192
12 Effects of Gravity Erosion on Particle Size Distribution …
(Xu et al. 2017). In the headwaters of the 2150 km
2 Waipaoa River Catchment, an
area well known for its severe erosion in New Zealand, the amount of mass movement
accounted for 41% of the total sediment load (Marden et al. 2014).
Particle size is one of the important characteristics of suspended sediment, because
it can reflect the sediment source and erosion processes (Wendling et al. 2016; Woodward and Walling 2007; Xu 2002), affect the entrainment, transport and deposition
of soils (Walling et al. 2000; Haritashya et al. 2010), and can be used to infer the
contaminant sources (Slattery and Burt 1997; Smith and Owens 2014; Abarca et al.
2016). Sediment usually carries the signature of upstream disturbances in runoff and
erosion to downstream channels (Sutherland et al. 2002). Previous studies used particle size characteristics as a means to trace suspended sediment in the river systems
(Walling and Woodward 1995). Walling and Moorehead (1989) found that considerable variation existed in the particle size characteristics of sediment from different
rivers in response to variations in source material and other physiographic controls.
Jia et al. (2016) also demonstrated that particle size characteristics were useful in
determining sediment provenance in the Yellow River basin based on the sediment
deposits in the river system which mainly comprised coarse sand particles larger
than 0.05 mm. Soil erosion causes sediments to be detached from their source materials and transported as suspended particles (Sadeghi and Harchegani 2012). In fact,
soil erosion is a size-selective process. Fine particle-size fractions are selectively
removed during inter-rill erosion process (Asadi et al. 2011; Wang et al. 2008; Stone
and Walling 1997; Govers 1985), resulting in generally finer sediments compared
to the source soil (Issa et al. 2006; Hao et al. 2016). Rill flows transport a greater
proportion of larger particles as compared with interrill flows (Alberts et al. 1980).
Research on particle size distribution caused by mass movements has been limited and almost exclusively been done in the field after their occurrence. Crosta et al.
(2007) and Davies and McSaveney (2009) reported that a large amount of smallsize debris was generated by fragmentation and collision of the large-size debris
of landslides. Cochrane and Acharya (2011) found that the largest particle in the
runoff affected by shallow landslides was significantly smaller than that of the original material. However, it is unknown how particle size distribution changes during
rainfall events in which both gravity erosion and water erosion occur. Such processbased data are hard, if not impossible, to obtain under natural rainfall conditions due
to the unpredictable nature of gravity erosion in timing and location. Thus, experimental studies have become increasingly important in understanding the processes
and mechanisms of gravity erosion. Chorley (1964) identified three broad classes of
physical models in experiments, namely, segments of unscaled reality, scale models,
and analog models. The monitoring of segments of unscaled reality is a widely used
experimental approach, and it has a long history and has arrived at a good understanding of surficial erosion processes on hill slopes (Schumm et al. 1987). Xu et al.
(2015b), for example, conducted a series of experiments in the laboratory to test
the stability of slopes under different slope geometries and rainfall conditions, and
then performed a sensitivity analysis to quantitatively explore the triggering mechanisms of mass failure on the steep loess slopes. Nevertheless, with respect to particle
selectivity, previous rainfall simulation experimental studies are mostly confined to
12 Effects of Gravity Erosion on Particle Size Distribution …
(Xu et al. 2017). In the headwaters of the 2150 km
2 Waipaoa River Catchment, an
area well known for its severe erosion in New Zealand, the amount of mass movement
accounted for 41% of the total sediment load (Marden et al. 2014).
Particle size is one of the important characteristics of suspended sediment, because
it can reflect the sediment source and erosion processes (Wendling et al. 2016; Woodward and Walling 2007; Xu 2002), affect the entrainment, transport and deposition
of soils (Walling et al. 2000; Haritashya et al. 2010), and can be used to infer the
contaminant sources (Slattery and Burt 1997; Smith and Owens 2014; Abarca et al.
2016). Sediment usually carries the signature of upstream disturbances in runoff and
erosion to downstream channels (Sutherland et al. 2002). Previous studies used particle size characteristics as a means to trace suspended sediment in the river systems
(Walling and Woodward 1995). Walling and Moorehead (1989) found that considerable variation existed in the particle size characteristics of sediment from different
rivers in response to variations in source material and other physiographic controls.
Jia et al. (2016) also demonstrated that particle size characteristics were useful in
determining sediment provenance in the Yellow River basin based on the sediment
deposits in the river system which mainly comprised coarse sand particles larger
than 0.05 mm. Soil erosion causes sediments to be detached from their source materials and transported as suspended particles (Sadeghi and Harchegani 2012). In fact,
soil erosion is a size-selective process. Fine particle-size fractions are selectively
removed during inter-rill erosion process (Asadi et al. 2011; Wang et al. 2008; Stone
and Walling 1997; Govers 1985), resulting in generally finer sediments compared
to the source soil (Issa et al. 2006; Hao et al. 2016). Rill flows transport a greater
proportion of larger particles as compared with interrill flows (Alberts et al. 1980).
Research on particle size distribution caused by mass movements has been limited and almost exclusively been done in the field after their occurrence. Crosta et al.
(2007) and Davies and McSaveney (2009) reported that a large amount of smallsize debris was generated by fragmentation and collision of the large-size debris
of landslides. Cochrane and Acharya (2011) found that the largest particle in the
runoff affected by shallow landslides was significantly smaller than that of the original material. However, it is unknown how particle size distribution changes during
rainfall events in which both gravity erosion and water erosion occur. Such processbased data are hard, if not impossible, to obtain under natural rainfall conditions due
to the unpredictable nature of gravity erosion in timing and location. Thus, experimental studies have become increasingly important in understanding the processes
and mechanisms of gravity erosion. Chorley (1964) identified three broad classes of
physical models in experiments, namely, segments of unscaled reality, scale models,
and analog models. The monitoring of segments of unscaled reality is a widely used
experimental approach, and it has a long history and has arrived at a good understanding of surficial erosion processes on hill slopes (Schumm et al. 1987). Xu et al.
(2015b), for example, conducted a series of experiments in the laboratory to test
the stability of slopes under different slope geometries and rainfall conditions, and
then performed a sensitivity analysis to quantitatively explore the triggering mechanisms of mass failure on the steep loess slopes. Nevertheless, with respect to particle
selectivity, previous rainfall simulation experimental studies are mostly confined to