The correlation between soil loss and recurrence interval was best fitted by linear function on SSP and by polynomial function on LSP. Also, a higher correlation coefficient between rainfall recurrence interval and soil loss exists on SSP than on LSP. The correlation between rainfall and runoff follows the same pattern as the one between rainfall and

soil loss, though the former generally had higher correlation coefficients than the latter. Fu et al. (2011) summarized Ganetespib datasheet the studies on the relationship between soil loss and slope gradients into three categories: power functions (e.g., Zingg, 1940 and Musgrave, 1947); linear functions (e.g. McCool et al., 1987 and Liu et al., 1994); and polynomial functions (e.g. Wischmeier and Smith, 1978). Nevertheless, all of these studies have been limited to relatively gentle slopes. The following are the supplementary data to this article. To assess the relative contributions of storms with various recurrence intervals to total soil and water loss, we divided recurrence intervals into five categories: less than 1, 1–2, 2–5, 5–10 and greater than 10 years. Supplementary Table 5 listed the contributions

of each category of storms to total soil and water loss at different slope angles. On SSP, rainstorms with recurrence intervals less than 1 year contributed to an average of 9.6% of total runoff and 12.4% of total soil loss; storms with recurrence intervals greater than 2 years were responsible for 68.6% of total runoff and 69.2% of total soil loss; the single SCH772984 nmr largest rainstorm with a recurrence interval of 21.5 years contributed to 19.6% of total runoff and 31.5% of total soil loss. On LSP, storms with recurrence intervals less than one year next contributed to an average 25.4%

of total runoff and 24.8% of total soil loss; storms with recurrence intervals greater than 2 years were responsible for 66% of total runoff and 66. 1% of total soil loss; the single largest storm with a recurrence interval of 10 years produced 23.3% of total runoff and 32% of total soil loss. It is interesting to notice that the contributions of storms with recurrence intervals greater than 2 years to total runoff and soil loss were comparable between SSP and LSP. The following are the supplementary data to this article. The slope factor used in the USLE was calculated in Eq. (2) (Wischmeier and Smith, 1978): equation(2) S=65.42sinθ+4.56sinθ+0.0654S=65.4sin2θ+4.56sinθ+0.0654 The above equation was modified in RUSLE as following (McCool et al., 1987): equation(3) S=10.8sinθ+0.03, for   q<9%S=10.8sinθ+0.03, for   q<9% equation(4) Or S=16.8sinθ−0.50 for   q>9%Or S=16.8sinθ−0.50 for   q>9%Where S is slope factor and θ is slope angle in per cent. The S values calculated using the equations in USLE and RUSLE were compared with the scaled ratio based on the measured annual soil loss data on both SSP and LSP ( Fig. 7).