scholarly journals Addressing numerical challenges in introducing a reactive transport code into a land surface model: a biogeochemical modeling proof-of-concept with CLM–PFLOTRAN 1.0

2016 ◽  
Vol 9 (3) ◽  
pp. 927-946 ◽  
Author(s):  
Guoping Tang ◽  
Fengming Yuan ◽  
Gautam Bisht ◽  
Glenn E. Hammond ◽  
Peter C. Lichtner ◽  
...  
Keyword(s):  
Land Surface ◽  
Reactive Transport ◽  
Computing Time ◽  
Reaction Network ◽  
Transformation Method ◽  
Residual Concentration ◽  
Log Transformation ◽  
Time Stepping ◽  
Transport Code ◽  
Scaling Methods

Abstract. We explore coupling to a configurable subsurface reactive transport code as a flexible and extensible approach to biogeochemistry in land surface models. A reaction network with the Community Land Model carbon–nitrogen (CLM-CN) decomposition, nitrification, denitrification, and plant uptake is used as an example. We implement the reactions in the open-source PFLOTRAN (massively parallel subsurface flow and reactive transport) code and couple it with the CLM. To make the rate formulae designed for use in explicit time stepping in CLMs compatible with the implicit time stepping used in PFLOTRAN, the Monod substrate rate-limiting function with a residual concentration is used to represent the limitation of nitrogen availability on plant uptake and immobilization. We demonstrate that CLM–PFLOTRAN predictions (without invoking PFLOTRAN transport) are consistent with CLM4.5 for Arctic, temperate, and tropical sites.Switching from explicit to implicit method increases rigor but introduces numerical challenges. Care needs to be taken to use scaling, clipping, or log transformation to avoid negative concentrations during the Newton iterations. With a tight relative update tolerance (STOL) to avoid false convergence, an accurate solution can be achieved with about 50 % more computing time than CLM in point mode site simulations using either the scaling or clipping methods. The log transformation method takes 60–100 % more computing time than CLM. The computing time increases slightly for clipping and scaling; it increases substantially for log transformation for half saturation decrease from 10−3 to 10−9 mol m−3, which normally results in decreasing nitrogen concentrations. The frequent occurrence of very low concentrations (e.g. below nanomolar) can increase the computing time for clipping or scaling by about 20 %, double for log transformation. Overall, the log transformation method is accurate and robust, and the clipping and scaling methods are efficient. When the reaction network is highly nonlinear or the half saturation or residual concentration is very low, the allowable time-step cuts may need to be increased for robustness for the log transformation method, or STOL may need to be tightened for the clipping and scaling methods to avoid false convergence.As some biogeochemical processes (e.g., methane and nitrous oxide reactions) involve very low half saturation and thresholds, this work provides insights for addressing nonphysical negativity issues and facilitates the representation of a mechanistic biogeochemical description in Earth system models to reduce climate prediction uncertainty.

scholarly journals Using reactive transport codes to provide mechanistic biogeochemistry representations in global land surface models: CLM-PFLOTRAN 1.0

2015 ◽  
Vol 8 (12) ◽  
pp. 10627-10676 ◽  
Author(s):  
G. Tang ◽  
F. Yuan ◽  
G. Bisht ◽  
G. E. Hammond ◽  
P. C. Lichtner ◽  
...  
Keyword(s):  
Plant Uptake ◽  
Land Surface ◽  
Reactive Transport ◽  
Computing Time ◽  
Reaction Network ◽  
Implicit Time ◽  
Land Surface Models ◽  
Log Transformation ◽  
Time Stepping ◽  
Surface Models

Abstract. We explore coupling to a configurable subsurface reactive transport code as a flexible and extensible approach to biogeochemistry in land surface models; our goal is to facilitate testing of alternative models and incorporation of new understanding. A reaction network with the CLM-CN decomposition, nitrification, denitrification, and plant uptake is used as an example. We implement the reactions in the open-source PFLOTRAN code, coupled with the Community Land Model (CLM), and test at Arctic, temperate, and tropical sites. To make the reaction network designed for use in explicit time stepping in CLM compatible with the implicit time stepping used in PFLOTRAN, the Monod substrate rate-limiting function with a residual concentration is used to represent the limitation of nitrogen availability on plant uptake and immobilization. To achieve accurate, efficient, and robust numerical solutions, care needs to be taken to use scaling, clipping, or log transformation to avoid negative concentrations during the Newton iterations. With a tight relative update tolerance to avoid false convergence, an accurate solution can be achieved with about 50 % more computing time than CLM in point mode site simulations using either the scaling or clipping methods. The log transformation method takes 60–100 % more computing time than CLM. The computing time increases slightly for clipping and scaling; it increases substantially for log transformation for half saturation decrease from 10−3 to 10−9 mol m−3, which normally results in decreasing nitrogen concentrations. The frequent occurrence of very low concentrations (e.g. below nanomolar) can increase the computing time for clipping or scaling by about 20 %; computing time can be doubled for log transformation. Caution needs to be taken in choosing the appropriate scaling factor because a small value caused by a negative update to a small concentration may diminish the update and result in false convergence even with very tight relative update tolerance. As some biogeochemical processes (e.g., methane and nitrous oxide production and consumption) involve very low half saturation and threshold concentrations, this work provides insights for addressing nonphysical negativity issues and facilitates the representation of a mechanistic biogeochemical description in earth system models to reduce climate prediction uncertainty.

scholarly journals BioRT-Flux-PIHM v1.0: a watershed biogeochemical reactive transport model

2020 ◽  
Author(s):  
Wei Zhi ◽  
Yuning Shi ◽  
Hang Wen ◽  
Leila Saberi ◽  
Gene-Hua Crystal Ng ◽  
...  
Keyword(s):  
Deep Water ◽  
Land Surface ◽  
Reactive Transport ◽  
Transport Model ◽  
Transport Processes ◽  
Spatially Explicit ◽  
Watershed Scale ◽  
Nitrogen And Phosphorus ◽  
Reactive Transport Model ◽  
Component Representation

Abstract. Watersheds are the fundamental Earth surface functioning unit that connects the land to aquatic systems. Existing watershed-scale models typically have physics-based representation of hydrology process but often lack mechanism-based, multi-component representation of reaction thermodynamics and kinetics. This lack of watershed reactive transport models has limited our ability to understand and predict solute export and water quality, particularly under changing climate and anthropogenic conditions. Here we present a recently developed BioRT-Flux-PIHM (BFP) v1.0, a watershed-scale biogeochemical reactive transport model. Augmenting the previously developed RT-Flux-PIHM that integrates land-surface interactions, surface hydrology, and abiotic geochemical reactions (Bao et al., 2017, WRR), the new development enables the simulation of 1) biotic processes including plant uptake and microbe-mediated biogeochemical reactions that are relevant to the transformation of organic matter that involve carbon, nitrogen, and phosphorus; and 2) shallow and deep water partitioning to represent surface and groundwater interactions. The reactive transport part of the code has been verified against the widely used reactive transport code CrunchTope. BioRT-Flux-PIHM v1.0 has recently been applied to understand reactive transport processes in multiple watersheds across different climate, vegetation, and geology conditions. This paper introduces the governing equations and model structure of the code. It also demonstrates examples that simulate shallow and deep water interactions, and biogeochemical reactive transport relevant to nitrate and dissolved organic carbon (DOC). These examples were illustrated in two simulation modes of varying complexity. One is the spatially implicit mode that focuses on processes and average behavior of a watershed. Another is in a spatially explicit mode that includes details of topography, land cover, and soil property conditions. The spatially explicit mode can be used to understand the impacts of spatial structure and identify hot spots of biogeochemical reactions.

Reliability-based design optimization for vertical vibrations of a modified electric vehicle using fourth-moment polynomial standard transformation method

2020 ◽  
Vol 234 (10-11) ◽  
pp. 2649-2664
Author(s):  
Sheng Wang ◽  
Lin Hua ◽  
Xinghui Han ◽  
Zhuoyu Su
Keyword(s):  
Design Optimization ◽  
Electric Vehicle ◽  
Optimization Problems ◽  
Computing Time ◽  
Optimization Procedure ◽  
Transformation Method ◽  
Vertical Vibration ◽  
Fourth Moment ◽  
Reliability Based Design Optimization ◽  
Reliability Based Design

This article presents a new reliability-based design optimization procedure for the vertical vibration issues raised by a modified electric vehicle using fourth-moment polynomial standard transformation method. First, the fourth-moment polynomial standard transformation method with polynomial chaos expansion is used to obtain the reliability index of uncertain constraints in the reliability-based design optimization which is highly precise and saves computing time compared with other common methods. Next, the half-car model with nonlinear suspension parameters for the modified electric vehicle is investigated, and the response surface methodology is adopted to approximate the complex and time-consuming vertical vibration calculation to the polynomial expressions, and the approximation is validated for reliability-based design optimization results within permissible error level. Then, reliability-based design optimization results under both deterministic and uncertain load parameters are shown and analyzed. Unlike the traditional vertical vibration optimization that only considers one or several sets of load parameters, which lacks versatility, this article presents the reliability-based design optimization with uncertain load parameters which is more suitable for engineering. The results show that the proposed reliability-based design optimization procedure is an effective and efficient way to solve vertical vibration optimization problems for the modified electric vehicle, and the optimization statistics, including the maximum probability interval, can provide references for other suspension dynamical optimization.

Coupling between Thermo-Hydro-Chemical reactive transport and Gibbs minimisation: magma evolution in evolving multiphase porous media

2020 ◽  
Author(s):  
Annelore Bessat ◽  
Sébastien Pilet ◽  
Stefan M. Schmalholz ◽  
Yuri Podladchikov
Keyword(s):  
Gibbs Energy ◽  
Reactive Transport ◽  
Viscous Medium ◽  
Physical Aspect ◽  
Magma Evolution ◽  
Energy Minimisation ◽  
Transport Code ◽  
Low Degree ◽  
Melt Percolation ◽  
Alkaline Magmas

<p>The formation of alkaline magmas observed worldwide requires that low degree-melts, potentially formed in the asthenosphere, were able to cross the overlying lithosphere. Fracturing in the upper, brittle part of the lithosphere may help to extract this melt to the surface. However, the mechanism of extraction in the lower, ductile part of the lithosphere is still contentious. Metasomatic enrichment of the lithospheric mantle demonstrates that such low-degree melts interact with the lithosphere, but the physical aspect of this process remains unclear. The aim of this study is to better understand the percolation of magma in a porous viscous medium at pressure (P) and temperature (T) conditions relevant for the base of the lithosphere. We study such melt percolation numerically with a Thermo-Hydro-Chemical model of reactive transport coupled with thermodynamic data obtained via Gibbs energy minimisation. We perform Gibbs energy minimisation with Matlab using the linprog algorithm. We start with a simple ternary system of Forsterite/Fayalite/Enstatite solids and melts. All variables are a function of T, P and composition of the system (C), and are computed in both the Gibbs energy minimisation and in the reactive transport code, and can therefore vary freely.</p>

Speeding up reactive transport simulations: statistical surrogates and caching of simulation results in lookup tables

2020 ◽  
Author(s):  
Marco De Lucia ◽  
Robert Engelmann ◽  
Michael Kühn ◽  
Alexander Lindemann ◽  
Max Lübke ◽  
...  
Keyword(s):  
Reactive Transport ◽  
Large Scale ◽  
Computing Time ◽  
Real Life ◽  
Data Driven ◽  
Support Vector ◽  
Coupled Transport ◽  
Charge Balance ◽  
Lookup Tables ◽  
Simulation Results

<p>A successful strategy for speeding up coupled reactive transport simulations at price of acceptable accuracy loss is to compute geochemistry, which represents the bottleneck of these simulations, through data-driven surrogates instead of ‘full physics‘ equation-based models [1]. A surrogate is a multivariate regressor trained on a set of pre-calculated geochemical simulations or potentially even at runtime during the coupled simulations. Many available algorithms and implementations are available from the thriving Machine Learning community: tree-based regressors such as Random Forests or xgboost, Artificial Neural Networks, Gaussian Processes and Support Vector Machines just to name a few. Given the ‘black-box‘ nature of the surrogates, however, they generally disregard physical constraints such as mass and charge balance, which are of course of paramount importance for coupled transport simulations. A runtime check of error of balances in the surrogate outcomes is therefore necessary: predictions offending a given tolerance must be rejected and the full physics chemical simulations run instead. Thus the practical speedup of this strategy is a tradeoff between careful training of the surrogate and run-time efficiency.</p><p><br>In this contribution we demonstrate that the use of surrogates can lead to a dramatic decrease of required computing time, with speedup factors in the order of 10 or even 100 in the most favorable cases. Thus, large scale simulations with some 10<sup>6</sup> grid elements are feasible on common workstations without requiring computation on HPC clusters [2].</p><p><span><br>Furthermore, we showcase our implementation of Distributed Hash Tables caching geochemical simulation results for further reuse in subsequent time steps. The computational advantage here stems from the fact that query and retrieval from lookup tables is much faster than both full physics geochemical simulations and surrogate predictions. Another advantage of this algorithm is that virtually no loss of accuracy is introduced in the simulations. Enabling the caching of geochemical simulations through DHT speeds up large scale reactive transport simulations up to a factor of four even when computing on several hundred </span><span>cores</span><span>.</span></p><p><br>These algorithmical developments are demonstrated in comparison with published reactive transport benchmarks and on a real-life scenario of CO<sub>2</sub> storage.</p><p> </p><p> </p><p><span>[1] </span><span>Jatnieks, J., De Lucia, M., Dransch, D., Sips, M. (2016): Data-driven surrogate model approach for improving the performance of reactive transport simulations. Energy Procedia </span><span>97</span><span>, pp. 447-453. DOI: 10.1016/j.egypro.2016.10.047</span></p><p>[2] De Lucia, M., Kempka, T., Jatnieks, J., Kühn, M. (2017): Integrating surrogate models into subsurface simulation framework allows computation of complex reactive transport scenarios. Energy Procedia 125, pp. 580-587. DOI: 10.1016/j.egypro.2017.08.200</p>

Presentation and use of a reactive transport code in porous media

2007 ◽  
Vol 32 (1-7) ◽  
pp. 507-517 ◽  
Author(s):  
Ph. Montarnal ◽  
C. Mügler ◽  
J. Colin ◽  
M. Descostes ◽  
A. Dimier ◽  
...  
Keyword(s):  
Porous Media ◽  
Reactive Transport ◽  
Transport Code

scholarly journals Theory and Application of the Numerical Simulation in the Frozen Soil Problems

2021 ◽  
Vol 13 (4) ◽  
pp. 30
Author(s):  
Liye Song ◽  
Yirang Yuan
Keyword(s):  
Numerical Simulation ◽  
Land Surface ◽  
Environmental Science ◽  
Transformation Method ◽  
Frozen Soil ◽  
Free Boundaries ◽  
Important Indicator ◽  
Thaw Depth ◽  
Terrestrial Hydrology ◽  
Coordinate Transformation Method

The freezing-thawing processes in soils are important components of terrestrial hydrology, which significantly influence energy and water exchanges between land surface and sub-surface. Long-term changes in frost and thaw depths are also an important indicator of climate change. A water-heat coupled movements model is established with frozen soil in this paper, which treats the freezing/thawing front as a moving interface governed by some Stefan problems with two free boundaries. The numerical simulation is conducted by using the modified finite difference method. The model is validated to compare its predictions with GEWEX Asian Monsoon Experiment(GAME)-Tibet observations at D66 site in Tibetan Plateau. The results show that the simulated soil temperature, soil water content and frost/thaw depth are in excellent agreement with the measured values. Finally, optimal error estimation for L^∞ norm is derived on the model problem by using coordinate transformation method. The numerical simulation system is established on the basis of rigorous mathematics and mechanics, which successfully solved the important and difficult problems of environmental science.

scholarly journals Low rank compression in the numerical solution of the nonequilibrium Dyson equation

2021 ◽  
Vol 10 (4) ◽  
Author(s):  
Jason Kaye ◽  
Denis Golez
Keyword(s):  
Computing Time ◽  
Mean Field Theory ◽  
Mean Field ◽  
Dyson Equation ◽  
Low Rank ◽  
Strong Coupling Regime ◽  
Gw Approximation ◽  
Empirical Observation ◽  
Time Stepping ◽  
Computational Bottleneck

We propose a method to improve the computational and memory efficiency of numerical solvers for the nonequilibrium Dyson equation in the Keldysh formalism. It is based on the empirical observation that the nonequilibrium Green's functions and self energies arising in many problems of physical interest, discretized as matrices, have low rank off-diagonal blocks, and can therefore be compressed using a hierarchical low rank data structure. We describe an efficient algorithm to build this compressed representation on the fly during the course of time stepping, and use the representation to reduce the cost of computing history integrals, which is the main computational bottleneck. For systems with the hierarchical low rank property, our method reduces the computational complexity of solving the nonequilibrium Dyson equation from cubic to near quadratic, and the memory complexity from quadratic to near linear. We demonstrate the full solver for the Falicov-Kimball model exposed to a rapid ramp and Floquet driving of system parameters, and are able to increase feasible propagation times substantially. We present examples with 262 144 time steps, which would require approximately five months of computing time and 2.2 TB of memory using the direct time stepping method, but can be completed in just over a day on a laptop with less than 4 GB of memory using our method. We also confirm the hierarchical low rank property for the driven Hubbard model in the weak coupling regime within the GW approximation, and in the strong coupling regime within dynamical mean-field theory.

Sign in / Sign up

Export Citation Format

Share Document