The Experts below are selected from a list of 189 Experts worldwide ranked by ideXlab platform
Adrian Sandu - One of the best experts on this subject based on the ideXlab platform.
-
SpringSim - Development and acceleration of parallel Chemical Transport models
Proceedings of the 2010 Spring Simulation Multiconference on - SpringSim '10, 2010Co-Authors: Paul R. Eller, Kumaresh Singh, Adrian SanduAbstract:Improving Chemical Transport models for atmospheric simulations relies on future developments of mathematical methods and parallelization methods. Better mathematical methods allow simulations to more accurately model realistic processes and/or to run in a shorter amount of time. Parallelization methods allow simulations to run in less time, allowing scientists to use more accurate or more detailed simulations (higher resolution grids, smaller time steps). The STEM Chemical Transport model provides a large scale end-to-end application to experiment with running Chemical integration methods and Transport methods on GPUs. GPUs provide high computational power at a fairly cheap cost. The CUDA programming environment simplifies the GPU development process by providing access to powerful functions to execute parallel code. This work demonstrates the acceleration of a large scale end-to-end application on GPUs showing significant speedups. This is achieved by implementing all relevant kernels on the GPU using CUDA. Nevertheless, further improvements to GPUs are needed to allow these applications to fully exploit the power of GPUs.
-
Implementation and Evaluation of an Array of Chemical Solvers in a Global Chemical Transport Model
Geoscientific Model Development Discussions, 2009Co-Authors: Paul R. Eller, Adrian Sandu, Kumaresh Singh, Kevin W. Bowman, Daven K. HenzeAbstract:This paper discusses the implementation and per- formance of an array of gas-phase chemistry solvers for the state-of-the-science GEOS-Chem global Chemical Transport model. The implementation is based on the Kinetic PrePro- cessor (KPP). Two perl parsers automatically generate the needed interfaces between GEOS-Chem and KPP, and al- low access to the Chemical simulation code without any addi- tional programming effort. This work illustrates the potential of KPP to positively impact global Chemical Transport mod- eling by providing additional functionality as follows. (1 ) The user can select a highly efficient numerical integration method from an array of solvers available in the KPP library. (2) KPP offers extreme flexibility for studies that involve changing the Chemical mechanism (e.g., a set of additional reactions is automatically translated into efficient code a nd incorporated into a modified global model). (3) This work provides immediate access to tangent linear, continuous ad- joint, and discrete adjoint Chemical models, with applicat ions to sensitivity analysis and data assimilation.
-
SpringSim - Optimizing large scale Chemical Transport models for multicore platforms
2008Co-Authors: John C. Linford, Adrian SanduAbstract:The performance of a typical Chemical Transport model is determined on two multicore processors: the heterogeneous Cell Broadband Engine and the homogeneous Intel Quad-Core Xeon shared-memory multiprocessor. Two problem decomposition techniques are discussed: dimension splitting for promoting parallelization in Chemical Transport models, and time splitting, for reducing truncation error. Additionally, a scalable method for accessing random rows or columns of a matrix of arbitrary size from the accelerator units of the Cell Broadband Engine is presented. This scalable access method increases Chemical Transport model efficiency by an average of 30% and significantly improves the scalability of dimension-splitting techniques on the Cell Broadband Engine. Experiments show that Chemical Transport models are 31% more efficient on the Cell Broadband Engine when only six accelerator units are used than on a shared-memory multiprocessor with eight executing cores. Our fully-optimized models achieve an average 118% speedup on the Cell Broadband Engine, and an average 87.5% speedup on a shared-memory multiprocessor with OpenMP.
-
singular vector analysis for atmospheric Chemical Transport models
Monthly Weather Review, 2006Co-Authors: Wenyuan Liao, Adrian Sandu, Gregory R Carmichael, Tianfeng ChaiAbstract:Abstract The singular vectors of a Chemical Transport model are the directions of maximum perturbation growth over a finite time interval. They have proved useful for the estimation of error growth, the initialization of ensemble forecasts, and the optimal placement of adaptive observations. The aim of this paper is to address computational aspects of singular vector analysis for atmospheric Chemical Transport models. The distinguishing feature of these models is the presence of stiff Chemical interactions. A projection approach to preserve the symmetry of the tangent linear–adjoint operator for stiff systems is discussed, and extended to 3D Chemical Transport simulations. Numerical results are presented for a simulation of atmospheric pollution in East Asia in March 2001. The singular values and the structure of the singular vectors depend on the length of the simulation interval, the meteorological data, the location of the optimization region and the selection of optimization species, the choice of err...
Wei Kuo Tao - One of the best experts on this subject based on the ideXlab platform.
-
Production of lightning NOx and its vertical distribution calculated from three-dimensional cloud-scale Chemical Transport model simulations
Journal of Geophysical Research Atmospheres, 2010Co-Authors: Lesley E. Ott, Dale J. Allen, Alex J. Decaria, Brian Ridley, Ruei Fong Lin, Kenneth E Pickering, Georgiy L Stenchikov, Stephen Lang, Wei Kuo TaoAbstract:A three-dimensional (3-D) cloud-scale Chemical Transport model that includes a parameterized source of lightning NOx on the basis of observed flash rates has been used to simulate six midlatitude and subtropical thunderstorms observed during four field projects. Production per intracloud (PIC) and cloud-to-ground (PCG) flash is estimated by assuming various values of PIC and PCG for each storm and determining which production scenario yields NOx mixing ratios that compare most favorably with in-cloud aircraft observations. We obtain a mean PCG value of 500 moles NO (7 kg N) per flash. The results of this analysis also suggest that on average, PIC may be nearly equal to PCG, which is contrary to the common assumption that intracloud flashes are significantly less productive of NO than are cloud-to-ground flashes. This study also presents vertical profiles of the mass of lightning NOx after convection based on 3-D cloud-scale model simulations. The results suggest that following convection, a large percentage of lightning NOx remains in the middle and upper troposphere where it originated, while only a small percentage is found near the surface. The results of this work differ from profiles calculated from 2-D cloud-scale model simulations with a simpler lightning parameterization that were peaked near the surface and in the upper troposphere (referred to as a “C-shaped” profile). The new model results (a backward C-shaped profile) suggest that Chemical Transport models that assume a C-shaped vertical profile of lightning NOx mass may place too much mass near the surface and too little in the middle troposphere.
Dan B. Jaynes - One of the best experts on this subject based on the ideXlab platform.
-
Field Measurement of Soil Surface Chemical Transport Properties for Comparison of Management Zones
Soil Science Society of America Journal, 2007Co-Authors: Joshua L. Heitman, Robert Horton, Dan B. Jaynes, Anju Gaur, T. C. KasparAbstract:Management of Chemicals in soil is important, yet the complexity of field soils limits prediction of management effects on Transport. To date, few methods have been available for field measurement of Chemical Transport properties, but a recently developed dripper-time domain reflectometry technique allows rapid collection of data for determining these properties. The objective of this work was to apply this technique for comparison of Chemical Transport properties for different soil management zones. Experiments were conducted comparing four interrow management zones: no-till nontrafficked, no-till trafficked, chisel plow nontrafficked, and chisel plow trafficked. Drip emitters were positioned at 12 locations in each zone and used to apply water followed by a step input of CaCl 2 tracer solution. Breakthrough curves were measured via electrical conductivity with time domain reflectometry probes. The mobile-immobile model was fit to the breakthrough curves to determine Chemical Transport properties. Mean Chemical Transport properties were 0.34, 0.1 h -1 , 10 cm h -1 , 164 cm 2 h -1 , and 5 cm, for the immobile water fraction, mass exchange coefficient, average pore-water velocity, mobile dispersion coefficient, and dispersivity, respectively. All five properties showed significant differences between management zones. Differences in mass exchange and mobile dispersion coefficients coincided with differences in tillage, while differences in mean pore water velocities coincided with differences in traffic. The immobile water fraction was largest for the no-till nontrafficked zone. These results represent one of very few reports for field measurement of Chemical Transport properties and the first application of this approach for comparison of Chemical Transport properties across management zones.
-
Field determination of soil hydraulic and Chemical Transport properties
Soil Science, 2002Co-Authors: Salem A. Al-jabri, Robert Horton, Dan B. Jaynes, Anju GaurAbstract:Hydraulic and Chemical Transport properties are the major inputs in predictive models that simulate the movement of water and Chemicals through the vadose zone. However, there is a lack of field measurements of such properties to verify models describing water and Chemical movement through the soil. One of the objectives of this study was to use a point source method to determine simultaneously the hydraulic and Chemical Transport properties at multiple field locations. A second objective was to determine the spatial distribution of such properties across a field. A total of 50 field locations within a 7 × 15-m area were rapidly and simultaneously evaluated for such properties. The hydraulic properties were the saturated hydraulic conductivity (K s ) and the macroscopic capillary length (λ c ). The Chemical Transport properties were the immobile water content, expressed as a fraction of water content (θ im /θ) and the mass exchange coefficient (a). The hydraulic properties were determined by applying three discharge rates from irrigation dripper lines and measuring the resultant steady-state flux densities at the soil surface beneath each emitter. The Chemical Transport properties were determined by applying a sequence of three conservative tracers at a steady-state infiltration rate and measuring their resident concentration in the soil. The K s values ranged from 7.5 to 79.0 cm h -1 , with a median of 27.4 cm h -1 (± 16.8). The λ c values ranged from 0.03 to 13.1 cm, with a median of 2.6 cm (± 3.6). The θ im /θ values ranged from 0.36 to 0.88, with a median of 0.57 (± 0.098). The a values ranged from 0.002 to 0.12 h -1 , with a median of0.034 h -1 (± 0.027). The values of the hydraulic and Chemical Transport parameters were found to be comparable with values reported by studies conducted on nearby field locations on similar soil. Based on semi-variogram analysis, the measured properties were not spatially correlated. Because the method required only 2 days to collect data it should prove useful for future studies that require extensive field measurements of hydraulic and Chemical Transport properties.
-
A Point-Source Method for Rapid Simultaneous Estimation of Soil Hydraulic and Chemical Transport Properties
Soil Science Society of America Journal, 2002Co-Authors: Salem A. Al-jabri, Robert Horton, Dan B. JaynesAbstract:Hydraulic and Chemical Transport properties are needed for accurate prediction of water and Chemical movement through the vadose zone. Field methods used to estimate such properties are often hampered by extensive labor and time constraints. One of the objectives of this study was to develop an experimental setup and a procedure for a point-source method that fadlitates rapid and simultaneous measurements of soil hydraulic and Chemical Transport properties at multiple locations. Another objective was to evaluate the point-source method by comparing the parameters with those produced by ponded and tension infiltrometers. The experimental setup consisted of three drip-per lines equipped with pressure-compensating drippers. The setup was evaluated on a greenhouse soil pit. Determined hydraulic properties were the saturated hydraulic conductivity (K,) and the macroscopic capillary length (λ c ). Hydraulic properties (from the point-source method) were determined by applying four consecutive discharge rates on the soil surface and measuring their corresponding steady-state saturated areas. Determined Chemical Transport parameters were the immobile water fraction (θ im /θ) and the mass exchange coefficient (a). They were determined by applying a sequence of conservative fluorobenzoate tracers. The point-source method gave consistent and reliable estimates for both sets of properties. Except for a, there was no significant difference between the two procedures (point source vs. infiltrometers) in determining both sets of properties. The study showed that the point-source setup could be utilized for rapid and simultaneous estimation of soil hydraulic and Chemical Transport properties at multiple locations with minimum labor requirements.
Lesley E. Ott - One of the best experts on this subject based on the ideXlab platform.
-
Production of lightning NOx and its vertical distribution calculated from three-dimensional cloud-scale Chemical Transport model simulations
Journal of Geophysical Research Atmospheres, 2010Co-Authors: Lesley E. Ott, Dale J. Allen, Alex J. Decaria, Brian Ridley, Ruei Fong Lin, Kenneth E Pickering, Georgiy L Stenchikov, Stephen Lang, Wei Kuo TaoAbstract:A three-dimensional (3-D) cloud-scale Chemical Transport model that includes a parameterized source of lightning NOx on the basis of observed flash rates has been used to simulate six midlatitude and subtropical thunderstorms observed during four field projects. Production per intracloud (PIC) and cloud-to-ground (PCG) flash is estimated by assuming various values of PIC and PCG for each storm and determining which production scenario yields NOx mixing ratios that compare most favorably with in-cloud aircraft observations. We obtain a mean PCG value of 500 moles NO (7 kg N) per flash. The results of this analysis also suggest that on average, PIC may be nearly equal to PCG, which is contrary to the common assumption that intracloud flashes are significantly less productive of NO than are cloud-to-ground flashes. This study also presents vertical profiles of the mass of lightning NOx after convection based on 3-D cloud-scale model simulations. The results suggest that following convection, a large percentage of lightning NOx remains in the middle and upper troposphere where it originated, while only a small percentage is found near the surface. The results of this work differ from profiles calculated from 2-D cloud-scale model simulations with a simpler lightning parameterization that were peaked near the surface and in the upper troposphere (referred to as a “C-shaped” profile). The new model results (a backward C-shaped profile) suggest that Chemical Transport models that assume a C-shaped vertical profile of lightning NOx mass may place too much mass near the surface and too little in the middle troposphere.
P Prati - One of the best experts on this subject based on the ideXlab platform.
-
an integrated pm2 5 source apportionment study positive matrix factorisation vs the Chemical Transport model camx
Atmospheric Environment, 2014Co-Authors: M C Bove, P Brotto, Federico Cassola, E Cuccia, D Massabo, A Mazzino, A Piazzalunga, P PratiAbstract:Abstract Receptor and Chemical Transport Models are commonly used tools in source apportionment studies, even if different expertise is required. We describe an experiment using both approaches to apportion the PM2.5 (i.e., particulate matter with aerodynamic diameters below 2.5 μm) sources in the city of Genoa (Italy). A sampling campaign was carried out to collect PM2.5 samples daily for approximately six month during 2011 in three sites. The subsequent compositional analyses included the speciation of elements, major ions and both organic and elemental carbon; these data produced a large database for receptor modelling through Positive Matrix Factorisation (PMF). In the same period, a meteorological and air quality modelling system was implemented based on the mesoscale numerical weather prediction model WRF and the Chemical Transport model CAMx to obtain meteorological and pollutant concentrations up to a resolution of 1.1 km. The source apportionment was evaluated by CAMx over the same period that was used for the monitoring campaign using the Particulate Source Apportionment Technology tool. Even if the source categorisations were changed (i.e., groups of time-correlated compounds in PMF vs. activity categories in CAMx), the PM2.5 source apportionment by PMF and CAMx produced comparable results. The different information provided by the two approaches (e.g., real-world factor profile by PMF and apportionment of a secondary aerosol by CAMx) was used jointly to elucidate the composition and origin of PM2.5 and to develop a more general methodology. When studying the primary and secondary components of PM, the main anthropogenic sources in the area were road Transportation, energy production/industry and maritime emissions, accounting for 40%–50%, 20%–30% and 10%–15%, of PM2.5, respectively.