Bare

7.6 ± 0.4b,c

3.0 ± 0.8a

1.2 ± 0.3b

150

~ 20°

Purple soil

Bare

7.7 ± 0.2c

3.3 ± 0.2a

0.8 ± 0.1a,b

140

Sediment

Underwater

8.6 ± 0.1d

5.2 ± 0.3b

4.3 ± 0.5c

aSOM soil organic matter

Different letters indicate significantly different variables between different altitudes at the P < 0.05 level

Sampling

Each year, the water level in the TGR starts to decrease gradually from 175 m asl in late March to 145 m asl in June. The water level of 145 m asl then will be kept till September when it starts to increase to 175 m asl in October (Bao et al. 2015). The WLF zone will be totally exposed in July. So we collected soil samples in July 2015 to ensure that soil and sediment samples across the whole WLF zone gradients could be collected. Samples were collected in five altitudes where the 180-m asl sites were considered as the background soil (Table 1). Six pits with a horizontal distance of > 20 m from each other were dug at each altitude (except the 140-m asl sites) (Table 1). Samples were collected from the walls of the approximately 0.5-m wide soil pits to the depth of 30 cm. Sediment samples were collected at the sites with slope gradients < 10°. To collect sediments (0–30 cm) underwater (140 m asl), six sediment cores were sampled using a gravity sediment core sampler (100-cm long and 6-cm internal diameter). Two random samples at each altitude were well mixed to yield a composition sample. And then three composition samples were obtained for each altitude.

Analysis

Samples were air-dried and sieved (< 2 mm) before analysis. The pH values were measured in a slurry of 1:2.5 soil to H2O by a glass electrode (7252101B) and a potentiometer (pH 6+, EUTECHTM). The contents of soil organic matter (SOM) and carbonate (CO32−) were determined by sample loss on ignition in a muffle furnace (SOM: combusting for 4 h after the sample was slowly brought up to 500 °C over 4 h). The contents of CO32− were calculated by difference of weights of soil samples ignited after 500 °C and 950 °C. The particle size distribution was determined by a laser particle size analyzer (Mastersizer 2000) after samples were treated with H2O2 solutions (3%) (clay, < 2 μm; silt, 2–20 μm; sand, 20–2000 μm). The concentration of total P was determined using a Profile DV (USA Teledyne Leeman Labs) Inductively Coupled Plasma Atomic Emission Spectrometer (ICP-AES) (detection limit, 0.02 mg P L−1), after digestion by refluxing with the nitric acid, hydrofluoric acid, and perchloric acid. Certified reference materials (GBW074134 and GBW07414) from the National Quality and Technology Supervision Agency of China were measured alongside the soil samples to validate the accuracy (10%) of the concentration measurements.

Because most of the samples were soil, a modified Hedley sequential extraction procedure (Hedley et al. 1982) was applied to distinguish soil P forms. Briefly, ~ 0.5 g sample was sequentially treated with (1) 20 mL deionized water and two resin strips (exchangeable P); (2) 30 mL 0.5 M NaHCO3 (pH 8.2) (weakly absorbed Al/Fe-P; NaHCO3-P); (3) 30 mL 0.1 M NaOH (strongly bound Al/Fe-P; NaOH-P); and (4) 30 mL 1 M HCl (Ca-bound P; HCl-P). The organic P was measured as the difference between the undigested and digested (121 °C) sample P concentrations. The suspended soil solutions from each step were shaken for 16 h before centrifuging at 10,000×g for 10 min at 0 °C and passing through Millipore filters (pore size 0.45 μm). All extraction solutions were analyzed with a UV–Vis spectrophotometer (SHIMAZU UV 2450) by the phosphomolybdate blue method (Murphy and Riley 1962). In all cases, blanks were mixed with the solvents to estimate matrix interferences during the extraction processes. The concentration of total organic P was the sum of the organic P in NaHCO3 and NaOH extracts. Available P was the sum of exchangeable P, NaHCO3-P, and NaOH-P (Liu et al. 2015). The difference between total P and the sum of all other fractions represents residual P. For the samples at the 140-m sites, only the concentrations of total and exchangeable P were measured.

Calculation

The pool of total P in the upper 30 cm of soil (TP30, g m−2) was calculated by the following formula:
$$ {\mathrm{TP}}_{30}=\left(\mathrm{P}\times \mathrm{BD}\times 30\;\mathrm{cm}\right)/100 $$
(1)

where P (mg kg−1) represents the concentration of total soil P. BD (g cm−3) represents the bulk density in the top 30 cm.

No fertilizer was added to the background soils after the full impoundment since these areas were abandoned cropland and orchard, and shrubs; therefore, the pools of total P in the background soils were considered as the initial P pools in soil in the WLF zone. The difference between the P pools in the background soils and in the 150–170-m sites were considered as the amounts of soil P discharge caused by the 5-year annual cyclic impoundment-exposure operation. The amount of soil P discharge was set to 0 in the sites covered by sediments.

Statistics

Soil properties and concentrations of P forms were subjected to a one-way analysis of variance (ANOVA) to test differences between different altitudes. A test of homogeneity of variances was applied to these variables before the ANOVA. The normality of residuals of the variables was tested by Q–Q plots. These variables were tested for normality (Shapiro-Wilk) before the Pearson’s correlation analysis to determine their correlations. A significance level of P < 0.05 was used in this study (except where noted).

Results

Soil properties

Soil pH in all the three transects showed an increasing trend downslope with a range of 6.4–8.6 in the TGR (Table 1). The concentrations of SOM at the 180- and 140-m sites were higher than those within the WLF zone (150–170-m sites), while the concentrations of inorganic C showed different altitudinal patterns in the three transects (Table 1). The average proportions of silt (60.6%) and clay (12.2%) in the LJ gradient were higher than those in the XS (40.5 and 10.5%) and DZ (48.1 and 11.8%) transects (Fig. 2a, b, c). The proportions of clay at the 170-m sites in the three transects were the lowest among the five altitudes varying between 8.9 and 9.6%. The percentages of silt (61.7–66.8%) and clay (16.8–18.1%) in the sediments (140-m sites) in the three gradients were higher than those in the 150–180-m sites.
Fig. 2

The variations in particle size with altitude in the water-level fluctuation zone in the Three Gorges Reservoir, China. a LJ transect; b XS transect; c DZ transect

Variations in P forms along WLF zone gradients

The concentrations of total P at the 170-m sites (140–241 mg kg−1) were significantly lower than those in the background soils (229–726 mg kg−1, 180-m sites) in the three gradients, although the contents of total P in the background soils varied greatly among the three different soil types (Fig. 3a, b, c). Altitudinal variations in the concentration of total P showed different patterns from the 170- to 150-m sites. Specifically, the concentration of total P showed an increasing trend in the LJ and XS transects, while no spatial trend in the DZ transects. The concentrations of total P in the sediments (140-m site) in the three sites were higher than those in the 150–180-m sites. The pools of total P (0–30 cm) at the 170-m sites were lower than those at other altitudes in all transects (Fig. 4).
Fig. 3

Comparison between the concentrations of total soil P and exchangeable P in the water-level fluctuation zone (150–170 m asl), the background soil (180 m asl), and the underwater sediments (140 m asl) in the Three Gorges Reservoir, China. a, d LJ transect; b, e XS transect; c, f DZ transect

Fig. 4

Comparison between the pool of total soil P in the water-level fluctuation zone (150–170 m asl), the background soil (180 m asl), and the underwater sediments (140 m asl) in the Three Gorges Reservoir, China. a LJ transect; b XS transect; c DZ transect

Similar to the total P, the highest and lowest concentrations of exchangeable P occurred at the 140- and 170-m sites in each gradient, respectively (Fig. 3d, e, f). Different from the total P, similar one-way increasing trends of exchangeable P from the 170- to 150-m sites were observed in the three transects (Fig. 3d, e, f). The concentrations of NaHCO3-P, NaOH-P, available P, and HCl-P at the 170-m sites were significantly lower than those in the background soils (180-m sites) in all gradients (Fig. 5a, b, c, d, e and f). The concentrations of residual P and organic P at the 170-m sites were lower than those at the 180-m sites in the LJ and XS gradients, while no spatial trend for the two P forms was observed in the DZ gradient (Fig. 5g, h and i). Distinct decreases in the proportions of available P and HCl-P to the total P and significant increases in the residual P were found at the 170-m sites compared with those at the background soils.
Fig. 5

Comparison between the concentrations of different P forms in the water-level fluctuation zone (150–170 m asl) and the background soil (180 m asl) in the Three Gorges Reservoir, China. a, d, g LJ transect; b, e, h XS transect; c, f, i DZ transect

Altitudinal pattern of the soil P discharge along WLF zone transects

An altitudinal pattern of the P loss was observed in the WLF zone. Specifically, the amount of the soil P loss significantly declined downslope in the WLF zone (Fig. 7). The largest discharge of soil P (171.1 g m−2) occurred at the 170-m sites in the LJ gradient, whereas no discharge was observed at the 160- and 150-m sites. The total amount of P discharge in the XS gradient was 141.7 g m−2, including 128.7 g m−2 at the 170-m site and 13.0 g m−2 at the 160-m site. The discharge of P occurred at all altitudes in the DZ transect, although the amount was significantly smaller than that in the LJ and XS transects. The discharge of soil P was 35.2, 14.6, and 24.1 g m−2 at the 170-, 160-, and 150-m sites, respectively, in the DZ transect.

Discussion

Amount of soil P discharge in the WLF zone

There are several reasons for considering the difference between the P pools in the background and the WLF zone soils as the amount of the soil P discharge from the WLF zone. First, the background soils shared same soil type and land use with the soils in the WLF zone before the impoundment. Agricultural practices are prohibited near the WLF zone after the impoundment. And because the background soils were only less than 10 m away from the WLF zone, it is reasonable to take the P pools in the background soils as the initial P pools in the WLF zone. The concentrations of total P in the background soils in the present study were within the range of previous studies, whose soil samples were collected from 2008 to 2011 in the same area (e.g., Cao 2011; Li et al. 2014; Zhang et al. 2012), indicating relative small changes in the concentrations of total P in our background soils.

Second, Wang et al. (2010) found the concentrations of total soil P at 165–175 m asl were almost two times of those at the 185-m asl under shrubs and 1.5 times of those at 145–165 m asl in July 2008 when the area in 156–175 m asl had not been inundated. Ye et al. (2015) reported that the concentration of total P in the 165–175 m asl was significantly higher than that in the 145–155 m asl in September 2008. Sun (2010) found that there was no distinct altitudinal pattern of total soil P from 175 to 145 m asl in July 2009. Li et al. (2014) also observed no significant difference of total P between soils at > 175 m and 165–175 m from January to November of 2011. These results are significantly different from our observation that the concentration of total soil P at the 170-m site was lower than those at the background soils and at 145–165 m in all the three gradients (Fig. 3a, b and c). In addition, Xu et al. (2009) and Cao (2011) found that the concentrations of total P in soils in the WLF zone were significantly higher than those in the underwater sediments in October 2007 and from May to July of 2008. This was also inconsistent with our findings that the concentration and pool of total soil P within the WLF zone (145–175 m asl) were lower than those in the sediments (140 m asl) in the three transects (Figs. 3a, b and c and 4a, b and c). Although season and temperature may influence the altitudinal pattern of soil P in the WLF zone given that the samples of the above studies were collected in various seasons, their results showed a similar trend that is different from ours, indicating the ignorable effects of season and temperature on the distribution of soil P in the WLF zone. Therefore, the possible interpretation for the difference between our results and others should be that considerable amounts of soil P discharge have occurred after 5-year annual impoundment-exposure cycles in the WLF zone of the TGR. And it is reasonable to consider the differences of P pools between the background and the WLF zone soils as a proxy of the amount of soil P discharge.

Factors for divergent patterns of P discharge along the WLF zone transects

Difference in the amounts of P discharge

The soil and land use types are mainly responsible to the differences in the amounts of P discharge among the three transects because of their different initial content of soil P which represents the potential source of P discharge. The facts that the land use types are abandoned cropland and citrus orchard in the LJ and XS (Table 1) show that chemical P fertilizers were applied in the two sites before the impoundment. In addition, the initial content of P in the WLF zone soil was in the order of yellow earth > fluvo-aquic soil > purple soil in the TGR (Guo 2012; Xiao 2010; Zhang et al. 2012). The largest P discharge occurred in the LJ transect where the initial content of P was highest among the three transects (Fig. 7). And the smallest P loss and the lowest initial P content were observed in the DZ transect (Fig. 7). Therefore, the different amounts of P discharge in the three transects mainly depended on the soil and land use types.

Altitudinal pattern of soil P discharge

The altitudinal pattern of soil P discharge along the WLF zone transects (Fig. 7) was first affected by the slope gradient. As shown in Table 1, the slope gradient is a major factor controlling the processes of deposition and erosion. More than 30 cm depth sediments were found in the sampling sites with a slope gradient of < 10° while no sediment was found in the sites > 15° (Table 1). Bao et al. (2018) suggested that the slope gradient was a control of soil erosion in the WLF zone based on a 9-year observation. Wang et al. (2016) reported a similar result that the slope gradient was the main factor affecting sediment deposition in the middle reaches of the TGR and most sediment deposited where the slope gradient was < 15°. Soils buried by the sediments are prevented from the interaction with water, and thus P in these soils is hard to be released to the water. However, because soils in the sites with a slope gradient > 15° are exposed to either overlying water during the submerging period or runoff during the low water-level period (wet season), therefore, considerable amounts of P discharge were observed in those sites with steep slope gradients (Table 1, Fig. 7). Ma et al. (2016) also found that there were the highest ratios of P discharge from surface runoff in 15° plots of sloping farmlands in the TGR watershed.

The hydrological regime of the TGR is another major factor influencing the altitudinal pattern of the soil P discharge. The submerging period reduces with the increasing altitude in the WLF zone due to this hydrological regime (Bao et al. 2015), resulting in a longer exposure period in the 160- and 170-m sites than in the 150-m site. The differences in submerging and exposure period among different altitudes, together with the effects of slope gradient and initial content of soil P, govern the altitudinal pattern of the soil P loss by influencing geomorphological and geochemical processes within the WLF zone.

In the 170-m sites, transformation of Al/Fe-P (NaHCO3-P and NaOH-P) and release of exchangeable P are likely the dominant geochemical processes during the submerging period (November to February). Compared with the background soils, the distinct lower NaHCO3-P and NaOH-P at the 170-m sites in all transects (Fig. 5a, b and c) reflect the release of the two P forms after their transformations into exchangeable P. In 2010, the contents of Al/Fe-P in the WLF zone soils were higher than those in background soils (Guo 2012). The inverse pattern of Al/Fe-P between the two studies indicates the discharge of much Al/Fe-P from the WLF zone 5 years after the full impoundment of the TGR. Moreover, the distinct increase in the proportion of residual P and decrease in NaHCO3-P and NaOH-P at the 170-m sites (Fig. 6a, b and c) demonstrate the loss of a large number of Al/Fe-P and thus the immobile residual P became the largest contributor to the TP. In addition, by an experiment that soil samples were collected in the WLF zone, Zhang et al. (2012) found that the amount of P released from the soils was positively correlated to the Fe/Al-P and that it contributed the largest part to the P-release source in the soils and to the sources for the overlying water. Ma et al. (2008) found that in the middle reaches of the TGR, the release of Fe-P due to the reduction from Fe3+ to Fe2+ during the submerging period was a major process of the release of soil P.
Fig. 6

Comparison between the contributions of different P forms in the water-level fluctuation zone (150–170 m asl) and the background soil (180 m asl) in the Three Gorges Reservoir, China. a LJ transect; b XS transect; c DZ transect

Although the adsorption of soils to PO43− and the deposition of particle P in the overlying water may increase the content of soil P during the submerging period, these reactions unlikely dominated the geochemical process because the concentrations of PO43−-P in water near our sites were not high enough (~ 0.07 mg L−1, Han et al. 2018) to be adsorbed to soils. Jia et al. (2007) suggested that adsorption occurred only when the concentration of PO43−-P in water was 0.1–1.6 mg L−1 in the TGR. Zhang (2013) also reported that soil P would be released when the concentration of PO43−-P in the overlying water was less than 0.12 mg L−1. In addition, the relative large slope gradient at the 170-m sites does not favor the deposition of particle P from the overlying water.

During the exposure period, downward transport of exchangeable P with runoff contributed to the discharge of soil P at the 170-m sites. This can be reflected by the significant lower concentration of exchangeable P at the 170-m sites than those at the background soils and at the 150- and 160-m sites (Fig. 3d, e and f). Another path of soil P loss was the erosion of P included in silt and clay, because more P was associated to finer soils due to their greater adsorption capacity (e.g., Wu et al. 2016). The observation that the lowest contents of clay, bioavailable P, and HCl-P occurred at the 170-m sites among the four altitudes in all transects shows the spontaneous loss of clay and P (Fig. 2a–c, 5d–f). And the significant positive correlations between most P forms and clay (Table 2) further demonstrate the role of erosion in facilitating the P discharge at the 170-m sites during the exposure period. A continuous field observation also found that the highest rate of erosion was in the altitude range of 170–175 m in the main stream of the TGR (Bao et al. 2018), highlighting the major influence of the anti-seasonal submerging period on the erosion of the soils and associated P in the upper section of the WLF zone.
Table 2

Pearson’s correlation coefficients between P forms and soil properties along the water-level fluctuation zone gradients in the Three Gorges Dam, China (n = 27).

 

pH

SOMa

CO32−

Sand

Silt

Clay

Exchangeable P

0.56

0.28

0.66

−0.09

0.03

0.41

Total P

0.63

0.62

0.90

0.43

0.40

0.41

NaHCO3-Pt

0.38

0.61

0.58

0.45

0.41

0.51

NaOH-Pt

0.49

0.71

0.62

0.56

0.53

0.53

HCl-P

0.63

0.56

0.91

−0.31

0.28

0.36

Residual P

0.44

0.50

0.65

0.55

0.57

0.23

Available P

0.49

0.66

0.64

0.50

0.46

0.53

Organic P

0.18

0.40

0.16

0.40

0.39

0.31

aSOM soil organic matter

Italics indicate significant correlations between two variables at the P < 0.05 level

In the bottom section (145–155 m) with a slope gradient < 15°, deposition of suspended particles from the overlying water was a major process during the submerging period, as indicated by the thick sediments (Table 1) and the similar proportions of silt and clay to those at the 140-m sites (Fig. 2). During the exposure period, deposition of silt and clay from the areas > 155 m asl dominated the bottom section. Although erosion happened during this period, most of the eroded materials in the 145–155 m asl were soils from the upper section and particle materials from the overlying water. Several previous studies also suggested that in the area between 145 and 155 m with gentle slopes, deposition process was a dominant process with the sedimentation rate varying between 1 and 40 cm (Bao et al. 2010; Tang et al. 2014). Han et al. (2018) suggested that as suspended particles mainly deposited in the tail and middle reaches of the TGR, the concentrations of total particle P in water showed a significantly decreasing trend downstream. Therefore, the amounts of P discharge from the soils underlying sediments in parts of the sampling sites were considered ignorable and were not estimated (Fig. 7). For the bottom section at the DZ transect, erosion was always dominant in both the submerging and exposure periods because of its relatively large slope gradient (Table 1). During the exposure period, this section is subject to the erosion by runoff and waves, therefore the amount of P discharge and the proportion of residual P at the 150-m sites were larger than those at the 160-m sites in the DZ transect (Fig. 7).
Fig. 7

The amounts of soil P discharge from the water-level fluctuation zone in the Three Gorges Reservoir, China

In summary, the divergent patterns of soil P discharge along the WLF zone gradients in the TGR (Fig. 7) are mainly attributed to the soil and land use types, slope gradient, and the hydrological regime of the TGR.

Conclusions

Compared with the initial soil P pool, considerable soil P discharge was observed in the WLF zone of the TGR 5 years after the water level first time reached 175 m asl. The differences in the amounts of soil P discharge among the yellow earth (171.1 g m−2), fluvo-aquic soil (141.7 g m−2), and purple soil (73.8 g m−2) were mainly induced by the initial soil P pools in the different soil and land use types. Generally, the amount of the soil P discharge increased with altitudes within the WLF zone. This altitudinal pattern of the soil P discharge was regulated by the slope gradient and the anti-seasonal impoundment-exposure operation of the TGR through regulating the geomorphological (erosion and deposition) and geochemical processes (downward transport of exchangeable P and transformation and release of Al/Fe-P). The upper section (165–175 m asl) of the WLF zone is a critical area to reduce the P load into the water of the TGR.

Notes

Funding information

This work was supported by the National Natural Science Foundation of China (Grant No. 41630751), the 135 Strategic Program of the Institute of Mountain Hazards and Environment, CAS (Grant No. SDS-135-1702), the Start-up Funds for Doctoral Research of China West Normal University (Grant No. 412654), the CAS “Light of West China” Program, and the China Scholarship Council.

References

  1. Bao YH, He XB (2011) Preliminary study on soil erosion at the water-level-fluctuating zone of the Three-Gorges Reservoir. Res Soil Water Conserv 18:190–195. (in Chinese with English abstract)Google Scholar
  2. Bao YH, Nan HW, He XB, Long Y, Zhang XB, (2010) Sedimentation in the riparian zone of the Three Gorges Reservoir, China. In: Banasik K, Horowitz AJ, Owens PN, Stone M, Walling DE (eds) Sediment Dynamics for a Changing Future. IAHS Publication. Int Assoc Hydrological Sciences, Wallingford, pp. 224–228Google Scholar
  3. Bao YH, Gao P, He XB (2015) The water-level fluctuation zone of Three Gorges Reservoir—a unique geomorphological unit. Earth Sci Rev 150:14–24.  https://doi.org/10.1016/j.earscirev.2015.07.005 CrossRefGoogle Scholar
  4. Bao YH, He XB, Wen AB, Gao P, Tang Q, Yan DC, Long Y (2018) Dynamic changes of soil erosion in a typical disturbance zone of China's Three Gorges Reservoir. Catena 169:128–139.  https://doi.org/10.1016/j.catena.2018.05.032 CrossRefGoogle Scholar
  5. Bouraima AK, He BH, Tian TQ (2016) Runoff, nitrogen (N) and phosphorus (P) losses from purple slope cropland soil under rating fertilization in Three Gorges Region. Environ Sci Pollut Res 23:4541–4550.  https://doi.org/10.1007/s11356-015-5488-1 CrossRefGoogle Scholar
  6. Cao L. (2011) The distribution characteristics and transformation mechanism of phosphorus research on water/sediments wet-dry alternation in water level fluctuating zone of Three Gorges Reservoir Area PhD. Chongqing University, Chongqing, China, pp. 120. (in Chinese with English abstract)Google Scholar
  7. Fu BJ, Wu BF, Lu YH, Xu ZH, Cao JH, Niu D et al (2010) Three Gorges Project: Efforts and challenges for the environment. Prog Phys Geogr 34:741–754.  https://doi.org/10.1177/0309133310370286 CrossRefGoogle Scholar
  8. Gao Q, Li Y, Cheng QY, Yu MX, Hu B, Wang ZG, Yu Z (2016) Analysis and assessment of the nutrients, biochemical indexes and heavy metals in the Three Gorges Reservoir, China, from 2008 to 2013. Water Res 92:262–274.  https://doi.org/10.1016/j.watres.2015.12.055 CrossRefGoogle Scholar
  9. Guo S. (2012) Phosphorus fractions and phosphate sorption-release characteristics of the surface soil in water-level-fluctuating zone of Three Gorges Reservoir Master Degree. Chongqing University, Chongqing, China, pp. 82. (in Chinese with English abstract)Google Scholar
  10. Guo N, Jiang T, Wei SQ, Yan JL, Liang J, Lu S, Gao J (2014) Effects of light irradiation on phosphorous releases from typical submerged soils of water-level fluctuation zones of Three Gorges Reservoirs areas. Environ Sci 35:4544–4552 (in Chinese with English abstract)Google Scholar
  11. Han CN, Zheng BH, Qin YW, Ma YQ, Yang CC, Liu ZC, Cao W, Chi M (2018) Impact of upstream river inputs and reservoir operation on phosphorus fractions in water-particulate phases in the Three Gorges Reservoir. Sci Total Environ 610:1546–1556.  https://doi.org/10.1016/j.scitotenv.2017.06.109 CrossRefGoogle Scholar
  12. Hedley MJ, Stewart J, Chauhan BS (1982) Changes in inorganic and organic soil phosphorus fractions induced by cultivation practices and by laboratory incubations. Soil Sci Soc Am J 46:970–976CrossRefGoogle Scholar
  13. Huang YL, Zhang P, Liu DF, Yang ZJ, Ji DB (2014) Nutrient spatial pattern of the upstream, mainstream and tributaries of the Three Gorges Reservoir in China. Environ Monit Assess 186:6833–6847.  https://doi.org/10.1007/s10661-014-3892-5 CrossRefGoogle Scholar
  14. Ji DB, Wells SA, Yang ZJ, Liu DF, Huang YL, Ma J, Berger CJ (2017) Impacts of water level rise on algal bloom prevention in the tributary of Three Gorges Reservoir, China. Ecol Eng 98:70–81.  https://doi.org/10.1016/j.ecoleng.2016.10.019 CrossRefGoogle Scholar
  15. Jia H, Lei A, Ye M, Lei J, Zhao J (2007) Assessment of phosphorus release from typical soil types in the zone of fluctuating water level in the Three Gorges Reservoir region. Adv Water Sci 18:433–438 (in Chinese with English abstract)Google Scholar
  16. Li C, Zhang J, Wang D, Yu Y, Sun R, Zhang C, et al. Dynamic variations of nitrogen and phosphorous of the soil in hydro⁃fluctuation of Three Gorges Reservoir. Environ Chem 2014; 33: 2169-2175. (in Chinese with English abstract)Google Scholar
  17. Liu Q, Liu SL, Zhao HD, Deng L, Wang C, Zhao QH, Dong S (2015) The phosphorus speciations in the sediments up- and down-stream of cascade dams along the middle Lancang River. Chemosphere 120:653–659.  https://doi.org/10.1016/j.chemosphere.2014.10.012 CrossRefGoogle Scholar
  18. Ma L, Zhang M, Teng Y, Zhao J (2008) Characteristics of phosphorous release from soil in periodic alternately waterlogged and drained environments at WFZ of the Three Gorges Reservoir. Environ Sci 29:1035–1039 (in Chinese with English abstract)Google Scholar
  19. Ma X, Li Y, Li BL, Han WY, Liu DB, Gan XZ (2016) Nitrogen and phosphorus losses by runoff erosion: field data monitored under natural rainfall in Three Gorges Reservoir Area, China. Catena 147:797–808.  https://doi.org/10.1016/j.catena.2016.09.004 CrossRefGoogle Scholar
  20. Murphy J, Riley JP (1962) A modified single solution method for the determination of phosphate in natural waters. Anal Chim Acta 27:31–36CrossRefGoogle Scholar
  21. Ouyang WJ, Li Z, Liu J, Guo JS, Fang F, Xiao Y, Lu L (2017) Inventory of apparent nitrogen and phosphorus balance and risk of potential pollution in typical sloping cropland of purple soil in China - a case study in the Three Gorges Reservoir region. Ecol Eng 106:620–628.  https://doi.org/10.1016/j.ecoleng.2017.06.044 CrossRefGoogle Scholar
  22. Shen ZY, Qiu JL, Hong Q, Chen L (2014) Simulation of spatial and temporal distributions of non-point source pollution load in the Three Gorges Reservoir Region. Sci Total Environ 493:138–146.  https://doi.org/10.1016/j.scitotenv.2014.05.109 CrossRefGoogle Scholar
  23. Sun L. (2010) Effects of phosphorus on water environment in Xiaojiang River basin Hyro-fluctuation belt in the Three Gorges Reservoir. Master Degree. Southwest University, Chongqing, China, 2010. (in Chinese with English abstract)Google Scholar
  24. Tang Q, Bao YH, He XB, Zhou HD, Cao ZJ, Gao P, Zhong RH, Hu YH, Zhang XB (2014) Sedimentation and associated trace metal enrichment in the riparian zone of the Three Gorges Reservoir, China. Sci Total Environ 479:258–266.  https://doi.org/10.1016/j.scitotenv.2014.01.122
  25. Wang X, Cheng R, Xiao W, Guo Q, Feng X (2010) Soil nutrient characteristics in juvenile water-level-fluctuating zone of Three Gorges Reservoir. Chinese Journal of Ecology 29:281–289 (in Chinese with English abstract)Google Scholar
  26. Wang BY, Yan DC, Wen AB, Chen JC (2016) Influencing factors of sediment deposition and their spatial variability in riparian zone of the Three Gorges Reservoir, China. J Mt Sci 13:1387–1396.  https://doi.org/10.1007/s11629-015-3806-1 CrossRefGoogle Scholar
  27. Wu YH, Wang XX, Zhou J, Bing HJ, Sun HY, Wang JP (2016) The fate of phosphorus in sediments after the full operation of the Three Gorges Reservoir, China. Environ Pollut 214:282–289.  https://doi.org/10.1016/j.envpol.2016.04.029 CrossRefGoogle Scholar
  28. Xiao GQ. (2010) Research on mobilization characteristics of phosphorous from the Three Gorges Reservoir zone and its control. PhD. Southwest University, Chongqing, China, pp. 106. (in Chinese with English abstract)Google Scholar
  29. Xu D, Qin Y, Zheng B, Hai R (2009) Phosphorus forms and its distribution characteristics in sediments and soils of water-level-fluctuating zone of the backwater reach from input river of Three Gorges Reservoir. Environ Sci 30:1337–1344 (in Chinese with English abstract)Google Scholar
  30. Ye L, Xu YY, Han XQ, Cai QH (2006) Daily dynamics of nutrients and chlorophyll a during a spring phytoplankton bloom in Xiangxi Bay of the Three Gorges Reservoir. J Freshw Ecol 21:315–321.  https://doi.org/10.1080/02705060.2006.9665001 CrossRefGoogle Scholar
  31. Ye C, Li SY, Yang YY, Shu X, Zhang JQ, Zhang QF (2015) Advancing analysis of spatio-temporal variations of soil nutrients in the water level fluctuation zone of China's Three Gorges Reservoir using self-organizing map. PLoS One 10:e0121210.  https://doi.org/10.1371/journal.pone.0121210 CrossRefGoogle Scholar
  32. Yuan H, Wang L, Gang H, Yanhui Z (2008) Release of N, P from submerged soil in the shore- area of Three Gorges Reservoir. Res Environ Sci 21:103–106 (in Chinese with English abstract)Google Scholar
  33. Zhang B. (2013) Study on distribution characteristics and flux of organic matter, nitrogen and phosphorus in the soil of WLFZ of Three Gorges Reservoir PhD. Chongqing University, Chongqing, China, (in Chinese with English abstract)Google Scholar
  34. Zhang B, Fang F, Guo JS, Chen YP, Li Z, Guo SS (2012) Phosphorus fractions and phosphate sorption-release characteristics relevant to the soil composition of water-level-fluctuating zone of Three Gorges Reservoir. Ecol Eng 40:153–159.  https://doi.org/10.1016/j.ecoleng.2011.12.024 CrossRefGoogle Scholar
  35. Zhou J, Li CH, Zhang ZY, Hu HQ, Wan CY, Hu L, Pan XJ (2018) Effects of flooding and drying on the transformation of soil inorganic phosphorus in the water-level-fluctuating zone of the Three Gorges Reservoir, China. Environ Sci 39:130–136 (in Chinese with English abstract)Google Scholar

Copyright information

© Springer-Verlag GmbH Germany, part of Springer Nature 2018

Authors and Affiliations

  1. 1.Key Laboratory of Mountain Surface Processes and Ecological RegulationInstitute of Mountain Hazards and Environment, Chinese Academy of SciencesChengduChina
  2. 2.School of Land and ResourcesChina West Normal UniversityNanchongChina

Personalised recommendations