Nitrogen and phosphorus losses by runoff erosion: Field data monitored under natural rainfall in Three Gorges Reservoir Area, China

Catena 147 (2016) 797–808

Contents lists available at ScienceDirect

Catena journal homepage: www.elsevier.com/locate/catena

Nitrogen and phosphorus losses by runoff erosion: Field data monitored under natural rainfall in Three Gorges Reservoir Area, China Xiao Ma a,b,⁎, Ye Li c,⁎⁎, Bolin Li c, Weiyi Han c, Dongbin Liu c, Xiaoze Gan d a

College of Urban and Environmental Sciences, Hubei Normal University, 11 Cihu Road, Huangshi 435002, PR China Hubei Key Laboratory of Pollutant Analysis & Reuse Technology, Hubei Normal University, 11 Cihu Road, Huangshi 435002, PR China c School of Resource and Environmental Engineering, Wuhan University of Technology, 122 Luoshi Road, Wuchang District, Wuhan 430070, PR China d Hubei Province Agriculture Ecological Environmental Protection Station, Department of Agriculture of Hubei, 519 Wuluo Road, Wuchang District, Wuhan 430070, PR China b

a r t i c l e

i n f o

Article history: Received 22 January 2016 Received in revised form 10 August 2016 Accepted 5 September 2016 Available online 17 September 2016 Keywords: Agricultural non-point source Runoff Natural rainfall Three Gorges Reservoir Area Nitrogen Phosphorus

a b s t r a c t Nitrogen (N) and phosphorous (P) from agricultural non-point source (ANPS) pollution are the great threats to the regional water quality in the Three Gorges Reservoir Area (TGRA) of China. To explore the pollution of N and P loss from the sloping land in TGRA agricultural areas, field-scale plot experiment under natural rainfall was conducted from May 1 to October 31 in 2012. We monitored and analyzed rainfall, runoff volume, sediment yield, and nutrient concentration under different conditions (3 soil types, 3 surface slopes, and 5 cropping systems) in 30 experimental plots of 20 m2 during the rainy season. The results indicated that average loss ratio of N was 1.90%, while that of P was 1.54%. N and P loss ratios from surface runoff in 15° plots of sloping farmland were the highest. Purple soil (PS) was the most severe soil and nutrient loss. Within all the 5 cropping systems, the loss ratios of N and P from intercropping of citrus and grass (C-G) were the lowest. The practice of C-G, which was the most beneficial cropping system, should be further encouraged especially in TGRA to cope with the serious issues of N and P losses. These findings provide useful and valuable information for decision makers and planners to take sustainable measures for the control of ANPS pollution in TGRA, and are beneficial for the agro-ecological environment management of the local agricultural watersheds. © 2016 Elsevier B.V. All rights reserved.

1. Introduction Eutrophication caused by uncontrolled discharges of nitrogen (N) and phosphorus (P) has become one of the most rapidly growing environmental issues (Xu et al., 2013; Ma et al., 2015). The situation is getting worse and worse in Three Gorges Reservoir Area (TGRA), People's Republic of China (Ma et al., 2011; Shen et al., 2013). Recently, the significant improvement in point source depuration technologies has highlighted problems regarding N and P pollution of surface and groundwater caused by agricultural non-point sources (ANPS) (De Wit and Giuseppe, 2001; Chen et al., 2009). In TGRA, the low quality of the available land and the shortage of easily cultivated land have driven cultivation to spill over into marginal lands, which results in conflict to the official policy of de-farming less productive lands (Jim and Yang, 2006). Moreover, 38.3% of the area is utilized for traditional agriculture, with a high cropping index (average times of crops grown in 1 year) and frequent tillage, which has led to serious soil erosion and ecological deterioration (Wei et al., 2008; Xia et al., 2014). ⁎ Corresponding author at: College of Urban and Environmental Sciences, Hubei Normal University, 11 Cihu Road, Huangshi 435002, PR China. ⁎⁎ Corresponding author. E-mail addresses: [email protected] (X. Ma), [email protected] (Y. Li).

http://dx.doi.org/10.1016/j.catena.2016.09.004 0341-8162/© 2016 Elsevier B.V. All rights reserved.

The widespread use of fertilizers, pesticides, composting is not the sole reason responsible for such pollution, but the same agricultural practices that favour the release of the above mentioned sub-products into the environment are an equal contributor (Kuhn et al., 2012; Recanatesi et al., 2013). Indeed nutrient carried by eroding sediments and water runoff may degrade surface water quality while those leached into soil and through the crop root zone by infiltration may eventually contaminate groundwater. Extensive research efforts have identified and quantified factors contributing to N and P losses in runoff including the amount and type of fertilizer or manure applied (Kleinman and Sharpley, 2003; Brennan et al., 2012), timing of the rainfall event after application of fertilizer or manure (Smith et al., 2007; Allen and Mallarino, 2008), the volume of runoff generated, antecedent hydrologic conditions and field position, flow path length (McDowell and Sharpley, 2002), vegetative cover (Zhang et al., 2003a) and surface slope (Wang et al., 2013). The length of time that has elapsed since manure application was also found to affect runoff nutrient concentrations (Gilley and Eghball, 2002). For example, Randall and Mulla (2001) reported that the largest losses of N in an agricultural field occurred after frequent precipitation events with above normal precipitation and when crops were not actively growing. The heavy rainfall increased both surface runoff and loss of particulate N with the simultaneous erosion of topsoil (Kwong

798

X. Ma et al. / Catena 147 (2016) 797–808

et al., 2002). Simultaneously, the highest soil P loss occurred in years with above normal precipitation and during fallow periods (Udawatta et al., 2004). Besides, Eghball and Gilley (1999) found that no-till application of manure resulted in greater dissolved P but less particulate P loss in runoff than when manure was incorporated by disking. Particulate P transport increases with increasing runoff and erosion, which is attributed to the stronger rainfall intensities (Quinton et al., 2001). Soil erosion can be reduced below the desirable threshold and soil properties greatly improved by selecting appropriate tillage methods and land cover patterns, among which mulching has been widely studied and proven to be effective (Abrisqueta et al., 2007; Liu et al., 2012). These research findings elaborate the mechanistic approach regarding the fate of N and P from soil to water bodies and help the development of management practices capable of minimizing the excessive nutrient problem. Following construction of the Three Gorges Dam, many farmers resettled in surrounding mountain areas and cultivated marginal lands, which are mostly on steep slopes with soil of poor structure. In addition, the topography is largely rugged and steep which speeds up erosion in some small watersheds of TGRA, transport of eroded soil and runoff into rivers while giving less chance for deposition. Due to high operational costs and measurement difficulties, there are few in situ monitoring data suitable for calibrating an ANPS pollution loading model in TGRA. Hence, a field-scale plot study was conducted to investigate the characteristic of N and P losses from the sloping land in TGRA agricultural land, and to evaluate the impact of the contributing factors (soil type, surface slope, and cropping system) on runoff, sediment yield, and nutrient depletion.

and hilly areas in Chongqing Municipality. It consists of two cities (Chongqing Urban and Yichang Urban) and 19 counties or districts, with a total area of 58,000 km2 (Fig. 1). In 2012, the rural population was 11.5 million, accounted about 68% of the total population in TGRA (MEPPRC, 2013). Geographically, the TGRA is bordered by the foothills of the Daba Mountains in the north and the margin of the Yunnan–Guizhou Plateau in the south. Due to the monsoon climate, there is obvious seasonality at the TGRA. The precipitation ranges from 1000 to 1400 mm, with 80% of the rainfall occurring between April to October. The temperature is approximately 10.8–18 °C. After the water level of TGRA was raised, many slopes along the Changjiang River and its tributaries began to deform, and a number of mass movements were reactivated or newly produced (Ehret et al., 2010). At the beginning of the resettlement, some farmers need to cultivate the sloping land due to the lack of arable lands. TGRA has a complex and rugged topography, with 95.7% of the total area consisting of mountains and hills at elevations of 175 m to about 2000 m. The main soil types in this area are purple soil (PS, Regosols in FAO taxonomy, Entisol in USD A taxonomy), calcareous soil (CS, Cambisols in FAO taxonomy, Inceptisols in USD A taxonomy) and yellow soil (YS, Ferralosols in FAO taxonomy, Oxisols in USD A taxonomy), which accounting for 47.8%, 34.1% and 16.3%, respectively. Agriculture and crop farming is the principal economic activity in the region. Traditionally the TGRA is well-known for rice, wheat, peanut, corn, citrus, tea, and medicinal plant (Fig. 2). About 70% of farmland in TGRA is sloping land, and 17.60% is located on N25° slopes (Fig. 2), through field investigated by Hubei Province Agriculture Ecological Environmental Protection Station, P.R. China.

2. Materials and methods

2.2. Experimental design and treatment

2.1. Study area

Based on agriculture land characteristics of TGRA, including cultivated slopes, soil types, and cropping systems, 30 field runoff plots were designed and constructed in 2004, 2005, and 2009, located in Xingshan (XS), Zigui (ZG), Badong (BD), Kaixian (KX), Wanzhou (WZ), and Shizhu (SZ), with 5 plots in each county of TGRA (Table 1). Data were

TGRA lies between 28°56′–31°44′N and 106°16′–111°28′E (Fig. 1), covers the lower section of the upper Changjiang River, including a low–mountain canyon in Hubei Province and a parallel valley of ridged

Fig. 1. Location of experimental plots in TGRA.

X. Ma et al. / Catena 147 (2016) 797–808

799

Fig. 2. The characteristic of agriculture land in TGRA.

collected during the rainy season from May 1 to October 31 in 2012 based on field monitoring of 30 plots of TGRA under natural rainfall, including 3 soil types: PS, YS, and CS; 3 surface slopes: 5°, 15°, and 25°; 5 cropping systems: intercropping of citrus and grass (C-G), rotation of Table 1 Description and fertilizer application amount of experimental plots. Fertilizer (kg/20 m2) Number

County

Slope (°)

Soil type

Cropping system

N

P

K

XS-01 XS-02 XS-03 XS-04 XS-05 ZG-06 ZG-07 ZG-08 ZG-09 ZG-10 BD-11 BD-12 BD-13 BD-14 BD-15 KX-16 KX-17 KX-18 KX-19 KX-20 WZ-21 WZ-22 WZ-23 WZ-24 WZ-25 SZ-26 SZ-27 SZ-28 SZ-29 SZ-30

Xingshan

5 5 5 25 25 5 15 25 25 5 5 5 15 25 25 5 5 15 25 25 5 5 15 15 25 5 5 15 15 25

CS CS CS YS YS YS PS YS PS PS YS PS YS YS PS YS PS PS YS PS PS CS PS PS CS YS CS CS YS PS

C-W C-G V-P C-G V-P C-W V-P R-S V-P C-G R-S R-S C-W V-P R-S R-S V-P B-C-S B-C-S R-S V-P R-S R-S C-W C-W C-W V-P V-P C-G C-G

0.53 0.76 0.53 0.45 0.38 0.88 0.80 0.76 0.88 0.80 1.41 1.24 2.17 1.25 1.24 0.73 0.30 0.74 1.41 0.54 1.18 1.54 1.33 1.14 1.14 2.16 2.16 2.16 2.16 2.16

0.30 0.30 0.30 0.45 0.38 0.44 0.40 0.38 0.44 0.40 0.57 0.54 0.87 0.87 0.54 0.16 0.20 0.28 0.53 0.20 0.78 0.60 0.93 0.68 0.40 1.08 1.08 1.08 1.08 1.08

0.30 0.30 0.30 0.45 0.38 0.33 0.30 0.29 0.33 0.30 0.30 0.30 0.60 0.70 0.48 0.12 0.20 0.21 0.43 0.15 0.68 0.30 0.83 0.30 0.30 0.81 0.81 0.81 0.81 0.81

Zigui

Badong

Kaixian

Wanzhou

Shizhu

vegetable and potato (V-P), rotation of rape and sweet potato (R-S), rotation of corn and wheat (C-W), and rotation of bean and corn and sweet potato (B-C-S). Dimensions of experimental plots were 10 m × 2 m. The plots were hydrologically isolated with partition walls filled with cement, inserted at least vertically 15 cm inserted into soil and 20 cm above soil surface to avoid unexpected seepage to the individual plot. At the bottom of each plot, a water-collection tank (width, length, and depth of 1 m each) was constructed to collect runoff and eroded soil after rainstorms. After site formation was completed, plots were delineated by concrete perimeters. All the plots had been built before December 2009. Characteristics of each runoff plot were given in Table 1 and Fig. 3. Information on tillage practices, fertilizer application rates and application timing at the study site was obtained from the landowners. For all plots, compound fertilizer (N 15%, P2O5 10%, K2O 10%) and manure (N 0.8%, P2O5 0.4%, K2O 0.3%) were applied as Table 1. Herbicide was applied as needed in 2012 to prevent weed growth. 2.3. Sampling and analysis Monitoring of surface runoff caused by natural rainfall in these plots was carried out from May 1 to October 31 in 2012. A total of 227 water samples and 29 sediment samples were taken from the 30 plots. At each plot site a self-recording rain gauge was installed to record daily rainfall amount. No facility was, however, available to measure rainfall intensity. Samples collected on the same day were mixed to record the data on daily basis for days with more than one rainfall event. Measurement of rainfall amount, runoff volume and sediment yield were made during the study period. In April 2012, before the rainy season, the soils of each plot were collected from 0 to 20 cm soil depth using soil core (70 mm diameter and 200 mm in depth). Before collecting soil sample each time, the surface aboveground litter and organic residues were removed. Each soil sample was a composite of 4 sampling cores, which was along the slope from the top to its bottom for every 2.5-m distance. After each rainfall, the eroded soil and the water captured in the tank were collected

800

X. Ma et al. / Catena 147 (2016) 797–808

Fig. 3. Photographs of experimental plots in TGRA (Xingshan).

when they are N 2.5 kg. Runoff volumes for each event were measured, and they were calibrated according to the volume of water and dried sediment contained in wet soil removed. a) After 15 min of sedimentation in tank, runoff water was sampled and sealed in plastic bottles. Then it was put in coolers at b 4 °C and immediately taken to the laboratory. b) Sediment concentration was tested by heat drying weighing method. The runoff water was carefully mixed with the eroded soil in the bottom of tank. And then, sediment amount was determined by multiplying the volume of runoff and the sediment concentration. After collected, all the sediment samples of each plot were mixed separately and stored in a cooler, then transported into laboratory for further chemical analysis. Chemical analysis was made on composite samples of water and eroded soil collected in each plot over the entire rainy season. For runoff water, the total nitrogen (TN) concentration in the filtrate was measured by the colorimetric method following sulfuric acid digestion in a block digester (Xie and Wang, 1998), while the total phosphorous (TP) concentration in the filtrate was determined colorimetrically (710 nm) using the procedure of molybdenum-antimony after digestion with K2S2O8. All the above measurements were based on the standard analytical methods (Xie and Wang, 1998; American Public Health Association et al., 1985), and analyzed by UV-220 spectrophotometry. The soil and eroded soil samples were air dried, ground through 1mm sieve, to measure pH, soil organic matter (SOM) or sediment organic matter (OMsed), soil total nitrogen (STN) or sediment total nitrogen (TNsed), soil total phosphorus (STP) or sediment total phosphorus

(TPsed), soil available (SAN) or sediment available nitrogen (ANsed), and soil available phosphorus (SAP) or sediment available phosphorus (APsed). pH were measured at a soil to solution ratio of 1:2.5. SOM and OMsed were determined using the dichromate oxidation method (Lu, 2000). STN content and TNsed content were measured by the Kjeldahl method (Lu, 2000), while SAN and ANsed were measured by the alkaline hydrolysis method described by Lu (2000). Extraction of SAP (APsed) and STP (TPsed) were carried out using 0.5 M NaHCO3 and HClO4–H2SO4, respectively, and assayed using the ammonium molybdate method (Olsen and Sommers, 1982). 2.4. Statistical analyses Eroded soil N and P losses were calculated by multiplying each of the contents by the corresponding sediment yields, while runoff water TN and TP losses were calculated by multiplying each of the concentrations by the corresponding runoff volumes. The N and P loss loads of runoff water (Lw, kg/m2) and sediment (Ls, kg/m2) were converted to area basis using the following equation: Lw ¼ 10‐6 

n n X X ðCwi V i Þ=S or Ls ¼ 10‐6  ðCsi Mi Þ=S i¼1

i¼1

where Cwi (mg/L) or Csi (mg/kg) represent the average nutrient concentration of water or sediment in i sampling duration, Vi (L) or Mi (kg) represent runoff volume or sediment yield, S (20 m2) represents

Table 2 Soil characteristics of the experimental plots in 2012.

Soil type

Slope (°)

Cropping system

PS (n = 12)⁎⁎⁎ YS (n = 11) CS (n = 7) 5 (n = 13) 15 (n = 7) 25 (n = 10) C-G (n = 5) V-P (n = 9) R-S (n = 8) C-W (n = 6) B-C-S (n = 2)

pH

SOM (g/kg)

STN (g/kg)

SAN (mg/kg)

STP (g/kg)

SAP (mg/kg)

6.4 (0.22)⁎ 6.0 ~ 6.7⁎⁎ 6.4 (0.20) 6.0 ~ 6.7 6.3 (0.21) 5.9 ~ 6.5 6.4 (0.22) 5.9 ~ 6.7 6.4 (0.22) 6.0 ~ 6.7 6.4 (0.23) 6.0 ~ 6.6 6.3 (0.28)a 6.0 ~ 6.6 6.4 (0.12)ab 6.2 ~ 6.6 6.3 (0.28)a 5.9 ~ 6.7 6.3 (0.15)a 6.0 ~ 6.4 6.7 (0.07)b 6.6 ~ 6.7

2.70 (0.96) 0.73 ~ 3.90 2.59 (0.59) 1.52 ~ 3.44 2.81 (0.24) 2.45 ~ 3.17 2.60 (0.68) 0.73 ~ 3.44 2.98 (0.85) 1.52 ~ 3.90 2.58 (0.64) 1.13 ~ 3.25 2.55 (0.39) 2.13 ~ 2.98 2.77 (0.28) 2.33 ~ 3.17 2.48 (1.15) 0.73 ~ 3.90 2.71 (0.66) 1.52 ~ 3.50 3.33 (0.25) 3.15 ~ 3.51

1.19 (0.56) 0.49 ~ 2.24 1.29 (0.42) 0.74 ~ 2.02 1.18 (0.21) 0.91 ~ 1.56 1.27 (0.41) 0.74 ~ 2.17 1.20 (0.47) 0.49 ~ 2.02 1.19 (0.49) 0.52 ~ 2.24 1.28 (0.15) 1.14 ~ 1.49 1.16 (0.32) 0.52 ~ 1.56 1.34 (0.69) 0.49 ~ 2.24 1.22 (0.44) 0.74 ~ 2.02 0.96 (0.01) 0.95 ~ 0.96

97.67 (36.51) 23.07 ~ 146.13 102.64 (25.73) 60.75 ~ 139.11 85.00 (13.11) 68.86 ~ 97.66 96.71 (22.25) 60.75 ~ 133.80 85.45 (44.90) 23.07 ~ 139.11 104.07 (21.63) 70.39 ~ 146.13 115.20 (32.43) 70.39 ~ 146.13 97.13 (18.44) 68.86 ~ 120.89 90.76 (36.27) 23.07 ~ 133.80 92.37 (31.26) 32.90 ~ 123.80 82.82 (12.72) 73.82 ~ 91.81

0.34 (0.20) 0.15 ~ 0.69 0.38 (0.16) 0.14 ~ 0.65 0.40 (0.17) 0.19 ~ 0.62 0.36 (0.19) 0.14 ~ 0.65 0.39 (0.19) 0.17 ~ 0.69 0.36 (0.16) 0.19 ~ 0.64 0.41 (0.18) 0.22 ~ 0.64 0.41 (0.19) 0.15 ~ 0.62 0.30 (0.17) 0.16 ~ 0.65 0.31 (0.12) 0.14 ~ 0.47 0.48 (0.30) 0.27 ~ 0.69

42.40 (29.21) 12.43 ~ 107.93 50.68 (65.44) 13.68 ~ 232.35 80.53 (70.24) 25.95 ~ 197.80 56.17 (59.46) 12.43 ~ 197.80 28.27 (7.77) 18.35 ~ 42.33 70.18 (65.29) 15.02 ~ 232.35 74.06 (52.36) 18.35 ~ 154.70 60.80 (61.81) 12.43 ~ 197.80 56.50 (72.49) 13.68 ~ 232.35 34.83 (27.78) 18.21 ~ 90.85 25.73 (2.87) 23.70 ~ 27.76

Note: ⁎ standard deviations; ⁎⁎ numerical range; ⁎⁎⁎sample numbers; small letters (a, b) represent statistical differences among various treatments (one-way ANOVA, Duncan's multiple range test, P b 0.05), and similarly hereinafter.

X. Ma et al. / Catena 147 (2016) 797–808

801

distribution. Then the transformed data were used for one-way analysis of variance. Duncan's test was used to compare the statistical difference at a significance level of P b 0.05. Back transformed data were recorded and graphed using SigmaPlot 12.5 (SPSS Inc., Chicago, IL, USA). 3. Results 3.1. Soil characteristics

Fig. 4. Rainfall of experimental plots in TGRA in 2012.

experimental plot area, n represents number of samples. The total loss loads of nutrient (L) are the sum of Lw and Ls. The application of compost or fertilizer at rates that exceed crop nutrient requirements can result in N and P accumulation in soil. Loss ratios of N and P were used to evaluate the transport of nutrient in runoff from manure and fertilizer applied land. The loss ratios of N (RN) and P (RP) were calculated using the following equation:

R¼

The soil chemical properties of the 30 plots, including 3 soil types, 3 surface slopes, and 5 cropping systems, are presented in Table 2. The results indicated that the soil pH was in range from 5.9 to 6.7. Among the 3 soil types, the SOM content of PS was 2.70 g/kg and that of CS was 2.81 g/kg. STN and SAN contents of YS were 1.29 g/kg and 102.64 mg/ kg, which were higher than others. STP and SAP contents of PS was 0.34 g/kg and 42.40 mg/kg, which were the lowest of all. Among the 3 surface slopes, the SOM content of 15° plots was 2.98 g/kg. STN and SAN contents of 15° plots were 1.20 g/kg and 85.45 mg/kg, which were the lowest. STP content of 15° plots was 0.39 g/kg, which was the highest of all. Nevertheless, SAP content of 15° plots was 28.27 mg/kg. Among the 5 cropping systems, the SOM content of R-S was 2.48 g/kg and that of B-C-S was 3.33 g/kg. STN content of R-S was 1.34 g/kg, which was the highest. While STP content of R-S was 0.30 g/kg, which was lowest of all.

n X ðCwi V i þ Csi M i Þ=F  100%

3.2. Surface runoff and nutrient loss

i¼1

The average annual rainfall of the study area is about 760 mm, while the count of altogether was about 37 events which produced water ≥ 30 mm during monitoring period in 2012 (Fig. 4). As shown in Fig. 4, the rainfall of KX is 986 mm in 2012, which is much greater than that of ZG, with 289 mm. The frequency of rainfall above 30 mm is 9 in KX, which is higher than that in ZG. The volumes, pH and nutrient concentrates of surface runoff in 30 experimental plots were shown in Table 3. The results indicated that the runoff pH was in range from 6.2 to 9.5. Among the 3 soil types, the runoff volume of PS was 94.97 L and that of CS was 27.08 L. However, the runoff TN concentration of CS was the highest, which was 1.71 mg/L. The runoff TP concentration of PS was the highest, which was 0.52 mg/L. Analysis of variance (ANOVA) demonstrated significative differences between runoff

where F (mg) represents N or P application amount of experimental plot. Enrichment ratio (ER) used to estimate the nutrient loss associated with soil erosion, could be obtained by the nutrient content in sediment divided by the content in initial soil (Massey and Jackson, 1952). The formula would be defined in general, i.e. enrichment ratios of STN and STP (ERSTN and ERSTP), and enrichment ratios of SAN and SAP (ERSAN and ERSAP), were calculated using the concentration of nutrients in the sediment/total concentration in soil. Raw data were subjected to normality test using PASW 18.0 software (SPSS Inc., Chicago, IL, USA), and relevant transformations were conducted so that all data set had the feature of normal

Table 3 Nutrient concentration of runoff in the experimental plots in 2012.

Soil type

Slope (°)

Cropping system

PS (n = 94) YS (n = 72) CS (n = 54) 5 (n = 92) 15 (n = 56) 25 (n = 72) C-G (n = 26) V-P (n = 66) R-S (n = 70) C-W (n = 40) B-C-S (n = 18)

Runoff volume (L)

pH

TN (mg/L)

TP (mg/L)

94.97 (160.05)b 2.50 ~ 940.40 69.77 (59.94)a 1.20 ~ 250.20 27.08 (31.57)b 2.50 ~ 155.40 44.56 (46.96)a 1.20 ~ 181.40 126.13 (198.56)b 2.50 ~ 940.40 59.02 (57.47)a 2.50 ~ 250.20 40.07 (34.74) 6.00 ~ 134.20 47.17 (45.28) 2.50 ~ 182.30 98.19 (155.80) 1.20 ~ 940.40 85.84 (142.61) 2.50 ~ 718.80 52.77 (71.97) 2.50 ~ 250.20

8.1 (0.39) 6.2 ~ 8.7 8.0 (0.35) 6.9 ~ 9.5 8.0 (0.34) 6.9 ~ 8.5 8.0 (0.44) 6.2 ~ 9.5 8.1 (0.33) 6.7 ~ 8.7 8.0 (0.25) 7.2 ~ 8.5 8.0 (0.31) 7.5 ~ 8.3 8.1 (0.41) 6.2 ~ 8.7 8.1 (0.42) 6.7 ~ 9.5 8.1 (0.26) 7.5 ~ 8.7 8.0 (0.10) 7.83 ~ 8.23

1.71 (2.44)a 0.02 ~ 10.56 1.60 (1.81)b 0.02 ~ 7.19 4.57 (3.58)a 0.37 ~ 11.68 2.94 (3.36)b 0.02 ~ 11.68 2.03 (2.59)ab 0.02 ~ 10.56 1.92 (2.29)a 0.05 ~ 10.38 5.06 (3.76)b 0.08 ~ 10.75 2.27 (2.94)a 0.05 ~ 11.68 1.72 (2.39)a 0.02 ~ 10.56 2.08 (2.43)a 0.02 ~ 9.18 2.09 (1.78)a 0.20 ~ 5.73

0.52 (0.50)b 0.01 ~ 2.45 0.37 (0.23)a 0.01 ~ 1.27 0.34 (0.29)b 0.03 ~ 1.12 0.46 (0.39) 0.01 ~ 2.45 0.39 (0.29) 0.01 ~ 1.27 0.42 (0.44) 0.01 ~ 2.39 0.67 (0.64)b 0.01 ~ 2.39 0.46 (0.46)a 0.01 ~ 2.45 0.37 (0.24)a 0.01 ~ 1.25 0.37 (0.23)a 0.04 ~ 1.00 0.35 (0.23)a 0.03 ~ 0.92

802

X. Ma et al. / Catena 147 (2016) 797–808

Table 4 Nutrient concentration of sediment in the experimental plots in 2012.

Soil type

Slope (°)

Cropping system

PS (n = 12) YS (n = 10) CS (n = 7) 5 (n = 12) 15 (n = 7) 25 (n = 10) C-G (n = 5) V-P (n = 9) R-S (n = 8) C-W (n = 5) B-C-S (n = 2)

Amount (kg)

pH

OMsed (g/kg)

TNsed (g/kg)

ANsed (mg/kg)

TPsed (g/kg)

APsed (mg/kg)

18.84 (12.75) 5.30 ~ 37.40 19.27 (12.31) 3.72 ~ 39.10 14.61 (11.93) 3.50 ~ 32.40 14.81 (11.61) 3.50 ~ 35.50 25.28 (11.44) 7.60 ~ 37.20 16.64 (12.16) 3.72 ~ 39.10 8.91 (6.38) 3.50 ~ 17.10 18.97 (12.05) 5.30 ~ 39.10 22.72 (13.69) 6.90 ~ 37.40 19.83 (13.67) 5.10 ~ 36.90 12.45 (6.86) 7.60 ~ 17.30

6.8 (0.42)ab 6.1 ~ 7.5 7.0 (0.21)b 6.7 ~ 7.4 6.5 (0.32)a 6.1 ~ 6.9 6.8 (0.44) 6.1 ~ 7.5 6.8 (0.38) 6.2 ~ 7.2 6.9 (0.29) 6.2 ~ 7.2 6.8 (0.19)ab 6.5 ~ 7.0 6.9 (0.38)ab 6.1 ~ 7.5 6.8 (0.44)ab 6.1 ~ 7.4 6.6 (0.36)a 6.2 ~ 7.0 7.2 (0.07)b 7.1 ~ 7.2

2.92 (0.78)b 1.84 ~ 4.06 2.47 (0.65)ab 0.76 ~ 3.13 1.81 (1.23)a 0.22 ~ 3.23 2.25 (1.05) 0.22 ~ 3.73 3.00 (0.82) 1.86 ~ 4.06 2.44 (0.83) 0.37 ~ 3.13 2.72 (0.49) 1.89 ~ 3.10 2.83 (0.67) 1.32 ~ 3.73 2.39 (0.97) 0.76 ~ 4.06 1.87 (1.57) 0.22 ~ 3.95 2.42 (0.78) 1.86 ~ 2.97

1.18 (0.39)a 0.70 ~ 2.07 1.17 (0.24)a 0.90 ~ 1.50 1.54 (0.29)b 1.20 ~ 1.90 1.26 (0.33) 0.70 ~ 1.90 1.37 (0.45) 0.80 ~ 2.07 1.20 (0.31) 0.70 ~ 1.80 1.18 (0.15)a 1.00 ~ 1.40 1.29 (0.28)ab 0.90 ~ 1.80 1.06 (0.34)a 0.70 ~ 1.50 1.46 (0.40)ab 1.00 ~ 1.90 1.69 (0.54)b 1.31 ~ 2.07

105.99 (68.69) 22.95 ~ 258.60 143.89 (64.82) 52.55 ~ 285.95 159.29 (120.46) 35.19 ~ 393.37 140.10 (97.31) 23.96 ~ 393.37 118.73 (92.37) 22.95 ~ 258.60 131.35 (61.60) 52.55 ~ 285.95 129.90 (14.32)a 106.58 ~ 142.30 99.36 (62.52)a 23.96 ~ 239.58 111.83 (48.86)a 33.66 ~ 153.02 168.56 (144.44)a 22.95 ~ 393.37 272.28 (19.34)b 258.60 ~ 285.95

0.45 (0.10) 0.27 ~ 0.58 0.51 (0.16) 0.35 ~ 0.75 0.54 (0.08) 0.43 ~ 0.68 0.47 (0.12) 0.29 ~ 0.68 0.51 (0.06) 0.42 ~ 0.58 0.50 (0.13) 0.27 ~ 0.75 0.48 (0.16)ab 0.33 ~ 0.75 0.54 (0.06)ab 0.40 ~ 0.63 0.40 (0.96)a 0.27 ~ 0.54 0.56 (0.08)b 0.46 ~ 0.68 0.49 (0.01)ab 0.48 ~ 0.50

51.83 (29.33)ab 16.96 ~ 105.24 39.50 (19.51)a 15.46 ~ 65.21 77.99 (51.46)b 34.20 ~ 157.61 62.59 (44.16) 15.46 ~ 157.61 38.74 (19.69) 19.10 ~ 66.39 54.06 (31.04) 15.65 ~ 105.24 48.14 (23.95) 19.10 ~ 84.30 51.96 (45.87) 16.96 ~ 157.61 49.32 (26.54) 15.46 ~ 105.24 75.58 (40.52) 30.20 ~ 137.71 41.02 (35.88) 15.65 ~ 66.39

nutrient concentrations of PS and those of YS. The runoff volume of 15° plots was the highest, with 126.13 L. While the TN and TP concentrations of 15° plots were the lowest, with 2.03 mg/L and

0.39 mg/L. Significative differences (P b 0.05) among runoff TN concentration of 15° plots, those of 5° plots and 25° plots could be noticed. No significant differences in TP concentrations in

Fig. 5. Nutrient loss loads and loss ratios of runoff from different soil types.

X. Ma et al. / Catena 147 (2016) 797–808

runoff water among 3 surface slopes were observed. For the runoff volume of 5 cropping system, C-G was the least and R-S was the most. The TN and TP concentrations of C-G were the highest of all, with 5.06 mg/L and 0.67 mg/L. There existed significative differences among runoff nutrient concentrations of C-G and those of other four cropping systems. The amount, pH, OMsed, and nutrient concentrates of soil removed in 30 experimental plots were shown in Table 4. The results indicated that the sediment pH was in range from 6.1 to 7.5. Among the 3 soil types, the eroded soil amount of YS was 19.27 kg and that of CS was 14.61 kg. The OMsed content of PS was 2.92 g/kg and that of CS was 1.81 g/kg. The TNsed, ANsed, TPsed, and APsed contents of CS were 1.54 g/kg, 159.29 mg/kg, 0.54 g/kg, and 77.99 mg/kg, which were higher than others. ANOVA showed that there are some significative differences among TNsed concentration of CS and that of PS and YS. No significant differences in TPsed concentrations among the soil types were observed. Simultaneously, the eroded soil amount, OMsed, TNsed, and TPsed contents of 15° plots were the most of all the three slope gradient plots. While the ANsed and APsed contents of 15° plots were the least, with 118.73 mg/kg and 38.74 mg/kg. Among the 5 cropping systems, as with runoff volume, the eroded soil amount of C-G was the least, with 8.91 kg, while that of R-S was the most of all. The OMsed content of V-P was the highest, following with that of C-G, with 2.72 g/kg. TNsed content of R-S was the least, with 1.06 g/kg and, while ANsed content of V-P was the least, with 99.36 mg/kg. However, the TPsed

803

and APsed contents of C-W were the highest, with 0.56 g/kg and 75.58 mg/kg. 3.3. Nutrient loss of different soil types N and P loss loads of surface runoff from PS plots are the most, with 1027 kg/km2 and 368 kg/km2, while those from CS plots, with 827 kg/ km2 and 266 kg/km2, are the least (Fig. 5). N and P loss ratios of surface runoff from PS (2.10% and 1.72%) and YS (2.11% and 1.74%) are higher than those from CS (1.88% and 1.24%) (Fig. 5). In addition, ERSTN and ERSAN of CS are 1.22 and 1.39, which are higher than those of PS and YS, while ERSTN and ERSTP, ERSAP and ERSAP of PS are the highest and those of YS are the lowest (Fig. 6). ERSTP is higher than ERSAP for all the plots of different soil types. ERSTN of PS and CS is lower than ERSAN of those, while ERSTN of YS is higher than ERSAN of YS. 3.4. Nutrient loss of different surface slopes Both loads and ratios of nutrient loss from surface runoff of 15° plots are more than those of 25° plots and 5° plots. The N and P loss loads of 15° plots are the most, with 1577 kg/km2 and 732 kg/km2, while those from 5° plots are the least, with 593 kg/km2 and 266 kg/km2 (Fig. 7). Significative differences were observed among LN/LP of 15° plots, those of 5° plots and 25° plots. N and P loss ratios of surface runoff from 15° plots (2.63% and 2.17%) are higher than those from 5° plots

Fig. 6. Nutrient enrichment ratios of sediments from different soil types.

804

X. Ma et al. / Catena 147 (2016) 797–808

Fig. 7. Nutrient loss loads and loss ratios of runoff from different slops. Note: small letters (a, b) represent statistical differences among various treatments (one-way ANOVA, Duncan's multiple range test, P b 0.05), and similarly hereinafter.

(1.88% and 1.61%) and 25° plots (1.96% and 1.56%) (Fig. 7). Significative differences could be noticed among RN of 15° plots, 5° plots and 25° plots. In addition, ERSAN and ERSAP of 15° plots are 1.42 and 1.08, which are the highest, while those of 5° plots are the lowest (Fig. 8). ERSTN of 25° plots is 0.99, which is lower than that of 5° plots and 15° plots, while ERSTP of 25° plots is 1.70, which is higher than that of 5° plots and 15° plots (Fig. 8). For all the plots of different slopes, ERSTP is higher than ERSAP, while ERSTN is lower than ERSAN.

3.5. Nutrient loss of different cropping systems The total load of N from C-G is 353 kg/km2, corresponding to 20% of that from R-S, and the total load of P is 144 kg/km2, corresponding to 27% of that from R-S (Fig. 9). The loss ratios of N and P from C-G are 0.91% and 0.64% (Fig. 9). There existed significative differences among RN/RP of C-G and those of other four cropping systems. ERSTN and ERSTP from C-G are 0.86 and 1.36, while ERSAN and ERSAP are 1.01 and 0.96 (Fig. 10). Significative differences were observed among ERSAN of B-C-S and that of other four cropping systems, while there were some significative differences among ERSAP of C-G and that of other four cropping systems. Additionally, the LN and LP from R-S are more than those of other cropping systems, with 1117 kg/km2 and 464 kg/km2. The N and P loss ratios of surface runoff from C-W are the highest of all, with 2.23% and 2.13%, while ERSTN

and ERSTP (1.08 and 1.89) and ERSAN and ERSAP (1.31 and 1.12) are more than all the others. 4. Discussion The experiment results indicated that nutrient concentration was directly affected by the runoff volume and eroded soil amount. Additionally, it could be calculated and concluded that most N and P transported with the runoff water while only a small portion of those were lost in eroded soil. This result was much different from that of Sharpley et al. (1987), who found that soil removed contributed average 64% of N and 75% of P in surface runoff. It has been reported that uncommon storm events are the main cause leading to serious soil erosion, and also nutrient loss (Zhang et al., 2003b). Particularly in rained agriculture areas, the main problem may be associated with greater irregularities in rainfall distribution and their concentration in a small number of events. The processes of N and P loss from surface runoff diverse from each other in sloping land of different soil types, slopes, and cropping systems. As a matter of fact, the nutrient losses in agricultural land depend on characteristics of the rainfall (quantity, intensity, and duration). Simultaneously, surface soil conditions could influence infiltration (residue cover and soil physical properties, including water content), and subsoil properties will also affect hydrologic conductivity (soil structure, texture, water content). Additionally, it still include factors like fertilizers and manure

X. Ma et al. / Catena 147 (2016) 797–808

805

Fig. 8. Nutrient enrichment ratios of sediments from different slops.

application (amount, method and timing), and water table depth (Havlin, 2004). These different contributing factors can impose effect on N and P losses by the means of affecting quantities, times, and processes of runoff water and soil removed. On the other side, LN and LP of 15° plots are the most of all three slope gradient plots. It is consistent with the findings of Geng et al. (2010) and Wang et al. (2013), regarding the effect of slope gradient on soil erosion process on purple soil hill slopes. Slope gradient affects the content of N and P losses by affecting the runoff volume in slope. Factors such as hydraulic friction, percussion angle of rain drops and rain area all change with increased slope gradient (Gilley et al., 2007; Wang et al., 2011). The change is related with water speed distribution in runoff, runoff volumes in the slope and the interaction intensity between runoff and soil removed (Wang et al., 2013). Slope farming is commonly practiced in TGRA where the topography is hilly and the majority of population relay on agriculture. Due to limited suitable farming land resources, a large amount of chemical fertilizer has been applied to the sloping land to meet the expectation of increasing unit productivity, which resulting in increasing the risk of N and P loss from surface runoff of ANPS. Engineering and biological measures are highly desirable to meet the challenging issue of N and P loss load, especially in sloping land under or about 15° of TGRA. However, the farmland above 15° should be continued restoring arable land to nature firstly, or it can be cultivated after remedial measures were taken. In fact, the magnitude of nutrient loss from agroforestry systems associated with the canopy interception of rainfall, the infiltration and

water holding capacity of soil by changing microclimate and soil structure (Girmay et al., 2009). Plants play an important role in improving soil anti-erosion capability and in preventing soil losses by crusts. The amount of crop residue on the soil surface has been found to have a minimal effect on nutrient concentrations in runoff (Nicolaisen et al., 2007; Gilley et al., 2007). Conventional slope farming is not considered a sustainable agricultural practice, because no soil conservation measures are adopted to tackle problems of soil erosion and nutrient depletion associated with the local steep topography (Ng et al., 2008). Agricultural cultivation depletes soil fertility and causes structural degradation, compaction and reduces permeability (Igwe, 2001; Wei et al., 2013). This leads to higher ratios of overland flow, soil erosion and an associated increase in nutrient loss. R-S and C-W are commonly used by farmers in the TGRA. Under conventional management, tillage using spades and hoes is performed twice a year during a rotation, in June for maize seedlings and in November for wheat seedlings. Fertilizers nutrient are then applied to the soil surface after tillage. The topsoil may be distorted when the sweet potato or potato are digging up, resulting in elevated eroded soil and runoff. This type of cultivation has been confirmed to lead to serious soil erosion and nutrient losses in sloping arable land in the TGRA (Cha and Li, 1998; Otero et al., 2011). Additionally, LN and LP, RN and RP, ERSTN and ERSTP, ERSAN and ERSAP of surface runoff from C-G are the lowest of all the 5 cropping systems (Fig. 9 and Fig. 10). Citrus is perennial and evergreen plant with many flowers, requires a large amount of nutrients. Crass has large roots system to consolidate soil. C-G intercropping could increase surface vegetation coverage and decrease

806

X. Ma et al. / Catena 147 (2016) 797–808

Fig. 9. Nutrient loss loads and loss ratios of runoff from different cropping systems.

soil disturbance. Hence, C-G is the most beneficial cropping system to decrease the loss loads of N and P from surface runoff of ANPS. To cope with the serious issues of N and P losses, the practice of C-G should be further encouraged especially in TGRA. Some studies have shown that inorganic forms of N (nitrate, NO− 3 and ammonium, NH+ 4 ) are readily available for uptake by plants, and dissolved phosphate (PO3− 4 ) is considered to be the most plant-available form of P (Cotner and Wetzel, 1992; Turnbull et al., 2011). For another, eroded soil nutrient account for most of the total nutrient loss in runoff, serving as the long-term sources in aquatic ecosystems, which can be transformed to soluble, bioavailable form in aquatic ecosystems, depending on their relative concentrations, water temperature, and dissolved oxygen (Sharpley and Withers, 1994; Heathwaite and Johnes, 1996; Yang et al., 2014). However, a change in nutrient dynamics during runoff events may have implications for ratios of nitrification and denitrification, and may therefore have broader consequences for biogeochemical cycling (Peterjohn and Schlesinger, 1990; Schlesinger et al., 1990). N and P losses from agriculture land are closely depended on properties such as forming environment, structure, and physico-chemical properties of the soil. In general, N and P are easy to loss in low OMsed content and bad soil structure agriculture land. The selective erosion of clay particles was the main reason for N and P enrichments. Sloping land in the region accounts for a large proportion of the whole purple soil area (Chen et al., 2011). Therefore, N and P losses from surface runoff in sloping farmland of purple soil are one of the most important ANPSs effecting water quality of TGRA. Many researches (Jia et al., 2007; Geng et al., 2010; Wang et al., 2011) have indicated that slope hydrological processes and nutrient transformation of purple topsoil

associated with serious erosion. Purple soil P retention by eroded soils adsorption plays a pivotal ecological role in P buffering in freshwater ecosystems (Luo et al., 2009).

5. Conclusions Based on the field observation and monitoring in TGRA, this study has increased the ability to develop reliable models for estimating the impact of agricultural production on nutrient runoff, which will be valuable in estimating N and P losses from sloping land and identifying areas on different soil types where nutrient loss remediation should be targeted. Our project monitored eroded soil and nutrient loss in runoff from 30 field-scale plots in agricultural sloping land of TGRA in 2012, and analyzed the impact of the contributing factors on nutrient loss, including soil type, surface slope, and cropping system. Results indicate that most N and P transported with the runoff water while only a small portion of those were lost in soil removed. To protect the agro-ecological environment and to search for and extend sustainable agriculture production methods in TGRA, engineering and biological measures are highly desirable to reduce nutrient loss in sloping land under or about 15°, especially in purple soil. In addition, C-G is the most beneficial cropping system to decrease the nutrient loss loads from surface runoff of ANPS. The practice of C-G should be further encouraged especially in TGRA, to cope with the serious issues of N and P losses. Furthermore, determining the accurate amount of soil nutrient of fertilizers transfer into surface runoff and reducing fertilizers loss are essential for the

X. Ma et al. / Catena 147 (2016) 797–808

807

Fig. 10. Nutrient enrichment ratios of sediments from different cropping systems.

agricultural management of ANPS, which will be studied in the next step. Acknowledgements This research was jointly supported by Major Science and Technology Programs of National Water Pollution Control and Management, China (2012ZX07104-002), Open Fund of Hubei Key Laboratory of Pollutant Analysis & Reuse Technology, China (PA160205) and the Fundamental Research Funds for the Central Universities, China (2012-YB17). The authors gratefully acknowledge the valuable datum provided by Hubei Province Agriculture Ecological Environmental Protection Station, P.R. China. Appendix A. Supplementary data Supplementary data associated with this article can be found in the online version, at doi: http://dx.doi.org/10.1016/j.catena.2016.09.004. These data include the Google maps of the most important areas described in this article. References Abrisqueta, J.M., Plana, V., Mounzer, O.H., Mendez, J., Ruiz-Sánchez, M.C., 2007. Effects of soil tillage on runoff generation in a Mediterranean apricot orchard. Agric. Water Manag. 93, 11–18. Allen, B.L., Mallarino, A.P., 2008. Effect of liquid swine manure rate, incorporation, and timing of rainfall on phosphorus loss with surface runoff. J. Environ. Qual. 37, 125–137.

American Public Health Association, American Water Works Association, Water Environment Federation, Water Environment Federation, 1985. Standard Methods for the Examination of Water and Wastewater. 18th ed. APHA, Washington DC. Brennan, R.B., Healy, M.G., Grant, J., Ibrahim, T.G., Fenton, O., 2012. Incidental phosphorus and nitrogen loss from grassland plots receiving chemically amended dairy cattle slurry. Sci. Total Environ. 441, 132–140. Cha, S.Y., Li, Q.H., 1998. Soil erosion from sloping arable land in TGA and the countermeasures to deal with. Agro-Eviron. Develop. 15, 30–33. Chen, D.J., Lu, J., Shen, Y.N., Dahlgren, R.A., Jin, S.Q., 2009. Estimation of critical nutrient amounts based on input-output analysis in an agriculture watershed of eastern China. Agric. Ecosyst. Environ. 134 (3), 159–167. Chen, L.G., Qian, X., Shi, Y., 2011. Critical area identification of potential soil loss in a typical watershed of the three gorges reservoir region. Water Resour. Manag. 25, 3445–3463. Cotner, J.B., Wetzel, R.G., 1992. Uptake of dissolved inorganic and organic phosphorus compounds by phytoplankton and bacterioplankton. Limnol. Oceanogr. 37, 232–243. De Wit, M., Giuseppe, B., 2001. Nutrient fluxes in the Po basin. Sci. Total Environ. 273 (1), 147–161. Eghball, B., Gilley, J.E., 1999. Phosphorus and nitrogen in runoff following beef cattle manure or compost application. J. Environ. Qual. 28 (4), 1201–1210. Ehret, D., Rohn, J., Dumperth, C., Eckstein, S., Ernstberger, S., Otte, K., Rudolph, R., Wiedenmann, J., Wei, X., Renneng, B., 2010. Frequency ratio analysis of mass movements in the Xiangxi catchment, Three Gorges Reservoir area, China. J. Earth Sci. 21, 824–834. Geng, X., Zheng, F., Liu, L., 2010. Effect of rainfall intensity and slope gradient on soil erosion process on purple soil hill slopes. J. Sediment. Res. 6, 48–53. Gilley, J.E., Eghball, B., 2002. Residual effects of compost and fertilizer applications on nutrients in runoff. Trans. ASAE 45 (6), 1905–1910. Gilley, J.E., Eghball, B., Marx, D.B., 2007. Nutrient concentrations of runoff during the year following manure application. Trans. ASABE 50 (6), 1987–1999. Girmay, G., Singh, B.R., Nyssen, J., Borrosen, T., 2009. Runoff and sediment-associated nutrient losses under different land uses in Tigray, Northern Ethiopia. J. Hydrol. 376 (1– 2), 70–80. Havlin, J.L., 2004. Technical basis for quantifying phosphorus transport to surface and groundwaters. J. Anim. Sci. 82, 277–291. Heathwaite, A.L., Johnes, P.J., 1996. Contribution of nitrogen species and phosphorus fractions to stream water quality in agricultural catchments. Hydrol. Process. 10, 971–983.

808

X. Ma et al. / Catena 147 (2016) 797–808

Igwe, A.C., 2001. Effects of land use on some structural properties of an Ultisol in southeastern Nigeria. Int. Agrophys. 15, 237–241. Jia, H.Y., Lei, A.L., Lei, J.S., Ye, M., Zhao, J.Z., 2007. Effects of hydrological processes on nitrogen loss in purple soil. [J]. Agr. Water Manage. 89 (1–2), 89–97. Jim, C.Y., Yang, F.Y., 2006. Local responses to inundation and defarming in the Reservoir Region of the Three Gorges Project (China). Environ. Manag. 38, 618–637. Kleinman, P.J.A., Sharpley, A.N., 2003. Effect of broadcast manure on runoff phosphorus concentrations over successive rainfall events. J. Environ. Qual. 32 (3), 1072–1081. Kuhn, N.J., Armstrong, E.K., Ling, A.C., Connolly, K.L., Heckrath, G., 2012. Intertill erosion of carbon and phosphorus from conventionally and organically farmed Devon silt soils. Catena 91, 94–103. Kwong, K.F.N.K., Bholah, A., Volcy, L., Pynee, K., 2002. Nitrogen and phosphorus transport by surface runoff from a silty clay loam soil under sugarcane in the humid tropical environment of Mauritius. Agric. Ecosyst. Environ. 91 (1–3), 147–157. Liu, Y., Tao, Y., Wan, K.Y., Zhang, G.S., Liu, D.B., Xiong, G.Y., Chen, F., 2012. Runoff and nutrient losses in citrus orchards on sloping land subjected to different surface mulching practices in the Danjiangkou Reservoir area of China. Agric. Water Manag. 110, 34–40. Lu, R.K., 2000. Soil Agrochemical Analysis. China Agricultural, Soil Science Society of China, Beijing, pp. 146–195 (in Chinese). Luo, Z.X., Zhu, B., Tang, J.L., Wang, T., 2009. Phosphorus retention capacity of agricultural headwater ditch sediments under alkaline condition in purple soils area, China. Ecol. Eng. 35, 57–64. Ma, X., Li, Y., Zhang, M., Zheng, F.Z., Du, S., 2011. Assessment and analysis of non-point source nitrogen and phosphorus loads in the Three Gorges Reservoir Area of Hubei Province. China. Sci. Total. Environ. 412, 154–161. Ma, X., Li, Y., Li, B.L., Han, W.Y., Liu, D.B., Liu, X., 2015. Evaluation of nitrogen and phosphorus loads from agricultural nonpoint source in relation to water quality in Three Gorges Reservoir Area, China. Desalin. Water Treat. 1–18. Massey, H.F., Jackson, M.L., 1952. Selective erosion of soil fertility constituents. Soil Sci. Soc. Am. J. 16, 353–356. McDowell, R., Sharpley, A.N., 2002. Phosphorus transport in overlandflow in response to position of manure application. J. Environ. Qual. 31, 217–227. Ministry of Environmental Protection of the People's Republic of China (MEPPRC), 2013l. Gazette of Eco-environmental Monitoring of Three Gorges Project, Yangzi River, China 1997–2011 (in Chinese). Ng, S.L., Cai, Q.G., Ding, S.W., Chau, K.C., Qin, J., 2008. Effects of contour hedgerows on water and soil conservation, crop productivity and nutrient budget for slope farmland in the Three Gorges Region (TGR) of China. Agrofor. Syst. 74 (3), 279–291. Nicolaisen, J.E., Gilley, J.E., Eghball, B., Marx, D.B., 2007. Crop residue effects on runoff nutrient concentrations following manure application. Trans. ASABE 50 (3), 939–944. Olsen, S.R., Sommers, L.E., 1982. Phosphorus. In: Page, A.L., Miller, R.H., Keeny, D.R. (Eds.), Method of Soil Analysis. II. Chemical and Microbiological Properties, Agronomic Monograph, second ed. vol. 9. SSSA, Madison, Wisconsin. Otero, J.D., Figueroa, A., Munoz, F.A., Pena, M.R., 2011. Loss of soil and nutrients by surface runoff in two agro-ecosystems within an Andean paramo area. Ecol. Eng. 37, 2035–2043. Peterjohn, W.T., Schlesinger, W.H., 1990. Nitrogen loss from deserts in the southwestern United States. Biogeochemistry 10, 67–79. Quinton, J.N., Catt, J.A., Hess, T.M., 2001. The selective removal of phosphorus from soil: is event size important. J. Environ. Qual. 30, 538–545.

Randall, G.W., Mulla, D.J., 2001. Nitrate nitrogen in surface waters as influenced by climatic conditions and agricultural practices. J. Environ. Qual. 30, 337–344. Recanatesi, F., Ripa, M.N., Leone, A., Luigi, P., Luca, S., 2013. Land use, climate and transport of nutrients: evidence emerging from the Lake Vicocase study. J. Environ. Manag. 52 (2), 503–513. Schlesinger, W.H., Reynolds, J.F., Cunningham, G.L., Huenneke, L.F., Jarrell, W.M., Virginia, R.A., Whitford, W.G., 1990. Biological feedbacks inglobal desertification. Science 247, 1043–1048. Sharpley, A.N., Withers, P.J.A., 1994. The environmentally-sound management of agricultural phosphorus. Fert. Res. 39, 133–146. Sharpley, A.N., Smith, S.J., Naney, J.W., 1987. Environmental impact of agricultural nitrogen and phosphorus use. J. Agric. Food Chem. 35, 812–817. Shen, Z.Y., Chen, L., Hong, Q., Qiu, J., Xie, H., Liu, R.M., 2013. Assessment of nitrogen and phosphorus loads and causal factors from different land use and soil types in the Three Gorges Reservoir Area. Sci. Total Environ. 454, 383–392. Smith, D.R., Owens, P.R., Leytem, A.B., Warnemuende, E.A., 2007. Nutrient losses from manure and fertilizer applications as impacted by time to first runoff event. Environ. Pollut. 147, 131–137. Turnbull, L., Wainwright, J., Brazier, R.E., 2011. Nitrogen and phosphorus dynamics during runoff events over a transition from grassland to shrubland in the south-western United States. Hydrol. Process. 25 (1), 1–17. Udawatta, R.P., Motavalli, P.P., Garrett, H.E., 2004. Phosphorus loss and runoff characteristics in three adjacent agricultural watersheds with claypan soils. J. Environ. Qual. 33, 1709–1719. Wang, L.Q., Liang, T., Chong, Z.Y., Zhang, C.S., 2011. Effects of soil type on leaching and runoff transport of rare earth elements and phosphorous in laboratory experiments. Environ. Sci. Pollut. Res. 18, 38–45. Wang, L.Q., Liang, T., Zhang, Q., 2013. Laboratory experiments of phosphorus loss with surface runoff during simulated rainfall. [J]. Environ. Earth Sci. 70 (6), 2839–2846. Wei, G.X., Wang, Y.B., Wang, Y.L., 2008. Using 137Cs to quantify the redistribution of soil organic carbon and total N affected by intensive soil erosion in the headwaters of the Yangtze River, China. Appl. Radiat. Isotopes 66, 2007–2012. Wei, L.H., Cheng, X.Q., Cai, Y.F., 2013. Nutrient export via overland flow from a cultivated field of an Ultisol in southern China. Hydrol. Process. 27, 421–432. Xia, L.Z., Liu, G.H., Ma, L., Yang, L.Z., Li, Y.D., 2014. The effects of contour hedges and reduced tillage with ridge furrow cultivation on nitrogen and phosphorus losses from sloping arable land. J. Soils Sediments 14 (3), 462–470. Xie, X.Q., Wang, L.J., 1998. Standard Method of Observation and Analysis of Chinese Ecosystem Research Network-Observation and Analysis of Water Environment Elements. Chinese Standard Press, Beijing, pp. 104–279. Xu, X.B., Tan, Y., Yang, G.S., 2013. Environmental impact assessments of the Three Gorges Project in China: issues and interventions. Earth-Sci. Rev. 124, 115–125. Yang, Y., Ye, Z.H., Liu, B.Y., Zeng, X.Q., Fu, S.H., Lu, B.J., 2014. Nitrogen enrichment in runoff sediments as affected by soil texture in Beijing mountain area. Environ. Monit. Assess. 186, 971–978. Zhang, Y., Liu, B.Y., Zhang, Q.C., Xie, Y., 2003a. Effect of different vegetation types on soil erosion by water. Acta Bot. Sin. 1204–1209. Zhang, G.L., Yang, J.L., Zhao, Y.G., 2003b. Nutrient discharge from a typical watershed in the hilly area of subtropical chain. Pedosphere 13 (1), 23–30.