[Objective] Production and construction activities, like digging, excavating, dumping and piling-up of earth, form a large number of engineering mounds, which are ordinarily complex in underlying and surface structure and poor in erosion resistance. The mounds are mainly in the form of flat tops and slopes, whereon, rill erosion is easily occurring on slope surface with scouring flow running down from the top of the mound. In the process of rill erosion, rill morphology is closely related to hydrodynamic characteristics, and the relationship tends to be affected by soil texture. So it is essential to explore the relationships of slope rill morphology with hydrodynamic parameters, sediment yield and soil texture of the engineering mound.[Method] To that end, a field scouring-erosion experiment was carried out on an engineering mound of red soil and another of lou soil. Changes in rill morphology parameters, including rill width, rill depth and width-depth ratio, were monitored on slopes 32° in gradient, with scouring water flowing down at 8, 12, 16 and 20 L·min^-1(designated as Treatment I, II, III and IV), separately.[Result] Results show:(1) In the first 24 min, rill width and rill depth on the slope of the two kinds of engineering mounds developed rapidly. With increasing volume of the scouring water (8-20 L·min^-1), both exhibited a rising trend; compared with Treatment I, Treatment II, III and IV was 0.8%, 19.3% and 27.2% higher in mean rill width, and 24.5%, 126.2% and 95.1% higher in mean rill depth, respectively, on the mound of red soil, and 3.7%, 62.4% and 108.2% higher and 18.2%, 87.9% and 95.5% higher, respectively, on the mound of lou soil. The relationships of rill width and depth with duration of scouring was statistically significant on the mound of red soil (R²>0.858, P<0.01), and of lou soil (R²>0.689, P<0.01), too; (2) Rill width-depth ratios on the two types of mounds both fluctuated during the initial period (0-24min) and then levelled off during the late period (24~45min); compared with Treatment I, Treatment III was the lowest in mean width/depth ratio on either mound, being 115% lower on the mound of red soil and 20% lower on the mound of lou soil; (3) The relationships of runoff shear force and power of rill flow on slope surface with width and depth of rill could be described with a power function (R²>0.602, P<0.01); runoff shear force was a parameter more suitable for describing development of rill depth on the mound of red soil, and runoff power was for describing development of rill width on the mound of lou soil; and (4) Extremely significant linear function relationships were observed of cumulative sediment yield of the slope with rill width and rill depth (R²>0.897, P<0.01); rill width contributed 84.9% and rill depth did 15.1% to the cumulative sediment yield on the mound of red slope, and 22.5% and 77.5%, respectively, on the mound of lou soil.[Conclusion] This study may provide a theoretical reference for research on rill erosion mechanism on engineering mounds different in texture.