【正文】 cture was the Redcliffe MEL culvert pleted in 1960. Since about 150 structures were built in Eastern Australia with discharge capacities ranging from less than 2m3 s?1 to more than 800m3 s?1 . Several structures were observed operating at design ?ows and for ?oods larger than design. Inspections during and after ?ood events demonstrated a sound operation associated with little maintenance (Fig. 2b).While McKay (1971) outlined general guidelines, Professor Colin APELT stressed that a successful MEL design must follow closely two basic design concepts: streamlining of the ?ow and nearcritical ?ow conditions at design ?ow (Apelt, 1983). Both inlet and outlet must be streamlined to avoid signi?cant form losses. In one structure, separation was observed in the inlet associated with ?ow recirculation in the barrel (Cornwall St, Brisbane). The barrel invert is often lowered to increase the discharge capacity (Fig. 2a). MEL culverts are usually designed to operate at design ?ow with Fr = to and supercritical ?ow conditions must be avoided. This is particularly important in the outlet where separation must be averted as well. The successful operation of large MEL culverts for over 45 years has highlighted further practical considerations. An adequate drainage is essential to prevent water ponding in the barrel invert. Drainage channels must be preferred to drainage pipes. For example, the MEL waterway shown in Fig. 2 is equipped with a welldesigned drainage system (Fig. 2a bottom left). One issue has been a loss of expertise in MEL culvert design. In Brisbane, two culvert structures were adversely affected by the construction of a new busway 25 years later. As a result, a major arterial will be overtopped during a design ?ood (Marshall Rd, Brisbane). For pleteness, MEL culverts may be designed for nonzero af?ux. The design process is similar (. Chanson, 1999, 2004a). The MEL culvert design received strong interests in Canada, USA, and UK. For example, Lowe (1970), Loveless (1984), Federal Highway Administration (1985, p. 114), Cottman and McKay (1990). Two pertinent studies in Canada (Lowe, 1970) and UK (Loveless, 1984) demonstrated that MEL culverts can pass successfully ice and sediment load without clogging nor silting. These laboratory ?ndings were con?rmed by inspections of MEL culverts after major ?ood events demonstrating the absence of siltation. Stepped spillways for embankment damsIn the last four decades, the regain of interest for stepped spillways has been associated with the development of new construction and design techniques. An innovative design is the embankment overtopping protection system (ASCE, 1994。 Chanson, 2001). The downstream slope is typically reinforced with precast concrete blocks, conventional concrete or RCC placed in a stepped fashion (Fig. 3). At large ?ow rates, these structures operate in a skimming ?ow regime that is characterised by plicated hydrodynamic interactions between the main stream, the step cavity recirculation zones and the freesurface. Figure 3 Melton dam stepped spillway on 30 January 2000—Completed in 1916, the Melton dam was equipped in 1994 with a secondary stepped spillway (Qdes = 2, 800 m3/s, h = ).Experimental observations highlighted strong interactions between the freesurface and the ?ow turbulence (. Chanson and Toombes, 2002a。Yasuda and Chanson, 2003。 Gonzalez and Chanson, 2004). At the upstream end, the ?ow is nonaerated and the freesurface exhibits an undular pro?le in phase with the stepped invert pro?le. Freesurface instabilities are however observed and strong air–water mixing occurs downstream of the inception point of freesurface aeration. Detailed air–water ?ow measurements demonstrate large amounts of entrained air (Fig. 4). Figure 4 shows experimental data for one ?ow rate down a 16? stepped chute (1V:) illustrating longitudinal oscillations of air–water ?ow properties downstream of the inception point of freesurface aeration. Such oscillations were observed on both steep and ?at slopes (. Matos, 2000。 Chanson and Toombes, 2002b) and it is believed that the seesaw patterns result from strong interference between vortex shedding behind each step edge and freesurface. Cavity recirculation and ?uid exchange between cavities and main stream are very energetic and contribute to form drag. Energy considerations provide a relationship between cavity ejection frequency, form drag and energy dissipation. At uniform equilibrium, the head loss between adjacent step edges equals the step height, while the energy is dissipated in the recirculation cavity at a rate proportional to the ejection frequency Fej , the volume of ejected ?uid and the main ?ow velocity V. It yields:[Fej (h cos θ)]/ V ≈ f /5 (1) where f is the Darcy–Weisbach friction factor, h is the step height and θ is the chute slope (Chanson et al., 2002b). Figure 4 Longitudinal distributions of mean air contents Cmean, dimensionless air–water depth Y90/dc, clearwater depth d/dc, air–wate velocity V90/Vc and mean ?ow velocity Uw/Vc—Stepped chute: 16 slope, h = , dc/h = (Yasuda and Chanson, 2003).Observed longitudinal oscillations of depthaveraged ?ow properties as shown in Fig. 4 affect in turn physical modelling analyses and ?ow property estimates Flow resistance may be grossly underestimated or overestimated when calculated between two adjacent step edges. In Fig. 4, the friction slope between adjacent steps ranged between + + for an average value of Sf = corresponding to an average Darcy friction factor f = . The latter pares favourably with an analytical solution of the form drag generated by step cavity ?ows (Chanson et al., 2002b。 Gonzalez and Chanson, 2004).中文翻譯Hydraulic engineering in the 21st century:Where to? 21世紀(jì)的水利工程：在哪里呢？13th Arthur Ippen awardee, IAHR Member ABSTRAC摘要For centuries, hydraulic engineers were at the forefront of science. 幾個(gè)世紀(jì)以來，水利工程師們行進(jìn)于科學(xué)前沿。The l