Approximately 70% of the aquatic-based production of animals is fed aquaculture, whereby animals are provided with high-protein aquafeeds. Currently, aquafeeds are reliant on fish meal and fish oil sourced from wild-captured forage fish. However, increasing use of forage fish is unsustainable and, because an additional 37.4 million tons of aquafeeds will be required by 2025, alternative protein sources are needed. Beyond plantbased ingredients, fishery and aquaculture byproducts and insect meals have the greatest potential to supply the protein required by aquafeeds over the next 10-20 years. Food waste also has potential through the biotransformation and/or bioconversion of raw waste materials, whereas microbial and macroalgal biomass have limitations regarding their scalability and protein content, respectively. In this review, we describe the considerable scope for improved efficiency in fed aquaculture and discuss the development and optimization of alternative protein sources for aquafeeds to ensure a socially and environmentally sustainable future for the aquaculture industry.

Introduction 287How widespread is corallivory? 288What corals do corallivores eat? 288Facultative and obligate corallivory 288Diet and feeding preferences 292Spatial variation in dietary composition 296Impact of corallivores on the coral communities 297Influence of corals on corallivore abundance and condition 299Role of corallivory in coral reef ecosystems 301 AbstractFishes that feed from live corals (corallivores) are a conspicuous component of healthy coral reef environments. However, knowledge of the occurrence and ecological significance of this feeding mode is fragmentary. Historically, very few fish were considered capable of feeding from live coral, and those few that did were considered ecologically insignificant. More recently, the role of corallivores has been re-evaluated; published records document 128 corallivorous fish species from 11 different families, with 69 of these belonging to the family Chaetodontidae. Other families, including the Labridae, Tetraodontidae, Balistidae, Monacanthidae, Pomacentridae and Scaridae, all have between seven and ten coral-feeding species. Onethird of coral-feeding fishes feed almost exclusively on corals, with more than 80% of their diet based on coral. Corallivorous fish show distinct prey preferences and consume only a small subset of available corals, usually the genera Acropora, Pocillopora and Porites. This selective predation by corallivores can limit abundance and distribution of preferred corals. Chronic predation by corallivores may also exacerbate effects of coral disturbance (e.g. climate-induced coral bleaching), impeding reef recovery and causing further coral loss. Conversely, the cover of preferred corals can be a primary determinant of corallivore abundance and physiological condition. Owing to this close association, obligate corallivores invariably decline in response to loss of coral cover. Increased knowledge of the number of corallivores and their diets suggest that this feeding mode is more important to coral reef food webs than traditionally thought.

A previously published magnetohydrodynamical ͑MHD͒ model of error-field penetration in tokamak plasmas is extended to take drift-MHD physics into account. In particular, diamagnetic and semicollisional effects are both fully incorporated into the analysis. The new model is used to examine the scaling of the penetration threshold in ohmic tokamak plasmas.

A model for field error penetration is developed that includes nonresonant as well as the usual resonant field error effects. The nonresonant components cause a neoclassical toroidal viscous torque that tries to keep the plasma rotating at a rate comparable to the ion diamagnetic frequency. The new theory is used to examine resonant error-field penetration threshold scaling in ohmic tokamak plasmas. Compared to previous theoretical results, the plasma is found to be less susceptible to error-field penetration and locking, by a factor that depends on the nonresonant error-field amplitude.

A model for field-error penetration is developed that includes nonresonant as well as the usual resonant field-error effects. The nonresonant components cause a neoclassical toroidal viscous torque that keeps the plasma rotating at a rate comparable to the ion diamagnetic frequency. The new theory is used to examine resonant error-field penetration threshold scaling in Ohmic tokamak plasmas. Compared to previous theoretical results, we find the plasma is less susceptible to error-field penetration and locking, by a factor that depends on the nonresonant error-field amplitude.

Recent experiments on DIII-D ͓J. L. Luxon, Nucl. Fusion 42, 614 ͑2002͔͒ and National Spherical Torus Experiment ͑NSTX͒ ͓M. Ono et al., Nucl. Fusion 40, 557 ͑2000͔͒ have focused on investigating mechanisms of driving rotation in fusion plasmas. The so-called intrinsic rotation is generated by an effective torque, driven by residual stresses in the plasma, which appears to originate in the plasma edge. A clear scaling of this intrinsic drive with the H-mode pressure gradient is observed. Coupled with the experimentally inferred pinch of angular momentum, such an edge source is capable of producing sheared rotation profiles. Intrinsic drive is also possible directly in the core, although the physics mechanisms are much more complex. Another option which is being explored is the use of nonresonant magnetic fields for spinning the plasma. It is found beneficially that the torque from these fields can be enhanced at low rotation, which assists in spinning the plasma from rest, and offers increased resistance against plasma slowing.

Abstract. Non-linear reduced MHD modelling of the response of a toroidally rotating plasma on Resonant Magnetic Perturbations (RMPs) is presented for DIII-D and ITER typical parameter. The non-linear cylindrical reduced MHD (RMHD) code was adapted to take into account toroidal rotation and plasma braking mechanisms such as resonant braking (~jxB) and the Neoclassical Toroidal Viscosity (NTV) calculated for low collisionality regimes ("1/ν" and "ν"). It was demonstrated that magnetic flux perturbation can be effectively screened by toroidal plasma rotation. This screening is larger for stronger rotation (V tor ) and lower resistivity In present modelling the central islands are screened by rotation. The pedestal region (r>0.9) is expected to be ergodic both for DIII-D and ITER parameters. Characteristic time for island formation at zero rotation increases for lower resistivity and for the pedestal top (r~0.9) is roughly estimated ~50ms for DIII-D and ~1500ms for ITER The non-resonant helical harmonics ( / q m n ≠ ) do not produce magnetic islands, penetrate on Alfven-like time, are not screened by plasma rotation, but produce NTV. If the "1/ν" low collisionality NTV regime is dominant in ITER, as it is suggested by dedicated DIII-D experiments and modelling, a counter (with respect to the plasma current direction ) rotation is predicted for ITER.

Recent experiments using DIII-D's capability to vary the injected torque at constant power have focused on developing the physics basis for understanding rotation through the detailed study of momentum sources, sinks and transport. Non-resonant magnetic braking has generally been considered a sink of momentum; however, recent results from DIII-D suggest that it may also act as a source. The torque applied by the field depends on the rotation relative to a non-zero 'offset' rotation. Therefore, at low initial rotation, the application of non-resonant magnetic fields can actually result in a spin-up of the plasma. Direct evidence of the effect of reverse shear Alfvén eigenmodes on plasma rotation has been observed, which has been explained through a redistribution of the fast ions and subsequent modification to the neutral beam torque profile. An effective momentum source has been identified by varying the input torque from neutral beam injection at fixed β N , until the plasma rotation across the entire profile is essentially zero. This torque profile is largest near the edge, but is still non-negligible in the core, qualitatively consistent with models for a so-called 'residual stress'. Perturbative studies of the rotation using combinations of co-and counter-neutral beams have uncovered the existence of a momentum pinch in DIII-D H-mode plasmas, which is quantitatively similar to theoretical predictions resulting from consideration of low-k turbulence.