Scholarly article on topic 'Effects of collapsing gully erosion on soil qualities of farm fields in the hilly granitic region of South China'

Effects of collapsing gully erosion on soil qualities of farm fields in the hilly granitic region of South China Academic research paper on "Earth and related environmental sciences"

CC BY-NC-ND
0
0
Share paper
Keywords
{"collapsing gully erosion" / farmland / "hilly granitic region" / "soil nutrient" / "soil properties" / "South China"}

Abstract of research paper on Earth and related environmental sciences, author of scientific article — Dong XIA, Shu-wen DING, Li LONG, Yu-song DENG, Qiu-xia WANG, et al.

Abstract Collapsing gully erosion is a specific form of soil erosion types in the hilly granitic region of tropical and subtropical South China, and can result in extremely rapid water and soil loss. Knowledge of the soil physical and chemical properties of farmland influenced by collapsing gully erosion is important in understanding the development of soil quality. This study was conducted at the Wuli Watershed of the Tongcheng County, south of Hubei Province, China. The aim is to investigate soil physical and chemical properties of three soil layers (0–20, 20–40 and 40–60 cm) for two farmland types (paddy field and upland field) in three regions influenced by collapsing gully erosion. The three regions are described as follows: strongly influenced region (SIR), weakly influenced region (WIR) and non-influenced region (NIR). The results show that collapsing gully erosion significantly increased the soil gravel and sand content in paddy and upland fields, especially the surface soil in the SIR and WIR. In the 0–20 cm layer of the paddy field, the highest gravel content (250.94 g kg−1) was in the SIR and the lowest (78.67 g kg−1) was in the NIR, but in the upland filed, the surface soil (0–20 cm) of the SIR and the 40–60 cm soil layer for the NIR had the highest (177.13 g kg−1) and the lowest (59.96 g kg−1) values of gravel content, respectively. The distribution of gravel and sand decreased with depth in the three influenced regions, but silt and clay showed the inverse change. In the paddy field, the average of sand content decreased from 58.6 (in the SIR) to 49.0% (in the NIR), but the silt content was in a reverse order, increasing from 27.9 to 36.9%, and the average of the clay content of three regions showed no significant variation (P<0.05). But in the upland filed, the sand, silt and clay fluctuated in the NIR and the WIR. Soils in the paddy and upland field were highly acidic (pH<5.2) in the SIR and WIR; moreover lower nutrient contents (soil organic matter (SOM), total N and available N, P, K) existed in the SIR. In the 0–20 cm soil layer of the paddy field, compared with the NIR and the WIR, collapsing gully erosion caused a very sharp decrease in the SOM and total N of the SIR (5.23 and 0.56 g kg−1, respectively). But in the surface soil (0–20 cm) of the upland field, the highest SOM, total N, available N, available P and available K occurred in the NIR, and the lowest ones were in the SIR. Compared with the NIR, the cation exchange capacity (CEC) in the SIR and WIR was found to be relatively lower. These results suggest that collapsing gully erosion seriously affect the soil physical and chemical properties of farmland, lead to coarse particles accumulation in the field and decrease pH and nutrient levels.

Academic research paper on topic "Effects of collapsing gully erosion on soil qualities of farm fields in the hilly granitic region of South China"

ELSEVIER

RESEARCH ARTICLE

Available online at www.sciencedirect.com

ScienceDirect

Effects of collapsing gully erosion on soil qualities of farm fields in the hilly granitic region of South China

CrossMark

XIA Dong1' 2, DING Shu-wen1, 2, LONG Li3, DENG Yu-song1' 2, WANG Qiu-xia1, 2, WANG Shu-ling1' 2, CAI Chong-fa1, 2

1 College of Resources and Environment, Huazhong Agricultural University, Wuhan 430070, P.R.China

2 Key Laboratory of Arable Land Conservation (Middle and Lower Reaches of Yangtze River), Ministry of Agriculture/Huazhong Agricultural University, Wuhan 430070, P.R.China

3Sichuan Research Institute of Water Conservancy Science, Chengdu 610072, P.R.China

Abstract

Collapsing gully erosion is a specific form of soil erosion types in the hilly granitic region of tropical and subtropical South China, and can result in extremely rapid water and soil loss. Knowledge of the soil physical and chemical properties of farmland influenced by collapsing gully erosion is important in understanding the development of soil quality. This study was conducted at the Wuli Watershed of the Tongcheng County, south of Hubei Province, China. The aim is to investigate soil physical and chemical properties of three soil layers (0-20, 20-40 and 40-60 cm) for two farmland types (paddy field and upland field) in three regions influenced by collapsing gully erosion. The three regions are described as follows: strongly influenced region (SIR), weakly influenced region (WIR) and non-influenced region (NIR). The results show that collapsing gully erosion significantly increased the soil gravel and sand content in paddy and upland fields, especially the surface soil in the SIR and WIR. In the 0-20 cm layer of the paddy field, the highest gravel content (250.94 g kg-1) was in the SIR and the lowest (78.67 g kg-1) was in the NIR, but in the upland filed, the surface soil (0-20 cm) of the SIR and the 40-60 cm soil layer for the NIR had the highest (177.13 g kg-1) and the lowest (59.96 g kg-1) values of gravel content, respectively. The distribution of gravel and sand decreased with depth in the three influenced regions, but silt and clay showed the inverse change. In the paddy field, the average of sand content decreased from 58.6 (in the SIR) to 49.0% (in the NIR), but the silt content was in a reverse order, increasing from 27.9 to 36.9%, and the average of the clay content of three regions showed no significant variation (P<0.05). But in the upland filed, the sand, silt and clay fluctuated in the NIR and the WIR. Soils in the paddy and upland field were highly acidic (pH<5.2) in the SIR and WIR; moreover lower nutrient contents (soil organic matter (SOM), total N and available N, P, K) existed in the SIR. In the 0-20 cm soil layer of the paddy field, compared with the NIR and the WIR, collapsing gully erosion caused a very sharp decrease in the SOM and total N of the SIR (5.23 and 0.56 g kg-1, respectively). But in the surface soil (0-20 cm) of the upland field, the highest SOM, total N, available N, available P and available K occurred in the NIR, and the lowest ones were in the SIR. Compared with the NIR, the cation

Received 26 October, 2015 Accepted 28 March, 2016 XIA Dong, E-mail: xiadongsanxia@163.com; Correspondence CAI Chong-fa, Tel/Fax: +86-27-87288249,

E-mail: cfcai@mail.hzau.edu.cn

© 2016, CAAS. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http:// creativecommons.org/licenses/by-nc-nd/4.0/) doi: 10.1016/S2095-3119(16)61348-5

exchange capacity (CEC) in the SIR and WIR was found to be relatively lower. These results suggest that collapsing gully erosion seriously affect the soil physical and chemical properties of farmland, lead to coarse particles accumulation in the field and decrease pH and nutrient levels.

Keywords: collapsing gully erosion, farmland, hilly granitic region, soil nutrient, soil properties, South China

1. Introduction

Soil erosion represents one of the most important, but poorly quantified common global environmental problems (Xia et al. 2009; Park et al. 2014). In the hilly granitic region of tropical and subtropical South China, there is a serious soil erosion type called collapsing gully erosion. Collapsing gully defined as an erosional landform with a very high rate of sediment transfer, is formed in the hillslopes with the cover of thick granite weathering crust and caused by the joint operation of mass wasting and flowing water erosion with the former playing the dominant role (Xu and Zeng 1992; Xu 1996). The granite weathering crust, which has undergone intense chemical weathering, with thicknesses in the range of 20-60 m is widely distributed (Lan et al. 2003). Several factors such as geological structure, soil properties, vegetation cover conditions and slope geometry strongly influence the rate of collapsing gully erosion (Scott Munro and Huang 1997; Woo et al. 1997). In the hilly granitic region, the impact of collapsing gully erosion is devastating. From 1950 to 2005, collapsing gully erosion affected 1 220 km2 in this region (Zhong et al. 2013), and monitoring data show that the numbers of the collapsing gullies are more than 239 000 and widely distributed in 7 regions of South China, including Guangdong, Jiangxi, Hubei, Hunan, Fuji-an, Anhui, and Guangxi (Deng et al. 2015). The collapsing gullies are similar to gullies in China's Loess Plateau and the badland landscape occurring in other humid tropical and subtropical areas of the world, but the collapsing gully has its own essential feature (Xu 1996). The deposition of eroded materials from the collapsing gullies has buried dozens of hectares of farmland and silted up the streams. Coarse and sandy materials smear over the fields, leading to sandification of land, marked decline in land productivity and sometimes farmland abandonment (Lam et al. 1997; Luk et al. 1997a; Sheng and Liao 1997).

Most of the relevant researches during the past decades focused on the influence factors of collapsing gully erosion (Xu and Zeng 1992; Xu 1996; Luk et al. 1997a) and the variations in soil physical properties in different soil profiles of the collapsing gully wall (Wu and Wang 2000; Xia et al. 2015) and the effects of vegetation cover (Woo and Luk 1990; Woo et al. 1997b; Zhang et al. 2004) and human activities (Kimoto

et al. 2002) on soil and water loss and sediment discharge in the hilly granitic regions. Woo and Luk (1990) reported that most slopes consist of an unknown combination of loose materials and weathered granite and the resistance to sediment entrainment cannot be easily determined. The results also indicated that the potential sediment yield from the collapsing gullies increases as vegetation cover decreases. Monitoring the water and sediment yield from collapsing gully is also the main aspect of the researches (diCenzo and Luk 1997; Luk et al. 1997b; Uchida et al. 2000). By using composite fingerprinting technique, the research shown that in the alluvial fan, 10% of the alluvial fan's sediment originated from the surface layer soils in the active collapsing gully, and the soil surface layer is more easily washed away in an active collapsing gully than in a stable collapsing gully (Lin et al. 2015). Some literatures assessed the nutrient status and nutrient fluxes of the soils in the upland and the lowland areas of collapsing gully (Sioh et al. 1990; Lam et al. 1997). Deng et al. (2014) found that collapsing gully erosion causes serious desertification and structural deterioration of the soil in the alluvial fan farmland. Based on the real-time kinematic global position system (RTK GPS) positioning technology, Zhang et al. (2015) studied the basic physical and chemical properties and nutrient distribution of the collapsing alluvial fan. The results found that the content of alkali hydrolysable nitrogen and available potassium were risen regularly with the increase of distance from the fan starting point and available phosphorus was increased enormously. Deng et al. (2015) evaluated the effects of different land uses on the soil physic-chemical properties and erodibility of collapsing gully alluvial fan, and reported that the soil physic-chemical properties were increased in different degrees under the different land uses. However, the effects of collapsing gully erosion on soil properties of farmland are poorly documented in the hilly granitic region, South China. The objectives of this study include (a) to describe and assess the soil properties status of the paddy and upland field in different regions influenced by collapsing gully erosion and (b) to determine the differences and state of the soil nutrient between different soil layers of the field in the different influenced regions. Results from the present study may broaden our knowledge of collapsing gully erosion effects on field soils and provide solutions to the amelioration and utilization of farmlands.

2. Materials and methods

2.1. Study area description

The study area was located in Wuli Watershed of the Tong-cheng County (29°2-29°24'N, 113°36-114°4'E) in southeast of Hubei Province, China (Fig. 1). Climate in the study area is subtropical with a cold and dry winter and a relatively hot and moist summer. The mean annual precipitation is around 1 521 mm, with more than 75% of rainfall occurring during March to September. The average frost-free period is 260 days. The annual mean temperature is 17°C, with the maximum monthly mean temperature in July and the minimum in January. The granites in the area were formed in the Mesozoic Yanshanian, many granitic joints formed by the Yanshan Movement and inherited from the bedrock facilitate slope failure. Weathering often produces a thick weathered granitic crust sometimes exceeding 30 m in the area. According to our survey in 2005, it showed that there were 1 102 collapsing gullies in Tongcheng County, and 86.9% of these were categorized as active type. The active collapsing gully is the gully in the stage with rapidly increasing erosion. The mass-wasted materials are rapidly removed by runoff, and the slope stability is not very high and the erosion rate reaches a maximum. The stable collapsing

gully is the gully in the stage with gradually declining erosion. There is a much lower rate of sediment transfer than that in the active collapsing gully, and the frequency of occurrence of the gully wall's slumps, slides and falls becomes very low. In most cases, the collapsing gully tends to transform from active to stable.

2.2. Soil sampling

Two farmland types (paddy field and upland field) were selected within the study area according to three regions influenced by collapsing gully erosion (Fig. 2). The three regions are described as follows: (a) strongly influenced region (SIR). In this region, more than 65% of collapsing gullies are categorized as active type, and farmland directly affected by the sediments from collapsing gullies; (b) weakly influenced region (WIR). In this region, more than 70% of collapsing gullies are categorized as stable type, several control measures were practiced to reduce sediment production and to prevent sediments from reaching farmland; (c) non-influenced region (NIR). There are no collapsing gullies surrounding the farmland, but it has similar soil-forming conditions with other regions, and collapsing gullies occurred in the region in history. The paddy field areas in the SIR, WIR and NIR encompass approximately 1.1, 1.5

Fig. 1 Location of study area.

Fig. 2 A typical curved collapsing gully in the hilly granitic region, Tongcheng County, Hubei Province, China, with thick weathered granitic crust, steep collapsing gully wall and sediments were carried by running water and smeared over the field (photo provided by Ding Shuwen, Associate Professor, Huazhong Agricultural University, China).

and 2.1 km2, respectively. The upland field areas in the SIR, WIR and NIR encompass approximately 0.8, 0.8 and 1.3 km2, respectively. In our survey, in the SIR and WIR, the paddy fields existed before the collapsing gullies occur, and the upland fields were cultivated after the collapsing gullies occur. In the study area, rice (Oryza sativa) and rape (Bras-sica campestris) rotation was adopted in the paddy field in the WIR and NIR and the paddy field in the WIR had been abandoned in 2011. Peanuts (Arachis hypogaea), cotton (Gossypium hirsutum) and soybean (Glycine max) were planted in the upland field in the three regions (Table 1).

For each type of farmland in the different regions, three plots were selected as sampling units. Six soil sampling sites were located within each plot for replications, along the length of an S-shaped soil sampling line. At each sampling site, soil samples were collected in May 2012 at three soil layers: 0-20, 20-40 and 40-60 cm (Fig. 3 and Table 2). The soil samples at the same soil layer were completely mixed into one composite fresh soil sample. Three soil samples at each soil layer (0-20, 20-40 and 40-60 cm) were obtained by using cutting ring to determine soil bulk density. Roots, stones and debris were removed from the soil, and 1 kg soil sample was obtained by means of quartering and transported to the laboratory. The soil was air-dried and then sieved on a 2-mm mesh for lab analyses.

2.3. Soil analysis

The soil bulk density was determined using the cutting ring method (Zhang and Gong 2012). The gravel content (>2 mm) was tested by sieving method (Zhang and Gong 2012).

ai o c ai

Tl ^ c

</) +J

o v> ai

IE ™ O

<u > «

'55 ® 5

CD O ^ T3

QD c ^ ra <u

q. ra ra

Si a §

<D O CP

ra g "to 5

<D 3 tB «

£ IS ■c .a <0 S.

Ï3 fr

E is SS m ¡=

■5 CD CD "

co m co r-"v ^ V-

^T CN ^T 00

? != o ra

è? . u

<5 CD <u ^ CL

Particle size distribution (clay, silt, and sand) was measured by the hydrometer method after dispersion with sodium hex-ametaphosphate (Kroetsch and Wang 2008). Soil pH was measured in a soil-water suspension (1:2.5 of soil:water) using an automatic acid-base titrator. Soil organic matter (SOM), soil total nitrogen, available nitrogen, available phosphorus and available potassium and cation exchange capacity (CEC) were tested according to the procedures described by the Institute of Soil Science of Chinese Academy of Sciences (ISSCAS 1978).

2.4. Statistical analysis

A one-way analysis of variance (ANOVA) was carried out with all properties to assess the differences among soil layers. The least square difference (LSD) test (at P<0.05) was used to compare means of soil variables when the results of ANOVA were significant at P<0.05. Pearson correlation analyses were performed to correlate the soil properties across soil depth in different regions under two farmland types. All statistical analyses were conducted using the SPSS 15.0 statistical software package, and all of the results are reported as the means±SD (standard deviation).

3. Results

3.1. Changes in soil physical properties

Soil bulk density Collapsing gully erosion had varying effects (P<0.05) on the mean values of the soil bulk density both in the paddy field and upland field in the different influence regions (Tables 3 and 4). Compared with the NIR, the average soil bulk density of the paddy fields in the SIR and WIR increased by 9.23 and 5.38%, respectively (Table 3) and the upland field average of soil bulk density increased by 1.36 and 0.68%, respectively (Table 4).

Additionally, collapsing gully erosion also had different effects on the soil bulk density of the different soil layers (Tables 3 and 4). In the paddy field, the 0-20 cm soil layer for the NIR and the SIR had the lowest (1.19 g cm-3) and the highest (1.61 g cm-3) bulk density values, respectively (Table 3). Significant differences (P<0.05) in the soil bulk density of the different soil layers were found both in the SIR and in the 0-20 and 20-40 cm soil layers in the NIR, whereas no significant changes were observed in the different soil layers in the WIR (Table 3). In the upland field, soil bulk density showed the highest value (1.58 g cm-3) and the lowest value (1.34 g cm-3) in the 0-20 and the 20-40 cm soil layers of the SIR, respectively (Table 4). Soil bulk density for all soil layers differed not significantly (P<0.01), with the exception of the 20-40 cm soil layer for the SIR (Table 4). Gravel content In the paddy field, significant differences

Fig. 3 Soil layers of paddy field (A, B and C) and upland filed (D, E and F) in the three regions. A and D, strongly influenced region (SIR). B and E, weakly influenced region (WIR). C and F, non-influenced region (NIR), showing lots of sediments smeared over the field in the SIR (photo provided by Long Li, Assistant Engineer, Sichuan Research Institute of Water Conservancy Science, China).

(P<0.05) for averages of gravel contents were observed among the different regions (Table 3). With the descent of the influence degree by collapsing gully erosion (from the SIR to the NIR), the mean gravel contents decreased sharply (P<0.01). In the 0-20 cm layer, the highest gravel content (250.94 g kg-1) was in the SIR and the lowest (78.67 g kg-1) was in the NIR (Table 3). The distribution of gravel content in soil profile decreased with depth in the SIR. However, the gravel content in the 40-60 cm layer for the WIR was higher than the value in 20-40 cm soil layer, but in the WIR, the results showed the inversed order (Table 3).

Collapsing gully erosion also had a notable effect on the average of gravel content among the regions and/or the soil layers in the upland field (Table 4). The changing trends of gravel content were similar with that in the paddy field. The surface soil (0-20 cm) of the SIR and the 40-60 cm soil layer for the NIR had the highest (177.13 g kg-1) and the lowest (59.96 g kg-1) values of gravel content, respectively (Table 4). The differences of all soil layers in different regions were remarkable (Table 4), except the 20-40 and 40-60 cm soil layers in the SIR.

Soil texture In the paddy field, collapsing gully erosion had varying effects on the soil texture of different regions and the soil depths. According to the data analysis, the

Table 2 Field survey description of the different soil layers in different influenced regions

Farmland type Region Soil layer (cm) Field survey description of the different soil layers1'

Paddy SIR 0- -20 Color: 7.5YR 7/8; sandy soil; granular structure and loose; soil with high content of mica and quartz; no roots; medium porosity, mainly intergranular porosity; no adhesion and no plasticity

20- 40 Color: 10BG 4/1; loam; sheet structure and compaction; soil with low content of mica and quartz; small amount of rice roots; low porosity, mainly root pores; adhesion and medium plasticity

40- -60 Color: 10YR 5/1; loam; sheet structure and compaction; soil with low content of mica and quartz; no roots; low porosity, mainly intergranular porosity; slight adhesion and no plasticity

WIR 0- 20 Color: 2.5Y 4/4; loam; sheet structure and loose; soil with medium content of mica and quartz; medium amount of rice roots; medium porosity, mainly root pores and intergranular porosity; adhesion and slight plasticity

20- 40 Color: 10BG 4/1; loam; sheet structure and compaction; soil with medium content of mica and quartz; no roots; low porosity, mainly intergranular porosity; slight adhesion and slight plasticity

40- 60 Color: 10BG 4/1; loam; sheet structure and compaction; soil with medium content of mica and quartz; no roots; low porosity, mainly intergranular porosity; high content of iron manganese markings; slight adhesion and slight plasticity

NIR 0- 20 Color: 5YR 4/2; loam; granular structure and compaction; soil with low content of mica and quartz; small amount of rice roots; low porosity, mainly root pores; slight adhesion and medium plasticity

20- 40 Color: 10YR 5/1; loam; granular structure and compaction; soil with low content of mica and quartz; small amount of rice roots; low porosity, mainly intergranular porosity and root pores; low content of iron manganese markings; slight adhesion and medium plasticity

40- 60 Color: 10YR 5/1; loam; granular structure and compaction; soil with low content of mica and quartz; no roots; low porosity, mainly intergranular porosity; high content of iron manganese markings; adhesion and slight plasticity

Upland SIR 0- 20 Color: 7.5YR 5/6; sandy loam; granular structure and extremely loose; soil with high content of mica and quartz; small amount of roots; medium porosity, mainly intergranular porosity; no adhesion and no plasticity

20- 40 Color: 2.5Y 6/3; loam sandy, clastic structure and loose; soil with high content of non weathered granite debris; no roots; medium porosity, mainly intergranular porosity; no adhesion and no plasticity

40- 60 Color: 2.5Y 6/3; sandy soil, clastic structure and loose; soil with high content of non weathered granite debris; no roots; medium porosity, mainly intergranular porosity; no adhesion and no plasticity

WIR 0- 20 Color: 5YR 5/8; loam sandy; granular structure and loose; soil with medium content of mica and quartz; small amount of roots; medium porosity, mainly intergranular porosity and root pores; no adhesion and no plasticity

20- 40 Color: 5YR 5/8; sandy loam; granular structure and extremely loose; soil with medium content of mica and quartz; no roots; medium porosity, mainly intergranular porosity; no adhesion and no plasticity

40- 60 Color: 2.5YR 4/8; sandy loam; granular structure and extremely loose; soil with medium content of mica and quartz; no roots; medium porosity, mainly intergranular porosity; slight adhesion and no plasticity

NIR 0- 20 Color: 2.5YR 4/4; loam; sheet structure and loose; soil with low content of mica and quartz; no roots; medium porosity, mainly intergranular porosity; slight adhesion and no plasticity

20- 40 Color: 5YR 5/8; sandy clay; sheet structure and loose; soil with low content of mica and quartz; no roots; low porosity, mainly intergranular porosity; slight adhesion and slight plasticity

40- 60 Color: 7.5YR 5/6; loam; sheet structure and compaction; soil with low content of mica and quartz; no roots; low porosity, mainly intergranular porosity; slight adhesion and slight plasticity

11 The colors are determinated basically on the Munsell Color Chart and the description in the Soil Survey Laboratory Methods (Zhang and Gong 2012).

average of sand content decreased from 58.6 (in the SIR) to 49.0% (in the NTR), but the silt content was in a reverse order, increasing from 27.9 to 36.9% (Table 3). However, a slight increase of the clay content was surveyed with the descent of the influence degree by collapsing gully erosion. The average of the clay content of three regions showed no significant variation (P<0.05) (Table 3). From top soil layer to deep soil layer, the sand content decreased noticeably but the silt content increased dramatically (P<0.01) in all

regions (Table 3). The data demonstrated that there was more clay accumulation in the deep layers, indicating the 40-60 cm layer of the SIR and the 20-40 cm layer of the WIR and NIR, respectively (Table 3). The surface soil (0-20 cm) of the SIR had the greatest amount of sand (67.3%) and the lowest amount of clay (10.8%). The highest amount of silt (33.8%) and the clay (13.9%) in the surface soil (0-20 cm) were in the NIR and the WIR, respectively (Table 3).

Table 3 Soil physical properties among the different layers and the mean of 0-60 cm layer for paddy field in the three influenced regions

Region Soil layer (cm) Bulk density (g cm 3) Gravel (g kg 1) Sand (%) Silt (%) Clay (%)

SIR 0-20 1.61±0.14 a 250.9±16.4 a 67.3±4.1 a 21.9±1.6 c 10.8±0.7 c

20-40 1.41±0.16 b 158.2±9.7 b 60.1±3.8 b 26.2±2.2 b 13.7±0.8 b

40-60 1.24±0.16 c 96.9±2.4 c 48.4±2.4 c 35.8±3.5 a 15.8±0.2 a

Mean 1.42 A 168.7 A 58.6 A 27.9 B 13.5 A

WIR 0-20 1.37±0.06 a 112.6±19.4 a 57.9±5.2 a 28.2±1.1 c 13.9±0.6 b

20-40 1.36±0.04 a 99.9±2.4 b 54.0±4.2 ab 30.9±1.1 b 15.1±0.5 a

40-60 1.37±0.02 a 106.9±17.2 ab 51.4±2.5 b 36.7±2.7 a 11.9±0.8 c

Mean 1.37 B 106.5 B 54.4 B 31.9 B 13.7 A

NIR 0-20 1.19±0.06 b 78.7±14.2 ab 52.8±2.0 a 33.8±1.6 b 13.4±0.6 b

20-40 1.37±0.03 a 82.1±4.1 a 48.5±1.2 b 35.6±0.9 b 15.9±0.5 a

40-60 1.35±0.04 a 57.7±12.4 b 45.8±2.5 c 41.3±1.9 a 12.9±0.5 b

Mean 1.30 C 72.8 C 49.0 C 36.9 A 14.1 A

Data are means±SD. Different lowercase letters indicate significant difference at the P<0.05 level among the different soil layers in the same region; different capital letters indicate significant difference at the P<0.05 level between the mean values for the different region. The same as below.

Table 4 Soil physical properties among the different layers and the mean of 0-60 cm layer for upland field in the three influenced regions

Region Soil layer (cm) Bulk density (g cm 3) Gravel (g kg 1) Sand (%) Silt (%) Clay (%)

SIR 0-20 1.58±0.07 a 177.1±15.1 a 63.8±2.9 b 20.0±0.7 b 16.3±0.6 b

20-40 1.34±0.03 b 153.3±29.9 a 72.3±5.2 a 18.2±0.5 b 9.5±0.7 c

40-60 1.55±0.05 a 156.5±14.7 a 53.0±4.2 c 29.1±1.0 a 18.0±0.9 a

Mean 1.49 A 162.3 A 63.0 A 22.4 C 14.6 B

WIR 0-20 1.47±0.02 ab 134.0±15.3 a 62.3±3.7 a 29.3±3.6 a 8.4±0.5 c

20-40 1.43±0.03 b 116.2±6.3 b 53.6±2.8 b 30.8±4.8 a 15.7±0.4 a

40-60 1.53±0.08 a 93.5±13.1 c 58.4 ±3.7 a 27.1±3.9 a 14.5±0.5 b

Mean 1.48 A 114.6 B 58.1 A 29.1 B 12.8 B

NIR 0-20 1.43±0.02 b 85.2±12.7 b 46.4±1.0 a 40.1±0.6 b 13.5±0.6 c

20-40 1.47±0.04 ab 104.6±6.7 a 48.7±2.1 a 27.5±1.8 c 23.8±0.5 a

40-60 1.49±0.05 a 60.0±10.3 c 28.4±0.6 b 49.7±0.4 a 21.9±0.2 b

Mean 1.47 A 83.3 C 41.1 B 39.1 A 19.8 A

The data in the Table 3 showed that the amount of sand, silt and clay in most soil layers differed markedly in the upland field of the SIR, the NTR and the WIR. In the SIR, sand increased from 0-20 to 20-40 cm soil layer, and then decreased in the 40-60 cm layer, but the changes of silt and clay were contrary to those of sand (Table 4). The sand, silt and clay fluctuated in the NIR and the WIR. Sand content of these regions ranged from 28.4 to 72.3% (the 40-60 cm layer of the NIR and the 20-40 cm layer of the SIR, respectively). 20-40 cm layer of the SIR had the lowest silt content (18.2%), but the highest (40.1%) was in the 0-20 cm layer of the NIR (Table 4). However, the biggest clay content (23.8%) existed in the 20-40 cm soil layer of the NIR, while the lowest value of clay (9.5%) occurred in the 20-40 cm soil layer of the SIR (Table 4).

3.2. Changes in soil chemical properties

Soil chemical properties in paddy field According to the

results, it was clear that the soil in the NIR had a higher pH value compared with that in the SIR and the WIR; meanwhile, there was a significant difference (P<0.05) in the average of three regions (pH ranged from 4.78 to 6.29) (Fig. 4-A). In the 0-20 cm soil layer, compared with the NIR and the WIR, collapsing gully erosion caused a very sharp decrease in the SOM and total N of the SIR (5.23 and 0.56 g kg-1, respectively) (Fig. 5-A and B). In the WIR and the NIR, SOM decreased notably with the soil depths (ranged from 21.48 to 6.92 g kg-1 and 27.13 to 4.84 g kg-1, respectively), and a similar trend was found for total N, which ranged from 1.97 to 0.62 g kg-1 and 2.42 to 0.58 g kg-1, respectively (Fig. 5-A and B). In the WIR and the NIR, both the highest available N (95.10 and 154.05 mg kg-1) and the available P (8.28 and 10.77 mg kg-1) were presented in the 0-20 cm soil layer, and declined significantly (P<0.05) with soil depths. But in the SIR, the lowest available N and the available P were found in the 0-20 cm soil layer and the 20-40 cm soil layer (21.56 and 5.03 mg kg-1, respec-

tively) (Fig. 5-C and D). With the soil layer being deeper, the available K decreased markedly except the 40-60 cm layers of the SIR and the NIR (varying from 68.09 to 13.04 mg kg-1) (Fig. 5-E). On average, soil chemical properties in the NIR were significantly higher (P<0.05) than those in the WIR, especially those in the SIR (Fig. 5). The values of CEC at different soil layers and the means for these regions differed significantly (P<0.01). The lowest (2.19 cmol kg-1) and the highest CEC (9.97 cmol kg-1) were observed in the 0-20 cm soil layer of the SIR and the NIR, respectively. The CEC increased with the soil depths in the SIR, but the results showed the inverse order in the WIR and the NIR (Fig. 5-F). Soil chemical properties in the upland field From Fig. 4, the soil pH of different layers in the same region didn't differ significantly (P<0.05). In general, the pH average value in the NIR was significantly higher (P<0.05) than that in the SIR and the WIR (Fig. 4-B). It was found that collapsing gully erosion also had varying influence on the soil chemical properties of the soil layers and the regions in the upland field (Fig. 6). In the surface soil (0-20 cm), the highest SOM, total N, available N, available P and available K occurred in the NIR, and the lowest ones were in the SIR, except for that the lowest available K was in the WIR (Fig. 6). With the soil layer being deeper in the SIR, the SOM decreased significantly (P<0.01) firstly and then increased (Fig. 6-A). The similar pattern was found for the total N and available N (Fig. 6-B and C). However, the available P increased slightly (from 2.95 to 3.57 mg kg-1), the available K showed just on the contrary (from 68.56 to 32.03 mg kg-1) (Fig. 6-D and E). In the WIR, the SOM, total N and available N declined markedly with the soil depth. Meanwhile, the changing tendency of the available P was in contrast to the available

K (Fig. 6). In the NIR, the SOM, total N and available P declined sharply (P<0.01) from the 0-20 cm soil layer to 20-40 cm soil layer, and then increased slightly. But the available N and the available K showed the falling tendency from the top soil layer to deep soil layer (Fig. 6). According to the average, the highest SOM, total N, available N, available P and available K appeared in the NIR, followed by the WIR and the SIR (Fig. 6). No significant difference (P<0.05) was observed between the averages of CEC of the WIR and the NIR; nevertheless, the CEC of the WIR and the NIR was markedly higher (P<0.01) than that of the SIR (CEC varying from 3.62 to 10.40 cmol kg-1) (Fig. 6-F).

4. Discussion

4.1. Soil physical properties changes

This study demonstrates that collapsing gully erosion has significantly affected soil physical properties of paddy and upland field in the WIR and the SIR. The rising influence degree by collapsing gully erosion has significantly increased soil gravel and sand content and significantly decreased soil silt and clay content (Tables 3 and 4). These results may be caused by the following reasons: collapsing gully erosion has a very high rate of sediment transfer, and the sediment material will be transported by flowing water in a rather short time. The near-by farm fields were buried by eroded material from the collapsing gully, leading to sandification of land, marked decline in land productivity and farmland abandonment (Xu 1996; diCenzo and Luk 1997). The sediments bring about the coarse sediments of the farmland and finer particles are carried away to the far distance by

□ 0-20 cm □ 20-40 cm □ 40-60 cm -♦- Mean of 0-60 cm layer

Fig. 4 Soil pH among the different layers and the mean of 0-60 cm layer for the paddy field (A) and upland field (B) in the three influenced regions. Different lowercase letters indicate significant difference at the P<0.05 level among the different soil layers in the same region; different capital letters indicate significant difference at the P<0.05 level between the mean values for the different region. The same as below.

□ 0-20 cm

A 30 24

180 150 120 90

E 80 r

1 60 O)

S 20 A

] 20-40 cm □ 40-60 cm B 2.8

-♦— Mean of 0-60 cm layer

Fig. 5 Soil organic matter (SOM; A), total N (B), available N (C), available P (D), available K (E) and cation exchange capacity (CEC; F) among the different layers and the mean of 0-60 cm layer for paddy field in the three influenced regions.

running water (Lam et al. 1997). Additionally, splash erosion and raindrops on the soil surface can create fine soil particle to be detached and carried away and increase soil coarseness (Roth and Eggert 1994; Ahn et al. 2013; Wang et al. 2013). Compared with the NIR, the averages of soil bulk density were higher in the SIR and WIR; meanwhile collapsing gully erosion also had different effects on the soil bulk density of the different soil layer, the surface soil for the SIR had higher bulk density, and the NIR had lower values (Tables 3 and 4). The weathered granitic crust is the primary source of eroded materials, it mainly consists of coarse sand

(mainly silica) and clay (Sioh et al. 1990; Lan et al. 2003). While the finer soil particles are washed out, the remaining coarser sand smear over the fields. In the SIR and WIR, the soil bulk density appears to be higher, especially in the surface soil layer, because the coarser soil particles can increase soil bulk density. That accords with our field survey, which shows alternate layers of coarse and fine deposits in the stratigraphic profile of the fields influenced by collapsing gully erosion (Figs. 2 and 3). Meanwhile, the relatively small amount of organic matter exists, resulting in a decline in soil structural properties and thus increasing bulk density

□ 0-20 cm □ 20-40 cm □ 40-60 cm -♦- Mean of 0-60 cm layer

B 1.2r

C 56 48

D 28 24 1 20 I 16 Î 12

en 120 £

Fig. 6 SOM (A), total N (B), available N (C), available P (D), available K (E) and CEC (F) among the different layers and the mean of the 0-60 cm layer for the upland field in the three influenced regions.

(Hajabbasi et al. 1997). In addition, tillage practices (Lam-purlanes and Cantero-Martinez 2003), cropping patterns and other forms of agroforestry (Van Noordwijk et al. 1991; Reintam et al. 2008) can also affect soil compaction and bulk density. The 0-30 cm depth was where bulk density was mostly affected by soil management (Benites et al. 2007). In the study area, rice (O. sativa) and rape (B. campestris) were cultivated in the paddy field, peanuts (A. hypogaea), cotton (G. hirsutum) and soybean (G. max) were planted

in the upland in the different regions. Due to the numerous factors, soil physical properties in different regions and different soil layers take on different forms of change.

4.2. Soil chemical properties analysis

Soils of the paddy and upland field in our study area are characterized by highly acidic (Fig. 4), with the mean pH ranging from 4.85 in the SIR to 5.72 in the NIR for the

paddy field and from 4.74 in the SIR to 5.57 in the NIR for the upland field, respectively. These results agree with the findings of Lam et al. (1997) who demonstrated that soils of the fields covered by sediment transported out of collapsing gully were highly acidic. Soils derived from granite are rich in potassium, magnesium, iron and aluminum. The abundance of ferric and aluminum oxides, coupled with a paucity of base cations that has been replaced by hydro-nium ions, has caused these soils to be highly acidic (Sioh et al. 1990). Such low pH is likely to affect the availability of many nutrients and is considered to be unfavorable for crop cultivation. For instance, highly acidic inhibited nitrogen fixing capacity of legumes.

SOM is chosen as the most important indicator of soil quality and agro-ecosystem sustainability because of its impact on other physical, chemical and biological indicators of soil quality (Reeves 1997; Li et al. 2015). Compared with the mean of soil organic matter in both the paddy and upland fields of the NIR and the WIR, the mean of soil organic matter shows lower levels in the SIR (Figs. 5-A and 6-A). One explanation is that loss of soil fine particles could result in loss of soil organic matter because organic matter is often combined with fine soil particles (Lopez 1998; Zhao et al.

2006). Some studies also show that erosion is a selective process which preferentially removes the finer and nutrient-enriched soil particles (Lam et al. 1997). In the SIR, organic fertilizers are needed to improve and ameliorate the soil structure and conditions.

It is well known that N is essential for plant growth and P plays an important role in photosynthesis and plant-protein production (Ma et al. 2015). The soils in the paddy and upland fields of the SIR, especially in the top soil layer processes lower total N, available N and available P (Figs. 5 and 6), which result from mineral nitrogen leaching losses and the lower amount and losses of soil organic matte (Mo et al. 2003). Runoff and erosion have long been recognized as the main phosphorus loss pathways (Pizzeghello et al. 2014), and mobility of phosphorus is strongly affected by its reactions with soil constituents which can alter the concentration of phosphorus in soils and sediments (Devau et al. 2011; Pizzeghello et al. 2011). Many attempts have been made to describe the sequence of reactions that leads to phosphate retention in acid soils (Borling et al. 2004; Allen and Mallarino 2006). In acid soils, phosphorus is fixed into slightly soluble forms by precipitation and sorption reactions with Fe and Al compounds to form highly insoluble P-containing minerals such as strengite and veriscite (Chairat et al. 2007; Spiteri et al.

2007) because pH governs the absorption properties of major P-fixing minerals, i.e., 1:1 and 2:1 clay minerals, aluminum and iron (hydro) oxides (Ioannou and Dimirkou 1997; Hiemstra and Van Riemsdijk 1999; Gustafsson

2001). Borling et al. (2001) also reported that P sorption was highly correlated with the clay content and amorphous Fe and Al oxides content. The result in the present study illustrates that N and P are lost in the sediments-associated forms as the nutrient-rich finer particles are washed and the nutrient-poor coarse materials are left in the fields. In terms of potassium, although the soil parent material granite is rich in potassium, the available K is not very high in the paddy and upland field. That is also expected because of runoff and erosion. Nutrient levels in the NIR and WIR soils are not much higher than those found in the SIR. These may be affected by the time of tillage practices, soil management and fertilizer input. Soil CEC could be created by either of two mechanisms: (a) by a higher charge density per unit surface area which means a higher degree of oxidation of SOM; or (b) by a higher surface area for cation adsorption sites, or a combined effect of both (Liang et al. 2006). CEC is also affected by both pH and the ionic strength of the soil solution, especially in highly weathered soils and other soils rich in Al and Fe oxides, hydroxides and amorphous clays (Robertson et al. 1999). So the lower amount of CEC in the SIR and WIR might be due to the lower pH, SOM and fine soil particles.

5. Conclusion

Collapsing gully erosion has potentially affected soil physical and chemical properties of farmland in the granitic hilly region, South China. To conclude, our research reveals that, both in the paddy and upland field soils of the SIR, the gravel and sand mean content proves the highest, but in the NIR, those are found to be the lowest. With the descent of the influence degree by collapsing gully erosion (from the SIR to the NIR), the coarse particles in the different layers decreased. The data demonstrates that coarse particles decreased with depth, and there are more fine particles accumulation in the lower depths, in the SIR and WIR. The soil bulk density shows irregular changes in the paddy and upland field. Soils in the paddy and upland field in the granitic areas of our study region are characterized by high acid; moreover lower SOM, total N, available N, available P and available K stay in the SIR, especially in the top soil layer. However, nutrient contents in the NIR and WIR soils refuse to be higher than those found in the SIR. This study shows that collapsing gully erosion has serious influence on the soil physical and chemical properties of farmland in the granitic hilly region, but more soil property indicators and long-term researches are needed to proceed further in this study.

Acknowledgements

This research was financially supported by the National

Natural Science Foundation of China (41630858).

References

Ahn S, Doerr S H, Douglas P, Bryant R, Hamlett C A E, McHale G, Newton M I, Shirtcliffe N J. 2013. Effects of hydrophobicity on splash erosion of model soil particles by a single water drop impact. Earth Surface Processes and Landforms, 38, 1225-1233.

Allen B L, Mallarino A P. 2006. Relationships between extractable soil phosphorus and phosphorus saturation after long term fertilizer or manure application. Soil Science Society of America Journal, 70, 454-463.

Benites V M, Machado P L, Fidalgo E C, Coelho M R, Madari B E. 2007. Pedotransfer functions for estimating soil bulk density from existing soil survey reports in Brazil. Geoderma, 139, 90-97.

Borling K, Barberis E, Otabbong E. 2004. Impact of long-term inorganic phosphorus fertilization on accumulation, sorption and release of phosphorus in Ave Swedish soil profiles. Nutrient Cycling in Agroecosystems, 69, 11-21.

Borling K, Otabbong E, Barberis E. 2001. Phosphorus sorption in relation to soil properties in some cultivated Swedish soils. Nutrient Cycling in Agroecosystems, 59, 39-46.

Chaírat C, Schott J, Oelkers E H, Lartigue J E, Harouiya N. 2007. Kinetics and mechanism of natural fluorapatite dissolution at 25°C and pH from 3 to 12. Geochimica et Cosmochimica Acta, 71, 5901-5912.

Deng Y S, Ding S W, Cai C F, Lü G A, Xia D, Zhu Y. 2014. Spatial distribution of the collapsing alluvial soil physical properties in Southeastern Hubei. Scientia Agricultura Sinica, 47, 4850-4857. (in Chinese)

Deng Y S, Xia D, Cai C F, Ding S W. 2016. Effects of land uses on soil physic-chemical properties and erodibility in collapsing-gully alluvial fan of Anxi County, China. Journal of Integrative Agriculture, 8, 1863-1873.

Devau N, Hinsinger P, Le Cadre E, Colomb B, Gérard F. 2011. Fertilization and pH effects on processes and mechanisms controlling dissolved inorganic phosphorus in soils. Geochimica et Cosmochimica Acta, 75, 2980-2996.

diCenzo P D, Luk S H. 1997. Gully erosion and sediment transport in a small subtropical catchment, South China. Catena, 29, 161-176.

Gustafsson J P. 2001. Modeling competitive anion adsorption on oxide minerals and an allophone-containing soil. European Journal of Soil Science, 52, 639-653.

Hajabbasi M A, Jalalian A, Karimzadeh H R. 1997. Deforestation effects on soil physical and chemical properties, Lordegan, Iran. Plant and Soil, 190, 301-308.

Hiemstra T, Van Riemsdijk W H. 1999. Surface structural ion adsorption modeling of competitive binding of oxyanions by metal (hydr) oxides. Journal of Colloid and Interface Science, 210,182-193.

loannou A, Dimirkou A. 1997. Phosphate adsorption an hematite, kaolinite, and kaolinite-hematite (k-h) systems as described by a constant capacitance model. Journal of

Colloid and Interface Science, 192, 119-128.

ISSCAS (Institute of Soil Science, Chinese Academy of Science). 1978. Physical and Chemical Analysis Methods of Soils. Shanghai Science and Technology Press, Shanghai, China. pp. 136-153, 169-176. (in Chinese)

Kimoto A, Uchida T, Mizuyama T, Li C. 2002. Influences of human activities on sediment discharge from devastated weathered granite hills of southern China effects of 4-year elimination of human activities. Catena, 48, 217-233.

Kroetsch D, Wang C. 2008. Particle size distribution. In: Carter M R, Gregorich E G, eds., Soil Sampling and Methods of Analysis. 2nd ed. Boca Raton CRC, USA. pp. 713-725.

Lam K C, Leung Y F, Yao Q. 1997. Nutrient fluxes in the Shenchong Basin, Deqing County, South China. Catena, 29, 191-210.

Lampurlanes J, Cantero-Martinez C. 2003. Soil bulk density and penetration resistance under different tillage and crop management systems and their relationship with barley root growth. Agronomy Journal, 95, 526-536.

Lan H X, Hu R L, Yue Z Q, Lee C F, Wang S J. 2003. Engineering and geological characteristics of granite weathering profiles in South China. Journal of Asian Earth Sciences, 21 , 353-364.

Li Z W, Nie X D, Chen X L, Lu Y M, Jiang W G, Zeng G M. 2015. The effects of land use and landscape position on labile organic carbon and carbon management index in red soil hilly region, southern china. Journal of Mountain Science, 3, 626-636.

Liang B, Lehmann J, Solomon D, Kinyangi J, Grossman J, O'neill B, Neves E G. 2006. Black carbon increases cation exchange capacity in soils. Soil Science Society of America Journal, 70, 1719-1730.

Lin J S, Huang Y H, Wang M K, Jiang F S, Zhang X B, Ge H L. 2015. Assessing the sources of sediment transported in gully systems using a fingerprinting approach: An example from south-east china. Catena, 129, 9-17.

Lopez M V. 1998. Wind erosion in agricultural soil: an example of limited supply of particles available for erosion. Catena, 33, 17-28.

Luk S H, diCenzo P D, Liu X Z. 1997b. Water and sediment yield from a small catchment in the hilly granitic region, South China. Catena, 29, 177-189.

Luk S H, Yao Q Y, Gao J Q, Zhang J Q, He Y G, Huang S M. 1997a. Environmental analysis of soil erosion in Guangdong Province: A Deqing case study. Catena, 29, 97-113.

Ma X, Xu W N, Xia Z Y, Wang Y J, Guan S F, Zhang L L. 2015. The soil inorganic phosphorus distribution of vegetation-growing concrete substrate in the disturbed engineering area. In: International Conference on Materials, Environmental and Biological Engineering (MEBE 2015). Atlantis Press, French. pp. 424-428.

Mo J, Brown S, Peng S, Kong G. 2003. Nitrogen availability in disturbed, rehabilitated and mature forests of tropical China. Forest Ecology and Management, 175, 573-583.

Van Noordwijk M, Widianto H M, Hairiah K. 1991. Old tree root channels in acid soils in the humid tropics: Important for

crop production, water infiltration and nitrogen management. Plant and Soil, 134, 37-44.

Park J H, Meusburger K, Jang I, Kang H, Alewell C. 2014. Erosion-induced changes in soil biogeochemical and microbiological properties in Swiss Alpine grasslands. Soil Biology & Biochemistry, 69, 382-392.

Pizzeghello D, Berti A, Nardi S, Morari F. 2011. Phosphorus forms and P-sorption properties in three alkaline soils after long-term mineral and manure applications in north-eastern Italy. Agriculture Ecosystems & Environment, 141, 58-66.

Pizzeghello D, Berti A, Nardi S, Morari F. 2014. Phosphorus-related properties in the profiles of three Italian soils after long-term mineral and manure applications. Agriculture Ecosystems & Environment, 189, 216-228.

Reeves D W. 1997. The role of soil organic matter in maintaining soil quality in continuous cropping systems. Soil & Tillage Research, 43, 131-167.

Reintam E, Trukmann K, Kuht J, Toomsoo A, Teesalu T, Koster T, Edesi L, Nugis E. 2008. Effect of Cirsium arvense L. on soil physical properties and crop growth. Agricultural and Food Science, 17, 153-164.

Robertson G P, Sollins P, Ellis B G, Lajtha K. 1999. Exchangeable ions, pH, and cation exchange capacity. In: Robertson G P, Bledsoe C S, Coleman D C, Soilins P, eds., Standard Soil Methods for Long-Term Ecological Research. Oxford University Press, New York. pp. 106-114.

Roth C H, Eggert T. 1994. Mechanisms of aggregate breakdown involved in surface sealing, runoff generation and sediment concentration on loess soils. Soil & Tillage Research, 32, 253-268.

Scott Munro D, Huang L J. 1997. Rainfall, evaporation and runoff responses to hillslope aspect in the Shenchong Basin. Catena, 29, 131-144.

Sheng J A, Liao A Z. 1997. Erosion control in South China. Catena, 29, 211-221.

Sioh M, Woo M K, Lain K C. 1990. Soil nutrients in eroded granitic areas of South China. Physical Geography, 11, 260-276.

Spiteri C, Slomp C P, Regnier P, Meile C, Van Cappellen P. 2007. Modeling the geochemical fate and transport of wastewater-derived phosphorus in contrasting groundwater systems. Journal of Contaminant Hydrology, 92, 87-108.

Uchida T, Ohte N, Kimoto A, Mizuyama T, Li C. 2000. Sediment yield on a devastated hill in southern China: Effects of microbiotic crust on surface erosion process. Geomorphology, 32, 129-145.

Wang B, Zheng F L, Romkens M J M, Darboux F. 2013. Soil erodibility for water erosion: A perspective and Chinese experiences. Geomorphology, 187, 1-10.

Woo M K, Fang G, diCenzo P D. 1997a. The role of vegetation in the retardation of rill erosion. Catena, 29, 145-159.

Woo M K, Huang L, Zhang S, Li Y. 1997b. Rainfall in Guangdong Province, South China. Catena, 29, 115-129.

Woo M K, Luk S H. 1990. Vegetation effects on soil and water losses on weathered granitic hillslopes, south China. Physical Geography, 11, 1-16.

Wu Z F, Wang J Z. 2000. Relationship between slope disintegration and rock-soil characteristics of granite weathering mantle in south China. Research of Soil Water Conservation, 14, 31-35. (in Chinese)

Xia D, Deng Y S, Wang S L, Ding S W, Cai C F. 2015. Fractal features of soil particle-size distribution of different weathering profiles of the collapsing gullies in the hilly granitic region, south china. Natural Hazards, 79, 455-478.

Xia Z Y, Xu W N, Wang J Z. 2009. Ecological characteristics of artificial vegetation communities on excavated slopes at the Xiangjiaba hydroelectric power station. Journal of Chongqing University (English Edition), 2, 75-81.

Xu J. 1996. Benggang erosion: The influencing factors. Catena, 27, 249-263.

Xu J, Zeng G. 1992. Benggang erosion in sub-tropical granite weathering crust geo-ecosystems: An example from guangdong province. In Walling D E, Davies T R, Hasholt B P, eds., Erosion, Debris Fows and Environment in Mountain Regions. IAHS Press Publication, UK. pp. 455-463.

Zhang B, Yang Y S, Zepp H. 2004. Effect of vegetation restoration on soil and water erosion and nutrient losses of a severely eroded clayey Plinthudult in southeastern China. Catena, 57, 77-90.

Zhang G L, Gong Z T. 2012. Soil Survey Laboratory Methods. Science Press, Beijing, China. pp. 8-30. (in Chinese)

Zhang Y, Ding S W, Wei Y J, Wang Q X. 2015. Transfer rules of soil nutrients in collapsing pluvial fan. Transactions of the CSAM, 46, 216-222. (in Chinese)

Zhao H L, Zhou R L, Zhang T H, Zhao X Y. 2006. Effects of desertification on soil and crop growth properties in Horqin sandy cropland of Inner Mongolia, north China. Soil & Tillage Research, 87, 175-185.

Zhong B, Peng S, Zhang Q, Ma H, Cao S. 2013. Using an ecological economics approach to support the restoration of collapsing gullies in southern China. Land Use Policy, 32, 119-124.

(Managing editor SUN Lu-juan)