In the case of the tarp treatment, Zhang et al concluded that steady state soil loss rates for section 1 during each 15 minute interval could be estimated from

, S1_0-15 = L_0-15 /A1 (4a)

S1_15-30 = (L₁₅ -₃₀ - L_0-15 ) / A1 (4b)

S1_30-45 = (L₃₀-₄₅ - L_15-30 ) / A1 (4c)

S1_45-60 = (L_45-60 -L_30-45 ) / A1 (4c)

where L is the average steady state delivery at the outlet (g min^-1) for the respective time interval, and A1is the projected area of section 1. Eqs 4a-4c provide estimates of soil loss rates (g m^-2 min^-1) that are consistent with the approach that generated the sediment discharges (g m^-1 min^-1) presented in Table 1. However, in analysing the data for the screen experiments, Zhang et al concluded that the steady state soil loss rates (g m^-2min^-1) from each section could be estimated from

S1_0-15 = L_0-15 / A1(5a)

S2_15-30 = (L₁₅ -₃₀ - L_0-15 ) / A2 (5b)

S3_30-45 = (L₃₀-₄₅ - L_15-30 ) / A3 (5c)

S4_45-60 = (L_45-60 -L_30-45 ) / A4 (5d)

where S1 , S2 , S3 , and S4 are the soil loss rates from sections 1, 2, 3, and 4 for the corresponding 15-minute interval in g min^-1 m^-2,, andA1 , A2 , A3 , and A4 are the projected areas of the respective sections. No theoretical or logical reason was given for Eq. 5. As noted above, the scheme shown in Figure 4 applies to both the tarp and screen experiments when all sections are exposed to rain, Logically, when comparable runoff rates from the screen and tarp experiments occur during the final 15 mins of each rainfall event the spatial variations in soil loss rates in the screen experiments should follow that same pattern as seen in the tarp experiments. However, as shown in Table 2, that does not occur when Eq. 5 is used to estimate the soil loss rates in the screen experiments. Although it is possible that flow detachment did occur in some of the Zhang at al experiments, the conclusion that situation flow detachment occurred in all the screen experiments and in all the tarp experiments on the 27 % slope and the tarp experiment with 120 mm h^-1 rain on the 18 % slope was based not only an inability to measure detachment by raindrop impact but on dubious data soil loss rates in the screen experiments.