An-bang Wen's research while affiliated with Chinese Academy of Sciences and other places

Accurate assessment of soil erosion is an important prerequisite for controlling soil erosion. The engineering-control (E) and tillage (T) factors are the keys for Chinese Soil Loss Equation (CSLE) to accurately evaluate water erosion in China. Besides, the E and T factors can reflect the water and soil conservation effects of engineering-control and tillage practices. But in the current full coverage of soil erosion surveys in China (such as soil erosion dynamic monitoring), for the same practice, the E or T factors are assigned the same value across the country. We selected 469 E and T factors data based on runoff plots from 73 publications, and they came from six soil and water conservation regions. Correlation analysis, regression analysis, and nonparametric tests were used to determine the comparability of the data, and it was proved that the runoff plots dimensions are consistent with the local topography. The results of one-way ANOVA and nonparametric tests for E and T factors in different regions showed that the engineering-control practices have good soil and water conservation effects and weaken the regional differences of other environmental factors, so there were no significant differences in E factors between different regions. However, there were significant differences in T factors between different regions, and the geodetector was applied to explore the intrinsic driving force of the spatial distribution of T factors. The results of the geodetector showed that the dominant driving forces of the spatial distribution of different types of tillage practices were not completely the same. When using CSLE to calculate water erosion, the E factor of the same practice can be used uniformly throughout the country, and the T factor needs to be considered and selected according to regional differences. At the same time, when choosing tillage practices in each water and soil conservation region, practices with better sediment reduction benefits should also be selected according to the regional environmental conditions.

Characterizing soil particle-size distribution is a key measure towards soil property. The purpose of this study was to evaluate the multifractal characteristics of soil particle-size distribution among different land-use from a purple soil catchment and to generalize the spatial variation trend of multifractal parameters across the catchment. A total of 84 soil samples were collected from four kinds of land use patterns (dry land, orchard, paddy, and forest) in an agricultural catchment in the Three Gorges Reservoir Region, China. The multifractal analysis method was applied to quantitatively characterize the soil particle size distribution. Six soil particle size distribution (PSD) multifractal parameters (D(0), D(1), D(2), Δα(q), Δf[α(q)], α(0)) were computed. Additionally, a geostatistical analysis was employed to reveal the spatial differentiation and map the spatial distribution of these parameters. Evident multifractal characteristics were found. The trend of generalized dimension spectrum of four land use patterns was basically consistent with the range of 0.8 to 2.0. However, orchard showed the largest monotonic decline, while the forest demonstrated the smallest decrease. D(0) of the four land use patterns were ranked as: dry land < orchard < forest < paddy, the order of D(1) was: dry land < paddy < orchard < forest, D(2) presented a rand-size relationship as dry land < forest < paddy < orchard. Furthermore, all land-use patterns presented as Δf[α(q)] < 0. The rand-size relationship of α(0) was same as D(0). The best-fitting model for D(0), D(1), D(2) and Δf[α(q)] was spherical model, for Δα(q) was gaussian model, and for α(0) was exponential model with structure variance ratio was 1.03%, 49.83%, 0.84%, 1.48%, 22.20% and 10.60%, respectively. The results showed that soil particles of each land use pattern were distributed unevenly. The multifractal parameters under different land use have significant differences, except for Δα(q). Differences in the composition of soil particles lead to differences in the multifractal properties even though they belong to the same soil texture. Farming behavior may refine particles and enhance the heterogeneity of soil particle distribution. Our results provide an effective reference for quantifying the impact of human activities on soil system in the Three Gorges Reservoir region.

Most studies have evaluated the effects of soil redistribution under water erosion or tillage erosion within hill slope landscapes, yet few studies have demonstrated soil particle redistribution from the sloping lands with a hedge berm. In this study, soil particle redistribution and fractal dimension from sloping lands with or without a hedge berm were investigated. The main radius of soil particles being redistributed from the sloping lands ranged between 0.002 and 0.02 mm, whether or not a hedge berm existed. However, there were significant differences between sloping land with a hedge berm and without a hedge berm (P < 0.05). The values of the bias coefficient ranged from −0.893 to −0.817, and the values of peak convex coefficient ranged from −1.619 to −1.371. And the order of values of the fractal dimension for land with a hedge berm of 10 cm height was as follows: upper slope > middle slope > lower slope, and the sequence of the values of fractal dimension from sloping land without a hedge berm was upper slope >middle slope > lower slope. There was also a significant difference between the upper (US) and middle slope (MS) position (P < 0.05), and results also showed highly significant correlations (R2=0.9088 without hedge berm, R2 = 0.9048 with hedge berm, respectively). A regression analysis was performed to establish the logarithmic relationship between peak convex coefficient and soil depth (R2 = 0.9461 without hedge berm, R2 = 0.9780 with hedge berm, respectively). The correlation between fractal dimension and bias coefficient was significant (P = < 0.05). Our results suggest that fractal dimension, bias coefficient and peak convex coefficient were sensitive indices to evaluate the soil particle redistribution and the farming activities, such as building a hedge berm, could better describe the characters of soil structure and soil particle redistribution from sloping cultivated land.

Surface soil/sediment samples were collected from the Water-Level Fluctuation Zone (WLFZ), cultivated land and forest land at 50 different grid points from Shenjia watershed, the Three Gorges Reservoir area in August 2013. The spatial distribution, sources and ecological risk assessment for Arsenic (As), Cadmium (Cd), Chromium (Cr), Copper (Cu), Nickel (Ni), Lead (Pb) and Zinc (Zn) were analyzed in this study. The results showed all tested metals had similar distribution patterns except Ni and Cr, with areas of high concentrations distributed in the southwest (WLFZ and watershed outlet) of the study area. Ni and Cr, which were highly positively correlated and present in high concentrations, were primarily distributed in the south and middle zones of the study area. Lower concentration areas of all metals were uniformly distributed west of the high-elevation zones and forest land. Factor analysis (FA) and factor analysis-multiple linear regression (FA-MLR) showed that the major sources of Cd were fertilizer and traffic sources, which together accounted for 87% of Cd. As, Zn and Cu levels were primarily supplied by industrial and domestic sources, accounting for 76% of As, 75% of Cu and 67% of Zn. Surface soils/sediments of the study watershed contaminated by Cd represent a high ecological risk, whereas other metals represent low ecological risks. The potential ecological risk index (PERI) analysis indicated that it had a low (wide-range) ecological risk and a moderate (small-range) ecological risk primarily distributed in the outlet of the study watershed. Fertilizers and traffic are the primary sources of Cd pollution, which should be more closely controlled for the purposes of water quality and ecological conservation.

Many factors can affect the sediment deposition and soil erosion process in riparian zone, including terrain, sediment transport and water level fluctuations. Clarifying the factors influencing sediment deposition process in the riparian zone of the Three Gorges Reservoirs is an important problem to determine the key area of sediment deposition and its trend of development in the study area. In order to reveal the influence of these environmental factors on the sediment deposition in riparian zone of the Three Gorges Reservoir, this study investigated 1) the amount of deposited sediment in different environmental conditions, 2) the potential factors affecting sediment deposition in riparian zone of the Three Gorges Reservoir, 3) the relationship between the deposited sediment amount and these factors previously mentioned using correlation analysis, and 4) the influence of human activities considered as an additional factor. This study found that 1) slope gradient, elevation, inundating duration and human activities were the main factors influencing sedimentation in riparian zone of the Three Gorges Reservoir, and 2) the impact of each factor varied with spaces. Specifically, in the upper reach from Jiangjin to Fuling, human activities such as gravel dredging, bank revetment and ports and wharfs constructing disturbed considerable amounts of deposited sediment, as a result, there was no natural law to dictate the distribution. In the middle reach from Fuling to Fengjie, slope gradient and inundating duration were the controlling factors, and the sediment deposition amount was greater in the areas with a gentler slope or lower elevation. Water flow on gentler slopes generally had lower velocity, resulting in more sediment to deposit. Sites with lower elevations would be drowned by sediment-laden flow with a longer duration resulting from hydrologic regime controlled by the operating strategy of the Three Gorges Reservoir, leading to a larger amount of sediment deposition. In the lower reach from Fengjie to Zigui, slope gradient was similar to the middle reach, performing a primary factor, while other factors showed little relationship with sediment amount.

In order to analyze the spatial variation characteristics of grain diameter, nutrient elements and heavy metal pollution with deposition sediment in tributaries bay of the Three Gorges Reservoir, we selected 9 typical tributaries bay, 54 deposited sediment samples were collected from the riparian zone for analyzing grain diameter distribution, capacity, organic matter, nutrient elements of TN, TP and K, heavy metal elements of Cr, Cu, Ni, Pb, and Zn. The results indicated that particle size distribution from Wujiang River in Fuling to the Modao stream in Yunyang presented a trend of fluctuation, deposited sediment at 160-165 m elevation was coarser than that at 165-175 m elevation,volume percent of sand and clay presented a moderate variation at both altitudes, while silt had small variation. Independent sample t test showed that characteristics difference between the upper and lower sediments in riparian zone was not significant. The geo-accumulation index of heavy metal pollutants in the sediment from riparian zone of the Three Gorges Reservoir tributaries bay indicated that, only Zn element in Zhenxi River, Longdong River and Long River, Pb element in the Modao Stream belonged to non-moderate pollution levels, whereas there were no pollution of all other elements in tributaries.

Soil erosion becomes a serious environmental problem in the world, especially in western China. An effective management practice called the Grain for Green Program (GGP), which was launched in 1999, aims to reduce soil and water loss and alleviate the ecological environment problem in western China. Two typical counties in western China, the Zhongxian (in Chongqing Municipality) and Ansai (in Shaanxi Province) were chosen to evaluate the dynamic changes of land use and agricultural production structure before and after the implementation of the Program in this paper. The results showed that the cultivated land area was reduced by 7.08% from 1989 to 2003. The cultivated land per person was decreased by 8.42% during 1999–2003. Moreover, the stability index of the secondary sector of the economy was increased from 0.91 in the period 1990–1999 to 0.94 in the following ten years. In addition, the stability index of tertiary economic sector increased from 0.88 to 0.92 in Zhongxian county. Meanwhile, the cultivated land area was reduced by 15.48% from 1990 to 1999. The soil erosion modulus was decreased by 33.33% from 1999 to 2006. Also, the stability index of secondary and tertiary economic sectors was 0.86 in the period 1998–2002. However, it decreased by 77% during 2002 to 2007 in Ansai County. These results imply that the Grain for Green Program had different impact on the two regions. Several effective strategies of soil and water conservation have been carried out to ameliorate the sustainable development of ecological environment and economy in these two counties of western China.

Soil erosion and associated off-site sedimentation are threatening the sustainable use of the Three Gorges Dam. To initiate management intervention to reduce sediment yields, there is an increasing need for reliable information on soil erosion in the Three Gorges Reservoir Region (TGRR). The purpose of this study is to use 137Cs tracing methods to construct a sediment budget for a small agricultural catchment in the TGRR. Cores were taken from a pond and from paddy fields, for 137Cs measurements. The results show that the average sedimentation rate in the pond since 1963 is 1.50 g cm−2 yr−1 and the corresponding amount of sediment deposited is 1,553 t. The surface erosion rate for the sloping cultivated lands and the sedimentation rate in the paddy fields were estimated to be 3,770 t km−2 yr−1 and 2,600 t km−2 yr−1, respectively. Based on the estimated erosion and deposition rates, and the area of each unit, the post 1970 sediment budget for the catchment has been constructed. A sediment delivery ratio of 0.5 has been estimated for the past 42 years. The data indicate that the sloping cultivated lands are the primary sediment source areas, and that the paddy fields are deposition zones. The typical land use pattern (with the upper parts characterized by sloping cultivated land and the lower parts by paddy fields) plays an important role in reducing sediment yield from agricultural catchments in the TGRR. A 137Cs profile for the sediment deposited in a pond is shown to provide an effective means of estimating the land surface erosion rate in the upstream catchment.

There is very thin soil layer in karst rocky desertification areas in Southwest China, sediment deposition and sediment yield in the karst area affects the development of vegetation greatly. In the present study, the 137Cs technique was used to assess the rate of sediment deposition and sediment yield in a small karst catchment. The 137Cs inventory within the depression varied between 800 m−2 and 2,200 Bq m−2, with the mean value of 1,500.1 Bq m−2. The 137Cs reference inventory at a nearby reference site was 805.9 Bq m−2. It could be inferred to that sediment deposition had occurred in the catchment. The mean depth of sediments deposition in the depression was 6 cm and the deposition rate was approximately 0.13 cm yr−1. The analysis of the topographic characteristics of the catchment revealed that the sediment deposition occurred mainly at the lower part of the small catchment. Although, there was a sinkhole in the depression, little sediment had drained out with runoff through the sinkhole, because the local people built ridges around the sinkholes for storing water. According to this, sediment yield rate in the small catchment was estimated to be approximately 19.25 to 27.5 t km−2 yr−1, and the extremely low sediment yield was maybe the main obstacle to vegetation restoration in karst rocky desertification areas.

The fallout radionuclide cesium-137 (137Cs) has been widely employed as a tracer for assessment of soil loss from thick uniform soils; however, few studies have been conducted on thin stony soils on slopes underlain by carbonate rocks which are widely distributed in karst areas. Information derived from 137Cs measurement of soil samples collected along a carbonate rock slope with thin stony soil where neither soil erosion nor deposition occurred was used to investigate the characteristics of 137Cs redistribution in a karst area of Southwest China. The results indicated that the 137Cs inventories of the surface soil on the slope studied were much lower than that of the local 137Cs reference inventory and the 137Cs activities were much higher than those on slopes with thick uniform soils. The spatial distribution of 137Cs inventories was characterized by considerable variation. The high 137Cs depletion in the stony soil of the slope studied was mainly because a considerable proportion of the fallout input of 137Cs could be lost with runoff and the dissolution of carbonate particles in the soil promoted the loss of 137Cs. These demonstrated that the rates of soil loss could not be estimated from the degree of depletion of the 137Cs inventory relative to the local reference inventory for the thin stony soil of the rocky slope underlain by carbonate rocks in the study area in the way that has been widely used in areas with thick uniform soils.