Figures
Abstract
A large amount of soil loss is caused by a small number of extreme events that are mainly responsible for the time compression of geomorphic processes. The aim of this study was to analyze suspended sediment transport during extreme erosion events in a mountainous watershed. Field measurements were conducted in Wangjiaqiao, a small agricultural watershed (16.7 km2) in the Three Gorges Area (TGA) of China. Continuous records were used to analyze suspended sediment transport regimes and assess the sediment loads of 205 rainfall–runoff events during a period of 16 hydrological years (1989–2004). Extreme events were defined as the largest events, ranked in order of their absolute magnitude (representing the 95th percentile). Ten extreme erosion events from 205 erosion events, representing 83.8% of the total suspended sediment load, were selected for study. The results of canonical discriminant analysis indicated that extreme erosion events are characterized by high maximum flood-suspended sediment concentrations, high runoff coefficients, and high flood peak discharge, which could possibly be explained by the transport of deposited sediment within the stream bed during previous events or bank collapses.
Citation: Fang N-F, Shi Z-H, Yue B-J, Wang L (2013) The Characteristics of Extreme Erosion Events in a Small Mountainous Watershed. PLoS ONE 8(10): e76610. https://doi.org/10.1371/journal.pone.0076610
Editor: Vanesa Magar, Plymouth University, United Kingdom
Received: April 30, 2013; Accepted: August 27, 2013; Published: October 11, 2013
Copyright: © 2013 Fang et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: Financial support for this research was provided by the National Natural Science Foundation of China (41071190) and the Program for New Century Excellent Talents in University (NCET-10-0423). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Introduction
Soil erosion poses a serious problem for sustainable agriculture and the environment [1], [2]. The soil and water conservationist is typically more interested in flood or erosion damage caused by large events than in the damage caused by smaller, more common hydrological events. Many studies have clearly emphasized that large-magnitude, low-frequency events are assumed to be dominant with respect to soil erosion [3]–[9]. Estimation of the role of sediment transport produced by extreme events is necessary for the calculation of sediment yields from basins, as a single event may represent the transport of several “normal” years [10]. However, soil erosion is a temporally compressed process; the effect of one event must be isolated when designing erosion control technologies and conservation planning [11], [12].
Previous projects have documented the effects of extreme events on total soil erosion and sediment transport on catchment scales [13]–[17]. These studies have mainly focused on extreme rainfall events and have discussed erosion based on those events. Analysis of the relationships between suspended sediment transport and rainfall characteristics during erosion events can help in understanding the factors and processes that determine sediment responses [18], [19]. However, runoff and erosion processes are strongly affected by many other factors in addition to rainfall [20]–[22], and few systematic attempts have been made to distinguish extreme rainfall events and erosion events.
The Three Gorges Project (TGP) on the Yangtze River in China is the world’s largest hydropower project. Following construction of the Three Gorges Dam, many farmers resettled in surrounding mountain areas and cultivated marginal lands, which are mostly on steep slopes with soil of poor structure. The TGP is controversial for several reasons, including the likely impact of sedimentation on the operation and useful life of the reservoir [23], [24]. The TGA refers to the riparian counties along the Yangtze valley between Yichang and Chongqing (Fig. 1). This area is periodically impacted by catastrophic floods, and soil erosion is a major environmental problem [25]. High levels of suspended sediment may result in high sediment deposition rates, thereby reducing the useful life of the Three Gorges Reservoir. Thus, understanding the temporal variations of suspended sediment transport and sediment loads during extreme events is essential for future watershed management plans.
Photos are of an overall view, the main channel, and the outlet of the Wangjiaqiao watershed. Photos are of an overall view, the main channel, and the outlet of the Wangjiaqiao watershed during an extreme event.
In this study, we investigated the runoff and erosion processes of the Wangjiaqiao watershed, which is a representative watershed within the TGA [26], [27]. The main objectives of this work were to (1) characterize extreme erosion events and the hydrological and sediment responses in a small watershed in the TGA and (2) improve our understanding of the factors that control sediment transport patterns and loads during extreme erosion events.
Study Area and Methods
Study Area
The study was conducted in the Wangjiaqiao watershed (31°5′N–31°9′N, 110°40′E–110°43′E), which lies in Zigui County in Hubei Province, China. It is approximately 50 km northwest of the Three Gorges Dam and covers an area of 16.7 km2 (Fig. 1). Elevations within the watershed range from 184 to 1,180 m and slopes range from 2° to 58°, with an average slope of 23°. The parent materials of this area are mostly Cretaceous and Tertiary purple shale, sandy shale, and sandstone, which contain large quantities of iron and manganese oxides in specific mineralogical forms. Two main soil great groups occur in the study watershed: purple soils derived from purple sandy shale and paddy soils developed from the purple soil. According to the USDA Soil Taxonomy, purple soils and paddy soils are classified as entisols and aquepts, respectively. The climate is subtropical, with mean temperatures between 11°C and 18°C. The annual precipitation averages 1,016 mm, 70% of which occurs between May and September. Land use is mainly a function of elevation and topography. Remnant forest patches exist primarily on steep, inaccessible peaks and slopes. Little natural vegetation is observed, and most areas are covered by secondary vegetation under human influence. A field survey was carried out in 1995 using a topographic map (scale 1∶10,000) and aerial photos. The results indicated that forest covered 44.5% of the study area, whereas cropland covered 23.3% (389.4 ha) and paddy fields covered 19.8% (330.7 ha). The other land use types were relatively minor and consisted of shrubland (3.2%), orchards (4.2%), rural residential land (3.9%), and water bodies (0.7%). The main agricultural crops are rice (Oryza sativa L.), maize (Zea mays L.), and wheat (Triticum aestivum L.). The streams in this watershed have a pinnate drainage pattern, and the length of the main channel is approximately 6,500 m [26].
Field and Lab Methods
Suspended sediment yields represent the sum of the erosion produced by all active sources within a watershed, although yields cannot be used as reliable indicators of total hill slope erosion due to the difficulty of quantifying sediment storage and identifying the mixture of sources [3]. A set of instruments consisting of a continuous recording rain gauge, water-level stage recorder, and bottle-type silt sampler were used to record rainfall, stream flow, and sediment load, respectively. The water stage was measured every 15 min and transformed into discharge using a calibrated rating curve obtained from periodic flow measurements. Suspended sediment concentrations (SSCs) were determined by the gravimetric method. Water samples were vacuum filtered through a 0.45-µm filter, and the residue was oven dried at 105°C for 24 h. Suspended sediment samples were taken only during rainfall–runoff events, and more than 10 samples were required during each event based on the “Technical code of practice on water and soil conservation monitoring” [28]. In practice, samples were collected based on the variation of the discharge and the magnitude of the SSC. Generally, samples were collected frequently during events with high SSC values and were collected infrequently during events with relatively small SSC values. The weight of each dried sample of residue and the sample volume were used to determine the SSC (g m−3). The suspended sediment load was then calculated from the SSC and water discharge data. Watershed runoff and rainfall data have been collected since 1989. No specific permissions were required for these sampling activities because the location is not privately-owned or protected in any way and the field activities did not involve endangered or protected species.
Data Processes
Hydrograph separation was conducted for 205 events during the 1989–2004 period, for which we have fairly complete records of sediment concentrations or hydrographs. Floods were identified in cases in which the increase in stream discharge exceeded 1.5 times the base flow recorded at the beginning of the rainfall event [29], [30]. Runoff was separated between storm flow and base flow using the classical hydrograph separation method of Hewlett and Hibbert (1967) [31]. Some continuous events were excluded because they were hard to separate as “an event” using our separation method. Some other events with complete hydrograph records were included even though their SSC values were below the threshold for our monitoring method. Total suspended sediment load of an event is the accumulation of sediment yield during each time step. The combined SS load of the 205 events represented 68% of the total suspended sediment load during the study years.
Statistical Analysis
The flood events were characterized using several variables (Table 1).where RC, R, P, Q, and BF are the runoff coefficient, surface runoff, total discharge, precipitation, and base flow, respectively.
Where SYi, SSCi and Qi are the suspended sediment yield, Suspended sediment concentrations, and discharge during a time term i. TL, Qmean, and SSCmean are the total suspended sediment load, mean discharge, and mean flood-suspended sediment concentration, respectively.
After evaluating the descriptive statistics, canonical discriminant analysis (CDA) was performed with the SPSS13.0 statistical software package using the variables mentioned above.
Results
Events Characteristics and Arrangement
An extreme event was defined classically as a rare, low-probability event, typically in relation to the exceedance of certain threshold values (e.g., means, percentiles). For this study, extreme erosion events are events that cause large amounts of SS load; no quantitative definition, e.g., in kg, was used. A partial definition might refer to an SS amount of a different order of magnitude than that caused by normal events. The definition of an extreme event in terms of damage caused by the event is discussed in [32]. A total of 10 events (approximately representing the 95th percentile) were identified as extreme erosion events based on the SS load caused by those 10 events having been clearly greater than that caused by the other events.
Based on the ranking of SS loads, 205 events were classified into three regimes; extreme erosion events, normal events, and no-SS load events. Four variables were compared: P, D, I30, and TL (Table 2). An extreme erosion event has a destructive effect on the soil surface. The mean SS load of the extreme erosion events that occurred during the study period was almost 68 times that of normal events, and the mean P and mean I30 were 2.5 and 2.1 times greater, respectively. Non-SS load events had observable hydrograph processes with a mean P of 19.6 mm but did not produce SS loads or their SSCs were below the threshold of our monitoring method. These results emphasize the importance of the rainfall type as a major cause of erosion. From this standpoint, the rainfall depth is an important factor in the degree of soil erosion in the study area. Meanwhile, the variance of TL is much higher than the variance of P for both extreme erosion events and normal events, which confirms the complex nature of sediment response in the study area.
Characterization of Extreme Erosion Events
The greatest proportion of the SS load was produced by the extreme erosion events. This finding is illustrated in Figure 2, which illustrates the percentage of the accumulated sediment load as a function of the percentage of events. Extreme erosion events were responsible for 83.8% of the SS loads. For each extreme erosion event, the erosive characteristics of individual storms were evaluated. The I30 ranged from 6.7 to 24.7 mm h−1. Antecedent rainfall amounts were highly variable, ranging from 0 to 12.2, 0 to 33.2, and 0 to 15.8 mm of precipitation during the previous one-, five-, and seven-day periods, respectively. The maximum SS load was 33,987,157 kg on 7/6/1999, which was 4.2 times greater than the mean annual load during the study years (Fig. 3). This event was caused by a Qmax of 144 m3/s. The curve representing sediment concentration vs. discharge over time exhibited a clockwise hysteretic loop [33]. The maximum SS occurred on 30/6/1991 and was 62,138 g/m3. This flood load was 2.68 times greater than the mean annual load and was generated by the highest precipitation recorded during the study period (P = 153.7 mm), which created a flood peak discharge of 87.9 m3 s−1 and a total flood runoff depth of 84.8 mm. The minimum SS load of an extreme erosion event was 1,033,247 kg. The RC of extreme erosion events ranged from 0.94 to 0.36. The peak discharge ranged from 16.2 to 144 m3 s−1. The direct SSCmax ranged from 3,539 to 62,138 g m−3. I30 ranged from 6.7 to 24.7 mm h−1. The antecedent rainfall amounts were highly variable, ranging from 0 to 12.2, 0 to 33.2, and 0 to 15.8 mm of precipitation during the previous one-, five-, and seven-day periods, respectively.
Events were independently ranked from largest to smallest for sediment.
Large Cumulative Rainfalls
The study of extreme erosion events is almost always associated with the study of precipitation [34]. To examine the relationship between rainfall and extreme erosion events, we identified the 10 events with the highest P values as large cumulative rainfalls (using the same method that was used to identify extreme erosion events). We then compared large cumulative rainfalls with extreme erosion events and evaluated the descriptive statistics of the extreme erosion and rainfall events, extreme erosion events, and large cumulative rainfalls shown in Table 3. Some studies have found that the most large cumulative rainfalls do not necessarily produce extreme fluvial discharge or the maximum soil erosion [7], [35], [36]. Our results confirm that the most extreme erosion events were not necessarily produced by the most large cumulative rainfalls. Only three extreme erosion events were found to be associated with large cumulative rainfalls. Fig. 3 illustrates the distributions of extreme erosion events and large cumulative rainfalls. A rainfall of 53.4 mm caused extreme erosion (117,504 kg) on 2/8/1998, whereas a rainfall of 74.3 mm caused only 53,428 kg of SS load on 9/4/1994. Extreme erosion events were greater than 16.6% if ranked by P, and large cumulative rainfalls were greater than 50% if ranked by SS load (Fig. 4).
Canonical Discriminant Analysis
To identify factors (excluded TL and P) that might explain the measured hydrological and sedimentological responses of extreme events, we conducted discriminant analyses using the data shown in Table 3. We created two canonical discriminant functions that included eight of the input variables to separate the groups (Table 4 and Fig. 5). The first canonical discriminant function (f1) explained 73.7% of the variance (Table 4), was highly positively correlated with SSCmax (0.411), RC (0.409), and Qmax (0.393), and was more weakly correlated with I30 and AP1d. For this reason, f1 expresses increasing SSCmax, RC, and Qmax in the watershed at the time of the floods with higher function values. The second discriminant function (f2) correlated with the remaining parameters (−0.481 for AP7d and −0.417 for SSCmax), as expected, given the lower proportion of variance explained. In addition, f2 was found to increase with increasing SSCmax, although with low intensity. Table 3 also presents the discriminant function values at the centroids of the three event types. The distribution of the flood events in the two dimensions created by the functions is shown in Fig. 6 (discriminant function coefficients are shown in Table 4). The centroids of the extreme erosion and rainfall group were high for both functions (f1 = 2.783; f2 = 2.056). The centroids of the extreme erosion events had relatively small values for f2 but high values for f1. These events were generated for very high values of RC, SSCmax, and Qmax. The large cumulative rainfalls have very low values for f1 and values close to zero for f2 (0.408). That is, the large cumulative rainfalls were generated for particularly low values of SSCmax, RC, and Qmax but with no special antecedent rainfall characteristics.
Discussion
A comparison of the accumulated suspended sediment transport with the accumulated rainfall and the number of single events indicates that 83.3% of sediment transport occurred during 5% of the erosion events and was caused by 13% of the precipitation. Fig. 7 illustrates the importance of the SS loads of extreme events to the total accumulated SS load. Two sharp rises in the curve were caused by the events of 1991/06/30 and 1999/07/06. The soil thickness of the Wangjiaqiao watershed ranges predominantly from 30 to 50 cm. The most extreme erosion event caused a lot of soil loss in the watershed. In the TGA, the sediment delivery rate on a small watershed scale has been thought to be very low in “common events,” ranging from 0.1 to 0.4 [37], [38], but mean sediment delivery radios are inappropriate for characterizing extreme erosion events. A study by Zhang et al. [39] reported a long-term natural sediment delivery radios close to 1 for a 10.88-km2 watershed.
The results confirm the complexity of erosion processes and the relationships between TL and P in particular. The most cumulative rainfalls do not necessarily produce the most extreme erosion. Events of low rainfall depth and short duration typically cause very limited hydrological responses and almost no sediment transport. Other studies [17], [40] have suggested that sediment yield rates may change in response to changes in rainfall, including changes in both total precipitation and intensity.
The geomorphologic impacts of an extreme event are closely related to such factors as the erodibility of the parent material, topography, vegetation cover, and land use [41]–[43]. Moreover, soil erosion is largely determined by on-site sediment production and the connectivity of sediment sources and streams [44]. Lu and Higgitt [45] found that 60% of sediment is contributed from arable land in 32 catchments in the TGA. Slopes with gradients in excess of 30% comprise 76% of the area of the Wangjiaqiao watershed. Cultivated sloping lands are major contributors to sediment yield. Tillage activities are generally carried out between April and September. Soil is eroded and then transported to the stream networks [27]. The soil parent material of the watershed is predominantly purple sandy shale, and bedrock is typically exposed in the channel bed; thus, channel erosion is rare. Sediments stored in the channel and distributed within tributaries are transported after flood events with sufficient transport capacity [46]. Therefore, the extreme erosion events characterized by the largest values of SSCmax, RC, and Qmax could be understood to result from the transport of deposited sediment within the stream bed during previous events. The clockwise hysteretic loop of the largest erosion event confirms this explanation [47]. Moreover, bank collapse has been identified as an important source of extreme erosion. Investigations in the Wangjiaqiao watershed have indicated that bank collapse along the stream results in bank erosion that contributes to sediment transport, particularly during major floods of high intensity [48]. The soils of the watershed are mainly purple soils developed from purple sandstone with a content of rock fragments and are rich in macrospores [49]. The macrospores of shallow soils become saturated easily during large rainfall events, the soil bulk density increases rapidly, and the combination of poor drainage in the soil and high-intensity rainfall easily causes bank collapse [50].
Conclusions
This study investigated the important issue of extreme erosion events in a typical watershed in the TGA. These events, which occur primarily during the summer season, can result in serious soil loss and great damage. The results confirm the complex and heterogeneous nature of hydrological and sediment responses in the watershed. A total of 83.3% of the sediment transport occurred during 5% of the erosion events that occurred in the TGA and was caused by 13% of the precipitation that the TGA received. The maximum TL of an event was 4.2 times the mean annual load during the study period. Large rainfall depth does not necessarily produce large fluvial discharge or large soil erosion. Extreme erosion events were generated under very high RC, SSCmax, and Qmax. The results indicate that the use of event average values or mean index values may not be suitable for analyzing soil erosion processes in the TGA. The effects of a small number of events appear to determine the magnitudes of soil loss; catastrophism may be more suitable for explaining soil erosion processes. Although our work provides a considerable database of extreme sediment load during rainfall events, we provide a speculative explanation concerning extreme erosion events, subject to the constraints of the field methods. Further investigations should be performed to assess the sediment yield during extreme events, which is expected to be of significant value in environmental management and development of strategies to control sediment dynamics at the catchment scale.
Acknowledgments
We would like to acknowledge members of the Soil and Water Conservation Bureau of Zigui County in Hubei Province for assistance in the field and in the laboratory. We thank Cheng Jiang, who provided valuable help in data collection.
Author Contributions
Conceived and designed the experiments: NFF ZHS. Performed the experiments: NFF ZHS. Analyzed the data: NFF ZHS BJY LW. Contributed reagents/materials/analysis tools: NFF BJY LW. Wrote the paper: NFF ZHS.
References
- 1. Ng SL, Cai QG, Ding SW, Chau KC, Qin J (2008) Effects of contour hedgerows on water and soil conservation, crop productivity and nutrient budget for slope farmland in the Three Gorges Region (TGR) of China. Agroforestry Systems 74: 279–291.
- 2. Shi ZH, Chen LD, Fang NF, Qin DF, Cai CF (2009) Research on the SCS-CN initial abstraction ratio using rainfall-runoff event analysis in the Three Gorges Area, China. Catena 77: 1–7.
- 3. Estrany J, Garcia C, Batalla RJ (2010) Hydrological response of a small mediterranean agricultural catchment. Journal of Hydrology 380: 180–190.
- 4. Hooke JM, Mant JM (2000) Geomorphological impacts of a flood event on ephemeral channels in SE Spain. Geomorphology 34: 163–180.
- 5. Lana-Renault N, Regues D, Marti-Bono C, Begueria S, Latron J, et al. (2007) Temporal variability in the relationships between precipitation, discharge and suspended sediment concentration in a small Mediterranean mountain catchment. Nordic Hydrology 38: 139–150.
- 6. Mano V, Nemery J, Belleudy P, Poirel A (2009) Assessment of suspended sediment transport in four alpine watersheds (France): influence of the climatic regime. Hydrological Processes 23: 777–792.
- 7. Osterkamp WR, Friedman JM (2000) The disparity between extreme rainfall events and rare floods - with emphasis on the semi-arid American West. Hydrological Processes 14: 2817–2829.
- 8. Poesen JWA, Hooke JM (1997) Erosion, flooding and channel management in Mediterranean environments of southern Europe. Progress in Physical Geography 21: 157–199.
- 9. White S, Garcia-Ruiz JM (1998) Extreme erosional events and their role in mountain areas of northern Spain. Ambio 27: 300–305.
- 10.
Wolman MG, Miller JP (1960) Magnitude and frequency of forces in geomorphic processes. The Journal of Geology: 54–74.
- 11. Seeger M, Errea MP, Begueria S, Arnaez J, Marti C, et al. (2004) Catchment soil moisture and rainfall characteristics as determinant factors for discharge/suspended sediment hysteretic loops in a small headwater catchment in the Spanish pyrenees. Journal of Hydrology 288: 299–311.
- 12. Stroosnijder L (2005) Measurement of erosion: Is it possible? Catena 64: 162–173.
- 13. Gonzalez-Hidalgo JC, Batalla RJ, Cerda A, de Luis M (2012) A regional analysis of the effects of largest events on soil erosion. Catena 95: 85–90.
- 14. Larson WE, Lindstrom MJ, Schumacher TE (1997) The role of severe storms in soil erosion: A problem needing consideration. Journal of Soil and Water Conservation 52: 90–95.
- 15. Lenzi MA, Mao L, Comiti F (2003) Interannual variation of suspended sediment load and sediment yield in an alpine catchment. Hydrological Sciences Journal-Journal Des Sciences Hydrologiques 48: 899–915.
- 16. McBroom MW, Beasley RS, Chang M, Ice GG (2008) Water quality effects of clearcut harvesting and forest fertilization with best management practices. Journal of Environmental Quality 37: 114–124.
- 17. Nearing MA, Jetten V, Baffaut C, Cerdan O, Couturier A, et al. (2005) Modeling response of soil erosion and runoff to changes in precipitation and cover. Catena 61: 131–154.
- 18. Nadal-Romero E, Regues D, Latron J (2008) Relationships among rainfall, runoff, and suspended sediment in a small catchment with badlands. Catena 74: 127–136.
- 19. Zabaleta A, Martinez M, Uriarte JA, Antiguedad I (2007) Factors controlling suspended sediment yield during runoff events in small headwater catchments of the Basque Country. Catena 71: 179–190.
- 20. Chen LD, Wang J, Fu BJ, Qiu Y (2001) Land-use change in a small catchment of northern Loess Plateau, China. Agriculture Ecosystems & Environment 86: 163–172.
- 21. Kirkby MJ, Bracken LJ, Shannon J (2005) The influence of rainfall distribution and morphological factors on runoff delivery from dryland catchments in SE Spain. Catena 62: 136–156.
- 22. Siriwardena L, Finlayson BL, McMahon TA (2006) The impact of land use change on catchment hydrology in large catchments: The Comet River, Central Queensland, Australia. Journal of Hydrology 326: 199–214.
- 23. Chu ZX, Zhai SK (2008) Yangtze River sediment: In response to Three Gorges Reservoir (TGR) water impoundment in June 2003. Journal of Coastal Research 24: 30–39.
- 24. Dai ZJ, Chu A, Stive M, Zhang XL, Yan H (2011) Unusual Salinity Conditions in the Yangtze Estuary in 2006: Impacts of an Extreme Drought or of the Three Gorges Dam? Ambio 40: 496–505.
- 25.
Yang YS, Shi DM (1994) Study on soil erosion in the Three Gorge area of the Changjiang River. Southeast University Press, Nanjing (in Chinese).
- 26. Fang NF, Shi ZH, Li L, Jiang C (2011) Rainfall, runoff, and suspended sediment delivery relationships in a small agricultural watershed of the Three Gorges area, China. Geomorphology 135: 158–166.
- 27. Shi ZH, Cai CF, Ding SW, Wang TW, Chow TL (2004) Soil conservation planning at the small watershed level using RUSLE with GIS: a case study in the Three Gorge Area of China. Catena 55: 33–48.
- 28.
Ministry of Water Resources of the People’s Republic of China (MWRPRC) (2002) Technical code of practice on water and soil conservation monitoring. 7 p. (in Chinese).
- 29. Garcia-Ruiz JM, Arnaez J, Begueria S, Seeger M, Marti-Bono C, et al. (2005) Runoff generation in an intensively disturbed, abandoned farmland catchment, Central Spanish Pyrenees. Catena 59: 79–92.
- 30. Lana-Renault N, Regues D (2009) Seasonal patterns of suspended sediment transport in an abandoned farmland catchment in the Central Spanish Pyrenees. Earth Surface Processes and Landforms 34: 1291–1301.
- 31.
Hewlett JD, Hibbert AR (1967) Factors affecting the response of small watersheds to precipitation in humid areas. Forest hydrology: 275–290.
- 32. Bathurst JC, Amezaga J, Cisneros F, Novillo MG, Iroume A, et al. (2010) Forests and floods in Latin America: science, management, policy and the EPIC FORCE project. Water International 35: 114–131.
- 33. Williams GP (1989) Sediment concentration versus water discharge during single hydrologic events in rivers. Journal of Hydrology 111: 89–106.
- 34. Soler M, Regüés D, Latron J, Gallart F (2007) Frequency–magnitude relationships for precipitation, stream flow and sediment load events in a small Mediterranean basin (Vallcebre basin, Eastern Pyrenees). Catena 71: 164–171.
- 35. Gonzalez-Hidalgo JC, Pena-Monne JL, de Luis M (2007) A review of daily soil erosion in Western Mediterranean areas. Catena 71: 193–199.
- 36. Nunes JP, Vieira GN, Seixas J, Gonclaves P, Carvalhais N (2005) Evaluating the MEFIDIS model for runoff and soil erosion prediction during rainfall events. Catena 61: 210–228.
- 37. Shi DM (1992) Soil and water loss and its control in the Yangtze river basin. Resources and Environment in the Yangtze Valley 1: 62–70 (in Chinese)..
- 38. Tu EQ, Liao XY, Chen ZJ, Wang HM (2010) Effects of rainfall on runoff from a small watershed in three gorges reservoir area. Bulletin of Soil and Water Conservation 30: 7–11 (in Chinese)..
- 39. Zhang XB, Wen AB, Zhang YQ, Wang YC, Qi YQ, et al. (2006) Perliminary study on natural erosion rates in small watershed of Sichuan hilly basin. Joural of Soil and Water Conservation 20: 1–5 (in Chinese)..
- 40. Old GH, Leeks GJL, Packman JC, Smith BPG, Lewis S, et al. (2003) The impact of a convectional summer rainfall event on river flow and fine sediment transport in a highly urbanised catchment: Bradford, West Yorkshire. Science of the Total Environment 314: 495–512.
- 41. Coppus R, Imeson AC (2002) Extreme events controlling erosion and sediment transport in a semi-arid sub-andean valley. Earth Surface Processes and Landforms 27: 1365–1375.
- 42. Belmonte AMC, Beltran FS (2001) Flood events in Mediterranean ephemeral streams (ramblas) in Valencia region, Spain. Catena 45: 229–249.
- 43. Bull LJ, Kirkby MJ, Shannon J, Hooke JM (2000) The impact of rainstorms on floods in ephemeral channels in southeast Spain. Catena 38: 191–209.
- 44. Bakker MM, Govers G, van Doom A, Quetier F, Chouvardas D, et al. (2008) The response of soil erosion and sediment export to land-use change in four areas of Europe: The importance of landscape pattern. Geomorphology 98: 213–226.
- 45. Lu XX, Higgitt DL (2001) Sediment delivery to the Three Gorges 2: Local response. Geomorphology 41: 157–169.
- 46. Oeurng C, Sauvage S, Sanchez-Perez JM (2010) Dynamics of suspended sediment transport and yield in a large agricultural catchment, southwest France. Earth Surface Processes and Landforms 35: 1289–1301.
- 47. Klein M (1984) Anti clockwise hysteresis in suspended sediment concentration during individual storms: Holbeck Catchment; Yorkshire, England. Catena 11: 251–257.
- 48.
Si JJ (2009) Research of man–made accelerated erosion and farm households’ behavior of Water&Soil conservation in small watershed. Master dissertation of Huazhong agricultural university. China academic journal (cd-rom version) electronic magazine,Wuhan. (in Chinese).
- 49. Fu ZY, Li ZX, Cai CF, Shi ZH, Xu QX, et al. (2011) Soil thickness effect on hydrological and erosion characteristics under sloping lands: A hydropedological perspective. Geoderma 167–68: 41–53.
- 50. Wu CL, Zhang PC (2002) primary study on soil erosion factors of the wangjiaqiao watershed in three gorges reservoir area. Resources and Env ironment in the Yangtze Basin 11: 165–170 (in Chinese)..