RESEARCH PUBLICATIONS & EXPERIENCES

Publications

Chen M., A.A. Keller, and Z. Lu (2007), Stochastic Analysis of three-phase flow in heterogeneous porous media, Stochastic Environmental Research and Risk Assessment, doi: 10.1007/s00477-007-0198-y. [PDF]

Chen M., Z. Lu, and G.A. Zyvoloski (2007), Conditional simulation of water-oil flow in heterogeneous porous media, Stochastic Environmental Research and Risk Assessment, doi: 10.1007/s00477-007-0178-2. [PDF]

Chen M., D. Zhang, A.A. Keller, Z. Lu, and G.A. Zyvoloski (2006), A stochastic analysis of transient two phase flow in heterogeneous media, Water Resources Research, 42, W03425, doi:10.1029/2005WR004257. [PDF]

Chen M., G.A. Zyvoloski (2006), Model coupling with control volume finite elements, MODFLOW and more 2006: managing ground water systems, Golden, CO, May 21-24, 2006, International Ground Water Modeling Center (IGWMC), Colorado School of Mines, Conference Proceedings Vol. II, 732-736. [PDF]

Chen M., D. Zhang, A.A. Keller, Z. Lu (2005), A Stochastic Analysis of Steady-State Two Phase Flow in Heterogeneous Media. Water Resources Research, 41(1): W01006. [PDF]

AA Keller, M. Chen (2003), Effect of Spreading Coefficient on Three-Phase Relative Permeability of NAPL. Water Resources Research, 39(10): 1288. [PDF]

AA Keller, M. Chen (2002), Seasonal variation in bioavailability of residual NAPL in the vadose zone, in Proceedings of the International Groundwater Symposium, March 25-28, 2002 in Berkeley, CA, USA. [PDF]

Chen M. , J.R. Ni, A. Xue, G.H. Huang, A. Chakama (2003), Comparative study on typical feed-forward ANNs for tidal simulation, Journal of Sediment Research, in China, No.5, 2003, 41-48. [PDF]

Chen M. , J.R. Ni, A. Chakama, G.H. Huang, A. Xue (2003), Application of genetic algorithm-based artificial neural networks in 2D tidal flow simulation, Journal of Hydraulic Engineering, in China, No.10, 2003, 87-95. [PDF]

7/2005 – present Postdoc Research Associate, Los Alamos National Laboratory (LANL)

My major research field is on modeling multiphase fluids and heat flow in OIL SHALE, and Piceance Basin modeling funded by Chevron Company supervised under George A. Zyvoloski. Most oil shale are fine-grained sedimentary rocks containing relatively large amounts of organic matter from which significant amounts of shale oil and combustible gas can be extracted by destructive distillation, which is the potential for burning as a fuel, in addition to the traditional petroleum.

I also cooperated with Elizabeth H. Keating on a project related to estimation of aquifer recharge.

Chevron 3-D piceance basin flow modeling. The part of the project was just embarked in June, 2007. The Piceance basin of northwestern Colorado contains rich deposits of kerogen-bearing sediments, which reserves 1.2E12 barrels of shale oil. The basic idea of in-situ extracting shale oil is to inject supercritical CO2, heat and rubble kerogen, and CO2 will bring oil out. This process is going to change the subsurface hydrogeologic properties ultimately. The major concern is the impact on groundwater. I am responsible for the construction of Piceance basin model, and model calibration using the field data from PNL, and INL. More details are yet to come.
. I am the key member of the developing team assembled in LANL responsible for the modeling part of Chevron-LANL oilshale project. The work includes subsurface process models, equation of state (flash calculation) subgrid scale models, upscaling, anisotropy, and coupled fluid and stress models. I am in charge of subsurface process models using Newton-Raphson method for outer nonlinear iterations and preconditional krylov method for the inner linear iterations. The flowchart of the model includes: formulate multiphase, multicomponent mass and energy conservation equations; discretize these equation for linear system using Control Volume Finite Element (CVFE) method; prefactorize, simplify, or eliminate degrees of freedom of linear system as appropriate; solve linear system. Flash calculations and state of art solvers are also deployed. LANL developed unstructured grid generator LaGrit are coupled with FEHM.

. Model coupling has become increasing important in recent years because of the need for more quantitative and realistic models. Major motivating factors include the need to simulate the complete water cycle, the necessity for high resolution grids to minimize numerical dispersion in contaminant transport calculations, the need to have fine grids for accuracy near production wells, and the requirement to use high resolution grids with resolution sufficient to capture correlation lengths in stochastic models. Model coupling also can be very important for situations where a relatively coarse basin model has been developed with a large investment in data collection and numerical model development. In this situation, it is easily conceived that a high resolution submodel might be required a posteriori for the simulation of a new water well, simulation of the flow and transport near a newly discovered contaminant source or a detailed analysis of aquifer stream interaction. The ability to include submodels with different flow physics is important. Our work present results for a variety of locally refined (called hybrid in this paper) and iteratively coupled control volume finite element models. In addition we compare the coupling algorithms for models with different physics. Error analysis will include comparisons to finite difference methods, non-iterative finite element methods, and fine grid solutions. Example simulations include coupled models with different flow physics. Discussion of iteration algorithms, strategies in choosing child regions and transition zones, and computational challenges presented in a parallel computing environment are also given.

In this study, we propose to directly measure the amount of groundwater change occurring in the Espanola Basin by using time sequence of gravity surveys acquired over the basin. Groundwater resources in many portions of the United States, but particularly in the southwestern U.S., are becoming increasingly in demand, primarily due to population increase, but also due to increased industrial and recreational use. Drought conditions which persisted in recent years caused concern that surface water supplied to some metropolitan areas (e.g. Santa Fe, NM) might be endangered, forcing such areas to rely very heavily upon groundwater supplies. Some areas (e.g. Los Alamos, Rio Rancho, NM) rely completely on groundwater for water supplies. In addition, population increase in Albuquerque, NM, which completely relies upon groundwater for its water supply, are forcing the city to try to accurately evaluate what its groundwater resources actually are. Standard hydrologic methods of analyzing pumping rates and aquifer water pressure data are generally insufficient to evaluate sustainability of the hydrologic system because of transient effects and large uncertainties in estimates of aquifer recharge rates. Within the Espanola Basin, which includes Los Alamos and Santa Fe, NM, water levels in wells have been declining for several decades. A number of hydrologic studies within the basin have reported widely disparate estimated of aquifer recharge. This reflects the difficulty of making these estimates and, consequently, the difficulty of evaluating the sustainability of current pumping rates in the basin. There is evidence that significant surface subsidence has occurred, observed using satellite-based, interferometric Synthetic Aperture Radar (US Geologic Survey). In this study, we propose to use surface gravity measurements to estimate the change in mass of the water in the Espanola Basin aquifer and aquifer subsidence, and, combined with other water budget components in the basin, provide a more accurate estimated of total recharge. The ability of make such the extremely accurate gravity measurements necessary of such as study has only come about in the part 2-3 years; thus, this study is an innovative use of surface gravity measurements to an important hydrologic problem. The results of study not only provide an important constraints upon the aquifers that we investigate, but will also establish procedures which may be sued in investigation s of other aquifers. It will also serve to introduce hydrologists with the new geophysical tools that may be applied to hydrologic problems.

lang="EN-US">Develop multiphase, multi-component transfer and constitutive models, mathematically lang="EN-US">Model style="mso-fareast-language: run:yes"> model. style="mso-fareast-language: lang="EN-US">Estimation

6/2003 – 6/2005    Graduate Research Assistant, Los Alamos National Laboratory (LANL)

As a Graduate Research Assistant (GRA) in LANL, I conducted my PhD dissertation research on Stochastic Modeling of Multiphase Phase Flow In Heterogeneous Porous Media under instruction by Dongxiao Zhang (former senior scientist in LANL, now is a professor of petroleum engineering department in University of Oklahoma, and also the Associate Dean of college of engineering at Peking Univ. in China ) and Arturo Keller (Professor of Bren school of environmental sciences and management).

In addition, I attend the calibration work of site-scale flow modeling of Yucca mountain area for Yucca Mountain Project (YMP) supervised under George Zyvoloski (senior scientist in LANL，Developer of FEHM). The YMP is related to assessment of a radioactive waste repository in Yucca Mountain west of Las Vegas , Nevada .

I also review the work on Groundwater Pathway Model for the 2004/2005 MDA G area of Los Alamos National lab by Philip Stauffer et al. (scientist in LANL). MDA G area is a disposition site of low-radioactive waste from LANL. The accurate prediction of possible radioactive waste access of groundwater is very important for drinking water supply of Los Alamos County .

Stochastic modeling of multiphase flow in heterogeneous porous media. This research topic is for my PhD dissertation. Two papers have been published in Water Resources Research (WRR), and another two papers related to this research are under review. Nonaqueous phase liquids (NAPLs), such as chlorinated solvents, hydrocarbon fuels, have been used extensively in private industry, military installations and Department of Energy (DOE) facilities. NAPLs may be leaking from a damaged or decaying storage vessel (e.g. in a gasoline station, refinery, dry-cleaning operation), improperly constructed storage and distribution systems, a waster disposal lagoon, or may be spilt during transport and use in a manufacturing process (e.g., during degreasing of metal parts, in the electronics industry to clean semiconductors, or in an airfield for cleaning jet engines). NAPL spills during transport and leaks from underground storage tanks have inevitably occurred and represent a major risk to water supply, since even a small amount of NAPLs can contaminate large volumes of groundwater. NAPL blobs trapped in the porous soil or rock matrix at residual saturation are a continuous source of contamination to the aquifer. To design a remediation scheme for a polluted aquifer, it is important to understand how the NAPLs move and transfer in soil. It is known that the heterogeneity of soil properties (e.g. permeability, porosity) plays a major role in the distribution and transfer of the spill. Many stochastic approaches have been studies to deal with the heterogeneous variables, but few have been able to be applied in multiphase flow system (water, NAPL, gas phases) due to the nonlinear character of mathematic equations governing multiphase flow. In my research, a novel stochastic analysis method was applied in multiphase flow, and a numerical code was developed to demonstrate the applicability and validity of this method. This cutting-edge achievement is not only an important contribution to the pure academic research area of stochastic multiphase flow, but also provide a powerful tool for engineers in environmental pollution remediation designing, and petroleum reservoir simulations, where multiphase flow in heterogeneous porous media conditions are encountered.

　

Calibration of the site-scale saturated-zone flow model for Yucca Mountain . This work has been described in a report for LANL documentation, and presented in GSA 2004 meeting: Yucca Mountain Site-Scale Flow Model: Sensitivity to Hydrostratigraphy Model, Boundary Flux and Head Measurements. The Geological Society of America , 2004 Denver Annual Meeting (November 7-10, 2004), Colorado . Paper No. 38-9. The U.S. Department of Energy began studying Yucca Mountain , Nevada , in 1978 to determine whether it would be suitable for the nation's first long-term geologic repository for spent nuclear fuel and high-level radioactive waste. Currently stored at 126 sites around the nation, these materials are a result of nuclear power generation and national defense programs. On July 23, 2002, President Bush signed House Joint Resolution 87, allowing the DOE (Department of Energy) to take the next step in establishing a safe repository in which to store our nation's nuclear waste. Yucca Mountain is located in the Great Basin about 150 km northwest of Las Vegas , Nevada . To assess the performance of Yucca Mountain as a potential radionuclides repository site, a site-scale saturated-zone model is needed to understand the groundwater flow regime and to provide a basis for performance assessment calculations evaluating the potential risks to groundwater users down gradient from the site. Performance assessment results are likely to depend strongly on the specific discharge of groundwater leaving the potential repository area, as well as on the flow paths and the distribution of flow among the various hydrogeologic units. Finite Element Heat and Mass transfer code (FEHM) has been developed by George Zyvoloski et al. for Yucca Mountain Program. FEHM is designed to provide an analysis tool that facilitates understanding of flow in the aquifer beneath and down gradient from the repository. The flow model is also a computational tool to provide the flow fields for performing radionuclide migration predictions in the saturated zone. With the new field data (e.g. new discharge and recharge data) available in Yucca Mountain , this code has to be calibrated to reflect the latest information for its intended use. I passed several tough trainings to be qualified to participate in Yucca Mountain Program.

Review the LANL peer’s work titled Groundwater Pathway Model for the 2004/2005 MDA G PA. Phillip Stauffer et al. model the groundwater pathway of the radiological performance assessment (PA) for Material Disposal Area G (MDA G) at Los Alamos National Laboratory. Since its inception, LANL has disposed of radioactive and hazardous wastes in MDAs on the Pajarito Plateau where the laboratory is located. MDAs are mesa-top disposal facilities where waste was buried in pits and/or shafts with the intention that disposal was permanent. The purpose of this project is to provide the regulators and stakeholders with information about alternatives to (1) determine what corrective action will be proposed for an MDA, and (2) ensure that human health and the environment remain protected into the future. I was the primary reviewer of this project following nuclear quality assurance procedure and standards (AP-WFM-044,R.O.FMU6-SWO Quality Assurance Procedures). Cooperated by Phillip Stauffer, I checked the report systemically and address every key points (e.g. model inputs, liability of the field data) of the project. Several errors, which could results in potential assessments mistakes, were corrected. My work assured the quality and validity of this project.

9/2000 – 6/2003    Research Assistant, University of California , Santa Barbara

As a research assistant with my PhD advisor Arturo A. Keller at Bren school of environmental sciences and management, I modified a software code UTCHEM in research of flow and transport of organic pollutants in soil, which are presented in two papers:

Seasonal variation in bioavailability of residual NAPL in the vadose zone. AA Keller, M.J. Chen. Presented in Proceedings of the International Groundwater Symposium, March 25-28, 2002 in Berkeley , CA . I used modified simulator UTCHEM to investigate the influence of soil temperature and moisture in the natural attenuation of Non-Aqueous Phase Liquids (NAPL, such as gasoline) in subsurface area. Natural attenuation has become the management strategy most commonly considered for many pollutants due to the potential cost savings. Although many studies have shown natural attenuation to be an acceptable solution in many circumstances, some of the factors that control natural attenuation are not fully understood. For instance, it is common to assume constant site conditions throughout the life of the remediation project, without taking into consideration that the biogeochemical processes involved are strong functions of soil temperature and moisture. There could be considerable error in the time estimated to bring the site to closure, or pollutants might migrate beyond the expected boundaries, increasing the risk to potential receptors. To protect human and ecosystem health, the mechanisms involved in natural biodegradation must be understood, so that the necessary predictability and reliability can be achieved. This work is one the first attempts to achieve this goal.

Effect of spreading coefficient on three-phase relative permeability of non-aqueous phase liquids. Published in Water Resources Research: 39(10), 2003. Three-phase (water, oil, gas) flow conditions are encountered, for example, as light or dense non-aqueous phase liquid (LANPL or DNAPL, organic pollutants) migrates through the unsaturated zone, displacing air and water from soil pore space. Three-phase flow occurs in a number of environmental situations that are relevant for risk assessment, conceptual model definition, active remediation design, as well as monitored natural attenuation. It is important to consider three-phase flow conditions. There is a need to make accurate prediction of the time needed for a NAPL spill to reach groundwater, so that appropriate risk management actions can be undertaken. Lack of adequate methods for estimating the three-phase residual saturation often leads to inaccurate estimates and incorrect assessments. In this paper, we develop a novel approach to determine the long-term behavior of a NAPL spill. To illustrate it, we conducted numerical simulations using a modified version of UTCHEM.

9/1997-7/2000    Research Assistant, Peking (Beijing) University, Beijing, China

Projects

·        Three-Gorge Dam Project (Yangzi River, China, 1998)

·        Dapeng Bay Pollution Control Project (Shenzhen, South China, 1999)

Degree Thesis

Master Degree: ANN and its application to simulation of 2D fluid field (2000)

            Abstract (html)      Full text (pdf)    Modeling area (contour plot)

　