Influence of raindrop size on rainfall intensity, kinetic energy, and erosivity in a sub-humid tropical area: a case study in the northern highlands of Ethiopia

Rainfall kinetic energy (KE) is an important factor in soil erosion models. Particle detachment from the soil surface depends on raindrop KE. Therefore, it is essential to evaluate the drop-size distribution (DSD) and KE of rainfall to understand soil erosion and runoff generation. In this study, we used an optical disdrometer to investigate the influence of DSD on the intensity (I), KE, and erosivity of rainfall at Bahir Dar, in northwestern Ethiopia. We recorded 1-min rainfall observations during 42 events, with I ranging from 0.82 to 46.27mmh(-1). The median raindrop diameter (D-50), which ranged between 1.14 and 4.33mm, was significantly correlated with I (R-2=0.96; P<0.001). We developed indices of rainfall KE as a function of time (KEtime) and of the KE content (KEcon). The best-fit relationships between KEtime and I were equally strong: R-2=0.96 (P<0.001) for both a linear function and a polynomial function. KEcon and I were most strongly related for a logarithmic function (R-2=0.98; P<0.001), followed by power (R-2=0.95; P<0.001) and polynomial (R-2=0.93; P<0.001) functions. The KEcon measured at Bahir Dar ranged from 7.4 to 32.43Jm (-2)mm(-1), whereas KEtime ranged from 38.34 to 1992.64Jm(-2)h(-1) for the observed range of I. The potential erosivity of rainfall events was found to be well correlated to smaller rainfalls depths (R-2=0.60, P<0.05). Our results suggest that, though empirical models are easy to use since they require readily available rainfall data, KE rather than rainfall depth should be used to estimate erosivity in the study area and regions of northwestern Ethiopia with similar characteristics. Moreover, the reasons why different measuring methods in the same area and similar methods in different areas provide different kinetic energy results are analyzed and discussed.