It has been more than two decades since the discovery of nanofluids-mixtures of common liquids and solid nanoparticles less than 100 nm in size. As a type of colloidal suspension, nanofluids are typically employed as heat transfer fluids due to their favorable thermal and fluid properties. There have been numerous numerical studies of nanofluids in recent years (more than 1000 in both 2016 and 2017, based on Scopus statistics). Due to the small size and large numbers of nanoparticles that interact with the surrounding fluid in nanofluid flows, it has been a major challenge to capture both the macro-scale and the nano-scale effects of these systems without incurring extraordinarily high computational costs.To help understand the state of the art in modeling nanofluids and to discuss the challenges that remain in this field, the present article reviews the latest developments in modeling of nanofluid flows and heat transfer with an emphasis on 3D simulations. In part I, a brief overview of nanofluids (fabrication, applications, and their achievable thermo-physical properties) will be presented first.Next, various forces that exist in particulate flows such as drag, lift (Magnus and Saffman), Brownian, thermophoretic, van der Waals, and electrostatic double layer forces and their significance in nanofluid flows are discussed. Afterwards, the main models used to calculate the thermophysical properties of nanofluids are reviewed. This will be followed with the description of the main physical models presented for nanofluid flows and heat transfer, from single-phase to Eulerian and Lagrangian two-phase models. In part II, various computational fluid dynamics (CFD) techniques will be presented. Next, the latest studies on 3D simulation of nanofluid flow in various regimes and configurations are reviewed. The present review is expected to be helpful for researchers working on numerical simulation of nanofluids and also for scholars who work on experimental aspects of nanofluids to understand the underlying physical phenomena occurring during their experiments.

The Nicoya Peninsula in Costa Rica is one of the few places on Earth where the seismically active plate interface of a subduction zone is directly overlaid by land rather than ocean. At this plate interface, large megathrust earthquakes with magnitudes greater than 7 occur approximately every 50 years. Such quakes occurred in 1853, 1900 and 1950, so another large earthquake had been anticipated 1,2 . Land-based Global Positioning System 3,4 (GPS) and seismic 5-7 measurements revealed a region where the plate interface was locked and hence accumulated seismic strain that could be released in future earthquakes. On 5 September 2012, the longanticipated Nicoya earthquake occurred in the heart of the previously identified locked patch. Here we report observations of coseismic deformation from GPS and geomorphic data along the Nicoya Peninsula and show that the magnitude 7.6 Nicoya earthquake ruptured the lateral and down-dip extent of the previously locked region of the plate interface. We also identify a previously locked part of the plate interface, located immediately offshore, that may not have slipped during the 2012 earthquake, where monitoring should continue. By pairing observations of the spatial extent of interseismic locking and subsequent coseismic rupture, we demonstrate the use of detailed near-field geodetic investigations during the late interseismic period for identifying future earthquake potential.The interface between convergent plates produces most of the world's largest earthquakes, threatening local inhabitants and global populations through destructive shaking and tsunami generation, as demonstrated by the recent 2011 M w 9.0 Tohoku-Oki and 2004 M w 9.15 Sumatra-Andaman earthquakes and tsunami. Owing to the significant societal impacts, geoscientists endeavour to understand the driving and locking mechanisms controlling subduction zone seismicity. The shallow earthquakegenerating portion of the subduction interface, hereafter referred to as the megathrust, is difficult to characterize because it is relatively inaccessible, spans great lengths of continental margins and requires detailed near-field observations primarily in the marine environment.

No abstract

No abstract

No abstract

Figure 1. Two topographic data sets combined to form digital elevation model for Costa Rica and Middle America Trench (viewed to the north). Offshore data from von Huene et al. (1995). Onland data from National Imagery and Mapping Agency of U.S. Department of Defense (30 arc second grid). Symbols: EB

Orthogonal subduction of bathymetrically rough oceanic lithosphere along the northwestern fl ank of the Cocos Ridge imprints a distinctive style of deformation on the overriding Costa Rican forearc. We divide the Costa Rican forearc into three 100-160-kmlong deformational domains based on the bathymetric roughness and thickness of the Cocos plate entering the Middle American Trench, the dip of the subducting plate, the variation in surface uplift rates of late Quaternary coastal deposits, and the orientations and types of faults deforming Paleogene and Neogene sedimentary rocks. In the ~100-km-long Nicoya domain, coastal deposits show localized surface uplift and arcward tilting above the downdip projections of the fossil trace of the Cocos-Nazca-Panama (CO-NZ-PA) triple junction and the Fisher seamount and ridge. In the ~120-kmlong central Pacifi c forearc domain between the Nicoya Peninsula and Quepos, shallower (~60°) subduction of seamounts and plateaus is accompanied by trench-perpendicular late Quaternary normal faults. Steeply dipping, northeast-striking, margin-perpendicular faults accommodate differential uplift associated with seamount subduction. Uplift and faulting differ between the segments of the forearc facing subducting seamounts and ridges. Inner forearc uplift along the seamount-dominated segment is greatest inboard of the largest furrows across the lower slope. Localized uplift and arcward tilting of coastal deposits is present adjacent to subducting seamounts. In contrast, inboard of the underthrusting aseismic Cocos Ridge, along the ~160-km-long Fila Costeña domain between Quepos and the Burica Peninsula, mesoscale fault populations record active shortening related to the ~100-km-long Fila Costeña fold-andthrust belt. The observed patterns of faulting and permanent uplift are best explained by crustal thickening. The uplifted terraces provide a fi rst-order estimate of permanent strain along the forearc in Costa Rica. The permanent strain recorded by uplift of these Quaternary surfaces exceeds the predicted rebound of stored elastic strain released during subduction-zone earthquakes.

No abstract