Development of a GPGPU-parallelized hybrid finite-discrete element method for modeling rock fracture

The hybrid finite-discrete element method (FDEM) is widely used for engineering applications, which, however, is computationally expensive and needs further development, especially when rock fracture process is modeled. This study aims to further develop a sequential hybrid FDEM code formerly proposed by the authors and parallelize it using compute unified device architecture (CUDA) C/C++ on the basis of a general-purpose graphics processing unit (GPGPU) for rock engineering applications. Because the contact detection algorithm in the sequential code is not suitable for GPGPU parallelization, a different contact detection algorithm is implemented in the GPGPU-parallelized hybrid FDEM. Moreover, a number of new features are implemented in the hybrid FDEM code, including the local damping technique for efficient geostatic stress analysis, contact damping, contact friction, and the absorbing boundary. Then, a number of simulations with both quasi-static and dynamic loading conditions are conducted using the GPGPU-parallelized hybrid FDEM, and the obtained results are compared both quantitatively and qualitatively with those from either theoretical analysis or the literature to calibrate the implementations. Finally, the speed-up performance of the hybrid FDEM is discussed in terms of its performance on various GPGPU accelerators and a comparison with the sequential code, which reveals that the GPGPU-parallelized hybrid FDEM can run more than 128 times faster than the sequential code if it is run on appropriate GPGPU accelerators, such as the Quadro GP100. It is concluded that the GPGPU-parallelized hybrid FDEM developed in this study is a valuable and powerful numerical tool for rock engineering applications.     

Sensitivity analysis of a coastal marine ecosystem

An ecosystem with four compartments, i.e. PO₄-P, phytoplankton, zooplankton and detritus, in a coastal region was considered. Sensitivity analysis of a phosphorus flow model in well-mixed water without currents concludes that (1) the maximal photosynthetic rate,V _m , in Michaelis-Menten relation plays an important role in the distribution of biomass among compartments but the half saturation constant,K _s , is not so important, (2) the natural death rate of phytoplankton is important for the ecosystem, (3) the natural death rate of zooplankton is also effective on the ecosystem.A numerical experiment was also performed on the ecosystem with four compartments are also studied using a dynamical barotropic model of tidal currents of Mikawa Bay (Japan). Diffusion coefficient in diffusion equation plays the role of a linear smoothing parameter in the horizontal distribution of compartment. On the other hand, perturbations of biological parameters cause nonlinear variations in the horizontal distribution of compartment.     

Application of a 3-D chemical fate prediction model (FATE3D) to predict dioxin concentrations in the Tokyo Bay

A 3-D chemical fate prediction model (FATE3D) was applied to predict the dioxin concentrations in the seawater of Tokyo Bay, Japan. The simulations were carried out for a period of one year (from September 2002 to August 2003). Parameters such as meteorological data, flow field conditions, concentrations and sinking rates of organic particulate matter, initial and boundary conditions, and loading fluxes and physico-chemical properties of dioxins were used as the model inputs.The simulation results compared favorably with the field measurements of dioxin concentrations in the bay for both the particulate and dissolved phases, indicating the validity and predictive capability of the model. Furthermore, the differences in the seasonal cycles and distributions between the particulate- and dissolved-phase dioxins in the bay were estimated from the simulation results.However, the particulate-phase dioxin concentrations in the bottom layers (+1 m from the bottom) were underestimated, probably because the resuspension process was not taken into account in the model. The improvement of the model's predictive capability, including the resuspension process, shall be the focus of our next study.     

Sensitivity of CFCs and Anthropogenic CO2 Uptake in a North Pacific GCM to Mixing Parameterization and Surface Forcing

Distributions and characteristics of water mass and chlorofluorocarbons (CFCs) in the North Pacific are investigated by using a General Circulation Model (GCM). The anthropogenic CO₂ uptake by the ocean is estimated with velocity fields derived from the GCM experiments. The sensitivity of the uptake to different diffusion parameterizations and different surface forcing used in the GCM is investigated by conducting the three GCM experiments; the diffusive processes are parameterized by horizontal and vertical eddy diffusion which is used in many previous models (RUN1), parameterized by isopycnal diffusion (RUN2), and isopycnal diffusion and perpetual winter forcing for surface temperature and salinity (RUN3). Realistic features for water masses and CFCs can be simulated by the isopycnal diffusion models. The horizontal and vertical diffusion model fails to simulate the salinity minimum and realistic penetration of CFCs into the ocean. The depth of the salinity minimum layer is better simulated under the winter forcing. The results suggest that both isopycnal parameterization and winter forcing are crucial for the model water masses and CFCs simulations. The oceanic uptake of anthropogenic CO₂ in RUN3 is about 19.8 GtC in 1990, which is larger by about 10% than that in RUN1 with horizontal and vertical diffusive parameterization. RUN3 well simulates the realistic water mass structure of the intermediate layer considered as a candidate of oceanic sink for anthropogenic CO₂. The results suggest that the previous models with horizontal and vertical diffusive parameterization may give the oceanic uptake of anthropogenic CO₂ underestimated. This revised version was published online in August 2006 with corrections to the Cover Date.     

Recovery of surface albedo and plant cover after wildfire in a Picea mariana forest in interior Alaska

Albedo influences vegetation structure, permafrost thawing, etc., in particular, after wildfires in Picea mariana forests in Alaska, USA, while albedo changes with plant succession. To understand interactions between albedo and ecosystem recovery after wildfire, surface albedo was measured in the spring and summer of 2005 at Poker Flat, interior Alaska, where P. mariana forest was dominant. The ground surface was mostly covered with Sphagnum moss before the 2004 wildfire, and was variously burned by the fire. The measured wavelengths ranged from 0.3 to 3.0 μm. We measured four independent variables, incidence, plant cover on the forest floor, cover of burned ground surface, canopy openness and incidence, to examine the determinants on surface albedo. Multiple regression analysis showed that total plant cover positively and mostly determines albedo, indicating that plant recovery is prerequisite to return high albedo. When the ground surface was damaged by fire, changes in albedo were mostly derived from decrease in reflectance wavelengths between 0.7 and 1.4 μm. The fluctuations of reflectance wavelengths did not differ greatly between damaged-moss and burned surfaces. We must mention the dynamics of Sphagnum to understand various environmental changes including surface albedo.     

The paleolimnology of a small waterbody in the Pocono Mountains of Pennsylvania,USA: reconstructing 19th–20th century specific conductivity trends in relation to changing land use

Remains of scaled chrysophytes, magnetic minerals and pollen were used to analyze the recent paleolimnological history of a small lake, Lake Waynewood, in the Pocono Mountains of Pennsylvania. Important shifts in all three variables were observed over the 133 years represented by the core. The most significant changes occurred near the turn of the century when the watershed was heavily logged. Before the logging event, species such asSynura sphagnicola, S. spinosa, Mallomonas galeiformis andM. duerrschmidtiae were co-dominant members of the flora. Subsequent to the deforestation of the watershed other taxa, includingM. crassisquama, M. caudata andS. petersenii, increased in relative importance. Concurrent with changes in the scaled chrysophytes was a six-fold increase in the concentrations of magnetic material, presumably the result of increased erosion caused by the logging. Changes in pollen grains also correlate well with the onset of the deforestation event. The scaled chrysophyte inferred specific conductivity of the lake has more than tripled, with the primary increase occurring concurrent with the commencement of logging and the increase in magnetic mineral material. The effects of other human-related disturbances are also discussed.     

Environmental magnetic and magnetic fabric studies in LakeWaynewood, northeastern Pennsylvania, USA: Evidence for changes inwatershed dynamics

An environmental magnetic and magnetic fabric study of sedimentscollected from Lake Waynewood, a post-glacial lake in the PoconoMountains of Pennsylvania, USA, provided a history of the lake's watersheddynamics over the past 3500 years. Two 5 m long, Mackereth coresof lake sediments and three watershed soil profiles were analyzed magnetically.Paleosecular variation and ¹⁴C measurements allowed timing ofchanges in the lake's watershed which are documented by changes in ARM,SIRM,, S-ratio, SIRM/, ARM/ and ARM/SIRMdowncore. Prior to 2900 years BP, there is little evidence for allogenicinflux. Dramatic changes in mineral magnetic properties and a strong AARmagnetic fabric appear approximately 2900 years BP, suggesting major changes inwatershed conditions, either in the hydrologic regime or in vegetative cover.Between 2900 and 1800 yrs BP, large variations in magnetic mineralogy areapparent, whereas about 1800 years BP, a single sediment source began todominate the magnetic mineralogy. About 100–200 years ago, conditionsagain changed, probably due to clearcutting and settlement of the watershed.Topsoil erosion appears to have dominated the magnetic signal. S and Mnconcentration downcore indicate that there is little evidence for reductiondiagenesis having caused the changes observed in magnetic mineral type andconcentration, except in the top 10 centimeters of the sediment column.     

Magnetic interaction in biotites and oxidised biotites

Mössbauer spectra of biotites (1) and (2) with relative iron concentrations ～1:1·6 and of their oxidation products are recorded at 4 K in zero field and in applied fields up to 5 T. Magnetic susceptibility data are also reported. The results show that Fe III spins are in a ferromagnetic configuration in the c-plane in both biotites. Partial oxidation of biotite (1) leads to a canted ferromagnetic structure, while complete oxidation of biotite (2) yields an antiferromagnetic spin configuration. Nearest-neighbour antiferromagnetic Fe III-O-Fe III, and ferromagnetic Fe III-O-Fe II and Fe II-O-Fe II superexchange can account for the results. For biotites with higher concentrations of iron, Fe II and Fe III seem to be distributed randomly in the triangular lattice. From susceptibility results in biotites dilute in iron, an estimate of the ratio of nearest-neighbour and next-nearest neighbour magnetic interaction in the triangular lattice is derived.