Search

Hydrological characteristics are expected to be affected by climate variability and vegetation dynamics which are interconnected and coupled in most cases. Few studies have decoupled the effects of vegetation dynamics and climate variability on hydrological characteristics, a process that benefit both water resources management and agricultural water allocation. In this study, we used partial least squares-structural equation modeling (PLS-SEM) to decouple the effects of climate variability and vegetation on the temporal variations in hydrological characteristics on a monthly scale in the Upper Du watershed (8973 km2) in subtropical China. Monthly hydrometeorological and vegetation cover data were collected from 2000 to 2010. Moderate Resolution Imaging Spectroradiometer (MODIS) derived vegetation indices were used to represent the vegetation status of the watershed. The results showed that vegetation dynamics and climate variability account for up to 67% of the temporal variation in runoff, whereas the combined effects of vegetation, climate and runoff explain 62% of the variation in sediment. Climate variability both directly affects monthly hydrological characteristics and indirectly affects these characteristics through its effects on vegetation. Vegetation is negatively correlated with both runoff and sediment, and the net effect of vegetation on the sediment load (total effect = −0.20) is greater than its effect on runoff (total effect = −0.12). Our results indicate that the direct effect of vegetation on sediment (direct effect = −0.09) is smaller than the indirect effect (indirect effect = −0.11) of vegetation on sediment through its influence on runoff; thus, vegetation reduces both the sediment concentration and the sediment load mainly by reducing runoff. Compared to conventional multivariate statistical methodologies, PLS-SEM provides additional valuable information, including the direct and indirect impacts of climatic factors and vegetation on hydrological characteristics.

Large-scale vegetation restoration generally reduces local water yield and influences river ecosystem health. Thus, scientific evaluation of vegetation restoration consequences is necessary for maintaining the stability of the surface water system and water cycle in semi-arid regions. In this study, we compared the efficiencies of different typical grasslands in regulating runoff and sediment yields and proposed feasible suggestions suiting for local environments. Four grasslands, including two Gramineae species (Elymus dahuricus and Bromus inermis) and two legume species (Medicago sativa and Trifolium repens), were tested during a two-year period with simulated rainfall experiments, and using bare land as control. Three replicates were done for each treatment, and fifteen plots with a slope of 20° were constructed. Three indices were used to assess the runoff and sediment yields reduction capacity of the grasslands, including runoff reduction benefit (RRB), sediment reduction benefit (SRB), and soil infiltration rate (SIR). The results showed that RRB and SRB were significantly different (P

The frequency of harmful algal blooms caused by eutrophication is increasing globally, posing serious threats to human health and economic development. Reservoir bays, affected by water environment and local watershed landscape, are more prone to eutrophication and algal blooms. The chlorophyll a (Chl a) concentration is an important indicator for the degree of eutrophication and algal bloom. Exploring the complex relationships between water environment and landscape background, and Chl a concentration in the reservoir bays are crucial for ensuring high-quality drinking water from reservoirs. In this study, we monitored Chl a concentrations of 66 bays in Danjiangkou Reservoir and the related water quality parameters (e.g., water temperature, turbidity, nutrients) in waterbodies of these reservoir bays in the storage and discharge periods from 2015 to 2018. Partial least squares-structural equation modeling (PLS-SEM) was used to quantify the relationship between water environmental factors and watershed landscapes, and Chl a concentrations in reservoir bays. The results showed that mean Chl a concentration was higher in storage period than that in discharge period. Two optimal PLS-SEMs explained 66.8% and 53.6% of Chl a concentration variation in the storage and discharge periods, respectively. The net effect of water chemistry on Chl a concentration was more pronounced during the discharge period (total effect = 0.61, 37% of the total effect on Chl a), while the net effect of land-use composition on Chl a concentration was more significant during the storage period (total effect = 0.57, 30% of the total effect on Chl a). The landscape pattern had significant indirect effects on Chl a concentration, especially during the discharge period (indirect effect = −0.31, 19% of the total effect on Chl a). Our results provide valuable information for managers to make rational decisions, thereby contributing to the prevention of eutrophication and algal blooms in reservoir bays.

Terracing and rainfall characteristics notably influence the water erosion processes. However, an extensive long term in situ quantitative evaluation of the approaches to control the water erosion in different orchard terraces has not yet been performed, especially considering the increasing frequency and severity of extreme rainfall events due to the global climate change. In this study, six types of orchard terraces, including slope land as the control (SLck), level terrace with bare (LTb) and vegetation taluses (LTv), outward (OTv) and inward terrace (ITv) with vegetation taluses and level terrace having front mounds and back ditches with vegetation taluses (MDLTv), were used to analyze the effects of extreme and ordinary rainfall events on the surface runoff and soil erosion. According to the measured data for twelve consecutive years, 356 natural rainfall events were divided into extreme and ordinary rainfall, based on the World Meteorological Organization standard. The results indicated that more severe surface runoff and sediment loss occurred under extreme rainfall: the runoff coefficient and soil loss under extreme rainfall were 2.6 and 11.5 times those under ordinary rainfall, respectively. The sediment yield (contribution rate, 42.9%) exhibited a higher sensitivity to extreme rainfall events compared to that of the surface runoff generation (contribution rate, 16.4%). Moreover, the reduction in the surface runoff and sediment in the extreme rainfall case differed for different orchard terraces. The average surface runoff coefficient and soil loss amount decreased in the following order: SLck>LTb>OTv>LTv>ITv>MDLTv. Nevertheless, the highest and lowest contributions of the extreme rainfall to the sediment yield occurred in the LTb (64.8%) and MDLTv (21.7%) plots, respectively. Therefore, severe talus erosion caused by extreme rainfall should be monitored, and a combination of vegetation taluses and front mounds and back ditches on the platforms is recommended as a sustainable strategy to prevent extreme water erosion when transforming slope land into orchard terraces.

This paper aims to investigate the hydrological response of a large-scale (8973 km2) mountainous watershed to different rainstorm spatial patterns and reforestation. Based on 32 years of observations, measurements of 184 rainstorm events and 125 sediment-producing events with complete hydrographs were analyzed. The K-means clustering method was used to classify the spatial patterns of rainstorm events in accordance with their event-based spatial rainfall characteristics. The 184 rainstorm events were classified into four spatial patterns, among which the spatial features differ significantly: (I) Spatial Pattern I (SPI) includes rainstorms with a low amount of cumulative areal rainfall (27.4 mm), the highest spatial variability (0.986), and the highest frequency; (II) Spatial Pattern II (SPII) includes rainstorms of high spatial variability (0.759) and the largest amount of local maximum daily rainfall (106.8 mm); (III) Spatial Pattern III (SPIII) includes rainstorms with a medium amount of cumulative areal rainfall (58.7 mm) and low spatial variability (0.362); and (IV) Spatial Pattern IV (SPIV) includes rainstorms with the largest amount of cumulative areal rainfall (117.2 mm) and the lowest spatial variability (0.313). Vegetation cover in the upper Du watershed was significantly improved after the implementation of the Grain-for-Green project. The average area-specific sediment yields (SSY) for the four SPs were 15.4, 65.5, 55.8, and 286.2 t km−2 before reforestation and decreased to 6.0, 59.3, 43.7 and 89.9 t km−2, respectively, after reforestation. ANOVA (analysis of variance) indicated that reforestation resulted in a significant reduction in runoff coefficient under SPIII and SPIV and a significant reduction in SSY under SPI and SPIV. A hysteresis analysis suggested that the proportion of events with a clockwise loop increased from 64.9% before reforestation to 82.1% after reforestation and that complex loops became less common during 2000–2010 under SPIV, thereby implying a reduced sediment supply.

Neonicotinoids pollution poses a serious threat to aquatic ecosystems. However, there is currently little knowledge about how neonicotinoids are transferred from the agricultural environment to the aquatic environment. Here, we conducted in situ high-frequency monitoring of neonicotinoids in soil-water systems along the hydrological flow path during rainfall to explore the horizontal and vertical transport mechanisms of neonicotinoids. The collected samples included 240 surface runoff, 128 subsurface runoff, 60 eroded sediment, 120 soil and 144 soil solution, which were used to analyse neonicotinoids concentrations. Surface runoff, subsurface runoff and eroded sediment were the three main paths for the horizontal migration of neonicotinoids. In the CK (citrus orchards without grass cover) and grass-covered citrus orchards, there are 15.89% and 2.29% of the applied neonicotinoids were transported with surface runoff, respectively. While in the CK and grass-covered citrus orchards, there are only 1.23% and 0.19% of the applied neonicotinoids were transported with eroded sediment and subsurface runoff. Although the amount of neonicotinoids lost along with eroded sediment was small, the concentration of neonicotinoids in eroded sediment was two orders of magnitude higher than the concentration of neonicotinoids in sediments of the surface water. Meanwhile, neonicotinoids migrated vertically in soil due to water infiltration. In the CK and grass-covered citrus orchards, there are 57.64% and 24.36% of the applied neonicotinoids were retained in soil and soil solution, respectively, and their concentration decreased as soil depth increased. Another noteworthy phenomenon is that more neonicotinoids migrated to deeper soil layers under grass cover compared with no grass cover because grass roots promoted the formation of cracks and vertical preferential flow. Our results are expected to improve the accuracy of neonicotinoids pollution prediction by considering migration paths, including surface and subsurface runoff and eroded sediment.

Changes in natural rainfall regimes have taken place and are expected to become more pronounced in future decades. These changes are also likely to be accompanied by changes in crop management practices. The main purpose of this study was to analyze runoff and soil loss in relation to rainfall regimes and terracing in the Three Gorges area, China. Based on 10 years of field observation and k-mean clusters, 101 rainfall events were grouped into three rainfall regimes. Rainfall regime I was the group of events with strong rainfall intensity, high frequency, and short duration. Rainfall regime III consisted of events with low intensity, long duration, and high rainfall amount. Rainfall regime II was the aggregation of events of high intensity and amount, and less frequent occurrence. The results showed that event runoff coefficients were not significantly different among rainfall regimes. However, the average soil erosion rates in rainfall regimes I and II were significantly higher than that in regime III. The average erosion rates under rainfall regimes I, II, and III were 21.6, 39.7, and 9.8 g m−2, respectively. The effect of rainfall regime on soil erosion also was changed by terracing. On unterraced cropland, soil erosion rate in rainfall regime I is significantly higher than that in regime III. However, the situation did not exist in unterraced orchard. Terracing significantly reduced runoff and soil erosion, and compensated the effects of rainfall regime on soil erosion, which indicated that runoff and erosion in terraced system may be little influenced by climate change. Based on these results, it was suggested more attention should be paid to the timing of rainfall events in relation to crop development and the high erosion on unterraced citrus orchard to control soil erosion in this area.

Anthropogenic pressure coupled with strong precipitation events and a mountainous landscape have led to serious soil erosion and associated problems in the subtropical climate zone of China. This study analyzes 1576 rainfall-runoff-soil loss events at 36 experimental plots (a total of 148 plot-years of data) under a wide range of conditions in subtropical mountainous areas of China where slope farming is commonly practiced. The plots, which have standardized dimensions, represent five common types of land use and have four different slopes. Event-based analyses show that almost half of the total rainfall caused soil erosion in the study area. The dominant factor controlling the runoff coefficient is the slope gradient rather than the land use type. The maximum soil lossfor crop plots under steep tillage (35°) is 5004 t km−2 for a single event. Among the common local crops, the average soil loss values increase in the following order: buckwheat cropland (1179 t km−2 year−1) > terraced cropland (1083 t km−2 year−1) > orchard land (1020 t km−2 year−1) > grassland (762 t km−2 year−1) > terraced orchard land (297 t km−2 year−1) > forest and grassland (281 t km−2 year−1).

Although the fallout radionuclide 137Cs technique has been widely used to provide soil redistribution estimates in recent decades, some researchers have recently challenged the usefulness of this technique. Thus, it is necessary to evaluate the usefulness of the 137Cs technique to study soil redistribution, especially in areas that suffer from severe soil erosion and where the 137Cs inventory distributions vary. This study analyses the abundance of 137Cs deposited in sediment trapped by a check dam over 31 years in a small agriculture catchment in the Loess Plateau, which has experienced some of the most extensive agricultural soil erosion in the world. The 137Cs inventory and soil loss were evaluated, and the results indicated that the sediment deposition rate behind the check dam was approximately 38 cm/year. The specific soil erosion rate of the catchment is estimated to be 178 t ha−1 year−1. A geochemical properties fingerprinting method indicated that the ratio of the inter-gully erosion from cropland to gully erosion was 0·46:0·54 during the investigated period of 31-year period. Our results suggest that the 137Cs technique cannot be directly used in the study catchment. Gully erosion is the main type of soil erosion that occurs in the loess catchment, and the amount of gully erosion changes from year to year. Therefore, the abundance of 137Cs cannot be used to estimate soil redistribution where gully erosion is severe. However, after calibration, the inter-gully soil loss was consistently related to 137Cs abundance, which means that the 137Cs technique can be used at the plot or slope scale on the Loess Plateau. Our results also reveal the particle-size selectivity of soil detachment and sediment deposition during runoff-erosion events. The average 137Cs content of the fine group couplets is 63% greater than the 137Cs content of the coarse group. When the studied section spans more than 15 years, the 137Cs inventory is consistent (R2 = 0·95, p

Nitrate exports from diffuse sources constitute a major cause of eutrophication and episodic acidification in inland aquatic systems, and remedial action requires the identification of the influencing factors associated with these nitrate exports. This paper examines the combined effects of watershed complexity on nitrate concentration and flux in terms of the hydrological and environmental factors in heterogeneous nested subwatersheds in the Danjiangkou Reservoir Area (DRA), China. We established 15 sampling sites in the main stream and tributaries and conducted biweekly sampling in 2008–2012 to monitor the nitrate exports. The hydrological and environmental indices within the watershed were divided into subwatersheds and considered as potential influencing factors. In consideration of the high co-linearity of these influencing factors, we used partial least squares regression (PLSR) to determine the associations between the stream nitrate concentration or flux and 26 selected watershed characteristics. The number of components was unequal for the nitrate concentration and flux models. The optimal models explained 66.4%, 60.0% and 59.9% of the variability in nitrate concentration and 74.7%, 67.1% and 58.0% of the variability in nitrate flux annually, in the dry season, and in the wet season, respectively. According to the variable importance in the projection (VIP) values, the dominant first-order factors for the nitrate concentration were as follows: the areal percentages of agricultural, forest and residential areas; followed by the slope; the largest patch index (LPI); the flow path gradient (FPG); the slope gradient variance (SGV); and the splitting index (SPLIT). In addition to these factors, the runoff coefficient (RC), flashiness index (FI), and patch density (PD) affected the changes in the nitrate flux. This study illustrates the influence of hydrological and environmental factors on seasonal water quality and can serve as guidelines for better watershed modeling and effective water quality management.

Long-term sedimentary sequence research can reveal how human activities and climate interact to affect catchment vegetation, flooding, soil erosion, and sediment sources. In this study, a biomarker sediment fingerprinting technique based on n-alkanes was used to identify long timescale (decadal) sediment sources in a small agricultural catchment. However, the highly saline carbonate environment and bacterial and algal activities elevated the levels of even-chain n-alkanes in the sediments, leading to an obvious even-over-odd predominance of short and middle components (C 15 -C 26 ). Therefore, by analyzing three odd, long-chain n-alkanes (C 27 , C 29 and C 31 ) in 27 source samples from cropland, gully, and steep slope areas and one sediment sequence (one cultivated horizon and 47 flood couplets), a composite fingerprinting method and genetic algorithm optimization were applied to find the optimal source contributions to sediments. The biomarker fingerprinting results demonstrated that the primary sediment source is gullies, followed by cropland and steep slope areas. The average median source contributions associated with 47 flood couples collected from sediment core samples ranged from 0 ± 0.1% to 91.9 ± 0.4% with an average of 45.0% for gullies, 0 ± 0.4% to 95.6 ± 1.6% with an average of 38.2% for cropland, and 0 ± 2.1% to 60.7 ± 0.4% with an average of 16.8% for steep slopes. However, because farmers were highly motivated to manage the cropland after the 1980s, over half the sediments were derived from cropland in the 1980s. Biomarkers have significant advantages in the identification of sediments derived from different landscape units (e.g., gully and steep slope areas), and n-alkanes have considerable potential in high-resolution research of environmental change based on soil erosion in the hilly Loess Plateau region.

6 Pags.- 4 Figs.- Suppl. Data., The application of ecological restoration programs, related to water resources protection and soil erosion control, may have some undesirable outcomes. An important example is the effect that vegetation restoration may have in reducing surface water resources. After searching peer-reviewed articles, we selected 38 publications from 16 countries in comparable areas - semiarid conditions (aridity index 50% and fine soil texture - to evaluate the effectiveness of different types of vegetation (i.e., forestland, scrubland and grassland) in regulating runoff and sediment transport. In particular, we used three indices: the runoff reduction effect, the sediment reduction effect and the ratio between runoff and sediment reduction. These indices were calculated from measured data reported in the original articles. Results showed that scrubland had higher runoff reduction effect (59% in gentle slopes; 65% in steep slopes) than in grassland (39% on gentle slopes; 43% on steep slopes) and forestland (33% on gentle slopes; 51% on steep slopes). For the three types of vegetation, the sediment reduction effect was >70%. Concerning the ratios between runoff and sediment reduction, grassland showed the lowest ratios (56% on gentle slopes; 53% on steep slopes) compared to forestland (63% on gentle slopes; 65% on steep slopes) and scrubland (93% on gentle slopes; 81% on steep slopes). Our results indicate that low values of ratios between runoff and sediment reduction are the most suitable because they indicate an effective soil erosion and sediment delivery reduction but maintaining surface runoff. Overall, our study demonstrates that grassland may be the best choice for optimizing the trade-off between catchment water yield and soil conservation during the implementation of ecological restoration programs in semi-arid regions., This study was funded by Projects of the National Natural Science Foundation of China (NSFC41722107, 41525003, 41977063, 41930755), and the Youth Talent Plan Foundation of Northwest A & F University (2452018025).

Summary Theoretical explanations for vegetation pattern dynamic emphasized on banded pattern-forming systems on the dynamics of the spot pattern. In this context, we explore the patch pattern forming and development in the desertification land. We hypothesized that spatial heterogeneity of microtopography and soil properties with different patch sizes would determine vegetation pattern dynamics theory. The spatial heterogeneity of microtopography and soil properties with different patch sizes were studied. Differences between the inside and outside of the canopy of soil carbon content and soil total nitrogen content were significantly increasing with patches sizes. Sampling location across vegetation patch was the main factor controlling soil properties. Soil nutrient content and saturated hydraulic conductivity were the largest, while bulk density and the coarse sand content were the lowest at the sampling location of half-way between taproot and downslope edge of the canopy. The height of the mound relative to the adjacent soil interspace between shrubs increased as patches diameter increased at the upslope of the taproot. Hydrological and aeolian processes resulted in spatial distributions of soil moisture, nutrition properties, which lead to patch migrated to downslope rather than upslope. A conceptual model was integrated hydrological and nutrient facilitation and competition effects among the plant-soil in mosaic-pattern patch formation and succession process.

Non-point source nitrogen loss poses a risk to sustainable aquatic ecosystems. However, non-point sources, as well as impaired river segments with high nitrogen concentrations, are difficult to monitor and regulate because of their diffusive nature, budget constraints, and resource deficiencies. For the purpose of catchment management, the Bayesian maximum entropy approach and spatial regression models have been used to explore the spatiotemporal patterns of non-point source nitrogen loss. In this study, a total of 18 sampling sites were selected along the river network in the Hujiashan Catchment. Over the time period of 2008–2012, water samples were collected 116 times at each site and analyzed for non-point source nitrogen loss. The morphometric variables and soil drainage of different land cover types were studied and considered potential factors affecting nitrogen loss. The results revealed that, compared with the approach using the Euclidean distance, the Bayesian maximum entropy approach using the river distance led to an appreciable 10.1% reduction in the estimation error, and more than 53.3% and 44.7% of the river network in the dry and wet seasons, respectively, had a probability of non-point source nitrogen impairment. The proportion of the impaired river segments exhibited an overall decreasing trend in the study catchment from 2008 to 2012, and the reduction in the wet seasons was greater than that in the dry seasons. High nitrogen concentrations were primarily found in the downstream reaches and river segments close to the residential lands. Croplands and residential lands were the dominant factors affecting non-point source nitrogen loss, and explained up to 70.7% of total nitrogen in the dry seasons and 54.7% in the wet seasons. A thorough understanding of the location of impaired river segments and the dominant factors affecting total nitrogen concentration would have considerable importance for catchment management.

Human activity is thought to perturb the dynamic change of organic carbon (OC), but its impact on the transportation and redistribution of erosion-induced OC is still poorly understood. Here, we use multi-source field sampling data and long-term hydrological observation data to clarify the transportation and redistribution of OC in an agricultural watershed on the Loess Plateau. The endmember mixing model based on the radiocarbon isotopes (14C) and a budget equation were used to assess the composition and budget of erosion-induced OC in a 187 km2 watershed. The results showed that compared with the period of agricultural activities (1960−1969), soil conservation activities in 1970−1999 and 2000−2019 reduced soil erosion by 31.5 % and 75.4 %, respectively. Additionally, land use changes significantly reduced the mobilized OC, from 14,370 ± 1966 (1960−1969) to 3311 ± 431 Mg C yr−1 (2000−2019). Check dam construction led to the mobilized OC was effectively buried at a rate of 2125 ± 478 and 1420 ± 282 Mg C yr-1 in periods of 1970−1999 and 2000−2019, respectively. Further radiocarbon isotopes analysis shows that the land use changes and check dam construction effectively reduced the output of ancient petrogenic organic carbon (OCpetro) and young biospheric organic carbon (OCbio) from the watershed, which has a positive impact on the regulation of atmospheric CO2 level. Our results emphasize the positive effects of soil conservation activities on terrestrial carbon sequestration, which are of great significance to clarify the effects of soil erosion on the carbon cycle.

Socio-economic development greatly affects soil erosion and the implementation of soil and water conservation (SWC) techniques. Here, we proposed a new index, the potential for soil erosion control (P), and investigated socio-economic determinants for ΔP using Partial Least Squares Regression (PLSR) during 2005–2015 in the hilly red soil region of southern China. P was defined as 1 minus the ratio of the minimum possible soil erosion to the actual soil erosion, which can reflect the gap between the actual soil erosion and the minimum possible soil erosion. According to P, it would be impossible for 100% of the study area to experience a soil erosion rate of less than the tolerable soil erosion rate due to the biophysical limits, which would avoid waste caused by excessive governance. The two PLSR models were statistically significant and the selected socio-economic factors can account for 37% and 28% of the variations in the positive and negative |ΔP|, respectively. Base on the PLSR model for the positive |ΔP|, the VIP value of urban population growth (URBPOP_GR) and rural population growth (RURPOP_GR) were −1.31 and 1.71, respectively, demonstrating that URBPOP_GR suppressed erosion while RURPOP_GR increased erosion. Government investment in afforestation can significantly reduce soil erosion (VIP greater than 1). Increasing grain output per km2 would reduce the positive |ΔP|, indicating that advanced farming practices can balance production and soil protection. These results emphasized that policies for soil erosion control should be developed from not only a simplistic technological perspective but also a social-ecological perspective, and economic growth will control soil erosion effectively if it can provide more positive feedback to rural areas.

The cover-management factor (C-factor) in the revised universal soil loss equation (RUSLE) indicates the effects of vegetation cover and management practices on water erosion. The conditions governing the C-factor can be managed by farmers and managers to affect soil loss and soil carbon cycle. Currently, the most common approach for estimating the C-factor on large scales is using the normalized difference vegetation index (NDVI) individually, which cannot precisely characterize the differences in the C-factor in various land-use types with the same degree of vegetation coverage. We developed a multiple-land-use synthesized C-factor (MLUS-C) model to capture the dynamics of the C-factor with a high spatio-temporal resolution for mixed-landscape watersheds. In this model, space-time fusion was executed based on multi-source remotely sensed data to estimate the phenological succession of vegetation cover, and the impacts of farmland management on large-scale evaluation of the C-factor were considered. A case study was conducted in the Gongshui watershed, southern China, where the dataset of precipitation and sediment yield covered a period of 11 years. We validated the simulation from the MLUS-C model and compared it with the results from NDVI-only approaches via conversion from the C-factor to sediment yield. The validation results showed that the MLUS-C model significantly improved the simulation accuracy and model adaptability. The root mean square error of our model was 71.5-95.5% lower than those of NDVI-only approaches. Our model has an advantage in estimating the C-factor in heterogeneous landscapes and provides a basis for implementing measurements to efficiently reduce soil loss.

China is facing the challenge of the uncertain impacts of large-scale afforestation on regional water resources. However, the effects of vegetation cover changes on the variation in surface water at the regional scale are still controversial. Here, we focused on the 0.9 million km2 vegetation restoration region in China, where the highest significant vegetation cover changes on the earth. Multi-source remote sensing data were used to describe the characteristics of seasonality and transition of surface water and to analyse the causes of surface water changes from climate, vegetation cover and other human factors. Our results show that the annual maximum NDVI of Northeast region (NE) and Loess Plateau region (LP) increased significantly from 0.74 to 0.85 and 0.49 to 0.62 from 2000 to 2015, respectively. Meanwhile, permanent water, as a vital component of surface water, exhibited net increases of 695.6 km2 and 119.4 km2 in NE and LP from 2000 to 2015, respectively. The extension in permanent water and the implementation of ecological projects exhibited highly consistent spatiotemporal patterns. Statistical analysis indicated that vegetation cover is an important factor in controlling permanent water changes. Human activities such as building dams and reservoirs are also an important explanatory variable for permanent water increases. The newly built dams contributed 43% in NE and 25% in LP to the increase in permanent water. In addition, although climatic factors were not the main factor influencing permanent water, precipitation significantly affected the total surface water in NE. These findings have potential implications for understanding surface water and forest dynamics and formulating regional development plans in the vegetation restoration region in China.

Reservoir bays, which are affected by the reservoir and watershed characteristics, are the initial and most sensitive areas in the evolution process of reservoir water quality. However, the relationship between the watershed characteristics and nitrogen and phosphorus concentrations in reservoir bays is poorly understood. We selected 66 bays from the Danjiangkou Reservoir and sampled twice per year (storage and discharge periods) from 2015 to 2018 to monitor the total nitrogen (TN) and total phosphorus (TP) concentration in the waterbodies of the reservoir bays. Four types of watershed characteristic indices (topographic variables, soil variables, land-use composition, and landscape patterns) around these bays were obtained. We quantified the relationship between the TN and TP concentrations and watershed characteristics in the waterbodies of the reservoir bays using partial least squares regression (PLSR). The results showed that the mean concentrations of TN and TP in the storage period (TN:1.69 mg·L−1, TP:0.088 mg·L−1) were higher than those in the discharge period (TN:1.22 mg·L−1, TP:0.063 mg·L−1). The optimal PLSR models explained 67.9% and 82.5% of the TN concentration variability, and 65.4% and 67.2% of the TP concentration variability during the storage and discharge period, respectively. Based on the variable importance in the projection (VIP) values, soil erodibility had significant effects on the TN and TP concentrations. The key factors affecting the TN concentration were the slope gradient, basin relief, topographic wetness index, forest and agricultural land use, whereas the factors controlling the TP concentration were the landscape shape index, edge density, Shannon's diversity index and grass land use, although the TP concentration was also controlled by the patch density and contagion during the storage period, and by mean patch size and largest patch index during the discharge period. This study provides critical insights into sustainable landscape planning and effective reservoir water quality management.

The widespread use of neonicotinoids has increased the risk of non-point source pollution. Previous studies mainly focused on neonicotinoid transport by runoff on gentle slopes and ignored the role of sediment. Neonicotinoid migration due to erosion could be underestimated because a considerable amount of sediment migration is caused by erosion on steep slopes. Here, we focused on the transport of two typical neonicotinoids, imidacloprid (IMI) and clothianidin (CLO), via runoff and sediment in new citrus orchards on steep slopes. The total runoff from the bare land (BL), biological crust (BC) and Kummerowia striata (Thunb.) Schindl. (KS) plots was 77.73, 66.22 and 46.94 mm, respectively, and the corresponding total sediment was 162.47, 54.00 and 3.51 kg ha-1, respectively. The on-site pollution was mainly due to IMI and CLO retention in topsoil (0-2 cm). A linear relation between the loss of IMI and CLO and the loss of runoff and sediment (p

Extreme rainfall is becoming more frequent and intense due to climate change and increasing human activity. Changes in rainfall will increase the risks and uncertainty of water erosion from orchards. Groundcover management is an important factor affecting hydrological and erosive processes of orchards. However, the roles of different groundcover managements for controlling soil water erosion caused by extreme rainfall are not fully understood. In this study, four groundcover managements (i.e., control bare land, orchard with no cover, orchard with grass cover and orchard with cover crops) were used to analyze runoff and erosion characteristics related to extreme and ordinary rainfalls in the red soil region of China. Based on the rainfall-runoff data measured on runoff plots, 356 natural rainfall events from 2001 to 2012 were analyzed, and rainfall depths and maximum 30-min rainfall intensities were used to select extreme rainfall events, according to the criteria of the World Meteorological Organization for extreme rainfall. There were 25 extreme rainfall events with a probability of 7.0% during the study period. Extreme rainfall events played the destructive role in inducing soil loss. The average extreme runoff coefficient and soil loss amount were 2.8 and 11.1 times higher than the values from ordinary rainfall, respectively. The contribution of extreme rainfall to sediment yield was 44.2%, which far higher than those to runoff generation (15.8%). Moreover, the runoff coefficients and erosion amounts varied greatly among the different groundcover managements under extreme rainfall, and were ranked in the following order: control bare land > orchard with no cover > orchard with cover crops > orchard with grass cover. However, the highest percentage of soil loss amount caused by extreme rainfalls were 59.0% in orchard with cover crops, and the lowest in orchard with grass cover (26.0%). Therefore, extreme rainfall should cause more attention in improving the groundcover management strategy for orchard soil erosion control, and groundcover like grass cover was extremely needed even after interplanting crops for the serious soil loss.

The traditional hydrology method, stable hydrogen and oxygen isotope technology, and rainfall simulation method were combined to investigate the hydrological function of small experimental plots (2 m×1.2 m) of contrasting architecture in Northwest Guangxi dolomite area. There were four typical catenary soils along the dolomite peak-cluster slope, which were the whole-sand, up-loam and down-sand, the whole loam, up-clay and down-sand soil types, respectively. All the experimental plots generated little amounts of overland runoff and had a high surface infiltration rate, ranging from 41 to 48 mm·h综合运用传统水文学、氢氧稳定同位素和原位模拟降雨等方法,研究了桂西北白云岩坡地典型土体构型石灰土微区(2 m×1.2 m)土壤水文功能.白云岩峰丛坡地沿坡向下表现出通体砂、上壤下砂、通体壤、上黏下砂4种典型石灰土土链格局,所有构型土体微区地表稳定入渗速率高达41～48 mm·h

This paper aims to use the composite fingerprinting method to reconstruct the environmental history after the Grain-for-Green Project and to provide effective sediment management and soil erosion-control strategies. This study used a composite fingerprinting method based on 45 geochemical properties and a mixing model to investigate sediment core changes in the sediment source in an agricultural catchment with little native vegetation. The samples consisted of 77 source samples (i.e., gully, grassland, forest, cropland, and fallow land) and five sediment cores. Genetic algorithm (GA) optimization has been recently used to find the best optimum source contribution to sediments. The results demonstrate that gully is the main sediment source in this catchment, constituting 34.7 %, followed by cropland (28.2 %), forest (21.5 %), grassland (12.7 %), and fallow land (2.9 %). However, the relative contribution of each source type was variable in all five sediment cores. The sediment that derived from grassland was relatively stable in the five cores. The relative contribution of forest was higher in the downstream portion of the check dam and lower in the upstream portion and gradually increased in the direction of the runoff pathway. As the forest matured, the sediment that derived from the forest gradually decreased. Changes in the hydro-ecological environment would lead to the leaf litter and understory being poorly developed and the soil being bare in the forest, making it more vulnerable to erosion. Reforestation and fallow are the key ecological strategies for reducing soil erosion. However, at the beginning of the Grain-for-Green Project, the young forest contributed 21.5 % of the sediment, indicating that natural fallow may be a better-designed sediment management and soil erosion-control strategy.

Knowledge of soil moisture dynamics within soil profiles can improve hydrologic process modeling. The heterogeneity of environmental factors (e.g., topographic attributes, soil properties, land use types and antecedent precipitation) influences soil moisture dynamics. In this study, we measured the soil moisture content at 10-cm depth intervals between the soil surface and the soil–bedrock boundary at 39 sites within a small catchment in 2008 and 2009; we primarily analyzed the environmental controls on soil moisture dynamics over the entire soil profile (0–10, 10–20, 20–30, 30–40, 40–50, and > 50 cm). Soil properties, terrain attributes, land use types and precipitation were evaluated as the potential controls. Given the high co-dependence of environmental factors, partial least squares regression (PLSR) was used to understand the links between soil moisture dynamics and the 14 selected environmental indices. The results show that PLSR considerably reduced the dimensionality of the soil moisture data. The first two components explained 73.5%, 69.9%, 65.7%, 60.1%, 51.5% and 47.6% of the soil moisture variability at the measurement depths of 0–10 cm, 10–20 cm, 20–30 cm, 30–40 cm, 40–50 cm and > 50 cm, respectively. Antecedent precipitation dominates the first and second positive components of the PLSR models at all of the measurement depths. The first-order controls were determined by calculating the Variable Influence on Projection (VIP). The VIP values revealed that the soil moisture in the top layer (0–10 cm) is controlled by antecedent precipitation. In the intermediate layer (10–30 cm), the first-order controls of soil moisture are antecedent precipitation, soil thickness, topographic wetness index, and soil texture; while in the deeper layer (> 30 cm), antecedent precipitation, soil thickness, topographic wetness index, relative elevation and slope gradient control soil moisture dynamics. The results indicate that soil moisture is closely associated with environmental factors in measurement catchments at various soil layers and provide a deeper understanding of hydrological processes.

Summary Understanding and quantifying sediment loads is important in watersheds with highly erodible materials, which will eventually cause environmental and ecological problems. Within this context, suspended sediment ( SS ) transport and its temporal dynamics were studied in a small mountainous watershed with sloping lands containing rock fragments in subtropical China. Soils containing rock fragments with many macro-pores have a high permeability rate. Over a 7-year period, the mean runoff coefficient of this watershed was 0.65. Overall, 30 flood events were monitored and accounted for 95.5%, 27.3%, 17.1% of the total SS load, precipitation and total discharge, respectively, over a 5-year period. The presence of rock fragments in soils can affect soil loss. When comparing the soil loss in the studied watershed with that of other watersheds under similar climatic conditions, rock fragments negatively affect soil loss. However, an extreme event occurred on 14 August 1990, and the sediment load exhibited a phenomenon called “small deposits towards lump withdrawal”, which resulted in a soil loss of 20,499 t (4.6 times the mean yearly soil loss). This event exhausted most of the SS s stored by the rock fragments on the slope and channel. Following this event, the mean SS concentration ( SSC ) of the 11 events was 1.05 kg m −3 , and the mean SSC of the 18 previous events was 1.75 kg m −3 . Twelve variables were separated using the classical hydrograph separation method. Partial least-squares regression (PLSR) was used to determine the highly co-related variables of the discharge. The results indicated that PLSR could explain runoff well. The relationship between discharge and SSC was highly scattered. During 24 flood events, three types of hysteresis loops were observed: clockwise (17 events), figure-eight (3 events), and complex (4 events).

Multivariate statistics are commonly used to identify the factors that control the dynamics of runoff or sediment yields during hydrological processes. However, one issue with the use of conventional statistical methods to address relationships between variables and runoff or sediment yield is multicollinearity. The main objectives of this study were to apply a method for effectively identifying runoff and sediment control factors during hydrological processes and apply that method to a case study. The method combines the clustering approach and partial least squares regression (PLSR) models. The case study was conducted in a mountainous watershed in the Three Gorges Area. A total of 29 flood events in three hydrological years in areas with different land uses were obtained. In total, fourteen related variables were separated from hydrographs using the classical hydrograph separation method. Twenty-nine rainfall events were classified into two rainfall regimes (heavy Rainfall Regime I and moderate Rainfall Regime II) based on rainfall characteristics and K-means clustering. Four separate PLSR models were constructed to identify the main variables that control runoff and sediment yield for the two rainfall regimes. For Rainfall Regime I, the dominant first-order factors affecting the changes in sediment yield in our study were all of the four rainfall-related variables, flood peak discharge, maximum flood suspended sediment concentration, runoff, and the percentages of forest and farmland. For Rainfall Regime II, antecedent condition-related variables have more effects on both runoff and sediment yield than in Rainfall Regime I. The results suggest that the different control factors of the two rainfall regimes are determined by the rainfall characteristics and thus different runoff mechanisms.

Soil erosion is one of the most serious environmental problems affecting the quality of soil, land, and water resources upon which humans depend for their sustenance. A soil erosion hazard map is highly useful for environmental planning, soil conservation and management in soil erosion prone areas. In order to assess the soil erosion hazard, remote sensing (RS) and geographical information system (GIS) technologies were adopted, and a numerical model was developed using spatial principal component analysis (SPCA). Here, an integrated soil erosion hazard index (SEHI) was computed and classified into four levels of soil erosion hazard viz. low, average, high, and very high. In the process, nine factors were selected together with the degree of importance of the factors in hazard of the soil erosion. Integrated RS and GIS techniques and models were applied to generate the necessary factors for the SPCA approach. In addition, erosion hazard and calculated rate of soil erosion were used to find out the risk of soil erosion. Soil erosion risk was identified as actual occurrence and potential risk. Results show that, in general, an average hazardous condition of soil erosion was found in the area. The potential risk was more extensive in terms of involved area compared to the actual occurrence, and both actual occurrence and potential risk were higher at the mid-level elevation of the area. This study highlighted that priority should be given where the actual occurrence is high to very high and the probability of potential risk is average to high for protecting the land at present and in the future as well. Therefore, the application of SPCA combined with RS and GIS provided an effective methodology to solve the complex decisional problem for soil erosion hazard and risk assessment.

Summary Landscape characteristics are widely accepted as strongly influencing stream water quality in heterogeneous watersheds. Understanding the relationships between landscape and specific water contaminant can greatly improve the predictability of potential contamination and the assessment of contaminant export. In this work, we examined the combined effects of watershed complexity, in terms of land use and physiography, on specific water contaminant across watersheds close to the Danjiangkou Reservoir. The land use composition, land use pattern, morphometric variables and soil properties were calculated at the watershed scale and considered potential factors of influence. Due to high co-dependence of these watershed characteristics, partial least squares regression was used to elucidate the linkages between some specific water contaminants and the 16 selected watershed characteristic variables. Water contaminant maps revealed spatial and seasonal heterogeneity. The dissolved oxygen values in the dry season were higher than those in the wet season, whereas the other contaminant concentrations displayed the opposite trend. The studied watersheds which are influenced strongly by urbanization, showed higher levels of ammonia nitrogen, total phosphorus, potassium permanganate index and petroleum, and lower levels of dissolved oxygen. The urban land use, largest patch index and the hypsometric integral were the dominant factors affecting specific water contaminant.

The geomorphic characteristics of a watershed affect the energy fluxes, mass movement, and water and sediment dispersion within the watershed. This paper examines how watershed complexity affects sediment yield in terms of rainfall and geomorphic characteristics. The geomorphic characteristics include primary, secondary and compound topographic attributes; watershed shape characteristics; relief parameters; and stream network characteristics. Because of the high co-dependence among these characteristics, partial least-squares regression (PLSR) was used to identify the relationships between the sediment yield and 29 selected watershed characteristics. The PLSR combines the features of a principal component analysis and multiple linear regression and is a robust multivariate regression method that is appropriate when the predictors exhibit multiple co-linearity. The first-order factors were determined by calculating the variable importance for the projection (VIP). Those variables with high VIP values are the most relevant for explaining the dependent variable. The results showed that the watershed shape and relief parameters have large influences on the sediment yield. The VIP values revealed that the sediment yield is primarily controlled by the plan curvature (VIP = 1.87) and the highest order channel length (VIP = 1.53), followed by the hypsometric integral (VIP = 1.49), rainfall (VIP = 1.44), basin relief (VIP = 1.19), slope (VIP = 1.15), sediment transport capacity index (VIP = 1.13), length ratio (VIP = 1.06), profile curvature (VIP = 1.01) and divide average relief (VIP = 1.00). This paper quantified the effects and relative importance of different geomorphic attributes on sediment yield. The insight provided by these results can be used in the selection of appropriate geomorphic variables for watershed erosion and hydrological models. Thus, this study is intended to elucidate the internal dynamics of sediment transport and storage in a watershed and provide a guide for watershed management.

Sediment and turbidity are primary causes of impaired river ecosystems; remedial action for these impairments requires the identification of their sources and controlling factors. This paper examines the combined effects of watershed complexity in terms of land use and physiography on the specific sediment yield of the upper Du River watershed (8973 km2) in China. The land use composition, land use pattern, morphometric variables, and soil properties of the watershed were calculated at the subwatershed scale and considered to be potentially influential factors. Because these watershed characteristics are highly codependent, a partial least-squares regression (PLSR) was used to elucidate the linkages between the specific sediment yield and metrics composed of 19 selected watershed characteristics. The first-order factors were identified by calculating the variable importance for the projection (VIP). The results revealed that the land use composition and land use pattern exerted the largest effects on the specific sediment yield and explained 65.2% of the variation in the specific sediment yield. A set of physiographic indices was also found to have a large effect on the specific sediment yield and explained 17.7% of the observed variation in the specific sediment yield. The following are the dominant first-order factors of the specific sediment yield at the subwatershed scale: the areal percentages of agriculture and forest, patch density, value of the Shannon's diversity index, contagion, value of the hypsometric integral, and saturated soil hydraulic conductivity. The watershed size exerted a substantial effect on the sediment delivery ratio (SDR). The VIP values also suggested that the Shannon's diversity index, contagion, and hypsometric integral are important factors in the SDR. With a readily available digital spatial database and rapid developments in geographic information system (GIS) technology, this practical and simple PLSR approach could be applied to a variety of watersheds.

Moisture variation in upper soil layers following a precipitation event is recognized as an important process in the conversion of rainfall to soil water. To understand the variability of topsoil moisture following a period of precipitation, we explored changes in soil moisture after 12 precipitation events at 38 sampling locations in the Wulongchi catchment (1.92 km 2 ) located in the Danjiangkou Reservoir Area, China. Our results indicated that variation in soil moisture increased with decreasing soil moisture following each of the 12 precipitation events. Using a two-way indicator species analysis (TWINSPAN), the soil moisture measurement periods following each event were classified into three discrete groups: humid periods, moderate periods, and dry periods. For the humid periods, the total variance of soil moisture that could be explained by redundancy analysis was more consistent than those of the moderate and dry periods. For the humid periods on the first axis, soil thickness was the main significant factor, and the topographic wetness index, slope position and aspect were also significant environmental factors. For the moderate and dry periods, however, the effect of aspect on soil moisture decreased, whereas the effect of relative elevation increased. Each environmental factor that was tested for its influence on topsoil moisture variation presented different effects throughout all three of the studied soil moisture periods.

The potential of rain to generate soil erosion is known as the rainfall erosivity (R), and its estimation is fundamental for a better understanding of the erosive ability of certain rainfall events. In this paper, we investigated the temporal variations of rainfall erosivity using common daily rainfall data from four meteorological stations during 1956 to 1989 and 2008 to 2010 periods in the Yanhe River catchment of the Chinese Loess Plateau. The adaptability of several simplified calculation models for R was evaluated and compared with the results of previous studies. An exponential model based on the modified Fournier index (MFI) was considered as the optimum for our study area. By considering the monthly distribution and coefficient of variation of annual precipitation, equations based on two indices, the MFI and its modification F F , produced a higher calculation accuracy than mean annual precipitation. The rainfall erosivity in the Yanhe River catchment has a remarkable interannual difference, with a seasonality index ranging from 0.69 to 1.05 and a precipitation concentration index from 14.51 to 27.46. In addition to the annual rainfall amounts, the extreme wave of monthly rainfall distribution also has an effect on the magnitude and temporal variation of rainfall erosivity, especially interannual variation. For long time series of rainfall erosivity, a trend coefficient r of −0.07 indicated a slight decline in erosivity in the Yanhe River catchment from 1956 to 2010.

Summary Rainfall kinetic energy (KE) can break down aggregates in the soil surface. A better understanding of sediment sorting associated with various KEs is essential for the development and verification of soil erosion models. A clay loam soil was used in the experiments. Six KEs were obtained (76, 90, 105, 160, 270, and 518 J m−2 h−1) by covering wire screens located above the soil surface with different apertures to change the size of raindrops falling on the soil surface, while maintaining the same rainfall intensity (90 ± 3.5 mm h−1). For each rainfall simulation, runoff and sediment were collected at 3-min intervals to investigate the temporal variation of the sediment particle size distribution (PSD). Comparison of the sediment effective PSD (undispersed) and ultimate PSD (dispersed) was used to investigate the detachment and transport mechanisms involved in sediment mobilization. The effective–ultimate ratios of clay-sized particles were less than 1, whereas that of sand-sized particles were greater than 1, suggesting that these particles were transported as aggregates. Under higher KE, the effective–ultimate ratios were much closer to 1, indicating that sediments were more likely transported as primary particles at higher KE owing to an increased severity of aggregate disaggregation for the clay loam soil. The percentage of clay-sized particles and the relative importance of suspension–saltation increased with increasing KE when KE was greater than 105 J m−2 h−1, while decreased with increasing KE when KE was less than 105 J m−2 h−1. A KE of 105 J m−2 h−1 appeared to be a threshold level beyond which the disintegration of aggregates was severe and the influence of KE on erosion processes and sediment sorting may change. Results of this study demonstrate the need for considering KE-influenced sediment transport when predicting erosion.

Contour ridging is an effective soil conservation practice used throughout the world. Because of microtopographic relief on sloping land, rainwater concentrates in low areas along furrows where contouring failure can occur. To quantify the effects and interactions of factors that influence runoff and sediment yield induced by contouring failure, a total of 32 rainfall simulation experiments were conducted, with two microtopography indices (row grade, RG, and field slope, FS), two ridge geometry indices (ridge height, H , and ridge width, W ), and two levels of rainfall intensity (RI) arranged in an L 16 (2 5 ) orthogonal array with two replications. The results showed that all of the factors considered except for row grade exerted significant influences on runoff and sediment yield ( p = 0.01). Rainfall intensity was the most important factor for runoff, with a contribution of 68.1%, followed by ridge height, field slope, and ridge width. Field slope and rainfall interacted negatively, with a contribution of 5.4%, resulting in increased runoff with increasing field slope at lower rainfall intensities, while the opposite effect was observed at higher rainfall intensities. The negative interaction of ridge height and width and the positive interaction of field slope and ridge height also had significant effects on runoff. For sediment yield, the most important factor (21.4%) was ridge height, which had a negative effect. Rainfall intensity had less effect on sediment yield than on runoff, while the row grade and its interaction with ridge width had greater influences. The optimal combinations of factors for control of runoff were determined to be RG 1 , FS 1 , H 2 , and W 2 for lower rainfall intensity and RG 1 , FS 2 , H 2 , and W 2 for higher rainfall intensity, and the optimal combinations of factors for sediment yield conservation were determined to be RG 1 , FS 1 , H 1 , and W 2 , where in all cases, the subscripts 1 and 2 denote lower and higher factor levels.

Wind erosion is very pronounced in semiarid regions during late winter–early spring and has major impacts on regional desertification and agriculture. In order to identify the effects of freeze/thaw and soil moisture on wind erosion, wind tunnel experiments were conducted to compare wind erosion effects under various soil moisture gradients in frozen and thawed soil. The variation of surface soil moisture after wind erosion and the effective soil particle size distribution was tested to explain the differences. The results showed that surface soil moisture content decreased in thawed soil and increased in frozen soil after wind erosion. The mean weight diameter, which increased with increasing soil moisture, was smaller in thawed soil than in frozen soil. The wind-driven sediment flux of frozen and thawed soil both decreased with increasing moisture, owing to the heavier soil particle weight and stronger interparticle bonding forces. The critical soil moisture content for suppressing wind erosion was around 2.34% for frozen soil and around 2.61% for thawed soil. The wind-driven sediment flux of thawed soil was always larger than that of frozen soil at the same moisture content, but this difference became negligible at moisture contents above 3.38%. We may speculate that wind erosion will be more severe in the future because of the lower soil moisture content and fewer soil freezing days as a result of global warming.

Summary There are strong ties between land cover patterns and soil erosion and sediment yield in watersheds. The spatial configuration of land cover has recently become an important aspect of the study of geomorphological processes related to erosion within watersheds. Many studies have used multivariate regression techniques to explore the response of soil erosion and sediment yield to land cover patterns in watersheds. However, many landscape metrics are highly correlated and may result in redundancy, which violates the assumptions of a traditional least-squares approach, thus leading to singular solutions or otherwise biased parameter estimates and confidence intervals. Here, we investigated the landscape patterns within watersheds in the Upper Du River watershed (8973 km 2 ) in China and examined how the spatial patterns of land cover are related to the soil erosion and sediment yield of watersheds using hydrological modeling and partial least-squares regression (PLSR). The results indicate that the watershed soil erosion and sediment yield are closely associated with the land cover patterns. At the landscape level, landscape characteristics, such as Shannon’s diversity index (SHDI), aggregation index (AI), largest patch index (LPI), contagion (CONTAG), and patch cohesion index (COHESION), were identified as the primary metrics controlling the watershed soil erosion and sediment yield. The landscape characteristics in watersheds could account for as much as 65% and 74% of the variation in soil erosion and sediment yield, respectively. Greater interspersion and an increased number of patch land cover types may significantly accelerate soil erosion and increase sediment export. PLSR can be used to simply determine the relationships between land-cover patterns and watershed soil erosion and sediment yield, providing quantitative information to allow decision makers to make better choices regarding landscape planning. With readily available remote sensing data and rapid developments in geographic information system (GIS) technology, this practical and simple PLSR approach could be applied to a variety of other watersheds.

Land use and land cover change is a key driver of environmental change. To investigate the runoff and erosion responses to frequent land use change on the steep lands in the Three Gorges area, China, a rainfall simulation experiment was conducted in plots randomly selected at a Sloping Land Conversion Program site with three soil surface conditions: existing vegetation cover, vegetation removal, and freshly hoed. Simulated rainfall was applied at intensities of 60 (low), 90 (medium), and 120 mm h−1 (high) in each plot. The results indicated that vegetation removal and hoeing significantly changed runoff generation. The proportion of subsurface runoff in the total runoff decreased from 30.3% to 6.2% after vegetation removal. In the hoed plots, the subsurface runoff comprised 29.1% of the total runoff under low-intensity rainfall simulation and the proportion rapidly decreased with increasing rainfall intensity. Vegetation removal and tillage also significantly increased soil erosion. The average soil erosion rates from the vegetation removal and hoed plots were 3.0 and 10.2 times larger than that in the existing vegetation cover plots, respectively. These identified that both the runoff generation mechanism and soil erosion changed as a consequence of altering land use on steep lands. Thus, conservation practices with maximum vegetation cover and minimum tillage should be used to reduce surface runoff and soil erosion on steep lands.

Soil–water characteristics are necessary for water quality monitoring, solute migration and plant growth. Soil–water characteristic curve (SWCC) is a relationship between suction and water content or degree of saturation. However, little information is available concerning the impacts of grazing exclusion management on soil–water characteristics. Here, the soil–water characteristics of grasslands, which were excluded grazing for 5 (GE5) and 15 years (GE15), were studied. The saturated hydraulic conductivity (K s), SWCC, particle composition, field capacity and some other indexes were determined. Results showed that the clay content and K s of grassland soil were higher for GE15 than GE5. For both treatments, in low suction condition (≤100 kPa), the water holding capacity of 0–10 cm soil was the best. Water holding capacity of topsoil decreased gradually with the increasing of suction, and it reached the strongest when the suction reached 600 kPa. In all soil water suction, the water holding capacity of subsoil was the weakest. The van Genuchten expression was applicable for most of the samples, except 20–30 cm of GE5 and 10–20 cm of GE15. Dual porosity equation was applicable for all the samples. The soil–water capability and soil structure of which was fenced for 15 years is superior to that of 5 years. This study suggests that the enclosure management improved the soil structure and soil–water capability.

In the mountainous area of China, the rock fragments in top soils are often removed by farmers because of their side effects on tillage. To understand hydrological processes and to assess the risk of soil erosion in purple soil areas with sloping farmlands, this study investigated the effects of rock fragment cover on surface runoff, infiltration, subsurface runoff, and erosion under field conditions using a portable rainfall simulator. Experimental twin plots (two subplots, 1 m wide by 2 m long, 32% slope) with different rock fragment covers ranging from 0 to 42% were exposed to four rainfall intensities: 45.5 � 1.9 mm=h (I1), 60.3 � 3.6 mm=h (I2), 92.0 � 4.1 mm=h (I3), and 123.9 � 9.4 mm=h (I4). Surface runoff, subsurface runoff, soil moisture, and sediment were measured. The results show that the hydrological response was influenced by the rock fragment cover and rainfall intensity. The rate of surface runoff in bare soils was affected by the rainfall intensity and increased with increasing rainfall intensity. The values of the surface runoff rates in bare soils were lower than 70 mm=h under the I1, I2, and I3 rainfall intensities. As the rock fragment cover increased, the surface runoff rate decreased while the infiltration rate, the subsurface runoff rate and the deep percolation coefficient increased. On the other hand, when the surface runoff rates in the bare soils were higher than 70 mm=h (in rainfall of I4), there were no significant differences in these hydrological variables among soils with different rock fragment covers. The differences in hydrological variables among soils with varying rock fragment covers decreased with increasing rainfall intensity. The rock fragment cover determined the erosive response. As the rock fragment cover increased, the sediment concentration de- creased. The presence of surface rock fragments significantly reduces soil erosion, and the relationship between the soil erosion rate and rock fragment cover can be expressed by an exponential function with a high degree of reliability for different rainfall intensities. However, the effectiveness of the rock fragment cover in reducing soil erosion decreased with increasing rainfall intensity. DOI: 10.1061/(ASCE)HE .1943-5584.0000576. © 2013 American Society of Civil Engineers. CE Database subject headings: Runoff; Rainfall; Soil erosion; China; Rocks; Slopes; Agriculture; Hydrologic aspects. Author keywords: Surface runoff; Subsurface runoff; Deep percolation; Rainwater distribution; Soil erosion rate; Simulated rainfall; Field condition.

Summary Understanding how changes in individual land use types influence the dynamics of streamflow and sediment yield would greatly improve the predictability of the hydrological consequences of land use changes and could thus help stakeholders to make better decisions. Multivariate statistics are commonly used to compare individual land use types to control the dynamics of streamflow or sediment yields. However, one issue with the use of conventional statistical methods to address relationships between land use types and streamflow or sediment yield is multicollinearity. In this study, an integrated approach involving hydrological modelling and partial least squares regression (PLSR) was used to quantify the contributions of changes in individual land use types to changes in streamflow and sediment yield. In a case study, hydrological modelling was conducted using land use maps from four time periods (1978, 1987, 1999, and 2007) for the Upper Du watershed (8973 km 2 ) in China using the Soil and Water Assessment Tool (SWAT). Changes in streamflow and sediment yield across the two simulations conducted using the land use maps from 2007 to 1978 were found to be related to land use changes according to a PLSR, which was used to quantify the effect of this influence at the sub-basin scale. The major land use changes that affected streamflow in the studied catchment areas were related to changes in the farmland, forest and urban areas between 1978 and 2007; the corresponding regression coefficients were 0.232, −0.147 and 1.256, respectively, and the Variable Influence on Projection (VIP) was greater than 1. The dominant first-order factors affecting the changes in sediment yield in our study were: farmland (the VIP and regression coefficient were 1.762 and 14.343, respectively) and forest (the VIP and regression coefficient were 1.517 and −7.746, respectively). The PLSR methodology presented in this paper is beneficial and novel, as it partially eliminates the co-dependency of the variables and facilitates a more unbiased view of the contribution of the changes in individual land use types to changes in streamflow and sediment yield. This practicable and simple approach could be applied to a variety of other watersheds for which time-sequenced digital land use maps are available.

The complexity of the relationship between suspended sediment concentration (SSC) and river discharge (Q) remains a challenge for SSC prediction in hyperconcentrated rivers. In this study, the wavelet-artificial neural network model (WANN) was built to predict SSC in the Kuye River, a representative hyperconcentrated river in the middle Yellow River catchments of China. In the WANN model, the observed daily time series for Q and SSC of 2193 days (from 1967 to 1972) were decomposed into subseries at different scales using discrete wavelet analysis. Then, the effective subseries were selected to construct Q/SSC inputs to the feed-forward back-propagation artificial neural network (BP ANN) to predict SSC 1 day in advance (the time resolution of the observed data). The coefficient of determination (R2) and root-mean square error (RMSE) were adopted to evaluate the model's performance. The WANN model showed higher prediction accuracy (R2 = 0.846 and RMSE = 29.82) than the sediment rating curve (SRC) model (R2 = 0.537 and RMSE = 55.40) or the ANN model (R2 = 0.664 and RMSE = 43.13). The WANN model exhibited more robust performance than the SRC and ANN models, indicated by the appropriate values of error autocorrelation and input-error correlation. Negative values of predicted SSC occurred in ANN and in WANN models. By adjusting the negative values to zero, the WANN R2 was improved by 4.3% from 0.846 to 0.882. In general, the results illustrate that the WANN model better predicts SSC in a hyperconcentrated river setting, with highly nonlinear and nonstationary time series.

Mulching with vegetative residue is an effective soil conservation practice. A better understanding of sediment characteristics associated with various mulch rates would improve the use of this practice for soil conservation. An experiment was conducted to evaluate the effects of straw mulch on runoff, erosion, and the particle-size distribution (PSD) of eroded sediment. Straw mulch rates of 0, 15, 30, 50, 70, and 90% cover were tested using simulated rainfall. The effective PSD of sediment (undispersed) was compared with equivalent measurements of the same samples after dispersion (ultimate PSD) to investigate the detachment and transport mechanisms involved in sediment mobilization. The maximum stream occurred at a different time from the peak sediment concentration during rainstorms under low mulch rates, which indicated the predominance of supply-limited conditions. However, at higher mulch rates the erosion processes were typical of a transport-limited sediment regime. The ratio of the sediment transported as primary clay to the soil matrix clay content was always less than 1, meaning that most of the clay was eroded in the form of aggregates. Transport selectivity was reflected by the silt enrichment, and silt-sized particles were transported mainly as primary particles since their effective-ultimate ratio was close to 1. The enrichment ratios for the sand-sized fractions decreased from 0.98 to 0.38 with increased mulch rates, and effective-ultimate ratios for sand-sized particles were always greater than 1, indicating that most of these particles were predominantly aggregates of finer particles, especially at high mulch rates. The findings reported in this study have important implications for the assessment and modeling of interrill erosion processes.

This paper analyzes runoff and soil loss in relation to the rainfall regimes and land use changes in a small mountainous watershed in the Three Gorges Area (TGA) of China. Based on 10 years of rainfall measurements and K-means clustering, 152 rainfall events were classified into three rainfall regimes. The mean statistical features of different rainfall regimes display a marked difference. Rainfall Regime I is events of medium amounts (31.8 mm) and medium duration (1371 min). Rainfall Regime II is events with high amounts (54.0 mm), long duration (2548 min), and an infrequent occurrence. Rainfall Regime III is events of low amount (22.2 mm), short duration (494 min) and high frequency. Each rainfall regime results in differing levels of runoff and erosion and Rainfall Regime I causes the greatest proportion of accumulated discharge (368.7 mm) and soil loss (4283 t). In the different rainfall regimes, the values of the mean runoff coefficient and the mean sediment load were ordered as follows: Rainfall Regime II > Rainfall Regime I > Rainfall Regime III. These results suggest that greater attention should be paid to Rainfall Regimes I and II because they had the most erosive effect. In the Wangjiaqiao watershed, the changes in land use primarily affected the paddy fields, where the cropland decreased significantly and the forest and orchards increased by 9.9% and 7.7%, respectively, during 1995–2004. The ANOVA shows land use changes caused significant decreasing trends in the runoff coefficients (P Rainfall Regime > Rainfall Regime III. Rainfall characteristics are decisive for the relative importance of different storm runoff generation mechanisms. The land use changes in the study watershed have considerably decreased runoff and soil loss.

Summary Sediment size distribution greatly affects sediment transport and deposition. A better understanding of sediment sorting will improve understanding of erosion and sedimentation processes, which in turn will improve erosion modeling. To address this issue, a total of 12 rainfall simulation experiments were conducted in a 1 m by 5 m box with varying steep slopes (10°, 15°, 20° and 25°), and the simulated rainfall lasted for 1 h at a rate of 90 mm h−1. For each simulated event, runoff and sediment were sampled at 3-min intervals, which were performed to study in detail the temporal change in size distribution of the eroded materials. These data were used to interpret the real-time sequence of transport mechanisms acting in response to the simulated rainfall. Total soil loss is the sum of suspended, saltating and contact loads. The proportion of sediment

Controlling soil hydrological processes and combating soil erosion on agricultural fields are high priorities for crop production and ecological environments. In this research, the effect of land application of rock fragments on surface runoff, subsurface runoff, infiltration and soil loss on Regosols had been investigated under field conditions using a portable rainfall simulator. Experimental twin-plot (two subplot with 1-m wide by 2-m long, 32% slope) with different rock fragment covers ranging from 0–40% were exposed to three rainfall intensities, low (57 ± 2 mm h-1), medium (91 ± 3 mm h-1), and high (122 ± 5 mm h-1). Fifteen treatments were conducted in the experiments in duplicates. Surface runoff, subsurface runoff, soil moisture and sediment were measured. Results show that surface rock fragment cover retarded surface runoff, increased infiltration and subsurface runoff, and diminished sediment concentration and soil loss. The final infiltration rates, which increased with increasing rock fragment covers, were 16.5–42.7 mm h-1 at low intensity, 19.7–50.8 mm h-1 at medium intensity and 23.8–56.4 mm h-1 at high intensity. Compared with bare soils, surface runoff rates were reduced to 24%, 51% and 64% in soils with 40% rock fragment cover at low, medium and high intensity rainfall, respectively. Subsurface runoff volume was significantly higher and time to start subsurface runoff was shorter as surface rock fragment cover increased. Compared with bare soils, subsurface runoff coefficients increased to 3.5, 3.3 and 2.9 times in soils with 40% rock fragment cover at low, medium and high intensity, respectively. Sediment concentrations increased from 9.8 g L-1 for 40% rock fragment cover under high intensity rainfall to 51.5 g L-1 for bare soils with rain falling at high intensity. The erosion rate was 15 g m-2 min-1 for bare soils under low intensity. The relationship between rock fragment cover and soil loss ratio can be expressed by an exponential function with a high degree of reliability, regardless of rainfall intensities. Overall, results indicate the surface rock fragments can be influential in shaping hydrological processes and soil loss from sloping farmland.

Summary Soil erosion poses a serious problem for sustainable agriculture and the environment. Owing to long-term anthropic pressure including overuse and inappropriate development, soil erosion has become a serious issue in the Three Gorges Area (TGA), China. Recently, integrated small watershed management (ISWM) for soil conservation in the TGA was rapidly developed. This study was conducted to investigate the impact of ISWM on soil erosion and sediment delivery in the Wangjiaqiao watershed in the TGA. The WATEM/SEDEM distributed erosion and sediment transport model was used to evaluate the effectiveness of the ISWM project. The model was calibrated against long-term measured suspended sediments at the watershed outlet. Land use and conservation measures were mapped and analyzed for 1995 and 2005, paying particular attention to quantification of changes in soil erosion and sediment delivery due to ISWM. The results showed that a combination of decreased soil loss (from 18.5 t ha −1 y −1 in 1995 to 13.2 t ha −1 y −1 in 2005) and increased sediment deposition (from 7.7 to 12.4 t ha −1 y −1 ) has led to a strong decrease in sediment yield (from 8.4 to 3.9 t ha −1 y −1 ) and the sediment delivery ratio (from 0.454 to 0.295). The results of scenario analysis showed that soil conservation measures taken in fields effectively reduce on-site soil loss and sediment yield. However, off-site sediment control measures appear to be much less effective at reducing sediment yield. This diachronic comparison of soil erosion and sediment delivery revealed that ISWM is quite effective and efficient; therefore, it is the appropriate method to combat soil erosion in the TGA and similar areas.

The emerging interdisciplinary field of hydropedology promotes synergistic integration of pedology and hydrology to enhance the holistic study of soil–water interactions across space and time. Our study illustrated this integration, exemplifying from plot-scale hydrological processes investigations on the sloping lands with different soil thickness in the Three Gorges Area of China. Our aims were to deal with (i) the watershed scale soil thickness survey, soil profiles description and hydrological processes inferential analysis for the plots with typical soil thickness from the pedological perspective, and (ii) the identification of dominant hydrological processes of these plots based on hydrological monitoring under simulated rainfall (designed as 60 mm h−1) from the hydrological perspective. The main results can be summarised as follows: (i) soil thickness of the sloping lands exhibited a wide range of variability along hillslopes. Thin soils reflected weaker pedogenesis degrees and less intense human intervention as compared with thick soils. (ii) Deep percolation and subsurface flow were the dominant processes in thin soils, whereas in thick soils, surface flow, deep percolation and storage were the dominant processes. (iii) Regarding to the surface flow, the 23-cm plot was mainly in the form of saturated overland flow, whereas the other plots were Horton overland flow. As to the subsurface flow, both 23- and 31-cm plots mainly took the form of preferential flow, the 59- and 76-cm plots mainly in the form of matrix flow regardless of soil horizon and the 45-cm plot displayed mainly the matrix flow in the A horizon and mainly the preferential flow in the AC horizon. Our study suggested that, relative to its parent disciplines of both pedology and hydrology, hydropedology improved synergies between pedology and hydrology in the plot-scale hydrological processes investigations. Copyright © 2011 John Wiley & Sons, Ltd.

The assessment of soil erodibility to water erosion in the field is often expensive and time-consuming. This study was designed to reveal the effects of aggregate breakdown mechanisms on interrill erosion dynamics and develop an improved model for assessing interrill soil loss, which incorporated the soil aggregate stability tests as a substitute for the interrill erodibility parameter, from both disturbed and undisturbed samples for red soils in subtropical China. Six cultivated areas of sloping land with red soils were selected, and topsoil aggregate stability was analyzed using the Le Bissonnais method to determine the different disaggregation forces. Laboratory rainfall simulations were designed to distinguish the effects of slaking (at different wetting rates) and mechanical breakdown (with and without screening) on soil erosion characteristics. Field rainstorm simulations with medium and high rainfall intensities were conducted on runoff plots (2 m 1 m) with slope gradients varying from 10% to 20% for each soil type. A new instability index, Ka , which considers aggregate breakdown mechanisms in interrill erosion processes, was proposed based on the disturbed sample results. Ka showed a close relationship with erosion rates in both disturbed and undisturbed samples. Following from the results of undisturbed sample experiments, Ka was used as a substitute for the erodibility factor, and introduced into the WEPP model, establishing a new erosion predication formula for red soils which had a good correlation coefficient ( R 2 = 0.89**). This research made a good attempt at estimating the interrill erosion rate on the basis of aggregate stability from simple laboratory determinations. These results extend the validity of soil aggregation characterization as an appropriate indicator of soil susceptibility to interrill erosion in red soils from subtropical China. The formula based on the instability index, Ka , has the potential to improve the methodology used for assessing interrill erosion rates.